In [2]:
# Data
import numpy as np
from tensorflow.examples.tutorials.mnist import input_data
import impl.layer as l
# Dataset preparation and pre-processing
mnist = input_data.read_data_sets('data/MNIST_data/', one_hot=False)
X_train, y_train = mnist.train.images, mnist.train.labels
X_val, y_val = mnist.validation.images, mnist.validation.labels
X_test, y_test = mnist.test.images, mnist.test.labels
Extracting data/MNIST_data/train-images-idx3-ubyte.gz
Extracting data/MNIST_data/train-labels-idx1-ubyte.gz
Extracting data/MNIST_data/t10k-images-idx3-ubyte.gz
Extracting data/MNIST_data/t10k-labels-idx1-ubyte.gz
In [3]:
# Pre-processing: normalizing
def normalize(X):
# max scale for images 255= 2**8= 8 bit grayscale for each channel
return (X - X.mean(axis=0)) #/ X.std(axis=0)
X_train, X_val, X_test = normalize(X=X_train), normalize(X=X_val), normalize(X=X_test)
In [4]:
# Model
import impl.layer as l # or from impl.layer import *
from impl.loss import * # import all functions from impl.loss file # import impl.loss as loss_func
from sklearn.utils import shuffle as skshuffle
class FFNN:
def __init__(self, D, C, H, L):
self.L = L # layers
self.C = C # classes
self.losses = {'train':[], 'train_acc':[],
'valid':[], 'valid_acc':[],
'test':[], 'test_acc':[]}
self.model = []
self.W_fixed = []
self.grads = []
self.dy_prev = np.zeros((1, C))
self.y_prev = np.zeros((1, C))
low, high = -1, 1
# Input layer: weights/ biases
m = dict(W=np.random.uniform(size=(D, H), low=low, high=high) / np.sqrt(D / 2.), b=np.zeros((1, H)))
self.model.append(m)
# Fixed feedback weight
m = np.random.uniform(size=(D, H), low=low, high=high) / np.sqrt(D / 2.)
self.W_fixed.append(m)
# Input layer: gradients
self.grads.append({key: np.zeros_like(val) for key, val in self.model[0].items()})
# Hidden layers: weights/ biases
m_L = []
for _ in range(L):
m = dict(W=np.random.uniform(size=(H, H), low=low, high=high) / np.sqrt(H / 2.), b=np.zeros((1, H)))
m_L.append(m)
self.model.append(m_L)
# Fixed feedback weight
m_L = []
for _ in range(L):
m = np.random.uniform(size=(H, H), low=low, high=high) / np.sqrt(H / 2.)
m_L.append(m)
self.W_fixed.append(m_L)
# Hidden layer: gradients
grad_L = []
for _ in range(L):
grad_L.append({key: np.zeros_like(val) for key, val in self.model[1][0].items()})
self.grads.append(grad_L)
# Output layer: weights/ biases
m = dict(W=np.random.uniform(size=(H, C), low=low, high=high) / np.sqrt(H / 2.), b=np.zeros((1, C)))
self.model.append(m)
# Fixed feedback weight
m = np.random.uniform(size=(H, C), low=low, high=high) / np.sqrt(H / 2.)
self.W_fixed.append(m)
# Output layer: gradients
self.grads.append({key: np.zeros_like(val) for key, val in self.model[2].items()})
def fc_forward(self, X, W, b):
out = (X @ W) + b
cache = (W, X)
return out, cache
def fc_backward(self, dout, cache, W_fixed):
W, X = cache
dW = X.T @ dout
db = np.sum(dout, axis=0).reshape(1, -1) # db_1xn
# dX = dout @ W.T # vanilla Backprop
dX = dout @ W_fixed.T # fba backprop
return dX, dW, db
def train_forward(self, X, train):
caches, ys = [], []
# Input layer
y, fc_cache = self.fc_forward(X=X, W=self.model[0]['W'], b=self.model[0]['b']) # X_1xD, y_1xc
y, nl_cache = l.tanh_forward(X=y)
# y, nl_cache = l.sigmoid_forward(X=y) # non-linearity/ activation
# y -= l.sigmoid(0.0) # zero-centered/ mean
# y *= 2.0 # uni-var/ std
if train:
caches.append((fc_cache, nl_cache))
X = y.copy() # pass to the next layer
# Hidden layers
fc_caches, nl_caches = [], []
for layer in range(self.L):
y, fc_cache = self.fc_forward(X=X, W=self.model[1][layer]['W'], b=self.model[1][layer]['b'])
y, nl_cache = l.tanh_forward(X=y)
# y, nl_cache = l.sigmoid_forward(X=y) # non-linearity/ activation
# y -= l.sigmoid(0.0) # zero-centered/ mean
# y *= 2.0 # uni-var/ std
X = y.copy() # pass to next layer
if train:
fc_caches.append(fc_cache)
nl_caches.append(nl_cache)
if train:
caches.append((fc_caches, nl_caches)) # caches[1]
# Output layer
y, fc_cache = self.fc_forward(X=X, W=self.model[2]['W'], b=self.model[2]['b'])
y_prob = l.softmax(X=y)
if train:
caches.append(fc_cache)
return y_prob, caches # for backpropating the error
def cross_entropy(self, y_prob, y_train):
m = y_prob.shape[0]
# prob = l.softmax(y_pred)
log_like = -np.log(y_prob[range(m), y_train] + l.eps) # to avoid the devision by zero
data_loss = np.sum(log_like) / m
return data_loss
def dcross_entropy(self, y_prob, y_train): # this is equal for both since the reg_loss (noise) derivative is ZERO.
m = y_prob.shape[0]
# grad_y = l.softmax(y_pred)
grad_y = y_prob
grad_y[range(m), y_train] -= 1.
grad_y /= m
return grad_y
def loss_function(self, y_prob, y_train):
loss = self.cross_entropy(y_prob, y_train) # softmax is included
dy = self.dcross_entropy(y_prob, y_train) # dsoftmax is included
return loss, dy
def train_backward(self, dy, caches, y):
grads = self.grads.copy() # initialized by Zero in every iteration/epoch
# dy_prev = self.dy_prev.copy() # for temporal differencing
# self.dy_prev = dy.copy() # next iteration/ epoch
# y_prev = self.y_prev.copy() # for temporal differencing
# self.y_prev = y.copy() # next iteration/ epoch
# Output layer
fc_cache = caches[2]
# softmax_backward is included in dcross_entropy.
dX, dW, db = self.fc_backward(dout=dy, cache=fc_cache, W_fixed=self.W_fixed[2])
dy = dX.copy()
# # dy = dy @ self.W_fixed[2].T # done
# dy_prev = dy_prev @ self.W_fixed[2].T
# y = y @ self.W_fixed[2].T # done
# y_prev = y_prev @ self.W_fixed[2].T
grads[2]['W'] = dW
grads[2]['b'] = db
# Hidden layer
fc_caches, nl_caches = caches[1]
for layer in reversed(range(self.L)):
dy = l.tanh_backward(cache=nl_caches[layer], dout=dy) # diffable function
# dy = l.sigmoid_backward(cache=nl_caches[layer], dout=dy) # diffable function
# dy *= 2.0 # uni-var/ std
# dy *= dy - dy_prev # temporal diff instead of differentiable function
# dy *= y - y_prev # temporal diff instead of differentiable function
dX, dW, db = self.fc_backward(dout=dy, cache=fc_caches[layer], W_fixed=self.W_fixed[1][layer])
dy = dX.copy()
# # dy = dy @ self.W_fixed[2].T # done
# dy_prev = dy_prev @ self.W_fixed[1][layer].T
# y = y @ self.W_fixed[1][layer].T # done
# y_prev = y_prev @ self.W_fixed[1][layer].T
grads[1][layer]['W'] = dW
grads[1][layer]['b'] = db
# Input layer
fc_cache, nl_cache = caches[0]
dy = l.tanh_backward(cache=nl_cache, dout=dy) # diffable function
# dy = l.sigmoid_backward(cache=nl_caches[layer], dout=dy) # diffable function
# dy *= 2.0 # uni-var/ std
# dy *= dy - dy_prev # temporal diff instead of differentiable function
# dy *= y - y_prev # temporal diff instead of differentiable function
dX, dW, db = self.fc_backward(dout=dy, cache=fc_cache, W_fixed=self.W_fixed[0])
grads[0]['W'] = dW
grads[0]['b'] = db
return dX, grads
def test(self, X):
y_prob, _ = self.train_forward(X, train=False)
# if self.mode == 'classification':
y_pred = np.argmax(y_prob, axis=1) # for loss ==err
return y_pred, y_prob
def get_minibatch(self, X, y, minibatch_size, shuffle):
minibatches = []
if shuffle:
X, y = skshuffle(X, y)
for i in range(0, X.shape[0], minibatch_size):
X_mini = X[i:i + minibatch_size]
y_mini = y[i:i + minibatch_size]
minibatches.append((X_mini, y_mini))
return minibatches
def sgd(self, train_set, val_set, alpha, mb_size, n_iter, print_after):
X_train, y_train = train_set
X_val, y_val = val_set
# Epochs
for iter in range(1, n_iter + 1):
# Minibatches
minibatches = self.get_minibatch(X_train, y_train, mb_size, shuffle=True)
idx = np.random.randint(0, len(minibatches))
X_mini, y_mini = minibatches[idx]
# Train the model
y_prob, caches = self.train_forward(X_mini, train=True)
_, dy = self.loss_function(y_prob, y_mini)
_, grads = self.train_backward(dy, caches, y_prob)
# Update the model for input layer
for key in grads[0].keys():
self.model[0][key] -= alpha * grads[0][key]
# Update the model for the hidden layers
for layer in range(self.L):
for key in grads[1][layer].keys():
self.model[1][layer][key] -= alpha * grads[1][layer][key]
# Update the model for output layer
for key in grads[2].keys():
self.model[2][key] -= alpha * grads[2][key]
# Training accuracy
y_pred, y_prob = self.test(X_mini)
loss, _ = self.loss_function(y_prob, y_mini) # softmax is included in entropy loss function
self.losses['train'].append(loss)
acc = np.mean(y_pred == y_mini) # confusion matrix
self.losses['train_acc'].append(acc)
# Validate the updated model
y_pred, y_prob = self.test(X_val)
valid_loss, _ = self.loss_function(y_prob, y_val) # softmax is included in entropy loss function
self.losses['valid'].append(valid_loss)
valid_acc = np.mean(y_pred == y_val) # confusion matrix
self.losses['valid_acc'].append(valid_acc)
# Test the final model
y_pred, y_prob = nn.test(X_test)
test_loss, _ = self.loss_function(y_prob, y_test) # softmax is included in entropy loss function
self.losses['test'].append(test_loss)
test_acc = np.mean(y_pred == y_test)
self.losses['test_acc'].append(test_acc)
# print('Test accuracy mean: {:.4f}, std: {:.4f}, loss: {:.4f}'.
# format(acc.mean(), acc.std(), loss))
# Print the model info: loss & accuracy or err & acc
if iter % print_after == 0:
print('Iter-{}, train loss-{:.4f}, acc-{:.4f}, valid loss-{:.4f}, acc-{:.4f}, test loss-{:.4f}, acc-{:.4f}'.format(
iter, loss, acc, valid_loss, valid_acc, test_loss, test_acc))
In [5]:
# Hyper-parameters
n_iter = 100000 # number of epochs
alpha = 1e-3 # learning_rate
mb_size = 50 # 2**10==1024 # width, timestep for sequential data or minibatch size
print_after = 10 # n_iter//10 # print loss for train, valid, and test
num_hidden_units = 32 # number of kernels/ filters in each layer
num_input_units = X_train.shape[1] # noise added at the input lavel as input noise we can use dX or for more improvement
num_output_units = y_train.max() + 1 # number of classes in this classification problem
num_layers = 2 # depth
# Build the model/NN and learn it: running session.
nn = FFNN(C=num_output_units, D=num_input_units, H=num_hidden_units, L=num_layers)
nn.sgd(train_set=(X_train, y_train), val_set=(X_val, y_val), mb_size=mb_size, alpha=alpha,
n_iter=n_iter, print_after=print_after)
Iter-10, train loss-2.3087, acc-0.0800, valid loss-2.3034, acc-0.0894, test loss-2.3037, acc-0.0935
Iter-20, train loss-2.3167, acc-0.0400, valid loss-2.3032, acc-0.0902, test loss-2.3035, acc-0.0941
Iter-30, train loss-2.3211, acc-0.1000, valid loss-2.3031, acc-0.0904, test loss-2.3034, acc-0.0945
Iter-40, train loss-2.3090, acc-0.0400, valid loss-2.3029, acc-0.0900, test loss-2.3032, acc-0.0951
Iter-50, train loss-2.2886, acc-0.1000, valid loss-2.3028, acc-0.0904, test loss-2.3031, acc-0.0958
Iter-60, train loss-2.2755, acc-0.1600, valid loss-2.3026, acc-0.0908, test loss-2.3029, acc-0.0960
Iter-70, train loss-2.3107, acc-0.0400, valid loss-2.3024, acc-0.0914, test loss-2.3028, acc-0.0965
Iter-80, train loss-2.3248, acc-0.0800, valid loss-2.3023, acc-0.0920, test loss-2.3026, acc-0.0969
Iter-90, train loss-2.2951, acc-0.1400, valid loss-2.3021, acc-0.0924, test loss-2.3025, acc-0.0971
Iter-100, train loss-2.3158, acc-0.1200, valid loss-2.3020, acc-0.0928, test loss-2.3023, acc-0.0973
Iter-110, train loss-2.2853, acc-0.0800, valid loss-2.3017, acc-0.0926, test loss-2.3021, acc-0.0979
Iter-120, train loss-2.3339, acc-0.0200, valid loss-2.3015, acc-0.0936, test loss-2.3019, acc-0.0985
Iter-130, train loss-2.3035, acc-0.0600, valid loss-2.3014, acc-0.0940, test loss-2.3018, acc-0.0986
Iter-140, train loss-2.2974, acc-0.0800, valid loss-2.3012, acc-0.0940, test loss-2.3016, acc-0.0994
Iter-150, train loss-2.2914, acc-0.0400, valid loss-2.3010, acc-0.0946, test loss-2.3015, acc-0.1000
Iter-160, train loss-2.3146, acc-0.0800, valid loss-2.3009, acc-0.0946, test loss-2.3013, acc-0.0998
Iter-170, train loss-2.3023, acc-0.1200, valid loss-2.3008, acc-0.0952, test loss-2.3012, acc-0.1000
Iter-180, train loss-2.3064, acc-0.0800, valid loss-2.3007, acc-0.0952, test loss-2.3011, acc-0.1004
Iter-190, train loss-2.2980, acc-0.0600, valid loss-2.3006, acc-0.0952, test loss-2.3010, acc-0.1003
Iter-200, train loss-2.2935, acc-0.0800, valid loss-2.3004, acc-0.0950, test loss-2.3009, acc-0.0999
Iter-210, train loss-2.3385, acc-0.0800, valid loss-2.3002, acc-0.0954, test loss-2.3007, acc-0.1002
Iter-220, train loss-2.2961, acc-0.0800, valid loss-2.3001, acc-0.0952, test loss-2.3005, acc-0.1014
Iter-230, train loss-2.3010, acc-0.1000, valid loss-2.2999, acc-0.0966, test loss-2.3004, acc-0.1015
Iter-240, train loss-2.3017, acc-0.1200, valid loss-2.2998, acc-0.0970, test loss-2.3003, acc-0.1014
Iter-250, train loss-2.3145, acc-0.1600, valid loss-2.2997, acc-0.0976, test loss-2.3001, acc-0.1022
Iter-260, train loss-2.3161, acc-0.0000, valid loss-2.2995, acc-0.0984, test loss-2.3000, acc-0.1025
Iter-270, train loss-2.3089, acc-0.0600, valid loss-2.2993, acc-0.0992, test loss-2.2998, acc-0.1030
Iter-280, train loss-2.2921, acc-0.1400, valid loss-2.2992, acc-0.0990, test loss-2.2997, acc-0.1030
Iter-290, train loss-2.2853, acc-0.1600, valid loss-2.2990, acc-0.0990, test loss-2.2996, acc-0.1034
Iter-300, train loss-2.3082, acc-0.0800, valid loss-2.2988, acc-0.0992, test loss-2.2994, acc-0.1037
Iter-310, train loss-2.3059, acc-0.1400, valid loss-2.2986, acc-0.1004, test loss-2.2992, acc-0.1044
Iter-320, train loss-2.3028, acc-0.1000, valid loss-2.2985, acc-0.1010, test loss-2.2990, acc-0.1052
Iter-330, train loss-2.2986, acc-0.1200, valid loss-2.2983, acc-0.1012, test loss-2.2989, acc-0.1053
Iter-340, train loss-2.3267, acc-0.0200, valid loss-2.2982, acc-0.1014, test loss-2.2987, acc-0.1051
Iter-350, train loss-2.2711, acc-0.1200, valid loss-2.2980, acc-0.1010, test loss-2.2985, acc-0.1061
Iter-360, train loss-2.2926, acc-0.1400, valid loss-2.2978, acc-0.1008, test loss-2.2984, acc-0.1066
Iter-370, train loss-2.2860, acc-0.1400, valid loss-2.2976, acc-0.1014, test loss-2.2982, acc-0.1073
Iter-380, train loss-2.3043, acc-0.0800, valid loss-2.2974, acc-0.1022, test loss-2.2980, acc-0.1077
Iter-390, train loss-2.2965, acc-0.1000, valid loss-2.2972, acc-0.1032, test loss-2.2978, acc-0.1082
Iter-400, train loss-2.3420, acc-0.0800, valid loss-2.2970, acc-0.1038, test loss-2.2976, acc-0.1080
Iter-410, train loss-2.3014, acc-0.0800, valid loss-2.2969, acc-0.1036, test loss-2.2975, acc-0.1089
Iter-420, train loss-2.2828, acc-0.1000, valid loss-2.2967, acc-0.1036, test loss-2.2973, acc-0.1095
Iter-430, train loss-2.3298, acc-0.0600, valid loss-2.2966, acc-0.1042, test loss-2.2972, acc-0.1099
Iter-440, train loss-2.2887, acc-0.1400, valid loss-2.2964, acc-0.1042, test loss-2.2970, acc-0.1103
Iter-450, train loss-2.3116, acc-0.1000, valid loss-2.2962, acc-0.1046, test loss-2.2969, acc-0.1106
Iter-460, train loss-2.2823, acc-0.1800, valid loss-2.2961, acc-0.1050, test loss-2.2967, acc-0.1108
Iter-470, train loss-2.2986, acc-0.0800, valid loss-2.2959, acc-0.1050, test loss-2.2965, acc-0.1105
Iter-480, train loss-2.3110, acc-0.0400, valid loss-2.2957, acc-0.1058, test loss-2.2963, acc-0.1110
Iter-490, train loss-2.2814, acc-0.1000, valid loss-2.2955, acc-0.1056, test loss-2.2962, acc-0.1116
Iter-500, train loss-2.2997, acc-0.1400, valid loss-2.2953, acc-0.1052, test loss-2.2960, acc-0.1112
Iter-510, train loss-2.3006, acc-0.1000, valid loss-2.2952, acc-0.1052, test loss-2.2959, acc-0.1114
Iter-520, train loss-2.2967, acc-0.1200, valid loss-2.2951, acc-0.1052, test loss-2.2957, acc-0.1113
Iter-530, train loss-2.3086, acc-0.0600, valid loss-2.2948, acc-0.1056, test loss-2.2955, acc-0.1120
Iter-540, train loss-2.2803, acc-0.1800, valid loss-2.2946, acc-0.1056, test loss-2.2953, acc-0.1128
Iter-550, train loss-2.2837, acc-0.1800, valid loss-2.2945, acc-0.1062, test loss-2.2952, acc-0.1135
Iter-560, train loss-2.2870, acc-0.1200, valid loss-2.2943, acc-0.1072, test loss-2.2951, acc-0.1141
Iter-570, train loss-2.2958, acc-0.0800, valid loss-2.2941, acc-0.1082, test loss-2.2949, acc-0.1148
Iter-580, train loss-2.2945, acc-0.0800, valid loss-2.2939, acc-0.1092, test loss-2.2947, acc-0.1150
Iter-590, train loss-2.2815, acc-0.1000, valid loss-2.2938, acc-0.1104, test loss-2.2946, acc-0.1155
Iter-600, train loss-2.2847, acc-0.1800, valid loss-2.2936, acc-0.1112, test loss-2.2944, acc-0.1157
Iter-610, train loss-2.3059, acc-0.1400, valid loss-2.2934, acc-0.1124, test loss-2.2942, acc-0.1159
Iter-620, train loss-2.2914, acc-0.1000, valid loss-2.2932, acc-0.1136, test loss-2.2940, acc-0.1161
Iter-630, train loss-2.2894, acc-0.1000, valid loss-2.2931, acc-0.1134, test loss-2.2938, acc-0.1164
Iter-640, train loss-2.2978, acc-0.0800, valid loss-2.2928, acc-0.1138, test loss-2.2936, acc-0.1174
Iter-650, train loss-2.3020, acc-0.0400, valid loss-2.2926, acc-0.1140, test loss-2.2934, acc-0.1176
Iter-660, train loss-2.3051, acc-0.0400, valid loss-2.2924, acc-0.1146, test loss-2.2932, acc-0.1179
Iter-670, train loss-2.3004, acc-0.1400, valid loss-2.2923, acc-0.1144, test loss-2.2930, acc-0.1181
Iter-680, train loss-2.2806, acc-0.1200, valid loss-2.2920, acc-0.1154, test loss-2.2928, acc-0.1181
Iter-690, train loss-2.2858, acc-0.1400, valid loss-2.2918, acc-0.1166, test loss-2.2926, acc-0.1189
Iter-700, train loss-2.2899, acc-0.0800, valid loss-2.2917, acc-0.1172, test loss-2.2925, acc-0.1194
Iter-710, train loss-2.2912, acc-0.1000, valid loss-2.2914, acc-0.1168, test loss-2.2922, acc-0.1201
Iter-720, train loss-2.2699, acc-0.1600, valid loss-2.2912, acc-0.1182, test loss-2.2920, acc-0.1199
Iter-730, train loss-2.2705, acc-0.1800, valid loss-2.2910, acc-0.1190, test loss-2.2918, acc-0.1203
Iter-740, train loss-2.2843, acc-0.1200, valid loss-2.2907, acc-0.1196, test loss-2.2916, acc-0.1207
Iter-750, train loss-2.2977, acc-0.1000, valid loss-2.2905, acc-0.1202, test loss-2.2914, acc-0.1210
Iter-760, train loss-2.3022, acc-0.0800, valid loss-2.2903, acc-0.1204, test loss-2.2911, acc-0.1210
Iter-770, train loss-2.2885, acc-0.1600, valid loss-2.2901, acc-0.1212, test loss-2.2909, acc-0.1220
Iter-780, train loss-2.3183, acc-0.0600, valid loss-2.2899, acc-0.1214, test loss-2.2907, acc-0.1224
Iter-790, train loss-2.2771, acc-0.1200, valid loss-2.2897, acc-0.1222, test loss-2.2905, acc-0.1230
Iter-800, train loss-2.2976, acc-0.1000, valid loss-2.2895, acc-0.1222, test loss-2.2903, acc-0.1238
Iter-810, train loss-2.2672, acc-0.1600, valid loss-2.2893, acc-0.1224, test loss-2.2901, acc-0.1249
Iter-820, train loss-2.3045, acc-0.1200, valid loss-2.2891, acc-0.1226, test loss-2.2899, acc-0.1257
Iter-830, train loss-2.2870, acc-0.1400, valid loss-2.2889, acc-0.1234, test loss-2.2898, acc-0.1255
Iter-840, train loss-2.2806, acc-0.1400, valid loss-2.2886, acc-0.1250, test loss-2.2895, acc-0.1259
Iter-850, train loss-2.3080, acc-0.0600, valid loss-2.2884, acc-0.1264, test loss-2.2893, acc-0.1264
Iter-860, train loss-2.2623, acc-0.1400, valid loss-2.2881, acc-0.1274, test loss-2.2890, acc-0.1275
Iter-870, train loss-2.2885, acc-0.1600, valid loss-2.2879, acc-0.1286, test loss-2.2888, acc-0.1278
Iter-880, train loss-2.3148, acc-0.0600, valid loss-2.2876, acc-0.1300, test loss-2.2885, acc-0.1285
Iter-890, train loss-2.3312, acc-0.0800, valid loss-2.2874, acc-0.1304, test loss-2.2883, acc-0.1289
Iter-900, train loss-2.3034, acc-0.1400, valid loss-2.2872, acc-0.1312, test loss-2.2881, acc-0.1293
Iter-910, train loss-2.2864, acc-0.1200, valid loss-2.2869, acc-0.1316, test loss-2.2879, acc-0.1297
Iter-920, train loss-2.2692, acc-0.1400, valid loss-2.2867, acc-0.1320, test loss-2.2877, acc-0.1299
Iter-930, train loss-2.3073, acc-0.2200, valid loss-2.2865, acc-0.1328, test loss-2.2875, acc-0.1303
Iter-940, train loss-2.3002, acc-0.1600, valid loss-2.2863, acc-0.1342, test loss-2.2872, acc-0.1309
Iter-950, train loss-2.2860, acc-0.1600, valid loss-2.2860, acc-0.1350, test loss-2.2870, acc-0.1313
Iter-960, train loss-2.2949, acc-0.0600, valid loss-2.2858, acc-0.1352, test loss-2.2867, acc-0.1315
Iter-970, train loss-2.2881, acc-0.1000, valid loss-2.2855, acc-0.1368, test loss-2.2865, acc-0.1324
Iter-980, train loss-2.3042, acc-0.1000, valid loss-2.2853, acc-0.1370, test loss-2.2862, acc-0.1326
Iter-990, train loss-2.2806, acc-0.1200, valid loss-2.2850, acc-0.1372, test loss-2.2860, acc-0.1333
Iter-1000, train loss-2.2798, acc-0.1200, valid loss-2.2847, acc-0.1384, test loss-2.2857, acc-0.1332
Iter-1010, train loss-2.3017, acc-0.0600, valid loss-2.2845, acc-0.1394, test loss-2.2855, acc-0.1338
Iter-1020, train loss-2.2849, acc-0.0800, valid loss-2.2842, acc-0.1396, test loss-2.2852, acc-0.1341
Iter-1030, train loss-2.3234, acc-0.0600, valid loss-2.2839, acc-0.1410, test loss-2.2850, acc-0.1352
Iter-1040, train loss-2.2564, acc-0.2600, valid loss-2.2837, acc-0.1418, test loss-2.2847, acc-0.1357
Iter-1050, train loss-2.3013, acc-0.1000, valid loss-2.2834, acc-0.1420, test loss-2.2845, acc-0.1366
Iter-1060, train loss-2.2605, acc-0.2400, valid loss-2.2832, acc-0.1428, test loss-2.2842, acc-0.1377
Iter-1070, train loss-2.2765, acc-0.2000, valid loss-2.2829, acc-0.1428, test loss-2.2839, acc-0.1380
Iter-1080, train loss-2.2719, acc-0.1600, valid loss-2.2827, acc-0.1438, test loss-2.2837, acc-0.1388
Iter-1090, train loss-2.2588, acc-0.1400, valid loss-2.2823, acc-0.1448, test loss-2.2834, acc-0.1399
Iter-1100, train loss-2.2558, acc-0.1200, valid loss-2.2821, acc-0.1444, test loss-2.2831, acc-0.1409
Iter-1110, train loss-2.3203, acc-0.1600, valid loss-2.2818, acc-0.1446, test loss-2.2829, acc-0.1417
Iter-1120, train loss-2.2832, acc-0.1600, valid loss-2.2816, acc-0.1452, test loss-2.2826, acc-0.1420
Iter-1130, train loss-2.3109, acc-0.0800, valid loss-2.2813, acc-0.1458, test loss-2.2824, acc-0.1430
Iter-1140, train loss-2.2819, acc-0.1200, valid loss-2.2810, acc-0.1460, test loss-2.2821, acc-0.1433
Iter-1150, train loss-2.2933, acc-0.1000, valid loss-2.2808, acc-0.1462, test loss-2.2819, acc-0.1440
Iter-1160, train loss-2.2558, acc-0.1200, valid loss-2.2805, acc-0.1468, test loss-2.2815, acc-0.1445
Iter-1170, train loss-2.2662, acc-0.2400, valid loss-2.2802, acc-0.1480, test loss-2.2813, acc-0.1453
Iter-1180, train loss-2.2756, acc-0.1600, valid loss-2.2799, acc-0.1498, test loss-2.2810, acc-0.1464
Iter-1190, train loss-2.2819, acc-0.2000, valid loss-2.2795, acc-0.1516, test loss-2.2806, acc-0.1472
Iter-1200, train loss-2.3070, acc-0.1200, valid loss-2.2792, acc-0.1522, test loss-2.2803, acc-0.1475
Iter-1210, train loss-2.2550, acc-0.1800, valid loss-2.2789, acc-0.1540, test loss-2.2800, acc-0.1480
Iter-1220, train loss-2.2851, acc-0.2200, valid loss-2.2786, acc-0.1546, test loss-2.2797, acc-0.1489
Iter-1230, train loss-2.2751, acc-0.1200, valid loss-2.2782, acc-0.1556, test loss-2.2793, acc-0.1496
Iter-1240, train loss-2.2747, acc-0.1200, valid loss-2.2779, acc-0.1556, test loss-2.2790, acc-0.1500
Iter-1250, train loss-2.2898, acc-0.0800, valid loss-2.2776, acc-0.1556, test loss-2.2787, acc-0.1505
Iter-1260, train loss-2.2679, acc-0.1600, valid loss-2.2773, acc-0.1566, test loss-2.2784, acc-0.1512
Iter-1270, train loss-2.3143, acc-0.1000, valid loss-2.2769, acc-0.1572, test loss-2.2781, acc-0.1519
Iter-1280, train loss-2.2988, acc-0.1400, valid loss-2.2766, acc-0.1580, test loss-2.2777, acc-0.1524
Iter-1290, train loss-2.2894, acc-0.1200, valid loss-2.2763, acc-0.1586, test loss-2.2774, acc-0.1536
Iter-1300, train loss-2.2770, acc-0.0800, valid loss-2.2760, acc-0.1596, test loss-2.2771, acc-0.1547
Iter-1310, train loss-2.2513, acc-0.1800, valid loss-2.2756, acc-0.1606, test loss-2.2767, acc-0.1548
Iter-1320, train loss-2.2583, acc-0.1600, valid loss-2.2752, acc-0.1614, test loss-2.2764, acc-0.1556
Iter-1330, train loss-2.2959, acc-0.1200, valid loss-2.2749, acc-0.1616, test loss-2.2761, acc-0.1555
Iter-1340, train loss-2.2783, acc-0.1600, valid loss-2.2746, acc-0.1626, test loss-2.2757, acc-0.1568
Iter-1350, train loss-2.2863, acc-0.1600, valid loss-2.2742, acc-0.1636, test loss-2.2754, acc-0.1577
Iter-1360, train loss-2.2413, acc-0.2400, valid loss-2.2739, acc-0.1648, test loss-2.2751, acc-0.1579
Iter-1370, train loss-2.2719, acc-0.1200, valid loss-2.2736, acc-0.1666, test loss-2.2747, acc-0.1587
Iter-1380, train loss-2.2436, acc-0.2000, valid loss-2.2732, acc-0.1676, test loss-2.2744, acc-0.1590
Iter-1390, train loss-2.2534, acc-0.1800, valid loss-2.2728, acc-0.1678, test loss-2.2740, acc-0.1601
Iter-1400, train loss-2.2808, acc-0.1200, valid loss-2.2725, acc-0.1680, test loss-2.2737, acc-0.1609
Iter-1410, train loss-2.2366, acc-0.2800, valid loss-2.2721, acc-0.1690, test loss-2.2733, acc-0.1616
Iter-1420, train loss-2.2955, acc-0.1800, valid loss-2.2717, acc-0.1698, test loss-2.2729, acc-0.1622
Iter-1430, train loss-2.2595, acc-0.1600, valid loss-2.2713, acc-0.1702, test loss-2.2725, acc-0.1636
Iter-1440, train loss-2.2685, acc-0.1600, valid loss-2.2710, acc-0.1720, test loss-2.2722, acc-0.1639
Iter-1450, train loss-2.2778, acc-0.1800, valid loss-2.2706, acc-0.1724, test loss-2.2718, acc-0.1644
Iter-1460, train loss-2.2525, acc-0.1800, valid loss-2.2702, acc-0.1730, test loss-2.2714, acc-0.1650
Iter-1470, train loss-2.2887, acc-0.1200, valid loss-2.2698, acc-0.1738, test loss-2.2711, acc-0.1654
Iter-1480, train loss-2.2616, acc-0.1400, valid loss-2.2695, acc-0.1748, test loss-2.2707, acc-0.1659
Iter-1490, train loss-2.2659, acc-0.1400, valid loss-2.2692, acc-0.1752, test loss-2.2704, acc-0.1668
Iter-1500, train loss-2.2765, acc-0.1400, valid loss-2.2688, acc-0.1756, test loss-2.2701, acc-0.1673
Iter-1510, train loss-2.2721, acc-0.2400, valid loss-2.2685, acc-0.1754, test loss-2.2697, acc-0.1677
Iter-1520, train loss-2.2710, acc-0.1400, valid loss-2.2681, acc-0.1768, test loss-2.2693, acc-0.1683
Iter-1530, train loss-2.2258, acc-0.3200, valid loss-2.2677, acc-0.1768, test loss-2.2689, acc-0.1691
Iter-1540, train loss-2.2539, acc-0.2000, valid loss-2.2673, acc-0.1770, test loss-2.2686, acc-0.1699
Iter-1550, train loss-2.2775, acc-0.2200, valid loss-2.2670, acc-0.1778, test loss-2.2682, acc-0.1704
Iter-1560, train loss-2.2607, acc-0.2000, valid loss-2.2666, acc-0.1774, test loss-2.2678, acc-0.1708
Iter-1570, train loss-2.2581, acc-0.2000, valid loss-2.2662, acc-0.1782, test loss-2.2674, acc-0.1716
Iter-1580, train loss-2.2564, acc-0.1200, valid loss-2.2658, acc-0.1794, test loss-2.2670, acc-0.1725
Iter-1590, train loss-2.2581, acc-0.1200, valid loss-2.2653, acc-0.1800, test loss-2.2665, acc-0.1738
Iter-1600, train loss-2.2711, acc-0.1000, valid loss-2.2650, acc-0.1802, test loss-2.2662, acc-0.1744
Iter-1610, train loss-2.2771, acc-0.1600, valid loss-2.2646, acc-0.1804, test loss-2.2658, acc-0.1748
Iter-1620, train loss-2.3079, acc-0.0600, valid loss-2.2641, acc-0.1806, test loss-2.2654, acc-0.1750
Iter-1630, train loss-2.2719, acc-0.1400, valid loss-2.2636, acc-0.1818, test loss-2.2649, acc-0.1761
Iter-1640, train loss-2.2542, acc-0.2600, valid loss-2.2632, acc-0.1826, test loss-2.2644, acc-0.1769
Iter-1650, train loss-2.2474, acc-0.2000, valid loss-2.2627, acc-0.1826, test loss-2.2639, acc-0.1772
Iter-1660, train loss-2.2462, acc-0.2000, valid loss-2.2622, acc-0.1838, test loss-2.2635, acc-0.1786
Iter-1670, train loss-2.2654, acc-0.2200, valid loss-2.2618, acc-0.1850, test loss-2.2631, acc-0.1794
Iter-1680, train loss-2.2792, acc-0.0600, valid loss-2.2614, acc-0.1864, test loss-2.2626, acc-0.1803
Iter-1690, train loss-2.2890, acc-0.1400, valid loss-2.2610, acc-0.1874, test loss-2.2622, acc-0.1808
Iter-1700, train loss-2.2561, acc-0.2800, valid loss-2.2605, acc-0.1886, test loss-2.2617, acc-0.1814
Iter-1710, train loss-2.2712, acc-0.1600, valid loss-2.2600, acc-0.1894, test loss-2.2613, acc-0.1824
Iter-1720, train loss-2.2627, acc-0.1600, valid loss-2.2596, acc-0.1898, test loss-2.2609, acc-0.1831
Iter-1730, train loss-2.2527, acc-0.1600, valid loss-2.2592, acc-0.1902, test loss-2.2605, acc-0.1833
Iter-1740, train loss-2.2240, acc-0.2600, valid loss-2.2587, acc-0.1910, test loss-2.2600, acc-0.1842
Iter-1750, train loss-2.2708, acc-0.0800, valid loss-2.2582, acc-0.1924, test loss-2.2595, acc-0.1850
Iter-1760, train loss-2.2748, acc-0.2000, valid loss-2.2577, acc-0.1932, test loss-2.2590, acc-0.1863
Iter-1770, train loss-2.2630, acc-0.2400, valid loss-2.2572, acc-0.1934, test loss-2.2585, acc-0.1870
Iter-1780, train loss-2.2708, acc-0.1800, valid loss-2.2567, acc-0.1950, test loss-2.2580, acc-0.1877
Iter-1790, train loss-2.2463, acc-0.2800, valid loss-2.2562, acc-0.1958, test loss-2.2575, acc-0.1888
Iter-1800, train loss-2.2648, acc-0.1600, valid loss-2.2557, acc-0.1962, test loss-2.2570, acc-0.1896
Iter-1810, train loss-2.2578, acc-0.1600, valid loss-2.2553, acc-0.1974, test loss-2.2565, acc-0.1910
Iter-1820, train loss-2.2838, acc-0.1000, valid loss-2.2549, acc-0.1978, test loss-2.2561, acc-0.1912
Iter-1830, train loss-2.2497, acc-0.1600, valid loss-2.2543, acc-0.1982, test loss-2.2556, acc-0.1914
Iter-1840, train loss-2.2947, acc-0.1800, valid loss-2.2538, acc-0.1984, test loss-2.2551, acc-0.1917
Iter-1850, train loss-2.2539, acc-0.2000, valid loss-2.2533, acc-0.1992, test loss-2.2546, acc-0.1933
Iter-1860, train loss-2.2571, acc-0.2000, valid loss-2.2529, acc-0.2014, test loss-2.2541, acc-0.1938
Iter-1870, train loss-2.2742, acc-0.1600, valid loss-2.2524, acc-0.2014, test loss-2.2536, acc-0.1948
Iter-1880, train loss-2.2541, acc-0.2600, valid loss-2.2519, acc-0.2024, test loss-2.2531, acc-0.1953
Iter-1890, train loss-2.2308, acc-0.2400, valid loss-2.2513, acc-0.2034, test loss-2.2526, acc-0.1960
Iter-1900, train loss-2.2473, acc-0.2000, valid loss-2.2508, acc-0.2040, test loss-2.2521, acc-0.1964
Iter-1910, train loss-2.2728, acc-0.1400, valid loss-2.2502, acc-0.2050, test loss-2.2515, acc-0.1974
Iter-1920, train loss-2.2728, acc-0.1600, valid loss-2.2497, acc-0.2058, test loss-2.2510, acc-0.1982
Iter-1930, train loss-2.2333, acc-0.2400, valid loss-2.2492, acc-0.2066, test loss-2.2505, acc-0.1988
Iter-1940, train loss-2.2325, acc-0.2400, valid loss-2.2486, acc-0.2080, test loss-2.2498, acc-0.1994
Iter-1950, train loss-2.2105, acc-0.1800, valid loss-2.2480, acc-0.2084, test loss-2.2493, acc-0.2007
Iter-1960, train loss-2.2358, acc-0.2200, valid loss-2.2475, acc-0.2092, test loss-2.2488, acc-0.2017
Iter-1970, train loss-2.1978, acc-0.3000, valid loss-2.2470, acc-0.2100, test loss-2.2482, acc-0.2022
Iter-1980, train loss-2.2261, acc-0.2800, valid loss-2.2464, acc-0.2100, test loss-2.2477, acc-0.2031
Iter-1990, train loss-2.2728, acc-0.1600, valid loss-2.2459, acc-0.2108, test loss-2.2471, acc-0.2043
Iter-2000, train loss-2.2183, acc-0.3400, valid loss-2.2453, acc-0.2122, test loss-2.2466, acc-0.2050
Iter-2010, train loss-2.2544, acc-0.2600, valid loss-2.2448, acc-0.2138, test loss-2.2460, acc-0.2058
Iter-2020, train loss-2.2159, acc-0.2800, valid loss-2.2442, acc-0.2148, test loss-2.2455, acc-0.2067
Iter-2030, train loss-2.2450, acc-0.1800, valid loss-2.2437, acc-0.2160, test loss-2.2449, acc-0.2073
Iter-2040, train loss-2.2440, acc-0.2000, valid loss-2.2431, acc-0.2174, test loss-2.2443, acc-0.2088
Iter-2050, train loss-2.2404, acc-0.2200, valid loss-2.2425, acc-0.2176, test loss-2.2438, acc-0.2094
Iter-2060, train loss-2.2269, acc-0.2400, valid loss-2.2419, acc-0.2188, test loss-2.2432, acc-0.2101
Iter-2070, train loss-2.2653, acc-0.1000, valid loss-2.2414, acc-0.2194, test loss-2.2426, acc-0.2115
Iter-2080, train loss-2.2640, acc-0.1200, valid loss-2.2408, acc-0.2210, test loss-2.2421, acc-0.2118
Iter-2090, train loss-2.2699, acc-0.1200, valid loss-2.2402, acc-0.2218, test loss-2.2415, acc-0.2130
Iter-2100, train loss-2.2296, acc-0.2000, valid loss-2.2397, acc-0.2224, test loss-2.2409, acc-0.2137
Iter-2110, train loss-2.2322, acc-0.2000, valid loss-2.2390, acc-0.2238, test loss-2.2402, acc-0.2154
Iter-2120, train loss-2.2410, acc-0.2000, valid loss-2.2383, acc-0.2250, test loss-2.2396, acc-0.2162
Iter-2130, train loss-2.2684, acc-0.2000, valid loss-2.2378, acc-0.2252, test loss-2.2390, acc-0.2167
Iter-2140, train loss-2.2254, acc-0.2400, valid loss-2.2372, acc-0.2252, test loss-2.2384, acc-0.2179
Iter-2150, train loss-2.2414, acc-0.2000, valid loss-2.2365, acc-0.2264, test loss-2.2378, acc-0.2195
Iter-2160, train loss-2.2871, acc-0.1800, valid loss-2.2359, acc-0.2270, test loss-2.2371, acc-0.2202
Iter-2170, train loss-2.2761, acc-0.1000, valid loss-2.2352, acc-0.2274, test loss-2.2365, acc-0.2206
Iter-2180, train loss-2.2313, acc-0.2800, valid loss-2.2345, acc-0.2294, test loss-2.2358, acc-0.2223
Iter-2190, train loss-2.2323, acc-0.2000, valid loss-2.2339, acc-0.2300, test loss-2.2351, acc-0.2238
Iter-2200, train loss-2.2005, acc-0.2600, valid loss-2.2333, acc-0.2308, test loss-2.2345, acc-0.2246
Iter-2210, train loss-2.1839, acc-0.2800, valid loss-2.2326, acc-0.2316, test loss-2.2338, acc-0.2256
Iter-2220, train loss-2.2441, acc-0.2600, valid loss-2.2320, acc-0.2320, test loss-2.2332, acc-0.2264
Iter-2230, train loss-2.2147, acc-0.2600, valid loss-2.2313, acc-0.2330, test loss-2.2325, acc-0.2270
Iter-2240, train loss-2.2163, acc-0.2800, valid loss-2.2307, acc-0.2340, test loss-2.2319, acc-0.2278
Iter-2250, train loss-2.2150, acc-0.2400, valid loss-2.2301, acc-0.2348, test loss-2.2313, acc-0.2290
Iter-2260, train loss-2.2183, acc-0.2600, valid loss-2.2294, acc-0.2360, test loss-2.2306, acc-0.2301
Iter-2270, train loss-2.2415, acc-0.2400, valid loss-2.2288, acc-0.2366, test loss-2.2300, acc-0.2302
Iter-2280, train loss-2.2173, acc-0.2800, valid loss-2.2281, acc-0.2380, test loss-2.2293, acc-0.2311
Iter-2290, train loss-2.2450, acc-0.2600, valid loss-2.2274, acc-0.2384, test loss-2.2286, acc-0.2318
Iter-2300, train loss-2.2276, acc-0.2400, valid loss-2.2268, acc-0.2388, test loss-2.2280, acc-0.2324
Iter-2310, train loss-2.1915, acc-0.2800, valid loss-2.2261, acc-0.2402, test loss-2.2273, acc-0.2328
Iter-2320, train loss-2.2128, acc-0.2800, valid loss-2.2254, acc-0.2408, test loss-2.2266, acc-0.2347
Iter-2330, train loss-2.2118, acc-0.1000, valid loss-2.2248, acc-0.2416, test loss-2.2260, acc-0.2351
Iter-2340, train loss-2.1946, acc-0.2600, valid loss-2.2240, acc-0.2426, test loss-2.2252, acc-0.2365
Iter-2350, train loss-2.2293, acc-0.1800, valid loss-2.2234, acc-0.2428, test loss-2.2246, acc-0.2367
Iter-2360, train loss-2.2238, acc-0.2200, valid loss-2.2227, acc-0.2440, test loss-2.2239, acc-0.2370
Iter-2370, train loss-2.2296, acc-0.2600, valid loss-2.2220, acc-0.2446, test loss-2.2232, acc-0.2379
Iter-2380, train loss-2.2402, acc-0.2200, valid loss-2.2213, acc-0.2456, test loss-2.2225, acc-0.2382
Iter-2390, train loss-2.2055, acc-0.2400, valid loss-2.2207, acc-0.2466, test loss-2.2219, acc-0.2384
Iter-2400, train loss-2.1738, acc-0.2400, valid loss-2.2200, acc-0.2472, test loss-2.2212, acc-0.2398
Iter-2410, train loss-2.2004, acc-0.3400, valid loss-2.2192, acc-0.2480, test loss-2.2204, acc-0.2413
Iter-2420, train loss-2.2512, acc-0.1800, valid loss-2.2186, acc-0.2480, test loss-2.2198, acc-0.2426
Iter-2430, train loss-2.1800, acc-0.3000, valid loss-2.2178, acc-0.2486, test loss-2.2190, acc-0.2438
Iter-2440, train loss-2.2045, acc-0.2800, valid loss-2.2170, acc-0.2496, test loss-2.2182, acc-0.2455
Iter-2450, train loss-2.2005, acc-0.3000, valid loss-2.2162, acc-0.2506, test loss-2.2174, acc-0.2459
Iter-2460, train loss-2.2170, acc-0.2800, valid loss-2.2155, acc-0.2516, test loss-2.2167, acc-0.2467
Iter-2470, train loss-2.2467, acc-0.1200, valid loss-2.2148, acc-0.2524, test loss-2.2160, acc-0.2472
Iter-2480, train loss-2.2278, acc-0.2200, valid loss-2.2140, acc-0.2532, test loss-2.2152, acc-0.2478
Iter-2490, train loss-2.2161, acc-0.2200, valid loss-2.2132, acc-0.2542, test loss-2.2144, acc-0.2489
Iter-2500, train loss-2.2081, acc-0.2600, valid loss-2.2124, acc-0.2544, test loss-2.2136, acc-0.2497
Iter-2510, train loss-2.1748, acc-0.3600, valid loss-2.2117, acc-0.2552, test loss-2.2128, acc-0.2500
Iter-2520, train loss-2.2354, acc-0.1800, valid loss-2.2109, acc-0.2562, test loss-2.2120, acc-0.2510
Iter-2530, train loss-2.1859, acc-0.3800, valid loss-2.2101, acc-0.2562, test loss-2.2113, acc-0.2515
Iter-2540, train loss-2.1945, acc-0.2400, valid loss-2.2093, acc-0.2570, test loss-2.2105, acc-0.2526
Iter-2550, train loss-2.2456, acc-0.1800, valid loss-2.2086, acc-0.2572, test loss-2.2097, acc-0.2532
Iter-2560, train loss-2.2053, acc-0.2000, valid loss-2.2078, acc-0.2568, test loss-2.2089, acc-0.2542
Iter-2570, train loss-2.1509, acc-0.3600, valid loss-2.2070, acc-0.2586, test loss-2.2081, acc-0.2547
Iter-2580, train loss-2.1957, acc-0.3600, valid loss-2.2062, acc-0.2594, test loss-2.2073, acc-0.2557
Iter-2590, train loss-2.2307, acc-0.1800, valid loss-2.2053, acc-0.2610, test loss-2.2064, acc-0.2576
Iter-2600, train loss-2.2377, acc-0.2000, valid loss-2.2046, acc-0.2618, test loss-2.2057, acc-0.2583
Iter-2610, train loss-2.2207, acc-0.2800, valid loss-2.2038, acc-0.2624, test loss-2.2049, acc-0.2600
Iter-2620, train loss-2.1985, acc-0.3000, valid loss-2.2031, acc-0.2628, test loss-2.2042, acc-0.2612
Iter-2630, train loss-2.2202, acc-0.2000, valid loss-2.2023, acc-0.2630, test loss-2.2034, acc-0.2624
Iter-2640, train loss-2.1770, acc-0.3800, valid loss-2.2015, acc-0.2632, test loss-2.2025, acc-0.2636
Iter-2650, train loss-2.2214, acc-0.2000, valid loss-2.2006, acc-0.2650, test loss-2.2017, acc-0.2645
Iter-2660, train loss-2.1837, acc-0.3000, valid loss-2.1998, acc-0.2652, test loss-2.2009, acc-0.2651
Iter-2670, train loss-2.1726, acc-0.3600, valid loss-2.1989, acc-0.2660, test loss-2.2000, acc-0.2657
Iter-2680, train loss-2.1484, acc-0.3400, valid loss-2.1981, acc-0.2658, test loss-2.1992, acc-0.2662
Iter-2690, train loss-2.2023, acc-0.2200, valid loss-2.1973, acc-0.2668, test loss-2.1983, acc-0.2674
Iter-2700, train loss-2.1546, acc-0.3600, valid loss-2.1963, acc-0.2674, test loss-2.1974, acc-0.2687
Iter-2710, train loss-2.2108, acc-0.2200, valid loss-2.1955, acc-0.2688, test loss-2.1966, acc-0.2697
Iter-2720, train loss-2.1925, acc-0.2200, valid loss-2.1947, acc-0.2692, test loss-2.1957, acc-0.2701
Iter-2730, train loss-2.1897, acc-0.3200, valid loss-2.1939, acc-0.2704, test loss-2.1949, acc-0.2708
Iter-2740, train loss-2.2272, acc-0.1600, valid loss-2.1930, acc-0.2706, test loss-2.1941, acc-0.2713
Iter-2750, train loss-2.1616, acc-0.3600, valid loss-2.1922, acc-0.2708, test loss-2.1932, acc-0.2721
Iter-2760, train loss-2.1873, acc-0.2400, valid loss-2.1914, acc-0.2718, test loss-2.1924, acc-0.2733
Iter-2770, train loss-2.1390, acc-0.4600, valid loss-2.1905, acc-0.2730, test loss-2.1915, acc-0.2740
Iter-2780, train loss-2.2313, acc-0.2400, valid loss-2.1896, acc-0.2738, test loss-2.1906, acc-0.2747
Iter-2790, train loss-2.2146, acc-0.2000, valid loss-2.1888, acc-0.2744, test loss-2.1898, acc-0.2753
Iter-2800, train loss-2.2470, acc-0.2200, valid loss-2.1879, acc-0.2762, test loss-2.1889, acc-0.2755
Iter-2810, train loss-2.2103, acc-0.1600, valid loss-2.1871, acc-0.2770, test loss-2.1881, acc-0.2760
Iter-2820, train loss-2.2297, acc-0.1800, valid loss-2.1864, acc-0.2776, test loss-2.1874, acc-0.2765
Iter-2830, train loss-2.2227, acc-0.2200, valid loss-2.1855, acc-0.2782, test loss-2.1865, acc-0.2766
Iter-2840, train loss-2.1542, acc-0.3400, valid loss-2.1845, acc-0.2798, test loss-2.1856, acc-0.2779
Iter-2850, train loss-2.2085, acc-0.1600, valid loss-2.1836, acc-0.2808, test loss-2.1846, acc-0.2785
Iter-2860, train loss-2.1771, acc-0.2400, valid loss-2.1827, acc-0.2816, test loss-2.1837, acc-0.2790
Iter-2870, train loss-2.1791, acc-0.2400, valid loss-2.1818, acc-0.2822, test loss-2.1828, acc-0.2798
Iter-2880, train loss-2.1354, acc-0.3400, valid loss-2.1810, acc-0.2824, test loss-2.1819, acc-0.2798
Iter-2890, train loss-2.1677, acc-0.3000, valid loss-2.1801, acc-0.2830, test loss-2.1811, acc-0.2807
Iter-2900, train loss-2.1836, acc-0.2600, valid loss-2.1792, acc-0.2842, test loss-2.1801, acc-0.2817
Iter-2910, train loss-2.1649, acc-0.3200, valid loss-2.1783, acc-0.2846, test loss-2.1792, acc-0.2826
Iter-2920, train loss-2.1290, acc-0.3200, valid loss-2.1773, acc-0.2848, test loss-2.1783, acc-0.2833
Iter-2930, train loss-2.2012, acc-0.3000, valid loss-2.1765, acc-0.2854, test loss-2.1775, acc-0.2841
Iter-2940, train loss-2.1689, acc-0.3200, valid loss-2.1755, acc-0.2856, test loss-2.1764, acc-0.2847
Iter-2950, train loss-2.1912, acc-0.2600, valid loss-2.1746, acc-0.2864, test loss-2.1755, acc-0.2858
Iter-2960, train loss-2.2078, acc-0.2600, valid loss-2.1738, acc-0.2874, test loss-2.1747, acc-0.2869
Iter-2970, train loss-2.1829, acc-0.2800, valid loss-2.1729, acc-0.2884, test loss-2.1738, acc-0.2875
Iter-2980, train loss-2.1987, acc-0.2000, valid loss-2.1720, acc-0.2888, test loss-2.1729, acc-0.2878
Iter-2990, train loss-2.1380, acc-0.4200, valid loss-2.1711, acc-0.2902, test loss-2.1721, acc-0.2886
Iter-3000, train loss-2.2120, acc-0.1600, valid loss-2.1702, acc-0.2904, test loss-2.1712, acc-0.2895
Iter-3010, train loss-2.1685, acc-0.1800, valid loss-2.1693, acc-0.2906, test loss-2.1702, acc-0.2904
Iter-3020, train loss-2.1787, acc-0.3400, valid loss-2.1684, acc-0.2914, test loss-2.1694, acc-0.2913
Iter-3030, train loss-2.1784, acc-0.3000, valid loss-2.1676, acc-0.2922, test loss-2.1685, acc-0.2917
Iter-3040, train loss-2.2091, acc-0.2000, valid loss-2.1667, acc-0.2926, test loss-2.1677, acc-0.2923
Iter-3050, train loss-2.1548, acc-0.2600, valid loss-2.1658, acc-0.2930, test loss-2.1667, acc-0.2931
Iter-3060, train loss-2.2059, acc-0.2000, valid loss-2.1649, acc-0.2934, test loss-2.1658, acc-0.2938
Iter-3070, train loss-2.1539, acc-0.2600, valid loss-2.1639, acc-0.2944, test loss-2.1648, acc-0.2938
Iter-3080, train loss-2.1627, acc-0.3200, valid loss-2.1629, acc-0.2952, test loss-2.1638, acc-0.2948
Iter-3090, train loss-2.1836, acc-0.2800, valid loss-2.1619, acc-0.2964, test loss-2.1629, acc-0.2961
Iter-3100, train loss-2.1846, acc-0.2400, valid loss-2.1611, acc-0.2968, test loss-2.1620, acc-0.2966
Iter-3110, train loss-2.1615, acc-0.3200, valid loss-2.1600, acc-0.2972, test loss-2.1610, acc-0.2973
Iter-3120, train loss-2.1461, acc-0.3000, valid loss-2.1591, acc-0.2980, test loss-2.1600, acc-0.2984
Iter-3130, train loss-2.1781, acc-0.2400, valid loss-2.1582, acc-0.2990, test loss-2.1591, acc-0.2994
Iter-3140, train loss-2.1461, acc-0.3600, valid loss-2.1573, acc-0.2994, test loss-2.1582, acc-0.3006
Iter-3150, train loss-2.1125, acc-0.3000, valid loss-2.1562, acc-0.3000, test loss-2.1571, acc-0.3022
Iter-3160, train loss-2.1094, acc-0.3800, valid loss-2.1552, acc-0.3006, test loss-2.1561, acc-0.3031
Iter-3170, train loss-2.1055, acc-0.4400, valid loss-2.1542, acc-0.3016, test loss-2.1551, acc-0.3029
Iter-3180, train loss-2.1183, acc-0.3200, valid loss-2.1533, acc-0.3026, test loss-2.1542, acc-0.3038
Iter-3190, train loss-2.1565, acc-0.3200, valid loss-2.1522, acc-0.3030, test loss-2.1531, acc-0.3049
Iter-3200, train loss-2.1944, acc-0.2000, valid loss-2.1512, acc-0.3042, test loss-2.1521, acc-0.3057
Iter-3210, train loss-2.2125, acc-0.1400, valid loss-2.1503, acc-0.3050, test loss-2.1512, acc-0.3063
Iter-3220, train loss-2.1534, acc-0.2800, valid loss-2.1494, acc-0.3062, test loss-2.1503, acc-0.3073
Iter-3230, train loss-2.1751, acc-0.2800, valid loss-2.1484, acc-0.3066, test loss-2.1493, acc-0.3080
Iter-3240, train loss-2.1337, acc-0.3800, valid loss-2.1473, acc-0.3070, test loss-2.1482, acc-0.3086
Iter-3250, train loss-2.1486, acc-0.3600, valid loss-2.1463, acc-0.3074, test loss-2.1471, acc-0.3095
Iter-3260, train loss-2.1368, acc-0.2800, valid loss-2.1453, acc-0.3086, test loss-2.1462, acc-0.3103
Iter-3270, train loss-2.1541, acc-0.3400, valid loss-2.1443, acc-0.3098, test loss-2.1452, acc-0.3117
Iter-3280, train loss-2.1275, acc-0.3600, valid loss-2.1433, acc-0.3110, test loss-2.1441, acc-0.3129
Iter-3290, train loss-2.1588, acc-0.1800, valid loss-2.1423, acc-0.3130, test loss-2.1431, acc-0.3139
Iter-3300, train loss-2.1198, acc-0.3200, valid loss-2.1414, acc-0.3132, test loss-2.1422, acc-0.3150
Iter-3310, train loss-2.1923, acc-0.2400, valid loss-2.1404, acc-0.3144, test loss-2.1412, acc-0.3158
Iter-3320, train loss-2.2380, acc-0.2000, valid loss-2.1394, acc-0.3150, test loss-2.1402, acc-0.3170
Iter-3330, train loss-2.1609, acc-0.2600, valid loss-2.1384, acc-0.3166, test loss-2.1392, acc-0.3180
Iter-3340, train loss-2.0983, acc-0.3600, valid loss-2.1374, acc-0.3172, test loss-2.1383, acc-0.3194
Iter-3350, train loss-2.1048, acc-0.3600, valid loss-2.1364, acc-0.3184, test loss-2.1372, acc-0.3199
Iter-3360, train loss-2.1220, acc-0.3400, valid loss-2.1353, acc-0.3190, test loss-2.1361, acc-0.3216
Iter-3370, train loss-2.1327, acc-0.3400, valid loss-2.1343, acc-0.3208, test loss-2.1351, acc-0.3225
Iter-3380, train loss-2.1023, acc-0.4000, valid loss-2.1333, acc-0.3220, test loss-2.1341, acc-0.3240
Iter-3390, train loss-2.0900, acc-0.4000, valid loss-2.1322, acc-0.3234, test loss-2.1330, acc-0.3250
Iter-3400, train loss-2.1390, acc-0.2600, valid loss-2.1312, acc-0.3244, test loss-2.1320, acc-0.3261
Iter-3410, train loss-2.1247, acc-0.3600, valid loss-2.1302, acc-0.3258, test loss-2.1310, acc-0.3269
Iter-3420, train loss-2.1264, acc-0.3200, valid loss-2.1292, acc-0.3272, test loss-2.1300, acc-0.3284
Iter-3430, train loss-2.2245, acc-0.2000, valid loss-2.1281, acc-0.3292, test loss-2.1289, acc-0.3295
Iter-3440, train loss-2.0982, acc-0.3400, valid loss-2.1271, acc-0.3308, test loss-2.1279, acc-0.3306
Iter-3450, train loss-2.1879, acc-0.1800, valid loss-2.1261, acc-0.3326, test loss-2.1269, acc-0.3326
Iter-3460, train loss-2.1488, acc-0.2600, valid loss-2.1250, acc-0.3348, test loss-2.1258, acc-0.3331
Iter-3470, train loss-2.0860, acc-0.4200, valid loss-2.1239, acc-0.3362, test loss-2.1246, acc-0.3351
Iter-3480, train loss-2.1074, acc-0.2800, valid loss-2.1229, acc-0.3366, test loss-2.1236, acc-0.3361
Iter-3490, train loss-2.1187, acc-0.3400, valid loss-2.1219, acc-0.3374, test loss-2.1227, acc-0.3376
Iter-3500, train loss-2.1051, acc-0.3600, valid loss-2.1208, acc-0.3392, test loss-2.1216, acc-0.3396
Iter-3510, train loss-2.1149, acc-0.3600, valid loss-2.1198, acc-0.3408, test loss-2.1205, acc-0.3402
Iter-3520, train loss-2.1234, acc-0.4000, valid loss-2.1186, acc-0.3412, test loss-2.1193, acc-0.3409
Iter-3530, train loss-2.0712, acc-0.3000, valid loss-2.1174, acc-0.3426, test loss-2.1182, acc-0.3415
Iter-3540, train loss-2.1688, acc-0.2600, valid loss-2.1164, acc-0.3442, test loss-2.1171, acc-0.3424
Iter-3550, train loss-2.1849, acc-0.3000, valid loss-2.1153, acc-0.3446, test loss-2.1160, acc-0.3440
Iter-3560, train loss-2.0938, acc-0.4400, valid loss-2.1142, acc-0.3450, test loss-2.1149, acc-0.3443
Iter-3570, train loss-2.1397, acc-0.3400, valid loss-2.1131, acc-0.3462, test loss-2.1138, acc-0.3452
Iter-3580, train loss-2.1103, acc-0.3200, valid loss-2.1120, acc-0.3480, test loss-2.1127, acc-0.3473
Iter-3590, train loss-2.1951, acc-0.2800, valid loss-2.1109, acc-0.3490, test loss-2.1116, acc-0.3490
Iter-3600, train loss-2.1143, acc-0.3000, valid loss-2.1097, acc-0.3494, test loss-2.1104, acc-0.3493
Iter-3610, train loss-2.0942, acc-0.3200, valid loss-2.1086, acc-0.3494, test loss-2.1093, acc-0.3508
Iter-3620, train loss-2.0628, acc-0.3800, valid loss-2.1076, acc-0.3524, test loss-2.1083, acc-0.3511
Iter-3630, train loss-2.1327, acc-0.3200, valid loss-2.1066, acc-0.3532, test loss-2.1073, acc-0.3522
Iter-3640, train loss-2.1625, acc-0.2400, valid loss-2.1056, acc-0.3548, test loss-2.1062, acc-0.3535
Iter-3650, train loss-2.0898, acc-0.3000, valid loss-2.1045, acc-0.3558, test loss-2.1052, acc-0.3541
Iter-3660, train loss-2.1097, acc-0.3200, valid loss-2.1034, acc-0.3566, test loss-2.1041, acc-0.3551
Iter-3670, train loss-2.1139, acc-0.3400, valid loss-2.1023, acc-0.3580, test loss-2.1030, acc-0.3560
Iter-3680, train loss-2.0226, acc-0.5400, valid loss-2.1012, acc-0.3588, test loss-2.1019, acc-0.3569
Iter-3690, train loss-2.0825, acc-0.4600, valid loss-2.1001, acc-0.3594, test loss-2.1007, acc-0.3579
Iter-3700, train loss-2.0532, acc-0.4000, valid loss-2.0989, acc-0.3604, test loss-2.0996, acc-0.3590
Iter-3710, train loss-2.1328, acc-0.2600, valid loss-2.0979, acc-0.3616, test loss-2.0985, acc-0.3604
Iter-3720, train loss-2.0887, acc-0.4200, valid loss-2.0968, acc-0.3626, test loss-2.0974, acc-0.3617
Iter-3730, train loss-2.0970, acc-0.3000, valid loss-2.0957, acc-0.3630, test loss-2.0963, acc-0.3624
Iter-3740, train loss-2.1385, acc-0.3000, valid loss-2.0946, acc-0.3630, test loss-2.0952, acc-0.3638
Iter-3750, train loss-2.1717, acc-0.2600, valid loss-2.0935, acc-0.3634, test loss-2.0941, acc-0.3645
Iter-3760, train loss-2.0876, acc-0.3800, valid loss-2.0923, acc-0.3646, test loss-2.0929, acc-0.3654
Iter-3770, train loss-2.0576, acc-0.3400, valid loss-2.0911, acc-0.3658, test loss-2.0918, acc-0.3669
Iter-3780, train loss-2.1522, acc-0.3000, valid loss-2.0901, acc-0.3664, test loss-2.0907, acc-0.3666
Iter-3790, train loss-2.0906, acc-0.3600, valid loss-2.0889, acc-0.3664, test loss-2.0895, acc-0.3671
Iter-3800, train loss-2.0870, acc-0.4000, valid loss-2.0878, acc-0.3664, test loss-2.0884, acc-0.3678
Iter-3810, train loss-2.0301, acc-0.4400, valid loss-2.0867, acc-0.3676, test loss-2.0872, acc-0.3691
Iter-3820, train loss-2.0963, acc-0.3600, valid loss-2.0855, acc-0.3676, test loss-2.0860, acc-0.3699
Iter-3830, train loss-2.0546, acc-0.4000, valid loss-2.0842, acc-0.3684, test loss-2.0848, acc-0.3707
Iter-3840, train loss-2.0497, acc-0.4000, valid loss-2.0831, acc-0.3682, test loss-2.0836, acc-0.3714
Iter-3850, train loss-2.1011, acc-0.4400, valid loss-2.0819, acc-0.3684, test loss-2.0824, acc-0.3725
Iter-3860, train loss-2.0730, acc-0.3600, valid loss-2.0808, acc-0.3692, test loss-2.0814, acc-0.3735
Iter-3870, train loss-2.0719, acc-0.3600, valid loss-2.0797, acc-0.3698, test loss-2.0802, acc-0.3742
Iter-3880, train loss-2.0373, acc-0.4000, valid loss-2.0784, acc-0.3708, test loss-2.0789, acc-0.3758
Iter-3890, train loss-2.1495, acc-0.2200, valid loss-2.0772, acc-0.3724, test loss-2.0777, acc-0.3773
Iter-3900, train loss-2.0918, acc-0.3800, valid loss-2.0761, acc-0.3740, test loss-2.0766, acc-0.3774
Iter-3910, train loss-2.0560, acc-0.3600, valid loss-2.0749, acc-0.3748, test loss-2.0754, acc-0.3787
Iter-3920, train loss-2.1302, acc-0.3200, valid loss-2.0737, acc-0.3746, test loss-2.0742, acc-0.3795
Iter-3930, train loss-2.0657, acc-0.4600, valid loss-2.0726, acc-0.3756, test loss-2.0731, acc-0.3797
Iter-3940, train loss-2.1273, acc-0.3200, valid loss-2.0714, acc-0.3760, test loss-2.0719, acc-0.3804
Iter-3950, train loss-2.1720, acc-0.2400, valid loss-2.0703, acc-0.3772, test loss-2.0707, acc-0.3821
Iter-3960, train loss-2.0951, acc-0.3800, valid loss-2.0691, acc-0.3776, test loss-2.0696, acc-0.3831
Iter-3970, train loss-2.0692, acc-0.4000, valid loss-2.0680, acc-0.3776, test loss-2.0685, acc-0.3832
Iter-3980, train loss-2.0222, acc-0.4000, valid loss-2.0668, acc-0.3782, test loss-2.0672, acc-0.3839
Iter-3990, train loss-2.0094, acc-0.3200, valid loss-2.0656, acc-0.3790, test loss-2.0660, acc-0.3845
Iter-4000, train loss-2.0580, acc-0.3200, valid loss-2.0645, acc-0.3790, test loss-2.0649, acc-0.3847
Iter-4010, train loss-2.1160, acc-0.3800, valid loss-2.0633, acc-0.3794, test loss-2.0638, acc-0.3852
Iter-4020, train loss-2.0809, acc-0.4200, valid loss-2.0621, acc-0.3794, test loss-2.0626, acc-0.3853
Iter-4030, train loss-2.0344, acc-0.4000, valid loss-2.0610, acc-0.3802, test loss-2.0614, acc-0.3866
Iter-4040, train loss-2.0528, acc-0.3800, valid loss-2.0598, acc-0.3804, test loss-2.0602, acc-0.3876
Iter-4050, train loss-2.0206, acc-0.5000, valid loss-2.0587, acc-0.3814, test loss-2.0591, acc-0.3875
Iter-4060, train loss-2.0102, acc-0.4400, valid loss-2.0575, acc-0.3812, test loss-2.0580, acc-0.3877
Iter-4070, train loss-2.0351, acc-0.4800, valid loss-2.0564, acc-0.3814, test loss-2.0569, acc-0.3880
Iter-4080, train loss-2.1190, acc-0.3200, valid loss-2.0552, acc-0.3816, test loss-2.0557, acc-0.3884
Iter-4090, train loss-2.0532, acc-0.3400, valid loss-2.0540, acc-0.3814, test loss-2.0545, acc-0.3882
Iter-4100, train loss-2.0818, acc-0.3600, valid loss-2.0528, acc-0.3824, test loss-2.0533, acc-0.3898
Iter-4110, train loss-2.0366, acc-0.4200, valid loss-2.0516, acc-0.3834, test loss-2.0521, acc-0.3901
Iter-4120, train loss-2.0911, acc-0.3400, valid loss-2.0505, acc-0.3832, test loss-2.0509, acc-0.3903
Iter-4130, train loss-2.0465, acc-0.3600, valid loss-2.0493, acc-0.3848, test loss-2.0497, acc-0.3921
Iter-4140, train loss-2.0225, acc-0.3600, valid loss-2.0481, acc-0.3852, test loss-2.0485, acc-0.3924
Iter-4150, train loss-2.0431, acc-0.3800, valid loss-2.0468, acc-0.3856, test loss-2.0473, acc-0.3926
Iter-4160, train loss-2.0044, acc-0.4400, valid loss-2.0457, acc-0.3862, test loss-2.0461, acc-0.3933
Iter-4170, train loss-2.0425, acc-0.3600, valid loss-2.0445, acc-0.3868, test loss-2.0449, acc-0.3942
Iter-4180, train loss-1.9996, acc-0.4600, valid loss-2.0433, acc-0.3866, test loss-2.0437, acc-0.3946
Iter-4190, train loss-2.1092, acc-0.3200, valid loss-2.0421, acc-0.3870, test loss-2.0425, acc-0.3957
Iter-4200, train loss-2.0543, acc-0.3200, valid loss-2.0410, acc-0.3870, test loss-2.0414, acc-0.3961
Iter-4210, train loss-2.0546, acc-0.3600, valid loss-2.0397, acc-0.3868, test loss-2.0401, acc-0.3964
Iter-4220, train loss-2.0116, acc-0.3800, valid loss-2.0385, acc-0.3866, test loss-2.0389, acc-0.3958
Iter-4230, train loss-2.0183, acc-0.4200, valid loss-2.0373, acc-0.3866, test loss-2.0376, acc-0.3955
Iter-4240, train loss-2.0584, acc-0.3200, valid loss-2.0360, acc-0.3872, test loss-2.0364, acc-0.3956
Iter-4250, train loss-2.0104, acc-0.5000, valid loss-2.0348, acc-0.3872, test loss-2.0352, acc-0.3959
Iter-4260, train loss-2.0123, acc-0.3600, valid loss-2.0336, acc-0.3882, test loss-2.0340, acc-0.3972
Iter-4270, train loss-2.0052, acc-0.5400, valid loss-2.0324, acc-0.3876, test loss-2.0328, acc-0.3973
Iter-4280, train loss-2.0725, acc-0.3800, valid loss-2.0313, acc-0.3878, test loss-2.0316, acc-0.3976
Iter-4290, train loss-2.0658, acc-0.3600, valid loss-2.0301, acc-0.3884, test loss-2.0305, acc-0.3977
Iter-4300, train loss-1.9964, acc-0.4200, valid loss-2.0290, acc-0.3896, test loss-2.0293, acc-0.3978
Iter-4310, train loss-2.0544, acc-0.2600, valid loss-2.0278, acc-0.3898, test loss-2.0281, acc-0.3978
Iter-4320, train loss-1.9527, acc-0.4800, valid loss-2.0266, acc-0.3906, test loss-2.0269, acc-0.3982
Iter-4330, train loss-2.0369, acc-0.4000, valid loss-2.0253, acc-0.3914, test loss-2.0256, acc-0.3998
Iter-4340, train loss-1.9760, acc-0.4600, valid loss-2.0242, acc-0.3928, test loss-2.0245, acc-0.4003
Iter-4350, train loss-2.0262, acc-0.3600, valid loss-2.0229, acc-0.3934, test loss-2.0232, acc-0.4004
Iter-4360, train loss-2.0982, acc-0.3200, valid loss-2.0218, acc-0.3936, test loss-2.0221, acc-0.4003
Iter-4370, train loss-2.0455, acc-0.3600, valid loss-2.0207, acc-0.3940, test loss-2.0210, acc-0.4002
Iter-4380, train loss-2.0051, acc-0.3800, valid loss-2.0195, acc-0.3948, test loss-2.0198, acc-0.4020
Iter-4390, train loss-1.9756, acc-0.4800, valid loss-2.0182, acc-0.3956, test loss-2.0185, acc-0.4027
Iter-4400, train loss-1.9442, acc-0.5600, valid loss-2.0169, acc-0.3962, test loss-2.0172, acc-0.4032
Iter-4410, train loss-2.0068, acc-0.4400, valid loss-2.0157, acc-0.3968, test loss-2.0160, acc-0.4035
Iter-4420, train loss-2.1177, acc-0.3000, valid loss-2.0145, acc-0.3962, test loss-2.0147, acc-0.4037
Iter-4430, train loss-2.0737, acc-0.3000, valid loss-2.0133, acc-0.3958, test loss-2.0135, acc-0.4039
Iter-4440, train loss-1.9453, acc-0.4200, valid loss-2.0120, acc-0.3970, test loss-2.0122, acc-0.4044
Iter-4450, train loss-2.0190, acc-0.4400, valid loss-2.0108, acc-0.3966, test loss-2.0111, acc-0.4047
Iter-4460, train loss-1.9694, acc-0.4200, valid loss-2.0095, acc-0.3966, test loss-2.0098, acc-0.4047
Iter-4470, train loss-1.9898, acc-0.3600, valid loss-2.0083, acc-0.3970, test loss-2.0086, acc-0.4048
Iter-4480, train loss-2.0223, acc-0.4600, valid loss-2.0070, acc-0.3972, test loss-2.0072, acc-0.4047
Iter-4490, train loss-1.9881, acc-0.4000, valid loss-2.0057, acc-0.3972, test loss-2.0059, acc-0.4047
Iter-4500, train loss-1.9800, acc-0.4400, valid loss-2.0045, acc-0.3972, test loss-2.0047, acc-0.4050
Iter-4510, train loss-2.0105, acc-0.3600, valid loss-2.0033, acc-0.3982, test loss-2.0034, acc-0.4050
Iter-4520, train loss-2.0516, acc-0.3400, valid loss-2.0020, acc-0.3986, test loss-2.0022, acc-0.4057
Iter-4530, train loss-1.9273, acc-0.5200, valid loss-2.0009, acc-0.3988, test loss-2.0011, acc-0.4056
Iter-4540, train loss-1.9556, acc-0.4000, valid loss-1.9996, acc-0.3992, test loss-1.9998, acc-0.4069
Iter-4550, train loss-2.0356, acc-0.3200, valid loss-1.9984, acc-0.3996, test loss-1.9986, acc-0.4069
Iter-4560, train loss-1.9707, acc-0.3600, valid loss-1.9971, acc-0.3994, test loss-1.9973, acc-0.4074
Iter-4570, train loss-2.0047, acc-0.3200, valid loss-1.9959, acc-0.3996, test loss-1.9961, acc-0.4083
Iter-4580, train loss-2.0340, acc-0.4400, valid loss-1.9947, acc-0.3998, test loss-1.9948, acc-0.4084
Iter-4590, train loss-2.0799, acc-0.2600, valid loss-1.9936, acc-0.4000, test loss-1.9938, acc-0.4085
Iter-4600, train loss-1.9579, acc-0.5000, valid loss-1.9924, acc-0.4008, test loss-1.9925, acc-0.4088
Iter-4610, train loss-2.0147, acc-0.4400, valid loss-1.9912, acc-0.4012, test loss-1.9913, acc-0.4090
Iter-4620, train loss-1.8517, acc-0.5200, valid loss-1.9898, acc-0.4018, test loss-1.9899, acc-0.4096
Iter-4630, train loss-2.0082, acc-0.4400, valid loss-1.9886, acc-0.4016, test loss-1.9888, acc-0.4100
Iter-4640, train loss-2.0562, acc-0.3600, valid loss-1.9874, acc-0.4020, test loss-1.9875, acc-0.4100
Iter-4650, train loss-1.9651, acc-0.4200, valid loss-1.9862, acc-0.4020, test loss-1.9863, acc-0.4105
Iter-4660, train loss-1.9773, acc-0.3800, valid loss-1.9850, acc-0.4022, test loss-1.9852, acc-0.4109
Iter-4670, train loss-1.9928, acc-0.3200, valid loss-1.9838, acc-0.4038, test loss-1.9839, acc-0.4115
Iter-4680, train loss-1.9298, acc-0.3600, valid loss-1.9824, acc-0.4036, test loss-1.9826, acc-0.4117
Iter-4690, train loss-1.9788, acc-0.4200, valid loss-1.9812, acc-0.4038, test loss-1.9813, acc-0.4116
Iter-4700, train loss-1.9666, acc-0.3800, valid loss-1.9798, acc-0.4044, test loss-1.9799, acc-0.4128
Iter-4710, train loss-1.9801, acc-0.3400, valid loss-1.9787, acc-0.4046, test loss-1.9788, acc-0.4129
Iter-4720, train loss-1.9793, acc-0.4600, valid loss-1.9775, acc-0.4052, test loss-1.9776, acc-0.4139
Iter-4730, train loss-2.0486, acc-0.3400, valid loss-1.9762, acc-0.4048, test loss-1.9763, acc-0.4149
Iter-4740, train loss-1.9408, acc-0.3800, valid loss-1.9749, acc-0.4054, test loss-1.9750, acc-0.4158
Iter-4750, train loss-1.9870, acc-0.3800, valid loss-1.9736, acc-0.4056, test loss-1.9737, acc-0.4163
Iter-4760, train loss-1.9259, acc-0.3600, valid loss-1.9723, acc-0.4062, test loss-1.9724, acc-0.4167
Iter-4770, train loss-1.9598, acc-0.4400, valid loss-1.9711, acc-0.4062, test loss-1.9712, acc-0.4172
Iter-4780, train loss-1.9092, acc-0.4200, valid loss-1.9698, acc-0.4064, test loss-1.9698, acc-0.4172
Iter-4790, train loss-1.9518, acc-0.4000, valid loss-1.9686, acc-0.4072, test loss-1.9686, acc-0.4178
Iter-4800, train loss-1.9641, acc-0.4400, valid loss-1.9674, acc-0.4076, test loss-1.9674, acc-0.4181
Iter-4810, train loss-1.9819, acc-0.4200, valid loss-1.9662, acc-0.4088, test loss-1.9662, acc-0.4190
Iter-4820, train loss-2.0116, acc-0.4000, valid loss-1.9650, acc-0.4092, test loss-1.9651, acc-0.4194
Iter-4830, train loss-1.9768, acc-0.3400, valid loss-1.9639, acc-0.4092, test loss-1.9640, acc-0.4198
Iter-4840, train loss-1.9497, acc-0.4400, valid loss-1.9626, acc-0.4102, test loss-1.9627, acc-0.4204
Iter-4850, train loss-1.9106, acc-0.4800, valid loss-1.9613, acc-0.4104, test loss-1.9614, acc-0.4213
Iter-4860, train loss-1.9170, acc-0.5000, valid loss-1.9600, acc-0.4104, test loss-1.9601, acc-0.4220
Iter-4870, train loss-1.8606, acc-0.5400, valid loss-1.9587, acc-0.4110, test loss-1.9588, acc-0.4225
Iter-4880, train loss-1.9674, acc-0.3600, valid loss-1.9575, acc-0.4118, test loss-1.9576, acc-0.4230
Iter-4890, train loss-1.9626, acc-0.3600, valid loss-1.9562, acc-0.4120, test loss-1.9562, acc-0.4229
Iter-4900, train loss-1.9353, acc-0.4600, valid loss-1.9548, acc-0.4120, test loss-1.9549, acc-0.4228
Iter-4910, train loss-1.9590, acc-0.4600, valid loss-1.9535, acc-0.4120, test loss-1.9536, acc-0.4238
Iter-4920, train loss-1.9642, acc-0.4200, valid loss-1.9522, acc-0.4132, test loss-1.9523, acc-0.4244
Iter-4930, train loss-1.9430, acc-0.4800, valid loss-1.9511, acc-0.4130, test loss-1.9511, acc-0.4242
Iter-4940, train loss-2.0134, acc-0.3000, valid loss-1.9498, acc-0.4132, test loss-1.9498, acc-0.4241
Iter-4950, train loss-2.0539, acc-0.2600, valid loss-1.9486, acc-0.4138, test loss-1.9487, acc-0.4243
Iter-4960, train loss-1.8335, acc-0.4800, valid loss-1.9473, acc-0.4146, test loss-1.9473, acc-0.4246
Iter-4970, train loss-1.9934, acc-0.3200, valid loss-1.9460, acc-0.4148, test loss-1.9460, acc-0.4254
Iter-4980, train loss-1.9433, acc-0.4000, valid loss-1.9447, acc-0.4154, test loss-1.9447, acc-0.4263
Iter-4990, train loss-2.0643, acc-0.3200, valid loss-1.9434, acc-0.4156, test loss-1.9434, acc-0.4269
Iter-5000, train loss-1.9162, acc-0.4200, valid loss-1.9421, acc-0.4168, test loss-1.9421, acc-0.4277
Iter-5010, train loss-1.9453, acc-0.4800, valid loss-1.9408, acc-0.4172, test loss-1.9408, acc-0.4277
Iter-5020, train loss-1.9691, acc-0.4600, valid loss-1.9395, acc-0.4180, test loss-1.9394, acc-0.4280
Iter-5030, train loss-1.9034, acc-0.4400, valid loss-1.9382, acc-0.4186, test loss-1.9382, acc-0.4284
Iter-5040, train loss-1.8867, acc-0.4800, valid loss-1.9369, acc-0.4184, test loss-1.9368, acc-0.4290
Iter-5050, train loss-1.9602, acc-0.4000, valid loss-1.9357, acc-0.4186, test loss-1.9356, acc-0.4295
Iter-5060, train loss-1.9847, acc-0.4000, valid loss-1.9343, acc-0.4200, test loss-1.9342, acc-0.4302
Iter-5070, train loss-1.8950, acc-0.4200, valid loss-1.9330, acc-0.4206, test loss-1.9329, acc-0.4305
Iter-5080, train loss-1.8893, acc-0.4400, valid loss-1.9316, acc-0.4206, test loss-1.9315, acc-0.4310
Iter-5090, train loss-1.9529, acc-0.4000, valid loss-1.9303, acc-0.4208, test loss-1.9302, acc-0.4310
Iter-5100, train loss-1.9703, acc-0.3800, valid loss-1.9290, acc-0.4214, test loss-1.9289, acc-0.4315
Iter-5110, train loss-1.8939, acc-0.5200, valid loss-1.9277, acc-0.4218, test loss-1.9276, acc-0.4315
Iter-5120, train loss-1.8747, acc-0.4200, valid loss-1.9263, acc-0.4220, test loss-1.9262, acc-0.4318
Iter-5130, train loss-1.9075, acc-0.3800, valid loss-1.9251, acc-0.4224, test loss-1.9249, acc-0.4320
Iter-5140, train loss-1.8743, acc-0.4200, valid loss-1.9238, acc-0.4234, test loss-1.9237, acc-0.4326
Iter-5150, train loss-1.8145, acc-0.5600, valid loss-1.9225, acc-0.4248, test loss-1.9223, acc-0.4327
Iter-5160, train loss-1.9275, acc-0.4400, valid loss-1.9212, acc-0.4252, test loss-1.9210, acc-0.4331
Iter-5170, train loss-1.9298, acc-0.4000, valid loss-1.9198, acc-0.4256, test loss-1.9197, acc-0.4335
Iter-5180, train loss-1.9368, acc-0.4000, valid loss-1.9186, acc-0.4262, test loss-1.9184, acc-0.4337
Iter-5190, train loss-1.9641, acc-0.3800, valid loss-1.9173, acc-0.4268, test loss-1.9171, acc-0.4337
Iter-5200, train loss-1.8497, acc-0.4600, valid loss-1.9161, acc-0.4274, test loss-1.9158, acc-0.4335
Iter-5210, train loss-1.8310, acc-0.5400, valid loss-1.9147, acc-0.4278, test loss-1.9145, acc-0.4341
Iter-5220, train loss-1.8897, acc-0.4600, valid loss-1.9135, acc-0.4278, test loss-1.9133, acc-0.4346
Iter-5230, train loss-1.9007, acc-0.4600, valid loss-1.9123, acc-0.4280, test loss-1.9121, acc-0.4350
Iter-5240, train loss-1.9209, acc-0.4000, valid loss-1.9110, acc-0.4292, test loss-1.9108, acc-0.4354
Iter-5250, train loss-1.9140, acc-0.4000, valid loss-1.9098, acc-0.4296, test loss-1.9095, acc-0.4354
Iter-5260, train loss-1.9383, acc-0.3800, valid loss-1.9085, acc-0.4298, test loss-1.9082, acc-0.4355
Iter-5270, train loss-1.9476, acc-0.5000, valid loss-1.9072, acc-0.4304, test loss-1.9070, acc-0.4355
Iter-5280, train loss-1.8618, acc-0.4600, valid loss-1.9059, acc-0.4310, test loss-1.9057, acc-0.4359
Iter-5290, train loss-1.9481, acc-0.4200, valid loss-1.9046, acc-0.4318, test loss-1.9044, acc-0.4365
Iter-5300, train loss-1.9209, acc-0.3600, valid loss-1.9033, acc-0.4322, test loss-1.9031, acc-0.4370
Iter-5310, train loss-1.8594, acc-0.5000, valid loss-1.9021, acc-0.4322, test loss-1.9018, acc-0.4371
Iter-5320, train loss-1.9212, acc-0.4200, valid loss-1.9008, acc-0.4328, test loss-1.9006, acc-0.4380
Iter-5330, train loss-1.8700, acc-0.5400, valid loss-1.8996, acc-0.4330, test loss-1.8993, acc-0.4380
Iter-5340, train loss-1.8279, acc-0.5400, valid loss-1.8982, acc-0.4332, test loss-1.8980, acc-0.4387
Iter-5350, train loss-1.8914, acc-0.4200, valid loss-1.8970, acc-0.4330, test loss-1.8967, acc-0.4388
Iter-5360, train loss-1.9064, acc-0.3800, valid loss-1.8958, acc-0.4330, test loss-1.8955, acc-0.4389
Iter-5370, train loss-1.8735, acc-0.4400, valid loss-1.8945, acc-0.4330, test loss-1.8942, acc-0.4387
Iter-5380, train loss-1.8520, acc-0.4600, valid loss-1.8933, acc-0.4332, test loss-1.8930, acc-0.4389
Iter-5390, train loss-1.9508, acc-0.3800, valid loss-1.8920, acc-0.4332, test loss-1.8917, acc-0.4392
Iter-5400, train loss-1.8657, acc-0.4800, valid loss-1.8908, acc-0.4336, test loss-1.8904, acc-0.4393
Iter-5410, train loss-1.9140, acc-0.3200, valid loss-1.8895, acc-0.4344, test loss-1.8892, acc-0.4397
Iter-5420, train loss-1.9379, acc-0.4200, valid loss-1.8885, acc-0.4348, test loss-1.8881, acc-0.4396
Iter-5430, train loss-1.9028, acc-0.4200, valid loss-1.8872, acc-0.4356, test loss-1.8868, acc-0.4399
Iter-5440, train loss-1.9186, acc-0.4000, valid loss-1.8858, acc-0.4366, test loss-1.8855, acc-0.4399
Iter-5450, train loss-1.8519, acc-0.4800, valid loss-1.8846, acc-0.4374, test loss-1.8843, acc-0.4405
Iter-5460, train loss-1.8872, acc-0.3600, valid loss-1.8832, acc-0.4380, test loss-1.8829, acc-0.4410
Iter-5470, train loss-1.8782, acc-0.5200, valid loss-1.8818, acc-0.4380, test loss-1.8815, acc-0.4416
Iter-5480, train loss-1.8313, acc-0.4600, valid loss-1.8806, acc-0.4386, test loss-1.8802, acc-0.4421
Iter-5490, train loss-1.7832, acc-0.5600, valid loss-1.8794, acc-0.4388, test loss-1.8790, acc-0.4421
Iter-5500, train loss-1.8794, acc-0.4200, valid loss-1.8782, acc-0.4396, test loss-1.8778, acc-0.4424
Iter-5510, train loss-1.9097, acc-0.5000, valid loss-1.8768, acc-0.4400, test loss-1.8765, acc-0.4427
Iter-5520, train loss-1.8881, acc-0.3800, valid loss-1.8757, acc-0.4402, test loss-1.8753, acc-0.4431
Iter-5530, train loss-1.8920, acc-0.4600, valid loss-1.8744, acc-0.4406, test loss-1.8740, acc-0.4432
Iter-5540, train loss-1.8678, acc-0.4200, valid loss-1.8731, acc-0.4408, test loss-1.8727, acc-0.4434
Iter-5550, train loss-1.9125, acc-0.4600, valid loss-1.8718, acc-0.4412, test loss-1.8714, acc-0.4439
Iter-5560, train loss-1.7998, acc-0.5600, valid loss-1.8706, acc-0.4406, test loss-1.8702, acc-0.4440
Iter-5570, train loss-1.8952, acc-0.4000, valid loss-1.8693, acc-0.4416, test loss-1.8689, acc-0.4443
Iter-5580, train loss-1.8908, acc-0.3400, valid loss-1.8681, acc-0.4418, test loss-1.8677, acc-0.4446
Iter-5590, train loss-1.8135, acc-0.4400, valid loss-1.8668, acc-0.4426, test loss-1.8664, acc-0.4450
Iter-5600, train loss-1.7701, acc-0.5400, valid loss-1.8656, acc-0.4434, test loss-1.8651, acc-0.4454
Iter-5610, train loss-1.8567, acc-0.4600, valid loss-1.8643, acc-0.4430, test loss-1.8638, acc-0.4458
Iter-5620, train loss-1.7770, acc-0.5400, valid loss-1.8630, acc-0.4440, test loss-1.8626, acc-0.4462
Iter-5630, train loss-1.7991, acc-0.4600, valid loss-1.8618, acc-0.4448, test loss-1.8613, acc-0.4464
Iter-5640, train loss-1.7865, acc-0.4800, valid loss-1.8605, acc-0.4450, test loss-1.8600, acc-0.4468
Iter-5650, train loss-1.8953, acc-0.3800, valid loss-1.8592, acc-0.4454, test loss-1.8587, acc-0.4474
Iter-5660, train loss-1.8275, acc-0.4800, valid loss-1.8579, acc-0.4454, test loss-1.8574, acc-0.4476
Iter-5670, train loss-1.8668, acc-0.4800, valid loss-1.8566, acc-0.4464, test loss-1.8561, acc-0.4479
Iter-5680, train loss-1.7938, acc-0.5600, valid loss-1.8554, acc-0.4466, test loss-1.8549, acc-0.4482
Iter-5690, train loss-1.8626, acc-0.5000, valid loss-1.8542, acc-0.4468, test loss-1.8537, acc-0.4486
Iter-5700, train loss-1.8950, acc-0.3800, valid loss-1.8530, acc-0.4468, test loss-1.8525, acc-0.4490
Iter-5710, train loss-1.8647, acc-0.4200, valid loss-1.8517, acc-0.4466, test loss-1.8512, acc-0.4494
Iter-5720, train loss-1.7402, acc-0.5200, valid loss-1.8505, acc-0.4464, test loss-1.8499, acc-0.4495
Iter-5730, train loss-1.8920, acc-0.5000, valid loss-1.8491, acc-0.4466, test loss-1.8486, acc-0.4501
Iter-5740, train loss-1.9451, acc-0.3000, valid loss-1.8480, acc-0.4468, test loss-1.8474, acc-0.4503
Iter-5750, train loss-1.8445, acc-0.4800, valid loss-1.8468, acc-0.4466, test loss-1.8462, acc-0.4505
Iter-5760, train loss-1.9211, acc-0.3400, valid loss-1.8456, acc-0.4470, test loss-1.8450, acc-0.4509
Iter-5770, train loss-1.8676, acc-0.5200, valid loss-1.8443, acc-0.4470, test loss-1.8438, acc-0.4513
Iter-5780, train loss-1.6691, acc-0.6200, valid loss-1.8430, acc-0.4474, test loss-1.8424, acc-0.4515
Iter-5790, train loss-1.8095, acc-0.4800, valid loss-1.8417, acc-0.4480, test loss-1.8411, acc-0.4516
Iter-5800, train loss-1.7988, acc-0.4200, valid loss-1.8404, acc-0.4486, test loss-1.8398, acc-0.4523
Iter-5810, train loss-1.7962, acc-0.4600, valid loss-1.8391, acc-0.4488, test loss-1.8384, acc-0.4525
Iter-5820, train loss-1.8772, acc-0.4200, valid loss-1.8377, acc-0.4492, test loss-1.8371, acc-0.4531
Iter-5830, train loss-1.7606, acc-0.5600, valid loss-1.8364, acc-0.4494, test loss-1.8358, acc-0.4531
Iter-5840, train loss-1.8301, acc-0.5400, valid loss-1.8351, acc-0.4498, test loss-1.8344, acc-0.4536
Iter-5850, train loss-1.8376, acc-0.4800, valid loss-1.8339, acc-0.4498, test loss-1.8332, acc-0.4536
Iter-5860, train loss-1.7214, acc-0.5600, valid loss-1.8325, acc-0.4504, test loss-1.8319, acc-0.4538
Iter-5870, train loss-1.8523, acc-0.5000, valid loss-1.8312, acc-0.4510, test loss-1.8306, acc-0.4539
Iter-5880, train loss-1.8432, acc-0.4600, valid loss-1.8299, acc-0.4516, test loss-1.8292, acc-0.4540
Iter-5890, train loss-1.8041, acc-0.4800, valid loss-1.8286, acc-0.4518, test loss-1.8279, acc-0.4543
Iter-5900, train loss-1.8861, acc-0.4000, valid loss-1.8272, acc-0.4518, test loss-1.8266, acc-0.4550
Iter-5910, train loss-1.7648, acc-0.5600, valid loss-1.8259, acc-0.4520, test loss-1.8252, acc-0.4554
Iter-5920, train loss-1.8172, acc-0.4600, valid loss-1.8246, acc-0.4522, test loss-1.8239, acc-0.4556
Iter-5930, train loss-1.8612, acc-0.5000, valid loss-1.8232, acc-0.4520, test loss-1.8226, acc-0.4562
Iter-5940, train loss-1.7090, acc-0.4800, valid loss-1.8219, acc-0.4520, test loss-1.8212, acc-0.4563
Iter-5950, train loss-1.8866, acc-0.4400, valid loss-1.8206, acc-0.4526, test loss-1.8200, acc-0.4567
Iter-5960, train loss-1.7771, acc-0.4600, valid loss-1.8194, acc-0.4530, test loss-1.8187, acc-0.4573
Iter-5970, train loss-1.7715, acc-0.5400, valid loss-1.8181, acc-0.4542, test loss-1.8174, acc-0.4574
Iter-5980, train loss-1.8063, acc-0.4200, valid loss-1.8168, acc-0.4542, test loss-1.8161, acc-0.4577
Iter-5990, train loss-1.7812, acc-0.5000, valid loss-1.8155, acc-0.4548, test loss-1.8148, acc-0.4581
Iter-6000, train loss-1.8039, acc-0.5000, valid loss-1.8143, acc-0.4548, test loss-1.8135, acc-0.4587
Iter-6010, train loss-1.7796, acc-0.4600, valid loss-1.8130, acc-0.4556, test loss-1.8123, acc-0.4588
Iter-6020, train loss-1.7599, acc-0.5200, valid loss-1.8118, acc-0.4560, test loss-1.8110, acc-0.4593
Iter-6030, train loss-1.7319, acc-0.5600, valid loss-1.8105, acc-0.4564, test loss-1.8097, acc-0.4591
Iter-6040, train loss-1.7746, acc-0.5400, valid loss-1.8092, acc-0.4568, test loss-1.8084, acc-0.4588
Iter-6050, train loss-1.8878, acc-0.4400, valid loss-1.8079, acc-0.4572, test loss-1.8071, acc-0.4593
Iter-6060, train loss-1.8040, acc-0.5200, valid loss-1.8067, acc-0.4574, test loss-1.8059, acc-0.4598
Iter-6070, train loss-1.8534, acc-0.4000, valid loss-1.8054, acc-0.4576, test loss-1.8045, acc-0.4606
Iter-6080, train loss-1.6964, acc-0.4800, valid loss-1.8041, acc-0.4578, test loss-1.8032, acc-0.4610
Iter-6090, train loss-1.7560, acc-0.5800, valid loss-1.8029, acc-0.4582, test loss-1.8020, acc-0.4608
Iter-6100, train loss-1.8031, acc-0.4200, valid loss-1.8016, acc-0.4584, test loss-1.8007, acc-0.4616
Iter-6110, train loss-1.7642, acc-0.5200, valid loss-1.8003, acc-0.4588, test loss-1.7995, acc-0.4617
Iter-6120, train loss-1.7563, acc-0.5000, valid loss-1.7990, acc-0.4590, test loss-1.7982, acc-0.4621
Iter-6130, train loss-1.7524, acc-0.4800, valid loss-1.7977, acc-0.4596, test loss-1.7969, acc-0.4627
Iter-6140, train loss-1.7900, acc-0.5200, valid loss-1.7965, acc-0.4600, test loss-1.7956, acc-0.4630
Iter-6150, train loss-1.7498, acc-0.4400, valid loss-1.7952, acc-0.4604, test loss-1.7944, acc-0.4632
Iter-6160, train loss-1.7665, acc-0.4800, valid loss-1.7941, acc-0.4604, test loss-1.7932, acc-0.4635
Iter-6170, train loss-1.7650, acc-0.4600, valid loss-1.7928, acc-0.4604, test loss-1.7920, acc-0.4645
Iter-6180, train loss-1.7738, acc-0.5200, valid loss-1.7914, acc-0.4606, test loss-1.7905, acc-0.4645
Iter-6190, train loss-1.7041, acc-0.5000, valid loss-1.7902, acc-0.4606, test loss-1.7893, acc-0.4649
Iter-6200, train loss-1.8940, acc-0.3800, valid loss-1.7890, acc-0.4612, test loss-1.7881, acc-0.4652
Iter-6210, train loss-1.7209, acc-0.5800, valid loss-1.7877, acc-0.4614, test loss-1.7868, acc-0.4656
Iter-6220, train loss-1.8385, acc-0.5000, valid loss-1.7864, acc-0.4616, test loss-1.7856, acc-0.4661
Iter-6230, train loss-1.7514, acc-0.5400, valid loss-1.7852, acc-0.4618, test loss-1.7843, acc-0.4663
Iter-6240, train loss-1.7656, acc-0.4400, valid loss-1.7839, acc-0.4622, test loss-1.7829, acc-0.4667
Iter-6250, train loss-1.8025, acc-0.3800, valid loss-1.7826, acc-0.4622, test loss-1.7817, acc-0.4671
Iter-6260, train loss-1.7590, acc-0.4400, valid loss-1.7814, acc-0.4624, test loss-1.7805, acc-0.4671
Iter-6270, train loss-1.7987, acc-0.4200, valid loss-1.7801, acc-0.4626, test loss-1.7792, acc-0.4670
Iter-6280, train loss-1.7545, acc-0.4400, valid loss-1.7789, acc-0.4630, test loss-1.7780, acc-0.4673
Iter-6290, train loss-1.7794, acc-0.4600, valid loss-1.7776, acc-0.4634, test loss-1.7767, acc-0.4676
Iter-6300, train loss-1.7775, acc-0.5000, valid loss-1.7763, acc-0.4634, test loss-1.7754, acc-0.4682
Iter-6310, train loss-1.7487, acc-0.5800, valid loss-1.7751, acc-0.4638, test loss-1.7742, acc-0.4682
Iter-6320, train loss-1.7139, acc-0.4400, valid loss-1.7738, acc-0.4642, test loss-1.7729, acc-0.4689
Iter-6330, train loss-1.7542, acc-0.5000, valid loss-1.7726, acc-0.4648, test loss-1.7716, acc-0.4689
Iter-6340, train loss-1.7870, acc-0.4600, valid loss-1.7713, acc-0.4648, test loss-1.7704, acc-0.4694
Iter-6350, train loss-1.8309, acc-0.4800, valid loss-1.7700, acc-0.4650, test loss-1.7690, acc-0.4695
Iter-6360, train loss-1.7249, acc-0.4800, valid loss-1.7687, acc-0.4658, test loss-1.7678, acc-0.4700
Iter-6370, train loss-1.8172, acc-0.4200, valid loss-1.7675, acc-0.4664, test loss-1.7665, acc-0.4705
Iter-6380, train loss-1.7330, acc-0.5000, valid loss-1.7661, acc-0.4670, test loss-1.7652, acc-0.4709
Iter-6390, train loss-1.7651, acc-0.4400, valid loss-1.7649, acc-0.4674, test loss-1.7639, acc-0.4707
Iter-6400, train loss-1.8006, acc-0.4600, valid loss-1.7635, acc-0.4678, test loss-1.7625, acc-0.4713
Iter-6410, train loss-1.8186, acc-0.4000, valid loss-1.7622, acc-0.4686, test loss-1.7611, acc-0.4715
Iter-6420, train loss-1.7410, acc-0.5200, valid loss-1.7609, acc-0.4684, test loss-1.7599, acc-0.4718
Iter-6430, train loss-1.7380, acc-0.4800, valid loss-1.7597, acc-0.4692, test loss-1.7587, acc-0.4725
Iter-6440, train loss-1.7761, acc-0.4600, valid loss-1.7584, acc-0.4702, test loss-1.7573, acc-0.4728
Iter-6450, train loss-1.7822, acc-0.3800, valid loss-1.7570, acc-0.4710, test loss-1.7559, acc-0.4732
Iter-6460, train loss-1.6207, acc-0.5800, valid loss-1.7558, acc-0.4712, test loss-1.7547, acc-0.4737
Iter-6470, train loss-1.8584, acc-0.4200, valid loss-1.7546, acc-0.4710, test loss-1.7534, acc-0.4737
Iter-6480, train loss-1.7738, acc-0.4400, valid loss-1.7532, acc-0.4718, test loss-1.7521, acc-0.4746
Iter-6490, train loss-1.6919, acc-0.4800, valid loss-1.7520, acc-0.4724, test loss-1.7508, acc-0.4753
Iter-6500, train loss-1.6973, acc-0.5600, valid loss-1.7507, acc-0.4732, test loss-1.7496, acc-0.4761
Iter-6510, train loss-1.8736, acc-0.4000, valid loss-1.7495, acc-0.4742, test loss-1.7484, acc-0.4762
Iter-6520, train loss-1.8376, acc-0.3400, valid loss-1.7482, acc-0.4742, test loss-1.7471, acc-0.4764
Iter-6530, train loss-1.7338, acc-0.5400, valid loss-1.7470, acc-0.4746, test loss-1.7458, acc-0.4768
Iter-6540, train loss-1.5938, acc-0.6200, valid loss-1.7457, acc-0.4750, test loss-1.7445, acc-0.4771
Iter-6550, train loss-1.7404, acc-0.4800, valid loss-1.7444, acc-0.4750, test loss-1.7433, acc-0.4772
Iter-6560, train loss-1.7732, acc-0.5000, valid loss-1.7432, acc-0.4752, test loss-1.7420, acc-0.4775
Iter-6570, train loss-1.6677, acc-0.5200, valid loss-1.7419, acc-0.4752, test loss-1.7408, acc-0.4775
Iter-6580, train loss-1.7445, acc-0.5400, valid loss-1.7406, acc-0.4758, test loss-1.7395, acc-0.4778
Iter-6590, train loss-1.6115, acc-0.5800, valid loss-1.7393, acc-0.4762, test loss-1.7382, acc-0.4781
Iter-6600, train loss-1.6395, acc-0.6000, valid loss-1.7381, acc-0.4764, test loss-1.7369, acc-0.4785
Iter-6610, train loss-1.9306, acc-0.3800, valid loss-1.7368, acc-0.4768, test loss-1.7356, acc-0.4788
Iter-6620, train loss-1.6688, acc-0.5200, valid loss-1.7354, acc-0.4768, test loss-1.7342, acc-0.4792
Iter-6630, train loss-1.6534, acc-0.4400, valid loss-1.7342, acc-0.4772, test loss-1.7330, acc-0.4794
Iter-6640, train loss-1.7978, acc-0.4200, valid loss-1.7329, acc-0.4778, test loss-1.7317, acc-0.4798
Iter-6650, train loss-1.6740, acc-0.5600, valid loss-1.7317, acc-0.4776, test loss-1.7305, acc-0.4803
Iter-6660, train loss-1.7028, acc-0.4800, valid loss-1.7304, acc-0.4778, test loss-1.7292, acc-0.4807
Iter-6670, train loss-1.7027, acc-0.5400, valid loss-1.7291, acc-0.4784, test loss-1.7278, acc-0.4812
Iter-6680, train loss-1.7380, acc-0.5000, valid loss-1.7278, acc-0.4788, test loss-1.7265, acc-0.4814
Iter-6690, train loss-1.6198, acc-0.5600, valid loss-1.7265, acc-0.4788, test loss-1.7252, acc-0.4814
Iter-6700, train loss-1.6869, acc-0.5200, valid loss-1.7252, acc-0.4796, test loss-1.7239, acc-0.4814
Iter-6710, train loss-1.8944, acc-0.3200, valid loss-1.7239, acc-0.4802, test loss-1.7226, acc-0.4819
Iter-6720, train loss-1.7695, acc-0.3400, valid loss-1.7226, acc-0.4806, test loss-1.7213, acc-0.4823
Iter-6730, train loss-1.7101, acc-0.4400, valid loss-1.7214, acc-0.4806, test loss-1.7200, acc-0.4828
Iter-6740, train loss-1.7204, acc-0.5400, valid loss-1.7201, acc-0.4812, test loss-1.7187, acc-0.4833
Iter-6750, train loss-1.7736, acc-0.4200, valid loss-1.7188, acc-0.4818, test loss-1.7175, acc-0.4839
Iter-6760, train loss-1.7889, acc-0.5000, valid loss-1.7176, acc-0.4820, test loss-1.7162, acc-0.4845
Iter-6770, train loss-1.7548, acc-0.4600, valid loss-1.7163, acc-0.4818, test loss-1.7149, acc-0.4851
Iter-6780, train loss-1.6547, acc-0.4800, valid loss-1.7150, acc-0.4814, test loss-1.7136, acc-0.4851
Iter-6790, train loss-1.7779, acc-0.3800, valid loss-1.7137, acc-0.4822, test loss-1.7123, acc-0.4852
Iter-6800, train loss-1.7355, acc-0.4600, valid loss-1.7124, acc-0.4824, test loss-1.7110, acc-0.4852
Iter-6810, train loss-1.7359, acc-0.4200, valid loss-1.7111, acc-0.4826, test loss-1.7097, acc-0.4855
Iter-6820, train loss-1.8112, acc-0.3800, valid loss-1.7099, acc-0.4828, test loss-1.7085, acc-0.4860
Iter-6830, train loss-1.7086, acc-0.4800, valid loss-1.7086, acc-0.4834, test loss-1.7072, acc-0.4863
Iter-6840, train loss-1.5199, acc-0.6200, valid loss-1.7072, acc-0.4836, test loss-1.7058, acc-0.4866
Iter-6850, train loss-1.6646, acc-0.5600, valid loss-1.7060, acc-0.4838, test loss-1.7045, acc-0.4869
Iter-6860, train loss-1.6980, acc-0.5200, valid loss-1.7048, acc-0.4840, test loss-1.7033, acc-0.4877
Iter-6870, train loss-1.6348, acc-0.5800, valid loss-1.7035, acc-0.4844, test loss-1.7020, acc-0.4879
Iter-6880, train loss-1.7037, acc-0.6000, valid loss-1.7023, acc-0.4844, test loss-1.7009, acc-0.4880
Iter-6890, train loss-1.8580, acc-0.4400, valid loss-1.7011, acc-0.4844, test loss-1.6997, acc-0.4883
Iter-6900, train loss-1.6243, acc-0.5600, valid loss-1.6999, acc-0.4852, test loss-1.6985, acc-0.4887
Iter-6910, train loss-1.5485, acc-0.7400, valid loss-1.6986, acc-0.4858, test loss-1.6971, acc-0.4892
Iter-6920, train loss-1.6856, acc-0.5200, valid loss-1.6974, acc-0.4862, test loss-1.6959, acc-0.4895
Iter-6930, train loss-1.6583, acc-0.5400, valid loss-1.6960, acc-0.4874, test loss-1.6945, acc-0.4903
Iter-6940, train loss-1.7033, acc-0.5400, valid loss-1.6948, acc-0.4874, test loss-1.6933, acc-0.4911
Iter-6950, train loss-1.7081, acc-0.5000, valid loss-1.6935, acc-0.4878, test loss-1.6919, acc-0.4916
Iter-6960, train loss-1.6723, acc-0.5600, valid loss-1.6923, acc-0.4878, test loss-1.6908, acc-0.4918
Iter-6970, train loss-1.5366, acc-0.6000, valid loss-1.6909, acc-0.4882, test loss-1.6894, acc-0.4921
Iter-6980, train loss-1.7743, acc-0.4200, valid loss-1.6897, acc-0.4886, test loss-1.6881, acc-0.4924
Iter-6990, train loss-1.8004, acc-0.4200, valid loss-1.6884, acc-0.4884, test loss-1.6869, acc-0.4924
Iter-7000, train loss-1.6735, acc-0.5400, valid loss-1.6872, acc-0.4886, test loss-1.6856, acc-0.4931
Iter-7010, train loss-1.7152, acc-0.4400, valid loss-1.6858, acc-0.4900, test loss-1.6843, acc-0.4939
Iter-7020, train loss-1.6709, acc-0.4000, valid loss-1.6846, acc-0.4902, test loss-1.6830, acc-0.4944
Iter-7030, train loss-1.6377, acc-0.5600, valid loss-1.6834, acc-0.4904, test loss-1.6819, acc-0.4948
Iter-7040, train loss-1.7512, acc-0.3800, valid loss-1.6821, acc-0.4906, test loss-1.6806, acc-0.4953
Iter-7050, train loss-1.5046, acc-0.5600, valid loss-1.6808, acc-0.4908, test loss-1.6792, acc-0.4961
Iter-7060, train loss-1.6223, acc-0.5000, valid loss-1.6796, acc-0.4916, test loss-1.6780, acc-0.4963
Iter-7070, train loss-1.6228, acc-0.5600, valid loss-1.6783, acc-0.4922, test loss-1.6767, acc-0.4967
Iter-7080, train loss-1.6465, acc-0.5000, valid loss-1.6770, acc-0.4926, test loss-1.6753, acc-0.4971
Iter-7090, train loss-1.6228, acc-0.5400, valid loss-1.6758, acc-0.4928, test loss-1.6742, acc-0.4978
Iter-7100, train loss-1.7463, acc-0.5400, valid loss-1.6745, acc-0.4936, test loss-1.6728, acc-0.4980
Iter-7110, train loss-1.8127, acc-0.4200, valid loss-1.6731, acc-0.4938, test loss-1.6715, acc-0.4979
Iter-7120, train loss-1.7537, acc-0.5000, valid loss-1.6719, acc-0.4942, test loss-1.6703, acc-0.4983
Iter-7130, train loss-1.6594, acc-0.4800, valid loss-1.6707, acc-0.4954, test loss-1.6691, acc-0.4993
Iter-7140, train loss-1.6436, acc-0.5600, valid loss-1.6694, acc-0.4956, test loss-1.6678, acc-0.4999
Iter-7150, train loss-1.7457, acc-0.5200, valid loss-1.6682, acc-0.4962, test loss-1.6665, acc-0.5004
Iter-7160, train loss-1.6287, acc-0.5600, valid loss-1.6670, acc-0.4970, test loss-1.6653, acc-0.5010
Iter-7170, train loss-1.6486, acc-0.5000, valid loss-1.6657, acc-0.4976, test loss-1.6639, acc-0.5017
Iter-7180, train loss-1.6598, acc-0.5800, valid loss-1.6644, acc-0.4978, test loss-1.6627, acc-0.5020
Iter-7190, train loss-1.5514, acc-0.5600, valid loss-1.6632, acc-0.4982, test loss-1.6614, acc-0.5026
Iter-7200, train loss-1.6903, acc-0.4200, valid loss-1.6619, acc-0.4986, test loss-1.6601, acc-0.5032
Iter-7210, train loss-1.7602, acc-0.4800, valid loss-1.6607, acc-0.4988, test loss-1.6590, acc-0.5037
Iter-7220, train loss-1.7399, acc-0.4200, valid loss-1.6595, acc-0.4994, test loss-1.6577, acc-0.5041
Iter-7230, train loss-1.7714, acc-0.5000, valid loss-1.6583, acc-0.4998, test loss-1.6565, acc-0.5045
Iter-7240, train loss-1.5606, acc-0.5200, valid loss-1.6570, acc-0.5006, test loss-1.6553, acc-0.5053
Iter-7250, train loss-1.7169, acc-0.5200, valid loss-1.6558, acc-0.5010, test loss-1.6540, acc-0.5052
Iter-7260, train loss-1.7571, acc-0.3800, valid loss-1.6547, acc-0.5012, test loss-1.6529, acc-0.5056
Iter-7270, train loss-1.6720, acc-0.5200, valid loss-1.6535, acc-0.5010, test loss-1.6517, acc-0.5057
Iter-7280, train loss-1.6112, acc-0.5200, valid loss-1.6522, acc-0.5020, test loss-1.6504, acc-0.5061
Iter-7290, train loss-1.7408, acc-0.4200, valid loss-1.6510, acc-0.5022, test loss-1.6491, acc-0.5067
Iter-7300, train loss-1.7457, acc-0.4800, valid loss-1.6497, acc-0.5026, test loss-1.6478, acc-0.5073
Iter-7310, train loss-1.6243, acc-0.4600, valid loss-1.6484, acc-0.5032, test loss-1.6466, acc-0.5076
Iter-7320, train loss-1.6647, acc-0.5600, valid loss-1.6471, acc-0.5034, test loss-1.6453, acc-0.5076
Iter-7330, train loss-1.6617, acc-0.5000, valid loss-1.6458, acc-0.5034, test loss-1.6440, acc-0.5087
Iter-7340, train loss-1.6907, acc-0.5000, valid loss-1.6446, acc-0.5038, test loss-1.6428, acc-0.5089
Iter-7350, train loss-1.7082, acc-0.4800, valid loss-1.6435, acc-0.5038, test loss-1.6416, acc-0.5091
Iter-7360, train loss-1.6345, acc-0.5400, valid loss-1.6422, acc-0.5046, test loss-1.6403, acc-0.5095
Iter-7370, train loss-1.6199, acc-0.5400, valid loss-1.6410, acc-0.5054, test loss-1.6390, acc-0.5106
Iter-7380, train loss-1.6240, acc-0.5200, valid loss-1.6397, acc-0.5058, test loss-1.6378, acc-0.5109
Iter-7390, train loss-1.4962, acc-0.6200, valid loss-1.6385, acc-0.5058, test loss-1.6366, acc-0.5113
Iter-7400, train loss-1.5206, acc-0.6000, valid loss-1.6373, acc-0.5058, test loss-1.6354, acc-0.5114
Iter-7410, train loss-1.6422, acc-0.6200, valid loss-1.6360, acc-0.5060, test loss-1.6341, acc-0.5116
Iter-7420, train loss-1.6751, acc-0.4600, valid loss-1.6349, acc-0.5066, test loss-1.6330, acc-0.5122
Iter-7430, train loss-1.5695, acc-0.5800, valid loss-1.6336, acc-0.5076, test loss-1.6317, acc-0.5127
Iter-7440, train loss-1.6477, acc-0.4800, valid loss-1.6324, acc-0.5078, test loss-1.6305, acc-0.5130
Iter-7450, train loss-1.5719, acc-0.5200, valid loss-1.6312, acc-0.5086, test loss-1.6292, acc-0.5137
Iter-7460, train loss-1.7006, acc-0.4400, valid loss-1.6299, acc-0.5090, test loss-1.6279, acc-0.5142
Iter-7470, train loss-1.5521, acc-0.5800, valid loss-1.6285, acc-0.5096, test loss-1.6266, acc-0.5146
Iter-7480, train loss-1.7391, acc-0.3600, valid loss-1.6273, acc-0.5100, test loss-1.6253, acc-0.5151
Iter-7490, train loss-1.6415, acc-0.4400, valid loss-1.6260, acc-0.5100, test loss-1.6241, acc-0.5157
Iter-7500, train loss-1.7368, acc-0.4600, valid loss-1.6248, acc-0.5110, test loss-1.6228, acc-0.5168
Iter-7510, train loss-1.5956, acc-0.5600, valid loss-1.6236, acc-0.5112, test loss-1.6216, acc-0.5170
Iter-7520, train loss-1.6083, acc-0.4400, valid loss-1.6222, acc-0.5118, test loss-1.6202, acc-0.5173
Iter-7530, train loss-1.6096, acc-0.5400, valid loss-1.6210, acc-0.5118, test loss-1.6190, acc-0.5177
Iter-7540, train loss-1.7360, acc-0.4600, valid loss-1.6198, acc-0.5124, test loss-1.6178, acc-0.5187
Iter-7550, train loss-1.6879, acc-0.4400, valid loss-1.6186, acc-0.5130, test loss-1.6166, acc-0.5193
Iter-7560, train loss-1.6182, acc-0.4800, valid loss-1.6173, acc-0.5136, test loss-1.6153, acc-0.5202
Iter-7570, train loss-1.5958, acc-0.5600, valid loss-1.6160, acc-0.5142, test loss-1.6139, acc-0.5205
Iter-7580, train loss-1.5494, acc-0.5600, valid loss-1.6147, acc-0.5154, test loss-1.6126, acc-0.5211
Iter-7590, train loss-1.5827, acc-0.5600, valid loss-1.6135, acc-0.5154, test loss-1.6115, acc-0.5215
Iter-7600, train loss-1.6919, acc-0.4800, valid loss-1.6122, acc-0.5170, test loss-1.6102, acc-0.5223
Iter-7610, train loss-1.6024, acc-0.4800, valid loss-1.6110, acc-0.5172, test loss-1.6089, acc-0.5230
Iter-7620, train loss-1.6008, acc-0.4600, valid loss-1.6096, acc-0.5180, test loss-1.6075, acc-0.5233
Iter-7630, train loss-1.5912, acc-0.5200, valid loss-1.6084, acc-0.5192, test loss-1.6063, acc-0.5242
Iter-7640, train loss-1.6842, acc-0.5800, valid loss-1.6071, acc-0.5204, test loss-1.6050, acc-0.5247
Iter-7650, train loss-1.6100, acc-0.4600, valid loss-1.6059, acc-0.5206, test loss-1.6038, acc-0.5250
Iter-7660, train loss-1.5240, acc-0.5600, valid loss-1.6047, acc-0.5204, test loss-1.6026, acc-0.5254
Iter-7670, train loss-1.6500, acc-0.4800, valid loss-1.6034, acc-0.5204, test loss-1.6013, acc-0.5259
Iter-7680, train loss-1.5344, acc-0.5400, valid loss-1.6022, acc-0.5210, test loss-1.6000, acc-0.5264
Iter-7690, train loss-1.6218, acc-0.5200, valid loss-1.6010, acc-0.5208, test loss-1.5988, acc-0.5263
Iter-7700, train loss-1.4917, acc-0.5400, valid loss-1.5999, acc-0.5216, test loss-1.5977, acc-0.5264
Iter-7710, train loss-1.6691, acc-0.4400, valid loss-1.5987, acc-0.5222, test loss-1.5965, acc-0.5267
Iter-7720, train loss-1.7012, acc-0.5400, valid loss-1.5974, acc-0.5230, test loss-1.5952, acc-0.5272
Iter-7730, train loss-1.5553, acc-0.5600, valid loss-1.5961, acc-0.5236, test loss-1.5939, acc-0.5278
Iter-7740, train loss-1.5661, acc-0.5200, valid loss-1.5949, acc-0.5244, test loss-1.5926, acc-0.5287
Iter-7750, train loss-1.6588, acc-0.4600, valid loss-1.5936, acc-0.5246, test loss-1.5913, acc-0.5291
Iter-7760, train loss-1.6263, acc-0.5600, valid loss-1.5924, acc-0.5248, test loss-1.5902, acc-0.5298
Iter-7770, train loss-1.5835, acc-0.5400, valid loss-1.5913, acc-0.5258, test loss-1.5890, acc-0.5302
Iter-7780, train loss-1.6047, acc-0.5200, valid loss-1.5900, acc-0.5272, test loss-1.5877, acc-0.5311
Iter-7790, train loss-1.7107, acc-0.3600, valid loss-1.5888, acc-0.5278, test loss-1.5864, acc-0.5317
Iter-7800, train loss-1.6403, acc-0.5000, valid loss-1.5875, acc-0.5294, test loss-1.5852, acc-0.5319
Iter-7810, train loss-1.6047, acc-0.5400, valid loss-1.5864, acc-0.5296, test loss-1.5841, acc-0.5325
Iter-7820, train loss-1.4547, acc-0.7000, valid loss-1.5852, acc-0.5300, test loss-1.5828, acc-0.5326
Iter-7830, train loss-1.5530, acc-0.5200, valid loss-1.5840, acc-0.5304, test loss-1.5817, acc-0.5329
Iter-7840, train loss-1.7490, acc-0.5000, valid loss-1.5829, acc-0.5310, test loss-1.5805, acc-0.5332
Iter-7850, train loss-1.5592, acc-0.6600, valid loss-1.5816, acc-0.5318, test loss-1.5793, acc-0.5334
Iter-7860, train loss-1.7287, acc-0.4000, valid loss-1.5805, acc-0.5328, test loss-1.5781, acc-0.5337
Iter-7870, train loss-1.7012, acc-0.5400, valid loss-1.5794, acc-0.5336, test loss-1.5770, acc-0.5344
Iter-7880, train loss-1.6032, acc-0.5000, valid loss-1.5782, acc-0.5334, test loss-1.5758, acc-0.5347
Iter-7890, train loss-1.5678, acc-0.4600, valid loss-1.5770, acc-0.5346, test loss-1.5746, acc-0.5358
Iter-7900, train loss-1.5850, acc-0.5200, valid loss-1.5758, acc-0.5346, test loss-1.5733, acc-0.5366
Iter-7910, train loss-1.5244, acc-0.5600, valid loss-1.5746, acc-0.5354, test loss-1.5721, acc-0.5371
Iter-7920, train loss-1.5549, acc-0.5200, valid loss-1.5734, acc-0.5356, test loss-1.5709, acc-0.5374
Iter-7930, train loss-1.5399, acc-0.6200, valid loss-1.5722, acc-0.5358, test loss-1.5697, acc-0.5383
Iter-7940, train loss-1.5251, acc-0.5200, valid loss-1.5710, acc-0.5358, test loss-1.5685, acc-0.5389
Iter-7950, train loss-1.6120, acc-0.5200, valid loss-1.5697, acc-0.5362, test loss-1.5673, acc-0.5393
Iter-7960, train loss-1.5976, acc-0.5200, valid loss-1.5686, acc-0.5368, test loss-1.5661, acc-0.5398
Iter-7970, train loss-1.5517, acc-0.5600, valid loss-1.5673, acc-0.5376, test loss-1.5648, acc-0.5399
Iter-7980, train loss-1.4548, acc-0.6200, valid loss-1.5661, acc-0.5382, test loss-1.5636, acc-0.5406
Iter-7990, train loss-1.7282, acc-0.4000, valid loss-1.5650, acc-0.5388, test loss-1.5625, acc-0.5410
Iter-8000, train loss-1.6137, acc-0.4000, valid loss-1.5638, acc-0.5392, test loss-1.5613, acc-0.5415
Iter-8010, train loss-1.6277, acc-0.5200, valid loss-1.5627, acc-0.5400, test loss-1.5602, acc-0.5420
Iter-8020, train loss-1.6669, acc-0.4200, valid loss-1.5615, acc-0.5402, test loss-1.5590, acc-0.5425
Iter-8030, train loss-1.7085, acc-0.3800, valid loss-1.5604, acc-0.5410, test loss-1.5579, acc-0.5432
Iter-8040, train loss-1.4208, acc-0.6800, valid loss-1.5592, acc-0.5414, test loss-1.5567, acc-0.5434
Iter-8050, train loss-1.5736, acc-0.4600, valid loss-1.5580, acc-0.5416, test loss-1.5555, acc-0.5436
Iter-8060, train loss-1.5901, acc-0.5600, valid loss-1.5567, acc-0.5418, test loss-1.5542, acc-0.5443
Iter-8070, train loss-1.5069, acc-0.6400, valid loss-1.5555, acc-0.5418, test loss-1.5530, acc-0.5448
Iter-8080, train loss-1.4679, acc-0.6000, valid loss-1.5543, acc-0.5420, test loss-1.5518, acc-0.5451
Iter-8090, train loss-1.6376, acc-0.5200, valid loss-1.5531, acc-0.5428, test loss-1.5506, acc-0.5461
Iter-8100, train loss-1.7208, acc-0.5200, valid loss-1.5519, acc-0.5432, test loss-1.5494, acc-0.5470
Iter-8110, train loss-1.4249, acc-0.6400, valid loss-1.5508, acc-0.5438, test loss-1.5483, acc-0.5470
Iter-8120, train loss-1.6825, acc-0.4000, valid loss-1.5496, acc-0.5442, test loss-1.5472, acc-0.5478
Iter-8130, train loss-1.5298, acc-0.5600, valid loss-1.5485, acc-0.5450, test loss-1.5459, acc-0.5486
Iter-8140, train loss-1.4855, acc-0.6000, valid loss-1.5473, acc-0.5452, test loss-1.5448, acc-0.5488
Iter-8150, train loss-1.6370, acc-0.3600, valid loss-1.5462, acc-0.5452, test loss-1.5436, acc-0.5491
Iter-8160, train loss-1.6334, acc-0.4800, valid loss-1.5450, acc-0.5456, test loss-1.5425, acc-0.5495
Iter-8170, train loss-1.4916, acc-0.5000, valid loss-1.5438, acc-0.5464, test loss-1.5412, acc-0.5505
Iter-8180, train loss-1.5454, acc-0.5600, valid loss-1.5426, acc-0.5480, test loss-1.5400, acc-0.5514
Iter-8190, train loss-1.6018, acc-0.5200, valid loss-1.5415, acc-0.5488, test loss-1.5389, acc-0.5520
Iter-8200, train loss-1.6867, acc-0.4200, valid loss-1.5404, acc-0.5492, test loss-1.5377, acc-0.5519
Iter-8210, train loss-1.7330, acc-0.3600, valid loss-1.5392, acc-0.5500, test loss-1.5366, acc-0.5519
Iter-8220, train loss-1.5391, acc-0.6000, valid loss-1.5380, acc-0.5504, test loss-1.5354, acc-0.5525
Iter-8230, train loss-1.4740, acc-0.5600, valid loss-1.5369, acc-0.5510, test loss-1.5342, acc-0.5534
Iter-8240, train loss-1.4290, acc-0.6800, valid loss-1.5357, acc-0.5522, test loss-1.5330, acc-0.5534
Iter-8250, train loss-1.4734, acc-0.6000, valid loss-1.5346, acc-0.5530, test loss-1.5319, acc-0.5542
Iter-8260, train loss-1.5087, acc-0.4800, valid loss-1.5334, acc-0.5536, test loss-1.5307, acc-0.5546
Iter-8270, train loss-1.6092, acc-0.5400, valid loss-1.5323, acc-0.5538, test loss-1.5296, acc-0.5554
Iter-8280, train loss-1.4939, acc-0.5200, valid loss-1.5311, acc-0.5540, test loss-1.5285, acc-0.5557
Iter-8290, train loss-1.4535, acc-0.6800, valid loss-1.5300, acc-0.5544, test loss-1.5273, acc-0.5561
Iter-8300, train loss-1.4599, acc-0.6800, valid loss-1.5288, acc-0.5548, test loss-1.5262, acc-0.5565
Iter-8310, train loss-1.6064, acc-0.4600, valid loss-1.5278, acc-0.5550, test loss-1.5251, acc-0.5569
Iter-8320, train loss-1.4006, acc-0.6800, valid loss-1.5265, acc-0.5562, test loss-1.5239, acc-0.5575
Iter-8330, train loss-1.6196, acc-0.5800, valid loss-1.5254, acc-0.5566, test loss-1.5227, acc-0.5579
Iter-8340, train loss-1.5469, acc-0.5400, valid loss-1.5242, acc-0.5574, test loss-1.5215, acc-0.5584
Iter-8350, train loss-1.5541, acc-0.5400, valid loss-1.5231, acc-0.5582, test loss-1.5204, acc-0.5590
Iter-8360, train loss-1.6076, acc-0.4400, valid loss-1.5219, acc-0.5584, test loss-1.5192, acc-0.5593
Iter-8370, train loss-1.5325, acc-0.5600, valid loss-1.5208, acc-0.5588, test loss-1.5180, acc-0.5600
Iter-8380, train loss-1.5503, acc-0.5600, valid loss-1.5197, acc-0.5600, test loss-1.5169, acc-0.5605
Iter-8390, train loss-1.4645, acc-0.5800, valid loss-1.5184, acc-0.5608, test loss-1.5157, acc-0.5610
Iter-8400, train loss-1.6547, acc-0.4800, valid loss-1.5173, acc-0.5610, test loss-1.5145, acc-0.5618
Iter-8410, train loss-1.6128, acc-0.5200, valid loss-1.5162, acc-0.5612, test loss-1.5134, acc-0.5619
Iter-8420, train loss-1.5634, acc-0.6000, valid loss-1.5149, acc-0.5614, test loss-1.5121, acc-0.5621
Iter-8430, train loss-1.6475, acc-0.5400, valid loss-1.5138, acc-0.5622, test loss-1.5110, acc-0.5624
Iter-8440, train loss-1.4796, acc-0.4800, valid loss-1.5126, acc-0.5628, test loss-1.5098, acc-0.5623
Iter-8450, train loss-1.5664, acc-0.5400, valid loss-1.5115, acc-0.5632, test loss-1.5086, acc-0.5630
Iter-8460, train loss-1.6867, acc-0.5000, valid loss-1.5103, acc-0.5638, test loss-1.5074, acc-0.5632
Iter-8470, train loss-1.4384, acc-0.6600, valid loss-1.5092, acc-0.5646, test loss-1.5063, acc-0.5633
Iter-8480, train loss-1.3821, acc-0.7000, valid loss-1.5081, acc-0.5650, test loss-1.5052, acc-0.5638
Iter-8490, train loss-1.4703, acc-0.5400, valid loss-1.5070, acc-0.5650, test loss-1.5041, acc-0.5640
Iter-8500, train loss-1.5515, acc-0.5200, valid loss-1.5058, acc-0.5654, test loss-1.5029, acc-0.5645
Iter-8510, train loss-1.6623, acc-0.4000, valid loss-1.5046, acc-0.5658, test loss-1.5017, acc-0.5650
Iter-8520, train loss-1.5214, acc-0.5800, valid loss-1.5034, acc-0.5672, test loss-1.5005, acc-0.5658
Iter-8530, train loss-1.5124, acc-0.5400, valid loss-1.5023, acc-0.5678, test loss-1.4993, acc-0.5666
Iter-8540, train loss-1.5525, acc-0.5400, valid loss-1.5012, acc-0.5680, test loss-1.4982, acc-0.5674
Iter-8550, train loss-1.5232, acc-0.5800, valid loss-1.5001, acc-0.5684, test loss-1.4971, acc-0.5676
Iter-8560, train loss-1.4269, acc-0.6000, valid loss-1.4989, acc-0.5686, test loss-1.4959, acc-0.5685
Iter-8570, train loss-1.3593, acc-0.6600, valid loss-1.4977, acc-0.5692, test loss-1.4948, acc-0.5686
Iter-8580, train loss-1.5111, acc-0.5200, valid loss-1.4966, acc-0.5694, test loss-1.4936, acc-0.5695
Iter-8590, train loss-1.4700, acc-0.5200, valid loss-1.4954, acc-0.5698, test loss-1.4924, acc-0.5699
Iter-8600, train loss-1.3931, acc-0.6400, valid loss-1.4943, acc-0.5702, test loss-1.4913, acc-0.5704
Iter-8610, train loss-1.5119, acc-0.6200, valid loss-1.4931, acc-0.5704, test loss-1.4901, acc-0.5709
Iter-8620, train loss-1.5346, acc-0.6600, valid loss-1.4919, acc-0.5714, test loss-1.4889, acc-0.5713
Iter-8630, train loss-1.5769, acc-0.5600, valid loss-1.4908, acc-0.5712, test loss-1.4878, acc-0.5716
Iter-8640, train loss-1.4758, acc-0.7000, valid loss-1.4897, acc-0.5718, test loss-1.4867, acc-0.5721
Iter-8650, train loss-1.5000, acc-0.6000, valid loss-1.4886, acc-0.5726, test loss-1.4856, acc-0.5732
Iter-8660, train loss-1.5196, acc-0.5600, valid loss-1.4875, acc-0.5736, test loss-1.4845, acc-0.5741
Iter-8670, train loss-1.4324, acc-0.5800, valid loss-1.4863, acc-0.5752, test loss-1.4833, acc-0.5744
Iter-8680, train loss-1.4428, acc-0.5600, valid loss-1.4852, acc-0.5758, test loss-1.4822, acc-0.5745
Iter-8690, train loss-1.6355, acc-0.5400, valid loss-1.4841, acc-0.5758, test loss-1.4811, acc-0.5749
Iter-8700, train loss-1.5349, acc-0.4800, valid loss-1.4830, acc-0.5766, test loss-1.4799, acc-0.5753
Iter-8710, train loss-1.3966, acc-0.6600, valid loss-1.4818, acc-0.5774, test loss-1.4788, acc-0.5756
Iter-8720, train loss-1.4434, acc-0.5800, valid loss-1.4807, acc-0.5778, test loss-1.4776, acc-0.5757
Iter-8730, train loss-1.4661, acc-0.6000, valid loss-1.4795, acc-0.5784, test loss-1.4765, acc-0.5762
Iter-8740, train loss-1.4967, acc-0.6000, valid loss-1.4784, acc-0.5780, test loss-1.4753, acc-0.5762
Iter-8750, train loss-1.3612, acc-0.6800, valid loss-1.4772, acc-0.5786, test loss-1.4742, acc-0.5768
Iter-8760, train loss-1.5359, acc-0.5400, valid loss-1.4761, acc-0.5798, test loss-1.4731, acc-0.5772
Iter-8770, train loss-1.5245, acc-0.5200, valid loss-1.4750, acc-0.5808, test loss-1.4720, acc-0.5781
Iter-8780, train loss-1.4807, acc-0.6000, valid loss-1.4738, acc-0.5820, test loss-1.4707, acc-0.5786
Iter-8790, train loss-1.6058, acc-0.4600, valid loss-1.4727, acc-0.5824, test loss-1.4697, acc-0.5794
Iter-8800, train loss-1.5460, acc-0.5200, valid loss-1.4716, acc-0.5834, test loss-1.4685, acc-0.5800
Iter-8810, train loss-1.4851, acc-0.6200, valid loss-1.4705, acc-0.5840, test loss-1.4674, acc-0.5806
Iter-8820, train loss-1.4753, acc-0.5600, valid loss-1.4694, acc-0.5838, test loss-1.4663, acc-0.5807
Iter-8830, train loss-1.4714, acc-0.5600, valid loss-1.4682, acc-0.5838, test loss-1.4652, acc-0.5821
Iter-8840, train loss-1.4064, acc-0.6400, valid loss-1.4671, acc-0.5844, test loss-1.4640, acc-0.5830
Iter-8850, train loss-1.5568, acc-0.5200, valid loss-1.4659, acc-0.5844, test loss-1.4629, acc-0.5832
Iter-8860, train loss-1.3295, acc-0.7200, valid loss-1.4648, acc-0.5846, test loss-1.4617, acc-0.5837
Iter-8870, train loss-1.4964, acc-0.5800, valid loss-1.4638, acc-0.5852, test loss-1.4607, acc-0.5838
Iter-8880, train loss-1.3359, acc-0.6800, valid loss-1.4626, acc-0.5856, test loss-1.4595, acc-0.5841
Iter-8890, train loss-1.4351, acc-0.5200, valid loss-1.4615, acc-0.5866, test loss-1.4585, acc-0.5843
Iter-8900, train loss-1.4946, acc-0.5600, valid loss-1.4604, acc-0.5868, test loss-1.4573, acc-0.5849
Iter-8910, train loss-1.4409, acc-0.6200, valid loss-1.4594, acc-0.5866, test loss-1.4563, acc-0.5847
Iter-8920, train loss-1.3616, acc-0.6800, valid loss-1.4583, acc-0.5872, test loss-1.4552, acc-0.5856
Iter-8930, train loss-1.5391, acc-0.4800, valid loss-1.4572, acc-0.5878, test loss-1.4541, acc-0.5863
Iter-8940, train loss-1.3577, acc-0.6400, valid loss-1.4561, acc-0.5886, test loss-1.4530, acc-0.5862
Iter-8950, train loss-1.3477, acc-0.6400, valid loss-1.4550, acc-0.5890, test loss-1.4519, acc-0.5873
Iter-8960, train loss-1.4633, acc-0.5200, valid loss-1.4539, acc-0.5896, test loss-1.4508, acc-0.5875
Iter-8970, train loss-1.4899, acc-0.5400, valid loss-1.4528, acc-0.5896, test loss-1.4497, acc-0.5878
Iter-8980, train loss-1.4737, acc-0.6600, valid loss-1.4517, acc-0.5896, test loss-1.4485, acc-0.5883
Iter-8990, train loss-1.3584, acc-0.6000, valid loss-1.4505, acc-0.5904, test loss-1.4473, acc-0.5889
Iter-9000, train loss-1.4973, acc-0.5800, valid loss-1.4494, acc-0.5918, test loss-1.4462, acc-0.5894
Iter-9010, train loss-1.3122, acc-0.6600, valid loss-1.4483, acc-0.5918, test loss-1.4451, acc-0.5900
Iter-9020, train loss-1.5333, acc-0.6000, valid loss-1.4472, acc-0.5918, test loss-1.4441, acc-0.5903
Iter-9030, train loss-1.4513, acc-0.5400, valid loss-1.4462, acc-0.5924, test loss-1.4430, acc-0.5908
Iter-9040, train loss-1.5282, acc-0.5200, valid loss-1.4450, acc-0.5928, test loss-1.4418, acc-0.5920
Iter-9050, train loss-1.4065, acc-0.6200, valid loss-1.4439, acc-0.5934, test loss-1.4407, acc-0.5923
Iter-9060, train loss-1.5569, acc-0.5400, valid loss-1.4428, acc-0.5944, test loss-1.4396, acc-0.5927
Iter-9070, train loss-1.4912, acc-0.5600, valid loss-1.4418, acc-0.5950, test loss-1.4386, acc-0.5927
Iter-9080, train loss-1.5321, acc-0.6000, valid loss-1.4408, acc-0.5948, test loss-1.4376, acc-0.5935
Iter-9090, train loss-1.3993, acc-0.5400, valid loss-1.4397, acc-0.5950, test loss-1.4365, acc-0.5940
Iter-9100, train loss-1.4312, acc-0.6000, valid loss-1.4387, acc-0.5954, test loss-1.4355, acc-0.5940
Iter-9110, train loss-1.3746, acc-0.6200, valid loss-1.4377, acc-0.5952, test loss-1.4345, acc-0.5945
Iter-9120, train loss-1.4158, acc-0.6000, valid loss-1.4366, acc-0.5952, test loss-1.4334, acc-0.5949
Iter-9130, train loss-1.4065, acc-0.6000, valid loss-1.4355, acc-0.5956, test loss-1.4323, acc-0.5953
Iter-9140, train loss-1.3162, acc-0.7000, valid loss-1.4344, acc-0.5956, test loss-1.4312, acc-0.5963
Iter-9150, train loss-1.4377, acc-0.6600, valid loss-1.4333, acc-0.5958, test loss-1.4301, acc-0.5971
Iter-9160, train loss-1.2602, acc-0.7000, valid loss-1.4323, acc-0.5958, test loss-1.4291, acc-0.5977
Iter-9170, train loss-1.4875, acc-0.5200, valid loss-1.4312, acc-0.5968, test loss-1.4280, acc-0.5982
Iter-9180, train loss-1.4223, acc-0.6200, valid loss-1.4301, acc-0.5976, test loss-1.4269, acc-0.5984
Iter-9190, train loss-1.2685, acc-0.6600, valid loss-1.4290, acc-0.5978, test loss-1.4258, acc-0.5990
Iter-9200, train loss-1.4493, acc-0.6000, valid loss-1.4280, acc-0.5980, test loss-1.4247, acc-0.5993
Iter-9210, train loss-1.4268, acc-0.6200, valid loss-1.4269, acc-0.5998, test loss-1.4236, acc-0.6001
Iter-9220, train loss-1.3430, acc-0.6200, valid loss-1.4259, acc-0.6008, test loss-1.4226, acc-0.6010
Iter-9230, train loss-1.3697, acc-0.5600, valid loss-1.4248, acc-0.6012, test loss-1.4215, acc-0.6015
Iter-9240, train loss-1.3381, acc-0.7400, valid loss-1.4237, acc-0.6016, test loss-1.4204, acc-0.6020
Iter-9250, train loss-1.3606, acc-0.6000, valid loss-1.4226, acc-0.6022, test loss-1.4193, acc-0.6029
Iter-9260, train loss-1.4348, acc-0.7000, valid loss-1.4216, acc-0.6024, test loss-1.4183, acc-0.6030
Iter-9270, train loss-1.5220, acc-0.4800, valid loss-1.4206, acc-0.6026, test loss-1.4173, acc-0.6035
Iter-9280, train loss-1.3492, acc-0.6000, valid loss-1.4196, acc-0.6026, test loss-1.4162, acc-0.6047
Iter-9290, train loss-1.3950, acc-0.6000, valid loss-1.4186, acc-0.6028, test loss-1.4152, acc-0.6052
Iter-9300, train loss-1.5736, acc-0.5400, valid loss-1.4176, acc-0.6036, test loss-1.4143, acc-0.6055
Iter-9310, train loss-1.5217, acc-0.5400, valid loss-1.4164, acc-0.6040, test loss-1.4132, acc-0.6059
Iter-9320, train loss-1.3513, acc-0.6200, valid loss-1.4153, acc-0.6052, test loss-1.4120, acc-0.6070
Iter-9330, train loss-1.3613, acc-0.6200, valid loss-1.4143, acc-0.6058, test loss-1.4110, acc-0.6079
Iter-9340, train loss-1.3424, acc-0.6800, valid loss-1.4132, acc-0.6058, test loss-1.4099, acc-0.6080
Iter-9350, train loss-1.4625, acc-0.5600, valid loss-1.4122, acc-0.6070, test loss-1.4089, acc-0.6090
Iter-9360, train loss-1.3449, acc-0.5200, valid loss-1.4111, acc-0.6078, test loss-1.4078, acc-0.6099
Iter-9370, train loss-1.5416, acc-0.5400, valid loss-1.4100, acc-0.6082, test loss-1.4067, acc-0.6101
Iter-9380, train loss-1.4308, acc-0.5600, valid loss-1.4090, acc-0.6084, test loss-1.4056, acc-0.6109
Iter-9390, train loss-1.3586, acc-0.6800, valid loss-1.4080, acc-0.6086, test loss-1.4047, acc-0.6116
Iter-9400, train loss-1.4772, acc-0.5000, valid loss-1.4070, acc-0.6088, test loss-1.4036, acc-0.6122
Iter-9410, train loss-1.3164, acc-0.7000, valid loss-1.4059, acc-0.6094, test loss-1.4026, acc-0.6129
Iter-9420, train loss-1.4005, acc-0.5800, valid loss-1.4049, acc-0.6094, test loss-1.4016, acc-0.6133
Iter-9430, train loss-1.3825, acc-0.6200, valid loss-1.4039, acc-0.6100, test loss-1.4006, acc-0.6135
Iter-9440, train loss-1.3442, acc-0.6800, valid loss-1.4029, acc-0.6102, test loss-1.3996, acc-0.6139
Iter-9450, train loss-1.5305, acc-0.5600, valid loss-1.4020, acc-0.6116, test loss-1.3986, acc-0.6143
Iter-9460, train loss-1.4118, acc-0.6400, valid loss-1.4009, acc-0.6126, test loss-1.3976, acc-0.6148
Iter-9470, train loss-1.3684, acc-0.6800, valid loss-1.4000, acc-0.6128, test loss-1.3966, acc-0.6153
Iter-9480, train loss-1.3817, acc-0.6400, valid loss-1.3989, acc-0.6130, test loss-1.3956, acc-0.6155
Iter-9490, train loss-1.2699, acc-0.7200, valid loss-1.3979, acc-0.6134, test loss-1.3946, acc-0.6158
Iter-9500, train loss-1.3601, acc-0.6400, valid loss-1.3968, acc-0.6140, test loss-1.3935, acc-0.6164
Iter-9510, train loss-1.3412, acc-0.6200, valid loss-1.3959, acc-0.6142, test loss-1.3925, acc-0.6169
Iter-9520, train loss-1.4327, acc-0.5200, valid loss-1.3949, acc-0.6144, test loss-1.3916, acc-0.6175
Iter-9530, train loss-1.3252, acc-0.7800, valid loss-1.3939, acc-0.6144, test loss-1.3906, acc-0.6183
Iter-9540, train loss-1.5133, acc-0.5800, valid loss-1.3929, acc-0.6152, test loss-1.3896, acc-0.6184
Iter-9550, train loss-1.4253, acc-0.5800, valid loss-1.3918, acc-0.6160, test loss-1.3885, acc-0.6193
Iter-9560, train loss-1.3709, acc-0.6200, valid loss-1.3908, acc-0.6166, test loss-1.3875, acc-0.6196
Iter-9570, train loss-1.4873, acc-0.6200, valid loss-1.3898, acc-0.6170, test loss-1.3864, acc-0.6198
Iter-9580, train loss-1.3928, acc-0.6000, valid loss-1.3888, acc-0.6176, test loss-1.3854, acc-0.6199
Iter-9590, train loss-1.4026, acc-0.6000, valid loss-1.3878, acc-0.6182, test loss-1.3844, acc-0.6203
Iter-9600, train loss-1.2837, acc-0.6800, valid loss-1.3868, acc-0.6186, test loss-1.3834, acc-0.6207
Iter-9610, train loss-1.4498, acc-0.5600, valid loss-1.3858, acc-0.6192, test loss-1.3824, acc-0.6215
Iter-9620, train loss-1.4450, acc-0.5200, valid loss-1.3848, acc-0.6196, test loss-1.3814, acc-0.6218
Iter-9630, train loss-1.3996, acc-0.6600, valid loss-1.3838, acc-0.6198, test loss-1.3804, acc-0.6225
Iter-9640, train loss-1.2795, acc-0.7600, valid loss-1.3828, acc-0.6200, test loss-1.3793, acc-0.6228
Iter-9650, train loss-1.4266, acc-0.7000, valid loss-1.3818, acc-0.6204, test loss-1.3783, acc-0.6230
Iter-9660, train loss-1.4335, acc-0.5800, valid loss-1.3808, acc-0.6204, test loss-1.3774, acc-0.6233
Iter-9670, train loss-1.5314, acc-0.5600, valid loss-1.3799, acc-0.6214, test loss-1.3764, acc-0.6235
Iter-9680, train loss-1.3292, acc-0.6600, valid loss-1.3789, acc-0.6220, test loss-1.3754, acc-0.6242
Iter-9690, train loss-1.3443, acc-0.6600, valid loss-1.3779, acc-0.6226, test loss-1.3744, acc-0.6251
Iter-9700, train loss-1.3899, acc-0.5800, valid loss-1.3768, acc-0.6230, test loss-1.3734, acc-0.6253
Iter-9710, train loss-1.2461, acc-0.7600, valid loss-1.3758, acc-0.6232, test loss-1.3724, acc-0.6254
Iter-9720, train loss-1.3550, acc-0.6800, valid loss-1.3749, acc-0.6236, test loss-1.3715, acc-0.6258
Iter-9730, train loss-1.3429, acc-0.6600, valid loss-1.3739, acc-0.6240, test loss-1.3705, acc-0.6257
Iter-9740, train loss-1.4151, acc-0.5600, valid loss-1.3729, acc-0.6254, test loss-1.3695, acc-0.6262
Iter-9750, train loss-1.4121, acc-0.6600, valid loss-1.3720, acc-0.6260, test loss-1.3685, acc-0.6266
Iter-9760, train loss-1.3076, acc-0.6400, valid loss-1.3710, acc-0.6268, test loss-1.3676, acc-0.6272
Iter-9770, train loss-1.4431, acc-0.5600, valid loss-1.3700, acc-0.6264, test loss-1.3666, acc-0.6276
Iter-9780, train loss-1.3828, acc-0.5600, valid loss-1.3690, acc-0.6268, test loss-1.3656, acc-0.6282
Iter-9790, train loss-1.4310, acc-0.5600, valid loss-1.3680, acc-0.6268, test loss-1.3646, acc-0.6285
Iter-9800, train loss-1.2968, acc-0.7000, valid loss-1.3670, acc-0.6276, test loss-1.3636, acc-0.6286
Iter-9810, train loss-1.1958, acc-0.7000, valid loss-1.3660, acc-0.6280, test loss-1.3626, acc-0.6291
Iter-9820, train loss-1.3550, acc-0.6400, valid loss-1.3650, acc-0.6292, test loss-1.3616, acc-0.6297
Iter-9830, train loss-1.4228, acc-0.5600, valid loss-1.3641, acc-0.6292, test loss-1.3606, acc-0.6299
Iter-9840, train loss-1.5233, acc-0.4400, valid loss-1.3631, acc-0.6298, test loss-1.3596, acc-0.6304
Iter-9850, train loss-1.3897, acc-0.6200, valid loss-1.3621, acc-0.6298, test loss-1.3586, acc-0.6308
Iter-9860, train loss-1.3360, acc-0.7000, valid loss-1.3611, acc-0.6306, test loss-1.3577, acc-0.6311
Iter-9870, train loss-1.2961, acc-0.6600, valid loss-1.3601, acc-0.6318, test loss-1.3566, acc-0.6314
Iter-9880, train loss-1.3663, acc-0.6200, valid loss-1.3591, acc-0.6326, test loss-1.3556, acc-0.6322
Iter-9890, train loss-1.4167, acc-0.5600, valid loss-1.3581, acc-0.6330, test loss-1.3547, acc-0.6324
Iter-9900, train loss-1.3884, acc-0.5600, valid loss-1.3572, acc-0.6332, test loss-1.3537, acc-0.6330
Iter-9910, train loss-1.3009, acc-0.6200, valid loss-1.3563, acc-0.6332, test loss-1.3528, acc-0.6336
Iter-9920, train loss-1.3757, acc-0.5800, valid loss-1.3553, acc-0.6336, test loss-1.3519, acc-0.6343
Iter-9930, train loss-1.5914, acc-0.4400, valid loss-1.3543, acc-0.6344, test loss-1.3509, acc-0.6346
Iter-9940, train loss-1.3919, acc-0.6200, valid loss-1.3533, acc-0.6354, test loss-1.3499, acc-0.6353
Iter-9950, train loss-1.4574, acc-0.5400, valid loss-1.3523, acc-0.6356, test loss-1.3489, acc-0.6357
Iter-9960, train loss-1.4257, acc-0.5000, valid loss-1.3514, acc-0.6362, test loss-1.3479, acc-0.6366
Iter-9970, train loss-1.4207, acc-0.5400, valid loss-1.3504, acc-0.6362, test loss-1.3469, acc-0.6370
Iter-9980, train loss-1.3490, acc-0.6600, valid loss-1.3494, acc-0.6370, test loss-1.3460, acc-0.6376
Iter-9990, train loss-1.3640, acc-0.6600, valid loss-1.3484, acc-0.6378, test loss-1.3450, acc-0.6379
Iter-10000, train loss-1.3265, acc-0.6600, valid loss-1.3475, acc-0.6382, test loss-1.3440, acc-0.6383
Iter-10010, train loss-1.3147, acc-0.7200, valid loss-1.3465, acc-0.6388, test loss-1.3431, acc-0.6392
Iter-10020, train loss-1.5493, acc-0.4600, valid loss-1.3456, acc-0.6388, test loss-1.3421, acc-0.6391
Iter-10030, train loss-1.4405, acc-0.6400, valid loss-1.3447, acc-0.6392, test loss-1.3412, acc-0.6398
Iter-10040, train loss-1.3013, acc-0.6400, valid loss-1.3438, acc-0.6396, test loss-1.3403, acc-0.6407
Iter-10050, train loss-1.3513, acc-0.6000, valid loss-1.3428, acc-0.6402, test loss-1.3394, acc-0.6406
Iter-10060, train loss-1.2173, acc-0.7200, valid loss-1.3419, acc-0.6404, test loss-1.3384, acc-0.6411
Iter-10070, train loss-1.4892, acc-0.4800, valid loss-1.3410, acc-0.6406, test loss-1.3375, acc-0.6415
Iter-10080, train loss-1.3330, acc-0.6600, valid loss-1.3399, acc-0.6414, test loss-1.3365, acc-0.6422
Iter-10090, train loss-1.3604, acc-0.6600, valid loss-1.3390, acc-0.6418, test loss-1.3355, acc-0.6430
Iter-10100, train loss-1.3702, acc-0.6400, valid loss-1.3381, acc-0.6418, test loss-1.3346, acc-0.6435
Iter-10110, train loss-1.2179, acc-0.7400, valid loss-1.3372, acc-0.6418, test loss-1.3337, acc-0.6434
Iter-10120, train loss-1.3415, acc-0.7000, valid loss-1.3362, acc-0.6420, test loss-1.3327, acc-0.6436
Iter-10130, train loss-1.3299, acc-0.6000, valid loss-1.3353, acc-0.6420, test loss-1.3318, acc-0.6438
Iter-10140, train loss-1.3732, acc-0.6800, valid loss-1.3343, acc-0.6422, test loss-1.3308, acc-0.6444
Iter-10150, train loss-1.3025, acc-0.7000, valid loss-1.3334, acc-0.6426, test loss-1.3299, acc-0.6447
Iter-10160, train loss-1.1355, acc-0.6800, valid loss-1.3324, acc-0.6440, test loss-1.3289, acc-0.6449
Iter-10170, train loss-1.5585, acc-0.5000, valid loss-1.3314, acc-0.6446, test loss-1.3280, acc-0.6449
Iter-10180, train loss-1.2350, acc-0.7000, valid loss-1.3305, acc-0.6446, test loss-1.3271, acc-0.6452
Iter-10190, train loss-1.5165, acc-0.5000, valid loss-1.3295, acc-0.6446, test loss-1.3261, acc-0.6453
Iter-10200, train loss-1.2629, acc-0.6600, valid loss-1.3286, acc-0.6444, test loss-1.3252, acc-0.6458
Iter-10210, train loss-1.4066, acc-0.7000, valid loss-1.3277, acc-0.6452, test loss-1.3243, acc-0.6459
Iter-10220, train loss-1.1637, acc-0.7800, valid loss-1.3268, acc-0.6452, test loss-1.3234, acc-0.6463
Iter-10230, train loss-1.4244, acc-0.5600, valid loss-1.3259, acc-0.6458, test loss-1.3225, acc-0.6470
Iter-10240, train loss-1.2871, acc-0.6600, valid loss-1.3249, acc-0.6470, test loss-1.3215, acc-0.6479
Iter-10250, train loss-1.2836, acc-0.6600, valid loss-1.3240, acc-0.6474, test loss-1.3205, acc-0.6486
Iter-10260, train loss-1.3456, acc-0.6200, valid loss-1.3230, acc-0.6478, test loss-1.3196, acc-0.6490
Iter-10270, train loss-1.3342, acc-0.6400, valid loss-1.3221, acc-0.6478, test loss-1.3187, acc-0.6494
Iter-10280, train loss-1.2217, acc-0.6800, valid loss-1.3212, acc-0.6480, test loss-1.3178, acc-0.6501
Iter-10290, train loss-1.4255, acc-0.5400, valid loss-1.3203, acc-0.6484, test loss-1.3169, acc-0.6500
Iter-10300, train loss-1.2695, acc-0.7200, valid loss-1.3193, acc-0.6492, test loss-1.3159, acc-0.6505
Iter-10310, train loss-1.3506, acc-0.5600, valid loss-1.3184, acc-0.6498, test loss-1.3150, acc-0.6509
Iter-10320, train loss-1.2626, acc-0.7400, valid loss-1.3175, acc-0.6502, test loss-1.3141, acc-0.6513
Iter-10330, train loss-1.4469, acc-0.6200, valid loss-1.3166, acc-0.6504, test loss-1.3132, acc-0.6515
Iter-10340, train loss-1.2314, acc-0.6800, valid loss-1.3156, acc-0.6516, test loss-1.3122, acc-0.6521
Iter-10350, train loss-1.2649, acc-0.6800, valid loss-1.3147, acc-0.6516, test loss-1.3113, acc-0.6526
Iter-10360, train loss-1.3545, acc-0.6200, valid loss-1.3137, acc-0.6524, test loss-1.3103, acc-0.6534
Iter-10370, train loss-1.1555, acc-0.7200, valid loss-1.3128, acc-0.6526, test loss-1.3094, acc-0.6541
Iter-10380, train loss-1.3356, acc-0.5200, valid loss-1.3119, acc-0.6532, test loss-1.3085, acc-0.6547
Iter-10390, train loss-1.2431, acc-0.7000, valid loss-1.3110, acc-0.6542, test loss-1.3076, acc-0.6551
Iter-10400, train loss-1.2226, acc-0.6600, valid loss-1.3101, acc-0.6546, test loss-1.3067, acc-0.6560
Iter-10410, train loss-1.2661, acc-0.6600, valid loss-1.3091, acc-0.6552, test loss-1.3057, acc-0.6568
Iter-10420, train loss-1.3582, acc-0.6800, valid loss-1.3081, acc-0.6558, test loss-1.3047, acc-0.6571
Iter-10430, train loss-1.1966, acc-0.7200, valid loss-1.3072, acc-0.6558, test loss-1.3038, acc-0.6574
Iter-10440, train loss-1.3156, acc-0.6200, valid loss-1.3063, acc-0.6562, test loss-1.3029, acc-0.6578
Iter-10450, train loss-1.3587, acc-0.6200, valid loss-1.3054, acc-0.6566, test loss-1.3020, acc-0.6579
Iter-10460, train loss-1.2975, acc-0.6800, valid loss-1.3045, acc-0.6572, test loss-1.3011, acc-0.6584
Iter-10470, train loss-1.2276, acc-0.6600, valid loss-1.3036, acc-0.6572, test loss-1.3002, acc-0.6587
Iter-10480, train loss-1.3190, acc-0.6800, valid loss-1.3027, acc-0.6574, test loss-1.2993, acc-0.6592
Iter-10490, train loss-1.4584, acc-0.4800, valid loss-1.3017, acc-0.6574, test loss-1.2983, acc-0.6596
Iter-10500, train loss-1.2332, acc-0.7000, valid loss-1.3008, acc-0.6578, test loss-1.2974, acc-0.6598
Iter-10510, train loss-1.1287, acc-0.7800, valid loss-1.3000, acc-0.6584, test loss-1.2966, acc-0.6597
Iter-10520, train loss-1.3389, acc-0.6200, valid loss-1.2991, acc-0.6584, test loss-1.2957, acc-0.6604
Iter-10530, train loss-1.2869, acc-0.7400, valid loss-1.2982, acc-0.6588, test loss-1.2948, acc-0.6604
Iter-10540, train loss-1.1486, acc-0.7600, valid loss-1.2973, acc-0.6592, test loss-1.2939, acc-0.6607
Iter-10550, train loss-1.1927, acc-0.7200, valid loss-1.2964, acc-0.6598, test loss-1.2930, acc-0.6611
Iter-10560, train loss-1.2256, acc-0.7000, valid loss-1.2955, acc-0.6598, test loss-1.2921, acc-0.6615
Iter-10570, train loss-1.3535, acc-0.6200, valid loss-1.2946, acc-0.6600, test loss-1.2911, acc-0.6618
Iter-10580, train loss-1.4112, acc-0.6800, valid loss-1.2936, acc-0.6602, test loss-1.2902, acc-0.6624
Iter-10590, train loss-1.2759, acc-0.6400, valid loss-1.2928, acc-0.6604, test loss-1.2893, acc-0.6625
Iter-10600, train loss-1.2615, acc-0.7000, valid loss-1.2919, acc-0.6604, test loss-1.2885, acc-0.6629
Iter-10610, train loss-1.2293, acc-0.6600, valid loss-1.2910, acc-0.6608, test loss-1.2876, acc-0.6636
Iter-10620, train loss-1.4199, acc-0.5600, valid loss-1.2901, acc-0.6606, test loss-1.2867, acc-0.6637
Iter-10630, train loss-1.3337, acc-0.6800, valid loss-1.2892, acc-0.6606, test loss-1.2858, acc-0.6640
Iter-10640, train loss-1.1251, acc-0.7200, valid loss-1.2883, acc-0.6608, test loss-1.2849, acc-0.6647
Iter-10650, train loss-1.2216, acc-0.7200, valid loss-1.2874, acc-0.6608, test loss-1.2841, acc-0.6647
Iter-10660, train loss-1.2368, acc-0.6200, valid loss-1.2865, acc-0.6610, test loss-1.2831, acc-0.6658
Iter-10670, train loss-1.3111, acc-0.6600, valid loss-1.2856, acc-0.6618, test loss-1.2822, acc-0.6668
Iter-10680, train loss-1.2831, acc-0.6400, valid loss-1.2847, acc-0.6620, test loss-1.2813, acc-0.6670
Iter-10690, train loss-1.3524, acc-0.6400, valid loss-1.2839, acc-0.6626, test loss-1.2805, acc-0.6675
Iter-10700, train loss-1.2494, acc-0.7400, valid loss-1.2830, acc-0.6630, test loss-1.2796, acc-0.6679
Iter-10710, train loss-1.3085, acc-0.6800, valid loss-1.2821, acc-0.6632, test loss-1.2788, acc-0.6681
Iter-10720, train loss-1.2038, acc-0.7000, valid loss-1.2813, acc-0.6636, test loss-1.2780, acc-0.6684
Iter-10730, train loss-1.2766, acc-0.6400, valid loss-1.2804, acc-0.6636, test loss-1.2771, acc-0.6689
Iter-10740, train loss-1.3085, acc-0.6400, valid loss-1.2796, acc-0.6642, test loss-1.2763, acc-0.6691
Iter-10750, train loss-1.2000, acc-0.7000, valid loss-1.2787, acc-0.6638, test loss-1.2754, acc-0.6691
Iter-10760, train loss-1.4599, acc-0.6200, valid loss-1.2779, acc-0.6638, test loss-1.2746, acc-0.6696
Iter-10770, train loss-1.2764, acc-0.6600, valid loss-1.2770, acc-0.6648, test loss-1.2737, acc-0.6697
Iter-10780, train loss-1.4168, acc-0.5600, valid loss-1.2762, acc-0.6646, test loss-1.2729, acc-0.6701
Iter-10790, train loss-1.2791, acc-0.6800, valid loss-1.2752, acc-0.6652, test loss-1.2719, acc-0.6711
Iter-10800, train loss-1.3567, acc-0.6200, valid loss-1.2744, acc-0.6656, test loss-1.2711, acc-0.6715
Iter-10810, train loss-1.1835, acc-0.7600, valid loss-1.2735, acc-0.6662, test loss-1.2702, acc-0.6716
Iter-10820, train loss-1.3596, acc-0.5800, valid loss-1.2726, acc-0.6668, test loss-1.2693, acc-0.6724
Iter-10830, train loss-1.2558, acc-0.6400, valid loss-1.2718, acc-0.6672, test loss-1.2685, acc-0.6726
Iter-10840, train loss-1.3126, acc-0.6800, valid loss-1.2708, acc-0.6676, test loss-1.2676, acc-0.6729
Iter-10850, train loss-1.3729, acc-0.6800, valid loss-1.2700, acc-0.6680, test loss-1.2667, acc-0.6730
Iter-10860, train loss-1.2924, acc-0.6400, valid loss-1.2691, acc-0.6686, test loss-1.2659, acc-0.6740
Iter-10870, train loss-1.2790, acc-0.6600, valid loss-1.2683, acc-0.6688, test loss-1.2650, acc-0.6743
Iter-10880, train loss-1.2710, acc-0.6800, valid loss-1.2674, acc-0.6694, test loss-1.2642, acc-0.6745
Iter-10890, train loss-1.3054, acc-0.6200, valid loss-1.2665, acc-0.6696, test loss-1.2633, acc-0.6748
Iter-10900, train loss-1.3135, acc-0.6600, valid loss-1.2657, acc-0.6696, test loss-1.2624, acc-0.6750
Iter-10910, train loss-1.3494, acc-0.6400, valid loss-1.2648, acc-0.6698, test loss-1.2616, acc-0.6750
Iter-10920, train loss-1.2510, acc-0.7200, valid loss-1.2640, acc-0.6696, test loss-1.2608, acc-0.6752
Iter-10930, train loss-1.4114, acc-0.6600, valid loss-1.2631, acc-0.6700, test loss-1.2599, acc-0.6754
Iter-10940, train loss-1.2388, acc-0.7000, valid loss-1.2622, acc-0.6706, test loss-1.2591, acc-0.6759
Iter-10950, train loss-1.2401, acc-0.6200, valid loss-1.2613, acc-0.6706, test loss-1.2582, acc-0.6762
Iter-10960, train loss-1.3519, acc-0.6600, valid loss-1.2606, acc-0.6708, test loss-1.2574, acc-0.6768
Iter-10970, train loss-1.2393, acc-0.6400, valid loss-1.2597, acc-0.6708, test loss-1.2565, acc-0.6773
Iter-10980, train loss-1.3578, acc-0.5000, valid loss-1.2588, acc-0.6710, test loss-1.2557, acc-0.6778
Iter-10990, train loss-1.3832, acc-0.6400, valid loss-1.2580, acc-0.6710, test loss-1.2548, acc-0.6782
Iter-11000, train loss-1.1847, acc-0.7200, valid loss-1.2572, acc-0.6712, test loss-1.2540, acc-0.6786
Iter-11010, train loss-1.2562, acc-0.6600, valid loss-1.2563, acc-0.6714, test loss-1.2532, acc-0.6787
Iter-11020, train loss-1.2070, acc-0.7000, valid loss-1.2554, acc-0.6722, test loss-1.2523, acc-0.6791
Iter-11030, train loss-1.3272, acc-0.6200, valid loss-1.2546, acc-0.6728, test loss-1.2515, acc-0.6793
Iter-11040, train loss-1.2860, acc-0.6000, valid loss-1.2537, acc-0.6738, test loss-1.2506, acc-0.6793
Iter-11050, train loss-1.1329, acc-0.6000, valid loss-1.2529, acc-0.6744, test loss-1.2497, acc-0.6792
Iter-11060, train loss-1.3755, acc-0.5600, valid loss-1.2520, acc-0.6744, test loss-1.2488, acc-0.6795
Iter-11070, train loss-1.1073, acc-0.7800, valid loss-1.2512, acc-0.6750, test loss-1.2480, acc-0.6797
Iter-11080, train loss-1.1922, acc-0.7000, valid loss-1.2503, acc-0.6750, test loss-1.2471, acc-0.6802
Iter-11090, train loss-1.1334, acc-0.7200, valid loss-1.2494, acc-0.6762, test loss-1.2463, acc-0.6803
Iter-11100, train loss-1.2337, acc-0.7000, valid loss-1.2486, acc-0.6762, test loss-1.2455, acc-0.6807
Iter-11110, train loss-1.3295, acc-0.6400, valid loss-1.2477, acc-0.6764, test loss-1.2446, acc-0.6816
Iter-11120, train loss-1.1083, acc-0.7600, valid loss-1.2469, acc-0.6770, test loss-1.2438, acc-0.6815
Iter-11130, train loss-1.4078, acc-0.5800, valid loss-1.2461, acc-0.6772, test loss-1.2430, acc-0.6817
Iter-11140, train loss-1.3058, acc-0.6000, valid loss-1.2453, acc-0.6774, test loss-1.2422, acc-0.6822
Iter-11150, train loss-1.2951, acc-0.6400, valid loss-1.2444, acc-0.6774, test loss-1.2413, acc-0.6826
Iter-11160, train loss-1.1855, acc-0.6800, valid loss-1.2436, acc-0.6780, test loss-1.2405, acc-0.6831
Iter-11170, train loss-1.1786, acc-0.7200, valid loss-1.2428, acc-0.6782, test loss-1.2397, acc-0.6833
Iter-11180, train loss-1.1506, acc-0.6800, valid loss-1.2420, acc-0.6786, test loss-1.2389, acc-0.6837
Iter-11190, train loss-1.1145, acc-0.7400, valid loss-1.2411, acc-0.6792, test loss-1.2381, acc-0.6842
Iter-11200, train loss-1.3591, acc-0.6400, valid loss-1.2403, acc-0.6796, test loss-1.2372, acc-0.6844
Iter-11210, train loss-1.1592, acc-0.7600, valid loss-1.2394, acc-0.6802, test loss-1.2364, acc-0.6848
Iter-11220, train loss-1.3102, acc-0.6000, valid loss-1.2387, acc-0.6802, test loss-1.2356, acc-0.6851
Iter-11230, train loss-1.2694, acc-0.6400, valid loss-1.2378, acc-0.6808, test loss-1.2348, acc-0.6860
Iter-11240, train loss-1.3102, acc-0.6400, valid loss-1.2370, acc-0.6808, test loss-1.2340, acc-0.6862
Iter-11250, train loss-1.2425, acc-0.6800, valid loss-1.2361, acc-0.6812, test loss-1.2331, acc-0.6865
Iter-11260, train loss-1.2351, acc-0.7600, valid loss-1.2353, acc-0.6816, test loss-1.2323, acc-0.6870
Iter-11270, train loss-1.1919, acc-0.7000, valid loss-1.2345, acc-0.6820, test loss-1.2315, acc-0.6872
Iter-11280, train loss-1.1589, acc-0.6800, valid loss-1.2337, acc-0.6828, test loss-1.2307, acc-0.6873
Iter-11290, train loss-1.2460, acc-0.5800, valid loss-1.2329, acc-0.6830, test loss-1.2299, acc-0.6878
Iter-11300, train loss-1.2349, acc-0.7400, valid loss-1.2320, acc-0.6824, test loss-1.2290, acc-0.6882
Iter-11310, train loss-1.2549, acc-0.6800, valid loss-1.2312, acc-0.6828, test loss-1.2282, acc-0.6883
Iter-11320, train loss-1.2134, acc-0.7400, valid loss-1.2304, acc-0.6830, test loss-1.2273, acc-0.6889
Iter-11330, train loss-1.3278, acc-0.6400, valid loss-1.2295, acc-0.6832, test loss-1.2265, acc-0.6896
Iter-11340, train loss-1.2039, acc-0.7000, valid loss-1.2287, acc-0.6840, test loss-1.2257, acc-0.6900
Iter-11350, train loss-1.2052, acc-0.6800, valid loss-1.2279, acc-0.6842, test loss-1.2249, acc-0.6901
Iter-11360, train loss-1.2345, acc-0.6200, valid loss-1.2271, acc-0.6846, test loss-1.2241, acc-0.6906
Iter-11370, train loss-1.2446, acc-0.7000, valid loss-1.2263, acc-0.6848, test loss-1.2233, acc-0.6910
Iter-11380, train loss-1.1660, acc-0.7000, valid loss-1.2255, acc-0.6846, test loss-1.2225, acc-0.6917
Iter-11390, train loss-1.4222, acc-0.6000, valid loss-1.2247, acc-0.6848, test loss-1.2217, acc-0.6921
Iter-11400, train loss-1.2451, acc-0.6400, valid loss-1.2239, acc-0.6852, test loss-1.2209, acc-0.6924
Iter-11410, train loss-1.2739, acc-0.6400, valid loss-1.2231, acc-0.6854, test loss-1.2201, acc-0.6928
Iter-11420, train loss-1.3266, acc-0.6000, valid loss-1.2223, acc-0.6858, test loss-1.2192, acc-0.6928
Iter-11430, train loss-1.2370, acc-0.6400, valid loss-1.2215, acc-0.6860, test loss-1.2184, acc-0.6930
Iter-11440, train loss-1.2560, acc-0.6400, valid loss-1.2207, acc-0.6864, test loss-1.2177, acc-0.6938
Iter-11450, train loss-1.1633, acc-0.7400, valid loss-1.2200, acc-0.6870, test loss-1.2169, acc-0.6943
Iter-11460, train loss-1.2554, acc-0.6600, valid loss-1.2192, acc-0.6868, test loss-1.2162, acc-0.6943
Iter-11470, train loss-1.1639, acc-0.7400, valid loss-1.2184, acc-0.6870, test loss-1.2154, acc-0.6951
Iter-11480, train loss-1.2274, acc-0.6800, valid loss-1.2176, acc-0.6874, test loss-1.2146, acc-0.6956
Iter-11490, train loss-1.1430, acc-0.7200, valid loss-1.2168, acc-0.6876, test loss-1.2138, acc-0.6955
Iter-11500, train loss-1.1224, acc-0.7200, valid loss-1.2160, acc-0.6878, test loss-1.2130, acc-0.6960
Iter-11510, train loss-1.2666, acc-0.6600, valid loss-1.2152, acc-0.6884, test loss-1.2122, acc-0.6965
Iter-11520, train loss-1.0192, acc-0.7600, valid loss-1.2144, acc-0.6888, test loss-1.2114, acc-0.6966
Iter-11530, train loss-1.1627, acc-0.6600, valid loss-1.2136, acc-0.6890, test loss-1.2106, acc-0.6973
Iter-11540, train loss-1.3211, acc-0.6600, valid loss-1.2129, acc-0.6894, test loss-1.2098, acc-0.6974
Iter-11550, train loss-1.1542, acc-0.7400, valid loss-1.2121, acc-0.6894, test loss-1.2091, acc-0.6981
Iter-11560, train loss-1.1916, acc-0.6800, valid loss-1.2113, acc-0.6900, test loss-1.2082, acc-0.6984
Iter-11570, train loss-1.0952, acc-0.7400, valid loss-1.2104, acc-0.6900, test loss-1.2074, acc-0.6988
Iter-11580, train loss-1.2962, acc-0.6600, valid loss-1.2096, acc-0.6900, test loss-1.2066, acc-0.6990
Iter-11590, train loss-1.2115, acc-0.6800, valid loss-1.2089, acc-0.6898, test loss-1.2058, acc-0.6993
Iter-11600, train loss-1.1748, acc-0.6800, valid loss-1.2081, acc-0.6900, test loss-1.2051, acc-0.6993
Iter-11610, train loss-1.1329, acc-0.7000, valid loss-1.2073, acc-0.6902, test loss-1.2042, acc-0.6998
Iter-11620, train loss-1.2159, acc-0.6400, valid loss-1.2065, acc-0.6910, test loss-1.2034, acc-0.6998
Iter-11630, train loss-1.1848, acc-0.6800, valid loss-1.2057, acc-0.6914, test loss-1.2026, acc-0.6999
Iter-11640, train loss-1.2369, acc-0.7800, valid loss-1.2049, acc-0.6920, test loss-1.2019, acc-0.7002
Iter-11650, train loss-1.2639, acc-0.6200, valid loss-1.2042, acc-0.6922, test loss-1.2012, acc-0.7003
Iter-11660, train loss-1.0455, acc-0.7600, valid loss-1.2034, acc-0.6922, test loss-1.2004, acc-0.7004
Iter-11670, train loss-1.1469, acc-0.7200, valid loss-1.2026, acc-0.6926, test loss-1.1996, acc-0.7009
Iter-11680, train loss-1.2492, acc-0.7000, valid loss-1.2018, acc-0.6928, test loss-1.1988, acc-0.7011
Iter-11690, train loss-1.2441, acc-0.7200, valid loss-1.2010, acc-0.6930, test loss-1.1980, acc-0.7014
Iter-11700, train loss-1.2743, acc-0.6600, valid loss-1.2002, acc-0.6932, test loss-1.1972, acc-0.7015
Iter-11710, train loss-1.1547, acc-0.7400, valid loss-1.1994, acc-0.6934, test loss-1.1964, acc-0.7017
Iter-11720, train loss-1.2077, acc-0.6200, valid loss-1.1986, acc-0.6940, test loss-1.1957, acc-0.7018
Iter-11730, train loss-1.2586, acc-0.6600, valid loss-1.1979, acc-0.6944, test loss-1.1950, acc-0.7023
Iter-11740, train loss-1.1630, acc-0.7200, valid loss-1.1971, acc-0.6942, test loss-1.1942, acc-0.7027
Iter-11750, train loss-1.1586, acc-0.7200, valid loss-1.1963, acc-0.6946, test loss-1.1934, acc-0.7031
Iter-11760, train loss-1.1006, acc-0.8200, valid loss-1.1954, acc-0.6946, test loss-1.1925, acc-0.7031
Iter-11770, train loss-1.2277, acc-0.6800, valid loss-1.1947, acc-0.6950, test loss-1.1918, acc-0.7032
Iter-11780, train loss-1.3492, acc-0.6200, valid loss-1.1940, acc-0.6952, test loss-1.1911, acc-0.7032
Iter-11790, train loss-1.0699, acc-0.7200, valid loss-1.1932, acc-0.6954, test loss-1.1903, acc-0.7031
Iter-11800, train loss-1.2176, acc-0.6800, valid loss-1.1924, acc-0.6954, test loss-1.1895, acc-0.7038
Iter-11810, train loss-1.2009, acc-0.6800, valid loss-1.1916, acc-0.6956, test loss-1.1887, acc-0.7042
Iter-11820, train loss-1.1384, acc-0.7000, valid loss-1.1908, acc-0.6962, test loss-1.1879, acc-0.7045
Iter-11830, train loss-1.1982, acc-0.6800, valid loss-1.1900, acc-0.6968, test loss-1.1872, acc-0.7046
Iter-11840, train loss-1.3073, acc-0.6400, valid loss-1.1893, acc-0.6970, test loss-1.1864, acc-0.7053
Iter-11850, train loss-1.1441, acc-0.7600, valid loss-1.1885, acc-0.6970, test loss-1.1857, acc-0.7057
Iter-11860, train loss-1.3285, acc-0.6200, valid loss-1.1878, acc-0.6976, test loss-1.1849, acc-0.7059
Iter-11870, train loss-1.3128, acc-0.6600, valid loss-1.1871, acc-0.6976, test loss-1.1842, acc-0.7060
Iter-11880, train loss-1.0734, acc-0.7000, valid loss-1.1863, acc-0.6976, test loss-1.1835, acc-0.7062
Iter-11890, train loss-1.3178, acc-0.6600, valid loss-1.1855, acc-0.6980, test loss-1.1827, acc-0.7063
Iter-11900, train loss-1.2682, acc-0.6600, valid loss-1.1848, acc-0.6982, test loss-1.1819, acc-0.7065
Iter-11910, train loss-1.1372, acc-0.7000, valid loss-1.1840, acc-0.6984, test loss-1.1811, acc-0.7068
Iter-11920, train loss-1.0886, acc-0.7800, valid loss-1.1832, acc-0.6984, test loss-1.1803, acc-0.7070
Iter-11930, train loss-1.3090, acc-0.6600, valid loss-1.1824, acc-0.6982, test loss-1.1796, acc-0.7073
Iter-11940, train loss-1.1946, acc-0.6800, valid loss-1.1816, acc-0.6984, test loss-1.1788, acc-0.7073
Iter-11950, train loss-1.2202, acc-0.6400, valid loss-1.1809, acc-0.6988, test loss-1.1781, acc-0.7079
Iter-11960, train loss-1.2613, acc-0.7200, valid loss-1.1801, acc-0.6994, test loss-1.1773, acc-0.7084
Iter-11970, train loss-1.1236, acc-0.7200, valid loss-1.1793, acc-0.7000, test loss-1.1765, acc-0.7087
Iter-11980, train loss-1.3307, acc-0.6200, valid loss-1.1786, acc-0.7000, test loss-1.1758, acc-0.7086
Iter-11990, train loss-1.1269, acc-0.7200, valid loss-1.1778, acc-0.7004, test loss-1.1750, acc-0.7089
Iter-12000, train loss-1.2446, acc-0.6400, valid loss-1.1771, acc-0.7002, test loss-1.1743, acc-0.7095
Iter-12010, train loss-1.0686, acc-0.7400, valid loss-1.1763, acc-0.7006, test loss-1.1735, acc-0.7097
Iter-12020, train loss-1.3192, acc-0.6200, valid loss-1.1755, acc-0.7012, test loss-1.1727, acc-0.7098
Iter-12030, train loss-1.1676, acc-0.7200, valid loss-1.1747, acc-0.7012, test loss-1.1720, acc-0.7100
Iter-12040, train loss-1.1439, acc-0.7200, valid loss-1.1740, acc-0.7018, test loss-1.1713, acc-0.7102
Iter-12050, train loss-1.0752, acc-0.7800, valid loss-1.1733, acc-0.7022, test loss-1.1706, acc-0.7103
Iter-12060, train loss-1.1594, acc-0.7600, valid loss-1.1725, acc-0.7028, test loss-1.1698, acc-0.7107
Iter-12070, train loss-1.1643, acc-0.6800, valid loss-1.1717, acc-0.7030, test loss-1.1690, acc-0.7108
Iter-12080, train loss-1.2528, acc-0.6800, valid loss-1.1710, acc-0.7026, test loss-1.1683, acc-0.7113
Iter-12090, train loss-1.0543, acc-0.7200, valid loss-1.1702, acc-0.7038, test loss-1.1675, acc-0.7113
Iter-12100, train loss-1.1896, acc-0.6400, valid loss-1.1694, acc-0.7044, test loss-1.1667, acc-0.7116
Iter-12110, train loss-1.1227, acc-0.6800, valid loss-1.1687, acc-0.7044, test loss-1.1660, acc-0.7116
Iter-12120, train loss-1.2399, acc-0.6800, valid loss-1.1680, acc-0.7048, test loss-1.1653, acc-0.7121
Iter-12130, train loss-1.1619, acc-0.6400, valid loss-1.1673, acc-0.7054, test loss-1.1646, acc-0.7128
Iter-12140, train loss-1.0862, acc-0.7200, valid loss-1.1665, acc-0.7056, test loss-1.1639, acc-0.7135
Iter-12150, train loss-1.1775, acc-0.7200, valid loss-1.1657, acc-0.7058, test loss-1.1631, acc-0.7142
Iter-12160, train loss-1.1992, acc-0.6200, valid loss-1.1650, acc-0.7054, test loss-1.1623, acc-0.7139
Iter-12170, train loss-1.0745, acc-0.8400, valid loss-1.1643, acc-0.7060, test loss-1.1616, acc-0.7139
Iter-12180, train loss-1.3207, acc-0.5600, valid loss-1.1635, acc-0.7062, test loss-1.1609, acc-0.7143
Iter-12190, train loss-1.0330, acc-0.8000, valid loss-1.1628, acc-0.7064, test loss-1.1602, acc-0.7142
Iter-12200, train loss-1.0903, acc-0.7400, valid loss-1.1621, acc-0.7066, test loss-1.1595, acc-0.7144
Iter-12210, train loss-1.0573, acc-0.7400, valid loss-1.1614, acc-0.7066, test loss-1.1587, acc-0.7149
Iter-12220, train loss-1.2172, acc-0.7200, valid loss-1.1605, acc-0.7068, test loss-1.1580, acc-0.7151
Iter-12230, train loss-1.2937, acc-0.5000, valid loss-1.1599, acc-0.7066, test loss-1.1573, acc-0.7151
Iter-12240, train loss-1.1806, acc-0.6600, valid loss-1.1591, acc-0.7070, test loss-1.1565, acc-0.7155
Iter-12250, train loss-1.1016, acc-0.7400, valid loss-1.1583, acc-0.7072, test loss-1.1558, acc-0.7161
Iter-12260, train loss-1.1791, acc-0.7200, valid loss-1.1576, acc-0.7074, test loss-1.1550, acc-0.7159
Iter-12270, train loss-1.1321, acc-0.6400, valid loss-1.1569, acc-0.7076, test loss-1.1544, acc-0.7160
Iter-12280, train loss-1.0641, acc-0.7400, valid loss-1.1562, acc-0.7086, test loss-1.1536, acc-0.7160
Iter-12290, train loss-1.1924, acc-0.7000, valid loss-1.1555, acc-0.7084, test loss-1.1529, acc-0.7162
Iter-12300, train loss-1.0841, acc-0.8200, valid loss-1.1547, acc-0.7090, test loss-1.1521, acc-0.7165
Iter-12310, train loss-1.0611, acc-0.7600, valid loss-1.1540, acc-0.7092, test loss-1.1514, acc-0.7166
Iter-12320, train loss-1.1384, acc-0.7000, valid loss-1.1533, acc-0.7094, test loss-1.1508, acc-0.7168
Iter-12330, train loss-0.9436, acc-0.8400, valid loss-1.1525, acc-0.7098, test loss-1.1500, acc-0.7173
Iter-12340, train loss-1.0642, acc-0.7800, valid loss-1.1518, acc-0.7102, test loss-1.1493, acc-0.7175
Iter-12350, train loss-1.1837, acc-0.6800, valid loss-1.1511, acc-0.7104, test loss-1.1485, acc-0.7181
Iter-12360, train loss-1.0663, acc-0.7200, valid loss-1.1503, acc-0.7104, test loss-1.1478, acc-0.7181
Iter-12370, train loss-1.2160, acc-0.6800, valid loss-1.1496, acc-0.7102, test loss-1.1471, acc-0.7181
Iter-12380, train loss-1.2019, acc-0.6200, valid loss-1.1489, acc-0.7110, test loss-1.1464, acc-0.7184
Iter-12390, train loss-1.2378, acc-0.6600, valid loss-1.1481, acc-0.7104, test loss-1.1456, acc-0.7183
Iter-12400, train loss-1.0208, acc-0.7600, valid loss-1.1474, acc-0.7112, test loss-1.1449, acc-0.7183
Iter-12410, train loss-1.3022, acc-0.6000, valid loss-1.1467, acc-0.7116, test loss-1.1442, acc-0.7186
Iter-12420, train loss-1.0902, acc-0.7000, valid loss-1.1460, acc-0.7110, test loss-1.1434, acc-0.7184
Iter-12430, train loss-1.2501, acc-0.7400, valid loss-1.1452, acc-0.7114, test loss-1.1427, acc-0.7185
Iter-12440, train loss-1.1440, acc-0.6200, valid loss-1.1445, acc-0.7120, test loss-1.1420, acc-0.7185
Iter-12450, train loss-1.0193, acc-0.7600, valid loss-1.1438, acc-0.7122, test loss-1.1413, acc-0.7187
Iter-12460, train loss-1.1156, acc-0.7200, valid loss-1.1431, acc-0.7122, test loss-1.1406, acc-0.7190
Iter-12470, train loss-1.2041, acc-0.7400, valid loss-1.1424, acc-0.7128, test loss-1.1399, acc-0.7192
Iter-12480, train loss-1.2234, acc-0.6600, valid loss-1.1416, acc-0.7136, test loss-1.1392, acc-0.7194
Iter-12490, train loss-1.1340, acc-0.7800, valid loss-1.1409, acc-0.7138, test loss-1.1384, acc-0.7196
Iter-12500, train loss-1.1680, acc-0.7000, valid loss-1.1402, acc-0.7138, test loss-1.1377, acc-0.7197
Iter-12510, train loss-1.1672, acc-0.7200, valid loss-1.1395, acc-0.7140, test loss-1.1370, acc-0.7198
Iter-12520, train loss-1.0107, acc-0.8200, valid loss-1.1387, acc-0.7142, test loss-1.1362, acc-0.7198
Iter-12530, train loss-1.1882, acc-0.6800, valid loss-1.1380, acc-0.7142, test loss-1.1355, acc-0.7203
Iter-12540, train loss-1.1457, acc-0.7200, valid loss-1.1372, acc-0.7144, test loss-1.1348, acc-0.7205
Iter-12550, train loss-1.0161, acc-0.8000, valid loss-1.1365, acc-0.7146, test loss-1.1340, acc-0.7205
Iter-12560, train loss-1.1498, acc-0.7000, valid loss-1.1358, acc-0.7146, test loss-1.1333, acc-0.7207
Iter-12570, train loss-1.1733, acc-0.7200, valid loss-1.1350, acc-0.7148, test loss-1.1326, acc-0.7208
Iter-12580, train loss-1.2305, acc-0.6200, valid loss-1.1343, acc-0.7156, test loss-1.1319, acc-0.7211
Iter-12590, train loss-0.9617, acc-0.7600, valid loss-1.1336, acc-0.7158, test loss-1.1312, acc-0.7214
Iter-12600, train loss-1.1460, acc-0.6600, valid loss-1.1329, acc-0.7162, test loss-1.1305, acc-0.7216
Iter-12610, train loss-1.1199, acc-0.7600, valid loss-1.1322, acc-0.7162, test loss-1.1298, acc-0.7218
Iter-12620, train loss-1.0940, acc-0.7400, valid loss-1.1315, acc-0.7164, test loss-1.1291, acc-0.7218
Iter-12630, train loss-1.2368, acc-0.6600, valid loss-1.1309, acc-0.7166, test loss-1.1285, acc-0.7224
Iter-12640, train loss-1.1134, acc-0.7000, valid loss-1.1302, acc-0.7166, test loss-1.1278, acc-0.7228
Iter-12650, train loss-1.0523, acc-0.8000, valid loss-1.1294, acc-0.7168, test loss-1.1271, acc-0.7228
Iter-12660, train loss-1.1477, acc-0.6600, valid loss-1.1287, acc-0.7176, test loss-1.1264, acc-0.7231
Iter-12670, train loss-1.1285, acc-0.7200, valid loss-1.1281, acc-0.7176, test loss-1.1257, acc-0.7230
Iter-12680, train loss-1.1917, acc-0.7000, valid loss-1.1274, acc-0.7184, test loss-1.1250, acc-0.7228
Iter-12690, train loss-1.2060, acc-0.7200, valid loss-1.1267, acc-0.7184, test loss-1.1244, acc-0.7230
Iter-12700, train loss-1.0697, acc-0.7600, valid loss-1.1259, acc-0.7188, test loss-1.1236, acc-0.7231
Iter-12710, train loss-1.3496, acc-0.6000, valid loss-1.1253, acc-0.7192, test loss-1.1230, acc-0.7231
Iter-12720, train loss-1.1002, acc-0.7000, valid loss-1.1245, acc-0.7194, test loss-1.1223, acc-0.7233
Iter-12730, train loss-1.2090, acc-0.6800, valid loss-1.1238, acc-0.7196, test loss-1.1215, acc-0.7236
Iter-12740, train loss-0.9843, acc-0.7800, valid loss-1.1231, acc-0.7200, test loss-1.1208, acc-0.7237
Iter-12750, train loss-1.2050, acc-0.6800, valid loss-1.1224, acc-0.7204, test loss-1.1201, acc-0.7240
Iter-12760, train loss-1.1219, acc-0.7600, valid loss-1.1217, acc-0.7210, test loss-1.1195, acc-0.7240
Iter-12770, train loss-1.3071, acc-0.6200, valid loss-1.1211, acc-0.7210, test loss-1.1188, acc-0.7243
Iter-12780, train loss-1.1651, acc-0.7000, valid loss-1.1203, acc-0.7214, test loss-1.1181, acc-0.7246
Iter-12790, train loss-1.1425, acc-0.6800, valid loss-1.1197, acc-0.7214, test loss-1.1174, acc-0.7246
Iter-12800, train loss-1.1029, acc-0.6800, valid loss-1.1189, acc-0.7216, test loss-1.1167, acc-0.7249
Iter-12810, train loss-1.1358, acc-0.7200, valid loss-1.1182, acc-0.7218, test loss-1.1160, acc-0.7250
Iter-12820, train loss-1.0925, acc-0.7800, valid loss-1.1176, acc-0.7218, test loss-1.1154, acc-0.7257
Iter-12830, train loss-0.9090, acc-0.8400, valid loss-1.1169, acc-0.7218, test loss-1.1147, acc-0.7258
Iter-12840, train loss-1.1083, acc-0.6000, valid loss-1.1162, acc-0.7222, test loss-1.1140, acc-0.7259
Iter-12850, train loss-1.1623, acc-0.6800, valid loss-1.1155, acc-0.7226, test loss-1.1133, acc-0.7261
Iter-12860, train loss-1.1107, acc-0.7200, valid loss-1.1149, acc-0.7230, test loss-1.1126, acc-0.7264
Iter-12870, train loss-1.2373, acc-0.6400, valid loss-1.1142, acc-0.7230, test loss-1.1120, acc-0.7263
Iter-12880, train loss-1.0258, acc-0.7600, valid loss-1.1135, acc-0.7228, test loss-1.1113, acc-0.7265
Iter-12890, train loss-0.9784, acc-0.7800, valid loss-1.1128, acc-0.7232, test loss-1.1106, acc-0.7270
Iter-12900, train loss-1.2036, acc-0.6200, valid loss-1.1121, acc-0.7232, test loss-1.1099, acc-0.7269
Iter-12910, train loss-1.0061, acc-0.7800, valid loss-1.1115, acc-0.7234, test loss-1.1093, acc-0.7271
Iter-12920, train loss-1.0805, acc-0.7400, valid loss-1.1108, acc-0.7232, test loss-1.1086, acc-0.7273
Iter-12930, train loss-1.2344, acc-0.6200, valid loss-1.1101, acc-0.7234, test loss-1.1079, acc-0.7271
Iter-12940, train loss-1.1705, acc-0.7200, valid loss-1.1095, acc-0.7234, test loss-1.1073, acc-0.7270
Iter-12950, train loss-1.2272, acc-0.6400, valid loss-1.1088, acc-0.7236, test loss-1.1066, acc-0.7273
Iter-12960, train loss-1.3007, acc-0.6600, valid loss-1.1082, acc-0.7236, test loss-1.1060, acc-0.7276
Iter-12970, train loss-1.0632, acc-0.7000, valid loss-1.1075, acc-0.7236, test loss-1.1053, acc-0.7275
Iter-12980, train loss-1.1919, acc-0.6800, valid loss-1.1068, acc-0.7240, test loss-1.1046, acc-0.7272
Iter-12990, train loss-1.0421, acc-0.6800, valid loss-1.1062, acc-0.7248, test loss-1.1039, acc-0.7274
Iter-13000, train loss-1.1806, acc-0.7200, valid loss-1.1055, acc-0.7250, test loss-1.1033, acc-0.7274
Iter-13010, train loss-1.1559, acc-0.7200, valid loss-1.1049, acc-0.7252, test loss-1.1027, acc-0.7278
Iter-13020, train loss-1.1552, acc-0.7000, valid loss-1.1043, acc-0.7250, test loss-1.1021, acc-0.7278
Iter-13030, train loss-1.1023, acc-0.7800, valid loss-1.1036, acc-0.7254, test loss-1.1014, acc-0.7285
Iter-13040, train loss-1.0688, acc-0.7200, valid loss-1.1029, acc-0.7252, test loss-1.1007, acc-0.7289
Iter-13050, train loss-1.0395, acc-0.7800, valid loss-1.1022, acc-0.7252, test loss-1.1001, acc-0.7296
Iter-13060, train loss-1.0767, acc-0.7400, valid loss-1.1016, acc-0.7256, test loss-1.0994, acc-0.7296
Iter-13070, train loss-1.0942, acc-0.7600, valid loss-1.1010, acc-0.7258, test loss-1.0988, acc-0.7298
Iter-13080, train loss-1.0655, acc-0.7000, valid loss-1.1003, acc-0.7258, test loss-1.0981, acc-0.7300
Iter-13090, train loss-1.1424, acc-0.7000, valid loss-1.0996, acc-0.7264, test loss-1.0974, acc-0.7298
Iter-13100, train loss-1.1398, acc-0.6200, valid loss-1.0989, acc-0.7266, test loss-1.0967, acc-0.7300
Iter-13110, train loss-1.1028, acc-0.7200, valid loss-1.0982, acc-0.7266, test loss-1.0961, acc-0.7305
Iter-13120, train loss-1.0630, acc-0.7400, valid loss-1.0976, acc-0.7266, test loss-1.0954, acc-0.7306
Iter-13130, train loss-1.1951, acc-0.6800, valid loss-1.0969, acc-0.7272, test loss-1.0948, acc-0.7310
Iter-13140, train loss-1.1185, acc-0.8200, valid loss-1.0963, acc-0.7272, test loss-1.0942, acc-0.7308
Iter-13150, train loss-1.1467, acc-0.6200, valid loss-1.0956, acc-0.7274, test loss-1.0935, acc-0.7308
Iter-13160, train loss-1.1760, acc-0.6400, valid loss-1.0950, acc-0.7280, test loss-1.0929, acc-0.7310
Iter-13170, train loss-1.2507, acc-0.6000, valid loss-1.0943, acc-0.7280, test loss-1.0922, acc-0.7311
Iter-13180, train loss-1.1119, acc-0.6800, valid loss-1.0937, acc-0.7286, test loss-1.0916, acc-0.7312
Iter-13190, train loss-1.2546, acc-0.6600, valid loss-1.0931, acc-0.7288, test loss-1.0910, acc-0.7312
Iter-13200, train loss-1.1005, acc-0.7000, valid loss-1.0924, acc-0.7286, test loss-1.0903, acc-0.7314
Iter-13210, train loss-1.0517, acc-0.6800, valid loss-1.0918, acc-0.7292, test loss-1.0897, acc-0.7319
Iter-13220, train loss-1.0841, acc-0.7200, valid loss-1.0911, acc-0.7294, test loss-1.0891, acc-0.7318
Iter-13230, train loss-0.9088, acc-0.8600, valid loss-1.0905, acc-0.7296, test loss-1.0884, acc-0.7319
Iter-13240, train loss-1.1144, acc-0.6800, valid loss-1.0899, acc-0.7300, test loss-1.0878, acc-0.7320
Iter-13250, train loss-1.0131, acc-0.8400, valid loss-1.0892, acc-0.7302, test loss-1.0872, acc-0.7324
Iter-13260, train loss-1.2166, acc-0.7400, valid loss-1.0886, acc-0.7306, test loss-1.0865, acc-0.7325
Iter-13270, train loss-1.1581, acc-0.6600, valid loss-1.0879, acc-0.7310, test loss-1.0859, acc-0.7326
Iter-13280, train loss-1.1473, acc-0.7000, valid loss-1.0872, acc-0.7312, test loss-1.0852, acc-0.7329
Iter-13290, train loss-1.1264, acc-0.7400, valid loss-1.0866, acc-0.7310, test loss-1.0846, acc-0.7330
Iter-13300, train loss-1.1153, acc-0.7800, valid loss-1.0860, acc-0.7318, test loss-1.0839, acc-0.7330
Iter-13310, train loss-1.0827, acc-0.8000, valid loss-1.0853, acc-0.7324, test loss-1.0833, acc-0.7330
Iter-13320, train loss-1.2316, acc-0.7200, valid loss-1.0847, acc-0.7326, test loss-1.0826, acc-0.7331
Iter-13330, train loss-0.9169, acc-0.7800, valid loss-1.0840, acc-0.7332, test loss-1.0820, acc-0.7335
Iter-13340, train loss-1.1799, acc-0.6800, valid loss-1.0834, acc-0.7332, test loss-1.0814, acc-0.7336
Iter-13350, train loss-1.0368, acc-0.6600, valid loss-1.0827, acc-0.7334, test loss-1.0808, acc-0.7334
Iter-13360, train loss-1.0925, acc-0.6800, valid loss-1.0821, acc-0.7332, test loss-1.0801, acc-0.7334
Iter-13370, train loss-1.0865, acc-0.7600, valid loss-1.0815, acc-0.7334, test loss-1.0795, acc-0.7335
Iter-13380, train loss-0.9920, acc-0.8000, valid loss-1.0809, acc-0.7338, test loss-1.0789, acc-0.7337
Iter-13390, train loss-1.0159, acc-0.8000, valid loss-1.0802, acc-0.7340, test loss-1.0782, acc-0.7340
Iter-13400, train loss-1.0998, acc-0.7200, valid loss-1.0795, acc-0.7340, test loss-1.0776, acc-0.7344
Iter-13410, train loss-1.1439, acc-0.6800, valid loss-1.0789, acc-0.7340, test loss-1.0770, acc-0.7349
Iter-13420, train loss-1.1845, acc-0.7000, valid loss-1.0782, acc-0.7340, test loss-1.0763, acc-0.7352
Iter-13430, train loss-1.1877, acc-0.7400, valid loss-1.0776, acc-0.7344, test loss-1.0756, acc-0.7353
Iter-13440, train loss-1.1653, acc-0.6800, valid loss-1.0769, acc-0.7344, test loss-1.0750, acc-0.7358
Iter-13450, train loss-1.1492, acc-0.7200, valid loss-1.0763, acc-0.7340, test loss-1.0743, acc-0.7360
Iter-13460, train loss-1.1628, acc-0.6600, valid loss-1.0756, acc-0.7344, test loss-1.0737, acc-0.7361
Iter-13470, train loss-1.0473, acc-0.7600, valid loss-1.0750, acc-0.7344, test loss-1.0730, acc-0.7361
Iter-13480, train loss-1.1301, acc-0.7200, valid loss-1.0743, acc-0.7346, test loss-1.0724, acc-0.7361
Iter-13490, train loss-1.0073, acc-0.7200, valid loss-1.0737, acc-0.7350, test loss-1.0717, acc-0.7370
Iter-13500, train loss-0.9816, acc-0.7600, valid loss-1.0730, acc-0.7354, test loss-1.0711, acc-0.7368
Iter-13510, train loss-1.0269, acc-0.7800, valid loss-1.0724, acc-0.7354, test loss-1.0705, acc-0.7372
Iter-13520, train loss-1.2252, acc-0.6200, valid loss-1.0718, acc-0.7358, test loss-1.0699, acc-0.7373
Iter-13530, train loss-1.2008, acc-0.6400, valid loss-1.0711, acc-0.7358, test loss-1.0692, acc-0.7379
Iter-13540, train loss-1.1649, acc-0.7800, valid loss-1.0705, acc-0.7362, test loss-1.0686, acc-0.7382
Iter-13550, train loss-1.1859, acc-0.6600, valid loss-1.0698, acc-0.7364, test loss-1.0680, acc-0.7381
Iter-13560, train loss-0.9147, acc-0.8600, valid loss-1.0692, acc-0.7366, test loss-1.0673, acc-0.7383
Iter-13570, train loss-1.1060, acc-0.6400, valid loss-1.0686, acc-0.7372, test loss-1.0668, acc-0.7384
Iter-13580, train loss-1.0273, acc-0.7200, valid loss-1.0680, acc-0.7370, test loss-1.0661, acc-0.7387
Iter-13590, train loss-0.9808, acc-0.7400, valid loss-1.0673, acc-0.7372, test loss-1.0655, acc-0.7385
Iter-13600, train loss-1.0217, acc-0.7800, valid loss-1.0666, acc-0.7374, test loss-1.0648, acc-0.7387
Iter-13610, train loss-1.1522, acc-0.6200, valid loss-1.0660, acc-0.7374, test loss-1.0642, acc-0.7388
Iter-13620, train loss-1.0411, acc-0.8200, valid loss-1.0653, acc-0.7380, test loss-1.0635, acc-0.7393
Iter-13630, train loss-1.0782, acc-0.7000, valid loss-1.0647, acc-0.7380, test loss-1.0629, acc-0.7395
Iter-13640, train loss-1.2333, acc-0.6600, valid loss-1.0641, acc-0.7382, test loss-1.0623, acc-0.7396
Iter-13650, train loss-1.1253, acc-0.6600, valid loss-1.0635, acc-0.7382, test loss-1.0616, acc-0.7398
Iter-13660, train loss-0.9739, acc-0.8400, valid loss-1.0629, acc-0.7382, test loss-1.0611, acc-0.7400
Iter-13670, train loss-1.0881, acc-0.7600, valid loss-1.0623, acc-0.7382, test loss-1.0605, acc-0.7398
Iter-13680, train loss-1.0822, acc-0.6800, valid loss-1.0617, acc-0.7384, test loss-1.0599, acc-0.7400
Iter-13690, train loss-0.8698, acc-0.8200, valid loss-1.0611, acc-0.7382, test loss-1.0593, acc-0.7403
Iter-13700, train loss-0.9514, acc-0.7800, valid loss-1.0605, acc-0.7384, test loss-1.0587, acc-0.7403
Iter-13710, train loss-1.1494, acc-0.6600, valid loss-1.0599, acc-0.7384, test loss-1.0581, acc-0.7402
Iter-13720, train loss-1.2816, acc-0.6400, valid loss-1.0592, acc-0.7384, test loss-1.0574, acc-0.7407
Iter-13730, train loss-0.9954, acc-0.7400, valid loss-1.0586, acc-0.7384, test loss-1.0568, acc-0.7407
Iter-13740, train loss-0.9678, acc-0.7600, valid loss-1.0580, acc-0.7382, test loss-1.0562, acc-0.7409
Iter-13750, train loss-1.1445, acc-0.6600, valid loss-1.0574, acc-0.7384, test loss-1.0556, acc-0.7414
Iter-13760, train loss-1.2173, acc-0.6400, valid loss-1.0568, acc-0.7384, test loss-1.0550, acc-0.7417
Iter-13770, train loss-0.9840, acc-0.7800, valid loss-1.0562, acc-0.7382, test loss-1.0544, acc-0.7415
Iter-13780, train loss-1.1272, acc-0.7400, valid loss-1.0555, acc-0.7384, test loss-1.0538, acc-0.7414
Iter-13790, train loss-1.1550, acc-0.7000, valid loss-1.0549, acc-0.7388, test loss-1.0531, acc-0.7418
Iter-13800, train loss-1.2764, acc-0.5800, valid loss-1.0543, acc-0.7392, test loss-1.0525, acc-0.7419
Iter-13810, train loss-1.0879, acc-0.6600, valid loss-1.0537, acc-0.7390, test loss-1.0519, acc-0.7420
Iter-13820, train loss-0.9556, acc-0.7800, valid loss-1.0530, acc-0.7394, test loss-1.0513, acc-0.7421
Iter-13830, train loss-1.0901, acc-0.7000, valid loss-1.0524, acc-0.7394, test loss-1.0507, acc-0.7426
Iter-13840, train loss-1.1302, acc-0.6400, valid loss-1.0519, acc-0.7398, test loss-1.0501, acc-0.7424
Iter-13850, train loss-1.1397, acc-0.6400, valid loss-1.0513, acc-0.7400, test loss-1.0496, acc-0.7426
Iter-13860, train loss-1.0172, acc-0.7600, valid loss-1.0507, acc-0.7400, test loss-1.0490, acc-0.7427
Iter-13870, train loss-0.9584, acc-0.8600, valid loss-1.0501, acc-0.7404, test loss-1.0484, acc-0.7428
Iter-13880, train loss-1.1233, acc-0.6600, valid loss-1.0495, acc-0.7406, test loss-1.0478, acc-0.7430
Iter-13890, train loss-1.1012, acc-0.7400, valid loss-1.0489, acc-0.7406, test loss-1.0472, acc-0.7432
Iter-13900, train loss-1.0389, acc-0.7200, valid loss-1.0483, acc-0.7410, test loss-1.0466, acc-0.7438
Iter-13910, train loss-1.1363, acc-0.7600, valid loss-1.0477, acc-0.7410, test loss-1.0460, acc-0.7441
Iter-13920, train loss-0.9164, acc-0.8000, valid loss-1.0471, acc-0.7410, test loss-1.0454, acc-0.7443
Iter-13930, train loss-1.2480, acc-0.6400, valid loss-1.0465, acc-0.7412, test loss-1.0448, acc-0.7442
Iter-13940, train loss-1.0450, acc-0.7200, valid loss-1.0459, acc-0.7412, test loss-1.0442, acc-0.7444
Iter-13950, train loss-1.0109, acc-0.7400, valid loss-1.0453, acc-0.7412, test loss-1.0436, acc-0.7446
Iter-13960, train loss-0.9639, acc-0.8400, valid loss-1.0447, acc-0.7408, test loss-1.0430, acc-0.7446
Iter-13970, train loss-1.2313, acc-0.6600, valid loss-1.0441, acc-0.7410, test loss-1.0424, acc-0.7449
Iter-13980, train loss-1.0031, acc-0.8000, valid loss-1.0435, acc-0.7416, test loss-1.0419, acc-0.7451
Iter-13990, train loss-1.0163, acc-0.7600, valid loss-1.0430, acc-0.7420, test loss-1.0413, acc-0.7457
Iter-14000, train loss-1.2029, acc-0.6400, valid loss-1.0424, acc-0.7422, test loss-1.0407, acc-0.7461
Iter-14010, train loss-1.1373, acc-0.7000, valid loss-1.0418, acc-0.7424, test loss-1.0402, acc-0.7463
Iter-14020, train loss-1.1849, acc-0.6400, valid loss-1.0412, acc-0.7422, test loss-1.0396, acc-0.7465
Iter-14030, train loss-1.1525, acc-0.6800, valid loss-1.0407, acc-0.7426, test loss-1.0390, acc-0.7468
Iter-14040, train loss-1.1192, acc-0.7000, valid loss-1.0401, acc-0.7430, test loss-1.0384, acc-0.7473
Iter-14050, train loss-1.2104, acc-0.7000, valid loss-1.0395, acc-0.7428, test loss-1.0379, acc-0.7473
Iter-14060, train loss-0.9963, acc-0.8200, valid loss-1.0390, acc-0.7428, test loss-1.0374, acc-0.7470
Iter-14070, train loss-1.2586, acc-0.6400, valid loss-1.0384, acc-0.7428, test loss-1.0368, acc-0.7469
Iter-14080, train loss-1.0151, acc-0.7400, valid loss-1.0378, acc-0.7428, test loss-1.0362, acc-0.7471
Iter-14090, train loss-1.0072, acc-0.7600, valid loss-1.0371, acc-0.7428, test loss-1.0356, acc-0.7472
Iter-14100, train loss-1.0480, acc-0.7400, valid loss-1.0365, acc-0.7428, test loss-1.0350, acc-0.7473
Iter-14110, train loss-1.1722, acc-0.7000, valid loss-1.0360, acc-0.7426, test loss-1.0344, acc-0.7474
Iter-14120, train loss-1.0678, acc-0.7400, valid loss-1.0354, acc-0.7426, test loss-1.0338, acc-0.7480
Iter-14130, train loss-1.1508, acc-0.6600, valid loss-1.0348, acc-0.7430, test loss-1.0332, acc-0.7479
Iter-14140, train loss-0.9986, acc-0.7600, valid loss-1.0342, acc-0.7430, test loss-1.0327, acc-0.7480
Iter-14150, train loss-1.0486, acc-0.7000, valid loss-1.0336, acc-0.7436, test loss-1.0321, acc-0.7481
Iter-14160, train loss-0.9172, acc-0.8200, valid loss-1.0330, acc-0.7434, test loss-1.0315, acc-0.7481
Iter-14170, train loss-0.9804, acc-0.7800, valid loss-1.0324, acc-0.7434, test loss-1.0310, acc-0.7480
Iter-14180, train loss-0.8366, acc-0.8400, valid loss-1.0318, acc-0.7436, test loss-1.0304, acc-0.7481
Iter-14190, train loss-1.1051, acc-0.7400, valid loss-1.0313, acc-0.7436, test loss-1.0298, acc-0.7481
Iter-14200, train loss-0.9862, acc-0.7600, valid loss-1.0307, acc-0.7438, test loss-1.0292, acc-0.7482
Iter-14210, train loss-0.9578, acc-0.7800, valid loss-1.0301, acc-0.7446, test loss-1.0287, acc-0.7484
Iter-14220, train loss-1.0777, acc-0.8000, valid loss-1.0296, acc-0.7448, test loss-1.0281, acc-0.7484
Iter-14230, train loss-0.9526, acc-0.7400, valid loss-1.0290, acc-0.7450, test loss-1.0275, acc-0.7484
Iter-14240, train loss-0.9860, acc-0.7200, valid loss-1.0284, acc-0.7450, test loss-1.0270, acc-0.7485
Iter-14250, train loss-1.1946, acc-0.6000, valid loss-1.0278, acc-0.7458, test loss-1.0264, acc-0.7488
Iter-14260, train loss-0.9316, acc-0.7800, valid loss-1.0273, acc-0.7460, test loss-1.0259, acc-0.7491
Iter-14270, train loss-0.9516, acc-0.8200, valid loss-1.0267, acc-0.7462, test loss-1.0253, acc-0.7491
Iter-14280, train loss-1.1476, acc-0.6400, valid loss-1.0262, acc-0.7462, test loss-1.0248, acc-0.7491
Iter-14290, train loss-1.0537, acc-0.7200, valid loss-1.0257, acc-0.7466, test loss-1.0242, acc-0.7493
Iter-14300, train loss-0.9671, acc-0.8200, valid loss-1.0251, acc-0.7462, test loss-1.0236, acc-0.7495
Iter-14310, train loss-1.1155, acc-0.6800, valid loss-1.0245, acc-0.7468, test loss-1.0231, acc-0.7497
Iter-14320, train loss-0.9566, acc-0.8000, valid loss-1.0240, acc-0.7466, test loss-1.0225, acc-0.7503
Iter-14330, train loss-0.8897, acc-0.7800, valid loss-1.0233, acc-0.7468, test loss-1.0219, acc-0.7503
Iter-14340, train loss-1.0429, acc-0.7800, valid loss-1.0228, acc-0.7470, test loss-1.0213, acc-0.7503
Iter-14350, train loss-1.0527, acc-0.7400, valid loss-1.0222, acc-0.7474, test loss-1.0208, acc-0.7501
Iter-14360, train loss-1.0776, acc-0.7000, valid loss-1.0216, acc-0.7472, test loss-1.0203, acc-0.7507
Iter-14370, train loss-0.9519, acc-0.8400, valid loss-1.0210, acc-0.7474, test loss-1.0197, acc-0.7506
Iter-14380, train loss-0.8776, acc-0.7600, valid loss-1.0205, acc-0.7472, test loss-1.0191, acc-0.7507
Iter-14390, train loss-1.0682, acc-0.6800, valid loss-1.0199, acc-0.7480, test loss-1.0186, acc-0.7511
Iter-14400, train loss-1.0068, acc-0.7400, valid loss-1.0193, acc-0.7478, test loss-1.0180, acc-0.7513
Iter-14410, train loss-1.0192, acc-0.7400, valid loss-1.0188, acc-0.7478, test loss-1.0174, acc-0.7513
Iter-14420, train loss-0.9886, acc-0.6800, valid loss-1.0182, acc-0.7482, test loss-1.0168, acc-0.7516
Iter-14430, train loss-1.1011, acc-0.7200, valid loss-1.0176, acc-0.7484, test loss-1.0163, acc-0.7514
Iter-14440, train loss-1.0064, acc-0.7600, valid loss-1.0170, acc-0.7488, test loss-1.0157, acc-0.7520
Iter-14450, train loss-0.9698, acc-0.7600, valid loss-1.0164, acc-0.7488, test loss-1.0151, acc-0.7521
Iter-14460, train loss-1.0717, acc-0.7400, valid loss-1.0159, acc-0.7486, test loss-1.0146, acc-0.7522
Iter-14470, train loss-0.9799, acc-0.8000, valid loss-1.0153, acc-0.7492, test loss-1.0140, acc-0.7523
Iter-14480, train loss-1.0078, acc-0.8000, valid loss-1.0148, acc-0.7494, test loss-1.0135, acc-0.7521
Iter-14490, train loss-0.9869, acc-0.7400, valid loss-1.0142, acc-0.7492, test loss-1.0129, acc-0.7522
Iter-14500, train loss-0.8773, acc-0.8400, valid loss-1.0136, acc-0.7492, test loss-1.0124, acc-0.7522
Iter-14510, train loss-0.8792, acc-0.7400, valid loss-1.0131, acc-0.7492, test loss-1.0118, acc-0.7523
Iter-14520, train loss-1.0515, acc-0.7600, valid loss-1.0125, acc-0.7494, test loss-1.0113, acc-0.7527
Iter-14530, train loss-0.9901, acc-0.7400, valid loss-1.0120, acc-0.7496, test loss-1.0108, acc-0.7527
Iter-14540, train loss-1.0953, acc-0.7000, valid loss-1.0115, acc-0.7498, test loss-1.0102, acc-0.7527
Iter-14550, train loss-1.1074, acc-0.6600, valid loss-1.0109, acc-0.7502, test loss-1.0097, acc-0.7528
Iter-14560, train loss-0.7980, acc-0.9000, valid loss-1.0103, acc-0.7502, test loss-1.0091, acc-0.7528
Iter-14570, train loss-0.9687, acc-0.7400, valid loss-1.0097, acc-0.7504, test loss-1.0085, acc-0.7530
Iter-14580, train loss-1.0682, acc-0.7000, valid loss-1.0092, acc-0.7506, test loss-1.0079, acc-0.7533
Iter-14590, train loss-0.9354, acc-0.8200, valid loss-1.0086, acc-0.7510, test loss-1.0074, acc-0.7534
Iter-14600, train loss-1.0232, acc-0.7200, valid loss-1.0081, acc-0.7510, test loss-1.0068, acc-0.7535
Iter-14610, train loss-1.0025, acc-0.7200, valid loss-1.0075, acc-0.7510, test loss-1.0063, acc-0.7539
Iter-14620, train loss-1.0389, acc-0.7400, valid loss-1.0070, acc-0.7512, test loss-1.0058, acc-0.7541
Iter-14630, train loss-1.0682, acc-0.6800, valid loss-1.0065, acc-0.7514, test loss-1.0053, acc-0.7541
Iter-14640, train loss-0.9874, acc-0.7400, valid loss-1.0059, acc-0.7514, test loss-1.0047, acc-0.7541
Iter-14650, train loss-1.0736, acc-0.7000, valid loss-1.0054, acc-0.7514, test loss-1.0042, acc-0.7541
Iter-14660, train loss-0.9404, acc-0.7800, valid loss-1.0048, acc-0.7514, test loss-1.0036, acc-0.7543
Iter-14670, train loss-0.9253, acc-0.7800, valid loss-1.0043, acc-0.7516, test loss-1.0031, acc-0.7545
Iter-14680, train loss-0.8282, acc-0.8400, valid loss-1.0037, acc-0.7514, test loss-1.0025, acc-0.7546
Iter-14690, train loss-1.0512, acc-0.7000, valid loss-1.0032, acc-0.7516, test loss-1.0020, acc-0.7546
Iter-14700, train loss-0.9263, acc-0.8200, valid loss-1.0027, acc-0.7512, test loss-1.0015, acc-0.7548
Iter-14710, train loss-1.0708, acc-0.7200, valid loss-1.0021, acc-0.7518, test loss-1.0009, acc-0.7548
Iter-14720, train loss-0.9941, acc-0.7200, valid loss-1.0016, acc-0.7522, test loss-1.0004, acc-0.7552
Iter-14730, train loss-0.9535, acc-0.7800, valid loss-1.0010, acc-0.7522, test loss-0.9998, acc-0.7553
Iter-14740, train loss-0.9984, acc-0.7600, valid loss-1.0005, acc-0.7522, test loss-0.9993, acc-0.7549
Iter-14750, train loss-0.9148, acc-0.8200, valid loss-0.9999, acc-0.7522, test loss-0.9987, acc-0.7551
Iter-14760, train loss-0.9797, acc-0.7200, valid loss-0.9993, acc-0.7526, test loss-0.9982, acc-0.7556
Iter-14770, train loss-0.8944, acc-0.8600, valid loss-0.9988, acc-0.7526, test loss-0.9976, acc-0.7556
Iter-14780, train loss-0.9467, acc-0.7200, valid loss-0.9982, acc-0.7524, test loss-0.9971, acc-0.7560
Iter-14790, train loss-0.9753, acc-0.7600, valid loss-0.9977, acc-0.7526, test loss-0.9965, acc-0.7559
Iter-14800, train loss-1.0615, acc-0.7400, valid loss-0.9971, acc-0.7526, test loss-0.9960, acc-0.7562
Iter-14810, train loss-1.1178, acc-0.7200, valid loss-0.9966, acc-0.7528, test loss-0.9955, acc-0.7564
Iter-14820, train loss-1.0011, acc-0.7600, valid loss-0.9962, acc-0.7530, test loss-0.9950, acc-0.7568
Iter-14830, train loss-0.8344, acc-0.9200, valid loss-0.9956, acc-0.7530, test loss-0.9944, acc-0.7566
Iter-14840, train loss-1.0836, acc-0.7200, valid loss-0.9951, acc-0.7532, test loss-0.9939, acc-0.7564
Iter-14850, train loss-0.8594, acc-0.8000, valid loss-0.9946, acc-0.7532, test loss-0.9934, acc-0.7568
Iter-14860, train loss-0.9899, acc-0.7800, valid loss-0.9940, acc-0.7534, test loss-0.9929, acc-0.7567
Iter-14870, train loss-1.1655, acc-0.6600, valid loss-0.9935, acc-0.7540, test loss-0.9924, acc-0.7568
Iter-14880, train loss-0.9571, acc-0.7200, valid loss-0.9930, acc-0.7540, test loss-0.9919, acc-0.7570
Iter-14890, train loss-1.0118, acc-0.7200, valid loss-0.9925, acc-0.7538, test loss-0.9913, acc-0.7567
Iter-14900, train loss-0.9820, acc-0.7800, valid loss-0.9919, acc-0.7538, test loss-0.9908, acc-0.7570
Iter-14910, train loss-1.1252, acc-0.5600, valid loss-0.9914, acc-0.7538, test loss-0.9903, acc-0.7570
Iter-14920, train loss-0.9218, acc-0.8200, valid loss-0.9909, acc-0.7536, test loss-0.9898, acc-0.7571
Iter-14930, train loss-1.0591, acc-0.7600, valid loss-0.9903, acc-0.7546, test loss-0.9892, acc-0.7570
Iter-14940, train loss-0.9777, acc-0.8200, valid loss-0.9898, acc-0.7546, test loss-0.9887, acc-0.7573
Iter-14950, train loss-1.0301, acc-0.7600, valid loss-0.9892, acc-0.7548, test loss-0.9882, acc-0.7576
Iter-14960, train loss-0.9560, acc-0.8400, valid loss-0.9887, acc-0.7548, test loss-0.9877, acc-0.7577
Iter-14970, train loss-0.9572, acc-0.8200, valid loss-0.9882, acc-0.7552, test loss-0.9871, acc-0.7579
Iter-14980, train loss-1.1297, acc-0.7400, valid loss-0.9877, acc-0.7548, test loss-0.9866, acc-0.7583
Iter-14990, train loss-1.0222, acc-0.7000, valid loss-0.9871, acc-0.7546, test loss-0.9861, acc-0.7586
Iter-15000, train loss-1.0733, acc-0.6400, valid loss-0.9866, acc-0.7552, test loss-0.9856, acc-0.7587
Iter-15010, train loss-0.9348, acc-0.7600, valid loss-0.9861, acc-0.7556, test loss-0.9850, acc-0.7589
Iter-15020, train loss-0.8641, acc-0.8000, valid loss-0.9856, acc-0.7554, test loss-0.9845, acc-0.7589
Iter-15030, train loss-1.1243, acc-0.7000, valid loss-0.9850, acc-0.7554, test loss-0.9840, acc-0.7589
Iter-15040, train loss-0.8396, acc-0.8000, valid loss-0.9845, acc-0.7554, test loss-0.9834, acc-0.7593
Iter-15050, train loss-0.9947, acc-0.7800, valid loss-0.9840, acc-0.7560, test loss-0.9829, acc-0.7592
Iter-15060, train loss-1.1937, acc-0.7000, valid loss-0.9835, acc-0.7558, test loss-0.9825, acc-0.7599
Iter-15070, train loss-1.0629, acc-0.7000, valid loss-0.9830, acc-0.7558, test loss-0.9820, acc-0.7600
Iter-15080, train loss-0.9125, acc-0.8200, valid loss-0.9825, acc-0.7558, test loss-0.9815, acc-0.7601
Iter-15090, train loss-1.1053, acc-0.7800, valid loss-0.9820, acc-0.7562, test loss-0.9810, acc-0.7601
Iter-15100, train loss-0.9267, acc-0.8000, valid loss-0.9815, acc-0.7562, test loss-0.9804, acc-0.7602
Iter-15110, train loss-1.1237, acc-0.7600, valid loss-0.9809, acc-0.7564, test loss-0.9799, acc-0.7602
Iter-15120, train loss-1.1408, acc-0.7000, valid loss-0.9804, acc-0.7568, test loss-0.9794, acc-0.7605
Iter-15130, train loss-0.9720, acc-0.7800, valid loss-0.9799, acc-0.7568, test loss-0.9788, acc-0.7607
Iter-15140, train loss-0.9699, acc-0.8200, valid loss-0.9794, acc-0.7570, test loss-0.9783, acc-0.7610
Iter-15150, train loss-0.8986, acc-0.8400, valid loss-0.9789, acc-0.7570, test loss-0.9778, acc-0.7610
Iter-15160, train loss-1.0188, acc-0.7600, valid loss-0.9783, acc-0.7570, test loss-0.9773, acc-0.7611
Iter-15170, train loss-1.0698, acc-0.6400, valid loss-0.9778, acc-0.7574, test loss-0.9768, acc-0.7611
Iter-15180, train loss-1.0476, acc-0.6000, valid loss-0.9773, acc-0.7570, test loss-0.9762, acc-0.7614
Iter-15190, train loss-1.1159, acc-0.6800, valid loss-0.9768, acc-0.7572, test loss-0.9757, acc-0.7613
Iter-15200, train loss-1.1037, acc-0.6600, valid loss-0.9763, acc-0.7572, test loss-0.9752, acc-0.7614
Iter-15210, train loss-1.0272, acc-0.7400, valid loss-0.9757, acc-0.7574, test loss-0.9746, acc-0.7614
Iter-15220, train loss-0.9725, acc-0.7200, valid loss-0.9752, acc-0.7574, test loss-0.9741, acc-0.7619
Iter-15230, train loss-0.9352, acc-0.7400, valid loss-0.9747, acc-0.7574, test loss-0.9736, acc-0.7621
Iter-15240, train loss-0.9779, acc-0.7200, valid loss-0.9742, acc-0.7574, test loss-0.9731, acc-0.7622
Iter-15250, train loss-0.9253, acc-0.8000, valid loss-0.9737, acc-0.7574, test loss-0.9726, acc-0.7624
Iter-15260, train loss-0.9364, acc-0.7600, valid loss-0.9732, acc-0.7574, test loss-0.9721, acc-0.7627
Iter-15270, train loss-1.0959, acc-0.7000, valid loss-0.9726, acc-0.7576, test loss-0.9715, acc-0.7630
Iter-15280, train loss-0.9774, acc-0.7400, valid loss-0.9721, acc-0.7576, test loss-0.9710, acc-0.7630
Iter-15290, train loss-1.0035, acc-0.7200, valid loss-0.9716, acc-0.7576, test loss-0.9705, acc-0.7633
Iter-15300, train loss-0.9779, acc-0.7600, valid loss-0.9711, acc-0.7580, test loss-0.9700, acc-0.7634
Iter-15310, train loss-0.9355, acc-0.7600, valid loss-0.9706, acc-0.7580, test loss-0.9695, acc-0.7633
Iter-15320, train loss-1.1627, acc-0.7200, valid loss-0.9701, acc-0.7582, test loss-0.9690, acc-0.7634
Iter-15330, train loss-0.9856, acc-0.7000, valid loss-0.9696, acc-0.7584, test loss-0.9686, acc-0.7634
Iter-15340, train loss-0.8894, acc-0.8000, valid loss-0.9691, acc-0.7586, test loss-0.9681, acc-0.7636
Iter-15350, train loss-0.9678, acc-0.7600, valid loss-0.9685, acc-0.7586, test loss-0.9675, acc-0.7638
Iter-15360, train loss-0.9828, acc-0.7800, valid loss-0.9680, acc-0.7588, test loss-0.9670, acc-0.7638
Iter-15370, train loss-0.8814, acc-0.7600, valid loss-0.9675, acc-0.7590, test loss-0.9665, acc-0.7641
Iter-15380, train loss-1.1720, acc-0.6600, valid loss-0.9670, acc-0.7590, test loss-0.9660, acc-0.7643
Iter-15390, train loss-0.9684, acc-0.7600, valid loss-0.9666, acc-0.7590, test loss-0.9656, acc-0.7646
Iter-15400, train loss-0.9388, acc-0.8200, valid loss-0.9660, acc-0.7590, test loss-0.9651, acc-0.7645
Iter-15410, train loss-0.8169, acc-0.8600, valid loss-0.9655, acc-0.7590, test loss-0.9646, acc-0.7644
Iter-15420, train loss-0.9487, acc-0.7200, valid loss-0.9650, acc-0.7588, test loss-0.9641, acc-0.7646
Iter-15430, train loss-0.8449, acc-0.8200, valid loss-0.9645, acc-0.7590, test loss-0.9636, acc-0.7646
Iter-15440, train loss-0.9318, acc-0.7400, valid loss-0.9640, acc-0.7590, test loss-0.9631, acc-0.7647
Iter-15450, train loss-0.9526, acc-0.8000, valid loss-0.9635, acc-0.7592, test loss-0.9626, acc-0.7650
Iter-15460, train loss-0.9037, acc-0.8200, valid loss-0.9631, acc-0.7592, test loss-0.9621, acc-0.7648
Iter-15470, train loss-1.0130, acc-0.6600, valid loss-0.9626, acc-0.7592, test loss-0.9617, acc-0.7648
Iter-15480, train loss-0.9259, acc-0.8000, valid loss-0.9621, acc-0.7592, test loss-0.9611, acc-0.7648
Iter-15490, train loss-1.0190, acc-0.7400, valid loss-0.9616, acc-0.7596, test loss-0.9606, acc-0.7649
Iter-15500, train loss-0.7922, acc-0.8800, valid loss-0.9610, acc-0.7596, test loss-0.9602, acc-0.7649
Iter-15510, train loss-0.9045, acc-0.7600, valid loss-0.9605, acc-0.7598, test loss-0.9597, acc-0.7650
Iter-15520, train loss-0.9810, acc-0.7600, valid loss-0.9600, acc-0.7596, test loss-0.9591, acc-0.7650
Iter-15530, train loss-0.8811, acc-0.7800, valid loss-0.9595, acc-0.7598, test loss-0.9586, acc-0.7650
Iter-15540, train loss-0.7359, acc-0.9000, valid loss-0.9590, acc-0.7602, test loss-0.9581, acc-0.7653
Iter-15550, train loss-0.8084, acc-0.8000, valid loss-0.9585, acc-0.7604, test loss-0.9576, acc-0.7657
Iter-15560, train loss-0.9958, acc-0.7800, valid loss-0.9580, acc-0.7604, test loss-0.9571, acc-0.7659
Iter-15570, train loss-0.8727, acc-0.8200, valid loss-0.9575, acc-0.7606, test loss-0.9567, acc-0.7659
Iter-15580, train loss-0.9472, acc-0.7400, valid loss-0.9570, acc-0.7608, test loss-0.9562, acc-0.7660
Iter-15590, train loss-0.8666, acc-0.8000, valid loss-0.9566, acc-0.7608, test loss-0.9557, acc-0.7661
Iter-15600, train loss-0.9911, acc-0.7400, valid loss-0.9561, acc-0.7608, test loss-0.9552, acc-0.7661
Iter-15610, train loss-0.9801, acc-0.7800, valid loss-0.9556, acc-0.7612, test loss-0.9547, acc-0.7660
Iter-15620, train loss-0.9685, acc-0.7600, valid loss-0.9551, acc-0.7612, test loss-0.9542, acc-0.7660
Iter-15630, train loss-0.9066, acc-0.7800, valid loss-0.9546, acc-0.7614, test loss-0.9537, acc-0.7661
Iter-15640, train loss-0.9559, acc-0.7400, valid loss-0.9540, acc-0.7614, test loss-0.9532, acc-0.7662
Iter-15650, train loss-0.9939, acc-0.6600, valid loss-0.9536, acc-0.7618, test loss-0.9527, acc-0.7664
Iter-15660, train loss-0.9857, acc-0.7200, valid loss-0.9531, acc-0.7622, test loss-0.9522, acc-0.7665
Iter-15670, train loss-0.8220, acc-0.8400, valid loss-0.9526, acc-0.7622, test loss-0.9517, acc-0.7668
Iter-15680, train loss-1.0011, acc-0.7000, valid loss-0.9521, acc-0.7622, test loss-0.9513, acc-0.7670
Iter-15690, train loss-0.9558, acc-0.8200, valid loss-0.9516, acc-0.7622, test loss-0.9508, acc-0.7669
Iter-15700, train loss-0.9766, acc-0.7600, valid loss-0.9512, acc-0.7624, test loss-0.9503, acc-0.7672
Iter-15710, train loss-0.8810, acc-0.7600, valid loss-0.9507, acc-0.7630, test loss-0.9498, acc-0.7674
Iter-15720, train loss-0.8999, acc-0.7400, valid loss-0.9502, acc-0.7628, test loss-0.9493, acc-0.7675
Iter-15730, train loss-1.1368, acc-0.6600, valid loss-0.9497, acc-0.7628, test loss-0.9489, acc-0.7678
Iter-15740, train loss-1.0319, acc-0.6600, valid loss-0.9492, acc-0.7628, test loss-0.9483, acc-0.7680
Iter-15750, train loss-0.7615, acc-0.8200, valid loss-0.9487, acc-0.7626, test loss-0.9479, acc-0.7681
Iter-15760, train loss-1.0585, acc-0.7400, valid loss-0.9482, acc-0.7626, test loss-0.9474, acc-0.7678
Iter-15770, train loss-1.0058, acc-0.8200, valid loss-0.9477, acc-0.7628, test loss-0.9469, acc-0.7682
Iter-15780, train loss-0.9183, acc-0.7800, valid loss-0.9473, acc-0.7630, test loss-0.9465, acc-0.7679
Iter-15790, train loss-0.9010, acc-0.8400, valid loss-0.9468, acc-0.7632, test loss-0.9460, acc-0.7679
Iter-15800, train loss-0.9470, acc-0.7600, valid loss-0.9463, acc-0.7634, test loss-0.9455, acc-0.7679
Iter-15810, train loss-1.0597, acc-0.7200, valid loss-0.9458, acc-0.7634, test loss-0.9450, acc-0.7682
Iter-15820, train loss-0.9446, acc-0.8200, valid loss-0.9453, acc-0.7632, test loss-0.9445, acc-0.7681
Iter-15830, train loss-1.0049, acc-0.7600, valid loss-0.9448, acc-0.7636, test loss-0.9441, acc-0.7680
Iter-15840, train loss-1.1420, acc-0.6200, valid loss-0.9444, acc-0.7638, test loss-0.9436, acc-0.7681
Iter-15850, train loss-0.7799, acc-0.8600, valid loss-0.9439, acc-0.7640, test loss-0.9431, acc-0.7684
Iter-15860, train loss-1.3467, acc-0.5200, valid loss-0.9434, acc-0.7642, test loss-0.9426, acc-0.7686
Iter-15870, train loss-1.0391, acc-0.7200, valid loss-0.9429, acc-0.7646, test loss-0.9421, acc-0.7688
Iter-15880, train loss-1.0678, acc-0.6400, valid loss-0.9424, acc-0.7646, test loss-0.9417, acc-0.7689
Iter-15890, train loss-0.9858, acc-0.7600, valid loss-0.9419, acc-0.7646, test loss-0.9412, acc-0.7688
Iter-15900, train loss-0.9200, acc-0.7800, valid loss-0.9415, acc-0.7646, test loss-0.9407, acc-0.7690
Iter-15910, train loss-0.8642, acc-0.8400, valid loss-0.9410, acc-0.7646, test loss-0.9402, acc-0.7688
Iter-15920, train loss-0.9981, acc-0.6600, valid loss-0.9405, acc-0.7650, test loss-0.9398, acc-0.7690
Iter-15930, train loss-0.8334, acc-0.7800, valid loss-0.9400, acc-0.7650, test loss-0.9393, acc-0.7690
Iter-15940, train loss-0.8548, acc-0.7800, valid loss-0.9395, acc-0.7650, test loss-0.9388, acc-0.7696
Iter-15950, train loss-0.9368, acc-0.7800, valid loss-0.9391, acc-0.7650, test loss-0.9384, acc-0.7696
Iter-15960, train loss-0.8225, acc-0.8000, valid loss-0.9386, acc-0.7648, test loss-0.9380, acc-0.7696
Iter-15970, train loss-1.0995, acc-0.6800, valid loss-0.9381, acc-0.7652, test loss-0.9375, acc-0.7700
Iter-15980, train loss-1.0315, acc-0.7000, valid loss-0.9377, acc-0.7656, test loss-0.9370, acc-0.7702
Iter-15990, train loss-0.8034, acc-0.8000, valid loss-0.9372, acc-0.7658, test loss-0.9365, acc-0.7703
Iter-16000, train loss-0.8288, acc-0.8200, valid loss-0.9367, acc-0.7658, test loss-0.9360, acc-0.7704
Iter-16010, train loss-0.8323, acc-0.7600, valid loss-0.9362, acc-0.7658, test loss-0.9356, acc-0.7707
Iter-16020, train loss-0.9468, acc-0.7800, valid loss-0.9357, acc-0.7658, test loss-0.9351, acc-0.7708
Iter-16030, train loss-1.0764, acc-0.7400, valid loss-0.9353, acc-0.7658, test loss-0.9347, acc-0.7707
Iter-16040, train loss-0.7770, acc-0.8600, valid loss-0.9347, acc-0.7658, test loss-0.9342, acc-0.7708
Iter-16050, train loss-0.9680, acc-0.7400, valid loss-0.9343, acc-0.7660, test loss-0.9337, acc-0.7710
Iter-16060, train loss-0.9935, acc-0.7000, valid loss-0.9338, acc-0.7662, test loss-0.9332, acc-0.7712
Iter-16070, train loss-0.7809, acc-0.8600, valid loss-0.9333, acc-0.7664, test loss-0.9327, acc-0.7713
Iter-16080, train loss-0.8755, acc-0.7800, valid loss-0.9328, acc-0.7664, test loss-0.9323, acc-0.7712
Iter-16090, train loss-0.9567, acc-0.7400, valid loss-0.9324, acc-0.7668, test loss-0.9318, acc-0.7713
Iter-16100, train loss-0.8572, acc-0.7800, valid loss-0.9319, acc-0.7670, test loss-0.9314, acc-0.7713
Iter-16110, train loss-0.9560, acc-0.7600, valid loss-0.9315, acc-0.7668, test loss-0.9310, acc-0.7715
Iter-16120, train loss-1.0641, acc-0.7600, valid loss-0.9310, acc-0.7666, test loss-0.9305, acc-0.7719
Iter-16130, train loss-1.0481, acc-0.7000, valid loss-0.9306, acc-0.7664, test loss-0.9301, acc-0.7720
Iter-16140, train loss-0.9645, acc-0.7400, valid loss-0.9302, acc-0.7662, test loss-0.9296, acc-0.7720
Iter-16150, train loss-0.9927, acc-0.7200, valid loss-0.9297, acc-0.7668, test loss-0.9292, acc-0.7721
Iter-16160, train loss-0.8578, acc-0.8400, valid loss-0.9292, acc-0.7670, test loss-0.9287, acc-0.7721
Iter-16170, train loss-0.9029, acc-0.8200, valid loss-0.9287, acc-0.7678, test loss-0.9282, acc-0.7723
Iter-16180, train loss-0.9494, acc-0.8400, valid loss-0.9283, acc-0.7678, test loss-0.9278, acc-0.7724
Iter-16190, train loss-0.8236, acc-0.8200, valid loss-0.9278, acc-0.7680, test loss-0.9273, acc-0.7721
Iter-16200, train loss-0.8995, acc-0.7600, valid loss-0.9274, acc-0.7680, test loss-0.9269, acc-0.7724
Iter-16210, train loss-0.8115, acc-0.7600, valid loss-0.9269, acc-0.7686, test loss-0.9264, acc-0.7729
Iter-16220, train loss-0.9234, acc-0.7400, valid loss-0.9265, acc-0.7686, test loss-0.9260, acc-0.7729
Iter-16230, train loss-0.9301, acc-0.8400, valid loss-0.9260, acc-0.7686, test loss-0.9255, acc-0.7729
Iter-16240, train loss-0.7129, acc-0.8800, valid loss-0.9255, acc-0.7688, test loss-0.9251, acc-0.7729
Iter-16250, train loss-0.9585, acc-0.7200, valid loss-0.9250, acc-0.7688, test loss-0.9246, acc-0.7732
Iter-16260, train loss-0.9664, acc-0.7600, valid loss-0.9246, acc-0.7692, test loss-0.9241, acc-0.7731
Iter-16270, train loss-0.9273, acc-0.8400, valid loss-0.9241, acc-0.7696, test loss-0.9237, acc-0.7728
Iter-16280, train loss-0.8193, acc-0.8200, valid loss-0.9236, acc-0.7696, test loss-0.9232, acc-0.7729
Iter-16290, train loss-0.9058, acc-0.8200, valid loss-0.9232, acc-0.7698, test loss-0.9228, acc-0.7728
Iter-16300, train loss-0.9527, acc-0.7800, valid loss-0.9228, acc-0.7702, test loss-0.9223, acc-0.7729
Iter-16310, train loss-0.8172, acc-0.8000, valid loss-0.9223, acc-0.7702, test loss-0.9219, acc-0.7729
Iter-16320, train loss-0.8681, acc-0.8000, valid loss-0.9219, acc-0.7702, test loss-0.9215, acc-0.7731
Iter-16330, train loss-0.7395, acc-0.9400, valid loss-0.9214, acc-0.7706, test loss-0.9210, acc-0.7734
Iter-16340, train loss-0.8972, acc-0.8400, valid loss-0.9210, acc-0.7708, test loss-0.9205, acc-0.7734
Iter-16350, train loss-1.1283, acc-0.6200, valid loss-0.9205, acc-0.7710, test loss-0.9201, acc-0.7734
Iter-16360, train loss-0.8981, acc-0.8400, valid loss-0.9201, acc-0.7712, test loss-0.9197, acc-0.7734
Iter-16370, train loss-0.7967, acc-0.7800, valid loss-0.9196, acc-0.7712, test loss-0.9192, acc-0.7737
Iter-16380, train loss-0.9029, acc-0.8000, valid loss-0.9192, acc-0.7714, test loss-0.9188, acc-0.7737
Iter-16390, train loss-0.8105, acc-0.7800, valid loss-0.9187, acc-0.7716, test loss-0.9183, acc-0.7736
Iter-16400, train loss-0.8502, acc-0.8600, valid loss-0.9183, acc-0.7716, test loss-0.9179, acc-0.7733
Iter-16410, train loss-0.8983, acc-0.7400, valid loss-0.9178, acc-0.7716, test loss-0.9175, acc-0.7735
Iter-16420, train loss-1.0967, acc-0.6400, valid loss-0.9174, acc-0.7718, test loss-0.9170, acc-0.7737
Iter-16430, train loss-0.9802, acc-0.7400, valid loss-0.9170, acc-0.7716, test loss-0.9166, acc-0.7738
Iter-16440, train loss-0.9448, acc-0.7400, valid loss-0.9166, acc-0.7718, test loss-0.9162, acc-0.7739
Iter-16450, train loss-1.1142, acc-0.6400, valid loss-0.9161, acc-0.7722, test loss-0.9157, acc-0.7737
Iter-16460, train loss-0.9827, acc-0.7000, valid loss-0.9157, acc-0.7726, test loss-0.9153, acc-0.7741
Iter-16470, train loss-0.8795, acc-0.8000, valid loss-0.9152, acc-0.7724, test loss-0.9148, acc-0.7742
Iter-16480, train loss-0.9370, acc-0.7600, valid loss-0.9148, acc-0.7726, test loss-0.9144, acc-0.7741
Iter-16490, train loss-0.9079, acc-0.8200, valid loss-0.9143, acc-0.7728, test loss-0.9140, acc-0.7742
Iter-16500, train loss-0.9844, acc-0.8400, valid loss-0.9139, acc-0.7728, test loss-0.9136, acc-0.7746
Iter-16510, train loss-1.0271, acc-0.7400, valid loss-0.9135, acc-0.7726, test loss-0.9131, acc-0.7746
Iter-16520, train loss-0.8463, acc-0.7800, valid loss-0.9130, acc-0.7732, test loss-0.9127, acc-0.7747
Iter-16530, train loss-0.8967, acc-0.7800, valid loss-0.9126, acc-0.7732, test loss-0.9122, acc-0.7747
Iter-16540, train loss-0.9827, acc-0.8400, valid loss-0.9122, acc-0.7732, test loss-0.9118, acc-0.7748
Iter-16550, train loss-0.9462, acc-0.7400, valid loss-0.9117, acc-0.7732, test loss-0.9114, acc-0.7748
Iter-16560, train loss-0.9447, acc-0.7200, valid loss-0.9113, acc-0.7732, test loss-0.9110, acc-0.7749
Iter-16570, train loss-0.8759, acc-0.8400, valid loss-0.9108, acc-0.7732, test loss-0.9105, acc-0.7750
Iter-16580, train loss-0.8393, acc-0.8800, valid loss-0.9104, acc-0.7736, test loss-0.9101, acc-0.7751
Iter-16590, train loss-0.8774, acc-0.8000, valid loss-0.9100, acc-0.7732, test loss-0.9097, acc-0.7751
Iter-16600, train loss-0.9730, acc-0.7400, valid loss-0.9096, acc-0.7736, test loss-0.9092, acc-0.7752
Iter-16610, train loss-0.8648, acc-0.8200, valid loss-0.9091, acc-0.7736, test loss-0.9088, acc-0.7752
Iter-16620, train loss-0.9751, acc-0.7800, valid loss-0.9087, acc-0.7736, test loss-0.9084, acc-0.7754
Iter-16630, train loss-0.9340, acc-0.7600, valid loss-0.9083, acc-0.7738, test loss-0.9080, acc-0.7756
Iter-16640, train loss-0.7631, acc-0.8400, valid loss-0.9078, acc-0.7740, test loss-0.9076, acc-0.7756
Iter-16650, train loss-0.7783, acc-0.9000, valid loss-0.9074, acc-0.7742, test loss-0.9071, acc-0.7758
Iter-16660, train loss-0.8950, acc-0.8000, valid loss-0.9070, acc-0.7748, test loss-0.9067, acc-0.7760
Iter-16670, train loss-1.0127, acc-0.7400, valid loss-0.9065, acc-0.7746, test loss-0.9062, acc-0.7762
Iter-16680, train loss-0.8591, acc-0.8200, valid loss-0.9061, acc-0.7750, test loss-0.9058, acc-0.7762
Iter-16690, train loss-0.7983, acc-0.8600, valid loss-0.9056, acc-0.7746, test loss-0.9054, acc-0.7762
Iter-16700, train loss-0.9376, acc-0.7400, valid loss-0.9052, acc-0.7748, test loss-0.9050, acc-0.7763
Iter-16710, train loss-0.9618, acc-0.7600, valid loss-0.9048, acc-0.7750, test loss-0.9046, acc-0.7762
Iter-16720, train loss-1.0243, acc-0.6600, valid loss-0.9044, acc-0.7756, test loss-0.9041, acc-0.7769
Iter-16730, train loss-0.9464, acc-0.7000, valid loss-0.9039, acc-0.7750, test loss-0.9037, acc-0.7766
Iter-16740, train loss-0.9897, acc-0.7200, valid loss-0.9035, acc-0.7752, test loss-0.9033, acc-0.7767
Iter-16750, train loss-0.8170, acc-0.7600, valid loss-0.9031, acc-0.7754, test loss-0.9029, acc-0.7769
Iter-16760, train loss-0.8109, acc-0.8600, valid loss-0.9027, acc-0.7754, test loss-0.9024, acc-0.7769
Iter-16770, train loss-1.1216, acc-0.6400, valid loss-0.9023, acc-0.7756, test loss-0.9020, acc-0.7770
Iter-16780, train loss-0.9834, acc-0.7000, valid loss-0.9019, acc-0.7758, test loss-0.9016, acc-0.7775
Iter-16790, train loss-0.9683, acc-0.7600, valid loss-0.9014, acc-0.7762, test loss-0.9012, acc-0.7774
Iter-16800, train loss-0.9111, acc-0.7800, valid loss-0.9010, acc-0.7762, test loss-0.9008, acc-0.7777
Iter-16810, train loss-0.9992, acc-0.7600, valid loss-0.9006, acc-0.7764, test loss-0.9004, acc-0.7779
Iter-16820, train loss-0.9341, acc-0.7000, valid loss-0.9001, acc-0.7762, test loss-0.8999, acc-0.7779
Iter-16830, train loss-1.0476, acc-0.6600, valid loss-0.8997, acc-0.7764, test loss-0.8995, acc-0.7778
Iter-16840, train loss-0.8809, acc-0.7800, valid loss-0.8993, acc-0.7770, test loss-0.8991, acc-0.7779
Iter-16850, train loss-0.8586, acc-0.8000, valid loss-0.8989, acc-0.7770, test loss-0.8987, acc-0.7781
Iter-16860, train loss-0.7747, acc-0.8200, valid loss-0.8985, acc-0.7768, test loss-0.8982, acc-0.7783
Iter-16870, train loss-0.8844, acc-0.8200, valid loss-0.8981, acc-0.7770, test loss-0.8978, acc-0.7786
Iter-16880, train loss-1.0549, acc-0.7000, valid loss-0.8976, acc-0.7772, test loss-0.8974, acc-0.7785
Iter-16890, train loss-0.9893, acc-0.7400, valid loss-0.8972, acc-0.7774, test loss-0.8970, acc-0.7787
Iter-16900, train loss-0.6956, acc-0.9000, valid loss-0.8968, acc-0.7776, test loss-0.8965, acc-0.7788
Iter-16910, train loss-0.9618, acc-0.7600, valid loss-0.8963, acc-0.7784, test loss-0.8961, acc-0.7791
Iter-16920, train loss-1.0351, acc-0.7200, valid loss-0.8959, acc-0.7786, test loss-0.8957, acc-0.7790
Iter-16930, train loss-0.9794, acc-0.7800, valid loss-0.8955, acc-0.7780, test loss-0.8953, acc-0.7791
Iter-16940, train loss-0.8753, acc-0.7600, valid loss-0.8951, acc-0.7786, test loss-0.8949, acc-0.7793
Iter-16950, train loss-0.9289, acc-0.7200, valid loss-0.8947, acc-0.7786, test loss-0.8945, acc-0.7794
Iter-16960, train loss-0.8331, acc-0.8000, valid loss-0.8943, acc-0.7786, test loss-0.8941, acc-0.7793
Iter-16970, train loss-0.8231, acc-0.8000, valid loss-0.8938, acc-0.7788, test loss-0.8936, acc-0.7794
Iter-16980, train loss-0.7578, acc-0.8400, valid loss-0.8934, acc-0.7790, test loss-0.8933, acc-0.7796
Iter-16990, train loss-0.9986, acc-0.7600, valid loss-0.8930, acc-0.7788, test loss-0.8928, acc-0.7795
Iter-17000, train loss-0.8537, acc-0.7400, valid loss-0.8926, acc-0.7790, test loss-0.8924, acc-0.7795
Iter-17010, train loss-0.8792, acc-0.8000, valid loss-0.8922, acc-0.7790, test loss-0.8920, acc-0.7797
Iter-17020, train loss-0.8853, acc-0.6800, valid loss-0.8918, acc-0.7790, test loss-0.8916, acc-0.7795
Iter-17030, train loss-0.8645, acc-0.8600, valid loss-0.8914, acc-0.7784, test loss-0.8912, acc-0.7794
Iter-17040, train loss-1.1061, acc-0.6800, valid loss-0.8909, acc-0.7784, test loss-0.8908, acc-0.7798
Iter-17050, train loss-1.0959, acc-0.7000, valid loss-0.8905, acc-0.7790, test loss-0.8903, acc-0.7800
Iter-17060, train loss-0.9046, acc-0.7800, valid loss-0.8901, acc-0.7790, test loss-0.8899, acc-0.7803
Iter-17070, train loss-0.8738, acc-0.7800, valid loss-0.8897, acc-0.7788, test loss-0.8895, acc-0.7805
Iter-17080, train loss-0.8461, acc-0.8200, valid loss-0.8892, acc-0.7790, test loss-0.8891, acc-0.7805
Iter-17090, train loss-0.8234, acc-0.7800, valid loss-0.8888, acc-0.7790, test loss-0.8887, acc-0.7807
Iter-17100, train loss-1.0769, acc-0.7400, valid loss-0.8884, acc-0.7790, test loss-0.8883, acc-0.7805
Iter-17110, train loss-0.9296, acc-0.7800, valid loss-0.8880, acc-0.7792, test loss-0.8879, acc-0.7807
Iter-17120, train loss-0.7715, acc-0.8600, valid loss-0.8876, acc-0.7792, test loss-0.8874, acc-0.7809
Iter-17130, train loss-0.8128, acc-0.8000, valid loss-0.8871, acc-0.7790, test loss-0.8870, acc-0.7811
Iter-17140, train loss-0.8781, acc-0.7800, valid loss-0.8867, acc-0.7790, test loss-0.8866, acc-0.7810
Iter-17150, train loss-0.8885, acc-0.7600, valid loss-0.8863, acc-0.7790, test loss-0.8862, acc-0.7808
Iter-17160, train loss-0.8312, acc-0.8400, valid loss-0.8859, acc-0.7792, test loss-0.8858, acc-0.7810
Iter-17170, train loss-1.1600, acc-0.6200, valid loss-0.8855, acc-0.7794, test loss-0.8854, acc-0.7812
Iter-17180, train loss-0.9758, acc-0.7000, valid loss-0.8851, acc-0.7796, test loss-0.8851, acc-0.7811
Iter-17190, train loss-0.7936, acc-0.8000, valid loss-0.8847, acc-0.7798, test loss-0.8847, acc-0.7812
Iter-17200, train loss-0.9328, acc-0.7600, valid loss-0.8843, acc-0.7798, test loss-0.8843, acc-0.7812
Iter-17210, train loss-0.9496, acc-0.7200, valid loss-0.8839, acc-0.7800, test loss-0.8839, acc-0.7813
Iter-17220, train loss-0.9330, acc-0.7600, valid loss-0.8835, acc-0.7800, test loss-0.8835, acc-0.7816
Iter-17230, train loss-1.0254, acc-0.7200, valid loss-0.8831, acc-0.7798, test loss-0.8831, acc-0.7817
Iter-17240, train loss-0.8772, acc-0.8400, valid loss-0.8827, acc-0.7798, test loss-0.8827, acc-0.7818
Iter-17250, train loss-0.8733, acc-0.7400, valid loss-0.8823, acc-0.7800, test loss-0.8823, acc-0.7819
Iter-17260, train loss-0.7513, acc-0.8600, valid loss-0.8819, acc-0.7802, test loss-0.8819, acc-0.7821
Iter-17270, train loss-0.8283, acc-0.7600, valid loss-0.8814, acc-0.7802, test loss-0.8814, acc-0.7819
Iter-17280, train loss-0.6983, acc-0.8600, valid loss-0.8811, acc-0.7806, test loss-0.8811, acc-0.7820
Iter-17290, train loss-0.9522, acc-0.6800, valid loss-0.8807, acc-0.7802, test loss-0.8807, acc-0.7820
Iter-17300, train loss-0.8374, acc-0.8200, valid loss-0.8803, acc-0.7804, test loss-0.8803, acc-0.7820
Iter-17310, train loss-0.7921, acc-0.7600, valid loss-0.8799, acc-0.7806, test loss-0.8799, acc-0.7821
Iter-17320, train loss-0.8635, acc-0.8200, valid loss-0.8795, acc-0.7808, test loss-0.8795, acc-0.7822
Iter-17330, train loss-0.7557, acc-0.8400, valid loss-0.8790, acc-0.7810, test loss-0.8791, acc-0.7823
Iter-17340, train loss-1.0773, acc-0.6400, valid loss-0.8786, acc-0.7820, test loss-0.8786, acc-0.7824
Iter-17350, train loss-0.8410, acc-0.8800, valid loss-0.8782, acc-0.7820, test loss-0.8782, acc-0.7824
Iter-17360, train loss-1.0560, acc-0.7600, valid loss-0.8778, acc-0.7822, test loss-0.8779, acc-0.7825
Iter-17370, train loss-0.8209, acc-0.7600, valid loss-0.8774, acc-0.7822, test loss-0.8774, acc-0.7827
Iter-17380, train loss-0.9793, acc-0.7200, valid loss-0.8770, acc-0.7828, test loss-0.8770, acc-0.7828
Iter-17390, train loss-0.8889, acc-0.7800, valid loss-0.8766, acc-0.7830, test loss-0.8766, acc-0.7828
Iter-17400, train loss-0.7016, acc-0.9200, valid loss-0.8762, acc-0.7830, test loss-0.8762, acc-0.7831
Iter-17410, train loss-0.9248, acc-0.6800, valid loss-0.8757, acc-0.7828, test loss-0.8758, acc-0.7832
Iter-17420, train loss-0.7210, acc-0.9200, valid loss-0.8753, acc-0.7826, test loss-0.8754, acc-0.7830
Iter-17430, train loss-0.8517, acc-0.7200, valid loss-0.8750, acc-0.7828, test loss-0.8750, acc-0.7833
Iter-17440, train loss-0.9429, acc-0.7800, valid loss-0.8746, acc-0.7830, test loss-0.8747, acc-0.7833
Iter-17450, train loss-0.9987, acc-0.7200, valid loss-0.8742, acc-0.7830, test loss-0.8743, acc-0.7836
Iter-17460, train loss-0.8020, acc-0.7600, valid loss-0.8738, acc-0.7832, test loss-0.8739, acc-0.7837
Iter-17470, train loss-0.6950, acc-0.8800, valid loss-0.8735, acc-0.7832, test loss-0.8736, acc-0.7840
Iter-17480, train loss-0.9055, acc-0.7600, valid loss-0.8731, acc-0.7834, test loss-0.8732, acc-0.7841
Iter-17490, train loss-0.7589, acc-0.8600, valid loss-0.8727, acc-0.7836, test loss-0.8728, acc-0.7841
Iter-17500, train loss-0.9808, acc-0.6400, valid loss-0.8724, acc-0.7840, test loss-0.8725, acc-0.7841
Iter-17510, train loss-0.8206, acc-0.7800, valid loss-0.8720, acc-0.7840, test loss-0.8720, acc-0.7842
Iter-17520, train loss-0.9094, acc-0.8200, valid loss-0.8716, acc-0.7840, test loss-0.8716, acc-0.7840
Iter-17530, train loss-1.1326, acc-0.7200, valid loss-0.8712, acc-0.7842, test loss-0.8713, acc-0.7843
Iter-17540, train loss-0.8820, acc-0.7400, valid loss-0.8708, acc-0.7842, test loss-0.8709, acc-0.7845
Iter-17550, train loss-0.8714, acc-0.8200, valid loss-0.8704, acc-0.7844, test loss-0.8705, acc-0.7847
Iter-17560, train loss-0.7432, acc-0.8200, valid loss-0.8700, acc-0.7844, test loss-0.8701, acc-0.7847
Iter-17570, train loss-0.8277, acc-0.8000, valid loss-0.8697, acc-0.7844, test loss-0.8698, acc-0.7847
Iter-17580, train loss-0.8346, acc-0.8200, valid loss-0.8693, acc-0.7842, test loss-0.8694, acc-0.7847
Iter-17590, train loss-0.8782, acc-0.7800, valid loss-0.8689, acc-0.7844, test loss-0.8690, acc-0.7847
Iter-17600, train loss-0.7497, acc-0.8400, valid loss-0.8685, acc-0.7844, test loss-0.8686, acc-0.7849
Iter-17610, train loss-1.1230, acc-0.6800, valid loss-0.8681, acc-0.7838, test loss-0.8683, acc-0.7850
Iter-17620, train loss-0.7948, acc-0.8000, valid loss-0.8678, acc-0.7836, test loss-0.8679, acc-0.7850
Iter-17630, train loss-0.9921, acc-0.6800, valid loss-0.8674, acc-0.7838, test loss-0.8675, acc-0.7850
Iter-17640, train loss-0.7662, acc-0.8800, valid loss-0.8670, acc-0.7838, test loss-0.8672, acc-0.7851
Iter-17650, train loss-0.8660, acc-0.8000, valid loss-0.8666, acc-0.7836, test loss-0.8668, acc-0.7852
Iter-17660, train loss-0.8613, acc-0.8000, valid loss-0.8662, acc-0.7838, test loss-0.8664, acc-0.7855
Iter-17670, train loss-0.9608, acc-0.7600, valid loss-0.8658, acc-0.7836, test loss-0.8660, acc-0.7858
Iter-17680, train loss-0.8995, acc-0.7200, valid loss-0.8654, acc-0.7840, test loss-0.8656, acc-0.7859
Iter-17690, train loss-0.8713, acc-0.7600, valid loss-0.8651, acc-0.7842, test loss-0.8652, acc-0.7857
Iter-17700, train loss-1.0750, acc-0.6400, valid loss-0.8646, acc-0.7844, test loss-0.8648, acc-0.7858
Iter-17710, train loss-0.8471, acc-0.7400, valid loss-0.8642, acc-0.7844, test loss-0.8644, acc-0.7859
Iter-17720, train loss-0.8650, acc-0.7800, valid loss-0.8638, acc-0.7848, test loss-0.8640, acc-0.7860
Iter-17730, train loss-0.8414, acc-0.8200, valid loss-0.8635, acc-0.7850, test loss-0.8637, acc-0.7861
Iter-17740, train loss-0.7998, acc-0.8400, valid loss-0.8631, acc-0.7854, test loss-0.8633, acc-0.7863
Iter-17750, train loss-0.8680, acc-0.8000, valid loss-0.8627, acc-0.7854, test loss-0.8629, acc-0.7866
Iter-17760, train loss-0.8949, acc-0.7000, valid loss-0.8623, acc-0.7860, test loss-0.8625, acc-0.7871
Iter-17770, train loss-0.8127, acc-0.8400, valid loss-0.8619, acc-0.7862, test loss-0.8621, acc-0.7870
Iter-17780, train loss-0.8748, acc-0.8200, valid loss-0.8615, acc-0.7860, test loss-0.8617, acc-0.7872
Iter-17790, train loss-0.8378, acc-0.7600, valid loss-0.8611, acc-0.7860, test loss-0.8613, acc-0.7871
Iter-17800, train loss-0.8654, acc-0.7600, valid loss-0.8607, acc-0.7860, test loss-0.8610, acc-0.7875
Iter-17810, train loss-0.9437, acc-0.7400, valid loss-0.8604, acc-0.7862, test loss-0.8606, acc-0.7876
Iter-17820, train loss-0.9139, acc-0.8000, valid loss-0.8600, acc-0.7864, test loss-0.8603, acc-0.7874
Iter-17830, train loss-0.8975, acc-0.8000, valid loss-0.8596, acc-0.7864, test loss-0.8599, acc-0.7871
Iter-17840, train loss-0.7884, acc-0.8200, valid loss-0.8592, acc-0.7866, test loss-0.8595, acc-0.7872
Iter-17850, train loss-1.0042, acc-0.7600, valid loss-0.8589, acc-0.7868, test loss-0.8591, acc-0.7872
Iter-17860, train loss-0.8826, acc-0.7800, valid loss-0.8585, acc-0.7870, test loss-0.8587, acc-0.7873
Iter-17870, train loss-0.7820, acc-0.8000, valid loss-0.8581, acc-0.7874, test loss-0.8584, acc-0.7875
Iter-17880, train loss-1.0125, acc-0.7200, valid loss-0.8577, acc-0.7872, test loss-0.8580, acc-0.7875
Iter-17890, train loss-0.9643, acc-0.6600, valid loss-0.8573, acc-0.7878, test loss-0.8576, acc-0.7879
Iter-17900, train loss-0.9388, acc-0.8000, valid loss-0.8570, acc-0.7876, test loss-0.8572, acc-0.7878
Iter-17910, train loss-1.1025, acc-0.6600, valid loss-0.8566, acc-0.7880, test loss-0.8569, acc-0.7879
Iter-17920, train loss-0.7957, acc-0.8200, valid loss-0.8562, acc-0.7884, test loss-0.8565, acc-0.7883
Iter-17930, train loss-0.8991, acc-0.8000, valid loss-0.8558, acc-0.7884, test loss-0.8561, acc-0.7884
Iter-17940, train loss-0.8945, acc-0.7600, valid loss-0.8554, acc-0.7882, test loss-0.8557, acc-0.7885
Iter-17950, train loss-0.9669, acc-0.7000, valid loss-0.8550, acc-0.7882, test loss-0.8553, acc-0.7885
Iter-17960, train loss-0.7808, acc-0.8600, valid loss-0.8546, acc-0.7884, test loss-0.8549, acc-0.7888
Iter-17970, train loss-0.7778, acc-0.8000, valid loss-0.8542, acc-0.7884, test loss-0.8546, acc-0.7890
Iter-17980, train loss-0.8292, acc-0.8000, valid loss-0.8539, acc-0.7884, test loss-0.8542, acc-0.7893
Iter-17990, train loss-0.8581, acc-0.7400, valid loss-0.8535, acc-0.7884, test loss-0.8538, acc-0.7893
Iter-18000, train loss-0.7624, acc-0.8000, valid loss-0.8531, acc-0.7886, test loss-0.8535, acc-0.7892
Iter-18010, train loss-0.8067, acc-0.7600, valid loss-0.8527, acc-0.7886, test loss-0.8531, acc-0.7894
Iter-18020, train loss-0.9013, acc-0.7600, valid loss-0.8523, acc-0.7882, test loss-0.8527, acc-0.7894
Iter-18030, train loss-0.7932, acc-0.8200, valid loss-0.8520, acc-0.7884, test loss-0.8524, acc-0.7894
Iter-18040, train loss-0.8948, acc-0.7400, valid loss-0.8516, acc-0.7884, test loss-0.8520, acc-0.7897
Iter-18050, train loss-0.7847, acc-0.7800, valid loss-0.8512, acc-0.7886, test loss-0.8517, acc-0.7896
Iter-18060, train loss-0.7857, acc-0.8400, valid loss-0.8509, acc-0.7884, test loss-0.8513, acc-0.7896
Iter-18070, train loss-0.7883, acc-0.7400, valid loss-0.8505, acc-0.7886, test loss-0.8509, acc-0.7897
Iter-18080, train loss-0.7692, acc-0.8600, valid loss-0.8502, acc-0.7886, test loss-0.8506, acc-0.7899
Iter-18090, train loss-0.9264, acc-0.7800, valid loss-0.8498, acc-0.7884, test loss-0.8502, acc-0.7900
Iter-18100, train loss-0.9237, acc-0.8000, valid loss-0.8495, acc-0.7884, test loss-0.8499, acc-0.7900
Iter-18110, train loss-0.6433, acc-0.8800, valid loss-0.8491, acc-0.7888, test loss-0.8495, acc-0.7904
Iter-18120, train loss-0.8535, acc-0.7400, valid loss-0.8487, acc-0.7890, test loss-0.8492, acc-0.7906
Iter-18130, train loss-0.7633, acc-0.8000, valid loss-0.8484, acc-0.7888, test loss-0.8488, acc-0.7906
Iter-18140, train loss-0.8460, acc-0.7600, valid loss-0.8480, acc-0.7890, test loss-0.8485, acc-0.7907
Iter-18150, train loss-0.8186, acc-0.7800, valid loss-0.8477, acc-0.7886, test loss-0.8481, acc-0.7906
Iter-18160, train loss-0.8279, acc-0.7800, valid loss-0.8473, acc-0.7886, test loss-0.8478, acc-0.7908
Iter-18170, train loss-1.0031, acc-0.7400, valid loss-0.8470, acc-0.7888, test loss-0.8474, acc-0.7908
Iter-18180, train loss-0.8824, acc-0.7400, valid loss-0.8466, acc-0.7888, test loss-0.8470, acc-0.7910
Iter-18190, train loss-0.8089, acc-0.7800, valid loss-0.8462, acc-0.7890, test loss-0.8467, acc-0.7911
Iter-18200, train loss-1.0231, acc-0.7200, valid loss-0.8459, acc-0.7888, test loss-0.8463, acc-0.7909
Iter-18210, train loss-0.8078, acc-0.8000, valid loss-0.8455, acc-0.7890, test loss-0.8460, acc-0.7907
Iter-18220, train loss-0.8428, acc-0.7600, valid loss-0.8451, acc-0.7892, test loss-0.8456, acc-0.7909
Iter-18230, train loss-0.7963, acc-0.7600, valid loss-0.8448, acc-0.7890, test loss-0.8453, acc-0.7909
Iter-18240, train loss-0.8769, acc-0.7600, valid loss-0.8444, acc-0.7890, test loss-0.8449, acc-0.7908
Iter-18250, train loss-0.9474, acc-0.7400, valid loss-0.8441, acc-0.7890, test loss-0.8446, acc-0.7909
Iter-18260, train loss-0.8005, acc-0.8200, valid loss-0.8437, acc-0.7892, test loss-0.8442, acc-0.7909
Iter-18270, train loss-0.7254, acc-0.8800, valid loss-0.8434, acc-0.7890, test loss-0.8439, acc-0.7910
Iter-18280, train loss-0.9121, acc-0.7400, valid loss-0.8430, acc-0.7890, test loss-0.8435, acc-0.7909
Iter-18290, train loss-0.7591, acc-0.8000, valid loss-0.8427, acc-0.7892, test loss-0.8432, acc-0.7913
Iter-18300, train loss-0.8930, acc-0.8200, valid loss-0.8423, acc-0.7896, test loss-0.8428, acc-0.7917
Iter-18310, train loss-0.8759, acc-0.7400, valid loss-0.8420, acc-0.7894, test loss-0.8425, acc-0.7918
Iter-18320, train loss-0.8120, acc-0.8000, valid loss-0.8416, acc-0.7894, test loss-0.8422, acc-0.7920
Iter-18330, train loss-0.7473, acc-0.8400, valid loss-0.8413, acc-0.7894, test loss-0.8418, acc-0.7920
Iter-18340, train loss-0.8760, acc-0.7600, valid loss-0.8410, acc-0.7896, test loss-0.8415, acc-0.7919
Iter-18350, train loss-0.8274, acc-0.7400, valid loss-0.8406, acc-0.7896, test loss-0.8411, acc-0.7917
Iter-18360, train loss-0.8997, acc-0.7600, valid loss-0.8402, acc-0.7894, test loss-0.8407, acc-0.7923
Iter-18370, train loss-0.6467, acc-0.9200, valid loss-0.8399, acc-0.7894, test loss-0.8404, acc-0.7921
Iter-18380, train loss-0.7951, acc-0.8600, valid loss-0.8395, acc-0.7894, test loss-0.8400, acc-0.7920
Iter-18390, train loss-0.9336, acc-0.7600, valid loss-0.8392, acc-0.7898, test loss-0.8397, acc-0.7921
Iter-18400, train loss-0.7890, acc-0.7800, valid loss-0.8388, acc-0.7900, test loss-0.8394, acc-0.7923
Iter-18410, train loss-0.9120, acc-0.7800, valid loss-0.8385, acc-0.7900, test loss-0.8390, acc-0.7925
Iter-18420, train loss-0.9666, acc-0.8000, valid loss-0.8381, acc-0.7902, test loss-0.8387, acc-0.7927
Iter-18430, train loss-0.7304, acc-0.8200, valid loss-0.8378, acc-0.7904, test loss-0.8383, acc-0.7928
Iter-18440, train loss-0.9566, acc-0.7000, valid loss-0.8374, acc-0.7904, test loss-0.8380, acc-0.7927
Iter-18450, train loss-0.8164, acc-0.8400, valid loss-0.8371, acc-0.7906, test loss-0.8376, acc-0.7928
Iter-18460, train loss-0.8943, acc-0.7800, valid loss-0.8367, acc-0.7904, test loss-0.8373, acc-0.7928
Iter-18470, train loss-1.0707, acc-0.7000, valid loss-0.8364, acc-0.7906, test loss-0.8370, acc-0.7931
Iter-18480, train loss-0.7930, acc-0.7800, valid loss-0.8361, acc-0.7906, test loss-0.8366, acc-0.7929
Iter-18490, train loss-0.9135, acc-0.8200, valid loss-0.8357, acc-0.7902, test loss-0.8362, acc-0.7928
Iter-18500, train loss-0.9110, acc-0.8000, valid loss-0.8353, acc-0.7904, test loss-0.8359, acc-0.7928
Iter-18510, train loss-0.8640, acc-0.7600, valid loss-0.8350, acc-0.7910, test loss-0.8356, acc-0.7929
Iter-18520, train loss-0.8310, acc-0.7800, valid loss-0.8347, acc-0.7910, test loss-0.8352, acc-0.7932
Iter-18530, train loss-1.1013, acc-0.6600, valid loss-0.8343, acc-0.7912, test loss-0.8349, acc-0.7933
Iter-18540, train loss-0.8818, acc-0.7800, valid loss-0.8340, acc-0.7912, test loss-0.8346, acc-0.7933
Iter-18550, train loss-0.7633, acc-0.8800, valid loss-0.8336, acc-0.7912, test loss-0.8342, acc-0.7933
Iter-18560, train loss-0.8554, acc-0.7600, valid loss-0.8333, acc-0.7924, test loss-0.8339, acc-0.7931
Iter-18570, train loss-0.9690, acc-0.7000, valid loss-0.8330, acc-0.7928, test loss-0.8336, acc-0.7931
Iter-18580, train loss-0.8288, acc-0.8000, valid loss-0.8326, acc-0.7928, test loss-0.8333, acc-0.7936
Iter-18590, train loss-0.8469, acc-0.7600, valid loss-0.8323, acc-0.7926, test loss-0.8329, acc-0.7933
Iter-18600, train loss-0.8577, acc-0.7600, valid loss-0.8320, acc-0.7928, test loss-0.8326, acc-0.7936
Iter-18610, train loss-0.6933, acc-0.8600, valid loss-0.8316, acc-0.7928, test loss-0.8322, acc-0.7936
Iter-18620, train loss-0.8658, acc-0.7800, valid loss-0.8313, acc-0.7928, test loss-0.8319, acc-0.7940
Iter-18630, train loss-0.8004, acc-0.8200, valid loss-0.8309, acc-0.7932, test loss-0.8315, acc-0.7940
Iter-18640, train loss-0.8931, acc-0.8400, valid loss-0.8305, acc-0.7932, test loss-0.8312, acc-0.7940
Iter-18650, train loss-0.8456, acc-0.7600, valid loss-0.8302, acc-0.7932, test loss-0.8308, acc-0.7939
Iter-18660, train loss-0.8461, acc-0.7400, valid loss-0.8299, acc-0.7930, test loss-0.8305, acc-0.7940
Iter-18670, train loss-0.9490, acc-0.8000, valid loss-0.8295, acc-0.7932, test loss-0.8301, acc-0.7941
Iter-18680, train loss-0.7714, acc-0.8200, valid loss-0.8292, acc-0.7936, test loss-0.8298, acc-0.7940
Iter-18690, train loss-0.9300, acc-0.7400, valid loss-0.8288, acc-0.7936, test loss-0.8295, acc-0.7941
Iter-18700, train loss-0.7830, acc-0.8200, valid loss-0.8285, acc-0.7940, test loss-0.8291, acc-0.7942
Iter-18710, train loss-0.7345, acc-0.8000, valid loss-0.8282, acc-0.7940, test loss-0.8288, acc-0.7941
Iter-18720, train loss-0.6058, acc-0.9200, valid loss-0.8278, acc-0.7940, test loss-0.8285, acc-0.7941
Iter-18730, train loss-0.7683, acc-0.7800, valid loss-0.8275, acc-0.7944, test loss-0.8281, acc-0.7941
Iter-18740, train loss-0.8301, acc-0.7800, valid loss-0.8271, acc-0.7944, test loss-0.8277, acc-0.7942
Iter-18750, train loss-0.7014, acc-0.8200, valid loss-0.8267, acc-0.7944, test loss-0.8274, acc-0.7942
Iter-18760, train loss-0.8005, acc-0.8200, valid loss-0.8264, acc-0.7944, test loss-0.8270, acc-0.7948
Iter-18770, train loss-0.8359, acc-0.7600, valid loss-0.8261, acc-0.7944, test loss-0.8267, acc-0.7947
Iter-18780, train loss-0.8541, acc-0.8000, valid loss-0.8257, acc-0.7944, test loss-0.8263, acc-0.7948
Iter-18790, train loss-0.7831, acc-0.8400, valid loss-0.8253, acc-0.7944, test loss-0.8260, acc-0.7948
Iter-18800, train loss-0.8177, acc-0.8000, valid loss-0.8250, acc-0.7944, test loss-0.8257, acc-0.7949
Iter-18810, train loss-0.8061, acc-0.8400, valid loss-0.8246, acc-0.7944, test loss-0.8253, acc-0.7950
Iter-18820, train loss-0.9934, acc-0.7000, valid loss-0.8243, acc-0.7944, test loss-0.8250, acc-0.7951
Iter-18830, train loss-0.9117, acc-0.8000, valid loss-0.8239, acc-0.7944, test loss-0.8246, acc-0.7954
Iter-18840, train loss-0.8280, acc-0.7600, valid loss-0.8236, acc-0.7944, test loss-0.8243, acc-0.7954
Iter-18850, train loss-0.8389, acc-0.7400, valid loss-0.8233, acc-0.7946, test loss-0.8240, acc-0.7954
Iter-18860, train loss-1.0850, acc-0.6200, valid loss-0.8230, acc-0.7946, test loss-0.8236, acc-0.7956
Iter-18870, train loss-0.7968, acc-0.8400, valid loss-0.8226, acc-0.7946, test loss-0.8233, acc-0.7955
Iter-18880, train loss-0.9645, acc-0.7400, valid loss-0.8223, acc-0.7946, test loss-0.8230, acc-0.7958
Iter-18890, train loss-0.8376, acc-0.8400, valid loss-0.8219, acc-0.7946, test loss-0.8227, acc-0.7961
Iter-18900, train loss-0.8948, acc-0.7400, valid loss-0.8216, acc-0.7948, test loss-0.8223, acc-0.7960
Iter-18910, train loss-0.7573, acc-0.7800, valid loss-0.8212, acc-0.7952, test loss-0.8220, acc-0.7962
Iter-18920, train loss-0.8135, acc-0.8000, valid loss-0.8209, acc-0.7958, test loss-0.8216, acc-0.7963
Iter-18930, train loss-0.7826, acc-0.8200, valid loss-0.8205, acc-0.7956, test loss-0.8213, acc-0.7966
Iter-18940, train loss-0.7600, acc-0.8400, valid loss-0.8202, acc-0.7956, test loss-0.8210, acc-0.7966
Iter-18950, train loss-0.7398, acc-0.8000, valid loss-0.8198, acc-0.7956, test loss-0.8206, acc-0.7968
Iter-18960, train loss-0.8195, acc-0.8000, valid loss-0.8195, acc-0.7958, test loss-0.8203, acc-0.7970
Iter-18970, train loss-0.8389, acc-0.7800, valid loss-0.8192, acc-0.7958, test loss-0.8200, acc-0.7970
Iter-18980, train loss-1.0113, acc-0.7200, valid loss-0.8189, acc-0.7964, test loss-0.8197, acc-0.7969
Iter-18990, train loss-0.7742, acc-0.7800, valid loss-0.8185, acc-0.7964, test loss-0.8193, acc-0.7971
Iter-19000, train loss-0.6341, acc-0.8000, valid loss-0.8182, acc-0.7964, test loss-0.8190, acc-0.7972
Iter-19010, train loss-0.9247, acc-0.7200, valid loss-0.8178, acc-0.7964, test loss-0.8187, acc-0.7970
Iter-19020, train loss-0.8251, acc-0.8400, valid loss-0.8175, acc-0.7964, test loss-0.8184, acc-0.7971
Iter-19030, train loss-0.8217, acc-0.7800, valid loss-0.8172, acc-0.7964, test loss-0.8180, acc-0.7971
Iter-19040, train loss-1.0173, acc-0.6800, valid loss-0.8168, acc-0.7966, test loss-0.8177, acc-0.7972
Iter-19050, train loss-0.8716, acc-0.7800, valid loss-0.8165, acc-0.7968, test loss-0.8173, acc-0.7973
Iter-19060, train loss-0.8332, acc-0.8000, valid loss-0.8162, acc-0.7970, test loss-0.8170, acc-0.7974
Iter-19070, train loss-0.9768, acc-0.6600, valid loss-0.8159, acc-0.7970, test loss-0.8167, acc-0.7972
Iter-19080, train loss-0.6933, acc-0.8800, valid loss-0.8155, acc-0.7972, test loss-0.8164, acc-0.7972
Iter-19090, train loss-0.8217, acc-0.7800, valid loss-0.8152, acc-0.7970, test loss-0.8161, acc-0.7973
Iter-19100, train loss-0.8543, acc-0.7600, valid loss-0.8149, acc-0.7970, test loss-0.8157, acc-0.7977
Iter-19110, train loss-0.9780, acc-0.7200, valid loss-0.8146, acc-0.7976, test loss-0.8154, acc-0.7979
Iter-19120, train loss-0.6928, acc-0.8200, valid loss-0.8142, acc-0.7976, test loss-0.8151, acc-0.7979
Iter-19130, train loss-0.9539, acc-0.7600, valid loss-0.8139, acc-0.7974, test loss-0.8148, acc-0.7979
Iter-19140, train loss-0.8573, acc-0.8200, valid loss-0.8136, acc-0.7974, test loss-0.8144, acc-0.7980
Iter-19150, train loss-1.0300, acc-0.7200, valid loss-0.8133, acc-0.7972, test loss-0.8141, acc-0.7980
Iter-19160, train loss-0.6673, acc-0.8800, valid loss-0.8129, acc-0.7972, test loss-0.8138, acc-0.7983
Iter-19170, train loss-1.0634, acc-0.6600, valid loss-0.8126, acc-0.7974, test loss-0.8135, acc-0.7982
Iter-19180, train loss-0.7446, acc-0.8800, valid loss-0.8123, acc-0.7972, test loss-0.8132, acc-0.7984
Iter-19190, train loss-0.8116, acc-0.8600, valid loss-0.8120, acc-0.7974, test loss-0.8129, acc-0.7984
Iter-19200, train loss-0.8995, acc-0.7400, valid loss-0.8117, acc-0.7974, test loss-0.8126, acc-0.7985
Iter-19210, train loss-0.7812, acc-0.8600, valid loss-0.8114, acc-0.7970, test loss-0.8122, acc-0.7987
Iter-19220, train loss-0.9266, acc-0.7200, valid loss-0.8111, acc-0.7972, test loss-0.8119, acc-0.7988
Iter-19230, train loss-0.8427, acc-0.7800, valid loss-0.8107, acc-0.7974, test loss-0.8116, acc-0.7988
Iter-19240, train loss-0.8569, acc-0.7800, valid loss-0.8104, acc-0.7970, test loss-0.8113, acc-0.7988
Iter-19250, train loss-0.8846, acc-0.6800, valid loss-0.8101, acc-0.7972, test loss-0.8109, acc-0.7988
Iter-19260, train loss-0.7992, acc-0.8400, valid loss-0.8097, acc-0.7976, test loss-0.8106, acc-0.7989
Iter-19270, train loss-0.6701, acc-0.9400, valid loss-0.8094, acc-0.7982, test loss-0.8103, acc-0.7989
Iter-19280, train loss-0.7831, acc-0.8200, valid loss-0.8091, acc-0.7986, test loss-0.8100, acc-0.7989
Iter-19290, train loss-0.7013, acc-0.8600, valid loss-0.8087, acc-0.7986, test loss-0.8096, acc-0.7991
Iter-19300, train loss-0.9880, acc-0.6800, valid loss-0.8084, acc-0.7986, test loss-0.8093, acc-0.7991
Iter-19310, train loss-0.8923, acc-0.8000, valid loss-0.8080, acc-0.7986, test loss-0.8089, acc-0.7992
Iter-19320, train loss-0.9358, acc-0.8000, valid loss-0.8077, acc-0.7988, test loss-0.8087, acc-0.7992
Iter-19330, train loss-0.8347, acc-0.8000, valid loss-0.8074, acc-0.7988, test loss-0.8083, acc-0.7993
Iter-19340, train loss-0.8534, acc-0.8000, valid loss-0.8071, acc-0.7992, test loss-0.8080, acc-0.7997
Iter-19350, train loss-0.8961, acc-0.8000, valid loss-0.8068, acc-0.7992, test loss-0.8077, acc-0.7996
Iter-19360, train loss-0.9694, acc-0.7200, valid loss-0.8065, acc-0.7990, test loss-0.8074, acc-0.7998
Iter-19370, train loss-0.8938, acc-0.7600, valid loss-0.8062, acc-0.7992, test loss-0.8071, acc-0.7998
Iter-19380, train loss-0.6950, acc-0.8400, valid loss-0.8058, acc-0.7996, test loss-0.8068, acc-0.7998
Iter-19390, train loss-0.8781, acc-0.8000, valid loss-0.8055, acc-0.7996, test loss-0.8065, acc-0.7999
Iter-19400, train loss-0.7455, acc-0.8200, valid loss-0.8052, acc-0.7996, test loss-0.8061, acc-0.8001
Iter-19410, train loss-0.6249, acc-0.9200, valid loss-0.8049, acc-0.7996, test loss-0.8058, acc-0.8003
Iter-19420, train loss-0.6877, acc-0.8400, valid loss-0.8046, acc-0.7998, test loss-0.8055, acc-0.8003
Iter-19430, train loss-0.8199, acc-0.8400, valid loss-0.8042, acc-0.7996, test loss-0.8052, acc-0.8002
Iter-19440, train loss-0.8677, acc-0.7400, valid loss-0.8039, acc-0.7998, test loss-0.8049, acc-0.8004
Iter-19450, train loss-1.0315, acc-0.6800, valid loss-0.8036, acc-0.8002, test loss-0.8046, acc-0.8004
Iter-19460, train loss-0.6928, acc-0.8400, valid loss-0.8034, acc-0.8002, test loss-0.8043, acc-0.8004
Iter-19470, train loss-0.9063, acc-0.7200, valid loss-0.8031, acc-0.8000, test loss-0.8040, acc-0.8004
Iter-19480, train loss-0.8381, acc-0.8200, valid loss-0.8028, acc-0.8000, test loss-0.8037, acc-0.8003
Iter-19490, train loss-0.7121, acc-0.8800, valid loss-0.8025, acc-0.8004, test loss-0.8034, acc-0.8005
Iter-19500, train loss-0.7761, acc-0.7800, valid loss-0.8022, acc-0.8004, test loss-0.8031, acc-0.8005
Iter-19510, train loss-0.7873, acc-0.7600, valid loss-0.8019, acc-0.8004, test loss-0.8028, acc-0.8008
Iter-19520, train loss-0.7685, acc-0.8200, valid loss-0.8015, acc-0.8004, test loss-0.8024, acc-0.8009
Iter-19530, train loss-0.6757, acc-0.8200, valid loss-0.8012, acc-0.8004, test loss-0.8021, acc-0.8007
Iter-19540, train loss-0.6898, acc-0.8800, valid loss-0.8009, acc-0.8004, test loss-0.8018, acc-0.8007
Iter-19550, train loss-0.9285, acc-0.7800, valid loss-0.8006, acc-0.8006, test loss-0.8015, acc-0.8007
Iter-19560, train loss-0.8853, acc-0.7400, valid loss-0.8003, acc-0.8004, test loss-0.8012, acc-0.8008
Iter-19570, train loss-0.6928, acc-0.8200, valid loss-0.8000, acc-0.8004, test loss-0.8009, acc-0.8010
Iter-19580, train loss-0.7754, acc-0.8400, valid loss-0.7997, acc-0.8006, test loss-0.8006, acc-0.8008
Iter-19590, train loss-0.7042, acc-0.8600, valid loss-0.7994, acc-0.8006, test loss-0.8003, acc-0.8013
Iter-19600, train loss-0.7201, acc-0.8200, valid loss-0.7991, acc-0.8004, test loss-0.8000, acc-0.8011
Iter-19610, train loss-0.7949, acc-0.7400, valid loss-0.7988, acc-0.8008, test loss-0.7997, acc-0.8012
Iter-19620, train loss-0.7164, acc-0.8400, valid loss-0.7985, acc-0.8004, test loss-0.7994, acc-0.8013
Iter-19630, train loss-0.8145, acc-0.8200, valid loss-0.7982, acc-0.8006, test loss-0.7991, acc-0.8012
Iter-19640, train loss-0.8938, acc-0.7400, valid loss-0.7979, acc-0.8006, test loss-0.7988, acc-0.8012
Iter-19650, train loss-0.7102, acc-0.8400, valid loss-0.7976, acc-0.8008, test loss-0.7985, acc-0.8012
Iter-19660, train loss-0.8029, acc-0.8200, valid loss-0.7972, acc-0.8006, test loss-0.7982, acc-0.8015
Iter-19670, train loss-0.8079, acc-0.8000, valid loss-0.7969, acc-0.8010, test loss-0.7979, acc-0.8015
Iter-19680, train loss-0.6082, acc-0.8800, valid loss-0.7966, acc-0.8012, test loss-0.7976, acc-0.8018
Iter-19690, train loss-0.7239, acc-0.8000, valid loss-0.7963, acc-0.8012, test loss-0.7973, acc-0.8021
Iter-19700, train loss-0.8338, acc-0.7400, valid loss-0.7961, acc-0.8012, test loss-0.7970, acc-0.8022
Iter-19710, train loss-0.9348, acc-0.7000, valid loss-0.7957, acc-0.8012, test loss-0.7966, acc-0.8022
Iter-19720, train loss-0.9888, acc-0.6600, valid loss-0.7954, acc-0.8012, test loss-0.7963, acc-0.8022
Iter-19730, train loss-0.8353, acc-0.7800, valid loss-0.7951, acc-0.8012, test loss-0.7961, acc-0.8023
Iter-19740, train loss-0.9011, acc-0.7800, valid loss-0.7948, acc-0.8016, test loss-0.7958, acc-0.8025
Iter-19750, train loss-0.5922, acc-0.8800, valid loss-0.7945, acc-0.8016, test loss-0.7955, acc-0.8025
Iter-19760, train loss-0.7680, acc-0.8000, valid loss-0.7942, acc-0.8016, test loss-0.7952, acc-0.8026
Iter-19770, train loss-0.6531, acc-0.8200, valid loss-0.7939, acc-0.8018, test loss-0.7949, acc-0.8027
Iter-19780, train loss-1.0734, acc-0.6800, valid loss-0.7937, acc-0.8018, test loss-0.7946, acc-0.8029
Iter-19790, train loss-0.6714, acc-0.8000, valid loss-0.7934, acc-0.8016, test loss-0.7943, acc-0.8028
Iter-19800, train loss-0.7127, acc-0.8800, valid loss-0.7930, acc-0.8020, test loss-0.7940, acc-0.8029
Iter-19810, train loss-0.7375, acc-0.8800, valid loss-0.7927, acc-0.8020, test loss-0.7937, acc-0.8026
Iter-19820, train loss-0.7282, acc-0.8400, valid loss-0.7924, acc-0.8020, test loss-0.7934, acc-0.8024
Iter-19830, train loss-0.7678, acc-0.8200, valid loss-0.7921, acc-0.8022, test loss-0.7931, acc-0.8023
Iter-19840, train loss-0.8505, acc-0.8000, valid loss-0.7918, acc-0.8022, test loss-0.7928, acc-0.8023
Iter-19850, train loss-0.9453, acc-0.7400, valid loss-0.7915, acc-0.8022, test loss-0.7925, acc-0.8024
Iter-19860, train loss-0.7271, acc-0.8200, valid loss-0.7912, acc-0.8024, test loss-0.7922, acc-0.8023
Iter-19870, train loss-0.8777, acc-0.8200, valid loss-0.7910, acc-0.8026, test loss-0.7919, acc-0.8023
Iter-19880, train loss-0.8377, acc-0.7600, valid loss-0.7906, acc-0.8024, test loss-0.7916, acc-0.8021
Iter-19890, train loss-0.8151, acc-0.8200, valid loss-0.7903, acc-0.8024, test loss-0.7913, acc-0.8022
Iter-19900, train loss-0.8375, acc-0.8400, valid loss-0.7901, acc-0.8026, test loss-0.7911, acc-0.8023
Iter-19910, train loss-0.7822, acc-0.8200, valid loss-0.7898, acc-0.8026, test loss-0.7908, acc-0.8022
Iter-19920, train loss-0.8886, acc-0.7800, valid loss-0.7895, acc-0.8026, test loss-0.7905, acc-0.8022
Iter-19930, train loss-0.8997, acc-0.7400, valid loss-0.7892, acc-0.8026, test loss-0.7902, acc-0.8023
Iter-19940, train loss-0.9988, acc-0.6800, valid loss-0.7889, acc-0.8030, test loss-0.7899, acc-0.8024
Iter-19950, train loss-0.6406, acc-0.8800, valid loss-0.7886, acc-0.8030, test loss-0.7896, acc-0.8026
Iter-19960, train loss-0.6839, acc-0.8400, valid loss-0.7883, acc-0.8036, test loss-0.7894, acc-0.8026
Iter-19970, train loss-0.6735, acc-0.8400, valid loss-0.7880, acc-0.8036, test loss-0.7890, acc-0.8025
Iter-19980, train loss-0.8622, acc-0.7800, valid loss-0.7877, acc-0.8036, test loss-0.7888, acc-0.8027
Iter-19990, train loss-0.7645, acc-0.8200, valid loss-0.7874, acc-0.8036, test loss-0.7885, acc-0.8028
Iter-20000, train loss-1.0292, acc-0.7400, valid loss-0.7871, acc-0.8036, test loss-0.7882, acc-0.8031
Iter-20010, train loss-0.9421, acc-0.7800, valid loss-0.7868, acc-0.8034, test loss-0.7879, acc-0.8030
Iter-20020, train loss-0.8311, acc-0.8000, valid loss-0.7866, acc-0.8036, test loss-0.7876, acc-0.8031
Iter-20030, train loss-0.7638, acc-0.8000, valid loss-0.7862, acc-0.8032, test loss-0.7873, acc-0.8032
Iter-20040, train loss-0.7407, acc-0.7600, valid loss-0.7860, acc-0.8036, test loss-0.7870, acc-0.8032
Iter-20050, train loss-0.9931, acc-0.7400, valid loss-0.7857, acc-0.8034, test loss-0.7867, acc-0.8031
Iter-20060, train loss-0.8717, acc-0.7600, valid loss-0.7854, acc-0.8034, test loss-0.7865, acc-0.8032
Iter-20070, train loss-0.8449, acc-0.7600, valid loss-0.7851, acc-0.8032, test loss-0.7862, acc-0.8030
Iter-20080, train loss-1.0969, acc-0.6400, valid loss-0.7848, acc-0.8034, test loss-0.7859, acc-0.8027
Iter-20090, train loss-0.8201, acc-0.8000, valid loss-0.7845, acc-0.8042, test loss-0.7856, acc-0.8029
Iter-20100, train loss-0.9382, acc-0.7000, valid loss-0.7842, acc-0.8042, test loss-0.7853, acc-0.8031
Iter-20110, train loss-0.8413, acc-0.8000, valid loss-0.7839, acc-0.8044, test loss-0.7850, acc-0.8029
Iter-20120, train loss-0.8647, acc-0.8000, valid loss-0.7836, acc-0.8046, test loss-0.7847, acc-0.8033
Iter-20130, train loss-0.8910, acc-0.7600, valid loss-0.7833, acc-0.8046, test loss-0.7844, acc-0.8032
Iter-20140, train loss-1.0299, acc-0.7000, valid loss-0.7831, acc-0.8046, test loss-0.7842, acc-0.8032
Iter-20150, train loss-0.7316, acc-0.8200, valid loss-0.7828, acc-0.8044, test loss-0.7839, acc-0.8032
Iter-20160, train loss-0.7821, acc-0.7800, valid loss-0.7825, acc-0.8042, test loss-0.7836, acc-0.8034
Iter-20170, train loss-0.8678, acc-0.8000, valid loss-0.7822, acc-0.8044, test loss-0.7833, acc-0.8035
Iter-20180, train loss-0.7827, acc-0.8200, valid loss-0.7819, acc-0.8044, test loss-0.7830, acc-0.8035
Iter-20190, train loss-0.8462, acc-0.7800, valid loss-0.7816, acc-0.8046, test loss-0.7827, acc-0.8037
Iter-20200, train loss-0.7797, acc-0.8600, valid loss-0.7814, acc-0.8046, test loss-0.7825, acc-0.8037
Iter-20210, train loss-1.0220, acc-0.6400, valid loss-0.7811, acc-0.8048, test loss-0.7822, acc-0.8038
Iter-20220, train loss-0.8505, acc-0.7600, valid loss-0.7808, acc-0.8046, test loss-0.7819, acc-0.8040
Iter-20230, train loss-0.7486, acc-0.7400, valid loss-0.7805, acc-0.8046, test loss-0.7816, acc-0.8040
Iter-20240, train loss-0.8486, acc-0.8000, valid loss-0.7802, acc-0.8046, test loss-0.7813, acc-0.8040
Iter-20250, train loss-0.7257, acc-0.8600, valid loss-0.7799, acc-0.8048, test loss-0.7811, acc-0.8038
Iter-20260, train loss-0.8490, acc-0.7600, valid loss-0.7796, acc-0.8050, test loss-0.7808, acc-0.8042
Iter-20270, train loss-0.9154, acc-0.7800, valid loss-0.7793, acc-0.8050, test loss-0.7805, acc-0.8044
Iter-20280, train loss-0.8150, acc-0.7800, valid loss-0.7791, acc-0.8050, test loss-0.7802, acc-0.8044
Iter-20290, train loss-0.8026, acc-0.7800, valid loss-0.7788, acc-0.8052, test loss-0.7799, acc-0.8043
Iter-20300, train loss-0.9141, acc-0.6800, valid loss-0.7785, acc-0.8052, test loss-0.7796, acc-0.8045
Iter-20310, train loss-0.7690, acc-0.8000, valid loss-0.7782, acc-0.8054, test loss-0.7793, acc-0.8047
Iter-20320, train loss-0.9456, acc-0.7400, valid loss-0.7780, acc-0.8054, test loss-0.7791, acc-0.8047
Iter-20330, train loss-0.7491, acc-0.8000, valid loss-0.7777, acc-0.8052, test loss-0.7788, acc-0.8047
Iter-20340, train loss-0.9961, acc-0.7200, valid loss-0.7774, acc-0.8052, test loss-0.7785, acc-0.8048
Iter-20350, train loss-0.9442, acc-0.7400, valid loss-0.7771, acc-0.8052, test loss-0.7782, acc-0.8049
Iter-20360, train loss-0.8968, acc-0.7600, valid loss-0.7768, acc-0.8052, test loss-0.7780, acc-0.8049
Iter-20370, train loss-0.6624, acc-0.9200, valid loss-0.7765, acc-0.8056, test loss-0.7777, acc-0.8049
Iter-20380, train loss-0.8974, acc-0.7200, valid loss-0.7763, acc-0.8056, test loss-0.7774, acc-0.8049
Iter-20390, train loss-0.7425, acc-0.8200, valid loss-0.7760, acc-0.8058, test loss-0.7771, acc-0.8052
Iter-20400, train loss-0.7249, acc-0.8200, valid loss-0.7757, acc-0.8062, test loss-0.7769, acc-0.8052
Iter-20410, train loss-0.8278, acc-0.7600, valid loss-0.7754, acc-0.8062, test loss-0.7766, acc-0.8052
Iter-20420, train loss-0.7798, acc-0.7800, valid loss-0.7752, acc-0.8062, test loss-0.7764, acc-0.8056
Iter-20430, train loss-0.7408, acc-0.8200, valid loss-0.7749, acc-0.8064, test loss-0.7761, acc-0.8056
Iter-20440, train loss-0.7747, acc-0.8400, valid loss-0.7746, acc-0.8066, test loss-0.7758, acc-0.8058
Iter-20450, train loss-0.6308, acc-0.8800, valid loss-0.7743, acc-0.8066, test loss-0.7755, acc-0.8057
Iter-20460, train loss-0.7089, acc-0.9000, valid loss-0.7740, acc-0.8068, test loss-0.7752, acc-0.8057
Iter-20470, train loss-0.7612, acc-0.7600, valid loss-0.7738, acc-0.8068, test loss-0.7750, acc-0.8058
Iter-20480, train loss-0.6801, acc-0.8600, valid loss-0.7735, acc-0.8068, test loss-0.7747, acc-0.8062
Iter-20490, train loss-0.7746, acc-0.8000, valid loss-0.7732, acc-0.8070, test loss-0.7744, acc-0.8063
Iter-20500, train loss-0.7892, acc-0.8200, valid loss-0.7729, acc-0.8070, test loss-0.7742, acc-0.8065
Iter-20510, train loss-0.7148, acc-0.8000, valid loss-0.7726, acc-0.8070, test loss-0.7739, acc-0.8067
Iter-20520, train loss-0.7515, acc-0.8400, valid loss-0.7723, acc-0.8070, test loss-0.7736, acc-0.8066
Iter-20530, train loss-0.7687, acc-0.7800, valid loss-0.7720, acc-0.8072, test loss-0.7733, acc-0.8068
Iter-20540, train loss-0.8344, acc-0.7200, valid loss-0.7717, acc-0.8074, test loss-0.7730, acc-0.8070
Iter-20550, train loss-0.8885, acc-0.7200, valid loss-0.7715, acc-0.8076, test loss-0.7728, acc-0.8070
Iter-20560, train loss-0.5299, acc-0.9600, valid loss-0.7712, acc-0.8076, test loss-0.7725, acc-0.8071
Iter-20570, train loss-0.7275, acc-0.8200, valid loss-0.7710, acc-0.8076, test loss-0.7723, acc-0.8072
Iter-20580, train loss-0.8035, acc-0.8200, valid loss-0.7707, acc-0.8076, test loss-0.7720, acc-0.8073
Iter-20590, train loss-0.6318, acc-0.8600, valid loss-0.7704, acc-0.8076, test loss-0.7717, acc-0.8074
Iter-20600, train loss-0.6210, acc-0.8600, valid loss-0.7702, acc-0.8076, test loss-0.7715, acc-0.8075
Iter-20610, train loss-0.8590, acc-0.7600, valid loss-0.7699, acc-0.8078, test loss-0.7712, acc-0.8073
Iter-20620, train loss-0.8180, acc-0.7800, valid loss-0.7696, acc-0.8078, test loss-0.7709, acc-0.8074
Iter-20630, train loss-0.7072, acc-0.8400, valid loss-0.7693, acc-0.8078, test loss-0.7707, acc-0.8074
Iter-20640, train loss-0.6962, acc-0.8000, valid loss-0.7690, acc-0.8084, test loss-0.7704, acc-0.8077
Iter-20650, train loss-0.8451, acc-0.7800, valid loss-0.7688, acc-0.8084, test loss-0.7701, acc-0.8076
Iter-20660, train loss-0.7180, acc-0.8400, valid loss-0.7685, acc-0.8084, test loss-0.7698, acc-0.8078
Iter-20670, train loss-0.7910, acc-0.8000, valid loss-0.7682, acc-0.8084, test loss-0.7695, acc-0.8079
Iter-20680, train loss-0.7679, acc-0.7800, valid loss-0.7679, acc-0.8086, test loss-0.7692, acc-0.8077
Iter-20690, train loss-0.7108, acc-0.8400, valid loss-0.7676, acc-0.8088, test loss-0.7690, acc-0.8079
Iter-20700, train loss-0.7665, acc-0.8200, valid loss-0.7674, acc-0.8090, test loss-0.7687, acc-0.8078
Iter-20710, train loss-0.7898, acc-0.7600, valid loss-0.7671, acc-0.8090, test loss-0.7684, acc-0.8080
Iter-20720, train loss-0.7268, acc-0.8400, valid loss-0.7668, acc-0.8090, test loss-0.7682, acc-0.8077
Iter-20730, train loss-0.8561, acc-0.8200, valid loss-0.7666, acc-0.8088, test loss-0.7679, acc-0.8081
Iter-20740, train loss-0.8904, acc-0.7200, valid loss-0.7663, acc-0.8088, test loss-0.7677, acc-0.8082
Iter-20750, train loss-0.7832, acc-0.7400, valid loss-0.7661, acc-0.8088, test loss-0.7674, acc-0.8081
Iter-20760, train loss-0.7236, acc-0.8600, valid loss-0.7658, acc-0.8088, test loss-0.7672, acc-0.8081
Iter-20770, train loss-0.8544, acc-0.7800, valid loss-0.7655, acc-0.8088, test loss-0.7669, acc-0.8082
Iter-20780, train loss-0.6982, acc-0.8400, valid loss-0.7653, acc-0.8090, test loss-0.7666, acc-0.8082
Iter-20790, train loss-0.8283, acc-0.7200, valid loss-0.7650, acc-0.8094, test loss-0.7663, acc-0.8083
Iter-20800, train loss-0.8729, acc-0.7600, valid loss-0.7647, acc-0.8094, test loss-0.7661, acc-0.8080
Iter-20810, train loss-0.6319, acc-0.8400, valid loss-0.7644, acc-0.8094, test loss-0.7658, acc-0.8083
Iter-20820, train loss-0.6849, acc-0.7800, valid loss-0.7642, acc-0.8098, test loss-0.7655, acc-0.8082
Iter-20830, train loss-0.7668, acc-0.8400, valid loss-0.7639, acc-0.8100, test loss-0.7653, acc-0.8083
Iter-20840, train loss-0.8989, acc-0.7400, valid loss-0.7636, acc-0.8102, test loss-0.7650, acc-0.8082
Iter-20850, train loss-0.7341, acc-0.8200, valid loss-0.7634, acc-0.8102, test loss-0.7648, acc-0.8083
Iter-20860, train loss-0.8807, acc-0.8200, valid loss-0.7631, acc-0.8102, test loss-0.7645, acc-0.8083
Iter-20870, train loss-0.6439, acc-0.8200, valid loss-0.7628, acc-0.8100, test loss-0.7642, acc-0.8082
Iter-20880, train loss-0.7490, acc-0.8400, valid loss-0.7626, acc-0.8102, test loss-0.7640, acc-0.8082
Iter-20890, train loss-0.7811, acc-0.8200, valid loss-0.7623, acc-0.8100, test loss-0.7637, acc-0.8083
Iter-20900, train loss-0.6067, acc-0.9200, valid loss-0.7620, acc-0.8102, test loss-0.7635, acc-0.8084
Iter-20910, train loss-0.7489, acc-0.8400, valid loss-0.7618, acc-0.8100, test loss-0.7632, acc-0.8085
Iter-20920, train loss-0.8475, acc-0.7200, valid loss-0.7615, acc-0.8108, test loss-0.7630, acc-0.8085
Iter-20930, train loss-1.0802, acc-0.7400, valid loss-0.7612, acc-0.8106, test loss-0.7627, acc-0.8087
Iter-20940, train loss-0.7000, acc-0.8600, valid loss-0.7610, acc-0.8108, test loss-0.7625, acc-0.8087
Iter-20950, train loss-0.6707, acc-0.8400, valid loss-0.7608, acc-0.8106, test loss-0.7622, acc-0.8086
Iter-20960, train loss-0.7892, acc-0.7800, valid loss-0.7605, acc-0.8108, test loss-0.7620, acc-0.8088
Iter-20970, train loss-0.8391, acc-0.7400, valid loss-0.7602, acc-0.8108, test loss-0.7617, acc-0.8085
Iter-20980, train loss-0.7948, acc-0.8200, valid loss-0.7599, acc-0.8110, test loss-0.7614, acc-0.8087
Iter-20990, train loss-0.9254, acc-0.6800, valid loss-0.7597, acc-0.8110, test loss-0.7612, acc-0.8086
Iter-21000, train loss-0.9971, acc-0.8000, valid loss-0.7594, acc-0.8110, test loss-0.7609, acc-0.8086
Iter-21010, train loss-0.8821, acc-0.8200, valid loss-0.7592, acc-0.8114, test loss-0.7607, acc-0.8086
Iter-21020, train loss-0.7843, acc-0.8200, valid loss-0.7589, acc-0.8112, test loss-0.7604, acc-0.8088
Iter-21030, train loss-0.7903, acc-0.8600, valid loss-0.7587, acc-0.8114, test loss-0.7602, acc-0.8087
Iter-21040, train loss-0.8205, acc-0.7400, valid loss-0.7584, acc-0.8116, test loss-0.7599, acc-0.8087
Iter-21050, train loss-0.7352, acc-0.8000, valid loss-0.7581, acc-0.8116, test loss-0.7597, acc-0.8086
Iter-21060, train loss-0.7820, acc-0.7400, valid loss-0.7579, acc-0.8116, test loss-0.7594, acc-0.8086
Iter-21070, train loss-0.6624, acc-0.8200, valid loss-0.7576, acc-0.8114, test loss-0.7592, acc-0.8086
Iter-21080, train loss-0.8809, acc-0.7800, valid loss-0.7574, acc-0.8114, test loss-0.7589, acc-0.8087
Iter-21090, train loss-0.6450, acc-0.8600, valid loss-0.7571, acc-0.8114, test loss-0.7587, acc-0.8087
Iter-21100, train loss-0.8284, acc-0.7400, valid loss-0.7568, acc-0.8116, test loss-0.7584, acc-0.8090
Iter-21110, train loss-0.7070, acc-0.9000, valid loss-0.7566, acc-0.8116, test loss-0.7581, acc-0.8089
Iter-21120, train loss-0.8231, acc-0.7800, valid loss-0.7563, acc-0.8114, test loss-0.7579, acc-0.8089
Iter-21130, train loss-0.9127, acc-0.7600, valid loss-0.7561, acc-0.8116, test loss-0.7576, acc-0.8088
Iter-21140, train loss-0.6597, acc-0.8600, valid loss-0.7558, acc-0.8116, test loss-0.7574, acc-0.8087
Iter-21150, train loss-0.8155, acc-0.8400, valid loss-0.7556, acc-0.8120, test loss-0.7571, acc-0.8088
Iter-21160, train loss-0.7412, acc-0.8800, valid loss-0.7553, acc-0.8120, test loss-0.7569, acc-0.8089
Iter-21170, train loss-0.6938, acc-0.8400, valid loss-0.7551, acc-0.8122, test loss-0.7566, acc-0.8091
Iter-21180, train loss-0.6812, acc-0.8200, valid loss-0.7548, acc-0.8120, test loss-0.7564, acc-0.8093
Iter-21190, train loss-0.6590, acc-0.8200, valid loss-0.7546, acc-0.8118, test loss-0.7562, acc-0.8094
Iter-21200, train loss-0.8940, acc-0.7200, valid loss-0.7544, acc-0.8118, test loss-0.7559, acc-0.8096
Iter-21210, train loss-0.7242, acc-0.8400, valid loss-0.7541, acc-0.8120, test loss-0.7556, acc-0.8095
Iter-21220, train loss-0.8723, acc-0.7800, valid loss-0.7538, acc-0.8124, test loss-0.7554, acc-0.8099
Iter-21230, train loss-0.6699, acc-0.8600, valid loss-0.7536, acc-0.8122, test loss-0.7551, acc-0.8098
Iter-21240, train loss-0.8185, acc-0.7400, valid loss-0.7533, acc-0.8120, test loss-0.7549, acc-0.8098
Iter-21250, train loss-0.8389, acc-0.7200, valid loss-0.7531, acc-0.8122, test loss-0.7547, acc-0.8099
Iter-21260, train loss-0.8416, acc-0.7600, valid loss-0.7529, acc-0.8124, test loss-0.7544, acc-0.8098
Iter-21270, train loss-0.9912, acc-0.6800, valid loss-0.7526, acc-0.8126, test loss-0.7541, acc-0.8101
Iter-21280, train loss-0.7814, acc-0.8000, valid loss-0.7524, acc-0.8124, test loss-0.7539, acc-0.8100
Iter-21290, train loss-0.6827, acc-0.7800, valid loss-0.7521, acc-0.8124, test loss-0.7537, acc-0.8100
Iter-21300, train loss-0.7569, acc-0.8400, valid loss-0.7518, acc-0.8124, test loss-0.7534, acc-0.8098
Iter-21310, train loss-0.8115, acc-0.7600, valid loss-0.7516, acc-0.8124, test loss-0.7531, acc-0.8099
Iter-21320, train loss-0.7842, acc-0.7800, valid loss-0.7514, acc-0.8124, test loss-0.7529, acc-0.8099
Iter-21330, train loss-0.6456, acc-0.8600, valid loss-0.7511, acc-0.8124, test loss-0.7527, acc-0.8099
Iter-21340, train loss-0.9065, acc-0.7200, valid loss-0.7509, acc-0.8124, test loss-0.7525, acc-0.8101
Iter-21350, train loss-0.6915, acc-0.8200, valid loss-0.7506, acc-0.8122, test loss-0.7522, acc-0.8101
Iter-21360, train loss-0.6209, acc-0.8400, valid loss-0.7504, acc-0.8122, test loss-0.7520, acc-0.8100
Iter-21370, train loss-0.7311, acc-0.7600, valid loss-0.7501, acc-0.8122, test loss-0.7517, acc-0.8102
Iter-21380, train loss-0.6961, acc-0.8200, valid loss-0.7499, acc-0.8124, test loss-0.7514, acc-0.8102
Iter-21390, train loss-0.6169, acc-0.8600, valid loss-0.7496, acc-0.8124, test loss-0.7512, acc-0.8103
Iter-21400, train loss-0.8732, acc-0.7400, valid loss-0.7494, acc-0.8124, test loss-0.7510, acc-0.8103
Iter-21410, train loss-0.5816, acc-0.8600, valid loss-0.7491, acc-0.8124, test loss-0.7507, acc-0.8105
Iter-21420, train loss-0.5994, acc-0.8400, valid loss-0.7489, acc-0.8124, test loss-0.7505, acc-0.8106
Iter-21430, train loss-0.7077, acc-0.8800, valid loss-0.7486, acc-0.8124, test loss-0.7502, acc-0.8108
Iter-21440, train loss-0.8171, acc-0.8600, valid loss-0.7484, acc-0.8126, test loss-0.7500, acc-0.8110
Iter-21450, train loss-0.8129, acc-0.8200, valid loss-0.7481, acc-0.8126, test loss-0.7497, acc-0.8109
Iter-21460, train loss-0.8407, acc-0.7400, valid loss-0.7478, acc-0.8126, test loss-0.7495, acc-0.8110
Iter-21470, train loss-0.7892, acc-0.7600, valid loss-0.7476, acc-0.8126, test loss-0.7492, acc-0.8111
Iter-21480, train loss-0.6919, acc-0.8000, valid loss-0.7473, acc-0.8124, test loss-0.7490, acc-0.8113
Iter-21490, train loss-0.7414, acc-0.7600, valid loss-0.7471, acc-0.8128, test loss-0.7487, acc-0.8112
Iter-21500, train loss-0.8879, acc-0.7000, valid loss-0.7469, acc-0.8126, test loss-0.7485, acc-0.8115
Iter-21510, train loss-0.6899, acc-0.8600, valid loss-0.7466, acc-0.8130, test loss-0.7483, acc-0.8116
Iter-21520, train loss-0.7139, acc-0.8400, valid loss-0.7464, acc-0.8132, test loss-0.7480, acc-0.8116
Iter-21530, train loss-0.7239, acc-0.8600, valid loss-0.7461, acc-0.8134, test loss-0.7478, acc-0.8116
Iter-21540, train loss-0.7109, acc-0.8000, valid loss-0.7458, acc-0.8134, test loss-0.7475, acc-0.8116
Iter-21550, train loss-0.7634, acc-0.8200, valid loss-0.7456, acc-0.8134, test loss-0.7473, acc-0.8119
Iter-21560, train loss-0.7587, acc-0.7800, valid loss-0.7453, acc-0.8134, test loss-0.7470, acc-0.8116
Iter-21570, train loss-0.6420, acc-0.9000, valid loss-0.7451, acc-0.8134, test loss-0.7468, acc-0.8116
Iter-21580, train loss-0.6256, acc-0.8400, valid loss-0.7448, acc-0.8134, test loss-0.7465, acc-0.8118
Iter-21590, train loss-0.7303, acc-0.8000, valid loss-0.7446, acc-0.8134, test loss-0.7463, acc-0.8121
Iter-21600, train loss-0.6314, acc-0.8800, valid loss-0.7443, acc-0.8134, test loss-0.7460, acc-0.8121
Iter-21610, train loss-0.7536, acc-0.7800, valid loss-0.7441, acc-0.8134, test loss-0.7458, acc-0.8121
Iter-21620, train loss-0.7307, acc-0.7800, valid loss-0.7438, acc-0.8134, test loss-0.7455, acc-0.8120
Iter-21630, train loss-0.8502, acc-0.7800, valid loss-0.7436, acc-0.8134, test loss-0.7453, acc-0.8121
Iter-21640, train loss-0.6168, acc-0.8400, valid loss-0.7433, acc-0.8134, test loss-0.7451, acc-0.8121
Iter-21650, train loss-0.9184, acc-0.6800, valid loss-0.7431, acc-0.8132, test loss-0.7448, acc-0.8121
Iter-21660, train loss-0.7534, acc-0.8400, valid loss-0.7428, acc-0.8132, test loss-0.7446, acc-0.8125
Iter-21670, train loss-0.7146, acc-0.8200, valid loss-0.7425, acc-0.8134, test loss-0.7443, acc-0.8126
Iter-21680, train loss-0.5513, acc-0.9200, valid loss-0.7423, acc-0.8132, test loss-0.7441, acc-0.8129
Iter-21690, train loss-0.6880, acc-0.8600, valid loss-0.7421, acc-0.8134, test loss-0.7438, acc-0.8128
Iter-21700, train loss-0.5831, acc-0.8800, valid loss-0.7418, acc-0.8134, test loss-0.7436, acc-0.8130
Iter-21710, train loss-0.8225, acc-0.7800, valid loss-0.7415, acc-0.8136, test loss-0.7433, acc-0.8133
Iter-21720, train loss-0.7413, acc-0.8200, valid loss-0.7413, acc-0.8136, test loss-0.7431, acc-0.8137
Iter-21730, train loss-0.6514, acc-0.8400, valid loss-0.7410, acc-0.8138, test loss-0.7428, acc-0.8137
Iter-21740, train loss-0.8469, acc-0.7600, valid loss-0.7408, acc-0.8138, test loss-0.7426, acc-0.8134
Iter-21750, train loss-0.6563, acc-0.8600, valid loss-0.7405, acc-0.8138, test loss-0.7424, acc-0.8135
Iter-21760, train loss-0.5872, acc-0.9000, valid loss-0.7403, acc-0.8140, test loss-0.7421, acc-0.8137
Iter-21770, train loss-0.6371, acc-0.8600, valid loss-0.7400, acc-0.8138, test loss-0.7419, acc-0.8139
Iter-21780, train loss-0.7442, acc-0.8000, valid loss-0.7398, acc-0.8140, test loss-0.7416, acc-0.8138
Iter-21790, train loss-0.7261, acc-0.8200, valid loss-0.7396, acc-0.8142, test loss-0.7414, acc-0.8138
Iter-21800, train loss-0.8945, acc-0.6800, valid loss-0.7394, acc-0.8142, test loss-0.7412, acc-0.8141
Iter-21810, train loss-0.6854, acc-0.8200, valid loss-0.7391, acc-0.8142, test loss-0.7409, acc-0.8141
Iter-21820, train loss-0.9487, acc-0.6800, valid loss-0.7389, acc-0.8144, test loss-0.7407, acc-0.8141
Iter-21830, train loss-0.6437, acc-0.8600, valid loss-0.7387, acc-0.8144, test loss-0.7405, acc-0.8143
Iter-21840, train loss-0.8397, acc-0.7000, valid loss-0.7384, acc-0.8142, test loss-0.7402, acc-0.8142
Iter-21850, train loss-0.7264, acc-0.8000, valid loss-0.7382, acc-0.8144, test loss-0.7400, acc-0.8142
Iter-21860, train loss-0.8118, acc-0.7200, valid loss-0.7379, acc-0.8144, test loss-0.7398, acc-0.8143
Iter-21870, train loss-0.7002, acc-0.8600, valid loss-0.7377, acc-0.8146, test loss-0.7395, acc-0.8144
Iter-21880, train loss-0.7123, acc-0.8600, valid loss-0.7375, acc-0.8146, test loss-0.7393, acc-0.8146
Iter-21890, train loss-0.7171, acc-0.7800, valid loss-0.7373, acc-0.8146, test loss-0.7391, acc-0.8146
Iter-21900, train loss-0.4968, acc-0.9200, valid loss-0.7370, acc-0.8148, test loss-0.7388, acc-0.8146
Iter-21910, train loss-0.7525, acc-0.8200, valid loss-0.7368, acc-0.8150, test loss-0.7386, acc-0.8147
Iter-21920, train loss-0.6929, acc-0.8200, valid loss-0.7365, acc-0.8150, test loss-0.7384, acc-0.8146
Iter-21930, train loss-0.6677, acc-0.7600, valid loss-0.7363, acc-0.8150, test loss-0.7381, acc-0.8149
Iter-21940, train loss-0.7484, acc-0.8200, valid loss-0.7361, acc-0.8152, test loss-0.7379, acc-0.8150
Iter-21950, train loss-0.6238, acc-0.8200, valid loss-0.7359, acc-0.8154, test loss-0.7377, acc-0.8149
Iter-21960, train loss-0.7797, acc-0.7600, valid loss-0.7356, acc-0.8152, test loss-0.7374, acc-0.8150
Iter-21970, train loss-0.7337, acc-0.8200, valid loss-0.7354, acc-0.8152, test loss-0.7372, acc-0.8151
Iter-21980, train loss-0.7275, acc-0.8400, valid loss-0.7351, acc-0.8156, test loss-0.7370, acc-0.8151
Iter-21990, train loss-0.6239, acc-0.8600, valid loss-0.7349, acc-0.8154, test loss-0.7368, acc-0.8150
Iter-22000, train loss-0.7391, acc-0.8400, valid loss-0.7347, acc-0.8154, test loss-0.7365, acc-0.8150
Iter-22010, train loss-0.6818, acc-0.8200, valid loss-0.7344, acc-0.8154, test loss-0.7363, acc-0.8153
Iter-22020, train loss-0.7392, acc-0.8400, valid loss-0.7342, acc-0.8150, test loss-0.7361, acc-0.8153
Iter-22030, train loss-0.9234, acc-0.7600, valid loss-0.7340, acc-0.8152, test loss-0.7359, acc-0.8154
Iter-22040, train loss-0.7435, acc-0.8200, valid loss-0.7337, acc-0.8152, test loss-0.7356, acc-0.8153
Iter-22050, train loss-0.8630, acc-0.8000, valid loss-0.7335, acc-0.8158, test loss-0.7354, acc-0.8156
Iter-22060, train loss-0.8879, acc-0.6600, valid loss-0.7332, acc-0.8158, test loss-0.7351, acc-0.8156
Iter-22070, train loss-0.6559, acc-0.9000, valid loss-0.7330, acc-0.8160, test loss-0.7349, acc-0.8157
Iter-22080, train loss-0.6394, acc-0.8600, valid loss-0.7327, acc-0.8162, test loss-0.7346, acc-0.8157
Iter-22090, train loss-0.7297, acc-0.8200, valid loss-0.7325, acc-0.8162, test loss-0.7344, acc-0.8157
Iter-22100, train loss-0.8098, acc-0.7800, valid loss-0.7323, acc-0.8164, test loss-0.7342, acc-0.8156
Iter-22110, train loss-0.6301, acc-0.8400, valid loss-0.7320, acc-0.8168, test loss-0.7340, acc-0.8158
Iter-22120, train loss-0.8110, acc-0.8600, valid loss-0.7318, acc-0.8170, test loss-0.7337, acc-0.8160
Iter-22130, train loss-0.8357, acc-0.7600, valid loss-0.7316, acc-0.8170, test loss-0.7335, acc-0.8159
Iter-22140, train loss-0.7172, acc-0.8600, valid loss-0.7314, acc-0.8170, test loss-0.7333, acc-0.8159
Iter-22150, train loss-0.6488, acc-0.8200, valid loss-0.7312, acc-0.8172, test loss-0.7331, acc-0.8159
Iter-22160, train loss-0.7356, acc-0.7800, valid loss-0.7310, acc-0.8172, test loss-0.7329, acc-0.8159
Iter-22170, train loss-0.7260, acc-0.7600, valid loss-0.7307, acc-0.8172, test loss-0.7327, acc-0.8159
Iter-22180, train loss-0.7190, acc-0.8400, valid loss-0.7305, acc-0.8176, test loss-0.7324, acc-0.8159
Iter-22190, train loss-0.7318, acc-0.7600, valid loss-0.7303, acc-0.8174, test loss-0.7322, acc-0.8161
Iter-22200, train loss-0.8267, acc-0.7600, valid loss-0.7301, acc-0.8176, test loss-0.7320, acc-0.8161
Iter-22210, train loss-0.6829, acc-0.8200, valid loss-0.7298, acc-0.8172, test loss-0.7317, acc-0.8162
Iter-22220, train loss-0.7316, acc-0.8400, valid loss-0.7296, acc-0.8174, test loss-0.7315, acc-0.8166
Iter-22230, train loss-0.8490, acc-0.7200, valid loss-0.7294, acc-0.8172, test loss-0.7313, acc-0.8165
Iter-22240, train loss-1.0621, acc-0.7000, valid loss-0.7291, acc-0.8174, test loss-0.7311, acc-0.8165
Iter-22250, train loss-0.6816, acc-0.8400, valid loss-0.7289, acc-0.8174, test loss-0.7308, acc-0.8164
Iter-22260, train loss-0.8457, acc-0.7600, valid loss-0.7286, acc-0.8172, test loss-0.7306, acc-0.8166
Iter-22270, train loss-0.6978, acc-0.8400, valid loss-0.7284, acc-0.8172, test loss-0.7303, acc-0.8168
Iter-22280, train loss-0.7685, acc-0.7400, valid loss-0.7282, acc-0.8172, test loss-0.7301, acc-0.8167
Iter-22290, train loss-0.8819, acc-0.7800, valid loss-0.7280, acc-0.8172, test loss-0.7299, acc-0.8166
Iter-22300, train loss-0.7480, acc-0.7800, valid loss-0.7277, acc-0.8172, test loss-0.7297, acc-0.8167
Iter-22310, train loss-0.6490, acc-0.9000, valid loss-0.7275, acc-0.8172, test loss-0.7294, acc-0.8169
Iter-22320, train loss-0.8772, acc-0.7600, valid loss-0.7272, acc-0.8172, test loss-0.7292, acc-0.8170
Iter-22330, train loss-0.7754, acc-0.8200, valid loss-0.7270, acc-0.8172, test loss-0.7290, acc-0.8169
Iter-22340, train loss-0.7537, acc-0.7800, valid loss-0.7268, acc-0.8174, test loss-0.7288, acc-0.8169
Iter-22350, train loss-0.5810, acc-0.9400, valid loss-0.7266, acc-0.8174, test loss-0.7286, acc-0.8170
Iter-22360, train loss-0.5490, acc-0.9200, valid loss-0.7264, acc-0.8176, test loss-0.7284, acc-0.8172
Iter-22370, train loss-0.7242, acc-0.8400, valid loss-0.7261, acc-0.8176, test loss-0.7281, acc-0.8173
Iter-22380, train loss-0.6905, acc-0.8600, valid loss-0.7259, acc-0.8174, test loss-0.7279, acc-0.8173
Iter-22390, train loss-0.8602, acc-0.7400, valid loss-0.7257, acc-0.8178, test loss-0.7277, acc-0.8172
Iter-22400, train loss-0.5870, acc-0.8800, valid loss-0.7255, acc-0.8178, test loss-0.7275, acc-0.8174
Iter-22410, train loss-0.8565, acc-0.7000, valid loss-0.7252, acc-0.8180, test loss-0.7273, acc-0.8174
Iter-22420, train loss-0.8280, acc-0.8200, valid loss-0.7250, acc-0.8180, test loss-0.7271, acc-0.8175
Iter-22430, train loss-0.6911, acc-0.8400, valid loss-0.7247, acc-0.8180, test loss-0.7268, acc-0.8176
Iter-22440, train loss-0.7153, acc-0.8200, valid loss-0.7245, acc-0.8180, test loss-0.7266, acc-0.8174
Iter-22450, train loss-0.7124, acc-0.7800, valid loss-0.7243, acc-0.8180, test loss-0.7264, acc-0.8174
Iter-22460, train loss-0.5038, acc-0.9400, valid loss-0.7241, acc-0.8180, test loss-0.7261, acc-0.8175
Iter-22470, train loss-0.9143, acc-0.8000, valid loss-0.7239, acc-0.8180, test loss-0.7259, acc-0.8174
Iter-22480, train loss-0.7499, acc-0.8200, valid loss-0.7236, acc-0.8180, test loss-0.7257, acc-0.8176
Iter-22490, train loss-0.8157, acc-0.8000, valid loss-0.7234, acc-0.8178, test loss-0.7255, acc-0.8178
Iter-22500, train loss-0.7159, acc-0.8000, valid loss-0.7232, acc-0.8178, test loss-0.7253, acc-0.8177
Iter-22510, train loss-0.6369, acc-0.9200, valid loss-0.7230, acc-0.8178, test loss-0.7250, acc-0.8179
Iter-22520, train loss-0.8977, acc-0.7400, valid loss-0.7227, acc-0.8176, test loss-0.7248, acc-0.8180
Iter-22530, train loss-0.6563, acc-0.8400, valid loss-0.7225, acc-0.8178, test loss-0.7246, acc-0.8181
Iter-22540, train loss-0.8976, acc-0.7200, valid loss-0.7223, acc-0.8178, test loss-0.7244, acc-0.8181
Iter-22550, train loss-0.6774, acc-0.8200, valid loss-0.7221, acc-0.8178, test loss-0.7242, acc-0.8180
Iter-22560, train loss-0.6741, acc-0.8600, valid loss-0.7219, acc-0.8178, test loss-0.7239, acc-0.8181
Iter-22570, train loss-0.7536, acc-0.7800, valid loss-0.7217, acc-0.8180, test loss-0.7237, acc-0.8183
Iter-22580, train loss-0.6237, acc-0.8800, valid loss-0.7214, acc-0.8180, test loss-0.7235, acc-0.8182
Iter-22590, train loss-0.7263, acc-0.8400, valid loss-0.7212, acc-0.8182, test loss-0.7233, acc-0.8181
Iter-22600, train loss-0.8803, acc-0.7200, valid loss-0.7210, acc-0.8182, test loss-0.7231, acc-0.8183
Iter-22610, train loss-0.7997, acc-0.7800, valid loss-0.7208, acc-0.8182, test loss-0.7229, acc-0.8183
Iter-22620, train loss-0.5806, acc-0.9000, valid loss-0.7206, acc-0.8180, test loss-0.7226, acc-0.8182
Iter-22630, train loss-0.6514, acc-0.8800, valid loss-0.7204, acc-0.8180, test loss-0.7224, acc-0.8181
Iter-22640, train loss-0.8693, acc-0.7400, valid loss-0.7201, acc-0.8182, test loss-0.7222, acc-0.8181
Iter-22650, train loss-0.7803, acc-0.8000, valid loss-0.7199, acc-0.8182, test loss-0.7220, acc-0.8180
Iter-22660, train loss-0.7717, acc-0.8200, valid loss-0.7197, acc-0.8182, test loss-0.7218, acc-0.8180
Iter-22670, train loss-0.7399, acc-0.8200, valid loss-0.7195, acc-0.8182, test loss-0.7215, acc-0.8180
Iter-22680, train loss-0.7131, acc-0.7800, valid loss-0.7193, acc-0.8182, test loss-0.7213, acc-0.8180
Iter-22690, train loss-0.6453, acc-0.8600, valid loss-0.7191, acc-0.8182, test loss-0.7211, acc-0.8180
Iter-22700, train loss-0.8398, acc-0.7800, valid loss-0.7188, acc-0.8184, test loss-0.7209, acc-0.8181
Iter-22710, train loss-0.8029, acc-0.8000, valid loss-0.7186, acc-0.8184, test loss-0.7207, acc-0.8182
Iter-22720, train loss-0.7530, acc-0.7800, valid loss-0.7184, acc-0.8182, test loss-0.7204, acc-0.8183
Iter-22730, train loss-0.6017, acc-0.8600, valid loss-0.7182, acc-0.8188, test loss-0.7202, acc-0.8181
Iter-22740, train loss-0.7318, acc-0.8400, valid loss-0.7179, acc-0.8186, test loss-0.7200, acc-0.8180
Iter-22750, train loss-0.7825, acc-0.8400, valid loss-0.7177, acc-0.8186, test loss-0.7198, acc-0.8181
Iter-22760, train loss-0.9177, acc-0.6800, valid loss-0.7175, acc-0.8184, test loss-0.7196, acc-0.8182
Iter-22770, train loss-0.6708, acc-0.8800, valid loss-0.7173, acc-0.8184, test loss-0.7194, acc-0.8182
Iter-22780, train loss-0.5604, acc-0.9200, valid loss-0.7171, acc-0.8184, test loss-0.7192, acc-0.8182
Iter-22790, train loss-0.7271, acc-0.8400, valid loss-0.7169, acc-0.8186, test loss-0.7190, acc-0.8184
Iter-22800, train loss-0.9236, acc-0.7800, valid loss-0.7167, acc-0.8186, test loss-0.7187, acc-0.8185
Iter-22810, train loss-0.7337, acc-0.8200, valid loss-0.7164, acc-0.8186, test loss-0.7185, acc-0.8186
Iter-22820, train loss-0.8206, acc-0.8000, valid loss-0.7162, acc-0.8188, test loss-0.7183, acc-0.8188
Iter-22830, train loss-0.6871, acc-0.8400, valid loss-0.7160, acc-0.8188, test loss-0.7181, acc-0.8186
Iter-22840, train loss-0.8496, acc-0.8200, valid loss-0.7158, acc-0.8188, test loss-0.7179, acc-0.8188
Iter-22850, train loss-0.6239, acc-0.9400, valid loss-0.7155, acc-0.8188, test loss-0.7177, acc-0.8190
Iter-22860, train loss-0.8083, acc-0.8400, valid loss-0.7153, acc-0.8190, test loss-0.7174, acc-0.8192
Iter-22870, train loss-0.8126, acc-0.8200, valid loss-0.7151, acc-0.8188, test loss-0.7172, acc-0.8192
Iter-22880, train loss-0.8924, acc-0.6800, valid loss-0.7149, acc-0.8188, test loss-0.7170, acc-0.8192
Iter-22890, train loss-0.6960, acc-0.8400, valid loss-0.7147, acc-0.8188, test loss-0.7168, acc-0.8190
Iter-22900, train loss-0.8016, acc-0.7800, valid loss-0.7144, acc-0.8188, test loss-0.7166, acc-0.8191
Iter-22910, train loss-0.5692, acc-0.8800, valid loss-0.7142, acc-0.8190, test loss-0.7163, acc-0.8192
Iter-22920, train loss-0.7953, acc-0.8000, valid loss-0.7140, acc-0.8192, test loss-0.7161, acc-0.8190
Iter-22930, train loss-0.7334, acc-0.8200, valid loss-0.7138, acc-0.8194, test loss-0.7159, acc-0.8192
Iter-22940, train loss-0.5536, acc-0.9600, valid loss-0.7136, acc-0.8194, test loss-0.7157, acc-0.8191
Iter-22950, train loss-0.7142, acc-0.7800, valid loss-0.7134, acc-0.8192, test loss-0.7155, acc-0.8192
Iter-22960, train loss-0.8802, acc-0.7600, valid loss-0.7132, acc-0.8192, test loss-0.7153, acc-0.8193
Iter-22970, train loss-0.7947, acc-0.7600, valid loss-0.7130, acc-0.8192, test loss-0.7151, acc-0.8193
Iter-22980, train loss-0.8087, acc-0.8200, valid loss-0.7128, acc-0.8192, test loss-0.7149, acc-0.8191
Iter-22990, train loss-0.6688, acc-0.8600, valid loss-0.7125, acc-0.8192, test loss-0.7147, acc-0.8194
Iter-23000, train loss-0.6717, acc-0.8200, valid loss-0.7123, acc-0.8192, test loss-0.7145, acc-0.8196
Iter-23010, train loss-0.7557, acc-0.8200, valid loss-0.7121, acc-0.8192, test loss-0.7143, acc-0.8198
Iter-23020, train loss-0.8163, acc-0.7200, valid loss-0.7119, acc-0.8192, test loss-0.7140, acc-0.8199
Iter-23030, train loss-0.7767, acc-0.7800, valid loss-0.7117, acc-0.8192, test loss-0.7138, acc-0.8201
Iter-23040, train loss-0.7672, acc-0.8400, valid loss-0.7115, acc-0.8192, test loss-0.7136, acc-0.8200
Iter-23050, train loss-0.8476, acc-0.7400, valid loss-0.7113, acc-0.8192, test loss-0.7134, acc-0.8202
Iter-23060, train loss-0.7911, acc-0.7600, valid loss-0.7111, acc-0.8194, test loss-0.7132, acc-0.8203
Iter-23070, train loss-0.6638, acc-0.8400, valid loss-0.7108, acc-0.8194, test loss-0.7130, acc-0.8203
Iter-23080, train loss-0.5979, acc-0.9000, valid loss-0.7106, acc-0.8194, test loss-0.7128, acc-0.8205
Iter-23090, train loss-0.6533, acc-0.8600, valid loss-0.7104, acc-0.8194, test loss-0.7126, acc-0.8205
Iter-23100, train loss-0.7643, acc-0.8000, valid loss-0.7102, acc-0.8194, test loss-0.7124, acc-0.8204
Iter-23110, train loss-0.5512, acc-0.9200, valid loss-0.7100, acc-0.8192, test loss-0.7122, acc-0.8204
Iter-23120, train loss-0.9373, acc-0.7400, valid loss-0.7098, acc-0.8194, test loss-0.7120, acc-0.8206
Iter-23130, train loss-0.6224, acc-0.8200, valid loss-0.7096, acc-0.8198, test loss-0.7118, acc-0.8206
Iter-23140, train loss-0.7030, acc-0.8600, valid loss-0.7094, acc-0.8198, test loss-0.7116, acc-0.8209
Iter-23150, train loss-0.8734, acc-0.7600, valid loss-0.7092, acc-0.8198, test loss-0.7114, acc-0.8210
Iter-23160, train loss-0.6078, acc-0.8800, valid loss-0.7090, acc-0.8202, test loss-0.7112, acc-0.8211
Iter-23170, train loss-0.6076, acc-0.9000, valid loss-0.7088, acc-0.8200, test loss-0.7110, acc-0.8212
Iter-23180, train loss-0.8243, acc-0.8400, valid loss-0.7086, acc-0.8200, test loss-0.7108, acc-0.8211
Iter-23190, train loss-0.8204, acc-0.7600, valid loss-0.7084, acc-0.8198, test loss-0.7106, acc-0.8211
Iter-23200, train loss-0.7272, acc-0.7800, valid loss-0.7082, acc-0.8198, test loss-0.7104, acc-0.8212
Iter-23210, train loss-0.6034, acc-0.8400, valid loss-0.7080, acc-0.8198, test loss-0.7102, acc-0.8213
Iter-23220, train loss-0.8211, acc-0.7400, valid loss-0.7078, acc-0.8200, test loss-0.7100, acc-0.8213
Iter-23230, train loss-0.8948, acc-0.8200, valid loss-0.7076, acc-0.8202, test loss-0.7098, acc-0.8215
Iter-23240, train loss-0.7611, acc-0.7600, valid loss-0.7074, acc-0.8200, test loss-0.7096, acc-0.8216
Iter-23250, train loss-0.7394, acc-0.8400, valid loss-0.7072, acc-0.8200, test loss-0.7094, acc-0.8216
Iter-23260, train loss-0.8787, acc-0.8000, valid loss-0.7070, acc-0.8202, test loss-0.7092, acc-0.8218
Iter-23270, train loss-0.6545, acc-0.7800, valid loss-0.7068, acc-0.8202, test loss-0.7090, acc-0.8216
Iter-23280, train loss-0.8065, acc-0.7400, valid loss-0.7066, acc-0.8202, test loss-0.7088, acc-0.8216
Iter-23290, train loss-0.8722, acc-0.7200, valid loss-0.7064, acc-0.8204, test loss-0.7086, acc-0.8217
Iter-23300, train loss-0.5844, acc-0.8600, valid loss-0.7062, acc-0.8204, test loss-0.7084, acc-0.8220
Iter-23310, train loss-0.7898, acc-0.8200, valid loss-0.7060, acc-0.8204, test loss-0.7082, acc-0.8222
Iter-23320, train loss-0.7671, acc-0.8200, valid loss-0.7058, acc-0.8204, test loss-0.7080, acc-0.8223
Iter-23330, train loss-0.9908, acc-0.7600, valid loss-0.7056, acc-0.8204, test loss-0.7078, acc-0.8225
Iter-23340, train loss-0.7626, acc-0.7400, valid loss-0.7054, acc-0.8204, test loss-0.7076, acc-0.8225
Iter-23350, train loss-0.6978, acc-0.8200, valid loss-0.7052, acc-0.8206, test loss-0.7073, acc-0.8225
Iter-23360, train loss-0.7463, acc-0.7800, valid loss-0.7049, acc-0.8206, test loss-0.7071, acc-0.8225
Iter-23370, train loss-0.7666, acc-0.7800, valid loss-0.7047, acc-0.8208, test loss-0.7069, acc-0.8226
Iter-23380, train loss-0.7069, acc-0.7600, valid loss-0.7045, acc-0.8208, test loss-0.7067, acc-0.8227
Iter-23390, train loss-0.6092, acc-0.8800, valid loss-0.7043, acc-0.8208, test loss-0.7065, acc-0.8227
Iter-23400, train loss-0.7043, acc-0.8000, valid loss-0.7042, acc-0.8208, test loss-0.7063, acc-0.8229
Iter-23410, train loss-0.7891, acc-0.7800, valid loss-0.7039, acc-0.8208, test loss-0.7061, acc-0.8229
Iter-23420, train loss-0.6870, acc-0.8200, valid loss-0.7038, acc-0.8208, test loss-0.7059, acc-0.8228
Iter-23430, train loss-0.8217, acc-0.7600, valid loss-0.7036, acc-0.8206, test loss-0.7057, acc-0.8230
Iter-23440, train loss-0.8419, acc-0.7200, valid loss-0.7034, acc-0.8208, test loss-0.7055, acc-0.8232
Iter-23450, train loss-0.9324, acc-0.7600, valid loss-0.7032, acc-0.8208, test loss-0.7053, acc-0.8232
Iter-23460, train loss-0.7326, acc-0.8000, valid loss-0.7030, acc-0.8208, test loss-0.7052, acc-0.8233
Iter-23470, train loss-0.6584, acc-0.8200, valid loss-0.7028, acc-0.8210, test loss-0.7049, acc-0.8232
Iter-23480, train loss-0.7043, acc-0.8000, valid loss-0.7026, acc-0.8210, test loss-0.7048, acc-0.8234
Iter-23490, train loss-0.7539, acc-0.7800, valid loss-0.7024, acc-0.8212, test loss-0.7046, acc-0.8234
Iter-23500, train loss-0.6856, acc-0.8000, valid loss-0.7022, acc-0.8212, test loss-0.7044, acc-0.8235
Iter-23510, train loss-0.6657, acc-0.8600, valid loss-0.7020, acc-0.8212, test loss-0.7042, acc-0.8235
Iter-23520, train loss-0.8320, acc-0.7400, valid loss-0.7018, acc-0.8212, test loss-0.7040, acc-0.8236
Iter-23530, train loss-0.7705, acc-0.8000, valid loss-0.7016, acc-0.8216, test loss-0.7038, acc-0.8235
Iter-23540, train loss-0.6087, acc-0.8400, valid loss-0.7014, acc-0.8218, test loss-0.7036, acc-0.8236
Iter-23550, train loss-0.6153, acc-0.9000, valid loss-0.7012, acc-0.8218, test loss-0.7034, acc-0.8235
Iter-23560, train loss-0.7537, acc-0.7400, valid loss-0.7010, acc-0.8218, test loss-0.7032, acc-0.8236
Iter-23570, train loss-0.8819, acc-0.8000, valid loss-0.7008, acc-0.8218, test loss-0.7030, acc-0.8237
Iter-23580, train loss-0.7702, acc-0.8200, valid loss-0.7006, acc-0.8216, test loss-0.7028, acc-0.8238
Iter-23590, train loss-0.7185, acc-0.8000, valid loss-0.7004, acc-0.8216, test loss-0.7026, acc-0.8236
Iter-23600, train loss-0.6148, acc-0.9000, valid loss-0.7002, acc-0.8216, test loss-0.7024, acc-0.8238
Iter-23610, train loss-0.7492, acc-0.7800, valid loss-0.7000, acc-0.8218, test loss-0.7022, acc-0.8238
Iter-23620, train loss-0.7887, acc-0.7200, valid loss-0.6998, acc-0.8216, test loss-0.7020, acc-0.8239
Iter-23630, train loss-0.6804, acc-0.8800, valid loss-0.6996, acc-0.8216, test loss-0.7018, acc-0.8240
Iter-23640, train loss-0.8208, acc-0.7200, valid loss-0.6994, acc-0.8218, test loss-0.7016, acc-0.8240
Iter-23650, train loss-0.5194, acc-0.9600, valid loss-0.6992, acc-0.8216, test loss-0.7014, acc-0.8240
Iter-23660, train loss-0.6155, acc-0.8600, valid loss-0.6990, acc-0.8218, test loss-0.7012, acc-0.8241
Iter-23670, train loss-0.6800, acc-0.8000, valid loss-0.6988, acc-0.8218, test loss-0.7010, acc-0.8240
Iter-23680, train loss-0.5221, acc-0.9600, valid loss-0.6986, acc-0.8218, test loss-0.7008, acc-0.8241
Iter-23690, train loss-0.7948, acc-0.7600, valid loss-0.6985, acc-0.8218, test loss-0.7007, acc-0.8240
Iter-23700, train loss-0.6069, acc-0.8800, valid loss-0.6982, acc-0.8218, test loss-0.7004, acc-0.8240
Iter-23710, train loss-0.7250, acc-0.8000, valid loss-0.6980, acc-0.8218, test loss-0.7002, acc-0.8240
Iter-23720, train loss-0.7054, acc-0.8400, valid loss-0.6978, acc-0.8222, test loss-0.7000, acc-0.8240
Iter-23730, train loss-0.6233, acc-0.8200, valid loss-0.6977, acc-0.8220, test loss-0.6999, acc-0.8241
Iter-23740, train loss-0.7835, acc-0.8400, valid loss-0.6975, acc-0.8220, test loss-0.6997, acc-0.8240
Iter-23750, train loss-0.9092, acc-0.7200, valid loss-0.6973, acc-0.8222, test loss-0.6995, acc-0.8240
Iter-23760, train loss-0.7963, acc-0.8000, valid loss-0.6971, acc-0.8222, test loss-0.6993, acc-0.8241
Iter-23770, train loss-0.5907, acc-0.8400, valid loss-0.6969, acc-0.8222, test loss-0.6991, acc-0.8242
Iter-23780, train loss-0.4719, acc-0.9200, valid loss-0.6967, acc-0.8222, test loss-0.6989, acc-0.8243
Iter-23790, train loss-0.6905, acc-0.8600, valid loss-0.6965, acc-0.8222, test loss-0.6987, acc-0.8243
Iter-23800, train loss-0.7805, acc-0.8000, valid loss-0.6963, acc-0.8224, test loss-0.6985, acc-0.8243
Iter-23810, train loss-0.7149, acc-0.8800, valid loss-0.6961, acc-0.8226, test loss-0.6983, acc-0.8244
Iter-23820, train loss-0.6451, acc-0.8400, valid loss-0.6959, acc-0.8224, test loss-0.6981, acc-0.8243
Iter-23830, train loss-0.6698, acc-0.8600, valid loss-0.6957, acc-0.8224, test loss-0.6979, acc-0.8243
Iter-23840, train loss-0.6047, acc-0.8600, valid loss-0.6955, acc-0.8226, test loss-0.6977, acc-0.8244
Iter-23850, train loss-0.9542, acc-0.6600, valid loss-0.6953, acc-0.8228, test loss-0.6975, acc-0.8245
Iter-23860, train loss-0.6530, acc-0.8800, valid loss-0.6951, acc-0.8228, test loss-0.6973, acc-0.8244
Iter-23870, train loss-0.6294, acc-0.8600, valid loss-0.6949, acc-0.8228, test loss-0.6971, acc-0.8245
Iter-23880, train loss-0.6106, acc-0.8400, valid loss-0.6947, acc-0.8228, test loss-0.6969, acc-0.8247
Iter-23890, train loss-0.5265, acc-0.9600, valid loss-0.6945, acc-0.8226, test loss-0.6967, acc-0.8247
Iter-23900, train loss-0.8454, acc-0.7000, valid loss-0.6943, acc-0.8228, test loss-0.6965, acc-0.8249
Iter-23910, train loss-0.9070, acc-0.6800, valid loss-0.6941, acc-0.8228, test loss-0.6963, acc-0.8249
Iter-23920, train loss-0.6584, acc-0.8600, valid loss-0.6939, acc-0.8228, test loss-0.6961, acc-0.8248
Iter-23930, train loss-0.8446, acc-0.7600, valid loss-0.6937, acc-0.8228, test loss-0.6960, acc-0.8248
Iter-23940, train loss-0.6346, acc-0.9000, valid loss-0.6935, acc-0.8228, test loss-0.6958, acc-0.8250
Iter-23950, train loss-0.7981, acc-0.7600, valid loss-0.6933, acc-0.8228, test loss-0.6956, acc-0.8252
Iter-23960, train loss-0.6051, acc-0.8400, valid loss-0.6931, acc-0.8230, test loss-0.6954, acc-0.8251
Iter-23970, train loss-0.6576, acc-0.8200, valid loss-0.6929, acc-0.8230, test loss-0.6952, acc-0.8251
Iter-23980, train loss-0.7554, acc-0.7400, valid loss-0.6927, acc-0.8228, test loss-0.6950, acc-0.8251
Iter-23990, train loss-0.7028, acc-0.8400, valid loss-0.6925, acc-0.8228, test loss-0.6948, acc-0.8253
Iter-24000, train loss-0.5274, acc-0.8800, valid loss-0.6923, acc-0.8228, test loss-0.6946, acc-0.8253
Iter-24010, train loss-0.8031, acc-0.8200, valid loss-0.6922, acc-0.8228, test loss-0.6944, acc-0.8253
Iter-24020, train loss-0.7750, acc-0.8000, valid loss-0.6920, acc-0.8228, test loss-0.6942, acc-0.8252
Iter-24030, train loss-0.8042, acc-0.8000, valid loss-0.6918, acc-0.8228, test loss-0.6940, acc-0.8253
Iter-24040, train loss-0.7472, acc-0.7800, valid loss-0.6916, acc-0.8230, test loss-0.6938, acc-0.8253
Iter-24050, train loss-0.5318, acc-0.9000, valid loss-0.6914, acc-0.8230, test loss-0.6936, acc-0.8256
Iter-24060, train loss-0.5420, acc-0.8600, valid loss-0.6912, acc-0.8232, test loss-0.6935, acc-0.8256
Iter-24070, train loss-0.7776, acc-0.8200, valid loss-0.6910, acc-0.8232, test loss-0.6933, acc-0.8257
Iter-24080, train loss-0.8013, acc-0.7400, valid loss-0.6908, acc-0.8232, test loss-0.6931, acc-0.8258
Iter-24090, train loss-0.6984, acc-0.8600, valid loss-0.6907, acc-0.8232, test loss-0.6929, acc-0.8257
Iter-24100, train loss-0.8951, acc-0.7800, valid loss-0.6905, acc-0.8232, test loss-0.6927, acc-0.8258
Iter-24110, train loss-0.6522, acc-0.8200, valid loss-0.6903, acc-0.8232, test loss-0.6926, acc-0.8258
Iter-24120, train loss-0.7465, acc-0.8000, valid loss-0.6901, acc-0.8230, test loss-0.6924, acc-0.8257
Iter-24130, train loss-0.7117, acc-0.8000, valid loss-0.6899, acc-0.8230, test loss-0.6922, acc-0.8257
Iter-24140, train loss-0.6624, acc-0.8000, valid loss-0.6898, acc-0.8230, test loss-0.6920, acc-0.8257
Iter-24150, train loss-0.6828, acc-0.8400, valid loss-0.6896, acc-0.8230, test loss-0.6919, acc-0.8259
Iter-24160, train loss-0.8224, acc-0.7000, valid loss-0.6894, acc-0.8230, test loss-0.6917, acc-0.8258
Iter-24170, train loss-0.8080, acc-0.7400, valid loss-0.6892, acc-0.8230, test loss-0.6915, acc-0.8257
Iter-24180, train loss-0.5237, acc-0.9400, valid loss-0.6890, acc-0.8228, test loss-0.6913, acc-0.8258
Iter-24190, train loss-0.8206, acc-0.7600, valid loss-0.6888, acc-0.8230, test loss-0.6911, acc-0.8256
Iter-24200, train loss-0.5341, acc-0.8800, valid loss-0.6886, acc-0.8230, test loss-0.6909, acc-0.8258
Iter-24210, train loss-0.5958, acc-0.8200, valid loss-0.6884, acc-0.8230, test loss-0.6907, acc-0.8259
Iter-24220, train loss-0.5936, acc-0.9000, valid loss-0.6882, acc-0.8230, test loss-0.6905, acc-0.8259
Iter-24230, train loss-0.7949, acc-0.7800, valid loss-0.6880, acc-0.8230, test loss-0.6903, acc-0.8261
Iter-24240, train loss-0.7205, acc-0.8400, valid loss-0.6878, acc-0.8228, test loss-0.6902, acc-0.8258
Iter-24250, train loss-0.8050, acc-0.7200, valid loss-0.6876, acc-0.8230, test loss-0.6900, acc-0.8260
Iter-24260, train loss-0.7048, acc-0.8200, valid loss-0.6875, acc-0.8230, test loss-0.6898, acc-0.8257
Iter-24270, train loss-0.7213, acc-0.8000, valid loss-0.6873, acc-0.8230, test loss-0.6896, acc-0.8261
Iter-24280, train loss-0.6745, acc-0.8400, valid loss-0.6871, acc-0.8230, test loss-0.6894, acc-0.8260
Iter-24290, train loss-0.8429, acc-0.6800, valid loss-0.6869, acc-0.8230, test loss-0.6892, acc-0.8263
Iter-24300, train loss-0.6636, acc-0.7800, valid loss-0.6867, acc-0.8228, test loss-0.6891, acc-0.8262
Iter-24310, train loss-0.9329, acc-0.6600, valid loss-0.6865, acc-0.8232, test loss-0.6889, acc-0.8261
Iter-24320, train loss-0.9242, acc-0.7800, valid loss-0.6863, acc-0.8230, test loss-0.6887, acc-0.8261
Iter-24330, train loss-0.7265, acc-0.7600, valid loss-0.6861, acc-0.8226, test loss-0.6885, acc-0.8261
Iter-24340, train loss-0.7529, acc-0.7800, valid loss-0.6859, acc-0.8228, test loss-0.6883, acc-0.8262
Iter-24350, train loss-0.6223, acc-0.9000, valid loss-0.6858, acc-0.8228, test loss-0.6882, acc-0.8261
Iter-24360, train loss-0.7524, acc-0.8000, valid loss-0.6856, acc-0.8228, test loss-0.6880, acc-0.8263
Iter-24370, train loss-0.6975, acc-0.8000, valid loss-0.6854, acc-0.8226, test loss-0.6878, acc-0.8262
Iter-24380, train loss-0.5882, acc-0.8800, valid loss-0.6852, acc-0.8226, test loss-0.6876, acc-0.8264
Iter-24390, train loss-0.8221, acc-0.8400, valid loss-0.6850, acc-0.8226, test loss-0.6874, acc-0.8264
Iter-24400, train loss-0.6878, acc-0.8400, valid loss-0.6848, acc-0.8226, test loss-0.6872, acc-0.8264
Iter-24410, train loss-0.7464, acc-0.8000, valid loss-0.6847, acc-0.8228, test loss-0.6871, acc-0.8262
Iter-24420, train loss-0.7976, acc-0.7600, valid loss-0.6845, acc-0.8226, test loss-0.6869, acc-0.8263
Iter-24430, train loss-0.8212, acc-0.7600, valid loss-0.6843, acc-0.8230, test loss-0.6867, acc-0.8264
Iter-24440, train loss-0.5585, acc-0.8600, valid loss-0.6841, acc-0.8230, test loss-0.6865, acc-0.8265
Iter-24450, train loss-0.9140, acc-0.7400, valid loss-0.6839, acc-0.8232, test loss-0.6863, acc-0.8265
Iter-24460, train loss-0.6626, acc-0.8600, valid loss-0.6837, acc-0.8234, test loss-0.6862, acc-0.8267
Iter-24470, train loss-0.6490, acc-0.7800, valid loss-0.6835, acc-0.8234, test loss-0.6860, acc-0.8267
Iter-24480, train loss-0.7038, acc-0.8000, valid loss-0.6834, acc-0.8236, test loss-0.6858, acc-0.8267
Iter-24490, train loss-0.6962, acc-0.7800, valid loss-0.6832, acc-0.8236, test loss-0.6856, acc-0.8267
Iter-24500, train loss-0.5925, acc-0.8000, valid loss-0.6830, acc-0.8238, test loss-0.6854, acc-0.8270
Iter-24510, train loss-0.6701, acc-0.8000, valid loss-0.6828, acc-0.8236, test loss-0.6852, acc-0.8270
Iter-24520, train loss-0.7799, acc-0.8200, valid loss-0.6826, acc-0.8236, test loss-0.6851, acc-0.8270
Iter-24530, train loss-0.6914, acc-0.8400, valid loss-0.6825, acc-0.8236, test loss-0.6849, acc-0.8267
Iter-24540, train loss-0.9198, acc-0.8000, valid loss-0.6823, acc-0.8236, test loss-0.6847, acc-0.8268
Iter-24550, train loss-0.4863, acc-0.9400, valid loss-0.6822, acc-0.8236, test loss-0.6845, acc-0.8270
Iter-24560, train loss-0.6346, acc-0.8200, valid loss-0.6820, acc-0.8240, test loss-0.6844, acc-0.8270
Iter-24570, train loss-0.7163, acc-0.8000, valid loss-0.6818, acc-0.8238, test loss-0.6842, acc-0.8271
Iter-24580, train loss-0.7089, acc-0.8400, valid loss-0.6816, acc-0.8240, test loss-0.6840, acc-0.8269
Iter-24590, train loss-0.6930, acc-0.8800, valid loss-0.6815, acc-0.8238, test loss-0.6838, acc-0.8270
Iter-24600, train loss-0.6989, acc-0.7800, valid loss-0.6813, acc-0.8240, test loss-0.6837, acc-0.8270
Iter-24610, train loss-0.6983, acc-0.8000, valid loss-0.6811, acc-0.8240, test loss-0.6835, acc-0.8269
Iter-24620, train loss-0.8914, acc-0.8000, valid loss-0.6810, acc-0.8238, test loss-0.6833, acc-0.8269
Iter-24630, train loss-0.7508, acc-0.7800, valid loss-0.6808, acc-0.8236, test loss-0.6831, acc-0.8269
Iter-24640, train loss-0.7672, acc-0.8000, valid loss-0.6806, acc-0.8240, test loss-0.6829, acc-0.8272
Iter-24650, train loss-0.7076, acc-0.8600, valid loss-0.6804, acc-0.8240, test loss-0.6827, acc-0.8273
Iter-24660, train loss-0.7537, acc-0.7800, valid loss-0.6802, acc-0.8240, test loss-0.6826, acc-0.8273
Iter-24670, train loss-0.7023, acc-0.8600, valid loss-0.6800, acc-0.8238, test loss-0.6824, acc-0.8274
Iter-24680, train loss-0.4992, acc-0.8800, valid loss-0.6799, acc-0.8240, test loss-0.6822, acc-0.8275
Iter-24690, train loss-0.9815, acc-0.7200, valid loss-0.6797, acc-0.8238, test loss-0.6820, acc-0.8274
Iter-24700, train loss-0.7360, acc-0.8200, valid loss-0.6795, acc-0.8242, test loss-0.6819, acc-0.8273
Iter-24710, train loss-0.4762, acc-0.9400, valid loss-0.6793, acc-0.8242, test loss-0.6817, acc-0.8277
Iter-24720, train loss-0.7178, acc-0.8200, valid loss-0.6791, acc-0.8244, test loss-0.6815, acc-0.8278
Iter-24730, train loss-0.7733, acc-0.7800, valid loss-0.6790, acc-0.8244, test loss-0.6813, acc-0.8279
Iter-24740, train loss-0.5794, acc-0.8800, valid loss-0.6788, acc-0.8244, test loss-0.6812, acc-0.8278
Iter-24750, train loss-0.6542, acc-0.8400, valid loss-0.6786, acc-0.8242, test loss-0.6810, acc-0.8280
Iter-24760, train loss-0.7129, acc-0.7800, valid loss-0.6784, acc-0.8242, test loss-0.6808, acc-0.8278
Iter-24770, train loss-0.6522, acc-0.8200, valid loss-0.6783, acc-0.8242, test loss-0.6807, acc-0.8280
Iter-24780, train loss-0.6322, acc-0.8200, valid loss-0.6781, acc-0.8246, test loss-0.6805, acc-0.8280
Iter-24790, train loss-0.5243, acc-0.9400, valid loss-0.6779, acc-0.8244, test loss-0.6803, acc-0.8283
Iter-24800, train loss-0.9377, acc-0.6800, valid loss-0.6777, acc-0.8248, test loss-0.6801, acc-0.8281
Iter-24810, train loss-0.6567, acc-0.8000, valid loss-0.6776, acc-0.8248, test loss-0.6799, acc-0.8281
Iter-24820, train loss-0.6808, acc-0.8800, valid loss-0.6774, acc-0.8248, test loss-0.6797, acc-0.8283
Iter-24830, train loss-0.7575, acc-0.7600, valid loss-0.6772, acc-0.8248, test loss-0.6796, acc-0.8282
Iter-24840, train loss-0.6603, acc-0.8400, valid loss-0.6771, acc-0.8248, test loss-0.6794, acc-0.8280
Iter-24850, train loss-0.7516, acc-0.8000, valid loss-0.6769, acc-0.8248, test loss-0.6792, acc-0.8281
Iter-24860, train loss-0.6278, acc-0.8600, valid loss-0.6767, acc-0.8250, test loss-0.6791, acc-0.8281
Iter-24870, train loss-0.6454, acc-0.8200, valid loss-0.6765, acc-0.8248, test loss-0.6789, acc-0.8283
Iter-24880, train loss-0.8415, acc-0.7600, valid loss-0.6764, acc-0.8248, test loss-0.6787, acc-0.8283
Iter-24890, train loss-0.7574, acc-0.7400, valid loss-0.6762, acc-0.8248, test loss-0.6785, acc-0.8282
Iter-24900, train loss-0.5794, acc-0.8400, valid loss-0.6760, acc-0.8252, test loss-0.6784, acc-0.8285
Iter-24910, train loss-0.7719, acc-0.7800, valid loss-0.6758, acc-0.8250, test loss-0.6782, acc-0.8286
Iter-24920, train loss-0.5577, acc-0.8800, valid loss-0.6757, acc-0.8250, test loss-0.6780, acc-0.8286
Iter-24930, train loss-0.5928, acc-0.8600, valid loss-0.6755, acc-0.8252, test loss-0.6778, acc-0.8287
Iter-24940, train loss-0.6213, acc-0.8200, valid loss-0.6753, acc-0.8252, test loss-0.6776, acc-0.8287
Iter-24950, train loss-0.6184, acc-0.8600, valid loss-0.6751, acc-0.8252, test loss-0.6775, acc-0.8288
Iter-24960, train loss-0.7733, acc-0.8200, valid loss-0.6750, acc-0.8254, test loss-0.6773, acc-0.8292
Iter-24970, train loss-0.6406, acc-0.8800, valid loss-0.6748, acc-0.8256, test loss-0.6771, acc-0.8292
Iter-24980, train loss-0.5468, acc-0.9200, valid loss-0.6746, acc-0.8258, test loss-0.6770, acc-0.8294
Iter-24990, train loss-0.7086, acc-0.8600, valid loss-0.6744, acc-0.8260, test loss-0.6768, acc-0.8294
Iter-25000, train loss-0.6656, acc-0.8200, valid loss-0.6743, acc-0.8260, test loss-0.6766, acc-0.8296
Iter-25010, train loss-0.5277, acc-0.9000, valid loss-0.6741, acc-0.8258, test loss-0.6765, acc-0.8295
Iter-25020, train loss-0.7656, acc-0.7400, valid loss-0.6739, acc-0.8262, test loss-0.6763, acc-0.8295
Iter-25030, train loss-0.7891, acc-0.7800, valid loss-0.6738, acc-0.8258, test loss-0.6761, acc-0.8295
Iter-25040, train loss-0.6888, acc-0.8000, valid loss-0.6736, acc-0.8260, test loss-0.6759, acc-0.8294
Iter-25050, train loss-0.8160, acc-0.8000, valid loss-0.6734, acc-0.8260, test loss-0.6758, acc-0.8294
Iter-25060, train loss-0.8372, acc-0.7600, valid loss-0.6732, acc-0.8258, test loss-0.6756, acc-0.8295
Iter-25070, train loss-0.7996, acc-0.8000, valid loss-0.6731, acc-0.8258, test loss-0.6755, acc-0.8295
Iter-25080, train loss-0.7053, acc-0.8200, valid loss-0.6729, acc-0.8262, test loss-0.6753, acc-0.8294
Iter-25090, train loss-0.6169, acc-0.8200, valid loss-0.6727, acc-0.8262, test loss-0.6751, acc-0.8295
Iter-25100, train loss-0.6022, acc-0.8400, valid loss-0.6726, acc-0.8262, test loss-0.6749, acc-0.8295
Iter-25110, train loss-0.7498, acc-0.7800, valid loss-0.6724, acc-0.8262, test loss-0.6748, acc-0.8293
Iter-25120, train loss-0.6140, acc-0.8600, valid loss-0.6722, acc-0.8262, test loss-0.6746, acc-0.8293
Iter-25130, train loss-0.7745, acc-0.7400, valid loss-0.6720, acc-0.8262, test loss-0.6744, acc-0.8294
Iter-25140, train loss-0.6398, acc-0.8600, valid loss-0.6718, acc-0.8260, test loss-0.6742, acc-0.8296
Iter-25150, train loss-1.0277, acc-0.7000, valid loss-0.6717, acc-0.8260, test loss-0.6741, acc-0.8296
Iter-25160, train loss-0.5631, acc-0.8800, valid loss-0.6715, acc-0.8262, test loss-0.6739, acc-0.8296
Iter-25170, train loss-0.6875, acc-0.8000, valid loss-0.6713, acc-0.8262, test loss-0.6737, acc-0.8299
Iter-25180, train loss-0.7153, acc-0.7800, valid loss-0.6711, acc-0.8264, test loss-0.6735, acc-0.8299
Iter-25190, train loss-0.5034, acc-0.9000, valid loss-0.6710, acc-0.8262, test loss-0.6734, acc-0.8298
Iter-25200, train loss-0.6108, acc-0.8800, valid loss-0.6708, acc-0.8264, test loss-0.6732, acc-0.8298
Iter-25210, train loss-0.5070, acc-0.8600, valid loss-0.6706, acc-0.8264, test loss-0.6730, acc-0.8299
Iter-25220, train loss-0.7591, acc-0.8000, valid loss-0.6704, acc-0.8264, test loss-0.6728, acc-0.8299
Iter-25230, train loss-0.5538, acc-0.8400, valid loss-0.6702, acc-0.8262, test loss-0.6727, acc-0.8300
Iter-25240, train loss-0.7075, acc-0.8000, valid loss-0.6701, acc-0.8264, test loss-0.6725, acc-0.8300
Iter-25250, train loss-0.6225, acc-0.8400, valid loss-0.6699, acc-0.8266, test loss-0.6723, acc-0.8299
Iter-25260, train loss-0.6571, acc-0.8800, valid loss-0.6697, acc-0.8266, test loss-0.6722, acc-0.8301
Iter-25270, train loss-0.6935, acc-0.8200, valid loss-0.6696, acc-0.8266, test loss-0.6720, acc-0.8299
Iter-25280, train loss-0.6810, acc-0.7600, valid loss-0.6694, acc-0.8262, test loss-0.6718, acc-0.8298
Iter-25290, train loss-0.6839, acc-0.8200, valid loss-0.6692, acc-0.8262, test loss-0.6717, acc-0.8301
Iter-25300, train loss-0.4928, acc-0.9200, valid loss-0.6691, acc-0.8262, test loss-0.6715, acc-0.8301
Iter-25310, train loss-0.8069, acc-0.7200, valid loss-0.6689, acc-0.8262, test loss-0.6713, acc-0.8300
Iter-25320, train loss-0.6448, acc-0.8400, valid loss-0.6688, acc-0.8266, test loss-0.6712, acc-0.8300
Iter-25330, train loss-0.7153, acc-0.8000, valid loss-0.6686, acc-0.8264, test loss-0.6710, acc-0.8300
Iter-25340, train loss-0.6477, acc-0.8200, valid loss-0.6684, acc-0.8264, test loss-0.6708, acc-0.8302
Iter-25350, train loss-0.6361, acc-0.9000, valid loss-0.6683, acc-0.8264, test loss-0.6707, acc-0.8302
Iter-25360, train loss-0.7126, acc-0.7800, valid loss-0.6681, acc-0.8262, test loss-0.6705, acc-0.8304
Iter-25370, train loss-0.6156, acc-0.8600, valid loss-0.6679, acc-0.8262, test loss-0.6703, acc-0.8304
Iter-25380, train loss-0.6577, acc-0.8200, valid loss-0.6678, acc-0.8264, test loss-0.6702, acc-0.8303
Iter-25390, train loss-0.7469, acc-0.8000, valid loss-0.6676, acc-0.8266, test loss-0.6700, acc-0.8303
Iter-25400, train loss-0.6630, acc-0.8200, valid loss-0.6674, acc-0.8266, test loss-0.6698, acc-0.8305
Iter-25410, train loss-0.6561, acc-0.8000, valid loss-0.6673, acc-0.8266, test loss-0.6696, acc-0.8305
Iter-25420, train loss-0.6904, acc-0.8400, valid loss-0.6671, acc-0.8268, test loss-0.6695, acc-0.8305
Iter-25430, train loss-0.6939, acc-0.8000, valid loss-0.6669, acc-0.8276, test loss-0.6693, acc-0.8304
Iter-25440, train loss-0.5260, acc-0.8800, valid loss-0.6668, acc-0.8276, test loss-0.6692, acc-0.8305
Iter-25450, train loss-0.5741, acc-0.8600, valid loss-0.6666, acc-0.8274, test loss-0.6690, acc-0.8306
Iter-25460, train loss-0.8057, acc-0.7400, valid loss-0.6664, acc-0.8278, test loss-0.6688, acc-0.8304
Iter-25470, train loss-0.6995, acc-0.8200, valid loss-0.6662, acc-0.8278, test loss-0.6687, acc-0.8305
Iter-25480, train loss-0.5109, acc-0.8800, valid loss-0.6661, acc-0.8276, test loss-0.6685, acc-0.8305
Iter-25490, train loss-0.6560, acc-0.8600, valid loss-0.6659, acc-0.8278, test loss-0.6683, acc-0.8305
Iter-25500, train loss-0.6550, acc-0.7800, valid loss-0.6658, acc-0.8276, test loss-0.6681, acc-0.8304
Iter-25510, train loss-0.7309, acc-0.8000, valid loss-0.6656, acc-0.8276, test loss-0.6680, acc-0.8305
Iter-25520, train loss-0.6570, acc-0.8400, valid loss-0.6654, acc-0.8276, test loss-0.6678, acc-0.8306
Iter-25530, train loss-0.9250, acc-0.7400, valid loss-0.6652, acc-0.8276, test loss-0.6676, acc-0.8305
Iter-25540, train loss-0.5822, acc-0.8600, valid loss-0.6651, acc-0.8276, test loss-0.6675, acc-0.8305
Iter-25550, train loss-0.5705, acc-0.9000, valid loss-0.6649, acc-0.8276, test loss-0.6673, acc-0.8306
Iter-25560, train loss-0.6633, acc-0.7600, valid loss-0.6647, acc-0.8276, test loss-0.6671, acc-0.8307
Iter-25570, train loss-0.7837, acc-0.8000, valid loss-0.6645, acc-0.8276, test loss-0.6669, acc-0.8307
Iter-25580, train loss-0.4831, acc-0.9400, valid loss-0.6644, acc-0.8276, test loss-0.6668, acc-0.8309
Iter-25590, train loss-0.5708, acc-0.8600, valid loss-0.6642, acc-0.8278, test loss-0.6666, acc-0.8308
Iter-25600, train loss-0.7449, acc-0.7800, valid loss-0.6641, acc-0.8276, test loss-0.6665, acc-0.8307
Iter-25610, train loss-0.7465, acc-0.8000, valid loss-0.6639, acc-0.8278, test loss-0.6663, acc-0.8307
Iter-25620, train loss-0.8097, acc-0.8400, valid loss-0.6637, acc-0.8278, test loss-0.6661, acc-0.8307
Iter-25630, train loss-0.5783, acc-0.8400, valid loss-0.6636, acc-0.8278, test loss-0.6660, acc-0.8308
Iter-25640, train loss-0.5495, acc-0.8400, valid loss-0.6634, acc-0.8284, test loss-0.6658, acc-0.8308
Iter-25650, train loss-0.7291, acc-0.7800, valid loss-0.6632, acc-0.8286, test loss-0.6657, acc-0.8310
Iter-25660, train loss-0.7538, acc-0.7600, valid loss-0.6631, acc-0.8292, test loss-0.6655, acc-0.8312
Iter-25670, train loss-0.7189, acc-0.8000, valid loss-0.6629, acc-0.8292, test loss-0.6654, acc-0.8310
Iter-25680, train loss-0.6278, acc-0.8400, valid loss-0.6627, acc-0.8292, test loss-0.6652, acc-0.8308
Iter-25690, train loss-0.6463, acc-0.8200, valid loss-0.6626, acc-0.8290, test loss-0.6650, acc-0.8309
Iter-25700, train loss-0.7839, acc-0.7800, valid loss-0.6624, acc-0.8294, test loss-0.6648, acc-0.8310
Iter-25710, train loss-0.6926, acc-0.8000, valid loss-0.6622, acc-0.8296, test loss-0.6647, acc-0.8309
Iter-25720, train loss-0.8420, acc-0.7600, valid loss-0.6621, acc-0.8296, test loss-0.6645, acc-0.8310
Iter-25730, train loss-0.7109, acc-0.8400, valid loss-0.6619, acc-0.8294, test loss-0.6643, acc-0.8310
Iter-25740, train loss-0.6109, acc-0.8400, valid loss-0.6617, acc-0.8296, test loss-0.6642, acc-0.8307
Iter-25750, train loss-0.7306, acc-0.8200, valid loss-0.6616, acc-0.8294, test loss-0.6641, acc-0.8310
Iter-25760, train loss-0.6840, acc-0.8000, valid loss-0.6614, acc-0.8296, test loss-0.6639, acc-0.8309
Iter-25770, train loss-0.7803, acc-0.7600, valid loss-0.6612, acc-0.8296, test loss-0.6637, acc-0.8309
Iter-25780, train loss-0.4920, acc-0.8600, valid loss-0.6611, acc-0.8298, test loss-0.6636, acc-0.8309
Iter-25790, train loss-0.7467, acc-0.8400, valid loss-0.6609, acc-0.8298, test loss-0.6634, acc-0.8309
Iter-25800, train loss-0.7694, acc-0.7800, valid loss-0.6607, acc-0.8296, test loss-0.6632, acc-0.8309
Iter-25810, train loss-0.7159, acc-0.7600, valid loss-0.6606, acc-0.8296, test loss-0.6631, acc-0.8309
Iter-25820, train loss-0.6117, acc-0.9400, valid loss-0.6604, acc-0.8298, test loss-0.6629, acc-0.8310
Iter-25830, train loss-0.6651, acc-0.8000, valid loss-0.6602, acc-0.8298, test loss-0.6628, acc-0.8310
Iter-25840, train loss-0.7026, acc-0.8400, valid loss-0.6601, acc-0.8296, test loss-0.6626, acc-0.8309
Iter-25850, train loss-0.8618, acc-0.7400, valid loss-0.6600, acc-0.8300, test loss-0.6625, acc-0.8311
Iter-25860, train loss-0.6582, acc-0.7600, valid loss-0.6598, acc-0.8302, test loss-0.6623, acc-0.8312
Iter-25870, train loss-0.6375, acc-0.8600, valid loss-0.6596, acc-0.8300, test loss-0.6622, acc-0.8312
Iter-25880, train loss-0.5072, acc-0.9000, valid loss-0.6595, acc-0.8300, test loss-0.6620, acc-0.8311
Iter-25890, train loss-0.4534, acc-0.9600, valid loss-0.6593, acc-0.8298, test loss-0.6619, acc-0.8311
Iter-25900, train loss-0.7959, acc-0.8200, valid loss-0.6592, acc-0.8300, test loss-0.6617, acc-0.8309
Iter-25910, train loss-0.5829, acc-0.8400, valid loss-0.6590, acc-0.8296, test loss-0.6616, acc-0.8311
Iter-25920, train loss-0.8179, acc-0.6800, valid loss-0.6589, acc-0.8296, test loss-0.6614, acc-0.8312
Iter-25930, train loss-0.6296, acc-0.8400, valid loss-0.6587, acc-0.8294, test loss-0.6612, acc-0.8312
Iter-25940, train loss-0.9053, acc-0.7400, valid loss-0.6585, acc-0.8300, test loss-0.6611, acc-0.8310
Iter-25950, train loss-0.8814, acc-0.7400, valid loss-0.6584, acc-0.8298, test loss-0.6609, acc-0.8313
Iter-25960, train loss-0.7493, acc-0.8000, valid loss-0.6582, acc-0.8298, test loss-0.6608, acc-0.8313
Iter-25970, train loss-1.0326, acc-0.7200, valid loss-0.6581, acc-0.8294, test loss-0.6606, acc-0.8313
Iter-25980, train loss-0.4814, acc-0.8800, valid loss-0.6579, acc-0.8296, test loss-0.6604, acc-0.8313
Iter-25990, train loss-0.6416, acc-0.7600, valid loss-0.6577, acc-0.8298, test loss-0.6603, acc-0.8313
Iter-26000, train loss-0.8981, acc-0.7000, valid loss-0.6576, acc-0.8294, test loss-0.6601, acc-0.8313
Iter-26010, train loss-0.5747, acc-0.8800, valid loss-0.6574, acc-0.8296, test loss-0.6600, acc-0.8313
Iter-26020, train loss-0.7449, acc-0.7600, valid loss-0.6572, acc-0.8294, test loss-0.6598, acc-0.8313
Iter-26030, train loss-0.7111, acc-0.8800, valid loss-0.6571, acc-0.8300, test loss-0.6597, acc-0.8312
Iter-26040, train loss-0.7174, acc-0.8400, valid loss-0.6569, acc-0.8302, test loss-0.6595, acc-0.8314
Iter-26050, train loss-0.4929, acc-0.9000, valid loss-0.6567, acc-0.8304, test loss-0.6593, acc-0.8316
Iter-26060, train loss-0.6812, acc-0.8400, valid loss-0.6566, acc-0.8310, test loss-0.6592, acc-0.8316
Iter-26070, train loss-0.7025, acc-0.8000, valid loss-0.6564, acc-0.8306, test loss-0.6590, acc-0.8314
Iter-26080, train loss-0.6217, acc-0.8600, valid loss-0.6563, acc-0.8306, test loss-0.6589, acc-0.8315
Iter-26090, train loss-0.5582, acc-0.8400, valid loss-0.6561, acc-0.8308, test loss-0.6587, acc-0.8315
Iter-26100, train loss-0.8443, acc-0.7400, valid loss-0.6560, acc-0.8312, test loss-0.6586, acc-0.8317
Iter-26110, train loss-0.7737, acc-0.7400, valid loss-0.6558, acc-0.8306, test loss-0.6584, acc-0.8317
Iter-26120, train loss-0.7435, acc-0.7800, valid loss-0.6557, acc-0.8306, test loss-0.6583, acc-0.8317
Iter-26130, train loss-0.5826, acc-0.8000, valid loss-0.6555, acc-0.8310, test loss-0.6581, acc-0.8317
Iter-26140, train loss-0.5520, acc-0.9000, valid loss-0.6553, acc-0.8310, test loss-0.6579, acc-0.8315
Iter-26150, train loss-0.6277, acc-0.8200, valid loss-0.6552, acc-0.8310, test loss-0.6578, acc-0.8314
Iter-26160, train loss-0.7908, acc-0.8000, valid loss-0.6550, acc-0.8312, test loss-0.6576, acc-0.8316
Iter-26170, train loss-0.7373, acc-0.7400, valid loss-0.6549, acc-0.8312, test loss-0.6575, acc-0.8317
Iter-26180, train loss-0.7770, acc-0.8200, valid loss-0.6547, acc-0.8312, test loss-0.6573, acc-0.8317
Iter-26190, train loss-0.5747, acc-0.8800, valid loss-0.6545, acc-0.8314, test loss-0.6572, acc-0.8316
Iter-26200, train loss-0.4027, acc-0.9400, valid loss-0.6544, acc-0.8312, test loss-0.6570, acc-0.8315
Iter-26210, train loss-0.7772, acc-0.7200, valid loss-0.6542, acc-0.8312, test loss-0.6569, acc-0.8317
Iter-26220, train loss-0.6172, acc-0.8200, valid loss-0.6541, acc-0.8314, test loss-0.6567, acc-0.8317
Iter-26230, train loss-0.8817, acc-0.7200, valid loss-0.6539, acc-0.8314, test loss-0.6565, acc-0.8319
Iter-26240, train loss-0.6093, acc-0.8600, valid loss-0.6537, acc-0.8318, test loss-0.6564, acc-0.8319
Iter-26250, train loss-0.5344, acc-0.9000, valid loss-0.6536, acc-0.8318, test loss-0.6562, acc-0.8319
Iter-26260, train loss-0.4490, acc-0.9000, valid loss-0.6535, acc-0.8318, test loss-0.6561, acc-0.8319
Iter-26270, train loss-0.5378, acc-0.8600, valid loss-0.6533, acc-0.8318, test loss-0.6559, acc-0.8318
Iter-26280, train loss-0.6620, acc-0.8800, valid loss-0.6531, acc-0.8318, test loss-0.6557, acc-0.8321
Iter-26290, train loss-0.6472, acc-0.8200, valid loss-0.6530, acc-0.8318, test loss-0.6556, acc-0.8319
Iter-26300, train loss-0.7229, acc-0.8000, valid loss-0.6528, acc-0.8318, test loss-0.6554, acc-0.8321
Iter-26310, train loss-0.7313, acc-0.7600, valid loss-0.6526, acc-0.8320, test loss-0.6553, acc-0.8321
Iter-26320, train loss-0.4933, acc-0.8600, valid loss-0.6525, acc-0.8318, test loss-0.6551, acc-0.8320
Iter-26330, train loss-0.6739, acc-0.8000, valid loss-0.6523, acc-0.8318, test loss-0.6550, acc-0.8319
Iter-26340, train loss-0.7329, acc-0.7800, valid loss-0.6522, acc-0.8318, test loss-0.6548, acc-0.8320
Iter-26350, train loss-0.5926, acc-0.8600, valid loss-0.6520, acc-0.8318, test loss-0.6547, acc-0.8319
Iter-26360, train loss-0.6612, acc-0.8000, valid loss-0.6518, acc-0.8318, test loss-0.6545, acc-0.8319
Iter-26370, train loss-0.6416, acc-0.8200, valid loss-0.6517, acc-0.8316, test loss-0.6543, acc-0.8318
Iter-26380, train loss-0.6365, acc-0.8000, valid loss-0.6515, acc-0.8316, test loss-0.6542, acc-0.8322
Iter-26390, train loss-0.6510, acc-0.8200, valid loss-0.6514, acc-0.8318, test loss-0.6540, acc-0.8322
Iter-26400, train loss-0.7022, acc-0.8000, valid loss-0.6512, acc-0.8316, test loss-0.6539, acc-0.8322
Iter-26410, train loss-0.6634, acc-0.8200, valid loss-0.6511, acc-0.8316, test loss-0.6537, acc-0.8323
Iter-26420, train loss-0.8353, acc-0.8200, valid loss-0.6509, acc-0.8316, test loss-0.6536, acc-0.8323
Iter-26430, train loss-0.5895, acc-0.8800, valid loss-0.6508, acc-0.8318, test loss-0.6534, acc-0.8323
Iter-26440, train loss-0.6130, acc-0.8200, valid loss-0.6506, acc-0.8316, test loss-0.6533, acc-0.8322
Iter-26450, train loss-0.5885, acc-0.8800, valid loss-0.6505, acc-0.8314, test loss-0.6531, acc-0.8323
Iter-26460, train loss-0.4454, acc-0.9200, valid loss-0.6503, acc-0.8314, test loss-0.6530, acc-0.8323
Iter-26470, train loss-0.7981, acc-0.7600, valid loss-0.6502, acc-0.8314, test loss-0.6528, acc-0.8325
Iter-26480, train loss-0.5755, acc-0.8400, valid loss-0.6500, acc-0.8316, test loss-0.6527, acc-0.8324
Iter-26490, train loss-0.6907, acc-0.7800, valid loss-0.6498, acc-0.8318, test loss-0.6525, acc-0.8324
Iter-26500, train loss-0.7326, acc-0.7800, valid loss-0.6497, acc-0.8318, test loss-0.6524, acc-0.8324
Iter-26510, train loss-0.5636, acc-0.8800, valid loss-0.6495, acc-0.8320, test loss-0.6522, acc-0.8327
Iter-26520, train loss-0.6053, acc-0.8400, valid loss-0.6493, acc-0.8320, test loss-0.6520, acc-0.8326
Iter-26530, train loss-0.8029, acc-0.8200, valid loss-0.6492, acc-0.8322, test loss-0.6519, acc-0.8326
Iter-26540, train loss-0.6856, acc-0.8200, valid loss-0.6490, acc-0.8324, test loss-0.6518, acc-0.8326
Iter-26550, train loss-0.5734, acc-0.8600, valid loss-0.6489, acc-0.8328, test loss-0.6516, acc-0.8327
Iter-26560, train loss-0.7058, acc-0.8600, valid loss-0.6487, acc-0.8328, test loss-0.6514, acc-0.8328
Iter-26570, train loss-0.7576, acc-0.7600, valid loss-0.6486, acc-0.8326, test loss-0.6513, acc-0.8327
Iter-26580, train loss-0.6515, acc-0.7200, valid loss-0.6484, acc-0.8326, test loss-0.6511, acc-0.8328
Iter-26590, train loss-0.6361, acc-0.9200, valid loss-0.6483, acc-0.8324, test loss-0.6510, acc-0.8329
Iter-26600, train loss-0.6825, acc-0.7800, valid loss-0.6481, acc-0.8324, test loss-0.6508, acc-0.8328
Iter-26610, train loss-0.7214, acc-0.8400, valid loss-0.6480, acc-0.8324, test loss-0.6507, acc-0.8328
Iter-26620, train loss-0.8133, acc-0.7600, valid loss-0.6478, acc-0.8324, test loss-0.6505, acc-0.8328
Iter-26630, train loss-0.6609, acc-0.8200, valid loss-0.6476, acc-0.8326, test loss-0.6504, acc-0.8329
Iter-26640, train loss-0.6688, acc-0.8200, valid loss-0.6475, acc-0.8324, test loss-0.6502, acc-0.8328
Iter-26650, train loss-0.5560, acc-0.8400, valid loss-0.6473, acc-0.8324, test loss-0.6500, acc-0.8328
Iter-26660, train loss-0.6445, acc-0.8200, valid loss-0.6472, acc-0.8326, test loss-0.6499, acc-0.8330
Iter-26670, train loss-0.6516, acc-0.8400, valid loss-0.6470, acc-0.8328, test loss-0.6497, acc-0.8330
Iter-26680, train loss-0.8535, acc-0.7400, valid loss-0.6469, acc-0.8328, test loss-0.6496, acc-0.8330
Iter-26690, train loss-0.5980, acc-0.8800, valid loss-0.6467, acc-0.8326, test loss-0.6495, acc-0.8330
Iter-26700, train loss-0.5878, acc-0.8200, valid loss-0.6465, acc-0.8326, test loss-0.6493, acc-0.8330
Iter-26710, train loss-0.6286, acc-0.9000, valid loss-0.6464, acc-0.8328, test loss-0.6491, acc-0.8330
Iter-26720, train loss-0.6914, acc-0.7800, valid loss-0.6462, acc-0.8328, test loss-0.6490, acc-0.8332
Iter-26730, train loss-0.6399, acc-0.8400, valid loss-0.6461, acc-0.8328, test loss-0.6489, acc-0.8331
Iter-26740, train loss-0.6568, acc-0.8400, valid loss-0.6459, acc-0.8326, test loss-0.6487, acc-0.8331
Iter-26750, train loss-0.7063, acc-0.7400, valid loss-0.6458, acc-0.8326, test loss-0.6486, acc-0.8330
Iter-26760, train loss-0.7472, acc-0.7200, valid loss-0.6456, acc-0.8326, test loss-0.6484, acc-0.8329
Iter-26770, train loss-0.4959, acc-0.8400, valid loss-0.6455, acc-0.8326, test loss-0.6483, acc-0.8329
Iter-26780, train loss-0.5669, acc-0.8400, valid loss-0.6453, acc-0.8326, test loss-0.6481, acc-0.8331
Iter-26790, train loss-0.5188, acc-0.9000, valid loss-0.6451, acc-0.8328, test loss-0.6480, acc-0.8332
Iter-26800, train loss-0.5171, acc-0.9200, valid loss-0.6450, acc-0.8328, test loss-0.6478, acc-0.8331
Iter-26810, train loss-0.8042, acc-0.8000, valid loss-0.6448, acc-0.8330, test loss-0.6477, acc-0.8330
Iter-26820, train loss-0.7926, acc-0.7600, valid loss-0.6447, acc-0.8330, test loss-0.6475, acc-0.8332
Iter-26830, train loss-0.6873, acc-0.7800, valid loss-0.6445, acc-0.8332, test loss-0.6473, acc-0.8332
Iter-26840, train loss-0.5443, acc-0.8600, valid loss-0.6443, acc-0.8330, test loss-0.6472, acc-0.8333
Iter-26850, train loss-0.8187, acc-0.7000, valid loss-0.6442, acc-0.8332, test loss-0.6470, acc-0.8333
Iter-26860, train loss-0.4876, acc-0.8400, valid loss-0.6441, acc-0.8330, test loss-0.6469, acc-0.8335
Iter-26870, train loss-0.7554, acc-0.7800, valid loss-0.6439, acc-0.8330, test loss-0.6467, acc-0.8334
Iter-26880, train loss-0.5725, acc-0.8400, valid loss-0.6438, acc-0.8330, test loss-0.6466, acc-0.8338
Iter-26890, train loss-0.5554, acc-0.8600, valid loss-0.6436, acc-0.8330, test loss-0.6464, acc-0.8338
Iter-26900, train loss-0.5340, acc-0.9400, valid loss-0.6434, acc-0.8330, test loss-0.6463, acc-0.8339
Iter-26910, train loss-0.7588, acc-0.7800, valid loss-0.6433, acc-0.8330, test loss-0.6461, acc-0.8338
Iter-26920, train loss-0.5782, acc-0.8600, valid loss-0.6431, acc-0.8332, test loss-0.6460, acc-0.8337
Iter-26930, train loss-0.6997, acc-0.8000, valid loss-0.6430, acc-0.8334, test loss-0.6459, acc-0.8340
Iter-26940, train loss-0.7889, acc-0.7400, valid loss-0.6429, acc-0.8334, test loss-0.6457, acc-0.8340
Iter-26950, train loss-0.6881, acc-0.8000, valid loss-0.6427, acc-0.8334, test loss-0.6456, acc-0.8339
Iter-26960, train loss-0.6591, acc-0.8200, valid loss-0.6426, acc-0.8334, test loss-0.6454, acc-0.8340
Iter-26970, train loss-0.8649, acc-0.7200, valid loss-0.6424, acc-0.8334, test loss-0.6453, acc-0.8338
Iter-26980, train loss-0.5756, acc-0.8800, valid loss-0.6423, acc-0.8336, test loss-0.6451, acc-0.8339
Iter-26990, train loss-0.7684, acc-0.7800, valid loss-0.6421, acc-0.8336, test loss-0.6450, acc-0.8343
Iter-27000, train loss-0.6079, acc-0.8800, valid loss-0.6420, acc-0.8338, test loss-0.6448, acc-0.8344
Iter-27010, train loss-0.7554, acc-0.7600, valid loss-0.6418, acc-0.8336, test loss-0.6447, acc-0.8344
Iter-27020, train loss-0.8293, acc-0.7400, valid loss-0.6417, acc-0.8338, test loss-0.6445, acc-0.8345
Iter-27030, train loss-0.6559, acc-0.8400, valid loss-0.6415, acc-0.8338, test loss-0.6444, acc-0.8346
Iter-27040, train loss-0.6008, acc-0.8400, valid loss-0.6414, acc-0.8338, test loss-0.6443, acc-0.8346
Iter-27050, train loss-0.7001, acc-0.7800, valid loss-0.6412, acc-0.8338, test loss-0.6441, acc-0.8345
Iter-27060, train loss-0.5164, acc-0.9000, valid loss-0.6411, acc-0.8338, test loss-0.6440, acc-0.8348
Iter-27070, train loss-0.5619, acc-0.8600, valid loss-0.6409, acc-0.8338, test loss-0.6438, acc-0.8345
Iter-27080, train loss-0.5732, acc-0.9000, valid loss-0.6408, acc-0.8336, test loss-0.6437, acc-0.8345
Iter-27090, train loss-0.6536, acc-0.8400, valid loss-0.6406, acc-0.8338, test loss-0.6435, acc-0.8345
Iter-27100, train loss-1.0052, acc-0.7200, valid loss-0.6405, acc-0.8338, test loss-0.6434, acc-0.8346
Iter-27110, train loss-0.6107, acc-0.8600, valid loss-0.6403, acc-0.8338, test loss-0.6432, acc-0.8346
Iter-27120, train loss-0.6900, acc-0.8400, valid loss-0.6402, acc-0.8338, test loss-0.6431, acc-0.8345
Iter-27130, train loss-0.7101, acc-0.8200, valid loss-0.6400, acc-0.8338, test loss-0.6429, acc-0.8345
Iter-27140, train loss-0.5371, acc-0.8800, valid loss-0.6399, acc-0.8338, test loss-0.6428, acc-0.8347
Iter-27150, train loss-0.9068, acc-0.7400, valid loss-0.6398, acc-0.8338, test loss-0.6427, acc-0.8347
Iter-27160, train loss-0.4448, acc-0.9400, valid loss-0.6396, acc-0.8338, test loss-0.6425, acc-0.8347
Iter-27170, train loss-0.5984, acc-0.8400, valid loss-0.6395, acc-0.8338, test loss-0.6424, acc-0.8349
Iter-27180, train loss-0.7796, acc-0.7400, valid loss-0.6393, acc-0.8338, test loss-0.6422, acc-0.8349
Iter-27190, train loss-0.6907, acc-0.7800, valid loss-0.6392, acc-0.8338, test loss-0.6421, acc-0.8350
Iter-27200, train loss-0.5004, acc-0.8600, valid loss-0.6390, acc-0.8338, test loss-0.6419, acc-0.8350
Iter-27210, train loss-0.6454, acc-0.8000, valid loss-0.6389, acc-0.8338, test loss-0.6418, acc-0.8350
Iter-27220, train loss-0.7184, acc-0.7600, valid loss-0.6388, acc-0.8340, test loss-0.6416, acc-0.8349
Iter-27230, train loss-0.7385, acc-0.8400, valid loss-0.6386, acc-0.8338, test loss-0.6415, acc-0.8351
Iter-27240, train loss-0.6942, acc-0.7600, valid loss-0.6385, acc-0.8338, test loss-0.6413, acc-0.8351
Iter-27250, train loss-0.6428, acc-0.8600, valid loss-0.6383, acc-0.8340, test loss-0.6412, acc-0.8351
Iter-27260, train loss-0.6620, acc-0.8600, valid loss-0.6382, acc-0.8344, test loss-0.6410, acc-0.8350
Iter-27270, train loss-0.6072, acc-0.8800, valid loss-0.6380, acc-0.8342, test loss-0.6409, acc-0.8350
Iter-27280, train loss-0.5589, acc-0.8000, valid loss-0.6379, acc-0.8342, test loss-0.6408, acc-0.8350
Iter-27290, train loss-0.6977, acc-0.8200, valid loss-0.6377, acc-0.8342, test loss-0.6406, acc-0.8350
Iter-27300, train loss-0.7155, acc-0.8000, valid loss-0.6376, acc-0.8344, test loss-0.6405, acc-0.8350
Iter-27310, train loss-0.6092, acc-0.8200, valid loss-0.6374, acc-0.8342, test loss-0.6403, acc-0.8350
Iter-27320, train loss-0.7558, acc-0.8000, valid loss-0.6373, acc-0.8344, test loss-0.6402, acc-0.8350
Iter-27330, train loss-0.6611, acc-0.8200, valid loss-0.6372, acc-0.8342, test loss-0.6401, acc-0.8352
Iter-27340, train loss-0.7764, acc-0.8200, valid loss-0.6370, acc-0.8346, test loss-0.6399, acc-0.8352
Iter-27350, train loss-0.5610, acc-0.8200, valid loss-0.6369, acc-0.8344, test loss-0.6398, acc-0.8353
Iter-27360, train loss-0.5117, acc-0.8600, valid loss-0.6367, acc-0.8344, test loss-0.6396, acc-0.8353
Iter-27370, train loss-0.5164, acc-0.9600, valid loss-0.6366, acc-0.8344, test loss-0.6395, acc-0.8352
Iter-27380, train loss-0.7009, acc-0.8200, valid loss-0.6364, acc-0.8344, test loss-0.6394, acc-0.8354
Iter-27390, train loss-0.6985, acc-0.7800, valid loss-0.6363, acc-0.8344, test loss-0.6392, acc-0.8352
Iter-27400, train loss-0.6744, acc-0.8400, valid loss-0.6361, acc-0.8344, test loss-0.6391, acc-0.8352
Iter-27410, train loss-0.6536, acc-0.8400, valid loss-0.6360, acc-0.8344, test loss-0.6390, acc-0.8353
Iter-27420, train loss-0.4927, acc-0.9400, valid loss-0.6358, acc-0.8344, test loss-0.6388, acc-0.8353
Iter-27430, train loss-0.6729, acc-0.8600, valid loss-0.6357, acc-0.8344, test loss-0.6387, acc-0.8351
Iter-27440, train loss-0.6071, acc-0.8000, valid loss-0.6355, acc-0.8344, test loss-0.6385, acc-0.8353
Iter-27450, train loss-0.5783, acc-0.8200, valid loss-0.6354, acc-0.8344, test loss-0.6384, acc-0.8354
Iter-27460, train loss-0.6859, acc-0.8200, valid loss-0.6353, acc-0.8346, test loss-0.6383, acc-0.8353
Iter-27470, train loss-0.8477, acc-0.7800, valid loss-0.6351, acc-0.8346, test loss-0.6381, acc-0.8352
Iter-27480, train loss-0.5923, acc-0.7800, valid loss-0.6350, acc-0.8348, test loss-0.6380, acc-0.8355
Iter-27490, train loss-0.5095, acc-0.8800, valid loss-0.6349, acc-0.8350, test loss-0.6378, acc-0.8356
Iter-27500, train loss-0.7017, acc-0.8600, valid loss-0.6347, acc-0.8350, test loss-0.6377, acc-0.8354
Iter-27510, train loss-0.6698, acc-0.7600, valid loss-0.6346, acc-0.8352, test loss-0.6376, acc-0.8355
Iter-27520, train loss-0.7867, acc-0.7800, valid loss-0.6344, acc-0.8350, test loss-0.6374, acc-0.8354
Iter-27530, train loss-0.6611, acc-0.8200, valid loss-0.6343, acc-0.8352, test loss-0.6373, acc-0.8355
Iter-27540, train loss-0.6716, acc-0.8600, valid loss-0.6342, acc-0.8352, test loss-0.6371, acc-0.8355
Iter-27550, train loss-0.4787, acc-0.9400, valid loss-0.6340, acc-0.8352, test loss-0.6370, acc-0.8354
Iter-27560, train loss-0.5818, acc-0.8800, valid loss-0.6339, acc-0.8352, test loss-0.6368, acc-0.8356
Iter-27570, train loss-0.7365, acc-0.8200, valid loss-0.6337, acc-0.8356, test loss-0.6367, acc-0.8355
Iter-27580, train loss-0.5503, acc-0.8800, valid loss-0.6336, acc-0.8356, test loss-0.6366, acc-0.8356
Iter-27590, train loss-0.7467, acc-0.8000, valid loss-0.6335, acc-0.8358, test loss-0.6365, acc-0.8356
Iter-27600, train loss-0.6339, acc-0.8200, valid loss-0.6334, acc-0.8354, test loss-0.6363, acc-0.8356
Iter-27610, train loss-0.6636, acc-0.7600, valid loss-0.6332, acc-0.8356, test loss-0.6362, acc-0.8356
Iter-27620, train loss-0.6613, acc-0.8200, valid loss-0.6330, acc-0.8356, test loss-0.6360, acc-0.8356
Iter-27630, train loss-0.6993, acc-0.7400, valid loss-0.6329, acc-0.8358, test loss-0.6359, acc-0.8356
Iter-27640, train loss-0.4261, acc-0.9000, valid loss-0.6328, acc-0.8358, test loss-0.6358, acc-0.8356
Iter-27650, train loss-0.6990, acc-0.7600, valid loss-0.6326, acc-0.8356, test loss-0.6356, acc-0.8355
Iter-27660, train loss-0.5899, acc-0.8600, valid loss-0.6325, acc-0.8356, test loss-0.6355, acc-0.8354
Iter-27670, train loss-0.7312, acc-0.7800, valid loss-0.6324, acc-0.8356, test loss-0.6353, acc-0.8354
Iter-27680, train loss-0.5593, acc-0.8800, valid loss-0.6322, acc-0.8356, test loss-0.6352, acc-0.8357
Iter-27690, train loss-0.5504, acc-0.8800, valid loss-0.6321, acc-0.8360, test loss-0.6351, acc-0.8354
Iter-27700, train loss-0.4418, acc-0.9000, valid loss-0.6320, acc-0.8358, test loss-0.6349, acc-0.8353
Iter-27710, train loss-0.4928, acc-0.8800, valid loss-0.6319, acc-0.8358, test loss-0.6348, acc-0.8354
Iter-27720, train loss-0.5257, acc-0.8600, valid loss-0.6317, acc-0.8358, test loss-0.6347, acc-0.8354
Iter-27730, train loss-0.4457, acc-0.9000, valid loss-0.6316, acc-0.8360, test loss-0.6345, acc-0.8355
Iter-27740, train loss-0.5905, acc-0.8800, valid loss-0.6315, acc-0.8360, test loss-0.6344, acc-0.8353
Iter-27750, train loss-0.4344, acc-0.9200, valid loss-0.6313, acc-0.8360, test loss-0.6343, acc-0.8354
Iter-27760, train loss-0.7242, acc-0.8400, valid loss-0.6312, acc-0.8364, test loss-0.6341, acc-0.8354
Iter-27770, train loss-0.5318, acc-0.8600, valid loss-0.6311, acc-0.8362, test loss-0.6340, acc-0.8356
Iter-27780, train loss-0.9176, acc-0.7200, valid loss-0.6309, acc-0.8366, test loss-0.6339, acc-0.8357
Iter-27790, train loss-0.6281, acc-0.8000, valid loss-0.6308, acc-0.8366, test loss-0.6337, acc-0.8358
Iter-27800, train loss-0.5728, acc-0.8600, valid loss-0.6306, acc-0.8368, test loss-0.6336, acc-0.8357
Iter-27810, train loss-0.6467, acc-0.8000, valid loss-0.6305, acc-0.8368, test loss-0.6334, acc-0.8358
Iter-27820, train loss-0.6452, acc-0.8000, valid loss-0.6303, acc-0.8368, test loss-0.6333, acc-0.8360
Iter-27830, train loss-0.6157, acc-0.8800, valid loss-0.6302, acc-0.8366, test loss-0.6332, acc-0.8360
Iter-27840, train loss-0.7873, acc-0.7800, valid loss-0.6301, acc-0.8368, test loss-0.6331, acc-0.8360
Iter-27850, train loss-0.6659, acc-0.8400, valid loss-0.6299, acc-0.8370, test loss-0.6329, acc-0.8359
Iter-27860, train loss-0.5386, acc-0.8800, valid loss-0.6298, acc-0.8368, test loss-0.6328, acc-0.8359
Iter-27870, train loss-0.6188, acc-0.8600, valid loss-0.6297, acc-0.8368, test loss-0.6327, acc-0.8359
Iter-27880, train loss-0.7212, acc-0.8400, valid loss-0.6295, acc-0.8370, test loss-0.6325, acc-0.8358
Iter-27890, train loss-0.7635, acc-0.7600, valid loss-0.6294, acc-0.8370, test loss-0.6324, acc-0.8358
Iter-27900, train loss-0.8497, acc-0.7600, valid loss-0.6293, acc-0.8372, test loss-0.6323, acc-0.8358
Iter-27910, train loss-0.5791, acc-0.8600, valid loss-0.6291, acc-0.8372, test loss-0.6321, acc-0.8358
Iter-27920, train loss-0.5779, acc-0.8600, valid loss-0.6290, acc-0.8372, test loss-0.6320, acc-0.8360
Iter-27930, train loss-0.7526, acc-0.8000, valid loss-0.6289, acc-0.8372, test loss-0.6319, acc-0.8360
Iter-27940, train loss-0.6656, acc-0.8400, valid loss-0.6287, acc-0.8372, test loss-0.6317, acc-0.8361
Iter-27950, train loss-0.7522, acc-0.7800, valid loss-0.6286, acc-0.8370, test loss-0.6316, acc-0.8360
Iter-27960, train loss-0.6186, acc-0.8600, valid loss-0.6285, acc-0.8372, test loss-0.6315, acc-0.8359
Iter-27970, train loss-0.6674, acc-0.7800, valid loss-0.6283, acc-0.8374, test loss-0.6313, acc-0.8361
Iter-27980, train loss-0.6380, acc-0.8000, valid loss-0.6282, acc-0.8368, test loss-0.6312, acc-0.8361
Iter-27990, train loss-0.7373, acc-0.8600, valid loss-0.6281, acc-0.8374, test loss-0.6311, acc-0.8361
Iter-28000, train loss-0.9604, acc-0.8000, valid loss-0.6279, acc-0.8374, test loss-0.6309, acc-0.8361
Iter-28010, train loss-0.5410, acc-0.8600, valid loss-0.6278, acc-0.8372, test loss-0.6308, acc-0.8361
Iter-28020, train loss-0.6917, acc-0.7800, valid loss-0.6277, acc-0.8372, test loss-0.6306, acc-0.8359
Iter-28030, train loss-0.7277, acc-0.8200, valid loss-0.6275, acc-0.8376, test loss-0.6305, acc-0.8360
Iter-28040, train loss-0.6348, acc-0.8200, valid loss-0.6274, acc-0.8374, test loss-0.6304, acc-0.8360
Iter-28050, train loss-0.5730, acc-0.9200, valid loss-0.6273, acc-0.8378, test loss-0.6303, acc-0.8360
Iter-28060, train loss-0.6504, acc-0.8200, valid loss-0.6271, acc-0.8376, test loss-0.6302, acc-0.8360
Iter-28070, train loss-0.6463, acc-0.8400, valid loss-0.6270, acc-0.8378, test loss-0.6300, acc-0.8360
Iter-28080, train loss-0.7889, acc-0.7600, valid loss-0.6269, acc-0.8376, test loss-0.6299, acc-0.8359
Iter-28090, train loss-0.6655, acc-0.8000, valid loss-0.6267, acc-0.8378, test loss-0.6298, acc-0.8359
Iter-28100, train loss-0.6223, acc-0.8400, valid loss-0.6266, acc-0.8380, test loss-0.6296, acc-0.8359
Iter-28110, train loss-0.6190, acc-0.8400, valid loss-0.6265, acc-0.8376, test loss-0.6295, acc-0.8359
Iter-28120, train loss-0.6666, acc-0.8800, valid loss-0.6263, acc-0.8378, test loss-0.6293, acc-0.8358
Iter-28130, train loss-0.7713, acc-0.7400, valid loss-0.6262, acc-0.8376, test loss-0.6292, acc-0.8358
Iter-28140, train loss-0.4796, acc-0.9000, valid loss-0.6261, acc-0.8376, test loss-0.6291, acc-0.8356
Iter-28150, train loss-0.6213, acc-0.8400, valid loss-0.6259, acc-0.8376, test loss-0.6289, acc-0.8357
Iter-28160, train loss-0.6534, acc-0.8600, valid loss-0.6258, acc-0.8384, test loss-0.6288, acc-0.8356
Iter-28170, train loss-0.6505, acc-0.8200, valid loss-0.6257, acc-0.8376, test loss-0.6287, acc-0.8356
Iter-28180, train loss-0.6059, acc-0.7800, valid loss-0.6255, acc-0.8374, test loss-0.6285, acc-0.8357
Iter-28190, train loss-0.5240, acc-0.8400, valid loss-0.6254, acc-0.8376, test loss-0.6284, acc-0.8358
Iter-28200, train loss-0.6341, acc-0.8600, valid loss-0.6252, acc-0.8374, test loss-0.6282, acc-0.8359
Iter-28210, train loss-0.5884, acc-0.8600, valid loss-0.6251, acc-0.8378, test loss-0.6281, acc-0.8358
Iter-28220, train loss-0.5201, acc-0.8400, valid loss-0.6250, acc-0.8378, test loss-0.6280, acc-0.8357
Iter-28230, train loss-0.5185, acc-0.8600, valid loss-0.6249, acc-0.8378, test loss-0.6279, acc-0.8357
Iter-28240, train loss-0.5123, acc-0.9200, valid loss-0.6247, acc-0.8378, test loss-0.6277, acc-0.8358
Iter-28250, train loss-0.5193, acc-0.8600, valid loss-0.6246, acc-0.8378, test loss-0.6276, acc-0.8358
Iter-28260, train loss-0.6753, acc-0.7800, valid loss-0.6245, acc-0.8378, test loss-0.6274, acc-0.8357
Iter-28270, train loss-0.8036, acc-0.7600, valid loss-0.6243, acc-0.8378, test loss-0.6273, acc-0.8358
Iter-28280, train loss-0.6834, acc-0.8600, valid loss-0.6242, acc-0.8378, test loss-0.6272, acc-0.8356
Iter-28290, train loss-0.6526, acc-0.8400, valid loss-0.6240, acc-0.8380, test loss-0.6270, acc-0.8356
Iter-28300, train loss-0.6851, acc-0.8400, valid loss-0.6239, acc-0.8382, test loss-0.6269, acc-0.8357
Iter-28310, train loss-0.6749, acc-0.8400, valid loss-0.6238, acc-0.8384, test loss-0.6268, acc-0.8356
Iter-28320, train loss-0.7278, acc-0.8200, valid loss-0.6237, acc-0.8386, test loss-0.6266, acc-0.8356
Iter-28330, train loss-0.6232, acc-0.8200, valid loss-0.6235, acc-0.8386, test loss-0.6265, acc-0.8357
Iter-28340, train loss-0.7404, acc-0.7600, valid loss-0.6234, acc-0.8386, test loss-0.6264, acc-0.8358
Iter-28350, train loss-0.5425, acc-0.9200, valid loss-0.6232, acc-0.8386, test loss-0.6262, acc-0.8358
Iter-28360, train loss-0.6610, acc-0.8400, valid loss-0.6231, acc-0.8386, test loss-0.6261, acc-0.8358
Iter-28370, train loss-0.7519, acc-0.7800, valid loss-0.6230, acc-0.8386, test loss-0.6260, acc-0.8361
Iter-28380, train loss-0.4913, acc-0.9200, valid loss-0.6228, acc-0.8390, test loss-0.6259, acc-0.8359
Iter-28390, train loss-0.7009, acc-0.8000, valid loss-0.6227, acc-0.8388, test loss-0.6257, acc-0.8359
Iter-28400, train loss-0.7492, acc-0.9000, valid loss-0.6226, acc-0.8388, test loss-0.6256, acc-0.8358
Iter-28410, train loss-0.5174, acc-0.8600, valid loss-0.6225, acc-0.8390, test loss-0.6255, acc-0.8359
Iter-28420, train loss-0.6784, acc-0.7800, valid loss-0.6223, acc-0.8388, test loss-0.6253, acc-0.8361
Iter-28430, train loss-0.6668, acc-0.8200, valid loss-0.6222, acc-0.8388, test loss-0.6252, acc-0.8363
Iter-28440, train loss-0.5445, acc-0.8200, valid loss-0.6220, acc-0.8388, test loss-0.6251, acc-0.8363
Iter-28450, train loss-0.5895, acc-0.8400, valid loss-0.6219, acc-0.8386, test loss-0.6249, acc-0.8362
Iter-28460, train loss-0.5317, acc-0.8400, valid loss-0.6218, acc-0.8386, test loss-0.6248, acc-0.8362
Iter-28470, train loss-0.8137, acc-0.7400, valid loss-0.6217, acc-0.8386, test loss-0.6247, acc-0.8364
Iter-28480, train loss-0.8268, acc-0.7800, valid loss-0.6215, acc-0.8388, test loss-0.6245, acc-0.8365
Iter-28490, train loss-0.5625, acc-0.8400, valid loss-0.6214, acc-0.8388, test loss-0.6244, acc-0.8364
Iter-28500, train loss-0.6150, acc-0.8600, valid loss-0.6213, acc-0.8390, test loss-0.6243, acc-0.8365
Iter-28510, train loss-0.6029, acc-0.8400, valid loss-0.6211, acc-0.8386, test loss-0.6241, acc-0.8364
Iter-28520, train loss-0.5782, acc-0.9000, valid loss-0.6210, acc-0.8388, test loss-0.6240, acc-0.8365
Iter-28530, train loss-0.6911, acc-0.8400, valid loss-0.6208, acc-0.8388, test loss-0.6238, acc-0.8366
Iter-28540, train loss-0.6079, acc-0.8600, valid loss-0.6207, acc-0.8390, test loss-0.6237, acc-0.8364
Iter-28550, train loss-0.5362, acc-0.8800, valid loss-0.6206, acc-0.8390, test loss-0.6236, acc-0.8366
Iter-28560, train loss-0.6487, acc-0.8800, valid loss-0.6204, acc-0.8388, test loss-0.6235, acc-0.8366
Iter-28570, train loss-0.5011, acc-0.9000, valid loss-0.6203, acc-0.8390, test loss-0.6233, acc-0.8365
Iter-28580, train loss-0.7500, acc-0.7800, valid loss-0.6202, acc-0.8392, test loss-0.6232, acc-0.8364
Iter-28590, train loss-0.7266, acc-0.8400, valid loss-0.6200, acc-0.8392, test loss-0.6231, acc-0.8365
Iter-28600, train loss-0.5908, acc-0.8400, valid loss-0.6199, acc-0.8394, test loss-0.6229, acc-0.8365
Iter-28610, train loss-0.7801, acc-0.7000, valid loss-0.6198, acc-0.8392, test loss-0.6228, acc-0.8365
Iter-28620, train loss-0.5377, acc-0.8200, valid loss-0.6196, acc-0.8392, test loss-0.6227, acc-0.8365
Iter-28630, train loss-0.7892, acc-0.6800, valid loss-0.6195, acc-0.8392, test loss-0.6225, acc-0.8365
Iter-28640, train loss-0.6254, acc-0.8200, valid loss-0.6194, acc-0.8392, test loss-0.6224, acc-0.8365
Iter-28650, train loss-0.4466, acc-0.9000, valid loss-0.6193, acc-0.8392, test loss-0.6223, acc-0.8365
Iter-28660, train loss-0.5206, acc-0.8800, valid loss-0.6191, acc-0.8392, test loss-0.6222, acc-0.8366
Iter-28670, train loss-0.5566, acc-0.8800, valid loss-0.6190, acc-0.8392, test loss-0.6220, acc-0.8365
Iter-28680, train loss-0.5477, acc-0.9000, valid loss-0.6189, acc-0.8392, test loss-0.6219, acc-0.8367
Iter-28690, train loss-0.5370, acc-0.8600, valid loss-0.6188, acc-0.8392, test loss-0.6218, acc-0.8367
Iter-28700, train loss-0.6020, acc-0.8400, valid loss-0.6187, acc-0.8392, test loss-0.6217, acc-0.8367
Iter-28710, train loss-0.5087, acc-0.8600, valid loss-0.6185, acc-0.8392, test loss-0.6215, acc-0.8367
Iter-28720, train loss-0.6730, acc-0.8600, valid loss-0.6184, acc-0.8390, test loss-0.6214, acc-0.8369
Iter-28730, train loss-0.6434, acc-0.8200, valid loss-0.6183, acc-0.8390, test loss-0.6213, acc-0.8370
Iter-28740, train loss-0.5327, acc-0.8800, valid loss-0.6181, acc-0.8390, test loss-0.6212, acc-0.8373
Iter-28750, train loss-0.5078, acc-0.8800, valid loss-0.6180, acc-0.8390, test loss-0.6210, acc-0.8370
Iter-28760, train loss-0.4284, acc-0.9400, valid loss-0.6179, acc-0.8392, test loss-0.6209, acc-0.8371
Iter-28770, train loss-0.4495, acc-0.9000, valid loss-0.6177, acc-0.8392, test loss-0.6208, acc-0.8371
Iter-28780, train loss-0.6731, acc-0.7800, valid loss-0.6176, acc-0.8396, test loss-0.6206, acc-0.8371
Iter-28790, train loss-0.6404, acc-0.8000, valid loss-0.6175, acc-0.8392, test loss-0.6205, acc-0.8373
Iter-28800, train loss-0.5667, acc-0.8800, valid loss-0.6174, acc-0.8394, test loss-0.6204, acc-0.8373
Iter-28810, train loss-0.8094, acc-0.8000, valid loss-0.6173, acc-0.8392, test loss-0.6203, acc-0.8373
Iter-28820, train loss-0.4004, acc-0.9600, valid loss-0.6172, acc-0.8392, test loss-0.6202, acc-0.8374
Iter-28830, train loss-0.5605, acc-0.8800, valid loss-0.6170, acc-0.8392, test loss-0.6200, acc-0.8373
Iter-28840, train loss-0.5725, acc-0.8600, valid loss-0.6169, acc-0.8394, test loss-0.6199, acc-0.8373
Iter-28850, train loss-0.4984, acc-0.8800, valid loss-0.6168, acc-0.8394, test loss-0.6198, acc-0.8375
Iter-28860, train loss-0.6516, acc-0.8400, valid loss-0.6167, acc-0.8392, test loss-0.6197, acc-0.8376
Iter-28870, train loss-0.5894, acc-0.8400, valid loss-0.6166, acc-0.8396, test loss-0.6196, acc-0.8375
Iter-28880, train loss-0.6283, acc-0.8400, valid loss-0.6164, acc-0.8396, test loss-0.6195, acc-0.8377
Iter-28890, train loss-0.6116, acc-0.8200, valid loss-0.6163, acc-0.8394, test loss-0.6193, acc-0.8377
Iter-28900, train loss-0.4044, acc-0.9400, valid loss-0.6162, acc-0.8396, test loss-0.6192, acc-0.8376
Iter-28910, train loss-0.5772, acc-0.8600, valid loss-0.6160, acc-0.8396, test loss-0.6191, acc-0.8377
Iter-28920, train loss-0.5504, acc-0.8800, valid loss-0.6159, acc-0.8396, test loss-0.6190, acc-0.8379
Iter-28930, train loss-0.5571, acc-0.8800, valid loss-0.6158, acc-0.8396, test loss-0.6189, acc-0.8378
Iter-28940, train loss-0.5896, acc-0.7800, valid loss-0.6156, acc-0.8396, test loss-0.6187, acc-0.8379
Iter-28950, train loss-0.5748, acc-0.8400, valid loss-0.6155, acc-0.8400, test loss-0.6186, acc-0.8379
Iter-28960, train loss-0.5250, acc-0.8800, valid loss-0.6154, acc-0.8398, test loss-0.6185, acc-0.8379
Iter-28970, train loss-0.5734, acc-0.8600, valid loss-0.6152, acc-0.8398, test loss-0.6183, acc-0.8379
Iter-28980, train loss-0.7198, acc-0.8000, valid loss-0.6151, acc-0.8398, test loss-0.6182, acc-0.8380
Iter-28990, train loss-0.6508, acc-0.8200, valid loss-0.6150, acc-0.8400, test loss-0.6181, acc-0.8380
Iter-29000, train loss-0.6203, acc-0.8200, valid loss-0.6148, acc-0.8400, test loss-0.6179, acc-0.8380
Iter-29010, train loss-0.6789, acc-0.8000, valid loss-0.6147, acc-0.8400, test loss-0.6178, acc-0.8380
Iter-29020, train loss-0.9467, acc-0.7600, valid loss-0.6146, acc-0.8404, test loss-0.6177, acc-0.8380
Iter-29030, train loss-0.5501, acc-0.8800, valid loss-0.6145, acc-0.8406, test loss-0.6176, acc-0.8380
Iter-29040, train loss-0.6702, acc-0.8200, valid loss-0.6143, acc-0.8406, test loss-0.6174, acc-0.8380
Iter-29050, train loss-0.6926, acc-0.8400, valid loss-0.6142, acc-0.8406, test loss-0.6173, acc-0.8381
Iter-29060, train loss-0.7154, acc-0.8200, valid loss-0.6141, acc-0.8404, test loss-0.6172, acc-0.8382
Iter-29070, train loss-0.5753, acc-0.8800, valid loss-0.6140, acc-0.8402, test loss-0.6171, acc-0.8381
Iter-29080, train loss-0.6910, acc-0.8000, valid loss-0.6138, acc-0.8402, test loss-0.6169, acc-0.8382
Iter-29090, train loss-0.7277, acc-0.8000, valid loss-0.6137, acc-0.8398, test loss-0.6168, acc-0.8382
Iter-29100, train loss-0.5187, acc-0.8800, valid loss-0.6136, acc-0.8398, test loss-0.6167, acc-0.8384
Iter-29110, train loss-0.4958, acc-0.8800, valid loss-0.6135, acc-0.8400, test loss-0.6166, acc-0.8383
Iter-29120, train loss-0.6683, acc-0.7600, valid loss-0.6133, acc-0.8400, test loss-0.6164, acc-0.8384
Iter-29130, train loss-0.6730, acc-0.8000, valid loss-0.6132, acc-0.8402, test loss-0.6163, acc-0.8383
Iter-29140, train loss-0.8066, acc-0.8000, valid loss-0.6131, acc-0.8402, test loss-0.6162, acc-0.8382
Iter-29150, train loss-0.5867, acc-0.8000, valid loss-0.6130, acc-0.8402, test loss-0.6161, acc-0.8383
Iter-29160, train loss-0.6277, acc-0.8600, valid loss-0.6129, acc-0.8404, test loss-0.6160, acc-0.8383
Iter-29170, train loss-0.5593, acc-0.9000, valid loss-0.6127, acc-0.8404, test loss-0.6158, acc-0.8382
Iter-29180, train loss-0.4847, acc-0.8600, valid loss-0.6126, acc-0.8404, test loss-0.6157, acc-0.8382
Iter-29190, train loss-0.7285, acc-0.8200, valid loss-0.6125, acc-0.8404, test loss-0.6156, acc-0.8384
Iter-29200, train loss-0.4238, acc-0.8800, valid loss-0.6124, acc-0.8404, test loss-0.6155, acc-0.8384
Iter-29210, train loss-0.6919, acc-0.7600, valid loss-0.6122, acc-0.8404, test loss-0.6153, acc-0.8384
Iter-29220, train loss-0.7242, acc-0.8000, valid loss-0.6121, acc-0.8406, test loss-0.6152, acc-0.8385
Iter-29230, train loss-0.5229, acc-0.9000, valid loss-0.6120, acc-0.8404, test loss-0.6151, acc-0.8386
Iter-29240, train loss-0.7568, acc-0.8000, valid loss-0.6119, acc-0.8406, test loss-0.6150, acc-0.8387
Iter-29250, train loss-0.7913, acc-0.8000, valid loss-0.6118, acc-0.8406, test loss-0.6148, acc-0.8385
Iter-29260, train loss-0.8826, acc-0.8000, valid loss-0.6116, acc-0.8406, test loss-0.6147, acc-0.8385
Iter-29270, train loss-0.3947, acc-0.9600, valid loss-0.6115, acc-0.8406, test loss-0.6146, acc-0.8385
Iter-29280, train loss-0.7591, acc-0.7600, valid loss-0.6114, acc-0.8408, test loss-0.6145, acc-0.8387
Iter-29290, train loss-0.6362, acc-0.8400, valid loss-0.6112, acc-0.8408, test loss-0.6143, acc-0.8386
Iter-29300, train loss-0.6117, acc-0.8200, valid loss-0.6111, acc-0.8408, test loss-0.6142, acc-0.8387
Iter-29310, train loss-0.6117, acc-0.7600, valid loss-0.6110, acc-0.8406, test loss-0.6141, acc-0.8388
Iter-29320, train loss-0.7426, acc-0.8400, valid loss-0.6109, acc-0.8408, test loss-0.6140, acc-0.8389
Iter-29330, train loss-0.4671, acc-0.8800, valid loss-0.6108, acc-0.8406, test loss-0.6138, acc-0.8389
Iter-29340, train loss-0.7158, acc-0.8000, valid loss-0.6106, acc-0.8408, test loss-0.6137, acc-0.8389
Iter-29350, train loss-0.5875, acc-0.8600, valid loss-0.6105, acc-0.8408, test loss-0.6136, acc-0.8389
Iter-29360, train loss-0.8022, acc-0.8200, valid loss-0.6104, acc-0.8406, test loss-0.6135, acc-0.8389
Iter-29370, train loss-0.6874, acc-0.7800, valid loss-0.6103, acc-0.8408, test loss-0.6134, acc-0.8389
Iter-29380, train loss-0.5313, acc-0.8800, valid loss-0.6101, acc-0.8406, test loss-0.6132, acc-0.8390
Iter-29390, train loss-0.6189, acc-0.8400, valid loss-0.6100, acc-0.8404, test loss-0.6131, acc-0.8389
Iter-29400, train loss-0.4711, acc-0.8800, valid loss-0.6099, acc-0.8404, test loss-0.6130, acc-0.8389
Iter-29410, train loss-0.6558, acc-0.8000, valid loss-0.6098, acc-0.8404, test loss-0.6128, acc-0.8389
Iter-29420, train loss-0.6475, acc-0.8800, valid loss-0.6097, acc-0.8404, test loss-0.6127, acc-0.8389
Iter-29430, train loss-0.5578, acc-0.8400, valid loss-0.6096, acc-0.8404, test loss-0.6126, acc-0.8389
Iter-29440, train loss-0.6427, acc-0.8400, valid loss-0.6094, acc-0.8406, test loss-0.6125, acc-0.8389
Iter-29450, train loss-0.6160, acc-0.8800, valid loss-0.6093, acc-0.8408, test loss-0.6124, acc-0.8389
Iter-29460, train loss-0.5182, acc-0.9200, valid loss-0.6092, acc-0.8408, test loss-0.6123, acc-0.8389
Iter-29470, train loss-0.4723, acc-0.9400, valid loss-0.6091, acc-0.8408, test loss-0.6121, acc-0.8389
Iter-29480, train loss-0.6085, acc-0.8400, valid loss-0.6090, acc-0.8410, test loss-0.6120, acc-0.8388
Iter-29490, train loss-0.5553, acc-0.8000, valid loss-0.6088, acc-0.8412, test loss-0.6119, acc-0.8388
Iter-29500, train loss-0.6395, acc-0.8400, valid loss-0.6087, acc-0.8412, test loss-0.6118, acc-0.8388
Iter-29510, train loss-0.4627, acc-0.9200, valid loss-0.6086, acc-0.8412, test loss-0.6117, acc-0.8388
Iter-29520, train loss-0.5528, acc-0.9000, valid loss-0.6085, acc-0.8416, test loss-0.6115, acc-0.8390
Iter-29530, train loss-0.5978, acc-0.8200, valid loss-0.6083, acc-0.8414, test loss-0.6114, acc-0.8390
Iter-29540, train loss-0.7611, acc-0.7600, valid loss-0.6082, acc-0.8416, test loss-0.6113, acc-0.8391
Iter-29550, train loss-0.6045, acc-0.8800, valid loss-0.6081, acc-0.8416, test loss-0.6112, acc-0.8392
Iter-29560, train loss-0.6499, acc-0.8200, valid loss-0.6080, acc-0.8416, test loss-0.6110, acc-0.8390
Iter-29570, train loss-0.5373, acc-0.8400, valid loss-0.6078, acc-0.8416, test loss-0.6109, acc-0.8392
Iter-29580, train loss-0.6373, acc-0.8000, valid loss-0.6077, acc-0.8414, test loss-0.6108, acc-0.8392
Iter-29590, train loss-0.6070, acc-0.8200, valid loss-0.6076, acc-0.8412, test loss-0.6106, acc-0.8392
Iter-29600, train loss-0.6915, acc-0.7800, valid loss-0.6075, acc-0.8414, test loss-0.6105, acc-0.8392
Iter-29610, train loss-0.6809, acc-0.7400, valid loss-0.6074, acc-0.8416, test loss-0.6104, acc-0.8394
Iter-29620, train loss-0.4925, acc-0.9000, valid loss-0.6072, acc-0.8412, test loss-0.6103, acc-0.8393
Iter-29630, train loss-0.7701, acc-0.8000, valid loss-0.6071, acc-0.8412, test loss-0.6101, acc-0.8394
Iter-29640, train loss-0.5897, acc-0.8600, valid loss-0.6070, acc-0.8414, test loss-0.6100, acc-0.8393
Iter-29650, train loss-0.6376, acc-0.7800, valid loss-0.6069, acc-0.8414, test loss-0.6099, acc-0.8395
Iter-29660, train loss-0.7623, acc-0.7800, valid loss-0.6068, acc-0.8412, test loss-0.6098, acc-0.8395
Iter-29670, train loss-0.5966, acc-0.8200, valid loss-0.6067, acc-0.8412, test loss-0.6097, acc-0.8394
Iter-29680, train loss-0.5470, acc-0.8200, valid loss-0.6065, acc-0.8412, test loss-0.6096, acc-0.8396
Iter-29690, train loss-0.7146, acc-0.7800, valid loss-0.6065, acc-0.8410, test loss-0.6094, acc-0.8397
Iter-29700, train loss-0.4772, acc-0.9000, valid loss-0.6063, acc-0.8410, test loss-0.6093, acc-0.8397
Iter-29710, train loss-0.5834, acc-0.8400, valid loss-0.6062, acc-0.8410, test loss-0.6092, acc-0.8397
Iter-29720, train loss-0.6011, acc-0.8400, valid loss-0.6061, acc-0.8410, test loss-0.6090, acc-0.8396
Iter-29730, train loss-0.5364, acc-0.8200, valid loss-0.6059, acc-0.8410, test loss-0.6089, acc-0.8397
Iter-29740, train loss-0.6182, acc-0.8600, valid loss-0.6058, acc-0.8410, test loss-0.6088, acc-0.8397
Iter-29750, train loss-0.5779, acc-0.8400, valid loss-0.6057, acc-0.8412, test loss-0.6087, acc-0.8397
Iter-29760, train loss-0.6330, acc-0.8400, valid loss-0.6055, acc-0.8412, test loss-0.6086, acc-0.8397
Iter-29770, train loss-0.7105, acc-0.7600, valid loss-0.6054, acc-0.8414, test loss-0.6084, acc-0.8398
Iter-29780, train loss-0.6688, acc-0.8000, valid loss-0.6053, acc-0.8412, test loss-0.6083, acc-0.8399
Iter-29790, train loss-0.6832, acc-0.8400, valid loss-0.6052, acc-0.8412, test loss-0.6082, acc-0.8399
Iter-29800, train loss-0.8303, acc-0.7400, valid loss-0.6051, acc-0.8412, test loss-0.6081, acc-0.8399
Iter-29810, train loss-0.6073, acc-0.7800, valid loss-0.6050, acc-0.8414, test loss-0.6080, acc-0.8398
Iter-29820, train loss-0.5417, acc-0.7800, valid loss-0.6048, acc-0.8414, test loss-0.6079, acc-0.8399
Iter-29830, train loss-0.4040, acc-0.9800, valid loss-0.6047, acc-0.8414, test loss-0.6078, acc-0.8400
Iter-29840, train loss-0.6851, acc-0.8400, valid loss-0.6046, acc-0.8412, test loss-0.6076, acc-0.8400
Iter-29850, train loss-0.5490, acc-0.8400, valid loss-0.6045, acc-0.8414, test loss-0.6075, acc-0.8401
Iter-29860, train loss-0.5755, acc-0.8600, valid loss-0.6044, acc-0.8414, test loss-0.6074, acc-0.8401
Iter-29870, train loss-0.5580, acc-0.8600, valid loss-0.6043, acc-0.8414, test loss-0.6073, acc-0.8400
Iter-29880, train loss-0.6624, acc-0.8200, valid loss-0.6042, acc-0.8416, test loss-0.6072, acc-0.8400
Iter-29890, train loss-0.6254, acc-0.8400, valid loss-0.6041, acc-0.8416, test loss-0.6071, acc-0.8400
Iter-29900, train loss-0.5894, acc-0.8200, valid loss-0.6040, acc-0.8418, test loss-0.6070, acc-0.8400
Iter-29910, train loss-0.5771, acc-0.8600, valid loss-0.6039, acc-0.8418, test loss-0.6069, acc-0.8401
Iter-29920, train loss-0.6766, acc-0.8200, valid loss-0.6037, acc-0.8418, test loss-0.6068, acc-0.8402
Iter-29930, train loss-0.5546, acc-0.8800, valid loss-0.6036, acc-0.8418, test loss-0.6066, acc-0.8402
Iter-29940, train loss-0.7017, acc-0.8600, valid loss-0.6035, acc-0.8418, test loss-0.6065, acc-0.8401
Iter-29950, train loss-0.6967, acc-0.7600, valid loss-0.6034, acc-0.8416, test loss-0.6064, acc-0.8401
Iter-29960, train loss-0.6806, acc-0.7600, valid loss-0.6033, acc-0.8414, test loss-0.6062, acc-0.8402
Iter-29970, train loss-0.7445, acc-0.7400, valid loss-0.6031, acc-0.8416, test loss-0.6061, acc-0.8401
Iter-29980, train loss-0.5913, acc-0.8200, valid loss-0.6030, acc-0.8414, test loss-0.6060, acc-0.8401
Iter-29990, train loss-0.3528, acc-0.9400, valid loss-0.6029, acc-0.8414, test loss-0.6059, acc-0.8400
Iter-30000, train loss-0.5102, acc-0.8800, valid loss-0.6028, acc-0.8414, test loss-0.6058, acc-0.8402
Iter-30010, train loss-0.7983, acc-0.7400, valid loss-0.6027, acc-0.8414, test loss-0.6056, acc-0.8401
Iter-30020, train loss-0.6607, acc-0.8000, valid loss-0.6025, acc-0.8416, test loss-0.6055, acc-0.8402
Iter-30030, train loss-0.5612, acc-0.8800, valid loss-0.6024, acc-0.8416, test loss-0.6054, acc-0.8403
Iter-30040, train loss-0.7794, acc-0.7800, valid loss-0.6023, acc-0.8414, test loss-0.6053, acc-0.8403
Iter-30050, train loss-0.7443, acc-0.8200, valid loss-0.6022, acc-0.8412, test loss-0.6052, acc-0.8402
Iter-30060, train loss-0.5881, acc-0.8600, valid loss-0.6021, acc-0.8416, test loss-0.6050, acc-0.8402
Iter-30070, train loss-0.6524, acc-0.8200, valid loss-0.6019, acc-0.8414, test loss-0.6049, acc-0.8401
Iter-30080, train loss-0.5382, acc-0.7800, valid loss-0.6018, acc-0.8416, test loss-0.6048, acc-0.8402
Iter-30090, train loss-0.7401, acc-0.8000, valid loss-0.6017, acc-0.8416, test loss-0.6047, acc-0.8403
Iter-30100, train loss-0.5806, acc-0.8400, valid loss-0.6016, acc-0.8416, test loss-0.6045, acc-0.8403
Iter-30110, train loss-0.5007, acc-0.8600, valid loss-0.6015, acc-0.8416, test loss-0.6044, acc-0.8403
Iter-30120, train loss-0.5973, acc-0.8600, valid loss-0.6013, acc-0.8420, test loss-0.6043, acc-0.8403
Iter-30130, train loss-0.8304, acc-0.7200, valid loss-0.6012, acc-0.8420, test loss-0.6041, acc-0.8404
Iter-30140, train loss-0.6891, acc-0.8400, valid loss-0.6011, acc-0.8420, test loss-0.6040, acc-0.8405
Iter-30150, train loss-0.6578, acc-0.7800, valid loss-0.6010, acc-0.8420, test loss-0.6039, acc-0.8404
Iter-30160, train loss-0.7022, acc-0.7800, valid loss-0.6009, acc-0.8420, test loss-0.6038, acc-0.8405
Iter-30170, train loss-0.5681, acc-0.8600, valid loss-0.6007, acc-0.8420, test loss-0.6037, acc-0.8405
Iter-30180, train loss-0.6532, acc-0.8000, valid loss-0.6006, acc-0.8418, test loss-0.6035, acc-0.8405
Iter-30190, train loss-0.5389, acc-0.8800, valid loss-0.6005, acc-0.8422, test loss-0.6034, acc-0.8405
Iter-30200, train loss-0.7182, acc-0.8600, valid loss-0.6004, acc-0.8422, test loss-0.6033, acc-0.8405
Iter-30210, train loss-0.5987, acc-0.8400, valid loss-0.6003, acc-0.8422, test loss-0.6032, acc-0.8405
Iter-30220, train loss-0.5674, acc-0.8000, valid loss-0.6001, acc-0.8420, test loss-0.6031, acc-0.8405
Iter-30230, train loss-0.6852, acc-0.8000, valid loss-0.6000, acc-0.8420, test loss-0.6030, acc-0.8406
Iter-30240, train loss-0.7472, acc-0.8200, valid loss-0.5999, acc-0.8422, test loss-0.6028, acc-0.8405
Iter-30250, train loss-0.6057, acc-0.8400, valid loss-0.5998, acc-0.8424, test loss-0.6028, acc-0.8406
Iter-30260, train loss-0.5545, acc-0.8200, valid loss-0.5997, acc-0.8426, test loss-0.6026, acc-0.8406
Iter-30270, train loss-0.5708, acc-0.8400, valid loss-0.5996, acc-0.8426, test loss-0.6025, acc-0.8406
Iter-30280, train loss-0.5096, acc-0.8600, valid loss-0.5995, acc-0.8428, test loss-0.6024, acc-0.8407
Iter-30290, train loss-0.6677, acc-0.8200, valid loss-0.5994, acc-0.8428, test loss-0.6023, acc-0.8406
Iter-30300, train loss-0.5083, acc-0.9000, valid loss-0.5992, acc-0.8426, test loss-0.6022, acc-0.8409
Iter-30310, train loss-0.9423, acc-0.7000, valid loss-0.5991, acc-0.8424, test loss-0.6020, acc-0.8407
Iter-30320, train loss-0.8170, acc-0.7600, valid loss-0.5990, acc-0.8424, test loss-0.6019, acc-0.8409
Iter-30330, train loss-0.5598, acc-0.8800, valid loss-0.5989, acc-0.8424, test loss-0.6018, acc-0.8409
Iter-30340, train loss-0.7477, acc-0.7800, valid loss-0.5988, acc-0.8426, test loss-0.6017, acc-0.8409
Iter-30350, train loss-0.7378, acc-0.7600, valid loss-0.5987, acc-0.8426, test loss-0.6016, acc-0.8409
Iter-30360, train loss-0.5445, acc-0.9000, valid loss-0.5985, acc-0.8428, test loss-0.6015, acc-0.8409
Iter-30370, train loss-0.5529, acc-0.8400, valid loss-0.5984, acc-0.8426, test loss-0.6013, acc-0.8407
Iter-30380, train loss-0.6161, acc-0.8400, valid loss-0.5983, acc-0.8430, test loss-0.6012, acc-0.8408
Iter-30390, train loss-0.5701, acc-0.8800, valid loss-0.5982, acc-0.8428, test loss-0.6011, acc-0.8407
Iter-30400, train loss-0.7095, acc-0.7000, valid loss-0.5981, acc-0.8428, test loss-0.6010, acc-0.8406
Iter-30410, train loss-0.6522, acc-0.8000, valid loss-0.5979, acc-0.8430, test loss-0.6009, acc-0.8409
Iter-30420, train loss-0.7340, acc-0.8600, valid loss-0.5978, acc-0.8432, test loss-0.6008, acc-0.8408
Iter-30430, train loss-0.5278, acc-0.9400, valid loss-0.5977, acc-0.8430, test loss-0.6006, acc-0.8409
Iter-30440, train loss-0.5891, acc-0.8600, valid loss-0.5976, acc-0.8432, test loss-0.6005, acc-0.8407
Iter-30450, train loss-0.6403, acc-0.8400, valid loss-0.5975, acc-0.8432, test loss-0.6004, acc-0.8408
Iter-30460, train loss-0.7136, acc-0.8200, valid loss-0.5973, acc-0.8432, test loss-0.6003, acc-0.8408
Iter-30470, train loss-0.8654, acc-0.7800, valid loss-0.5972, acc-0.8432, test loss-0.6001, acc-0.8408
Iter-30480, train loss-0.6765, acc-0.7800, valid loss-0.5971, acc-0.8432, test loss-0.6000, acc-0.8408
Iter-30490, train loss-0.9210, acc-0.7800, valid loss-0.5970, acc-0.8430, test loss-0.5999, acc-0.8407
Iter-30500, train loss-0.4792, acc-0.9400, valid loss-0.5969, acc-0.8430, test loss-0.5998, acc-0.8410
Iter-30510, train loss-0.6566, acc-0.8400, valid loss-0.5968, acc-0.8436, test loss-0.5997, acc-0.8410
Iter-30520, train loss-0.7262, acc-0.7800, valid loss-0.5967, acc-0.8432, test loss-0.5996, acc-0.8413
Iter-30530, train loss-0.5018, acc-0.8800, valid loss-0.5965, acc-0.8430, test loss-0.5994, acc-0.8413
Iter-30540, train loss-0.5604, acc-0.8600, valid loss-0.5964, acc-0.8434, test loss-0.5994, acc-0.8414
Iter-30550, train loss-0.5052, acc-0.8600, valid loss-0.5963, acc-0.8434, test loss-0.5992, acc-0.8417
Iter-30560, train loss-0.6054, acc-0.8600, valid loss-0.5962, acc-0.8434, test loss-0.5991, acc-0.8415
Iter-30570, train loss-0.8094, acc-0.8000, valid loss-0.5961, acc-0.8434, test loss-0.5990, acc-0.8416
Iter-30580, train loss-0.6881, acc-0.7800, valid loss-0.5960, acc-0.8434, test loss-0.5989, acc-0.8418
Iter-30590, train loss-0.3766, acc-0.9400, valid loss-0.5959, acc-0.8434, test loss-0.5988, acc-0.8418
Iter-30600, train loss-0.5092, acc-0.8200, valid loss-0.5958, acc-0.8438, test loss-0.5987, acc-0.8418
Iter-30610, train loss-0.6977, acc-0.7600, valid loss-0.5957, acc-0.8436, test loss-0.5986, acc-0.8421
Iter-30620, train loss-0.5935, acc-0.8000, valid loss-0.5956, acc-0.8436, test loss-0.5985, acc-0.8419
Iter-30630, train loss-0.4923, acc-0.9000, valid loss-0.5955, acc-0.8436, test loss-0.5984, acc-0.8421
Iter-30640, train loss-0.5660, acc-0.8400, valid loss-0.5953, acc-0.8438, test loss-0.5983, acc-0.8421
Iter-30650, train loss-0.6340, acc-0.8000, valid loss-0.5952, acc-0.8440, test loss-0.5981, acc-0.8420
Iter-30660, train loss-0.6901, acc-0.8200, valid loss-0.5951, acc-0.8440, test loss-0.5980, acc-0.8421
Iter-30670, train loss-0.7850, acc-0.7800, valid loss-0.5950, acc-0.8438, test loss-0.5979, acc-0.8421
Iter-30680, train loss-0.5569, acc-0.9000, valid loss-0.5949, acc-0.8440, test loss-0.5978, acc-0.8420
Iter-30690, train loss-0.7043, acc-0.7600, valid loss-0.5948, acc-0.8438, test loss-0.5977, acc-0.8420
Iter-30700, train loss-0.7566, acc-0.7000, valid loss-0.5947, acc-0.8438, test loss-0.5976, acc-0.8421
Iter-30710, train loss-0.7300, acc-0.7800, valid loss-0.5946, acc-0.8442, test loss-0.5975, acc-0.8422
Iter-30720, train loss-0.6300, acc-0.8800, valid loss-0.5944, acc-0.8440, test loss-0.5974, acc-0.8422
Iter-30730, train loss-0.7889, acc-0.7000, valid loss-0.5943, acc-0.8438, test loss-0.5973, acc-0.8422
Iter-30740, train loss-0.4131, acc-0.9000, valid loss-0.5942, acc-0.8438, test loss-0.5971, acc-0.8423
Iter-30750, train loss-0.6839, acc-0.8200, valid loss-0.5941, acc-0.8442, test loss-0.5970, acc-0.8422
Iter-30760, train loss-0.5904, acc-0.8800, valid loss-0.5939, acc-0.8440, test loss-0.5969, acc-0.8424
Iter-30770, train loss-0.8002, acc-0.7600, valid loss-0.5938, acc-0.8438, test loss-0.5968, acc-0.8425
Iter-30780, train loss-0.5841, acc-0.8200, valid loss-0.5937, acc-0.8442, test loss-0.5967, acc-0.8425
Iter-30790, train loss-0.5291, acc-0.8800, valid loss-0.5936, acc-0.8442, test loss-0.5966, acc-0.8427
Iter-30800, train loss-0.7469, acc-0.7400, valid loss-0.5935, acc-0.8444, test loss-0.5965, acc-0.8426
Iter-30810, train loss-0.4903, acc-0.8800, valid loss-0.5934, acc-0.8446, test loss-0.5964, acc-0.8426
Iter-30820, train loss-0.7499, acc-0.8600, valid loss-0.5933, acc-0.8444, test loss-0.5962, acc-0.8424
Iter-30830, train loss-0.7171, acc-0.8600, valid loss-0.5932, acc-0.8444, test loss-0.5961, acc-0.8424
Iter-30840, train loss-0.6614, acc-0.8400, valid loss-0.5931, acc-0.8444, test loss-0.5960, acc-0.8421
Iter-30850, train loss-0.6068, acc-0.8600, valid loss-0.5929, acc-0.8444, test loss-0.5959, acc-0.8424
Iter-30860, train loss-0.5824, acc-0.8400, valid loss-0.5928, acc-0.8444, test loss-0.5958, acc-0.8426
Iter-30870, train loss-0.5527, acc-0.8600, valid loss-0.5927, acc-0.8446, test loss-0.5957, acc-0.8425
Iter-30880, train loss-0.9596, acc-0.7400, valid loss-0.5926, acc-0.8446, test loss-0.5956, acc-0.8423
Iter-30890, train loss-0.7545, acc-0.8000, valid loss-0.5925, acc-0.8446, test loss-0.5955, acc-0.8424
Iter-30900, train loss-0.5091, acc-0.9200, valid loss-0.5924, acc-0.8446, test loss-0.5954, acc-0.8425
Iter-30910, train loss-0.5075, acc-0.8600, valid loss-0.5923, acc-0.8448, test loss-0.5952, acc-0.8426
Iter-30920, train loss-0.6756, acc-0.8000, valid loss-0.5922, acc-0.8448, test loss-0.5951, acc-0.8425
Iter-30930, train loss-0.6899, acc-0.8200, valid loss-0.5921, acc-0.8448, test loss-0.5950, acc-0.8425
Iter-30940, train loss-0.5125, acc-0.8800, valid loss-0.5920, acc-0.8448, test loss-0.5949, acc-0.8426
Iter-30950, train loss-0.5866, acc-0.8600, valid loss-0.5919, acc-0.8450, test loss-0.5948, acc-0.8425
Iter-30960, train loss-0.4008, acc-0.9000, valid loss-0.5918, acc-0.8448, test loss-0.5947, acc-0.8425
Iter-30970, train loss-0.5357, acc-0.8800, valid loss-0.5917, acc-0.8448, test loss-0.5946, acc-0.8426
Iter-30980, train loss-0.6661, acc-0.7800, valid loss-0.5916, acc-0.8446, test loss-0.5945, acc-0.8426
Iter-30990, train loss-0.6497, acc-0.8000, valid loss-0.5915, acc-0.8444, test loss-0.5944, acc-0.8426
Iter-31000, train loss-0.5946, acc-0.9000, valid loss-0.5914, acc-0.8448, test loss-0.5943, acc-0.8427
Iter-31010, train loss-0.7554, acc-0.7800, valid loss-0.5912, acc-0.8450, test loss-0.5941, acc-0.8427
Iter-31020, train loss-0.9663, acc-0.7400, valid loss-0.5911, acc-0.8448, test loss-0.5940, acc-0.8427
Iter-31030, train loss-0.6270, acc-0.8200, valid loss-0.5910, acc-0.8446, test loss-0.5939, acc-0.8427
Iter-31040, train loss-0.6296, acc-0.8000, valid loss-0.5909, acc-0.8446, test loss-0.5938, acc-0.8428
Iter-31050, train loss-0.7742, acc-0.7000, valid loss-0.5908, acc-0.8446, test loss-0.5937, acc-0.8429
Iter-31060, train loss-0.6339, acc-0.8200, valid loss-0.5907, acc-0.8450, test loss-0.5936, acc-0.8430
Iter-31070, train loss-0.9084, acc-0.6800, valid loss-0.5906, acc-0.8450, test loss-0.5935, acc-0.8430
Iter-31080, train loss-0.6050, acc-0.8200, valid loss-0.5905, acc-0.8450, test loss-0.5934, acc-0.8430
Iter-31090, train loss-0.5721, acc-0.9000, valid loss-0.5903, acc-0.8452, test loss-0.5933, acc-0.8430
Iter-31100, train loss-0.5550, acc-0.9000, valid loss-0.5902, acc-0.8450, test loss-0.5932, acc-0.8430
Iter-31110, train loss-0.5792, acc-0.8600, valid loss-0.5901, acc-0.8450, test loss-0.5931, acc-0.8431
Iter-31120, train loss-0.4999, acc-0.8400, valid loss-0.5900, acc-0.8450, test loss-0.5930, acc-0.8431
Iter-31130, train loss-0.6351, acc-0.8800, valid loss-0.5899, acc-0.8452, test loss-0.5929, acc-0.8432
Iter-31140, train loss-0.7266, acc-0.8000, valid loss-0.5898, acc-0.8452, test loss-0.5928, acc-0.8431
Iter-31150, train loss-0.8233, acc-0.8000, valid loss-0.5897, acc-0.8452, test loss-0.5926, acc-0.8432
Iter-31160, train loss-0.7692, acc-0.7600, valid loss-0.5896, acc-0.8452, test loss-0.5925, acc-0.8431
Iter-31170, train loss-0.6333, acc-0.8400, valid loss-0.5895, acc-0.8454, test loss-0.5924, acc-0.8432
Iter-31180, train loss-0.4952, acc-0.9200, valid loss-0.5894, acc-0.8454, test loss-0.5923, acc-0.8432
Iter-31190, train loss-0.6942, acc-0.8200, valid loss-0.5893, acc-0.8454, test loss-0.5922, acc-0.8433
Iter-31200, train loss-0.5387, acc-0.9200, valid loss-0.5892, acc-0.8454, test loss-0.5921, acc-0.8433
Iter-31210, train loss-0.5714, acc-0.8200, valid loss-0.5891, acc-0.8452, test loss-0.5920, acc-0.8434
Iter-31220, train loss-0.4790, acc-0.9000, valid loss-0.5889, acc-0.8452, test loss-0.5919, acc-0.8434
Iter-31230, train loss-0.7566, acc-0.7800, valid loss-0.5888, acc-0.8452, test loss-0.5918, acc-0.8434
Iter-31240, train loss-0.5100, acc-0.8200, valid loss-0.5887, acc-0.8454, test loss-0.5917, acc-0.8434
Iter-31250, train loss-0.5170, acc-0.8800, valid loss-0.5886, acc-0.8454, test loss-0.5916, acc-0.8435
Iter-31260, train loss-0.7257, acc-0.7600, valid loss-0.5885, acc-0.8458, test loss-0.5915, acc-0.8436
Iter-31270, train loss-0.6817, acc-0.7800, valid loss-0.5884, acc-0.8456, test loss-0.5913, acc-0.8435
Iter-31280, train loss-0.7014, acc-0.8000, valid loss-0.5883, acc-0.8456, test loss-0.5912, acc-0.8434
Iter-31290, train loss-0.5009, acc-0.8600, valid loss-0.5882, acc-0.8456, test loss-0.5911, acc-0.8435
Iter-31300, train loss-0.4314, acc-0.8800, valid loss-0.5881, acc-0.8454, test loss-0.5910, acc-0.8436
Iter-31310, train loss-0.5266, acc-0.9000, valid loss-0.5880, acc-0.8456, test loss-0.5909, acc-0.8434
Iter-31320, train loss-0.6142, acc-0.8400, valid loss-0.5878, acc-0.8456, test loss-0.5908, acc-0.8432
Iter-31330, train loss-0.7293, acc-0.8400, valid loss-0.5877, acc-0.8458, test loss-0.5907, acc-0.8434
Iter-31340, train loss-0.7048, acc-0.8000, valid loss-0.5876, acc-0.8458, test loss-0.5905, acc-0.8434
Iter-31350, train loss-0.9613, acc-0.6800, valid loss-0.5875, acc-0.8456, test loss-0.5904, acc-0.8434
Iter-31360, train loss-0.6612, acc-0.8200, valid loss-0.5874, acc-0.8456, test loss-0.5903, acc-0.8434
Iter-31370, train loss-0.6243, acc-0.8600, valid loss-0.5873, acc-0.8458, test loss-0.5902, acc-0.8434
Iter-31380, train loss-0.4638, acc-0.9400, valid loss-0.5872, acc-0.8458, test loss-0.5901, acc-0.8436
Iter-31390, train loss-0.6560, acc-0.9000, valid loss-0.5870, acc-0.8456, test loss-0.5900, acc-0.8437
Iter-31400, train loss-0.4056, acc-0.9200, valid loss-0.5869, acc-0.8456, test loss-0.5899, acc-0.8437
Iter-31410, train loss-0.6755, acc-0.7400, valid loss-0.5868, acc-0.8458, test loss-0.5898, acc-0.8438
Iter-31420, train loss-0.5505, acc-0.8400, valid loss-0.5867, acc-0.8458, test loss-0.5896, acc-0.8438
Iter-31430, train loss-0.5202, acc-0.8800, valid loss-0.5866, acc-0.8460, test loss-0.5895, acc-0.8439
Iter-31440, train loss-0.4679, acc-0.9200, valid loss-0.5865, acc-0.8460, test loss-0.5894, acc-0.8438
Iter-31450, train loss-0.5273, acc-0.8600, valid loss-0.5864, acc-0.8460, test loss-0.5893, acc-0.8438
Iter-31460, train loss-0.7243, acc-0.7800, valid loss-0.5863, acc-0.8460, test loss-0.5892, acc-0.8438
Iter-31470, train loss-0.7487, acc-0.8000, valid loss-0.5862, acc-0.8460, test loss-0.5891, acc-0.8438
Iter-31480, train loss-0.5216, acc-0.8400, valid loss-0.5861, acc-0.8460, test loss-0.5890, acc-0.8437
Iter-31490, train loss-0.5595, acc-0.8200, valid loss-0.5860, acc-0.8460, test loss-0.5889, acc-0.8437
Iter-31500, train loss-0.5617, acc-0.8600, valid loss-0.5858, acc-0.8460, test loss-0.5888, acc-0.8437
Iter-31510, train loss-0.6765, acc-0.8200, valid loss-0.5857, acc-0.8460, test loss-0.5887, acc-0.8437
Iter-31520, train loss-0.5361, acc-0.8600, valid loss-0.5856, acc-0.8460, test loss-0.5886, acc-0.8437
Iter-31530, train loss-0.5458, acc-0.9000, valid loss-0.5855, acc-0.8460, test loss-0.5885, acc-0.8437
Iter-31540, train loss-0.7964, acc-0.8200, valid loss-0.5854, acc-0.8460, test loss-0.5884, acc-0.8439
Iter-31550, train loss-0.5730, acc-0.8200, valid loss-0.5853, acc-0.8460, test loss-0.5883, acc-0.8439
Iter-31560, train loss-0.6080, acc-0.8200, valid loss-0.5852, acc-0.8460, test loss-0.5882, acc-0.8442
Iter-31570, train loss-0.4574, acc-0.8800, valid loss-0.5851, acc-0.8460, test loss-0.5880, acc-0.8440
Iter-31580, train loss-0.4158, acc-0.9400, valid loss-0.5850, acc-0.8460, test loss-0.5879, acc-0.8439
Iter-31590, train loss-0.5280, acc-0.8200, valid loss-0.5849, acc-0.8460, test loss-0.5878, acc-0.8439
Iter-31600, train loss-0.6301, acc-0.8000, valid loss-0.5848, acc-0.8460, test loss-0.5877, acc-0.8441
Iter-31610, train loss-0.5977, acc-0.8800, valid loss-0.5847, acc-0.8460, test loss-0.5876, acc-0.8443
Iter-31620, train loss-0.5467, acc-0.8800, valid loss-0.5846, acc-0.8460, test loss-0.5875, acc-0.8443
Iter-31630, train loss-0.7171, acc-0.8200, valid loss-0.5845, acc-0.8460, test loss-0.5874, acc-0.8443
Iter-31640, train loss-0.5946, acc-0.9000, valid loss-0.5844, acc-0.8460, test loss-0.5873, acc-0.8441
Iter-31650, train loss-0.4591, acc-0.9000, valid loss-0.5842, acc-0.8460, test loss-0.5872, acc-0.8441
Iter-31660, train loss-0.6315, acc-0.8600, valid loss-0.5841, acc-0.8462, test loss-0.5871, acc-0.8443
Iter-31670, train loss-0.6654, acc-0.8000, valid loss-0.5840, acc-0.8462, test loss-0.5870, acc-0.8445
Iter-31680, train loss-0.7227, acc-0.8000, valid loss-0.5839, acc-0.8464, test loss-0.5869, acc-0.8444
Iter-31690, train loss-0.6271, acc-0.8600, valid loss-0.5838, acc-0.8462, test loss-0.5868, acc-0.8443
Iter-31700, train loss-0.7175, acc-0.7800, valid loss-0.5837, acc-0.8464, test loss-0.5867, acc-0.8442
Iter-31710, train loss-0.6985, acc-0.8000, valid loss-0.5836, acc-0.8464, test loss-0.5866, acc-0.8444
Iter-31720, train loss-0.5163, acc-0.8800, valid loss-0.5835, acc-0.8464, test loss-0.5865, acc-0.8445
Iter-31730, train loss-0.4896, acc-0.9000, valid loss-0.5834, acc-0.8464, test loss-0.5864, acc-0.8443
Iter-31740, train loss-0.4927, acc-0.8600, valid loss-0.5833, acc-0.8464, test loss-0.5863, acc-0.8446
Iter-31750, train loss-0.5996, acc-0.8600, valid loss-0.5832, acc-0.8464, test loss-0.5862, acc-0.8446
Iter-31760, train loss-0.7053, acc-0.8200, valid loss-0.5831, acc-0.8464, test loss-0.5861, acc-0.8447
Iter-31770, train loss-0.6960, acc-0.8200, valid loss-0.5829, acc-0.8464, test loss-0.5859, acc-0.8446
Iter-31780, train loss-0.8094, acc-0.7800, valid loss-0.5828, acc-0.8464, test loss-0.5858, acc-0.8447
Iter-31790, train loss-0.7926, acc-0.7000, valid loss-0.5827, acc-0.8466, test loss-0.5857, acc-0.8446
Iter-31800, train loss-0.5895, acc-0.8400, valid loss-0.5826, acc-0.8468, test loss-0.5856, acc-0.8446
Iter-31810, train loss-0.4973, acc-0.8800, valid loss-0.5825, acc-0.8468, test loss-0.5855, acc-0.8447
Iter-31820, train loss-0.5319, acc-0.8600, valid loss-0.5824, acc-0.8468, test loss-0.5854, acc-0.8447
Iter-31830, train loss-0.4447, acc-0.9400, valid loss-0.5823, acc-0.8468, test loss-0.5853, acc-0.8449
Iter-31840, train loss-0.5850, acc-0.8200, valid loss-0.5822, acc-0.8468, test loss-0.5852, acc-0.8448
Iter-31850, train loss-0.4840, acc-0.9000, valid loss-0.5821, acc-0.8466, test loss-0.5851, acc-0.8448
Iter-31860, train loss-0.4241, acc-0.9000, valid loss-0.5820, acc-0.8466, test loss-0.5850, acc-0.8447
Iter-31870, train loss-0.6158, acc-0.8600, valid loss-0.5818, acc-0.8468, test loss-0.5849, acc-0.8448
Iter-31880, train loss-0.7297, acc-0.7600, valid loss-0.5817, acc-0.8468, test loss-0.5847, acc-0.8449
Iter-31890, train loss-0.7770, acc-0.8000, valid loss-0.5816, acc-0.8468, test loss-0.5847, acc-0.8449
Iter-31900, train loss-0.6064, acc-0.8400, valid loss-0.5815, acc-0.8470, test loss-0.5846, acc-0.8447
Iter-31910, train loss-0.6305, acc-0.8400, valid loss-0.5814, acc-0.8472, test loss-0.5844, acc-0.8444
Iter-31920, train loss-0.7119, acc-0.7600, valid loss-0.5814, acc-0.8474, test loss-0.5843, acc-0.8446
Iter-31930, train loss-0.8195, acc-0.8400, valid loss-0.5812, acc-0.8476, test loss-0.5842, acc-0.8445
Iter-31940, train loss-0.4677, acc-0.9200, valid loss-0.5811, acc-0.8476, test loss-0.5841, acc-0.8446
Iter-31950, train loss-0.6807, acc-0.7800, valid loss-0.5810, acc-0.8476, test loss-0.5840, acc-0.8447
Iter-31960, train loss-0.6884, acc-0.7600, valid loss-0.5809, acc-0.8476, test loss-0.5839, acc-0.8447
Iter-31970, train loss-0.7038, acc-0.8600, valid loss-0.5808, acc-0.8476, test loss-0.5838, acc-0.8445
Iter-31980, train loss-0.6007, acc-0.8800, valid loss-0.5807, acc-0.8476, test loss-0.5837, acc-0.8448
Iter-31990, train loss-0.4948, acc-0.8400, valid loss-0.5806, acc-0.8476, test loss-0.5837, acc-0.8449
Iter-32000, train loss-0.4393, acc-0.8800, valid loss-0.5805, acc-0.8476, test loss-0.5836, acc-0.8449
Iter-32010, train loss-0.5058, acc-0.8800, valid loss-0.5804, acc-0.8476, test loss-0.5835, acc-0.8449
Iter-32020, train loss-0.4279, acc-0.9000, valid loss-0.5803, acc-0.8478, test loss-0.5833, acc-0.8449
Iter-32030, train loss-0.6224, acc-0.8400, valid loss-0.5802, acc-0.8478, test loss-0.5832, acc-0.8449
Iter-32040, train loss-0.7120, acc-0.8200, valid loss-0.5801, acc-0.8478, test loss-0.5831, acc-0.8449
Iter-32050, train loss-0.5835, acc-0.8200, valid loss-0.5800, acc-0.8478, test loss-0.5830, acc-0.8449
Iter-32060, train loss-0.5884, acc-0.8200, valid loss-0.5799, acc-0.8478, test loss-0.5829, acc-0.8449
Iter-32070, train loss-0.5343, acc-0.9000, valid loss-0.5798, acc-0.8480, test loss-0.5828, acc-0.8449
Iter-32080, train loss-0.6287, acc-0.8400, valid loss-0.5796, acc-0.8480, test loss-0.5827, acc-0.8452
Iter-32090, train loss-0.5998, acc-0.8200, valid loss-0.5795, acc-0.8482, test loss-0.5826, acc-0.8452
Iter-32100, train loss-0.5531, acc-0.8600, valid loss-0.5794, acc-0.8482, test loss-0.5825, acc-0.8452
Iter-32110, train loss-0.7266, acc-0.7400, valid loss-0.5793, acc-0.8484, test loss-0.5824, acc-0.8451
Iter-32120, train loss-0.6767, acc-0.7800, valid loss-0.5792, acc-0.8482, test loss-0.5823, acc-0.8454
Iter-32130, train loss-0.7709, acc-0.7600, valid loss-0.5791, acc-0.8482, test loss-0.5822, acc-0.8452
Iter-32140, train loss-0.8505, acc-0.7800, valid loss-0.5790, acc-0.8480, test loss-0.5821, acc-0.8453
Iter-32150, train loss-0.6128, acc-0.8400, valid loss-0.5789, acc-0.8484, test loss-0.5820, acc-0.8454
Iter-32160, train loss-0.5473, acc-0.8400, valid loss-0.5789, acc-0.8484, test loss-0.5819, acc-0.8452
Iter-32170, train loss-0.5301, acc-0.8800, valid loss-0.5788, acc-0.8482, test loss-0.5818, acc-0.8454
Iter-32180, train loss-0.5698, acc-0.8200, valid loss-0.5786, acc-0.8482, test loss-0.5817, acc-0.8456
Iter-32190, train loss-0.5344, acc-0.8800, valid loss-0.5785, acc-0.8486, test loss-0.5816, acc-0.8455
Iter-32200, train loss-0.6222, acc-0.8000, valid loss-0.5784, acc-0.8486, test loss-0.5815, acc-0.8456
Iter-32210, train loss-0.6747, acc-0.8000, valid loss-0.5783, acc-0.8486, test loss-0.5814, acc-0.8457
Iter-32220, train loss-0.5790, acc-0.8400, valid loss-0.5782, acc-0.8484, test loss-0.5813, acc-0.8457
Iter-32230, train loss-0.5583, acc-0.8400, valid loss-0.5781, acc-0.8486, test loss-0.5812, acc-0.8455
Iter-32240, train loss-0.5224, acc-0.8600, valid loss-0.5780, acc-0.8486, test loss-0.5811, acc-0.8458
Iter-32250, train loss-0.6313, acc-0.8800, valid loss-0.5779, acc-0.8488, test loss-0.5810, acc-0.8457
Iter-32260, train loss-0.6693, acc-0.8200, valid loss-0.5778, acc-0.8488, test loss-0.5809, acc-0.8457
Iter-32270, train loss-0.6650, acc-0.7800, valid loss-0.5777, acc-0.8486, test loss-0.5808, acc-0.8457
Iter-32280, train loss-0.4046, acc-0.9200, valid loss-0.5776, acc-0.8486, test loss-0.5806, acc-0.8458
Iter-32290, train loss-0.5393, acc-0.8200, valid loss-0.5775, acc-0.8486, test loss-0.5806, acc-0.8460
Iter-32300, train loss-0.8335, acc-0.7400, valid loss-0.5774, acc-0.8486, test loss-0.5805, acc-0.8460
Iter-32310, train loss-0.5498, acc-0.8200, valid loss-0.5773, acc-0.8486, test loss-0.5804, acc-0.8460
Iter-32320, train loss-0.7813, acc-0.7800, valid loss-0.5772, acc-0.8488, test loss-0.5803, acc-0.8460
Iter-32330, train loss-0.5572, acc-0.8400, valid loss-0.5771, acc-0.8492, test loss-0.5802, acc-0.8461
Iter-32340, train loss-0.6883, acc-0.7400, valid loss-0.5769, acc-0.8492, test loss-0.5801, acc-0.8460
Iter-32350, train loss-0.6878, acc-0.8400, valid loss-0.5768, acc-0.8492, test loss-0.5800, acc-0.8461
Iter-32360, train loss-0.3999, acc-0.9000, valid loss-0.5767, acc-0.8494, test loss-0.5798, acc-0.8460
Iter-32370, train loss-0.6574, acc-0.7600, valid loss-0.5767, acc-0.8494, test loss-0.5798, acc-0.8461
Iter-32380, train loss-0.6481, acc-0.7800, valid loss-0.5765, acc-0.8494, test loss-0.5797, acc-0.8459
Iter-32390, train loss-0.6041, acc-0.8400, valid loss-0.5764, acc-0.8494, test loss-0.5796, acc-0.8459
Iter-32400, train loss-0.5742, acc-0.8800, valid loss-0.5763, acc-0.8494, test loss-0.5795, acc-0.8459
Iter-32410, train loss-0.4343, acc-0.8600, valid loss-0.5762, acc-0.8494, test loss-0.5793, acc-0.8458
Iter-32420, train loss-0.5294, acc-0.9200, valid loss-0.5761, acc-0.8494, test loss-0.5792, acc-0.8460
Iter-32430, train loss-0.5673, acc-0.8200, valid loss-0.5760, acc-0.8496, test loss-0.5791, acc-0.8459
Iter-32440, train loss-0.6191, acc-0.8400, valid loss-0.5760, acc-0.8494, test loss-0.5791, acc-0.8460
Iter-32450, train loss-0.5855, acc-0.8400, valid loss-0.5759, acc-0.8492, test loss-0.5790, acc-0.8462
Iter-32460, train loss-0.5755, acc-0.8200, valid loss-0.5758, acc-0.8492, test loss-0.5789, acc-0.8464
Iter-32470, train loss-0.5089, acc-0.8800, valid loss-0.5756, acc-0.8492, test loss-0.5788, acc-0.8461
Iter-32480, train loss-0.5241, acc-0.9200, valid loss-0.5755, acc-0.8492, test loss-0.5787, acc-0.8462
Iter-32490, train loss-0.6229, acc-0.8200, valid loss-0.5754, acc-0.8494, test loss-0.5786, acc-0.8463
Iter-32500, train loss-0.4414, acc-0.9400, valid loss-0.5753, acc-0.8494, test loss-0.5784, acc-0.8463
Iter-32510, train loss-0.5329, acc-0.8400, valid loss-0.5753, acc-0.8494, test loss-0.5784, acc-0.8464
Iter-32520, train loss-0.4101, acc-0.8800, valid loss-0.5752, acc-0.8496, test loss-0.5783, acc-0.8464
Iter-32530, train loss-0.6858, acc-0.8400, valid loss-0.5751, acc-0.8492, test loss-0.5782, acc-0.8465
Iter-32540, train loss-0.7169, acc-0.7800, valid loss-0.5750, acc-0.8494, test loss-0.5781, acc-0.8465
Iter-32550, train loss-0.5785, acc-0.8200, valid loss-0.5749, acc-0.8496, test loss-0.5780, acc-0.8464
Iter-32560, train loss-0.4437, acc-0.9600, valid loss-0.5747, acc-0.8496, test loss-0.5779, acc-0.8465
Iter-32570, train loss-0.6078, acc-0.8200, valid loss-0.5746, acc-0.8498, test loss-0.5778, acc-0.8464
Iter-32580, train loss-0.7005, acc-0.7600, valid loss-0.5746, acc-0.8496, test loss-0.5777, acc-0.8464
Iter-32590, train loss-0.7623, acc-0.7600, valid loss-0.5745, acc-0.8496, test loss-0.5776, acc-0.8465
Iter-32600, train loss-0.9679, acc-0.7200, valid loss-0.5744, acc-0.8498, test loss-0.5775, acc-0.8466
Iter-32610, train loss-0.6593, acc-0.8000, valid loss-0.5743, acc-0.8498, test loss-0.5774, acc-0.8466
Iter-32620, train loss-0.6176, acc-0.8200, valid loss-0.5742, acc-0.8498, test loss-0.5773, acc-0.8466
Iter-32630, train loss-0.6174, acc-0.8200, valid loss-0.5741, acc-0.8498, test loss-0.5772, acc-0.8464
Iter-32640, train loss-0.4405, acc-0.9000, valid loss-0.5740, acc-0.8498, test loss-0.5771, acc-0.8466
Iter-32650, train loss-0.4584, acc-0.8800, valid loss-0.5739, acc-0.8498, test loss-0.5770, acc-0.8466
Iter-32660, train loss-0.6126, acc-0.8200, valid loss-0.5738, acc-0.8496, test loss-0.5769, acc-0.8467
Iter-32670, train loss-0.7083, acc-0.8400, valid loss-0.5737, acc-0.8496, test loss-0.5768, acc-0.8467
Iter-32680, train loss-0.5166, acc-0.8200, valid loss-0.5736, acc-0.8496, test loss-0.5767, acc-0.8468
Iter-32690, train loss-0.5907, acc-0.8600, valid loss-0.5735, acc-0.8496, test loss-0.5766, acc-0.8467
Iter-32700, train loss-0.5655, acc-0.8400, valid loss-0.5734, acc-0.8494, test loss-0.5765, acc-0.8467
Iter-32710, train loss-0.6149, acc-0.8200, valid loss-0.5733, acc-0.8496, test loss-0.5764, acc-0.8467
Iter-32720, train loss-0.7536, acc-0.8400, valid loss-0.5732, acc-0.8494, test loss-0.5764, acc-0.8466
Iter-32730, train loss-0.7100, acc-0.8200, valid loss-0.5731, acc-0.8494, test loss-0.5763, acc-0.8467
Iter-32740, train loss-0.6232, acc-0.8400, valid loss-0.5730, acc-0.8494, test loss-0.5762, acc-0.8469
Iter-32750, train loss-0.4644, acc-0.9200, valid loss-0.5729, acc-0.8494, test loss-0.5761, acc-0.8467
Iter-32760, train loss-0.5764, acc-0.8800, valid loss-0.5728, acc-0.8492, test loss-0.5760, acc-0.8469
Iter-32770, train loss-0.5572, acc-0.7800, valid loss-0.5727, acc-0.8492, test loss-0.5759, acc-0.8470
Iter-32780, train loss-0.4118, acc-0.9200, valid loss-0.5726, acc-0.8494, test loss-0.5758, acc-0.8468
Iter-32790, train loss-0.5142, acc-0.8200, valid loss-0.5725, acc-0.8494, test loss-0.5757, acc-0.8467
Iter-32800, train loss-0.5948, acc-0.8600, valid loss-0.5724, acc-0.8494, test loss-0.5756, acc-0.8469
Iter-32810, train loss-0.6336, acc-0.8600, valid loss-0.5723, acc-0.8494, test loss-0.5755, acc-0.8470
Iter-32820, train loss-0.7452, acc-0.8400, valid loss-0.5722, acc-0.8496, test loss-0.5754, acc-0.8471
Iter-32830, train loss-0.5094, acc-0.8600, valid loss-0.5721, acc-0.8496, test loss-0.5753, acc-0.8472
Iter-32840, train loss-0.8237, acc-0.7400, valid loss-0.5720, acc-0.8496, test loss-0.5752, acc-0.8473
Iter-32850, train loss-0.7643, acc-0.8400, valid loss-0.5719, acc-0.8498, test loss-0.5751, acc-0.8472
Iter-32860, train loss-0.7355, acc-0.7600, valid loss-0.5718, acc-0.8496, test loss-0.5750, acc-0.8471
Iter-32870, train loss-0.6379, acc-0.8200, valid loss-0.5717, acc-0.8496, test loss-0.5749, acc-0.8471
Iter-32880, train loss-0.5023, acc-0.8600, valid loss-0.5716, acc-0.8496, test loss-0.5748, acc-0.8470
Iter-32890, train loss-0.7329, acc-0.7600, valid loss-0.5715, acc-0.8498, test loss-0.5747, acc-0.8470
Iter-32900, train loss-0.4940, acc-0.9000, valid loss-0.5714, acc-0.8500, test loss-0.5746, acc-0.8470
Iter-32910, train loss-0.5596, acc-0.9000, valid loss-0.5713, acc-0.8500, test loss-0.5745, acc-0.8470
Iter-32920, train loss-0.5238, acc-0.8200, valid loss-0.5712, acc-0.8500, test loss-0.5744, acc-0.8470
Iter-32930, train loss-0.6081, acc-0.8600, valid loss-0.5712, acc-0.8498, test loss-0.5743, acc-0.8470
Iter-32940, train loss-0.5782, acc-0.8200, valid loss-0.5711, acc-0.8500, test loss-0.5742, acc-0.8470
Iter-32950, train loss-0.5868, acc-0.7800, valid loss-0.5710, acc-0.8500, test loss-0.5741, acc-0.8471
Iter-32960, train loss-0.4474, acc-0.8800, valid loss-0.5709, acc-0.8500, test loss-0.5740, acc-0.8471
Iter-32970, train loss-0.7019, acc-0.7600, valid loss-0.5708, acc-0.8500, test loss-0.5739, acc-0.8470
Iter-32980, train loss-0.4754, acc-0.8800, valid loss-0.5707, acc-0.8500, test loss-0.5738, acc-0.8472
Iter-32990, train loss-0.5877, acc-0.8200, valid loss-0.5706, acc-0.8500, test loss-0.5737, acc-0.8473
Iter-33000, train loss-0.7974, acc-0.7600, valid loss-0.5705, acc-0.8500, test loss-0.5737, acc-0.8473
Iter-33010, train loss-0.5235, acc-0.8400, valid loss-0.5704, acc-0.8500, test loss-0.5735, acc-0.8473
Iter-33020, train loss-0.3787, acc-0.9200, valid loss-0.5703, acc-0.8502, test loss-0.5735, acc-0.8474
Iter-33030, train loss-0.6145, acc-0.8600, valid loss-0.5702, acc-0.8502, test loss-0.5734, acc-0.8476
Iter-33040, train loss-0.4878, acc-0.8600, valid loss-0.5701, acc-0.8502, test loss-0.5733, acc-0.8476
Iter-33050, train loss-0.7760, acc-0.6800, valid loss-0.5700, acc-0.8504, test loss-0.5732, acc-0.8475
Iter-33060, train loss-0.6659, acc-0.7800, valid loss-0.5699, acc-0.8504, test loss-0.5731, acc-0.8476
Iter-33070, train loss-0.4577, acc-0.8800, valid loss-0.5698, acc-0.8502, test loss-0.5730, acc-0.8476
Iter-33080, train loss-0.6397, acc-0.8400, valid loss-0.5697, acc-0.8502, test loss-0.5729, acc-0.8476
Iter-33090, train loss-0.5202, acc-0.8800, valid loss-0.5697, acc-0.8504, test loss-0.5728, acc-0.8476
Iter-33100, train loss-0.5470, acc-0.8600, valid loss-0.5696, acc-0.8506, test loss-0.5727, acc-0.8476
Iter-33110, train loss-0.6361, acc-0.8400, valid loss-0.5695, acc-0.8504, test loss-0.5726, acc-0.8476
Iter-33120, train loss-0.7543, acc-0.8000, valid loss-0.5694, acc-0.8506, test loss-0.5726, acc-0.8477
Iter-33130, train loss-0.5761, acc-0.8800, valid loss-0.5693, acc-0.8506, test loss-0.5725, acc-0.8477
Iter-33140, train loss-0.6327, acc-0.8200, valid loss-0.5692, acc-0.8506, test loss-0.5724, acc-0.8477
Iter-33150, train loss-0.6475, acc-0.8400, valid loss-0.5691, acc-0.8506, test loss-0.5723, acc-0.8476
Iter-33160, train loss-0.5421, acc-0.8600, valid loss-0.5690, acc-0.8508, test loss-0.5722, acc-0.8475
Iter-33170, train loss-0.6622, acc-0.7600, valid loss-0.5689, acc-0.8506, test loss-0.5721, acc-0.8476
Iter-33180, train loss-0.6964, acc-0.8200, valid loss-0.5688, acc-0.8506, test loss-0.5720, acc-0.8476
Iter-33190, train loss-0.6897, acc-0.8000, valid loss-0.5687, acc-0.8506, test loss-0.5719, acc-0.8476
Iter-33200, train loss-0.4598, acc-0.8200, valid loss-0.5686, acc-0.8506, test loss-0.5718, acc-0.8474
Iter-33210, train loss-0.7082, acc-0.8000, valid loss-0.5685, acc-0.8506, test loss-0.5717, acc-0.8474
Iter-33220, train loss-0.5856, acc-0.8200, valid loss-0.5684, acc-0.8506, test loss-0.5716, acc-0.8476
Iter-33230, train loss-0.6620, acc-0.8400, valid loss-0.5683, acc-0.8504, test loss-0.5715, acc-0.8475
Iter-33240, train loss-0.6712, acc-0.8200, valid loss-0.5682, acc-0.8508, test loss-0.5714, acc-0.8476
Iter-33250, train loss-0.4939, acc-0.8800, valid loss-0.5681, acc-0.8508, test loss-0.5713, acc-0.8477
Iter-33260, train loss-0.5668, acc-0.8400, valid loss-0.5680, acc-0.8510, test loss-0.5712, acc-0.8477
Iter-33270, train loss-0.6777, acc-0.8400, valid loss-0.5680, acc-0.8510, test loss-0.5711, acc-0.8477
Iter-33280, train loss-0.5224, acc-0.8600, valid loss-0.5679, acc-0.8508, test loss-0.5710, acc-0.8477
Iter-33290, train loss-0.5496, acc-0.8800, valid loss-0.5678, acc-0.8510, test loss-0.5709, acc-0.8477
Iter-33300, train loss-0.4998, acc-0.8600, valid loss-0.5677, acc-0.8506, test loss-0.5708, acc-0.8476
Iter-33310, train loss-0.4413, acc-0.8600, valid loss-0.5676, acc-0.8508, test loss-0.5707, acc-0.8474
Iter-33320, train loss-0.5402, acc-0.8200, valid loss-0.5675, acc-0.8510, test loss-0.5706, acc-0.8475
Iter-33330, train loss-0.6021, acc-0.7600, valid loss-0.5674, acc-0.8510, test loss-0.5705, acc-0.8475
Iter-33340, train loss-0.7859, acc-0.7600, valid loss-0.5673, acc-0.8510, test loss-0.5704, acc-0.8476
Iter-33350, train loss-0.5350, acc-0.8400, valid loss-0.5672, acc-0.8508, test loss-0.5703, acc-0.8476
Iter-33360, train loss-0.5292, acc-0.8400, valid loss-0.5671, acc-0.8506, test loss-0.5702, acc-0.8477
Iter-33370, train loss-0.6575, acc-0.8200, valid loss-0.5670, acc-0.8508, test loss-0.5701, acc-0.8477
Iter-33380, train loss-0.5854, acc-0.8600, valid loss-0.5669, acc-0.8508, test loss-0.5700, acc-0.8478
Iter-33390, train loss-0.5376, acc-0.8800, valid loss-0.5668, acc-0.8510, test loss-0.5699, acc-0.8479
Iter-33400, train loss-0.4950, acc-0.9200, valid loss-0.5667, acc-0.8510, test loss-0.5698, acc-0.8479
Iter-33410, train loss-0.4661, acc-0.8800, valid loss-0.5666, acc-0.8510, test loss-0.5697, acc-0.8478
Iter-33420, train loss-0.4080, acc-0.9000, valid loss-0.5665, acc-0.8510, test loss-0.5696, acc-0.8477
Iter-33430, train loss-0.4109, acc-0.8800, valid loss-0.5664, acc-0.8510, test loss-0.5695, acc-0.8477
Iter-33440, train loss-0.6575, acc-0.8000, valid loss-0.5664, acc-0.8510, test loss-0.5694, acc-0.8478
Iter-33450, train loss-0.5409, acc-0.8600, valid loss-0.5663, acc-0.8508, test loss-0.5693, acc-0.8477
Iter-33460, train loss-0.6129, acc-0.8200, valid loss-0.5661, acc-0.8508, test loss-0.5692, acc-0.8477
Iter-33470, train loss-0.5375, acc-0.9000, valid loss-0.5660, acc-0.8508, test loss-0.5691, acc-0.8478
Iter-33480, train loss-0.6446, acc-0.8400, valid loss-0.5659, acc-0.8510, test loss-0.5690, acc-0.8477
Iter-33490, train loss-0.5028, acc-0.8400, valid loss-0.5658, acc-0.8510, test loss-0.5689, acc-0.8479
Iter-33500, train loss-0.4865, acc-0.8600, valid loss-0.5658, acc-0.8510, test loss-0.5688, acc-0.8479
Iter-33510, train loss-0.5976, acc-0.8800, valid loss-0.5656, acc-0.8514, test loss-0.5687, acc-0.8480
Iter-33520, train loss-0.3681, acc-0.9600, valid loss-0.5656, acc-0.8514, test loss-0.5686, acc-0.8481
Iter-33530, train loss-0.6547, acc-0.8000, valid loss-0.5655, acc-0.8516, test loss-0.5685, acc-0.8481
Iter-33540, train loss-0.6727, acc-0.8200, valid loss-0.5654, acc-0.8516, test loss-0.5685, acc-0.8481
Iter-33550, train loss-0.4107, acc-0.8800, valid loss-0.5653, acc-0.8516, test loss-0.5684, acc-0.8481
Iter-33560, train loss-0.5677, acc-0.8400, valid loss-0.5652, acc-0.8514, test loss-0.5683, acc-0.8481
Iter-33570, train loss-0.7049, acc-0.7800, valid loss-0.5651, acc-0.8516, test loss-0.5682, acc-0.8481
Iter-33580, train loss-0.4550, acc-0.8600, valid loss-0.5650, acc-0.8516, test loss-0.5681, acc-0.8482
Iter-33590, train loss-0.5174, acc-0.9000, valid loss-0.5649, acc-0.8516, test loss-0.5680, acc-0.8482
Iter-33600, train loss-0.4281, acc-0.9200, valid loss-0.5648, acc-0.8518, test loss-0.5679, acc-0.8481
Iter-33610, train loss-0.5181, acc-0.8400, valid loss-0.5647, acc-0.8516, test loss-0.5678, acc-0.8481
Iter-33620, train loss-0.6350, acc-0.8400, valid loss-0.5646, acc-0.8518, test loss-0.5677, acc-0.8483
Iter-33630, train loss-0.4854, acc-0.8400, valid loss-0.5645, acc-0.8518, test loss-0.5677, acc-0.8482
Iter-33640, train loss-0.4812, acc-0.8600, valid loss-0.5644, acc-0.8518, test loss-0.5676, acc-0.8482
Iter-33650, train loss-0.6265, acc-0.8400, valid loss-0.5644, acc-0.8516, test loss-0.5675, acc-0.8483
Iter-33660, train loss-0.4524, acc-0.9400, valid loss-0.5643, acc-0.8516, test loss-0.5674, acc-0.8484
Iter-33670, train loss-0.5136, acc-0.8800, valid loss-0.5642, acc-0.8516, test loss-0.5673, acc-0.8484
Iter-33680, train loss-0.4837, acc-0.8600, valid loss-0.5641, acc-0.8514, test loss-0.5672, acc-0.8486
Iter-33690, train loss-0.4536, acc-0.8600, valid loss-0.5640, acc-0.8516, test loss-0.5671, acc-0.8485
Iter-33700, train loss-0.6009, acc-0.7800, valid loss-0.5639, acc-0.8516, test loss-0.5670, acc-0.8486
Iter-33710, train loss-0.8155, acc-0.7600, valid loss-0.5638, acc-0.8516, test loss-0.5669, acc-0.8486
Iter-33720, train loss-0.4749, acc-0.9200, valid loss-0.5637, acc-0.8516, test loss-0.5668, acc-0.8486
Iter-33730, train loss-0.5092, acc-0.8800, valid loss-0.5636, acc-0.8516, test loss-0.5667, acc-0.8486
Iter-33740, train loss-0.4381, acc-0.9000, valid loss-0.5635, acc-0.8516, test loss-0.5666, acc-0.8486
Iter-33750, train loss-0.7670, acc-0.8400, valid loss-0.5635, acc-0.8516, test loss-0.5665, acc-0.8486
Iter-33760, train loss-0.6147, acc-0.8600, valid loss-0.5634, acc-0.8516, test loss-0.5664, acc-0.8486
Iter-33770, train loss-0.6321, acc-0.7800, valid loss-0.5633, acc-0.8516, test loss-0.5663, acc-0.8486
Iter-33780, train loss-0.5017, acc-0.8800, valid loss-0.5632, acc-0.8516, test loss-0.5663, acc-0.8486
Iter-33790, train loss-0.4382, acc-0.8800, valid loss-0.5631, acc-0.8516, test loss-0.5662, acc-0.8487
Iter-33800, train loss-0.6364, acc-0.8600, valid loss-0.5630, acc-0.8516, test loss-0.5661, acc-0.8486
Iter-33810, train loss-0.4701, acc-0.8600, valid loss-0.5629, acc-0.8516, test loss-0.5660, acc-0.8487
Iter-33820, train loss-0.6600, acc-0.8200, valid loss-0.5628, acc-0.8516, test loss-0.5659, acc-0.8485
Iter-33830, train loss-0.5404, acc-0.8400, valid loss-0.5628, acc-0.8516, test loss-0.5658, acc-0.8485
Iter-33840, train loss-0.4975, acc-0.8400, valid loss-0.5627, acc-0.8516, test loss-0.5657, acc-0.8486
Iter-33850, train loss-0.6098, acc-0.8200, valid loss-0.5626, acc-0.8516, test loss-0.5656, acc-0.8486
Iter-33860, train loss-0.6008, acc-0.8200, valid loss-0.5625, acc-0.8516, test loss-0.5655, acc-0.8487
Iter-33870, train loss-0.5286, acc-0.9000, valid loss-0.5624, acc-0.8516, test loss-0.5654, acc-0.8487
Iter-33880, train loss-0.5602, acc-0.8400, valid loss-0.5623, acc-0.8516, test loss-0.5653, acc-0.8488
Iter-33890, train loss-0.5996, acc-0.8400, valid loss-0.5622, acc-0.8516, test loss-0.5652, acc-0.8486
Iter-33900, train loss-0.8307, acc-0.7600, valid loss-0.5621, acc-0.8516, test loss-0.5652, acc-0.8486
Iter-33910, train loss-0.6236, acc-0.8200, valid loss-0.5620, acc-0.8516, test loss-0.5651, acc-0.8489
Iter-33920, train loss-0.3954, acc-0.9200, valid loss-0.5619, acc-0.8516, test loss-0.5650, acc-0.8490
Iter-33930, train loss-0.6873, acc-0.8000, valid loss-0.5618, acc-0.8516, test loss-0.5649, acc-0.8489
Iter-33940, train loss-0.4709, acc-0.8800, valid loss-0.5617, acc-0.8516, test loss-0.5648, acc-0.8488
Iter-33950, train loss-0.4799, acc-0.8600, valid loss-0.5616, acc-0.8516, test loss-0.5647, acc-0.8490
Iter-33960, train loss-0.7636, acc-0.8000, valid loss-0.5615, acc-0.8516, test loss-0.5646, acc-0.8490
Iter-33970, train loss-0.5140, acc-0.8600, valid loss-0.5614, acc-0.8516, test loss-0.5645, acc-0.8489
Iter-33980, train loss-0.5464, acc-0.8400, valid loss-0.5613, acc-0.8516, test loss-0.5644, acc-0.8490
Iter-33990, train loss-0.4554, acc-0.9000, valid loss-0.5613, acc-0.8518, test loss-0.5643, acc-0.8490
Iter-34000, train loss-0.6192, acc-0.8200, valid loss-0.5612, acc-0.8518, test loss-0.5642, acc-0.8490
Iter-34010, train loss-0.3913, acc-0.8600, valid loss-0.5610, acc-0.8518, test loss-0.5641, acc-0.8490
Iter-34020, train loss-0.4576, acc-0.9000, valid loss-0.5609, acc-0.8518, test loss-0.5641, acc-0.8490
Iter-34030, train loss-0.8761, acc-0.7600, valid loss-0.5609, acc-0.8518, test loss-0.5640, acc-0.8490
Iter-34040, train loss-0.6516, acc-0.8200, valid loss-0.5608, acc-0.8520, test loss-0.5639, acc-0.8490
Iter-34050, train loss-0.6210, acc-0.7800, valid loss-0.5607, acc-0.8522, test loss-0.5638, acc-0.8490
Iter-34060, train loss-0.4977, acc-0.8400, valid loss-0.5606, acc-0.8522, test loss-0.5637, acc-0.8490
Iter-34070, train loss-0.6080, acc-0.8200, valid loss-0.5605, acc-0.8522, test loss-0.5636, acc-0.8492
Iter-34080, train loss-0.6321, acc-0.8200, valid loss-0.5604, acc-0.8520, test loss-0.5635, acc-0.8492
Iter-34090, train loss-0.6407, acc-0.8200, valid loss-0.5603, acc-0.8524, test loss-0.5634, acc-0.8492
Iter-34100, train loss-0.3543, acc-0.9600, valid loss-0.5602, acc-0.8524, test loss-0.5633, acc-0.8492
Iter-34110, train loss-0.7399, acc-0.7400, valid loss-0.5601, acc-0.8524, test loss-0.5632, acc-0.8492
Iter-34120, train loss-0.5336, acc-0.8800, valid loss-0.5600, acc-0.8524, test loss-0.5631, acc-0.8491
Iter-34130, train loss-0.7746, acc-0.7400, valid loss-0.5599, acc-0.8524, test loss-0.5630, acc-0.8491
Iter-34140, train loss-0.4922, acc-0.8800, valid loss-0.5598, acc-0.8524, test loss-0.5629, acc-0.8492
Iter-34150, train loss-0.5159, acc-0.8600, valid loss-0.5598, acc-0.8526, test loss-0.5628, acc-0.8493
Iter-34160, train loss-0.7383, acc-0.7600, valid loss-0.5597, acc-0.8526, test loss-0.5628, acc-0.8492
Iter-34170, train loss-0.6353, acc-0.7800, valid loss-0.5596, acc-0.8526, test loss-0.5627, acc-0.8492
Iter-34180, train loss-0.8980, acc-0.6600, valid loss-0.5595, acc-0.8526, test loss-0.5626, acc-0.8492
Iter-34190, train loss-0.8283, acc-0.7800, valid loss-0.5594, acc-0.8524, test loss-0.5625, acc-0.8492
Iter-34200, train loss-0.5362, acc-0.8600, valid loss-0.5593, acc-0.8526, test loss-0.5624, acc-0.8493
Iter-34210, train loss-0.3371, acc-0.9200, valid loss-0.5592, acc-0.8526, test loss-0.5623, acc-0.8494
Iter-34220, train loss-0.6750, acc-0.7600, valid loss-0.5591, acc-0.8526, test loss-0.5622, acc-0.8494
Iter-34230, train loss-0.6896, acc-0.7800, valid loss-0.5590, acc-0.8526, test loss-0.5621, acc-0.8495
Iter-34240, train loss-0.6781, acc-0.8400, valid loss-0.5589, acc-0.8526, test loss-0.5621, acc-0.8495
Iter-34250, train loss-0.6603, acc-0.8400, valid loss-0.5588, acc-0.8524, test loss-0.5620, acc-0.8495
Iter-34260, train loss-0.5367, acc-0.8000, valid loss-0.5587, acc-0.8526, test loss-0.5619, acc-0.8496
Iter-34270, train loss-0.8721, acc-0.7600, valid loss-0.5586, acc-0.8524, test loss-0.5618, acc-0.8496
Iter-34280, train loss-0.5684, acc-0.8400, valid loss-0.5585, acc-0.8524, test loss-0.5617, acc-0.8497
Iter-34290, train loss-0.4838, acc-0.8400, valid loss-0.5584, acc-0.8524, test loss-0.5616, acc-0.8497
Iter-34300, train loss-0.5346, acc-0.8800, valid loss-0.5584, acc-0.8524, test loss-0.5615, acc-0.8497
Iter-34310, train loss-0.6119, acc-0.7800, valid loss-0.5583, acc-0.8524, test loss-0.5614, acc-0.8496
Iter-34320, train loss-0.7062, acc-0.8200, valid loss-0.5582, acc-0.8522, test loss-0.5613, acc-0.8496
Iter-34330, train loss-0.4249, acc-0.9000, valid loss-0.5581, acc-0.8526, test loss-0.5612, acc-0.8496
Iter-34340, train loss-0.6442, acc-0.8200, valid loss-0.5580, acc-0.8526, test loss-0.5611, acc-0.8496
Iter-34350, train loss-0.4701, acc-0.8800, valid loss-0.5579, acc-0.8524, test loss-0.5610, acc-0.8495
Iter-34360, train loss-0.7281, acc-0.7600, valid loss-0.5578, acc-0.8524, test loss-0.5610, acc-0.8497
Iter-34370, train loss-0.6302, acc-0.8000, valid loss-0.5577, acc-0.8524, test loss-0.5608, acc-0.8499
Iter-34380, train loss-0.4757, acc-0.9200, valid loss-0.5576, acc-0.8524, test loss-0.5607, acc-0.8499
Iter-34390, train loss-0.5606, acc-0.8800, valid loss-0.5575, acc-0.8524, test loss-0.5607, acc-0.8499
Iter-34400, train loss-0.6978, acc-0.7800, valid loss-0.5575, acc-0.8524, test loss-0.5606, acc-0.8498
Iter-34410, train loss-0.6759, acc-0.8200, valid loss-0.5574, acc-0.8524, test loss-0.5605, acc-0.8498
Iter-34420, train loss-0.4890, acc-0.8400, valid loss-0.5573, acc-0.8524, test loss-0.5604, acc-0.8498
Iter-34430, train loss-0.6191, acc-0.7600, valid loss-0.5572, acc-0.8524, test loss-0.5604, acc-0.8499
Iter-34440, train loss-0.7468, acc-0.7400, valid loss-0.5572, acc-0.8524, test loss-0.5603, acc-0.8498
Iter-34450, train loss-0.5780, acc-0.8200, valid loss-0.5571, acc-0.8522, test loss-0.5602, acc-0.8498
Iter-34460, train loss-0.5237, acc-0.8200, valid loss-0.5570, acc-0.8522, test loss-0.5601, acc-0.8499
Iter-34470, train loss-0.6022, acc-0.8400, valid loss-0.5569, acc-0.8522, test loss-0.5600, acc-0.8500
Iter-34480, train loss-0.4748, acc-0.8600, valid loss-0.5568, acc-0.8522, test loss-0.5599, acc-0.8498
Iter-34490, train loss-0.4459, acc-0.8600, valid loss-0.5567, acc-0.8522, test loss-0.5598, acc-0.8499
Iter-34500, train loss-0.3711, acc-0.9400, valid loss-0.5566, acc-0.8520, test loss-0.5597, acc-0.8501
Iter-34510, train loss-0.6130, acc-0.8200, valid loss-0.5566, acc-0.8520, test loss-0.5596, acc-0.8501
Iter-34520, train loss-0.5939, acc-0.7600, valid loss-0.5565, acc-0.8520, test loss-0.5596, acc-0.8502
Iter-34530, train loss-0.5801, acc-0.8400, valid loss-0.5564, acc-0.8520, test loss-0.5595, acc-0.8502
Iter-34540, train loss-0.6113, acc-0.7600, valid loss-0.5563, acc-0.8522, test loss-0.5594, acc-0.8503
Iter-34550, train loss-0.6839, acc-0.8200, valid loss-0.5562, acc-0.8522, test loss-0.5593, acc-0.8502
Iter-34560, train loss-0.6024, acc-0.8600, valid loss-0.5561, acc-0.8524, test loss-0.5592, acc-0.8503
Iter-34570, train loss-0.7617, acc-0.7000, valid loss-0.5560, acc-0.8524, test loss-0.5591, acc-0.8504
Iter-34580, train loss-0.8138, acc-0.7400, valid loss-0.5559, acc-0.8526, test loss-0.5590, acc-0.8504
Iter-34590, train loss-0.6990, acc-0.7600, valid loss-0.5558, acc-0.8522, test loss-0.5589, acc-0.8504
Iter-34600, train loss-0.5418, acc-0.8400, valid loss-0.5557, acc-0.8524, test loss-0.5588, acc-0.8503
Iter-34610, train loss-0.5764, acc-0.8200, valid loss-0.5556, acc-0.8524, test loss-0.5587, acc-0.8503
Iter-34620, train loss-0.3812, acc-0.9400, valid loss-0.5555, acc-0.8524, test loss-0.5587, acc-0.8500
Iter-34630, train loss-0.9024, acc-0.7600, valid loss-0.5554, acc-0.8524, test loss-0.5586, acc-0.8500
Iter-34640, train loss-0.3803, acc-0.9200, valid loss-0.5553, acc-0.8524, test loss-0.5585, acc-0.8500
Iter-34650, train loss-0.8548, acc-0.7200, valid loss-0.5553, acc-0.8526, test loss-0.5584, acc-0.8500
Iter-34660, train loss-0.4358, acc-0.9200, valid loss-0.5552, acc-0.8524, test loss-0.5583, acc-0.8500
Iter-34670, train loss-0.6786, acc-0.7400, valid loss-0.5551, acc-0.8524, test loss-0.5582, acc-0.8500
Iter-34680, train loss-0.6657, acc-0.7600, valid loss-0.5550, acc-0.8526, test loss-0.5581, acc-0.8502
Iter-34690, train loss-0.8551, acc-0.7400, valid loss-0.5549, acc-0.8524, test loss-0.5581, acc-0.8502
Iter-34700, train loss-0.6417, acc-0.8800, valid loss-0.5548, acc-0.8524, test loss-0.5580, acc-0.8502
Iter-34710, train loss-0.4693, acc-0.8600, valid loss-0.5547, acc-0.8526, test loss-0.5579, acc-0.8502
Iter-34720, train loss-0.4690, acc-0.8600, valid loss-0.5547, acc-0.8524, test loss-0.5578, acc-0.8503
Iter-34730, train loss-0.7326, acc-0.8200, valid loss-0.5545, acc-0.8524, test loss-0.5577, acc-0.8505
Iter-34740, train loss-0.6737, acc-0.8000, valid loss-0.5545, acc-0.8524, test loss-0.5576, acc-0.8503
Iter-34750, train loss-0.6994, acc-0.7800, valid loss-0.5544, acc-0.8526, test loss-0.5575, acc-0.8504
Iter-34760, train loss-0.5182, acc-0.9000, valid loss-0.5543, acc-0.8526, test loss-0.5574, acc-0.8503
Iter-34770, train loss-0.6499, acc-0.8200, valid loss-0.5542, acc-0.8526, test loss-0.5573, acc-0.8504
Iter-34780, train loss-0.6167, acc-0.8600, valid loss-0.5541, acc-0.8528, test loss-0.5572, acc-0.8504
Iter-34790, train loss-0.5606, acc-0.8600, valid loss-0.5540, acc-0.8528, test loss-0.5572, acc-0.8505
Iter-34800, train loss-0.4352, acc-0.8800, valid loss-0.5539, acc-0.8528, test loss-0.5571, acc-0.8505
Iter-34810, train loss-0.6393, acc-0.8400, valid loss-0.5538, acc-0.8528, test loss-0.5570, acc-0.8506
Iter-34820, train loss-0.4937, acc-0.8200, valid loss-0.5537, acc-0.8528, test loss-0.5569, acc-0.8506
Iter-34830, train loss-0.6320, acc-0.8200, valid loss-0.5537, acc-0.8528, test loss-0.5568, acc-0.8506
Iter-34840, train loss-0.5126, acc-0.8800, valid loss-0.5536, acc-0.8526, test loss-0.5568, acc-0.8505
Iter-34850, train loss-0.5566, acc-0.9000, valid loss-0.5535, acc-0.8526, test loss-0.5567, acc-0.8504
Iter-34860, train loss-0.8127, acc-0.7200, valid loss-0.5534, acc-0.8528, test loss-0.5566, acc-0.8505
Iter-34870, train loss-0.5030, acc-0.8600, valid loss-0.5533, acc-0.8528, test loss-0.5565, acc-0.8505
Iter-34880, train loss-0.5659, acc-0.8200, valid loss-0.5532, acc-0.8528, test loss-0.5564, acc-0.8506
Iter-34890, train loss-0.7985, acc-0.7800, valid loss-0.5532, acc-0.8526, test loss-0.5563, acc-0.8505
Iter-34900, train loss-0.4360, acc-0.9000, valid loss-0.5531, acc-0.8528, test loss-0.5562, acc-0.8506
Iter-34910, train loss-0.4700, acc-0.9000, valid loss-0.5530, acc-0.8528, test loss-0.5561, acc-0.8506
Iter-34920, train loss-0.5181, acc-0.8600, valid loss-0.5529, acc-0.8530, test loss-0.5561, acc-0.8506
Iter-34930, train loss-0.5995, acc-0.7800, valid loss-0.5528, acc-0.8530, test loss-0.5560, acc-0.8506
Iter-34940, train loss-0.4242, acc-0.8800, valid loss-0.5527, acc-0.8530, test loss-0.5559, acc-0.8505
Iter-34950, train loss-0.4859, acc-0.8400, valid loss-0.5526, acc-0.8526, test loss-0.5558, acc-0.8505
Iter-34960, train loss-0.4107, acc-0.9200, valid loss-0.5525, acc-0.8528, test loss-0.5557, acc-0.8505
Iter-34970, train loss-0.4722, acc-0.8800, valid loss-0.5525, acc-0.8528, test loss-0.5556, acc-0.8505
Iter-34980, train loss-0.5566, acc-0.8000, valid loss-0.5524, acc-0.8528, test loss-0.5556, acc-0.8505
Iter-34990, train loss-0.5189, acc-0.8200, valid loss-0.5523, acc-0.8528, test loss-0.5555, acc-0.8504
Iter-35000, train loss-0.5679, acc-0.8400, valid loss-0.5522, acc-0.8528, test loss-0.5554, acc-0.8504
Iter-35010, train loss-0.4815, acc-0.9400, valid loss-0.5521, acc-0.8528, test loss-0.5553, acc-0.8504
Iter-35020, train loss-0.6396, acc-0.8400, valid loss-0.5520, acc-0.8528, test loss-0.5552, acc-0.8504
Iter-35030, train loss-0.4788, acc-0.9400, valid loss-0.5519, acc-0.8526, test loss-0.5551, acc-0.8505
Iter-35040, train loss-0.3707, acc-0.9400, valid loss-0.5519, acc-0.8528, test loss-0.5550, acc-0.8505
Iter-35050, train loss-0.5305, acc-0.8200, valid loss-0.5518, acc-0.8528, test loss-0.5550, acc-0.8505
Iter-35060, train loss-0.4202, acc-0.9200, valid loss-0.5517, acc-0.8526, test loss-0.5549, acc-0.8505
Iter-35070, train loss-0.4877, acc-0.7800, valid loss-0.5516, acc-0.8528, test loss-0.5548, acc-0.8505
Iter-35080, train loss-0.5366, acc-0.8200, valid loss-0.5515, acc-0.8528, test loss-0.5547, acc-0.8505
Iter-35090, train loss-0.5384, acc-0.8600, valid loss-0.5514, acc-0.8528, test loss-0.5546, acc-0.8506
Iter-35100, train loss-0.6319, acc-0.8000, valid loss-0.5513, acc-0.8530, test loss-0.5545, acc-0.8506
Iter-35110, train loss-0.5303, acc-0.8600, valid loss-0.5513, acc-0.8530, test loss-0.5544, acc-0.8507
Iter-35120, train loss-0.3662, acc-0.9400, valid loss-0.5512, acc-0.8532, test loss-0.5543, acc-0.8508
Iter-35130, train loss-0.5870, acc-0.8400, valid loss-0.5511, acc-0.8530, test loss-0.5542, acc-0.8510
Iter-35140, train loss-0.8368, acc-0.8200, valid loss-0.5510, acc-0.8530, test loss-0.5542, acc-0.8510
Iter-35150, train loss-0.6156, acc-0.8400, valid loss-0.5509, acc-0.8530, test loss-0.5541, acc-0.8510
Iter-35160, train loss-0.3897, acc-0.9000, valid loss-0.5508, acc-0.8528, test loss-0.5540, acc-0.8509
Iter-35170, train loss-0.2407, acc-0.9600, valid loss-0.5508, acc-0.8528, test loss-0.5539, acc-0.8509
Iter-35180, train loss-0.5355, acc-0.9000, valid loss-0.5507, acc-0.8528, test loss-0.5539, acc-0.8509
Iter-35190, train loss-0.5174, acc-0.8800, valid loss-0.5506, acc-0.8528, test loss-0.5538, acc-0.8509
Iter-35200, train loss-0.4779, acc-0.8600, valid loss-0.5505, acc-0.8528, test loss-0.5537, acc-0.8510
Iter-35210, train loss-0.4316, acc-0.8800, valid loss-0.5504, acc-0.8528, test loss-0.5536, acc-0.8511
Iter-35220, train loss-0.4908, acc-0.8800, valid loss-0.5503, acc-0.8528, test loss-0.5535, acc-0.8510
Iter-35230, train loss-0.5346, acc-0.8600, valid loss-0.5502, acc-0.8528, test loss-0.5534, acc-0.8510
Iter-35240, train loss-0.4882, acc-0.9200, valid loss-0.5501, acc-0.8528, test loss-0.5534, acc-0.8507
Iter-35250, train loss-0.6209, acc-0.8200, valid loss-0.5501, acc-0.8534, test loss-0.5533, acc-0.8509
Iter-35260, train loss-0.6273, acc-0.8000, valid loss-0.5500, acc-0.8532, test loss-0.5532, acc-0.8509
Iter-35270, train loss-0.5008, acc-0.8800, valid loss-0.5499, acc-0.8532, test loss-0.5531, acc-0.8511
Iter-35280, train loss-0.5099, acc-0.8600, valid loss-0.5498, acc-0.8534, test loss-0.5530, acc-0.8511
Iter-35290, train loss-0.6147, acc-0.8400, valid loss-0.5497, acc-0.8532, test loss-0.5529, acc-0.8513
Iter-35300, train loss-0.7386, acc-0.7800, valid loss-0.5496, acc-0.8534, test loss-0.5529, acc-0.8512
Iter-35310, train loss-0.5067, acc-0.9000, valid loss-0.5495, acc-0.8532, test loss-0.5528, acc-0.8513
Iter-35320, train loss-0.5265, acc-0.8800, valid loss-0.5494, acc-0.8530, test loss-0.5527, acc-0.8513
Iter-35330, train loss-0.5139, acc-0.8400, valid loss-0.5493, acc-0.8532, test loss-0.5526, acc-0.8514
Iter-35340, train loss-0.6155, acc-0.8400, valid loss-0.5492, acc-0.8528, test loss-0.5525, acc-0.8513
Iter-35350, train loss-0.5649, acc-0.8800, valid loss-0.5492, acc-0.8528, test loss-0.5524, acc-0.8513
Iter-35360, train loss-0.5584, acc-0.8600, valid loss-0.5491, acc-0.8528, test loss-0.5523, acc-0.8514
Iter-35370, train loss-0.7508, acc-0.8000, valid loss-0.5490, acc-0.8530, test loss-0.5523, acc-0.8512
Iter-35380, train loss-0.4530, acc-0.9200, valid loss-0.5489, acc-0.8530, test loss-0.5522, acc-0.8514
Iter-35390, train loss-0.6712, acc-0.7800, valid loss-0.5488, acc-0.8528, test loss-0.5521, acc-0.8515
Iter-35400, train loss-0.5854, acc-0.8600, valid loss-0.5487, acc-0.8532, test loss-0.5520, acc-0.8514
Iter-35410, train loss-0.5620, acc-0.8800, valid loss-0.5487, acc-0.8532, test loss-0.5519, acc-0.8515
Iter-35420, train loss-0.6490, acc-0.8000, valid loss-0.5486, acc-0.8530, test loss-0.5518, acc-0.8516
Iter-35430, train loss-0.6340, acc-0.7800, valid loss-0.5485, acc-0.8534, test loss-0.5518, acc-0.8515
Iter-35440, train loss-0.4539, acc-0.8600, valid loss-0.5484, acc-0.8532, test loss-0.5517, acc-0.8515
Iter-35450, train loss-0.6580, acc-0.7800, valid loss-0.5483, acc-0.8534, test loss-0.5516, acc-0.8515
Iter-35460, train loss-0.4163, acc-0.8800, valid loss-0.5483, acc-0.8534, test loss-0.5515, acc-0.8517
Iter-35470, train loss-0.4681, acc-0.9000, valid loss-0.5482, acc-0.8534, test loss-0.5514, acc-0.8518
Iter-35480, train loss-0.6331, acc-0.8600, valid loss-0.5481, acc-0.8532, test loss-0.5514, acc-0.8518
Iter-35490, train loss-0.6913, acc-0.8000, valid loss-0.5480, acc-0.8532, test loss-0.5513, acc-0.8518
Iter-35500, train loss-0.6331, acc-0.8000, valid loss-0.5479, acc-0.8532, test loss-0.5512, acc-0.8519
Iter-35510, train loss-0.4828, acc-0.8600, valid loss-0.5478, acc-0.8532, test loss-0.5511, acc-0.8519
Iter-35520, train loss-0.3928, acc-0.9400, valid loss-0.5477, acc-0.8534, test loss-0.5511, acc-0.8519
Iter-35530, train loss-0.4802, acc-0.8800, valid loss-0.5477, acc-0.8534, test loss-0.5510, acc-0.8518
Iter-35540, train loss-0.3921, acc-0.9200, valid loss-0.5476, acc-0.8534, test loss-0.5509, acc-0.8520
Iter-35550, train loss-0.5773, acc-0.8200, valid loss-0.5475, acc-0.8534, test loss-0.5508, acc-0.8519
Iter-35560, train loss-0.6252, acc-0.7800, valid loss-0.5474, acc-0.8532, test loss-0.5507, acc-0.8518
Iter-35570, train loss-0.4161, acc-0.9000, valid loss-0.5473, acc-0.8532, test loss-0.5507, acc-0.8519
Iter-35580, train loss-0.5005, acc-0.9000, valid loss-0.5473, acc-0.8532, test loss-0.5506, acc-0.8520
Iter-35590, train loss-0.6899, acc-0.7400, valid loss-0.5472, acc-0.8536, test loss-0.5505, acc-0.8520
Iter-35600, train loss-0.3783, acc-0.9400, valid loss-0.5471, acc-0.8534, test loss-0.5504, acc-0.8521
Iter-35610, train loss-0.6571, acc-0.8800, valid loss-0.5470, acc-0.8532, test loss-0.5503, acc-0.8522
Iter-35620, train loss-0.4886, acc-0.8800, valid loss-0.5469, acc-0.8534, test loss-0.5502, acc-0.8523
Iter-35630, train loss-0.5328, acc-0.8000, valid loss-0.5469, acc-0.8536, test loss-0.5501, acc-0.8523
Iter-35640, train loss-0.6466, acc-0.8400, valid loss-0.5468, acc-0.8532, test loss-0.5501, acc-0.8522
Iter-35650, train loss-0.5171, acc-0.8600, valid loss-0.5467, acc-0.8532, test loss-0.5500, acc-0.8523
Iter-35660, train loss-0.7814, acc-0.7800, valid loss-0.5466, acc-0.8536, test loss-0.5499, acc-0.8522
Iter-35670, train loss-0.4417, acc-0.9200, valid loss-0.5465, acc-0.8536, test loss-0.5498, acc-0.8523
Iter-35680, train loss-0.5929, acc-0.8000, valid loss-0.5464, acc-0.8536, test loss-0.5497, acc-0.8522
Iter-35690, train loss-0.4727, acc-0.8800, valid loss-0.5464, acc-0.8536, test loss-0.5496, acc-0.8523
Iter-35700, train loss-0.5736, acc-0.8400, valid loss-0.5463, acc-0.8536, test loss-0.5496, acc-0.8522
Iter-35710, train loss-0.5154, acc-0.8600, valid loss-0.5462, acc-0.8536, test loss-0.5495, acc-0.8521
Iter-35720, train loss-0.6729, acc-0.8000, valid loss-0.5461, acc-0.8536, test loss-0.5494, acc-0.8520
Iter-35730, train loss-0.6025, acc-0.8000, valid loss-0.5460, acc-0.8536, test loss-0.5493, acc-0.8521
Iter-35740, train loss-0.5946, acc-0.8000, valid loss-0.5459, acc-0.8540, test loss-0.5492, acc-0.8522
Iter-35750, train loss-0.4715, acc-0.8800, valid loss-0.5459, acc-0.8536, test loss-0.5492, acc-0.8521
Iter-35760, train loss-0.5276, acc-0.8400, valid loss-0.5458, acc-0.8538, test loss-0.5491, acc-0.8522
Iter-35770, train loss-0.5177, acc-0.8800, valid loss-0.5457, acc-0.8538, test loss-0.5490, acc-0.8521
Iter-35780, train loss-0.5958, acc-0.7800, valid loss-0.5456, acc-0.8540, test loss-0.5489, acc-0.8521
Iter-35790, train loss-0.4656, acc-0.8800, valid loss-0.5456, acc-0.8542, test loss-0.5488, acc-0.8522
Iter-35800, train loss-0.3490, acc-0.9200, valid loss-0.5455, acc-0.8544, test loss-0.5488, acc-0.8521
Iter-35810, train loss-0.5138, acc-0.8600, valid loss-0.5454, acc-0.8544, test loss-0.5487, acc-0.8522
Iter-35820, train loss-0.4842, acc-0.9000, valid loss-0.5453, acc-0.8546, test loss-0.5486, acc-0.8523
Iter-35830, train loss-0.8617, acc-0.7200, valid loss-0.5452, acc-0.8546, test loss-0.5485, acc-0.8522
Iter-35840, train loss-0.5090, acc-0.8600, valid loss-0.5452, acc-0.8544, test loss-0.5484, acc-0.8522
Iter-35850, train loss-0.5830, acc-0.8400, valid loss-0.5451, acc-0.8546, test loss-0.5484, acc-0.8523
Iter-35860, train loss-0.6411, acc-0.8200, valid loss-0.5450, acc-0.8544, test loss-0.5483, acc-0.8522
Iter-35870, train loss-0.5413, acc-0.8400, valid loss-0.5449, acc-0.8546, test loss-0.5482, acc-0.8522
Iter-35880, train loss-0.5235, acc-0.8200, valid loss-0.5448, acc-0.8550, test loss-0.5481, acc-0.8524
Iter-35890, train loss-0.7204, acc-0.7800, valid loss-0.5447, acc-0.8550, test loss-0.5481, acc-0.8524
Iter-35900, train loss-0.7135, acc-0.7800, valid loss-0.5446, acc-0.8550, test loss-0.5480, acc-0.8525
Iter-35910, train loss-0.7857, acc-0.7600, valid loss-0.5446, acc-0.8550, test loss-0.5479, acc-0.8524
Iter-35920, train loss-0.5800, acc-0.8200, valid loss-0.5445, acc-0.8552, test loss-0.5478, acc-0.8525
Iter-35930, train loss-0.7297, acc-0.8000, valid loss-0.5444, acc-0.8552, test loss-0.5477, acc-0.8525
Iter-35940, train loss-0.8887, acc-0.7400, valid loss-0.5443, acc-0.8552, test loss-0.5476, acc-0.8526
Iter-35950, train loss-0.5038, acc-0.9000, valid loss-0.5442, acc-0.8552, test loss-0.5475, acc-0.8526
Iter-35960, train loss-0.5866, acc-0.8800, valid loss-0.5441, acc-0.8552, test loss-0.5474, acc-0.8526
Iter-35970, train loss-0.5374, acc-0.8000, valid loss-0.5441, acc-0.8552, test loss-0.5474, acc-0.8526
Iter-35980, train loss-0.5395, acc-0.8800, valid loss-0.5440, acc-0.8552, test loss-0.5473, acc-0.8526
Iter-35990, train loss-0.4881, acc-0.8800, valid loss-0.5439, acc-0.8552, test loss-0.5472, acc-0.8526
Iter-36000, train loss-0.6678, acc-0.8000, valid loss-0.5439, acc-0.8552, test loss-0.5472, acc-0.8527
Iter-36010, train loss-0.5716, acc-0.8200, valid loss-0.5438, acc-0.8552, test loss-0.5471, acc-0.8527
Iter-36020, train loss-0.5998, acc-0.7800, valid loss-0.5437, acc-0.8552, test loss-0.5470, acc-0.8528
Iter-36030, train loss-0.5978, acc-0.8200, valid loss-0.5436, acc-0.8554, test loss-0.5469, acc-0.8528
Iter-36040, train loss-0.5962, acc-0.8600, valid loss-0.5435, acc-0.8554, test loss-0.5468, acc-0.8526
Iter-36050, train loss-0.6517, acc-0.8200, valid loss-0.5435, acc-0.8554, test loss-0.5467, acc-0.8527
Iter-36060, train loss-0.5819, acc-0.8400, valid loss-0.5434, acc-0.8554, test loss-0.5466, acc-0.8527
Iter-36070, train loss-0.4198, acc-0.9000, valid loss-0.5433, acc-0.8554, test loss-0.5466, acc-0.8526
Iter-36080, train loss-0.4257, acc-0.8800, valid loss-0.5432, acc-0.8554, test loss-0.5465, acc-0.8527
Iter-36090, train loss-0.6215, acc-0.8200, valid loss-0.5431, acc-0.8554, test loss-0.5464, acc-0.8528
Iter-36100, train loss-0.6531, acc-0.8000, valid loss-0.5431, acc-0.8554, test loss-0.5463, acc-0.8527
Iter-36110, train loss-0.4430, acc-0.9400, valid loss-0.5430, acc-0.8554, test loss-0.5462, acc-0.8526
Iter-36120, train loss-0.4433, acc-0.9200, valid loss-0.5429, acc-0.8554, test loss-0.5462, acc-0.8527
Iter-36130, train loss-0.4181, acc-0.9200, valid loss-0.5428, acc-0.8554, test loss-0.5461, acc-0.8526
Iter-36140, train loss-0.6141, acc-0.7600, valid loss-0.5427, acc-0.8554, test loss-0.5460, acc-0.8525
Iter-36150, train loss-0.6217, acc-0.7800, valid loss-0.5426, acc-0.8554, test loss-0.5459, acc-0.8526
Iter-36160, train loss-0.8417, acc-0.8000, valid loss-0.5425, acc-0.8554, test loss-0.5458, acc-0.8526
Iter-36170, train loss-0.6548, acc-0.8000, valid loss-0.5425, acc-0.8554, test loss-0.5457, acc-0.8527
Iter-36180, train loss-0.6178, acc-0.8400, valid loss-0.5424, acc-0.8554, test loss-0.5457, acc-0.8527
Iter-36190, train loss-0.4876, acc-0.9200, valid loss-0.5423, acc-0.8554, test loss-0.5456, acc-0.8527
Iter-36200, train loss-0.5884, acc-0.8600, valid loss-0.5422, acc-0.8554, test loss-0.5455, acc-0.8527
Iter-36210, train loss-0.4602, acc-0.8600, valid loss-0.5421, acc-0.8552, test loss-0.5454, acc-0.8527
Iter-36220, train loss-0.6307, acc-0.8200, valid loss-0.5420, acc-0.8552, test loss-0.5453, acc-0.8525
Iter-36230, train loss-0.5493, acc-0.8200, valid loss-0.5420, acc-0.8550, test loss-0.5453, acc-0.8528
Iter-36240, train loss-0.7353, acc-0.7000, valid loss-0.5419, acc-0.8548, test loss-0.5452, acc-0.8526
Iter-36250, train loss-0.5294, acc-0.8400, valid loss-0.5418, acc-0.8548, test loss-0.5451, acc-0.8527
Iter-36260, train loss-0.6567, acc-0.8600, valid loss-0.5417, acc-0.8550, test loss-0.5450, acc-0.8527
Iter-36270, train loss-0.3985, acc-0.8800, valid loss-0.5416, acc-0.8550, test loss-0.5449, acc-0.8527
Iter-36280, train loss-0.3593, acc-0.9000, valid loss-0.5416, acc-0.8550, test loss-0.5448, acc-0.8527
Iter-36290, train loss-0.5590, acc-0.8800, valid loss-0.5415, acc-0.8550, test loss-0.5448, acc-0.8526
Iter-36300, train loss-0.4494, acc-0.9200, valid loss-0.5414, acc-0.8550, test loss-0.5447, acc-0.8526
Iter-36310, train loss-0.5843, acc-0.8800, valid loss-0.5413, acc-0.8550, test loss-0.5446, acc-0.8526
Iter-36320, train loss-0.3949, acc-0.9000, valid loss-0.5412, acc-0.8550, test loss-0.5445, acc-0.8527
Iter-36330, train loss-0.6625, acc-0.8000, valid loss-0.5411, acc-0.8552, test loss-0.5444, acc-0.8526
Iter-36340, train loss-0.4717, acc-0.8800, valid loss-0.5411, acc-0.8554, test loss-0.5443, acc-0.8527
Iter-36350, train loss-0.5280, acc-0.8800, valid loss-0.5410, acc-0.8556, test loss-0.5442, acc-0.8528
Iter-36360, train loss-0.4878, acc-0.9000, valid loss-0.5409, acc-0.8554, test loss-0.5441, acc-0.8528
Iter-36370, train loss-0.4922, acc-0.9200, valid loss-0.5408, acc-0.8556, test loss-0.5441, acc-0.8529
Iter-36380, train loss-0.5665, acc-0.8800, valid loss-0.5408, acc-0.8556, test loss-0.5440, acc-0.8529
Iter-36390, train loss-0.5510, acc-0.8200, valid loss-0.5407, acc-0.8556, test loss-0.5439, acc-0.8529
Iter-36400, train loss-0.4639, acc-0.8800, valid loss-0.5406, acc-0.8556, test loss-0.5439, acc-0.8530
Iter-36410, train loss-0.5242, acc-0.8200, valid loss-0.5405, acc-0.8556, test loss-0.5438, acc-0.8530
Iter-36420, train loss-0.5938, acc-0.7600, valid loss-0.5405, acc-0.8556, test loss-0.5437, acc-0.8529
Iter-36430, train loss-0.6361, acc-0.8000, valid loss-0.5404, acc-0.8556, test loss-0.5436, acc-0.8529
Iter-36440, train loss-0.5595, acc-0.8600, valid loss-0.5403, acc-0.8556, test loss-0.5436, acc-0.8529
Iter-36450, train loss-0.5350, acc-0.8800, valid loss-0.5402, acc-0.8558, test loss-0.5435, acc-0.8528
Iter-36460, train loss-0.4358, acc-0.8800, valid loss-0.5401, acc-0.8558, test loss-0.5434, acc-0.8528
Iter-36470, train loss-0.3717, acc-0.9200, valid loss-0.5401, acc-0.8558, test loss-0.5433, acc-0.8528
Iter-36480, train loss-0.5934, acc-0.8600, valid loss-0.5400, acc-0.8558, test loss-0.5432, acc-0.8528
Iter-36490, train loss-0.4602, acc-0.8600, valid loss-0.5399, acc-0.8558, test loss-0.5431, acc-0.8528
Iter-36500, train loss-0.5385, acc-0.8400, valid loss-0.5398, acc-0.8556, test loss-0.5431, acc-0.8528
Iter-36510, train loss-0.6881, acc-0.8400, valid loss-0.5397, acc-0.8556, test loss-0.5430, acc-0.8530
Iter-36520, train loss-0.7330, acc-0.8200, valid loss-0.5396, acc-0.8556, test loss-0.5429, acc-0.8530
Iter-36530, train loss-0.5683, acc-0.7800, valid loss-0.5395, acc-0.8556, test loss-0.5428, acc-0.8530
Iter-36540, train loss-0.5916, acc-0.8400, valid loss-0.5395, acc-0.8556, test loss-0.5427, acc-0.8531
Iter-36550, train loss-0.6086, acc-0.8400, valid loss-0.5394, acc-0.8558, test loss-0.5427, acc-0.8532
Iter-36560, train loss-0.5813, acc-0.8800, valid loss-0.5393, acc-0.8560, test loss-0.5426, acc-0.8532
Iter-36570, train loss-0.4076, acc-0.8800, valid loss-0.5392, acc-0.8558, test loss-0.5425, acc-0.8533
Iter-36580, train loss-0.4658, acc-0.9200, valid loss-0.5391, acc-0.8560, test loss-0.5424, acc-0.8533
Iter-36590, train loss-0.6604, acc-0.8400, valid loss-0.5391, acc-0.8560, test loss-0.5423, acc-0.8531
Iter-36600, train loss-0.5518, acc-0.8600, valid loss-0.5390, acc-0.8560, test loss-0.5423, acc-0.8532
Iter-36610, train loss-0.4519, acc-0.9200, valid loss-0.5389, acc-0.8560, test loss-0.5422, acc-0.8531
Iter-36620, train loss-0.7563, acc-0.8000, valid loss-0.5389, acc-0.8560, test loss-0.5421, acc-0.8532
Iter-36630, train loss-0.5072, acc-0.8400, valid loss-0.5388, acc-0.8560, test loss-0.5421, acc-0.8531
Iter-36640, train loss-0.5709, acc-0.8400, valid loss-0.5387, acc-0.8560, test loss-0.5420, acc-0.8531
Iter-36650, train loss-0.5002, acc-0.8600, valid loss-0.5386, acc-0.8560, test loss-0.5419, acc-0.8533
Iter-36660, train loss-0.7021, acc-0.7800, valid loss-0.5385, acc-0.8558, test loss-0.5418, acc-0.8534
Iter-36670, train loss-0.6696, acc-0.8000, valid loss-0.5385, acc-0.8558, test loss-0.5417, acc-0.8536
Iter-36680, train loss-0.6190, acc-0.8600, valid loss-0.5384, acc-0.8558, test loss-0.5417, acc-0.8536
Iter-36690, train loss-0.3797, acc-0.9400, valid loss-0.5383, acc-0.8560, test loss-0.5416, acc-0.8536
Iter-36700, train loss-0.5412, acc-0.8600, valid loss-0.5382, acc-0.8562, test loss-0.5415, acc-0.8536
Iter-36710, train loss-0.4670, acc-0.8800, valid loss-0.5382, acc-0.8562, test loss-0.5414, acc-0.8536
Iter-36720, train loss-0.4898, acc-0.8800, valid loss-0.5381, acc-0.8560, test loss-0.5414, acc-0.8534
Iter-36730, train loss-0.5782, acc-0.8200, valid loss-0.5380, acc-0.8562, test loss-0.5413, acc-0.8535
Iter-36740, train loss-0.4122, acc-0.8600, valid loss-0.5379, acc-0.8560, test loss-0.5412, acc-0.8535
Iter-36750, train loss-0.7292, acc-0.8000, valid loss-0.5379, acc-0.8560, test loss-0.5411, acc-0.8536
Iter-36760, train loss-0.6551, acc-0.8800, valid loss-0.5378, acc-0.8562, test loss-0.5411, acc-0.8535
Iter-36770, train loss-0.3997, acc-0.8800, valid loss-0.5377, acc-0.8562, test loss-0.5410, acc-0.8535
Iter-36780, train loss-0.3940, acc-0.9200, valid loss-0.5376, acc-0.8562, test loss-0.5409, acc-0.8536
Iter-36790, train loss-0.8366, acc-0.8000, valid loss-0.5375, acc-0.8562, test loss-0.5408, acc-0.8537
Iter-36800, train loss-0.5011, acc-0.8800, valid loss-0.5375, acc-0.8564, test loss-0.5407, acc-0.8534
Iter-36810, train loss-0.4888, acc-0.8600, valid loss-0.5374, acc-0.8564, test loss-0.5407, acc-0.8536
Iter-36820, train loss-0.5677, acc-0.8000, valid loss-0.5373, acc-0.8562, test loss-0.5406, acc-0.8535
Iter-36830, train loss-0.6965, acc-0.8400, valid loss-0.5372, acc-0.8560, test loss-0.5405, acc-0.8535
Iter-36840, train loss-0.6213, acc-0.7600, valid loss-0.5371, acc-0.8560, test loss-0.5404, acc-0.8535
Iter-36850, train loss-0.4146, acc-0.9200, valid loss-0.5371, acc-0.8560, test loss-0.5403, acc-0.8535
Iter-36860, train loss-0.6197, acc-0.8200, valid loss-0.5370, acc-0.8562, test loss-0.5402, acc-0.8535
Iter-36870, train loss-0.4368, acc-0.9000, valid loss-0.5369, acc-0.8562, test loss-0.5402, acc-0.8538
Iter-36880, train loss-0.6217, acc-0.8000, valid loss-0.5368, acc-0.8560, test loss-0.5401, acc-0.8538
Iter-36890, train loss-0.6109, acc-0.8200, valid loss-0.5367, acc-0.8560, test loss-0.5400, acc-0.8538
Iter-36900, train loss-0.5645, acc-0.8400, valid loss-0.5367, acc-0.8560, test loss-0.5399, acc-0.8541
Iter-36910, train loss-0.6539, acc-0.8400, valid loss-0.5366, acc-0.8560, test loss-0.5399, acc-0.8540
Iter-36920, train loss-0.3484, acc-0.9200, valid loss-0.5365, acc-0.8560, test loss-0.5398, acc-0.8540
Iter-36930, train loss-0.4718, acc-0.8600, valid loss-0.5364, acc-0.8560, test loss-0.5397, acc-0.8540
Iter-36940, train loss-0.5211, acc-0.8000, valid loss-0.5363, acc-0.8560, test loss-0.5396, acc-0.8539
Iter-36950, train loss-0.4803, acc-0.9000, valid loss-0.5363, acc-0.8560, test loss-0.5396, acc-0.8540
Iter-36960, train loss-0.9109, acc-0.8000, valid loss-0.5362, acc-0.8558, test loss-0.5395, acc-0.8540
Iter-36970, train loss-0.6656, acc-0.8400, valid loss-0.5361, acc-0.8560, test loss-0.5394, acc-0.8542
Iter-36980, train loss-0.4951, acc-0.8800, valid loss-0.5361, acc-0.8562, test loss-0.5393, acc-0.8541
Iter-36990, train loss-0.4956, acc-0.8600, valid loss-0.5360, acc-0.8560, test loss-0.5392, acc-0.8542
Iter-37000, train loss-0.5368, acc-0.8400, valid loss-0.5359, acc-0.8560, test loss-0.5392, acc-0.8542
Iter-37010, train loss-0.4608, acc-0.9000, valid loss-0.5358, acc-0.8560, test loss-0.5391, acc-0.8542
Iter-37020, train loss-0.7271, acc-0.8000, valid loss-0.5357, acc-0.8560, test loss-0.5390, acc-0.8542
Iter-37030, train loss-0.5327, acc-0.8600, valid loss-0.5357, acc-0.8560, test loss-0.5389, acc-0.8542
Iter-37040, train loss-0.6954, acc-0.8400, valid loss-0.5356, acc-0.8560, test loss-0.5388, acc-0.8542
Iter-37050, train loss-0.6356, acc-0.8200, valid loss-0.5355, acc-0.8560, test loss-0.5388, acc-0.8542
Iter-37060, train loss-0.4862, acc-0.8200, valid loss-0.5354, acc-0.8560, test loss-0.5387, acc-0.8542
Iter-37070, train loss-0.4637, acc-0.8400, valid loss-0.5353, acc-0.8560, test loss-0.5386, acc-0.8542
Iter-37080, train loss-0.6185, acc-0.8200, valid loss-0.5352, acc-0.8560, test loss-0.5385, acc-0.8543
Iter-37090, train loss-0.6000, acc-0.8800, valid loss-0.5352, acc-0.8560, test loss-0.5384, acc-0.8542
Iter-37100, train loss-0.4585, acc-0.8600, valid loss-0.5351, acc-0.8560, test loss-0.5383, acc-0.8543
Iter-37110, train loss-0.5234, acc-0.8600, valid loss-0.5350, acc-0.8560, test loss-0.5382, acc-0.8543
Iter-37120, train loss-0.4369, acc-0.9000, valid loss-0.5349, acc-0.8560, test loss-0.5381, acc-0.8543
Iter-37130, train loss-0.5590, acc-0.8200, valid loss-0.5348, acc-0.8560, test loss-0.5381, acc-0.8542
Iter-37140, train loss-0.5303, acc-0.8600, valid loss-0.5347, acc-0.8560, test loss-0.5380, acc-0.8542
Iter-37150, train loss-0.3666, acc-0.9000, valid loss-0.5347, acc-0.8560, test loss-0.5379, acc-0.8543
Iter-37160, train loss-0.4690, acc-0.9400, valid loss-0.5346, acc-0.8560, test loss-0.5378, acc-0.8543
Iter-37170, train loss-0.5604, acc-0.8400, valid loss-0.5345, acc-0.8560, test loss-0.5378, acc-0.8543
Iter-37180, train loss-0.6617, acc-0.8000, valid loss-0.5344, acc-0.8562, test loss-0.5377, acc-0.8543
Iter-37190, train loss-0.4503, acc-0.9200, valid loss-0.5343, acc-0.8560, test loss-0.5376, acc-0.8543
Iter-37200, train loss-0.7205, acc-0.8000, valid loss-0.5343, acc-0.8560, test loss-0.5375, acc-0.8544
Iter-37210, train loss-0.8623, acc-0.7200, valid loss-0.5342, acc-0.8562, test loss-0.5374, acc-0.8544
Iter-37220, train loss-0.3453, acc-0.9400, valid loss-0.5341, acc-0.8562, test loss-0.5373, acc-0.8544
Iter-37230, train loss-0.7822, acc-0.7400, valid loss-0.5340, acc-0.8562, test loss-0.5373, acc-0.8544
Iter-37240, train loss-0.4304, acc-0.9000, valid loss-0.5339, acc-0.8562, test loss-0.5372, acc-0.8544
Iter-37250, train loss-0.4393, acc-0.9200, valid loss-0.5339, acc-0.8562, test loss-0.5371, acc-0.8544
Iter-37260, train loss-0.4622, acc-0.8400, valid loss-0.5338, acc-0.8562, test loss-0.5371, acc-0.8544
Iter-37270, train loss-0.3977, acc-0.9200, valid loss-0.5338, acc-0.8562, test loss-0.5370, acc-0.8543
Iter-37280, train loss-0.4440, acc-0.8800, valid loss-0.5337, acc-0.8562, test loss-0.5369, acc-0.8543
Iter-37290, train loss-0.8001, acc-0.7600, valid loss-0.5336, acc-0.8562, test loss-0.5369, acc-0.8543
Iter-37300, train loss-0.5672, acc-0.8200, valid loss-0.5335, acc-0.8562, test loss-0.5368, acc-0.8545
Iter-37310, train loss-0.4408, acc-0.9200, valid loss-0.5334, acc-0.8562, test loss-0.5367, acc-0.8544
Iter-37320, train loss-0.4223, acc-0.8800, valid loss-0.5334, acc-0.8560, test loss-0.5367, acc-0.8543
Iter-37330, train loss-0.5113, acc-0.8400, valid loss-0.5333, acc-0.8560, test loss-0.5366, acc-0.8545
Iter-37340, train loss-0.3651, acc-0.9400, valid loss-0.5332, acc-0.8560, test loss-0.5365, acc-0.8545
Iter-37350, train loss-0.5209, acc-0.8400, valid loss-0.5331, acc-0.8560, test loss-0.5364, acc-0.8546
Iter-37360, train loss-0.5502, acc-0.8800, valid loss-0.5330, acc-0.8560, test loss-0.5363, acc-0.8546
Iter-37370, train loss-0.4378, acc-0.8800, valid loss-0.5329, acc-0.8560, test loss-0.5363, acc-0.8547
Iter-37380, train loss-0.7151, acc-0.8000, valid loss-0.5329, acc-0.8560, test loss-0.5362, acc-0.8548
Iter-37390, train loss-0.6049, acc-0.8600, valid loss-0.5328, acc-0.8560, test loss-0.5361, acc-0.8549
Iter-37400, train loss-0.3375, acc-0.9000, valid loss-0.5327, acc-0.8560, test loss-0.5360, acc-0.8550
Iter-37410, train loss-0.8254, acc-0.7200, valid loss-0.5326, acc-0.8562, test loss-0.5359, acc-0.8550
Iter-37420, train loss-0.6004, acc-0.8400, valid loss-0.5326, acc-0.8562, test loss-0.5359, acc-0.8550
Iter-37430, train loss-0.7370, acc-0.8200, valid loss-0.5325, acc-0.8560, test loss-0.5358, acc-0.8551
Iter-37440, train loss-0.3077, acc-0.9200, valid loss-0.5324, acc-0.8560, test loss-0.5357, acc-0.8550
Iter-37450, train loss-0.6115, acc-0.8200, valid loss-0.5323, acc-0.8564, test loss-0.5357, acc-0.8551
Iter-37460, train loss-0.5936, acc-0.8800, valid loss-0.5323, acc-0.8566, test loss-0.5356, acc-0.8551
Iter-37470, train loss-0.5847, acc-0.8400, valid loss-0.5322, acc-0.8564, test loss-0.5355, acc-0.8551
Iter-37480, train loss-0.6587, acc-0.8200, valid loss-0.5321, acc-0.8564, test loss-0.5355, acc-0.8550
Iter-37490, train loss-0.5692, acc-0.7800, valid loss-0.5320, acc-0.8564, test loss-0.5354, acc-0.8548
Iter-37500, train loss-0.5615, acc-0.8600, valid loss-0.5319, acc-0.8564, test loss-0.5353, acc-0.8550
Iter-37510, train loss-0.4004, acc-0.8600, valid loss-0.5319, acc-0.8564, test loss-0.5352, acc-0.8549
Iter-37520, train loss-0.7821, acc-0.8000, valid loss-0.5318, acc-0.8564, test loss-0.5351, acc-0.8549
Iter-37530, train loss-0.5032, acc-0.8400, valid loss-0.5317, acc-0.8564, test loss-0.5351, acc-0.8548
Iter-37540, train loss-0.7168, acc-0.8200, valid loss-0.5316, acc-0.8564, test loss-0.5350, acc-0.8549
Iter-37550, train loss-0.5173, acc-0.8600, valid loss-0.5315, acc-0.8564, test loss-0.5349, acc-0.8550
Iter-37560, train loss-0.4357, acc-0.8800, valid loss-0.5315, acc-0.8564, test loss-0.5348, acc-0.8550
Iter-37570, train loss-0.7473, acc-0.8600, valid loss-0.5314, acc-0.8564, test loss-0.5348, acc-0.8550
Iter-37580, train loss-0.6260, acc-0.7800, valid loss-0.5313, acc-0.8564, test loss-0.5347, acc-0.8550
Iter-37590, train loss-0.6098, acc-0.8600, valid loss-0.5312, acc-0.8564, test loss-0.5346, acc-0.8548
Iter-37600, train loss-0.4921, acc-0.8800, valid loss-0.5312, acc-0.8564, test loss-0.5346, acc-0.8548
Iter-37610, train loss-0.5237, acc-0.8600, valid loss-0.5311, acc-0.8564, test loss-0.5345, acc-0.8548
Iter-37620, train loss-0.6394, acc-0.8000, valid loss-0.5310, acc-0.8564, test loss-0.5344, acc-0.8549
Iter-37630, train loss-0.5925, acc-0.8200, valid loss-0.5309, acc-0.8564, test loss-0.5343, acc-0.8551
Iter-37640, train loss-0.5455, acc-0.8600, valid loss-0.5309, acc-0.8564, test loss-0.5343, acc-0.8550
Iter-37650, train loss-0.6501, acc-0.8600, valid loss-0.5308, acc-0.8564, test loss-0.5342, acc-0.8552
Iter-37660, train loss-0.4076, acc-0.8800, valid loss-0.5307, acc-0.8564, test loss-0.5341, acc-0.8552
Iter-37670, train loss-0.5452, acc-0.8200, valid loss-0.5306, acc-0.8564, test loss-0.5340, acc-0.8552
Iter-37680, train loss-0.4068, acc-0.9000, valid loss-0.5306, acc-0.8564, test loss-0.5340, acc-0.8551
Iter-37690, train loss-0.7680, acc-0.7800, valid loss-0.5305, acc-0.8564, test loss-0.5339, acc-0.8551
Iter-37700, train loss-0.4442, acc-0.9400, valid loss-0.5304, acc-0.8566, test loss-0.5338, acc-0.8551
Iter-37710, train loss-0.5114, acc-0.8800, valid loss-0.5304, acc-0.8566, test loss-0.5337, acc-0.8551
Iter-37720, train loss-0.4796, acc-0.9000, valid loss-0.5303, acc-0.8566, test loss-0.5336, acc-0.8551
Iter-37730, train loss-0.4854, acc-0.9000, valid loss-0.5302, acc-0.8566, test loss-0.5336, acc-0.8550
Iter-37740, train loss-0.4674, acc-0.8600, valid loss-0.5302, acc-0.8566, test loss-0.5335, acc-0.8552
Iter-37750, train loss-0.6082, acc-0.8200, valid loss-0.5301, acc-0.8566, test loss-0.5334, acc-0.8552
Iter-37760, train loss-0.5686, acc-0.8200, valid loss-0.5300, acc-0.8566, test loss-0.5333, acc-0.8551
Iter-37770, train loss-0.5740, acc-0.8400, valid loss-0.5299, acc-0.8568, test loss-0.5332, acc-0.8552
Iter-37780, train loss-0.3651, acc-0.9200, valid loss-0.5298, acc-0.8568, test loss-0.5332, acc-0.8552
Iter-37790, train loss-0.5986, acc-0.8000, valid loss-0.5298, acc-0.8568, test loss-0.5331, acc-0.8552
Iter-37800, train loss-0.4717, acc-0.9200, valid loss-0.5297, acc-0.8568, test loss-0.5330, acc-0.8552
Iter-37810, train loss-0.7175, acc-0.7600, valid loss-0.5296, acc-0.8570, test loss-0.5330, acc-0.8553
Iter-37820, train loss-0.5366, acc-0.9000, valid loss-0.5296, acc-0.8570, test loss-0.5329, acc-0.8552
Iter-37830, train loss-0.3822, acc-0.9400, valid loss-0.5295, acc-0.8570, test loss-0.5328, acc-0.8552
Iter-37840, train loss-0.4651, acc-0.9000, valid loss-0.5294, acc-0.8568, test loss-0.5327, acc-0.8551
Iter-37850, train loss-0.6042, acc-0.8400, valid loss-0.5293, acc-0.8568, test loss-0.5326, acc-0.8551
Iter-37860, train loss-0.5010, acc-0.8800, valid loss-0.5293, acc-0.8568, test loss-0.5326, acc-0.8553
Iter-37870, train loss-0.5719, acc-0.8600, valid loss-0.5292, acc-0.8568, test loss-0.5325, acc-0.8553
Iter-37880, train loss-0.7682, acc-0.8200, valid loss-0.5291, acc-0.8570, test loss-0.5324, acc-0.8555
Iter-37890, train loss-0.5961, acc-0.8400, valid loss-0.5290, acc-0.8570, test loss-0.5323, acc-0.8555
Iter-37900, train loss-0.4787, acc-0.9000, valid loss-0.5290, acc-0.8572, test loss-0.5323, acc-0.8554
Iter-37910, train loss-0.8092, acc-0.7600, valid loss-0.5289, acc-0.8572, test loss-0.5322, acc-0.8554
Iter-37920, train loss-0.4739, acc-0.9200, valid loss-0.5289, acc-0.8572, test loss-0.5321, acc-0.8555
Iter-37930, train loss-0.5897, acc-0.7600, valid loss-0.5288, acc-0.8572, test loss-0.5321, acc-0.8554
Iter-37940, train loss-0.3922, acc-0.9200, valid loss-0.5287, acc-0.8572, test loss-0.5320, acc-0.8554
Iter-37950, train loss-0.3979, acc-0.9200, valid loss-0.5286, acc-0.8572, test loss-0.5319, acc-0.8554
Iter-37960, train loss-0.3724, acc-0.9400, valid loss-0.5285, acc-0.8572, test loss-0.5319, acc-0.8552
Iter-37970, train loss-0.3879, acc-0.9200, valid loss-0.5285, acc-0.8570, test loss-0.5318, acc-0.8552
Iter-37980, train loss-0.5893, acc-0.7800, valid loss-0.5284, acc-0.8570, test loss-0.5317, acc-0.8553
Iter-37990, train loss-0.5230, acc-0.8400, valid loss-0.5283, acc-0.8570, test loss-0.5316, acc-0.8555
Iter-38000, train loss-0.5811, acc-0.8200, valid loss-0.5282, acc-0.8572, test loss-0.5316, acc-0.8555
Iter-38010, train loss-0.5415, acc-0.7600, valid loss-0.5282, acc-0.8574, test loss-0.5315, acc-0.8555
Iter-38020, train loss-0.6900, acc-0.8200, valid loss-0.5281, acc-0.8574, test loss-0.5314, acc-0.8555
Iter-38030, train loss-0.4948, acc-0.9200, valid loss-0.5280, acc-0.8574, test loss-0.5313, acc-0.8555
Iter-38040, train loss-0.4909, acc-0.8600, valid loss-0.5279, acc-0.8574, test loss-0.5313, acc-0.8555
Iter-38050, train loss-0.6843, acc-0.8400, valid loss-0.5278, acc-0.8574, test loss-0.5312, acc-0.8555
Iter-38060, train loss-0.5616, acc-0.7800, valid loss-0.5278, acc-0.8574, test loss-0.5311, acc-0.8555
Iter-38070, train loss-0.5526, acc-0.8600, valid loss-0.5277, acc-0.8574, test loss-0.5310, acc-0.8555
Iter-38080, train loss-0.5140, acc-0.8800, valid loss-0.5276, acc-0.8574, test loss-0.5310, acc-0.8556
Iter-38090, train loss-0.7050, acc-0.8400, valid loss-0.5275, acc-0.8574, test loss-0.5309, acc-0.8557
Iter-38100, train loss-0.5744, acc-0.8400, valid loss-0.5274, acc-0.8574, test loss-0.5308, acc-0.8557
Iter-38110, train loss-0.7607, acc-0.8200, valid loss-0.5273, acc-0.8574, test loss-0.5307, acc-0.8555
Iter-38120, train loss-0.4922, acc-0.9000, valid loss-0.5273, acc-0.8576, test loss-0.5306, acc-0.8557
Iter-38130, train loss-0.4535, acc-0.9200, valid loss-0.5272, acc-0.8578, test loss-0.5306, acc-0.8556
Iter-38140, train loss-0.5635, acc-0.8400, valid loss-0.5271, acc-0.8578, test loss-0.5304, acc-0.8556
Iter-38150, train loss-0.5145, acc-0.8400, valid loss-0.5270, acc-0.8578, test loss-0.5304, acc-0.8556
Iter-38160, train loss-0.5762, acc-0.8800, valid loss-0.5270, acc-0.8578, test loss-0.5303, acc-0.8557
Iter-38170, train loss-0.6435, acc-0.8400, valid loss-0.5269, acc-0.8578, test loss-0.5302, acc-0.8557
Iter-38180, train loss-0.4008, acc-0.8800, valid loss-0.5268, acc-0.8580, test loss-0.5302, acc-0.8557
Iter-38190, train loss-0.3787, acc-0.9200, valid loss-0.5267, acc-0.8580, test loss-0.5301, acc-0.8556
Iter-38200, train loss-0.5241, acc-0.8400, valid loss-0.5266, acc-0.8582, test loss-0.5300, acc-0.8557
Iter-38210, train loss-0.3776, acc-0.9200, valid loss-0.5266, acc-0.8582, test loss-0.5299, acc-0.8557
Iter-38220, train loss-0.4753, acc-0.8800, valid loss-0.5265, acc-0.8582, test loss-0.5299, acc-0.8558
Iter-38230, train loss-0.5264, acc-0.8600, valid loss-0.5264, acc-0.8582, test loss-0.5298, acc-0.8557
Iter-38240, train loss-0.5383, acc-0.8600, valid loss-0.5263, acc-0.8582, test loss-0.5297, acc-0.8557
Iter-38250, train loss-0.5143, acc-0.8400, valid loss-0.5263, acc-0.8584, test loss-0.5297, acc-0.8559
Iter-38260, train loss-0.5577, acc-0.8000, valid loss-0.5262, acc-0.8582, test loss-0.5296, acc-0.8561
Iter-38270, train loss-0.4052, acc-0.8800, valid loss-0.5261, acc-0.8582, test loss-0.5295, acc-0.8561
Iter-38280, train loss-0.3935, acc-0.9200, valid loss-0.5260, acc-0.8582, test loss-0.5294, acc-0.8561
Iter-38290, train loss-0.5320, acc-0.8400, valid loss-0.5259, acc-0.8582, test loss-0.5293, acc-0.8561
Iter-38300, train loss-0.5525, acc-0.8200, valid loss-0.5259, acc-0.8582, test loss-0.5293, acc-0.8561
Iter-38310, train loss-0.4580, acc-0.8400, valid loss-0.5258, acc-0.8582, test loss-0.5292, acc-0.8561
Iter-38320, train loss-0.4678, acc-0.9200, valid loss-0.5257, acc-0.8582, test loss-0.5291, acc-0.8561
Iter-38330, train loss-0.8075, acc-0.7800, valid loss-0.5257, acc-0.8582, test loss-0.5290, acc-0.8561
Iter-38340, train loss-0.5293, acc-0.8800, valid loss-0.5256, acc-0.8582, test loss-0.5290, acc-0.8561
Iter-38350, train loss-0.5743, acc-0.8400, valid loss-0.5255, acc-0.8582, test loss-0.5289, acc-0.8561
Iter-38360, train loss-0.5113, acc-0.8600, valid loss-0.5254, acc-0.8582, test loss-0.5288, acc-0.8562
Iter-38370, train loss-0.4353, acc-0.8800, valid loss-0.5254, acc-0.8582, test loss-0.5288, acc-0.8562
Iter-38380, train loss-0.5078, acc-0.9000, valid loss-0.5253, acc-0.8582, test loss-0.5287, acc-0.8562
Iter-38390, train loss-0.5596, acc-0.7600, valid loss-0.5252, acc-0.8582, test loss-0.5286, acc-0.8562
Iter-38400, train loss-0.5791, acc-0.8400, valid loss-0.5251, acc-0.8582, test loss-0.5286, acc-0.8561
Iter-38410, train loss-0.5662, acc-0.8400, valid loss-0.5251, acc-0.8584, test loss-0.5285, acc-0.8561
Iter-38420, train loss-0.5540, acc-0.8400, valid loss-0.5250, acc-0.8586, test loss-0.5284, acc-0.8562
Iter-38430, train loss-0.6515, acc-0.7600, valid loss-0.5249, acc-0.8586, test loss-0.5284, acc-0.8563
Iter-38440, train loss-0.3843, acc-0.9600, valid loss-0.5249, acc-0.8586, test loss-0.5283, acc-0.8561
Iter-38450, train loss-0.5155, acc-0.8400, valid loss-0.5248, acc-0.8586, test loss-0.5282, acc-0.8561
Iter-38460, train loss-0.6033, acc-0.8200, valid loss-0.5247, acc-0.8586, test loss-0.5282, acc-0.8559
Iter-38470, train loss-0.7372, acc-0.8000, valid loss-0.5247, acc-0.8586, test loss-0.5281, acc-0.8561
Iter-38480, train loss-0.5612, acc-0.8600, valid loss-0.5246, acc-0.8586, test loss-0.5280, acc-0.8560
Iter-38490, train loss-0.4360, acc-0.8400, valid loss-0.5245, acc-0.8586, test loss-0.5279, acc-0.8559
Iter-38500, train loss-0.4522, acc-0.9000, valid loss-0.5244, acc-0.8586, test loss-0.5279, acc-0.8559
Iter-38510, train loss-0.3738, acc-0.9000, valid loss-0.5243, acc-0.8586, test loss-0.5278, acc-0.8559
Iter-38520, train loss-0.6071, acc-0.8200, valid loss-0.5243, acc-0.8586, test loss-0.5277, acc-0.8560
Iter-38530, train loss-0.6640, acc-0.8000, valid loss-0.5242, acc-0.8586, test loss-0.5277, acc-0.8560
Iter-38540, train loss-0.4897, acc-0.8400, valid loss-0.5241, acc-0.8586, test loss-0.5276, acc-0.8559
Iter-38550, train loss-0.5616, acc-0.8600, valid loss-0.5241, acc-0.8586, test loss-0.5275, acc-0.8561
Iter-38560, train loss-0.6654, acc-0.7600, valid loss-0.5240, acc-0.8586, test loss-0.5274, acc-0.8561
Iter-38570, train loss-0.4933, acc-0.8400, valid loss-0.5239, acc-0.8586, test loss-0.5274, acc-0.8561
Iter-38580, train loss-0.4615, acc-0.8800, valid loss-0.5239, acc-0.8586, test loss-0.5273, acc-0.8563
Iter-38590, train loss-0.6427, acc-0.8200, valid loss-0.5238, acc-0.8586, test loss-0.5272, acc-0.8563
Iter-38600, train loss-0.5446, acc-0.8200, valid loss-0.5238, acc-0.8586, test loss-0.5272, acc-0.8564
Iter-38610, train loss-0.5636, acc-0.8200, valid loss-0.5237, acc-0.8588, test loss-0.5271, acc-0.8565
Iter-38620, train loss-0.6140, acc-0.8600, valid loss-0.5236, acc-0.8586, test loss-0.5270, acc-0.8564
Iter-38630, train loss-0.5729, acc-0.8600, valid loss-0.5236, acc-0.8586, test loss-0.5270, acc-0.8563
Iter-38640, train loss-0.4111, acc-0.9000, valid loss-0.5235, acc-0.8586, test loss-0.5269, acc-0.8564
Iter-38650, train loss-0.6997, acc-0.7800, valid loss-0.5234, acc-0.8588, test loss-0.5268, acc-0.8564
Iter-38660, train loss-0.4846, acc-0.8600, valid loss-0.5233, acc-0.8590, test loss-0.5267, acc-0.8565
Iter-38670, train loss-0.4723, acc-0.8400, valid loss-0.5232, acc-0.8590, test loss-0.5266, acc-0.8566
Iter-38680, train loss-0.7548, acc-0.7800, valid loss-0.5232, acc-0.8590, test loss-0.5266, acc-0.8566
Iter-38690, train loss-0.4428, acc-0.8600, valid loss-0.5231, acc-0.8590, test loss-0.5265, acc-0.8565
Iter-38700, train loss-0.5880, acc-0.8000, valid loss-0.5230, acc-0.8590, test loss-0.5265, acc-0.8565
Iter-38710, train loss-0.7031, acc-0.7800, valid loss-0.5230, acc-0.8590, test loss-0.5264, acc-0.8565
Iter-38720, train loss-0.7385, acc-0.8400, valid loss-0.5229, acc-0.8590, test loss-0.5263, acc-0.8565
Iter-38730, train loss-0.5988, acc-0.8400, valid loss-0.5228, acc-0.8590, test loss-0.5262, acc-0.8563
Iter-38740, train loss-0.4273, acc-0.8600, valid loss-0.5227, acc-0.8590, test loss-0.5262, acc-0.8563
Iter-38750, train loss-0.5387, acc-0.8400, valid loss-0.5226, acc-0.8590, test loss-0.5261, acc-0.8563
Iter-38760, train loss-0.4970, acc-0.8600, valid loss-0.5225, acc-0.8592, test loss-0.5260, acc-0.8564
Iter-38770, train loss-0.4797, acc-0.9000, valid loss-0.5225, acc-0.8590, test loss-0.5259, acc-0.8564
Iter-38780, train loss-0.4932, acc-0.8400, valid loss-0.5224, acc-0.8592, test loss-0.5259, acc-0.8563
Iter-38790, train loss-0.7036, acc-0.7600, valid loss-0.5224, acc-0.8592, test loss-0.5258, acc-0.8564
Iter-38800, train loss-0.5285, acc-0.8600, valid loss-0.5223, acc-0.8592, test loss-0.5257, acc-0.8565
Iter-38810, train loss-0.5869, acc-0.8800, valid loss-0.5222, acc-0.8590, test loss-0.5256, acc-0.8564
Iter-38820, train loss-0.5843, acc-0.8200, valid loss-0.5222, acc-0.8592, test loss-0.5256, acc-0.8566
Iter-38830, train loss-0.8814, acc-0.7400, valid loss-0.5221, acc-0.8592, test loss-0.5255, acc-0.8565
Iter-38840, train loss-0.4489, acc-0.8400, valid loss-0.5220, acc-0.8592, test loss-0.5254, acc-0.8565
Iter-38850, train loss-0.3514, acc-0.9600, valid loss-0.5220, acc-0.8596, test loss-0.5254, acc-0.8565
Iter-38860, train loss-0.7700, acc-0.7800, valid loss-0.5219, acc-0.8594, test loss-0.5253, acc-0.8565
Iter-38870, train loss-0.3432, acc-0.9200, valid loss-0.5219, acc-0.8594, test loss-0.5252, acc-0.8565
Iter-38880, train loss-0.8002, acc-0.7200, valid loss-0.5218, acc-0.8594, test loss-0.5252, acc-0.8566
Iter-38890, train loss-0.6608, acc-0.7800, valid loss-0.5217, acc-0.8592, test loss-0.5251, acc-0.8566
Iter-38900, train loss-0.5554, acc-0.8400, valid loss-0.5216, acc-0.8592, test loss-0.5250, acc-0.8568
Iter-38910, train loss-0.6124, acc-0.8400, valid loss-0.5216, acc-0.8592, test loss-0.5250, acc-0.8568
Iter-38920, train loss-0.2689, acc-0.9600, valid loss-0.5215, acc-0.8596, test loss-0.5249, acc-0.8568
Iter-38930, train loss-0.4508, acc-0.9000, valid loss-0.5215, acc-0.8594, test loss-0.5248, acc-0.8569
Iter-38940, train loss-0.4997, acc-0.9000, valid loss-0.5214, acc-0.8594, test loss-0.5247, acc-0.8570
Iter-38950, train loss-0.7108, acc-0.8200, valid loss-0.5213, acc-0.8594, test loss-0.5247, acc-0.8569
Iter-38960, train loss-0.5230, acc-0.8200, valid loss-0.5212, acc-0.8592, test loss-0.5246, acc-0.8570
Iter-38970, train loss-0.5036, acc-0.8400, valid loss-0.5212, acc-0.8594, test loss-0.5245, acc-0.8569
Iter-38980, train loss-0.5409, acc-0.8600, valid loss-0.5211, acc-0.8594, test loss-0.5245, acc-0.8570
Iter-38990, train loss-0.4396, acc-0.9000, valid loss-0.5210, acc-0.8598, test loss-0.5244, acc-0.8570
Iter-39000, train loss-0.4529, acc-0.9200, valid loss-0.5209, acc-0.8598, test loss-0.5243, acc-0.8569
Iter-39010, train loss-0.5918, acc-0.8600, valid loss-0.5209, acc-0.8592, test loss-0.5243, acc-0.8569
Iter-39020, train loss-0.6136, acc-0.8400, valid loss-0.5208, acc-0.8596, test loss-0.5242, acc-0.8568
Iter-39030, train loss-0.7795, acc-0.7600, valid loss-0.5207, acc-0.8596, test loss-0.5241, acc-0.8569
Iter-39040, train loss-0.4961, acc-0.8800, valid loss-0.5207, acc-0.8594, test loss-0.5240, acc-0.8570
Iter-39050, train loss-0.5732, acc-0.8600, valid loss-0.5206, acc-0.8596, test loss-0.5240, acc-0.8569
Iter-39060, train loss-0.5915, acc-0.8000, valid loss-0.5205, acc-0.8596, test loss-0.5239, acc-0.8568
Iter-39070, train loss-0.7367, acc-0.8000, valid loss-0.5204, acc-0.8598, test loss-0.5238, acc-0.8568
Iter-39080, train loss-0.4224, acc-0.9000, valid loss-0.5203, acc-0.8598, test loss-0.5237, acc-0.8567
Iter-39090, train loss-0.5506, acc-0.8200, valid loss-0.5203, acc-0.8604, test loss-0.5237, acc-0.8567
Iter-39100, train loss-0.6672, acc-0.8200, valid loss-0.5202, acc-0.8602, test loss-0.5236, acc-0.8567
Iter-39110, train loss-0.6413, acc-0.8400, valid loss-0.5201, acc-0.8602, test loss-0.5235, acc-0.8567
Iter-39120, train loss-0.3209, acc-0.9400, valid loss-0.5201, acc-0.8602, test loss-0.5235, acc-0.8568
Iter-39130, train loss-0.7430, acc-0.7400, valid loss-0.5200, acc-0.8600, test loss-0.5234, acc-0.8567
Iter-39140, train loss-0.3814, acc-0.9200, valid loss-0.5200, acc-0.8600, test loss-0.5233, acc-0.8568
Iter-39150, train loss-0.3169, acc-0.9200, valid loss-0.5199, acc-0.8602, test loss-0.5233, acc-0.8568
Iter-39160, train loss-0.5536, acc-0.8200, valid loss-0.5198, acc-0.8602, test loss-0.5232, acc-0.8567
Iter-39170, train loss-0.5616, acc-0.8400, valid loss-0.5197, acc-0.8602, test loss-0.5231, acc-0.8567
Iter-39180, train loss-0.5629, acc-0.8800, valid loss-0.5197, acc-0.8602, test loss-0.5231, acc-0.8567
Iter-39190, train loss-0.6067, acc-0.8200, valid loss-0.5196, acc-0.8602, test loss-0.5230, acc-0.8567
Iter-39200, train loss-0.5139, acc-0.8200, valid loss-0.5195, acc-0.8604, test loss-0.5229, acc-0.8566
Iter-39210, train loss-0.6635, acc-0.8200, valid loss-0.5195, acc-0.8602, test loss-0.5229, acc-0.8566
Iter-39220, train loss-0.3884, acc-0.9000, valid loss-0.5194, acc-0.8604, test loss-0.5228, acc-0.8566
Iter-39230, train loss-0.3847, acc-0.9200, valid loss-0.5193, acc-0.8602, test loss-0.5227, acc-0.8566
Iter-39240, train loss-0.5867, acc-0.8200, valid loss-0.5192, acc-0.8602, test loss-0.5226, acc-0.8565
Iter-39250, train loss-0.7104, acc-0.7800, valid loss-0.5191, acc-0.8602, test loss-0.5226, acc-0.8566
Iter-39260, train loss-0.3397, acc-0.9000, valid loss-0.5191, acc-0.8604, test loss-0.5225, acc-0.8567
Iter-39270, train loss-0.4694, acc-0.8800, valid loss-0.5190, acc-0.8604, test loss-0.5224, acc-0.8567
Iter-39280, train loss-0.5911, acc-0.9000, valid loss-0.5189, acc-0.8602, test loss-0.5223, acc-0.8569
Iter-39290, train loss-0.4629, acc-0.9000, valid loss-0.5188, acc-0.8602, test loss-0.5222, acc-0.8567
Iter-39300, train loss-0.4298, acc-0.9200, valid loss-0.5187, acc-0.8600, test loss-0.5222, acc-0.8567
Iter-39310, train loss-0.3266, acc-0.9200, valid loss-0.5187, acc-0.8602, test loss-0.5221, acc-0.8567
Iter-39320, train loss-0.4822, acc-0.8400, valid loss-0.5186, acc-0.8602, test loss-0.5220, acc-0.8570
Iter-39330, train loss-0.4312, acc-0.9000, valid loss-0.5185, acc-0.8602, test loss-0.5219, acc-0.8568
Iter-39340, train loss-0.4441, acc-0.9200, valid loss-0.5185, acc-0.8602, test loss-0.5219, acc-0.8569
Iter-39350, train loss-0.5735, acc-0.8400, valid loss-0.5184, acc-0.8604, test loss-0.5218, acc-0.8569
Iter-39360, train loss-0.5036, acc-0.8400, valid loss-0.5183, acc-0.8604, test loss-0.5217, acc-0.8570
Iter-39370, train loss-0.6576, acc-0.8000, valid loss-0.5182, acc-0.8602, test loss-0.5217, acc-0.8570
Iter-39380, train loss-0.6680, acc-0.8400, valid loss-0.5182, acc-0.8602, test loss-0.5216, acc-0.8571
Iter-39390, train loss-0.7990, acc-0.8000, valid loss-0.5181, acc-0.8600, test loss-0.5215, acc-0.8571
Iter-39400, train loss-0.4575, acc-0.8600, valid loss-0.5181, acc-0.8602, test loss-0.5215, acc-0.8571
Iter-39410, train loss-0.3358, acc-0.9400, valid loss-0.5180, acc-0.8600, test loss-0.5214, acc-0.8571
Iter-39420, train loss-0.6139, acc-0.8600, valid loss-0.5179, acc-0.8602, test loss-0.5213, acc-0.8572
Iter-39430, train loss-0.4202, acc-0.9000, valid loss-0.5179, acc-0.8600, test loss-0.5213, acc-0.8573
Iter-39440, train loss-0.3193, acc-0.9600, valid loss-0.5178, acc-0.8600, test loss-0.5212, acc-0.8572
Iter-39450, train loss-0.4830, acc-0.8600, valid loss-0.5177, acc-0.8600, test loss-0.5211, acc-0.8573
Iter-39460, train loss-0.3839, acc-0.9000, valid loss-0.5177, acc-0.8602, test loss-0.5210, acc-0.8572
Iter-39470, train loss-0.5966, acc-0.8600, valid loss-0.5176, acc-0.8602, test loss-0.5210, acc-0.8573
Iter-39480, train loss-0.3656, acc-0.9200, valid loss-0.5175, acc-0.8604, test loss-0.5209, acc-0.8573
Iter-39490, train loss-0.5220, acc-0.8800, valid loss-0.5175, acc-0.8602, test loss-0.5208, acc-0.8573
Iter-39500, train loss-0.4372, acc-0.8800, valid loss-0.5174, acc-0.8602, test loss-0.5207, acc-0.8573
Iter-39510, train loss-0.5348, acc-0.9000, valid loss-0.5173, acc-0.8602, test loss-0.5207, acc-0.8572
Iter-39520, train loss-0.3781, acc-0.9400, valid loss-0.5173, acc-0.8600, test loss-0.5206, acc-0.8573
Iter-39530, train loss-0.4667, acc-0.8800, valid loss-0.5172, acc-0.8602, test loss-0.5206, acc-0.8573
Iter-39540, train loss-0.4592, acc-0.8600, valid loss-0.5171, acc-0.8602, test loss-0.5205, acc-0.8575
Iter-39550, train loss-0.6256, acc-0.8200, valid loss-0.5170, acc-0.8604, test loss-0.5204, acc-0.8574
Iter-39560, train loss-0.6552, acc-0.9000, valid loss-0.5170, acc-0.8604, test loss-0.5204, acc-0.8573
Iter-39570, train loss-0.4564, acc-0.8600, valid loss-0.5169, acc-0.8604, test loss-0.5203, acc-0.8574
Iter-39580, train loss-0.5806, acc-0.8400, valid loss-0.5168, acc-0.8602, test loss-0.5202, acc-0.8574
Iter-39590, train loss-0.7041, acc-0.7800, valid loss-0.5168, acc-0.8602, test loss-0.5202, acc-0.8573
Iter-39600, train loss-0.5797, acc-0.8200, valid loss-0.5167, acc-0.8604, test loss-0.5201, acc-0.8573
Iter-39610, train loss-0.5140, acc-0.9000, valid loss-0.5166, acc-0.8608, test loss-0.5201, acc-0.8574
Iter-39620, train loss-0.5594, acc-0.8800, valid loss-0.5166, acc-0.8608, test loss-0.5200, acc-0.8574
Iter-39630, train loss-0.4761, acc-0.8600, valid loss-0.5165, acc-0.8608, test loss-0.5199, acc-0.8574
Iter-39640, train loss-0.5885, acc-0.8200, valid loss-0.5164, acc-0.8606, test loss-0.5198, acc-0.8575
Iter-39650, train loss-0.5684, acc-0.8400, valid loss-0.5164, acc-0.8604, test loss-0.5198, acc-0.8575
Iter-39660, train loss-0.5298, acc-0.8400, valid loss-0.5163, acc-0.8604, test loss-0.5197, acc-0.8577
Iter-39670, train loss-0.6959, acc-0.7800, valid loss-0.5162, acc-0.8604, test loss-0.5196, acc-0.8575
Iter-39680, train loss-0.6098, acc-0.7400, valid loss-0.5162, acc-0.8602, test loss-0.5195, acc-0.8577
Iter-39690, train loss-0.6338, acc-0.7600, valid loss-0.5161, acc-0.8606, test loss-0.5195, acc-0.8578
Iter-39700, train loss-0.5242, acc-0.8600, valid loss-0.5161, acc-0.8606, test loss-0.5194, acc-0.8578
Iter-39710, train loss-0.4702, acc-0.9000, valid loss-0.5160, acc-0.8604, test loss-0.5194, acc-0.8577
Iter-39720, train loss-0.5687, acc-0.8400, valid loss-0.5159, acc-0.8602, test loss-0.5193, acc-0.8575
Iter-39730, train loss-0.6301, acc-0.8000, valid loss-0.5159, acc-0.8604, test loss-0.5192, acc-0.8578
Iter-39740, train loss-0.3366, acc-0.9200, valid loss-0.5158, acc-0.8606, test loss-0.5192, acc-0.8579
Iter-39750, train loss-0.5389, acc-0.9000, valid loss-0.5157, acc-0.8606, test loss-0.5191, acc-0.8578
Iter-39760, train loss-0.5743, acc-0.8200, valid loss-0.5157, acc-0.8606, test loss-0.5190, acc-0.8579
Iter-39770, train loss-0.4886, acc-0.9000, valid loss-0.5156, acc-0.8606, test loss-0.5189, acc-0.8580
Iter-39780, train loss-0.7944, acc-0.7800, valid loss-0.5155, acc-0.8606, test loss-0.5189, acc-0.8577
Iter-39790, train loss-0.6070, acc-0.8400, valid loss-0.5155, acc-0.8606, test loss-0.5188, acc-0.8578
Iter-39800, train loss-0.5404, acc-0.8000, valid loss-0.5154, acc-0.8606, test loss-0.5187, acc-0.8581
Iter-39810, train loss-0.3169, acc-0.9600, valid loss-0.5153, acc-0.8608, test loss-0.5187, acc-0.8581
Iter-39820, train loss-0.6075, acc-0.8200, valid loss-0.5153, acc-0.8608, test loss-0.5186, acc-0.8580
Iter-39830, train loss-0.6911, acc-0.7400, valid loss-0.5152, acc-0.8608, test loss-0.5185, acc-0.8580
Iter-39840, train loss-0.7148, acc-0.7600, valid loss-0.5151, acc-0.8608, test loss-0.5185, acc-0.8579
Iter-39850, train loss-0.4764, acc-0.8800, valid loss-0.5151, acc-0.8608, test loss-0.5184, acc-0.8579
Iter-39860, train loss-0.5461, acc-0.8600, valid loss-0.5150, acc-0.8610, test loss-0.5183, acc-0.8580
Iter-39870, train loss-0.4962, acc-0.9000, valid loss-0.5149, acc-0.8610, test loss-0.5183, acc-0.8582
Iter-39880, train loss-0.5773, acc-0.8400, valid loss-0.5149, acc-0.8610, test loss-0.5182, acc-0.8581
Iter-39890, train loss-0.4518, acc-0.8600, valid loss-0.5148, acc-0.8612, test loss-0.5182, acc-0.8581
Iter-39900, train loss-0.6184, acc-0.8400, valid loss-0.5147, acc-0.8612, test loss-0.5181, acc-0.8581
Iter-39910, train loss-0.3998, acc-0.9400, valid loss-0.5147, acc-0.8610, test loss-0.5180, acc-0.8581
Iter-39920, train loss-0.4439, acc-0.8600, valid loss-0.5146, acc-0.8612, test loss-0.5179, acc-0.8581
Iter-39930, train loss-0.5738, acc-0.8400, valid loss-0.5145, acc-0.8612, test loss-0.5179, acc-0.8582
Iter-39940, train loss-0.8647, acc-0.7600, valid loss-0.5144, acc-0.8610, test loss-0.5178, acc-0.8582
Iter-39950, train loss-0.6346, acc-0.8400, valid loss-0.5144, acc-0.8610, test loss-0.5177, acc-0.8583
Iter-39960, train loss-0.3950, acc-0.8400, valid loss-0.5143, acc-0.8610, test loss-0.5177, acc-0.8582
Iter-39970, train loss-0.5970, acc-0.8200, valid loss-0.5143, acc-0.8610, test loss-0.5176, acc-0.8586
Iter-39980, train loss-0.5510, acc-0.8800, valid loss-0.5142, acc-0.8610, test loss-0.5175, acc-0.8587
Iter-39990, train loss-0.5444, acc-0.8800, valid loss-0.5141, acc-0.8610, test loss-0.5175, acc-0.8585
Iter-40000, train loss-0.5752, acc-0.8600, valid loss-0.5140, acc-0.8610, test loss-0.5174, acc-0.8585
Iter-40010, train loss-0.4645, acc-0.8200, valid loss-0.5140, acc-0.8610, test loss-0.5173, acc-0.8589
Iter-40020, train loss-0.5884, acc-0.8000, valid loss-0.5139, acc-0.8610, test loss-0.5172, acc-0.8588
Iter-40030, train loss-0.6134, acc-0.8400, valid loss-0.5139, acc-0.8610, test loss-0.5172, acc-0.8590
Iter-40040, train loss-0.6023, acc-0.8200, valid loss-0.5138, acc-0.8610, test loss-0.5171, acc-0.8590
Iter-40050, train loss-0.5084, acc-0.8200, valid loss-0.5137, acc-0.8610, test loss-0.5170, acc-0.8591
Iter-40060, train loss-0.4986, acc-0.8400, valid loss-0.5137, acc-0.8610, test loss-0.5170, acc-0.8591
Iter-40070, train loss-0.5845, acc-0.8000, valid loss-0.5136, acc-0.8610, test loss-0.5169, acc-0.8590
Iter-40080, train loss-0.8260, acc-0.7800, valid loss-0.5135, acc-0.8610, test loss-0.5168, acc-0.8590
Iter-40090, train loss-0.8767, acc-0.7600, valid loss-0.5135, acc-0.8610, test loss-0.5168, acc-0.8592
Iter-40100, train loss-0.4818, acc-0.8600, valid loss-0.5134, acc-0.8610, test loss-0.5167, acc-0.8592
Iter-40110, train loss-0.5378, acc-0.8800, valid loss-0.5133, acc-0.8610, test loss-0.5166, acc-0.8592
Iter-40120, train loss-0.4867, acc-0.9000, valid loss-0.5133, acc-0.8610, test loss-0.5165, acc-0.8592
Iter-40130, train loss-0.5226, acc-0.8400, valid loss-0.5132, acc-0.8610, test loss-0.5165, acc-0.8592
Iter-40140, train loss-0.4207, acc-0.8800, valid loss-0.5131, acc-0.8608, test loss-0.5164, acc-0.8592
Iter-40150, train loss-0.5558, acc-0.8200, valid loss-0.5131, acc-0.8610, test loss-0.5163, acc-0.8592
Iter-40160, train loss-0.6488, acc-0.7400, valid loss-0.5130, acc-0.8610, test loss-0.5162, acc-0.8592
Iter-40170, train loss-0.5562, acc-0.8400, valid loss-0.5129, acc-0.8610, test loss-0.5162, acc-0.8593
Iter-40180, train loss-0.6126, acc-0.8800, valid loss-0.5129, acc-0.8610, test loss-0.5161, acc-0.8592
Iter-40190, train loss-0.6976, acc-0.8400, valid loss-0.5128, acc-0.8610, test loss-0.5161, acc-0.8592
Iter-40200, train loss-0.7311, acc-0.7600, valid loss-0.5128, acc-0.8610, test loss-0.5160, acc-0.8592
Iter-40210, train loss-0.5108, acc-0.8600, valid loss-0.5127, acc-0.8610, test loss-0.5159, acc-0.8593
Iter-40220, train loss-0.4498, acc-0.8600, valid loss-0.5126, acc-0.8610, test loss-0.5159, acc-0.8592
Iter-40230, train loss-0.4098, acc-0.8800, valid loss-0.5125, acc-0.8610, test loss-0.5158, acc-0.8594
Iter-40240, train loss-0.3959, acc-0.9000, valid loss-0.5125, acc-0.8610, test loss-0.5157, acc-0.8594
Iter-40250, train loss-0.4907, acc-0.9000, valid loss-0.5124, acc-0.8608, test loss-0.5157, acc-0.8594
Iter-40260, train loss-0.4813, acc-0.8400, valid loss-0.5123, acc-0.8610, test loss-0.5156, acc-0.8595
Iter-40270, train loss-0.5521, acc-0.8600, valid loss-0.5122, acc-0.8610, test loss-0.5155, acc-0.8595
Iter-40280, train loss-0.5346, acc-0.8000, valid loss-0.5122, acc-0.8610, test loss-0.5155, acc-0.8595
Iter-40290, train loss-0.4124, acc-0.9200, valid loss-0.5121, acc-0.8610, test loss-0.5154, acc-0.8596
Iter-40300, train loss-0.4275, acc-0.8800, valid loss-0.5120, acc-0.8610, test loss-0.5153, acc-0.8596
Iter-40310, train loss-0.4419, acc-0.9000, valid loss-0.5120, acc-0.8610, test loss-0.5153, acc-0.8596
Iter-40320, train loss-0.7279, acc-0.8200, valid loss-0.5119, acc-0.8610, test loss-0.5152, acc-0.8595
Iter-40330, train loss-0.4921, acc-0.8400, valid loss-0.5118, acc-0.8610, test loss-0.5151, acc-0.8595
Iter-40340, train loss-0.6466, acc-0.7200, valid loss-0.5117, acc-0.8610, test loss-0.5150, acc-0.8595
Iter-40350, train loss-0.5963, acc-0.8400, valid loss-0.5117, acc-0.8610, test loss-0.5150, acc-0.8594
Iter-40360, train loss-0.4137, acc-0.9000, valid loss-0.5116, acc-0.8608, test loss-0.5149, acc-0.8597
Iter-40370, train loss-0.5825, acc-0.8000, valid loss-0.5115, acc-0.8608, test loss-0.5148, acc-0.8597
Iter-40380, train loss-0.4261, acc-0.8800, valid loss-0.5115, acc-0.8608, test loss-0.5148, acc-0.8597
Iter-40390, train loss-0.4648, acc-0.8800, valid loss-0.5114, acc-0.8608, test loss-0.5147, acc-0.8593
Iter-40400, train loss-0.5452, acc-0.8600, valid loss-0.5113, acc-0.8608, test loss-0.5146, acc-0.8593
Iter-40410, train loss-0.5695, acc-0.8800, valid loss-0.5113, acc-0.8610, test loss-0.5146, acc-0.8594
Iter-40420, train loss-0.5551, acc-0.8000, valid loss-0.5112, acc-0.8612, test loss-0.5145, acc-0.8594
Iter-40430, train loss-0.3948, acc-0.9200, valid loss-0.5112, acc-0.8612, test loss-0.5144, acc-0.8594
Iter-40440, train loss-0.5353, acc-0.8400, valid loss-0.5111, acc-0.8612, test loss-0.5144, acc-0.8593
Iter-40450, train loss-0.3904, acc-0.8800, valid loss-0.5110, acc-0.8612, test loss-0.5143, acc-0.8593
Iter-40460, train loss-0.5129, acc-0.8400, valid loss-0.5110, acc-0.8612, test loss-0.5143, acc-0.8593
Iter-40470, train loss-0.4230, acc-0.8800, valid loss-0.5109, acc-0.8612, test loss-0.5142, acc-0.8594
Iter-40480, train loss-0.7122, acc-0.7600, valid loss-0.5108, acc-0.8612, test loss-0.5142, acc-0.8593
Iter-40490, train loss-0.5853, acc-0.8800, valid loss-0.5108, acc-0.8612, test loss-0.5141, acc-0.8593
Iter-40500, train loss-0.5523, acc-0.8400, valid loss-0.5107, acc-0.8612, test loss-0.5140, acc-0.8593
Iter-40510, train loss-0.6272, acc-0.7800, valid loss-0.5106, acc-0.8612, test loss-0.5140, acc-0.8593
Iter-40520, train loss-0.5542, acc-0.8200, valid loss-0.5106, acc-0.8612, test loss-0.5139, acc-0.8593
Iter-40530, train loss-0.4903, acc-0.8600, valid loss-0.5105, acc-0.8612, test loss-0.5138, acc-0.8593
Iter-40540, train loss-0.4437, acc-0.9000, valid loss-0.5105, acc-0.8612, test loss-0.5137, acc-0.8595
Iter-40550, train loss-0.6559, acc-0.8000, valid loss-0.5104, acc-0.8612, test loss-0.5137, acc-0.8593
Iter-40560, train loss-0.5336, acc-0.8400, valid loss-0.5103, acc-0.8612, test loss-0.5136, acc-0.8593
Iter-40570, train loss-0.4481, acc-0.8800, valid loss-0.5102, acc-0.8612, test loss-0.5135, acc-0.8593
Iter-40580, train loss-0.6429, acc-0.8000, valid loss-0.5102, acc-0.8612, test loss-0.5135, acc-0.8596
Iter-40590, train loss-0.3968, acc-0.8800, valid loss-0.5101, acc-0.8612, test loss-0.5134, acc-0.8596
Iter-40600, train loss-0.4769, acc-0.9200, valid loss-0.5100, acc-0.8612, test loss-0.5134, acc-0.8596
Iter-40610, train loss-0.4102, acc-0.8600, valid loss-0.5100, acc-0.8612, test loss-0.5133, acc-0.8596
Iter-40620, train loss-0.5732, acc-0.8200, valid loss-0.5099, acc-0.8612, test loss-0.5132, acc-0.8596
Iter-40630, train loss-0.5498, acc-0.8600, valid loss-0.5098, acc-0.8614, test loss-0.5131, acc-0.8598
Iter-40640, train loss-0.3636, acc-0.9400, valid loss-0.5098, acc-0.8612, test loss-0.5131, acc-0.8598
Iter-40650, train loss-0.7026, acc-0.8600, valid loss-0.5097, acc-0.8614, test loss-0.5130, acc-0.8598
Iter-40660, train loss-0.4927, acc-0.8800, valid loss-0.5096, acc-0.8614, test loss-0.5129, acc-0.8598
Iter-40670, train loss-0.5330, acc-0.8200, valid loss-0.5095, acc-0.8612, test loss-0.5129, acc-0.8598
Iter-40680, train loss-0.4073, acc-0.9200, valid loss-0.5095, acc-0.8612, test loss-0.5128, acc-0.8599
Iter-40690, train loss-0.5934, acc-0.8400, valid loss-0.5094, acc-0.8612, test loss-0.5127, acc-0.8599
Iter-40700, train loss-0.5624, acc-0.8800, valid loss-0.5093, acc-0.8612, test loss-0.5127, acc-0.8599
Iter-40710, train loss-0.4572, acc-0.8800, valid loss-0.5093, acc-0.8612, test loss-0.5126, acc-0.8599
Iter-40720, train loss-0.3870, acc-0.9400, valid loss-0.5092, acc-0.8612, test loss-0.5125, acc-0.8598
Iter-40730, train loss-0.6138, acc-0.8200, valid loss-0.5091, acc-0.8612, test loss-0.5125, acc-0.8600
Iter-40740, train loss-0.4825, acc-0.8600, valid loss-0.5091, acc-0.8612, test loss-0.5124, acc-0.8600
Iter-40750, train loss-0.4887, acc-0.8400, valid loss-0.5090, acc-0.8614, test loss-0.5123, acc-0.8599
Iter-40760, train loss-0.5264, acc-0.8800, valid loss-0.5090, acc-0.8614, test loss-0.5123, acc-0.8600
Iter-40770, train loss-0.7723, acc-0.8000, valid loss-0.5089, acc-0.8614, test loss-0.5122, acc-0.8600
Iter-40780, train loss-0.5060, acc-0.8400, valid loss-0.5088, acc-0.8614, test loss-0.5121, acc-0.8600
Iter-40790, train loss-0.4904, acc-0.9000, valid loss-0.5088, acc-0.8614, test loss-0.5121, acc-0.8600
Iter-40800, train loss-0.4686, acc-0.8600, valid loss-0.5087, acc-0.8614, test loss-0.5120, acc-0.8601
Iter-40810, train loss-0.3960, acc-0.8800, valid loss-0.5087, acc-0.8614, test loss-0.5120, acc-0.8601
Iter-40820, train loss-0.6457, acc-0.8600, valid loss-0.5086, acc-0.8614, test loss-0.5119, acc-0.8602
Iter-40830, train loss-0.6301, acc-0.8400, valid loss-0.5085, acc-0.8614, test loss-0.5118, acc-0.8601
Iter-40840, train loss-0.3912, acc-0.9000, valid loss-0.5084, acc-0.8614, test loss-0.5118, acc-0.8602
Iter-40850, train loss-0.5493, acc-0.8200, valid loss-0.5084, acc-0.8614, test loss-0.5117, acc-0.8601
Iter-40860, train loss-0.3058, acc-0.9400, valid loss-0.5083, acc-0.8614, test loss-0.5117, acc-0.8602
Iter-40870, train loss-0.4825, acc-0.8200, valid loss-0.5083, acc-0.8614, test loss-0.5116, acc-0.8602
Iter-40880, train loss-0.4553, acc-0.8800, valid loss-0.5082, acc-0.8614, test loss-0.5115, acc-0.8601
Iter-40890, train loss-0.5121, acc-0.8600, valid loss-0.5081, acc-0.8614, test loss-0.5115, acc-0.8602
Iter-40900, train loss-0.5910, acc-0.8200, valid loss-0.5081, acc-0.8614, test loss-0.5114, acc-0.8602
Iter-40910, train loss-0.5923, acc-0.8000, valid loss-0.5080, acc-0.8616, test loss-0.5113, acc-0.8603
Iter-40920, train loss-0.4047, acc-0.9200, valid loss-0.5079, acc-0.8616, test loss-0.5113, acc-0.8603
Iter-40930, train loss-0.5747, acc-0.8400, valid loss-0.5079, acc-0.8616, test loss-0.5112, acc-0.8602
Iter-40940, train loss-0.4035, acc-0.9000, valid loss-0.5078, acc-0.8614, test loss-0.5111, acc-0.8603
Iter-40950, train loss-0.5257, acc-0.8600, valid loss-0.5077, acc-0.8616, test loss-0.5111, acc-0.8602
Iter-40960, train loss-0.3953, acc-0.8800, valid loss-0.5077, acc-0.8616, test loss-0.5110, acc-0.8603
Iter-40970, train loss-0.3197, acc-0.9200, valid loss-0.5076, acc-0.8616, test loss-0.5109, acc-0.8603
Iter-40980, train loss-0.3349, acc-0.9200, valid loss-0.5075, acc-0.8616, test loss-0.5108, acc-0.8602
Iter-40990, train loss-0.4866, acc-0.8400, valid loss-0.5074, acc-0.8616, test loss-0.5108, acc-0.8602
Iter-41000, train loss-0.4720, acc-0.8400, valid loss-0.5073, acc-0.8616, test loss-0.5107, acc-0.8602
Iter-41010, train loss-0.4659, acc-0.9200, valid loss-0.5073, acc-0.8616, test loss-0.5106, acc-0.8602
Iter-41020, train loss-0.5031, acc-0.8200, valid loss-0.5072, acc-0.8616, test loss-0.5106, acc-0.8602
Iter-41030, train loss-0.6101, acc-0.8200, valid loss-0.5072, acc-0.8616, test loss-0.5105, acc-0.8601
Iter-41040, train loss-0.4843, acc-0.8800, valid loss-0.5071, acc-0.8616, test loss-0.5105, acc-0.8601
Iter-41050, train loss-0.5911, acc-0.9200, valid loss-0.5070, acc-0.8616, test loss-0.5104, acc-0.8602
Iter-41060, train loss-0.5095, acc-0.9000, valid loss-0.5070, acc-0.8616, test loss-0.5103, acc-0.8606
Iter-41070, train loss-0.4013, acc-0.9000, valid loss-0.5069, acc-0.8616, test loss-0.5103, acc-0.8604
Iter-41080, train loss-0.5169, acc-0.8800, valid loss-0.5068, acc-0.8616, test loss-0.5102, acc-0.8605
Iter-41090, train loss-0.4477, acc-0.8600, valid loss-0.5068, acc-0.8618, test loss-0.5101, acc-0.8605
Iter-41100, train loss-0.4974, acc-0.8400, valid loss-0.5067, acc-0.8618, test loss-0.5101, acc-0.8606
Iter-41110, train loss-0.3819, acc-0.9200, valid loss-0.5066, acc-0.8618, test loss-0.5100, acc-0.8605
Iter-41120, train loss-0.5299, acc-0.8200, valid loss-0.5066, acc-0.8616, test loss-0.5100, acc-0.8605
Iter-41130, train loss-0.5434, acc-0.8000, valid loss-0.5065, acc-0.8618, test loss-0.5099, acc-0.8605
Iter-41140, train loss-0.4749, acc-0.9000, valid loss-0.5064, acc-0.8618, test loss-0.5098, acc-0.8607
Iter-41150, train loss-0.5390, acc-0.8400, valid loss-0.5064, acc-0.8618, test loss-0.5097, acc-0.8607
Iter-41160, train loss-0.5608, acc-0.8200, valid loss-0.5063, acc-0.8618, test loss-0.5097, acc-0.8606
Iter-41170, train loss-0.5118, acc-0.8000, valid loss-0.5062, acc-0.8618, test loss-0.5096, acc-0.8606
Iter-41180, train loss-0.4992, acc-0.8800, valid loss-0.5062, acc-0.8618, test loss-0.5095, acc-0.8605
Iter-41190, train loss-0.6727, acc-0.7800, valid loss-0.5061, acc-0.8618, test loss-0.5095, acc-0.8606
Iter-41200, train loss-0.4912, acc-0.8600, valid loss-0.5061, acc-0.8618, test loss-0.5094, acc-0.8603
Iter-41210, train loss-0.4867, acc-0.8600, valid loss-0.5060, acc-0.8618, test loss-0.5094, acc-0.8604
Iter-41220, train loss-0.4223, acc-0.8800, valid loss-0.5059, acc-0.8618, test loss-0.5093, acc-0.8605
Iter-41230, train loss-0.4603, acc-0.8600, valid loss-0.5059, acc-0.8618, test loss-0.5092, acc-0.8605
Iter-41240, train loss-0.4841, acc-0.9400, valid loss-0.5058, acc-0.8620, test loss-0.5092, acc-0.8605
Iter-41250, train loss-0.6511, acc-0.7800, valid loss-0.5057, acc-0.8620, test loss-0.5091, acc-0.8604
Iter-41260, train loss-0.5125, acc-0.8200, valid loss-0.5057, acc-0.8620, test loss-0.5090, acc-0.8604
Iter-41270, train loss-0.5078, acc-0.8800, valid loss-0.5056, acc-0.8620, test loss-0.5090, acc-0.8605
Iter-41280, train loss-0.4760, acc-0.9000, valid loss-0.5055, acc-0.8620, test loss-0.5089, acc-0.8606
Iter-41290, train loss-0.5989, acc-0.8400, valid loss-0.5055, acc-0.8620, test loss-0.5088, acc-0.8605
Iter-41300, train loss-0.5931, acc-0.8400, valid loss-0.5054, acc-0.8620, test loss-0.5088, acc-0.8605
Iter-41310, train loss-0.5030, acc-0.8200, valid loss-0.5053, acc-0.8620, test loss-0.5087, acc-0.8605
Iter-41320, train loss-0.4112, acc-0.8800, valid loss-0.5053, acc-0.8620, test loss-0.5087, acc-0.8603
Iter-41330, train loss-0.4491, acc-0.9000, valid loss-0.5052, acc-0.8620, test loss-0.5086, acc-0.8603
Iter-41340, train loss-0.5486, acc-0.8400, valid loss-0.5052, acc-0.8620, test loss-0.5085, acc-0.8603
Iter-41350, train loss-0.7187, acc-0.8400, valid loss-0.5051, acc-0.8620, test loss-0.5084, acc-0.8602
Iter-41360, train loss-0.5046, acc-0.8600, valid loss-0.5050, acc-0.8622, test loss-0.5084, acc-0.8601
Iter-41370, train loss-0.4725, acc-0.8800, valid loss-0.5050, acc-0.8624, test loss-0.5083, acc-0.8603
Iter-41380, train loss-0.6145, acc-0.7600, valid loss-0.5049, acc-0.8624, test loss-0.5083, acc-0.8602
Iter-41390, train loss-0.6223, acc-0.8200, valid loss-0.5049, acc-0.8622, test loss-0.5082, acc-0.8603
Iter-41400, train loss-0.6336, acc-0.8200, valid loss-0.5048, acc-0.8622, test loss-0.5081, acc-0.8602
Iter-41410, train loss-0.5123, acc-0.8800, valid loss-0.5048, acc-0.8622, test loss-0.5081, acc-0.8602
Iter-41420, train loss-0.5102, acc-0.8600, valid loss-0.5047, acc-0.8624, test loss-0.5080, acc-0.8602
Iter-41430, train loss-0.5709, acc-0.8200, valid loss-0.5046, acc-0.8624, test loss-0.5080, acc-0.8601
Iter-41440, train loss-0.3755, acc-0.9200, valid loss-0.5046, acc-0.8624, test loss-0.5079, acc-0.8602
Iter-41450, train loss-0.5355, acc-0.8800, valid loss-0.5045, acc-0.8624, test loss-0.5078, acc-0.8603
Iter-41460, train loss-0.6165, acc-0.9000, valid loss-0.5044, acc-0.8624, test loss-0.5078, acc-0.8603
Iter-41470, train loss-0.7418, acc-0.7000, valid loss-0.5044, acc-0.8624, test loss-0.5077, acc-0.8604
Iter-41480, train loss-0.5139, acc-0.8600, valid loss-0.5043, acc-0.8622, test loss-0.5077, acc-0.8604
Iter-41490, train loss-0.5354, acc-0.8400, valid loss-0.5043, acc-0.8624, test loss-0.5076, acc-0.8603
Iter-41500, train loss-0.5150, acc-0.8800, valid loss-0.5042, acc-0.8626, test loss-0.5076, acc-0.8603
Iter-41510, train loss-0.5108, acc-0.9000, valid loss-0.5041, acc-0.8626, test loss-0.5075, acc-0.8603
Iter-41520, train loss-0.5864, acc-0.8200, valid loss-0.5041, acc-0.8626, test loss-0.5074, acc-0.8605
Iter-41530, train loss-0.6321, acc-0.8200, valid loss-0.5040, acc-0.8626, test loss-0.5073, acc-0.8604
Iter-41540, train loss-0.6485, acc-0.8000, valid loss-0.5039, acc-0.8626, test loss-0.5073, acc-0.8603
Iter-41550, train loss-0.4214, acc-0.8800, valid loss-0.5039, acc-0.8626, test loss-0.5072, acc-0.8605
Iter-41560, train loss-0.3637, acc-0.9600, valid loss-0.5038, acc-0.8626, test loss-0.5072, acc-0.8604
Iter-41570, train loss-0.6093, acc-0.8200, valid loss-0.5037, acc-0.8626, test loss-0.5071, acc-0.8605
Iter-41580, train loss-0.5676, acc-0.8000, valid loss-0.5037, acc-0.8628, test loss-0.5071, acc-0.8605
Iter-41590, train loss-0.5161, acc-0.8200, valid loss-0.5036, acc-0.8628, test loss-0.5070, acc-0.8605
Iter-41600, train loss-0.5496, acc-0.8400, valid loss-0.5036, acc-0.8628, test loss-0.5069, acc-0.8603
Iter-41610, train loss-0.3847, acc-0.8800, valid loss-0.5035, acc-0.8628, test loss-0.5069, acc-0.8605
Iter-41620, train loss-0.5434, acc-0.8800, valid loss-0.5034, acc-0.8628, test loss-0.5068, acc-0.8606
Iter-41630, train loss-0.5285, acc-0.8800, valid loss-0.5034, acc-0.8628, test loss-0.5067, acc-0.8605
Iter-41640, train loss-0.5570, acc-0.8400, valid loss-0.5033, acc-0.8628, test loss-0.5067, acc-0.8606
Iter-41650, train loss-0.4671, acc-0.8600, valid loss-0.5033, acc-0.8628, test loss-0.5066, acc-0.8606
Iter-41660, train loss-0.6671, acc-0.8000, valid loss-0.5032, acc-0.8628, test loss-0.5066, acc-0.8606
Iter-41670, train loss-0.5706, acc-0.8400, valid loss-0.5031, acc-0.8628, test loss-0.5065, acc-0.8606
Iter-41680, train loss-0.5325, acc-0.8800, valid loss-0.5031, acc-0.8628, test loss-0.5064, acc-0.8607
Iter-41690, train loss-0.5384, acc-0.8400, valid loss-0.5030, acc-0.8628, test loss-0.5064, acc-0.8607
Iter-41700, train loss-0.3097, acc-0.9600, valid loss-0.5029, acc-0.8628, test loss-0.5063, acc-0.8608
Iter-41710, train loss-0.5966, acc-0.8000, valid loss-0.5029, acc-0.8628, test loss-0.5062, acc-0.8607
Iter-41720, train loss-0.4620, acc-0.8800, valid loss-0.5028, acc-0.8628, test loss-0.5062, acc-0.8607
Iter-41730, train loss-0.5441, acc-0.8800, valid loss-0.5028, acc-0.8628, test loss-0.5061, acc-0.8607
Iter-41740, train loss-0.5540, acc-0.8600, valid loss-0.5027, acc-0.8628, test loss-0.5060, acc-0.8607
Iter-41750, train loss-0.6084, acc-0.8600, valid loss-0.5026, acc-0.8628, test loss-0.5060, acc-0.8607
Iter-41760, train loss-0.3632, acc-0.9200, valid loss-0.5025, acc-0.8628, test loss-0.5059, acc-0.8607
Iter-41770, train loss-0.5223, acc-0.8400, valid loss-0.5025, acc-0.8628, test loss-0.5058, acc-0.8607
Iter-41780, train loss-0.4412, acc-0.8800, valid loss-0.5024, acc-0.8628, test loss-0.5058, acc-0.8606
Iter-41790, train loss-0.3865, acc-0.9200, valid loss-0.5024, acc-0.8628, test loss-0.5057, acc-0.8606
Iter-41800, train loss-0.4685, acc-0.9000, valid loss-0.5023, acc-0.8628, test loss-0.5057, acc-0.8606
Iter-41810, train loss-0.4734, acc-0.8800, valid loss-0.5023, acc-0.8628, test loss-0.5056, acc-0.8606
Iter-41820, train loss-0.5414, acc-0.8200, valid loss-0.5022, acc-0.8630, test loss-0.5056, acc-0.8606
Iter-41830, train loss-0.4757, acc-0.8600, valid loss-0.5022, acc-0.8630, test loss-0.5055, acc-0.8606
Iter-41840, train loss-0.6922, acc-0.8200, valid loss-0.5021, acc-0.8630, test loss-0.5054, acc-0.8607
Iter-41850, train loss-0.4107, acc-0.8400, valid loss-0.5020, acc-0.8630, test loss-0.5054, acc-0.8607
Iter-41860, train loss-0.4667, acc-0.8800, valid loss-0.5020, acc-0.8628, test loss-0.5053, acc-0.8606
Iter-41870, train loss-0.3836, acc-0.9000, valid loss-0.5019, acc-0.8630, test loss-0.5052, acc-0.8606
Iter-41880, train loss-0.3513, acc-0.9800, valid loss-0.5018, acc-0.8630, test loss-0.5052, acc-0.8606
Iter-41890, train loss-0.9269, acc-0.7600, valid loss-0.5018, acc-0.8628, test loss-0.5051, acc-0.8606
Iter-41900, train loss-0.5388, acc-0.8600, valid loss-0.5017, acc-0.8632, test loss-0.5051, acc-0.8607
Iter-41910, train loss-0.5164, acc-0.8800, valid loss-0.5017, acc-0.8632, test loss-0.5050, acc-0.8607
Iter-41920, train loss-0.4595, acc-0.8600, valid loss-0.5016, acc-0.8630, test loss-0.5049, acc-0.8607
Iter-41930, train loss-0.5898, acc-0.8400, valid loss-0.5015, acc-0.8630, test loss-0.5049, acc-0.8607
Iter-41940, train loss-0.4058, acc-0.9000, valid loss-0.5015, acc-0.8630, test loss-0.5048, acc-0.8607
Iter-41950, train loss-0.4576, acc-0.9200, valid loss-0.5014, acc-0.8632, test loss-0.5047, acc-0.8607
Iter-41960, train loss-0.4234, acc-0.9200, valid loss-0.5013, acc-0.8628, test loss-0.5046, acc-0.8607
Iter-41970, train loss-0.5044, acc-0.8400, valid loss-0.5013, acc-0.8628, test loss-0.5046, acc-0.8607
Iter-41980, train loss-0.6316, acc-0.7800, valid loss-0.5012, acc-0.8628, test loss-0.5045, acc-0.8607
Iter-41990, train loss-0.4161, acc-0.8800, valid loss-0.5012, acc-0.8628, test loss-0.5045, acc-0.8607
Iter-42000, train loss-0.4882, acc-0.8600, valid loss-0.5011, acc-0.8628, test loss-0.5044, acc-0.8607
Iter-42010, train loss-0.4769, acc-0.8600, valid loss-0.5010, acc-0.8628, test loss-0.5043, acc-0.8607
Iter-42020, train loss-0.4790, acc-0.8800, valid loss-0.5009, acc-0.8630, test loss-0.5043, acc-0.8608
Iter-42030, train loss-0.4243, acc-0.8800, valid loss-0.5009, acc-0.8630, test loss-0.5042, acc-0.8607
Iter-42040, train loss-0.4184, acc-0.8800, valid loss-0.5008, acc-0.8630, test loss-0.5042, acc-0.8609
Iter-42050, train loss-0.3092, acc-0.9600, valid loss-0.5008, acc-0.8630, test loss-0.5041, acc-0.8609
Iter-42060, train loss-0.4553, acc-0.9200, valid loss-0.5007, acc-0.8630, test loss-0.5041, acc-0.8609
Iter-42070, train loss-0.3214, acc-0.9400, valid loss-0.5006, acc-0.8630, test loss-0.5040, acc-0.8609
Iter-42080, train loss-0.6829, acc-0.8000, valid loss-0.5006, acc-0.8630, test loss-0.5039, acc-0.8609
Iter-42090, train loss-0.5479, acc-0.8400, valid loss-0.5005, acc-0.8630, test loss-0.5039, acc-0.8610
Iter-42100, train loss-0.5110, acc-0.8800, valid loss-0.5005, acc-0.8630, test loss-0.5038, acc-0.8610
Iter-42110, train loss-0.5884, acc-0.8000, valid loss-0.5004, acc-0.8630, test loss-0.5037, acc-0.8610
Iter-42120, train loss-0.8770, acc-0.7000, valid loss-0.5003, acc-0.8630, test loss-0.5037, acc-0.8611
Iter-42130, train loss-0.5696, acc-0.8400, valid loss-0.5003, acc-0.8632, test loss-0.5036, acc-0.8611
Iter-42140, train loss-0.3720, acc-0.9200, valid loss-0.5002, acc-0.8632, test loss-0.5035, acc-0.8611
Iter-42150, train loss-0.8372, acc-0.7200, valid loss-0.5001, acc-0.8630, test loss-0.5035, acc-0.8610
Iter-42160, train loss-0.4252, acc-0.9200, valid loss-0.5001, acc-0.8628, test loss-0.5034, acc-0.8610
Iter-42170, train loss-0.5623, acc-0.8400, valid loss-0.5000, acc-0.8630, test loss-0.5034, acc-0.8610
Iter-42180, train loss-0.6384, acc-0.8400, valid loss-0.5000, acc-0.8632, test loss-0.5033, acc-0.8610
Iter-42190, train loss-0.5273, acc-0.9000, valid loss-0.4999, acc-0.8632, test loss-0.5033, acc-0.8610
Iter-42200, train loss-0.3100, acc-0.9600, valid loss-0.4999, acc-0.8632, test loss-0.5032, acc-0.8610
Iter-42210, train loss-0.6729, acc-0.8200, valid loss-0.4998, acc-0.8632, test loss-0.5031, acc-0.8611
Iter-42220, train loss-0.2979, acc-0.9200, valid loss-0.4998, acc-0.8634, test loss-0.5031, acc-0.8610
Iter-42230, train loss-0.4398, acc-0.8800, valid loss-0.4997, acc-0.8632, test loss-0.5030, acc-0.8611
Iter-42240, train loss-0.5869, acc-0.8200, valid loss-0.4996, acc-0.8632, test loss-0.5030, acc-0.8612
Iter-42250, train loss-0.5181, acc-0.8600, valid loss-0.4996, acc-0.8634, test loss-0.5029, acc-0.8613
Iter-42260, train loss-0.4283, acc-0.9000, valid loss-0.4995, acc-0.8634, test loss-0.5028, acc-0.8613
Iter-42270, train loss-0.3957, acc-0.9000, valid loss-0.4994, acc-0.8632, test loss-0.5028, acc-0.8614
Iter-42280, train loss-0.5356, acc-0.8400, valid loss-0.4994, acc-0.8632, test loss-0.5027, acc-0.8614
Iter-42290, train loss-0.4430, acc-0.8200, valid loss-0.4993, acc-0.8634, test loss-0.5027, acc-0.8613
Iter-42300, train loss-0.4410, acc-0.9000, valid loss-0.4992, acc-0.8634, test loss-0.5026, acc-0.8613
Iter-42310, train loss-0.6243, acc-0.7600, valid loss-0.4992, acc-0.8634, test loss-0.5026, acc-0.8613
Iter-42320, train loss-0.4637, acc-0.8400, valid loss-0.4991, acc-0.8634, test loss-0.5025, acc-0.8613
Iter-42330, train loss-0.4076, acc-0.8200, valid loss-0.4991, acc-0.8634, test loss-0.5024, acc-0.8613
Iter-42340, train loss-0.5225, acc-0.8600, valid loss-0.4990, acc-0.8634, test loss-0.5023, acc-0.8614
Iter-42350, train loss-0.5783, acc-0.8200, valid loss-0.4989, acc-0.8634, test loss-0.5023, acc-0.8613
Iter-42360, train loss-0.4157, acc-0.9000, valid loss-0.4989, acc-0.8634, test loss-0.5022, acc-0.8614
Iter-42370, train loss-0.4456, acc-0.8800, valid loss-0.4988, acc-0.8634, test loss-0.5022, acc-0.8615
Iter-42380, train loss-0.7168, acc-0.7800, valid loss-0.4988, acc-0.8634, test loss-0.5021, acc-0.8615
Iter-42390, train loss-0.4822, acc-0.8400, valid loss-0.4987, acc-0.8632, test loss-0.5021, acc-0.8615
Iter-42400, train loss-0.5872, acc-0.8200, valid loss-0.4986, acc-0.8632, test loss-0.5020, acc-0.8615
Iter-42410, train loss-0.5194, acc-0.8800, valid loss-0.4986, acc-0.8632, test loss-0.5019, acc-0.8614
Iter-42420, train loss-0.4477, acc-0.8200, valid loss-0.4985, acc-0.8632, test loss-0.5019, acc-0.8614
Iter-42430, train loss-0.5386, acc-0.8400, valid loss-0.4984, acc-0.8634, test loss-0.5018, acc-0.8614
Iter-42440, train loss-0.4562, acc-0.9200, valid loss-0.4984, acc-0.8634, test loss-0.5017, acc-0.8614
Iter-42450, train loss-0.3995, acc-0.9000, valid loss-0.4983, acc-0.8634, test loss-0.5017, acc-0.8614
Iter-42460, train loss-0.6474, acc-0.8000, valid loss-0.4983, acc-0.8634, test loss-0.5016, acc-0.8614
Iter-42470, train loss-0.5893, acc-0.8800, valid loss-0.4982, acc-0.8634, test loss-0.5016, acc-0.8613
Iter-42480, train loss-0.5243, acc-0.8200, valid loss-0.4981, acc-0.8634, test loss-0.5015, acc-0.8614
Iter-42490, train loss-0.3842, acc-0.9200, valid loss-0.4981, acc-0.8634, test loss-0.5014, acc-0.8615
Iter-42500, train loss-0.6412, acc-0.7400, valid loss-0.4980, acc-0.8634, test loss-0.5014, acc-0.8615
Iter-42510, train loss-0.4039, acc-0.8800, valid loss-0.4979, acc-0.8634, test loss-0.5013, acc-0.8616
Iter-42520, train loss-0.5868, acc-0.8400, valid loss-0.4979, acc-0.8632, test loss-0.5012, acc-0.8616
Iter-42530, train loss-0.3513, acc-0.8800, valid loss-0.4978, acc-0.8634, test loss-0.5012, acc-0.8616
Iter-42540, train loss-0.3279, acc-0.9000, valid loss-0.4978, acc-0.8634, test loss-0.5011, acc-0.8617
Iter-42550, train loss-0.4089, acc-0.8800, valid loss-0.4977, acc-0.8634, test loss-0.5011, acc-0.8616
Iter-42560, train loss-0.4125, acc-0.9200, valid loss-0.4976, acc-0.8632, test loss-0.5010, acc-0.8617
Iter-42570, train loss-0.4498, acc-0.9200, valid loss-0.4976, acc-0.8632, test loss-0.5010, acc-0.8618
Iter-42580, train loss-0.3456, acc-0.9400, valid loss-0.4976, acc-0.8634, test loss-0.5009, acc-0.8616
Iter-42590, train loss-0.4280, acc-0.8600, valid loss-0.4975, acc-0.8636, test loss-0.5008, acc-0.8616
Iter-42600, train loss-0.6021, acc-0.8800, valid loss-0.4974, acc-0.8634, test loss-0.5008, acc-0.8618
Iter-42610, train loss-0.4767, acc-0.8800, valid loss-0.4974, acc-0.8634, test loss-0.5007, acc-0.8618
Iter-42620, train loss-0.5880, acc-0.8000, valid loss-0.4973, acc-0.8634, test loss-0.5007, acc-0.8618
Iter-42630, train loss-0.4736, acc-0.8800, valid loss-0.4973, acc-0.8634, test loss-0.5006, acc-0.8618
Iter-42640, train loss-0.7803, acc-0.7600, valid loss-0.4972, acc-0.8642, test loss-0.5006, acc-0.8619
Iter-42650, train loss-0.5207, acc-0.8000, valid loss-0.4971, acc-0.8640, test loss-0.5005, acc-0.8620
Iter-42660, train loss-0.4676, acc-0.8800, valid loss-0.4971, acc-0.8640, test loss-0.5004, acc-0.8617
Iter-42670, train loss-0.4174, acc-0.9200, valid loss-0.4970, acc-0.8638, test loss-0.5004, acc-0.8617
Iter-42680, train loss-0.5433, acc-0.8400, valid loss-0.4970, acc-0.8640, test loss-0.5003, acc-0.8620
Iter-42690, train loss-0.4962, acc-0.8600, valid loss-0.4970, acc-0.8640, test loss-0.5003, acc-0.8618
Iter-42700, train loss-0.5135, acc-0.8600, valid loss-0.4969, acc-0.8638, test loss-0.5002, acc-0.8618
Iter-42710, train loss-0.4016, acc-0.8800, valid loss-0.4968, acc-0.8638, test loss-0.5002, acc-0.8618
Iter-42720, train loss-0.4426, acc-0.8600, valid loss-0.4967, acc-0.8638, test loss-0.5001, acc-0.8617
Iter-42730, train loss-0.5708, acc-0.8800, valid loss-0.4967, acc-0.8638, test loss-0.5001, acc-0.8618
Iter-42740, train loss-0.6591, acc-0.8600, valid loss-0.4966, acc-0.8636, test loss-0.5000, acc-0.8618
Iter-42750, train loss-0.4707, acc-0.8000, valid loss-0.4966, acc-0.8638, test loss-0.4999, acc-0.8618
Iter-42760, train loss-0.4396, acc-0.8000, valid loss-0.4965, acc-0.8638, test loss-0.4999, acc-0.8619
Iter-42770, train loss-0.5317, acc-0.8600, valid loss-0.4965, acc-0.8638, test loss-0.4998, acc-0.8620
Iter-42780, train loss-0.6372, acc-0.8200, valid loss-0.4964, acc-0.8638, test loss-0.4998, acc-0.8620
Iter-42790, train loss-0.4516, acc-0.8800, valid loss-0.4964, acc-0.8638, test loss-0.4997, acc-0.8620
Iter-42800, train loss-0.3535, acc-0.9000, valid loss-0.4963, acc-0.8638, test loss-0.4997, acc-0.8620
Iter-42810, train loss-0.5685, acc-0.8600, valid loss-0.4963, acc-0.8638, test loss-0.4996, acc-0.8621
Iter-42820, train loss-0.5477, acc-0.8000, valid loss-0.4962, acc-0.8636, test loss-0.4996, acc-0.8620
Iter-42830, train loss-0.3939, acc-0.9000, valid loss-0.4962, acc-0.8636, test loss-0.4995, acc-0.8620
Iter-42840, train loss-0.6033, acc-0.8400, valid loss-0.4961, acc-0.8640, test loss-0.4995, acc-0.8620
Iter-42850, train loss-0.4445, acc-0.8800, valid loss-0.4960, acc-0.8638, test loss-0.4994, acc-0.8621
Iter-42860, train loss-0.6757, acc-0.8200, valid loss-0.4960, acc-0.8642, test loss-0.4993, acc-0.8620
Iter-42870, train loss-0.4485, acc-0.9000, valid loss-0.4959, acc-0.8640, test loss-0.4993, acc-0.8620
Iter-42880, train loss-0.4877, acc-0.9000, valid loss-0.4958, acc-0.8642, test loss-0.4992, acc-0.8620
Iter-42890, train loss-0.3993, acc-0.9000, valid loss-0.4958, acc-0.8640, test loss-0.4992, acc-0.8620
Iter-42900, train loss-0.4788, acc-0.8400, valid loss-0.4957, acc-0.8642, test loss-0.4991, acc-0.8620
Iter-42910, train loss-0.5382, acc-0.8400, valid loss-0.4956, acc-0.8642, test loss-0.4990, acc-0.8620
Iter-42920, train loss-0.3877, acc-0.9200, valid loss-0.4956, acc-0.8642, test loss-0.4990, acc-0.8620
Iter-42930, train loss-0.6712, acc-0.7400, valid loss-0.4955, acc-0.8644, test loss-0.4989, acc-0.8619
Iter-42940, train loss-0.4205, acc-0.8600, valid loss-0.4955, acc-0.8642, test loss-0.4989, acc-0.8619
Iter-42950, train loss-0.4370, acc-0.8800, valid loss-0.4954, acc-0.8642, test loss-0.4988, acc-0.8619
Iter-42960, train loss-0.4184, acc-0.9200, valid loss-0.4953, acc-0.8642, test loss-0.4988, acc-0.8619
Iter-42970, train loss-0.4030, acc-0.9000, valid loss-0.4953, acc-0.8642, test loss-0.4987, acc-0.8619
Iter-42980, train loss-0.3713, acc-0.9000, valid loss-0.4953, acc-0.8642, test loss-0.4987, acc-0.8619
Iter-42990, train loss-0.6915, acc-0.8200, valid loss-0.4952, acc-0.8644, test loss-0.4986, acc-0.8619
Iter-43000, train loss-0.3839, acc-0.9400, valid loss-0.4951, acc-0.8644, test loss-0.4985, acc-0.8619
Iter-43010, train loss-0.4606, acc-0.8600, valid loss-0.4951, acc-0.8644, test loss-0.4985, acc-0.8619
Iter-43020, train loss-0.4639, acc-0.8600, valid loss-0.4951, acc-0.8644, test loss-0.4984, acc-0.8619
Iter-43030, train loss-0.5000, acc-0.8400, valid loss-0.4950, acc-0.8644, test loss-0.4984, acc-0.8619
Iter-43040, train loss-0.5565, acc-0.8400, valid loss-0.4949, acc-0.8644, test loss-0.4983, acc-0.8619
Iter-43050, train loss-0.4827, acc-0.9000, valid loss-0.4949, acc-0.8644, test loss-0.4983, acc-0.8619
Iter-43060, train loss-0.4359, acc-0.8800, valid loss-0.4948, acc-0.8644, test loss-0.4982, acc-0.8620
Iter-43070, train loss-0.6206, acc-0.8000, valid loss-0.4948, acc-0.8646, test loss-0.4981, acc-0.8619
Iter-43080, train loss-0.4048, acc-0.9400, valid loss-0.4947, acc-0.8644, test loss-0.4981, acc-0.8619
Iter-43090, train loss-0.6827, acc-0.8400, valid loss-0.4947, acc-0.8642, test loss-0.4980, acc-0.8619
Iter-43100, train loss-0.4102, acc-0.9000, valid loss-0.4946, acc-0.8640, test loss-0.4980, acc-0.8619
Iter-43110, train loss-0.6371, acc-0.8400, valid loss-0.4946, acc-0.8642, test loss-0.4979, acc-0.8619
Iter-43120, train loss-0.7975, acc-0.7600, valid loss-0.4945, acc-0.8644, test loss-0.4978, acc-0.8620
Iter-43130, train loss-0.4952, acc-0.8400, valid loss-0.4945, acc-0.8644, test loss-0.4978, acc-0.8620
Iter-43140, train loss-0.5449, acc-0.8200, valid loss-0.4944, acc-0.8644, test loss-0.4977, acc-0.8621
Iter-43150, train loss-0.4501, acc-0.9000, valid loss-0.4943, acc-0.8644, test loss-0.4977, acc-0.8623
Iter-43160, train loss-0.6419, acc-0.8600, valid loss-0.4942, acc-0.8644, test loss-0.4976, acc-0.8623
Iter-43170, train loss-0.4073, acc-0.8800, valid loss-0.4942, acc-0.8644, test loss-0.4976, acc-0.8623
Iter-43180, train loss-0.5274, acc-0.8600, valid loss-0.4941, acc-0.8644, test loss-0.4975, acc-0.8622
Iter-43190, train loss-0.3456, acc-0.9400, valid loss-0.4941, acc-0.8644, test loss-0.4974, acc-0.8621
Iter-43200, train loss-0.5110, acc-0.8400, valid loss-0.4940, acc-0.8644, test loss-0.4974, acc-0.8622
Iter-43210, train loss-0.3343, acc-0.9200, valid loss-0.4939, acc-0.8646, test loss-0.4973, acc-0.8622
Iter-43220, train loss-0.8159, acc-0.7600, valid loss-0.4939, acc-0.8646, test loss-0.4973, acc-0.8622
Iter-43230, train loss-0.4524, acc-0.8800, valid loss-0.4938, acc-0.8646, test loss-0.4972, acc-0.8622
Iter-43240, train loss-0.5352, acc-0.8200, valid loss-0.4938, acc-0.8646, test loss-0.4972, acc-0.8621
Iter-43250, train loss-0.6977, acc-0.7600, valid loss-0.4937, acc-0.8646, test loss-0.4971, acc-0.8621
Iter-43260, train loss-0.4848, acc-0.8600, valid loss-0.4937, acc-0.8648, test loss-0.4971, acc-0.8620
Iter-43270, train loss-0.5550, acc-0.8400, valid loss-0.4936, acc-0.8648, test loss-0.4970, acc-0.8621
Iter-43280, train loss-0.3727, acc-0.9200, valid loss-0.4935, acc-0.8648, test loss-0.4970, acc-0.8622
Iter-43290, train loss-0.5007, acc-0.9000, valid loss-0.4935, acc-0.8650, test loss-0.4969, acc-0.8622
Iter-43300, train loss-0.7288, acc-0.7400, valid loss-0.4934, acc-0.8650, test loss-0.4969, acc-0.8623
Iter-43310, train loss-0.5322, acc-0.8400, valid loss-0.4934, acc-0.8648, test loss-0.4968, acc-0.8624
Iter-43320, train loss-0.3371, acc-0.8800, valid loss-0.4933, acc-0.8650, test loss-0.4967, acc-0.8621
Iter-43330, train loss-0.4147, acc-0.9000, valid loss-0.4932, acc-0.8650, test loss-0.4967, acc-0.8624
Iter-43340, train loss-0.3491, acc-0.8600, valid loss-0.4932, acc-0.8650, test loss-0.4966, acc-0.8623
Iter-43350, train loss-0.4433, acc-0.8800, valid loss-0.4931, acc-0.8650, test loss-0.4966, acc-0.8624
Iter-43360, train loss-0.5137, acc-0.8000, valid loss-0.4930, acc-0.8650, test loss-0.4965, acc-0.8623
Iter-43370, train loss-0.3746, acc-0.8800, valid loss-0.4930, acc-0.8650, test loss-0.4965, acc-0.8623
Iter-43380, train loss-0.4776, acc-0.9000, valid loss-0.4929, acc-0.8650, test loss-0.4964, acc-0.8623
Iter-43390, train loss-0.4617, acc-0.8600, valid loss-0.4929, acc-0.8650, test loss-0.4964, acc-0.8623
Iter-43400, train loss-0.6673, acc-0.8000, valid loss-0.4928, acc-0.8650, test loss-0.4963, acc-0.8623
Iter-43410, train loss-0.4680, acc-0.8600, valid loss-0.4927, acc-0.8650, test loss-0.4962, acc-0.8625
Iter-43420, train loss-0.5282, acc-0.8600, valid loss-0.4927, acc-0.8650, test loss-0.4962, acc-0.8625
Iter-43430, train loss-0.3139, acc-0.9400, valid loss-0.4926, acc-0.8650, test loss-0.4961, acc-0.8625
Iter-43440, train loss-0.4181, acc-0.9000, valid loss-0.4926, acc-0.8650, test loss-0.4960, acc-0.8624
Iter-43450, train loss-0.3466, acc-0.9000, valid loss-0.4925, acc-0.8652, test loss-0.4960, acc-0.8624
Iter-43460, train loss-0.5185, acc-0.8200, valid loss-0.4925, acc-0.8652, test loss-0.4959, acc-0.8625
Iter-43470, train loss-0.3936, acc-0.8800, valid loss-0.4924, acc-0.8652, test loss-0.4959, acc-0.8626
Iter-43480, train loss-0.3051, acc-0.9400, valid loss-0.4923, acc-0.8652, test loss-0.4958, acc-0.8625
Iter-43490, train loss-0.6106, acc-0.8200, valid loss-0.4923, acc-0.8652, test loss-0.4958, acc-0.8626
Iter-43500, train loss-0.6282, acc-0.8000, valid loss-0.4922, acc-0.8654, test loss-0.4957, acc-0.8626
Iter-43510, train loss-0.5147, acc-0.8200, valid loss-0.4922, acc-0.8654, test loss-0.4956, acc-0.8625
Iter-43520, train loss-0.3895, acc-0.9600, valid loss-0.4921, acc-0.8654, test loss-0.4956, acc-0.8625
Iter-43530, train loss-0.5352, acc-0.8800, valid loss-0.4920, acc-0.8654, test loss-0.4955, acc-0.8626
Iter-43540, train loss-0.3608, acc-0.9000, valid loss-0.4920, acc-0.8654, test loss-0.4955, acc-0.8626
Iter-43550, train loss-0.4512, acc-0.9200, valid loss-0.4919, acc-0.8654, test loss-0.4954, acc-0.8626
Iter-43560, train loss-0.5630, acc-0.8000, valid loss-0.4919, acc-0.8654, test loss-0.4954, acc-0.8627
Iter-43570, train loss-0.5267, acc-0.8400, valid loss-0.4918, acc-0.8652, test loss-0.4953, acc-0.8627
Iter-43580, train loss-0.3091, acc-0.9400, valid loss-0.4917, acc-0.8654, test loss-0.4952, acc-0.8627
Iter-43590, train loss-0.4991, acc-0.8800, valid loss-0.4917, acc-0.8654, test loss-0.4952, acc-0.8626
Iter-43600, train loss-0.4747, acc-0.8800, valid loss-0.4916, acc-0.8654, test loss-0.4951, acc-0.8627
Iter-43610, train loss-0.5692, acc-0.8400, valid loss-0.4916, acc-0.8654, test loss-0.4951, acc-0.8626
Iter-43620, train loss-0.4131, acc-0.8800, valid loss-0.4915, acc-0.8654, test loss-0.4950, acc-0.8625
Iter-43630, train loss-0.4258, acc-0.9000, valid loss-0.4915, acc-0.8654, test loss-0.4950, acc-0.8626
Iter-43640, train loss-0.5101, acc-0.8200, valid loss-0.4914, acc-0.8654, test loss-0.4949, acc-0.8626
Iter-43650, train loss-0.3949, acc-0.9000, valid loss-0.4913, acc-0.8654, test loss-0.4948, acc-0.8626
Iter-43660, train loss-0.3512, acc-0.9000, valid loss-0.4913, acc-0.8656, test loss-0.4948, acc-0.8625
Iter-43670, train loss-0.4561, acc-0.8600, valid loss-0.4912, acc-0.8656, test loss-0.4947, acc-0.8626
Iter-43680, train loss-0.7363, acc-0.7400, valid loss-0.4911, acc-0.8656, test loss-0.4947, acc-0.8626
Iter-43690, train loss-0.5654, acc-0.7800, valid loss-0.4911, acc-0.8656, test loss-0.4946, acc-0.8626
Iter-43700, train loss-0.4315, acc-0.9200, valid loss-0.4911, acc-0.8656, test loss-0.4946, acc-0.8628
Iter-43710, train loss-0.5843, acc-0.8200, valid loss-0.4910, acc-0.8656, test loss-0.4945, acc-0.8627
Iter-43720, train loss-0.4640, acc-0.9000, valid loss-0.4909, acc-0.8656, test loss-0.4945, acc-0.8627
Iter-43730, train loss-0.4194, acc-0.9000, valid loss-0.4909, acc-0.8656, test loss-0.4944, acc-0.8627
Iter-43740, train loss-0.7290, acc-0.7800, valid loss-0.4908, acc-0.8656, test loss-0.4944, acc-0.8627
Iter-43750, train loss-0.5834, acc-0.8400, valid loss-0.4908, acc-0.8656, test loss-0.4943, acc-0.8627
Iter-43760, train loss-0.5354, acc-0.8400, valid loss-0.4907, acc-0.8656, test loss-0.4943, acc-0.8628
Iter-43770, train loss-0.5904, acc-0.7800, valid loss-0.4907, acc-0.8656, test loss-0.4942, acc-0.8625
Iter-43780, train loss-0.3376, acc-0.9400, valid loss-0.4906, acc-0.8656, test loss-0.4941, acc-0.8625
Iter-43790, train loss-0.3764, acc-0.8800, valid loss-0.4905, acc-0.8658, test loss-0.4940, acc-0.8629
Iter-43800, train loss-0.4068, acc-0.9200, valid loss-0.4905, acc-0.8658, test loss-0.4940, acc-0.8628
Iter-43810, train loss-0.4929, acc-0.8600, valid loss-0.4904, acc-0.8658, test loss-0.4939, acc-0.8629
Iter-43820, train loss-0.3989, acc-0.9200, valid loss-0.4904, acc-0.8658, test loss-0.4939, acc-0.8629
Iter-43830, train loss-0.4162, acc-0.8600, valid loss-0.4903, acc-0.8656, test loss-0.4938, acc-0.8629
Iter-43840, train loss-0.5255, acc-0.9200, valid loss-0.4902, acc-0.8656, test loss-0.4938, acc-0.8629
Iter-43850, train loss-0.5959, acc-0.8600, valid loss-0.4902, acc-0.8658, test loss-0.4937, acc-0.8630
Iter-43860, train loss-0.4807, acc-0.8800, valid loss-0.4901, acc-0.8656, test loss-0.4936, acc-0.8629
Iter-43870, train loss-0.4475, acc-0.9000, valid loss-0.4901, acc-0.8658, test loss-0.4936, acc-0.8629
Iter-43880, train loss-0.6766, acc-0.7800, valid loss-0.4900, acc-0.8656, test loss-0.4935, acc-0.8628
Iter-43890, train loss-0.4946, acc-0.8800, valid loss-0.4900, acc-0.8656, test loss-0.4935, acc-0.8628
Iter-43900, train loss-0.3927, acc-0.9200, valid loss-0.4899, acc-0.8656, test loss-0.4934, acc-0.8628
Iter-43910, train loss-0.7444, acc-0.7800, valid loss-0.4899, acc-0.8654, test loss-0.4934, acc-0.8629
Iter-43920, train loss-0.4018, acc-0.8800, valid loss-0.4898, acc-0.8656, test loss-0.4933, acc-0.8629
Iter-43930, train loss-0.5254, acc-0.8600, valid loss-0.4898, acc-0.8656, test loss-0.4933, acc-0.8629
Iter-43940, train loss-0.6058, acc-0.7800, valid loss-0.4897, acc-0.8656, test loss-0.4932, acc-0.8630
Iter-43950, train loss-0.4119, acc-0.8400, valid loss-0.4897, acc-0.8656, test loss-0.4932, acc-0.8630
Iter-43960, train loss-0.4013, acc-0.9200, valid loss-0.4896, acc-0.8654, test loss-0.4931, acc-0.8630
Iter-43970, train loss-0.6009, acc-0.8200, valid loss-0.4895, acc-0.8658, test loss-0.4931, acc-0.8631
Iter-43980, train loss-0.4630, acc-0.8600, valid loss-0.4895, acc-0.8654, test loss-0.4930, acc-0.8630
Iter-43990, train loss-0.4379, acc-0.8800, valid loss-0.4894, acc-0.8658, test loss-0.4930, acc-0.8631
Iter-44000, train loss-0.4559, acc-0.8400, valid loss-0.4894, acc-0.8656, test loss-0.4929, acc-0.8630
Iter-44010, train loss-0.5024, acc-0.8400, valid loss-0.4893, acc-0.8654, test loss-0.4928, acc-0.8630
Iter-44020, train loss-0.4747, acc-0.8800, valid loss-0.4892, acc-0.8656, test loss-0.4928, acc-0.8631
Iter-44030, train loss-0.4682, acc-0.8800, valid loss-0.4892, acc-0.8656, test loss-0.4927, acc-0.8631
Iter-44040, train loss-0.3898, acc-0.9400, valid loss-0.4891, acc-0.8656, test loss-0.4927, acc-0.8631
Iter-44050, train loss-0.6328, acc-0.8400, valid loss-0.4891, acc-0.8656, test loss-0.4926, acc-0.8632
Iter-44060, train loss-0.5544, acc-0.8200, valid loss-0.4890, acc-0.8658, test loss-0.4926, acc-0.8632
Iter-44070, train loss-0.3544, acc-0.9000, valid loss-0.4890, acc-0.8656, test loss-0.4925, acc-0.8633
Iter-44080, train loss-0.6712, acc-0.8800, valid loss-0.4890, acc-0.8660, test loss-0.4925, acc-0.8633
Iter-44090, train loss-0.6567, acc-0.8000, valid loss-0.4889, acc-0.8660, test loss-0.4924, acc-0.8633
Iter-44100, train loss-0.4417, acc-0.8800, valid loss-0.4889, acc-0.8660, test loss-0.4924, acc-0.8633
Iter-44110, train loss-0.5028, acc-0.8400, valid loss-0.4888, acc-0.8658, test loss-0.4923, acc-0.8633
Iter-44120, train loss-0.5382, acc-0.8600, valid loss-0.4887, acc-0.8660, test loss-0.4922, acc-0.8633
Iter-44130, train loss-0.3066, acc-0.9200, valid loss-0.4887, acc-0.8660, test loss-0.4922, acc-0.8633
Iter-44140, train loss-0.3644, acc-0.9200, valid loss-0.4886, acc-0.8662, test loss-0.4921, acc-0.8633
Iter-44150, train loss-0.4413, acc-0.8800, valid loss-0.4886, acc-0.8662, test loss-0.4921, acc-0.8633
Iter-44160, train loss-0.3857, acc-0.9200, valid loss-0.4885, acc-0.8664, test loss-0.4920, acc-0.8633
Iter-44170, train loss-0.5326, acc-0.8400, valid loss-0.4885, acc-0.8664, test loss-0.4919, acc-0.8633
Iter-44180, train loss-0.5589, acc-0.7800, valid loss-0.4884, acc-0.8668, test loss-0.4919, acc-0.8634
Iter-44190, train loss-0.5458, acc-0.8800, valid loss-0.4883, acc-0.8666, test loss-0.4918, acc-0.8633
Iter-44200, train loss-0.5619, acc-0.8600, valid loss-0.4883, acc-0.8668, test loss-0.4918, acc-0.8634
Iter-44210, train loss-0.5522, acc-0.7600, valid loss-0.4882, acc-0.8668, test loss-0.4917, acc-0.8633
Iter-44220, train loss-0.4978, acc-0.8600, valid loss-0.4882, acc-0.8664, test loss-0.4917, acc-0.8634
Iter-44230, train loss-0.4316, acc-0.8800, valid loss-0.4881, acc-0.8664, test loss-0.4916, acc-0.8636
Iter-44240, train loss-0.4771, acc-0.8800, valid loss-0.4880, acc-0.8664, test loss-0.4915, acc-0.8636
Iter-44250, train loss-0.4121, acc-0.8600, valid loss-0.4880, acc-0.8664, test loss-0.4915, acc-0.8636
Iter-44260, train loss-0.4820, acc-0.8400, valid loss-0.4879, acc-0.8664, test loss-0.4914, acc-0.8636
Iter-44270, train loss-0.6984, acc-0.8200, valid loss-0.4879, acc-0.8664, test loss-0.4914, acc-0.8635
Iter-44280, train loss-0.5510, acc-0.7800, valid loss-0.4878, acc-0.8666, test loss-0.4913, acc-0.8635
Iter-44290, train loss-0.5551, acc-0.8800, valid loss-0.4878, acc-0.8666, test loss-0.4913, acc-0.8635
Iter-44300, train loss-0.4106, acc-0.8400, valid loss-0.4877, acc-0.8666, test loss-0.4912, acc-0.8635
Iter-44310, train loss-0.4145, acc-0.9000, valid loss-0.4877, acc-0.8664, test loss-0.4912, acc-0.8636
Iter-44320, train loss-0.5730, acc-0.8600, valid loss-0.4876, acc-0.8666, test loss-0.4911, acc-0.8635
Iter-44330, train loss-0.6683, acc-0.8600, valid loss-0.4875, acc-0.8666, test loss-0.4911, acc-0.8635
Iter-44340, train loss-0.4827, acc-0.8400, valid loss-0.4875, acc-0.8666, test loss-0.4910, acc-0.8637
Iter-44350, train loss-0.6439, acc-0.8200, valid loss-0.4874, acc-0.8664, test loss-0.4910, acc-0.8637
Iter-44360, train loss-0.4519, acc-0.8600, valid loss-0.4874, acc-0.8666, test loss-0.4909, acc-0.8637
Iter-44370, train loss-0.3617, acc-0.9000, valid loss-0.4873, acc-0.8664, test loss-0.4909, acc-0.8638
Iter-44380, train loss-0.7021, acc-0.8400, valid loss-0.4872, acc-0.8664, test loss-0.4908, acc-0.8639
Iter-44390, train loss-0.3866, acc-0.8600, valid loss-0.4872, acc-0.8662, test loss-0.4907, acc-0.8638
Iter-44400, train loss-0.4355, acc-0.9000, valid loss-0.4871, acc-0.8666, test loss-0.4907, acc-0.8638
Iter-44410, train loss-0.4776, acc-0.9000, valid loss-0.4871, acc-0.8666, test loss-0.4906, acc-0.8638
Iter-44420, train loss-0.4752, acc-0.8800, valid loss-0.4870, acc-0.8666, test loss-0.4906, acc-0.8638
Iter-44430, train loss-0.5113, acc-0.8600, valid loss-0.4870, acc-0.8666, test loss-0.4905, acc-0.8638
Iter-44440, train loss-0.5467, acc-0.8600, valid loss-0.4869, acc-0.8666, test loss-0.4905, acc-0.8639
Iter-44450, train loss-0.6605, acc-0.7800, valid loss-0.4869, acc-0.8668, test loss-0.4904, acc-0.8640
Iter-44460, train loss-0.4032, acc-0.8400, valid loss-0.4868, acc-0.8664, test loss-0.4903, acc-0.8640
Iter-44470, train loss-0.4783, acc-0.8800, valid loss-0.4868, acc-0.8668, test loss-0.4903, acc-0.8642
Iter-44480, train loss-0.5513, acc-0.8400, valid loss-0.4867, acc-0.8664, test loss-0.4902, acc-0.8642
Iter-44490, train loss-0.6329, acc-0.8200, valid loss-0.4866, acc-0.8668, test loss-0.4902, acc-0.8642
Iter-44500, train loss-0.6335, acc-0.7600, valid loss-0.4866, acc-0.8670, test loss-0.4901, acc-0.8643
Iter-44510, train loss-0.4252, acc-0.8400, valid loss-0.4865, acc-0.8672, test loss-0.4900, acc-0.8642
Iter-44520, train loss-0.6217, acc-0.8000, valid loss-0.4865, acc-0.8672, test loss-0.4900, acc-0.8644
Iter-44530, train loss-0.6230, acc-0.8400, valid loss-0.4865, acc-0.8672, test loss-0.4899, acc-0.8644
Iter-44540, train loss-0.3770, acc-0.9200, valid loss-0.4864, acc-0.8672, test loss-0.4899, acc-0.8645
Iter-44550, train loss-0.5504, acc-0.8400, valid loss-0.4864, acc-0.8672, test loss-0.4898, acc-0.8645
Iter-44560, train loss-0.4966, acc-0.8600, valid loss-0.4863, acc-0.8674, test loss-0.4897, acc-0.8645
Iter-44570, train loss-0.3326, acc-0.9000, valid loss-0.4862, acc-0.8672, test loss-0.4897, acc-0.8644
Iter-44580, train loss-0.4862, acc-0.8800, valid loss-0.4862, acc-0.8672, test loss-0.4896, acc-0.8644
Iter-44590, train loss-0.4817, acc-0.8800, valid loss-0.4861, acc-0.8674, test loss-0.4896, acc-0.8644
Iter-44600, train loss-0.4699, acc-0.9200, valid loss-0.4861, acc-0.8674, test loss-0.4895, acc-0.8646
Iter-44610, train loss-0.6032, acc-0.7600, valid loss-0.4860, acc-0.8674, test loss-0.4895, acc-0.8646
Iter-44620, train loss-0.5848, acc-0.8200, valid loss-0.4860, acc-0.8674, test loss-0.4894, acc-0.8644
Iter-44630, train loss-0.5770, acc-0.8600, valid loss-0.4859, acc-0.8672, test loss-0.4894, acc-0.8643
Iter-44640, train loss-0.5186, acc-0.8800, valid loss-0.4859, acc-0.8670, test loss-0.4893, acc-0.8645
Iter-44650, train loss-0.3860, acc-0.9400, valid loss-0.4858, acc-0.8674, test loss-0.4893, acc-0.8646
Iter-44660, train loss-0.5358, acc-0.8400, valid loss-0.4858, acc-0.8672, test loss-0.4892, acc-0.8647
Iter-44670, train loss-0.6209, acc-0.8400, valid loss-0.4857, acc-0.8674, test loss-0.4891, acc-0.8647
Iter-44680, train loss-0.5911, acc-0.8400, valid loss-0.4856, acc-0.8674, test loss-0.4891, acc-0.8645
Iter-44690, train loss-0.3674, acc-0.9000, valid loss-0.4856, acc-0.8674, test loss-0.4890, acc-0.8646
Iter-44700, train loss-0.5114, acc-0.9400, valid loss-0.4855, acc-0.8674, test loss-0.4890, acc-0.8647
Iter-44710, train loss-0.5400, acc-0.8400, valid loss-0.4855, acc-0.8674, test loss-0.4889, acc-0.8648
Iter-44720, train loss-0.7808, acc-0.7800, valid loss-0.4854, acc-0.8674, test loss-0.4889, acc-0.8648
Iter-44730, train loss-0.3341, acc-0.9200, valid loss-0.4854, acc-0.8674, test loss-0.4888, acc-0.8647
Iter-44740, train loss-0.3772, acc-0.9000, valid loss-0.4853, acc-0.8674, test loss-0.4888, acc-0.8648
Iter-44750, train loss-0.5539, acc-0.8200, valid loss-0.4853, acc-0.8674, test loss-0.4887, acc-0.8646
Iter-44760, train loss-0.5007, acc-0.8800, valid loss-0.4852, acc-0.8674, test loss-0.4887, acc-0.8647
Iter-44770, train loss-0.6477, acc-0.7800, valid loss-0.4851, acc-0.8674, test loss-0.4886, acc-0.8647
Iter-44780, train loss-0.4578, acc-0.8800, valid loss-0.4851, acc-0.8674, test loss-0.4885, acc-0.8647
Iter-44790, train loss-0.3812, acc-0.9200, valid loss-0.4850, acc-0.8674, test loss-0.4885, acc-0.8647
Iter-44800, train loss-0.3337, acc-0.9400, valid loss-0.4850, acc-0.8672, test loss-0.4884, acc-0.8647
Iter-44810, train loss-0.5952, acc-0.7800, valid loss-0.4849, acc-0.8674, test loss-0.4884, acc-0.8647
Iter-44820, train loss-0.8059, acc-0.7400, valid loss-0.4849, acc-0.8674, test loss-0.4883, acc-0.8647
Iter-44830, train loss-0.6515, acc-0.8400, valid loss-0.4848, acc-0.8674, test loss-0.4882, acc-0.8648
Iter-44840, train loss-0.4268, acc-0.8400, valid loss-0.4848, acc-0.8674, test loss-0.4882, acc-0.8649
Iter-44850, train loss-0.2694, acc-0.9400, valid loss-0.4847, acc-0.8674, test loss-0.4881, acc-0.8648
Iter-44860, train loss-0.5629, acc-0.8800, valid loss-0.4847, acc-0.8674, test loss-0.4881, acc-0.8649
Iter-44870, train loss-0.4728, acc-0.8800, valid loss-0.4846, acc-0.8674, test loss-0.4880, acc-0.8649
Iter-44880, train loss-0.5335, acc-0.8600, valid loss-0.4845, acc-0.8674, test loss-0.4879, acc-0.8648
Iter-44890, train loss-0.4553, acc-0.8600, valid loss-0.4845, acc-0.8674, test loss-0.4879, acc-0.8648
Iter-44900, train loss-0.4031, acc-0.9000, valid loss-0.4844, acc-0.8674, test loss-0.4878, acc-0.8649
Iter-44910, train loss-0.4622, acc-0.8800, valid loss-0.4844, acc-0.8674, test loss-0.4878, acc-0.8649
Iter-44920, train loss-0.5984, acc-0.8200, valid loss-0.4843, acc-0.8674, test loss-0.4877, acc-0.8648
Iter-44930, train loss-0.5296, acc-0.8200, valid loss-0.4843, acc-0.8672, test loss-0.4877, acc-0.8649
Iter-44940, train loss-0.6674, acc-0.8200, valid loss-0.4842, acc-0.8672, test loss-0.4876, acc-0.8649
Iter-44950, train loss-0.5647, acc-0.8400, valid loss-0.4842, acc-0.8672, test loss-0.4875, acc-0.8649
Iter-44960, train loss-0.3681, acc-0.9200, valid loss-0.4841, acc-0.8674, test loss-0.4875, acc-0.8649
Iter-44970, train loss-0.4310, acc-0.8800, valid loss-0.4841, acc-0.8674, test loss-0.4874, acc-0.8650
Iter-44980, train loss-0.5047, acc-0.8400, valid loss-0.4840, acc-0.8674, test loss-0.4874, acc-0.8650
Iter-44990, train loss-0.5722, acc-0.8400, valid loss-0.4840, acc-0.8674, test loss-0.4873, acc-0.8650
Iter-45000, train loss-0.5147, acc-0.8800, valid loss-0.4839, acc-0.8672, test loss-0.4873, acc-0.8650
Iter-45010, train loss-0.5681, acc-0.8000, valid loss-0.4839, acc-0.8672, test loss-0.4872, acc-0.8651
Iter-45020, train loss-0.4784, acc-0.8400, valid loss-0.4838, acc-0.8674, test loss-0.4872, acc-0.8651
Iter-45030, train loss-0.3682, acc-0.8800, valid loss-0.4837, acc-0.8676, test loss-0.4871, acc-0.8651
Iter-45040, train loss-0.5193, acc-0.8800, valid loss-0.4837, acc-0.8676, test loss-0.4871, acc-0.8651
Iter-45050, train loss-0.4150, acc-0.9000, valid loss-0.4836, acc-0.8676, test loss-0.4870, acc-0.8651
Iter-45060, train loss-0.5672, acc-0.8400, valid loss-0.4836, acc-0.8676, test loss-0.4870, acc-0.8651
Iter-45070, train loss-0.3821, acc-0.8800, valid loss-0.4835, acc-0.8676, test loss-0.4869, acc-0.8651
Iter-45080, train loss-0.6091, acc-0.8200, valid loss-0.4835, acc-0.8676, test loss-0.4869, acc-0.8651
Iter-45090, train loss-0.3471, acc-0.9000, valid loss-0.4834, acc-0.8676, test loss-0.4868, acc-0.8651
Iter-45100, train loss-0.5124, acc-0.8000, valid loss-0.4834, acc-0.8676, test loss-0.4867, acc-0.8651
Iter-45110, train loss-0.4709, acc-0.8600, valid loss-0.4833, acc-0.8676, test loss-0.4867, acc-0.8651
Iter-45120, train loss-0.4906, acc-0.8600, valid loss-0.4832, acc-0.8678, test loss-0.4866, acc-0.8651
Iter-45130, train loss-0.6007, acc-0.8600, valid loss-0.4832, acc-0.8676, test loss-0.4866, acc-0.8651
Iter-45140, train loss-0.4583, acc-0.8600, valid loss-0.4831, acc-0.8676, test loss-0.4865, acc-0.8651
Iter-45150, train loss-0.5525, acc-0.8600, valid loss-0.4831, acc-0.8678, test loss-0.4865, acc-0.8652
Iter-45160, train loss-0.5030, acc-0.8400, valid loss-0.4830, acc-0.8676, test loss-0.4864, acc-0.8651
Iter-45170, train loss-0.5409, acc-0.8400, valid loss-0.4830, acc-0.8676, test loss-0.4863, acc-0.8651
Iter-45180, train loss-0.4458, acc-0.9000, valid loss-0.4829, acc-0.8676, test loss-0.4863, acc-0.8651
Iter-45190, train loss-0.5007, acc-0.8800, valid loss-0.4829, acc-0.8676, test loss-0.4862, acc-0.8650
Iter-45200, train loss-0.4369, acc-0.8800, valid loss-0.4828, acc-0.8676, test loss-0.4862, acc-0.8651
Iter-45210, train loss-0.5685, acc-0.8000, valid loss-0.4828, acc-0.8676, test loss-0.4862, acc-0.8651
Iter-45220, train loss-0.3862, acc-0.9200, valid loss-0.4827, acc-0.8676, test loss-0.4861, acc-0.8650
Iter-45230, train loss-0.6042, acc-0.8800, valid loss-0.4826, acc-0.8674, test loss-0.4861, acc-0.8651
Iter-45240, train loss-0.3361, acc-0.9200, valid loss-0.4826, acc-0.8676, test loss-0.4860, acc-0.8651
Iter-45250, train loss-0.5335, acc-0.8400, valid loss-0.4825, acc-0.8676, test loss-0.4859, acc-0.8650
Iter-45260, train loss-0.4773, acc-0.8600, valid loss-0.4825, acc-0.8676, test loss-0.4859, acc-0.8652
Iter-45270, train loss-0.3475, acc-0.8400, valid loss-0.4824, acc-0.8676, test loss-0.4858, acc-0.8651
Iter-45280, train loss-0.5328, acc-0.8200, valid loss-0.4824, acc-0.8676, test loss-0.4858, acc-0.8651
Iter-45290, train loss-0.3602, acc-0.9200, valid loss-0.4823, acc-0.8676, test loss-0.4857, acc-0.8652
Iter-45300, train loss-0.5049, acc-0.8800, valid loss-0.4823, acc-0.8676, test loss-0.4857, acc-0.8652
Iter-45310, train loss-0.6801, acc-0.8400, valid loss-0.4822, acc-0.8676, test loss-0.4856, acc-0.8652
Iter-45320, train loss-0.4182, acc-0.8800, valid loss-0.4822, acc-0.8678, test loss-0.4856, acc-0.8653
Iter-45330, train loss-0.4861, acc-0.8800, valid loss-0.4821, acc-0.8678, test loss-0.4855, acc-0.8654
Iter-45340, train loss-0.7125, acc-0.8000, valid loss-0.4821, acc-0.8678, test loss-0.4855, acc-0.8652
Iter-45350, train loss-0.4826, acc-0.8600, valid loss-0.4820, acc-0.8680, test loss-0.4854, acc-0.8653
Iter-45360, train loss-0.5728, acc-0.8400, valid loss-0.4819, acc-0.8680, test loss-0.4853, acc-0.8653
Iter-45370, train loss-0.4523, acc-0.9000, valid loss-0.4819, acc-0.8680, test loss-0.4853, acc-0.8652
Iter-45380, train loss-0.5348, acc-0.8800, valid loss-0.4818, acc-0.8682, test loss-0.4852, acc-0.8653
Iter-45390, train loss-0.4458, acc-0.8400, valid loss-0.4818, acc-0.8682, test loss-0.4852, acc-0.8655
Iter-45400, train loss-0.5993, acc-0.8000, valid loss-0.4817, acc-0.8680, test loss-0.4851, acc-0.8654
Iter-45410, train loss-0.5018, acc-0.8800, valid loss-0.4817, acc-0.8680, test loss-0.4851, acc-0.8653
Iter-45420, train loss-0.3764, acc-0.9000, valid loss-0.4816, acc-0.8678, test loss-0.4850, acc-0.8654
Iter-45430, train loss-0.5986, acc-0.8200, valid loss-0.4816, acc-0.8680, test loss-0.4850, acc-0.8653
Iter-45440, train loss-0.5017, acc-0.8400, valid loss-0.4815, acc-0.8680, test loss-0.4849, acc-0.8653
Iter-45450, train loss-0.4268, acc-0.8400, valid loss-0.4815, acc-0.8680, test loss-0.4849, acc-0.8654
Iter-45460, train loss-0.5979, acc-0.8200, valid loss-0.4814, acc-0.8680, test loss-0.4848, acc-0.8654
Iter-45470, train loss-0.5026, acc-0.8000, valid loss-0.4814, acc-0.8680, test loss-0.4848, acc-0.8654
Iter-45480, train loss-0.4499, acc-0.8600, valid loss-0.4813, acc-0.8680, test loss-0.4847, acc-0.8653
Iter-45490, train loss-0.7052, acc-0.8000, valid loss-0.4813, acc-0.8682, test loss-0.4847, acc-0.8653
Iter-45500, train loss-0.3910, acc-0.8800, valid loss-0.4812, acc-0.8680, test loss-0.4846, acc-0.8652
Iter-45510, train loss-0.3184, acc-0.9400, valid loss-0.4812, acc-0.8678, test loss-0.4846, acc-0.8652
Iter-45520, train loss-0.5904, acc-0.8800, valid loss-0.4811, acc-0.8678, test loss-0.4845, acc-0.8652
Iter-45530, train loss-0.4785, acc-0.8600, valid loss-0.4811, acc-0.8680, test loss-0.4845, acc-0.8653
Iter-45540, train loss-0.3926, acc-0.9600, valid loss-0.4810, acc-0.8680, test loss-0.4844, acc-0.8653
Iter-45550, train loss-0.5375, acc-0.9000, valid loss-0.4809, acc-0.8680, test loss-0.4843, acc-0.8654
Iter-45560, train loss-0.3232, acc-0.9200, valid loss-0.4809, acc-0.8678, test loss-0.4843, acc-0.8655
Iter-45570, train loss-0.6462, acc-0.8200, valid loss-0.4808, acc-0.8680, test loss-0.4842, acc-0.8654
Iter-45580, train loss-0.4040, acc-0.9200, valid loss-0.4808, acc-0.8684, test loss-0.4842, acc-0.8656
Iter-45590, train loss-0.6723, acc-0.8200, valid loss-0.4807, acc-0.8686, test loss-0.4841, acc-0.8656
Iter-45600, train loss-0.5723, acc-0.8200, valid loss-0.4807, acc-0.8684, test loss-0.4840, acc-0.8657
Iter-45610, train loss-0.4603, acc-0.8400, valid loss-0.4806, acc-0.8686, test loss-0.4840, acc-0.8657
Iter-45620, train loss-0.7418, acc-0.8200, valid loss-0.4805, acc-0.8684, test loss-0.4839, acc-0.8657
Iter-45630, train loss-0.5968, acc-0.8600, valid loss-0.4805, acc-0.8686, test loss-0.4839, acc-0.8659
Iter-45640, train loss-0.5905, acc-0.8200, valid loss-0.4804, acc-0.8686, test loss-0.4838, acc-0.8658
Iter-45650, train loss-0.5467, acc-0.8400, valid loss-0.4804, acc-0.8686, test loss-0.4838, acc-0.8658
Iter-45660, train loss-0.4973, acc-0.9200, valid loss-0.4803, acc-0.8686, test loss-0.4837, acc-0.8657
Iter-45670, train loss-0.7022, acc-0.7800, valid loss-0.4803, acc-0.8688, test loss-0.4837, acc-0.8657
Iter-45680, train loss-0.5428, acc-0.8400, valid loss-0.4802, acc-0.8690, test loss-0.4836, acc-0.8657
Iter-45690, train loss-0.6182, acc-0.8200, valid loss-0.4802, acc-0.8688, test loss-0.4836, acc-0.8659
Iter-45700, train loss-0.3800, acc-0.8800, valid loss-0.4801, acc-0.8690, test loss-0.4836, acc-0.8658
Iter-45710, train loss-0.5211, acc-0.8400, valid loss-0.4801, acc-0.8688, test loss-0.4835, acc-0.8658
Iter-45720, train loss-0.5977, acc-0.8600, valid loss-0.4800, acc-0.8688, test loss-0.4835, acc-0.8661
Iter-45730, train loss-0.4882, acc-0.8600, valid loss-0.4799, acc-0.8690, test loss-0.4834, acc-0.8660
Iter-45740, train loss-0.4755, acc-0.8400, valid loss-0.4799, acc-0.8688, test loss-0.4834, acc-0.8660
Iter-45750, train loss-0.4331, acc-0.8800, valid loss-0.4798, acc-0.8688, test loss-0.4833, acc-0.8660
Iter-45760, train loss-0.5733, acc-0.8400, valid loss-0.4798, acc-0.8690, test loss-0.4832, acc-0.8660
Iter-45770, train loss-0.4871, acc-0.8600, valid loss-0.4797, acc-0.8690, test loss-0.4832, acc-0.8659
Iter-45780, train loss-0.4037, acc-0.8400, valid loss-0.4797, acc-0.8688, test loss-0.4831, acc-0.8660
Iter-45790, train loss-0.3956, acc-0.8800, valid loss-0.4796, acc-0.8688, test loss-0.4831, acc-0.8660
Iter-45800, train loss-0.6777, acc-0.8000, valid loss-0.4796, acc-0.8690, test loss-0.4830, acc-0.8659
Iter-45810, train loss-0.4054, acc-0.8800, valid loss-0.4795, acc-0.8690, test loss-0.4830, acc-0.8661
Iter-45820, train loss-0.5699, acc-0.8200, valid loss-0.4795, acc-0.8690, test loss-0.4829, acc-0.8659
Iter-45830, train loss-0.6442, acc-0.8200, valid loss-0.4794, acc-0.8690, test loss-0.4829, acc-0.8660
Iter-45840, train loss-0.5369, acc-0.8800, valid loss-0.4794, acc-0.8690, test loss-0.4828, acc-0.8660
Iter-45850, train loss-0.5990, acc-0.7600, valid loss-0.4793, acc-0.8690, test loss-0.4828, acc-0.8660
Iter-45860, train loss-0.7192, acc-0.8200, valid loss-0.4793, acc-0.8690, test loss-0.4827, acc-0.8660
Iter-45870, train loss-0.3361, acc-0.9400, valid loss-0.4792, acc-0.8690, test loss-0.4827, acc-0.8661
Iter-45880, train loss-0.5256, acc-0.8800, valid loss-0.4791, acc-0.8688, test loss-0.4826, acc-0.8661
Iter-45890, train loss-0.4615, acc-0.8600, valid loss-0.4791, acc-0.8688, test loss-0.4825, acc-0.8662
Iter-45900, train loss-0.5964, acc-0.8400, valid loss-0.4790, acc-0.8686, test loss-0.4825, acc-0.8662
Iter-45910, train loss-0.5779, acc-0.8800, valid loss-0.4790, acc-0.8688, test loss-0.4824, acc-0.8662
Iter-45920, train loss-0.4584, acc-0.9200, valid loss-0.4789, acc-0.8688, test loss-0.4824, acc-0.8661
Iter-45930, train loss-0.7232, acc-0.8200, valid loss-0.4788, acc-0.8688, test loss-0.4823, acc-0.8662
Iter-45940, train loss-0.5227, acc-0.8400, valid loss-0.4788, acc-0.8688, test loss-0.4823, acc-0.8660
Iter-45950, train loss-0.6258, acc-0.8200, valid loss-0.4787, acc-0.8688, test loss-0.4822, acc-0.8661
Iter-45960, train loss-0.4532, acc-0.9000, valid loss-0.4787, acc-0.8690, test loss-0.4822, acc-0.8660
Iter-45970, train loss-0.4716, acc-0.9200, valid loss-0.4786, acc-0.8692, test loss-0.4821, acc-0.8661
Iter-45980, train loss-0.4523, acc-0.8400, valid loss-0.4786, acc-0.8692, test loss-0.4821, acc-0.8661
Iter-45990, train loss-0.5056, acc-0.8600, valid loss-0.4786, acc-0.8688, test loss-0.4820, acc-0.8661
Iter-46000, train loss-0.5200, acc-0.8400, valid loss-0.4785, acc-0.8686, test loss-0.4819, acc-0.8661
Iter-46010, train loss-0.4125, acc-0.9000, valid loss-0.4785, acc-0.8692, test loss-0.4819, acc-0.8661
Iter-46020, train loss-0.5892, acc-0.8400, valid loss-0.4784, acc-0.8692, test loss-0.4819, acc-0.8661
Iter-46030, train loss-0.7158, acc-0.7800, valid loss-0.4784, acc-0.8694, test loss-0.4818, acc-0.8662
Iter-46040, train loss-0.5319, acc-0.9000, valid loss-0.4783, acc-0.8694, test loss-0.4818, acc-0.8662
Iter-46050, train loss-0.6898, acc-0.8000, valid loss-0.4782, acc-0.8696, test loss-0.4817, acc-0.8662
Iter-46060, train loss-0.3238, acc-0.9600, valid loss-0.4782, acc-0.8692, test loss-0.4817, acc-0.8662
Iter-46070, train loss-0.2499, acc-0.9800, valid loss-0.4781, acc-0.8692, test loss-0.4816, acc-0.8662
Iter-46080, train loss-0.5812, acc-0.8400, valid loss-0.4781, acc-0.8692, test loss-0.4815, acc-0.8662
Iter-46090, train loss-0.4619, acc-0.9200, valid loss-0.4780, acc-0.8696, test loss-0.4815, acc-0.8662
Iter-46100, train loss-0.7982, acc-0.7400, valid loss-0.4780, acc-0.8694, test loss-0.4815, acc-0.8663
Iter-46110, train loss-0.3853, acc-0.9200, valid loss-0.4779, acc-0.8696, test loss-0.4814, acc-0.8662
Iter-46120, train loss-0.4846, acc-0.8600, valid loss-0.4779, acc-0.8694, test loss-0.4814, acc-0.8661
Iter-46130, train loss-0.3344, acc-0.9000, valid loss-0.4778, acc-0.8696, test loss-0.4813, acc-0.8660
Iter-46140, train loss-0.5268, acc-0.7800, valid loss-0.4778, acc-0.8694, test loss-0.4813, acc-0.8663
Iter-46150, train loss-0.4233, acc-0.9000, valid loss-0.4777, acc-0.8698, test loss-0.4812, acc-0.8661
Iter-46160, train loss-0.3045, acc-0.9400, valid loss-0.4777, acc-0.8700, test loss-0.4812, acc-0.8661
Iter-46170, train loss-0.5124, acc-0.8400, valid loss-0.4776, acc-0.8698, test loss-0.4811, acc-0.8662
Iter-46180, train loss-0.3580, acc-0.9400, valid loss-0.4776, acc-0.8694, test loss-0.4811, acc-0.8662
Iter-46190, train loss-0.5128, acc-0.8800, valid loss-0.4775, acc-0.8698, test loss-0.4810, acc-0.8662
Iter-46200, train loss-0.4886, acc-0.8000, valid loss-0.4775, acc-0.8694, test loss-0.4810, acc-0.8662
Iter-46210, train loss-0.5245, acc-0.8600, valid loss-0.4774, acc-0.8696, test loss-0.4809, acc-0.8663
Iter-46220, train loss-0.5358, acc-0.8200, valid loss-0.4774, acc-0.8696, test loss-0.4808, acc-0.8663
Iter-46230, train loss-0.5518, acc-0.8600, valid loss-0.4773, acc-0.8700, test loss-0.4808, acc-0.8662
Iter-46240, train loss-0.5653, acc-0.8000, valid loss-0.4773, acc-0.8700, test loss-0.4807, acc-0.8663
Iter-46250, train loss-0.5476, acc-0.8600, valid loss-0.4772, acc-0.8702, test loss-0.4807, acc-0.8663
Iter-46260, train loss-0.4244, acc-0.8800, valid loss-0.4772, acc-0.8702, test loss-0.4806, acc-0.8663
Iter-46270, train loss-0.5971, acc-0.8000, valid loss-0.4771, acc-0.8700, test loss-0.4806, acc-0.8664
Iter-46280, train loss-0.5822, acc-0.8400, valid loss-0.4771, acc-0.8698, test loss-0.4805, acc-0.8665
Iter-46290, train loss-0.4032, acc-0.8800, valid loss-0.4770, acc-0.8698, test loss-0.4805, acc-0.8663
Iter-46300, train loss-0.4817, acc-0.8600, valid loss-0.4770, acc-0.8698, test loss-0.4804, acc-0.8664
Iter-46310, train loss-0.6219, acc-0.8400, valid loss-0.4769, acc-0.8698, test loss-0.4804, acc-0.8666
Iter-46320, train loss-0.3637, acc-0.9000, valid loss-0.4769, acc-0.8696, test loss-0.4804, acc-0.8664
Iter-46330, train loss-0.6372, acc-0.8200, valid loss-0.4768, acc-0.8702, test loss-0.4803, acc-0.8663
Iter-46340, train loss-0.5748, acc-0.8400, valid loss-0.4768, acc-0.8700, test loss-0.4802, acc-0.8663
Iter-46350, train loss-0.5583, acc-0.8400, valid loss-0.4767, acc-0.8702, test loss-0.4802, acc-0.8663
Iter-46360, train loss-0.4377, acc-0.8600, valid loss-0.4767, acc-0.8700, test loss-0.4801, acc-0.8663
Iter-46370, train loss-0.3462, acc-0.9200, valid loss-0.4766, acc-0.8700, test loss-0.4801, acc-0.8664
Iter-46380, train loss-0.4624, acc-0.8600, valid loss-0.4766, acc-0.8698, test loss-0.4800, acc-0.8663
Iter-46390, train loss-0.4415, acc-0.9000, valid loss-0.4765, acc-0.8700, test loss-0.4799, acc-0.8665
Iter-46400, train loss-0.5531, acc-0.8600, valid loss-0.4765, acc-0.8700, test loss-0.4799, acc-0.8666
Iter-46410, train loss-0.5115, acc-0.8400, valid loss-0.4764, acc-0.8700, test loss-0.4798, acc-0.8666
Iter-46420, train loss-0.5941, acc-0.8800, valid loss-0.4763, acc-0.8702, test loss-0.4798, acc-0.8666
Iter-46430, train loss-0.4344, acc-0.8800, valid loss-0.4763, acc-0.8702, test loss-0.4797, acc-0.8665
Iter-46440, train loss-0.3963, acc-0.9000, valid loss-0.4762, acc-0.8702, test loss-0.4797, acc-0.8665
Iter-46450, train loss-0.5664, acc-0.7800, valid loss-0.4762, acc-0.8702, test loss-0.4796, acc-0.8665
Iter-46460, train loss-0.3990, acc-0.9000, valid loss-0.4761, acc-0.8702, test loss-0.4796, acc-0.8665
Iter-46470, train loss-0.3952, acc-0.9000, valid loss-0.4761, acc-0.8702, test loss-0.4795, acc-0.8667
Iter-46480, train loss-0.5842, acc-0.8600, valid loss-0.4760, acc-0.8704, test loss-0.4795, acc-0.8665
Iter-46490, train loss-0.5522, acc-0.8600, valid loss-0.4760, acc-0.8704, test loss-0.4794, acc-0.8666
Iter-46500, train loss-0.5093, acc-0.9200, valid loss-0.4759, acc-0.8704, test loss-0.4794, acc-0.8666
Iter-46510, train loss-0.4555, acc-0.8400, valid loss-0.4759, acc-0.8704, test loss-0.4793, acc-0.8666
Iter-46520, train loss-0.3492, acc-0.8800, valid loss-0.4759, acc-0.8704, test loss-0.4793, acc-0.8667
Iter-46530, train loss-0.3563, acc-0.9000, valid loss-0.4758, acc-0.8704, test loss-0.4793, acc-0.8665
Iter-46540, train loss-0.4903, acc-0.9000, valid loss-0.4757, acc-0.8706, test loss-0.4792, acc-0.8665
Iter-46550, train loss-0.4036, acc-0.8800, valid loss-0.4757, acc-0.8708, test loss-0.4791, acc-0.8664
Iter-46560, train loss-0.3907, acc-0.9000, valid loss-0.4756, acc-0.8706, test loss-0.4791, acc-0.8668
Iter-46570, train loss-0.4040, acc-0.9200, valid loss-0.4756, acc-0.8708, test loss-0.4790, acc-0.8667
Iter-46580, train loss-0.6479, acc-0.7600, valid loss-0.4755, acc-0.8704, test loss-0.4790, acc-0.8667
Iter-46590, train loss-0.5951, acc-0.8600, valid loss-0.4755, acc-0.8702, test loss-0.4789, acc-0.8670
Iter-46600, train loss-0.3874, acc-0.8600, valid loss-0.4754, acc-0.8704, test loss-0.4789, acc-0.8669
Iter-46610, train loss-0.4609, acc-0.9000, valid loss-0.4754, acc-0.8710, test loss-0.4788, acc-0.8671
Iter-46620, train loss-0.5699, acc-0.8600, valid loss-0.4754, acc-0.8708, test loss-0.4788, acc-0.8671
Iter-46630, train loss-0.4522, acc-0.8800, valid loss-0.4753, acc-0.8708, test loss-0.4787, acc-0.8670
Iter-46640, train loss-0.3491, acc-0.8800, valid loss-0.4753, acc-0.8708, test loss-0.4787, acc-0.8669
Iter-46650, train loss-0.7322, acc-0.7800, valid loss-0.4752, acc-0.8706, test loss-0.4786, acc-0.8669
Iter-46660, train loss-0.4356, acc-0.8800, valid loss-0.4752, acc-0.8706, test loss-0.4786, acc-0.8668
Iter-46670, train loss-0.6302, acc-0.7800, valid loss-0.4751, acc-0.8710, test loss-0.4785, acc-0.8666
Iter-46680, train loss-0.5494, acc-0.8000, valid loss-0.4751, acc-0.8708, test loss-0.4785, acc-0.8667
Iter-46690, train loss-0.4731, acc-0.8600, valid loss-0.4750, acc-0.8708, test loss-0.4784, acc-0.8666
Iter-46700, train loss-0.3889, acc-0.9000, valid loss-0.4749, acc-0.8708, test loss-0.4784, acc-0.8666
Iter-46710, train loss-0.4838, acc-0.8800, valid loss-0.4749, acc-0.8708, test loss-0.4783, acc-0.8667
Iter-46720, train loss-0.5015, acc-0.8600, valid loss-0.4748, acc-0.8708, test loss-0.4783, acc-0.8667
Iter-46730, train loss-0.4927, acc-0.8800, valid loss-0.4748, acc-0.8708, test loss-0.4782, acc-0.8667
Iter-46740, train loss-0.5269, acc-0.8400, valid loss-0.4748, acc-0.8708, test loss-0.4782, acc-0.8667
Iter-46750, train loss-0.6036, acc-0.8200, valid loss-0.4747, acc-0.8708, test loss-0.4781, acc-0.8667
Iter-46760, train loss-0.6142, acc-0.8400, valid loss-0.4746, acc-0.8708, test loss-0.4781, acc-0.8667
Iter-46770, train loss-0.5166, acc-0.8400, valid loss-0.4746, acc-0.8708, test loss-0.4780, acc-0.8667
Iter-46780, train loss-0.7736, acc-0.6800, valid loss-0.4745, acc-0.8708, test loss-0.4780, acc-0.8668
Iter-46790, train loss-0.4701, acc-0.8600, valid loss-0.4745, acc-0.8708, test loss-0.4779, acc-0.8667
Iter-46800, train loss-0.5345, acc-0.8200, valid loss-0.4744, acc-0.8708, test loss-0.4779, acc-0.8669
Iter-46810, train loss-0.6711, acc-0.8000, valid loss-0.4744, acc-0.8708, test loss-0.4778, acc-0.8670
Iter-46820, train loss-0.4421, acc-0.8800, valid loss-0.4743, acc-0.8708, test loss-0.4778, acc-0.8672
Iter-46830, train loss-0.3591, acc-0.8800, valid loss-0.4743, acc-0.8708, test loss-0.4777, acc-0.8671
Iter-46840, train loss-0.3372, acc-0.8600, valid loss-0.4742, acc-0.8708, test loss-0.4777, acc-0.8671
Iter-46850, train loss-0.5715, acc-0.8800, valid loss-0.4742, acc-0.8710, test loss-0.4776, acc-0.8671
Iter-46860, train loss-0.7076, acc-0.7400, valid loss-0.4742, acc-0.8710, test loss-0.4776, acc-0.8671
Iter-46870, train loss-0.6995, acc-0.7800, valid loss-0.4741, acc-0.8710, test loss-0.4775, acc-0.8672
Iter-46880, train loss-0.4009, acc-0.8800, valid loss-0.4740, acc-0.8710, test loss-0.4775, acc-0.8672
Iter-46890, train loss-0.4280, acc-0.9000, valid loss-0.4740, acc-0.8708, test loss-0.4774, acc-0.8671
Iter-46900, train loss-0.5346, acc-0.8400, valid loss-0.4739, acc-0.8708, test loss-0.4774, acc-0.8673
Iter-46910, train loss-0.4740, acc-0.8400, valid loss-0.4739, acc-0.8708, test loss-0.4774, acc-0.8674
Iter-46920, train loss-0.5328, acc-0.8600, valid loss-0.4739, acc-0.8710, test loss-0.4773, acc-0.8673
Iter-46930, train loss-0.5224, acc-0.8600, valid loss-0.4738, acc-0.8710, test loss-0.4773, acc-0.8673
Iter-46940, train loss-0.3883, acc-0.9200, valid loss-0.4738, acc-0.8712, test loss-0.4772, acc-0.8675
Iter-46950, train loss-0.5400, acc-0.8000, valid loss-0.4737, acc-0.8710, test loss-0.4772, acc-0.8674
Iter-46960, train loss-0.5806, acc-0.9200, valid loss-0.4737, acc-0.8710, test loss-0.4771, acc-0.8673
Iter-46970, train loss-0.4252, acc-0.8400, valid loss-0.4736, acc-0.8712, test loss-0.4771, acc-0.8673
Iter-46980, train loss-0.3976, acc-0.8800, valid loss-0.4736, acc-0.8710, test loss-0.4770, acc-0.8673
Iter-46990, train loss-0.3834, acc-0.9000, valid loss-0.4735, acc-0.8710, test loss-0.4770, acc-0.8674
Iter-47000, train loss-0.4970, acc-0.8400, valid loss-0.4735, acc-0.8708, test loss-0.4769, acc-0.8675
Iter-47010, train loss-0.5501, acc-0.8800, valid loss-0.4734, acc-0.8708, test loss-0.4769, acc-0.8675
Iter-47020, train loss-0.5071, acc-0.8600, valid loss-0.4734, acc-0.8708, test loss-0.4768, acc-0.8675
Iter-47030, train loss-0.4313, acc-0.8600, valid loss-0.4733, acc-0.8708, test loss-0.4768, acc-0.8675
Iter-47040, train loss-0.3960, acc-0.8600, valid loss-0.4733, acc-0.8708, test loss-0.4767, acc-0.8676
Iter-47050, train loss-0.5066, acc-0.8800, valid loss-0.4732, acc-0.8708, test loss-0.4767, acc-0.8676
Iter-47060, train loss-0.5601, acc-0.8800, valid loss-0.4731, acc-0.8710, test loss-0.4767, acc-0.8676
Iter-47070, train loss-0.3402, acc-0.9400, valid loss-0.4731, acc-0.8710, test loss-0.4766, acc-0.8677
Iter-47080, train loss-0.4393, acc-0.8800, valid loss-0.4730, acc-0.8710, test loss-0.4766, acc-0.8677
Iter-47090, train loss-0.3308, acc-0.9000, valid loss-0.4730, acc-0.8710, test loss-0.4765, acc-0.8677
Iter-47100, train loss-0.4114, acc-0.9000, valid loss-0.4729, acc-0.8710, test loss-0.4765, acc-0.8677
Iter-47110, train loss-0.4847, acc-0.8400, valid loss-0.4729, acc-0.8710, test loss-0.4764, acc-0.8677
Iter-47120, train loss-0.4298, acc-0.8800, valid loss-0.4729, acc-0.8712, test loss-0.4764, acc-0.8676
Iter-47130, train loss-0.5583, acc-0.8800, valid loss-0.4728, acc-0.8710, test loss-0.4763, acc-0.8675
Iter-47140, train loss-0.6420, acc-0.8000, valid loss-0.4728, acc-0.8710, test loss-0.4763, acc-0.8675
Iter-47150, train loss-0.4460, acc-0.8800, valid loss-0.4727, acc-0.8710, test loss-0.4762, acc-0.8674
Iter-47160, train loss-0.3942, acc-0.9400, valid loss-0.4727, acc-0.8710, test loss-0.4762, acc-0.8675
Iter-47170, train loss-0.6205, acc-0.8200, valid loss-0.4726, acc-0.8706, test loss-0.4761, acc-0.8677
Iter-47180, train loss-0.5168, acc-0.8600, valid loss-0.4726, acc-0.8712, test loss-0.4761, acc-0.8676
Iter-47190, train loss-0.6616, acc-0.8600, valid loss-0.4725, acc-0.8712, test loss-0.4760, acc-0.8676
Iter-47200, train loss-0.3773, acc-0.9200, valid loss-0.4725, acc-0.8710, test loss-0.4760, acc-0.8676
Iter-47210, train loss-0.3373, acc-0.9000, valid loss-0.4724, acc-0.8708, test loss-0.4759, acc-0.8677
Iter-47220, train loss-0.5394, acc-0.8600, valid loss-0.4724, acc-0.8710, test loss-0.4759, acc-0.8678
Iter-47230, train loss-0.5731, acc-0.8000, valid loss-0.4723, acc-0.8708, test loss-0.4759, acc-0.8677
Iter-47240, train loss-0.3093, acc-0.9400, valid loss-0.4723, acc-0.8708, test loss-0.4758, acc-0.8677
Iter-47250, train loss-0.4857, acc-0.8600, valid loss-0.4722, acc-0.8706, test loss-0.4758, acc-0.8676
Iter-47260, train loss-0.5509, acc-0.8600, valid loss-0.4722, acc-0.8708, test loss-0.4757, acc-0.8676
Iter-47270, train loss-0.4108, acc-0.8800, valid loss-0.4722, acc-0.8708, test loss-0.4757, acc-0.8676
Iter-47280, train loss-0.6742, acc-0.8200, valid loss-0.4721, acc-0.8712, test loss-0.4757, acc-0.8676
Iter-47290, train loss-0.5255, acc-0.8200, valid loss-0.4721, acc-0.8712, test loss-0.4756, acc-0.8676
Iter-47300, train loss-0.5452, acc-0.7200, valid loss-0.4720, acc-0.8710, test loss-0.4755, acc-0.8676
Iter-47310, train loss-0.3363, acc-0.9000, valid loss-0.4720, acc-0.8710, test loss-0.4755, acc-0.8676
Iter-47320, train loss-0.6384, acc-0.8400, valid loss-0.4719, acc-0.8710, test loss-0.4755, acc-0.8677
Iter-47330, train loss-0.5579, acc-0.8400, valid loss-0.4719, acc-0.8710, test loss-0.4754, acc-0.8678
Iter-47340, train loss-0.4180, acc-0.8600, valid loss-0.4718, acc-0.8710, test loss-0.4754, acc-0.8677
Iter-47350, train loss-0.4993, acc-0.9000, valid loss-0.4718, acc-0.8710, test loss-0.4753, acc-0.8676
Iter-47360, train loss-0.4122, acc-0.8800, valid loss-0.4718, acc-0.8710, test loss-0.4753, acc-0.8678
Iter-47370, train loss-0.4132, acc-0.8800, valid loss-0.4717, acc-0.8708, test loss-0.4753, acc-0.8677
Iter-47380, train loss-0.4099, acc-0.8800, valid loss-0.4717, acc-0.8710, test loss-0.4752, acc-0.8677
Iter-47390, train loss-0.5482, acc-0.8600, valid loss-0.4716, acc-0.8708, test loss-0.4752, acc-0.8679
Iter-47400, train loss-0.3379, acc-0.9000, valid loss-0.4716, acc-0.8708, test loss-0.4751, acc-0.8678
Iter-47410, train loss-0.4719, acc-0.9000, valid loss-0.4715, acc-0.8708, test loss-0.4751, acc-0.8678
Iter-47420, train loss-0.3981, acc-0.8800, valid loss-0.4715, acc-0.8708, test loss-0.4750, acc-0.8678
Iter-47430, train loss-0.4045, acc-0.9000, valid loss-0.4714, acc-0.8708, test loss-0.4750, acc-0.8678
Iter-47440, train loss-0.4647, acc-0.8800, valid loss-0.4714, acc-0.8708, test loss-0.4749, acc-0.8678
Iter-47450, train loss-0.3029, acc-0.9400, valid loss-0.4713, acc-0.8708, test loss-0.4749, acc-0.8679
Iter-47460, train loss-0.3578, acc-0.9000, valid loss-0.4713, acc-0.8708, test loss-0.4748, acc-0.8679
Iter-47470, train loss-0.4613, acc-0.9000, valid loss-0.4713, acc-0.8708, test loss-0.4748, acc-0.8679
Iter-47480, train loss-0.3744, acc-0.9000, valid loss-0.4712, acc-0.8708, test loss-0.4747, acc-0.8679
Iter-47490, train loss-0.5813, acc-0.8200, valid loss-0.4711, acc-0.8708, test loss-0.4747, acc-0.8679
Iter-47500, train loss-0.4660, acc-0.8600, valid loss-0.4711, acc-0.8708, test loss-0.4746, acc-0.8679
Iter-47510, train loss-0.4808, acc-0.8800, valid loss-0.4711, acc-0.8708, test loss-0.4746, acc-0.8678
Iter-47520, train loss-0.2936, acc-0.9800, valid loss-0.4710, acc-0.8708, test loss-0.4746, acc-0.8678
Iter-47530, train loss-0.4599, acc-0.9000, valid loss-0.4710, acc-0.8704, test loss-0.4745, acc-0.8678
Iter-47540, train loss-0.3543, acc-0.9000, valid loss-0.4709, acc-0.8704, test loss-0.4745, acc-0.8678
Iter-47550, train loss-0.5506, acc-0.8000, valid loss-0.4709, acc-0.8706, test loss-0.4744, acc-0.8678
Iter-47560, train loss-0.3475, acc-0.9200, valid loss-0.4708, acc-0.8706, test loss-0.4744, acc-0.8678
Iter-47570, train loss-0.4470, acc-0.8600, valid loss-0.4708, acc-0.8708, test loss-0.4743, acc-0.8678
Iter-47580, train loss-0.5314, acc-0.8400, valid loss-0.4707, acc-0.8706, test loss-0.4743, acc-0.8678
Iter-47590, train loss-0.3195, acc-0.8800, valid loss-0.4707, acc-0.8706, test loss-0.4742, acc-0.8679
Iter-47600, train loss-0.3948, acc-0.9000, valid loss-0.4706, acc-0.8706, test loss-0.4741, acc-0.8679
Iter-47610, train loss-0.5508, acc-0.7800, valid loss-0.4706, acc-0.8706, test loss-0.4741, acc-0.8679
Iter-47620, train loss-0.5515, acc-0.8000, valid loss-0.4706, acc-0.8708, test loss-0.4741, acc-0.8679
Iter-47630, train loss-0.3845, acc-0.9000, valid loss-0.4705, acc-0.8708, test loss-0.4740, acc-0.8678
Iter-47640, train loss-0.5536, acc-0.8400, valid loss-0.4705, acc-0.8706, test loss-0.4740, acc-0.8678
Iter-47650, train loss-0.3215, acc-0.9400, valid loss-0.4705, acc-0.8706, test loss-0.4739, acc-0.8679
Iter-47660, train loss-0.4355, acc-0.9000, valid loss-0.4704, acc-0.8708, test loss-0.4739, acc-0.8679
Iter-47670, train loss-0.4279, acc-0.9000, valid loss-0.4704, acc-0.8706, test loss-0.4738, acc-0.8679
Iter-47680, train loss-0.4716, acc-0.8600, valid loss-0.4703, acc-0.8708, test loss-0.4738, acc-0.8679
Iter-47690, train loss-0.3713, acc-0.9400, valid loss-0.4703, acc-0.8708, test loss-0.4737, acc-0.8680
Iter-47700, train loss-0.6737, acc-0.7400, valid loss-0.4702, acc-0.8710, test loss-0.4737, acc-0.8680
Iter-47710, train loss-0.5227, acc-0.8400, valid loss-0.4702, acc-0.8710, test loss-0.4737, acc-0.8681
Iter-47720, train loss-0.6474, acc-0.8200, valid loss-0.4701, acc-0.8710, test loss-0.4736, acc-0.8681
Iter-47730, train loss-0.3466, acc-0.9000, valid loss-0.4701, acc-0.8708, test loss-0.4736, acc-0.8680
Iter-47740, train loss-0.6404, acc-0.7800, valid loss-0.4700, acc-0.8708, test loss-0.4735, acc-0.8681
Iter-47750, train loss-0.4502, acc-0.8400, valid loss-0.4700, acc-0.8710, test loss-0.4735, acc-0.8682
Iter-47760, train loss-0.4483, acc-0.8800, valid loss-0.4699, acc-0.8708, test loss-0.4734, acc-0.8681
Iter-47770, train loss-0.3930, acc-0.9400, valid loss-0.4699, acc-0.8710, test loss-0.4733, acc-0.8681
Iter-47780, train loss-0.5080, acc-0.8600, valid loss-0.4698, acc-0.8708, test loss-0.4733, acc-0.8683
Iter-47790, train loss-0.5471, acc-0.8200, valid loss-0.4697, acc-0.8710, test loss-0.4732, acc-0.8683
Iter-47800, train loss-0.6786, acc-0.8200, valid loss-0.4697, acc-0.8708, test loss-0.4732, acc-0.8683
Iter-47810, train loss-0.5281, acc-0.8400, valid loss-0.4696, acc-0.8710, test loss-0.4731, acc-0.8684
Iter-47820, train loss-0.4216, acc-0.9000, valid loss-0.4696, acc-0.8708, test loss-0.4731, acc-0.8684
Iter-47830, train loss-0.4352, acc-0.9400, valid loss-0.4695, acc-0.8706, test loss-0.4730, acc-0.8684
Iter-47840, train loss-0.4526, acc-0.8600, valid loss-0.4695, acc-0.8706, test loss-0.4730, acc-0.8684
Iter-47850, train loss-0.4161, acc-0.9200, valid loss-0.4694, acc-0.8706, test loss-0.4729, acc-0.8683
Iter-47860, train loss-0.4860, acc-0.8400, valid loss-0.4694, acc-0.8708, test loss-0.4729, acc-0.8683
Iter-47870, train loss-0.4341, acc-0.8800, valid loss-0.4693, acc-0.8706, test loss-0.4728, acc-0.8684
Iter-47880, train loss-0.5135, acc-0.8600, valid loss-0.4693, acc-0.8706, test loss-0.4728, acc-0.8683
Iter-47890, train loss-0.5359, acc-0.8200, valid loss-0.4692, acc-0.8706, test loss-0.4727, acc-0.8683
Iter-47900, train loss-0.5604, acc-0.8600, valid loss-0.4692, acc-0.8708, test loss-0.4727, acc-0.8684
Iter-47910, train loss-0.3511, acc-0.9000, valid loss-0.4691, acc-0.8708, test loss-0.4727, acc-0.8683
Iter-47920, train loss-0.3587, acc-0.9400, valid loss-0.4691, acc-0.8708, test loss-0.4726, acc-0.8683
Iter-47930, train loss-0.4691, acc-0.8800, valid loss-0.4690, acc-0.8710, test loss-0.4726, acc-0.8683
Iter-47940, train loss-0.4845, acc-0.9000, valid loss-0.4690, acc-0.8710, test loss-0.4725, acc-0.8684
Iter-47950, train loss-0.5303, acc-0.8400, valid loss-0.4689, acc-0.8710, test loss-0.4725, acc-0.8685
Iter-47960, train loss-0.4979, acc-0.8600, valid loss-0.4689, acc-0.8710, test loss-0.4724, acc-0.8685
Iter-47970, train loss-0.6499, acc-0.8400, valid loss-0.4688, acc-0.8710, test loss-0.4724, acc-0.8685
Iter-47980, train loss-0.3845, acc-0.9000, valid loss-0.4688, acc-0.8712, test loss-0.4723, acc-0.8685
Iter-47990, train loss-0.7022, acc-0.8400, valid loss-0.4687, acc-0.8712, test loss-0.4723, acc-0.8685
Iter-48000, train loss-0.6063, acc-0.8200, valid loss-0.4687, acc-0.8712, test loss-0.4722, acc-0.8685
Iter-48010, train loss-0.3791, acc-0.9000, valid loss-0.4686, acc-0.8712, test loss-0.4722, acc-0.8685
Iter-48020, train loss-0.4726, acc-0.8600, valid loss-0.4686, acc-0.8712, test loss-0.4721, acc-0.8685
Iter-48030, train loss-0.4418, acc-0.9000, valid loss-0.4685, acc-0.8712, test loss-0.4721, acc-0.8685
Iter-48040, train loss-0.5251, acc-0.8800, valid loss-0.4685, acc-0.8714, test loss-0.4720, acc-0.8685
Iter-48050, train loss-0.5448, acc-0.8000, valid loss-0.4685, acc-0.8714, test loss-0.4720, acc-0.8686
Iter-48060, train loss-0.4430, acc-0.8600, valid loss-0.4684, acc-0.8716, test loss-0.4720, acc-0.8688
Iter-48070, train loss-0.1837, acc-0.9800, valid loss-0.4684, acc-0.8716, test loss-0.4719, acc-0.8689
Iter-48080, train loss-0.3661, acc-0.9200, valid loss-0.4683, acc-0.8716, test loss-0.4719, acc-0.8688
Iter-48090, train loss-0.4272, acc-0.8800, valid loss-0.4683, acc-0.8716, test loss-0.4718, acc-0.8687
Iter-48100, train loss-0.4327, acc-0.8400, valid loss-0.4682, acc-0.8714, test loss-0.4718, acc-0.8688
Iter-48110, train loss-0.3144, acc-0.9200, valid loss-0.4682, acc-0.8714, test loss-0.4717, acc-0.8688
Iter-48120, train loss-0.5343, acc-0.8600, valid loss-0.4681, acc-0.8714, test loss-0.4717, acc-0.8687
Iter-48130, train loss-0.3983, acc-0.9200, valid loss-0.4681, acc-0.8714, test loss-0.4717, acc-0.8688
Iter-48140, train loss-0.6066, acc-0.7800, valid loss-0.4681, acc-0.8714, test loss-0.4716, acc-0.8687
Iter-48150, train loss-0.5011, acc-0.8600, valid loss-0.4680, acc-0.8714, test loss-0.4716, acc-0.8687
Iter-48160, train loss-0.6729, acc-0.8000, valid loss-0.4680, acc-0.8716, test loss-0.4715, acc-0.8687
Iter-48170, train loss-0.3944, acc-0.9000, valid loss-0.4679, acc-0.8716, test loss-0.4715, acc-0.8687
Iter-48180, train loss-0.5356, acc-0.9000, valid loss-0.4679, acc-0.8718, test loss-0.4715, acc-0.8687
Iter-48190, train loss-0.4499, acc-0.9000, valid loss-0.4678, acc-0.8720, test loss-0.4714, acc-0.8687
Iter-48200, train loss-0.4282, acc-0.9000, valid loss-0.4678, acc-0.8720, test loss-0.4714, acc-0.8687
Iter-48210, train loss-0.4350, acc-0.8400, valid loss-0.4677, acc-0.8720, test loss-0.4713, acc-0.8687
Iter-48220, train loss-0.3541, acc-0.9400, valid loss-0.4677, acc-0.8718, test loss-0.4713, acc-0.8687
Iter-48230, train loss-0.3837, acc-0.9000, valid loss-0.4676, acc-0.8718, test loss-0.4712, acc-0.8687
Iter-48240, train loss-0.5641, acc-0.8400, valid loss-0.4676, acc-0.8720, test loss-0.4712, acc-0.8687
Iter-48250, train loss-0.5719, acc-0.8000, valid loss-0.4675, acc-0.8718, test loss-0.4711, acc-0.8687
Iter-48260, train loss-0.3725, acc-0.9400, valid loss-0.4675, acc-0.8718, test loss-0.4711, acc-0.8688
Iter-48270, train loss-0.4329, acc-0.9000, valid loss-0.4675, acc-0.8718, test loss-0.4710, acc-0.8686
Iter-48280, train loss-0.7409, acc-0.7600, valid loss-0.4674, acc-0.8718, test loss-0.4710, acc-0.8686
Iter-48290, train loss-0.4502, acc-0.9000, valid loss-0.4674, acc-0.8718, test loss-0.4709, acc-0.8686
Iter-48300, train loss-0.3479, acc-0.9200, valid loss-0.4673, acc-0.8718, test loss-0.4709, acc-0.8686
Iter-48310, train loss-0.5796, acc-0.8000, valid loss-0.4672, acc-0.8716, test loss-0.4708, acc-0.8687
Iter-48320, train loss-0.4136, acc-0.9000, valid loss-0.4672, acc-0.8718, test loss-0.4707, acc-0.8688
Iter-48330, train loss-0.3546, acc-0.9400, valid loss-0.4671, acc-0.8718, test loss-0.4707, acc-0.8688
Iter-48340, train loss-0.6432, acc-0.7800, valid loss-0.4671, acc-0.8718, test loss-0.4706, acc-0.8687
Iter-48350, train loss-0.4154, acc-0.8600, valid loss-0.4670, acc-0.8716, test loss-0.4706, acc-0.8687
Iter-48360, train loss-0.5186, acc-0.8400, valid loss-0.4670, acc-0.8718, test loss-0.4705, acc-0.8687
Iter-48370, train loss-0.5160, acc-0.8000, valid loss-0.4669, acc-0.8716, test loss-0.4705, acc-0.8688
Iter-48380, train loss-0.4813, acc-0.9000, valid loss-0.4669, acc-0.8718, test loss-0.4704, acc-0.8687
Iter-48390, train loss-0.4779, acc-0.8800, valid loss-0.4668, acc-0.8718, test loss-0.4704, acc-0.8687
Iter-48400, train loss-0.5167, acc-0.8600, valid loss-0.4668, acc-0.8716, test loss-0.4703, acc-0.8687
Iter-48410, train loss-0.4499, acc-0.8000, valid loss-0.4667, acc-0.8716, test loss-0.4703, acc-0.8687
Iter-48420, train loss-0.4737, acc-0.8600, valid loss-0.4667, acc-0.8716, test loss-0.4702, acc-0.8688
Iter-48430, train loss-0.4331, acc-0.8400, valid loss-0.4666, acc-0.8716, test loss-0.4702, acc-0.8689
Iter-48440, train loss-0.6304, acc-0.8800, valid loss-0.4666, acc-0.8716, test loss-0.4702, acc-0.8689
Iter-48450, train loss-0.4611, acc-0.9000, valid loss-0.4665, acc-0.8718, test loss-0.4701, acc-0.8688
Iter-48460, train loss-0.4365, acc-0.9000, valid loss-0.4665, acc-0.8718, test loss-0.4701, acc-0.8689
Iter-48470, train loss-0.3596, acc-0.8800, valid loss-0.4664, acc-0.8718, test loss-0.4700, acc-0.8689
Iter-48480, train loss-0.4706, acc-0.8000, valid loss-0.4664, acc-0.8720, test loss-0.4700, acc-0.8689
Iter-48490, train loss-0.5108, acc-0.8400, valid loss-0.4663, acc-0.8718, test loss-0.4699, acc-0.8690
Iter-48500, train loss-0.5645, acc-0.7800, valid loss-0.4663, acc-0.8718, test loss-0.4699, acc-0.8690
Iter-48510, train loss-0.6583, acc-0.8400, valid loss-0.4662, acc-0.8720, test loss-0.4698, acc-0.8690
Iter-48520, train loss-0.3954, acc-0.8200, valid loss-0.4662, acc-0.8720, test loss-0.4698, acc-0.8691
Iter-48530, train loss-0.4652, acc-0.8600, valid loss-0.4661, acc-0.8718, test loss-0.4697, acc-0.8690
Iter-48540, train loss-0.5196, acc-0.8200, valid loss-0.4661, acc-0.8722, test loss-0.4697, acc-0.8690
Iter-48550, train loss-0.3225, acc-0.9200, valid loss-0.4660, acc-0.8720, test loss-0.4697, acc-0.8691
Iter-48560, train loss-0.4971, acc-0.8400, valid loss-0.4660, acc-0.8724, test loss-0.4696, acc-0.8690
Iter-48570, train loss-0.3138, acc-0.9000, valid loss-0.4659, acc-0.8720, test loss-0.4696, acc-0.8691
Iter-48580, train loss-0.5355, acc-0.8400, valid loss-0.4659, acc-0.8724, test loss-0.4695, acc-0.8691
Iter-48590, train loss-0.3689, acc-0.9000, valid loss-0.4658, acc-0.8724, test loss-0.4695, acc-0.8691
Iter-48600, train loss-0.5615, acc-0.8000, valid loss-0.4658, acc-0.8730, test loss-0.4694, acc-0.8691
Iter-48610, train loss-0.4193, acc-0.9000, valid loss-0.4657, acc-0.8728, test loss-0.4694, acc-0.8691
Iter-48620, train loss-0.6546, acc-0.8000, valid loss-0.4657, acc-0.8728, test loss-0.4693, acc-0.8691
Iter-48630, train loss-0.3423, acc-0.9200, valid loss-0.4656, acc-0.8728, test loss-0.4693, acc-0.8690
Iter-48640, train loss-0.4461, acc-0.8600, valid loss-0.4656, acc-0.8730, test loss-0.4692, acc-0.8690
Iter-48650, train loss-0.5315, acc-0.8600, valid loss-0.4655, acc-0.8728, test loss-0.4692, acc-0.8690
Iter-48660, train loss-0.5148, acc-0.8400, valid loss-0.4655, acc-0.8728, test loss-0.4691, acc-0.8690
Iter-48670, train loss-0.6074, acc-0.8200, valid loss-0.4655, acc-0.8728, test loss-0.4691, acc-0.8690
Iter-48680, train loss-0.3301, acc-0.9200, valid loss-0.4654, acc-0.8724, test loss-0.4690, acc-0.8691
Iter-48690, train loss-0.3864, acc-0.8400, valid loss-0.4654, acc-0.8724, test loss-0.4690, acc-0.8691
Iter-48700, train loss-0.3691, acc-0.9200, valid loss-0.4653, acc-0.8724, test loss-0.4689, acc-0.8690
Iter-48710, train loss-0.4628, acc-0.8600, valid loss-0.4653, acc-0.8724, test loss-0.4689, acc-0.8693
Iter-48720, train loss-0.5211, acc-0.8600, valid loss-0.4653, acc-0.8724, test loss-0.4689, acc-0.8692
Iter-48730, train loss-0.4972, acc-0.8400, valid loss-0.4652, acc-0.8724, test loss-0.4688, acc-0.8693
Iter-48740, train loss-0.2845, acc-0.9600, valid loss-0.4651, acc-0.8724, test loss-0.4688, acc-0.8693
Iter-48750, train loss-0.4669, acc-0.8600, valid loss-0.4651, acc-0.8724, test loss-0.4687, acc-0.8693
Iter-48760, train loss-0.4985, acc-0.9000, valid loss-0.4650, acc-0.8726, test loss-0.4686, acc-0.8691
Iter-48770, train loss-0.5413, acc-0.8000, valid loss-0.4650, acc-0.8724, test loss-0.4686, acc-0.8692
Iter-48780, train loss-0.5901, acc-0.8000, valid loss-0.4650, acc-0.8724, test loss-0.4686, acc-0.8692
Iter-48790, train loss-0.3476, acc-0.9200, valid loss-0.4649, acc-0.8724, test loss-0.4685, acc-0.8692
Iter-48800, train loss-0.5331, acc-0.8400, valid loss-0.4649, acc-0.8724, test loss-0.4685, acc-0.8692
Iter-48810, train loss-0.7953, acc-0.8200, valid loss-0.4648, acc-0.8724, test loss-0.4684, acc-0.8692
Iter-48820, train loss-0.5184, acc-0.8600, valid loss-0.4648, acc-0.8724, test loss-0.4684, acc-0.8692
Iter-48830, train loss-0.6521, acc-0.7800, valid loss-0.4647, acc-0.8724, test loss-0.4683, acc-0.8692
Iter-48840, train loss-0.4553, acc-0.8800, valid loss-0.4647, acc-0.8724, test loss-0.4683, acc-0.8692
Iter-48850, train loss-0.4671, acc-0.8400, valid loss-0.4646, acc-0.8726, test loss-0.4682, acc-0.8692
Iter-48860, train loss-0.4892, acc-0.8000, valid loss-0.4646, acc-0.8724, test loss-0.4682, acc-0.8692
Iter-48870, train loss-0.4711, acc-0.8400, valid loss-0.4645, acc-0.8726, test loss-0.4682, acc-0.8693
Iter-48880, train loss-0.3851, acc-0.9400, valid loss-0.4645, acc-0.8726, test loss-0.4681, acc-0.8693
Iter-48890, train loss-0.3701, acc-0.9000, valid loss-0.4644, acc-0.8726, test loss-0.4681, acc-0.8692
Iter-48900, train loss-0.2443, acc-0.9600, valid loss-0.4644, acc-0.8726, test loss-0.4680, acc-0.8692
Iter-48910, train loss-0.3254, acc-0.8800, valid loss-0.4644, acc-0.8726, test loss-0.4680, acc-0.8692
Iter-48920, train loss-0.7931, acc-0.8200, valid loss-0.4643, acc-0.8726, test loss-0.4680, acc-0.8693
Iter-48930, train loss-0.4899, acc-0.8800, valid loss-0.4643, acc-0.8726, test loss-0.4679, acc-0.8693
Iter-48940, train loss-0.3418, acc-0.9200, valid loss-0.4642, acc-0.8726, test loss-0.4679, acc-0.8694
Iter-48950, train loss-0.6042, acc-0.8200, valid loss-0.4642, acc-0.8726, test loss-0.4678, acc-0.8694
Iter-48960, train loss-0.4784, acc-0.8800, valid loss-0.4641, acc-0.8726, test loss-0.4678, acc-0.8694
Iter-48970, train loss-0.4154, acc-0.8600, valid loss-0.4641, acc-0.8726, test loss-0.4677, acc-0.8694
Iter-48980, train loss-0.3022, acc-0.9400, valid loss-0.4641, acc-0.8728, test loss-0.4677, acc-0.8694
Iter-48990, train loss-0.3955, acc-0.8800, valid loss-0.4640, acc-0.8728, test loss-0.4677, acc-0.8694
Iter-49000, train loss-0.4153, acc-0.9200, valid loss-0.4640, acc-0.8728, test loss-0.4676, acc-0.8695
Iter-49010, train loss-0.3186, acc-0.9200, valid loss-0.4639, acc-0.8728, test loss-0.4676, acc-0.8695
Iter-49020, train loss-0.6232, acc-0.8400, valid loss-0.4639, acc-0.8728, test loss-0.4675, acc-0.8694
Iter-49030, train loss-0.5828, acc-0.8400, valid loss-0.4638, acc-0.8728, test loss-0.4675, acc-0.8694
Iter-49040, train loss-0.4219, acc-0.9200, valid loss-0.4638, acc-0.8728, test loss-0.4674, acc-0.8695
Iter-49050, train loss-0.5631, acc-0.8400, valid loss-0.4637, acc-0.8728, test loss-0.4674, acc-0.8694
Iter-49060, train loss-0.5320, acc-0.8200, valid loss-0.4637, acc-0.8728, test loss-0.4673, acc-0.8693
Iter-49070, train loss-0.3156, acc-0.8800, valid loss-0.4636, acc-0.8728, test loss-0.4673, acc-0.8693
Iter-49080, train loss-0.5499, acc-0.8600, valid loss-0.4636, acc-0.8728, test loss-0.4672, acc-0.8693
Iter-49090, train loss-0.3775, acc-0.9200, valid loss-0.4635, acc-0.8728, test loss-0.4672, acc-0.8693
Iter-49100, train loss-0.5088, acc-0.8400, valid loss-0.4635, acc-0.8728, test loss-0.4671, acc-0.8693
Iter-49110, train loss-0.4965, acc-0.8200, valid loss-0.4634, acc-0.8728, test loss-0.4671, acc-0.8695
Iter-49120, train loss-0.3916, acc-0.9000, valid loss-0.4634, acc-0.8728, test loss-0.4670, acc-0.8694
Iter-49130, train loss-0.6741, acc-0.8000, valid loss-0.4634, acc-0.8728, test loss-0.4670, acc-0.8694
Iter-49140, train loss-0.4713, acc-0.8200, valid loss-0.4633, acc-0.8728, test loss-0.4670, acc-0.8696
Iter-49150, train loss-0.4851, acc-0.8800, valid loss-0.4632, acc-0.8728, test loss-0.4669, acc-0.8696
Iter-49160, train loss-0.6329, acc-0.8000, valid loss-0.4632, acc-0.8730, test loss-0.4669, acc-0.8696
Iter-49170, train loss-0.6098, acc-0.8400, valid loss-0.4632, acc-0.8730, test loss-0.4668, acc-0.8696
Iter-49180, train loss-0.3981, acc-0.9000, valid loss-0.4631, acc-0.8730, test loss-0.4668, acc-0.8696
Iter-49190, train loss-0.4403, acc-0.8800, valid loss-0.4631, acc-0.8730, test loss-0.4667, acc-0.8696
Iter-49200, train loss-0.5045, acc-0.8400, valid loss-0.4630, acc-0.8732, test loss-0.4667, acc-0.8696
Iter-49210, train loss-0.5059, acc-0.8400, valid loss-0.4630, acc-0.8730, test loss-0.4666, acc-0.8696
Iter-49220, train loss-0.6369, acc-0.8400, valid loss-0.4629, acc-0.8730, test loss-0.4666, acc-0.8696
Iter-49230, train loss-0.4576, acc-0.8800, valid loss-0.4629, acc-0.8730, test loss-0.4665, acc-0.8696
Iter-49240, train loss-0.5745, acc-0.8000, valid loss-0.4628, acc-0.8730, test loss-0.4665, acc-0.8697
Iter-49250, train loss-0.8552, acc-0.7800, valid loss-0.4628, acc-0.8732, test loss-0.4665, acc-0.8695
Iter-49260, train loss-0.5526, acc-0.8200, valid loss-0.4628, acc-0.8734, test loss-0.4664, acc-0.8697
Iter-49270, train loss-0.5089, acc-0.7800, valid loss-0.4627, acc-0.8734, test loss-0.4664, acc-0.8694
Iter-49280, train loss-0.7968, acc-0.8000, valid loss-0.4627, acc-0.8734, test loss-0.4663, acc-0.8695
Iter-49290, train loss-0.5455, acc-0.8600, valid loss-0.4626, acc-0.8734, test loss-0.4663, acc-0.8695
Iter-49300, train loss-0.6857, acc-0.7400, valid loss-0.4626, acc-0.8732, test loss-0.4662, acc-0.8695
Iter-49310, train loss-0.6284, acc-0.8000, valid loss-0.4625, acc-0.8734, test loss-0.4662, acc-0.8696
Iter-49320, train loss-0.4120, acc-0.9200, valid loss-0.4625, acc-0.8734, test loss-0.4662, acc-0.8696
Iter-49330, train loss-0.3688, acc-0.8800, valid loss-0.4624, acc-0.8734, test loss-0.4661, acc-0.8696
Iter-49340, train loss-0.4523, acc-0.9200, valid loss-0.4624, acc-0.8734, test loss-0.4661, acc-0.8696
Iter-49350, train loss-0.4445, acc-0.8600, valid loss-0.4624, acc-0.8734, test loss-0.4660, acc-0.8696
Iter-49360, train loss-0.5175, acc-0.8600, valid loss-0.4623, acc-0.8734, test loss-0.4660, acc-0.8697
Iter-49370, train loss-0.5963, acc-0.7600, valid loss-0.4623, acc-0.8734, test loss-0.4659, acc-0.8697
Iter-49380, train loss-0.4680, acc-0.8000, valid loss-0.4622, acc-0.8732, test loss-0.4659, acc-0.8697
Iter-49390, train loss-0.6100, acc-0.7800, valid loss-0.4622, acc-0.8732, test loss-0.4658, acc-0.8696
Iter-49400, train loss-0.5225, acc-0.8200, valid loss-0.4621, acc-0.8732, test loss-0.4658, acc-0.8696
Iter-49410, train loss-0.3428, acc-0.9200, valid loss-0.4621, acc-0.8732, test loss-0.4657, acc-0.8698
Iter-49420, train loss-0.2768, acc-0.9600, valid loss-0.4620, acc-0.8730, test loss-0.4657, acc-0.8698
Iter-49430, train loss-0.6955, acc-0.8000, valid loss-0.4620, acc-0.8730, test loss-0.4656, acc-0.8698
Iter-49440, train loss-0.5374, acc-0.8000, valid loss-0.4619, acc-0.8732, test loss-0.4656, acc-0.8699
Iter-49450, train loss-0.6284, acc-0.8000, valid loss-0.4619, acc-0.8730, test loss-0.4655, acc-0.8699
Iter-49460, train loss-0.4018, acc-0.8600, valid loss-0.4618, acc-0.8734, test loss-0.4655, acc-0.8699
Iter-49470, train loss-0.4459, acc-0.8600, valid loss-0.4618, acc-0.8732, test loss-0.4655, acc-0.8700
Iter-49480, train loss-0.4384, acc-0.8600, valid loss-0.4617, acc-0.8734, test loss-0.4654, acc-0.8700
Iter-49490, train loss-0.3350, acc-0.9400, valid loss-0.4617, acc-0.8732, test loss-0.4654, acc-0.8700
Iter-49500, train loss-0.3880, acc-0.8600, valid loss-0.4617, acc-0.8732, test loss-0.4653, acc-0.8700
Iter-49510, train loss-0.6079, acc-0.8800, valid loss-0.4616, acc-0.8732, test loss-0.4653, acc-0.8701
Iter-49520, train loss-0.5108, acc-0.8800, valid loss-0.4616, acc-0.8732, test loss-0.4652, acc-0.8701
Iter-49530, train loss-0.3385, acc-0.9200, valid loss-0.4615, acc-0.8732, test loss-0.4652, acc-0.8701
Iter-49540, train loss-0.9137, acc-0.7400, valid loss-0.4615, acc-0.8732, test loss-0.4651, acc-0.8701
Iter-49550, train loss-0.4135, acc-0.8800, valid loss-0.4614, acc-0.8732, test loss-0.4651, acc-0.8702
Iter-49560, train loss-0.3303, acc-0.9400, valid loss-0.4614, acc-0.8732, test loss-0.4650, acc-0.8702
Iter-49570, train loss-0.5013, acc-0.8800, valid loss-0.4613, acc-0.8730, test loss-0.4650, acc-0.8702
Iter-49580, train loss-0.4333, acc-0.9200, valid loss-0.4613, acc-0.8730, test loss-0.4650, acc-0.8703
Iter-49590, train loss-0.3476, acc-0.9400, valid loss-0.4612, acc-0.8730, test loss-0.4649, acc-0.8703
Iter-49600, train loss-0.5622, acc-0.8800, valid loss-0.4612, acc-0.8732, test loss-0.4649, acc-0.8704
Iter-49610, train loss-0.5375, acc-0.8400, valid loss-0.4612, acc-0.8734, test loss-0.4648, acc-0.8703
Iter-49620, train loss-0.4974, acc-0.8000, valid loss-0.4611, acc-0.8734, test loss-0.4648, acc-0.8704
Iter-49630, train loss-0.6594, acc-0.7600, valid loss-0.4611, acc-0.8734, test loss-0.4647, acc-0.8703
Iter-49640, train loss-0.5485, acc-0.8400, valid loss-0.4610, acc-0.8734, test loss-0.4647, acc-0.8703
Iter-49650, train loss-0.4124, acc-0.8600, valid loss-0.4610, acc-0.8734, test loss-0.4647, acc-0.8703
Iter-49660, train loss-0.4752, acc-0.8400, valid loss-0.4609, acc-0.8732, test loss-0.4646, acc-0.8703
Iter-49670, train loss-0.4123, acc-0.9200, valid loss-0.4609, acc-0.8734, test loss-0.4646, acc-0.8703
Iter-49680, train loss-0.4426, acc-0.8400, valid loss-0.4608, acc-0.8734, test loss-0.4646, acc-0.8703
Iter-49690, train loss-0.5723, acc-0.8200, valid loss-0.4608, acc-0.8734, test loss-0.4645, acc-0.8703
Iter-49700, train loss-0.4237, acc-0.8600, valid loss-0.4607, acc-0.8732, test loss-0.4645, acc-0.8702
Iter-49710, train loss-0.5586, acc-0.8600, valid loss-0.4607, acc-0.8730, test loss-0.4644, acc-0.8702
Iter-49720, train loss-0.3876, acc-0.9200, valid loss-0.4607, acc-0.8730, test loss-0.4644, acc-0.8703
Iter-49730, train loss-0.5407, acc-0.8400, valid loss-0.4606, acc-0.8730, test loss-0.4643, acc-0.8703
Iter-49740, train loss-0.4482, acc-0.8800, valid loss-0.4606, acc-0.8730, test loss-0.4643, acc-0.8703
Iter-49750, train loss-0.5006, acc-0.9200, valid loss-0.4605, acc-0.8732, test loss-0.4642, acc-0.8703
Iter-49760, train loss-0.3926, acc-0.8600, valid loss-0.4605, acc-0.8732, test loss-0.4642, acc-0.8703
Iter-49770, train loss-0.4720, acc-0.8200, valid loss-0.4605, acc-0.8732, test loss-0.4641, acc-0.8703
Iter-49780, train loss-0.3743, acc-0.9200, valid loss-0.4604, acc-0.8732, test loss-0.4641, acc-0.8703
Iter-49790, train loss-0.5720, acc-0.8200, valid loss-0.4604, acc-0.8732, test loss-0.4641, acc-0.8703
Iter-49800, train loss-0.4752, acc-0.9000, valid loss-0.4603, acc-0.8730, test loss-0.4640, acc-0.8703
Iter-49810, train loss-0.6421, acc-0.8800, valid loss-0.4603, acc-0.8732, test loss-0.4640, acc-0.8703
Iter-49820, train loss-0.4216, acc-0.8600, valid loss-0.4602, acc-0.8730, test loss-0.4639, acc-0.8703
Iter-49830, train loss-0.4447, acc-0.8800, valid loss-0.4602, acc-0.8732, test loss-0.4639, acc-0.8703
Iter-49840, train loss-0.5423, acc-0.8600, valid loss-0.4601, acc-0.8734, test loss-0.4638, acc-0.8703
Iter-49850, train loss-0.6247, acc-0.8600, valid loss-0.4601, acc-0.8734, test loss-0.4638, acc-0.8704
Iter-49860, train loss-0.4937, acc-0.8200, valid loss-0.4601, acc-0.8734, test loss-0.4637, acc-0.8704
Iter-49870, train loss-0.3350, acc-0.9400, valid loss-0.4600, acc-0.8738, test loss-0.4637, acc-0.8704
Iter-49880, train loss-0.7124, acc-0.8000, valid loss-0.4600, acc-0.8740, test loss-0.4636, acc-0.8704
Iter-49890, train loss-0.5878, acc-0.8600, valid loss-0.4599, acc-0.8740, test loss-0.4636, acc-0.8705
Iter-49900, train loss-0.3415, acc-0.9600, valid loss-0.4599, acc-0.8740, test loss-0.4635, acc-0.8705
Iter-49910, train loss-0.6282, acc-0.8400, valid loss-0.4598, acc-0.8742, test loss-0.4635, acc-0.8705
Iter-49920, train loss-0.4079, acc-0.8600, valid loss-0.4598, acc-0.8740, test loss-0.4635, acc-0.8705
Iter-49930, train loss-0.4162, acc-0.8600, valid loss-0.4598, acc-0.8742, test loss-0.4634, acc-0.8706
Iter-49940, train loss-0.4871, acc-0.8600, valid loss-0.4597, acc-0.8740, test loss-0.4634, acc-0.8706
Iter-49950, train loss-0.5086, acc-0.8800, valid loss-0.4597, acc-0.8742, test loss-0.4634, acc-0.8705
Iter-49960, train loss-0.3626, acc-0.9600, valid loss-0.4597, acc-0.8740, test loss-0.4633, acc-0.8705
Iter-49970, train loss-0.4834, acc-0.8800, valid loss-0.4596, acc-0.8740, test loss-0.4633, acc-0.8704
Iter-49980, train loss-0.5071, acc-0.8200, valid loss-0.4596, acc-0.8742, test loss-0.4632, acc-0.8704
Iter-49990, train loss-0.3298, acc-0.9400, valid loss-0.4595, acc-0.8742, test loss-0.4632, acc-0.8706
Iter-50000, train loss-0.6405, acc-0.8000, valid loss-0.4595, acc-0.8738, test loss-0.4631, acc-0.8704
Iter-50010, train loss-0.7687, acc-0.7400, valid loss-0.4594, acc-0.8742, test loss-0.4631, acc-0.8705
Iter-50020, train loss-0.4641, acc-0.8800, valid loss-0.4594, acc-0.8738, test loss-0.4631, acc-0.8706
Iter-50030, train loss-0.4154, acc-0.8800, valid loss-0.4594, acc-0.8740, test loss-0.4630, acc-0.8707
Iter-50040, train loss-0.4681, acc-0.9000, valid loss-0.4593, acc-0.8740, test loss-0.4630, acc-0.8707
Iter-50050, train loss-0.5833, acc-0.7800, valid loss-0.4593, acc-0.8742, test loss-0.4629, acc-0.8708
Iter-50060, train loss-0.3605, acc-0.9200, valid loss-0.4592, acc-0.8742, test loss-0.4629, acc-0.8708
Iter-50070, train loss-0.6520, acc-0.8000, valid loss-0.4592, acc-0.8742, test loss-0.4628, acc-0.8708
Iter-50080, train loss-0.7191, acc-0.8000, valid loss-0.4591, acc-0.8742, test loss-0.4628, acc-0.8708
Iter-50090, train loss-0.6071, acc-0.7600, valid loss-0.4591, acc-0.8744, test loss-0.4627, acc-0.8708
Iter-50100, train loss-0.3599, acc-0.8800, valid loss-0.4590, acc-0.8744, test loss-0.4627, acc-0.8708
Iter-50110, train loss-0.4223, acc-0.8800, valid loss-0.4590, acc-0.8744, test loss-0.4627, acc-0.8709
Iter-50120, train loss-0.4309, acc-0.8600, valid loss-0.4590, acc-0.8744, test loss-0.4626, acc-0.8708
Iter-50130, train loss-0.2830, acc-0.9000, valid loss-0.4589, acc-0.8744, test loss-0.4625, acc-0.8709
Iter-50140, train loss-0.7740, acc-0.7400, valid loss-0.4589, acc-0.8746, test loss-0.4625, acc-0.8709
Iter-50150, train loss-0.5236, acc-0.8600, valid loss-0.4588, acc-0.8746, test loss-0.4625, acc-0.8709
Iter-50160, train loss-0.3889, acc-0.8800, valid loss-0.4588, acc-0.8746, test loss-0.4624, acc-0.8709
Iter-50170, train loss-0.4204, acc-0.9000, valid loss-0.4587, acc-0.8746, test loss-0.4624, acc-0.8709
Iter-50180, train loss-0.5094, acc-0.8600, valid loss-0.4587, acc-0.8746, test loss-0.4623, acc-0.8709
Iter-50190, train loss-0.5299, acc-0.8000, valid loss-0.4586, acc-0.8746, test loss-0.4623, acc-0.8709
Iter-50200, train loss-0.6717, acc-0.8800, valid loss-0.4586, acc-0.8744, test loss-0.4623, acc-0.8709
Iter-50210, train loss-0.3417, acc-0.9000, valid loss-0.4585, acc-0.8744, test loss-0.4622, acc-0.8709
Iter-50220, train loss-0.6672, acc-0.7600, valid loss-0.4585, acc-0.8744, test loss-0.4622, acc-0.8709
Iter-50230, train loss-0.3454, acc-0.9200, valid loss-0.4584, acc-0.8744, test loss-0.4621, acc-0.8709
Iter-50240, train loss-0.3968, acc-0.9200, valid loss-0.4584, acc-0.8744, test loss-0.4621, acc-0.8710
Iter-50250, train loss-0.4941, acc-0.8800, valid loss-0.4584, acc-0.8742, test loss-0.4621, acc-0.8709
Iter-50260, train loss-0.4630, acc-0.9000, valid loss-0.4583, acc-0.8740, test loss-0.4620, acc-0.8709
Iter-50270, train loss-0.4294, acc-0.9000, valid loss-0.4583, acc-0.8742, test loss-0.4620, acc-0.8709
Iter-50280, train loss-0.4291, acc-0.8800, valid loss-0.4582, acc-0.8742, test loss-0.4619, acc-0.8709
Iter-50290, train loss-0.4730, acc-0.9000, valid loss-0.4582, acc-0.8742, test loss-0.4619, acc-0.8709
Iter-50300, train loss-0.3870, acc-0.8800, valid loss-0.4581, acc-0.8742, test loss-0.4619, acc-0.8709
Iter-50310, train loss-0.4784, acc-0.8800, valid loss-0.4581, acc-0.8742, test loss-0.4618, acc-0.8709
Iter-50320, train loss-0.4446, acc-0.9000, valid loss-0.4580, acc-0.8740, test loss-0.4618, acc-0.8709
Iter-50330, train loss-0.4226, acc-0.8800, valid loss-0.4580, acc-0.8740, test loss-0.4617, acc-0.8708
Iter-50340, train loss-0.3282, acc-0.9200, valid loss-0.4579, acc-0.8740, test loss-0.4617, acc-0.8707
Iter-50350, train loss-0.5439, acc-0.8800, valid loss-0.4579, acc-0.8740, test loss-0.4616, acc-0.8708
Iter-50360, train loss-0.4054, acc-0.8800, valid loss-0.4578, acc-0.8742, test loss-0.4616, acc-0.8711
Iter-50370, train loss-0.3694, acc-0.9000, valid loss-0.4578, acc-0.8740, test loss-0.4616, acc-0.8711
Iter-50380, train loss-0.5317, acc-0.9200, valid loss-0.4577, acc-0.8740, test loss-0.4615, acc-0.8712
Iter-50390, train loss-0.4190, acc-0.8800, valid loss-0.4577, acc-0.8742, test loss-0.4615, acc-0.8713
Iter-50400, train loss-0.6495, acc-0.7800, valid loss-0.4577, acc-0.8744, test loss-0.4614, acc-0.8713
Iter-50410, train loss-0.6067, acc-0.8400, valid loss-0.4576, acc-0.8742, test loss-0.4614, acc-0.8713
Iter-50420, train loss-0.6502, acc-0.8600, valid loss-0.4576, acc-0.8742, test loss-0.4613, acc-0.8713
Iter-50430, train loss-0.4822, acc-0.9200, valid loss-0.4575, acc-0.8742, test loss-0.4613, acc-0.8712
Iter-50440, train loss-0.3769, acc-0.9000, valid loss-0.4575, acc-0.8742, test loss-0.4613, acc-0.8713
Iter-50450, train loss-0.3671, acc-0.8800, valid loss-0.4574, acc-0.8742, test loss-0.4612, acc-0.8712
Iter-50460, train loss-0.4975, acc-0.8400, valid loss-0.4574, acc-0.8742, test loss-0.4612, acc-0.8713
Iter-50470, train loss-0.4058, acc-0.8600, valid loss-0.4573, acc-0.8742, test loss-0.4611, acc-0.8713
Iter-50480, train loss-0.6552, acc-0.8000, valid loss-0.4573, acc-0.8742, test loss-0.4611, acc-0.8713
Iter-50490, train loss-0.4914, acc-0.9000, valid loss-0.4572, acc-0.8742, test loss-0.4610, acc-0.8712
Iter-50500, train loss-0.7507, acc-0.7600, valid loss-0.4572, acc-0.8744, test loss-0.4610, acc-0.8712
Iter-50510, train loss-0.4095, acc-0.9200, valid loss-0.4572, acc-0.8742, test loss-0.4609, acc-0.8713
Iter-50520, train loss-0.6681, acc-0.7800, valid loss-0.4571, acc-0.8742, test loss-0.4609, acc-0.8713
Iter-50530, train loss-0.6648, acc-0.8000, valid loss-0.4571, acc-0.8740, test loss-0.4609, acc-0.8713
Iter-50540, train loss-0.3272, acc-0.9000, valid loss-0.4570, acc-0.8742, test loss-0.4608, acc-0.8711
Iter-50550, train loss-0.3573, acc-0.8800, valid loss-0.4570, acc-0.8742, test loss-0.4608, acc-0.8711
Iter-50560, train loss-0.5420, acc-0.9000, valid loss-0.4570, acc-0.8744, test loss-0.4607, acc-0.8712
Iter-50570, train loss-0.7246, acc-0.8200, valid loss-0.4569, acc-0.8744, test loss-0.4607, acc-0.8712
Iter-50580, train loss-0.3367, acc-0.9200, valid loss-0.4569, acc-0.8744, test loss-0.4606, acc-0.8712
Iter-50590, train loss-0.6455, acc-0.8200, valid loss-0.4568, acc-0.8742, test loss-0.4606, acc-0.8713
Iter-50600, train loss-0.4654, acc-0.8400, valid loss-0.4568, acc-0.8742, test loss-0.4606, acc-0.8714
Iter-50610, train loss-0.4922, acc-0.8600, valid loss-0.4567, acc-0.8742, test loss-0.4605, acc-0.8712
Iter-50620, train loss-0.6634, acc-0.8000, valid loss-0.4567, acc-0.8744, test loss-0.4605, acc-0.8712
Iter-50630, train loss-0.4989, acc-0.7800, valid loss-0.4566, acc-0.8742, test loss-0.4604, acc-0.8712
Iter-50640, train loss-0.4672, acc-0.8800, valid loss-0.4566, acc-0.8746, test loss-0.4604, acc-0.8711
Iter-50650, train loss-0.2694, acc-0.9600, valid loss-0.4566, acc-0.8746, test loss-0.4603, acc-0.8712
Iter-50660, train loss-0.3049, acc-0.9200, valid loss-0.4565, acc-0.8748, test loss-0.4603, acc-0.8711
Iter-50670, train loss-0.5876, acc-0.7800, valid loss-0.4565, acc-0.8746, test loss-0.4603, acc-0.8712
Iter-50680, train loss-0.4847, acc-0.9000, valid loss-0.4564, acc-0.8746, test loss-0.4602, acc-0.8712
Iter-50690, train loss-0.7449, acc-0.7600, valid loss-0.4564, acc-0.8746, test loss-0.4602, acc-0.8712
Iter-50700, train loss-0.4738, acc-0.9000, valid loss-0.4563, acc-0.8746, test loss-0.4601, acc-0.8712
Iter-50710, train loss-0.3848, acc-0.9000, valid loss-0.4563, acc-0.8746, test loss-0.4601, acc-0.8712
Iter-50720, train loss-0.7263, acc-0.7800, valid loss-0.4563, acc-0.8744, test loss-0.4600, acc-0.8712
Iter-50730, train loss-0.4372, acc-0.9000, valid loss-0.4562, acc-0.8744, test loss-0.4600, acc-0.8712
Iter-50740, train loss-0.7191, acc-0.7800, valid loss-0.4562, acc-0.8744, test loss-0.4600, acc-0.8714
Iter-50750, train loss-0.9142, acc-0.7800, valid loss-0.4561, acc-0.8748, test loss-0.4599, acc-0.8714
Iter-50760, train loss-0.4477, acc-0.9200, valid loss-0.4561, acc-0.8746, test loss-0.4599, acc-0.8713
Iter-50770, train loss-0.4752, acc-0.8200, valid loss-0.4560, acc-0.8748, test loss-0.4598, acc-0.8713
Iter-50780, train loss-0.3937, acc-0.9000, valid loss-0.4560, acc-0.8748, test loss-0.4598, acc-0.8714
Iter-50790, train loss-0.5223, acc-0.8200, valid loss-0.4560, acc-0.8748, test loss-0.4598, acc-0.8715
Iter-50800, train loss-0.4149, acc-0.8800, valid loss-0.4559, acc-0.8748, test loss-0.4597, acc-0.8716
Iter-50810, train loss-0.4333, acc-0.9000, valid loss-0.4559, acc-0.8746, test loss-0.4597, acc-0.8716
Iter-50820, train loss-0.4138, acc-0.9200, valid loss-0.4559, acc-0.8746, test loss-0.4596, acc-0.8716
Iter-50830, train loss-0.3978, acc-0.8800, valid loss-0.4558, acc-0.8748, test loss-0.4596, acc-0.8714
Iter-50840, train loss-0.4946, acc-0.9000, valid loss-0.4558, acc-0.8748, test loss-0.4595, acc-0.8714
Iter-50850, train loss-0.5399, acc-0.7800, valid loss-0.4558, acc-0.8748, test loss-0.4595, acc-0.8714
Iter-50860, train loss-0.5461, acc-0.8000, valid loss-0.4557, acc-0.8750, test loss-0.4594, acc-0.8713
Iter-50870, train loss-0.6718, acc-0.8000, valid loss-0.4557, acc-0.8750, test loss-0.4594, acc-0.8714
Iter-50880, train loss-0.3831, acc-0.9200, valid loss-0.4556, acc-0.8750, test loss-0.4594, acc-0.8714
Iter-50890, train loss-0.4395, acc-0.8800, valid loss-0.4556, acc-0.8750, test loss-0.4593, acc-0.8714
Iter-50900, train loss-0.4172, acc-0.8800, valid loss-0.4555, acc-0.8750, test loss-0.4593, acc-0.8713
Iter-50910, train loss-0.5760, acc-0.8400, valid loss-0.4555, acc-0.8750, test loss-0.4592, acc-0.8713
Iter-50920, train loss-0.5601, acc-0.8200, valid loss-0.4554, acc-0.8750, test loss-0.4592, acc-0.8713
Iter-50930, train loss-0.4416, acc-0.9200, valid loss-0.4554, acc-0.8752, test loss-0.4591, acc-0.8713
Iter-50940, train loss-0.2760, acc-0.9400, valid loss-0.4554, acc-0.8752, test loss-0.4591, acc-0.8714
Iter-50950, train loss-0.4841, acc-0.8400, valid loss-0.4553, acc-0.8752, test loss-0.4590, acc-0.8715
Iter-50960, train loss-0.3173, acc-0.9600, valid loss-0.4553, acc-0.8752, test loss-0.4590, acc-0.8715
Iter-50970, train loss-0.6188, acc-0.8000, valid loss-0.4552, acc-0.8752, test loss-0.4590, acc-0.8714
Iter-50980, train loss-0.2531, acc-0.9200, valid loss-0.4552, acc-0.8752, test loss-0.4589, acc-0.8714
Iter-50990, train loss-0.3172, acc-0.9400, valid loss-0.4552, acc-0.8752, test loss-0.4589, acc-0.8714
Iter-51000, train loss-0.4665, acc-0.8600, valid loss-0.4551, acc-0.8754, test loss-0.4588, acc-0.8715
Iter-51010, train loss-0.4007, acc-0.9000, valid loss-0.4551, acc-0.8754, test loss-0.4588, acc-0.8714
Iter-51020, train loss-0.3948, acc-0.9000, valid loss-0.4550, acc-0.8754, test loss-0.4587, acc-0.8716
Iter-51030, train loss-0.4367, acc-0.8400, valid loss-0.4550, acc-0.8754, test loss-0.4587, acc-0.8717
Iter-51040, train loss-0.5146, acc-0.8400, valid loss-0.4550, acc-0.8752, test loss-0.4586, acc-0.8716
Iter-51050, train loss-0.5285, acc-0.8200, valid loss-0.4549, acc-0.8752, test loss-0.4586, acc-0.8717
Iter-51060, train loss-0.3697, acc-0.9200, valid loss-0.4549, acc-0.8752, test loss-0.4586, acc-0.8717
Iter-51070, train loss-0.4784, acc-0.8600, valid loss-0.4548, acc-0.8752, test loss-0.4585, acc-0.8718
Iter-51080, train loss-0.4734, acc-0.8600, valid loss-0.4548, acc-0.8752, test loss-0.4585, acc-0.8717
Iter-51090, train loss-0.5617, acc-0.8600, valid loss-0.4548, acc-0.8752, test loss-0.4584, acc-0.8717
Iter-51100, train loss-0.4491, acc-0.9000, valid loss-0.4547, acc-0.8752, test loss-0.4584, acc-0.8717
Iter-51110, train loss-0.5474, acc-0.8600, valid loss-0.4547, acc-0.8752, test loss-0.4583, acc-0.8717
Iter-51120, train loss-0.4725, acc-0.9200, valid loss-0.4546, acc-0.8752, test loss-0.4583, acc-0.8717
Iter-51130, train loss-0.3976, acc-0.9000, valid loss-0.4546, acc-0.8752, test loss-0.4583, acc-0.8718
Iter-51140, train loss-0.3391, acc-0.9400, valid loss-0.4545, acc-0.8752, test loss-0.4582, acc-0.8717
Iter-51150, train loss-0.4713, acc-0.8800, valid loss-0.4545, acc-0.8752, test loss-0.4582, acc-0.8718
Iter-51160, train loss-0.4715, acc-0.8600, valid loss-0.4544, acc-0.8752, test loss-0.4581, acc-0.8718
Iter-51170, train loss-0.5593, acc-0.8600, valid loss-0.4544, acc-0.8752, test loss-0.4581, acc-0.8718
Iter-51180, train loss-0.6713, acc-0.8200, valid loss-0.4544, acc-0.8752, test loss-0.4580, acc-0.8719
Iter-51190, train loss-0.6826, acc-0.8400, valid loss-0.4543, acc-0.8752, test loss-0.4580, acc-0.8719
Iter-51200, train loss-0.4083, acc-0.9200, valid loss-0.4543, acc-0.8752, test loss-0.4580, acc-0.8719
Iter-51210, train loss-0.5245, acc-0.8200, valid loss-0.4542, acc-0.8752, test loss-0.4579, acc-0.8720
Iter-51220, train loss-0.4433, acc-0.8400, valid loss-0.4542, acc-0.8754, test loss-0.4579, acc-0.8720
Iter-51230, train loss-0.3588, acc-0.8800, valid loss-0.4541, acc-0.8754, test loss-0.4578, acc-0.8721
Iter-51240, train loss-0.4746, acc-0.8600, valid loss-0.4541, acc-0.8754, test loss-0.4578, acc-0.8720
Iter-51250, train loss-0.4588, acc-0.8800, valid loss-0.4541, acc-0.8752, test loss-0.4578, acc-0.8720
Iter-51260, train loss-0.5546, acc-0.8200, valid loss-0.4540, acc-0.8752, test loss-0.4577, acc-0.8720
Iter-51270, train loss-0.4230, acc-0.9000, valid loss-0.4540, acc-0.8752, test loss-0.4577, acc-0.8720
Iter-51280, train loss-0.4575, acc-0.8400, valid loss-0.4540, acc-0.8752, test loss-0.4576, acc-0.8720
Iter-51290, train loss-0.2756, acc-0.9400, valid loss-0.4539, acc-0.8752, test loss-0.4576, acc-0.8720
Iter-51300, train loss-0.5643, acc-0.8000, valid loss-0.4539, acc-0.8754, test loss-0.4576, acc-0.8720
Iter-51310, train loss-0.6119, acc-0.8000, valid loss-0.4538, acc-0.8754, test loss-0.4575, acc-0.8720
Iter-51320, train loss-0.4000, acc-0.9000, valid loss-0.4538, acc-0.8754, test loss-0.4575, acc-0.8722
Iter-51330, train loss-0.6561, acc-0.8000, valid loss-0.4537, acc-0.8754, test loss-0.4574, acc-0.8722
Iter-51340, train loss-0.4561, acc-0.8600, valid loss-0.4537, acc-0.8754, test loss-0.4574, acc-0.8723
Iter-51350, train loss-0.3748, acc-0.8600, valid loss-0.4537, acc-0.8754, test loss-0.4573, acc-0.8721
Iter-51360, train loss-0.3891, acc-0.9000, valid loss-0.4536, acc-0.8754, test loss-0.4573, acc-0.8721
Iter-51370, train loss-0.4078, acc-0.9200, valid loss-0.4536, acc-0.8754, test loss-0.4573, acc-0.8721
Iter-51380, train loss-0.6438, acc-0.8600, valid loss-0.4535, acc-0.8754, test loss-0.4572, acc-0.8721
Iter-51390, train loss-0.5439, acc-0.8200, valid loss-0.4535, acc-0.8754, test loss-0.4572, acc-0.8723
Iter-51400, train loss-0.4925, acc-0.8800, valid loss-0.4535, acc-0.8754, test loss-0.4571, acc-0.8722
Iter-51410, train loss-0.4097, acc-0.9000, valid loss-0.4534, acc-0.8754, test loss-0.4571, acc-0.8721
Iter-51420, train loss-0.3459, acc-0.9200, valid loss-0.4534, acc-0.8754, test loss-0.4570, acc-0.8719
Iter-51430, train loss-0.4557, acc-0.8600, valid loss-0.4533, acc-0.8754, test loss-0.4570, acc-0.8720
Iter-51440, train loss-0.5113, acc-0.8200, valid loss-0.4533, acc-0.8754, test loss-0.4570, acc-0.8722
Iter-51450, train loss-0.5205, acc-0.8400, valid loss-0.4533, acc-0.8754, test loss-0.4569, acc-0.8721
Iter-51460, train loss-0.5452, acc-0.8400, valid loss-0.4532, acc-0.8754, test loss-0.4569, acc-0.8722
Iter-51470, train loss-0.7780, acc-0.7600, valid loss-0.4532, acc-0.8754, test loss-0.4569, acc-0.8724
Iter-51480, train loss-0.5162, acc-0.8800, valid loss-0.4531, acc-0.8754, test loss-0.4568, acc-0.8725
Iter-51490, train loss-0.3683, acc-0.9400, valid loss-0.4531, acc-0.8752, test loss-0.4568, acc-0.8725
Iter-51500, train loss-0.4081, acc-0.9000, valid loss-0.4531, acc-0.8754, test loss-0.4567, acc-0.8725
Iter-51510, train loss-0.5267, acc-0.8400, valid loss-0.4530, acc-0.8754, test loss-0.4567, acc-0.8725
Iter-51520, train loss-0.3712, acc-0.9000, valid loss-0.4530, acc-0.8752, test loss-0.4567, acc-0.8725
Iter-51530, train loss-0.5415, acc-0.8000, valid loss-0.4529, acc-0.8754, test loss-0.4566, acc-0.8725
Iter-51540, train loss-0.3626, acc-0.9200, valid loss-0.4529, acc-0.8754, test loss-0.4566, acc-0.8724
Iter-51550, train loss-0.5306, acc-0.8600, valid loss-0.4529, acc-0.8754, test loss-0.4565, acc-0.8724
Iter-51560, train loss-0.3809, acc-0.9200, valid loss-0.4528, acc-0.8754, test loss-0.4565, acc-0.8725
Iter-51570, train loss-0.4724, acc-0.8600, valid loss-0.4527, acc-0.8754, test loss-0.4565, acc-0.8724
Iter-51580, train loss-0.3656, acc-0.8800, valid loss-0.4527, acc-0.8754, test loss-0.4564, acc-0.8725
Iter-51590, train loss-0.5554, acc-0.8600, valid loss-0.4527, acc-0.8754, test loss-0.4564, acc-0.8725
Iter-51600, train loss-0.4468, acc-0.8600, valid loss-0.4526, acc-0.8754, test loss-0.4563, acc-0.8725
Iter-51610, train loss-0.6057, acc-0.8200, valid loss-0.4526, acc-0.8754, test loss-0.4563, acc-0.8725
Iter-51620, train loss-0.4666, acc-0.8600, valid loss-0.4525, acc-0.8754, test loss-0.4563, acc-0.8726
Iter-51630, train loss-0.3661, acc-0.9000, valid loss-0.4525, acc-0.8754, test loss-0.4562, acc-0.8724
Iter-51640, train loss-0.5097, acc-0.8600, valid loss-0.4525, acc-0.8754, test loss-0.4562, acc-0.8725
Iter-51650, train loss-0.7068, acc-0.7600, valid loss-0.4524, acc-0.8754, test loss-0.4562, acc-0.8725
Iter-51660, train loss-0.2881, acc-0.9400, valid loss-0.4524, acc-0.8754, test loss-0.4561, acc-0.8725
Iter-51670, train loss-0.5781, acc-0.8000, valid loss-0.4523, acc-0.8754, test loss-0.4561, acc-0.8725
Iter-51680, train loss-0.5068, acc-0.8400, valid loss-0.4523, acc-0.8754, test loss-0.4560, acc-0.8725
Iter-51690, train loss-0.6723, acc-0.8000, valid loss-0.4522, acc-0.8754, test loss-0.4560, acc-0.8725
Iter-51700, train loss-0.5332, acc-0.8400, valid loss-0.4522, acc-0.8754, test loss-0.4559, acc-0.8725
Iter-51710, train loss-0.4916, acc-0.8600, valid loss-0.4521, acc-0.8754, test loss-0.4559, acc-0.8725
Iter-51720, train loss-0.5739, acc-0.8400, valid loss-0.4521, acc-0.8754, test loss-0.4558, acc-0.8725
Iter-51730, train loss-0.4309, acc-0.8400, valid loss-0.4521, acc-0.8754, test loss-0.4558, acc-0.8725
Iter-51740, train loss-0.4414, acc-0.9000, valid loss-0.4520, acc-0.8754, test loss-0.4557, acc-0.8725
Iter-51750, train loss-0.4649, acc-0.8800, valid loss-0.4520, acc-0.8754, test loss-0.4557, acc-0.8726
Iter-51760, train loss-0.6312, acc-0.8200, valid loss-0.4519, acc-0.8754, test loss-0.4557, acc-0.8727
Iter-51770, train loss-0.5290, acc-0.8600, valid loss-0.4519, acc-0.8754, test loss-0.4556, acc-0.8727
Iter-51780, train loss-0.4249, acc-0.8400, valid loss-0.4518, acc-0.8754, test loss-0.4556, acc-0.8727
Iter-51790, train loss-0.5528, acc-0.8200, valid loss-0.4518, acc-0.8754, test loss-0.4555, acc-0.8727
Iter-51800, train loss-0.4129, acc-0.9000, valid loss-0.4518, acc-0.8754, test loss-0.4555, acc-0.8727
Iter-51810, train loss-0.3656, acc-0.8800, valid loss-0.4517, acc-0.8754, test loss-0.4555, acc-0.8727
Iter-51820, train loss-0.4044, acc-0.9000, valid loss-0.4517, acc-0.8754, test loss-0.4554, acc-0.8727
Iter-51830, train loss-0.6493, acc-0.8400, valid loss-0.4516, acc-0.8754, test loss-0.4554, acc-0.8725
Iter-51840, train loss-0.4634, acc-0.8200, valid loss-0.4516, acc-0.8754, test loss-0.4553, acc-0.8726
Iter-51850, train loss-0.6689, acc-0.7800, valid loss-0.4516, acc-0.8754, test loss-0.4553, acc-0.8728
Iter-51860, train loss-0.3180, acc-0.9000, valid loss-0.4515, acc-0.8754, test loss-0.4553, acc-0.8725
Iter-51870, train loss-0.4703, acc-0.8600, valid loss-0.4515, acc-0.8754, test loss-0.4552, acc-0.8725
Iter-51880, train loss-0.5239, acc-0.8200, valid loss-0.4514, acc-0.8756, test loss-0.4552, acc-0.8726
Iter-51890, train loss-0.3961, acc-0.9000, valid loss-0.4514, acc-0.8754, test loss-0.4551, acc-0.8726
Iter-51900, train loss-0.3936, acc-0.9000, valid loss-0.4514, acc-0.8754, test loss-0.4551, acc-0.8726
Iter-51910, train loss-0.4397, acc-0.9000, valid loss-0.4513, acc-0.8754, test loss-0.4550, acc-0.8726
Iter-51920, train loss-0.6738, acc-0.8200, valid loss-0.4513, acc-0.8756, test loss-0.4550, acc-0.8726
Iter-51930, train loss-0.4604, acc-0.8800, valid loss-0.4513, acc-0.8754, test loss-0.4549, acc-0.8727
Iter-51940, train loss-0.6899, acc-0.8200, valid loss-0.4512, acc-0.8754, test loss-0.4549, acc-0.8726
Iter-51950, train loss-0.3778, acc-0.9000, valid loss-0.4512, acc-0.8754, test loss-0.4549, acc-0.8726
Iter-51960, train loss-0.5409, acc-0.8600, valid loss-0.4511, acc-0.8752, test loss-0.4548, acc-0.8727
Iter-51970, train loss-0.5466, acc-0.8600, valid loss-0.4511, acc-0.8754, test loss-0.4548, acc-0.8727
Iter-51980, train loss-0.5081, acc-0.8600, valid loss-0.4511, acc-0.8752, test loss-0.4548, acc-0.8727
Iter-51990, train loss-0.4967, acc-0.8800, valid loss-0.4510, acc-0.8752, test loss-0.4547, acc-0.8727
Iter-52000, train loss-0.8598, acc-0.8200, valid loss-0.4510, acc-0.8752, test loss-0.4547, acc-0.8727
Iter-52010, train loss-0.4471, acc-0.8800, valid loss-0.4509, acc-0.8752, test loss-0.4546, acc-0.8727
Iter-52020, train loss-0.5132, acc-0.8400, valid loss-0.4509, acc-0.8756, test loss-0.4546, acc-0.8727
Iter-52030, train loss-0.7235, acc-0.7600, valid loss-0.4509, acc-0.8756, test loss-0.4546, acc-0.8727
Iter-52040, train loss-0.5175, acc-0.8000, valid loss-0.4508, acc-0.8758, test loss-0.4545, acc-0.8727
Iter-52050, train loss-0.5733, acc-0.8200, valid loss-0.4508, acc-0.8758, test loss-0.4545, acc-0.8726
Iter-52060, train loss-0.4838, acc-0.8400, valid loss-0.4507, acc-0.8758, test loss-0.4544, acc-0.8726
Iter-52070, train loss-0.6006, acc-0.7600, valid loss-0.4507, acc-0.8758, test loss-0.4544, acc-0.8727
Iter-52080, train loss-0.4403, acc-0.9000, valid loss-0.4507, acc-0.8758, test loss-0.4543, acc-0.8727
Iter-52090, train loss-0.5903, acc-0.8000, valid loss-0.4506, acc-0.8758, test loss-0.4543, acc-0.8726
Iter-52100, train loss-0.5481, acc-0.7800, valid loss-0.4506, acc-0.8756, test loss-0.4543, acc-0.8726
Iter-52110, train loss-0.2910, acc-0.9400, valid loss-0.4505, acc-0.8756, test loss-0.4542, acc-0.8726
Iter-52120, train loss-0.2807, acc-0.9400, valid loss-0.4505, acc-0.8756, test loss-0.4542, acc-0.8726
Iter-52130, train loss-0.3628, acc-0.9200, valid loss-0.4505, acc-0.8756, test loss-0.4542, acc-0.8727
Iter-52140, train loss-0.4432, acc-0.8600, valid loss-0.4504, acc-0.8756, test loss-0.4541, acc-0.8726
Iter-52150, train loss-0.4496, acc-0.8200, valid loss-0.4504, acc-0.8754, test loss-0.4541, acc-0.8728
Iter-52160, train loss-0.6539, acc-0.7600, valid loss-0.4503, acc-0.8754, test loss-0.4540, acc-0.8728
Iter-52170, train loss-0.4780, acc-0.8400, valid loss-0.4503, acc-0.8754, test loss-0.4540, acc-0.8729
Iter-52180, train loss-0.3807, acc-0.9000, valid loss-0.4502, acc-0.8754, test loss-0.4540, acc-0.8729
Iter-52190, train loss-0.4305, acc-0.8800, valid loss-0.4502, acc-0.8754, test loss-0.4539, acc-0.8729
Iter-52200, train loss-0.5171, acc-0.8600, valid loss-0.4502, acc-0.8754, test loss-0.4539, acc-0.8729
Iter-52210, train loss-0.5845, acc-0.8000, valid loss-0.4501, acc-0.8754, test loss-0.4538, acc-0.8727
Iter-52220, train loss-0.5421, acc-0.8400, valid loss-0.4501, acc-0.8752, test loss-0.4538, acc-0.8727
Iter-52230, train loss-0.5515, acc-0.8400, valid loss-0.4500, acc-0.8754, test loss-0.4538, acc-0.8727
Iter-52240, train loss-0.5704, acc-0.8200, valid loss-0.4500, acc-0.8752, test loss-0.4537, acc-0.8728
Iter-52250, train loss-0.3911, acc-0.8600, valid loss-0.4500, acc-0.8752, test loss-0.4537, acc-0.8728
Iter-52260, train loss-0.4152, acc-0.9200, valid loss-0.4499, acc-0.8752, test loss-0.4536, acc-0.8728
Iter-52270, train loss-0.2712, acc-0.9600, valid loss-0.4499, acc-0.8752, test loss-0.4536, acc-0.8728
Iter-52280, train loss-0.3999, acc-0.9000, valid loss-0.4499, acc-0.8752, test loss-0.4536, acc-0.8728
Iter-52290, train loss-0.5385, acc-0.8600, valid loss-0.4498, acc-0.8752, test loss-0.4535, acc-0.8729
Iter-52300, train loss-0.5524, acc-0.8200, valid loss-0.4498, acc-0.8752, test loss-0.4535, acc-0.8729
Iter-52310, train loss-0.4932, acc-0.8400, valid loss-0.4497, acc-0.8752, test loss-0.4535, acc-0.8730
Iter-52320, train loss-0.7015, acc-0.8000, valid loss-0.4497, acc-0.8752, test loss-0.4534, acc-0.8729
Iter-52330, train loss-0.5615, acc-0.8400, valid loss-0.4497, acc-0.8754, test loss-0.4534, acc-0.8730
Iter-52340, train loss-0.3480, acc-0.9200, valid loss-0.4496, acc-0.8752, test loss-0.4533, acc-0.8731
Iter-52350, train loss-0.5785, acc-0.8400, valid loss-0.4496, acc-0.8754, test loss-0.4533, acc-0.8730
Iter-52360, train loss-0.3644, acc-0.8600, valid loss-0.4495, acc-0.8754, test loss-0.4533, acc-0.8731
Iter-52370, train loss-0.3373, acc-0.9400, valid loss-0.4495, acc-0.8754, test loss-0.4532, acc-0.8731
Iter-52380, train loss-0.4210, acc-0.8600, valid loss-0.4494, acc-0.8754, test loss-0.4532, acc-0.8732
Iter-52390, train loss-0.4372, acc-0.9000, valid loss-0.4494, acc-0.8754, test loss-0.4531, acc-0.8732
Iter-52400, train loss-0.3675, acc-0.8800, valid loss-0.4494, acc-0.8754, test loss-0.4531, acc-0.8731
Iter-52410, train loss-0.5340, acc-0.8400, valid loss-0.4493, acc-0.8758, test loss-0.4531, acc-0.8731
Iter-52420, train loss-0.4824, acc-0.8000, valid loss-0.4493, acc-0.8756, test loss-0.4530, acc-0.8731
Iter-52430, train loss-0.8925, acc-0.7400, valid loss-0.4492, acc-0.8758, test loss-0.4530, acc-0.8731
Iter-52440, train loss-0.5676, acc-0.8200, valid loss-0.4492, acc-0.8756, test loss-0.4530, acc-0.8732
Iter-52450, train loss-0.5850, acc-0.8400, valid loss-0.4492, acc-0.8756, test loss-0.4529, acc-0.8732
Iter-52460, train loss-0.5589, acc-0.8400, valid loss-0.4491, acc-0.8756, test loss-0.4529, acc-0.8732
Iter-52470, train loss-0.6975, acc-0.8400, valid loss-0.4491, acc-0.8756, test loss-0.4528, acc-0.8733
Iter-52480, train loss-0.4852, acc-0.8200, valid loss-0.4490, acc-0.8756, test loss-0.4528, acc-0.8733
Iter-52490, train loss-0.4283, acc-0.8600, valid loss-0.4490, acc-0.8756, test loss-0.4527, acc-0.8733
Iter-52500, train loss-0.4158, acc-0.8600, valid loss-0.4490, acc-0.8756, test loss-0.4527, acc-0.8732
Iter-52510, train loss-0.4162, acc-0.9200, valid loss-0.4489, acc-0.8756, test loss-0.4527, acc-0.8733
Iter-52520, train loss-0.5924, acc-0.8000, valid loss-0.4489, acc-0.8756, test loss-0.4526, acc-0.8733
Iter-52530, train loss-0.4595, acc-0.8600, valid loss-0.4488, acc-0.8758, test loss-0.4526, acc-0.8734
Iter-52540, train loss-0.4719, acc-0.8800, valid loss-0.4488, acc-0.8756, test loss-0.4525, acc-0.8733
Iter-52550, train loss-0.5544, acc-0.8200, valid loss-0.4487, acc-0.8756, test loss-0.4525, acc-0.8733
Iter-52560, train loss-0.4928, acc-0.8800, valid loss-0.4487, acc-0.8756, test loss-0.4524, acc-0.8732
Iter-52570, train loss-0.3343, acc-0.9400, valid loss-0.4487, acc-0.8756, test loss-0.4524, acc-0.8733
Iter-52580, train loss-0.5214, acc-0.8600, valid loss-0.4486, acc-0.8756, test loss-0.4524, acc-0.8734
Iter-52590, train loss-0.3063, acc-0.9400, valid loss-0.4486, acc-0.8756, test loss-0.4523, acc-0.8733
Iter-52600, train loss-0.5474, acc-0.9000, valid loss-0.4486, acc-0.8756, test loss-0.4523, acc-0.8733
Iter-52610, train loss-0.7415, acc-0.8000, valid loss-0.4485, acc-0.8756, test loss-0.4522, acc-0.8733
Iter-52620, train loss-0.6107, acc-0.8400, valid loss-0.4485, acc-0.8756, test loss-0.4522, acc-0.8734
Iter-52630, train loss-0.3734, acc-0.9000, valid loss-0.4484, acc-0.8756, test loss-0.4522, acc-0.8734
Iter-52640, train loss-0.5760, acc-0.8400, valid loss-0.4484, acc-0.8756, test loss-0.4521, acc-0.8734
Iter-52650, train loss-0.4285, acc-0.8800, valid loss-0.4484, acc-0.8756, test loss-0.4521, acc-0.8735
Iter-52660, train loss-0.4486, acc-0.8800, valid loss-0.4483, acc-0.8756, test loss-0.4521, acc-0.8734
Iter-52670, train loss-0.5655, acc-0.7400, valid loss-0.4483, acc-0.8756, test loss-0.4520, acc-0.8735
Iter-52680, train loss-0.4880, acc-0.8200, valid loss-0.4482, acc-0.8756, test loss-0.4520, acc-0.8734
Iter-52690, train loss-0.4898, acc-0.8600, valid loss-0.4482, acc-0.8756, test loss-0.4520, acc-0.8734
Iter-52700, train loss-0.5435, acc-0.8000, valid loss-0.4482, acc-0.8756, test loss-0.4519, acc-0.8733
Iter-52710, train loss-0.5437, acc-0.8600, valid loss-0.4481, acc-0.8756, test loss-0.4519, acc-0.8734
Iter-52720, train loss-0.5360, acc-0.8400, valid loss-0.4481, acc-0.8756, test loss-0.4518, acc-0.8735
Iter-52730, train loss-0.2591, acc-0.9400, valid loss-0.4481, acc-0.8756, test loss-0.4518, acc-0.8735
Iter-52740, train loss-0.3184, acc-0.9400, valid loss-0.4480, acc-0.8754, test loss-0.4518, acc-0.8735
Iter-52750, train loss-0.4751, acc-0.8400, valid loss-0.4480, acc-0.8756, test loss-0.4517, acc-0.8735
Iter-52760, train loss-0.2976, acc-0.9200, valid loss-0.4480, acc-0.8756, test loss-0.4517, acc-0.8735
Iter-52770, train loss-0.5444, acc-0.8200, valid loss-0.4479, acc-0.8756, test loss-0.4517, acc-0.8736
Iter-52780, train loss-0.4485, acc-0.8400, valid loss-0.4479, acc-0.8756, test loss-0.4516, acc-0.8736
Iter-52790, train loss-0.3825, acc-0.9000, valid loss-0.4478, acc-0.8756, test loss-0.4516, acc-0.8737
Iter-52800, train loss-0.6504, acc-0.8600, valid loss-0.4478, acc-0.8756, test loss-0.4516, acc-0.8737
Iter-52810, train loss-0.3103, acc-0.9200, valid loss-0.4477, acc-0.8756, test loss-0.4515, acc-0.8736
Iter-52820, train loss-0.5322, acc-0.8400, valid loss-0.4477, acc-0.8756, test loss-0.4515, acc-0.8737
Iter-52830, train loss-0.6497, acc-0.7600, valid loss-0.4477, acc-0.8756, test loss-0.4514, acc-0.8737
Iter-52840, train loss-0.3460, acc-0.9400, valid loss-0.4476, acc-0.8756, test loss-0.4514, acc-0.8736
Iter-52850, train loss-0.4417, acc-0.8400, valid loss-0.4476, acc-0.8756, test loss-0.4514, acc-0.8736
Iter-52860, train loss-0.2479, acc-0.9600, valid loss-0.4475, acc-0.8756, test loss-0.4513, acc-0.8736
Iter-52870, train loss-0.3991, acc-0.9000, valid loss-0.4475, acc-0.8756, test loss-0.4513, acc-0.8737
Iter-52880, train loss-0.4380, acc-0.8800, valid loss-0.4474, acc-0.8756, test loss-0.4512, acc-0.8737
Iter-52890, train loss-0.6772, acc-0.7800, valid loss-0.4474, acc-0.8756, test loss-0.4512, acc-0.8736
Iter-52900, train loss-0.5185, acc-0.8800, valid loss-0.4474, acc-0.8756, test loss-0.4512, acc-0.8738
Iter-52910, train loss-0.6762, acc-0.8600, valid loss-0.4473, acc-0.8756, test loss-0.4511, acc-0.8738
Iter-52920, train loss-0.2697, acc-0.9600, valid loss-0.4473, acc-0.8758, test loss-0.4511, acc-0.8738
Iter-52930, train loss-0.5272, acc-0.8600, valid loss-0.4472, acc-0.8758, test loss-0.4510, acc-0.8739
Iter-52940, train loss-0.3244, acc-0.9400, valid loss-0.4472, acc-0.8758, test loss-0.4510, acc-0.8739
Iter-52950, train loss-0.4108, acc-0.8800, valid loss-0.4471, acc-0.8760, test loss-0.4509, acc-0.8741
Iter-52960, train loss-0.3824, acc-0.9000, valid loss-0.4471, acc-0.8762, test loss-0.4509, acc-0.8741
Iter-52970, train loss-0.5556, acc-0.8000, valid loss-0.4471, acc-0.8760, test loss-0.4509, acc-0.8740
Iter-52980, train loss-0.6197, acc-0.8200, valid loss-0.4470, acc-0.8760, test loss-0.4508, acc-0.8740
Iter-52990, train loss-0.6157, acc-0.8600, valid loss-0.4470, acc-0.8762, test loss-0.4508, acc-0.8739
Iter-53000, train loss-0.5960, acc-0.8200, valid loss-0.4470, acc-0.8760, test loss-0.4508, acc-0.8740
Iter-53010, train loss-0.5596, acc-0.8000, valid loss-0.4469, acc-0.8764, test loss-0.4507, acc-0.8739
Iter-53020, train loss-0.5745, acc-0.9000, valid loss-0.4469, acc-0.8764, test loss-0.4507, acc-0.8738
Iter-53030, train loss-0.4787, acc-0.8600, valid loss-0.4468, acc-0.8764, test loss-0.4507, acc-0.8739
Iter-53040, train loss-0.6615, acc-0.8000, valid loss-0.4468, acc-0.8764, test loss-0.4506, acc-0.8738
Iter-53050, train loss-0.5103, acc-0.9000, valid loss-0.4468, acc-0.8764, test loss-0.4506, acc-0.8740
Iter-53060, train loss-0.4882, acc-0.8800, valid loss-0.4467, acc-0.8766, test loss-0.4506, acc-0.8739
Iter-53070, train loss-0.4463, acc-0.9200, valid loss-0.4467, acc-0.8768, test loss-0.4505, acc-0.8740
Iter-53080, train loss-0.3547, acc-0.8600, valid loss-0.4467, acc-0.8766, test loss-0.4505, acc-0.8741
Iter-53090, train loss-0.2628, acc-0.9600, valid loss-0.4466, acc-0.8768, test loss-0.4504, acc-0.8740
Iter-53100, train loss-0.5019, acc-0.8400, valid loss-0.4466, acc-0.8768, test loss-0.4504, acc-0.8739
Iter-53110, train loss-0.4081, acc-0.9200, valid loss-0.4465, acc-0.8768, test loss-0.4503, acc-0.8740
Iter-53120, train loss-0.4631, acc-0.8800, valid loss-0.4465, acc-0.8766, test loss-0.4503, acc-0.8740
Iter-53130, train loss-0.3635, acc-0.9200, valid loss-0.4464, acc-0.8764, test loss-0.4503, acc-0.8739
Iter-53140, train loss-0.4963, acc-0.8600, valid loss-0.4464, acc-0.8764, test loss-0.4502, acc-0.8739
Iter-53150, train loss-0.5945, acc-0.8400, valid loss-0.4464, acc-0.8764, test loss-0.4502, acc-0.8739
Iter-53160, train loss-0.3924, acc-0.9000, valid loss-0.4463, acc-0.8764, test loss-0.4501, acc-0.8739
Iter-53170, train loss-0.7007, acc-0.7800, valid loss-0.4463, acc-0.8766, test loss-0.4501, acc-0.8739
Iter-53180, train loss-0.5474, acc-0.8000, valid loss-0.4463, acc-0.8764, test loss-0.4501, acc-0.8740
Iter-53190, train loss-0.5580, acc-0.8400, valid loss-0.4462, acc-0.8764, test loss-0.4500, acc-0.8740
Iter-53200, train loss-0.4264, acc-0.8800, valid loss-0.4462, acc-0.8764, test loss-0.4500, acc-0.8739
Iter-53210, train loss-0.3937, acc-0.9200, valid loss-0.4462, acc-0.8764, test loss-0.4500, acc-0.8740
Iter-53220, train loss-0.3644, acc-0.9200, valid loss-0.4461, acc-0.8766, test loss-0.4499, acc-0.8740
Iter-53230, train loss-0.4709, acc-0.8600, valid loss-0.4461, acc-0.8766, test loss-0.4499, acc-0.8740
Iter-53240, train loss-0.4732, acc-0.8400, valid loss-0.4460, acc-0.8766, test loss-0.4498, acc-0.8742
Iter-53250, train loss-0.3665, acc-0.9400, valid loss-0.4460, acc-0.8766, test loss-0.4498, acc-0.8742
Iter-53260, train loss-0.4083, acc-0.8800, valid loss-0.4460, acc-0.8766, test loss-0.4498, acc-0.8743
Iter-53270, train loss-0.4148, acc-0.8800, valid loss-0.4459, acc-0.8766, test loss-0.4497, acc-0.8743
Iter-53280, train loss-0.2981, acc-0.9600, valid loss-0.4459, acc-0.8766, test loss-0.4497, acc-0.8742
Iter-53290, train loss-0.5993, acc-0.8200, valid loss-0.4458, acc-0.8766, test loss-0.4497, acc-0.8743
Iter-53300, train loss-0.5584, acc-0.8400, valid loss-0.4458, acc-0.8766, test loss-0.4496, acc-0.8741
Iter-53310, train loss-0.3879, acc-0.8800, valid loss-0.4457, acc-0.8768, test loss-0.4496, acc-0.8742
Iter-53320, train loss-0.2877, acc-0.9400, valid loss-0.4457, acc-0.8768, test loss-0.4495, acc-0.8743
Iter-53330, train loss-0.4910, acc-0.8600, valid loss-0.4457, acc-0.8766, test loss-0.4495, acc-0.8741
Iter-53340, train loss-0.3126, acc-0.9200, valid loss-0.4457, acc-0.8768, test loss-0.4495, acc-0.8740
Iter-53350, train loss-0.4252, acc-0.8400, valid loss-0.4456, acc-0.8768, test loss-0.4494, acc-0.8739
Iter-53360, train loss-0.4120, acc-0.9400, valid loss-0.4456, acc-0.8768, test loss-0.4494, acc-0.8740
Iter-53370, train loss-0.6369, acc-0.8200, valid loss-0.4456, acc-0.8768, test loss-0.4494, acc-0.8741
Iter-53380, train loss-0.6790, acc-0.8200, valid loss-0.4455, acc-0.8768, test loss-0.4493, acc-0.8740
Iter-53390, train loss-0.3214, acc-0.9000, valid loss-0.4455, acc-0.8768, test loss-0.4493, acc-0.8740
Iter-53400, train loss-0.4878, acc-0.8400, valid loss-0.4454, acc-0.8768, test loss-0.4493, acc-0.8740
Iter-53410, train loss-0.3773, acc-0.8800, valid loss-0.4454, acc-0.8768, test loss-0.4492, acc-0.8741
Iter-53420, train loss-0.2619, acc-0.9400, valid loss-0.4454, acc-0.8768, test loss-0.4492, acc-0.8740
Iter-53430, train loss-0.5875, acc-0.8400, valid loss-0.4453, acc-0.8770, test loss-0.4491, acc-0.8740
Iter-53440, train loss-0.3125, acc-0.9600, valid loss-0.4453, acc-0.8772, test loss-0.4491, acc-0.8742
Iter-53450, train loss-0.4458, acc-0.8800, valid loss-0.4453, acc-0.8772, test loss-0.4491, acc-0.8742
Iter-53460, train loss-0.5406, acc-0.8800, valid loss-0.4452, acc-0.8772, test loss-0.4490, acc-0.8744
Iter-53470, train loss-0.3325, acc-0.9400, valid loss-0.4452, acc-0.8772, test loss-0.4490, acc-0.8742
Iter-53480, train loss-0.4796, acc-0.8600, valid loss-0.4452, acc-0.8772, test loss-0.4490, acc-0.8744
Iter-53490, train loss-0.3428, acc-0.9200, valid loss-0.4451, acc-0.8772, test loss-0.4489, acc-0.8743
Iter-53500, train loss-0.5154, acc-0.8400, valid loss-0.4451, acc-0.8772, test loss-0.4489, acc-0.8743
Iter-53510, train loss-0.3603, acc-0.9000, valid loss-0.4450, acc-0.8772, test loss-0.4488, acc-0.8743
Iter-53520, train loss-0.4983, acc-0.8600, valid loss-0.4450, acc-0.8772, test loss-0.4488, acc-0.8744
Iter-53530, train loss-0.4594, acc-0.8600, valid loss-0.4450, acc-0.8772, test loss-0.4488, acc-0.8744
Iter-53540, train loss-0.5497, acc-0.8800, valid loss-0.4449, acc-0.8772, test loss-0.4487, acc-0.8744
Iter-53550, train loss-0.5397, acc-0.8400, valid loss-0.4449, acc-0.8772, test loss-0.4487, acc-0.8742
Iter-53560, train loss-0.4123, acc-0.8400, valid loss-0.4448, acc-0.8772, test loss-0.4486, acc-0.8743
Iter-53570, train loss-0.5686, acc-0.8400, valid loss-0.4448, acc-0.8772, test loss-0.4486, acc-0.8743
Iter-53580, train loss-0.4737, acc-0.9200, valid loss-0.4447, acc-0.8774, test loss-0.4486, acc-0.8743
Iter-53590, train loss-0.3574, acc-0.9400, valid loss-0.4447, acc-0.8774, test loss-0.4485, acc-0.8743
Iter-53600, train loss-0.7460, acc-0.8000, valid loss-0.4447, acc-0.8774, test loss-0.4485, acc-0.8743
Iter-53610, train loss-0.4350, acc-0.9000, valid loss-0.4446, acc-0.8774, test loss-0.4485, acc-0.8744
Iter-53620, train loss-0.4309, acc-0.9400, valid loss-0.4446, acc-0.8774, test loss-0.4484, acc-0.8744
Iter-53630, train loss-0.5419, acc-0.8200, valid loss-0.4445, acc-0.8774, test loss-0.4484, acc-0.8743
Iter-53640, train loss-0.5070, acc-0.8000, valid loss-0.4445, acc-0.8774, test loss-0.4483, acc-0.8744
Iter-53650, train loss-0.3870, acc-0.9400, valid loss-0.4444, acc-0.8774, test loss-0.4483, acc-0.8742
Iter-53660, train loss-0.5436, acc-0.8200, valid loss-0.4444, acc-0.8774, test loss-0.4483, acc-0.8742
Iter-53670, train loss-0.6052, acc-0.8000, valid loss-0.4444, acc-0.8774, test loss-0.4482, acc-0.8742
Iter-53680, train loss-0.4571, acc-0.8400, valid loss-0.4443, acc-0.8774, test loss-0.4482, acc-0.8742
Iter-53690, train loss-0.3827, acc-0.9000, valid loss-0.4443, acc-0.8774, test loss-0.4481, acc-0.8743
Iter-53700, train loss-0.2440, acc-0.9400, valid loss-0.4443, acc-0.8774, test loss-0.4481, acc-0.8742
Iter-53710, train loss-0.5080, acc-0.8800, valid loss-0.4442, acc-0.8774, test loss-0.4481, acc-0.8745
Iter-53720, train loss-0.5537, acc-0.8200, valid loss-0.4442, acc-0.8774, test loss-0.4480, acc-0.8743
Iter-53730, train loss-0.3804, acc-0.9000, valid loss-0.4441, acc-0.8774, test loss-0.4480, acc-0.8744
Iter-53740, train loss-0.3216, acc-0.8800, valid loss-0.4441, acc-0.8774, test loss-0.4480, acc-0.8744
Iter-53750, train loss-0.4151, acc-0.9400, valid loss-0.4440, acc-0.8774, test loss-0.4479, acc-0.8747
Iter-53760, train loss-0.3682, acc-0.8800, valid loss-0.4440, acc-0.8774, test loss-0.4479, acc-0.8746
Iter-53770, train loss-0.7306, acc-0.8600, valid loss-0.4440, acc-0.8774, test loss-0.4479, acc-0.8746
Iter-53780, train loss-0.4264, acc-0.8800, valid loss-0.4439, acc-0.8774, test loss-0.4478, acc-0.8744
Iter-53790, train loss-0.4088, acc-0.9000, valid loss-0.4439, acc-0.8774, test loss-0.4478, acc-0.8745
Iter-53800, train loss-0.5680, acc-0.8200, valid loss-0.4439, acc-0.8772, test loss-0.4477, acc-0.8746
Iter-53810, train loss-0.4503, acc-0.8600, valid loss-0.4438, acc-0.8776, test loss-0.4477, acc-0.8746
Iter-53820, train loss-0.6056, acc-0.8200, valid loss-0.4438, acc-0.8774, test loss-0.4477, acc-0.8745
Iter-53830, train loss-0.4637, acc-0.8600, valid loss-0.4437, acc-0.8776, test loss-0.4476, acc-0.8747
Iter-53840, train loss-0.5537, acc-0.8600, valid loss-0.4437, acc-0.8776, test loss-0.4476, acc-0.8745
Iter-53850, train loss-0.8056, acc-0.7200, valid loss-0.4437, acc-0.8776, test loss-0.4475, acc-0.8744
Iter-53860, train loss-0.2127, acc-0.9800, valid loss-0.4436, acc-0.8776, test loss-0.4475, acc-0.8746
Iter-53870, train loss-0.4633, acc-0.8400, valid loss-0.4436, acc-0.8774, test loss-0.4475, acc-0.8748
Iter-53880, train loss-0.3617, acc-0.9200, valid loss-0.4436, acc-0.8774, test loss-0.4474, acc-0.8748
Iter-53890, train loss-0.3179, acc-0.9000, valid loss-0.4435, acc-0.8774, test loss-0.4474, acc-0.8749
Iter-53900, train loss-0.4842, acc-0.8800, valid loss-0.4435, acc-0.8774, test loss-0.4473, acc-0.8748
Iter-53910, train loss-0.3818, acc-0.9000, valid loss-0.4434, acc-0.8774, test loss-0.4473, acc-0.8748
Iter-53920, train loss-0.6042, acc-0.8200, valid loss-0.4434, acc-0.8774, test loss-0.4473, acc-0.8749
Iter-53930, train loss-0.4141, acc-0.8800, valid loss-0.4433, acc-0.8774, test loss-0.4472, acc-0.8749
Iter-53940, train loss-0.4056, acc-0.9400, valid loss-0.4433, acc-0.8774, test loss-0.4472, acc-0.8749
Iter-53950, train loss-0.3339, acc-0.9200, valid loss-0.4433, acc-0.8774, test loss-0.4472, acc-0.8748
Iter-53960, train loss-0.3940, acc-0.9000, valid loss-0.4432, acc-0.8774, test loss-0.4471, acc-0.8748
Iter-53970, train loss-0.3857, acc-0.9200, valid loss-0.4432, acc-0.8774, test loss-0.4471, acc-0.8749
Iter-53980, train loss-0.4998, acc-0.9000, valid loss-0.4432, acc-0.8774, test loss-0.4470, acc-0.8748
Iter-53990, train loss-0.6602, acc-0.7600, valid loss-0.4431, acc-0.8774, test loss-0.4470, acc-0.8749
Iter-54000, train loss-0.6529, acc-0.7800, valid loss-0.4431, acc-0.8774, test loss-0.4470, acc-0.8748
Iter-54010, train loss-0.3952, acc-0.9000, valid loss-0.4431, acc-0.8774, test loss-0.4469, acc-0.8748
Iter-54020, train loss-0.4516, acc-0.8800, valid loss-0.4431, acc-0.8774, test loss-0.4469, acc-0.8747
Iter-54030, train loss-0.2526, acc-0.9800, valid loss-0.4430, acc-0.8776, test loss-0.4469, acc-0.8748
Iter-54040, train loss-0.4746, acc-0.8000, valid loss-0.4430, acc-0.8776, test loss-0.4468, acc-0.8747
Iter-54050, train loss-0.3324, acc-0.9200, valid loss-0.4429, acc-0.8776, test loss-0.4468, acc-0.8746
Iter-54060, train loss-0.5988, acc-0.8800, valid loss-0.4429, acc-0.8776, test loss-0.4468, acc-0.8746
Iter-54070, train loss-0.3164, acc-0.9200, valid loss-0.4429, acc-0.8776, test loss-0.4467, acc-0.8747
Iter-54080, train loss-0.4925, acc-0.8800, valid loss-0.4428, acc-0.8776, test loss-0.4467, acc-0.8747
Iter-54090, train loss-0.4868, acc-0.8600, valid loss-0.4428, acc-0.8776, test loss-0.4466, acc-0.8748
Iter-54100, train loss-0.3427, acc-0.9400, valid loss-0.4427, acc-0.8776, test loss-0.4466, acc-0.8747
Iter-54110, train loss-0.2975, acc-0.9400, valid loss-0.4427, acc-0.8776, test loss-0.4466, acc-0.8748
Iter-54120, train loss-0.3614, acc-0.9400, valid loss-0.4427, acc-0.8776, test loss-0.4465, acc-0.8747
Iter-54130, train loss-0.2336, acc-0.9600, valid loss-0.4426, acc-0.8776, test loss-0.4465, acc-0.8749
Iter-54140, train loss-0.4033, acc-0.8800, valid loss-0.4426, acc-0.8776, test loss-0.4464, acc-0.8749
Iter-54150, train loss-0.5171, acc-0.8800, valid loss-0.4425, acc-0.8776, test loss-0.4464, acc-0.8749
Iter-54160, train loss-0.2187, acc-0.9400, valid loss-0.4425, acc-0.8776, test loss-0.4464, acc-0.8750
Iter-54170, train loss-0.3886, acc-0.8600, valid loss-0.4425, acc-0.8776, test loss-0.4463, acc-0.8749
Iter-54180, train loss-0.8220, acc-0.7600, valid loss-0.4424, acc-0.8776, test loss-0.4463, acc-0.8749
Iter-54190, train loss-0.6683, acc-0.8000, valid loss-0.4424, acc-0.8776, test loss-0.4463, acc-0.8750
Iter-54200, train loss-0.3873, acc-0.8800, valid loss-0.4424, acc-0.8776, test loss-0.4462, acc-0.8750
Iter-54210, train loss-0.5287, acc-0.8600, valid loss-0.4423, acc-0.8776, test loss-0.4462, acc-0.8750
Iter-54220, train loss-0.4429, acc-0.8200, valid loss-0.4423, acc-0.8776, test loss-0.4461, acc-0.8753
Iter-54230, train loss-0.4438, acc-0.8200, valid loss-0.4423, acc-0.8776, test loss-0.4461, acc-0.8753
Iter-54240, train loss-0.3925, acc-0.8800, valid loss-0.4422, acc-0.8776, test loss-0.4461, acc-0.8753
Iter-54250, train loss-0.5531, acc-0.8200, valid loss-0.4422, acc-0.8776, test loss-0.4460, acc-0.8753
Iter-54260, train loss-0.7538, acc-0.7800, valid loss-0.4422, acc-0.8776, test loss-0.4460, acc-0.8754
Iter-54270, train loss-0.3727, acc-0.8600, valid loss-0.4421, acc-0.8776, test loss-0.4459, acc-0.8753
Iter-54280, train loss-0.4339, acc-0.8600, valid loss-0.4421, acc-0.8776, test loss-0.4459, acc-0.8752
Iter-54290, train loss-0.3548, acc-0.9200, valid loss-0.4420, acc-0.8776, test loss-0.4459, acc-0.8753
Iter-54300, train loss-0.4856, acc-0.8600, valid loss-0.4420, acc-0.8774, test loss-0.4458, acc-0.8753
Iter-54310, train loss-0.4228, acc-0.8800, valid loss-0.4420, acc-0.8774, test loss-0.4458, acc-0.8752
Iter-54320, train loss-0.4731, acc-0.8200, valid loss-0.4419, acc-0.8776, test loss-0.4457, acc-0.8753
Iter-54330, train loss-0.3780, acc-0.9600, valid loss-0.4419, acc-0.8776, test loss-0.4457, acc-0.8754
Iter-54340, train loss-0.5140, acc-0.8800, valid loss-0.4418, acc-0.8776, test loss-0.4457, acc-0.8753
Iter-54350, train loss-0.7347, acc-0.7800, valid loss-0.4418, acc-0.8776, test loss-0.4456, acc-0.8753
Iter-54360, train loss-0.5853, acc-0.8400, valid loss-0.4418, acc-0.8776, test loss-0.4456, acc-0.8753
Iter-54370, train loss-0.4313, acc-0.9000, valid loss-0.4417, acc-0.8776, test loss-0.4456, acc-0.8752
Iter-54380, train loss-0.4816, acc-0.9000, valid loss-0.4417, acc-0.8776, test loss-0.4455, acc-0.8751
Iter-54390, train loss-0.4425, acc-0.8400, valid loss-0.4417, acc-0.8776, test loss-0.4455, acc-0.8754
Iter-54400, train loss-0.5553, acc-0.8200, valid loss-0.4416, acc-0.8776, test loss-0.4455, acc-0.8753
Iter-54410, train loss-0.4074, acc-0.8600, valid loss-0.4416, acc-0.8774, test loss-0.4454, acc-0.8753
Iter-54420, train loss-0.4427, acc-0.9000, valid loss-0.4416, acc-0.8774, test loss-0.4454, acc-0.8755
Iter-54430, train loss-0.4354, acc-0.8800, valid loss-0.4415, acc-0.8774, test loss-0.4454, acc-0.8755
Iter-54440, train loss-0.4844, acc-0.9000, valid loss-0.4415, acc-0.8774, test loss-0.4453, acc-0.8755
Iter-54450, train loss-0.4224, acc-0.8600, valid loss-0.4414, acc-0.8772, test loss-0.4453, acc-0.8755
Iter-54460, train loss-0.2265, acc-0.9400, valid loss-0.4414, acc-0.8774, test loss-0.4453, acc-0.8753
Iter-54470, train loss-0.3445, acc-0.9000, valid loss-0.4414, acc-0.8774, test loss-0.4452, acc-0.8752
Iter-54480, train loss-0.4072, acc-0.8600, valid loss-0.4413, acc-0.8772, test loss-0.4452, acc-0.8754
Iter-54490, train loss-0.6509, acc-0.8000, valid loss-0.4413, acc-0.8772, test loss-0.4452, acc-0.8753
Iter-54500, train loss-0.4220, acc-0.9000, valid loss-0.4413, acc-0.8772, test loss-0.4451, acc-0.8753
Iter-54510, train loss-0.5206, acc-0.8600, valid loss-0.4412, acc-0.8772, test loss-0.4451, acc-0.8753
Iter-54520, train loss-0.4156, acc-0.8600, valid loss-0.4412, acc-0.8774, test loss-0.4450, acc-0.8752
Iter-54530, train loss-0.5026, acc-0.7800, valid loss-0.4411, acc-0.8774, test loss-0.4450, acc-0.8752
Iter-54540, train loss-1.0057, acc-0.7600, valid loss-0.4411, acc-0.8774, test loss-0.4450, acc-0.8752
Iter-54550, train loss-0.2628, acc-0.9800, valid loss-0.4411, acc-0.8774, test loss-0.4450, acc-0.8753
Iter-54560, train loss-0.5017, acc-0.8800, valid loss-0.4410, acc-0.8774, test loss-0.4449, acc-0.8751
Iter-54570, train loss-0.4488, acc-0.8800, valid loss-0.4410, acc-0.8774, test loss-0.4449, acc-0.8754
Iter-54580, train loss-0.6327, acc-0.8000, valid loss-0.4409, acc-0.8774, test loss-0.4448, acc-0.8754
Iter-54590, train loss-0.5416, acc-0.8400, valid loss-0.4409, acc-0.8774, test loss-0.4448, acc-0.8752
Iter-54600, train loss-0.3446, acc-0.9400, valid loss-0.4409, acc-0.8774, test loss-0.4447, acc-0.8752
Iter-54610, train loss-0.3181, acc-0.9200, valid loss-0.4408, acc-0.8774, test loss-0.4447, acc-0.8751
Iter-54620, train loss-0.5057, acc-0.8000, valid loss-0.4408, acc-0.8776, test loss-0.4447, acc-0.8751
Iter-54630, train loss-0.6856, acc-0.8000, valid loss-0.4408, acc-0.8774, test loss-0.4446, acc-0.8752
Iter-54640, train loss-0.2738, acc-0.9200, valid loss-0.4408, acc-0.8774, test loss-0.4446, acc-0.8751
Iter-54650, train loss-0.6042, acc-0.8400, valid loss-0.4407, acc-0.8774, test loss-0.4446, acc-0.8750
Iter-54660, train loss-0.7000, acc-0.8000, valid loss-0.4407, acc-0.8774, test loss-0.4445, acc-0.8750
Iter-54670, train loss-0.5386, acc-0.8200, valid loss-0.4407, acc-0.8776, test loss-0.4445, acc-0.8752
Iter-54680, train loss-0.5567, acc-0.8200, valid loss-0.4406, acc-0.8776, test loss-0.4445, acc-0.8752
Iter-54690, train loss-0.4527, acc-0.8800, valid loss-0.4406, acc-0.8780, test loss-0.4444, acc-0.8753
Iter-54700, train loss-0.5306, acc-0.8800, valid loss-0.4406, acc-0.8778, test loss-0.4444, acc-0.8753
Iter-54710, train loss-0.4892, acc-0.8400, valid loss-0.4405, acc-0.8778, test loss-0.4444, acc-0.8753
Iter-54720, train loss-0.3771, acc-0.9000, valid loss-0.4405, acc-0.8778, test loss-0.4443, acc-0.8753
Iter-54730, train loss-0.2805, acc-0.9200, valid loss-0.4404, acc-0.8778, test loss-0.4443, acc-0.8754
Iter-54740, train loss-0.3911, acc-0.8600, valid loss-0.4404, acc-0.8778, test loss-0.4442, acc-0.8754
Iter-54750, train loss-0.4597, acc-0.8800, valid loss-0.4404, acc-0.8778, test loss-0.4442, acc-0.8754
Iter-54760, train loss-0.6021, acc-0.8400, valid loss-0.4403, acc-0.8778, test loss-0.4442, acc-0.8755
Iter-54770, train loss-0.4661, acc-0.8600, valid loss-0.4403, acc-0.8778, test loss-0.4441, acc-0.8757
Iter-54780, train loss-0.5213, acc-0.8800, valid loss-0.4402, acc-0.8780, test loss-0.4441, acc-0.8755
Iter-54790, train loss-0.4192, acc-0.8600, valid loss-0.4402, acc-0.8780, test loss-0.4441, acc-0.8754
Iter-54800, train loss-0.3686, acc-0.9000, valid loss-0.4402, acc-0.8780, test loss-0.4440, acc-0.8757
Iter-54810, train loss-0.4337, acc-0.8800, valid loss-0.4401, acc-0.8782, test loss-0.4440, acc-0.8755
Iter-54820, train loss-0.4391, acc-0.8800, valid loss-0.4401, acc-0.8782, test loss-0.4440, acc-0.8755
Iter-54830, train loss-0.4621, acc-0.9200, valid loss-0.4400, acc-0.8780, test loss-0.4439, acc-0.8756
Iter-54840, train loss-0.2735, acc-0.9400, valid loss-0.4400, acc-0.8782, test loss-0.4439, acc-0.8757
Iter-54850, train loss-0.3084, acc-0.9200, valid loss-0.4399, acc-0.8780, test loss-0.4439, acc-0.8758
Iter-54860, train loss-0.3876, acc-0.9000, valid loss-0.4399, acc-0.8782, test loss-0.4438, acc-0.8758
Iter-54870, train loss-0.4810, acc-0.8200, valid loss-0.4399, acc-0.8782, test loss-0.4438, acc-0.8758
Iter-54880, train loss-0.3619, acc-0.8800, valid loss-0.4398, acc-0.8782, test loss-0.4437, acc-0.8758
Iter-54890, train loss-0.7206, acc-0.7800, valid loss-0.4398, acc-0.8782, test loss-0.4437, acc-0.8758
Iter-54900, train loss-0.6481, acc-0.8600, valid loss-0.4398, acc-0.8780, test loss-0.4437, acc-0.8758
Iter-54910, train loss-0.3286, acc-0.9400, valid loss-0.4397, acc-0.8782, test loss-0.4436, acc-0.8759
Iter-54920, train loss-0.4679, acc-0.9000, valid loss-0.4397, acc-0.8780, test loss-0.4436, acc-0.8759
Iter-54930, train loss-0.4407, acc-0.9000, valid loss-0.4397, acc-0.8778, test loss-0.4436, acc-0.8756
Iter-54940, train loss-0.4596, acc-0.8600, valid loss-0.4396, acc-0.8778, test loss-0.4435, acc-0.8757
Iter-54950, train loss-0.6625, acc-0.8000, valid loss-0.4396, acc-0.8778, test loss-0.4435, acc-0.8758
Iter-54960, train loss-0.5520, acc-0.8600, valid loss-0.4396, acc-0.8780, test loss-0.4434, acc-0.8759
Iter-54970, train loss-0.4772, acc-0.8200, valid loss-0.4395, acc-0.8780, test loss-0.4434, acc-0.8759
Iter-54980, train loss-0.5181, acc-0.8800, valid loss-0.4395, acc-0.8780, test loss-0.4434, acc-0.8759
Iter-54990, train loss-0.3765, acc-0.8800, valid loss-0.4395, acc-0.8780, test loss-0.4433, acc-0.8759
Iter-55000, train loss-0.5385, acc-0.8800, valid loss-0.4394, acc-0.8780, test loss-0.4433, acc-0.8759
Iter-55010, train loss-0.3727, acc-0.9000, valid loss-0.4394, acc-0.8780, test loss-0.4433, acc-0.8759
Iter-55020, train loss-0.3527, acc-0.9000, valid loss-0.4394, acc-0.8782, test loss-0.4432, acc-0.8758
Iter-55030, train loss-0.4273, acc-0.8600, valid loss-0.4393, acc-0.8780, test loss-0.4432, acc-0.8758
Iter-55040, train loss-0.2030, acc-0.9600, valid loss-0.4393, acc-0.8780, test loss-0.4432, acc-0.8759
Iter-55050, train loss-0.5616, acc-0.8400, valid loss-0.4393, acc-0.8780, test loss-0.4431, acc-0.8758
Iter-55060, train loss-0.3707, acc-0.9400, valid loss-0.4392, acc-0.8780, test loss-0.4431, acc-0.8757
Iter-55070, train loss-0.2397, acc-0.9600, valid loss-0.4392, acc-0.8780, test loss-0.4431, acc-0.8758
Iter-55080, train loss-0.4369, acc-0.8600, valid loss-0.4391, acc-0.8780, test loss-0.4430, acc-0.8756
Iter-55090, train loss-0.4372, acc-0.8800, valid loss-0.4391, acc-0.8782, test loss-0.4430, acc-0.8755
Iter-55100, train loss-0.6114, acc-0.8600, valid loss-0.4391, acc-0.8782, test loss-0.4429, acc-0.8760
Iter-55110, train loss-0.3317, acc-0.9400, valid loss-0.4390, acc-0.8780, test loss-0.4429, acc-0.8758
Iter-55120, train loss-0.5638, acc-0.8400, valid loss-0.4390, acc-0.8780, test loss-0.4429, acc-0.8758
Iter-55130, train loss-0.4036, acc-0.8600, valid loss-0.4390, acc-0.8780, test loss-0.4429, acc-0.8759
Iter-55140, train loss-0.3676, acc-0.9000, valid loss-0.4389, acc-0.8778, test loss-0.4428, acc-0.8759
Iter-55150, train loss-0.6058, acc-0.8200, valid loss-0.4389, acc-0.8778, test loss-0.4428, acc-0.8761
Iter-55160, train loss-0.5372, acc-0.8800, valid loss-0.4388, acc-0.8778, test loss-0.4427, acc-0.8761
Iter-55170, train loss-0.4337, acc-0.8800, valid loss-0.4388, acc-0.8778, test loss-0.4427, acc-0.8761
Iter-55180, train loss-0.3721, acc-0.9200, valid loss-0.4388, acc-0.8780, test loss-0.4427, acc-0.8760
Iter-55190, train loss-0.5748, acc-0.8400, valid loss-0.4387, acc-0.8778, test loss-0.4426, acc-0.8761
Iter-55200, train loss-0.5276, acc-0.8400, valid loss-0.4387, acc-0.8778, test loss-0.4426, acc-0.8761
Iter-55210, train loss-0.4883, acc-0.8200, valid loss-0.4387, acc-0.8778, test loss-0.4426, acc-0.8761
Iter-55220, train loss-0.5563, acc-0.9000, valid loss-0.4386, acc-0.8778, test loss-0.4425, acc-0.8760
Iter-55230, train loss-0.5110, acc-0.8600, valid loss-0.4386, acc-0.8778, test loss-0.4425, acc-0.8762
Iter-55240, train loss-0.5811, acc-0.8600, valid loss-0.4385, acc-0.8780, test loss-0.4425, acc-0.8761
Iter-55250, train loss-0.5161, acc-0.8400, valid loss-0.4385, acc-0.8780, test loss-0.4424, acc-0.8761
Iter-55260, train loss-0.5778, acc-0.9000, valid loss-0.4385, acc-0.8780, test loss-0.4424, acc-0.8761
Iter-55270, train loss-0.3907, acc-0.8200, valid loss-0.4385, acc-0.8780, test loss-0.4424, acc-0.8760
Iter-55280, train loss-0.3057, acc-0.9400, valid loss-0.4384, acc-0.8780, test loss-0.4423, acc-0.8760
Iter-55290, train loss-0.5105, acc-0.8200, valid loss-0.4384, acc-0.8780, test loss-0.4423, acc-0.8761
Iter-55300, train loss-0.4315, acc-0.8600, valid loss-0.4383, acc-0.8782, test loss-0.4423, acc-0.8761
Iter-55310, train loss-0.3065, acc-0.9200, valid loss-0.4383, acc-0.8778, test loss-0.4422, acc-0.8761
Iter-55320, train loss-0.5427, acc-0.8200, valid loss-0.4383, acc-0.8780, test loss-0.4422, acc-0.8760
Iter-55330, train loss-0.4610, acc-0.9000, valid loss-0.4382, acc-0.8782, test loss-0.4421, acc-0.8759
Iter-55340, train loss-0.4550, acc-0.8200, valid loss-0.4382, acc-0.8782, test loss-0.4421, acc-0.8760
Iter-55350, train loss-0.4568, acc-0.8200, valid loss-0.4381, acc-0.8780, test loss-0.4421, acc-0.8760
Iter-55360, train loss-0.3442, acc-0.9000, valid loss-0.4381, acc-0.8780, test loss-0.4420, acc-0.8760
Iter-55370, train loss-0.3848, acc-0.9000, valid loss-0.4381, acc-0.8780, test loss-0.4420, acc-0.8760
Iter-55380, train loss-0.3733, acc-0.9600, valid loss-0.4380, acc-0.8782, test loss-0.4420, acc-0.8760
Iter-55390, train loss-0.2965, acc-0.9400, valid loss-0.4380, acc-0.8782, test loss-0.4419, acc-0.8759
Iter-55400, train loss-0.2817, acc-0.9200, valid loss-0.4379, acc-0.8782, test loss-0.4419, acc-0.8759
Iter-55410, train loss-0.4147, acc-0.8400, valid loss-0.4379, acc-0.8780, test loss-0.4419, acc-0.8759
Iter-55420, train loss-0.3856, acc-0.8200, valid loss-0.4379, acc-0.8780, test loss-0.4418, acc-0.8758
Iter-55430, train loss-0.5220, acc-0.8200, valid loss-0.4378, acc-0.8782, test loss-0.4418, acc-0.8760
Iter-55440, train loss-0.4932, acc-0.8600, valid loss-0.4378, acc-0.8780, test loss-0.4418, acc-0.8760
Iter-55450, train loss-0.4488, acc-0.8800, valid loss-0.4378, acc-0.8780, test loss-0.4417, acc-0.8762
Iter-55460, train loss-0.4332, acc-0.8600, valid loss-0.4377, acc-0.8782, test loss-0.4417, acc-0.8761
Iter-55470, train loss-0.4162, acc-0.9200, valid loss-0.4377, acc-0.8782, test loss-0.4416, acc-0.8760
Iter-55480, train loss-0.4838, acc-0.8600, valid loss-0.4376, acc-0.8780, test loss-0.4416, acc-0.8763
Iter-55490, train loss-0.4528, acc-0.8800, valid loss-0.4376, acc-0.8780, test loss-0.4416, acc-0.8761
Iter-55500, train loss-0.6468, acc-0.8000, valid loss-0.4376, acc-0.8780, test loss-0.4415, acc-0.8762
Iter-55510, train loss-0.6333, acc-0.8600, valid loss-0.4375, acc-0.8780, test loss-0.4415, acc-0.8762
Iter-55520, train loss-0.4506, acc-0.8600, valid loss-0.4375, acc-0.8780, test loss-0.4414, acc-0.8761
Iter-55530, train loss-0.6455, acc-0.9000, valid loss-0.4375, acc-0.8780, test loss-0.4414, acc-0.8761
Iter-55540, train loss-0.2844, acc-0.9400, valid loss-0.4374, acc-0.8778, test loss-0.4414, acc-0.8761
Iter-55550, train loss-0.3853, acc-0.9200, valid loss-0.4374, acc-0.8780, test loss-0.4413, acc-0.8761
Iter-55560, train loss-0.4256, acc-0.8400, valid loss-0.4374, acc-0.8780, test loss-0.4413, acc-0.8760
Iter-55570, train loss-0.4238, acc-0.8600, valid loss-0.4373, acc-0.8780, test loss-0.4413, acc-0.8760
Iter-55580, train loss-0.6009, acc-0.8200, valid loss-0.4373, acc-0.8780, test loss-0.4412, acc-0.8759
Iter-55590, train loss-0.5208, acc-0.8600, valid loss-0.4372, acc-0.8780, test loss-0.4412, acc-0.8760
Iter-55600, train loss-0.5899, acc-0.7800, valid loss-0.4372, acc-0.8780, test loss-0.4412, acc-0.8759
Iter-55610, train loss-0.3157, acc-0.9400, valid loss-0.4372, acc-0.8780, test loss-0.4411, acc-0.8758
Iter-55620, train loss-0.3733, acc-0.9000, valid loss-0.4372, acc-0.8780, test loss-0.4411, acc-0.8758
Iter-55630, train loss-0.3052, acc-0.9600, valid loss-0.4371, acc-0.8784, test loss-0.4410, acc-0.8758
Iter-55640, train loss-0.4151, acc-0.8600, valid loss-0.4371, acc-0.8784, test loss-0.4410, acc-0.8758
Iter-55650, train loss-0.4669, acc-0.8800, valid loss-0.4370, acc-0.8784, test loss-0.4410, acc-0.8759
Iter-55660, train loss-0.3549, acc-0.8800, valid loss-0.4370, acc-0.8784, test loss-0.4409, acc-0.8760
Iter-55670, train loss-0.5949, acc-0.8200, valid loss-0.4370, acc-0.8782, test loss-0.4409, acc-0.8760
Iter-55680, train loss-0.4954, acc-0.8800, valid loss-0.4369, acc-0.8782, test loss-0.4408, acc-0.8761
Iter-55690, train loss-0.4691, acc-0.8400, valid loss-0.4369, acc-0.8780, test loss-0.4408, acc-0.8761
Iter-55700, train loss-0.3456, acc-0.9200, valid loss-0.4369, acc-0.8782, test loss-0.4408, acc-0.8762
Iter-55710, train loss-0.6056, acc-0.8000, valid loss-0.4368, acc-0.8784, test loss-0.4407, acc-0.8761
Iter-55720, train loss-0.3979, acc-0.9000, valid loss-0.4368, acc-0.8784, test loss-0.4407, acc-0.8761
Iter-55730, train loss-0.4034, acc-0.8600, valid loss-0.4368, acc-0.8784, test loss-0.4406, acc-0.8761
Iter-55740, train loss-0.7079, acc-0.7800, valid loss-0.4368, acc-0.8782, test loss-0.4406, acc-0.8761
Iter-55750, train loss-0.4013, acc-0.9000, valid loss-0.4367, acc-0.8786, test loss-0.4406, acc-0.8762
Iter-55760, train loss-0.3076, acc-0.9400, valid loss-0.4367, acc-0.8786, test loss-0.4406, acc-0.8763
Iter-55770, train loss-0.4126, acc-0.9000, valid loss-0.4367, acc-0.8788, test loss-0.4405, acc-0.8762
Iter-55780, train loss-0.4234, acc-0.8600, valid loss-0.4366, acc-0.8786, test loss-0.4405, acc-0.8762
Iter-55790, train loss-0.6480, acc-0.8000, valid loss-0.4366, acc-0.8788, test loss-0.4405, acc-0.8759
Iter-55800, train loss-0.6211, acc-0.8000, valid loss-0.4366, acc-0.8788, test loss-0.4404, acc-0.8760
Iter-55810, train loss-0.4370, acc-0.8800, valid loss-0.4365, acc-0.8788, test loss-0.4404, acc-0.8760
Iter-55820, train loss-0.4353, acc-0.8800, valid loss-0.4365, acc-0.8786, test loss-0.4403, acc-0.8760
Iter-55830, train loss-0.3545, acc-0.9400, valid loss-0.4364, acc-0.8788, test loss-0.4403, acc-0.8761
Iter-55840, train loss-0.2965, acc-0.9000, valid loss-0.4364, acc-0.8790, test loss-0.4403, acc-0.8761
Iter-55850, train loss-0.2730, acc-0.9400, valid loss-0.4364, acc-0.8788, test loss-0.4402, acc-0.8760
Iter-55860, train loss-0.3268, acc-0.9400, valid loss-0.4363, acc-0.8790, test loss-0.4402, acc-0.8760
Iter-55870, train loss-0.3795, acc-0.9000, valid loss-0.4363, acc-0.8788, test loss-0.4401, acc-0.8760
Iter-55880, train loss-0.3144, acc-0.9000, valid loss-0.4363, acc-0.8790, test loss-0.4401, acc-0.8760
Iter-55890, train loss-0.3709, acc-0.9400, valid loss-0.4362, acc-0.8788, test loss-0.4401, acc-0.8761
Iter-55900, train loss-0.3842, acc-0.8800, valid loss-0.4362, acc-0.8792, test loss-0.4401, acc-0.8760
Iter-55910, train loss-0.6529, acc-0.7800, valid loss-0.4362, acc-0.8790, test loss-0.4400, acc-0.8759
Iter-55920, train loss-0.3107, acc-0.9000, valid loss-0.4361, acc-0.8792, test loss-0.4400, acc-0.8760
Iter-55930, train loss-0.5440, acc-0.9000, valid loss-0.4361, acc-0.8786, test loss-0.4400, acc-0.8760
Iter-55940, train loss-0.6319, acc-0.8000, valid loss-0.4361, acc-0.8790, test loss-0.4399, acc-0.8760
Iter-55950, train loss-0.5056, acc-0.9000, valid loss-0.4360, acc-0.8792, test loss-0.4399, acc-0.8759
Iter-55960, train loss-0.4442, acc-0.8800, valid loss-0.4360, acc-0.8788, test loss-0.4399, acc-0.8759
Iter-55970, train loss-0.3643, acc-0.9000, valid loss-0.4360, acc-0.8788, test loss-0.4398, acc-0.8759
Iter-55980, train loss-0.3981, acc-0.9000, valid loss-0.4359, acc-0.8788, test loss-0.4398, acc-0.8760
Iter-55990, train loss-0.5280, acc-0.8600, valid loss-0.4359, acc-0.8788, test loss-0.4398, acc-0.8759
Iter-56000, train loss-0.5787, acc-0.8000, valid loss-0.4359, acc-0.8788, test loss-0.4397, acc-0.8758
Iter-56010, train loss-0.3403, acc-0.9200, valid loss-0.4358, acc-0.8788, test loss-0.4397, acc-0.8760
Iter-56020, train loss-0.4681, acc-0.8600, valid loss-0.4358, acc-0.8788, test loss-0.4397, acc-0.8760
Iter-56030, train loss-0.3716, acc-0.9000, valid loss-0.4358, acc-0.8790, test loss-0.4397, acc-0.8760
Iter-56040, train loss-0.3826, acc-0.9000, valid loss-0.4357, acc-0.8790, test loss-0.4396, acc-0.8760
Iter-56050, train loss-0.5403, acc-0.8000, valid loss-0.4357, acc-0.8792, test loss-0.4396, acc-0.8760
Iter-56060, train loss-0.3770, acc-0.8800, valid loss-0.4357, acc-0.8792, test loss-0.4395, acc-0.8761
Iter-56070, train loss-0.4883, acc-0.8400, valid loss-0.4356, acc-0.8792, test loss-0.4395, acc-0.8761
Iter-56080, train loss-0.4048, acc-0.9000, valid loss-0.4356, acc-0.8792, test loss-0.4395, acc-0.8761
Iter-56090, train loss-0.5171, acc-0.8400, valid loss-0.4356, acc-0.8790, test loss-0.4395, acc-0.8762
Iter-56100, train loss-0.4664, acc-0.9000, valid loss-0.4355, acc-0.8790, test loss-0.4394, acc-0.8762
Iter-56110, train loss-0.3500, acc-0.9000, valid loss-0.4355, acc-0.8790, test loss-0.4394, acc-0.8762
Iter-56120, train loss-0.3249, acc-0.8800, valid loss-0.4355, acc-0.8792, test loss-0.4394, acc-0.8762
Iter-56130, train loss-0.4734, acc-0.8800, valid loss-0.4354, acc-0.8792, test loss-0.4393, acc-0.8762
Iter-56140, train loss-0.4001, acc-0.8600, valid loss-0.4354, acc-0.8792, test loss-0.4393, acc-0.8761
Iter-56150, train loss-0.5163, acc-0.8600, valid loss-0.4354, acc-0.8792, test loss-0.4392, acc-0.8760
Iter-56160, train loss-0.2520, acc-0.9400, valid loss-0.4353, acc-0.8792, test loss-0.4392, acc-0.8760
Iter-56170, train loss-0.3802, acc-0.9400, valid loss-0.4353, acc-0.8792, test loss-0.4392, acc-0.8760
Iter-56180, train loss-0.3457, acc-0.8800, valid loss-0.4353, acc-0.8792, test loss-0.4391, acc-0.8760
Iter-56190, train loss-0.3793, acc-0.9200, valid loss-0.4352, acc-0.8792, test loss-0.4391, acc-0.8760
Iter-56200, train loss-0.4489, acc-0.8600, valid loss-0.4352, acc-0.8792, test loss-0.4390, acc-0.8761
Iter-56210, train loss-0.4430, acc-0.8800, valid loss-0.4351, acc-0.8792, test loss-0.4390, acc-0.8761
Iter-56220, train loss-0.6851, acc-0.8200, valid loss-0.4351, acc-0.8794, test loss-0.4390, acc-0.8762
Iter-56230, train loss-0.4161, acc-0.8600, valid loss-0.4351, acc-0.8794, test loss-0.4389, acc-0.8762
Iter-56240, train loss-0.4713, acc-0.8600, valid loss-0.4350, acc-0.8794, test loss-0.4389, acc-0.8762
Iter-56250, train loss-0.5896, acc-0.8600, valid loss-0.4350, acc-0.8794, test loss-0.4389, acc-0.8762
Iter-56260, train loss-0.4418, acc-0.8400, valid loss-0.4349, acc-0.8794, test loss-0.4388, acc-0.8763
Iter-56270, train loss-0.6922, acc-0.8400, valid loss-0.4349, acc-0.8794, test loss-0.4388, acc-0.8764
Iter-56280, train loss-0.6019, acc-0.8800, valid loss-0.4349, acc-0.8796, test loss-0.4388, acc-0.8764
Iter-56290, train loss-0.5173, acc-0.8600, valid loss-0.4348, acc-0.8794, test loss-0.4387, acc-0.8763
Iter-56300, train loss-0.2944, acc-0.9000, valid loss-0.4348, acc-0.8794, test loss-0.4387, acc-0.8764
Iter-56310, train loss-0.3905, acc-0.9000, valid loss-0.4348, acc-0.8794, test loss-0.4387, acc-0.8765
Iter-56320, train loss-0.5279, acc-0.8600, valid loss-0.4347, acc-0.8796, test loss-0.4386, acc-0.8766
Iter-56330, train loss-0.4026, acc-0.8600, valid loss-0.4347, acc-0.8796, test loss-0.4386, acc-0.8766
Iter-56340, train loss-0.3539, acc-0.9200, valid loss-0.4346, acc-0.8796, test loss-0.4386, acc-0.8766
Iter-56350, train loss-0.3266, acc-0.8800, valid loss-0.4346, acc-0.8794, test loss-0.4385, acc-0.8765
Iter-56360, train loss-0.4598, acc-0.8400, valid loss-0.4346, acc-0.8796, test loss-0.4385, acc-0.8764
Iter-56370, train loss-0.5320, acc-0.8200, valid loss-0.4345, acc-0.8794, test loss-0.4385, acc-0.8764
Iter-56380, train loss-0.2502, acc-0.9600, valid loss-0.4345, acc-0.8796, test loss-0.4384, acc-0.8765
Iter-56390, train loss-0.3746, acc-0.9000, valid loss-0.4345, acc-0.8794, test loss-0.4384, acc-0.8765
Iter-56400, train loss-0.5392, acc-0.8600, valid loss-0.4344, acc-0.8794, test loss-0.4384, acc-0.8765
Iter-56410, train loss-0.4867, acc-0.8800, valid loss-0.4344, acc-0.8796, test loss-0.4384, acc-0.8765
Iter-56420, train loss-0.3780, acc-0.9000, valid loss-0.4344, acc-0.8796, test loss-0.4383, acc-0.8765
Iter-56430, train loss-0.4266, acc-0.8600, valid loss-0.4343, acc-0.8794, test loss-0.4383, acc-0.8766
Iter-56440, train loss-0.4568, acc-0.8800, valid loss-0.4343, acc-0.8796, test loss-0.4382, acc-0.8765
Iter-56450, train loss-0.4351, acc-0.9400, valid loss-0.4343, acc-0.8796, test loss-0.4382, acc-0.8765
Iter-56460, train loss-0.3814, acc-0.9200, valid loss-0.4343, acc-0.8796, test loss-0.4382, acc-0.8765
Iter-56470, train loss-0.3289, acc-0.9000, valid loss-0.4342, acc-0.8798, test loss-0.4382, acc-0.8765
Iter-56480, train loss-0.4232, acc-0.8800, valid loss-0.4342, acc-0.8798, test loss-0.4381, acc-0.8765
Iter-56490, train loss-0.5324, acc-0.7800, valid loss-0.4342, acc-0.8800, test loss-0.4381, acc-0.8765
Iter-56500, train loss-0.3841, acc-0.9400, valid loss-0.4341, acc-0.8800, test loss-0.4380, acc-0.8765
Iter-56510, train loss-0.4948, acc-0.8400, valid loss-0.4341, acc-0.8802, test loss-0.4380, acc-0.8766
Iter-56520, train loss-0.7074, acc-0.8000, valid loss-0.4341, acc-0.8802, test loss-0.4380, acc-0.8765
Iter-56530, train loss-0.3399, acc-0.8800, valid loss-0.4340, acc-0.8802, test loss-0.4379, acc-0.8765
Iter-56540, train loss-0.4316, acc-0.9000, valid loss-0.4340, acc-0.8800, test loss-0.4379, acc-0.8765
Iter-56550, train loss-0.2773, acc-0.9000, valid loss-0.4339, acc-0.8798, test loss-0.4379, acc-0.8765
Iter-56560, train loss-0.4193, acc-0.9000, valid loss-0.4339, acc-0.8798, test loss-0.4378, acc-0.8765
Iter-56570, train loss-0.6359, acc-0.8000, valid loss-0.4339, acc-0.8800, test loss-0.4378, acc-0.8765
Iter-56580, train loss-0.2665, acc-0.9400, valid loss-0.4338, acc-0.8798, test loss-0.4378, acc-0.8765
Iter-56590, train loss-0.4229, acc-0.8800, valid loss-0.4338, acc-0.8800, test loss-0.4377, acc-0.8765
Iter-56600, train loss-0.4387, acc-0.9000, valid loss-0.4338, acc-0.8802, test loss-0.4377, acc-0.8765
Iter-56610, train loss-0.4813, acc-0.8600, valid loss-0.4337, acc-0.8802, test loss-0.4376, acc-0.8765
Iter-56620, train loss-0.4389, acc-0.9000, valid loss-0.4337, acc-0.8802, test loss-0.4376, acc-0.8765
Iter-56630, train loss-0.4279, acc-0.8600, valid loss-0.4337, acc-0.8798, test loss-0.4376, acc-0.8767
Iter-56640, train loss-0.4490, acc-0.8400, valid loss-0.4336, acc-0.8800, test loss-0.4375, acc-0.8767
Iter-56650, train loss-0.6762, acc-0.8000, valid loss-0.4336, acc-0.8802, test loss-0.4375, acc-0.8766
Iter-56660, train loss-0.3821, acc-0.8800, valid loss-0.4336, acc-0.8802, test loss-0.4374, acc-0.8766
Iter-56670, train loss-0.5809, acc-0.8200, valid loss-0.4336, acc-0.8802, test loss-0.4374, acc-0.8766
Iter-56680, train loss-0.4473, acc-0.9000, valid loss-0.4335, acc-0.8802, test loss-0.4374, acc-0.8765
Iter-56690, train loss-0.3823, acc-0.8800, valid loss-0.4335, acc-0.8802, test loss-0.4374, acc-0.8765
Iter-56700, train loss-0.4108, acc-0.8800, valid loss-0.4334, acc-0.8804, test loss-0.4373, acc-0.8764
Iter-56710, train loss-0.2500, acc-0.9600, valid loss-0.4334, acc-0.8804, test loss-0.4373, acc-0.8764
Iter-56720, train loss-0.4501, acc-0.8600, valid loss-0.4334, acc-0.8804, test loss-0.4373, acc-0.8764
Iter-56730, train loss-0.5131, acc-0.8400, valid loss-0.4333, acc-0.8804, test loss-0.4372, acc-0.8764
Iter-56740, train loss-0.3571, acc-0.9000, valid loss-0.4333, acc-0.8802, test loss-0.4372, acc-0.8765
Iter-56750, train loss-0.4070, acc-0.8800, valid loss-0.4333, acc-0.8800, test loss-0.4372, acc-0.8766
Iter-56760, train loss-0.5652, acc-0.8400, valid loss-0.4332, acc-0.8802, test loss-0.4372, acc-0.8767
Iter-56770, train loss-0.4185, acc-0.9000, valid loss-0.4332, acc-0.8800, test loss-0.4371, acc-0.8767
Iter-56780, train loss-0.4308, acc-0.7800, valid loss-0.4332, acc-0.8800, test loss-0.4371, acc-0.8768
Iter-56790, train loss-0.2598, acc-0.9400, valid loss-0.4331, acc-0.8800, test loss-0.4371, acc-0.8768
Iter-56800, train loss-0.4298, acc-0.8800, valid loss-0.4331, acc-0.8804, test loss-0.4370, acc-0.8766
Iter-56810, train loss-0.2710, acc-0.9600, valid loss-0.4331, acc-0.8806, test loss-0.4370, acc-0.8767
Iter-56820, train loss-0.4503, acc-0.8800, valid loss-0.4330, acc-0.8804, test loss-0.4370, acc-0.8768
Iter-56830, train loss-0.4870, acc-0.8600, valid loss-0.4330, acc-0.8800, test loss-0.4369, acc-0.8768
Iter-56840, train loss-0.4870, acc-0.8600, valid loss-0.4330, acc-0.8806, test loss-0.4369, acc-0.8769
Iter-56850, train loss-0.5737, acc-0.8600, valid loss-0.4329, acc-0.8804, test loss-0.4368, acc-0.8768
Iter-56860, train loss-0.5161, acc-0.8600, valid loss-0.4329, acc-0.8804, test loss-0.4368, acc-0.8768
Iter-56870, train loss-0.3989, acc-0.9200, valid loss-0.4329, acc-0.8804, test loss-0.4368, acc-0.8768
Iter-56880, train loss-0.6148, acc-0.8800, valid loss-0.4328, acc-0.8806, test loss-0.4368, acc-0.8767
Iter-56890, train loss-0.5453, acc-0.9200, valid loss-0.4328, acc-0.8808, test loss-0.4367, acc-0.8769
Iter-56900, train loss-0.3750, acc-0.9000, valid loss-0.4328, acc-0.8804, test loss-0.4367, acc-0.8768
Iter-56910, train loss-0.3627, acc-0.9200, valid loss-0.4328, acc-0.8806, test loss-0.4367, acc-0.8768
Iter-56920, train loss-0.3078, acc-0.9200, valid loss-0.4327, acc-0.8806, test loss-0.4367, acc-0.8769
Iter-56930, train loss-0.5937, acc-0.8200, valid loss-0.4327, acc-0.8804, test loss-0.4366, acc-0.8771
Iter-56940, train loss-0.5976, acc-0.8200, valid loss-0.4326, acc-0.8810, test loss-0.4366, acc-0.8771
Iter-56950, train loss-0.4472, acc-0.8600, valid loss-0.4326, acc-0.8810, test loss-0.4366, acc-0.8772
Iter-56960, train loss-0.5433, acc-0.8400, valid loss-0.4326, acc-0.8808, test loss-0.4365, acc-0.8772
Iter-56970, train loss-0.4195, acc-0.8600, valid loss-0.4325, acc-0.8808, test loss-0.4365, acc-0.8771
Iter-56980, train loss-0.3141, acc-0.9400, valid loss-0.4325, acc-0.8808, test loss-0.4365, acc-0.8771
Iter-56990, train loss-0.4550, acc-0.8800, valid loss-0.4325, acc-0.8808, test loss-0.4364, acc-0.8771
Iter-57000, train loss-0.4836, acc-0.8600, valid loss-0.4324, acc-0.8810, test loss-0.4364, acc-0.8771
Iter-57010, train loss-0.3009, acc-0.9600, valid loss-0.4324, acc-0.8810, test loss-0.4364, acc-0.8771
Iter-57020, train loss-0.4687, acc-0.8800, valid loss-0.4324, acc-0.8810, test loss-0.4363, acc-0.8771
Iter-57030, train loss-0.2438, acc-0.9600, valid loss-0.4323, acc-0.8810, test loss-0.4363, acc-0.8771
Iter-57040, train loss-0.3876, acc-0.9200, valid loss-0.4323, acc-0.8810, test loss-0.4362, acc-0.8771
Iter-57050, train loss-0.3454, acc-0.8800, valid loss-0.4322, acc-0.8810, test loss-0.4362, acc-0.8773
Iter-57060, train loss-0.2717, acc-0.9400, valid loss-0.4322, acc-0.8810, test loss-0.4362, acc-0.8773
Iter-57070, train loss-0.5328, acc-0.8200, valid loss-0.4322, acc-0.8808, test loss-0.4361, acc-0.8772
Iter-57080, train loss-0.4499, acc-0.8800, valid loss-0.4321, acc-0.8808, test loss-0.4361, acc-0.8772
Iter-57090, train loss-0.4775, acc-0.8200, valid loss-0.4321, acc-0.8810, test loss-0.4361, acc-0.8773
Iter-57100, train loss-0.5271, acc-0.8200, valid loss-0.4321, acc-0.8808, test loss-0.4360, acc-0.8773
Iter-57110, train loss-0.4561, acc-0.8800, valid loss-0.4320, acc-0.8810, test loss-0.4360, acc-0.8775
Iter-57120, train loss-0.4267, acc-0.8600, valid loss-0.4320, acc-0.8810, test loss-0.4360, acc-0.8776
Iter-57130, train loss-0.3951, acc-0.9000, valid loss-0.4320, acc-0.8810, test loss-0.4359, acc-0.8774
Iter-57140, train loss-0.5048, acc-0.8800, valid loss-0.4320, acc-0.8812, test loss-0.4359, acc-0.8774
Iter-57150, train loss-0.5398, acc-0.8200, valid loss-0.4319, acc-0.8808, test loss-0.4359, acc-0.8774
Iter-57160, train loss-0.5463, acc-0.8400, valid loss-0.4319, acc-0.8810, test loss-0.4358, acc-0.8775
Iter-57170, train loss-0.3900, acc-0.9000, valid loss-0.4318, acc-0.8808, test loss-0.4358, acc-0.8776
Iter-57180, train loss-0.6039, acc-0.8000, valid loss-0.4318, acc-0.8810, test loss-0.4358, acc-0.8775
Iter-57190, train loss-0.3222, acc-0.9000, valid loss-0.4318, acc-0.8812, test loss-0.4357, acc-0.8775
Iter-57200, train loss-0.4761, acc-0.8800, valid loss-0.4317, acc-0.8808, test loss-0.4357, acc-0.8775
Iter-57210, train loss-0.4310, acc-0.8600, valid loss-0.4317, acc-0.8810, test loss-0.4357, acc-0.8776
Iter-57220, train loss-0.3517, acc-0.9000, valid loss-0.4317, acc-0.8812, test loss-0.4356, acc-0.8776
Iter-57230, train loss-0.2336, acc-0.9400, valid loss-0.4317, acc-0.8814, test loss-0.4356, acc-0.8775
Iter-57240, train loss-0.3790, acc-0.9200, valid loss-0.4316, acc-0.8814, test loss-0.4356, acc-0.8775
Iter-57250, train loss-0.5549, acc-0.8200, valid loss-0.4316, acc-0.8814, test loss-0.4355, acc-0.8776
Iter-57260, train loss-0.7985, acc-0.8200, valid loss-0.4316, acc-0.8814, test loss-0.4355, acc-0.8776
Iter-57270, train loss-0.3945, acc-0.9200, valid loss-0.4315, acc-0.8812, test loss-0.4355, acc-0.8776
Iter-57280, train loss-0.4206, acc-0.8200, valid loss-0.4315, acc-0.8812, test loss-0.4354, acc-0.8776
Iter-57290, train loss-0.4993, acc-0.8600, valid loss-0.4315, acc-0.8812, test loss-0.4354, acc-0.8776
Iter-57300, train loss-0.5061, acc-0.8400, valid loss-0.4314, acc-0.8812, test loss-0.4353, acc-0.8775
Iter-57310, train loss-0.6046, acc-0.8600, valid loss-0.4314, acc-0.8812, test loss-0.4353, acc-0.8775
Iter-57320, train loss-0.5010, acc-0.8400, valid loss-0.4314, acc-0.8812, test loss-0.4353, acc-0.8776
Iter-57330, train loss-0.2157, acc-0.9800, valid loss-0.4313, acc-0.8814, test loss-0.4352, acc-0.8775
Iter-57340, train loss-0.3799, acc-0.9000, valid loss-0.4313, acc-0.8814, test loss-0.4352, acc-0.8775
Iter-57350, train loss-0.5698, acc-0.8000, valid loss-0.4313, acc-0.8814, test loss-0.4352, acc-0.8775
Iter-57360, train loss-0.4084, acc-0.8800, valid loss-0.4313, acc-0.8814, test loss-0.4352, acc-0.8775
Iter-57370, train loss-0.6616, acc-0.8200, valid loss-0.4312, acc-0.8814, test loss-0.4351, acc-0.8776
Iter-57380, train loss-0.3361, acc-0.9000, valid loss-0.4312, acc-0.8814, test loss-0.4351, acc-0.8776
Iter-57390, train loss-0.4343, acc-0.8800, valid loss-0.4311, acc-0.8814, test loss-0.4351, acc-0.8776
Iter-57400, train loss-0.4953, acc-0.8600, valid loss-0.4311, acc-0.8818, test loss-0.4350, acc-0.8776
Iter-57410, train loss-0.7902, acc-0.7800, valid loss-0.4311, acc-0.8818, test loss-0.4350, acc-0.8777
Iter-57420, train loss-0.5792, acc-0.8400, valid loss-0.4311, acc-0.8818, test loss-0.4350, acc-0.8775
Iter-57430, train loss-0.4825, acc-0.8400, valid loss-0.4310, acc-0.8818, test loss-0.4349, acc-0.8775
Iter-57440, train loss-0.3822, acc-0.8600, valid loss-0.4310, acc-0.8818, test loss-0.4349, acc-0.8776
Iter-57450, train loss-0.4015, acc-0.9000, valid loss-0.4309, acc-0.8818, test loss-0.4349, acc-0.8776
Iter-57460, train loss-0.5858, acc-0.8200, valid loss-0.4309, acc-0.8818, test loss-0.4348, acc-0.8776
Iter-57470, train loss-0.4127, acc-0.8800, valid loss-0.4309, acc-0.8818, test loss-0.4348, acc-0.8776
Iter-57480, train loss-0.3738, acc-0.8600, valid loss-0.4308, acc-0.8818, test loss-0.4348, acc-0.8777
Iter-57490, train loss-0.3430, acc-0.9000, valid loss-0.4308, acc-0.8818, test loss-0.4348, acc-0.8776
Iter-57500, train loss-0.3884, acc-0.9200, valid loss-0.4308, acc-0.8820, test loss-0.4347, acc-0.8776
Iter-57510, train loss-0.5907, acc-0.7800, valid loss-0.4307, acc-0.8820, test loss-0.4347, acc-0.8777
Iter-57520, train loss-0.3450, acc-0.9200, valid loss-0.4307, acc-0.8820, test loss-0.4347, acc-0.8778
Iter-57530, train loss-0.3859, acc-0.8800, valid loss-0.4307, acc-0.8818, test loss-0.4346, acc-0.8777
Iter-57540, train loss-0.3965, acc-0.9200, valid loss-0.4306, acc-0.8820, test loss-0.4346, acc-0.8776
Iter-57550, train loss-0.8878, acc-0.7600, valid loss-0.4306, acc-0.8818, test loss-0.4345, acc-0.8778
Iter-57560, train loss-0.5808, acc-0.7800, valid loss-0.4306, acc-0.8818, test loss-0.4345, acc-0.8777
Iter-57570, train loss-0.3975, acc-0.9000, valid loss-0.4305, acc-0.8820, test loss-0.4345, acc-0.8777
Iter-57580, train loss-0.4774, acc-0.8800, valid loss-0.4305, acc-0.8818, test loss-0.4345, acc-0.8777
Iter-57590, train loss-0.4378, acc-0.8800, valid loss-0.4305, acc-0.8818, test loss-0.4344, acc-0.8777
Iter-57600, train loss-0.5740, acc-0.8600, valid loss-0.4304, acc-0.8816, test loss-0.4344, acc-0.8778
Iter-57610, train loss-0.2994, acc-0.9400, valid loss-0.4304, acc-0.8818, test loss-0.4344, acc-0.8776
Iter-57620, train loss-0.4029, acc-0.8800, valid loss-0.4304, acc-0.8818, test loss-0.4343, acc-0.8777
Iter-57630, train loss-0.3313, acc-0.9000, valid loss-0.4303, acc-0.8818, test loss-0.4343, acc-0.8779
Iter-57640, train loss-0.6023, acc-0.8200, valid loss-0.4303, acc-0.8816, test loss-0.4342, acc-0.8781
Iter-57650, train loss-0.5182, acc-0.7800, valid loss-0.4302, acc-0.8816, test loss-0.4342, acc-0.8782
Iter-57660, train loss-0.3872, acc-0.9400, valid loss-0.4302, acc-0.8814, test loss-0.4341, acc-0.8781
Iter-57670, train loss-0.3219, acc-0.9400, valid loss-0.4302, acc-0.8814, test loss-0.4341, acc-0.8778
Iter-57680, train loss-0.7064, acc-0.8400, valid loss-0.4301, acc-0.8816, test loss-0.4341, acc-0.8778
Iter-57690, train loss-0.6662, acc-0.8200, valid loss-0.4301, acc-0.8816, test loss-0.4340, acc-0.8779
Iter-57700, train loss-0.6472, acc-0.8000, valid loss-0.4301, acc-0.8818, test loss-0.4340, acc-0.8780
Iter-57710, train loss-0.3872, acc-0.9200, valid loss-0.4300, acc-0.8816, test loss-0.4340, acc-0.8780
Iter-57720, train loss-0.8240, acc-0.8200, valid loss-0.4300, acc-0.8816, test loss-0.4339, acc-0.8778
Iter-57730, train loss-0.4515, acc-0.8400, valid loss-0.4300, acc-0.8816, test loss-0.4339, acc-0.8778
Iter-57740, train loss-0.3940, acc-0.8800, valid loss-0.4299, acc-0.8816, test loss-0.4339, acc-0.8779
Iter-57750, train loss-0.5867, acc-0.8600, valid loss-0.4299, acc-0.8816, test loss-0.4338, acc-0.8779
Iter-57760, train loss-0.3799, acc-0.8600, valid loss-0.4299, acc-0.8818, test loss-0.4338, acc-0.8780
Iter-57770, train loss-0.2698, acc-0.9400, valid loss-0.4298, acc-0.8818, test loss-0.4338, acc-0.8779
Iter-57780, train loss-0.3401, acc-0.9000, valid loss-0.4298, acc-0.8818, test loss-0.4337, acc-0.8779
Iter-57790, train loss-0.3292, acc-0.9200, valid loss-0.4298, acc-0.8818, test loss-0.4337, acc-0.8780
Iter-57800, train loss-0.3511, acc-0.9600, valid loss-0.4297, acc-0.8818, test loss-0.4337, acc-0.8779
Iter-57810, train loss-0.2876, acc-0.9400, valid loss-0.4297, acc-0.8820, test loss-0.4336, acc-0.8779
Iter-57820, train loss-0.4007, acc-0.8400, valid loss-0.4297, acc-0.8820, test loss-0.4336, acc-0.8778
Iter-57830, train loss-0.4137, acc-0.9000, valid loss-0.4296, acc-0.8820, test loss-0.4336, acc-0.8778
Iter-57840, train loss-0.4370, acc-0.8000, valid loss-0.4296, acc-0.8820, test loss-0.4335, acc-0.8778
Iter-57850, train loss-0.5606, acc-0.8200, valid loss-0.4296, acc-0.8820, test loss-0.4335, acc-0.8778
Iter-57860, train loss-0.4298, acc-0.8600, valid loss-0.4295, acc-0.8818, test loss-0.4335, acc-0.8777
Iter-57870, train loss-0.3822, acc-0.9400, valid loss-0.4295, acc-0.8820, test loss-0.4334, acc-0.8778
Iter-57880, train loss-0.4416, acc-0.9000, valid loss-0.4294, acc-0.8820, test loss-0.4334, acc-0.8778
Iter-57890, train loss-0.4659, acc-0.8600, valid loss-0.4294, acc-0.8820, test loss-0.4334, acc-0.8777
Iter-57900, train loss-0.4524, acc-0.8400, valid loss-0.4293, acc-0.8818, test loss-0.4333, acc-0.8778
Iter-57910, train loss-0.3782, acc-0.9200, valid loss-0.4293, acc-0.8818, test loss-0.4333, acc-0.8780
Iter-57920, train loss-0.2490, acc-0.9400, valid loss-0.4293, acc-0.8818, test loss-0.4333, acc-0.8780
Iter-57930, train loss-0.6112, acc-0.8800, valid loss-0.4292, acc-0.8816, test loss-0.4332, acc-0.8781
Iter-57940, train loss-0.5454, acc-0.8200, valid loss-0.4292, acc-0.8818, test loss-0.4332, acc-0.8781
Iter-57950, train loss-0.5083, acc-0.8400, valid loss-0.4292, acc-0.8818, test loss-0.4332, acc-0.8780
Iter-57960, train loss-0.3626, acc-0.8800, valid loss-0.4291, acc-0.8818, test loss-0.4331, acc-0.8781
Iter-57970, train loss-0.4744, acc-0.9200, valid loss-0.4291, acc-0.8818, test loss-0.4331, acc-0.8779
Iter-57980, train loss-0.5676, acc-0.8600, valid loss-0.4291, acc-0.8818, test loss-0.4331, acc-0.8780
Iter-57990, train loss-0.6294, acc-0.8000, valid loss-0.4291, acc-0.8818, test loss-0.4330, acc-0.8780
Iter-58000, train loss-0.3404, acc-0.9000, valid loss-0.4291, acc-0.8818, test loss-0.4330, acc-0.8781
Iter-58010, train loss-0.5684, acc-0.7800, valid loss-0.4290, acc-0.8818, test loss-0.4330, acc-0.8781
Iter-58020, train loss-0.5148, acc-0.8800, valid loss-0.4290, acc-0.8818, test loss-0.4330, acc-0.8781
Iter-58030, train loss-0.5819, acc-0.8600, valid loss-0.4290, acc-0.8814, test loss-0.4329, acc-0.8781
Iter-58040, train loss-0.3843, acc-0.9000, valid loss-0.4289, acc-0.8818, test loss-0.4329, acc-0.8781
Iter-58050, train loss-0.4724, acc-0.8600, valid loss-0.4289, acc-0.8816, test loss-0.4329, acc-0.8781
Iter-58060, train loss-0.5099, acc-0.8600, valid loss-0.4289, acc-0.8820, test loss-0.4328, acc-0.8780
Iter-58070, train loss-0.5730, acc-0.7800, valid loss-0.4288, acc-0.8818, test loss-0.4328, acc-0.8780
Iter-58080, train loss-0.4083, acc-0.8600, valid loss-0.4288, acc-0.8816, test loss-0.4328, acc-0.8781
Iter-58090, train loss-0.4391, acc-0.8800, valid loss-0.4288, acc-0.8816, test loss-0.4327, acc-0.8781
Iter-58100, train loss-0.4705, acc-0.9200, valid loss-0.4287, acc-0.8816, test loss-0.4327, acc-0.8782
Iter-58110, train loss-0.2890, acc-0.9200, valid loss-0.4287, acc-0.8816, test loss-0.4327, acc-0.8782
Iter-58120, train loss-0.3354, acc-0.9000, valid loss-0.4287, acc-0.8818, test loss-0.4326, acc-0.8782
Iter-58130, train loss-0.5078, acc-0.8600, valid loss-0.4286, acc-0.8818, test loss-0.4326, acc-0.8783
Iter-58140, train loss-0.3295, acc-0.8800, valid loss-0.4286, acc-0.8818, test loss-0.4326, acc-0.8783
Iter-58150, train loss-0.6570, acc-0.8000, valid loss-0.4285, acc-0.8820, test loss-0.4326, acc-0.8783
Iter-58160, train loss-0.7048, acc-0.7600, valid loss-0.4285, acc-0.8820, test loss-0.4325, acc-0.8782
Iter-58170, train loss-0.6416, acc-0.8400, valid loss-0.4285, acc-0.8820, test loss-0.4325, acc-0.8783
Iter-58180, train loss-0.3946, acc-0.9000, valid loss-0.4284, acc-0.8820, test loss-0.4325, acc-0.8784
Iter-58190, train loss-0.3674, acc-0.8800, valid loss-0.4284, acc-0.8818, test loss-0.4324, acc-0.8784
Iter-58200, train loss-0.2964, acc-0.9600, valid loss-0.4284, acc-0.8820, test loss-0.4324, acc-0.8785
Iter-58210, train loss-0.4228, acc-0.9200, valid loss-0.4283, acc-0.8820, test loss-0.4324, acc-0.8784
Iter-58220, train loss-0.4525, acc-0.9000, valid loss-0.4283, acc-0.8820, test loss-0.4323, acc-0.8784
Iter-58230, train loss-0.3812, acc-0.8800, valid loss-0.4283, acc-0.8820, test loss-0.4323, acc-0.8784
Iter-58240, train loss-0.2778, acc-0.9200, valid loss-0.4283, acc-0.8822, test loss-0.4323, acc-0.8784
Iter-58250, train loss-0.3907, acc-0.9200, valid loss-0.4282, acc-0.8822, test loss-0.4323, acc-0.8784
Iter-58260, train loss-0.5657, acc-0.8400, valid loss-0.4282, acc-0.8822, test loss-0.4322, acc-0.8784
Iter-58270, train loss-0.4959, acc-0.8800, valid loss-0.4282, acc-0.8820, test loss-0.4322, acc-0.8785
Iter-58280, train loss-0.5801, acc-0.8000, valid loss-0.4281, acc-0.8822, test loss-0.4322, acc-0.8785
Iter-58290, train loss-0.4776, acc-0.9000, valid loss-0.4281, acc-0.8820, test loss-0.4321, acc-0.8786
Iter-58300, train loss-0.3678, acc-0.9200, valid loss-0.4281, acc-0.8822, test loss-0.4321, acc-0.8786
Iter-58310, train loss-0.6032, acc-0.8400, valid loss-0.4280, acc-0.8822, test loss-0.4321, acc-0.8786
Iter-58320, train loss-0.3923, acc-0.8800, valid loss-0.4280, acc-0.8822, test loss-0.4320, acc-0.8786
Iter-58330, train loss-0.3988, acc-0.9000, valid loss-0.4280, acc-0.8822, test loss-0.4320, acc-0.8786
Iter-58340, train loss-0.6394, acc-0.8400, valid loss-0.4280, acc-0.8822, test loss-0.4320, acc-0.8786
Iter-58350, train loss-0.4873, acc-0.8200, valid loss-0.4279, acc-0.8822, test loss-0.4319, acc-0.8787
Iter-58360, train loss-0.3159, acc-0.9000, valid loss-0.4279, acc-0.8822, test loss-0.4319, acc-0.8787
Iter-58370, train loss-0.4759, acc-0.8200, valid loss-0.4279, acc-0.8822, test loss-0.4319, acc-0.8787
Iter-58380, train loss-0.6257, acc-0.8400, valid loss-0.4278, acc-0.8822, test loss-0.4318, acc-0.8787
Iter-58390, train loss-0.4463, acc-0.8800, valid loss-0.4278, acc-0.8822, test loss-0.4318, acc-0.8788
Iter-58400, train loss-0.6059, acc-0.8600, valid loss-0.4278, acc-0.8822, test loss-0.4318, acc-0.8788
Iter-58410, train loss-0.3492, acc-0.9400, valid loss-0.4278, acc-0.8822, test loss-0.4317, acc-0.8788
Iter-58420, train loss-0.4212, acc-0.8400, valid loss-0.4277, acc-0.8822, test loss-0.4317, acc-0.8788
Iter-58430, train loss-0.3227, acc-0.9200, valid loss-0.4277, acc-0.8822, test loss-0.4317, acc-0.8788
Iter-58440, train loss-0.4268, acc-0.8800, valid loss-0.4277, acc-0.8822, test loss-0.4316, acc-0.8788
Iter-58450, train loss-0.3464, acc-0.9000, valid loss-0.4276, acc-0.8822, test loss-0.4316, acc-0.8788
Iter-58460, train loss-0.2745, acc-0.9200, valid loss-0.4276, acc-0.8822, test loss-0.4316, acc-0.8787
Iter-58470, train loss-0.4382, acc-0.8600, valid loss-0.4276, acc-0.8822, test loss-0.4315, acc-0.8787
Iter-58480, train loss-0.3731, acc-0.9000, valid loss-0.4275, acc-0.8822, test loss-0.4315, acc-0.8787
Iter-58490, train loss-0.5663, acc-0.8400, valid loss-0.4275, acc-0.8822, test loss-0.4315, acc-0.8787
Iter-58500, train loss-0.7541, acc-0.7400, valid loss-0.4275, acc-0.8822, test loss-0.4314, acc-0.8786
Iter-58510, train loss-0.4719, acc-0.8600, valid loss-0.4274, acc-0.8822, test loss-0.4314, acc-0.8786
Iter-58520, train loss-0.4534, acc-0.9200, valid loss-0.4274, acc-0.8822, test loss-0.4314, acc-0.8786
Iter-58530, train loss-0.3508, acc-0.8800, valid loss-0.4274, acc-0.8822, test loss-0.4313, acc-0.8786
Iter-58540, train loss-0.7001, acc-0.7600, valid loss-0.4273, acc-0.8820, test loss-0.4313, acc-0.8787
Iter-58550, train loss-0.3198, acc-0.9200, valid loss-0.4273, acc-0.8820, test loss-0.4313, acc-0.8787
Iter-58560, train loss-0.4727, acc-0.8800, valid loss-0.4273, acc-0.8822, test loss-0.4312, acc-0.8787
Iter-58570, train loss-0.3446, acc-0.9600, valid loss-0.4272, acc-0.8822, test loss-0.4312, acc-0.8787
Iter-58580, train loss-0.3913, acc-0.8600, valid loss-0.4272, acc-0.8824, test loss-0.4312, acc-0.8786
Iter-58590, train loss-0.5171, acc-0.8400, valid loss-0.4272, acc-0.8824, test loss-0.4311, acc-0.8787
Iter-58600, train loss-0.5797, acc-0.8600, valid loss-0.4271, acc-0.8824, test loss-0.4311, acc-0.8787
Iter-58610, train loss-0.4191, acc-0.9200, valid loss-0.4271, acc-0.8824, test loss-0.4311, acc-0.8788
Iter-58620, train loss-0.3507, acc-0.9000, valid loss-0.4271, acc-0.8824, test loss-0.4310, acc-0.8790
Iter-58630, train loss-0.2958, acc-0.9600, valid loss-0.4270, acc-0.8824, test loss-0.4310, acc-0.8788
Iter-58640, train loss-0.3669, acc-0.9200, valid loss-0.4270, acc-0.8824, test loss-0.4310, acc-0.8787
Iter-58650, train loss-0.2678, acc-0.9000, valid loss-0.4270, acc-0.8824, test loss-0.4309, acc-0.8787
Iter-58660, train loss-0.4613, acc-0.8400, valid loss-0.4270, acc-0.8824, test loss-0.4309, acc-0.8786
Iter-58670, train loss-0.5003, acc-0.8800, valid loss-0.4269, acc-0.8824, test loss-0.4309, acc-0.8789
Iter-58680, train loss-0.3846, acc-0.8400, valid loss-0.4269, acc-0.8824, test loss-0.4309, acc-0.8788
Iter-58690, train loss-0.5606, acc-0.8400, valid loss-0.4268, acc-0.8824, test loss-0.4308, acc-0.8791
Iter-58700, train loss-0.8224, acc-0.7600, valid loss-0.4268, acc-0.8824, test loss-0.4308, acc-0.8791
Iter-58710, train loss-0.4769, acc-0.8800, valid loss-0.4268, acc-0.8824, test loss-0.4308, acc-0.8790
Iter-58720, train loss-0.3805, acc-0.8600, valid loss-0.4267, acc-0.8824, test loss-0.4307, acc-0.8790
Iter-58730, train loss-0.6861, acc-0.8200, valid loss-0.4267, acc-0.8824, test loss-0.4307, acc-0.8791
Iter-58740, train loss-0.5860, acc-0.8200, valid loss-0.4267, acc-0.8822, test loss-0.4307, acc-0.8790
Iter-58750, train loss-0.3916, acc-0.9200, valid loss-0.4266, acc-0.8822, test loss-0.4306, acc-0.8789
Iter-58760, train loss-0.3395, acc-0.9000, valid loss-0.4266, acc-0.8822, test loss-0.4306, acc-0.8790
Iter-58770, train loss-0.2234, acc-0.9600, valid loss-0.4266, acc-0.8822, test loss-0.4306, acc-0.8790
Iter-58780, train loss-0.3968, acc-0.8600, valid loss-0.4266, acc-0.8820, test loss-0.4305, acc-0.8790
Iter-58790, train loss-0.4575, acc-0.8600, valid loss-0.4265, acc-0.8822, test loss-0.4305, acc-0.8790
Iter-58800, train loss-0.6682, acc-0.7800, valid loss-0.4265, acc-0.8822, test loss-0.4305, acc-0.8792
Iter-58810, train loss-0.4995, acc-0.8600, valid loss-0.4265, acc-0.8822, test loss-0.4304, acc-0.8789
Iter-58820, train loss-0.5227, acc-0.8000, valid loss-0.4265, acc-0.8820, test loss-0.4304, acc-0.8790
Iter-58830, train loss-0.3318, acc-0.8800, valid loss-0.4264, acc-0.8822, test loss-0.4304, acc-0.8790
Iter-58840, train loss-0.3974, acc-0.8400, valid loss-0.4264, acc-0.8820, test loss-0.4303, acc-0.8790
Iter-58850, train loss-0.4356, acc-0.9000, valid loss-0.4263, acc-0.8822, test loss-0.4303, acc-0.8791
Iter-58860, train loss-0.3318, acc-0.9200, valid loss-0.4263, acc-0.8822, test loss-0.4303, acc-0.8792
Iter-58870, train loss-0.4551, acc-0.8600, valid loss-0.4263, acc-0.8822, test loss-0.4302, acc-0.8792
Iter-58880, train loss-0.5211, acc-0.8400, valid loss-0.4262, acc-0.8822, test loss-0.4302, acc-0.8793
Iter-58890, train loss-0.5219, acc-0.8600, valid loss-0.4262, acc-0.8822, test loss-0.4301, acc-0.8793
Iter-58900, train loss-0.3167, acc-0.9200, valid loss-0.4262, acc-0.8824, test loss-0.4301, acc-0.8795
Iter-58910, train loss-0.2976, acc-0.9400, valid loss-0.4261, acc-0.8824, test loss-0.4301, acc-0.8795
Iter-58920, train loss-0.3613, acc-0.9000, valid loss-0.4261, acc-0.8824, test loss-0.4300, acc-0.8794
Iter-58930, train loss-0.3916, acc-0.9200, valid loss-0.4261, acc-0.8824, test loss-0.4300, acc-0.8794
Iter-58940, train loss-0.5478, acc-0.8400, valid loss-0.4261, acc-0.8824, test loss-0.4300, acc-0.8796
Iter-58950, train loss-0.4006, acc-0.8600, valid loss-0.4260, acc-0.8824, test loss-0.4299, acc-0.8796
Iter-58960, train loss-0.4389, acc-0.8600, valid loss-0.4260, acc-0.8826, test loss-0.4299, acc-0.8795
Iter-58970, train loss-0.3433, acc-0.9200, valid loss-0.4260, acc-0.8826, test loss-0.4299, acc-0.8796
Iter-58980, train loss-0.8206, acc-0.7200, valid loss-0.4259, acc-0.8826, test loss-0.4298, acc-0.8794
Iter-58990, train loss-0.5042, acc-0.8400, valid loss-0.4259, acc-0.8826, test loss-0.4298, acc-0.8794
Iter-59000, train loss-0.4392, acc-0.8800, valid loss-0.4259, acc-0.8824, test loss-0.4298, acc-0.8796
Iter-59010, train loss-0.4147, acc-0.8800, valid loss-0.4258, acc-0.8826, test loss-0.4297, acc-0.8796
Iter-59020, train loss-0.5460, acc-0.8400, valid loss-0.4258, acc-0.8826, test loss-0.4297, acc-0.8795
Iter-59030, train loss-0.3794, acc-0.9200, valid loss-0.4258, acc-0.8824, test loss-0.4297, acc-0.8796
Iter-59040, train loss-0.4379, acc-0.9000, valid loss-0.4257, acc-0.8824, test loss-0.4296, acc-0.8797
Iter-59050, train loss-0.3788, acc-0.9000, valid loss-0.4257, acc-0.8824, test loss-0.4296, acc-0.8796
Iter-59060, train loss-0.5982, acc-0.8200, valid loss-0.4257, acc-0.8824, test loss-0.4296, acc-0.8796
Iter-59070, train loss-0.3721, acc-0.9200, valid loss-0.4256, acc-0.8822, test loss-0.4295, acc-0.8796
Iter-59080, train loss-0.3955, acc-0.9200, valid loss-0.4256, acc-0.8822, test loss-0.4295, acc-0.8795
Iter-59090, train loss-0.6073, acc-0.8200, valid loss-0.4256, acc-0.8822, test loss-0.4295, acc-0.8796
Iter-59100, train loss-0.4364, acc-0.8800, valid loss-0.4256, acc-0.8826, test loss-0.4294, acc-0.8798
Iter-59110, train loss-0.5314, acc-0.8600, valid loss-0.4255, acc-0.8824, test loss-0.4294, acc-0.8799
Iter-59120, train loss-0.4115, acc-0.9000, valid loss-0.4255, acc-0.8828, test loss-0.4294, acc-0.8797
Iter-59130, train loss-0.3141, acc-0.9400, valid loss-0.4255, acc-0.8828, test loss-0.4293, acc-0.8798
Iter-59140, train loss-0.4886, acc-0.8600, valid loss-0.4254, acc-0.8828, test loss-0.4293, acc-0.8798
Iter-59150, train loss-0.4494, acc-0.8400, valid loss-0.4254, acc-0.8828, test loss-0.4293, acc-0.8797
Iter-59160, train loss-0.3636, acc-0.9200, valid loss-0.4254, acc-0.8826, test loss-0.4292, acc-0.8799
Iter-59170, train loss-0.3453, acc-0.9400, valid loss-0.4253, acc-0.8826, test loss-0.4292, acc-0.8799
Iter-59180, train loss-0.3312, acc-0.9200, valid loss-0.4253, acc-0.8826, test loss-0.4292, acc-0.8799
Iter-59190, train loss-0.4414, acc-0.8800, valid loss-0.4253, acc-0.8824, test loss-0.4291, acc-0.8800
Iter-59200, train loss-0.3659, acc-0.9200, valid loss-0.4252, acc-0.8826, test loss-0.4291, acc-0.8800
Iter-59210, train loss-0.3448, acc-0.8600, valid loss-0.4252, acc-0.8826, test loss-0.4291, acc-0.8800
Iter-59220, train loss-0.4105, acc-0.8800, valid loss-0.4252, acc-0.8826, test loss-0.4291, acc-0.8798
Iter-59230, train loss-0.3323, acc-0.9400, valid loss-0.4252, acc-0.8824, test loss-0.4290, acc-0.8798
Iter-59240, train loss-0.3511, acc-0.8600, valid loss-0.4251, acc-0.8824, test loss-0.4290, acc-0.8797
Iter-59250, train loss-0.4562, acc-0.8800, valid loss-0.4251, acc-0.8824, test loss-0.4290, acc-0.8797
Iter-59260, train loss-0.5934, acc-0.8600, valid loss-0.4251, acc-0.8824, test loss-0.4290, acc-0.8798
Iter-59270, train loss-0.3723, acc-0.8800, valid loss-0.4251, acc-0.8822, test loss-0.4289, acc-0.8799
Iter-59280, train loss-0.7186, acc-0.8000, valid loss-0.4250, acc-0.8822, test loss-0.4289, acc-0.8798
Iter-59290, train loss-0.3669, acc-0.9400, valid loss-0.4250, acc-0.8824, test loss-0.4289, acc-0.8798
Iter-59300, train loss-0.2933, acc-0.9400, valid loss-0.4250, acc-0.8822, test loss-0.4288, acc-0.8798
Iter-59310, train loss-0.3132, acc-0.9600, valid loss-0.4249, acc-0.8822, test loss-0.4288, acc-0.8798
Iter-59320, train loss-0.4193, acc-0.9000, valid loss-0.4249, acc-0.8822, test loss-0.4288, acc-0.8800
Iter-59330, train loss-0.3086, acc-0.9200, valid loss-0.4249, acc-0.8822, test loss-0.4287, acc-0.8800
Iter-59340, train loss-0.4048, acc-0.8800, valid loss-0.4248, acc-0.8822, test loss-0.4287, acc-0.8800
Iter-59350, train loss-0.3830, acc-0.9000, valid loss-0.4248, acc-0.8822, test loss-0.4287, acc-0.8799
Iter-59360, train loss-0.4917, acc-0.8600, valid loss-0.4248, acc-0.8822, test loss-0.4286, acc-0.8801
Iter-59370, train loss-0.5349, acc-0.8600, valid loss-0.4247, acc-0.8822, test loss-0.4286, acc-0.8801
Iter-59380, train loss-0.3519, acc-0.9400, valid loss-0.4247, acc-0.8822, test loss-0.4286, acc-0.8800
Iter-59390, train loss-0.3753, acc-0.9000, valid loss-0.4247, acc-0.8822, test loss-0.4286, acc-0.8799
Iter-59400, train loss-0.6362, acc-0.8600, valid loss-0.4246, acc-0.8822, test loss-0.4285, acc-0.8799
Iter-59410, train loss-0.5079, acc-0.8000, valid loss-0.4246, acc-0.8824, test loss-0.4285, acc-0.8799
Iter-59420, train loss-0.7180, acc-0.8400, valid loss-0.4246, acc-0.8820, test loss-0.4285, acc-0.8802
Iter-59430, train loss-0.2330, acc-0.9600, valid loss-0.4246, acc-0.8820, test loss-0.4284, acc-0.8801
Iter-59440, train loss-0.4313, acc-0.8600, valid loss-0.4245, acc-0.8820, test loss-0.4284, acc-0.8802
Iter-59450, train loss-0.2606, acc-0.9800, valid loss-0.4245, acc-0.8820, test loss-0.4284, acc-0.8802
Iter-59460, train loss-0.4002, acc-0.9000, valid loss-0.4245, acc-0.8822, test loss-0.4283, acc-0.8800
Iter-59470, train loss-0.6467, acc-0.8200, valid loss-0.4245, acc-0.8822, test loss-0.4283, acc-0.8803
Iter-59480, train loss-0.4176, acc-0.9000, valid loss-0.4244, acc-0.8822, test loss-0.4283, acc-0.8802
Iter-59490, train loss-0.3543, acc-0.8800, valid loss-0.4244, acc-0.8822, test loss-0.4283, acc-0.8801
Iter-59500, train loss-0.5363, acc-0.8200, valid loss-0.4244, acc-0.8822, test loss-0.4282, acc-0.8801
Iter-59510, train loss-0.3066, acc-0.9200, valid loss-0.4243, acc-0.8822, test loss-0.4282, acc-0.8801
Iter-59520, train loss-0.2563, acc-0.9400, valid loss-0.4243, acc-0.8820, test loss-0.4282, acc-0.8803
Iter-59530, train loss-0.6995, acc-0.7600, valid loss-0.4243, acc-0.8820, test loss-0.4281, acc-0.8803
Iter-59540, train loss-0.3351, acc-0.8600, valid loss-0.4243, acc-0.8820, test loss-0.4281, acc-0.8803
Iter-59550, train loss-0.3246, acc-0.9400, valid loss-0.4242, acc-0.8820, test loss-0.4281, acc-0.8801
Iter-59560, train loss-0.6171, acc-0.7800, valid loss-0.4242, acc-0.8820, test loss-0.4280, acc-0.8802
Iter-59570, train loss-0.5825, acc-0.8200, valid loss-0.4242, acc-0.8822, test loss-0.4280, acc-0.8804
Iter-59580, train loss-0.6418, acc-0.8000, valid loss-0.4241, acc-0.8820, test loss-0.4280, acc-0.8804
Iter-59590, train loss-0.4743, acc-0.8600, valid loss-0.4241, acc-0.8820, test loss-0.4279, acc-0.8804
Iter-59600, train loss-0.3603, acc-0.9000, valid loss-0.4241, acc-0.8820, test loss-0.4279, acc-0.8804
Iter-59610, train loss-0.3254, acc-0.9000, valid loss-0.4241, acc-0.8820, test loss-0.4279, acc-0.8805
Iter-59620, train loss-0.2720, acc-0.9400, valid loss-0.4240, acc-0.8820, test loss-0.4279, acc-0.8805
Iter-59630, train loss-0.2042, acc-0.9800, valid loss-0.4240, acc-0.8822, test loss-0.4278, acc-0.8805
Iter-59640, train loss-0.2908, acc-0.9200, valid loss-0.4240, acc-0.8822, test loss-0.4278, acc-0.8805
Iter-59650, train loss-0.3917, acc-0.9200, valid loss-0.4239, acc-0.8822, test loss-0.4278, acc-0.8804
Iter-59660, train loss-0.3313, acc-0.8800, valid loss-0.4239, acc-0.8822, test loss-0.4277, acc-0.8804
Iter-59670, train loss-0.2373, acc-0.9600, valid loss-0.4239, acc-0.8822, test loss-0.4277, acc-0.8804
Iter-59680, train loss-0.4239, acc-0.8800, valid loss-0.4238, acc-0.8822, test loss-0.4277, acc-0.8805
Iter-59690, train loss-0.4084, acc-0.8800, valid loss-0.4238, acc-0.8822, test loss-0.4277, acc-0.8806
Iter-59700, train loss-0.2713, acc-0.9600, valid loss-0.4238, acc-0.8822, test loss-0.4276, acc-0.8806
Iter-59710, train loss-0.6537, acc-0.8000, valid loss-0.4238, acc-0.8822, test loss-0.4276, acc-0.8808
Iter-59720, train loss-0.4328, acc-0.9000, valid loss-0.4237, acc-0.8822, test loss-0.4276, acc-0.8807
Iter-59730, train loss-0.3587, acc-0.9400, valid loss-0.4237, acc-0.8822, test loss-0.4276, acc-0.8807
Iter-59740, train loss-0.4289, acc-0.9000, valid loss-0.4237, acc-0.8822, test loss-0.4275, acc-0.8807
Iter-59750, train loss-0.4121, acc-0.8800, valid loss-0.4237, acc-0.8822, test loss-0.4275, acc-0.8807
Iter-59760, train loss-0.2915, acc-0.9200, valid loss-0.4236, acc-0.8822, test loss-0.4275, acc-0.8806
Iter-59770, train loss-0.5375, acc-0.8800, valid loss-0.4236, acc-0.8822, test loss-0.4274, acc-0.8806
Iter-59780, train loss-0.4821, acc-0.8600, valid loss-0.4236, acc-0.8822, test loss-0.4274, acc-0.8807
Iter-59790, train loss-0.5074, acc-0.8800, valid loss-0.4236, acc-0.8822, test loss-0.4274, acc-0.8807
Iter-59800, train loss-0.5918, acc-0.8200, valid loss-0.4235, acc-0.8822, test loss-0.4274, acc-0.8807
Iter-59810, train loss-0.3921, acc-0.8800, valid loss-0.4235, acc-0.8822, test loss-0.4273, acc-0.8807
Iter-59820, train loss-0.3225, acc-0.9400, valid loss-0.4235, acc-0.8820, test loss-0.4273, acc-0.8807
Iter-59830, train loss-0.4481, acc-0.8600, valid loss-0.4234, acc-0.8820, test loss-0.4273, acc-0.8807
Iter-59840, train loss-0.5822, acc-0.8200, valid loss-0.4234, acc-0.8820, test loss-0.4272, acc-0.8806
Iter-59850, train loss-0.3691, acc-0.9000, valid loss-0.4233, acc-0.8820, test loss-0.4272, acc-0.8806
Iter-59860, train loss-0.3384, acc-0.9200, valid loss-0.4233, acc-0.8820, test loss-0.4272, acc-0.8807
Iter-59870, train loss-0.4732, acc-0.8400, valid loss-0.4233, acc-0.8820, test loss-0.4271, acc-0.8808
Iter-59880, train loss-0.2577, acc-0.9600, valid loss-0.4233, acc-0.8820, test loss-0.4271, acc-0.8806
Iter-59890, train loss-0.4269, acc-0.9000, valid loss-0.4232, acc-0.8820, test loss-0.4271, acc-0.8806
Iter-59900, train loss-0.2422, acc-0.9600, valid loss-0.4232, acc-0.8820, test loss-0.4270, acc-0.8808
Iter-59910, train loss-0.3226, acc-0.8600, valid loss-0.4232, acc-0.8820, test loss-0.4270, acc-0.8807
Iter-59920, train loss-0.3476, acc-0.9000, valid loss-0.4231, acc-0.8820, test loss-0.4269, acc-0.8807
Iter-59930, train loss-0.3589, acc-0.9000, valid loss-0.4231, acc-0.8820, test loss-0.4269, acc-0.8807
Iter-59940, train loss-0.4741, acc-0.8800, valid loss-0.4231, acc-0.8822, test loss-0.4269, acc-0.8808
Iter-59950, train loss-0.4414, acc-0.8400, valid loss-0.4230, acc-0.8822, test loss-0.4268, acc-0.8806
Iter-59960, train loss-0.5201, acc-0.8800, valid loss-0.4230, acc-0.8822, test loss-0.4268, acc-0.8807
Iter-59970, train loss-0.4075, acc-0.8800, valid loss-0.4230, acc-0.8822, test loss-0.4268, acc-0.8807
Iter-59980, train loss-0.3775, acc-0.8800, valid loss-0.4229, acc-0.8822, test loss-0.4267, acc-0.8806
Iter-59990, train loss-0.3657, acc-0.9000, valid loss-0.4229, acc-0.8822, test loss-0.4267, acc-0.8807
Iter-60000, train loss-0.7260, acc-0.8000, valid loss-0.4229, acc-0.8822, test loss-0.4267, acc-0.8806
Iter-60010, train loss-0.4101, acc-0.8400, valid loss-0.4229, acc-0.8822, test loss-0.4267, acc-0.8807
Iter-60020, train loss-0.2916, acc-0.9400, valid loss-0.4228, acc-0.8822, test loss-0.4266, acc-0.8807
Iter-60030, train loss-0.4750, acc-0.8000, valid loss-0.4228, acc-0.8822, test loss-0.4266, acc-0.8807
Iter-60040, train loss-0.3715, acc-0.8800, valid loss-0.4228, acc-0.8822, test loss-0.4266, acc-0.8808
Iter-60050, train loss-0.4847, acc-0.8600, valid loss-0.4227, acc-0.8822, test loss-0.4265, acc-0.8808
Iter-60060, train loss-0.4326, acc-0.9000, valid loss-0.4227, acc-0.8824, test loss-0.4265, acc-0.8807
Iter-60070, train loss-0.5487, acc-0.7800, valid loss-0.4227, acc-0.8824, test loss-0.4265, acc-0.8807
Iter-60080, train loss-0.5284, acc-0.8200, valid loss-0.4226, acc-0.8824, test loss-0.4265, acc-0.8807
Iter-60090, train loss-0.3145, acc-0.9000, valid loss-0.4226, acc-0.8824, test loss-0.4264, acc-0.8809
Iter-60100, train loss-0.4238, acc-0.8800, valid loss-0.4226, acc-0.8824, test loss-0.4264, acc-0.8809
Iter-60110, train loss-0.4036, acc-0.8600, valid loss-0.4225, acc-0.8826, test loss-0.4264, acc-0.8809
Iter-60120, train loss-0.4348, acc-0.8600, valid loss-0.4225, acc-0.8826, test loss-0.4263, acc-0.8810
Iter-60130, train loss-0.5320, acc-0.8600, valid loss-0.4225, acc-0.8826, test loss-0.4263, acc-0.8809
Iter-60140, train loss-0.4768, acc-0.8600, valid loss-0.4224, acc-0.8826, test loss-0.4263, acc-0.8809
Iter-60150, train loss-0.2002, acc-0.9800, valid loss-0.4224, acc-0.8826, test loss-0.4262, acc-0.8809
Iter-60160, train loss-0.5251, acc-0.8600, valid loss-0.4224, acc-0.8826, test loss-0.4262, acc-0.8809
Iter-60170, train loss-0.6442, acc-0.7800, valid loss-0.4223, acc-0.8826, test loss-0.4262, acc-0.8809
Iter-60180, train loss-0.4895, acc-0.8600, valid loss-0.4223, acc-0.8824, test loss-0.4261, acc-0.8809
Iter-60190, train loss-0.5981, acc-0.8200, valid loss-0.4223, acc-0.8826, test loss-0.4261, acc-0.8809
Iter-60200, train loss-0.3327, acc-0.9000, valid loss-0.4222, acc-0.8826, test loss-0.4261, acc-0.8810
Iter-60210, train loss-0.4950, acc-0.8600, valid loss-0.4222, acc-0.8826, test loss-0.4260, acc-0.8811
Iter-60220, train loss-0.2715, acc-0.9400, valid loss-0.4222, acc-0.8826, test loss-0.4260, acc-0.8808
Iter-60230, train loss-0.3566, acc-0.8800, valid loss-0.4221, acc-0.8826, test loss-0.4260, acc-0.8808
Iter-60240, train loss-0.3700, acc-0.9200, valid loss-0.4221, acc-0.8828, test loss-0.4260, acc-0.8807
Iter-60250, train loss-0.4411, acc-0.8400, valid loss-0.4221, acc-0.8826, test loss-0.4259, acc-0.8807
Iter-60260, train loss-0.6067, acc-0.8400, valid loss-0.4220, acc-0.8826, test loss-0.4259, acc-0.8808
Iter-60270, train loss-0.4912, acc-0.8600, valid loss-0.4220, acc-0.8826, test loss-0.4259, acc-0.8808
Iter-60280, train loss-0.4101, acc-0.9200, valid loss-0.4220, acc-0.8824, test loss-0.4259, acc-0.8809
Iter-60290, train loss-0.3599, acc-0.9200, valid loss-0.4220, acc-0.8824, test loss-0.4258, acc-0.8809
Iter-60300, train loss-0.5939, acc-0.7800, valid loss-0.4219, acc-0.8822, test loss-0.4258, acc-0.8809
Iter-60310, train loss-0.5430, acc-0.8600, valid loss-0.4219, acc-0.8824, test loss-0.4258, acc-0.8809
Iter-60320, train loss-0.4693, acc-0.8600, valid loss-0.4219, acc-0.8824, test loss-0.4257, acc-0.8809
Iter-60330, train loss-0.5441, acc-0.8400, valid loss-0.4218, acc-0.8822, test loss-0.4257, acc-0.8809
Iter-60340, train loss-0.4042, acc-0.9000, valid loss-0.4218, acc-0.8822, test loss-0.4257, acc-0.8809
Iter-60350, train loss-0.2151, acc-0.9800, valid loss-0.4218, acc-0.8822, test loss-0.4256, acc-0.8809
Iter-60360, train loss-0.3138, acc-0.9600, valid loss-0.4218, acc-0.8822, test loss-0.4256, acc-0.8809
Iter-60370, train loss-0.4762, acc-0.8600, valid loss-0.4217, acc-0.8822, test loss-0.4256, acc-0.8809
Iter-60380, train loss-0.4351, acc-0.9200, valid loss-0.4217, acc-0.8822, test loss-0.4256, acc-0.8809
Iter-60390, train loss-0.3661, acc-0.9400, valid loss-0.4217, acc-0.8822, test loss-0.4255, acc-0.8808
Iter-60400, train loss-0.4017, acc-0.9400, valid loss-0.4217, acc-0.8822, test loss-0.4255, acc-0.8808
Iter-60410, train loss-0.4880, acc-0.9000, valid loss-0.4216, acc-0.8822, test loss-0.4255, acc-0.8808
Iter-60420, train loss-0.3928, acc-0.9200, valid loss-0.4216, acc-0.8822, test loss-0.4255, acc-0.8808
Iter-60430, train loss-0.3823, acc-0.9000, valid loss-0.4216, acc-0.8824, test loss-0.4254, acc-0.8809
Iter-60440, train loss-0.6313, acc-0.8000, valid loss-0.4216, acc-0.8824, test loss-0.4254, acc-0.8809
Iter-60450, train loss-0.4783, acc-0.8400, valid loss-0.4215, acc-0.8822, test loss-0.4254, acc-0.8809
Iter-60460, train loss-0.2417, acc-0.9400, valid loss-0.4215, acc-0.8826, test loss-0.4253, acc-0.8809
Iter-60470, train loss-0.2970, acc-0.8800, valid loss-0.4215, acc-0.8824, test loss-0.4253, acc-0.8810
Iter-60480, train loss-0.3624, acc-0.9200, valid loss-0.4214, acc-0.8826, test loss-0.4253, acc-0.8811
Iter-60490, train loss-0.4947, acc-0.8400, valid loss-0.4214, acc-0.8826, test loss-0.4253, acc-0.8812
Iter-60500, train loss-0.4579, acc-0.8800, valid loss-0.4214, acc-0.8822, test loss-0.4252, acc-0.8811
Iter-60510, train loss-0.4691, acc-0.9000, valid loss-0.4214, acc-0.8824, test loss-0.4252, acc-0.8812
Iter-60520, train loss-0.7705, acc-0.7600, valid loss-0.4213, acc-0.8824, test loss-0.4252, acc-0.8813
Iter-60530, train loss-0.4717, acc-0.8400, valid loss-0.4213, acc-0.8824, test loss-0.4251, acc-0.8813
Iter-60540, train loss-0.3149, acc-0.9400, valid loss-0.4213, acc-0.8824, test loss-0.4251, acc-0.8812
Iter-60550, train loss-0.4347, acc-0.8400, valid loss-0.4212, acc-0.8824, test loss-0.4251, acc-0.8812
Iter-60560, train loss-0.4856, acc-0.8400, valid loss-0.4212, acc-0.8824, test loss-0.4250, acc-0.8812
Iter-60570, train loss-0.2518, acc-0.9800, valid loss-0.4211, acc-0.8824, test loss-0.4250, acc-0.8812
Iter-60580, train loss-0.7031, acc-0.7400, valid loss-0.4211, acc-0.8824, test loss-0.4250, acc-0.8811
Iter-60590, train loss-0.4128, acc-0.9200, valid loss-0.4211, acc-0.8824, test loss-0.4249, acc-0.8812
Iter-60600, train loss-0.3766, acc-0.9200, valid loss-0.4210, acc-0.8824, test loss-0.4249, acc-0.8812
Iter-60610, train loss-0.3253, acc-0.9200, valid loss-0.4210, acc-0.8826, test loss-0.4249, acc-0.8812
Iter-60620, train loss-0.4966, acc-0.8600, valid loss-0.4210, acc-0.8824, test loss-0.4248, acc-0.8812
Iter-60630, train loss-0.3417, acc-0.9000, valid loss-0.4210, acc-0.8824, test loss-0.4248, acc-0.8812
Iter-60640, train loss-0.5160, acc-0.8400, valid loss-0.4209, acc-0.8824, test loss-0.4248, acc-0.8812
Iter-60650, train loss-0.2984, acc-0.9200, valid loss-0.4209, acc-0.8824, test loss-0.4247, acc-0.8812
Iter-60660, train loss-0.6134, acc-0.8200, valid loss-0.4208, acc-0.8826, test loss-0.4247, acc-0.8812
Iter-60670, train loss-0.4677, acc-0.8800, valid loss-0.4208, acc-0.8828, test loss-0.4247, acc-0.8810
Iter-60680, train loss-0.4117, acc-0.9000, valid loss-0.4208, acc-0.8828, test loss-0.4246, acc-0.8812
Iter-60690, train loss-0.3870, acc-0.9000, valid loss-0.4207, acc-0.8826, test loss-0.4246, acc-0.8811
Iter-60700, train loss-0.3436, acc-0.9200, valid loss-0.4207, acc-0.8826, test loss-0.4246, acc-0.8813
Iter-60710, train loss-0.4753, acc-0.8600, valid loss-0.4207, acc-0.8824, test loss-0.4246, acc-0.8812
Iter-60720, train loss-0.4154, acc-0.9000, valid loss-0.4207, acc-0.8824, test loss-0.4245, acc-0.8813
Iter-60730, train loss-0.2281, acc-0.9400, valid loss-0.4206, acc-0.8826, test loss-0.4245, acc-0.8813
Iter-60740, train loss-0.3922, acc-0.8400, valid loss-0.4206, acc-0.8824, test loss-0.4245, acc-0.8812
Iter-60750, train loss-0.3202, acc-0.9400, valid loss-0.4206, acc-0.8826, test loss-0.4245, acc-0.8813
Iter-60760, train loss-0.3544, acc-0.9000, valid loss-0.4205, acc-0.8828, test loss-0.4244, acc-0.8813
Iter-60770, train loss-0.4273, acc-0.8400, valid loss-0.4205, acc-0.8826, test loss-0.4244, acc-0.8814
Iter-60780, train loss-0.4748, acc-0.8800, valid loss-0.4205, acc-0.8828, test loss-0.4243, acc-0.8813
Iter-60790, train loss-0.4860, acc-0.9000, valid loss-0.4204, acc-0.8826, test loss-0.4243, acc-0.8813
Iter-60800, train loss-0.5310, acc-0.8400, valid loss-0.4204, acc-0.8828, test loss-0.4243, acc-0.8812
Iter-60810, train loss-0.2983, acc-0.9200, valid loss-0.4204, acc-0.8828, test loss-0.4243, acc-0.8812
Iter-60820, train loss-0.3063, acc-0.9600, valid loss-0.4203, acc-0.8830, test loss-0.4243, acc-0.8812
Iter-60830, train loss-0.4381, acc-0.8800, valid loss-0.4203, acc-0.8826, test loss-0.4242, acc-0.8812
Iter-60840, train loss-0.5031, acc-0.8600, valid loss-0.4203, acc-0.8830, test loss-0.4242, acc-0.8813
Iter-60850, train loss-0.5154, acc-0.8400, valid loss-0.4203, acc-0.8830, test loss-0.4242, acc-0.8813
Iter-60860, train loss-0.4909, acc-0.9000, valid loss-0.4202, acc-0.8830, test loss-0.4241, acc-0.8813
Iter-60870, train loss-0.4106, acc-0.8600, valid loss-0.4202, acc-0.8828, test loss-0.4241, acc-0.8813
Iter-60880, train loss-0.5214, acc-0.8600, valid loss-0.4201, acc-0.8828, test loss-0.4241, acc-0.8813
Iter-60890, train loss-0.5577, acc-0.8800, valid loss-0.4201, acc-0.8830, test loss-0.4240, acc-0.8813
Iter-60900, train loss-0.5330, acc-0.8000, valid loss-0.4201, acc-0.8832, test loss-0.4240, acc-0.8813
Iter-60910, train loss-0.7762, acc-0.7800, valid loss-0.4201, acc-0.8832, test loss-0.4240, acc-0.8813
Iter-60920, train loss-0.2877, acc-0.9400, valid loss-0.4200, acc-0.8830, test loss-0.4239, acc-0.8814
Iter-60930, train loss-0.3448, acc-0.9000, valid loss-0.4200, acc-0.8830, test loss-0.4239, acc-0.8814
Iter-60940, train loss-0.3380, acc-0.9400, valid loss-0.4200, acc-0.8830, test loss-0.4239, acc-0.8814
Iter-60950, train loss-0.3232, acc-0.9000, valid loss-0.4200, acc-0.8830, test loss-0.4239, acc-0.8814
Iter-60960, train loss-0.3628, acc-0.9400, valid loss-0.4199, acc-0.8832, test loss-0.4238, acc-0.8814
Iter-60970, train loss-0.4604, acc-0.8200, valid loss-0.4199, acc-0.8832, test loss-0.4238, acc-0.8814
Iter-60980, train loss-0.3591, acc-0.8800, valid loss-0.4199, acc-0.8832, test loss-0.4238, acc-0.8814
Iter-60990, train loss-0.3934, acc-0.8600, valid loss-0.4198, acc-0.8832, test loss-0.4238, acc-0.8814
Iter-61000, train loss-0.4448, acc-0.8800, valid loss-0.4198, acc-0.8830, test loss-0.4237, acc-0.8814
Iter-61010, train loss-0.3943, acc-0.9000, valid loss-0.4198, acc-0.8830, test loss-0.4237, acc-0.8814
Iter-61020, train loss-0.4948, acc-0.8000, valid loss-0.4197, acc-0.8832, test loss-0.4236, acc-0.8814
Iter-61030, train loss-0.5507, acc-0.8200, valid loss-0.4197, acc-0.8832, test loss-0.4236, acc-0.8814
Iter-61040, train loss-0.4255, acc-0.9000, valid loss-0.4197, acc-0.8832, test loss-0.4236, acc-0.8814
Iter-61050, train loss-0.5298, acc-0.8400, valid loss-0.4196, acc-0.8830, test loss-0.4235, acc-0.8814
Iter-61060, train loss-0.4560, acc-0.8600, valid loss-0.4196, acc-0.8828, test loss-0.4235, acc-0.8814
Iter-61070, train loss-0.3375, acc-0.9200, valid loss-0.4196, acc-0.8826, test loss-0.4235, acc-0.8814
Iter-61080, train loss-0.4855, acc-0.8600, valid loss-0.4196, acc-0.8824, test loss-0.4234, acc-0.8814
Iter-61090, train loss-0.5226, acc-0.8400, valid loss-0.4195, acc-0.8824, test loss-0.4234, acc-0.8814
Iter-61100, train loss-0.3998, acc-0.8800, valid loss-0.4195, acc-0.8824, test loss-0.4234, acc-0.8814
Iter-61110, train loss-0.5526, acc-0.7600, valid loss-0.4195, acc-0.8824, test loss-0.4233, acc-0.8814
Iter-61120, train loss-0.5362, acc-0.8800, valid loss-0.4194, acc-0.8826, test loss-0.4233, acc-0.8815
Iter-61130, train loss-0.4331, acc-0.8400, valid loss-0.4194, acc-0.8828, test loss-0.4233, acc-0.8815
Iter-61140, train loss-0.2397, acc-0.9200, valid loss-0.4194, acc-0.8824, test loss-0.4232, acc-0.8814
Iter-61150, train loss-0.3117, acc-0.9400, valid loss-0.4193, acc-0.8828, test loss-0.4232, acc-0.8815
Iter-61160, train loss-0.3759, acc-0.9000, valid loss-0.4193, acc-0.8824, test loss-0.4232, acc-0.8814
Iter-61170, train loss-0.3949, acc-0.8400, valid loss-0.4193, acc-0.8826, test loss-0.4231, acc-0.8814
Iter-61180, train loss-0.6389, acc-0.8800, valid loss-0.4192, acc-0.8830, test loss-0.4231, acc-0.8814
Iter-61190, train loss-0.5878, acc-0.8400, valid loss-0.4192, acc-0.8828, test loss-0.4231, acc-0.8814
Iter-61200, train loss-0.4530, acc-0.9000, valid loss-0.4192, acc-0.8828, test loss-0.4231, acc-0.8815
Iter-61210, train loss-0.4973, acc-0.8400, valid loss-0.4192, acc-0.8828, test loss-0.4230, acc-0.8815
Iter-61220, train loss-0.3684, acc-0.8400, valid loss-0.4191, acc-0.8828, test loss-0.4230, acc-0.8815
Iter-61230, train loss-0.3227, acc-0.8800, valid loss-0.4191, acc-0.8830, test loss-0.4230, acc-0.8815
Iter-61240, train loss-0.6043, acc-0.7400, valid loss-0.4191, acc-0.8830, test loss-0.4230, acc-0.8815
Iter-61250, train loss-0.5019, acc-0.8400, valid loss-0.4190, acc-0.8830, test loss-0.4229, acc-0.8815
Iter-61260, train loss-0.3413, acc-0.8800, valid loss-0.4190, acc-0.8828, test loss-0.4229, acc-0.8815
Iter-61270, train loss-0.4162, acc-0.9200, valid loss-0.4190, acc-0.8828, test loss-0.4229, acc-0.8815
Iter-61280, train loss-0.7505, acc-0.7400, valid loss-0.4190, acc-0.8828, test loss-0.4228, acc-0.8815
Iter-61290, train loss-0.4577, acc-0.8800, valid loss-0.4189, acc-0.8830, test loss-0.4228, acc-0.8815
Iter-61300, train loss-0.3617, acc-0.8400, valid loss-0.4189, acc-0.8830, test loss-0.4228, acc-0.8815
Iter-61310, train loss-0.5149, acc-0.8400, valid loss-0.4189, acc-0.8830, test loss-0.4228, acc-0.8815
Iter-61320, train loss-0.3051, acc-0.9000, valid loss-0.4188, acc-0.8830, test loss-0.4227, acc-0.8815
Iter-61330, train loss-0.5795, acc-0.7800, valid loss-0.4188, acc-0.8830, test loss-0.4227, acc-0.8815
Iter-61340, train loss-0.4383, acc-0.9000, valid loss-0.4188, acc-0.8830, test loss-0.4227, acc-0.8815
Iter-61350, train loss-0.4803, acc-0.8600, valid loss-0.4188, acc-0.8828, test loss-0.4226, acc-0.8815
Iter-61360, train loss-0.4779, acc-0.8400, valid loss-0.4187, acc-0.8828, test loss-0.4226, acc-0.8815
Iter-61370, train loss-0.4155, acc-0.9000, valid loss-0.4187, acc-0.8832, test loss-0.4226, acc-0.8814
Iter-61380, train loss-0.3626, acc-0.9000, valid loss-0.4187, acc-0.8830, test loss-0.4226, acc-0.8814
Iter-61390, train loss-0.3795, acc-0.9400, valid loss-0.4186, acc-0.8828, test loss-0.4225, acc-0.8814
Iter-61400, train loss-0.4357, acc-0.8600, valid loss-0.4186, acc-0.8828, test loss-0.4225, acc-0.8815
Iter-61410, train loss-0.4004, acc-0.9200, valid loss-0.4186, acc-0.8830, test loss-0.4225, acc-0.8815
Iter-61420, train loss-0.4108, acc-0.9200, valid loss-0.4185, acc-0.8830, test loss-0.4225, acc-0.8815
Iter-61430, train loss-0.4101, acc-0.9000, valid loss-0.4185, acc-0.8830, test loss-0.4224, acc-0.8816
Iter-61440, train loss-0.2744, acc-0.9200, valid loss-0.4185, acc-0.8832, test loss-0.4224, acc-0.8816
Iter-61450, train loss-0.4157, acc-0.9200, valid loss-0.4185, acc-0.8834, test loss-0.4224, acc-0.8818
Iter-61460, train loss-0.3830, acc-0.8600, valid loss-0.4184, acc-0.8834, test loss-0.4223, acc-0.8816
Iter-61470, train loss-0.3743, acc-0.8600, valid loss-0.4184, acc-0.8832, test loss-0.4223, acc-0.8816
Iter-61480, train loss-0.4230, acc-0.9000, valid loss-0.4184, acc-0.8832, test loss-0.4223, acc-0.8816
Iter-61490, train loss-0.4087, acc-0.9200, valid loss-0.4183, acc-0.8830, test loss-0.4223, acc-0.8816
Iter-61500, train loss-0.3867, acc-0.8600, valid loss-0.4183, acc-0.8830, test loss-0.4222, acc-0.8816
Iter-61510, train loss-0.4289, acc-0.8800, valid loss-0.4183, acc-0.8830, test loss-0.4222, acc-0.8816
Iter-61520, train loss-0.3918, acc-0.9000, valid loss-0.4182, acc-0.8832, test loss-0.4222, acc-0.8816
Iter-61530, train loss-0.3638, acc-0.8600, valid loss-0.4182, acc-0.8832, test loss-0.4221, acc-0.8816
Iter-61540, train loss-0.6084, acc-0.8400, valid loss-0.4182, acc-0.8832, test loss-0.4221, acc-0.8815
Iter-61550, train loss-0.2892, acc-0.9000, valid loss-0.4182, acc-0.8834, test loss-0.4221, acc-0.8815
Iter-61560, train loss-0.5437, acc-0.8400, valid loss-0.4181, acc-0.8834, test loss-0.4221, acc-0.8815
Iter-61570, train loss-0.4897, acc-0.8000, valid loss-0.4181, acc-0.8834, test loss-0.4220, acc-0.8815
Iter-61580, train loss-0.4601, acc-0.8600, valid loss-0.4181, acc-0.8832, test loss-0.4220, acc-0.8815
Iter-61590, train loss-0.2742, acc-0.9400, valid loss-0.4181, acc-0.8836, test loss-0.4220, acc-0.8815
Iter-61600, train loss-0.3996, acc-0.9200, valid loss-0.4180, acc-0.8834, test loss-0.4220, acc-0.8816
Iter-61610, train loss-0.5377, acc-0.9000, valid loss-0.4180, acc-0.8836, test loss-0.4219, acc-0.8816
Iter-61620, train loss-0.4610, acc-0.8400, valid loss-0.4180, acc-0.8838, test loss-0.4219, acc-0.8816
Iter-61630, train loss-0.5657, acc-0.8600, valid loss-0.4179, acc-0.8836, test loss-0.4219, acc-0.8816
Iter-61640, train loss-0.5542, acc-0.8000, valid loss-0.4179, acc-0.8834, test loss-0.4219, acc-0.8816
Iter-61650, train loss-0.3552, acc-0.9200, valid loss-0.4179, acc-0.8836, test loss-0.4218, acc-0.8816
Iter-61660, train loss-0.3652, acc-0.9200, valid loss-0.4179, acc-0.8836, test loss-0.4218, acc-0.8816
Iter-61670, train loss-0.3920, acc-0.8800, valid loss-0.4178, acc-0.8836, test loss-0.4218, acc-0.8817
Iter-61680, train loss-0.4122, acc-0.8800, valid loss-0.4178, acc-0.8836, test loss-0.4217, acc-0.8817
Iter-61690, train loss-0.4612, acc-0.8400, valid loss-0.4178, acc-0.8836, test loss-0.4217, acc-0.8818
Iter-61700, train loss-0.3413, acc-0.9200, valid loss-0.4178, acc-0.8836, test loss-0.4217, acc-0.8818
Iter-61710, train loss-0.2992, acc-0.9400, valid loss-0.4177, acc-0.8834, test loss-0.4216, acc-0.8818
Iter-61720, train loss-0.3860, acc-0.8800, valid loss-0.4177, acc-0.8836, test loss-0.4216, acc-0.8819
Iter-61730, train loss-0.4024, acc-0.9000, valid loss-0.4177, acc-0.8838, test loss-0.4216, acc-0.8819
Iter-61740, train loss-0.3656, acc-0.9000, valid loss-0.4177, acc-0.8838, test loss-0.4216, acc-0.8819
Iter-61750, train loss-0.4077, acc-0.8800, valid loss-0.4176, acc-0.8836, test loss-0.4215, acc-0.8819
Iter-61760, train loss-0.4323, acc-0.9200, valid loss-0.4176, acc-0.8836, test loss-0.4215, acc-0.8820
Iter-61770, train loss-0.3589, acc-0.8800, valid loss-0.4176, acc-0.8834, test loss-0.4215, acc-0.8820
Iter-61780, train loss-0.4163, acc-0.9000, valid loss-0.4175, acc-0.8836, test loss-0.4214, acc-0.8821
Iter-61790, train loss-0.2853, acc-0.9600, valid loss-0.4175, acc-0.8834, test loss-0.4214, acc-0.8820
Iter-61800, train loss-0.4977, acc-0.8400, valid loss-0.4175, acc-0.8834, test loss-0.4214, acc-0.8821
Iter-61810, train loss-0.4642, acc-0.9400, valid loss-0.4175, acc-0.8834, test loss-0.4214, acc-0.8820
Iter-61820, train loss-0.3411, acc-0.9000, valid loss-0.4174, acc-0.8834, test loss-0.4213, acc-0.8820
Iter-61830, train loss-0.4398, acc-0.8400, valid loss-0.4174, acc-0.8834, test loss-0.4213, acc-0.8821
Iter-61840, train loss-0.3883, acc-0.8600, valid loss-0.4174, acc-0.8832, test loss-0.4213, acc-0.8820
Iter-61850, train loss-0.4172, acc-0.9000, valid loss-0.4173, acc-0.8832, test loss-0.4212, acc-0.8819
Iter-61860, train loss-0.5046, acc-0.8800, valid loss-0.4173, acc-0.8832, test loss-0.4212, acc-0.8821
Iter-61870, train loss-0.4013, acc-0.8800, valid loss-0.4173, acc-0.8832, test loss-0.4212, acc-0.8821
Iter-61880, train loss-0.4959, acc-0.8200, valid loss-0.4172, acc-0.8832, test loss-0.4212, acc-0.8820
Iter-61890, train loss-0.5662, acc-0.8200, valid loss-0.4172, acc-0.8832, test loss-0.4211, acc-0.8820
Iter-61900, train loss-0.3482, acc-0.8800, valid loss-0.4172, acc-0.8830, test loss-0.4211, acc-0.8821
Iter-61910, train loss-0.3880, acc-0.9000, valid loss-0.4172, acc-0.8830, test loss-0.4211, acc-0.8820
Iter-61920, train loss-0.7334, acc-0.7800, valid loss-0.4171, acc-0.8832, test loss-0.4211, acc-0.8820
Iter-61930, train loss-0.4536, acc-0.8400, valid loss-0.4171, acc-0.8834, test loss-0.4211, acc-0.8820
Iter-61940, train loss-0.4260, acc-0.9000, valid loss-0.4171, acc-0.8836, test loss-0.4210, acc-0.8820
Iter-61950, train loss-0.3344, acc-0.8800, valid loss-0.4171, acc-0.8834, test loss-0.4210, acc-0.8820
Iter-61960, train loss-0.3094, acc-0.9400, valid loss-0.4170, acc-0.8836, test loss-0.4210, acc-0.8819
Iter-61970, train loss-0.3739, acc-0.9000, valid loss-0.4170, acc-0.8836, test loss-0.4210, acc-0.8818
Iter-61980, train loss-0.4859, acc-0.8600, valid loss-0.4169, acc-0.8836, test loss-0.4209, acc-0.8818
Iter-61990, train loss-0.4864, acc-0.8800, valid loss-0.4169, acc-0.8838, test loss-0.4209, acc-0.8818
Iter-62000, train loss-0.3992, acc-0.9200, valid loss-0.4169, acc-0.8838, test loss-0.4209, acc-0.8820
Iter-62010, train loss-0.4381, acc-0.9400, valid loss-0.4169, acc-0.8836, test loss-0.4208, acc-0.8819
Iter-62020, train loss-0.4818, acc-0.8400, valid loss-0.4169, acc-0.8836, test loss-0.4208, acc-0.8819
Iter-62030, train loss-0.4828, acc-0.8000, valid loss-0.4168, acc-0.8836, test loss-0.4208, acc-0.8821
Iter-62040, train loss-0.4737, acc-0.8600, valid loss-0.4168, acc-0.8836, test loss-0.4208, acc-0.8819
Iter-62050, train loss-0.3474, acc-0.8800, valid loss-0.4168, acc-0.8836, test loss-0.4207, acc-0.8820
Iter-62060, train loss-0.3657, acc-0.9200, valid loss-0.4167, acc-0.8836, test loss-0.4207, acc-0.8820
Iter-62070, train loss-0.3273, acc-0.9000, valid loss-0.4167, acc-0.8836, test loss-0.4207, acc-0.8819
Iter-62080, train loss-0.3713, acc-0.8800, valid loss-0.4167, acc-0.8834, test loss-0.4206, acc-0.8819
Iter-62090, train loss-0.5047, acc-0.9000, valid loss-0.4166, acc-0.8836, test loss-0.4206, acc-0.8820
Iter-62100, train loss-0.4502, acc-0.8200, valid loss-0.4166, acc-0.8834, test loss-0.4206, acc-0.8820
Iter-62110, train loss-0.4328, acc-0.9000, valid loss-0.4166, acc-0.8830, test loss-0.4206, acc-0.8819
Iter-62120, train loss-0.4852, acc-0.8400, valid loss-0.4165, acc-0.8830, test loss-0.4205, acc-0.8819
Iter-62130, train loss-0.5584, acc-0.8600, valid loss-0.4165, acc-0.8832, test loss-0.4205, acc-0.8819
Iter-62140, train loss-0.3536, acc-0.8600, valid loss-0.4165, acc-0.8836, test loss-0.4205, acc-0.8819
Iter-62150, train loss-0.5306, acc-0.8800, valid loss-0.4164, acc-0.8836, test loss-0.4205, acc-0.8819
Iter-62160, train loss-0.3158, acc-0.9400, valid loss-0.4164, acc-0.8836, test loss-0.4204, acc-0.8819
Iter-62170, train loss-0.4706, acc-0.9000, valid loss-0.4164, acc-0.8836, test loss-0.4204, acc-0.8820
Iter-62180, train loss-0.3856, acc-0.9000, valid loss-0.4163, acc-0.8836, test loss-0.4204, acc-0.8820
Iter-62190, train loss-0.6828, acc-0.8400, valid loss-0.4163, acc-0.8836, test loss-0.4204, acc-0.8819
Iter-62200, train loss-0.4573, acc-0.9200, valid loss-0.4163, acc-0.8836, test loss-0.4203, acc-0.8819
Iter-62210, train loss-0.3892, acc-0.8800, valid loss-0.4163, acc-0.8834, test loss-0.4203, acc-0.8819
Iter-62220, train loss-0.4540, acc-0.8600, valid loss-0.4162, acc-0.8834, test loss-0.4203, acc-0.8820
Iter-62230, train loss-0.4851, acc-0.8600, valid loss-0.4162, acc-0.8836, test loss-0.4203, acc-0.8820
Iter-62240, train loss-0.4630, acc-0.8800, valid loss-0.4162, acc-0.8832, test loss-0.4202, acc-0.8821
Iter-62250, train loss-0.5893, acc-0.8600, valid loss-0.4162, acc-0.8830, test loss-0.4202, acc-0.8821
Iter-62260, train loss-0.4314, acc-0.9000, valid loss-0.4161, acc-0.8834, test loss-0.4202, acc-0.8820
Iter-62270, train loss-0.4171, acc-0.8800, valid loss-0.4161, acc-0.8830, test loss-0.4201, acc-0.8820
Iter-62280, train loss-0.5492, acc-0.8200, valid loss-0.4161, acc-0.8836, test loss-0.4201, acc-0.8820
Iter-62290, train loss-0.2853, acc-0.9200, valid loss-0.4161, acc-0.8834, test loss-0.4201, acc-0.8820
Iter-62300, train loss-0.4878, acc-0.8600, valid loss-0.4160, acc-0.8834, test loss-0.4200, acc-0.8820
Iter-62310, train loss-0.3966, acc-0.9000, valid loss-0.4160, acc-0.8832, test loss-0.4200, acc-0.8820
Iter-62320, train loss-0.3200, acc-0.9200, valid loss-0.4160, acc-0.8832, test loss-0.4200, acc-0.8820
Iter-62330, train loss-0.5163, acc-0.8600, valid loss-0.4159, acc-0.8836, test loss-0.4200, acc-0.8820
Iter-62340, train loss-0.3493, acc-0.9200, valid loss-0.4159, acc-0.8836, test loss-0.4199, acc-0.8820
Iter-62350, train loss-0.4934, acc-0.8400, valid loss-0.4159, acc-0.8836, test loss-0.4199, acc-0.8820
Iter-62360, train loss-0.2515, acc-0.9200, valid loss-0.4159, acc-0.8838, test loss-0.4199, acc-0.8820
Iter-62370, train loss-0.3333, acc-0.9200, valid loss-0.4159, acc-0.8836, test loss-0.4199, acc-0.8819
Iter-62380, train loss-0.2041, acc-0.9600, valid loss-0.4158, acc-0.8838, test loss-0.4198, acc-0.8820
Iter-62390, train loss-0.5837, acc-0.8000, valid loss-0.4158, acc-0.8838, test loss-0.4198, acc-0.8819
Iter-62400, train loss-0.5340, acc-0.8600, valid loss-0.4158, acc-0.8836, test loss-0.4198, acc-0.8820
Iter-62410, train loss-0.7332, acc-0.8200, valid loss-0.4158, acc-0.8840, test loss-0.4198, acc-0.8820
Iter-62420, train loss-0.5335, acc-0.8600, valid loss-0.4157, acc-0.8836, test loss-0.4197, acc-0.8820
Iter-62430, train loss-0.4319, acc-0.9000, valid loss-0.4157, acc-0.8838, test loss-0.4197, acc-0.8820
Iter-62440, train loss-0.3193, acc-0.9200, valid loss-0.4157, acc-0.8834, test loss-0.4197, acc-0.8821
Iter-62450, train loss-0.5595, acc-0.8400, valid loss-0.4157, acc-0.8834, test loss-0.4196, acc-0.8821
Iter-62460, train loss-0.3701, acc-0.9400, valid loss-0.4156, acc-0.8834, test loss-0.4196, acc-0.8821
Iter-62470, train loss-0.3667, acc-0.8800, valid loss-0.4156, acc-0.8832, test loss-0.4196, acc-0.8821
Iter-62480, train loss-0.3559, acc-0.9200, valid loss-0.4156, acc-0.8832, test loss-0.4196, acc-0.8821
Iter-62490, train loss-0.3503, acc-0.8600, valid loss-0.4156, acc-0.8832, test loss-0.4195, acc-0.8821
Iter-62500, train loss-0.3081, acc-0.9400, valid loss-0.4155, acc-0.8832, test loss-0.4195, acc-0.8822
Iter-62510, train loss-0.3312, acc-0.8800, valid loss-0.4155, acc-0.8832, test loss-0.4195, acc-0.8821
Iter-62520, train loss-0.2627, acc-0.9600, valid loss-0.4155, acc-0.8832, test loss-0.4195, acc-0.8822
Iter-62530, train loss-0.5200, acc-0.8400, valid loss-0.4155, acc-0.8832, test loss-0.4194, acc-0.8822
Iter-62540, train loss-0.3812, acc-0.9200, valid loss-0.4154, acc-0.8832, test loss-0.4194, acc-0.8822
Iter-62550, train loss-0.5963, acc-0.7800, valid loss-0.4154, acc-0.8836, test loss-0.4194, acc-0.8823
Iter-62560, train loss-0.3214, acc-0.9000, valid loss-0.4154, acc-0.8836, test loss-0.4193, acc-0.8823
Iter-62570, train loss-0.3776, acc-0.9200, valid loss-0.4154, acc-0.8836, test loss-0.4193, acc-0.8824
Iter-62580, train loss-0.2732, acc-0.9600, valid loss-0.4153, acc-0.8840, test loss-0.4193, acc-0.8823
Iter-62590, train loss-0.5462, acc-0.8000, valid loss-0.4153, acc-0.8840, test loss-0.4193, acc-0.8823
Iter-62600, train loss-0.3521, acc-0.9000, valid loss-0.4153, acc-0.8840, test loss-0.4192, acc-0.8824
Iter-62610, train loss-0.4584, acc-0.9000, valid loss-0.4153, acc-0.8842, test loss-0.4192, acc-0.8825
Iter-62620, train loss-0.6191, acc-0.8600, valid loss-0.4152, acc-0.8840, test loss-0.4192, acc-0.8824
Iter-62630, train loss-0.5824, acc-0.8400, valid loss-0.4152, acc-0.8838, test loss-0.4191, acc-0.8825
Iter-62640, train loss-0.5035, acc-0.8000, valid loss-0.4152, acc-0.8838, test loss-0.4191, acc-0.8824
Iter-62650, train loss-0.4544, acc-0.8400, valid loss-0.4151, acc-0.8840, test loss-0.4191, acc-0.8824
Iter-62660, train loss-0.3484, acc-0.8800, valid loss-0.4151, acc-0.8840, test loss-0.4190, acc-0.8825
Iter-62670, train loss-0.4013, acc-0.9000, valid loss-0.4151, acc-0.8844, test loss-0.4190, acc-0.8825
Iter-62680, train loss-0.4802, acc-0.8400, valid loss-0.4150, acc-0.8844, test loss-0.4190, acc-0.8825
Iter-62690, train loss-0.3396, acc-0.9000, valid loss-0.4150, acc-0.8846, test loss-0.4189, acc-0.8824
Iter-62700, train loss-0.6442, acc-0.8200, valid loss-0.4150, acc-0.8844, test loss-0.4189, acc-0.8823
Iter-62710, train loss-0.2766, acc-0.9400, valid loss-0.4150, acc-0.8846, test loss-0.4189, acc-0.8823
Iter-62720, train loss-0.4295, acc-0.8200, valid loss-0.4149, acc-0.8846, test loss-0.4189, acc-0.8823
Iter-62730, train loss-0.5635, acc-0.8200, valid loss-0.4149, acc-0.8844, test loss-0.4188, acc-0.8823
Iter-62740, train loss-0.6095, acc-0.8800, valid loss-0.4149, acc-0.8842, test loss-0.4188, acc-0.8823
Iter-62750, train loss-0.4958, acc-0.8400, valid loss-0.4149, acc-0.8846, test loss-0.4188, acc-0.8823
Iter-62760, train loss-0.7239, acc-0.8400, valid loss-0.4148, acc-0.8844, test loss-0.4188, acc-0.8825
Iter-62770, train loss-0.3010, acc-0.9200, valid loss-0.4148, acc-0.8844, test loss-0.4188, acc-0.8824
Iter-62780, train loss-0.4579, acc-0.8600, valid loss-0.4148, acc-0.8844, test loss-0.4187, acc-0.8823
Iter-62790, train loss-0.4751, acc-0.8200, valid loss-0.4148, acc-0.8842, test loss-0.4187, acc-0.8822
Iter-62800, train loss-0.5028, acc-0.8600, valid loss-0.4147, acc-0.8842, test loss-0.4187, acc-0.8823
Iter-62810, train loss-0.4249, acc-0.8800, valid loss-0.4147, acc-0.8844, test loss-0.4187, acc-0.8823
Iter-62820, train loss-0.4608, acc-0.9000, valid loss-0.4147, acc-0.8844, test loss-0.4187, acc-0.8823
Iter-62830, train loss-0.5217, acc-0.8400, valid loss-0.4147, acc-0.8842, test loss-0.4186, acc-0.8825
Iter-62840, train loss-0.4160, acc-0.9000, valid loss-0.4146, acc-0.8842, test loss-0.4186, acc-0.8823
Iter-62850, train loss-0.3920, acc-0.8800, valid loss-0.4146, acc-0.8842, test loss-0.4186, acc-0.8825
Iter-62860, train loss-0.3330, acc-0.9000, valid loss-0.4146, acc-0.8844, test loss-0.4186, acc-0.8824
Iter-62870, train loss-0.3583, acc-0.9400, valid loss-0.4145, acc-0.8844, test loss-0.4185, acc-0.8825
Iter-62880, train loss-0.4454, acc-0.8200, valid loss-0.4145, acc-0.8844, test loss-0.4185, acc-0.8825
Iter-62890, train loss-0.3658, acc-0.9000, valid loss-0.4145, acc-0.8844, test loss-0.4185, acc-0.8825
Iter-62900, train loss-0.6833, acc-0.8000, valid loss-0.4145, acc-0.8844, test loss-0.4184, acc-0.8825
Iter-62910, train loss-0.4111, acc-0.8400, valid loss-0.4144, acc-0.8844, test loss-0.4184, acc-0.8825
Iter-62920, train loss-0.4483, acc-0.8800, valid loss-0.4144, acc-0.8844, test loss-0.4184, acc-0.8824
Iter-62930, train loss-0.4029, acc-0.8800, valid loss-0.4144, acc-0.8844, test loss-0.4183, acc-0.8825
Iter-62940, train loss-0.3302, acc-0.9000, valid loss-0.4144, acc-0.8844, test loss-0.4183, acc-0.8825
Iter-62950, train loss-0.4009, acc-0.9200, valid loss-0.4144, acc-0.8842, test loss-0.4183, acc-0.8825
Iter-62960, train loss-0.2845, acc-0.9200, valid loss-0.4143, acc-0.8842, test loss-0.4183, acc-0.8825
Iter-62970, train loss-0.2817, acc-0.9200, valid loss-0.4143, acc-0.8844, test loss-0.4182, acc-0.8824
Iter-62980, train loss-0.3875, acc-0.8600, valid loss-0.4143, acc-0.8842, test loss-0.4182, acc-0.8824
Iter-62990, train loss-0.6311, acc-0.8800, valid loss-0.4142, acc-0.8844, test loss-0.4182, acc-0.8824
Iter-63000, train loss-0.4452, acc-0.9200, valid loss-0.4142, acc-0.8844, test loss-0.4182, acc-0.8824
Iter-63010, train loss-0.3353, acc-0.9000, valid loss-0.4142, acc-0.8844, test loss-0.4181, acc-0.8824
Iter-63020, train loss-0.5334, acc-0.8200, valid loss-0.4142, acc-0.8844, test loss-0.4181, acc-0.8824
Iter-63030, train loss-0.4481, acc-0.8600, valid loss-0.4141, acc-0.8842, test loss-0.4181, acc-0.8824
Iter-63040, train loss-0.3796, acc-0.9200, valid loss-0.4141, acc-0.8844, test loss-0.4181, acc-0.8824
Iter-63050, train loss-0.5824, acc-0.8800, valid loss-0.4141, acc-0.8844, test loss-0.4180, acc-0.8824
Iter-63060, train loss-0.3872, acc-0.8600, valid loss-0.4140, acc-0.8846, test loss-0.4180, acc-0.8824
Iter-63070, train loss-0.6321, acc-0.8800, valid loss-0.4140, acc-0.8846, test loss-0.4180, acc-0.8824
Iter-63080, train loss-0.3017, acc-0.9200, valid loss-0.4140, acc-0.8846, test loss-0.4180, acc-0.8825
Iter-63090, train loss-0.4807, acc-0.8800, valid loss-0.4140, acc-0.8842, test loss-0.4180, acc-0.8824
Iter-63100, train loss-0.5397, acc-0.8800, valid loss-0.4139, acc-0.8842, test loss-0.4179, acc-0.8824
Iter-63110, train loss-0.5601, acc-0.7600, valid loss-0.4139, acc-0.8844, test loss-0.4179, acc-0.8824
Iter-63120, train loss-0.5254, acc-0.8200, valid loss-0.4139, acc-0.8846, test loss-0.4179, acc-0.8825
Iter-63130, train loss-0.3229, acc-0.8800, valid loss-0.4138, acc-0.8846, test loss-0.4179, acc-0.8825
Iter-63140, train loss-0.4214, acc-0.9000, valid loss-0.4138, acc-0.8846, test loss-0.4178, acc-0.8825
Iter-63150, train loss-0.4013, acc-0.9400, valid loss-0.4138, acc-0.8846, test loss-0.4178, acc-0.8825
Iter-63160, train loss-0.4269, acc-0.8600, valid loss-0.4137, acc-0.8846, test loss-0.4178, acc-0.8825
Iter-63170, train loss-0.5321, acc-0.8200, valid loss-0.4137, acc-0.8846, test loss-0.4178, acc-0.8825
Iter-63180, train loss-0.4398, acc-0.8400, valid loss-0.4137, acc-0.8846, test loss-0.4177, acc-0.8825
Iter-63190, train loss-0.3458, acc-0.8600, valid loss-0.4137, acc-0.8846, test loss-0.4177, acc-0.8825
Iter-63200, train loss-0.3441, acc-0.9200, valid loss-0.4136, acc-0.8848, test loss-0.4177, acc-0.8825
Iter-63210, train loss-0.3379, acc-0.9000, valid loss-0.4136, acc-0.8848, test loss-0.4176, acc-0.8825
Iter-63220, train loss-0.5930, acc-0.8600, valid loss-0.4136, acc-0.8848, test loss-0.4176, acc-0.8825
Iter-63230, train loss-0.5346, acc-0.8400, valid loss-0.4135, acc-0.8848, test loss-0.4176, acc-0.8825
Iter-63240, train loss-0.5194, acc-0.8600, valid loss-0.4135, acc-0.8848, test loss-0.4175, acc-0.8825
Iter-63250, train loss-0.3915, acc-0.8600, valid loss-0.4135, acc-0.8848, test loss-0.4175, acc-0.8825
Iter-63260, train loss-0.2233, acc-0.9200, valid loss-0.4134, acc-0.8848, test loss-0.4175, acc-0.8825
Iter-63270, train loss-0.4733, acc-0.8800, valid loss-0.4134, acc-0.8848, test loss-0.4175, acc-0.8825
Iter-63280, train loss-0.6687, acc-0.8200, valid loss-0.4134, acc-0.8848, test loss-0.4174, acc-0.8825
Iter-63290, train loss-0.5850, acc-0.8400, valid loss-0.4134, acc-0.8848, test loss-0.4174, acc-0.8826
Iter-63300, train loss-0.4987, acc-0.8800, valid loss-0.4133, acc-0.8848, test loss-0.4174, acc-0.8826
Iter-63310, train loss-0.4302, acc-0.8600, valid loss-0.4133, acc-0.8850, test loss-0.4173, acc-0.8826
Iter-63320, train loss-0.6977, acc-0.7800, valid loss-0.4133, acc-0.8850, test loss-0.4173, acc-0.8825
Iter-63330, train loss-0.2734, acc-0.9600, valid loss-0.4132, acc-0.8850, test loss-0.4173, acc-0.8825
Iter-63340, train loss-0.4568, acc-0.8800, valid loss-0.4132, acc-0.8850, test loss-0.4173, acc-0.8825
Iter-63350, train loss-0.2866, acc-0.9000, valid loss-0.4132, acc-0.8848, test loss-0.4172, acc-0.8825
Iter-63360, train loss-0.5992, acc-0.7600, valid loss-0.4132, acc-0.8850, test loss-0.4172, acc-0.8825
Iter-63370, train loss-0.4180, acc-0.8400, valid loss-0.4131, acc-0.8848, test loss-0.4172, acc-0.8824
Iter-63380, train loss-0.4708, acc-0.8400, valid loss-0.4131, acc-0.8848, test loss-0.4172, acc-0.8825
Iter-63390, train loss-0.5463, acc-0.8800, valid loss-0.4131, acc-0.8848, test loss-0.4171, acc-0.8825
Iter-63400, train loss-0.6401, acc-0.8400, valid loss-0.4131, acc-0.8848, test loss-0.4171, acc-0.8826
Iter-63410, train loss-0.3900, acc-0.8800, valid loss-0.4130, acc-0.8848, test loss-0.4171, acc-0.8826
Iter-63420, train loss-0.4011, acc-0.9000, valid loss-0.4130, acc-0.8848, test loss-0.4170, acc-0.8826
Iter-63430, train loss-0.3920, acc-0.9000, valid loss-0.4130, acc-0.8848, test loss-0.4170, acc-0.8825
Iter-63440, train loss-0.6358, acc-0.8200, valid loss-0.4130, acc-0.8850, test loss-0.4170, acc-0.8825
Iter-63450, train loss-0.3821, acc-0.8600, valid loss-0.4129, acc-0.8848, test loss-0.4169, acc-0.8826
Iter-63460, train loss-0.3441, acc-0.8800, valid loss-0.4129, acc-0.8850, test loss-0.4169, acc-0.8826
Iter-63470, train loss-0.3251, acc-0.9200, valid loss-0.4129, acc-0.8850, test loss-0.4169, acc-0.8826
Iter-63480, train loss-0.4238, acc-0.8800, valid loss-0.4128, acc-0.8850, test loss-0.4169, acc-0.8826
Iter-63490, train loss-0.2474, acc-0.9400, valid loss-0.4128, acc-0.8850, test loss-0.4168, acc-0.8826
Iter-63500, train loss-0.2342, acc-0.9400, valid loss-0.4128, acc-0.8850, test loss-0.4168, acc-0.8827
Iter-63510, train loss-0.5868, acc-0.8200, valid loss-0.4127, acc-0.8850, test loss-0.4168, acc-0.8827
Iter-63520, train loss-0.4892, acc-0.9000, valid loss-0.4127, acc-0.8852, test loss-0.4167, acc-0.8827
Iter-63530, train loss-0.5304, acc-0.8200, valid loss-0.4127, acc-0.8852, test loss-0.4167, acc-0.8828
Iter-63540, train loss-0.3726, acc-0.9000, valid loss-0.4127, acc-0.8852, test loss-0.4167, acc-0.8827
Iter-63550, train loss-0.5258, acc-0.8400, valid loss-0.4126, acc-0.8850, test loss-0.4167, acc-0.8828
Iter-63560, train loss-0.3800, acc-0.8600, valid loss-0.4126, acc-0.8850, test loss-0.4166, acc-0.8828
Iter-63570, train loss-0.5421, acc-0.8600, valid loss-0.4126, acc-0.8850, test loss-0.4166, acc-0.8828
Iter-63580, train loss-0.3280, acc-0.9000, valid loss-0.4125, acc-0.8852, test loss-0.4166, acc-0.8828
Iter-63590, train loss-0.5798, acc-0.8200, valid loss-0.4125, acc-0.8850, test loss-0.4166, acc-0.8828
Iter-63600, train loss-0.3130, acc-0.9400, valid loss-0.4125, acc-0.8850, test loss-0.4165, acc-0.8829
Iter-63610, train loss-0.5688, acc-0.8200, valid loss-0.4124, acc-0.8852, test loss-0.4165, acc-0.8829
Iter-63620, train loss-0.4437, acc-0.8600, valid loss-0.4124, acc-0.8852, test loss-0.4165, acc-0.8829
Iter-63630, train loss-0.3308, acc-0.9200, valid loss-0.4124, acc-0.8852, test loss-0.4165, acc-0.8829
Iter-63640, train loss-0.2626, acc-0.9400, valid loss-0.4124, acc-0.8852, test loss-0.4164, acc-0.8828
Iter-63650, train loss-0.3840, acc-0.9200, valid loss-0.4123, acc-0.8852, test loss-0.4164, acc-0.8828
Iter-63660, train loss-0.4366, acc-0.9000, valid loss-0.4123, acc-0.8852, test loss-0.4164, acc-0.8828
Iter-63670, train loss-0.4675, acc-0.9000, valid loss-0.4123, acc-0.8852, test loss-0.4164, acc-0.8828
Iter-63680, train loss-0.6021, acc-0.8200, valid loss-0.4123, acc-0.8850, test loss-0.4163, acc-0.8828
Iter-63690, train loss-0.3757, acc-0.8600, valid loss-0.4122, acc-0.8850, test loss-0.4163, acc-0.8828
Iter-63700, train loss-0.4289, acc-0.8600, valid loss-0.4122, acc-0.8852, test loss-0.4163, acc-0.8828
Iter-63710, train loss-0.5353, acc-0.8600, valid loss-0.4122, acc-0.8854, test loss-0.4163, acc-0.8829
Iter-63720, train loss-0.2815, acc-0.9000, valid loss-0.4122, acc-0.8854, test loss-0.4162, acc-0.8828
Iter-63730, train loss-0.4571, acc-0.9000, valid loss-0.4121, acc-0.8856, test loss-0.4162, acc-0.8828
Iter-63740, train loss-0.3979, acc-0.8600, valid loss-0.4121, acc-0.8854, test loss-0.4162, acc-0.8828
Iter-63750, train loss-0.3462, acc-0.9000, valid loss-0.4121, acc-0.8854, test loss-0.4162, acc-0.8828
Iter-63760, train loss-0.4762, acc-0.8400, valid loss-0.4121, acc-0.8854, test loss-0.4161, acc-0.8829
Iter-63770, train loss-0.3302, acc-0.8600, valid loss-0.4120, acc-0.8854, test loss-0.4161, acc-0.8829
Iter-63780, train loss-0.5035, acc-0.8000, valid loss-0.4120, acc-0.8854, test loss-0.4161, acc-0.8829
Iter-63790, train loss-0.4974, acc-0.8600, valid loss-0.4120, acc-0.8856, test loss-0.4161, acc-0.8829
Iter-63800, train loss-0.4673, acc-0.8400, valid loss-0.4120, acc-0.8856, test loss-0.4160, acc-0.8828
Iter-63810, train loss-0.5219, acc-0.8400, valid loss-0.4119, acc-0.8856, test loss-0.4160, acc-0.8828
Iter-63820, train loss-0.4083, acc-0.8800, valid loss-0.4119, acc-0.8858, test loss-0.4160, acc-0.8828
Iter-63830, train loss-0.2794, acc-0.9400, valid loss-0.4119, acc-0.8856, test loss-0.4160, acc-0.8828
Iter-63840, train loss-0.3413, acc-0.9000, valid loss-0.4119, acc-0.8858, test loss-0.4159, acc-0.8828
Iter-63850, train loss-0.5542, acc-0.8000, valid loss-0.4118, acc-0.8858, test loss-0.4159, acc-0.8829
Iter-63860, train loss-0.5300, acc-0.8400, valid loss-0.4118, acc-0.8858, test loss-0.4159, acc-0.8828
Iter-63870, train loss-0.7793, acc-0.7400, valid loss-0.4118, acc-0.8858, test loss-0.4159, acc-0.8828
Iter-63880, train loss-0.4527, acc-0.8800, valid loss-0.4118, acc-0.8858, test loss-0.4158, acc-0.8828
Iter-63890, train loss-0.4416, acc-0.8600, valid loss-0.4117, acc-0.8858, test loss-0.4158, acc-0.8828
Iter-63900, train loss-0.2966, acc-0.9200, valid loss-0.4117, acc-0.8858, test loss-0.4158, acc-0.8828
Iter-63910, train loss-0.5176, acc-0.8800, valid loss-0.4117, acc-0.8858, test loss-0.4158, acc-0.8829
Iter-63920, train loss-0.5373, acc-0.7600, valid loss-0.4116, acc-0.8858, test loss-0.4157, acc-0.8829
Iter-63930, train loss-0.4135, acc-0.9200, valid loss-0.4116, acc-0.8858, test loss-0.4157, acc-0.8829
Iter-63940, train loss-0.3930, acc-0.8400, valid loss-0.4116, acc-0.8858, test loss-0.4157, acc-0.8828
Iter-63950, train loss-0.2953, acc-0.9400, valid loss-0.4116, acc-0.8858, test loss-0.4156, acc-0.8828
Iter-63960, train loss-0.4407, acc-0.8600, valid loss-0.4115, acc-0.8856, test loss-0.4156, acc-0.8828
Iter-63970, train loss-0.3947, acc-0.9000, valid loss-0.4115, acc-0.8856, test loss-0.4156, acc-0.8829
Iter-63980, train loss-0.4402, acc-0.9000, valid loss-0.4115, acc-0.8856, test loss-0.4156, acc-0.8830
Iter-63990, train loss-0.7547, acc-0.7200, valid loss-0.4115, acc-0.8856, test loss-0.4155, acc-0.8831
Iter-64000, train loss-0.3019, acc-0.9400, valid loss-0.4115, acc-0.8856, test loss-0.4155, acc-0.8831
Iter-64010, train loss-0.4695, acc-0.8400, valid loss-0.4114, acc-0.8856, test loss-0.4155, acc-0.8831
Iter-64020, train loss-0.3771, acc-0.8800, valid loss-0.4114, acc-0.8856, test loss-0.4155, acc-0.8831
Iter-64030, train loss-0.4111, acc-0.9000, valid loss-0.4114, acc-0.8858, test loss-0.4154, acc-0.8832
Iter-64040, train loss-0.4979, acc-0.8200, valid loss-0.4114, acc-0.8858, test loss-0.4154, acc-0.8832
Iter-64050, train loss-0.7179, acc-0.7800, valid loss-0.4113, acc-0.8858, test loss-0.4154, acc-0.8832
Iter-64060, train loss-0.4136, acc-0.8600, valid loss-0.4113, acc-0.8856, test loss-0.4153, acc-0.8832
Iter-64070, train loss-0.3357, acc-0.9000, valid loss-0.4113, acc-0.8858, test loss-0.4153, acc-0.8832
Iter-64080, train loss-0.4139, acc-0.8600, valid loss-0.4113, acc-0.8858, test loss-0.4153, acc-0.8833
Iter-64090, train loss-0.4495, acc-0.8400, valid loss-0.4113, acc-0.8858, test loss-0.4153, acc-0.8833
Iter-64100, train loss-0.3271, acc-0.8800, valid loss-0.4112, acc-0.8858, test loss-0.4152, acc-0.8833
Iter-64110, train loss-0.3924, acc-0.8600, valid loss-0.4112, acc-0.8856, test loss-0.4152, acc-0.8833
Iter-64120, train loss-0.5020, acc-0.8600, valid loss-0.4112, acc-0.8856, test loss-0.4152, acc-0.8832
Iter-64130, train loss-0.4339, acc-0.8400, valid loss-0.4112, acc-0.8856, test loss-0.4152, acc-0.8833
Iter-64140, train loss-0.4239, acc-0.8800, valid loss-0.4111, acc-0.8856, test loss-0.4151, acc-0.8834
Iter-64150, train loss-0.6686, acc-0.8400, valid loss-0.4111, acc-0.8856, test loss-0.4151, acc-0.8834
Iter-64160, train loss-0.4732, acc-0.9000, valid loss-0.4111, acc-0.8858, test loss-0.4151, acc-0.8834
Iter-64170, train loss-0.3818, acc-0.8600, valid loss-0.4110, acc-0.8858, test loss-0.4151, acc-0.8833
Iter-64180, train loss-0.3744, acc-0.8600, valid loss-0.4110, acc-0.8858, test loss-0.4150, acc-0.8833
Iter-64190, train loss-0.4023, acc-0.9000, valid loss-0.4110, acc-0.8858, test loss-0.4150, acc-0.8833
Iter-64200, train loss-0.2738, acc-0.9600, valid loss-0.4110, acc-0.8858, test loss-0.4150, acc-0.8833
Iter-64210, train loss-0.5346, acc-0.8600, valid loss-0.4109, acc-0.8858, test loss-0.4150, acc-0.8834
Iter-64220, train loss-0.3228, acc-0.9000, valid loss-0.4109, acc-0.8858, test loss-0.4149, acc-0.8835
Iter-64230, train loss-0.3679, acc-0.9200, valid loss-0.4109, acc-0.8858, test loss-0.4149, acc-0.8835
Iter-64240, train loss-0.3790, acc-0.8800, valid loss-0.4109, acc-0.8858, test loss-0.4149, acc-0.8834
Iter-64250, train loss-0.3490, acc-0.9200, valid loss-0.4108, acc-0.8856, test loss-0.4148, acc-0.8834
Iter-64260, train loss-0.6862, acc-0.8400, valid loss-0.4108, acc-0.8856, test loss-0.4148, acc-0.8833
Iter-64270, train loss-0.3697, acc-0.8800, valid loss-0.4108, acc-0.8856, test loss-0.4148, acc-0.8834
Iter-64280, train loss-0.3471, acc-0.8800, valid loss-0.4107, acc-0.8856, test loss-0.4147, acc-0.8834
Iter-64290, train loss-0.5300, acc-0.8600, valid loss-0.4107, acc-0.8858, test loss-0.4147, acc-0.8834
Iter-64300, train loss-0.3317, acc-0.9200, valid loss-0.4107, acc-0.8856, test loss-0.4147, acc-0.8835
Iter-64310, train loss-0.5943, acc-0.8000, valid loss-0.4107, acc-0.8856, test loss-0.4147, acc-0.8835
Iter-64320, train loss-0.2687, acc-0.9600, valid loss-0.4106, acc-0.8860, test loss-0.4147, acc-0.8835
Iter-64330, train loss-0.2767, acc-0.9000, valid loss-0.4106, acc-0.8858, test loss-0.4146, acc-0.8835
Iter-64340, train loss-0.3531, acc-0.9600, valid loss-0.4106, acc-0.8858, test loss-0.4146, acc-0.8834
Iter-64350, train loss-0.4932, acc-0.9200, valid loss-0.4105, acc-0.8858, test loss-0.4146, acc-0.8835
Iter-64360, train loss-0.4948, acc-0.8600, valid loss-0.4105, acc-0.8858, test loss-0.4146, acc-0.8835
Iter-64370, train loss-0.2287, acc-0.9600, valid loss-0.4105, acc-0.8858, test loss-0.4145, acc-0.8835
Iter-64380, train loss-0.4598, acc-0.8800, valid loss-0.4105, acc-0.8858, test loss-0.4145, acc-0.8834
Iter-64390, train loss-0.5211, acc-0.8000, valid loss-0.4104, acc-0.8858, test loss-0.4145, acc-0.8834
Iter-64400, train loss-0.6081, acc-0.8000, valid loss-0.4104, acc-0.8858, test loss-0.4145, acc-0.8836
Iter-64410, train loss-0.5480, acc-0.7800, valid loss-0.4104, acc-0.8858, test loss-0.4144, acc-0.8836
Iter-64420, train loss-0.3911, acc-0.8600, valid loss-0.4103, acc-0.8858, test loss-0.4144, acc-0.8836
Iter-64430, train loss-0.4452, acc-0.8800, valid loss-0.4103, acc-0.8858, test loss-0.4144, acc-0.8836
Iter-64440, train loss-0.5544, acc-0.8800, valid loss-0.4102, acc-0.8858, test loss-0.4143, acc-0.8835
Iter-64450, train loss-0.5181, acc-0.8600, valid loss-0.4102, acc-0.8860, test loss-0.4143, acc-0.8835
Iter-64460, train loss-0.3134, acc-0.9000, valid loss-0.4102, acc-0.8862, test loss-0.4143, acc-0.8836
Iter-64470, train loss-0.2981, acc-0.9600, valid loss-0.4102, acc-0.8858, test loss-0.4143, acc-0.8836
Iter-64480, train loss-0.3313, acc-0.9200, valid loss-0.4102, acc-0.8862, test loss-0.4142, acc-0.8836
Iter-64490, train loss-0.4003, acc-0.8600, valid loss-0.4101, acc-0.8862, test loss-0.4142, acc-0.8836
Iter-64500, train loss-0.5456, acc-0.8400, valid loss-0.4101, acc-0.8862, test loss-0.4142, acc-0.8835
Iter-64510, train loss-0.4697, acc-0.8600, valid loss-0.4101, acc-0.8864, test loss-0.4141, acc-0.8836
Iter-64520, train loss-0.3573, acc-0.8800, valid loss-0.4100, acc-0.8864, test loss-0.4141, acc-0.8835
Iter-64530, train loss-0.4911, acc-0.8000, valid loss-0.4100, acc-0.8866, test loss-0.4141, acc-0.8835
Iter-64540, train loss-0.4503, acc-0.8800, valid loss-0.4100, acc-0.8866, test loss-0.4141, acc-0.8836
Iter-64550, train loss-0.4878, acc-0.8600, valid loss-0.4099, acc-0.8864, test loss-0.4140, acc-0.8836
Iter-64560, train loss-0.3405, acc-0.8400, valid loss-0.4099, acc-0.8862, test loss-0.4140, acc-0.8835
Iter-64570, train loss-0.3702, acc-0.8800, valid loss-0.4099, acc-0.8862, test loss-0.4140, acc-0.8835
Iter-64580, train loss-0.1993, acc-0.9600, valid loss-0.4099, acc-0.8862, test loss-0.4139, acc-0.8834
Iter-64590, train loss-0.3295, acc-0.8800, valid loss-0.4098, acc-0.8862, test loss-0.4139, acc-0.8834
Iter-64600, train loss-0.3799, acc-0.8800, valid loss-0.4098, acc-0.8860, test loss-0.4139, acc-0.8836
Iter-64610, train loss-0.2927, acc-0.9200, valid loss-0.4098, acc-0.8862, test loss-0.4139, acc-0.8835
Iter-64620, train loss-0.6755, acc-0.8200, valid loss-0.4097, acc-0.8862, test loss-0.4138, acc-0.8834
Iter-64630, train loss-0.5090, acc-0.8400, valid loss-0.4097, acc-0.8864, test loss-0.4138, acc-0.8834
Iter-64640, train loss-0.3157, acc-0.9400, valid loss-0.4097, acc-0.8862, test loss-0.4138, acc-0.8835
Iter-64650, train loss-0.3938, acc-0.8800, valid loss-0.4097, acc-0.8864, test loss-0.4137, acc-0.8835
Iter-64660, train loss-0.4863, acc-0.8400, valid loss-0.4096, acc-0.8862, test loss-0.4137, acc-0.8835
Iter-64670, train loss-0.3772, acc-0.8200, valid loss-0.4096, acc-0.8864, test loss-0.4137, acc-0.8835
Iter-64680, train loss-0.5569, acc-0.7800, valid loss-0.4096, acc-0.8866, test loss-0.4136, acc-0.8834
Iter-64690, train loss-0.4488, acc-0.8800, valid loss-0.4095, acc-0.8868, test loss-0.4136, acc-0.8834
Iter-64700, train loss-0.3566, acc-0.9000, valid loss-0.4095, acc-0.8868, test loss-0.4136, acc-0.8834
Iter-64710, train loss-0.4060, acc-0.8800, valid loss-0.4095, acc-0.8868, test loss-0.4136, acc-0.8834
Iter-64720, train loss-0.4774, acc-0.9000, valid loss-0.4095, acc-0.8866, test loss-0.4136, acc-0.8836
Iter-64730, train loss-0.2621, acc-0.9600, valid loss-0.4094, acc-0.8868, test loss-0.4135, acc-0.8834
Iter-64740, train loss-0.8164, acc-0.7200, valid loss-0.4094, acc-0.8868, test loss-0.4135, acc-0.8834
Iter-64750, train loss-0.5598, acc-0.8000, valid loss-0.4094, acc-0.8868, test loss-0.4135, acc-0.8835
Iter-64760, train loss-0.4675, acc-0.8400, valid loss-0.4094, acc-0.8866, test loss-0.4135, acc-0.8836
Iter-64770, train loss-0.7287, acc-0.8800, valid loss-0.4093, acc-0.8864, test loss-0.4134, acc-0.8836
Iter-64780, train loss-0.4933, acc-0.8600, valid loss-0.4093, acc-0.8866, test loss-0.4134, acc-0.8836
Iter-64790, train loss-0.3915, acc-0.9000, valid loss-0.4093, acc-0.8864, test loss-0.4134, acc-0.8836
Iter-64800, train loss-0.4723, acc-0.8800, valid loss-0.4093, acc-0.8864, test loss-0.4133, acc-0.8836
Iter-64810, train loss-0.3644, acc-0.8600, valid loss-0.4092, acc-0.8864, test loss-0.4133, acc-0.8837
Iter-64820, train loss-0.5931, acc-0.7800, valid loss-0.4092, acc-0.8864, test loss-0.4133, acc-0.8837
Iter-64830, train loss-0.2502, acc-0.9200, valid loss-0.4092, acc-0.8864, test loss-0.4132, acc-0.8837
Iter-64840, train loss-0.5082, acc-0.8600, valid loss-0.4091, acc-0.8866, test loss-0.4132, acc-0.8838
Iter-64850, train loss-0.4393, acc-0.8800, valid loss-0.4091, acc-0.8864, test loss-0.4132, acc-0.8838
Iter-64860, train loss-0.3871, acc-0.9000, valid loss-0.4091, acc-0.8866, test loss-0.4132, acc-0.8837
Iter-64870, train loss-0.4088, acc-0.8800, valid loss-0.4090, acc-0.8868, test loss-0.4131, acc-0.8837
Iter-64880, train loss-0.4570, acc-0.8600, valid loss-0.4090, acc-0.8864, test loss-0.4131, acc-0.8837
Iter-64890, train loss-0.7296, acc-0.8200, valid loss-0.4090, acc-0.8868, test loss-0.4131, acc-0.8838
Iter-64900, train loss-0.4458, acc-0.8400, valid loss-0.4090, acc-0.8868, test loss-0.4131, acc-0.8839
Iter-64910, train loss-0.5733, acc-0.8400, valid loss-0.4089, acc-0.8868, test loss-0.4130, acc-0.8840
Iter-64920, train loss-0.3898, acc-0.9200, valid loss-0.4089, acc-0.8868, test loss-0.4130, acc-0.8839
Iter-64930, train loss-0.4373, acc-0.8600, valid loss-0.4089, acc-0.8868, test loss-0.4130, acc-0.8840
Iter-64940, train loss-0.4555, acc-0.8400, valid loss-0.4089, acc-0.8868, test loss-0.4130, acc-0.8838
Iter-64950, train loss-0.7103, acc-0.8200, valid loss-0.4088, acc-0.8868, test loss-0.4129, acc-0.8839
Iter-64960, train loss-0.3556, acc-0.9200, valid loss-0.4088, acc-0.8868, test loss-0.4129, acc-0.8841
Iter-64970, train loss-0.3859, acc-0.9000, valid loss-0.4088, acc-0.8868, test loss-0.4129, acc-0.8840
Iter-64980, train loss-0.4270, acc-0.8200, valid loss-0.4087, acc-0.8868, test loss-0.4129, acc-0.8841
Iter-64990, train loss-0.4058, acc-0.9200, valid loss-0.4087, acc-0.8868, test loss-0.4128, acc-0.8840
Iter-65000, train loss-0.4019, acc-0.8800, valid loss-0.4087, acc-0.8868, test loss-0.4128, acc-0.8840
Iter-65010, train loss-0.4841, acc-0.8600, valid loss-0.4087, acc-0.8868, test loss-0.4128, acc-0.8840
Iter-65020, train loss-0.3542, acc-0.9200, valid loss-0.4086, acc-0.8868, test loss-0.4127, acc-0.8840
Iter-65030, train loss-0.4962, acc-0.8600, valid loss-0.4086, acc-0.8866, test loss-0.4127, acc-0.8840
Iter-65040, train loss-0.3018, acc-0.9200, valid loss-0.4086, acc-0.8864, test loss-0.4127, acc-0.8841
Iter-65050, train loss-0.3693, acc-0.9000, valid loss-0.4086, acc-0.8864, test loss-0.4127, acc-0.8841
Iter-65060, train loss-0.3490, acc-0.8800, valid loss-0.4085, acc-0.8862, test loss-0.4126, acc-0.8841
Iter-65070, train loss-0.4399, acc-0.8400, valid loss-0.4085, acc-0.8866, test loss-0.4126, acc-0.8841
Iter-65080, train loss-0.4711, acc-0.9000, valid loss-0.4085, acc-0.8862, test loss-0.4126, acc-0.8842
Iter-65090, train loss-0.3964, acc-0.9000, valid loss-0.4085, acc-0.8862, test loss-0.4126, acc-0.8842
Iter-65100, train loss-0.4027, acc-0.8600, valid loss-0.4084, acc-0.8862, test loss-0.4125, acc-0.8842
Iter-65110, train loss-0.5762, acc-0.8200, valid loss-0.4084, acc-0.8864, test loss-0.4125, acc-0.8842
Iter-65120, train loss-0.7407, acc-0.8000, valid loss-0.4084, acc-0.8864, test loss-0.4125, acc-0.8842
Iter-65130, train loss-0.3448, acc-0.8800, valid loss-0.4084, acc-0.8864, test loss-0.4125, acc-0.8842
Iter-65140, train loss-0.5106, acc-0.7800, valid loss-0.4083, acc-0.8864, test loss-0.4124, acc-0.8842
Iter-65150, train loss-0.3629, acc-0.9200, valid loss-0.4083, acc-0.8866, test loss-0.4124, acc-0.8841
Iter-65160, train loss-0.2774, acc-0.8800, valid loss-0.4083, acc-0.8866, test loss-0.4124, acc-0.8841
Iter-65170, train loss-0.5506, acc-0.8600, valid loss-0.4082, acc-0.8864, test loss-0.4124, acc-0.8841
Iter-65180, train loss-0.2739, acc-0.9400, valid loss-0.4082, acc-0.8866, test loss-0.4123, acc-0.8841
Iter-65190, train loss-0.3384, acc-0.8800, valid loss-0.4082, acc-0.8866, test loss-0.4123, acc-0.8841
Iter-65200, train loss-0.2461, acc-0.9600, valid loss-0.4082, acc-0.8866, test loss-0.4123, acc-0.8841
Iter-65210, train loss-0.2933, acc-0.9200, valid loss-0.4081, acc-0.8864, test loss-0.4122, acc-0.8840
Iter-65220, train loss-0.4297, acc-0.8800, valid loss-0.4081, acc-0.8866, test loss-0.4122, acc-0.8841
Iter-65230, train loss-0.4534, acc-0.8400, valid loss-0.4081, acc-0.8866, test loss-0.4122, acc-0.8840
Iter-65240, train loss-0.6718, acc-0.8000, valid loss-0.4080, acc-0.8866, test loss-0.4122, acc-0.8840
Iter-65250, train loss-0.5047, acc-0.9000, valid loss-0.4080, acc-0.8866, test loss-0.4121, acc-0.8841
Iter-65260, train loss-0.5160, acc-0.8200, valid loss-0.4080, acc-0.8866, test loss-0.4121, acc-0.8842
Iter-65270, train loss-0.5516, acc-0.8600, valid loss-0.4080, acc-0.8866, test loss-0.4121, acc-0.8842
Iter-65280, train loss-0.4338, acc-0.8800, valid loss-0.4080, acc-0.8866, test loss-0.4121, acc-0.8842
Iter-65290, train loss-0.1997, acc-0.9600, valid loss-0.4079, acc-0.8866, test loss-0.4120, acc-0.8842
Iter-65300, train loss-0.2929, acc-0.9200, valid loss-0.4079, acc-0.8866, test loss-0.4120, acc-0.8842
Iter-65310, train loss-0.3634, acc-0.9000, valid loss-0.4079, acc-0.8868, test loss-0.4120, acc-0.8842
Iter-65320, train loss-0.3549, acc-0.8600, valid loss-0.4078, acc-0.8868, test loss-0.4119, acc-0.8842
Iter-65330, train loss-0.5885, acc-0.8000, valid loss-0.4078, acc-0.8868, test loss-0.4119, acc-0.8843
Iter-65340, train loss-0.5627, acc-0.8200, valid loss-0.4078, acc-0.8868, test loss-0.4119, acc-0.8843
Iter-65350, train loss-0.3692, acc-0.9200, valid loss-0.4078, acc-0.8868, test loss-0.4119, acc-0.8843
Iter-65360, train loss-0.4112, acc-0.8600, valid loss-0.4078, acc-0.8868, test loss-0.4118, acc-0.8843
Iter-65370, train loss-0.5249, acc-0.8600, valid loss-0.4077, acc-0.8868, test loss-0.4118, acc-0.8843
Iter-65380, train loss-0.4037, acc-0.8600, valid loss-0.4077, acc-0.8868, test loss-0.4118, acc-0.8843
Iter-65390, train loss-0.5415, acc-0.8400, valid loss-0.4077, acc-0.8868, test loss-0.4117, acc-0.8843
Iter-65400, train loss-0.5099, acc-0.8600, valid loss-0.4077, acc-0.8868, test loss-0.4117, acc-0.8843
Iter-65410, train loss-0.5167, acc-0.8400, valid loss-0.4076, acc-0.8870, test loss-0.4117, acc-0.8843
Iter-65420, train loss-0.3368, acc-0.9400, valid loss-0.4076, acc-0.8868, test loss-0.4117, acc-0.8843
Iter-65430, train loss-0.2275, acc-0.9800, valid loss-0.4076, acc-0.8870, test loss-0.4116, acc-0.8843
Iter-65440, train loss-0.3888, acc-0.9000, valid loss-0.4076, acc-0.8868, test loss-0.4116, acc-0.8843
Iter-65450, train loss-0.5506, acc-0.9200, valid loss-0.4075, acc-0.8870, test loss-0.4116, acc-0.8843
Iter-65460, train loss-0.3103, acc-0.9400, valid loss-0.4075, acc-0.8870, test loss-0.4116, acc-0.8843
Iter-65470, train loss-0.3756, acc-0.8800, valid loss-0.4075, acc-0.8868, test loss-0.4116, acc-0.8844
Iter-65480, train loss-0.5590, acc-0.8000, valid loss-0.4074, acc-0.8868, test loss-0.4115, acc-0.8844
Iter-65490, train loss-0.3217, acc-0.8800, valid loss-0.4074, acc-0.8868, test loss-0.4115, acc-0.8843
Iter-65500, train loss-0.2870, acc-0.9000, valid loss-0.4074, acc-0.8868, test loss-0.4115, acc-0.8844
Iter-65510, train loss-0.5219, acc-0.8800, valid loss-0.4074, acc-0.8876, test loss-0.4114, acc-0.8844
Iter-65520, train loss-0.2904, acc-0.9400, valid loss-0.4073, acc-0.8874, test loss-0.4114, acc-0.8845
Iter-65530, train loss-0.5384, acc-0.8600, valid loss-0.4073, acc-0.8874, test loss-0.4114, acc-0.8844
Iter-65540, train loss-0.5492, acc-0.8600, valid loss-0.4073, acc-0.8876, test loss-0.4114, acc-0.8846
Iter-65550, train loss-0.5106, acc-0.8600, valid loss-0.4072, acc-0.8876, test loss-0.4113, acc-0.8846
Iter-65560, train loss-0.3894, acc-0.8200, valid loss-0.4072, acc-0.8874, test loss-0.4113, acc-0.8846
Iter-65570, train loss-0.4167, acc-0.8800, valid loss-0.4072, acc-0.8874, test loss-0.4113, acc-0.8846
Iter-65580, train loss-0.3807, acc-0.9200, valid loss-0.4072, acc-0.8874, test loss-0.4113, acc-0.8847
Iter-65590, train loss-0.3385, acc-0.8600, valid loss-0.4071, acc-0.8874, test loss-0.4112, acc-0.8847
Iter-65600, train loss-0.4094, acc-0.8800, valid loss-0.4071, acc-0.8876, test loss-0.4112, acc-0.8846
Iter-65610, train loss-0.6418, acc-0.8400, valid loss-0.4071, acc-0.8876, test loss-0.4112, acc-0.8847
Iter-65620, train loss-0.4254, acc-0.8400, valid loss-0.4071, acc-0.8876, test loss-0.4111, acc-0.8848
Iter-65630, train loss-0.5076, acc-0.9200, valid loss-0.4070, acc-0.8874, test loss-0.4111, acc-0.8848
Iter-65640, train loss-0.4700, acc-0.8800, valid loss-0.4070, acc-0.8876, test loss-0.4111, acc-0.8847
Iter-65650, train loss-0.3960, acc-0.8600, valid loss-0.4070, acc-0.8874, test loss-0.4111, acc-0.8847
Iter-65660, train loss-0.3494, acc-0.9200, valid loss-0.4070, acc-0.8874, test loss-0.4110, acc-0.8847
Iter-65670, train loss-0.1885, acc-0.9600, valid loss-0.4070, acc-0.8876, test loss-0.4110, acc-0.8847
Iter-65680, train loss-0.4848, acc-0.8400, valid loss-0.4069, acc-0.8876, test loss-0.4110, acc-0.8848
Iter-65690, train loss-0.3486, acc-0.9000, valid loss-0.4069, acc-0.8876, test loss-0.4110, acc-0.8848
Iter-65700, train loss-0.5723, acc-0.8800, valid loss-0.4069, acc-0.8876, test loss-0.4110, acc-0.8848
Iter-65710, train loss-0.3007, acc-0.9400, valid loss-0.4068, acc-0.8876, test loss-0.4109, acc-0.8848
Iter-65720, train loss-0.4505, acc-0.8400, valid loss-0.4068, acc-0.8876, test loss-0.4109, acc-0.8848
Iter-65730, train loss-0.5855, acc-0.8400, valid loss-0.4068, acc-0.8876, test loss-0.4109, acc-0.8847
Iter-65740, train loss-0.5736, acc-0.8400, valid loss-0.4067, acc-0.8876, test loss-0.4109, acc-0.8847
Iter-65750, train loss-0.1599, acc-0.9600, valid loss-0.4067, acc-0.8878, test loss-0.4108, acc-0.8846
Iter-65760, train loss-0.4027, acc-0.8400, valid loss-0.4067, acc-0.8878, test loss-0.4108, acc-0.8846
Iter-65770, train loss-0.3785, acc-0.9000, valid loss-0.4066, acc-0.8878, test loss-0.4108, acc-0.8846
Iter-65780, train loss-0.4225, acc-0.8800, valid loss-0.4066, acc-0.8878, test loss-0.4108, acc-0.8845
Iter-65790, train loss-0.7038, acc-0.8000, valid loss-0.4066, acc-0.8878, test loss-0.4108, acc-0.8846
Iter-65800, train loss-0.4437, acc-0.9000, valid loss-0.4066, acc-0.8878, test loss-0.4107, acc-0.8846
Iter-65810, train loss-0.4948, acc-0.8600, valid loss-0.4066, acc-0.8876, test loss-0.4107, acc-0.8847
Iter-65820, train loss-0.4655, acc-0.8600, valid loss-0.4065, acc-0.8878, test loss-0.4107, acc-0.8847
Iter-65830, train loss-0.4198, acc-0.8800, valid loss-0.4065, acc-0.8878, test loss-0.4106, acc-0.8847
Iter-65840, train loss-0.7333, acc-0.8000, valid loss-0.4065, acc-0.8876, test loss-0.4106, acc-0.8846
Iter-65850, train loss-0.3368, acc-0.9200, valid loss-0.4065, acc-0.8876, test loss-0.4106, acc-0.8846
Iter-65860, train loss-0.4710, acc-0.8600, valid loss-0.4064, acc-0.8878, test loss-0.4106, acc-0.8846
Iter-65870, train loss-0.3320, acc-0.9000, valid loss-0.4064, acc-0.8876, test loss-0.4105, acc-0.8847
Iter-65880, train loss-0.4074, acc-0.8600, valid loss-0.4064, acc-0.8878, test loss-0.4105, acc-0.8847
Iter-65890, train loss-0.3196, acc-0.9000, valid loss-0.4063, acc-0.8876, test loss-0.4105, acc-0.8847
Iter-65900, train loss-0.3883, acc-0.9400, valid loss-0.4063, acc-0.8876, test loss-0.4105, acc-0.8848
Iter-65910, train loss-0.3461, acc-0.9400, valid loss-0.4063, acc-0.8876, test loss-0.4104, acc-0.8848
Iter-65920, train loss-0.4177, acc-0.8800, valid loss-0.4063, acc-0.8876, test loss-0.4104, acc-0.8848
Iter-65930, train loss-0.3869, acc-0.8800, valid loss-0.4062, acc-0.8876, test loss-0.4104, acc-0.8847
Iter-65940, train loss-0.3073, acc-0.9000, valid loss-0.4062, acc-0.8876, test loss-0.4104, acc-0.8847
Iter-65950, train loss-0.3490, acc-0.9000, valid loss-0.4062, acc-0.8876, test loss-0.4103, acc-0.8847
Iter-65960, train loss-0.4751, acc-0.8400, valid loss-0.4061, acc-0.8876, test loss-0.4103, acc-0.8847
Iter-65970, train loss-0.2244, acc-0.9800, valid loss-0.4061, acc-0.8876, test loss-0.4103, acc-0.8848
Iter-65980, train loss-0.5315, acc-0.8400, valid loss-0.4061, acc-0.8876, test loss-0.4103, acc-0.8848
Iter-65990, train loss-0.3292, acc-0.9000, valid loss-0.4061, acc-0.8880, test loss-0.4102, acc-0.8848
Iter-66000, train loss-0.3759, acc-0.9000, valid loss-0.4060, acc-0.8880, test loss-0.4102, acc-0.8848
Iter-66010, train loss-0.5033, acc-0.8600, valid loss-0.4060, acc-0.8880, test loss-0.4102, acc-0.8848
Iter-66020, train loss-0.4048, acc-0.8800, valid loss-0.4060, acc-0.8880, test loss-0.4101, acc-0.8847
Iter-66030, train loss-0.3607, acc-0.8800, valid loss-0.4059, acc-0.8880, test loss-0.4101, acc-0.8847
Iter-66040, train loss-0.5928, acc-0.8600, valid loss-0.4059, acc-0.8880, test loss-0.4101, acc-0.8847
Iter-66050, train loss-0.5218, acc-0.8600, valid loss-0.4059, acc-0.8880, test loss-0.4101, acc-0.8847
Iter-66060, train loss-0.4373, acc-0.8400, valid loss-0.4059, acc-0.8880, test loss-0.4100, acc-0.8847
Iter-66070, train loss-0.3455, acc-0.9200, valid loss-0.4058, acc-0.8880, test loss-0.4100, acc-0.8849
Iter-66080, train loss-0.4330, acc-0.9000, valid loss-0.4058, acc-0.8880, test loss-0.4100, acc-0.8849
Iter-66090, train loss-0.3962, acc-0.8600, valid loss-0.4058, acc-0.8880, test loss-0.4100, acc-0.8848
Iter-66100, train loss-0.7237, acc-0.8200, valid loss-0.4058, acc-0.8880, test loss-0.4099, acc-0.8848
Iter-66110, train loss-0.4286, acc-0.8800, valid loss-0.4058, acc-0.8880, test loss-0.4099, acc-0.8848
Iter-66120, train loss-0.2565, acc-0.9200, valid loss-0.4057, acc-0.8880, test loss-0.4099, acc-0.8848
Iter-66130, train loss-0.3640, acc-0.8800, valid loss-0.4057, acc-0.8882, test loss-0.4099, acc-0.8848
Iter-66140, train loss-0.2272, acc-0.9400, valid loss-0.4057, acc-0.8880, test loss-0.4099, acc-0.8848
Iter-66150, train loss-0.3353, acc-0.9000, valid loss-0.4057, acc-0.8880, test loss-0.4098, acc-0.8848
Iter-66160, train loss-0.4115, acc-0.9000, valid loss-0.4056, acc-0.8878, test loss-0.4098, acc-0.8848
Iter-66170, train loss-0.3297, acc-0.9000, valid loss-0.4056, acc-0.8880, test loss-0.4098, acc-0.8848
Iter-66180, train loss-0.5207, acc-0.8400, valid loss-0.4056, acc-0.8880, test loss-0.4098, acc-0.8847
Iter-66190, train loss-0.5552, acc-0.8400, valid loss-0.4056, acc-0.8880, test loss-0.4097, acc-0.8849
Iter-66200, train loss-0.5727, acc-0.8600, valid loss-0.4055, acc-0.8880, test loss-0.4097, acc-0.8849
Iter-66210, train loss-0.5522, acc-0.8000, valid loss-0.4055, acc-0.8880, test loss-0.4097, acc-0.8849
Iter-66220, train loss-0.3498, acc-0.9200, valid loss-0.4055, acc-0.8880, test loss-0.4097, acc-0.8849
Iter-66230, train loss-0.4601, acc-0.8600, valid loss-0.4055, acc-0.8882, test loss-0.4097, acc-0.8849
Iter-66240, train loss-0.5177, acc-0.8800, valid loss-0.4054, acc-0.8880, test loss-0.4096, acc-0.8850
Iter-66250, train loss-0.3338, acc-0.8800, valid loss-0.4054, acc-0.8880, test loss-0.4096, acc-0.8849
Iter-66260, train loss-0.5429, acc-0.8600, valid loss-0.4054, acc-0.8880, test loss-0.4096, acc-0.8849
Iter-66270, train loss-0.4314, acc-0.8800, valid loss-0.4054, acc-0.8880, test loss-0.4096, acc-0.8850
Iter-66280, train loss-0.3973, acc-0.9000, valid loss-0.4054, acc-0.8880, test loss-0.4096, acc-0.8851
Iter-66290, train loss-0.5388, acc-0.8200, valid loss-0.4054, acc-0.8878, test loss-0.4095, acc-0.8850
Iter-66300, train loss-0.7245, acc-0.7600, valid loss-0.4054, acc-0.8878, test loss-0.4095, acc-0.8851
Iter-66310, train loss-0.4885, acc-0.8600, valid loss-0.4053, acc-0.8876, test loss-0.4095, acc-0.8849
Iter-66320, train loss-0.4698, acc-0.8600, valid loss-0.4053, acc-0.8878, test loss-0.4094, acc-0.8850
Iter-66330, train loss-0.4169, acc-0.8800, valid loss-0.4053, acc-0.8880, test loss-0.4094, acc-0.8851
Iter-66340, train loss-0.4515, acc-0.8400, valid loss-0.4053, acc-0.8878, test loss-0.4094, acc-0.8851
Iter-66350, train loss-0.4651, acc-0.8800, valid loss-0.4052, acc-0.8878, test loss-0.4094, acc-0.8851
Iter-66360, train loss-0.3381, acc-0.8600, valid loss-0.4052, acc-0.8876, test loss-0.4093, acc-0.8849
Iter-66370, train loss-0.5669, acc-0.8200, valid loss-0.4052, acc-0.8876, test loss-0.4093, acc-0.8848
Iter-66380, train loss-0.3408, acc-0.8800, valid loss-0.4051, acc-0.8876, test loss-0.4093, acc-0.8847
Iter-66390, train loss-0.5235, acc-0.8600, valid loss-0.4051, acc-0.8876, test loss-0.4092, acc-0.8848
Iter-66400, train loss-0.4202, acc-0.8600, valid loss-0.4051, acc-0.8876, test loss-0.4092, acc-0.8849
Iter-66410, train loss-0.5941, acc-0.8600, valid loss-0.4050, acc-0.8876, test loss-0.4092, acc-0.8849
Iter-66420, train loss-0.3415, acc-0.9000, valid loss-0.4050, acc-0.8876, test loss-0.4092, acc-0.8850
Iter-66430, train loss-0.4926, acc-0.8800, valid loss-0.4050, acc-0.8876, test loss-0.4091, acc-0.8851
Iter-66440, train loss-0.5455, acc-0.8000, valid loss-0.4050, acc-0.8876, test loss-0.4091, acc-0.8850
Iter-66450, train loss-0.3125, acc-0.9200, valid loss-0.4049, acc-0.8880, test loss-0.4091, acc-0.8850
Iter-66460, train loss-0.3935, acc-0.9000, valid loss-0.4049, acc-0.8880, test loss-0.4091, acc-0.8850
Iter-66470, train loss-0.5958, acc-0.8400, valid loss-0.4049, acc-0.8884, test loss-0.4091, acc-0.8850
Iter-66480, train loss-0.3324, acc-0.9200, valid loss-0.4049, acc-0.8886, test loss-0.4090, acc-0.8850
Iter-66490, train loss-0.2177, acc-0.9800, valid loss-0.4048, acc-0.8886, test loss-0.4090, acc-0.8850
Iter-66500, train loss-0.4533, acc-0.9200, valid loss-0.4048, acc-0.8886, test loss-0.4090, acc-0.8851
Iter-66510, train loss-0.5390, acc-0.8200, valid loss-0.4048, acc-0.8884, test loss-0.4090, acc-0.8851
Iter-66520, train loss-0.2732, acc-0.9000, valid loss-0.4048, acc-0.8884, test loss-0.4089, acc-0.8851
Iter-66530, train loss-0.3602, acc-0.9200, valid loss-0.4047, acc-0.8884, test loss-0.4089, acc-0.8850
Iter-66540, train loss-0.3255, acc-0.9400, valid loss-0.4047, acc-0.8884, test loss-0.4089, acc-0.8850
Iter-66550, train loss-0.5018, acc-0.8200, valid loss-0.4047, acc-0.8884, test loss-0.4089, acc-0.8850
Iter-66560, train loss-0.4788, acc-0.8800, valid loss-0.4047, acc-0.8884, test loss-0.4089, acc-0.8851
Iter-66570, train loss-0.3893, acc-0.8800, valid loss-0.4047, acc-0.8880, test loss-0.4089, acc-0.8851
Iter-66580, train loss-0.4450, acc-0.8800, valid loss-0.4047, acc-0.8880, test loss-0.4088, acc-0.8852
Iter-66590, train loss-0.3051, acc-0.9400, valid loss-0.4046, acc-0.8880, test loss-0.4088, acc-0.8852
Iter-66600, train loss-0.3396, acc-0.8600, valid loss-0.4046, acc-0.8882, test loss-0.4088, acc-0.8851
Iter-66610, train loss-0.2956, acc-0.9200, valid loss-0.4046, acc-0.8882, test loss-0.4088, acc-0.8851
Iter-66620, train loss-0.6046, acc-0.8800, valid loss-0.4046, acc-0.8882, test loss-0.4087, acc-0.8851
Iter-66630, train loss-0.3027, acc-0.9200, valid loss-0.4045, acc-0.8882, test loss-0.4087, acc-0.8852
Iter-66640, train loss-0.4304, acc-0.8800, valid loss-0.4045, acc-0.8880, test loss-0.4087, acc-0.8852
Iter-66650, train loss-0.4736, acc-0.8600, valid loss-0.4045, acc-0.8880, test loss-0.4087, acc-0.8851
Iter-66660, train loss-0.5156, acc-0.8800, valid loss-0.4045, acc-0.8880, test loss-0.4086, acc-0.8851
Iter-66670, train loss-0.3666, acc-0.8800, valid loss-0.4044, acc-0.8880, test loss-0.4086, acc-0.8851
Iter-66680, train loss-0.1754, acc-0.9800, valid loss-0.4044, acc-0.8882, test loss-0.4086, acc-0.8853
Iter-66690, train loss-0.4488, acc-0.9000, valid loss-0.4044, acc-0.8882, test loss-0.4086, acc-0.8853
Iter-66700, train loss-0.4929, acc-0.9000, valid loss-0.4043, acc-0.8882, test loss-0.4085, acc-0.8853
Iter-66710, train loss-0.3861, acc-0.8800, valid loss-0.4043, acc-0.8884, test loss-0.4085, acc-0.8853
Iter-66720, train loss-0.5724, acc-0.8200, valid loss-0.4043, acc-0.8884, test loss-0.4085, acc-0.8853
Iter-66730, train loss-0.5986, acc-0.8000, valid loss-0.4043, acc-0.8884, test loss-0.4085, acc-0.8853
Iter-66740, train loss-0.6651, acc-0.7800, valid loss-0.4042, acc-0.8884, test loss-0.4084, acc-0.8852
Iter-66750, train loss-0.3212, acc-0.9200, valid loss-0.4042, acc-0.8882, test loss-0.4084, acc-0.8852
Iter-66760, train loss-0.4087, acc-0.9000, valid loss-0.4042, acc-0.8882, test loss-0.4084, acc-0.8851
Iter-66770, train loss-0.4196, acc-0.8600, valid loss-0.4042, acc-0.8882, test loss-0.4084, acc-0.8853
Iter-66780, train loss-0.3099, acc-0.9200, valid loss-0.4041, acc-0.8880, test loss-0.4084, acc-0.8853
Iter-66790, train loss-0.3924, acc-0.9200, valid loss-0.4041, acc-0.8882, test loss-0.4083, acc-0.8853
Iter-66800, train loss-0.4054, acc-0.8600, valid loss-0.4041, acc-0.8884, test loss-0.4083, acc-0.8853
Iter-66810, train loss-0.4088, acc-0.8600, valid loss-0.4041, acc-0.8884, test loss-0.4083, acc-0.8852
Iter-66820, train loss-0.2667, acc-0.9000, valid loss-0.4040, acc-0.8884, test loss-0.4083, acc-0.8852
Iter-66830, train loss-0.4672, acc-0.9200, valid loss-0.4040, acc-0.8886, test loss-0.4082, acc-0.8853
Iter-66840, train loss-0.3837, acc-0.8600, valid loss-0.4040, acc-0.8886, test loss-0.4082, acc-0.8853
Iter-66850, train loss-0.4415, acc-0.8800, valid loss-0.4040, acc-0.8884, test loss-0.4082, acc-0.8852
Iter-66860, train loss-0.4457, acc-0.8600, valid loss-0.4040, acc-0.8884, test loss-0.4082, acc-0.8853
Iter-66870, train loss-0.3838, acc-0.9000, valid loss-0.4039, acc-0.8886, test loss-0.4082, acc-0.8853
Iter-66880, train loss-0.3673, acc-0.9000, valid loss-0.4039, acc-0.8886, test loss-0.4081, acc-0.8853
Iter-66890, train loss-0.4262, acc-0.8800, valid loss-0.4039, acc-0.8886, test loss-0.4081, acc-0.8853
Iter-66900, train loss-0.4482, acc-0.8800, valid loss-0.4038, acc-0.8886, test loss-0.4081, acc-0.8854
Iter-66910, train loss-0.6710, acc-0.8200, valid loss-0.4038, acc-0.8886, test loss-0.4081, acc-0.8854
Iter-66920, train loss-0.5261, acc-0.7800, valid loss-0.4038, acc-0.8888, test loss-0.4080, acc-0.8853
Iter-66930, train loss-0.3135, acc-0.9200, valid loss-0.4038, acc-0.8884, test loss-0.4080, acc-0.8854
Iter-66940, train loss-0.4053, acc-0.8800, valid loss-0.4038, acc-0.8884, test loss-0.4080, acc-0.8855
Iter-66950, train loss-0.4922, acc-0.8600, valid loss-0.4037, acc-0.8886, test loss-0.4080, acc-0.8854
Iter-66960, train loss-0.4065, acc-0.8600, valid loss-0.4037, acc-0.8886, test loss-0.4079, acc-0.8854
Iter-66970, train loss-0.3883, acc-0.8800, valid loss-0.4037, acc-0.8884, test loss-0.4079, acc-0.8854
Iter-66980, train loss-0.4177, acc-0.8800, valid loss-0.4037, acc-0.8884, test loss-0.4079, acc-0.8854
Iter-66990, train loss-0.5221, acc-0.8600, valid loss-0.4036, acc-0.8884, test loss-0.4079, acc-0.8854
Iter-67000, train loss-0.5124, acc-0.8400, valid loss-0.4036, acc-0.8884, test loss-0.4078, acc-0.8854
Iter-67010, train loss-0.3105, acc-0.9200, valid loss-0.4036, acc-0.8884, test loss-0.4078, acc-0.8854
Iter-67020, train loss-0.3039, acc-0.9400, valid loss-0.4035, acc-0.8884, test loss-0.4078, acc-0.8854
Iter-67030, train loss-0.7282, acc-0.7400, valid loss-0.4035, acc-0.8884, test loss-0.4078, acc-0.8855
Iter-67040, train loss-0.2289, acc-0.9600, valid loss-0.4035, acc-0.8886, test loss-0.4077, acc-0.8854
Iter-67050, train loss-0.4275, acc-0.8600, valid loss-0.4035, acc-0.8884, test loss-0.4077, acc-0.8853
Iter-67060, train loss-0.5821, acc-0.8200, valid loss-0.4035, acc-0.8882, test loss-0.4077, acc-0.8854
Iter-67070, train loss-0.5366, acc-0.7800, valid loss-0.4034, acc-0.8882, test loss-0.4077, acc-0.8854
Iter-67080, train loss-0.5266, acc-0.8800, valid loss-0.4034, acc-0.8884, test loss-0.4076, acc-0.8855
Iter-67090, train loss-0.4112, acc-0.8400, valid loss-0.4034, acc-0.8882, test loss-0.4076, acc-0.8855
Iter-67100, train loss-0.4299, acc-0.8400, valid loss-0.4034, acc-0.8882, test loss-0.4076, acc-0.8855
Iter-67110, train loss-0.4494, acc-0.8800, valid loss-0.4033, acc-0.8882, test loss-0.4076, acc-0.8854
Iter-67120, train loss-0.4010, acc-0.8600, valid loss-0.4033, acc-0.8882, test loss-0.4075, acc-0.8855
Iter-67130, train loss-0.4193, acc-0.9000, valid loss-0.4033, acc-0.8884, test loss-0.4075, acc-0.8855
Iter-67140, train loss-0.6078, acc-0.8000, valid loss-0.4033, acc-0.8884, test loss-0.4075, acc-0.8855
Iter-67150, train loss-0.4674, acc-0.8800, valid loss-0.4032, acc-0.8884, test loss-0.4074, acc-0.8854
Iter-67160, train loss-0.4511, acc-0.8600, valid loss-0.4032, acc-0.8884, test loss-0.4074, acc-0.8854
Iter-67170, train loss-0.5108, acc-0.8600, valid loss-0.4032, acc-0.8884, test loss-0.4074, acc-0.8856
Iter-67180, train loss-0.5009, acc-0.8200, valid loss-0.4032, acc-0.8884, test loss-0.4074, acc-0.8856
Iter-67190, train loss-0.4188, acc-0.8400, valid loss-0.4031, acc-0.8884, test loss-0.4073, acc-0.8856
Iter-67200, train loss-0.3319, acc-0.9000, valid loss-0.4031, acc-0.8884, test loss-0.4073, acc-0.8855
Iter-67210, train loss-0.2922, acc-0.9400, valid loss-0.4031, acc-0.8882, test loss-0.4073, acc-0.8856
Iter-67220, train loss-0.5921, acc-0.8600, valid loss-0.4031, acc-0.8886, test loss-0.4073, acc-0.8857
Iter-67230, train loss-0.2826, acc-0.9200, valid loss-0.4030, acc-0.8884, test loss-0.4072, acc-0.8857
Iter-67240, train loss-0.2965, acc-0.9400, valid loss-0.4030, acc-0.8886, test loss-0.4072, acc-0.8856
Iter-67250, train loss-0.3341, acc-0.9000, valid loss-0.4030, acc-0.8886, test loss-0.4072, acc-0.8857
Iter-67260, train loss-0.4045, acc-0.8600, valid loss-0.4030, acc-0.8886, test loss-0.4072, acc-0.8857
Iter-67270, train loss-0.2558, acc-0.9200, valid loss-0.4030, acc-0.8886, test loss-0.4072, acc-0.8857
Iter-67280, train loss-0.2889, acc-0.9400, valid loss-0.4029, acc-0.8888, test loss-0.4071, acc-0.8856
Iter-67290, train loss-0.4338, acc-0.8800, valid loss-0.4029, acc-0.8886, test loss-0.4071, acc-0.8857
Iter-67300, train loss-0.6629, acc-0.8200, valid loss-0.4029, acc-0.8886, test loss-0.4071, acc-0.8857
Iter-67310, train loss-0.4350, acc-0.8800, valid loss-0.4029, acc-0.8886, test loss-0.4071, acc-0.8857
Iter-67320, train loss-0.5226, acc-0.8600, valid loss-0.4028, acc-0.8886, test loss-0.4070, acc-0.8856
Iter-67330, train loss-0.4083, acc-0.8600, valid loss-0.4028, acc-0.8886, test loss-0.4070, acc-0.8856
Iter-67340, train loss-0.4964, acc-0.8200, valid loss-0.4028, acc-0.8886, test loss-0.4070, acc-0.8856
Iter-67350, train loss-0.5297, acc-0.8800, valid loss-0.4028, acc-0.8886, test loss-0.4070, acc-0.8856
Iter-67360, train loss-0.5699, acc-0.8200, valid loss-0.4027, acc-0.8886, test loss-0.4069, acc-0.8856
Iter-67370, train loss-0.4259, acc-0.8800, valid loss-0.4027, acc-0.8886, test loss-0.4069, acc-0.8856
Iter-67380, train loss-0.6582, acc-0.8400, valid loss-0.4027, acc-0.8886, test loss-0.4069, acc-0.8856
Iter-67390, train loss-0.3975, acc-0.9000, valid loss-0.4027, acc-0.8886, test loss-0.4069, acc-0.8856
Iter-67400, train loss-0.5638, acc-0.8600, valid loss-0.4027, acc-0.8886, test loss-0.4068, acc-0.8857
Iter-67410, train loss-0.4684, acc-0.8800, valid loss-0.4026, acc-0.8886, test loss-0.4068, acc-0.8857
Iter-67420, train loss-0.3442, acc-0.8800, valid loss-0.4026, acc-0.8886, test loss-0.4068, acc-0.8857
Iter-67430, train loss-0.3186, acc-0.9200, valid loss-0.4026, acc-0.8886, test loss-0.4068, acc-0.8857
Iter-67440, train loss-0.4453, acc-0.9000, valid loss-0.4026, acc-0.8886, test loss-0.4068, acc-0.8856
Iter-67450, train loss-0.3882, acc-0.8800, valid loss-0.4025, acc-0.8886, test loss-0.4067, acc-0.8857
Iter-67460, train loss-0.1918, acc-0.9600, valid loss-0.4025, acc-0.8886, test loss-0.4067, acc-0.8858
Iter-67470, train loss-0.4040, acc-0.8800, valid loss-0.4025, acc-0.8886, test loss-0.4067, acc-0.8859
Iter-67480, train loss-0.6597, acc-0.8600, valid loss-0.4025, acc-0.8886, test loss-0.4067, acc-0.8858
Iter-67490, train loss-0.3757, acc-0.8600, valid loss-0.4024, acc-0.8886, test loss-0.4066, acc-0.8858
Iter-67500, train loss-0.5310, acc-0.8000, valid loss-0.4024, acc-0.8886, test loss-0.4066, acc-0.8857
Iter-67510, train loss-0.4457, acc-0.9400, valid loss-0.4024, acc-0.8886, test loss-0.4066, acc-0.8857
Iter-67520, train loss-0.4655, acc-0.8600, valid loss-0.4024, acc-0.8884, test loss-0.4066, acc-0.8857
Iter-67530, train loss-0.3678, acc-0.9000, valid loss-0.4023, acc-0.8884, test loss-0.4065, acc-0.8857
Iter-67540, train loss-0.5338, acc-0.8600, valid loss-0.4023, acc-0.8884, test loss-0.4065, acc-0.8858
Iter-67550, train loss-0.4933, acc-0.8800, valid loss-0.4023, acc-0.8886, test loss-0.4065, acc-0.8858
Iter-67560, train loss-0.5780, acc-0.8600, valid loss-0.4023, acc-0.8886, test loss-0.4065, acc-0.8858
Iter-67570, train loss-0.2134, acc-0.9600, valid loss-0.4022, acc-0.8886, test loss-0.4064, acc-0.8858
Iter-67580, train loss-0.6344, acc-0.8000, valid loss-0.4022, acc-0.8886, test loss-0.4064, acc-0.8858
Iter-67590, train loss-0.2855, acc-0.9200, valid loss-0.4022, acc-0.8888, test loss-0.4064, acc-0.8858
Iter-67600, train loss-0.4129, acc-0.8800, valid loss-0.4022, acc-0.8886, test loss-0.4064, acc-0.8857
Iter-67610, train loss-0.5235, acc-0.8400, valid loss-0.4022, acc-0.8886, test loss-0.4064, acc-0.8855
Iter-67620, train loss-0.3667, acc-0.9400, valid loss-0.4021, acc-0.8884, test loss-0.4063, acc-0.8856
Iter-67630, train loss-0.3960, acc-0.8600, valid loss-0.4021, acc-0.8888, test loss-0.4063, acc-0.8856
Iter-67640, train loss-0.5428, acc-0.8000, valid loss-0.4021, acc-0.8886, test loss-0.4063, acc-0.8856
Iter-67650, train loss-0.2825, acc-0.8800, valid loss-0.4021, acc-0.8886, test loss-0.4062, acc-0.8856
Iter-67660, train loss-0.4324, acc-0.9000, valid loss-0.4021, acc-0.8886, test loss-0.4062, acc-0.8857
Iter-67670, train loss-0.4172, acc-0.8800, valid loss-0.4020, acc-0.8886, test loss-0.4062, acc-0.8858
Iter-67680, train loss-0.4711, acc-0.8000, valid loss-0.4020, acc-0.8886, test loss-0.4062, acc-0.8858
Iter-67690, train loss-0.5885, acc-0.8000, valid loss-0.4020, acc-0.8886, test loss-0.4061, acc-0.8859
Iter-67700, train loss-0.2827, acc-0.9400, valid loss-0.4020, acc-0.8886, test loss-0.4061, acc-0.8860
Iter-67710, train loss-0.4844, acc-0.8600, valid loss-0.4019, acc-0.8886, test loss-0.4061, acc-0.8858
Iter-67720, train loss-0.2841, acc-0.9400, valid loss-0.4019, acc-0.8886, test loss-0.4061, acc-0.8857
Iter-67730, train loss-0.4331, acc-0.8400, valid loss-0.4019, acc-0.8884, test loss-0.4061, acc-0.8858
Iter-67740, train loss-0.4088, acc-0.9000, valid loss-0.4019, acc-0.8886, test loss-0.4060, acc-0.8859
Iter-67750, train loss-0.5284, acc-0.8400, valid loss-0.4019, acc-0.8886, test loss-0.4060, acc-0.8859
Iter-67760, train loss-0.2754, acc-0.9400, valid loss-0.4018, acc-0.8886, test loss-0.4060, acc-0.8857
Iter-67770, train loss-0.2870, acc-0.9400, valid loss-0.4018, acc-0.8884, test loss-0.4060, acc-0.8858
Iter-67780, train loss-0.2823, acc-0.9200, valid loss-0.4018, acc-0.8886, test loss-0.4060, acc-0.8857
Iter-67790, train loss-0.6262, acc-0.8600, valid loss-0.4018, acc-0.8886, test loss-0.4059, acc-0.8858
Iter-67800, train loss-0.3828, acc-0.9000, valid loss-0.4017, acc-0.8886, test loss-0.4059, acc-0.8857
Iter-67810, train loss-0.4813, acc-0.8800, valid loss-0.4017, acc-0.8886, test loss-0.4059, acc-0.8857
Iter-67820, train loss-0.4120, acc-0.8800, valid loss-0.4017, acc-0.8886, test loss-0.4059, acc-0.8856
Iter-67830, train loss-0.2618, acc-0.9400, valid loss-0.4017, acc-0.8886, test loss-0.4058, acc-0.8856
Iter-67840, train loss-0.5398, acc-0.8800, valid loss-0.4017, acc-0.8886, test loss-0.4058, acc-0.8856
Iter-67850, train loss-0.2521, acc-0.9400, valid loss-0.4016, acc-0.8886, test loss-0.4058, acc-0.8857
Iter-67860, train loss-0.2771, acc-0.9600, valid loss-0.4016, acc-0.8886, test loss-0.4058, acc-0.8857
Iter-67870, train loss-0.3469, acc-0.9400, valid loss-0.4016, acc-0.8886, test loss-0.4057, acc-0.8857
Iter-67880, train loss-0.5864, acc-0.8000, valid loss-0.4016, acc-0.8888, test loss-0.4057, acc-0.8856
Iter-67890, train loss-0.5024, acc-0.9000, valid loss-0.4016, acc-0.8886, test loss-0.4057, acc-0.8857
Iter-67900, train loss-0.4264, acc-0.8600, valid loss-0.4015, acc-0.8886, test loss-0.4057, acc-0.8857
Iter-67910, train loss-0.1961, acc-0.9800, valid loss-0.4015, acc-0.8886, test loss-0.4056, acc-0.8856
Iter-67920, train loss-0.4576, acc-0.9000, valid loss-0.4015, acc-0.8886, test loss-0.4056, acc-0.8858
Iter-67930, train loss-0.3485, acc-0.8800, valid loss-0.4015, acc-0.8888, test loss-0.4056, acc-0.8858
Iter-67940, train loss-0.2883, acc-0.9000, valid loss-0.4014, acc-0.8886, test loss-0.4056, acc-0.8857
Iter-67950, train loss-0.3498, acc-0.9000, valid loss-0.4014, acc-0.8886, test loss-0.4056, acc-0.8858
Iter-67960, train loss-0.3745, acc-0.9200, valid loss-0.4014, acc-0.8886, test loss-0.4055, acc-0.8859
Iter-67970, train loss-0.3254, acc-0.9000, valid loss-0.4014, acc-0.8888, test loss-0.4055, acc-0.8858
Iter-67980, train loss-0.6580, acc-0.7800, valid loss-0.4013, acc-0.8888, test loss-0.4055, acc-0.8859
Iter-67990, train loss-0.5180, acc-0.8400, valid loss-0.4013, acc-0.8886, test loss-0.4055, acc-0.8859
Iter-68000, train loss-0.4608, acc-0.8600, valid loss-0.4013, acc-0.8886, test loss-0.4054, acc-0.8859
Iter-68010, train loss-0.2054, acc-0.9400, valid loss-0.4012, acc-0.8886, test loss-0.4054, acc-0.8859
Iter-68020, train loss-0.3014, acc-0.9400, valid loss-0.4012, acc-0.8888, test loss-0.4054, acc-0.8859
Iter-68030, train loss-0.6541, acc-0.8000, valid loss-0.4012, acc-0.8886, test loss-0.4054, acc-0.8859
Iter-68040, train loss-0.5120, acc-0.8200, valid loss-0.4012, acc-0.8888, test loss-0.4054, acc-0.8859
Iter-68050, train loss-0.2619, acc-0.9200, valid loss-0.4011, acc-0.8886, test loss-0.4053, acc-0.8859
Iter-68060, train loss-0.3373, acc-0.9400, valid loss-0.4011, acc-0.8886, test loss-0.4053, acc-0.8859
Iter-68070, train loss-0.2987, acc-0.9200, valid loss-0.4011, acc-0.8886, test loss-0.4053, acc-0.8860
Iter-68080, train loss-0.4319, acc-0.9000, valid loss-0.4011, acc-0.8886, test loss-0.4053, acc-0.8859
Iter-68090, train loss-0.2288, acc-0.9400, valid loss-0.4010, acc-0.8886, test loss-0.4052, acc-0.8859
Iter-68100, train loss-0.4110, acc-0.9200, valid loss-0.4010, acc-0.8886, test loss-0.4052, acc-0.8859
Iter-68110, train loss-0.4386, acc-0.9000, valid loss-0.4010, acc-0.8886, test loss-0.4052, acc-0.8860
Iter-68120, train loss-0.6698, acc-0.8400, valid loss-0.4010, acc-0.8886, test loss-0.4052, acc-0.8860
Iter-68130, train loss-0.4081, acc-0.8600, valid loss-0.4009, acc-0.8886, test loss-0.4051, acc-0.8860
Iter-68140, train loss-0.4324, acc-0.8400, valid loss-0.4009, acc-0.8888, test loss-0.4051, acc-0.8861
Iter-68150, train loss-0.3145, acc-0.8800, valid loss-0.4009, acc-0.8890, test loss-0.4051, acc-0.8861
Iter-68160, train loss-0.4550, acc-0.8600, valid loss-0.4009, acc-0.8890, test loss-0.4051, acc-0.8861
Iter-68170, train loss-0.4227, acc-0.9000, valid loss-0.4009, acc-0.8890, test loss-0.4050, acc-0.8861
Iter-68180, train loss-0.4940, acc-0.8000, valid loss-0.4008, acc-0.8890, test loss-0.4050, acc-0.8861
Iter-68190, train loss-0.3067, acc-0.9400, valid loss-0.4008, acc-0.8888, test loss-0.4050, acc-0.8860
Iter-68200, train loss-0.5278, acc-0.8400, valid loss-0.4008, acc-0.8888, test loss-0.4050, acc-0.8860
Iter-68210, train loss-0.3638, acc-0.8800, valid loss-0.4007, acc-0.8888, test loss-0.4049, acc-0.8860
Iter-68220, train loss-0.4117, acc-0.9000, valid loss-0.4007, acc-0.8888, test loss-0.4049, acc-0.8860
Iter-68230, train loss-0.5672, acc-0.8400, valid loss-0.4007, acc-0.8888, test loss-0.4049, acc-0.8860
Iter-68240, train loss-0.3063, acc-0.9400, valid loss-0.4007, acc-0.8888, test loss-0.4049, acc-0.8860
Iter-68250, train loss-0.4823, acc-0.9000, valid loss-0.4007, acc-0.8888, test loss-0.4049, acc-0.8860
Iter-68260, train loss-0.2977, acc-0.9000, valid loss-0.4006, acc-0.8888, test loss-0.4049, acc-0.8860
Iter-68270, train loss-0.4546, acc-0.8600, valid loss-0.4006, acc-0.8888, test loss-0.4048, acc-0.8861
Iter-68280, train loss-0.3766, acc-0.9200, valid loss-0.4005, acc-0.8888, test loss-0.4048, acc-0.8861
Iter-68290, train loss-0.4341, acc-0.8200, valid loss-0.4005, acc-0.8888, test loss-0.4048, acc-0.8861
Iter-68300, train loss-0.5093, acc-0.9000, valid loss-0.4005, acc-0.8886, test loss-0.4048, acc-0.8861
Iter-68310, train loss-0.3957, acc-0.9200, valid loss-0.4005, acc-0.8888, test loss-0.4047, acc-0.8861
Iter-68320, train loss-0.4578, acc-0.8800, valid loss-0.4005, acc-0.8888, test loss-0.4047, acc-0.8861
Iter-68330, train loss-0.2986, acc-0.9600, valid loss-0.4004, acc-0.8886, test loss-0.4047, acc-0.8861
Iter-68340, train loss-0.3489, acc-0.8800, valid loss-0.4004, acc-0.8886, test loss-0.4047, acc-0.8861
Iter-68350, train loss-0.6161, acc-0.7600, valid loss-0.4004, acc-0.8886, test loss-0.4047, acc-0.8862
Iter-68360, train loss-0.2103, acc-0.9600, valid loss-0.4004, acc-0.8886, test loss-0.4046, acc-0.8861
Iter-68370, train loss-0.4937, acc-0.9000, valid loss-0.4003, acc-0.8886, test loss-0.4046, acc-0.8863
Iter-68380, train loss-0.3384, acc-0.9000, valid loss-0.4003, acc-0.8886, test loss-0.4046, acc-0.8863
Iter-68390, train loss-0.6976, acc-0.7600, valid loss-0.4003, acc-0.8886, test loss-0.4046, acc-0.8863
Iter-68400, train loss-0.4887, acc-0.8800, valid loss-0.4002, acc-0.8886, test loss-0.4046, acc-0.8862
Iter-68410, train loss-0.6160, acc-0.7600, valid loss-0.4002, acc-0.8886, test loss-0.4045, acc-0.8863
Iter-68420, train loss-0.3770, acc-0.9400, valid loss-0.4002, acc-0.8886, test loss-0.4045, acc-0.8862
Iter-68430, train loss-0.2599, acc-0.9600, valid loss-0.4002, acc-0.8886, test loss-0.4045, acc-0.8862
Iter-68440, train loss-0.3074, acc-0.9200, valid loss-0.4002, acc-0.8886, test loss-0.4045, acc-0.8862
Iter-68450, train loss-0.3871, acc-0.8800, valid loss-0.4002, acc-0.8886, test loss-0.4044, acc-0.8863
Iter-68460, train loss-0.5563, acc-0.8400, valid loss-0.4001, acc-0.8886, test loss-0.4044, acc-0.8863
Iter-68470, train loss-0.3502, acc-0.9000, valid loss-0.4001, acc-0.8886, test loss-0.4044, acc-0.8863
Iter-68480, train loss-0.4363, acc-0.8000, valid loss-0.4001, acc-0.8886, test loss-0.4044, acc-0.8862
Iter-68490, train loss-0.6351, acc-0.8600, valid loss-0.4001, acc-0.8886, test loss-0.4043, acc-0.8862
Iter-68500, train loss-0.4929, acc-0.8400, valid loss-0.4001, acc-0.8886, test loss-0.4043, acc-0.8862
Iter-68510, train loss-0.4689, acc-0.9000, valid loss-0.4000, acc-0.8886, test loss-0.4043, acc-0.8862
Iter-68520, train loss-0.3069, acc-0.9200, valid loss-0.4000, acc-0.8886, test loss-0.4043, acc-0.8862
Iter-68530, train loss-0.3438, acc-0.9000, valid loss-0.4000, acc-0.8886, test loss-0.4043, acc-0.8863
Iter-68540, train loss-0.3316, acc-0.9200, valid loss-0.4000, acc-0.8886, test loss-0.4042, acc-0.8863
Iter-68550, train loss-0.4010, acc-0.8800, valid loss-0.3999, acc-0.8886, test loss-0.4042, acc-0.8863
Iter-68560, train loss-0.5570, acc-0.8800, valid loss-0.3999, acc-0.8886, test loss-0.4042, acc-0.8864
Iter-68570, train loss-0.3538, acc-0.9400, valid loss-0.3999, acc-0.8886, test loss-0.4042, acc-0.8863
Iter-68580, train loss-0.5757, acc-0.8600, valid loss-0.3999, acc-0.8888, test loss-0.4042, acc-0.8863
Iter-68590, train loss-0.3443, acc-0.9200, valid loss-0.3999, acc-0.8888, test loss-0.4041, acc-0.8863
Iter-68600, train loss-0.4388, acc-0.8800, valid loss-0.3998, acc-0.8886, test loss-0.4041, acc-0.8863
Iter-68610, train loss-0.5237, acc-0.8400, valid loss-0.3998, acc-0.8886, test loss-0.4041, acc-0.8863
Iter-68620, train loss-0.4391, acc-0.9000, valid loss-0.3998, acc-0.8888, test loss-0.4041, acc-0.8863
Iter-68630, train loss-0.5424, acc-0.8400, valid loss-0.3998, acc-0.8886, test loss-0.4040, acc-0.8863
Iter-68640, train loss-0.5436, acc-0.8400, valid loss-0.3997, acc-0.8888, test loss-0.4040, acc-0.8863
Iter-68650, train loss-0.3309, acc-0.9000, valid loss-0.3997, acc-0.8888, test loss-0.4040, acc-0.8863
Iter-68660, train loss-0.4132, acc-0.8800, valid loss-0.3997, acc-0.8888, test loss-0.4040, acc-0.8863
Iter-68670, train loss-0.3591, acc-0.8600, valid loss-0.3997, acc-0.8888, test loss-0.4039, acc-0.8863
Iter-68680, train loss-0.7496, acc-0.8200, valid loss-0.3996, acc-0.8888, test loss-0.4039, acc-0.8863
Iter-68690, train loss-0.4227, acc-0.8800, valid loss-0.3996, acc-0.8888, test loss-0.4039, acc-0.8863
Iter-68700, train loss-0.5920, acc-0.8000, valid loss-0.3996, acc-0.8888, test loss-0.4039, acc-0.8863
Iter-68710, train loss-0.3741, acc-0.9000, valid loss-0.3996, acc-0.8888, test loss-0.4039, acc-0.8863
Iter-68720, train loss-0.3714, acc-0.9200, valid loss-0.3996, acc-0.8888, test loss-0.4038, acc-0.8863
Iter-68730, train loss-0.4749, acc-0.8600, valid loss-0.3995, acc-0.8888, test loss-0.4038, acc-0.8862
Iter-68740, train loss-0.2767, acc-0.9400, valid loss-0.3995, acc-0.8888, test loss-0.4038, acc-0.8863
Iter-68750, train loss-0.4178, acc-0.8800, valid loss-0.3995, acc-0.8888, test loss-0.4038, acc-0.8864
Iter-68760, train loss-0.3992, acc-0.9000, valid loss-0.3995, acc-0.8888, test loss-0.4038, acc-0.8864
Iter-68770, train loss-0.3403, acc-0.9600, valid loss-0.3995, acc-0.8890, test loss-0.4037, acc-0.8864
Iter-68780, train loss-0.3913, acc-0.8800, valid loss-0.3995, acc-0.8890, test loss-0.4037, acc-0.8864
Iter-68790, train loss-0.2271, acc-0.9600, valid loss-0.3994, acc-0.8888, test loss-0.4037, acc-0.8866
Iter-68800, train loss-0.3181, acc-0.8800, valid loss-0.3994, acc-0.8888, test loss-0.4037, acc-0.8864
Iter-68810, train loss-0.3813, acc-0.8800, valid loss-0.3994, acc-0.8888, test loss-0.4037, acc-0.8864
Iter-68820, train loss-0.3989, acc-0.8800, valid loss-0.3993, acc-0.8888, test loss-0.4036, acc-0.8863
Iter-68830, train loss-0.3853, acc-0.9400, valid loss-0.3993, acc-0.8888, test loss-0.4036, acc-0.8864
Iter-68840, train loss-0.4226, acc-0.8600, valid loss-0.3993, acc-0.8888, test loss-0.4036, acc-0.8863
Iter-68850, train loss-0.4061, acc-0.9200, valid loss-0.3993, acc-0.8886, test loss-0.4036, acc-0.8865
Iter-68860, train loss-0.3405, acc-0.8600, valid loss-0.3993, acc-0.8890, test loss-0.4036, acc-0.8865
Iter-68870, train loss-0.4155, acc-0.9000, valid loss-0.3992, acc-0.8890, test loss-0.4035, acc-0.8864
Iter-68880, train loss-0.4372, acc-0.8200, valid loss-0.3992, acc-0.8888, test loss-0.4035, acc-0.8863
Iter-68890, train loss-0.3863, acc-0.8600, valid loss-0.3992, acc-0.8890, test loss-0.4035, acc-0.8864
Iter-68900, train loss-0.4832, acc-0.8400, valid loss-0.3992, acc-0.8890, test loss-0.4035, acc-0.8865
Iter-68910, train loss-0.5021, acc-0.8800, valid loss-0.3991, acc-0.8888, test loss-0.4034, acc-0.8865
Iter-68920, train loss-0.6363, acc-0.8200, valid loss-0.3991, acc-0.8890, test loss-0.4034, acc-0.8865
Iter-68930, train loss-0.4821, acc-0.8600, valid loss-0.3991, acc-0.8888, test loss-0.4034, acc-0.8865
Iter-68940, train loss-0.3549, acc-0.9200, valid loss-0.3991, acc-0.8888, test loss-0.4034, acc-0.8865
Iter-68950, train loss-0.3739, acc-0.8800, valid loss-0.3990, acc-0.8888, test loss-0.4033, acc-0.8865
Iter-68960, train loss-0.3138, acc-0.8800, valid loss-0.3990, acc-0.8890, test loss-0.4033, acc-0.8865
Iter-68970, train loss-0.4585, acc-0.8200, valid loss-0.3990, acc-0.8886, test loss-0.4033, acc-0.8864
Iter-68980, train loss-0.4375, acc-0.9400, valid loss-0.3990, acc-0.8886, test loss-0.4033, acc-0.8864
Iter-68990, train loss-0.3338, acc-0.9200, valid loss-0.3990, acc-0.8886, test loss-0.4032, acc-0.8864
Iter-69000, train loss-0.6790, acc-0.8000, valid loss-0.3989, acc-0.8886, test loss-0.4032, acc-0.8865
Iter-69010, train loss-0.4612, acc-0.8000, valid loss-0.3989, acc-0.8886, test loss-0.4032, acc-0.8866
Iter-69020, train loss-0.3593, acc-0.8600, valid loss-0.3989, acc-0.8886, test loss-0.4032, acc-0.8866
Iter-69030, train loss-0.4615, acc-0.8800, valid loss-0.3988, acc-0.8886, test loss-0.4031, acc-0.8866
Iter-69040, train loss-0.3750, acc-0.8400, valid loss-0.3988, acc-0.8886, test loss-0.4031, acc-0.8865
Iter-69050, train loss-0.4989, acc-0.8600, valid loss-0.3988, acc-0.8886, test loss-0.4031, acc-0.8864
Iter-69060, train loss-0.4807, acc-0.8200, valid loss-0.3988, acc-0.8886, test loss-0.4031, acc-0.8866
Iter-69070, train loss-0.3419, acc-0.9200, valid loss-0.3988, acc-0.8886, test loss-0.4031, acc-0.8865
Iter-69080, train loss-0.3408, acc-0.9400, valid loss-0.3988, acc-0.8886, test loss-0.4031, acc-0.8865
Iter-69090, train loss-0.3281, acc-0.9000, valid loss-0.3987, acc-0.8886, test loss-0.4030, acc-0.8864
Iter-69100, train loss-0.3748, acc-0.8600, valid loss-0.3987, acc-0.8886, test loss-0.4030, acc-0.8864
Iter-69110, train loss-0.3304, acc-0.9200, valid loss-0.3987, acc-0.8886, test loss-0.4030, acc-0.8864
Iter-69120, train loss-0.4021, acc-0.8400, valid loss-0.3987, acc-0.8886, test loss-0.4030, acc-0.8864
Iter-69130, train loss-0.4579, acc-0.8800, valid loss-0.3987, acc-0.8886, test loss-0.4030, acc-0.8863
Iter-69140, train loss-0.5836, acc-0.8200, valid loss-0.3986, acc-0.8886, test loss-0.4029, acc-0.8865
Iter-69150, train loss-0.3443, acc-0.9400, valid loss-0.3986, acc-0.8886, test loss-0.4029, acc-0.8864
Iter-69160, train loss-0.5045, acc-0.8600, valid loss-0.3986, acc-0.8886, test loss-0.4029, acc-0.8864
Iter-69170, train loss-0.7364, acc-0.7800, valid loss-0.3986, acc-0.8886, test loss-0.4029, acc-0.8864
Iter-69180, train loss-0.4964, acc-0.8600, valid loss-0.3986, acc-0.8886, test loss-0.4029, acc-0.8864
Iter-69190, train loss-0.6002, acc-0.8000, valid loss-0.3985, acc-0.8886, test loss-0.4028, acc-0.8864
Iter-69200, train loss-0.3840, acc-0.9000, valid loss-0.3985, acc-0.8886, test loss-0.4028, acc-0.8865
Iter-69210, train loss-0.3576, acc-0.9000, valid loss-0.3985, acc-0.8886, test loss-0.4028, acc-0.8865
Iter-69220, train loss-0.4454, acc-0.8600, valid loss-0.3985, acc-0.8886, test loss-0.4028, acc-0.8864
Iter-69230, train loss-0.4242, acc-0.8600, valid loss-0.3984, acc-0.8886, test loss-0.4027, acc-0.8864
Iter-69240, train loss-0.6487, acc-0.8000, valid loss-0.3984, acc-0.8886, test loss-0.4027, acc-0.8864
Iter-69250, train loss-0.2684, acc-0.9200, valid loss-0.3984, acc-0.8886, test loss-0.4027, acc-0.8863
Iter-69260, train loss-0.3575, acc-0.9000, valid loss-0.3984, acc-0.8886, test loss-0.4027, acc-0.8863
Iter-69270, train loss-0.5616, acc-0.8200, valid loss-0.3984, acc-0.8886, test loss-0.4027, acc-0.8863
Iter-69280, train loss-0.5751, acc-0.8000, valid loss-0.3983, acc-0.8886, test loss-0.4026, acc-0.8863
Iter-69290, train loss-0.3080, acc-0.8800, valid loss-0.3983, acc-0.8886, test loss-0.4026, acc-0.8865
Iter-69300, train loss-0.4173, acc-0.8600, valid loss-0.3983, acc-0.8886, test loss-0.4026, acc-0.8866
Iter-69310, train loss-0.3703, acc-0.9000, valid loss-0.3983, acc-0.8886, test loss-0.4026, acc-0.8865
Iter-69320, train loss-0.2394, acc-0.9400, valid loss-0.3982, acc-0.8886, test loss-0.4026, acc-0.8865
Iter-69330, train loss-0.5192, acc-0.8000, valid loss-0.3982, acc-0.8888, test loss-0.4025, acc-0.8865
Iter-69340, train loss-0.5875, acc-0.8600, valid loss-0.3981, acc-0.8886, test loss-0.4025, acc-0.8866
Iter-69350, train loss-0.3188, acc-0.9200, valid loss-0.3981, acc-0.8888, test loss-0.4025, acc-0.8865
Iter-69360, train loss-0.3944, acc-0.9000, valid loss-0.3981, acc-0.8888, test loss-0.4025, acc-0.8865
Iter-69370, train loss-0.4140, acc-0.9000, valid loss-0.3981, acc-0.8888, test loss-0.4024, acc-0.8866
Iter-69380, train loss-0.4559, acc-0.8400, valid loss-0.3981, acc-0.8888, test loss-0.4024, acc-0.8866
Iter-69390, train loss-0.4319, acc-0.9000, valid loss-0.3980, acc-0.8888, test loss-0.4024, acc-0.8865
Iter-69400, train loss-0.4429, acc-0.8600, valid loss-0.3980, acc-0.8888, test loss-0.4024, acc-0.8864
Iter-69410, train loss-0.3686, acc-0.8800, valid loss-0.3980, acc-0.8888, test loss-0.4023, acc-0.8865
Iter-69420, train loss-0.4749, acc-0.8400, valid loss-0.3980, acc-0.8888, test loss-0.4023, acc-0.8866
Iter-69430, train loss-0.4121, acc-0.9000, valid loss-0.3980, acc-0.8888, test loss-0.4023, acc-0.8865
Iter-69440, train loss-0.3858, acc-0.9200, valid loss-0.3979, acc-0.8890, test loss-0.4023, acc-0.8864
Iter-69450, train loss-0.4059, acc-0.8800, valid loss-0.3979, acc-0.8890, test loss-0.4023, acc-0.8865
Iter-69460, train loss-0.3203, acc-0.9200, valid loss-0.3979, acc-0.8890, test loss-0.4022, acc-0.8867
Iter-69470, train loss-0.3320, acc-0.9400, valid loss-0.3979, acc-0.8888, test loss-0.4022, acc-0.8867
Iter-69480, train loss-0.5154, acc-0.8600, valid loss-0.3978, acc-0.8890, test loss-0.4022, acc-0.8868
Iter-69490, train loss-0.3393, acc-0.9200, valid loss-0.3978, acc-0.8890, test loss-0.4022, acc-0.8867
Iter-69500, train loss-0.4125, acc-0.9000, valid loss-0.3978, acc-0.8890, test loss-0.4021, acc-0.8867
Iter-69510, train loss-0.3798, acc-0.8800, valid loss-0.3977, acc-0.8890, test loss-0.4021, acc-0.8867
Iter-69520, train loss-0.3000, acc-0.9200, valid loss-0.3977, acc-0.8892, test loss-0.4021, acc-0.8868
Iter-69530, train loss-0.4938, acc-0.8400, valid loss-0.3977, acc-0.8890, test loss-0.4021, acc-0.8868
Iter-69540, train loss-0.3212, acc-0.9400, valid loss-0.3977, acc-0.8888, test loss-0.4021, acc-0.8868
Iter-69550, train loss-0.3168, acc-0.9400, valid loss-0.3977, acc-0.8888, test loss-0.4020, acc-0.8868
Iter-69560, train loss-0.4464, acc-0.8800, valid loss-0.3977, acc-0.8888, test loss-0.4020, acc-0.8868
Iter-69570, train loss-0.4848, acc-0.8800, valid loss-0.3976, acc-0.8888, test loss-0.4020, acc-0.8868
Iter-69580, train loss-0.4302, acc-0.8800, valid loss-0.3976, acc-0.8888, test loss-0.4020, acc-0.8866
Iter-69590, train loss-0.4691, acc-0.9200, valid loss-0.3976, acc-0.8890, test loss-0.4019, acc-0.8866
Iter-69600, train loss-0.5592, acc-0.7800, valid loss-0.3976, acc-0.8890, test loss-0.4019, acc-0.8866
Iter-69610, train loss-0.2743, acc-0.9400, valid loss-0.3975, acc-0.8892, test loss-0.4019, acc-0.8865
Iter-69620, train loss-0.5105, acc-0.8400, valid loss-0.3975, acc-0.8892, test loss-0.4019, acc-0.8865
Iter-69630, train loss-0.3930, acc-0.9000, valid loss-0.3975, acc-0.8892, test loss-0.4018, acc-0.8865
Iter-69640, train loss-0.3787, acc-0.8600, valid loss-0.3975, acc-0.8890, test loss-0.4018, acc-0.8865
Iter-69650, train loss-0.4435, acc-0.8400, valid loss-0.3974, acc-0.8890, test loss-0.4018, acc-0.8865
Iter-69660, train loss-0.3605, acc-0.8800, valid loss-0.3974, acc-0.8890, test loss-0.4018, acc-0.8864
Iter-69670, train loss-0.4520, acc-0.9000, valid loss-0.3974, acc-0.8890, test loss-0.4018, acc-0.8863
Iter-69680, train loss-0.3845, acc-0.8600, valid loss-0.3974, acc-0.8890, test loss-0.4017, acc-0.8863
Iter-69690, train loss-0.6726, acc-0.8200, valid loss-0.3974, acc-0.8890, test loss-0.4017, acc-0.8864
Iter-69700, train loss-0.2228, acc-0.9600, valid loss-0.3974, acc-0.8890, test loss-0.4017, acc-0.8864
Iter-69710, train loss-0.2933, acc-0.8800, valid loss-0.3973, acc-0.8890, test loss-0.4017, acc-0.8864
Iter-69720, train loss-0.5502, acc-0.8000, valid loss-0.3973, acc-0.8890, test loss-0.4017, acc-0.8864
Iter-69730, train loss-0.3080, acc-0.8800, valid loss-0.3973, acc-0.8890, test loss-0.4016, acc-0.8866
Iter-69740, train loss-0.4822, acc-0.8400, valid loss-0.3973, acc-0.8892, test loss-0.4016, acc-0.8867
Iter-69750, train loss-0.3040, acc-0.9000, valid loss-0.3972, acc-0.8892, test loss-0.4016, acc-0.8867
Iter-69760, train loss-0.3139, acc-0.9000, valid loss-0.3972, acc-0.8892, test loss-0.4016, acc-0.8867
Iter-69770, train loss-0.4608, acc-0.9000, valid loss-0.3972, acc-0.8892, test loss-0.4016, acc-0.8867
Iter-69780, train loss-0.3590, acc-0.9200, valid loss-0.3972, acc-0.8892, test loss-0.4015, acc-0.8867
Iter-69790, train loss-0.5288, acc-0.8200, valid loss-0.3972, acc-0.8892, test loss-0.4015, acc-0.8867
Iter-69800, train loss-0.2817, acc-0.9800, valid loss-0.3971, acc-0.8892, test loss-0.4015, acc-0.8866
Iter-69810, train loss-0.5608, acc-0.8000, valid loss-0.3971, acc-0.8892, test loss-0.4015, acc-0.8865
Iter-69820, train loss-0.4668, acc-0.8200, valid loss-0.3971, acc-0.8892, test loss-0.4015, acc-0.8866
Iter-69830, train loss-0.4159, acc-0.8800, valid loss-0.3971, acc-0.8892, test loss-0.4015, acc-0.8866
Iter-69840, train loss-0.3915, acc-0.9000, valid loss-0.3970, acc-0.8892, test loss-0.4014, acc-0.8867
Iter-69850, train loss-0.5479, acc-0.8000, valid loss-0.3970, acc-0.8892, test loss-0.4014, acc-0.8867
Iter-69860, train loss-0.4263, acc-0.8600, valid loss-0.3970, acc-0.8892, test loss-0.4014, acc-0.8867
Iter-69870, train loss-0.2478, acc-0.9200, valid loss-0.3970, acc-0.8892, test loss-0.4014, acc-0.8867
Iter-69880, train loss-0.2989, acc-0.9000, valid loss-0.3970, acc-0.8892, test loss-0.4014, acc-0.8866
Iter-69890, train loss-0.3317, acc-0.9200, valid loss-0.3970, acc-0.8892, test loss-0.4013, acc-0.8866
Iter-69900, train loss-0.3503, acc-0.9000, valid loss-0.3970, acc-0.8892, test loss-0.4013, acc-0.8866
Iter-69910, train loss-0.2823, acc-0.9000, valid loss-0.3969, acc-0.8892, test loss-0.4013, acc-0.8866
Iter-69920, train loss-0.4151, acc-0.8600, valid loss-0.3969, acc-0.8894, test loss-0.4013, acc-0.8866
Iter-69930, train loss-0.3427, acc-0.9000, valid loss-0.3969, acc-0.8894, test loss-0.4012, acc-0.8864
Iter-69940, train loss-0.2924, acc-0.9200, valid loss-0.3969, acc-0.8894, test loss-0.4012, acc-0.8864
Iter-69950, train loss-0.3798, acc-0.9200, valid loss-0.3969, acc-0.8894, test loss-0.4012, acc-0.8865
Iter-69960, train loss-0.4822, acc-0.8800, valid loss-0.3968, acc-0.8894, test loss-0.4012, acc-0.8864
Iter-69970, train loss-0.3885, acc-0.9200, valid loss-0.3968, acc-0.8894, test loss-0.4011, acc-0.8864
Iter-69980, train loss-0.2751, acc-0.9400, valid loss-0.3968, acc-0.8894, test loss-0.4011, acc-0.8864
Iter-69990, train loss-0.5259, acc-0.8000, valid loss-0.3968, acc-0.8894, test loss-0.4011, acc-0.8864
Iter-70000, train loss-0.4284, acc-0.8800, valid loss-0.3967, acc-0.8894, test loss-0.4011, acc-0.8865
Iter-70010, train loss-0.4647, acc-0.8800, valid loss-0.3967, acc-0.8894, test loss-0.4011, acc-0.8864
Iter-70020, train loss-0.2620, acc-0.9600, valid loss-0.3967, acc-0.8896, test loss-0.4011, acc-0.8864
Iter-70030, train loss-0.4000, acc-0.9000, valid loss-0.3967, acc-0.8894, test loss-0.4010, acc-0.8866
Iter-70040, train loss-0.3188, acc-0.9200, valid loss-0.3967, acc-0.8896, test loss-0.4010, acc-0.8866
Iter-70050, train loss-0.2210, acc-0.9200, valid loss-0.3966, acc-0.8894, test loss-0.4010, acc-0.8865
Iter-70060, train loss-0.3600, acc-0.8400, valid loss-0.3966, acc-0.8894, test loss-0.4010, acc-0.8866
Iter-70070, train loss-0.3994, acc-0.9000, valid loss-0.3966, acc-0.8894, test loss-0.4009, acc-0.8866
Iter-70080, train loss-0.3121, acc-0.9200, valid loss-0.3966, acc-0.8896, test loss-0.4009, acc-0.8866
Iter-70090, train loss-0.2917, acc-0.8800, valid loss-0.3966, acc-0.8894, test loss-0.4009, acc-0.8866
Iter-70100, train loss-0.4886, acc-0.8600, valid loss-0.3965, acc-0.8894, test loss-0.4009, acc-0.8866
Iter-70110, train loss-0.3902, acc-0.8600, valid loss-0.3965, acc-0.8894, test loss-0.4008, acc-0.8866
Iter-70120, train loss-0.2848, acc-0.9400, valid loss-0.3965, acc-0.8894, test loss-0.4008, acc-0.8866
Iter-70130, train loss-0.5085, acc-0.8600, valid loss-0.3964, acc-0.8894, test loss-0.4008, acc-0.8866
Iter-70140, train loss-0.3529, acc-0.9200, valid loss-0.3964, acc-0.8894, test loss-0.4008, acc-0.8866
Iter-70150, train loss-0.4318, acc-0.8800, valid loss-0.3964, acc-0.8894, test loss-0.4007, acc-0.8866
Iter-70160, train loss-0.5016, acc-0.9000, valid loss-0.3964, acc-0.8894, test loss-0.4007, acc-0.8866
Iter-70170, train loss-0.3949, acc-0.8600, valid loss-0.3964, acc-0.8894, test loss-0.4007, acc-0.8865
Iter-70180, train loss-0.3838, acc-0.8600, valid loss-0.3964, acc-0.8894, test loss-0.4007, acc-0.8866
Iter-70190, train loss-0.3384, acc-0.9200, valid loss-0.3964, acc-0.8894, test loss-0.4007, acc-0.8866
Iter-70200, train loss-0.4704, acc-0.8800, valid loss-0.3963, acc-0.8896, test loss-0.4006, acc-0.8866
Iter-70210, train loss-0.6205, acc-0.8200, valid loss-0.3963, acc-0.8896, test loss-0.4006, acc-0.8866
Iter-70220, train loss-0.5064, acc-0.8800, valid loss-0.3963, acc-0.8894, test loss-0.4006, acc-0.8865
Iter-70230, train loss-0.3478, acc-0.9200, valid loss-0.3963, acc-0.8894, test loss-0.4006, acc-0.8866
Iter-70240, train loss-0.4306, acc-0.8600, valid loss-0.3962, acc-0.8896, test loss-0.4005, acc-0.8866
Iter-70250, train loss-0.3958, acc-0.8800, valid loss-0.3962, acc-0.8896, test loss-0.4005, acc-0.8866
Iter-70260, train loss-0.4324, acc-0.8800, valid loss-0.3962, acc-0.8896, test loss-0.4005, acc-0.8866
Iter-70270, train loss-0.2984, acc-0.9200, valid loss-0.3962, acc-0.8896, test loss-0.4005, acc-0.8866
Iter-70280, train loss-0.3811, acc-0.9000, valid loss-0.3961, acc-0.8896, test loss-0.4005, acc-0.8866
Iter-70290, train loss-0.2869, acc-0.9400, valid loss-0.3961, acc-0.8894, test loss-0.4004, acc-0.8866
Iter-70300, train loss-0.3713, acc-0.9400, valid loss-0.3961, acc-0.8894, test loss-0.4004, acc-0.8866
Iter-70310, train loss-0.3395, acc-0.9000, valid loss-0.3961, acc-0.8894, test loss-0.4004, acc-0.8866
Iter-70320, train loss-0.3592, acc-0.9200, valid loss-0.3960, acc-0.8894, test loss-0.4004, acc-0.8865
Iter-70330, train loss-0.3918, acc-0.8800, valid loss-0.3960, acc-0.8896, test loss-0.4003, acc-0.8866
Iter-70340, train loss-0.4215, acc-0.8800, valid loss-0.3960, acc-0.8896, test loss-0.4003, acc-0.8866
Iter-70350, train loss-0.5302, acc-0.8200, valid loss-0.3960, acc-0.8894, test loss-0.4003, acc-0.8866
Iter-70360, train loss-0.4899, acc-0.8400, valid loss-0.3959, acc-0.8896, test loss-0.4003, acc-0.8865
Iter-70370, train loss-0.5373, acc-0.8200, valid loss-0.3959, acc-0.8896, test loss-0.4003, acc-0.8865
Iter-70380, train loss-0.4041, acc-0.9000, valid loss-0.3959, acc-0.8896, test loss-0.4002, acc-0.8865
Iter-70390, train loss-0.3390, acc-0.9000, valid loss-0.3959, acc-0.8894, test loss-0.4002, acc-0.8865
Iter-70400, train loss-0.3032, acc-0.9000, valid loss-0.3958, acc-0.8894, test loss-0.4002, acc-0.8865
Iter-70410, train loss-0.3741, acc-0.9000, valid loss-0.3958, acc-0.8894, test loss-0.4002, acc-0.8865
Iter-70420, train loss-0.4027, acc-0.8600, valid loss-0.3958, acc-0.8894, test loss-0.4001, acc-0.8866
Iter-70430, train loss-0.2701, acc-0.9000, valid loss-0.3958, acc-0.8894, test loss-0.4001, acc-0.8865
Iter-70440, train loss-0.5003, acc-0.8600, valid loss-0.3957, acc-0.8894, test loss-0.4001, acc-0.8865
Iter-70450, train loss-0.4705, acc-0.8800, valid loss-0.3957, acc-0.8894, test loss-0.4001, acc-0.8865
Iter-70460, train loss-0.7950, acc-0.8400, valid loss-0.3957, acc-0.8894, test loss-0.4001, acc-0.8866
Iter-70470, train loss-0.4771, acc-0.8200, valid loss-0.3957, acc-0.8894, test loss-0.4000, acc-0.8865
Iter-70480, train loss-0.3889, acc-0.8800, valid loss-0.3956, acc-0.8894, test loss-0.4000, acc-0.8865
Iter-70490, train loss-0.5424, acc-0.8800, valid loss-0.3956, acc-0.8896, test loss-0.4000, acc-0.8865
Iter-70500, train loss-0.5685, acc-0.8400, valid loss-0.3956, acc-0.8896, test loss-0.4000, acc-0.8865
Iter-70510, train loss-0.4224, acc-0.9000, valid loss-0.3956, acc-0.8896, test loss-0.3999, acc-0.8865
Iter-70520, train loss-0.5197, acc-0.9200, valid loss-0.3956, acc-0.8896, test loss-0.3999, acc-0.8867
Iter-70530, train loss-0.5299, acc-0.8400, valid loss-0.3955, acc-0.8894, test loss-0.3999, acc-0.8867
Iter-70540, train loss-0.3636, acc-0.9000, valid loss-0.3955, acc-0.8896, test loss-0.3999, acc-0.8868
Iter-70550, train loss-0.3904, acc-0.9400, valid loss-0.3955, acc-0.8894, test loss-0.3999, acc-0.8868
Iter-70560, train loss-0.2838, acc-0.9600, valid loss-0.3955, acc-0.8894, test loss-0.3999, acc-0.8869
Iter-70570, train loss-0.4434, acc-0.8800, valid loss-0.3955, acc-0.8896, test loss-0.3998, acc-0.8868
Iter-70580, train loss-0.4995, acc-0.9000, valid loss-0.3954, acc-0.8896, test loss-0.3998, acc-0.8869
Iter-70590, train loss-0.3740, acc-0.9200, valid loss-0.3954, acc-0.8898, test loss-0.3998, acc-0.8868
Iter-70600, train loss-0.4250, acc-0.8400, valid loss-0.3954, acc-0.8900, test loss-0.3998, acc-0.8869
Iter-70610, train loss-0.4341, acc-0.9000, valid loss-0.3954, acc-0.8900, test loss-0.3997, acc-0.8868
Iter-70620, train loss-0.5331, acc-0.8600, valid loss-0.3953, acc-0.8900, test loss-0.3997, acc-0.8868
Iter-70630, train loss-0.2565, acc-0.9200, valid loss-0.3953, acc-0.8898, test loss-0.3997, acc-0.8869
Iter-70640, train loss-0.5296, acc-0.8200, valid loss-0.3953, acc-0.8900, test loss-0.3997, acc-0.8868
Iter-70650, train loss-0.4616, acc-0.8600, valid loss-0.3953, acc-0.8898, test loss-0.3997, acc-0.8867
Iter-70660, train loss-0.3675, acc-0.9000, valid loss-0.3953, acc-0.8900, test loss-0.3997, acc-0.8867
Iter-70670, train loss-0.3005, acc-0.9200, valid loss-0.3953, acc-0.8898, test loss-0.3996, acc-0.8867
Iter-70680, train loss-0.4187, acc-0.8600, valid loss-0.3953, acc-0.8900, test loss-0.3996, acc-0.8867
Iter-70690, train loss-0.3592, acc-0.9400, valid loss-0.3952, acc-0.8898, test loss-0.3996, acc-0.8867
Iter-70700, train loss-0.4887, acc-0.8800, valid loss-0.3952, acc-0.8898, test loss-0.3996, acc-0.8867
Iter-70710, train loss-0.4847, acc-0.8600, valid loss-0.3952, acc-0.8898, test loss-0.3995, acc-0.8867
Iter-70720, train loss-0.3836, acc-0.8800, valid loss-0.3952, acc-0.8900, test loss-0.3995, acc-0.8867
Iter-70730, train loss-0.4191, acc-0.8600, valid loss-0.3952, acc-0.8900, test loss-0.3995, acc-0.8867
Iter-70740, train loss-0.6801, acc-0.8400, valid loss-0.3951, acc-0.8900, test loss-0.3995, acc-0.8868
Iter-70750, train loss-0.4596, acc-0.8600, valid loss-0.3951, acc-0.8900, test loss-0.3994, acc-0.8868
Iter-70760, train loss-0.2090, acc-0.9600, valid loss-0.3951, acc-0.8900, test loss-0.3994, acc-0.8868
Iter-70770, train loss-0.9135, acc-0.7400, valid loss-0.3951, acc-0.8900, test loss-0.3994, acc-0.8868
Iter-70780, train loss-0.7465, acc-0.7400, valid loss-0.3950, acc-0.8900, test loss-0.3994, acc-0.8869
Iter-70790, train loss-0.4990, acc-0.8600, valid loss-0.3950, acc-0.8900, test loss-0.3994, acc-0.8869
Iter-70800, train loss-0.5125, acc-0.8400, valid loss-0.3950, acc-0.8900, test loss-0.3993, acc-0.8869
Iter-70810, train loss-0.3047, acc-0.9000, valid loss-0.3950, acc-0.8900, test loss-0.3993, acc-0.8869
Iter-70820, train loss-0.4742, acc-0.8000, valid loss-0.3950, acc-0.8900, test loss-0.3993, acc-0.8869
Iter-70830, train loss-0.3368, acc-0.9200, valid loss-0.3949, acc-0.8898, test loss-0.3993, acc-0.8869
Iter-70840, train loss-0.3824, acc-0.9200, valid loss-0.3949, acc-0.8900, test loss-0.3992, acc-0.8869
Iter-70850, train loss-0.3176, acc-0.9400, valid loss-0.3949, acc-0.8900, test loss-0.3992, acc-0.8869
Iter-70860, train loss-0.5937, acc-0.8000, valid loss-0.3949, acc-0.8900, test loss-0.3992, acc-0.8869
Iter-70870, train loss-0.3923, acc-0.9200, valid loss-0.3948, acc-0.8900, test loss-0.3992, acc-0.8869
Iter-70880, train loss-0.2556, acc-0.9600, valid loss-0.3948, acc-0.8900, test loss-0.3992, acc-0.8869
Iter-70890, train loss-0.3244, acc-0.9200, valid loss-0.3948, acc-0.8900, test loss-0.3991, acc-0.8868
Iter-70900, train loss-0.5528, acc-0.8600, valid loss-0.3948, acc-0.8900, test loss-0.3991, acc-0.8868
Iter-70910, train loss-0.4598, acc-0.8600, valid loss-0.3947, acc-0.8900, test loss-0.3991, acc-0.8868
Iter-70920, train loss-0.5636, acc-0.8400, valid loss-0.3947, acc-0.8898, test loss-0.3991, acc-0.8867
Iter-70930, train loss-0.2814, acc-0.9200, valid loss-0.3947, acc-0.8898, test loss-0.3991, acc-0.8867
Iter-70940, train loss-0.5609, acc-0.8800, valid loss-0.3947, acc-0.8898, test loss-0.3990, acc-0.8867
Iter-70950, train loss-0.4457, acc-0.8800, valid loss-0.3947, acc-0.8898, test loss-0.3990, acc-0.8867
Iter-70960, train loss-0.3819, acc-0.9000, valid loss-0.3946, acc-0.8898, test loss-0.3990, acc-0.8867
Iter-70970, train loss-0.3062, acc-0.9600, valid loss-0.3946, acc-0.8898, test loss-0.3990, acc-0.8867
Iter-70980, train loss-0.4838, acc-0.8800, valid loss-0.3946, acc-0.8898, test loss-0.3989, acc-0.8867
Iter-70990, train loss-0.4748, acc-0.8600, valid loss-0.3946, acc-0.8898, test loss-0.3989, acc-0.8867
Iter-71000, train loss-0.3529, acc-0.8800, valid loss-0.3946, acc-0.8898, test loss-0.3989, acc-0.8867
Iter-71010, train loss-0.2747, acc-0.8800, valid loss-0.3945, acc-0.8898, test loss-0.3989, acc-0.8867
Iter-71020, train loss-0.3616, acc-0.9200, valid loss-0.3945, acc-0.8898, test loss-0.3989, acc-0.8867
Iter-71030, train loss-0.5338, acc-0.8000, valid loss-0.3945, acc-0.8898, test loss-0.3989, acc-0.8867
Iter-71040, train loss-0.2502, acc-0.9400, valid loss-0.3945, acc-0.8898, test loss-0.3988, acc-0.8867
Iter-71050, train loss-0.4304, acc-0.9000, valid loss-0.3945, acc-0.8898, test loss-0.3988, acc-0.8867
Iter-71060, train loss-0.4775, acc-0.8200, valid loss-0.3945, acc-0.8898, test loss-0.3988, acc-0.8868
Iter-71070, train loss-0.2961, acc-0.9200, valid loss-0.3944, acc-0.8898, test loss-0.3988, acc-0.8868
Iter-71080, train loss-0.3816, acc-0.8800, valid loss-0.3944, acc-0.8898, test loss-0.3987, acc-0.8868
Iter-71090, train loss-0.2922, acc-0.9400, valid loss-0.3944, acc-0.8898, test loss-0.3987, acc-0.8868
Iter-71100, train loss-0.7233, acc-0.7800, valid loss-0.3944, acc-0.8898, test loss-0.3987, acc-0.8868
Iter-71110, train loss-0.2373, acc-0.9400, valid loss-0.3943, acc-0.8898, test loss-0.3987, acc-0.8867
Iter-71120, train loss-0.4309, acc-0.8800, valid loss-0.3943, acc-0.8898, test loss-0.3986, acc-0.8868
Iter-71130, train loss-0.3699, acc-0.8600, valid loss-0.3943, acc-0.8898, test loss-0.3986, acc-0.8868
Iter-71140, train loss-0.3391, acc-0.9600, valid loss-0.3943, acc-0.8898, test loss-0.3986, acc-0.8868
Iter-71150, train loss-0.4011, acc-0.8600, valid loss-0.3943, acc-0.8898, test loss-0.3986, acc-0.8868
Iter-71160, train loss-0.2481, acc-0.9400, valid loss-0.3942, acc-0.8898, test loss-0.3985, acc-0.8869
Iter-71170, train loss-0.4337, acc-0.9000, valid loss-0.3942, acc-0.8898, test loss-0.3985, acc-0.8869
Iter-71180, train loss-0.4967, acc-0.8400, valid loss-0.3942, acc-0.8898, test loss-0.3985, acc-0.8868
Iter-71190, train loss-0.4333, acc-0.8600, valid loss-0.3942, acc-0.8898, test loss-0.3985, acc-0.8869
Iter-71200, train loss-0.5628, acc-0.8600, valid loss-0.3942, acc-0.8898, test loss-0.3985, acc-0.8870
Iter-71210, train loss-0.5065, acc-0.8200, valid loss-0.3941, acc-0.8898, test loss-0.3985, acc-0.8870
Iter-71220, train loss-0.3275, acc-0.9000, valid loss-0.3941, acc-0.8900, test loss-0.3984, acc-0.8870
Iter-71230, train loss-0.4253, acc-0.8400, valid loss-0.3941, acc-0.8900, test loss-0.3984, acc-0.8870
Iter-71240, train loss-0.4290, acc-0.8600, valid loss-0.3941, acc-0.8902, test loss-0.3984, acc-0.8870
Iter-71250, train loss-0.3534, acc-0.8800, valid loss-0.3940, acc-0.8900, test loss-0.3984, acc-0.8870
Iter-71260, train loss-0.5956, acc-0.8400, valid loss-0.3940, acc-0.8900, test loss-0.3984, acc-0.8871
Iter-71270, train loss-0.5093, acc-0.8200, valid loss-0.3940, acc-0.8900, test loss-0.3983, acc-0.8869
Iter-71280, train loss-0.2565, acc-0.9400, valid loss-0.3940, acc-0.8900, test loss-0.3983, acc-0.8869
Iter-71290, train loss-0.3616, acc-0.8400, valid loss-0.3940, acc-0.8900, test loss-0.3983, acc-0.8870
Iter-71300, train loss-0.5289, acc-0.8400, valid loss-0.3940, acc-0.8900, test loss-0.3983, acc-0.8870
Iter-71310, train loss-0.5580, acc-0.7400, valid loss-0.3939, acc-0.8900, test loss-0.3982, acc-0.8870
Iter-71320, train loss-0.5935, acc-0.9000, valid loss-0.3939, acc-0.8902, test loss-0.3982, acc-0.8870
Iter-71330, train loss-0.3909, acc-0.8800, valid loss-0.3939, acc-0.8902, test loss-0.3982, acc-0.8870
Iter-71340, train loss-0.3941, acc-0.8600, valid loss-0.3939, acc-0.8900, test loss-0.3982, acc-0.8869
Iter-71350, train loss-0.6711, acc-0.8400, valid loss-0.3939, acc-0.8902, test loss-0.3982, acc-0.8869
Iter-71360, train loss-0.2908, acc-0.9200, valid loss-0.3938, acc-0.8902, test loss-0.3982, acc-0.8869
Iter-71370, train loss-0.6035, acc-0.8600, valid loss-0.3938, acc-0.8902, test loss-0.3981, acc-0.8870
Iter-71380, train loss-0.3095, acc-0.9400, valid loss-0.3938, acc-0.8900, test loss-0.3981, acc-0.8870
Iter-71390, train loss-0.4833, acc-0.8200, valid loss-0.3938, acc-0.8902, test loss-0.3981, acc-0.8869
Iter-71400, train loss-0.2780, acc-0.9400, valid loss-0.3937, acc-0.8902, test loss-0.3981, acc-0.8870
Iter-71410, train loss-0.3662, acc-0.8600, valid loss-0.3937, acc-0.8902, test loss-0.3980, acc-0.8870
Iter-71420, train loss-0.5631, acc-0.8400, valid loss-0.3937, acc-0.8902, test loss-0.3980, acc-0.8871
Iter-71430, train loss-0.3516, acc-0.8400, valid loss-0.3937, acc-0.8902, test loss-0.3980, acc-0.8871
Iter-71440, train loss-0.3770, acc-0.9200, valid loss-0.3936, acc-0.8902, test loss-0.3980, acc-0.8871
Iter-71450, train loss-0.3753, acc-0.8600, valid loss-0.3936, acc-0.8902, test loss-0.3979, acc-0.8869
Iter-71460, train loss-0.3464, acc-0.9200, valid loss-0.3936, acc-0.8902, test loss-0.3979, acc-0.8871
Iter-71470, train loss-0.2566, acc-0.9600, valid loss-0.3936, acc-0.8904, test loss-0.3979, acc-0.8870
Iter-71480, train loss-0.5427, acc-0.9000, valid loss-0.3935, acc-0.8902, test loss-0.3979, acc-0.8870
Iter-71490, train loss-0.3308, acc-0.9200, valid loss-0.3935, acc-0.8902, test loss-0.3979, acc-0.8870
Iter-71500, train loss-0.3787, acc-0.9200, valid loss-0.3935, acc-0.8902, test loss-0.3978, acc-0.8871
Iter-71510, train loss-0.4186, acc-0.9000, valid loss-0.3935, acc-0.8902, test loss-0.3978, acc-0.8871
Iter-71520, train loss-0.2377, acc-0.9400, valid loss-0.3934, acc-0.8904, test loss-0.3978, acc-0.8871
Iter-71530, train loss-0.6191, acc-0.7600, valid loss-0.3934, acc-0.8902, test loss-0.3978, acc-0.8872
Iter-71540, train loss-0.6654, acc-0.7600, valid loss-0.3934, acc-0.8902, test loss-0.3978, acc-0.8870
Iter-71550, train loss-0.5005, acc-0.8400, valid loss-0.3934, acc-0.8902, test loss-0.3978, acc-0.8870
Iter-71560, train loss-0.5353, acc-0.8200, valid loss-0.3934, acc-0.8902, test loss-0.3977, acc-0.8871
Iter-71570, train loss-0.5643, acc-0.8400, valid loss-0.3933, acc-0.8902, test loss-0.3977, acc-0.8871
Iter-71580, train loss-0.3852, acc-0.8800, valid loss-0.3933, acc-0.8902, test loss-0.3977, acc-0.8871
Iter-71590, train loss-0.4139, acc-0.8800, valid loss-0.3933, acc-0.8902, test loss-0.3977, acc-0.8871
Iter-71600, train loss-0.4383, acc-0.8800, valid loss-0.3933, acc-0.8904, test loss-0.3977, acc-0.8871
Iter-71610, train loss-0.4130, acc-0.8600, valid loss-0.3933, acc-0.8904, test loss-0.3976, acc-0.8873
Iter-71620, train loss-0.3516, acc-0.9000, valid loss-0.3932, acc-0.8904, test loss-0.3976, acc-0.8871
Iter-71630, train loss-0.8615, acc-0.7600, valid loss-0.3932, acc-0.8904, test loss-0.3976, acc-0.8871
Iter-71640, train loss-0.3808, acc-0.8200, valid loss-0.3932, acc-0.8904, test loss-0.3976, acc-0.8870
Iter-71650, train loss-0.4758, acc-0.8800, valid loss-0.3932, acc-0.8904, test loss-0.3975, acc-0.8870
Iter-71660, train loss-0.3305, acc-0.9000, valid loss-0.3932, acc-0.8904, test loss-0.3975, acc-0.8870
Iter-71670, train loss-0.4782, acc-0.8400, valid loss-0.3931, acc-0.8904, test loss-0.3975, acc-0.8871
Iter-71680, train loss-0.4518, acc-0.8600, valid loss-0.3931, acc-0.8904, test loss-0.3975, acc-0.8871
Iter-71690, train loss-0.6158, acc-0.8400, valid loss-0.3931, acc-0.8902, test loss-0.3974, acc-0.8871
Iter-71700, train loss-0.2295, acc-0.9600, valid loss-0.3930, acc-0.8904, test loss-0.3974, acc-0.8871
Iter-71710, train loss-0.5673, acc-0.8600, valid loss-0.3930, acc-0.8904, test loss-0.3974, acc-0.8871
Iter-71720, train loss-0.2548, acc-0.9600, valid loss-0.3930, acc-0.8904, test loss-0.3974, acc-0.8871
Iter-71730, train loss-0.4836, acc-0.8200, valid loss-0.3930, acc-0.8904, test loss-0.3974, acc-0.8871
Iter-71740, train loss-0.4358, acc-0.8800, valid loss-0.3930, acc-0.8904, test loss-0.3973, acc-0.8871
Iter-71750, train loss-0.3708, acc-0.9000, valid loss-0.3929, acc-0.8904, test loss-0.3973, acc-0.8872
Iter-71760, train loss-0.5455, acc-0.8400, valid loss-0.3929, acc-0.8904, test loss-0.3973, acc-0.8872
Iter-71770, train loss-0.2597, acc-0.9800, valid loss-0.3929, acc-0.8904, test loss-0.3973, acc-0.8872
Iter-71780, train loss-0.4449, acc-0.9000, valid loss-0.3929, acc-0.8904, test loss-0.3973, acc-0.8872
Iter-71790, train loss-0.3599, acc-0.9000, valid loss-0.3928, acc-0.8904, test loss-0.3972, acc-0.8871
Iter-71800, train loss-0.2573, acc-0.9400, valid loss-0.3928, acc-0.8904, test loss-0.3972, acc-0.8872
Iter-71810, train loss-0.3999, acc-0.9000, valid loss-0.3928, acc-0.8904, test loss-0.3972, acc-0.8872
Iter-71820, train loss-0.3884, acc-0.8800, valid loss-0.3928, acc-0.8904, test loss-0.3972, acc-0.8872
Iter-71830, train loss-0.2709, acc-0.9200, valid loss-0.3927, acc-0.8904, test loss-0.3971, acc-0.8872
Iter-71840, train loss-0.7579, acc-0.7600, valid loss-0.3927, acc-0.8904, test loss-0.3971, acc-0.8872
Iter-71850, train loss-0.4352, acc-0.8800, valid loss-0.3927, acc-0.8904, test loss-0.3971, acc-0.8873
Iter-71860, train loss-0.2838, acc-0.9400, valid loss-0.3927, acc-0.8904, test loss-0.3971, acc-0.8873
Iter-71870, train loss-0.4420, acc-0.8200, valid loss-0.3927, acc-0.8904, test loss-0.3971, acc-0.8873
Iter-71880, train loss-0.3667, acc-0.9200, valid loss-0.3926, acc-0.8904, test loss-0.3970, acc-0.8873
Iter-71890, train loss-0.4355, acc-0.8800, valid loss-0.3926, acc-0.8904, test loss-0.3970, acc-0.8873
Iter-71900, train loss-0.3420, acc-0.9200, valid loss-0.3926, acc-0.8904, test loss-0.3970, acc-0.8873
Iter-71910, train loss-0.3692, acc-0.9400, valid loss-0.3926, acc-0.8904, test loss-0.3970, acc-0.8873
Iter-71920, train loss-0.3343, acc-0.9400, valid loss-0.3925, acc-0.8904, test loss-0.3970, acc-0.8873
Iter-71930, train loss-0.5082, acc-0.8000, valid loss-0.3925, acc-0.8904, test loss-0.3970, acc-0.8873
Iter-71940, train loss-0.4877, acc-0.8400, valid loss-0.3925, acc-0.8904, test loss-0.3969, acc-0.8873
Iter-71950, train loss-0.3209, acc-0.9400, valid loss-0.3925, acc-0.8904, test loss-0.3969, acc-0.8874
Iter-71960, train loss-0.2826, acc-0.9200, valid loss-0.3924, acc-0.8904, test loss-0.3969, acc-0.8874
Iter-71970, train loss-0.5883, acc-0.7400, valid loss-0.3924, acc-0.8904, test loss-0.3969, acc-0.8874
Iter-71980, train loss-0.6387, acc-0.7800, valid loss-0.3924, acc-0.8904, test loss-0.3968, acc-0.8873
Iter-71990, train loss-0.4018, acc-0.8400, valid loss-0.3924, acc-0.8904, test loss-0.3968, acc-0.8873
Iter-72000, train loss-0.3722, acc-0.8800, valid loss-0.3924, acc-0.8904, test loss-0.3968, acc-0.8873
Iter-72010, train loss-0.4338, acc-0.8800, valid loss-0.3924, acc-0.8904, test loss-0.3968, acc-0.8873
Iter-72020, train loss-0.6137, acc-0.8200, valid loss-0.3923, acc-0.8904, test loss-0.3968, acc-0.8874
Iter-72030, train loss-0.3919, acc-0.9200, valid loss-0.3923, acc-0.8904, test loss-0.3967, acc-0.8874
Iter-72040, train loss-0.7231, acc-0.7200, valid loss-0.3923, acc-0.8904, test loss-0.3967, acc-0.8873
Iter-72050, train loss-0.3917, acc-0.9600, valid loss-0.3922, acc-0.8904, test loss-0.3967, acc-0.8873
Iter-72060, train loss-0.3380, acc-0.8800, valid loss-0.3922, acc-0.8904, test loss-0.3966, acc-0.8874
Iter-72070, train loss-0.6427, acc-0.8000, valid loss-0.3922, acc-0.8904, test loss-0.3966, acc-0.8873
Iter-72080, train loss-0.4297, acc-0.8600, valid loss-0.3922, acc-0.8904, test loss-0.3966, acc-0.8873
Iter-72090, train loss-0.4668, acc-0.8400, valid loss-0.3922, acc-0.8904, test loss-0.3966, acc-0.8874
Iter-72100, train loss-0.6230, acc-0.8600, valid loss-0.3921, acc-0.8904, test loss-0.3966, acc-0.8875
Iter-72110, train loss-0.4143, acc-0.9000, valid loss-0.3921, acc-0.8908, test loss-0.3965, acc-0.8875
Iter-72120, train loss-0.3612, acc-0.9000, valid loss-0.3921, acc-0.8908, test loss-0.3965, acc-0.8875
Iter-72130, train loss-0.2882, acc-0.9200, valid loss-0.3921, acc-0.8906, test loss-0.3965, acc-0.8875
Iter-72140, train loss-0.5309, acc-0.8400, valid loss-0.3920, acc-0.8908, test loss-0.3965, acc-0.8875
Iter-72150, train loss-0.3619, acc-0.9200, valid loss-0.3920, acc-0.8906, test loss-0.3965, acc-0.8874
Iter-72160, train loss-0.3696, acc-0.9200, valid loss-0.3920, acc-0.8904, test loss-0.3965, acc-0.8874
Iter-72170, train loss-0.3353, acc-0.9000, valid loss-0.3920, acc-0.8908, test loss-0.3964, acc-0.8873
Iter-72180, train loss-0.3085, acc-0.9200, valid loss-0.3920, acc-0.8906, test loss-0.3964, acc-0.8873
Iter-72190, train loss-0.5095, acc-0.8600, valid loss-0.3920, acc-0.8908, test loss-0.3964, acc-0.8874
Iter-72200, train loss-0.4269, acc-0.8800, valid loss-0.3919, acc-0.8904, test loss-0.3964, acc-0.8874
Iter-72210, train loss-0.2265, acc-0.9200, valid loss-0.3919, acc-0.8904, test loss-0.3964, acc-0.8874
Iter-72220, train loss-0.3012, acc-0.9400, valid loss-0.3919, acc-0.8904, test loss-0.3964, acc-0.8874
Iter-72230, train loss-0.4193, acc-0.8800, valid loss-0.3919, acc-0.8904, test loss-0.3963, acc-0.8873
Iter-72240, train loss-0.3286, acc-0.9200, valid loss-0.3919, acc-0.8904, test loss-0.3963, acc-0.8874
Iter-72250, train loss-0.4480, acc-0.8800, valid loss-0.3919, acc-0.8904, test loss-0.3963, acc-0.8873
Iter-72260, train loss-0.5018, acc-0.8600, valid loss-0.3918, acc-0.8904, test loss-0.3963, acc-0.8873
Iter-72270, train loss-0.2758, acc-0.9200, valid loss-0.3918, acc-0.8904, test loss-0.3962, acc-0.8873
Iter-72280, train loss-0.4474, acc-0.8400, valid loss-0.3918, acc-0.8908, test loss-0.3962, acc-0.8873
Iter-72290, train loss-0.2769, acc-0.9600, valid loss-0.3918, acc-0.8906, test loss-0.3962, acc-0.8873
Iter-72300, train loss-0.5091, acc-0.8000, valid loss-0.3918, acc-0.8908, test loss-0.3962, acc-0.8873
Iter-72310, train loss-0.2847, acc-0.9400, valid loss-0.3917, acc-0.8908, test loss-0.3962, acc-0.8874
Iter-72320, train loss-0.4426, acc-0.8600, valid loss-0.3917, acc-0.8908, test loss-0.3961, acc-0.8873
Iter-72330, train loss-0.2793, acc-0.9400, valid loss-0.3917, acc-0.8908, test loss-0.3961, acc-0.8872
Iter-72340, train loss-0.3289, acc-0.9400, valid loss-0.3917, acc-0.8904, test loss-0.3961, acc-0.8872
Iter-72350, train loss-0.2715, acc-0.9000, valid loss-0.3916, acc-0.8908, test loss-0.3961, acc-0.8872
Iter-72360, train loss-0.3954, acc-0.9000, valid loss-0.3916, acc-0.8908, test loss-0.3960, acc-0.8872
Iter-72370, train loss-0.2957, acc-0.9000, valid loss-0.3916, acc-0.8904, test loss-0.3960, acc-0.8873
Iter-72380, train loss-0.3733, acc-0.9200, valid loss-0.3916, acc-0.8904, test loss-0.3960, acc-0.8874
Iter-72390, train loss-0.2998, acc-0.9000, valid loss-0.3916, acc-0.8904, test loss-0.3960, acc-0.8874
Iter-72400, train loss-0.3636, acc-0.9200, valid loss-0.3915, acc-0.8904, test loss-0.3960, acc-0.8874
Iter-72410, train loss-0.3301, acc-0.8800, valid loss-0.3915, acc-0.8904, test loss-0.3959, acc-0.8874
Iter-72420, train loss-0.4442, acc-0.8200, valid loss-0.3915, acc-0.8904, test loss-0.3959, acc-0.8874
Iter-72430, train loss-0.3329, acc-0.9400, valid loss-0.3915, acc-0.8904, test loss-0.3959, acc-0.8874
Iter-72440, train loss-0.3295, acc-0.9000, valid loss-0.3915, acc-0.8904, test loss-0.3959, acc-0.8874
Iter-72450, train loss-0.4799, acc-0.8400, valid loss-0.3915, acc-0.8904, test loss-0.3958, acc-0.8874
Iter-72460, train loss-0.3979, acc-0.8800, valid loss-0.3914, acc-0.8906, test loss-0.3958, acc-0.8874
Iter-72470, train loss-0.4177, acc-0.9000, valid loss-0.3914, acc-0.8906, test loss-0.3958, acc-0.8874
Iter-72480, train loss-0.3021, acc-0.9200, valid loss-0.3914, acc-0.8906, test loss-0.3958, acc-0.8874
Iter-72490, train loss-0.6106, acc-0.8200, valid loss-0.3914, acc-0.8906, test loss-0.3958, acc-0.8874
Iter-72500, train loss-0.4423, acc-0.8400, valid loss-0.3914, acc-0.8904, test loss-0.3957, acc-0.8873
Iter-72510, train loss-0.3786, acc-0.9000, valid loss-0.3913, acc-0.8906, test loss-0.3957, acc-0.8873
Iter-72520, train loss-0.3889, acc-0.9200, valid loss-0.3913, acc-0.8904, test loss-0.3957, acc-0.8874
Iter-72530, train loss-0.2333, acc-0.9600, valid loss-0.3913, acc-0.8904, test loss-0.3957, acc-0.8875
Iter-72540, train loss-0.5659, acc-0.9000, valid loss-0.3912, acc-0.8904, test loss-0.3956, acc-0.8875
Iter-72550, train loss-0.3777, acc-0.8800, valid loss-0.3912, acc-0.8904, test loss-0.3956, acc-0.8875
Iter-72560, train loss-0.4363, acc-0.8800, valid loss-0.3912, acc-0.8906, test loss-0.3956, acc-0.8875
Iter-72570, train loss-0.5881, acc-0.8600, valid loss-0.3912, acc-0.8906, test loss-0.3956, acc-0.8874
Iter-72580, train loss-0.3341, acc-0.8800, valid loss-0.3912, acc-0.8906, test loss-0.3956, acc-0.8874
Iter-72590, train loss-0.2859, acc-0.9600, valid loss-0.3911, acc-0.8904, test loss-0.3955, acc-0.8874
Iter-72600, train loss-0.4910, acc-0.8200, valid loss-0.3911, acc-0.8904, test loss-0.3955, acc-0.8875
Iter-72610, train loss-0.6076, acc-0.8600, valid loss-0.3911, acc-0.8904, test loss-0.3955, acc-0.8875
Iter-72620, train loss-0.1813, acc-0.9600, valid loss-0.3911, acc-0.8904, test loss-0.3955, acc-0.8875
Iter-72630, train loss-0.4629, acc-0.8800, valid loss-0.3911, acc-0.8904, test loss-0.3955, acc-0.8875
Iter-72640, train loss-0.4390, acc-0.8600, valid loss-0.3911, acc-0.8904, test loss-0.3955, acc-0.8875
Iter-72650, train loss-0.5078, acc-0.8200, valid loss-0.3911, acc-0.8904, test loss-0.3954, acc-0.8875
Iter-72660, train loss-0.3626, acc-0.9000, valid loss-0.3910, acc-0.8904, test loss-0.3954, acc-0.8876
Iter-72670, train loss-0.3221, acc-0.9000, valid loss-0.3910, acc-0.8904, test loss-0.3954, acc-0.8875
Iter-72680, train loss-0.3965, acc-0.8800, valid loss-0.3910, acc-0.8904, test loss-0.3954, acc-0.8874
Iter-72690, train loss-0.5927, acc-0.9000, valid loss-0.3910, acc-0.8906, test loss-0.3954, acc-0.8876
Iter-72700, train loss-0.2601, acc-0.9400, valid loss-0.3909, acc-0.8906, test loss-0.3954, acc-0.8876
Iter-72710, train loss-0.5229, acc-0.8400, valid loss-0.3909, acc-0.8904, test loss-0.3953, acc-0.8876
Iter-72720, train loss-0.3358, acc-0.9200, valid loss-0.3909, acc-0.8904, test loss-0.3953, acc-0.8876
Iter-72730, train loss-0.5302, acc-0.7800, valid loss-0.3909, acc-0.8906, test loss-0.3953, acc-0.8876
Iter-72740, train loss-0.3206, acc-0.9200, valid loss-0.3908, acc-0.8904, test loss-0.3953, acc-0.8876
Iter-72750, train loss-0.6445, acc-0.8000, valid loss-0.3908, acc-0.8906, test loss-0.3953, acc-0.8876
Iter-72760, train loss-0.4529, acc-0.8200, valid loss-0.3908, acc-0.8908, test loss-0.3952, acc-0.8876
Iter-72770, train loss-0.2589, acc-0.9200, valid loss-0.3908, acc-0.8908, test loss-0.3952, acc-0.8876
Iter-72780, train loss-0.3036, acc-0.9200, valid loss-0.3908, acc-0.8908, test loss-0.3952, acc-0.8876
Iter-72790, train loss-0.3285, acc-0.9000, valid loss-0.3908, acc-0.8908, test loss-0.3952, acc-0.8876
Iter-72800, train loss-0.2069, acc-0.9800, valid loss-0.3907, acc-0.8908, test loss-0.3952, acc-0.8876
Iter-72810, train loss-0.4440, acc-0.8400, valid loss-0.3907, acc-0.8908, test loss-0.3951, acc-0.8876
Iter-72820, train loss-0.1876, acc-0.9800, valid loss-0.3907, acc-0.8908, test loss-0.3951, acc-0.8876
Iter-72830, train loss-0.2748, acc-0.9200, valid loss-0.3907, acc-0.8908, test loss-0.3951, acc-0.8876
Iter-72840, train loss-0.4271, acc-0.8200, valid loss-0.3907, acc-0.8908, test loss-0.3951, acc-0.8876
Iter-72850, train loss-0.4657, acc-0.8800, valid loss-0.3906, acc-0.8908, test loss-0.3950, acc-0.8876
Iter-72860, train loss-0.4218, acc-0.8600, valid loss-0.3906, acc-0.8908, test loss-0.3950, acc-0.8876
Iter-72870, train loss-0.4518, acc-0.8600, valid loss-0.3906, acc-0.8908, test loss-0.3950, acc-0.8876
Iter-72880, train loss-0.5243, acc-0.8000, valid loss-0.3906, acc-0.8908, test loss-0.3950, acc-0.8876
Iter-72890, train loss-0.4122, acc-0.8600, valid loss-0.3906, acc-0.8908, test loss-0.3950, acc-0.8876
Iter-72900, train loss-0.3691, acc-0.8600, valid loss-0.3905, acc-0.8908, test loss-0.3950, acc-0.8875
Iter-72910, train loss-0.3971, acc-0.8600, valid loss-0.3905, acc-0.8910, test loss-0.3950, acc-0.8876
Iter-72920, train loss-0.2606, acc-0.9400, valid loss-0.3905, acc-0.8908, test loss-0.3950, acc-0.8875
Iter-72930, train loss-0.3029, acc-0.9000, valid loss-0.3905, acc-0.8908, test loss-0.3949, acc-0.8875
Iter-72940, train loss-0.3191, acc-0.9400, valid loss-0.3905, acc-0.8908, test loss-0.3949, acc-0.8875
Iter-72950, train loss-0.7619, acc-0.7600, valid loss-0.3905, acc-0.8908, test loss-0.3949, acc-0.8875
Iter-72960, train loss-0.2353, acc-0.9200, valid loss-0.3904, acc-0.8908, test loss-0.3949, acc-0.8875
Iter-72970, train loss-0.5644, acc-0.8600, valid loss-0.3904, acc-0.8908, test loss-0.3949, acc-0.8875
Iter-72980, train loss-0.4323, acc-0.8400, valid loss-0.3904, acc-0.8908, test loss-0.3948, acc-0.8875
Iter-72990, train loss-0.4522, acc-0.9200, valid loss-0.3904, acc-0.8908, test loss-0.3948, acc-0.8875
Iter-73000, train loss-0.5501, acc-0.8200, valid loss-0.3903, acc-0.8908, test loss-0.3948, acc-0.8875
Iter-73010, train loss-0.2208, acc-0.9600, valid loss-0.3903, acc-0.8908, test loss-0.3948, acc-0.8875
Iter-73020, train loss-0.2250, acc-0.9600, valid loss-0.3903, acc-0.8908, test loss-0.3948, acc-0.8875
Iter-73030, train loss-0.4324, acc-0.8400, valid loss-0.3903, acc-0.8908, test loss-0.3947, acc-0.8875
Iter-73040, train loss-0.3251, acc-0.9000, valid loss-0.3902, acc-0.8908, test loss-0.3947, acc-0.8875
Iter-73050, train loss-0.5197, acc-0.8000, valid loss-0.3902, acc-0.8908, test loss-0.3947, acc-0.8876
Iter-73060, train loss-0.4571, acc-0.8800, valid loss-0.3902, acc-0.8908, test loss-0.3947, acc-0.8875
Iter-73070, train loss-0.6337, acc-0.8400, valid loss-0.3902, acc-0.8908, test loss-0.3947, acc-0.8875
Iter-73080, train loss-0.3567, acc-0.9000, valid loss-0.3901, acc-0.8908, test loss-0.3946, acc-0.8875
Iter-73090, train loss-0.4602, acc-0.8400, valid loss-0.3901, acc-0.8908, test loss-0.3946, acc-0.8875
Iter-73100, train loss-0.5226, acc-0.8600, valid loss-0.3901, acc-0.8910, test loss-0.3946, acc-0.8875
Iter-73110, train loss-0.3394, acc-0.9200, valid loss-0.3901, acc-0.8910, test loss-0.3946, acc-0.8875
Iter-73120, train loss-0.3309, acc-0.9000, valid loss-0.3901, acc-0.8910, test loss-0.3945, acc-0.8875
Iter-73130, train loss-0.2937, acc-0.9000, valid loss-0.3901, acc-0.8910, test loss-0.3945, acc-0.8875
Iter-73140, train loss-0.4469, acc-0.8600, valid loss-0.3900, acc-0.8910, test loss-0.3945, acc-0.8874
Iter-73150, train loss-0.6183, acc-0.8600, valid loss-0.3900, acc-0.8910, test loss-0.3944, acc-0.8874
Iter-73160, train loss-0.4609, acc-0.8800, valid loss-0.3900, acc-0.8910, test loss-0.3944, acc-0.8874
Iter-73170, train loss-0.3060, acc-0.8800, valid loss-0.3900, acc-0.8910, test loss-0.3944, acc-0.8874
Iter-73180, train loss-0.4652, acc-0.9200, valid loss-0.3899, acc-0.8910, test loss-0.3944, acc-0.8874
Iter-73190, train loss-0.5682, acc-0.8400, valid loss-0.3899, acc-0.8910, test loss-0.3944, acc-0.8874
Iter-73200, train loss-0.8258, acc-0.8200, valid loss-0.3899, acc-0.8912, test loss-0.3943, acc-0.8874
Iter-73210, train loss-0.2247, acc-0.9600, valid loss-0.3899, acc-0.8910, test loss-0.3943, acc-0.8874
Iter-73220, train loss-0.4596, acc-0.8600, valid loss-0.3898, acc-0.8912, test loss-0.3943, acc-0.8874
Iter-73230, train loss-0.3254, acc-0.8800, valid loss-0.3898, acc-0.8912, test loss-0.3943, acc-0.8874
Iter-73240, train loss-0.4132, acc-0.8600, valid loss-0.3898, acc-0.8912, test loss-0.3942, acc-0.8874
Iter-73250, train loss-0.2602, acc-0.9200, valid loss-0.3898, acc-0.8912, test loss-0.3942, acc-0.8874
Iter-73260, train loss-0.4075, acc-0.9200, valid loss-0.3898, acc-0.8912, test loss-0.3942, acc-0.8874
Iter-73270, train loss-0.4560, acc-0.8800, valid loss-0.3898, acc-0.8912, test loss-0.3942, acc-0.8874
Iter-73280, train loss-0.5076, acc-0.8400, valid loss-0.3897, acc-0.8912, test loss-0.3942, acc-0.8874
Iter-73290, train loss-0.4193, acc-0.8600, valid loss-0.3897, acc-0.8910, test loss-0.3942, acc-0.8874
Iter-73300, train loss-0.5378, acc-0.8200, valid loss-0.3897, acc-0.8912, test loss-0.3941, acc-0.8874
Iter-73310, train loss-0.3337, acc-0.8800, valid loss-0.3897, acc-0.8912, test loss-0.3941, acc-0.8874
Iter-73320, train loss-0.2642, acc-0.9400, valid loss-0.3896, acc-0.8912, test loss-0.3941, acc-0.8874
Iter-73330, train loss-0.1935, acc-0.9600, valid loss-0.3896, acc-0.8912, test loss-0.3941, acc-0.8874
Iter-73340, train loss-0.2809, acc-0.9600, valid loss-0.3896, acc-0.8912, test loss-0.3940, acc-0.8874
Iter-73350, train loss-0.3692, acc-0.8600, valid loss-0.3896, acc-0.8912, test loss-0.3940, acc-0.8874
Iter-73360, train loss-0.6176, acc-0.7200, valid loss-0.3896, acc-0.8912, test loss-0.3940, acc-0.8874
Iter-73370, train loss-0.4979, acc-0.8600, valid loss-0.3895, acc-0.8912, test loss-0.3940, acc-0.8874
Iter-73380, train loss-0.3546, acc-0.9400, valid loss-0.3895, acc-0.8912, test loss-0.3940, acc-0.8874
Iter-73390, train loss-0.3570, acc-0.9000, valid loss-0.3895, acc-0.8912, test loss-0.3939, acc-0.8875
Iter-73400, train loss-0.3601, acc-0.8800, valid loss-0.3895, acc-0.8912, test loss-0.3939, acc-0.8874
Iter-73410, train loss-0.5068, acc-0.9000, valid loss-0.3894, acc-0.8912, test loss-0.3939, acc-0.8875
Iter-73420, train loss-0.8115, acc-0.7400, valid loss-0.3894, acc-0.8912, test loss-0.3939, acc-0.8875
Iter-73430, train loss-0.3753, acc-0.9000, valid loss-0.3894, acc-0.8912, test loss-0.3939, acc-0.8875
Iter-73440, train loss-0.4606, acc-0.9200, valid loss-0.3894, acc-0.8912, test loss-0.3938, acc-0.8875
Iter-73450, train loss-0.5343, acc-0.8600, valid loss-0.3893, acc-0.8912, test loss-0.3938, acc-0.8875
Iter-73460, train loss-0.3271, acc-0.9200, valid loss-0.3893, acc-0.8912, test loss-0.3938, acc-0.8875
Iter-73470, train loss-0.3502, acc-0.8800, valid loss-0.3893, acc-0.8912, test loss-0.3938, acc-0.8874
Iter-73480, train loss-0.2869, acc-0.9000, valid loss-0.3893, acc-0.8912, test loss-0.3938, acc-0.8874
Iter-73490, train loss-0.4860, acc-0.9000, valid loss-0.3893, acc-0.8914, test loss-0.3937, acc-0.8874
Iter-73500, train loss-0.3894, acc-0.8800, valid loss-0.3892, acc-0.8916, test loss-0.3937, acc-0.8874
Iter-73510, train loss-0.4220, acc-0.8400, valid loss-0.3892, acc-0.8914, test loss-0.3937, acc-0.8876
Iter-73520, train loss-0.3533, acc-0.8400, valid loss-0.3892, acc-0.8914, test loss-0.3937, acc-0.8876
Iter-73530, train loss-0.3589, acc-0.9200, valid loss-0.3891, acc-0.8912, test loss-0.3936, acc-0.8876
Iter-73540, train loss-0.3862, acc-0.8800, valid loss-0.3891, acc-0.8912, test loss-0.3936, acc-0.8876
Iter-73550, train loss-0.5948, acc-0.8600, valid loss-0.3891, acc-0.8912, test loss-0.3936, acc-0.8876
Iter-73560, train loss-0.4534, acc-0.8800, valid loss-0.3891, acc-0.8912, test loss-0.3935, acc-0.8875
Iter-73570, train loss-0.4433, acc-0.9000, valid loss-0.3890, acc-0.8912, test loss-0.3935, acc-0.8876
Iter-73580, train loss-0.4970, acc-0.8000, valid loss-0.3890, acc-0.8912, test loss-0.3935, acc-0.8875
Iter-73590, train loss-0.2314, acc-0.9400, valid loss-0.3890, acc-0.8912, test loss-0.3935, acc-0.8875
Iter-73600, train loss-0.1765, acc-0.9600, valid loss-0.3890, acc-0.8912, test loss-0.3935, acc-0.8876
Iter-73610, train loss-0.4357, acc-0.8800, valid loss-0.3890, acc-0.8912, test loss-0.3935, acc-0.8875
Iter-73620, train loss-0.3731, acc-0.9000, valid loss-0.3890, acc-0.8912, test loss-0.3934, acc-0.8875
Iter-73630, train loss-0.2991, acc-0.9200, valid loss-0.3889, acc-0.8914, test loss-0.3934, acc-0.8876
Iter-73640, train loss-0.3772, acc-0.8800, valid loss-0.3889, acc-0.8914, test loss-0.3934, acc-0.8876
Iter-73650, train loss-0.3906, acc-0.9000, valid loss-0.3889, acc-0.8914, test loss-0.3934, acc-0.8876
Iter-73660, train loss-0.5749, acc-0.8200, valid loss-0.3889, acc-0.8914, test loss-0.3934, acc-0.8875
Iter-73670, train loss-0.5430, acc-0.8800, valid loss-0.3889, acc-0.8914, test loss-0.3933, acc-0.8876
Iter-73680, train loss-0.3719, acc-0.9000, valid loss-0.3889, acc-0.8914, test loss-0.3933, acc-0.8875
Iter-73690, train loss-0.2885, acc-0.9800, valid loss-0.3888, acc-0.8914, test loss-0.3933, acc-0.8875
Iter-73700, train loss-0.6181, acc-0.8000, valid loss-0.3888, acc-0.8914, test loss-0.3933, acc-0.8875
Iter-73710, train loss-0.4033, acc-0.8600, valid loss-0.3888, acc-0.8914, test loss-0.3933, acc-0.8874
Iter-73720, train loss-0.4258, acc-0.8400, valid loss-0.3888, acc-0.8916, test loss-0.3933, acc-0.8874
Iter-73730, train loss-0.2107, acc-0.9400, valid loss-0.3888, acc-0.8914, test loss-0.3932, acc-0.8874
Iter-73740, train loss-0.4470, acc-0.8800, valid loss-0.3887, acc-0.8914, test loss-0.3932, acc-0.8874
Iter-73750, train loss-0.3500, acc-0.9200, valid loss-0.3887, acc-0.8914, test loss-0.3932, acc-0.8875
Iter-73760, train loss-0.4320, acc-0.9000, valid loss-0.3887, acc-0.8914, test loss-0.3932, acc-0.8875
Iter-73770, train loss-0.3394, acc-0.9000, valid loss-0.3887, acc-0.8914, test loss-0.3932, acc-0.8875
Iter-73780, train loss-0.2997, acc-0.8800, valid loss-0.3887, acc-0.8918, test loss-0.3931, acc-0.8876
Iter-73790, train loss-0.5935, acc-0.8200, valid loss-0.3887, acc-0.8918, test loss-0.3931, acc-0.8875
Iter-73800, train loss-0.4485, acc-0.8400, valid loss-0.3887, acc-0.8920, test loss-0.3931, acc-0.8875
Iter-73810, train loss-0.3861, acc-0.9000, valid loss-0.3886, acc-0.8920, test loss-0.3931, acc-0.8875
Iter-73820, train loss-0.5763, acc-0.8200, valid loss-0.3886, acc-0.8920, test loss-0.3931, acc-0.8877
Iter-73830, train loss-0.5291, acc-0.8400, valid loss-0.3886, acc-0.8918, test loss-0.3930, acc-0.8877
Iter-73840, train loss-0.2708, acc-0.9600, valid loss-0.3886, acc-0.8918, test loss-0.3930, acc-0.8877
Iter-73850, train loss-0.4421, acc-0.9000, valid loss-0.3886, acc-0.8916, test loss-0.3930, acc-0.8876
Iter-73860, train loss-0.3450, acc-0.9200, valid loss-0.3885, acc-0.8920, test loss-0.3930, acc-0.8875
Iter-73870, train loss-0.3512, acc-0.9200, valid loss-0.3885, acc-0.8920, test loss-0.3930, acc-0.8876
Iter-73880, train loss-0.4685, acc-0.8800, valid loss-0.3885, acc-0.8920, test loss-0.3930, acc-0.8877
Iter-73890, train loss-0.1440, acc-0.9800, valid loss-0.3885, acc-0.8918, test loss-0.3929, acc-0.8877
Iter-73900, train loss-0.2483, acc-0.9200, valid loss-0.3884, acc-0.8918, test loss-0.3929, acc-0.8876
Iter-73910, train loss-0.4766, acc-0.8400, valid loss-0.3884, acc-0.8920, test loss-0.3929, acc-0.8877
Iter-73920, train loss-0.6110, acc-0.8400, valid loss-0.3884, acc-0.8918, test loss-0.3929, acc-0.8878
Iter-73930, train loss-0.3150, acc-0.9000, valid loss-0.3884, acc-0.8918, test loss-0.3929, acc-0.8878
Iter-73940, train loss-0.3443, acc-0.8800, valid loss-0.3884, acc-0.8918, test loss-0.3928, acc-0.8878
Iter-73950, train loss-0.4435, acc-0.8800, valid loss-0.3883, acc-0.8920, test loss-0.3928, acc-0.8878
Iter-73960, train loss-0.3319, acc-0.9000, valid loss-0.3883, acc-0.8920, test loss-0.3928, acc-0.8878
Iter-73970, train loss-0.4769, acc-0.8800, valid loss-0.3883, acc-0.8920, test loss-0.3928, acc-0.8878
Iter-73980, train loss-0.5516, acc-0.8600, valid loss-0.3883, acc-0.8920, test loss-0.3928, acc-0.8878
Iter-73990, train loss-0.4879, acc-0.8800, valid loss-0.3882, acc-0.8920, test loss-0.3927, acc-0.8877
Iter-74000, train loss-0.5352, acc-0.8600, valid loss-0.3882, acc-0.8920, test loss-0.3927, acc-0.8877
Iter-74010, train loss-0.3736, acc-0.9000, valid loss-0.3882, acc-0.8920, test loss-0.3927, acc-0.8878
Iter-74020, train loss-0.5109, acc-0.8600, valid loss-0.3882, acc-0.8920, test loss-0.3927, acc-0.8878
Iter-74030, train loss-0.6327, acc-0.8000, valid loss-0.3882, acc-0.8920, test loss-0.3927, acc-0.8878
Iter-74040, train loss-0.6244, acc-0.8200, valid loss-0.3881, acc-0.8920, test loss-0.3927, acc-0.8877
Iter-74050, train loss-0.3627, acc-0.9400, valid loss-0.3881, acc-0.8920, test loss-0.3926, acc-0.8877
Iter-74060, train loss-0.3881, acc-0.9200, valid loss-0.3881, acc-0.8920, test loss-0.3926, acc-0.8877
Iter-74070, train loss-0.3445, acc-0.8600, valid loss-0.3881, acc-0.8920, test loss-0.3926, acc-0.8877
Iter-74080, train loss-0.4059, acc-0.8600, valid loss-0.3881, acc-0.8920, test loss-0.3926, acc-0.8877
Iter-74090, train loss-0.3803, acc-0.9000, valid loss-0.3880, acc-0.8920, test loss-0.3926, acc-0.8877
Iter-74100, train loss-0.3157, acc-0.9400, valid loss-0.3880, acc-0.8920, test loss-0.3925, acc-0.8877
Iter-74110, train loss-0.5269, acc-0.8400, valid loss-0.3880, acc-0.8920, test loss-0.3925, acc-0.8877
Iter-74120, train loss-0.3764, acc-0.9400, valid loss-0.3880, acc-0.8920, test loss-0.3925, acc-0.8877
Iter-74130, train loss-0.2923, acc-0.9200, valid loss-0.3880, acc-0.8920, test loss-0.3925, acc-0.8877
Iter-74140, train loss-0.6963, acc-0.7800, valid loss-0.3879, acc-0.8920, test loss-0.3925, acc-0.8877
Iter-74150, train loss-0.4682, acc-0.8800, valid loss-0.3879, acc-0.8920, test loss-0.3924, acc-0.8877
Iter-74160, train loss-0.2841, acc-0.9400, valid loss-0.3879, acc-0.8920, test loss-0.3924, acc-0.8876
Iter-74170, train loss-0.2503, acc-0.9400, valid loss-0.3879, acc-0.8920, test loss-0.3924, acc-0.8877
Iter-74180, train loss-0.4837, acc-0.9000, valid loss-0.3879, acc-0.8920, test loss-0.3924, acc-0.8877
Iter-74190, train loss-0.4635, acc-0.9000, valid loss-0.3878, acc-0.8920, test loss-0.3924, acc-0.8877
Iter-74200, train loss-0.4907, acc-0.8600, valid loss-0.3878, acc-0.8920, test loss-0.3923, acc-0.8877
Iter-74210, train loss-0.3554, acc-0.8800, valid loss-0.3878, acc-0.8920, test loss-0.3923, acc-0.8877
Iter-74220, train loss-0.4931, acc-0.8600, valid loss-0.3878, acc-0.8920, test loss-0.3923, acc-0.8877
Iter-74230, train loss-0.6477, acc-0.7800, valid loss-0.3878, acc-0.8920, test loss-0.3923, acc-0.8877
Iter-74240, train loss-0.6808, acc-0.7800, valid loss-0.3877, acc-0.8920, test loss-0.3923, acc-0.8878
Iter-74250, train loss-0.2999, acc-0.9400, valid loss-0.3877, acc-0.8920, test loss-0.3923, acc-0.8878
Iter-74260, train loss-0.5541, acc-0.8400, valid loss-0.3877, acc-0.8920, test loss-0.3922, acc-0.8878
Iter-74270, train loss-0.4466, acc-0.9000, valid loss-0.3877, acc-0.8922, test loss-0.3922, acc-0.8878
Iter-74280, train loss-0.4247, acc-0.8800, valid loss-0.3877, acc-0.8922, test loss-0.3922, acc-0.8878
Iter-74290, train loss-0.3820, acc-0.9200, valid loss-0.3876, acc-0.8920, test loss-0.3922, acc-0.8877
Iter-74300, train loss-0.4952, acc-0.8000, valid loss-0.3876, acc-0.8920, test loss-0.3922, acc-0.8877
Iter-74310, train loss-0.3467, acc-0.8800, valid loss-0.3876, acc-0.8922, test loss-0.3921, acc-0.8878
Iter-74320, train loss-0.4412, acc-0.8600, valid loss-0.3876, acc-0.8922, test loss-0.3921, acc-0.8879
Iter-74330, train loss-0.3034, acc-0.9400, valid loss-0.3876, acc-0.8922, test loss-0.3921, acc-0.8877
Iter-74340, train loss-0.4340, acc-0.8800, valid loss-0.3876, acc-0.8920, test loss-0.3921, acc-0.8877
Iter-74350, train loss-0.3756, acc-0.9200, valid loss-0.3875, acc-0.8920, test loss-0.3921, acc-0.8877
Iter-74360, train loss-0.5671, acc-0.7800, valid loss-0.3875, acc-0.8920, test loss-0.3921, acc-0.8877
Iter-74370, train loss-0.5533, acc-0.8400, valid loss-0.3875, acc-0.8920, test loss-0.3920, acc-0.8879
Iter-74380, train loss-0.2639, acc-0.9600, valid loss-0.3875, acc-0.8920, test loss-0.3920, acc-0.8878
Iter-74390, train loss-0.5186, acc-0.8200, valid loss-0.3875, acc-0.8920, test loss-0.3920, acc-0.8879
Iter-74400, train loss-0.1850, acc-0.9600, valid loss-0.3875, acc-0.8920, test loss-0.3920, acc-0.8880
Iter-74410, train loss-0.5652, acc-0.8200, valid loss-0.3874, acc-0.8922, test loss-0.3920, acc-0.8879
Iter-74420, train loss-0.2874, acc-0.9200, valid loss-0.3874, acc-0.8922, test loss-0.3919, acc-0.8879
Iter-74430, train loss-0.3468, acc-0.9000, valid loss-0.3874, acc-0.8920, test loss-0.3919, acc-0.8879
Iter-74440, train loss-0.3223, acc-0.9200, valid loss-0.3874, acc-0.8918, test loss-0.3919, acc-0.8879
Iter-74450, train loss-0.3624, acc-0.9000, valid loss-0.3874, acc-0.8918, test loss-0.3919, acc-0.8879
Iter-74460, train loss-0.4157, acc-0.9200, valid loss-0.3874, acc-0.8918, test loss-0.3919, acc-0.8879
Iter-74470, train loss-0.5240, acc-0.8000, valid loss-0.3873, acc-0.8918, test loss-0.3918, acc-0.8879
Iter-74480, train loss-0.3698, acc-0.9000, valid loss-0.3873, acc-0.8918, test loss-0.3918, acc-0.8879
Iter-74490, train loss-0.2748, acc-0.9000, valid loss-0.3873, acc-0.8918, test loss-0.3918, acc-0.8879
Iter-74500, train loss-0.3026, acc-0.9400, valid loss-0.3873, acc-0.8918, test loss-0.3918, acc-0.8879
Iter-74510, train loss-0.2468, acc-0.9600, valid loss-0.3873, acc-0.8918, test loss-0.3918, acc-0.8879
Iter-74520, train loss-0.4736, acc-0.8400, valid loss-0.3872, acc-0.8918, test loss-0.3918, acc-0.8879
Iter-74530, train loss-0.5082, acc-0.8400, valid loss-0.3872, acc-0.8918, test loss-0.3917, acc-0.8879
Iter-74540, train loss-0.3804, acc-0.8800, valid loss-0.3872, acc-0.8918, test loss-0.3917, acc-0.8879
Iter-74550, train loss-0.3725, acc-0.9000, valid loss-0.3872, acc-0.8920, test loss-0.3917, acc-0.8879
Iter-74560, train loss-0.2884, acc-0.9200, valid loss-0.3872, acc-0.8920, test loss-0.3917, acc-0.8879
Iter-74570, train loss-0.4500, acc-0.8200, valid loss-0.3871, acc-0.8920, test loss-0.3917, acc-0.8879
Iter-74580, train loss-0.2604, acc-0.9400, valid loss-0.3871, acc-0.8920, test loss-0.3916, acc-0.8880
Iter-74590, train loss-0.2751, acc-0.9200, valid loss-0.3871, acc-0.8920, test loss-0.3916, acc-0.8879
Iter-74600, train loss-0.8817, acc-0.7000, valid loss-0.3871, acc-0.8920, test loss-0.3916, acc-0.8880
Iter-74610, train loss-0.5314, acc-0.8400, valid loss-0.3871, acc-0.8920, test loss-0.3916, acc-0.8879
Iter-74620, train loss-0.4507, acc-0.8200, valid loss-0.3870, acc-0.8920, test loss-0.3916, acc-0.8879
Iter-74630, train loss-0.1732, acc-0.9800, valid loss-0.3870, acc-0.8920, test loss-0.3916, acc-0.8879
Iter-74640, train loss-0.4987, acc-0.8400, valid loss-0.3870, acc-0.8918, test loss-0.3915, acc-0.8879
Iter-74650, train loss-0.3392, acc-0.9000, valid loss-0.3870, acc-0.8918, test loss-0.3915, acc-0.8879
Iter-74660, train loss-0.3193, acc-0.8800, valid loss-0.3870, acc-0.8918, test loss-0.3915, acc-0.8880
Iter-74670, train loss-0.3067, acc-0.9200, valid loss-0.3869, acc-0.8918, test loss-0.3915, acc-0.8881
Iter-74680, train loss-0.4514, acc-0.8400, valid loss-0.3869, acc-0.8920, test loss-0.3915, acc-0.8879
Iter-74690, train loss-0.3859, acc-0.8600, valid loss-0.3869, acc-0.8918, test loss-0.3914, acc-0.8879
Iter-74700, train loss-0.3607, acc-0.8600, valid loss-0.3869, acc-0.8918, test loss-0.3914, acc-0.8879
Iter-74710, train loss-0.5463, acc-0.8600, valid loss-0.3869, acc-0.8918, test loss-0.3914, acc-0.8879
Iter-74720, train loss-0.5298, acc-0.8400, valid loss-0.3869, acc-0.8920, test loss-0.3914, acc-0.8879
Iter-74730, train loss-0.3077, acc-0.9400, valid loss-0.3868, acc-0.8920, test loss-0.3914, acc-0.8879
Iter-74740, train loss-0.3447, acc-0.8800, valid loss-0.3868, acc-0.8920, test loss-0.3914, acc-0.8879
Iter-74750, train loss-0.4373, acc-0.8800, valid loss-0.3868, acc-0.8918, test loss-0.3913, acc-0.8880
Iter-74760, train loss-0.4854, acc-0.8600, valid loss-0.3868, acc-0.8918, test loss-0.3913, acc-0.8880
Iter-74770, train loss-0.3377, acc-0.9000, valid loss-0.3868, acc-0.8918, test loss-0.3913, acc-0.8879
Iter-74780, train loss-0.3707, acc-0.8800, valid loss-0.3867, acc-0.8918, test loss-0.3913, acc-0.8879
Iter-74790, train loss-0.3876, acc-0.8800, valid loss-0.3867, acc-0.8918, test loss-0.3913, acc-0.8880
Iter-74800, train loss-0.3621, acc-0.9200, valid loss-0.3867, acc-0.8918, test loss-0.3912, acc-0.8879
Iter-74810, train loss-0.5031, acc-0.8400, valid loss-0.3867, acc-0.8918, test loss-0.3912, acc-0.8879
Iter-74820, train loss-0.3429, acc-0.8800, valid loss-0.3867, acc-0.8918, test loss-0.3912, acc-0.8879
Iter-74830, train loss-0.5181, acc-0.8400, valid loss-0.3867, acc-0.8918, test loss-0.3912, acc-0.8880
Iter-74840, train loss-0.1606, acc-0.9800, valid loss-0.3866, acc-0.8918, test loss-0.3912, acc-0.8881
Iter-74850, train loss-0.3825, acc-0.9200, valid loss-0.3866, acc-0.8918, test loss-0.3911, acc-0.8881
Iter-74860, train loss-0.5683, acc-0.8400, valid loss-0.3866, acc-0.8918, test loss-0.3911, acc-0.8880
Iter-74870, train loss-0.3901, acc-0.9000, valid loss-0.3866, acc-0.8920, test loss-0.3911, acc-0.8881
Iter-74880, train loss-0.3943, acc-0.9000, valid loss-0.3866, acc-0.8920, test loss-0.3911, acc-0.8880
Iter-74890, train loss-0.2111, acc-0.9400, valid loss-0.3865, acc-0.8920, test loss-0.3911, acc-0.8881
Iter-74900, train loss-0.6284, acc-0.8200, valid loss-0.3865, acc-0.8918, test loss-0.3911, acc-0.8881
Iter-74910, train loss-0.4221, acc-0.8800, valid loss-0.3865, acc-0.8918, test loss-0.3910, acc-0.8881
Iter-74920, train loss-0.3819, acc-0.9000, valid loss-0.3865, acc-0.8918, test loss-0.3910, acc-0.8881
Iter-74930, train loss-0.3556, acc-0.9000, valid loss-0.3865, acc-0.8918, test loss-0.3910, acc-0.8880
Iter-74940, train loss-0.5359, acc-0.8400, valid loss-0.3865, acc-0.8918, test loss-0.3910, acc-0.8879
Iter-74950, train loss-0.3133, acc-0.9000, valid loss-0.3864, acc-0.8918, test loss-0.3910, acc-0.8881
Iter-74960, train loss-0.3213, acc-0.8800, valid loss-0.3864, acc-0.8918, test loss-0.3909, acc-0.8880
Iter-74970, train loss-0.3638, acc-0.9000, valid loss-0.3864, acc-0.8918, test loss-0.3909, acc-0.8882
Iter-74980, train loss-0.2768, acc-0.9200, valid loss-0.3864, acc-0.8918, test loss-0.3909, acc-0.8880
Iter-74990, train loss-0.5725, acc-0.8200, valid loss-0.3863, acc-0.8918, test loss-0.3909, acc-0.8881
Iter-75000, train loss-0.4505, acc-0.8800, valid loss-0.3863, acc-0.8916, test loss-0.3909, acc-0.8881
Iter-75010, train loss-0.4125, acc-0.8400, valid loss-0.3863, acc-0.8916, test loss-0.3909, acc-0.8879
Iter-75020, train loss-0.1536, acc-0.9600, valid loss-0.3863, acc-0.8918, test loss-0.3908, acc-0.8880
Iter-75030, train loss-0.4788, acc-0.8400, valid loss-0.3862, acc-0.8918, test loss-0.3908, acc-0.8880
Iter-75040, train loss-0.5714, acc-0.8600, valid loss-0.3862, acc-0.8918, test loss-0.3908, acc-0.8880
Iter-75050, train loss-0.3921, acc-0.8600, valid loss-0.3862, acc-0.8918, test loss-0.3908, acc-0.8880
Iter-75060, train loss-0.3176, acc-0.9200, valid loss-0.3862, acc-0.8918, test loss-0.3907, acc-0.8880
Iter-75070, train loss-0.5426, acc-0.8400, valid loss-0.3862, acc-0.8918, test loss-0.3907, acc-0.8881
Iter-75080, train loss-0.4356, acc-0.8800, valid loss-0.3862, acc-0.8918, test loss-0.3907, acc-0.8881
Iter-75090, train loss-0.3477, acc-0.9000, valid loss-0.3861, acc-0.8918, test loss-0.3907, acc-0.8881
Iter-75100, train loss-0.3764, acc-0.8600, valid loss-0.3861, acc-0.8918, test loss-0.3907, acc-0.8881
Iter-75110, train loss-0.4138, acc-0.8800, valid loss-0.3861, acc-0.8918, test loss-0.3907, acc-0.8880
Iter-75120, train loss-0.3419, acc-0.8800, valid loss-0.3861, acc-0.8918, test loss-0.3906, acc-0.8880
Iter-75130, train loss-0.3023, acc-0.9200, valid loss-0.3861, acc-0.8918, test loss-0.3906, acc-0.8880
Iter-75140, train loss-0.3997, acc-0.8800, valid loss-0.3860, acc-0.8920, test loss-0.3906, acc-0.8880
Iter-75150, train loss-0.3625, acc-0.8800, valid loss-0.3860, acc-0.8920, test loss-0.3906, acc-0.8880
Iter-75160, train loss-0.3651, acc-0.9200, valid loss-0.3860, acc-0.8916, test loss-0.3906, acc-0.8880
Iter-75170, train loss-0.4391, acc-0.9200, valid loss-0.3860, acc-0.8916, test loss-0.3905, acc-0.8881
Iter-75180, train loss-0.3875, acc-0.9400, valid loss-0.3860, acc-0.8916, test loss-0.3905, acc-0.8882
Iter-75190, train loss-0.3807, acc-0.9400, valid loss-0.3860, acc-0.8916, test loss-0.3905, acc-0.8882
Iter-75200, train loss-0.5815, acc-0.8600, valid loss-0.3860, acc-0.8916, test loss-0.3905, acc-0.8882
Iter-75210, train loss-0.3703, acc-0.9000, valid loss-0.3859, acc-0.8916, test loss-0.3905, acc-0.8881
Iter-75220, train loss-0.6864, acc-0.7800, valid loss-0.3859, acc-0.8916, test loss-0.3905, acc-0.8881
Iter-75230, train loss-0.5540, acc-0.8600, valid loss-0.3859, acc-0.8916, test loss-0.3905, acc-0.8881
Iter-75240, train loss-0.4444, acc-0.9000, valid loss-0.3859, acc-0.8916, test loss-0.3904, acc-0.8882
Iter-75250, train loss-0.4079, acc-0.8800, valid loss-0.3859, acc-0.8916, test loss-0.3904, acc-0.8881
Iter-75260, train loss-0.4155, acc-0.9000, valid loss-0.3858, acc-0.8916, test loss-0.3904, acc-0.8881
Iter-75270, train loss-0.4880, acc-0.9400, valid loss-0.3858, acc-0.8916, test loss-0.3904, acc-0.8880
Iter-75280, train loss-0.5502, acc-0.8000, valid loss-0.3858, acc-0.8918, test loss-0.3904, acc-0.8880
Iter-75290, train loss-0.3583, acc-0.9400, valid loss-0.3858, acc-0.8918, test loss-0.3903, acc-0.8881
Iter-75300, train loss-0.4763, acc-0.9000, valid loss-0.3857, acc-0.8918, test loss-0.3903, acc-0.8881
Iter-75310, train loss-0.7363, acc-0.8000, valid loss-0.3857, acc-0.8918, test loss-0.3903, acc-0.8881
Iter-75320, train loss-0.3737, acc-0.8800, valid loss-0.3857, acc-0.8918, test loss-0.3903, acc-0.8879
Iter-75330, train loss-0.5701, acc-0.8200, valid loss-0.3857, acc-0.8918, test loss-0.3903, acc-0.8881
Iter-75340, train loss-0.4375, acc-0.8800, valid loss-0.3857, acc-0.8918, test loss-0.3903, acc-0.8881
Iter-75350, train loss-0.3041, acc-0.9000, valid loss-0.3857, acc-0.8918, test loss-0.3902, acc-0.8881
Iter-75360, train loss-0.2589, acc-0.9200, valid loss-0.3856, acc-0.8918, test loss-0.3902, acc-0.8881
Iter-75370, train loss-0.4956, acc-0.9000, valid loss-0.3856, acc-0.8918, test loss-0.3902, acc-0.8881
Iter-75380, train loss-0.2491, acc-0.9600, valid loss-0.3856, acc-0.8918, test loss-0.3902, acc-0.8882
Iter-75390, train loss-0.3660, acc-0.9200, valid loss-0.3856, acc-0.8918, test loss-0.3902, acc-0.8882
Iter-75400, train loss-0.2273, acc-0.9200, valid loss-0.3855, acc-0.8918, test loss-0.3902, acc-0.8882
Iter-75410, train loss-0.4227, acc-0.8800, valid loss-0.3855, acc-0.8918, test loss-0.3901, acc-0.8882
Iter-75420, train loss-0.5601, acc-0.9000, valid loss-0.3855, acc-0.8916, test loss-0.3901, acc-0.8882
Iter-75430, train loss-0.2608, acc-0.9400, valid loss-0.3855, acc-0.8916, test loss-0.3901, acc-0.8882
Iter-75440, train loss-0.3319, acc-0.8600, valid loss-0.3855, acc-0.8916, test loss-0.3901, acc-0.8883
Iter-75450, train loss-0.4572, acc-0.9000, valid loss-0.3854, acc-0.8918, test loss-0.3901, acc-0.8883
Iter-75460, train loss-0.3240, acc-0.9000, valid loss-0.3854, acc-0.8916, test loss-0.3900, acc-0.8884
Iter-75470, train loss-0.2665, acc-0.9200, valid loss-0.3854, acc-0.8918, test loss-0.3900, acc-0.8883
Iter-75480, train loss-0.5715, acc-0.8400, valid loss-0.3854, acc-0.8918, test loss-0.3900, acc-0.8883
Iter-75490, train loss-0.3714, acc-0.8800, valid loss-0.3854, acc-0.8918, test loss-0.3900, acc-0.8882
Iter-75500, train loss-0.4307, acc-0.8800, valid loss-0.3854, acc-0.8918, test loss-0.3900, acc-0.8883
Iter-75510, train loss-0.3720, acc-0.8800, valid loss-0.3854, acc-0.8918, test loss-0.3900, acc-0.8883
Iter-75520, train loss-0.3874, acc-0.9000, valid loss-0.3853, acc-0.8918, test loss-0.3899, acc-0.8884
Iter-75530, train loss-0.3385, acc-0.8800, valid loss-0.3853, acc-0.8918, test loss-0.3899, acc-0.8884
Iter-75540, train loss-0.4413, acc-0.8200, valid loss-0.3853, acc-0.8916, test loss-0.3899, acc-0.8884
Iter-75550, train loss-0.3594, acc-0.8800, valid loss-0.3853, acc-0.8918, test loss-0.3899, acc-0.8884
Iter-75560, train loss-0.5776, acc-0.8600, valid loss-0.3852, acc-0.8918, test loss-0.3899, acc-0.8884
Iter-75570, train loss-0.3352, acc-0.9600, valid loss-0.3852, acc-0.8918, test loss-0.3898, acc-0.8884
Iter-75580, train loss-0.3263, acc-0.8800, valid loss-0.3852, acc-0.8916, test loss-0.3898, acc-0.8884
Iter-75590, train loss-0.3296, acc-0.9200, valid loss-0.3852, acc-0.8918, test loss-0.3898, acc-0.8883
Iter-75600, train loss-0.3684, acc-0.9000, valid loss-0.3852, acc-0.8918, test loss-0.3898, acc-0.8883
Iter-75610, train loss-0.4484, acc-0.9000, valid loss-0.3852, acc-0.8916, test loss-0.3897, acc-0.8886
Iter-75620, train loss-0.4092, acc-0.8800, valid loss-0.3851, acc-0.8916, test loss-0.3897, acc-0.8885
Iter-75630, train loss-0.2833, acc-0.9400, valid loss-0.3851, acc-0.8914, test loss-0.3897, acc-0.8886
Iter-75640, train loss-0.3845, acc-0.9200, valid loss-0.3851, acc-0.8916, test loss-0.3897, acc-0.8885
Iter-75650, train loss-0.6545, acc-0.8600, valid loss-0.3851, acc-0.8914, test loss-0.3896, acc-0.8885
Iter-75660, train loss-0.3298, acc-0.9000, valid loss-0.3851, acc-0.8916, test loss-0.3896, acc-0.8885
Iter-75670, train loss-0.3665, acc-0.9000, valid loss-0.3850, acc-0.8918, test loss-0.3896, acc-0.8889
Iter-75680, train loss-0.2769, acc-0.9400, valid loss-0.3850, acc-0.8918, test loss-0.3896, acc-0.8888
Iter-75690, train loss-0.4354, acc-0.8400, valid loss-0.3850, acc-0.8918, test loss-0.3896, acc-0.8888
Iter-75700, train loss-0.5973, acc-0.8600, valid loss-0.3850, acc-0.8918, test loss-0.3895, acc-0.8888
Iter-75710, train loss-0.4420, acc-0.8800, valid loss-0.3850, acc-0.8920, test loss-0.3895, acc-0.8888
Iter-75720, train loss-0.4245, acc-0.8800, valid loss-0.3849, acc-0.8920, test loss-0.3895, acc-0.8888
Iter-75730, train loss-0.3636, acc-0.8800, valid loss-0.3849, acc-0.8920, test loss-0.3895, acc-0.8888
Iter-75740, train loss-0.6075, acc-0.8600, valid loss-0.3849, acc-0.8920, test loss-0.3895, acc-0.8888
Iter-75750, train loss-0.3448, acc-0.9000, valid loss-0.3849, acc-0.8920, test loss-0.3895, acc-0.8887
Iter-75760, train loss-0.3125, acc-0.9600, valid loss-0.3849, acc-0.8920, test loss-0.3894, acc-0.8886
Iter-75770, train loss-0.4981, acc-0.8400, valid loss-0.3848, acc-0.8920, test loss-0.3894, acc-0.8886
Iter-75780, train loss-0.4335, acc-0.8600, valid loss-0.3848, acc-0.8920, test loss-0.3894, acc-0.8887
Iter-75790, train loss-0.6143, acc-0.7800, valid loss-0.3848, acc-0.8922, test loss-0.3894, acc-0.8887
Iter-75800, train loss-0.2917, acc-0.9200, valid loss-0.3848, acc-0.8920, test loss-0.3894, acc-0.8887
Iter-75810, train loss-0.5452, acc-0.8400, valid loss-0.3848, acc-0.8920, test loss-0.3893, acc-0.8888
Iter-75820, train loss-0.2503, acc-0.9000, valid loss-0.3848, acc-0.8920, test loss-0.3893, acc-0.8887
Iter-75830, train loss-0.2318, acc-0.9200, valid loss-0.3847, acc-0.8920, test loss-0.3893, acc-0.8889
Iter-75840, train loss-0.3279, acc-0.9200, valid loss-0.3847, acc-0.8920, test loss-0.3893, acc-0.8888
Iter-75850, train loss-0.5133, acc-0.8200, valid loss-0.3847, acc-0.8922, test loss-0.3893, acc-0.8888
Iter-75860, train loss-0.6032, acc-0.8000, valid loss-0.3847, acc-0.8920, test loss-0.3892, acc-0.8888
Iter-75870, train loss-0.4873, acc-0.8600, valid loss-0.3846, acc-0.8922, test loss-0.3892, acc-0.8888
Iter-75880, train loss-0.4085, acc-0.8400, valid loss-0.3846, acc-0.8924, test loss-0.3892, acc-0.8889
Iter-75890, train loss-0.5404, acc-0.8600, valid loss-0.3846, acc-0.8924, test loss-0.3892, acc-0.8889
Iter-75900, train loss-0.1928, acc-0.9400, valid loss-0.3846, acc-0.8924, test loss-0.3892, acc-0.8889
Iter-75910, train loss-0.4760, acc-0.8800, valid loss-0.3845, acc-0.8924, test loss-0.3892, acc-0.8889
Iter-75920, train loss-0.5913, acc-0.8400, valid loss-0.3845, acc-0.8924, test loss-0.3891, acc-0.8890
Iter-75930, train loss-0.3557, acc-0.9200, valid loss-0.3845, acc-0.8926, test loss-0.3891, acc-0.8890
Iter-75940, train loss-0.2116, acc-0.9800, valid loss-0.3845, acc-0.8922, test loss-0.3891, acc-0.8891
Iter-75950, train loss-0.5429, acc-0.8800, valid loss-0.3845, acc-0.8922, test loss-0.3891, acc-0.8891
Iter-75960, train loss-0.1514, acc-0.9800, valid loss-0.3844, acc-0.8922, test loss-0.3891, acc-0.8890
Iter-75970, train loss-0.2706, acc-0.9200, valid loss-0.3844, acc-0.8924, test loss-0.3890, acc-0.8889
Iter-75980, train loss-0.2593, acc-0.9400, valid loss-0.3844, acc-0.8924, test loss-0.3890, acc-0.8890
Iter-75990, train loss-0.3371, acc-0.9400, valid loss-0.3844, acc-0.8924, test loss-0.3890, acc-0.8890
Iter-76000, train loss-0.3362, acc-0.9200, valid loss-0.3844, acc-0.8922, test loss-0.3890, acc-0.8890
Iter-76010, train loss-0.2410, acc-0.9200, valid loss-0.3843, acc-0.8924, test loss-0.3890, acc-0.8891
Iter-76020, train loss-0.3741, acc-0.9000, valid loss-0.3843, acc-0.8922, test loss-0.3890, acc-0.8891
Iter-76030, train loss-0.4226, acc-0.8600, valid loss-0.3843, acc-0.8922, test loss-0.3889, acc-0.8890
Iter-76040, train loss-0.4800, acc-0.8000, valid loss-0.3843, acc-0.8922, test loss-0.3889, acc-0.8890
Iter-76050, train loss-0.5771, acc-0.8200, valid loss-0.3843, acc-0.8922, test loss-0.3889, acc-0.8892
Iter-76060, train loss-0.3118, acc-0.9200, valid loss-0.3842, acc-0.8922, test loss-0.3889, acc-0.8891
Iter-76070, train loss-0.6894, acc-0.7600, valid loss-0.3842, acc-0.8924, test loss-0.3889, acc-0.8892
Iter-76080, train loss-0.2876, acc-0.9000, valid loss-0.3842, acc-0.8924, test loss-0.3888, acc-0.8891
Iter-76090, train loss-0.3355, acc-0.9000, valid loss-0.3842, acc-0.8922, test loss-0.3888, acc-0.8891
Iter-76100, train loss-0.7571, acc-0.7800, valid loss-0.3842, acc-0.8922, test loss-0.3888, acc-0.8892
Iter-76110, train loss-0.4432, acc-0.9000, valid loss-0.3842, acc-0.8922, test loss-0.3888, acc-0.8892
Iter-76120, train loss-0.4609, acc-0.8800, valid loss-0.3842, acc-0.8922, test loss-0.3888, acc-0.8893
Iter-76130, train loss-0.4834, acc-0.8800, valid loss-0.3841, acc-0.8922, test loss-0.3888, acc-0.8893
Iter-76140, train loss-0.4605, acc-0.8800, valid loss-0.3841, acc-0.8922, test loss-0.3887, acc-0.8893
Iter-76150, train loss-0.3110, acc-0.9400, valid loss-0.3841, acc-0.8920, test loss-0.3887, acc-0.8892
Iter-76160, train loss-0.4552, acc-0.8200, valid loss-0.3841, acc-0.8920, test loss-0.3887, acc-0.8893
Iter-76170, train loss-0.3108, acc-0.9000, valid loss-0.3841, acc-0.8920, test loss-0.3887, acc-0.8891
Iter-76180, train loss-0.5335, acc-0.8200, valid loss-0.3841, acc-0.8920, test loss-0.3887, acc-0.8891
Iter-76190, train loss-0.2182, acc-0.9400, valid loss-0.3840, acc-0.8920, test loss-0.3886, acc-0.8891
Iter-76200, train loss-0.2493, acc-0.9600, valid loss-0.3840, acc-0.8920, test loss-0.3886, acc-0.8892
Iter-76210, train loss-0.2515, acc-0.9200, valid loss-0.3840, acc-0.8920, test loss-0.3886, acc-0.8891
Iter-76220, train loss-0.3194, acc-0.8800, valid loss-0.3840, acc-0.8922, test loss-0.3886, acc-0.8889
Iter-76230, train loss-0.6172, acc-0.8800, valid loss-0.3840, acc-0.8920, test loss-0.3886, acc-0.8890
Iter-76240, train loss-0.6047, acc-0.8600, valid loss-0.3840, acc-0.8920, test loss-0.3886, acc-0.8891
Iter-76250, train loss-0.3268, acc-0.8800, valid loss-0.3839, acc-0.8920, test loss-0.3885, acc-0.8891
Iter-76260, train loss-0.5349, acc-0.8400, valid loss-0.3839, acc-0.8920, test loss-0.3885, acc-0.8891
Iter-76270, train loss-0.5884, acc-0.8000, valid loss-0.3839, acc-0.8920, test loss-0.3885, acc-0.8891
Iter-76280, train loss-0.4007, acc-0.9000, valid loss-0.3839, acc-0.8922, test loss-0.3885, acc-0.8890
Iter-76290, train loss-0.5260, acc-0.8600, valid loss-0.3838, acc-0.8922, test loss-0.3885, acc-0.8892
Iter-76300, train loss-0.4777, acc-0.8200, valid loss-0.3838, acc-0.8922, test loss-0.3884, acc-0.8891
Iter-76310, train loss-0.4163, acc-0.8600, valid loss-0.3838, acc-0.8924, test loss-0.3884, acc-0.8893
Iter-76320, train loss-0.5023, acc-0.8400, valid loss-0.3838, acc-0.8922, test loss-0.3884, acc-0.8892
Iter-76330, train loss-0.3338, acc-0.9200, valid loss-0.3838, acc-0.8922, test loss-0.3884, acc-0.8893
Iter-76340, train loss-0.5842, acc-0.8200, valid loss-0.3837, acc-0.8922, test loss-0.3883, acc-0.8893
Iter-76350, train loss-0.2640, acc-0.9000, valid loss-0.3837, acc-0.8922, test loss-0.3883, acc-0.8894
Iter-76360, train loss-0.5249, acc-0.8200, valid loss-0.3837, acc-0.8922, test loss-0.3883, acc-0.8892
Iter-76370, train loss-0.3437, acc-0.8800, valid loss-0.3837, acc-0.8922, test loss-0.3883, acc-0.8895
Iter-76380, train loss-0.4377, acc-0.9000, valid loss-0.3837, acc-0.8924, test loss-0.3883, acc-0.8894
Iter-76390, train loss-0.2510, acc-0.9600, valid loss-0.3836, acc-0.8924, test loss-0.3882, acc-0.8894
Iter-76400, train loss-0.2757, acc-0.9400, valid loss-0.3836, acc-0.8924, test loss-0.3882, acc-0.8894
Iter-76410, train loss-0.3069, acc-0.9600, valid loss-0.3836, acc-0.8924, test loss-0.3882, acc-0.8894
Iter-76420, train loss-0.4925, acc-0.8800, valid loss-0.3836, acc-0.8924, test loss-0.3882, acc-0.8894
Iter-76430, train loss-0.3862, acc-0.9000, valid loss-0.3836, acc-0.8924, test loss-0.3882, acc-0.8894
Iter-76440, train loss-0.2058, acc-0.9800, valid loss-0.3835, acc-0.8924, test loss-0.3881, acc-0.8895
Iter-76450, train loss-0.7178, acc-0.7600, valid loss-0.3835, acc-0.8924, test loss-0.3881, acc-0.8895
Iter-76460, train loss-0.4376, acc-0.8200, valid loss-0.3835, acc-0.8924, test loss-0.3881, acc-0.8895
Iter-76470, train loss-0.4031, acc-0.9400, valid loss-0.3835, acc-0.8924, test loss-0.3881, acc-0.8895
Iter-76480, train loss-0.4939, acc-0.8200, valid loss-0.3835, acc-0.8924, test loss-0.3881, acc-0.8896
Iter-76490, train loss-0.3315, acc-0.9400, valid loss-0.3835, acc-0.8924, test loss-0.3881, acc-0.8896
Iter-76500, train loss-0.3238, acc-0.9600, valid loss-0.3835, acc-0.8924, test loss-0.3880, acc-0.8895
Iter-76510, train loss-0.3524, acc-0.9000, valid loss-0.3834, acc-0.8924, test loss-0.3880, acc-0.8894
Iter-76520, train loss-0.5929, acc-0.8000, valid loss-0.3834, acc-0.8924, test loss-0.3880, acc-0.8893
Iter-76530, train loss-0.2430, acc-0.9200, valid loss-0.3834, acc-0.8924, test loss-0.3880, acc-0.8895
Iter-76540, train loss-0.4347, acc-0.8200, valid loss-0.3834, acc-0.8924, test loss-0.3880, acc-0.8896
Iter-76550, train loss-0.3006, acc-0.9400, valid loss-0.3834, acc-0.8924, test loss-0.3880, acc-0.8896
Iter-76560, train loss-0.4810, acc-0.8800, valid loss-0.3833, acc-0.8924, test loss-0.3879, acc-0.8896
Iter-76570, train loss-0.5092, acc-0.8200, valid loss-0.3833, acc-0.8924, test loss-0.3879, acc-0.8896
Iter-76580, train loss-0.2225, acc-0.9400, valid loss-0.3833, acc-0.8924, test loss-0.3879, acc-0.8896
Iter-76590, train loss-0.3407, acc-0.9000, valid loss-0.3833, acc-0.8924, test loss-0.3879, acc-0.8896
Iter-76600, train loss-0.2767, acc-0.9200, valid loss-0.3833, acc-0.8924, test loss-0.3879, acc-0.8895
Iter-76610, train loss-0.2657, acc-0.9400, valid loss-0.3833, acc-0.8924, test loss-0.3879, acc-0.8895
Iter-76620, train loss-0.4313, acc-0.8800, valid loss-0.3833, acc-0.8924, test loss-0.3878, acc-0.8895
Iter-76630, train loss-0.3374, acc-0.8800, valid loss-0.3832, acc-0.8924, test loss-0.3878, acc-0.8895
Iter-76640, train loss-0.3224, acc-0.9000, valid loss-0.3832, acc-0.8924, test loss-0.3878, acc-0.8896
Iter-76650, train loss-0.4528, acc-0.8200, valid loss-0.3832, acc-0.8924, test loss-0.3878, acc-0.8896
Iter-76660, train loss-0.2583, acc-0.9600, valid loss-0.3832, acc-0.8924, test loss-0.3878, acc-0.8896
Iter-76670, train loss-0.4726, acc-0.8800, valid loss-0.3831, acc-0.8924, test loss-0.3877, acc-0.8896
Iter-76680, train loss-0.4703, acc-0.8600, valid loss-0.3831, acc-0.8926, test loss-0.3877, acc-0.8896
Iter-76690, train loss-0.5655, acc-0.8600, valid loss-0.3831, acc-0.8926, test loss-0.3877, acc-0.8896
Iter-76700, train loss-0.3304, acc-0.9200, valid loss-0.3831, acc-0.8924, test loss-0.3877, acc-0.8896
Iter-76710, train loss-0.6516, acc-0.8200, valid loss-0.3830, acc-0.8924, test loss-0.3877, acc-0.8896
Iter-76720, train loss-0.2523, acc-0.9200, valid loss-0.3830, acc-0.8924, test loss-0.3877, acc-0.8895
Iter-76730, train loss-0.3584, acc-0.9000, valid loss-0.3830, acc-0.8924, test loss-0.3876, acc-0.8895
Iter-76740, train loss-0.3822, acc-0.9200, valid loss-0.3830, acc-0.8926, test loss-0.3876, acc-0.8895
Iter-76750, train loss-0.6533, acc-0.8800, valid loss-0.3830, acc-0.8924, test loss-0.3876, acc-0.8896
Iter-76760, train loss-0.2092, acc-0.9600, valid loss-0.3829, acc-0.8926, test loss-0.3876, acc-0.8896
Iter-76770, train loss-0.3810, acc-0.9200, valid loss-0.3829, acc-0.8926, test loss-0.3876, acc-0.8895
Iter-76780, train loss-0.2981, acc-0.9400, valid loss-0.3829, acc-0.8924, test loss-0.3875, acc-0.8895
Iter-76790, train loss-0.2692, acc-0.9000, valid loss-0.3829, acc-0.8926, test loss-0.3875, acc-0.8895
Iter-76800, train loss-0.4685, acc-0.8400, valid loss-0.3828, acc-0.8926, test loss-0.3875, acc-0.8895
Iter-76810, train loss-0.4105, acc-0.9000, valid loss-0.3828, acc-0.8924, test loss-0.3875, acc-0.8895
Iter-76820, train loss-0.4384, acc-0.8600, valid loss-0.3828, acc-0.8924, test loss-0.3875, acc-0.8895
Iter-76830, train loss-0.3281, acc-0.9000, valid loss-0.3828, acc-0.8924, test loss-0.3874, acc-0.8894
Iter-76840, train loss-0.3016, acc-0.9000, valid loss-0.3828, acc-0.8924, test loss-0.3874, acc-0.8894
Iter-76850, train loss-0.4002, acc-0.8800, valid loss-0.3827, acc-0.8924, test loss-0.3874, acc-0.8894
Iter-76860, train loss-0.4289, acc-0.8400, valid loss-0.3827, acc-0.8924, test loss-0.3874, acc-0.8894
Iter-76870, train loss-0.3837, acc-0.8200, valid loss-0.3827, acc-0.8922, test loss-0.3874, acc-0.8894
Iter-76880, train loss-0.4211, acc-0.9000, valid loss-0.3827, acc-0.8922, test loss-0.3873, acc-0.8894
Iter-76890, train loss-0.3509, acc-0.9000, valid loss-0.3827, acc-0.8922, test loss-0.3873, acc-0.8895
Iter-76900, train loss-0.3921, acc-0.9000, valid loss-0.3827, acc-0.8922, test loss-0.3873, acc-0.8895
Iter-76910, train loss-0.2325, acc-0.9600, valid loss-0.3826, acc-0.8922, test loss-0.3873, acc-0.8895
Iter-76920, train loss-0.5821, acc-0.8400, valid loss-0.3826, acc-0.8922, test loss-0.3873, acc-0.8895
Iter-76930, train loss-0.2248, acc-0.9800, valid loss-0.3826, acc-0.8922, test loss-0.3873, acc-0.8895
Iter-76940, train loss-0.5799, acc-0.8400, valid loss-0.3826, acc-0.8922, test loss-0.3873, acc-0.8895
Iter-76950, train loss-0.2069, acc-0.9600, valid loss-0.3826, acc-0.8922, test loss-0.3872, acc-0.8895
Iter-76960, train loss-0.4346, acc-0.8600, valid loss-0.3825, acc-0.8926, test loss-0.3872, acc-0.8895
Iter-76970, train loss-0.3506, acc-0.9200, valid loss-0.3825, acc-0.8922, test loss-0.3872, acc-0.8896
Iter-76980, train loss-0.3002, acc-0.9200, valid loss-0.3825, acc-0.8922, test loss-0.3872, acc-0.8897
Iter-76990, train loss-0.4969, acc-0.8600, valid loss-0.3825, acc-0.8922, test loss-0.3872, acc-0.8898
Iter-77000, train loss-0.3097, acc-0.8800, valid loss-0.3825, acc-0.8924, test loss-0.3871, acc-0.8898
Iter-77010, train loss-0.1873, acc-0.9800, valid loss-0.3824, acc-0.8924, test loss-0.3871, acc-0.8898
Iter-77020, train loss-0.3544, acc-0.9000, valid loss-0.3824, acc-0.8924, test loss-0.3871, acc-0.8898
Iter-77030, train loss-0.3318, acc-0.9000, valid loss-0.3824, acc-0.8924, test loss-0.3871, acc-0.8898
Iter-77040, train loss-0.4977, acc-0.8600, valid loss-0.3824, acc-0.8924, test loss-0.3871, acc-0.8898
Iter-77050, train loss-0.5445, acc-0.8800, valid loss-0.3824, acc-0.8924, test loss-0.3871, acc-0.8898
Iter-77060, train loss-0.6740, acc-0.7800, valid loss-0.3824, acc-0.8924, test loss-0.3871, acc-0.8898
Iter-77070, train loss-0.4289, acc-0.8600, valid loss-0.3823, acc-0.8924, test loss-0.3870, acc-0.8897
Iter-77080, train loss-0.4561, acc-0.8000, valid loss-0.3823, acc-0.8924, test loss-0.3870, acc-0.8896
Iter-77090, train loss-0.2960, acc-0.9600, valid loss-0.3823, acc-0.8924, test loss-0.3870, acc-0.8896
Iter-77100, train loss-0.3032, acc-0.9200, valid loss-0.3823, acc-0.8924, test loss-0.3870, acc-0.8897
Iter-77110, train loss-0.3622, acc-0.9200, valid loss-0.3823, acc-0.8924, test loss-0.3870, acc-0.8897
Iter-77120, train loss-0.5871, acc-0.8400, valid loss-0.3822, acc-0.8924, test loss-0.3869, acc-0.8898
Iter-77130, train loss-0.2860, acc-0.9200, valid loss-0.3822, acc-0.8924, test loss-0.3869, acc-0.8897
Iter-77140, train loss-0.4191, acc-0.9000, valid loss-0.3822, acc-0.8926, test loss-0.3869, acc-0.8897
Iter-77150, train loss-0.5891, acc-0.8200, valid loss-0.3822, acc-0.8926, test loss-0.3869, acc-0.8897
Iter-77160, train loss-0.4554, acc-0.8600, valid loss-0.3822, acc-0.8926, test loss-0.3869, acc-0.8898
Iter-77170, train loss-0.5962, acc-0.8400, valid loss-0.3822, acc-0.8926, test loss-0.3869, acc-0.8898
Iter-77180, train loss-0.3189, acc-0.8800, valid loss-0.3822, acc-0.8926, test loss-0.3868, acc-0.8898
Iter-77190, train loss-0.3214, acc-0.9200, valid loss-0.3821, acc-0.8924, test loss-0.3868, acc-0.8897
Iter-77200, train loss-0.2683, acc-0.9400, valid loss-0.3821, acc-0.8926, test loss-0.3868, acc-0.8898
Iter-77210, train loss-0.4805, acc-0.9000, valid loss-0.3821, acc-0.8924, test loss-0.3868, acc-0.8897
Iter-77220, train loss-0.3504, acc-0.9000, valid loss-0.3821, acc-0.8924, test loss-0.3868, acc-0.8897
Iter-77230, train loss-0.2781, acc-0.9200, valid loss-0.3821, acc-0.8924, test loss-0.3867, acc-0.8897
Iter-77240, train loss-0.7028, acc-0.8000, valid loss-0.3820, acc-0.8924, test loss-0.3867, acc-0.8897
Iter-77250, train loss-0.4982, acc-0.8000, valid loss-0.3820, acc-0.8924, test loss-0.3867, acc-0.8898
Iter-77260, train loss-0.4519, acc-0.8400, valid loss-0.3820, acc-0.8924, test loss-0.3867, acc-0.8898
Iter-77270, train loss-0.3702, acc-0.8600, valid loss-0.3820, acc-0.8924, test loss-0.3867, acc-0.8899
Iter-77280, train loss-0.4325, acc-0.8200, valid loss-0.3820, acc-0.8924, test loss-0.3866, acc-0.8899
Iter-77290, train loss-0.2901, acc-0.9400, valid loss-0.3819, acc-0.8924, test loss-0.3866, acc-0.8900
Iter-77300, train loss-0.3632, acc-0.8600, valid loss-0.3819, acc-0.8926, test loss-0.3866, acc-0.8899
Iter-77310, train loss-0.3572, acc-0.9200, valid loss-0.3819, acc-0.8926, test loss-0.3866, acc-0.8900
Iter-77320, train loss-0.3207, acc-0.9000, valid loss-0.3819, acc-0.8926, test loss-0.3866, acc-0.8900
Iter-77330, train loss-0.4041, acc-0.8600, valid loss-0.3818, acc-0.8924, test loss-0.3865, acc-0.8899
Iter-77340, train loss-0.5715, acc-0.8800, valid loss-0.3818, acc-0.8926, test loss-0.3865, acc-0.8898
Iter-77350, train loss-0.5291, acc-0.8400, valid loss-0.3818, acc-0.8928, test loss-0.3865, acc-0.8898
Iter-77360, train loss-0.3582, acc-0.9600, valid loss-0.3818, acc-0.8928, test loss-0.3865, acc-0.8899
Iter-77370, train loss-0.4486, acc-0.8600, valid loss-0.3818, acc-0.8928, test loss-0.3865, acc-0.8899
Iter-77380, train loss-0.4910, acc-0.8800, valid loss-0.3818, acc-0.8928, test loss-0.3865, acc-0.8899
Iter-77390, train loss-0.6491, acc-0.8400, valid loss-0.3818, acc-0.8928, test loss-0.3864, acc-0.8899
Iter-77400, train loss-0.3083, acc-0.9400, valid loss-0.3817, acc-0.8928, test loss-0.3864, acc-0.8899
Iter-77410, train loss-0.2075, acc-0.9400, valid loss-0.3817, acc-0.8928, test loss-0.3864, acc-0.8899
Iter-77420, train loss-0.4115, acc-0.8400, valid loss-0.3817, acc-0.8928, test loss-0.3864, acc-0.8900
Iter-77430, train loss-0.4768, acc-0.8200, valid loss-0.3817, acc-0.8930, test loss-0.3864, acc-0.8900
Iter-77440, train loss-0.4052, acc-0.8800, valid loss-0.3817, acc-0.8928, test loss-0.3864, acc-0.8900
Iter-77450, train loss-0.2997, acc-0.9400, valid loss-0.3817, acc-0.8930, test loss-0.3864, acc-0.8899
Iter-77460, train loss-0.2730, acc-0.9400, valid loss-0.3817, acc-0.8930, test loss-0.3863, acc-0.8899
Iter-77470, train loss-0.2659, acc-0.9400, valid loss-0.3817, acc-0.8930, test loss-0.3863, acc-0.8899
Iter-77480, train loss-0.3446, acc-0.8800, valid loss-0.3816, acc-0.8930, test loss-0.3863, acc-0.8899
Iter-77490, train loss-0.2927, acc-0.8800, valid loss-0.3816, acc-0.8928, test loss-0.3863, acc-0.8898
Iter-77500, train loss-0.2283, acc-0.9400, valid loss-0.3816, acc-0.8930, test loss-0.3863, acc-0.8900
Iter-77510, train loss-0.4445, acc-0.9000, valid loss-0.3816, acc-0.8928, test loss-0.3863, acc-0.8899
Iter-77520, train loss-0.4164, acc-0.8800, valid loss-0.3816, acc-0.8928, test loss-0.3862, acc-0.8899
Iter-77530, train loss-0.4049, acc-0.9000, valid loss-0.3816, acc-0.8928, test loss-0.3862, acc-0.8899
Iter-77540, train loss-0.5046, acc-0.8600, valid loss-0.3815, acc-0.8928, test loss-0.3862, acc-0.8899
Iter-77550, train loss-0.3470, acc-0.9000, valid loss-0.3815, acc-0.8928, test loss-0.3862, acc-0.8899
Iter-77560, train loss-0.2752, acc-0.9400, valid loss-0.3815, acc-0.8930, test loss-0.3862, acc-0.8899
Iter-77570, train loss-0.3731, acc-0.9000, valid loss-0.3815, acc-0.8930, test loss-0.3862, acc-0.8899
Iter-77580, train loss-0.4322, acc-0.8400, valid loss-0.3815, acc-0.8930, test loss-0.3861, acc-0.8899
Iter-77590, train loss-0.6286, acc-0.8200, valid loss-0.3814, acc-0.8930, test loss-0.3861, acc-0.8899
Iter-77600, train loss-0.4979, acc-0.8800, valid loss-0.3814, acc-0.8932, test loss-0.3861, acc-0.8901
Iter-77610, train loss-0.5655, acc-0.8000, valid loss-0.3814, acc-0.8932, test loss-0.3861, acc-0.8899
Iter-77620, train loss-0.3102, acc-0.9400, valid loss-0.3814, acc-0.8932, test loss-0.3861, acc-0.8901
Iter-77630, train loss-0.4122, acc-0.8600, valid loss-0.3814, acc-0.8932, test loss-0.3860, acc-0.8902
Iter-77640, train loss-0.4834, acc-0.8200, valid loss-0.3814, acc-0.8932, test loss-0.3860, acc-0.8901
Iter-77650, train loss-0.2907, acc-0.8800, valid loss-0.3814, acc-0.8932, test loss-0.3860, acc-0.8902
Iter-77660, train loss-0.4396, acc-0.8800, valid loss-0.3814, acc-0.8930, test loss-0.3860, acc-0.8901
Iter-77670, train loss-0.4130, acc-0.8600, valid loss-0.3813, acc-0.8930, test loss-0.3860, acc-0.8902
Iter-77680, train loss-0.4844, acc-0.9000, valid loss-0.3813, acc-0.8930, test loss-0.3860, acc-0.8902
Iter-77690, train loss-0.1940, acc-0.9600, valid loss-0.3813, acc-0.8930, test loss-0.3860, acc-0.8901
Iter-77700, train loss-0.3773, acc-0.8600, valid loss-0.3813, acc-0.8930, test loss-0.3859, acc-0.8901
Iter-77710, train loss-0.3866, acc-0.8600, valid loss-0.3813, acc-0.8930, test loss-0.3859, acc-0.8900
Iter-77720, train loss-0.3873, acc-0.9000, valid loss-0.3813, acc-0.8930, test loss-0.3859, acc-0.8901
Iter-77730, train loss-0.4237, acc-0.8800, valid loss-0.3813, acc-0.8930, test loss-0.3859, acc-0.8902
Iter-77740, train loss-0.2992, acc-0.9000, valid loss-0.3812, acc-0.8930, test loss-0.3859, acc-0.8903
Iter-77750, train loss-0.5019, acc-0.8800, valid loss-0.3812, acc-0.8928, test loss-0.3858, acc-0.8904
Iter-77760, train loss-0.6224, acc-0.8600, valid loss-0.3812, acc-0.8930, test loss-0.3858, acc-0.8903
Iter-77770, train loss-0.3064, acc-0.9000, valid loss-0.3812, acc-0.8930, test loss-0.3858, acc-0.8903
Iter-77780, train loss-0.4075, acc-0.8800, valid loss-0.3812, acc-0.8930, test loss-0.3858, acc-0.8903
Iter-77790, train loss-0.3088, acc-0.9400, valid loss-0.3811, acc-0.8930, test loss-0.3857, acc-0.8903
Iter-77800, train loss-0.6618, acc-0.8400, valid loss-0.3811, acc-0.8930, test loss-0.3857, acc-0.8902
Iter-77810, train loss-0.5223, acc-0.8400, valid loss-0.3811, acc-0.8930, test loss-0.3857, acc-0.8902
Iter-77820, train loss-0.4445, acc-0.8800, valid loss-0.3811, acc-0.8930, test loss-0.3857, acc-0.8903
Iter-77830, train loss-0.2695, acc-0.9400, valid loss-0.3811, acc-0.8930, test loss-0.3857, acc-0.8902
Iter-77840, train loss-0.3341, acc-0.9200, valid loss-0.3811, acc-0.8930, test loss-0.3857, acc-0.8903
Iter-77850, train loss-0.3106, acc-0.9200, valid loss-0.3810, acc-0.8930, test loss-0.3857, acc-0.8904
Iter-77860, train loss-0.2007, acc-0.9800, valid loss-0.3810, acc-0.8930, test loss-0.3856, acc-0.8904
Iter-77870, train loss-0.2875, acc-0.9200, valid loss-0.3810, acc-0.8930, test loss-0.3856, acc-0.8903
Iter-77880, train loss-0.4179, acc-0.9200, valid loss-0.3810, acc-0.8930, test loss-0.3856, acc-0.8903
Iter-77890, train loss-0.2457, acc-0.9400, valid loss-0.3810, acc-0.8930, test loss-0.3856, acc-0.8902
Iter-77900, train loss-0.5403, acc-0.8400, valid loss-0.3809, acc-0.8930, test loss-0.3856, acc-0.8902
Iter-77910, train loss-0.3676, acc-0.9000, valid loss-0.3809, acc-0.8930, test loss-0.3855, acc-0.8902
Iter-77920, train loss-0.3523, acc-0.8600, valid loss-0.3809, acc-0.8930, test loss-0.3855, acc-0.8901
Iter-77930, train loss-0.4484, acc-0.8600, valid loss-0.3809, acc-0.8930, test loss-0.3855, acc-0.8901
Iter-77940, train loss-0.4746, acc-0.8800, valid loss-0.3809, acc-0.8930, test loss-0.3855, acc-0.8901
Iter-77950, train loss-0.3682, acc-0.8600, valid loss-0.3809, acc-0.8930, test loss-0.3855, acc-0.8902
Iter-77960, train loss-0.4667, acc-0.8800, valid loss-0.3808, acc-0.8930, test loss-0.3854, acc-0.8901
Iter-77970, train loss-0.5452, acc-0.7400, valid loss-0.3808, acc-0.8930, test loss-0.3854, acc-0.8901
Iter-77980, train loss-0.5467, acc-0.8000, valid loss-0.3808, acc-0.8930, test loss-0.3854, acc-0.8902
Iter-77990, train loss-0.3660, acc-0.8000, valid loss-0.3808, acc-0.8930, test loss-0.3854, acc-0.8903
Iter-78000, train loss-0.3336, acc-0.9000, valid loss-0.3808, acc-0.8930, test loss-0.3854, acc-0.8903
Iter-78010, train loss-0.5225, acc-0.9000, valid loss-0.3808, acc-0.8930, test loss-0.3854, acc-0.8903
Iter-78020, train loss-0.2753, acc-0.9400, valid loss-0.3808, acc-0.8930, test loss-0.3854, acc-0.8903
Iter-78030, train loss-0.3687, acc-0.8800, valid loss-0.3808, acc-0.8930, test loss-0.3853, acc-0.8903
Iter-78040, train loss-0.6556, acc-0.8400, valid loss-0.3807, acc-0.8928, test loss-0.3853, acc-0.8903
Iter-78050, train loss-0.2423, acc-0.9200, valid loss-0.3807, acc-0.8928, test loss-0.3853, acc-0.8902
Iter-78060, train loss-0.3803, acc-0.8600, valid loss-0.3807, acc-0.8928, test loss-0.3853, acc-0.8902
Iter-78070, train loss-0.3373, acc-0.9000, valid loss-0.3807, acc-0.8930, test loss-0.3853, acc-0.8902
Iter-78080, train loss-0.4587, acc-0.8200, valid loss-0.3807, acc-0.8930, test loss-0.3852, acc-0.8905
Iter-78090, train loss-0.2434, acc-0.9400, valid loss-0.3806, acc-0.8928, test loss-0.3852, acc-0.8904
Iter-78100, train loss-0.3101, acc-0.9400, valid loss-0.3806, acc-0.8928, test loss-0.3852, acc-0.8905
Iter-78110, train loss-0.4666, acc-0.8600, valid loss-0.3806, acc-0.8928, test loss-0.3852, acc-0.8904
Iter-78120, train loss-0.1860, acc-0.9800, valid loss-0.3806, acc-0.8930, test loss-0.3852, acc-0.8904
Iter-78130, train loss-0.8459, acc-0.7600, valid loss-0.3806, acc-0.8930, test loss-0.3852, acc-0.8903
Iter-78140, train loss-0.3369, acc-0.9200, valid loss-0.3806, acc-0.8930, test loss-0.3852, acc-0.8905
Iter-78150, train loss-0.2401, acc-0.9600, valid loss-0.3806, acc-0.8928, test loss-0.3851, acc-0.8903
Iter-78160, train loss-0.3354, acc-0.9000, valid loss-0.3806, acc-0.8930, test loss-0.3851, acc-0.8903
Iter-78170, train loss-0.3649, acc-0.9000, valid loss-0.3805, acc-0.8928, test loss-0.3851, acc-0.8904
Iter-78180, train loss-0.2592, acc-0.9200, valid loss-0.3805, acc-0.8928, test loss-0.3851, acc-0.8903
Iter-78190, train loss-0.3473, acc-0.8800, valid loss-0.3805, acc-0.8928, test loss-0.3851, acc-0.8904
Iter-78200, train loss-0.4022, acc-0.9000, valid loss-0.3805, acc-0.8928, test loss-0.3851, acc-0.8903
Iter-78210, train loss-0.3921, acc-0.9000, valid loss-0.3805, acc-0.8926, test loss-0.3850, acc-0.8903
Iter-78220, train loss-0.3772, acc-0.8800, valid loss-0.3805, acc-0.8928, test loss-0.3850, acc-0.8903
Iter-78230, train loss-0.3202, acc-0.9000, valid loss-0.3804, acc-0.8928, test loss-0.3850, acc-0.8903
Iter-78240, train loss-0.3983, acc-0.8800, valid loss-0.3804, acc-0.8928, test loss-0.3850, acc-0.8904
Iter-78250, train loss-0.4484, acc-0.8600, valid loss-0.3804, acc-0.8930, test loss-0.3850, acc-0.8904
Iter-78260, train loss-0.2563, acc-0.9200, valid loss-0.3804, acc-0.8930, test loss-0.3850, acc-0.8904
Iter-78270, train loss-0.5465, acc-0.8400, valid loss-0.3804, acc-0.8930, test loss-0.3849, acc-0.8904
Iter-78280, train loss-0.3728, acc-0.9000, valid loss-0.3803, acc-0.8928, test loss-0.3849, acc-0.8904
Iter-78290, train loss-0.3661, acc-0.8600, valid loss-0.3803, acc-0.8928, test loss-0.3849, acc-0.8905
Iter-78300, train loss-0.3080, acc-0.9200, valid loss-0.3803, acc-0.8928, test loss-0.3849, acc-0.8905
Iter-78310, train loss-0.3207, acc-0.8800, valid loss-0.3803, acc-0.8930, test loss-0.3848, acc-0.8904
Iter-78320, train loss-0.1718, acc-0.9800, valid loss-0.3803, acc-0.8930, test loss-0.3848, acc-0.8904
Iter-78330, train loss-0.3402, acc-0.8800, valid loss-0.3803, acc-0.8930, test loss-0.3848, acc-0.8905
Iter-78340, train loss-0.4300, acc-0.9000, valid loss-0.3802, acc-0.8930, test loss-0.3848, acc-0.8904
Iter-78350, train loss-0.3825, acc-0.9000, valid loss-0.3802, acc-0.8930, test loss-0.3848, acc-0.8904
Iter-78360, train loss-0.3068, acc-0.9200, valid loss-0.3802, acc-0.8930, test loss-0.3848, acc-0.8904
Iter-78370, train loss-0.4009, acc-0.8800, valid loss-0.3802, acc-0.8930, test loss-0.3847, acc-0.8904
Iter-78380, train loss-0.5050, acc-0.8200, valid loss-0.3802, acc-0.8928, test loss-0.3847, acc-0.8906
Iter-78390, train loss-0.6058, acc-0.8600, valid loss-0.3802, acc-0.8928, test loss-0.3847, acc-0.8906
Iter-78400, train loss-0.3661, acc-0.9200, valid loss-0.3801, acc-0.8928, test loss-0.3847, acc-0.8907
Iter-78410, train loss-0.4808, acc-0.8600, valid loss-0.3801, acc-0.8928, test loss-0.3847, acc-0.8906
Iter-78420, train loss-0.5089, acc-0.9000, valid loss-0.3801, acc-0.8928, test loss-0.3847, acc-0.8906
Iter-78430, train loss-0.5331, acc-0.8800, valid loss-0.3801, acc-0.8928, test loss-0.3846, acc-0.8907
Iter-78440, train loss-0.2675, acc-0.9400, valid loss-0.3801, acc-0.8930, test loss-0.3846, acc-0.8908
Iter-78450, train loss-0.5020, acc-0.8800, valid loss-0.3801, acc-0.8932, test loss-0.3846, acc-0.8909
Iter-78460, train loss-0.3919, acc-0.8800, valid loss-0.3800, acc-0.8930, test loss-0.3846, acc-0.8908
Iter-78470, train loss-0.3551, acc-0.9200, valid loss-0.3800, acc-0.8928, test loss-0.3846, acc-0.8909
Iter-78480, train loss-0.5553, acc-0.8400, valid loss-0.3800, acc-0.8928, test loss-0.3846, acc-0.8908
Iter-78490, train loss-0.3656, acc-0.8600, valid loss-0.3800, acc-0.8928, test loss-0.3845, acc-0.8909
Iter-78500, train loss-0.3286, acc-0.8800, valid loss-0.3799, acc-0.8928, test loss-0.3845, acc-0.8909
Iter-78510, train loss-0.3113, acc-0.9000, valid loss-0.3799, acc-0.8930, test loss-0.3845, acc-0.8909
Iter-78520, train loss-0.3956, acc-0.9000, valid loss-0.3799, acc-0.8930, test loss-0.3845, acc-0.8909
Iter-78530, train loss-0.3593, acc-0.9200, valid loss-0.3799, acc-0.8930, test loss-0.3845, acc-0.8908
Iter-78540, train loss-0.4496, acc-0.8600, valid loss-0.3799, acc-0.8930, test loss-0.3844, acc-0.8910
Iter-78550, train loss-0.4418, acc-0.8800, valid loss-0.3799, acc-0.8930, test loss-0.3844, acc-0.8910
Iter-78560, train loss-0.3929, acc-0.9000, valid loss-0.3799, acc-0.8930, test loss-0.3844, acc-0.8909
Iter-78570, train loss-0.4217, acc-0.8600, valid loss-0.3798, acc-0.8930, test loss-0.3844, acc-0.8908
Iter-78580, train loss-0.2327, acc-0.9400, valid loss-0.3798, acc-0.8928, test loss-0.3844, acc-0.8908
Iter-78590, train loss-0.3526, acc-0.8800, valid loss-0.3798, acc-0.8928, test loss-0.3843, acc-0.8908
Iter-78600, train loss-0.4872, acc-0.8800, valid loss-0.3798, acc-0.8928, test loss-0.3843, acc-0.8908
Iter-78610, train loss-0.2860, acc-0.9200, valid loss-0.3798, acc-0.8928, test loss-0.3843, acc-0.8909
Iter-78620, train loss-0.2381, acc-0.9400, valid loss-0.3798, acc-0.8932, test loss-0.3843, acc-0.8908
Iter-78630, train loss-0.4835, acc-0.8600, valid loss-0.3798, acc-0.8932, test loss-0.3843, acc-0.8907
Iter-78640, train loss-0.5833, acc-0.8000, valid loss-0.3797, acc-0.8932, test loss-0.3843, acc-0.8908
Iter-78650, train loss-0.4959, acc-0.8600, valid loss-0.3797, acc-0.8930, test loss-0.3842, acc-0.8908
Iter-78660, train loss-0.4604, acc-0.8400, valid loss-0.3797, acc-0.8928, test loss-0.3842, acc-0.8907
Iter-78670, train loss-0.6373, acc-0.8600, valid loss-0.3797, acc-0.8928, test loss-0.3842, acc-0.8908
Iter-78680, train loss-0.2445, acc-0.9400, valid loss-0.3796, acc-0.8932, test loss-0.3842, acc-0.8907
Iter-78690, train loss-0.4103, acc-0.8800, valid loss-0.3796, acc-0.8930, test loss-0.3842, acc-0.8907
Iter-78700, train loss-0.1699, acc-0.9600, valid loss-0.3796, acc-0.8930, test loss-0.3842, acc-0.8907
Iter-78710, train loss-0.3998, acc-0.8800, valid loss-0.3796, acc-0.8930, test loss-0.3842, acc-0.8907
Iter-78720, train loss-0.3833, acc-0.9400, valid loss-0.3796, acc-0.8930, test loss-0.3841, acc-0.8907
Iter-78730, train loss-0.2820, acc-0.9400, valid loss-0.3796, acc-0.8930, test loss-0.3841, acc-0.8907
Iter-78740, train loss-0.4500, acc-0.8600, valid loss-0.3795, acc-0.8930, test loss-0.3841, acc-0.8908
Iter-78750, train loss-0.3364, acc-0.9200, valid loss-0.3795, acc-0.8930, test loss-0.3841, acc-0.8907
Iter-78760, train loss-0.3201, acc-0.9000, valid loss-0.3795, acc-0.8930, test loss-0.3841, acc-0.8906
Iter-78770, train loss-0.4157, acc-0.9000, valid loss-0.3795, acc-0.8930, test loss-0.3840, acc-0.8906
Iter-78780, train loss-0.3328, acc-0.9200, valid loss-0.3795, acc-0.8930, test loss-0.3840, acc-0.8907
Iter-78790, train loss-0.2639, acc-0.9400, valid loss-0.3795, acc-0.8930, test loss-0.3840, acc-0.8907
Iter-78800, train loss-0.2685, acc-0.9400, valid loss-0.3795, acc-0.8932, test loss-0.3840, acc-0.8907
Iter-78810, train loss-0.3286, acc-0.9000, valid loss-0.3794, acc-0.8932, test loss-0.3840, acc-0.8907
Iter-78820, train loss-0.5312, acc-0.8800, valid loss-0.3794, acc-0.8932, test loss-0.3839, acc-0.8909
Iter-78830, train loss-0.4723, acc-0.8800, valid loss-0.3794, acc-0.8932, test loss-0.3839, acc-0.8908
Iter-78840, train loss-0.3964, acc-0.9200, valid loss-0.3794, acc-0.8932, test loss-0.3839, acc-0.8907
Iter-78850, train loss-0.3495, acc-0.9000, valid loss-0.3794, acc-0.8932, test loss-0.3839, acc-0.8905
Iter-78860, train loss-0.3521, acc-0.9000, valid loss-0.3793, acc-0.8930, test loss-0.3839, acc-0.8909
Iter-78870, train loss-0.2188, acc-0.9600, valid loss-0.3793, acc-0.8932, test loss-0.3838, acc-0.8908
Iter-78880, train loss-0.4260, acc-0.8800, valid loss-0.3793, acc-0.8932, test loss-0.3838, acc-0.8909
Iter-78890, train loss-0.4718, acc-0.8800, valid loss-0.3793, acc-0.8930, test loss-0.3838, acc-0.8908
Iter-78900, train loss-0.5057, acc-0.8800, valid loss-0.3793, acc-0.8930, test loss-0.3838, acc-0.8909
Iter-78910, train loss-0.5406, acc-0.8600, valid loss-0.3792, acc-0.8932, test loss-0.3838, acc-0.8909
Iter-78920, train loss-0.4199, acc-0.8600, valid loss-0.3792, acc-0.8930, test loss-0.3838, acc-0.8909
Iter-78930, train loss-0.3891, acc-0.8800, valid loss-0.3792, acc-0.8932, test loss-0.3837, acc-0.8909
Iter-78940, train loss-0.4272, acc-0.8600, valid loss-0.3792, acc-0.8932, test loss-0.3837, acc-0.8910
Iter-78950, train loss-0.3626, acc-0.9000, valid loss-0.3792, acc-0.8930, test loss-0.3837, acc-0.8910
Iter-78960, train loss-0.2428, acc-0.9400, valid loss-0.3792, acc-0.8930, test loss-0.3837, acc-0.8911
Iter-78970, train loss-0.3580, acc-0.9000, valid loss-0.3791, acc-0.8930, test loss-0.3837, acc-0.8910
Iter-78980, train loss-0.4366, acc-0.8400, valid loss-0.3791, acc-0.8930, test loss-0.3837, acc-0.8909
Iter-78990, train loss-0.5875, acc-0.8600, valid loss-0.3791, acc-0.8930, test loss-0.3836, acc-0.8911
Iter-79000, train loss-0.2311, acc-0.9600, valid loss-0.3791, acc-0.8930, test loss-0.3836, acc-0.8912
Iter-79010, train loss-0.4476, acc-0.9000, valid loss-0.3791, acc-0.8930, test loss-0.3836, acc-0.8911
Iter-79020, train loss-0.6828, acc-0.7600, valid loss-0.3791, acc-0.8930, test loss-0.3836, acc-0.8912
Iter-79030, train loss-0.2212, acc-0.9600, valid loss-0.3790, acc-0.8930, test loss-0.3836, acc-0.8911
Iter-79040, train loss-0.2299, acc-0.9200, valid loss-0.3790, acc-0.8930, test loss-0.3836, acc-0.8911
Iter-79050, train loss-0.4589, acc-0.8600, valid loss-0.3790, acc-0.8932, test loss-0.3836, acc-0.8910
Iter-79060, train loss-0.3137, acc-0.9200, valid loss-0.3790, acc-0.8930, test loss-0.3835, acc-0.8912
Iter-79070, train loss-0.3294, acc-0.9200, valid loss-0.3790, acc-0.8932, test loss-0.3835, acc-0.8911
Iter-79080, train loss-0.4236, acc-0.9000, valid loss-0.3790, acc-0.8932, test loss-0.3835, acc-0.8911
Iter-79090, train loss-0.4735, acc-0.8600, valid loss-0.3790, acc-0.8932, test loss-0.3835, acc-0.8911
Iter-79100, train loss-0.3573, acc-0.8600, valid loss-0.3789, acc-0.8932, test loss-0.3835, acc-0.8910
Iter-79110, train loss-0.5101, acc-0.8200, valid loss-0.3789, acc-0.8932, test loss-0.3835, acc-0.8911
Iter-79120, train loss-0.2231, acc-0.9200, valid loss-0.3789, acc-0.8932, test loss-0.3834, acc-0.8911
Iter-79130, train loss-0.4367, acc-0.9200, valid loss-0.3789, acc-0.8932, test loss-0.3834, acc-0.8911
Iter-79140, train loss-0.5012, acc-0.8400, valid loss-0.3788, acc-0.8930, test loss-0.3834, acc-0.8911
Iter-79150, train loss-0.4348, acc-0.8400, valid loss-0.3788, acc-0.8932, test loss-0.3834, acc-0.8911
Iter-79160, train loss-0.4584, acc-0.9000, valid loss-0.3788, acc-0.8932, test loss-0.3834, acc-0.8911
Iter-79170, train loss-0.3307, acc-0.9200, valid loss-0.3788, acc-0.8932, test loss-0.3834, acc-0.8911
Iter-79180, train loss-0.2990, acc-0.9200, valid loss-0.3787, acc-0.8932, test loss-0.3833, acc-0.8912
Iter-79190, train loss-0.5381, acc-0.8400, valid loss-0.3787, acc-0.8934, test loss-0.3833, acc-0.8911
Iter-79200, train loss-0.3242, acc-0.8800, valid loss-0.3787, acc-0.8932, test loss-0.3833, acc-0.8911
Iter-79210, train loss-0.3882, acc-0.9200, valid loss-0.3787, acc-0.8934, test loss-0.3833, acc-0.8911
Iter-79220, train loss-0.3311, acc-0.8600, valid loss-0.3787, acc-0.8938, test loss-0.3833, acc-0.8911
Iter-79230, train loss-0.4410, acc-0.9000, valid loss-0.3787, acc-0.8936, test loss-0.3832, acc-0.8911
Iter-79240, train loss-0.2844, acc-0.9000, valid loss-0.3787, acc-0.8936, test loss-0.3832, acc-0.8911
Iter-79250, train loss-0.5085, acc-0.8600, valid loss-0.3787, acc-0.8936, test loss-0.3832, acc-0.8911
Iter-79260, train loss-0.4427, acc-0.9000, valid loss-0.3786, acc-0.8936, test loss-0.3832, acc-0.8912
Iter-79270, train loss-0.4004, acc-0.8800, valid loss-0.3786, acc-0.8936, test loss-0.3832, acc-0.8912
Iter-79280, train loss-0.3673, acc-0.9000, valid loss-0.3786, acc-0.8934, test loss-0.3831, acc-0.8912
Iter-79290, train loss-0.4615, acc-0.8600, valid loss-0.3786, acc-0.8936, test loss-0.3831, acc-0.8912
Iter-79300, train loss-0.4790, acc-0.9200, valid loss-0.3785, acc-0.8936, test loss-0.3831, acc-0.8912
Iter-79310, train loss-0.5080, acc-0.8000, valid loss-0.3785, acc-0.8936, test loss-0.3831, acc-0.8913
Iter-79320, train loss-0.6371, acc-0.8200, valid loss-0.3785, acc-0.8936, test loss-0.3831, acc-0.8913
Iter-79330, train loss-0.3286, acc-0.9000, valid loss-0.3785, acc-0.8936, test loss-0.3831, acc-0.8913
Iter-79340, train loss-0.3131, acc-0.9400, valid loss-0.3785, acc-0.8936, test loss-0.3830, acc-0.8913
Iter-79350, train loss-0.4232, acc-0.8600, valid loss-0.3785, acc-0.8934, test loss-0.3830, acc-0.8913
Iter-79360, train loss-0.4792, acc-0.8600, valid loss-0.3784, acc-0.8934, test loss-0.3830, acc-0.8912
Iter-79370, train loss-0.6708, acc-0.8000, valid loss-0.3784, acc-0.8938, test loss-0.3830, acc-0.8912
Iter-79380, train loss-0.3837, acc-0.9000, valid loss-0.3784, acc-0.8938, test loss-0.3830, acc-0.8912
Iter-79390, train loss-0.3401, acc-0.9000, valid loss-0.3784, acc-0.8938, test loss-0.3830, acc-0.8912
Iter-79400, train loss-0.4723, acc-0.9000, valid loss-0.3784, acc-0.8938, test loss-0.3830, acc-0.8912
Iter-79410, train loss-0.4963, acc-0.8200, valid loss-0.3783, acc-0.8938, test loss-0.3829, acc-0.8912
Iter-79420, train loss-0.4776, acc-0.8600, valid loss-0.3783, acc-0.8938, test loss-0.3829, acc-0.8912
Iter-79430, train loss-0.4729, acc-0.9200, valid loss-0.3783, acc-0.8938, test loss-0.3829, acc-0.8911
Iter-79440, train loss-0.4203, acc-0.9000, valid loss-0.3783, acc-0.8938, test loss-0.3829, acc-0.8910
Iter-79450, train loss-0.5552, acc-0.8600, valid loss-0.3783, acc-0.8938, test loss-0.3829, acc-0.8910
Iter-79460, train loss-0.5720, acc-0.8600, valid loss-0.3783, acc-0.8938, test loss-0.3829, acc-0.8910
Iter-79470, train loss-0.4800, acc-0.8600, valid loss-0.3783, acc-0.8938, test loss-0.3829, acc-0.8910
Iter-79480, train loss-0.3685, acc-0.8200, valid loss-0.3782, acc-0.8936, test loss-0.3828, acc-0.8910
Iter-79490, train loss-0.4866, acc-0.8600, valid loss-0.3782, acc-0.8936, test loss-0.3828, acc-0.8910
Iter-79500, train loss-0.4475, acc-0.8400, valid loss-0.3782, acc-0.8938, test loss-0.3828, acc-0.8910
Iter-79510, train loss-0.2537, acc-0.9600, valid loss-0.3782, acc-0.8934, test loss-0.3828, acc-0.8910
Iter-79520, train loss-0.3606, acc-0.9000, valid loss-0.3781, acc-0.8934, test loss-0.3828, acc-0.8910
Iter-79530, train loss-0.4014, acc-0.8400, valid loss-0.3781, acc-0.8934, test loss-0.3828, acc-0.8910
Iter-79540, train loss-0.3870, acc-0.8800, valid loss-0.3781, acc-0.8934, test loss-0.3827, acc-0.8911
Iter-79550, train loss-0.3839, acc-0.9400, valid loss-0.3781, acc-0.8936, test loss-0.3827, acc-0.8911
Iter-79560, train loss-0.5051, acc-0.8400, valid loss-0.3781, acc-0.8938, test loss-0.3827, acc-0.8912
Iter-79570, train loss-0.5944, acc-0.8400, valid loss-0.3781, acc-0.8938, test loss-0.3827, acc-0.8912
Iter-79580, train loss-0.4474, acc-0.8800, valid loss-0.3781, acc-0.8938, test loss-0.3827, acc-0.8911
Iter-79590, train loss-0.3292, acc-0.9000, valid loss-0.3781, acc-0.8936, test loss-0.3827, acc-0.8911
Iter-79600, train loss-0.3702, acc-0.9000, valid loss-0.3780, acc-0.8934, test loss-0.3827, acc-0.8911
Iter-79610, train loss-0.2462, acc-0.9200, valid loss-0.3780, acc-0.8938, test loss-0.3826, acc-0.8911
Iter-79620, train loss-0.3523, acc-0.8800, valid loss-0.3780, acc-0.8936, test loss-0.3826, acc-0.8911
Iter-79630, train loss-0.3143, acc-0.9200, valid loss-0.3780, acc-0.8936, test loss-0.3826, acc-0.8912
Iter-79640, train loss-0.2990, acc-0.9200, valid loss-0.3780, acc-0.8934, test loss-0.3826, acc-0.8911
Iter-79650, train loss-0.2362, acc-0.9600, valid loss-0.3779, acc-0.8934, test loss-0.3826, acc-0.8913
Iter-79660, train loss-0.4109, acc-0.8600, valid loss-0.3779, acc-0.8934, test loss-0.3826, acc-0.8914
Iter-79670, train loss-0.3872, acc-0.8600, valid loss-0.3779, acc-0.8934, test loss-0.3825, acc-0.8911
Iter-79680, train loss-0.6013, acc-0.8000, valid loss-0.3779, acc-0.8934, test loss-0.3825, acc-0.8912
Iter-79690, train loss-0.4066, acc-0.9000, valid loss-0.3778, acc-0.8934, test loss-0.3825, acc-0.8913
Iter-79700, train loss-0.3840, acc-0.8800, valid loss-0.3778, acc-0.8934, test loss-0.3825, acc-0.8913
Iter-79710, train loss-0.6395, acc-0.8200, valid loss-0.3778, acc-0.8934, test loss-0.3825, acc-0.8914
Iter-79720, train loss-0.3183, acc-0.9400, valid loss-0.3778, acc-0.8934, test loss-0.3825, acc-0.8914
Iter-79730, train loss-0.3790, acc-0.9000, valid loss-0.3778, acc-0.8934, test loss-0.3825, acc-0.8913
Iter-79740, train loss-0.5733, acc-0.8400, valid loss-0.3778, acc-0.8934, test loss-0.3824, acc-0.8914
Iter-79750, train loss-0.3932, acc-0.9000, valid loss-0.3777, acc-0.8934, test loss-0.3824, acc-0.8914
Iter-79760, train loss-0.3185, acc-0.9000, valid loss-0.3777, acc-0.8934, test loss-0.3824, acc-0.8913
Iter-79770, train loss-0.4232, acc-0.8400, valid loss-0.3777, acc-0.8932, test loss-0.3824, acc-0.8914
Iter-79780, train loss-0.3857, acc-0.8800, valid loss-0.3777, acc-0.8932, test loss-0.3824, acc-0.8914
Iter-79790, train loss-0.3321, acc-0.9200, valid loss-0.3777, acc-0.8932, test loss-0.3824, acc-0.8913
Iter-79800, train loss-0.4625, acc-0.8400, valid loss-0.3777, acc-0.8932, test loss-0.3823, acc-0.8913
Iter-79810, train loss-0.3268, acc-0.8800, valid loss-0.3776, acc-0.8932, test loss-0.3823, acc-0.8913
Iter-79820, train loss-0.5565, acc-0.8800, valid loss-0.3776, acc-0.8932, test loss-0.3823, acc-0.8913
Iter-79830, train loss-0.3864, acc-0.9000, valid loss-0.3776, acc-0.8932, test loss-0.3823, acc-0.8913
Iter-79840, train loss-0.4234, acc-0.8000, valid loss-0.3776, acc-0.8932, test loss-0.3823, acc-0.8914
Iter-79850, train loss-0.2803, acc-0.9800, valid loss-0.3776, acc-0.8932, test loss-0.3822, acc-0.8914
Iter-79860, train loss-0.5983, acc-0.8600, valid loss-0.3775, acc-0.8932, test loss-0.3822, acc-0.8914
Iter-79870, train loss-0.1289, acc-0.9800, valid loss-0.3775, acc-0.8932, test loss-0.3822, acc-0.8914
Iter-79880, train loss-0.6349, acc-0.8200, valid loss-0.3775, acc-0.8932, test loss-0.3822, acc-0.8914
Iter-79890, train loss-0.2367, acc-0.9600, valid loss-0.3775, acc-0.8932, test loss-0.3822, acc-0.8914
Iter-79900, train loss-0.4247, acc-0.9000, valid loss-0.3775, acc-0.8932, test loss-0.3822, acc-0.8914
Iter-79910, train loss-0.3498, acc-0.9000, valid loss-0.3775, acc-0.8932, test loss-0.3822, acc-0.8914
Iter-79920, train loss-0.3540, acc-0.9000, valid loss-0.3775, acc-0.8932, test loss-0.3821, acc-0.8914
Iter-79930, train loss-0.4280, acc-0.9000, valid loss-0.3774, acc-0.8932, test loss-0.3821, acc-0.8914
Iter-79940, train loss-0.3436, acc-0.9200, valid loss-0.3774, acc-0.8932, test loss-0.3821, acc-0.8914
Iter-79950, train loss-0.3843, acc-0.8600, valid loss-0.3774, acc-0.8932, test loss-0.3821, acc-0.8914
Iter-79960, train loss-0.2952, acc-0.9000, valid loss-0.3774, acc-0.8934, test loss-0.3821, acc-0.8914
Iter-79970, train loss-0.3773, acc-0.8800, valid loss-0.3774, acc-0.8934, test loss-0.3820, acc-0.8913
Iter-79980, train loss-0.4194, acc-0.8600, valid loss-0.3773, acc-0.8934, test loss-0.3820, acc-0.8914
Iter-79990, train loss-0.2602, acc-0.9400, valid loss-0.3773, acc-0.8934, test loss-0.3820, acc-0.8914
Iter-80000, train loss-0.3531, acc-0.9000, valid loss-0.3773, acc-0.8934, test loss-0.3820, acc-0.8913
Iter-80010, train loss-0.3624, acc-0.9400, valid loss-0.3773, acc-0.8934, test loss-0.3820, acc-0.8914
Iter-80020, train loss-0.2325, acc-0.9600, valid loss-0.3773, acc-0.8934, test loss-0.3820, acc-0.8914
Iter-80030, train loss-0.4362, acc-0.8800, valid loss-0.3773, acc-0.8934, test loss-0.3819, acc-0.8914
Iter-80040, train loss-0.3690, acc-0.8800, valid loss-0.3772, acc-0.8932, test loss-0.3819, acc-0.8915
Iter-80050, train loss-0.4097, acc-0.8800, valid loss-0.3772, acc-0.8932, test loss-0.3819, acc-0.8915
Iter-80060, train loss-0.2020, acc-0.9400, valid loss-0.3772, acc-0.8934, test loss-0.3819, acc-0.8915
Iter-80070, train loss-0.5147, acc-0.9000, valid loss-0.3772, acc-0.8934, test loss-0.3819, acc-0.8916
Iter-80080, train loss-0.5790, acc-0.8400, valid loss-0.3772, acc-0.8932, test loss-0.3819, acc-0.8915
Iter-80090, train loss-0.3847, acc-0.9000, valid loss-0.3772, acc-0.8934, test loss-0.3818, acc-0.8916
Iter-80100, train loss-0.2665, acc-0.9400, valid loss-0.3771, acc-0.8934, test loss-0.3818, acc-0.8915
Iter-80110, train loss-0.3067, acc-0.9200, valid loss-0.3771, acc-0.8932, test loss-0.3818, acc-0.8915
Iter-80120, train loss-0.4435, acc-0.8400, valid loss-0.3771, acc-0.8932, test loss-0.3818, acc-0.8915
Iter-80130, train loss-0.4220, acc-0.9200, valid loss-0.3771, acc-0.8932, test loss-0.3818, acc-0.8915
Iter-80140, train loss-0.2637, acc-0.9400, valid loss-0.3771, acc-0.8932, test loss-0.3818, acc-0.8915
Iter-80150, train loss-0.2832, acc-0.9400, valid loss-0.3771, acc-0.8932, test loss-0.3818, acc-0.8916
Iter-80160, train loss-0.3145, acc-0.9400, valid loss-0.3771, acc-0.8932, test loss-0.3817, acc-0.8916
Iter-80170, train loss-0.3922, acc-0.9000, valid loss-0.3770, acc-0.8932, test loss-0.3817, acc-0.8915
Iter-80180, train loss-0.3091, acc-0.8800, valid loss-0.3770, acc-0.8932, test loss-0.3817, acc-0.8915
Iter-80190, train loss-0.3581, acc-0.8600, valid loss-0.3770, acc-0.8932, test loss-0.3817, acc-0.8915
Iter-80200, train loss-0.3357, acc-0.9000, valid loss-0.3770, acc-0.8932, test loss-0.3817, acc-0.8916
Iter-80210, train loss-0.3545, acc-0.9200, valid loss-0.3770, acc-0.8932, test loss-0.3817, acc-0.8916
Iter-80220, train loss-0.5532, acc-0.8200, valid loss-0.3770, acc-0.8934, test loss-0.3817, acc-0.8916
Iter-80230, train loss-0.3856, acc-0.8800, valid loss-0.3769, acc-0.8934, test loss-0.3816, acc-0.8916
Iter-80240, train loss-0.5068, acc-0.8800, valid loss-0.3769, acc-0.8934, test loss-0.3816, acc-0.8915
Iter-80250, train loss-0.3929, acc-0.8600, valid loss-0.3769, acc-0.8934, test loss-0.3816, acc-0.8915
Iter-80260, train loss-0.5699, acc-0.8400, valid loss-0.3769, acc-0.8934, test loss-0.3816, acc-0.8916
Iter-80270, train loss-0.5285, acc-0.9000, valid loss-0.3769, acc-0.8934, test loss-0.3816, acc-0.8916
Iter-80280, train loss-0.1808, acc-0.9600, valid loss-0.3768, acc-0.8932, test loss-0.3816, acc-0.8916
Iter-80290, train loss-0.4414, acc-0.8200, valid loss-0.3768, acc-0.8932, test loss-0.3815, acc-0.8916
Iter-80300, train loss-0.3912, acc-0.8800, valid loss-0.3768, acc-0.8932, test loss-0.3815, acc-0.8917
Iter-80310, train loss-0.7363, acc-0.7600, valid loss-0.3768, acc-0.8932, test loss-0.3815, acc-0.8917
Iter-80320, train loss-0.4173, acc-0.8800, valid loss-0.3768, acc-0.8932, test loss-0.3815, acc-0.8915
Iter-80330, train loss-0.5732, acc-0.8200, valid loss-0.3767, acc-0.8932, test loss-0.3815, acc-0.8915
Iter-80340, train loss-0.2378, acc-0.9600, valid loss-0.3767, acc-0.8932, test loss-0.3815, acc-0.8916
Iter-80350, train loss-0.5933, acc-0.8600, valid loss-0.3767, acc-0.8934, test loss-0.3814, acc-0.8916
Iter-80360, train loss-0.4586, acc-0.8200, valid loss-0.3767, acc-0.8936, test loss-0.3814, acc-0.8917
Iter-80370, train loss-0.6336, acc-0.8200, valid loss-0.3767, acc-0.8936, test loss-0.3814, acc-0.8916
Iter-80380, train loss-0.3343, acc-0.9000, valid loss-0.3767, acc-0.8936, test loss-0.3814, acc-0.8916
Iter-80390, train loss-0.2275, acc-0.9600, valid loss-0.3766, acc-0.8936, test loss-0.3814, acc-0.8916
Iter-80400, train loss-0.2367, acc-0.9000, valid loss-0.3766, acc-0.8936, test loss-0.3814, acc-0.8917
Iter-80410, train loss-0.3991, acc-0.8800, valid loss-0.3766, acc-0.8934, test loss-0.3813, acc-0.8915
Iter-80420, train loss-0.2607, acc-0.9000, valid loss-0.3766, acc-0.8934, test loss-0.3813, acc-0.8915
Iter-80430, train loss-0.3243, acc-0.9000, valid loss-0.3766, acc-0.8934, test loss-0.3813, acc-0.8916
Iter-80440, train loss-0.5476, acc-0.8400, valid loss-0.3766, acc-0.8934, test loss-0.3813, acc-0.8917
Iter-80450, train loss-0.3061, acc-0.9200, valid loss-0.3765, acc-0.8934, test loss-0.3813, acc-0.8916
Iter-80460, train loss-0.4481, acc-0.8600, valid loss-0.3765, acc-0.8934, test loss-0.3813, acc-0.8917
Iter-80470, train loss-0.3184, acc-0.9200, valid loss-0.3765, acc-0.8934, test loss-0.3813, acc-0.8918
Iter-80480, train loss-0.3368, acc-0.9200, valid loss-0.3765, acc-0.8934, test loss-0.3813, acc-0.8918
Iter-80490, train loss-0.2801, acc-0.9400, valid loss-0.3764, acc-0.8934, test loss-0.3812, acc-0.8917
Iter-80500, train loss-0.3178, acc-0.9200, valid loss-0.3764, acc-0.8934, test loss-0.3812, acc-0.8917
Iter-80510, train loss-0.4731, acc-0.8400, valid loss-0.3764, acc-0.8934, test loss-0.3812, acc-0.8919
Iter-80520, train loss-0.4316, acc-0.8600, valid loss-0.3764, acc-0.8934, test loss-0.3812, acc-0.8919
Iter-80530, train loss-0.4420, acc-0.8200, valid loss-0.3764, acc-0.8934, test loss-0.3812, acc-0.8919
Iter-80540, train loss-0.2288, acc-0.9600, valid loss-0.3764, acc-0.8934, test loss-0.3812, acc-0.8919
Iter-80550, train loss-0.4004, acc-0.9200, valid loss-0.3764, acc-0.8934, test loss-0.3812, acc-0.8919
Iter-80560, train loss-0.6407, acc-0.7800, valid loss-0.3763, acc-0.8934, test loss-0.3811, acc-0.8919
Iter-80570, train loss-0.3511, acc-0.8800, valid loss-0.3763, acc-0.8934, test loss-0.3811, acc-0.8918
Iter-80580, train loss-0.2173, acc-0.9600, valid loss-0.3763, acc-0.8934, test loss-0.3811, acc-0.8918
Iter-80590, train loss-0.3975, acc-0.9000, valid loss-0.3763, acc-0.8934, test loss-0.3811, acc-0.8919
Iter-80600, train loss-0.3557, acc-0.9000, valid loss-0.3763, acc-0.8934, test loss-0.3811, acc-0.8920
Iter-80610, train loss-0.3684, acc-0.9000, valid loss-0.3763, acc-0.8934, test loss-0.3811, acc-0.8920
Iter-80620, train loss-0.4619, acc-0.8600, valid loss-0.3762, acc-0.8934, test loss-0.3811, acc-0.8919
Iter-80630, train loss-0.5813, acc-0.8200, valid loss-0.3762, acc-0.8934, test loss-0.3810, acc-0.8920
Iter-80640, train loss-0.3310, acc-0.8400, valid loss-0.3762, acc-0.8934, test loss-0.3810, acc-0.8920
Iter-80650, train loss-0.3048, acc-0.9600, valid loss-0.3762, acc-0.8934, test loss-0.3810, acc-0.8921
Iter-80660, train loss-0.3219, acc-0.9000, valid loss-0.3762, acc-0.8934, test loss-0.3810, acc-0.8920
Iter-80670, train loss-0.2764, acc-0.9400, valid loss-0.3762, acc-0.8934, test loss-0.3810, acc-0.8920
Iter-80680, train loss-0.4954, acc-0.8600, valid loss-0.3762, acc-0.8932, test loss-0.3810, acc-0.8920
Iter-80690, train loss-0.4628, acc-0.9000, valid loss-0.3761, acc-0.8934, test loss-0.3810, acc-0.8920
Iter-80700, train loss-0.3432, acc-0.9000, valid loss-0.3761, acc-0.8934, test loss-0.3810, acc-0.8920
Iter-80710, train loss-0.4799, acc-0.8400, valid loss-0.3761, acc-0.8934, test loss-0.3809, acc-0.8920
Iter-80720, train loss-0.4863, acc-0.8800, valid loss-0.3761, acc-0.8936, test loss-0.3809, acc-0.8920
Iter-80730, train loss-0.3082, acc-0.9400, valid loss-0.3761, acc-0.8934, test loss-0.3809, acc-0.8920
Iter-80740, train loss-0.2779, acc-0.9200, valid loss-0.3761, acc-0.8932, test loss-0.3809, acc-0.8920
Iter-80750, train loss-0.4892, acc-0.8800, valid loss-0.3761, acc-0.8934, test loss-0.3809, acc-0.8920
Iter-80760, train loss-0.2378, acc-0.9800, valid loss-0.3760, acc-0.8932, test loss-0.3809, acc-0.8921
Iter-80770, train loss-0.5942, acc-0.8600, valid loss-0.3760, acc-0.8932, test loss-0.3809, acc-0.8920
Iter-80780, train loss-0.2336, acc-0.9400, valid loss-0.3760, acc-0.8932, test loss-0.3809, acc-0.8921
Iter-80790, train loss-0.4580, acc-0.8600, valid loss-0.3760, acc-0.8932, test loss-0.3808, acc-0.8921
Iter-80800, train loss-0.5393, acc-0.8400, valid loss-0.3760, acc-0.8932, test loss-0.3808, acc-0.8921
Iter-80810, train loss-0.6102, acc-0.8200, valid loss-0.3759, acc-0.8932, test loss-0.3808, acc-0.8921
Iter-80820, train loss-0.3746, acc-0.8600, valid loss-0.3759, acc-0.8932, test loss-0.3808, acc-0.8921
Iter-80830, train loss-0.3976, acc-0.8800, valid loss-0.3759, acc-0.8932, test loss-0.3808, acc-0.8921
Iter-80840, train loss-0.5032, acc-0.8600, valid loss-0.3759, acc-0.8932, test loss-0.3807, acc-0.8922
Iter-80850, train loss-0.6279, acc-0.8200, valid loss-0.3759, acc-0.8932, test loss-0.3807, acc-0.8922
Iter-80860, train loss-0.4547, acc-0.8800, valid loss-0.3759, acc-0.8932, test loss-0.3807, acc-0.8922
Iter-80870, train loss-0.3972, acc-0.8600, valid loss-0.3759, acc-0.8932, test loss-0.3807, acc-0.8921
Iter-80880, train loss-0.2985, acc-0.9600, valid loss-0.3758, acc-0.8932, test loss-0.3807, acc-0.8922
Iter-80890, train loss-0.4740, acc-0.8400, valid loss-0.3758, acc-0.8932, test loss-0.3807, acc-0.8921
Iter-80900, train loss-0.5128, acc-0.8400, valid loss-0.3758, acc-0.8932, test loss-0.3806, acc-0.8921
Iter-80910, train loss-0.2312, acc-0.9200, valid loss-0.3758, acc-0.8934, test loss-0.3806, acc-0.8921
Iter-80920, train loss-0.3713, acc-0.9200, valid loss-0.3758, acc-0.8936, test loss-0.3806, acc-0.8921
Iter-80930, train loss-0.2497, acc-0.9400, valid loss-0.3757, acc-0.8936, test loss-0.3806, acc-0.8921
Iter-80940, train loss-0.2625, acc-0.9400, valid loss-0.3757, acc-0.8934, test loss-0.3806, acc-0.8921
Iter-80950, train loss-0.3117, acc-0.9600, valid loss-0.3757, acc-0.8934, test loss-0.3806, acc-0.8920
Iter-80960, train loss-0.5258, acc-0.8400, valid loss-0.3757, acc-0.8934, test loss-0.3806, acc-0.8920
Iter-80970, train loss-0.4725, acc-0.8200, valid loss-0.3757, acc-0.8938, test loss-0.3805, acc-0.8922
Iter-80980, train loss-0.4973, acc-0.8600, valid loss-0.3756, acc-0.8940, test loss-0.3805, acc-0.8922
Iter-80990, train loss-0.2794, acc-0.9000, valid loss-0.3756, acc-0.8936, test loss-0.3805, acc-0.8922
Iter-81000, train loss-0.3246, acc-0.9200, valid loss-0.3756, acc-0.8940, test loss-0.3805, acc-0.8921
Iter-81010, train loss-0.4313, acc-0.9000, valid loss-0.3756, acc-0.8940, test loss-0.3805, acc-0.8921
Iter-81020, train loss-0.2147, acc-0.9400, valid loss-0.3756, acc-0.8938, test loss-0.3804, acc-0.8921
Iter-81030, train loss-0.1828, acc-0.9800, valid loss-0.3756, acc-0.8936, test loss-0.3804, acc-0.8922
Iter-81040, train loss-0.2104, acc-0.9600, valid loss-0.3756, acc-0.8938, test loss-0.3804, acc-0.8920
Iter-81050, train loss-0.3282, acc-0.9000, valid loss-0.3756, acc-0.8938, test loss-0.3804, acc-0.8920
Iter-81060, train loss-0.4036, acc-0.9000, valid loss-0.3755, acc-0.8940, test loss-0.3804, acc-0.8921
Iter-81070, train loss-0.4084, acc-0.9000, valid loss-0.3755, acc-0.8938, test loss-0.3804, acc-0.8920
Iter-81080, train loss-0.2688, acc-0.9400, valid loss-0.3755, acc-0.8938, test loss-0.3804, acc-0.8920
Iter-81090, train loss-0.5202, acc-0.8800, valid loss-0.3755, acc-0.8940, test loss-0.3803, acc-0.8921
Iter-81100, train loss-0.4232, acc-0.9200, valid loss-0.3755, acc-0.8940, test loss-0.3803, acc-0.8921
Iter-81110, train loss-0.5081, acc-0.8400, valid loss-0.3755, acc-0.8940, test loss-0.3803, acc-0.8920
Iter-81120, train loss-0.3568, acc-0.8800, valid loss-0.3755, acc-0.8940, test loss-0.3803, acc-0.8921
Iter-81130, train loss-0.5134, acc-0.8400, valid loss-0.3754, acc-0.8942, test loss-0.3803, acc-0.8921
Iter-81140, train loss-0.4656, acc-0.8400, valid loss-0.3754, acc-0.8940, test loss-0.3803, acc-0.8922
Iter-81150, train loss-0.3735, acc-0.9000, valid loss-0.3754, acc-0.8940, test loss-0.3803, acc-0.8922
Iter-81160, train loss-0.3567, acc-0.8600, valid loss-0.3754, acc-0.8940, test loss-0.3803, acc-0.8923
Iter-81170, train loss-0.3975, acc-0.8600, valid loss-0.3754, acc-0.8940, test loss-0.3802, acc-0.8923
Iter-81180, train loss-0.1893, acc-0.9400, valid loss-0.3754, acc-0.8940, test loss-0.3802, acc-0.8924
Iter-81190, train loss-0.4569, acc-0.9000, valid loss-0.3753, acc-0.8940, test loss-0.3802, acc-0.8923
Iter-81200, train loss-0.3348, acc-0.8800, valid loss-0.3753, acc-0.8938, test loss-0.3802, acc-0.8923
Iter-81210, train loss-0.3143, acc-0.8600, valid loss-0.3753, acc-0.8938, test loss-0.3802, acc-0.8923
Iter-81220, train loss-0.2417, acc-0.9600, valid loss-0.3753, acc-0.8938, test loss-0.3801, acc-0.8923
Iter-81230, train loss-0.3887, acc-0.8800, valid loss-0.3753, acc-0.8938, test loss-0.3801, acc-0.8924
Iter-81240, train loss-0.4901, acc-0.8200, valid loss-0.3752, acc-0.8938, test loss-0.3801, acc-0.8923
Iter-81250, train loss-0.1975, acc-0.9800, valid loss-0.3752, acc-0.8938, test loss-0.3801, acc-0.8923
Iter-81260, train loss-0.3320, acc-0.9000, valid loss-0.3752, acc-0.8940, test loss-0.3801, acc-0.8923
Iter-81270, train loss-0.5197, acc-0.8000, valid loss-0.3752, acc-0.8936, test loss-0.3801, acc-0.8924
Iter-81280, train loss-0.3399, acc-0.9000, valid loss-0.3752, acc-0.8938, test loss-0.3801, acc-0.8922
Iter-81290, train loss-0.4276, acc-0.8800, valid loss-0.3752, acc-0.8938, test loss-0.3801, acc-0.8922
Iter-81300, train loss-0.4582, acc-0.8600, valid loss-0.3751, acc-0.8938, test loss-0.3800, acc-0.8922
Iter-81310, train loss-0.4491, acc-0.8800, valid loss-0.3751, acc-0.8938, test loss-0.3800, acc-0.8923
Iter-81320, train loss-0.2219, acc-0.9600, valid loss-0.3751, acc-0.8938, test loss-0.3800, acc-0.8924
Iter-81330, train loss-0.3598, acc-0.9400, valid loss-0.3751, acc-0.8938, test loss-0.3800, acc-0.8925
Iter-81340, train loss-0.6678, acc-0.8400, valid loss-0.3751, acc-0.8938, test loss-0.3800, acc-0.8925
Iter-81350, train loss-0.2858, acc-0.9200, valid loss-0.3751, acc-0.8938, test loss-0.3800, acc-0.8924
Iter-81360, train loss-0.4024, acc-0.9000, valid loss-0.3751, acc-0.8938, test loss-0.3800, acc-0.8924
Iter-81370, train loss-0.4352, acc-0.8400, valid loss-0.3751, acc-0.8938, test loss-0.3799, acc-0.8924
Iter-81380, train loss-0.4639, acc-0.8200, valid loss-0.3751, acc-0.8938, test loss-0.3799, acc-0.8923
Iter-81390, train loss-0.5273, acc-0.8400, valid loss-0.3751, acc-0.8938, test loss-0.3799, acc-0.8923
Iter-81400, train loss-0.2737, acc-0.9400, valid loss-0.3750, acc-0.8938, test loss-0.3799, acc-0.8923
Iter-81410, train loss-0.4910, acc-0.9000, valid loss-0.3750, acc-0.8938, test loss-0.3799, acc-0.8923
Iter-81420, train loss-0.3050, acc-0.9400, valid loss-0.3750, acc-0.8938, test loss-0.3799, acc-0.8922
Iter-81430, train loss-0.3177, acc-0.9400, valid loss-0.3750, acc-0.8938, test loss-0.3799, acc-0.8922
Iter-81440, train loss-0.4068, acc-0.8800, valid loss-0.3750, acc-0.8938, test loss-0.3798, acc-0.8922
Iter-81450, train loss-0.6331, acc-0.8400, valid loss-0.3749, acc-0.8938, test loss-0.3798, acc-0.8922
Iter-81460, train loss-0.4705, acc-0.8400, valid loss-0.3749, acc-0.8938, test loss-0.3798, acc-0.8923
Iter-81470, train loss-0.5805, acc-0.9000, valid loss-0.3749, acc-0.8938, test loss-0.3798, acc-0.8923
Iter-81480, train loss-0.4953, acc-0.8600, valid loss-0.3749, acc-0.8938, test loss-0.3798, acc-0.8923
Iter-81490, train loss-0.4403, acc-0.9000, valid loss-0.3749, acc-0.8938, test loss-0.3798, acc-0.8924
Iter-81500, train loss-0.3632, acc-0.9000, valid loss-0.3749, acc-0.8938, test loss-0.3798, acc-0.8924
Iter-81510, train loss-0.2471, acc-0.9200, valid loss-0.3748, acc-0.8938, test loss-0.3797, acc-0.8924
Iter-81520, train loss-0.2419, acc-0.9200, valid loss-0.3748, acc-0.8938, test loss-0.3797, acc-0.8924
Iter-81530, train loss-0.3344, acc-0.9200, valid loss-0.3748, acc-0.8936, test loss-0.3797, acc-0.8924
Iter-81540, train loss-0.5590, acc-0.8600, valid loss-0.3748, acc-0.8936, test loss-0.3797, acc-0.8924
Iter-81550, train loss-0.4526, acc-0.8400, valid loss-0.3748, acc-0.8934, test loss-0.3797, acc-0.8924
Iter-81560, train loss-0.4545, acc-0.9000, valid loss-0.3747, acc-0.8936, test loss-0.3797, acc-0.8924
Iter-81570, train loss-0.4774, acc-0.8400, valid loss-0.3747, acc-0.8936, test loss-0.3796, acc-0.8924
Iter-81580, train loss-0.3052, acc-0.9200, valid loss-0.3747, acc-0.8936, test loss-0.3796, acc-0.8924
Iter-81590, train loss-0.2451, acc-0.9400, valid loss-0.3747, acc-0.8936, test loss-0.3796, acc-0.8924
Iter-81600, train loss-0.5791, acc-0.8800, valid loss-0.3747, acc-0.8936, test loss-0.3796, acc-0.8924
Iter-81610, train loss-0.8530, acc-0.7600, valid loss-0.3747, acc-0.8938, test loss-0.3796, acc-0.8924
Iter-81620, train loss-0.3126, acc-0.8800, valid loss-0.3746, acc-0.8938, test loss-0.3795, acc-0.8924
Iter-81630, train loss-0.4464, acc-0.8800, valid loss-0.3746, acc-0.8938, test loss-0.3795, acc-0.8924
Iter-81640, train loss-0.3589, acc-0.9200, valid loss-0.3746, acc-0.8938, test loss-0.3795, acc-0.8923
Iter-81650, train loss-0.4547, acc-0.8600, valid loss-0.3746, acc-0.8940, test loss-0.3795, acc-0.8923
Iter-81660, train loss-0.2985, acc-0.9400, valid loss-0.3746, acc-0.8940, test loss-0.3795, acc-0.8923
Iter-81670, train loss-0.5736, acc-0.8200, valid loss-0.3745, acc-0.8938, test loss-0.3795, acc-0.8923
Iter-81680, train loss-0.2710, acc-0.9200, valid loss-0.3745, acc-0.8940, test loss-0.3795, acc-0.8923
Iter-81690, train loss-0.4131, acc-0.8800, valid loss-0.3745, acc-0.8940, test loss-0.3794, acc-0.8923
Iter-81700, train loss-0.2433, acc-0.9400, valid loss-0.3745, acc-0.8940, test loss-0.3794, acc-0.8923
Iter-81710, train loss-0.4418, acc-0.8600, valid loss-0.3745, acc-0.8942, test loss-0.3794, acc-0.8923
Iter-81720, train loss-0.3104, acc-0.9200, valid loss-0.3744, acc-0.8940, test loss-0.3794, acc-0.8924
Iter-81730, train loss-0.3858, acc-0.9000, valid loss-0.3744, acc-0.8942, test loss-0.3794, acc-0.8923
Iter-81740, train loss-0.3070, acc-0.9600, valid loss-0.3744, acc-0.8942, test loss-0.3794, acc-0.8923
Iter-81750, train loss-0.2957, acc-0.9200, valid loss-0.3744, acc-0.8942, test loss-0.3793, acc-0.8923
Iter-81760, train loss-0.2960, acc-0.9400, valid loss-0.3744, acc-0.8942, test loss-0.3793, acc-0.8923
Iter-81770, train loss-0.4650, acc-0.8400, valid loss-0.3744, acc-0.8942, test loss-0.3793, acc-0.8923
Iter-81780, train loss-0.3899, acc-0.8400, valid loss-0.3744, acc-0.8942, test loss-0.3793, acc-0.8923
Iter-81790, train loss-0.4028, acc-0.8800, valid loss-0.3744, acc-0.8940, test loss-0.3793, acc-0.8924
Iter-81800, train loss-0.4326, acc-0.9200, valid loss-0.3743, acc-0.8940, test loss-0.3793, acc-0.8924
Iter-81810, train loss-0.6128, acc-0.8400, valid loss-0.3743, acc-0.8940, test loss-0.3793, acc-0.8924
Iter-81820, train loss-0.3808, acc-0.8800, valid loss-0.3743, acc-0.8940, test loss-0.3792, acc-0.8925
Iter-81830, train loss-0.3748, acc-0.8600, valid loss-0.3743, acc-0.8940, test loss-0.3792, acc-0.8925
Iter-81840, train loss-0.3880, acc-0.9000, valid loss-0.3743, acc-0.8942, test loss-0.3792, acc-0.8925
Iter-81850, train loss-0.5334, acc-0.8600, valid loss-0.3743, acc-0.8940, test loss-0.3792, acc-0.8924
Iter-81860, train loss-0.3144, acc-0.9000, valid loss-0.3742, acc-0.8940, test loss-0.3792, acc-0.8925
Iter-81870, train loss-0.3959, acc-0.9200, valid loss-0.3742, acc-0.8940, test loss-0.3792, acc-0.8925
Iter-81880, train loss-0.2449, acc-0.9600, valid loss-0.3742, acc-0.8940, test loss-0.3791, acc-0.8926
Iter-81890, train loss-0.3254, acc-0.9200, valid loss-0.3742, acc-0.8940, test loss-0.3791, acc-0.8926
Iter-81900, train loss-0.3602, acc-0.9000, valid loss-0.3742, acc-0.8940, test loss-0.3791, acc-0.8926
Iter-81910, train loss-0.4295, acc-0.8200, valid loss-0.3742, acc-0.8940, test loss-0.3791, acc-0.8926
Iter-81920, train loss-0.3893, acc-0.8600, valid loss-0.3742, acc-0.8938, test loss-0.3791, acc-0.8926
Iter-81930, train loss-0.4396, acc-0.8800, valid loss-0.3742, acc-0.8938, test loss-0.3791, acc-0.8926
Iter-81940, train loss-0.3215, acc-0.9000, valid loss-0.3741, acc-0.8938, test loss-0.3791, acc-0.8926
Iter-81950, train loss-0.4006, acc-0.8800, valid loss-0.3741, acc-0.8938, test loss-0.3790, acc-0.8926
Iter-81960, train loss-0.2636, acc-0.9200, valid loss-0.3741, acc-0.8938, test loss-0.3790, acc-0.8926
Iter-81970, train loss-0.4185, acc-0.8600, valid loss-0.3741, acc-0.8938, test loss-0.3790, acc-0.8926
Iter-81980, train loss-0.2562, acc-0.9600, valid loss-0.3741, acc-0.8938, test loss-0.3790, acc-0.8925
Iter-81990, train loss-0.4904, acc-0.8800, valid loss-0.3740, acc-0.8938, test loss-0.3790, acc-0.8925
Iter-82000, train loss-0.4729, acc-0.8400, valid loss-0.3740, acc-0.8938, test loss-0.3790, acc-0.8925
Iter-82010, train loss-0.5430, acc-0.8400, valid loss-0.3740, acc-0.8938, test loss-0.3790, acc-0.8925
Iter-82020, train loss-0.2803, acc-0.9000, valid loss-0.3740, acc-0.8936, test loss-0.3790, acc-0.8925
Iter-82030, train loss-0.4348, acc-0.8800, valid loss-0.3740, acc-0.8934, test loss-0.3789, acc-0.8925
Iter-82040, train loss-0.5106, acc-0.8000, valid loss-0.3740, acc-0.8938, test loss-0.3789, acc-0.8925
Iter-82050, train loss-0.2650, acc-0.9400, valid loss-0.3740, acc-0.8940, test loss-0.3789, acc-0.8925
Iter-82060, train loss-0.2839, acc-0.9600, valid loss-0.3739, acc-0.8942, test loss-0.3789, acc-0.8925
Iter-82070, train loss-0.4675, acc-0.8600, valid loss-0.3739, acc-0.8942, test loss-0.3789, acc-0.8925
Iter-82080, train loss-0.4313, acc-0.8200, valid loss-0.3739, acc-0.8942, test loss-0.3789, acc-0.8925
Iter-82090, train loss-0.4293, acc-0.9400, valid loss-0.3739, acc-0.8940, test loss-0.3789, acc-0.8925
Iter-82100, train loss-0.6741, acc-0.8200, valid loss-0.3739, acc-0.8940, test loss-0.3788, acc-0.8926
Iter-82110, train loss-0.1863, acc-0.9800, valid loss-0.3738, acc-0.8940, test loss-0.3788, acc-0.8926
Iter-82120, train loss-0.2887, acc-0.9000, valid loss-0.3738, acc-0.8942, test loss-0.3788, acc-0.8926
Iter-82130, train loss-0.2628, acc-0.9000, valid loss-0.3738, acc-0.8940, test loss-0.3788, acc-0.8927
Iter-82140, train loss-0.3085, acc-0.9200, valid loss-0.3738, acc-0.8938, test loss-0.3788, acc-0.8927
Iter-82150, train loss-0.5550, acc-0.8800, valid loss-0.3738, acc-0.8942, test loss-0.3788, acc-0.8926
Iter-82160, train loss-0.2499, acc-0.9200, valid loss-0.3738, acc-0.8940, test loss-0.3787, acc-0.8926
Iter-82170, train loss-0.4800, acc-0.8400, valid loss-0.3737, acc-0.8940, test loss-0.3787, acc-0.8926
Iter-82180, train loss-0.4338, acc-0.8600, valid loss-0.3737, acc-0.8946, test loss-0.3787, acc-0.8926
Iter-82190, train loss-0.4581, acc-0.8400, valid loss-0.3737, acc-0.8946, test loss-0.3787, acc-0.8926
Iter-82200, train loss-0.3784, acc-0.8800, valid loss-0.3737, acc-0.8940, test loss-0.3787, acc-0.8926
Iter-82210, train loss-0.3241, acc-0.9200, valid loss-0.3737, acc-0.8942, test loss-0.3786, acc-0.8926
Iter-82220, train loss-0.4761, acc-0.8400, valid loss-0.3736, acc-0.8942, test loss-0.3786, acc-0.8926
Iter-82230, train loss-0.4939, acc-0.9000, valid loss-0.3736, acc-0.8942, test loss-0.3786, acc-0.8925
Iter-82240, train loss-0.4672, acc-0.8600, valid loss-0.3736, acc-0.8942, test loss-0.3786, acc-0.8925
Iter-82250, train loss-0.4867, acc-0.8400, valid loss-0.3736, acc-0.8940, test loss-0.3786, acc-0.8925
Iter-82260, train loss-0.2496, acc-0.9400, valid loss-0.3736, acc-0.8946, test loss-0.3786, acc-0.8926
Iter-82270, train loss-0.4921, acc-0.8400, valid loss-0.3736, acc-0.8942, test loss-0.3785, acc-0.8925
Iter-82280, train loss-0.3990, acc-0.8800, valid loss-0.3735, acc-0.8942, test loss-0.3785, acc-0.8926
Iter-82290, train loss-0.2877, acc-0.9200, valid loss-0.3735, acc-0.8942, test loss-0.3785, acc-0.8926
Iter-82300, train loss-0.3798, acc-0.9000, valid loss-0.3735, acc-0.8944, test loss-0.3785, acc-0.8926
Iter-82310, train loss-0.2603, acc-0.9400, valid loss-0.3735, acc-0.8944, test loss-0.3785, acc-0.8926
Iter-82320, train loss-0.5561, acc-0.8200, valid loss-0.3735, acc-0.8944, test loss-0.3785, acc-0.8926
Iter-82330, train loss-0.4117, acc-0.8800, valid loss-0.3735, acc-0.8942, test loss-0.3785, acc-0.8926
Iter-82340, train loss-0.3115, acc-0.9400, valid loss-0.3735, acc-0.8942, test loss-0.3784, acc-0.8927
Iter-82350, train loss-0.3004, acc-0.9200, valid loss-0.3734, acc-0.8948, test loss-0.3784, acc-0.8927
Iter-82360, train loss-0.4584, acc-0.8400, valid loss-0.3734, acc-0.8948, test loss-0.3784, acc-0.8927
Iter-82370, train loss-0.4015, acc-0.8800, valid loss-0.3734, acc-0.8948, test loss-0.3784, acc-0.8927
Iter-82380, train loss-0.5055, acc-0.8600, valid loss-0.3734, acc-0.8946, test loss-0.3784, acc-0.8927
Iter-82390, train loss-0.2052, acc-0.9400, valid loss-0.3734, acc-0.8946, test loss-0.3783, acc-0.8927
Iter-82400, train loss-0.4813, acc-0.8400, valid loss-0.3733, acc-0.8946, test loss-0.3783, acc-0.8927
Iter-82410, train loss-0.4004, acc-0.8600, valid loss-0.3733, acc-0.8944, test loss-0.3783, acc-0.8927
Iter-82420, train loss-0.3770, acc-0.8600, valid loss-0.3733, acc-0.8944, test loss-0.3783, acc-0.8927
Iter-82430, train loss-0.4033, acc-0.8600, valid loss-0.3733, acc-0.8946, test loss-0.3783, acc-0.8928
Iter-82440, train loss-0.3548, acc-0.8800, valid loss-0.3733, acc-0.8944, test loss-0.3783, acc-0.8928
Iter-82450, train loss-0.2908, acc-0.9000, valid loss-0.3733, acc-0.8944, test loss-0.3783, acc-0.8928
Iter-82460, train loss-0.3780, acc-0.8800, valid loss-0.3733, acc-0.8946, test loss-0.3782, acc-0.8928
Iter-82470, train loss-0.2291, acc-0.9800, valid loss-0.3732, acc-0.8946, test loss-0.3782, acc-0.8928
Iter-82480, train loss-0.5580, acc-0.8400, valid loss-0.3732, acc-0.8948, test loss-0.3782, acc-0.8928
Iter-82490, train loss-0.3086, acc-0.9200, valid loss-0.3732, acc-0.8946, test loss-0.3782, acc-0.8928
Iter-82500, train loss-0.4156, acc-0.9000, valid loss-0.3732, acc-0.8946, test loss-0.3782, acc-0.8928
Iter-82510, train loss-0.3349, acc-0.9000, valid loss-0.3732, acc-0.8946, test loss-0.3782, acc-0.8928
Iter-82520, train loss-0.2457, acc-0.9200, valid loss-0.3732, acc-0.8946, test loss-0.3781, acc-0.8928
Iter-82530, train loss-0.4580, acc-0.8400, valid loss-0.3732, acc-0.8948, test loss-0.3781, acc-0.8928
Iter-82540, train loss-0.5345, acc-0.9000, valid loss-0.3731, acc-0.8948, test loss-0.3781, acc-0.8928
Iter-82550, train loss-0.2177, acc-0.9200, valid loss-0.3731, acc-0.8946, test loss-0.3781, acc-0.8928
Iter-82560, train loss-0.4008, acc-0.9000, valid loss-0.3731, acc-0.8950, test loss-0.3781, acc-0.8928
Iter-82570, train loss-0.2177, acc-0.9800, valid loss-0.3731, acc-0.8946, test loss-0.3781, acc-0.8927
Iter-82580, train loss-0.4206, acc-0.8600, valid loss-0.3731, acc-0.8946, test loss-0.3781, acc-0.8927
Iter-82590, train loss-0.3465, acc-0.9000, valid loss-0.3731, acc-0.8948, test loss-0.3781, acc-0.8927
Iter-82600, train loss-0.4433, acc-0.9000, valid loss-0.3731, acc-0.8948, test loss-0.3780, acc-0.8927
Iter-82610, train loss-0.4528, acc-0.8000, valid loss-0.3731, acc-0.8948, test loss-0.3780, acc-0.8927
Iter-82620, train loss-0.2744, acc-0.9200, valid loss-0.3730, acc-0.8948, test loss-0.3780, acc-0.8927
Iter-82630, train loss-0.2932, acc-0.9000, valid loss-0.3730, acc-0.8948, test loss-0.3780, acc-0.8927
Iter-82640, train loss-0.6028, acc-0.7800, valid loss-0.3730, acc-0.8946, test loss-0.3780, acc-0.8927
Iter-82650, train loss-0.4688, acc-0.8200, valid loss-0.3730, acc-0.8946, test loss-0.3780, acc-0.8927
Iter-82660, train loss-0.4097, acc-0.9000, valid loss-0.3730, acc-0.8946, test loss-0.3780, acc-0.8927
Iter-82670, train loss-0.2699, acc-0.9400, valid loss-0.3730, acc-0.8946, test loss-0.3779, acc-0.8927
Iter-82680, train loss-0.4375, acc-0.8800, valid loss-0.3730, acc-0.8944, test loss-0.3779, acc-0.8928
Iter-82690, train loss-0.4558, acc-0.8200, valid loss-0.3729, acc-0.8944, test loss-0.3779, acc-0.8928
Iter-82700, train loss-0.5926, acc-0.8400, valid loss-0.3729, acc-0.8944, test loss-0.3779, acc-0.8928
Iter-82710, train loss-0.4791, acc-0.8200, valid loss-0.3729, acc-0.8944, test loss-0.3779, acc-0.8928
Iter-82720, train loss-0.2694, acc-0.8800, valid loss-0.3729, acc-0.8946, test loss-0.3779, acc-0.8928
Iter-82730, train loss-0.5240, acc-0.8800, valid loss-0.3729, acc-0.8948, test loss-0.3779, acc-0.8928
Iter-82740, train loss-0.4705, acc-0.8600, valid loss-0.3729, acc-0.8952, test loss-0.3778, acc-0.8928
Iter-82750, train loss-0.4375, acc-0.8400, valid loss-0.3729, acc-0.8952, test loss-0.3778, acc-0.8928
Iter-82760, train loss-0.3038, acc-0.8800, valid loss-0.3728, acc-0.8952, test loss-0.3778, acc-0.8928
Iter-82770, train loss-0.3113, acc-0.9000, valid loss-0.3728, acc-0.8952, test loss-0.3778, acc-0.8928
Iter-82780, train loss-0.3782, acc-0.9200, valid loss-0.3728, acc-0.8952, test loss-0.3778, acc-0.8928
Iter-82790, train loss-0.5016, acc-0.8800, valid loss-0.3728, acc-0.8952, test loss-0.3778, acc-0.8928
Iter-82800, train loss-0.1897, acc-0.9800, valid loss-0.3728, acc-0.8946, test loss-0.3777, acc-0.8928
Iter-82810, train loss-0.3424, acc-0.8800, valid loss-0.3728, acc-0.8950, test loss-0.3777, acc-0.8928
Iter-82820, train loss-0.3489, acc-0.9200, valid loss-0.3728, acc-0.8948, test loss-0.3777, acc-0.8928
Iter-82830, train loss-0.3296, acc-0.9000, valid loss-0.3727, acc-0.8948, test loss-0.3777, acc-0.8928
Iter-82840, train loss-0.5125, acc-0.8200, valid loss-0.3727, acc-0.8950, test loss-0.3777, acc-0.8928
Iter-82850, train loss-0.3794, acc-0.9200, valid loss-0.3727, acc-0.8950, test loss-0.3777, acc-0.8928
Iter-82860, train loss-0.2427, acc-0.9400, valid loss-0.3727, acc-0.8948, test loss-0.3776, acc-0.8928
Iter-82870, train loss-0.4872, acc-0.8600, valid loss-0.3727, acc-0.8948, test loss-0.3776, acc-0.8928
Iter-82880, train loss-0.4643, acc-0.8600, valid loss-0.3726, acc-0.8948, test loss-0.3776, acc-0.8928
Iter-82890, train loss-0.5065, acc-0.8200, valid loss-0.3726, acc-0.8950, test loss-0.3776, acc-0.8928
Iter-82900, train loss-0.3773, acc-0.8800, valid loss-0.3726, acc-0.8948, test loss-0.3776, acc-0.8928
Iter-82910, train loss-0.2060, acc-0.9600, valid loss-0.3726, acc-0.8948, test loss-0.3776, acc-0.8928
Iter-82920, train loss-0.4652, acc-0.9000, valid loss-0.3726, acc-0.8948, test loss-0.3775, acc-0.8928
Iter-82930, train loss-0.4891, acc-0.8600, valid loss-0.3725, acc-0.8948, test loss-0.3775, acc-0.8928
Iter-82940, train loss-0.3794, acc-0.8800, valid loss-0.3725, acc-0.8948, test loss-0.3775, acc-0.8929
Iter-82950, train loss-0.3595, acc-0.9000, valid loss-0.3725, acc-0.8948, test loss-0.3775, acc-0.8928
Iter-82960, train loss-0.3765, acc-0.9200, valid loss-0.3725, acc-0.8948, test loss-0.3775, acc-0.8928
Iter-82970, train loss-0.5467, acc-0.8400, valid loss-0.3725, acc-0.8948, test loss-0.3775, acc-0.8928
Iter-82980, train loss-0.3550, acc-0.9000, valid loss-0.3724, acc-0.8952, test loss-0.3774, acc-0.8928
Iter-82990, train loss-0.3524, acc-0.9200, valid loss-0.3724, acc-0.8956, test loss-0.3774, acc-0.8928
Iter-83000, train loss-0.5062, acc-0.8400, valid loss-0.3724, acc-0.8952, test loss-0.3774, acc-0.8928
Iter-83010, train loss-0.3628, acc-0.8800, valid loss-0.3724, acc-0.8952, test loss-0.3774, acc-0.8928
Iter-83020, train loss-0.1773, acc-0.9600, valid loss-0.3724, acc-0.8948, test loss-0.3774, acc-0.8928
Iter-83030, train loss-0.4097, acc-0.9200, valid loss-0.3724, acc-0.8948, test loss-0.3774, acc-0.8928
Iter-83040, train loss-0.5262, acc-0.8800, valid loss-0.3723, acc-0.8950, test loss-0.3774, acc-0.8928
Iter-83050, train loss-0.3590, acc-0.9000, valid loss-0.3723, acc-0.8950, test loss-0.3774, acc-0.8928
Iter-83060, train loss-0.3177, acc-0.9200, valid loss-0.3723, acc-0.8950, test loss-0.3773, acc-0.8928
Iter-83070, train loss-0.4357, acc-0.9000, valid loss-0.3723, acc-0.8948, test loss-0.3773, acc-0.8928
Iter-83080, train loss-0.4536, acc-0.8800, valid loss-0.3723, acc-0.8948, test loss-0.3773, acc-0.8928
Iter-83090, train loss-0.3498, acc-0.9000, valid loss-0.3723, acc-0.8948, test loss-0.3773, acc-0.8928
Iter-83100, train loss-0.3903, acc-0.8800, valid loss-0.3722, acc-0.8950, test loss-0.3773, acc-0.8928
Iter-83110, train loss-0.4806, acc-0.8200, valid loss-0.3722, acc-0.8950, test loss-0.3772, acc-0.8928
Iter-83120, train loss-0.3597, acc-0.8600, valid loss-0.3722, acc-0.8950, test loss-0.3772, acc-0.8928
Iter-83130, train loss-0.3354, acc-0.9400, valid loss-0.3722, acc-0.8950, test loss-0.3772, acc-0.8928
Iter-83140, train loss-0.1924, acc-0.9400, valid loss-0.3722, acc-0.8950, test loss-0.3772, acc-0.8928
Iter-83150, train loss-0.5798, acc-0.8200, valid loss-0.3722, acc-0.8950, test loss-0.3772, acc-0.8928
Iter-83160, train loss-0.3071, acc-0.8800, valid loss-0.3721, acc-0.8950, test loss-0.3772, acc-0.8928
Iter-83170, train loss-0.2617, acc-0.9400, valid loss-0.3721, acc-0.8952, test loss-0.3772, acc-0.8928
Iter-83180, train loss-0.3375, acc-0.8800, valid loss-0.3721, acc-0.8950, test loss-0.3771, acc-0.8928
Iter-83190, train loss-0.4923, acc-0.8400, valid loss-0.3721, acc-0.8948, test loss-0.3771, acc-0.8928
Iter-83200, train loss-0.3156, acc-0.9200, valid loss-0.3721, acc-0.8948, test loss-0.3771, acc-0.8927
Iter-83210, train loss-0.4180, acc-0.8400, valid loss-0.3721, acc-0.8948, test loss-0.3771, acc-0.8927
Iter-83220, train loss-0.3588, acc-0.9400, valid loss-0.3720, acc-0.8948, test loss-0.3771, acc-0.8928
Iter-83230, train loss-0.4228, acc-0.8800, valid loss-0.3720, acc-0.8950, test loss-0.3771, acc-0.8927
Iter-83240, train loss-0.2469, acc-0.9600, valid loss-0.3720, acc-0.8950, test loss-0.3770, acc-0.8927
Iter-83250, train loss-0.4686, acc-0.9200, valid loss-0.3720, acc-0.8948, test loss-0.3770, acc-0.8927
Iter-83260, train loss-0.2628, acc-0.9200, valid loss-0.3720, acc-0.8948, test loss-0.3770, acc-0.8927
Iter-83270, train loss-0.4119, acc-0.8800, valid loss-0.3720, acc-0.8950, test loss-0.3770, acc-0.8927
Iter-83280, train loss-0.5142, acc-0.8600, valid loss-0.3720, acc-0.8950, test loss-0.3770, acc-0.8928
Iter-83290, train loss-0.3390, acc-0.8600, valid loss-0.3719, acc-0.8950, test loss-0.3770, acc-0.8928
Iter-83300, train loss-0.5559, acc-0.8400, valid loss-0.3719, acc-0.8952, test loss-0.3770, acc-0.8929
Iter-83310, train loss-0.3789, acc-0.8800, valid loss-0.3719, acc-0.8952, test loss-0.3770, acc-0.8929
Iter-83320, train loss-0.5188, acc-0.8800, valid loss-0.3719, acc-0.8952, test loss-0.3769, acc-0.8929
Iter-83330, train loss-0.3931, acc-0.8400, valid loss-0.3719, acc-0.8950, test loss-0.3769, acc-0.8929
Iter-83340, train loss-0.3798, acc-0.9600, valid loss-0.3719, acc-0.8948, test loss-0.3769, acc-0.8929
Iter-83350, train loss-0.5800, acc-0.8400, valid loss-0.3719, acc-0.8948, test loss-0.3769, acc-0.8929
Iter-83360, train loss-0.3621, acc-0.9000, valid loss-0.3719, acc-0.8946, test loss-0.3769, acc-0.8930
Iter-83370, train loss-0.3357, acc-0.9400, valid loss-0.3718, acc-0.8948, test loss-0.3769, acc-0.8929
Iter-83380, train loss-0.4584, acc-0.8800, valid loss-0.3718, acc-0.8948, test loss-0.3768, acc-0.8928
Iter-83390, train loss-0.3315, acc-0.9400, valid loss-0.3718, acc-0.8950, test loss-0.3768, acc-0.8927
Iter-83400, train loss-0.2534, acc-0.9200, valid loss-0.3718, acc-0.8952, test loss-0.3768, acc-0.8928
Iter-83410, train loss-0.2338, acc-0.9600, valid loss-0.3718, acc-0.8950, test loss-0.3768, acc-0.8928
Iter-83420, train loss-0.2234, acc-0.9000, valid loss-0.3718, acc-0.8948, test loss-0.3768, acc-0.8928
Iter-83430, train loss-0.5087, acc-0.8800, valid loss-0.3717, acc-0.8950, test loss-0.3768, acc-0.8930
Iter-83440, train loss-0.3643, acc-0.8800, valid loss-0.3717, acc-0.8952, test loss-0.3767, acc-0.8929
Iter-83450, train loss-0.3301, acc-0.9200, valid loss-0.3717, acc-0.8948, test loss-0.3767, acc-0.8929
Iter-83460, train loss-0.3124, acc-0.9000, valid loss-0.3717, acc-0.8950, test loss-0.3767, acc-0.8929
Iter-83470, train loss-0.2894, acc-0.9200, valid loss-0.3717, acc-0.8950, test loss-0.3767, acc-0.8929
Iter-83480, train loss-0.4339, acc-0.8400, valid loss-0.3716, acc-0.8950, test loss-0.3767, acc-0.8929
Iter-83490, train loss-0.2783, acc-0.9200, valid loss-0.3716, acc-0.8950, test loss-0.3767, acc-0.8930
Iter-83500, train loss-0.5220, acc-0.8400, valid loss-0.3716, acc-0.8950, test loss-0.3767, acc-0.8929
Iter-83510, train loss-0.3586, acc-0.9000, valid loss-0.3716, acc-0.8952, test loss-0.3767, acc-0.8931
Iter-83520, train loss-0.3597, acc-0.9000, valid loss-0.3716, acc-0.8950, test loss-0.3766, acc-0.8930
Iter-83530, train loss-0.2010, acc-0.9400, valid loss-0.3716, acc-0.8948, test loss-0.3766, acc-0.8929
Iter-83540, train loss-0.4456, acc-0.9200, valid loss-0.3715, acc-0.8950, test loss-0.3766, acc-0.8930
Iter-83550, train loss-0.3625, acc-0.9200, valid loss-0.3715, acc-0.8950, test loss-0.3766, acc-0.8930
Iter-83560, train loss-0.7035, acc-0.7800, valid loss-0.3715, acc-0.8950, test loss-0.3766, acc-0.8929
Iter-83570, train loss-0.4747, acc-0.8800, valid loss-0.3715, acc-0.8952, test loss-0.3766, acc-0.8931
Iter-83580, train loss-0.3031, acc-0.9400, valid loss-0.3715, acc-0.8952, test loss-0.3766, acc-0.8930
Iter-83590, train loss-0.4146, acc-0.8800, valid loss-0.3715, acc-0.8952, test loss-0.3766, acc-0.8929
Iter-83600, train loss-0.4206, acc-0.8400, valid loss-0.3714, acc-0.8952, test loss-0.3765, acc-0.8931
Iter-83610, train loss-0.4917, acc-0.8800, valid loss-0.3714, acc-0.8952, test loss-0.3765, acc-0.8931
Iter-83620, train loss-0.5619, acc-0.8200, valid loss-0.3714, acc-0.8952, test loss-0.3765, acc-0.8931
Iter-83630, train loss-0.2257, acc-0.9600, valid loss-0.3714, acc-0.8952, test loss-0.3765, acc-0.8931
Iter-83640, train loss-0.4234, acc-0.8800, valid loss-0.3714, acc-0.8952, test loss-0.3765, acc-0.8930
Iter-83650, train loss-0.4855, acc-0.8200, valid loss-0.3714, acc-0.8954, test loss-0.3765, acc-0.8929
Iter-83660, train loss-0.3164, acc-0.9200, valid loss-0.3713, acc-0.8954, test loss-0.3765, acc-0.8929
Iter-83670, train loss-0.4217, acc-0.9000, valid loss-0.3713, acc-0.8954, test loss-0.3765, acc-0.8929
Iter-83680, train loss-0.5580, acc-0.9200, valid loss-0.3713, acc-0.8954, test loss-0.3764, acc-0.8930
Iter-83690, train loss-0.1438, acc-0.9800, valid loss-0.3713, acc-0.8954, test loss-0.3764, acc-0.8930
Iter-83700, train loss-0.3166, acc-0.9200, valid loss-0.3713, acc-0.8954, test loss-0.3764, acc-0.8930
Iter-83710, train loss-0.2731, acc-0.9400, valid loss-0.3712, acc-0.8952, test loss-0.3764, acc-0.8930
Iter-83720, train loss-0.2733, acc-0.9400, valid loss-0.3712, acc-0.8952, test loss-0.3764, acc-0.8930
Iter-83730, train loss-0.5476, acc-0.8600, valid loss-0.3712, acc-0.8952, test loss-0.3764, acc-0.8931
Iter-83740, train loss-0.4646, acc-0.8600, valid loss-0.3712, acc-0.8952, test loss-0.3764, acc-0.8931
Iter-83750, train loss-0.4859, acc-0.9000, valid loss-0.3712, acc-0.8952, test loss-0.3763, acc-0.8931
Iter-83760, train loss-0.1683, acc-0.9600, valid loss-0.3711, acc-0.8952, test loss-0.3763, acc-0.8931
Iter-83770, train loss-0.4086, acc-0.8800, valid loss-0.3711, acc-0.8952, test loss-0.3763, acc-0.8931
Iter-83780, train loss-0.2775, acc-0.9200, valid loss-0.3711, acc-0.8952, test loss-0.3763, acc-0.8931
Iter-83790, train loss-0.3606, acc-0.9400, valid loss-0.3711, acc-0.8952, test loss-0.3762, acc-0.8931
Iter-83800, train loss-0.3076, acc-0.9000, valid loss-0.3711, acc-0.8952, test loss-0.3762, acc-0.8931
Iter-83810, train loss-0.3490, acc-0.9200, valid loss-0.3711, acc-0.8952, test loss-0.3762, acc-0.8931
Iter-83820, train loss-0.6294, acc-0.8200, valid loss-0.3710, acc-0.8954, test loss-0.3762, acc-0.8931
Iter-83830, train loss-0.6410, acc-0.7800, valid loss-0.3710, acc-0.8954, test loss-0.3762, acc-0.8931
Iter-83840, train loss-0.3918, acc-0.8600, valid loss-0.3710, acc-0.8952, test loss-0.3762, acc-0.8931
Iter-83850, train loss-0.3929, acc-0.9200, valid loss-0.3710, acc-0.8954, test loss-0.3762, acc-0.8931
Iter-83860, train loss-0.4318, acc-0.8600, valid loss-0.3710, acc-0.8954, test loss-0.3762, acc-0.8931
Iter-83870, train loss-0.3248, acc-0.8800, valid loss-0.3710, acc-0.8952, test loss-0.3761, acc-0.8931
Iter-83880, train loss-0.3147, acc-0.9200, valid loss-0.3709, acc-0.8954, test loss-0.3761, acc-0.8931
Iter-83890, train loss-0.4336, acc-0.8600, valid loss-0.3709, acc-0.8954, test loss-0.3761, acc-0.8931
Iter-83900, train loss-0.3097, acc-0.8800, valid loss-0.3709, acc-0.8954, test loss-0.3761, acc-0.8931
Iter-83910, train loss-0.3701, acc-0.9000, valid loss-0.3709, acc-0.8954, test loss-0.3761, acc-0.8931
Iter-83920, train loss-0.2770, acc-0.9400, valid loss-0.3709, acc-0.8954, test loss-0.3760, acc-0.8931
Iter-83930, train loss-0.3708, acc-0.8800, valid loss-0.3708, acc-0.8954, test loss-0.3760, acc-0.8932
Iter-83940, train loss-0.4158, acc-0.8600, valid loss-0.3708, acc-0.8952, test loss-0.3760, acc-0.8932
Iter-83950, train loss-0.4827, acc-0.9200, valid loss-0.3708, acc-0.8952, test loss-0.3760, acc-0.8932
Iter-83960, train loss-0.3673, acc-0.8800, valid loss-0.3708, acc-0.8954, test loss-0.3760, acc-0.8932
Iter-83970, train loss-0.3935, acc-0.9000, valid loss-0.3708, acc-0.8954, test loss-0.3759, acc-0.8932
Iter-83980, train loss-0.3595, acc-0.9400, valid loss-0.3707, acc-0.8952, test loss-0.3759, acc-0.8932
Iter-83990, train loss-0.4057, acc-0.9200, valid loss-0.3707, acc-0.8952, test loss-0.3759, acc-0.8932
Iter-84000, train loss-0.3005, acc-0.9400, valid loss-0.3707, acc-0.8954, test loss-0.3759, acc-0.8932
Iter-84010, train loss-0.3617, acc-0.8800, valid loss-0.3707, acc-0.8954, test loss-0.3759, acc-0.8932
Iter-84020, train loss-0.4310, acc-0.9000, valid loss-0.3707, acc-0.8954, test loss-0.3759, acc-0.8932
Iter-84030, train loss-0.5608, acc-0.9000, valid loss-0.3707, acc-0.8954, test loss-0.3758, acc-0.8932
Iter-84040, train loss-0.2919, acc-0.9200, valid loss-0.3707, acc-0.8954, test loss-0.3758, acc-0.8932
Iter-84050, train loss-0.4822, acc-0.9200, valid loss-0.3706, acc-0.8954, test loss-0.3758, acc-0.8931
Iter-84060, train loss-0.2888, acc-0.9200, valid loss-0.3706, acc-0.8954, test loss-0.3758, acc-0.8931
Iter-84070, train loss-0.5797, acc-0.8400, valid loss-0.3706, acc-0.8954, test loss-0.3758, acc-0.8931
Iter-84080, train loss-0.3048, acc-0.9200, valid loss-0.3706, acc-0.8954, test loss-0.3758, acc-0.8932
Iter-84090, train loss-0.2792, acc-0.9400, valid loss-0.3706, acc-0.8954, test loss-0.3758, acc-0.8932
Iter-84100, train loss-0.4861, acc-0.8400, valid loss-0.3706, acc-0.8954, test loss-0.3757, acc-0.8932
Iter-84110, train loss-0.2886, acc-0.9600, valid loss-0.3706, acc-0.8954, test loss-0.3757, acc-0.8930
Iter-84120, train loss-0.5522, acc-0.8600, valid loss-0.3706, acc-0.8954, test loss-0.3757, acc-0.8931
Iter-84130, train loss-0.4338, acc-0.9000, valid loss-0.3705, acc-0.8954, test loss-0.3757, acc-0.8931
Iter-84140, train loss-0.4111, acc-0.8800, valid loss-0.3705, acc-0.8954, test loss-0.3757, acc-0.8931
Iter-84150, train loss-0.4153, acc-0.9000, valid loss-0.3705, acc-0.8954, test loss-0.3757, acc-0.8931
Iter-84160, train loss-0.2587, acc-0.9600, valid loss-0.3705, acc-0.8956, test loss-0.3756, acc-0.8932
Iter-84170, train loss-0.4044, acc-0.8400, valid loss-0.3705, acc-0.8956, test loss-0.3756, acc-0.8932
Iter-84180, train loss-0.3830, acc-0.8600, valid loss-0.3705, acc-0.8956, test loss-0.3756, acc-0.8932
Iter-84190, train loss-0.4727, acc-0.8600, valid loss-0.3704, acc-0.8956, test loss-0.3756, acc-0.8932
Iter-84200, train loss-0.6397, acc-0.8200, valid loss-0.3704, acc-0.8956, test loss-0.3756, acc-0.8932
Iter-84210, train loss-0.6462, acc-0.8200, valid loss-0.3704, acc-0.8956, test loss-0.3756, acc-0.8932
Iter-84220, train loss-0.3214, acc-0.9200, valid loss-0.3704, acc-0.8956, test loss-0.3755, acc-0.8932
Iter-84230, train loss-0.5214, acc-0.8000, valid loss-0.3704, acc-0.8956, test loss-0.3755, acc-0.8932
Iter-84240, train loss-0.5631, acc-0.8400, valid loss-0.3704, acc-0.8956, test loss-0.3755, acc-0.8932
Iter-84250, train loss-0.2296, acc-0.9600, valid loss-0.3703, acc-0.8956, test loss-0.3755, acc-0.8932
Iter-84260, train loss-0.3206, acc-0.9200, valid loss-0.3703, acc-0.8956, test loss-0.3755, acc-0.8932
Iter-84270, train loss-0.3340, acc-0.9000, valid loss-0.3703, acc-0.8956, test loss-0.3755, acc-0.8932
Iter-84280, train loss-0.5345, acc-0.8000, valid loss-0.3703, acc-0.8956, test loss-0.3754, acc-0.8932
Iter-84290, train loss-0.2375, acc-0.9200, valid loss-0.3703, acc-0.8956, test loss-0.3754, acc-0.8932
Iter-84300, train loss-0.3501, acc-0.9200, valid loss-0.3703, acc-0.8956, test loss-0.3754, acc-0.8932
Iter-84310, train loss-0.4489, acc-0.9000, valid loss-0.3703, acc-0.8956, test loss-0.3754, acc-0.8932
Iter-84320, train loss-0.5309, acc-0.8200, valid loss-0.3703, acc-0.8956, test loss-0.3754, acc-0.8932
Iter-84330, train loss-0.2565, acc-0.9200, valid loss-0.3702, acc-0.8956, test loss-0.3754, acc-0.8932
Iter-84340, train loss-0.2618, acc-0.9200, valid loss-0.3702, acc-0.8956, test loss-0.3754, acc-0.8932
Iter-84350, train loss-0.3399, acc-0.9000, valid loss-0.3702, acc-0.8956, test loss-0.3754, acc-0.8932
Iter-84360, train loss-0.3890, acc-0.8800, valid loss-0.3702, acc-0.8956, test loss-0.3754, acc-0.8932
Iter-84370, train loss-0.3607, acc-0.9000, valid loss-0.3702, acc-0.8956, test loss-0.3753, acc-0.8933
Iter-84380, train loss-0.4717, acc-0.9000, valid loss-0.3702, acc-0.8956, test loss-0.3753, acc-0.8933
Iter-84390, train loss-0.2727, acc-0.9600, valid loss-0.3702, acc-0.8956, test loss-0.3753, acc-0.8933
Iter-84400, train loss-0.3562, acc-0.8800, valid loss-0.3701, acc-0.8956, test loss-0.3753, acc-0.8931
Iter-84410, train loss-0.3751, acc-0.8800, valid loss-0.3701, acc-0.8956, test loss-0.3753, acc-0.8933
Iter-84420, train loss-0.2113, acc-0.9600, valid loss-0.3701, acc-0.8956, test loss-0.3753, acc-0.8932
Iter-84430, train loss-0.2795, acc-0.9200, valid loss-0.3701, acc-0.8956, test loss-0.3752, acc-0.8933
Iter-84440, train loss-0.2014, acc-0.9600, valid loss-0.3701, acc-0.8956, test loss-0.3752, acc-0.8933
Iter-84450, train loss-0.5365, acc-0.8600, valid loss-0.3701, acc-0.8956, test loss-0.3752, acc-0.8933
Iter-84460, train loss-0.3517, acc-0.9200, valid loss-0.3701, acc-0.8956, test loss-0.3752, acc-0.8934
Iter-84470, train loss-0.5150, acc-0.9000, valid loss-0.3700, acc-0.8956, test loss-0.3752, acc-0.8933
Iter-84480, train loss-0.5123, acc-0.8600, valid loss-0.3700, acc-0.8956, test loss-0.3752, acc-0.8931
Iter-84490, train loss-0.6336, acc-0.8400, valid loss-0.3700, acc-0.8956, test loss-0.3751, acc-0.8932
Iter-84500, train loss-0.5805, acc-0.8200, valid loss-0.3700, acc-0.8956, test loss-0.3751, acc-0.8931
Iter-84510, train loss-0.3380, acc-0.8800, valid loss-0.3700, acc-0.8956, test loss-0.3751, acc-0.8931
Iter-84520, train loss-0.5177, acc-0.8600, valid loss-0.3700, acc-0.8956, test loss-0.3751, acc-0.8931
Iter-84530, train loss-0.3636, acc-0.9000, valid loss-0.3699, acc-0.8958, test loss-0.3751, acc-0.8932
Iter-84540, train loss-0.4136, acc-0.8600, valid loss-0.3699, acc-0.8956, test loss-0.3751, acc-0.8930
Iter-84550, train loss-0.3088, acc-0.9600, valid loss-0.3699, acc-0.8960, test loss-0.3750, acc-0.8931
Iter-84560, train loss-0.3570, acc-0.8600, valid loss-0.3699, acc-0.8960, test loss-0.3750, acc-0.8930
Iter-84570, train loss-0.5963, acc-0.8600, valid loss-0.3699, acc-0.8958, test loss-0.3750, acc-0.8930
Iter-84580, train loss-0.2420, acc-0.9400, valid loss-0.3699, acc-0.8960, test loss-0.3750, acc-0.8929
Iter-84590, train loss-0.3385, acc-0.9200, valid loss-0.3699, acc-0.8960, test loss-0.3750, acc-0.8929
Iter-84600, train loss-0.3858, acc-0.8200, valid loss-0.3699, acc-0.8960, test loss-0.3750, acc-0.8930
Iter-84610, train loss-0.6326, acc-0.8400, valid loss-0.3698, acc-0.8958, test loss-0.3749, acc-0.8929
Iter-84620, train loss-0.4116, acc-0.8600, valid loss-0.3698, acc-0.8960, test loss-0.3749, acc-0.8929
Iter-84630, train loss-0.3379, acc-0.8800, valid loss-0.3698, acc-0.8960, test loss-0.3749, acc-0.8928
Iter-84640, train loss-0.7305, acc-0.7600, valid loss-0.3698, acc-0.8960, test loss-0.3749, acc-0.8929
Iter-84650, train loss-0.4119, acc-0.8800, valid loss-0.3698, acc-0.8960, test loss-0.3749, acc-0.8928
Iter-84660, train loss-0.2931, acc-0.9200, valid loss-0.3697, acc-0.8960, test loss-0.3749, acc-0.8929
Iter-84670, train loss-0.3166, acc-0.9000, valid loss-0.3697, acc-0.8960, test loss-0.3748, acc-0.8929
Iter-84680, train loss-0.6165, acc-0.8400, valid loss-0.3697, acc-0.8960, test loss-0.3748, acc-0.8930
Iter-84690, train loss-0.3195, acc-0.9200, valid loss-0.3697, acc-0.8960, test loss-0.3748, acc-0.8930
Iter-84700, train loss-0.3657, acc-0.8800, valid loss-0.3697, acc-0.8960, test loss-0.3748, acc-0.8929
Iter-84710, train loss-0.3170, acc-0.9000, valid loss-0.3697, acc-0.8960, test loss-0.3748, acc-0.8929
Iter-84720, train loss-0.3810, acc-0.8600, valid loss-0.3696, acc-0.8960, test loss-0.3748, acc-0.8929
Iter-84730, train loss-0.5716, acc-0.8400, valid loss-0.3696, acc-0.8960, test loss-0.3748, acc-0.8929
Iter-84740, train loss-0.4086, acc-0.9000, valid loss-0.3696, acc-0.8960, test loss-0.3747, acc-0.8929
Iter-84750, train loss-0.6138, acc-0.7800, valid loss-0.3696, acc-0.8960, test loss-0.3747, acc-0.8929
Iter-84760, train loss-0.2417, acc-0.9400, valid loss-0.3696, acc-0.8960, test loss-0.3747, acc-0.8929
Iter-84770, train loss-0.5401, acc-0.8800, valid loss-0.3696, acc-0.8960, test loss-0.3747, acc-0.8929
Iter-84780, train loss-0.4993, acc-0.8400, valid loss-0.3696, acc-0.8960, test loss-0.3747, acc-0.8929
Iter-84790, train loss-0.2818, acc-0.9400, valid loss-0.3695, acc-0.8960, test loss-0.3747, acc-0.8929
Iter-84800, train loss-0.4582, acc-0.8400, valid loss-0.3695, acc-0.8960, test loss-0.3747, acc-0.8929
Iter-84810, train loss-0.6144, acc-0.8400, valid loss-0.3695, acc-0.8960, test loss-0.3746, acc-0.8929
Iter-84820, train loss-0.4367, acc-0.8800, valid loss-0.3695, acc-0.8960, test loss-0.3746, acc-0.8929
Iter-84830, train loss-0.2145, acc-0.9200, valid loss-0.3695, acc-0.8960, test loss-0.3746, acc-0.8929
Iter-84840, train loss-0.3327, acc-0.9000, valid loss-0.3695, acc-0.8960, test loss-0.3746, acc-0.8929
Iter-84850, train loss-0.5915, acc-0.8600, valid loss-0.3694, acc-0.8962, test loss-0.3746, acc-0.8929
Iter-84860, train loss-0.3764, acc-0.9000, valid loss-0.3694, acc-0.8960, test loss-0.3746, acc-0.8929
Iter-84870, train loss-0.2521, acc-0.9200, valid loss-0.3694, acc-0.8960, test loss-0.3745, acc-0.8929
Iter-84880, train loss-0.5445, acc-0.8000, valid loss-0.3694, acc-0.8962, test loss-0.3745, acc-0.8929
Iter-84890, train loss-0.6859, acc-0.8000, valid loss-0.3694, acc-0.8962, test loss-0.3745, acc-0.8929
Iter-84900, train loss-0.3272, acc-0.9000, valid loss-0.3694, acc-0.8962, test loss-0.3745, acc-0.8929
Iter-84910, train loss-0.3118, acc-0.9000, valid loss-0.3694, acc-0.8962, test loss-0.3745, acc-0.8929
Iter-84920, train loss-0.4403, acc-0.8800, valid loss-0.3693, acc-0.8962, test loss-0.3745, acc-0.8929
Iter-84930, train loss-0.1529, acc-0.9600, valid loss-0.3693, acc-0.8962, test loss-0.3745, acc-0.8929
Iter-84940, train loss-0.5512, acc-0.8600, valid loss-0.3693, acc-0.8962, test loss-0.3744, acc-0.8929
Iter-84950, train loss-0.5036, acc-0.8600, valid loss-0.3693, acc-0.8962, test loss-0.3744, acc-0.8929
Iter-84960, train loss-0.4686, acc-0.8800, valid loss-0.3693, acc-0.8962, test loss-0.3744, acc-0.8929
Iter-84970, train loss-0.3744, acc-0.9200, valid loss-0.3692, acc-0.8962, test loss-0.3744, acc-0.8929
Iter-84980, train loss-0.3925, acc-0.8800, valid loss-0.3692, acc-0.8962, test loss-0.3744, acc-0.8929
Iter-84990, train loss-0.6799, acc-0.7800, valid loss-0.3692, acc-0.8962, test loss-0.3744, acc-0.8929
Iter-85000, train loss-0.3106, acc-0.8800, valid loss-0.3692, acc-0.8962, test loss-0.3744, acc-0.8929
Iter-85010, train loss-0.2532, acc-0.9400, valid loss-0.3692, acc-0.8962, test loss-0.3744, acc-0.8929
Iter-85020, train loss-0.5545, acc-0.8400, valid loss-0.3692, acc-0.8962, test loss-0.3743, acc-0.8929
Iter-85030, train loss-0.3796, acc-0.9200, valid loss-0.3692, acc-0.8960, test loss-0.3743, acc-0.8929
Iter-85040, train loss-0.3328, acc-0.8800, valid loss-0.3691, acc-0.8960, test loss-0.3743, acc-0.8929
Iter-85050, train loss-0.3223, acc-0.9000, valid loss-0.3691, acc-0.8958, test loss-0.3743, acc-0.8929
Iter-85060, train loss-0.5348, acc-0.8000, valid loss-0.3691, acc-0.8958, test loss-0.3743, acc-0.8929
Iter-85070, train loss-0.2493, acc-0.9400, valid loss-0.3691, acc-0.8958, test loss-0.3743, acc-0.8929
Iter-85080, train loss-0.4044, acc-0.8600, valid loss-0.3691, acc-0.8960, test loss-0.3743, acc-0.8929
Iter-85090, train loss-0.5871, acc-0.8200, valid loss-0.3691, acc-0.8960, test loss-0.3742, acc-0.8929
Iter-85100, train loss-0.3660, acc-0.8800, valid loss-0.3691, acc-0.8960, test loss-0.3742, acc-0.8929
Iter-85110, train loss-0.3628, acc-0.9000, valid loss-0.3690, acc-0.8960, test loss-0.3742, acc-0.8928
Iter-85120, train loss-0.2286, acc-0.9600, valid loss-0.3690, acc-0.8960, test loss-0.3742, acc-0.8929
Iter-85130, train loss-0.3136, acc-0.9400, valid loss-0.3690, acc-0.8962, test loss-0.3742, acc-0.8929
Iter-85140, train loss-0.2281, acc-0.9600, valid loss-0.3690, acc-0.8962, test loss-0.3742, acc-0.8929
Iter-85150, train loss-0.4328, acc-0.8400, valid loss-0.3690, acc-0.8960, test loss-0.3741, acc-0.8929
Iter-85160, train loss-0.2931, acc-0.9400, valid loss-0.3689, acc-0.8962, test loss-0.3741, acc-0.8929
Iter-85170, train loss-0.4799, acc-0.8000, valid loss-0.3689, acc-0.8962, test loss-0.3741, acc-0.8928
Iter-85180, train loss-0.4732, acc-0.9000, valid loss-0.3689, acc-0.8960, test loss-0.3741, acc-0.8928
Iter-85190, train loss-0.3647, acc-0.8800, valid loss-0.3689, acc-0.8960, test loss-0.3741, acc-0.8929
Iter-85200, train loss-0.3467, acc-0.9200, valid loss-0.3689, acc-0.8960, test loss-0.3741, acc-0.8929
Iter-85210, train loss-0.3101, acc-0.9200, valid loss-0.3689, acc-0.8960, test loss-0.3741, acc-0.8928
Iter-85220, train loss-0.4097, acc-0.8400, valid loss-0.3688, acc-0.8960, test loss-0.3741, acc-0.8927
Iter-85230, train loss-0.3327, acc-0.9200, valid loss-0.3688, acc-0.8960, test loss-0.3741, acc-0.8927
Iter-85240, train loss-0.3197, acc-0.9600, valid loss-0.3688, acc-0.8960, test loss-0.3740, acc-0.8928
Iter-85250, train loss-0.5250, acc-0.9000, valid loss-0.3688, acc-0.8962, test loss-0.3740, acc-0.8928
Iter-85260, train loss-0.3122, acc-0.9200, valid loss-0.3688, acc-0.8960, test loss-0.3740, acc-0.8928
Iter-85270, train loss-0.2809, acc-0.9200, valid loss-0.3687, acc-0.8960, test loss-0.3740, acc-0.8928
Iter-85280, train loss-0.4891, acc-0.8000, valid loss-0.3687, acc-0.8958, test loss-0.3740, acc-0.8929
Iter-85290, train loss-0.6460, acc-0.8200, valid loss-0.3687, acc-0.8960, test loss-0.3740, acc-0.8929
Iter-85300, train loss-0.4004, acc-0.8400, valid loss-0.3687, acc-0.8958, test loss-0.3739, acc-0.8929
Iter-85310, train loss-0.3405, acc-0.8800, valid loss-0.3687, acc-0.8960, test loss-0.3739, acc-0.8929
Iter-85320, train loss-0.5064, acc-0.9000, valid loss-0.3687, acc-0.8960, test loss-0.3739, acc-0.8929
Iter-85330, train loss-0.4181, acc-0.8400, valid loss-0.3687, acc-0.8958, test loss-0.3739, acc-0.8929
Iter-85340, train loss-0.3412, acc-0.8800, valid loss-0.3686, acc-0.8958, test loss-0.3739, acc-0.8929
Iter-85350, train loss-0.4101, acc-0.9200, valid loss-0.3686, acc-0.8958, test loss-0.3738, acc-0.8929
Iter-85360, train loss-0.5924, acc-0.9000, valid loss-0.3686, acc-0.8958, test loss-0.3738, acc-0.8929
Iter-85370, train loss-0.2796, acc-0.9200, valid loss-0.3686, acc-0.8958, test loss-0.3738, acc-0.8929
Iter-85380, train loss-0.4786, acc-0.8600, valid loss-0.3686, acc-0.8958, test loss-0.3738, acc-0.8929
Iter-85390, train loss-0.4297, acc-0.8400, valid loss-0.3686, acc-0.8958, test loss-0.3738, acc-0.8929
Iter-85400, train loss-0.3561, acc-0.8600, valid loss-0.3686, acc-0.8958, test loss-0.3738, acc-0.8929
Iter-85410, train loss-0.6184, acc-0.8400, valid loss-0.3685, acc-0.8958, test loss-0.3738, acc-0.8929
Iter-85420, train loss-0.5044, acc-0.8600, valid loss-0.3685, acc-0.8956, test loss-0.3737, acc-0.8929
Iter-85430, train loss-0.6277, acc-0.8400, valid loss-0.3685, acc-0.8956, test loss-0.3737, acc-0.8929
Iter-85440, train loss-0.5665, acc-0.8800, valid loss-0.3685, acc-0.8956, test loss-0.3737, acc-0.8929
Iter-85450, train loss-0.3322, acc-0.9200, valid loss-0.3685, acc-0.8958, test loss-0.3737, acc-0.8929
Iter-85460, train loss-0.2612, acc-0.9400, valid loss-0.3685, acc-0.8956, test loss-0.3737, acc-0.8929
Iter-85470, train loss-0.3207, acc-0.9400, valid loss-0.3685, acc-0.8956, test loss-0.3737, acc-0.8929
Iter-85480, train loss-0.5855, acc-0.8600, valid loss-0.3685, acc-0.8958, test loss-0.3736, acc-0.8928
Iter-85490, train loss-0.3322, acc-0.9400, valid loss-0.3685, acc-0.8958, test loss-0.3736, acc-0.8929
Iter-85500, train loss-0.2359, acc-0.9400, valid loss-0.3684, acc-0.8958, test loss-0.3736, acc-0.8929
Iter-85510, train loss-0.3385, acc-0.9200, valid loss-0.3684, acc-0.8958, test loss-0.3736, acc-0.8929
Iter-85520, train loss-0.3622, acc-0.9200, valid loss-0.3684, acc-0.8958, test loss-0.3736, acc-0.8928
Iter-85530, train loss-0.2306, acc-0.9400, valid loss-0.3684, acc-0.8956, test loss-0.3736, acc-0.8927
Iter-85540, train loss-0.3275, acc-0.9000, valid loss-0.3684, acc-0.8956, test loss-0.3736, acc-0.8928
Iter-85550, train loss-0.3450, acc-0.9200, valid loss-0.3684, acc-0.8956, test loss-0.3735, acc-0.8928
Iter-85560, train loss-0.3938, acc-0.8400, valid loss-0.3683, acc-0.8956, test loss-0.3735, acc-0.8928
Iter-85570, train loss-0.4180, acc-0.8600, valid loss-0.3683, acc-0.8956, test loss-0.3735, acc-0.8929
Iter-85580, train loss-0.6069, acc-0.8400, valid loss-0.3683, acc-0.8958, test loss-0.3735, acc-0.8928
Iter-85590, train loss-0.3123, acc-0.9000, valid loss-0.3683, acc-0.8960, test loss-0.3735, acc-0.8929
Iter-85600, train loss-0.2293, acc-0.9600, valid loss-0.3683, acc-0.8958, test loss-0.3735, acc-0.8930
Iter-85610, train loss-0.3093, acc-0.9000, valid loss-0.3682, acc-0.8960, test loss-0.3735, acc-0.8928
Iter-85620, train loss-0.6888, acc-0.8000, valid loss-0.3682, acc-0.8962, test loss-0.3735, acc-0.8930
Iter-85630, train loss-0.3854, acc-0.9000, valid loss-0.3682, acc-0.8960, test loss-0.3735, acc-0.8930
Iter-85640, train loss-0.2812, acc-0.9400, valid loss-0.3682, acc-0.8962, test loss-0.3734, acc-0.8930
Iter-85650, train loss-0.4556, acc-0.8200, valid loss-0.3682, acc-0.8962, test loss-0.3734, acc-0.8929
Iter-85660, train loss-0.3690, acc-0.9000, valid loss-0.3682, acc-0.8962, test loss-0.3734, acc-0.8930
Iter-85670, train loss-0.2945, acc-0.9600, valid loss-0.3682, acc-0.8962, test loss-0.3734, acc-0.8930
Iter-85680, train loss-0.4097, acc-0.8600, valid loss-0.3682, acc-0.8960, test loss-0.3734, acc-0.8929
Iter-85690, train loss-0.4116, acc-0.9200, valid loss-0.3681, acc-0.8960, test loss-0.3734, acc-0.8928
Iter-85700, train loss-0.3616, acc-0.9400, valid loss-0.3681, acc-0.8958, test loss-0.3733, acc-0.8928
Iter-85710, train loss-0.4503, acc-0.8800, valid loss-0.3681, acc-0.8960, test loss-0.3733, acc-0.8929
Iter-85720, train loss-0.7583, acc-0.8600, valid loss-0.3681, acc-0.8960, test loss-0.3733, acc-0.8929
Iter-85730, train loss-0.4204, acc-0.8800, valid loss-0.3681, acc-0.8960, test loss-0.3733, acc-0.8929
Iter-85740, train loss-0.3825, acc-0.9000, valid loss-0.3681, acc-0.8960, test loss-0.3733, acc-0.8929
Iter-85750, train loss-0.3459, acc-0.9200, valid loss-0.3681, acc-0.8960, test loss-0.3733, acc-0.8930
Iter-85760, train loss-0.6408, acc-0.8200, valid loss-0.3680, acc-0.8960, test loss-0.3733, acc-0.8929
Iter-85770, train loss-0.5759, acc-0.8200, valid loss-0.3680, acc-0.8960, test loss-0.3733, acc-0.8928
Iter-85780, train loss-0.8734, acc-0.7600, valid loss-0.3680, acc-0.8960, test loss-0.3733, acc-0.8930
Iter-85790, train loss-0.4069, acc-0.8800, valid loss-0.3680, acc-0.8960, test loss-0.3733, acc-0.8930
Iter-85800, train loss-0.3324, acc-0.9000, valid loss-0.3680, acc-0.8960, test loss-0.3732, acc-0.8930
Iter-85810, train loss-0.2256, acc-0.9600, valid loss-0.3680, acc-0.8960, test loss-0.3732, acc-0.8929
Iter-85820, train loss-0.3100, acc-0.8800, valid loss-0.3680, acc-0.8960, test loss-0.3732, acc-0.8930
Iter-85830, train loss-0.3477, acc-0.9000, valid loss-0.3680, acc-0.8960, test loss-0.3732, acc-0.8929
Iter-85840, train loss-0.4814, acc-0.8400, valid loss-0.3680, acc-0.8960, test loss-0.3732, acc-0.8930
Iter-85850, train loss-0.3275, acc-0.9600, valid loss-0.3679, acc-0.8958, test loss-0.3732, acc-0.8928
Iter-85860, train loss-0.3432, acc-0.9000, valid loss-0.3679, acc-0.8958, test loss-0.3731, acc-0.8930
Iter-85870, train loss-0.3000, acc-0.8800, valid loss-0.3679, acc-0.8958, test loss-0.3731, acc-0.8930
Iter-85880, train loss-0.3815, acc-0.8600, valid loss-0.3679, acc-0.8958, test loss-0.3731, acc-0.8930
Iter-85890, train loss-0.4513, acc-0.8600, valid loss-0.3678, acc-0.8958, test loss-0.3731, acc-0.8930
Iter-85900, train loss-0.6407, acc-0.7600, valid loss-0.3678, acc-0.8958, test loss-0.3731, acc-0.8931
Iter-85910, train loss-0.4690, acc-0.8600, valid loss-0.3678, acc-0.8958, test loss-0.3731, acc-0.8930
Iter-85920, train loss-0.4027, acc-0.9000, valid loss-0.3678, acc-0.8958, test loss-0.3731, acc-0.8930
Iter-85930, train loss-0.2238, acc-0.9400, valid loss-0.3678, acc-0.8958, test loss-0.3731, acc-0.8929
Iter-85940, train loss-0.2820, acc-0.9600, valid loss-0.3678, acc-0.8958, test loss-0.3730, acc-0.8930
Iter-85950, train loss-0.2763, acc-0.9000, valid loss-0.3677, acc-0.8958, test loss-0.3730, acc-0.8931
Iter-85960, train loss-0.3078, acc-0.9000, valid loss-0.3677, acc-0.8958, test loss-0.3730, acc-0.8930
Iter-85970, train loss-0.3165, acc-0.9200, valid loss-0.3677, acc-0.8958, test loss-0.3730, acc-0.8930
Iter-85980, train loss-0.2878, acc-0.9200, valid loss-0.3677, acc-0.8960, test loss-0.3730, acc-0.8930
Iter-85990, train loss-0.3819, acc-0.8600, valid loss-0.3677, acc-0.8960, test loss-0.3730, acc-0.8931
Iter-86000, train loss-0.3737, acc-0.8600, valid loss-0.3677, acc-0.8960, test loss-0.3729, acc-0.8931
Iter-86010, train loss-0.2231, acc-0.9200, valid loss-0.3676, acc-0.8960, test loss-0.3729, acc-0.8931
Iter-86020, train loss-0.3917, acc-0.8800, valid loss-0.3676, acc-0.8960, test loss-0.3729, acc-0.8931
Iter-86030, train loss-0.5426, acc-0.8800, valid loss-0.3676, acc-0.8960, test loss-0.3729, acc-0.8931
Iter-86040, train loss-0.5126, acc-0.8600, valid loss-0.3676, acc-0.8960, test loss-0.3729, acc-0.8931
Iter-86050, train loss-0.3595, acc-0.8800, valid loss-0.3676, acc-0.8960, test loss-0.3729, acc-0.8931
Iter-86060, train loss-0.4930, acc-0.8800, valid loss-0.3676, acc-0.8960, test loss-0.3729, acc-0.8931
Iter-86070, train loss-0.3247, acc-0.9200, valid loss-0.3675, acc-0.8960, test loss-0.3728, acc-0.8931
Iter-86080, train loss-0.4312, acc-0.9000, valid loss-0.3675, acc-0.8960, test loss-0.3728, acc-0.8931
Iter-86090, train loss-0.4152, acc-0.8600, valid loss-0.3675, acc-0.8960, test loss-0.3728, acc-0.8931
Iter-86100, train loss-0.3357, acc-0.9200, valid loss-0.3675, acc-0.8960, test loss-0.3728, acc-0.8931
Iter-86110, train loss-0.2658, acc-0.9200, valid loss-0.3675, acc-0.8962, test loss-0.3728, acc-0.8931
Iter-86120, train loss-0.4722, acc-0.8800, valid loss-0.3675, acc-0.8960, test loss-0.3728, acc-0.8931
Iter-86130, train loss-0.3637, acc-0.9000, valid loss-0.3675, acc-0.8960, test loss-0.3728, acc-0.8931
Iter-86140, train loss-0.1728, acc-0.9600, valid loss-0.3674, acc-0.8962, test loss-0.3728, acc-0.8931
Iter-86150, train loss-0.3239, acc-0.9200, valid loss-0.3674, acc-0.8960, test loss-0.3727, acc-0.8931
Iter-86160, train loss-0.3265, acc-0.9200, valid loss-0.3674, acc-0.8960, test loss-0.3727, acc-0.8931
Iter-86170, train loss-0.3767, acc-0.9200, valid loss-0.3674, acc-0.8962, test loss-0.3727, acc-0.8931
Iter-86180, train loss-0.2469, acc-0.9200, valid loss-0.3674, acc-0.8960, test loss-0.3727, acc-0.8931
Iter-86190, train loss-0.3088, acc-0.9000, valid loss-0.3674, acc-0.8960, test loss-0.3727, acc-0.8931
Iter-86200, train loss-0.3881, acc-0.8800, valid loss-0.3673, acc-0.8962, test loss-0.3727, acc-0.8931
Iter-86210, train loss-0.3732, acc-0.9400, valid loss-0.3673, acc-0.8962, test loss-0.3727, acc-0.8931
Iter-86220, train loss-0.4216, acc-0.8600, valid loss-0.3673, acc-0.8962, test loss-0.3727, acc-0.8931
Iter-86230, train loss-0.4019, acc-0.8600, valid loss-0.3673, acc-0.8962, test loss-0.3726, acc-0.8931
Iter-86240, train loss-0.3550, acc-0.9200, valid loss-0.3673, acc-0.8962, test loss-0.3726, acc-0.8931
Iter-86250, train loss-0.2964, acc-0.9400, valid loss-0.3673, acc-0.8962, test loss-0.3726, acc-0.8931
Iter-86260, train loss-0.3049, acc-0.9400, valid loss-0.3673, acc-0.8962, test loss-0.3726, acc-0.8931
Iter-86270, train loss-0.4099, acc-0.8600, valid loss-0.3672, acc-0.8960, test loss-0.3726, acc-0.8932
Iter-86280, train loss-0.3594, acc-0.9200, valid loss-0.3672, acc-0.8962, test loss-0.3726, acc-0.8932
Iter-86290, train loss-0.3895, acc-0.8800, valid loss-0.3672, acc-0.8962, test loss-0.3726, acc-0.8932
Iter-86300, train loss-0.2941, acc-0.9000, valid loss-0.3672, acc-0.8962, test loss-0.3725, acc-0.8932
Iter-86310, train loss-0.5111, acc-0.8800, valid loss-0.3672, acc-0.8962, test loss-0.3725, acc-0.8932
Iter-86320, train loss-0.3807, acc-0.8800, valid loss-0.3671, acc-0.8964, test loss-0.3725, acc-0.8932
Iter-86330, train loss-0.6284, acc-0.7800, valid loss-0.3671, acc-0.8964, test loss-0.3725, acc-0.8932
Iter-86340, train loss-0.3492, acc-0.8800, valid loss-0.3671, acc-0.8962, test loss-0.3725, acc-0.8932
Iter-86350, train loss-0.2716, acc-0.9200, valid loss-0.3671, acc-0.8966, test loss-0.3725, acc-0.8931
Iter-86360, train loss-0.5128, acc-0.8600, valid loss-0.3671, acc-0.8964, test loss-0.3725, acc-0.8932
Iter-86370, train loss-0.4558, acc-0.8600, valid loss-0.3671, acc-0.8966, test loss-0.3724, acc-0.8933
Iter-86380, train loss-0.4446, acc-0.8800, valid loss-0.3670, acc-0.8966, test loss-0.3724, acc-0.8932
Iter-86390, train loss-0.2839, acc-0.9000, valid loss-0.3670, acc-0.8964, test loss-0.3724, acc-0.8932
Iter-86400, train loss-0.3679, acc-0.8800, valid loss-0.3670, acc-0.8964, test loss-0.3724, acc-0.8932
Iter-86410, train loss-0.4847, acc-0.8800, valid loss-0.3670, acc-0.8964, test loss-0.3724, acc-0.8932
Iter-86420, train loss-0.3339, acc-0.8800, valid loss-0.3670, acc-0.8964, test loss-0.3724, acc-0.8931
Iter-86430, train loss-0.5426, acc-0.8200, valid loss-0.3670, acc-0.8964, test loss-0.3724, acc-0.8932
Iter-86440, train loss-0.6327, acc-0.8400, valid loss-0.3670, acc-0.8964, test loss-0.3723, acc-0.8932
Iter-86450, train loss-0.4441, acc-0.8400, valid loss-0.3669, acc-0.8964, test loss-0.3723, acc-0.8932
Iter-86460, train loss-0.1952, acc-0.9600, valid loss-0.3669, acc-0.8964, test loss-0.3723, acc-0.8932
Iter-86470, train loss-0.5381, acc-0.8600, valid loss-0.3669, acc-0.8964, test loss-0.3723, acc-0.8932
Iter-86480, train loss-0.2945, acc-0.9200, valid loss-0.3669, acc-0.8964, test loss-0.3723, acc-0.8932
Iter-86490, train loss-0.3914, acc-0.8400, valid loss-0.3669, acc-0.8964, test loss-0.3723, acc-0.8933
Iter-86500, train loss-0.5009, acc-0.8600, valid loss-0.3669, acc-0.8964, test loss-0.3723, acc-0.8933
Iter-86510, train loss-0.4570, acc-0.8400, valid loss-0.3669, acc-0.8964, test loss-0.3723, acc-0.8933
Iter-86520, train loss-0.3923, acc-0.8400, valid loss-0.3669, acc-0.8964, test loss-0.3722, acc-0.8933
Iter-86530, train loss-0.2467, acc-0.9200, valid loss-0.3668, acc-0.8964, test loss-0.3722, acc-0.8933
Iter-86540, train loss-0.3827, acc-0.8600, valid loss-0.3668, acc-0.8966, test loss-0.3722, acc-0.8933
Iter-86550, train loss-0.5308, acc-0.8400, valid loss-0.3668, acc-0.8966, test loss-0.3722, acc-0.8933
Iter-86560, train loss-0.3679, acc-0.8800, valid loss-0.3668, acc-0.8966, test loss-0.3722, acc-0.8934
Iter-86570, train loss-0.1303, acc-0.9800, valid loss-0.3668, acc-0.8966, test loss-0.3722, acc-0.8934
Iter-86580, train loss-0.4830, acc-0.8800, valid loss-0.3668, acc-0.8966, test loss-0.3722, acc-0.8934
Iter-86590, train loss-0.3251, acc-0.8800, valid loss-0.3667, acc-0.8964, test loss-0.3721, acc-0.8934
Iter-86600, train loss-0.4943, acc-0.8600, valid loss-0.3667, acc-0.8964, test loss-0.3721, acc-0.8934
Iter-86610, train loss-0.3390, acc-0.9200, valid loss-0.3667, acc-0.8964, test loss-0.3721, acc-0.8934
Iter-86620, train loss-0.3832, acc-0.9200, valid loss-0.3667, acc-0.8966, test loss-0.3721, acc-0.8934
Iter-86630, train loss-0.5972, acc-0.8200, valid loss-0.3667, acc-0.8966, test loss-0.3721, acc-0.8934
Iter-86640, train loss-0.3652, acc-0.8400, valid loss-0.3667, acc-0.8966, test loss-0.3721, acc-0.8934
Iter-86650, train loss-0.3054, acc-0.9000, valid loss-0.3667, acc-0.8966, test loss-0.3721, acc-0.8934
Iter-86660, train loss-0.3047, acc-0.8800, valid loss-0.3667, acc-0.8968, test loss-0.3720, acc-0.8935
Iter-86670, train loss-0.4367, acc-0.8400, valid loss-0.3667, acc-0.8970, test loss-0.3720, acc-0.8935
Iter-86680, train loss-0.2204, acc-0.9400, valid loss-0.3666, acc-0.8970, test loss-0.3720, acc-0.8935
Iter-86690, train loss-0.3459, acc-0.9200, valid loss-0.3666, acc-0.8968, test loss-0.3720, acc-0.8935
Iter-86700, train loss-0.3508, acc-0.9000, valid loss-0.3666, acc-0.8970, test loss-0.3720, acc-0.8935
Iter-86710, train loss-0.4022, acc-0.8600, valid loss-0.3666, acc-0.8968, test loss-0.3720, acc-0.8935
Iter-86720, train loss-0.4375, acc-0.9000, valid loss-0.3666, acc-0.8966, test loss-0.3719, acc-0.8935
Iter-86730, train loss-0.5141, acc-0.7800, valid loss-0.3666, acc-0.8966, test loss-0.3719, acc-0.8936
Iter-86740, train loss-0.3463, acc-0.8800, valid loss-0.3666, acc-0.8968, test loss-0.3719, acc-0.8936
Iter-86750, train loss-0.4089, acc-0.8800, valid loss-0.3665, acc-0.8964, test loss-0.3719, acc-0.8936
Iter-86760, train loss-0.4997, acc-0.8600, valid loss-0.3665, acc-0.8966, test loss-0.3719, acc-0.8935
Iter-86770, train loss-0.2222, acc-0.9200, valid loss-0.3665, acc-0.8964, test loss-0.3719, acc-0.8935
Iter-86780, train loss-0.2916, acc-0.9400, valid loss-0.3665, acc-0.8962, test loss-0.3719, acc-0.8936
Iter-86790, train loss-0.2020, acc-0.9400, valid loss-0.3665, acc-0.8962, test loss-0.3719, acc-0.8935
Iter-86800, train loss-0.4699, acc-0.7800, valid loss-0.3665, acc-0.8962, test loss-0.3718, acc-0.8935
Iter-86810, train loss-0.1496, acc-1.0000, valid loss-0.3665, acc-0.8964, test loss-0.3718, acc-0.8935
Iter-86820, train loss-0.4394, acc-0.8400, valid loss-0.3665, acc-0.8962, test loss-0.3718, acc-0.8935
Iter-86830, train loss-0.3604, acc-0.9000, valid loss-0.3665, acc-0.8962, test loss-0.3718, acc-0.8935
Iter-86840, train loss-0.1996, acc-0.9600, valid loss-0.3665, acc-0.8964, test loss-0.3718, acc-0.8935
Iter-86850, train loss-0.2668, acc-0.9600, valid loss-0.3664, acc-0.8964, test loss-0.3718, acc-0.8935
Iter-86860, train loss-0.2391, acc-0.9400, valid loss-0.3664, acc-0.8964, test loss-0.3717, acc-0.8935
Iter-86870, train loss-0.4204, acc-0.9000, valid loss-0.3664, acc-0.8964, test loss-0.3717, acc-0.8935
Iter-86880, train loss-0.3045, acc-0.9400, valid loss-0.3664, acc-0.8964, test loss-0.3717, acc-0.8935
Iter-86890, train loss-0.4651, acc-0.8800, valid loss-0.3664, acc-0.8964, test loss-0.3717, acc-0.8934
Iter-86900, train loss-0.3995, acc-0.9200, valid loss-0.3664, acc-0.8962, test loss-0.3717, acc-0.8933
Iter-86910, train loss-0.4127, acc-0.8800, valid loss-0.3663, acc-0.8964, test loss-0.3717, acc-0.8934
Iter-86920, train loss-0.4367, acc-0.9200, valid loss-0.3663, acc-0.8962, test loss-0.3716, acc-0.8933
Iter-86930, train loss-0.3883, acc-0.9000, valid loss-0.3663, acc-0.8964, test loss-0.3716, acc-0.8935
Iter-86940, train loss-0.3984, acc-0.8800, valid loss-0.3663, acc-0.8960, test loss-0.3716, acc-0.8936
Iter-86950, train loss-0.5761, acc-0.8000, valid loss-0.3663, acc-0.8964, test loss-0.3716, acc-0.8935
Iter-86960, train loss-0.3849, acc-0.8800, valid loss-0.3663, acc-0.8964, test loss-0.3716, acc-0.8935
Iter-86970, train loss-0.4278, acc-0.8800, valid loss-0.3663, acc-0.8964, test loss-0.3716, acc-0.8936
Iter-86980, train loss-0.5026, acc-0.8400, valid loss-0.3663, acc-0.8964, test loss-0.3715, acc-0.8936
Iter-86990, train loss-0.1813, acc-0.9600, valid loss-0.3662, acc-0.8966, test loss-0.3715, acc-0.8936
Iter-87000, train loss-0.4966, acc-0.8600, valid loss-0.3662, acc-0.8966, test loss-0.3715, acc-0.8935
Iter-87010, train loss-0.4837, acc-0.9000, valid loss-0.3662, acc-0.8966, test loss-0.3715, acc-0.8936
Iter-87020, train loss-0.3109, acc-0.8600, valid loss-0.3662, acc-0.8966, test loss-0.3715, acc-0.8936
Iter-87030, train loss-0.2765, acc-0.9000, valid loss-0.3662, acc-0.8966, test loss-0.3715, acc-0.8935
Iter-87040, train loss-0.4556, acc-0.8800, valid loss-0.3662, acc-0.8966, test loss-0.3715, acc-0.8936
Iter-87050, train loss-0.4100, acc-0.9200, valid loss-0.3661, acc-0.8964, test loss-0.3714, acc-0.8935
Iter-87060, train loss-0.1310, acc-0.9800, valid loss-0.3661, acc-0.8964, test loss-0.3714, acc-0.8935
Iter-87070, train loss-0.2022, acc-0.9400, valid loss-0.3661, acc-0.8964, test loss-0.3714, acc-0.8934
Iter-87080, train loss-0.3196, acc-0.9200, valid loss-0.3661, acc-0.8962, test loss-0.3714, acc-0.8934
Iter-87090, train loss-0.3708, acc-0.9000, valid loss-0.3661, acc-0.8962, test loss-0.3714, acc-0.8935
Iter-87100, train loss-0.2208, acc-0.9400, valid loss-0.3661, acc-0.8962, test loss-0.3714, acc-0.8936
Iter-87110, train loss-0.1663, acc-0.9800, valid loss-0.3661, acc-0.8962, test loss-0.3714, acc-0.8935
Iter-87120, train loss-0.2820, acc-0.9400, valid loss-0.3660, acc-0.8964, test loss-0.3714, acc-0.8936
Iter-87130, train loss-0.4739, acc-0.8400, valid loss-0.3660, acc-0.8962, test loss-0.3713, acc-0.8936
Iter-87140, train loss-0.3589, acc-0.8800, valid loss-0.3660, acc-0.8962, test loss-0.3713, acc-0.8936
Iter-87150, train loss-0.3241, acc-0.9000, valid loss-0.3660, acc-0.8962, test loss-0.3713, acc-0.8936
Iter-87160, train loss-0.5265, acc-0.8000, valid loss-0.3660, acc-0.8960, test loss-0.3713, acc-0.8936
Iter-87170, train loss-0.4484, acc-0.8800, valid loss-0.3660, acc-0.8962, test loss-0.3713, acc-0.8935
Iter-87180, train loss-0.2206, acc-0.9400, valid loss-0.3660, acc-0.8962, test loss-0.3713, acc-0.8937
Iter-87190, train loss-0.5494, acc-0.8400, valid loss-0.3659, acc-0.8962, test loss-0.3712, acc-0.8937
Iter-87200, train loss-0.4723, acc-0.8800, valid loss-0.3659, acc-0.8962, test loss-0.3712, acc-0.8938
Iter-87210, train loss-0.2453, acc-0.9600, valid loss-0.3659, acc-0.8962, test loss-0.3712, acc-0.8936
Iter-87220, train loss-0.3626, acc-0.9000, valid loss-0.3659, acc-0.8964, test loss-0.3712, acc-0.8936
Iter-87230, train loss-0.4506, acc-0.8400, valid loss-0.3659, acc-0.8962, test loss-0.3712, acc-0.8936
Iter-87240, train loss-0.4887, acc-0.8600, valid loss-0.3659, acc-0.8962, test loss-0.3712, acc-0.8936
Iter-87250, train loss-0.5130, acc-0.8400, valid loss-0.3659, acc-0.8962, test loss-0.3712, acc-0.8936
Iter-87260, train loss-0.4236, acc-0.9000, valid loss-0.3659, acc-0.8964, test loss-0.3712, acc-0.8936
Iter-87270, train loss-0.2577, acc-0.9600, valid loss-0.3658, acc-0.8962, test loss-0.3711, acc-0.8937
Iter-87280, train loss-0.3492, acc-0.8400, valid loss-0.3658, acc-0.8962, test loss-0.3711, acc-0.8938
Iter-87290, train loss-0.4076, acc-0.8600, valid loss-0.3658, acc-0.8964, test loss-0.3711, acc-0.8938
Iter-87300, train loss-0.3168, acc-0.9400, valid loss-0.3658, acc-0.8964, test loss-0.3711, acc-0.8938
Iter-87310, train loss-0.2542, acc-0.9400, valid loss-0.3658, acc-0.8966, test loss-0.3711, acc-0.8936
Iter-87320, train loss-0.3961, acc-0.8800, valid loss-0.3658, acc-0.8966, test loss-0.3711, acc-0.8938
Iter-87330, train loss-0.5288, acc-0.8000, valid loss-0.3658, acc-0.8964, test loss-0.3711, acc-0.8937
Iter-87340, train loss-0.5892, acc-0.8400, valid loss-0.3658, acc-0.8964, test loss-0.3711, acc-0.8936
Iter-87350, train loss-0.4251, acc-0.9000, valid loss-0.3658, acc-0.8964, test loss-0.3711, acc-0.8936
Iter-87360, train loss-0.3353, acc-0.8600, valid loss-0.3657, acc-0.8964, test loss-0.3710, acc-0.8937
Iter-87370, train loss-0.4017, acc-0.8400, valid loss-0.3657, acc-0.8964, test loss-0.3710, acc-0.8937
Iter-87380, train loss-0.2309, acc-0.9200, valid loss-0.3657, acc-0.8964, test loss-0.3710, acc-0.8936
Iter-87390, train loss-0.5614, acc-0.8200, valid loss-0.3657, acc-0.8966, test loss-0.3710, acc-0.8937
Iter-87400, train loss-0.4045, acc-0.8600, valid loss-0.3657, acc-0.8964, test loss-0.3710, acc-0.8937
Iter-87410, train loss-0.3450, acc-0.8600, valid loss-0.3657, acc-0.8964, test loss-0.3710, acc-0.8937
Iter-87420, train loss-0.5819, acc-0.8400, valid loss-0.3657, acc-0.8968, test loss-0.3710, acc-0.8938
Iter-87430, train loss-0.3156, acc-0.9600, valid loss-0.3656, acc-0.8968, test loss-0.3709, acc-0.8938
Iter-87440, train loss-0.6946, acc-0.8200, valid loss-0.3656, acc-0.8966, test loss-0.3709, acc-0.8937
Iter-87450, train loss-0.4715, acc-0.8600, valid loss-0.3656, acc-0.8966, test loss-0.3709, acc-0.8937
Iter-87460, train loss-0.2729, acc-0.9400, valid loss-0.3656, acc-0.8966, test loss-0.3709, acc-0.8938
Iter-87470, train loss-0.3647, acc-0.8600, valid loss-0.3656, acc-0.8966, test loss-0.3709, acc-0.8937
Iter-87480, train loss-0.2714, acc-0.9200, valid loss-0.3656, acc-0.8966, test loss-0.3709, acc-0.8937
Iter-87490, train loss-0.2166, acc-0.9800, valid loss-0.3656, acc-0.8966, test loss-0.3709, acc-0.8936
Iter-87500, train loss-0.6351, acc-0.8200, valid loss-0.3656, acc-0.8968, test loss-0.3709, acc-0.8937
Iter-87510, train loss-0.6001, acc-0.7600, valid loss-0.3656, acc-0.8968, test loss-0.3708, acc-0.8937
Iter-87520, train loss-0.3131, acc-0.9200, valid loss-0.3655, acc-0.8968, test loss-0.3708, acc-0.8937
Iter-87530, train loss-0.3452, acc-0.9400, valid loss-0.3655, acc-0.8968, test loss-0.3708, acc-0.8938
Iter-87540, train loss-0.6234, acc-0.8200, valid loss-0.3655, acc-0.8966, test loss-0.3708, acc-0.8937
Iter-87550, train loss-0.4127, acc-0.8800, valid loss-0.3655, acc-0.8966, test loss-0.3708, acc-0.8937
Iter-87560, train loss-0.4011, acc-0.9200, valid loss-0.3655, acc-0.8966, test loss-0.3708, acc-0.8936
Iter-87570, train loss-0.3380, acc-0.8800, valid loss-0.3655, acc-0.8966, test loss-0.3707, acc-0.8936
Iter-87580, train loss-0.4256, acc-0.8600, valid loss-0.3654, acc-0.8968, test loss-0.3707, acc-0.8937
Iter-87590, train loss-0.1546, acc-1.0000, valid loss-0.3654, acc-0.8968, test loss-0.3707, acc-0.8937
Iter-87600, train loss-0.3640, acc-0.8800, valid loss-0.3654, acc-0.8970, test loss-0.3707, acc-0.8937
Iter-87610, train loss-0.2470, acc-0.9200, valid loss-0.3654, acc-0.8970, test loss-0.3707, acc-0.8937
Iter-87620, train loss-0.3901, acc-0.9200, valid loss-0.3654, acc-0.8968, test loss-0.3707, acc-0.8937
Iter-87630, train loss-0.3801, acc-0.9000, valid loss-0.3654, acc-0.8970, test loss-0.3707, acc-0.8937
Iter-87640, train loss-0.2258, acc-0.9600, valid loss-0.3653, acc-0.8968, test loss-0.3706, acc-0.8937
Iter-87650, train loss-0.3585, acc-0.8800, valid loss-0.3653, acc-0.8968, test loss-0.3706, acc-0.8937
Iter-87660, train loss-0.4142, acc-0.8600, valid loss-0.3653, acc-0.8970, test loss-0.3706, acc-0.8937
Iter-87670, train loss-0.4524, acc-0.8800, valid loss-0.3653, acc-0.8970, test loss-0.3706, acc-0.8936
Iter-87680, train loss-0.5943, acc-0.8400, valid loss-0.3653, acc-0.8970, test loss-0.3706, acc-0.8937
Iter-87690, train loss-0.2653, acc-0.9200, valid loss-0.3653, acc-0.8968, test loss-0.3706, acc-0.8939
Iter-87700, train loss-0.3340, acc-0.9200, valid loss-0.3653, acc-0.8966, test loss-0.3705, acc-0.8939
Iter-87710, train loss-0.2904, acc-0.8800, valid loss-0.3652, acc-0.8966, test loss-0.3705, acc-0.8939
Iter-87720, train loss-0.3073, acc-0.9200, valid loss-0.3652, acc-0.8968, test loss-0.3705, acc-0.8938
Iter-87730, train loss-0.3820, acc-0.9000, valid loss-0.3652, acc-0.8968, test loss-0.3705, acc-0.8938
Iter-87740, train loss-0.3490, acc-0.9200, valid loss-0.3652, acc-0.8966, test loss-0.3705, acc-0.8938
Iter-87750, train loss-0.4026, acc-0.9000, valid loss-0.3652, acc-0.8966, test loss-0.3705, acc-0.8937
Iter-87760, train loss-0.2948, acc-0.9600, valid loss-0.3651, acc-0.8966, test loss-0.3705, acc-0.8937
Iter-87770, train loss-0.5654, acc-0.8400, valid loss-0.3651, acc-0.8968, test loss-0.3704, acc-0.8937
Iter-87780, train loss-0.2887, acc-0.9000, valid loss-0.3651, acc-0.8968, test loss-0.3704, acc-0.8938
Iter-87790, train loss-0.4093, acc-0.9200, valid loss-0.3651, acc-0.8966, test loss-0.3704, acc-0.8938
Iter-87800, train loss-0.1414, acc-0.9800, valid loss-0.3651, acc-0.8966, test loss-0.3704, acc-0.8938
Iter-87810, train loss-0.3745, acc-0.9200, valid loss-0.3651, acc-0.8966, test loss-0.3703, acc-0.8937
Iter-87820, train loss-0.5488, acc-0.7600, valid loss-0.3650, acc-0.8966, test loss-0.3703, acc-0.8937
Iter-87830, train loss-0.3136, acc-0.9000, valid loss-0.3650, acc-0.8966, test loss-0.3703, acc-0.8937
Iter-87840, train loss-0.4762, acc-0.8600, valid loss-0.3650, acc-0.8968, test loss-0.3703, acc-0.8938
Iter-87850, train loss-0.2309, acc-0.9600, valid loss-0.3650, acc-0.8968, test loss-0.3703, acc-0.8938
Iter-87860, train loss-0.5259, acc-0.8200, valid loss-0.3650, acc-0.8968, test loss-0.3703, acc-0.8938
Iter-87870, train loss-0.2713, acc-0.9200, valid loss-0.3650, acc-0.8968, test loss-0.3702, acc-0.8937
Iter-87880, train loss-0.4711, acc-0.9000, valid loss-0.3650, acc-0.8968, test loss-0.3702, acc-0.8938
Iter-87890, train loss-0.2233, acc-0.9400, valid loss-0.3650, acc-0.8970, test loss-0.3702, acc-0.8937
Iter-87900, train loss-0.4451, acc-0.8400, valid loss-0.3649, acc-0.8970, test loss-0.3702, acc-0.8936
Iter-87910, train loss-0.4041, acc-0.8800, valid loss-0.3649, acc-0.8968, test loss-0.3702, acc-0.8936
Iter-87920, train loss-0.3552, acc-0.8400, valid loss-0.3649, acc-0.8966, test loss-0.3702, acc-0.8936
Iter-87930, train loss-0.1902, acc-0.9800, valid loss-0.3649, acc-0.8968, test loss-0.3702, acc-0.8936
Iter-87940, train loss-0.2709, acc-0.9400, valid loss-0.3649, acc-0.8968, test loss-0.3701, acc-0.8936
Iter-87950, train loss-0.2833, acc-0.9000, valid loss-0.3649, acc-0.8970, test loss-0.3701, acc-0.8937
Iter-87960, train loss-0.3815, acc-0.9000, valid loss-0.3648, acc-0.8968, test loss-0.3701, acc-0.8936
Iter-87970, train loss-0.3610, acc-0.8800, valid loss-0.3648, acc-0.8968, test loss-0.3701, acc-0.8937
Iter-87980, train loss-0.3461, acc-0.9400, valid loss-0.3648, acc-0.8970, test loss-0.3701, acc-0.8937
Iter-87990, train loss-0.2079, acc-0.9600, valid loss-0.3648, acc-0.8970, test loss-0.3701, acc-0.8937
Iter-88000, train loss-0.3616, acc-0.8600, valid loss-0.3648, acc-0.8968, test loss-0.3701, acc-0.8937
Iter-88010, train loss-0.3063, acc-0.8600, valid loss-0.3648, acc-0.8970, test loss-0.3701, acc-0.8937
Iter-88020, train loss-0.2548, acc-0.9000, valid loss-0.3648, acc-0.8970, test loss-0.3701, acc-0.8935
Iter-88030, train loss-0.4001, acc-0.8400, valid loss-0.3648, acc-0.8972, test loss-0.3700, acc-0.8935
Iter-88040, train loss-0.3731, acc-0.9000, valid loss-0.3647, acc-0.8974, test loss-0.3700, acc-0.8935
Iter-88050, train loss-0.2829, acc-0.8800, valid loss-0.3647, acc-0.8974, test loss-0.3700, acc-0.8936
Iter-88060, train loss-0.4732, acc-0.8800, valid loss-0.3647, acc-0.8976, test loss-0.3700, acc-0.8935
Iter-88070, train loss-0.3268, acc-0.9000, valid loss-0.3647, acc-0.8976, test loss-0.3700, acc-0.8935
Iter-88080, train loss-0.5944, acc-0.8600, valid loss-0.3647, acc-0.8976, test loss-0.3700, acc-0.8935
Iter-88090, train loss-0.3108, acc-0.9200, valid loss-0.3647, acc-0.8976, test loss-0.3699, acc-0.8936
Iter-88100, train loss-0.4888, acc-0.8400, valid loss-0.3647, acc-0.8976, test loss-0.3699, acc-0.8936
Iter-88110, train loss-0.4317, acc-0.8600, valid loss-0.3646, acc-0.8976, test loss-0.3699, acc-0.8937
Iter-88120, train loss-0.2595, acc-0.9600, valid loss-0.3646, acc-0.8976, test loss-0.3699, acc-0.8937
Iter-88130, train loss-0.2025, acc-0.9600, valid loss-0.3646, acc-0.8976, test loss-0.3699, acc-0.8937
Iter-88140, train loss-0.2127, acc-0.9800, valid loss-0.3646, acc-0.8974, test loss-0.3699, acc-0.8937
Iter-88150, train loss-0.3536, acc-0.8800, valid loss-0.3646, acc-0.8974, test loss-0.3699, acc-0.8937
Iter-88160, train loss-0.5052, acc-0.7800, valid loss-0.3646, acc-0.8974, test loss-0.3698, acc-0.8936
Iter-88170, train loss-0.5386, acc-0.8400, valid loss-0.3646, acc-0.8974, test loss-0.3698, acc-0.8936
Iter-88180, train loss-0.3130, acc-0.9200, valid loss-0.3645, acc-0.8974, test loss-0.3698, acc-0.8936
Iter-88190, train loss-0.1810, acc-0.9800, valid loss-0.3645, acc-0.8974, test loss-0.3698, acc-0.8936
Iter-88200, train loss-0.3681, acc-0.8800, valid loss-0.3645, acc-0.8974, test loss-0.3698, acc-0.8936
Iter-88210, train loss-0.7211, acc-0.7600, valid loss-0.3645, acc-0.8974, test loss-0.3698, acc-0.8936
Iter-88220, train loss-0.3935, acc-0.8400, valid loss-0.3645, acc-0.8974, test loss-0.3698, acc-0.8937
Iter-88230, train loss-0.4859, acc-0.9000, valid loss-0.3645, acc-0.8974, test loss-0.3698, acc-0.8937
Iter-88240, train loss-0.4911, acc-0.8600, valid loss-0.3645, acc-0.8974, test loss-0.3697, acc-0.8938
Iter-88250, train loss-0.1934, acc-0.9800, valid loss-0.3645, acc-0.8974, test loss-0.3697, acc-0.8936
Iter-88260, train loss-0.4715, acc-0.8600, valid loss-0.3644, acc-0.8974, test loss-0.3697, acc-0.8938
Iter-88270, train loss-0.2601, acc-0.9400, valid loss-0.3644, acc-0.8974, test loss-0.3697, acc-0.8937
Iter-88280, train loss-0.4200, acc-0.9200, valid loss-0.3644, acc-0.8974, test loss-0.3697, acc-0.8938
Iter-88290, train loss-0.5843, acc-0.8000, valid loss-0.3644, acc-0.8974, test loss-0.3697, acc-0.8938
Iter-88300, train loss-0.5329, acc-0.8200, valid loss-0.3644, acc-0.8974, test loss-0.3696, acc-0.8938
Iter-88310, train loss-0.2135, acc-0.9800, valid loss-0.3644, acc-0.8974, test loss-0.3696, acc-0.8938
Iter-88320, train loss-0.2060, acc-0.9400, valid loss-0.3644, acc-0.8974, test loss-0.3696, acc-0.8938
Iter-88330, train loss-0.2728, acc-0.9400, valid loss-0.3644, acc-0.8976, test loss-0.3696, acc-0.8938
Iter-88340, train loss-0.2565, acc-0.9000, valid loss-0.3644, acc-0.8974, test loss-0.3696, acc-0.8938
Iter-88350, train loss-0.5285, acc-0.8000, valid loss-0.3643, acc-0.8976, test loss-0.3696, acc-0.8938
Iter-88360, train loss-0.3659, acc-0.8800, valid loss-0.3643, acc-0.8978, test loss-0.3696, acc-0.8937
Iter-88370, train loss-0.3639, acc-0.9400, valid loss-0.3643, acc-0.8978, test loss-0.3696, acc-0.8937
Iter-88380, train loss-0.3978, acc-0.8800, valid loss-0.3643, acc-0.8978, test loss-0.3695, acc-0.8937
Iter-88390, train loss-0.5918, acc-0.8800, valid loss-0.3643, acc-0.8976, test loss-0.3695, acc-0.8936
Iter-88400, train loss-0.4785, acc-0.8600, valid loss-0.3643, acc-0.8976, test loss-0.3695, acc-0.8936
Iter-88410, train loss-0.3010, acc-0.9200, valid loss-0.3642, acc-0.8978, test loss-0.3695, acc-0.8937
Iter-88420, train loss-0.3535, acc-0.8400, valid loss-0.3643, acc-0.8978, test loss-0.3695, acc-0.8936
Iter-88430, train loss-0.6133, acc-0.9000, valid loss-0.3642, acc-0.8976, test loss-0.3695, acc-0.8937
Iter-88440, train loss-0.2530, acc-0.9200, valid loss-0.3642, acc-0.8978, test loss-0.3695, acc-0.8936
Iter-88450, train loss-0.2207, acc-0.9400, valid loss-0.3642, acc-0.8978, test loss-0.3694, acc-0.8937
Iter-88460, train loss-0.4732, acc-0.8800, valid loss-0.3642, acc-0.8976, test loss-0.3694, acc-0.8937
Iter-88470, train loss-0.5199, acc-0.8600, valid loss-0.3642, acc-0.8978, test loss-0.3694, acc-0.8937
Iter-88480, train loss-0.2030, acc-0.9400, valid loss-0.3642, acc-0.8976, test loss-0.3694, acc-0.8937
Iter-88490, train loss-0.3927, acc-0.9200, valid loss-0.3641, acc-0.8976, test loss-0.3694, acc-0.8937
Iter-88500, train loss-0.3554, acc-0.9000, valid loss-0.3641, acc-0.8976, test loss-0.3694, acc-0.8938
Iter-88510, train loss-0.4400, acc-0.9000, valid loss-0.3641, acc-0.8978, test loss-0.3694, acc-0.8939
Iter-88520, train loss-0.3632, acc-0.9000, valid loss-0.3641, acc-0.8976, test loss-0.3694, acc-0.8939
Iter-88530, train loss-0.4022, acc-0.8800, valid loss-0.3641, acc-0.8976, test loss-0.3693, acc-0.8938
Iter-88540, train loss-0.5098, acc-0.7800, valid loss-0.3640, acc-0.8978, test loss-0.3693, acc-0.8938
Iter-88550, train loss-0.5460, acc-0.8400, valid loss-0.3640, acc-0.8976, test loss-0.3693, acc-0.8938
Iter-88560, train loss-0.3428, acc-0.9200, valid loss-0.3640, acc-0.8976, test loss-0.3693, acc-0.8938
Iter-88570, train loss-0.3917, acc-0.9000, valid loss-0.3640, acc-0.8976, test loss-0.3693, acc-0.8938
Iter-88580, train loss-0.4414, acc-0.8000, valid loss-0.3640, acc-0.8978, test loss-0.3693, acc-0.8938
Iter-88590, train loss-0.3786, acc-0.8400, valid loss-0.3640, acc-0.8978, test loss-0.3693, acc-0.8939
Iter-88600, train loss-0.5440, acc-0.8200, valid loss-0.3640, acc-0.8974, test loss-0.3693, acc-0.8938
Iter-88610, train loss-0.4410, acc-0.8800, valid loss-0.3639, acc-0.8976, test loss-0.3692, acc-0.8938
Iter-88620, train loss-0.3130, acc-0.9000, valid loss-0.3639, acc-0.8976, test loss-0.3692, acc-0.8939
Iter-88630, train loss-0.5149, acc-0.8800, valid loss-0.3639, acc-0.8976, test loss-0.3692, acc-0.8938
Iter-88640, train loss-0.5820, acc-0.8400, valid loss-0.3639, acc-0.8978, test loss-0.3692, acc-0.8938
Iter-88650, train loss-0.3242, acc-0.9400, valid loss-0.3639, acc-0.8976, test loss-0.3692, acc-0.8939
Iter-88660, train loss-0.3981, acc-0.8600, valid loss-0.3639, acc-0.8976, test loss-0.3692, acc-0.8939
Iter-88670, train loss-0.3982, acc-0.8800, valid loss-0.3639, acc-0.8976, test loss-0.3692, acc-0.8940
Iter-88680, train loss-0.2441, acc-0.9800, valid loss-0.3638, acc-0.8976, test loss-0.3692, acc-0.8941
Iter-88690, train loss-0.1928, acc-0.9600, valid loss-0.3638, acc-0.8974, test loss-0.3691, acc-0.8941
Iter-88700, train loss-0.3528, acc-0.9000, valid loss-0.3638, acc-0.8974, test loss-0.3691, acc-0.8941
Iter-88710, train loss-0.5123, acc-0.8800, valid loss-0.3638, acc-0.8974, test loss-0.3691, acc-0.8941
Iter-88720, train loss-0.4560, acc-0.8800, valid loss-0.3638, acc-0.8976, test loss-0.3691, acc-0.8941
Iter-88730, train loss-0.3828, acc-0.8800, valid loss-0.3638, acc-0.8974, test loss-0.3691, acc-0.8941
Iter-88740, train loss-0.4077, acc-0.8800, valid loss-0.3638, acc-0.8978, test loss-0.3691, acc-0.8941
Iter-88750, train loss-0.3583, acc-0.8800, valid loss-0.3637, acc-0.8976, test loss-0.3691, acc-0.8941
Iter-88760, train loss-0.5391, acc-0.8800, valid loss-0.3637, acc-0.8978, test loss-0.3690, acc-0.8942
Iter-88770, train loss-0.2868, acc-0.9000, valid loss-0.3637, acc-0.8976, test loss-0.3690, acc-0.8942
Iter-88780, train loss-0.5256, acc-0.8200, valid loss-0.3637, acc-0.8978, test loss-0.3690, acc-0.8941
Iter-88790, train loss-0.2962, acc-0.9200, valid loss-0.3637, acc-0.8976, test loss-0.3690, acc-0.8941
Iter-88800, train loss-0.4583, acc-0.9200, valid loss-0.3637, acc-0.8978, test loss-0.3690, acc-0.8942
Iter-88810, train loss-0.3314, acc-0.8800, valid loss-0.3637, acc-0.8976, test loss-0.3690, acc-0.8941
Iter-88820, train loss-0.3215, acc-0.9400, valid loss-0.3637, acc-0.8976, test loss-0.3690, acc-0.8941
Iter-88830, train loss-0.2679, acc-0.9200, valid loss-0.3636, acc-0.8976, test loss-0.3690, acc-0.8942
Iter-88840, train loss-0.4450, acc-0.8400, valid loss-0.3636, acc-0.8976, test loss-0.3690, acc-0.8942
Iter-88850, train loss-0.4239, acc-0.8600, valid loss-0.3636, acc-0.8976, test loss-0.3689, acc-0.8942
Iter-88860, train loss-0.4688, acc-0.8400, valid loss-0.3636, acc-0.8976, test loss-0.3689, acc-0.8942
Iter-88870, train loss-0.2925, acc-0.9000, valid loss-0.3636, acc-0.8976, test loss-0.3689, acc-0.8941
Iter-88880, train loss-0.2470, acc-0.9400, valid loss-0.3636, acc-0.8976, test loss-0.3689, acc-0.8941
Iter-88890, train loss-0.3364, acc-0.9200, valid loss-0.3636, acc-0.8976, test loss-0.3689, acc-0.8941
Iter-88900, train loss-0.4202, acc-0.8400, valid loss-0.3636, acc-0.8972, test loss-0.3689, acc-0.8941
Iter-88910, train loss-0.4940, acc-0.8600, valid loss-0.3635, acc-0.8972, test loss-0.3689, acc-0.8940
Iter-88920, train loss-0.3763, acc-0.8800, valid loss-0.3635, acc-0.8970, test loss-0.3689, acc-0.8941
Iter-88930, train loss-0.3598, acc-0.9200, valid loss-0.3635, acc-0.8968, test loss-0.3689, acc-0.8941
Iter-88940, train loss-0.4957, acc-0.8800, valid loss-0.3635, acc-0.8970, test loss-0.3688, acc-0.8941
Iter-88950, train loss-0.4680, acc-0.8400, valid loss-0.3635, acc-0.8970, test loss-0.3688, acc-0.8941
Iter-88960, train loss-0.2548, acc-0.9400, valid loss-0.3635, acc-0.8970, test loss-0.3688, acc-0.8941
Iter-88970, train loss-0.2233, acc-0.9400, valid loss-0.3634, acc-0.8970, test loss-0.3688, acc-0.8941
Iter-88980, train loss-0.3374, acc-0.8800, valid loss-0.3634, acc-0.8970, test loss-0.3688, acc-0.8940
Iter-88990, train loss-0.1784, acc-0.9400, valid loss-0.3634, acc-0.8970, test loss-0.3688, acc-0.8940
Iter-89000, train loss-0.2460, acc-0.9200, valid loss-0.3634, acc-0.8974, test loss-0.3688, acc-0.8941
Iter-89010, train loss-0.4154, acc-0.8800, valid loss-0.3634, acc-0.8974, test loss-0.3688, acc-0.8941
Iter-89020, train loss-0.6016, acc-0.8000, valid loss-0.3633, acc-0.8972, test loss-0.3688, acc-0.8941
Iter-89030, train loss-0.3533, acc-0.8800, valid loss-0.3633, acc-0.8974, test loss-0.3687, acc-0.8942
Iter-89040, train loss-0.2966, acc-0.9000, valid loss-0.3633, acc-0.8974, test loss-0.3687, acc-0.8943
Iter-89050, train loss-0.2989, acc-0.9400, valid loss-0.3633, acc-0.8972, test loss-0.3687, acc-0.8944
Iter-89060, train loss-0.4548, acc-0.9000, valid loss-0.3633, acc-0.8972, test loss-0.3687, acc-0.8942
Iter-89070, train loss-0.3107, acc-0.9200, valid loss-0.3632, acc-0.8976, test loss-0.3687, acc-0.8942
Iter-89080, train loss-0.4987, acc-0.8600, valid loss-0.3632, acc-0.8980, test loss-0.3687, acc-0.8942
Iter-89090, train loss-0.3573, acc-0.8400, valid loss-0.3632, acc-0.8980, test loss-0.3686, acc-0.8942
Iter-89100, train loss-0.2516, acc-0.9200, valid loss-0.3632, acc-0.8980, test loss-0.3686, acc-0.8943
Iter-89110, train loss-0.3458, acc-0.9200, valid loss-0.3632, acc-0.8978, test loss-0.3686, acc-0.8943
Iter-89120, train loss-0.2829, acc-0.9200, valid loss-0.3632, acc-0.8982, test loss-0.3686, acc-0.8943
Iter-89130, train loss-0.4368, acc-0.8400, valid loss-0.3632, acc-0.8976, test loss-0.3686, acc-0.8943
Iter-89140, train loss-0.2963, acc-0.9600, valid loss-0.3631, acc-0.8978, test loss-0.3686, acc-0.8942
Iter-89150, train loss-0.4986, acc-0.8200, valid loss-0.3631, acc-0.8978, test loss-0.3686, acc-0.8942
Iter-89160, train loss-0.2586, acc-0.9200, valid loss-0.3631, acc-0.8978, test loss-0.3685, acc-0.8942
Iter-89170, train loss-0.2809, acc-0.9400, valid loss-0.3631, acc-0.8980, test loss-0.3685, acc-0.8942
Iter-89180, train loss-0.4679, acc-0.8600, valid loss-0.3631, acc-0.8980, test loss-0.3685, acc-0.8942
Iter-89190, train loss-0.3488, acc-0.8600, valid loss-0.3631, acc-0.8978, test loss-0.3685, acc-0.8942
Iter-89200, train loss-0.3929, acc-0.8800, valid loss-0.3631, acc-0.8974, test loss-0.3685, acc-0.8943
Iter-89210, train loss-0.4152, acc-0.9000, valid loss-0.3630, acc-0.8972, test loss-0.3685, acc-0.8943
Iter-89220, train loss-0.3209, acc-0.9400, valid loss-0.3630, acc-0.8974, test loss-0.3685, acc-0.8943
Iter-89230, train loss-0.3659, acc-0.9200, valid loss-0.3630, acc-0.8974, test loss-0.3685, acc-0.8942
Iter-89240, train loss-0.4693, acc-0.8000, valid loss-0.3630, acc-0.8974, test loss-0.3685, acc-0.8943
Iter-89250, train loss-0.5774, acc-0.8600, valid loss-0.3630, acc-0.8976, test loss-0.3685, acc-0.8943
Iter-89260, train loss-0.2405, acc-0.9200, valid loss-0.3630, acc-0.8974, test loss-0.3684, acc-0.8942
Iter-89270, train loss-0.4224, acc-0.8600, valid loss-0.3630, acc-0.8972, test loss-0.3684, acc-0.8942
Iter-89280, train loss-0.3983, acc-0.8600, valid loss-0.3629, acc-0.8970, test loss-0.3684, acc-0.8942
Iter-89290, train loss-0.5052, acc-0.9000, valid loss-0.3629, acc-0.8970, test loss-0.3684, acc-0.8942
Iter-89300, train loss-0.1885, acc-0.9600, valid loss-0.3629, acc-0.8972, test loss-0.3684, acc-0.8942
Iter-89310, train loss-0.3244, acc-0.9200, valid loss-0.3629, acc-0.8972, test loss-0.3684, acc-0.8943
Iter-89320, train loss-0.3205, acc-0.8600, valid loss-0.3629, acc-0.8974, test loss-0.3684, acc-0.8943
Iter-89330, train loss-0.2861, acc-0.9000, valid loss-0.3629, acc-0.8976, test loss-0.3684, acc-0.8943
Iter-89340, train loss-0.5290, acc-0.8400, valid loss-0.3629, acc-0.8976, test loss-0.3684, acc-0.8942
Iter-89350, train loss-0.4507, acc-0.8000, valid loss-0.3629, acc-0.8976, test loss-0.3683, acc-0.8942
Iter-89360, train loss-0.2488, acc-0.9400, valid loss-0.3629, acc-0.8978, test loss-0.3683, acc-0.8941
Iter-89370, train loss-0.3159, acc-0.9000, valid loss-0.3628, acc-0.8976, test loss-0.3683, acc-0.8942
Iter-89380, train loss-0.4963, acc-0.8400, valid loss-0.3628, acc-0.8976, test loss-0.3683, acc-0.8942
Iter-89390, train loss-0.2479, acc-0.9400, valid loss-0.3628, acc-0.8976, test loss-0.3683, acc-0.8943
Iter-89400, train loss-0.4218, acc-0.8800, valid loss-0.3628, acc-0.8976, test loss-0.3683, acc-0.8943
Iter-89410, train loss-0.4965, acc-0.8400, valid loss-0.3628, acc-0.8976, test loss-0.3683, acc-0.8941
Iter-89420, train loss-0.3116, acc-0.9000, valid loss-0.3628, acc-0.8974, test loss-0.3683, acc-0.8942
Iter-89430, train loss-0.5532, acc-0.8800, valid loss-0.3628, acc-0.8972, test loss-0.3682, acc-0.8942
Iter-89440, train loss-0.4250, acc-0.8600, valid loss-0.3627, acc-0.8976, test loss-0.3682, acc-0.8942
Iter-89450, train loss-0.3845, acc-0.9000, valid loss-0.3627, acc-0.8972, test loss-0.3682, acc-0.8942
Iter-89460, train loss-0.2511, acc-0.9400, valid loss-0.3627, acc-0.8972, test loss-0.3682, acc-0.8942
Iter-89470, train loss-0.3388, acc-0.9000, valid loss-0.3627, acc-0.8972, test loss-0.3682, acc-0.8942
Iter-89480, train loss-0.5366, acc-0.8400, valid loss-0.3627, acc-0.8972, test loss-0.3682, acc-0.8942
Iter-89490, train loss-0.3705, acc-0.9000, valid loss-0.3627, acc-0.8972, test loss-0.3682, acc-0.8942
Iter-89500, train loss-0.4330, acc-0.8600, valid loss-0.3626, acc-0.8972, test loss-0.3682, acc-0.8942
Iter-89510, train loss-0.4426, acc-0.9000, valid loss-0.3626, acc-0.8972, test loss-0.3682, acc-0.8943
Iter-89520, train loss-0.6120, acc-0.8800, valid loss-0.3626, acc-0.8972, test loss-0.3681, acc-0.8943
Iter-89530, train loss-0.2815, acc-0.9400, valid loss-0.3626, acc-0.8972, test loss-0.3681, acc-0.8944
Iter-89540, train loss-0.4745, acc-0.8800, valid loss-0.3626, acc-0.8972, test loss-0.3681, acc-0.8943
Iter-89550, train loss-0.3820, acc-0.9200, valid loss-0.3626, acc-0.8972, test loss-0.3681, acc-0.8943
Iter-89560, train loss-0.3022, acc-0.9200, valid loss-0.3626, acc-0.8972, test loss-0.3681, acc-0.8943
Iter-89570, train loss-0.3252, acc-0.9400, valid loss-0.3625, acc-0.8972, test loss-0.3681, acc-0.8943
Iter-89580, train loss-0.3287, acc-0.9400, valid loss-0.3625, acc-0.8972, test loss-0.3681, acc-0.8944
Iter-89590, train loss-0.3780, acc-0.9200, valid loss-0.3625, acc-0.8974, test loss-0.3681, acc-0.8944
Iter-89600, train loss-0.4686, acc-0.8400, valid loss-0.3625, acc-0.8972, test loss-0.3680, acc-0.8944
Iter-89610, train loss-0.4004, acc-0.8800, valid loss-0.3625, acc-0.8972, test loss-0.3680, acc-0.8944
Iter-89620, train loss-0.5160, acc-0.8600, valid loss-0.3624, acc-0.8972, test loss-0.3680, acc-0.8943
Iter-89630, train loss-0.5364, acc-0.8400, valid loss-0.3624, acc-0.8972, test loss-0.3680, acc-0.8944
Iter-89640, train loss-0.3355, acc-0.9200, valid loss-0.3624, acc-0.8972, test loss-0.3679, acc-0.8942
Iter-89650, train loss-0.3479, acc-0.8600, valid loss-0.3624, acc-0.8972, test loss-0.3679, acc-0.8943
Iter-89660, train loss-0.1502, acc-0.9600, valid loss-0.3624, acc-0.8972, test loss-0.3679, acc-0.8943
Iter-89670, train loss-0.3784, acc-0.8800, valid loss-0.3624, acc-0.8970, test loss-0.3679, acc-0.8943
Iter-89680, train loss-0.3428, acc-0.8600, valid loss-0.3623, acc-0.8970, test loss-0.3679, acc-0.8943
Iter-89690, train loss-0.4654, acc-0.8200, valid loss-0.3623, acc-0.8972, test loss-0.3679, acc-0.8943
Iter-89700, train loss-0.4348, acc-0.8800, valid loss-0.3623, acc-0.8972, test loss-0.3679, acc-0.8941
Iter-89710, train loss-0.5920, acc-0.9000, valid loss-0.3623, acc-0.8974, test loss-0.3679, acc-0.8941
Iter-89720, train loss-0.3336, acc-0.9200, valid loss-0.3623, acc-0.8974, test loss-0.3679, acc-0.8941
Iter-89730, train loss-0.4043, acc-0.9000, valid loss-0.3623, acc-0.8974, test loss-0.3679, acc-0.8941
Iter-89740, train loss-0.5525, acc-0.8000, valid loss-0.3623, acc-0.8974, test loss-0.3679, acc-0.8941
Iter-89750, train loss-0.5896, acc-0.8200, valid loss-0.3622, acc-0.8970, test loss-0.3678, acc-0.8941
Iter-89760, train loss-0.4448, acc-0.8600, valid loss-0.3622, acc-0.8970, test loss-0.3678, acc-0.8942
Iter-89770, train loss-0.4034, acc-0.9000, valid loss-0.3622, acc-0.8970, test loss-0.3678, acc-0.8943
Iter-89780, train loss-0.4334, acc-0.9200, valid loss-0.3622, acc-0.8970, test loss-0.3678, acc-0.8943
Iter-89790, train loss-0.4039, acc-0.8800, valid loss-0.3622, acc-0.8970, test loss-0.3678, acc-0.8943
Iter-89800, train loss-0.3364, acc-0.9000, valid loss-0.3622, acc-0.8970, test loss-0.3678, acc-0.8943
Iter-89810, train loss-0.6099, acc-0.8000, valid loss-0.3622, acc-0.8970, test loss-0.3677, acc-0.8941
Iter-89820, train loss-0.5105, acc-0.8400, valid loss-0.3622, acc-0.8970, test loss-0.3677, acc-0.8941
Iter-89830, train loss-0.4950, acc-0.8400, valid loss-0.3621, acc-0.8968, test loss-0.3677, acc-0.8942
Iter-89840, train loss-0.3602, acc-0.8800, valid loss-0.3621, acc-0.8968, test loss-0.3677, acc-0.8943
Iter-89850, train loss-0.3323, acc-0.9000, valid loss-0.3621, acc-0.8972, test loss-0.3677, acc-0.8941
Iter-89860, train loss-0.2402, acc-0.9000, valid loss-0.3621, acc-0.8972, test loss-0.3677, acc-0.8943
Iter-89870, train loss-0.4734, acc-0.8400, valid loss-0.3621, acc-0.8970, test loss-0.3677, acc-0.8942
Iter-89880, train loss-0.3777, acc-0.8400, valid loss-0.3621, acc-0.8972, test loss-0.3677, acc-0.8942
Iter-89890, train loss-0.4246, acc-0.8600, valid loss-0.3621, acc-0.8970, test loss-0.3676, acc-0.8942
Iter-89900, train loss-0.4169, acc-0.9000, valid loss-0.3621, acc-0.8970, test loss-0.3676, acc-0.8942
Iter-89910, train loss-0.4385, acc-0.8800, valid loss-0.3621, acc-0.8970, test loss-0.3676, acc-0.8940
Iter-89920, train loss-0.4273, acc-0.8600, valid loss-0.3620, acc-0.8972, test loss-0.3676, acc-0.8941
Iter-89930, train loss-0.4178, acc-0.9000, valid loss-0.3620, acc-0.8972, test loss-0.3676, acc-0.8940
Iter-89940, train loss-0.2862, acc-0.9200, valid loss-0.3620, acc-0.8974, test loss-0.3676, acc-0.8942
Iter-89950, train loss-0.3586, acc-0.9200, valid loss-0.3620, acc-0.8970, test loss-0.3676, acc-0.8942
Iter-89960, train loss-0.6079, acc-0.7600, valid loss-0.3620, acc-0.8972, test loss-0.3675, acc-0.8942
Iter-89970, train loss-0.2934, acc-0.9000, valid loss-0.3620, acc-0.8974, test loss-0.3675, acc-0.8942
Iter-89980, train loss-0.3546, acc-0.9400, valid loss-0.3620, acc-0.8976, test loss-0.3675, acc-0.8942
Iter-89990, train loss-0.2749, acc-0.9200, valid loss-0.3619, acc-0.8976, test loss-0.3675, acc-0.8942
Iter-90000, train loss-0.4547, acc-0.8400, valid loss-0.3619, acc-0.8976, test loss-0.3675, acc-0.8942
Iter-90010, train loss-0.3996, acc-0.8600, valid loss-0.3619, acc-0.8976, test loss-0.3675, acc-0.8941
Iter-90020, train loss-0.3616, acc-0.9000, valid loss-0.3619, acc-0.8974, test loss-0.3674, acc-0.8942
Iter-90030, train loss-0.4525, acc-0.8800, valid loss-0.3619, acc-0.8974, test loss-0.3674, acc-0.8943
Iter-90040, train loss-0.3866, acc-0.8800, valid loss-0.3619, acc-0.8974, test loss-0.3674, acc-0.8944
Iter-90050, train loss-0.2919, acc-0.9000, valid loss-0.3619, acc-0.8974, test loss-0.3674, acc-0.8943
Iter-90060, train loss-0.5963, acc-0.8600, valid loss-0.3619, acc-0.8976, test loss-0.3674, acc-0.8943
Iter-90070, train loss-0.4255, acc-0.9000, valid loss-0.3619, acc-0.8976, test loss-0.3674, acc-0.8943
Iter-90080, train loss-0.2311, acc-0.9200, valid loss-0.3618, acc-0.8976, test loss-0.3673, acc-0.8943
Iter-90090, train loss-0.5716, acc-0.9200, valid loss-0.3618, acc-0.8976, test loss-0.3673, acc-0.8942
Iter-90100, train loss-0.3032, acc-0.9000, valid loss-0.3618, acc-0.8976, test loss-0.3673, acc-0.8942
Iter-90110, train loss-0.2907, acc-0.9400, valid loss-0.3618, acc-0.8976, test loss-0.3673, acc-0.8942
Iter-90120, train loss-0.4430, acc-0.8600, valid loss-0.3618, acc-0.8976, test loss-0.3673, acc-0.8943
Iter-90130, train loss-0.4369, acc-0.8800, valid loss-0.3618, acc-0.8976, test loss-0.3673, acc-0.8942
Iter-90140, train loss-0.5091, acc-0.7800, valid loss-0.3618, acc-0.8976, test loss-0.3673, acc-0.8942
Iter-90150, train loss-0.3474, acc-0.9000, valid loss-0.3618, acc-0.8974, test loss-0.3673, acc-0.8943
Iter-90160, train loss-0.3538, acc-0.9200, valid loss-0.3617, acc-0.8974, test loss-0.3672, acc-0.8942
Iter-90170, train loss-0.3851, acc-0.8600, valid loss-0.3617, acc-0.8974, test loss-0.3672, acc-0.8943
Iter-90180, train loss-0.3394, acc-0.8800, valid loss-0.3617, acc-0.8976, test loss-0.3672, acc-0.8943
Iter-90190, train loss-0.2795, acc-0.9200, valid loss-0.3617, acc-0.8976, test loss-0.3672, acc-0.8943
Iter-90200, train loss-0.3329, acc-0.9000, valid loss-0.3617, acc-0.8976, test loss-0.3672, acc-0.8943
Iter-90210, train loss-0.3847, acc-0.8600, valid loss-0.3616, acc-0.8976, test loss-0.3672, acc-0.8942
Iter-90220, train loss-0.3153, acc-0.9400, valid loss-0.3616, acc-0.8976, test loss-0.3672, acc-0.8941
Iter-90230, train loss-0.5597, acc-0.8200, valid loss-0.3616, acc-0.8978, test loss-0.3671, acc-0.8941
Iter-90240, train loss-0.3981, acc-0.8600, valid loss-0.3616, acc-0.8978, test loss-0.3671, acc-0.8940
Iter-90250, train loss-0.2925, acc-0.9400, valid loss-0.3616, acc-0.8978, test loss-0.3671, acc-0.8940
Iter-90260, train loss-0.4842, acc-0.8800, valid loss-0.3616, acc-0.8980, test loss-0.3671, acc-0.8940
Iter-90270, train loss-0.4988, acc-0.9000, valid loss-0.3616, acc-0.8980, test loss-0.3671, acc-0.8938
Iter-90280, train loss-0.4144, acc-0.9000, valid loss-0.3615, acc-0.8978, test loss-0.3671, acc-0.8938
Iter-90290, train loss-0.3969, acc-0.8600, valid loss-0.3615, acc-0.8978, test loss-0.3671, acc-0.8941
Iter-90300, train loss-0.2472, acc-0.9400, valid loss-0.3615, acc-0.8978, test loss-0.3671, acc-0.8942
Iter-90310, train loss-0.3824, acc-0.8800, valid loss-0.3615, acc-0.8976, test loss-0.3670, acc-0.8942
Iter-90320, train loss-0.2164, acc-0.9400, valid loss-0.3615, acc-0.8976, test loss-0.3670, acc-0.8943
Iter-90330, train loss-0.2197, acc-0.9400, valid loss-0.3615, acc-0.8976, test loss-0.3670, acc-0.8941
Iter-90340, train loss-0.4620, acc-0.8600, valid loss-0.3615, acc-0.8976, test loss-0.3670, acc-0.8942
Iter-90350, train loss-0.4170, acc-0.8800, valid loss-0.3615, acc-0.8976, test loss-0.3670, acc-0.8942
Iter-90360, train loss-0.5694, acc-0.8800, valid loss-0.3614, acc-0.8976, test loss-0.3670, acc-0.8941
Iter-90370, train loss-0.3521, acc-0.9000, valid loss-0.3614, acc-0.8974, test loss-0.3670, acc-0.8941
Iter-90380, train loss-0.5457, acc-0.9000, valid loss-0.3614, acc-0.8972, test loss-0.3670, acc-0.8940
Iter-90390, train loss-0.3206, acc-0.9000, valid loss-0.3614, acc-0.8972, test loss-0.3670, acc-0.8941
Iter-90400, train loss-0.1903, acc-0.9800, valid loss-0.3614, acc-0.8974, test loss-0.3669, acc-0.8938
Iter-90410, train loss-0.2858, acc-0.9000, valid loss-0.3614, acc-0.8976, test loss-0.3669, acc-0.8939
Iter-90420, train loss-0.2944, acc-0.9000, valid loss-0.3614, acc-0.8974, test loss-0.3669, acc-0.8939
Iter-90430, train loss-0.3882, acc-0.9000, valid loss-0.3614, acc-0.8974, test loss-0.3669, acc-0.8938
Iter-90440, train loss-0.4507, acc-0.8800, valid loss-0.3613, acc-0.8976, test loss-0.3669, acc-0.8939
Iter-90450, train loss-0.2822, acc-0.9600, valid loss-0.3613, acc-0.8976, test loss-0.3668, acc-0.8942
Iter-90460, train loss-0.4139, acc-0.9000, valid loss-0.3613, acc-0.8976, test loss-0.3668, acc-0.8941
Iter-90470, train loss-0.3255, acc-0.9200, valid loss-0.3613, acc-0.8976, test loss-0.3668, acc-0.8941
Iter-90480, train loss-0.2966, acc-0.9200, valid loss-0.3613, acc-0.8976, test loss-0.3668, acc-0.8942
Iter-90490, train loss-0.6330, acc-0.8000, valid loss-0.3613, acc-0.8978, test loss-0.3668, acc-0.8942
Iter-90500, train loss-0.3855, acc-0.8800, valid loss-0.3613, acc-0.8978, test loss-0.3668, acc-0.8942
Iter-90510, train loss-0.4527, acc-0.8400, valid loss-0.3612, acc-0.8976, test loss-0.3668, acc-0.8941
Iter-90520, train loss-0.3121, acc-0.9400, valid loss-0.3612, acc-0.8978, test loss-0.3668, acc-0.8941
Iter-90530, train loss-0.2715, acc-0.9000, valid loss-0.3612, acc-0.8980, test loss-0.3668, acc-0.8941
Iter-90540, train loss-0.5032, acc-0.8600, valid loss-0.3612, acc-0.8980, test loss-0.3667, acc-0.8942
Iter-90550, train loss-0.4590, acc-0.8800, valid loss-0.3612, acc-0.8980, test loss-0.3667, acc-0.8942
Iter-90560, train loss-0.4978, acc-0.9000, valid loss-0.3612, acc-0.8980, test loss-0.3667, acc-0.8943
Iter-90570, train loss-0.7095, acc-0.7600, valid loss-0.3612, acc-0.8980, test loss-0.3667, acc-0.8943
Iter-90580, train loss-0.5573, acc-0.8400, valid loss-0.3611, acc-0.8980, test loss-0.3667, acc-0.8943
Iter-90590, train loss-0.4393, acc-0.8800, valid loss-0.3611, acc-0.8980, test loss-0.3667, acc-0.8943
Iter-90600, train loss-0.2692, acc-0.9400, valid loss-0.3611, acc-0.8982, test loss-0.3667, acc-0.8942
Iter-90610, train loss-0.3090, acc-0.9200, valid loss-0.3611, acc-0.8984, test loss-0.3666, acc-0.8942
Iter-90620, train loss-0.5939, acc-0.9000, valid loss-0.3611, acc-0.8984, test loss-0.3666, acc-0.8942
Iter-90630, train loss-0.1529, acc-0.9600, valid loss-0.3611, acc-0.8984, test loss-0.3666, acc-0.8942
Iter-90640, train loss-0.3775, acc-0.9000, valid loss-0.3611, acc-0.8984, test loss-0.3666, acc-0.8942
Iter-90650, train loss-0.2744, acc-0.9200, valid loss-0.3611, acc-0.8982, test loss-0.3666, acc-0.8942
Iter-90660, train loss-0.2328, acc-0.9400, valid loss-0.3611, acc-0.8980, test loss-0.3666, acc-0.8942
Iter-90670, train loss-0.4090, acc-0.8400, valid loss-0.3610, acc-0.8984, test loss-0.3666, acc-0.8942
Iter-90680, train loss-0.2036, acc-0.9800, valid loss-0.3610, acc-0.8982, test loss-0.3665, acc-0.8942
Iter-90690, train loss-0.7760, acc-0.7400, valid loss-0.3610, acc-0.8980, test loss-0.3665, acc-0.8942
Iter-90700, train loss-0.3675, acc-0.8800, valid loss-0.3610, acc-0.8980, test loss-0.3665, acc-0.8943
Iter-90710, train loss-0.2534, acc-0.9400, valid loss-0.3610, acc-0.8980, test loss-0.3665, acc-0.8942
Iter-90720, train loss-0.3206, acc-0.9600, valid loss-0.3610, acc-0.8980, test loss-0.3665, acc-0.8942
Iter-90730, train loss-0.3564, acc-0.9400, valid loss-0.3610, acc-0.8980, test loss-0.3665, acc-0.8942
Iter-90740, train loss-0.3465, acc-0.9200, valid loss-0.3610, acc-0.8980, test loss-0.3665, acc-0.8942
Iter-90750, train loss-0.2139, acc-0.9600, valid loss-0.3610, acc-0.8980, test loss-0.3665, acc-0.8942
Iter-90760, train loss-0.3426, acc-0.9200, valid loss-0.3609, acc-0.8980, test loss-0.3665, acc-0.8942
Iter-90770, train loss-0.3488, acc-0.8400, valid loss-0.3609, acc-0.8976, test loss-0.3664, acc-0.8942
Iter-90780, train loss-0.3417, acc-0.9400, valid loss-0.3609, acc-0.8976, test loss-0.3664, acc-0.8943
Iter-90790, train loss-0.4537, acc-0.8800, valid loss-0.3609, acc-0.8976, test loss-0.3664, acc-0.8943
Iter-90800, train loss-0.4052, acc-0.8800, valid loss-0.3609, acc-0.8976, test loss-0.3664, acc-0.8943
Iter-90810, train loss-0.3152, acc-0.9200, valid loss-0.3609, acc-0.8976, test loss-0.3664, acc-0.8942
Iter-90820, train loss-0.4726, acc-0.8800, valid loss-0.3609, acc-0.8978, test loss-0.3664, acc-0.8941
Iter-90830, train loss-0.2137, acc-0.9600, valid loss-0.3608, acc-0.8976, test loss-0.3664, acc-0.8942
Iter-90840, train loss-0.4204, acc-0.8800, valid loss-0.3608, acc-0.8978, test loss-0.3663, acc-0.8942
Iter-90850, train loss-0.3230, acc-0.9000, valid loss-0.3608, acc-0.8976, test loss-0.3663, acc-0.8941
Iter-90860, train loss-0.3815, acc-0.8800, valid loss-0.3608, acc-0.8978, test loss-0.3663, acc-0.8942
Iter-90870, train loss-0.4789, acc-0.8400, valid loss-0.3608, acc-0.8978, test loss-0.3663, acc-0.8943
Iter-90880, train loss-0.3616, acc-0.9200, valid loss-0.3608, acc-0.8976, test loss-0.3663, acc-0.8943
Iter-90890, train loss-0.3547, acc-0.8600, valid loss-0.3607, acc-0.8976, test loss-0.3663, acc-0.8943
Iter-90900, train loss-0.3981, acc-0.9000, valid loss-0.3607, acc-0.8976, test loss-0.3663, acc-0.8944
Iter-90910, train loss-0.4780, acc-0.8800, valid loss-0.3607, acc-0.8976, test loss-0.3663, acc-0.8944
Iter-90920, train loss-0.5246, acc-0.8600, valid loss-0.3607, acc-0.8978, test loss-0.3663, acc-0.8944
Iter-90930, train loss-0.3985, acc-0.9000, valid loss-0.3607, acc-0.8978, test loss-0.3663, acc-0.8943
Iter-90940, train loss-0.3372, acc-0.8800, valid loss-0.3607, acc-0.8978, test loss-0.3663, acc-0.8943
Iter-90950, train loss-0.2906, acc-0.9000, valid loss-0.3607, acc-0.8978, test loss-0.3662, acc-0.8942
Iter-90960, train loss-0.2298, acc-0.9600, valid loss-0.3607, acc-0.8976, test loss-0.3662, acc-0.8942
Iter-90970, train loss-0.4207, acc-0.8800, valid loss-0.3607, acc-0.8976, test loss-0.3662, acc-0.8942
Iter-90980, train loss-0.6580, acc-0.8200, valid loss-0.3607, acc-0.8978, test loss-0.3662, acc-0.8942
Iter-90990, train loss-0.2047, acc-0.9200, valid loss-0.3607, acc-0.8978, test loss-0.3662, acc-0.8942
Iter-91000, train loss-0.3252, acc-0.9000, valid loss-0.3607, acc-0.8982, test loss-0.3662, acc-0.8942
Iter-91010, train loss-0.4404, acc-0.8400, valid loss-0.3606, acc-0.8980, test loss-0.3662, acc-0.8942
Iter-91020, train loss-0.2405, acc-0.9400, valid loss-0.3606, acc-0.8978, test loss-0.3662, acc-0.8942
Iter-91030, train loss-0.2919, acc-0.9000, valid loss-0.3606, acc-0.8980, test loss-0.3661, acc-0.8942
Iter-91040, train loss-0.5399, acc-0.8000, valid loss-0.3606, acc-0.8976, test loss-0.3661, acc-0.8942
Iter-91050, train loss-0.3586, acc-0.9400, valid loss-0.3606, acc-0.8978, test loss-0.3661, acc-0.8942
Iter-91060, train loss-0.3870, acc-0.9400, valid loss-0.3606, acc-0.8976, test loss-0.3661, acc-0.8943
Iter-91070, train loss-0.3079, acc-0.9400, valid loss-0.3606, acc-0.8976, test loss-0.3661, acc-0.8942
Iter-91080, train loss-0.5828, acc-0.8800, valid loss-0.3606, acc-0.8976, test loss-0.3661, acc-0.8941
Iter-91090, train loss-0.3686, acc-0.9400, valid loss-0.3606, acc-0.8976, test loss-0.3661, acc-0.8941
Iter-91100, train loss-0.4421, acc-0.8200, valid loss-0.3605, acc-0.8978, test loss-0.3661, acc-0.8941
Iter-91110, train loss-0.3504, acc-0.9400, valid loss-0.3605, acc-0.8978, test loss-0.3661, acc-0.8942
Iter-91120, train loss-0.2901, acc-0.9600, valid loss-0.3605, acc-0.8978, test loss-0.3661, acc-0.8942
Iter-91130, train loss-0.2135, acc-0.9000, valid loss-0.3605, acc-0.8978, test loss-0.3661, acc-0.8942
Iter-91140, train loss-0.2863, acc-0.9600, valid loss-0.3604, acc-0.8978, test loss-0.3660, acc-0.8942
Iter-91150, train loss-0.4341, acc-0.8600, valid loss-0.3604, acc-0.8978, test loss-0.3660, acc-0.8943
Iter-91160, train loss-0.3593, acc-0.9400, valid loss-0.3604, acc-0.8978, test loss-0.3660, acc-0.8941
Iter-91170, train loss-0.3533, acc-0.9200, valid loss-0.3604, acc-0.8978, test loss-0.3660, acc-0.8942
Iter-91180, train loss-0.2787, acc-0.9000, valid loss-0.3604, acc-0.8978, test loss-0.3660, acc-0.8942
Iter-91190, train loss-0.6335, acc-0.8400, valid loss-0.3604, acc-0.8978, test loss-0.3659, acc-0.8942
Iter-91200, train loss-0.1586, acc-0.9800, valid loss-0.3604, acc-0.8978, test loss-0.3659, acc-0.8944
Iter-91210, train loss-0.3430, acc-0.9000, valid loss-0.3603, acc-0.8978, test loss-0.3659, acc-0.8944
Iter-91220, train loss-0.4118, acc-0.9000, valid loss-0.3603, acc-0.8978, test loss-0.3659, acc-0.8944
Iter-91230, train loss-0.2982, acc-0.9000, valid loss-0.3603, acc-0.8978, test loss-0.3659, acc-0.8943
Iter-91240, train loss-0.3555, acc-0.9200, valid loss-0.3603, acc-0.8978, test loss-0.3659, acc-0.8944
Iter-91250, train loss-0.7149, acc-0.8000, valid loss-0.3603, acc-0.8978, test loss-0.3659, acc-0.8943
Iter-91260, train loss-0.3709, acc-0.8800, valid loss-0.3602, acc-0.8978, test loss-0.3658, acc-0.8946
Iter-91270, train loss-0.6612, acc-0.8200, valid loss-0.3602, acc-0.8980, test loss-0.3658, acc-0.8946
Iter-91280, train loss-0.3554, acc-0.9000, valid loss-0.3602, acc-0.8980, test loss-0.3658, acc-0.8947
Iter-91290, train loss-0.3338, acc-0.8800, valid loss-0.3602, acc-0.8978, test loss-0.3658, acc-0.8945
Iter-91300, train loss-0.1415, acc-0.9800, valid loss-0.3602, acc-0.8978, test loss-0.3658, acc-0.8945
Iter-91310, train loss-0.4129, acc-0.8600, valid loss-0.3602, acc-0.8980, test loss-0.3658, acc-0.8946
Iter-91320, train loss-0.4446, acc-0.8200, valid loss-0.3602, acc-0.8980, test loss-0.3658, acc-0.8946
Iter-91330, train loss-0.3838, acc-0.8600, valid loss-0.3602, acc-0.8978, test loss-0.3658, acc-0.8946
Iter-91340, train loss-0.3697, acc-0.8800, valid loss-0.3601, acc-0.8978, test loss-0.3657, acc-0.8945
Iter-91350, train loss-0.3939, acc-0.9000, valid loss-0.3601, acc-0.8978, test loss-0.3657, acc-0.8945
Iter-91360, train loss-0.3353, acc-0.9000, valid loss-0.3601, acc-0.8978, test loss-0.3657, acc-0.8948
Iter-91370, train loss-0.4018, acc-0.8400, valid loss-0.3601, acc-0.8978, test loss-0.3657, acc-0.8947
Iter-91380, train loss-0.3619, acc-0.9000, valid loss-0.3601, acc-0.8980, test loss-0.3657, acc-0.8947
Iter-91390, train loss-0.3240, acc-0.9400, valid loss-0.3601, acc-0.8982, test loss-0.3657, acc-0.8947
Iter-91400, train loss-0.4843, acc-0.8400, valid loss-0.3600, acc-0.8982, test loss-0.3657, acc-0.8946
Iter-91410, train loss-0.3214, acc-0.9000, valid loss-0.3600, acc-0.8980, test loss-0.3657, acc-0.8946
Iter-91420, train loss-0.3959, acc-0.8800, valid loss-0.3600, acc-0.8980, test loss-0.3656, acc-0.8947
Iter-91430, train loss-0.3023, acc-0.8800, valid loss-0.3600, acc-0.8978, test loss-0.3656, acc-0.8948
Iter-91440, train loss-0.4524, acc-0.9000, valid loss-0.3600, acc-0.8978, test loss-0.3656, acc-0.8947
Iter-91450, train loss-0.4072, acc-0.8800, valid loss-0.3600, acc-0.8980, test loss-0.3656, acc-0.8946
Iter-91460, train loss-0.5683, acc-0.8600, valid loss-0.3600, acc-0.8978, test loss-0.3656, acc-0.8947
Iter-91470, train loss-0.2953, acc-0.9200, valid loss-0.3600, acc-0.8978, test loss-0.3656, acc-0.8945
Iter-91480, train loss-0.3381, acc-0.8600, valid loss-0.3599, acc-0.8978, test loss-0.3656, acc-0.8946
Iter-91490, train loss-0.2643, acc-0.9400, valid loss-0.3599, acc-0.8978, test loss-0.3656, acc-0.8946
Iter-91500, train loss-0.2873, acc-0.9200, valid loss-0.3599, acc-0.8978, test loss-0.3655, acc-0.8947
Iter-91510, train loss-0.2841, acc-0.9600, valid loss-0.3599, acc-0.8980, test loss-0.3655, acc-0.8946
Iter-91520, train loss-0.4260, acc-0.9200, valid loss-0.3599, acc-0.8978, test loss-0.3655, acc-0.8947
Iter-91530, train loss-0.4049, acc-0.8800, valid loss-0.3599, acc-0.8980, test loss-0.3655, acc-0.8947
Iter-91540, train loss-0.4364, acc-0.8800, valid loss-0.3599, acc-0.8980, test loss-0.3655, acc-0.8946
Iter-91550, train loss-0.6215, acc-0.8400, valid loss-0.3599, acc-0.8980, test loss-0.3655, acc-0.8946
Iter-91560, train loss-0.3044, acc-0.9200, valid loss-0.3598, acc-0.8980, test loss-0.3655, acc-0.8946
Iter-91570, train loss-0.4088, acc-0.9000, valid loss-0.3598, acc-0.8980, test loss-0.3654, acc-0.8944
Iter-91580, train loss-0.3250, acc-0.9200, valid loss-0.3598, acc-0.8980, test loss-0.3654, acc-0.8945
Iter-91590, train loss-0.3434, acc-0.9200, valid loss-0.3598, acc-0.8980, test loss-0.3654, acc-0.8945
Iter-91600, train loss-0.4154, acc-0.9200, valid loss-0.3598, acc-0.8982, test loss-0.3654, acc-0.8946
Iter-91610, train loss-0.3194, acc-0.9400, valid loss-0.3598, acc-0.8982, test loss-0.3654, acc-0.8945
Iter-91620, train loss-0.4443, acc-0.8800, valid loss-0.3598, acc-0.8982, test loss-0.3654, acc-0.8945
Iter-91630, train loss-0.5112, acc-0.8000, valid loss-0.3597, acc-0.8980, test loss-0.3654, acc-0.8947
Iter-91640, train loss-0.4729, acc-0.8200, valid loss-0.3597, acc-0.8980, test loss-0.3654, acc-0.8945
Iter-91650, train loss-0.1818, acc-0.9800, valid loss-0.3597, acc-0.8978, test loss-0.3653, acc-0.8946
Iter-91660, train loss-0.3698, acc-0.9200, valid loss-0.3597, acc-0.8978, test loss-0.3653, acc-0.8946
Iter-91670, train loss-0.4730, acc-0.8000, valid loss-0.3597, acc-0.8978, test loss-0.3653, acc-0.8945
Iter-91680, train loss-0.4970, acc-0.8200, valid loss-0.3597, acc-0.8978, test loss-0.3653, acc-0.8945
Iter-91690, train loss-0.3210, acc-0.9200, valid loss-0.3597, acc-0.8978, test loss-0.3653, acc-0.8945
Iter-91700, train loss-0.1964, acc-0.9600, valid loss-0.3597, acc-0.8978, test loss-0.3653, acc-0.8945
Iter-91710, train loss-0.3411, acc-0.8800, valid loss-0.3597, acc-0.8978, test loss-0.3653, acc-0.8945
Iter-91720, train loss-0.4975, acc-0.8200, valid loss-0.3596, acc-0.8978, test loss-0.3653, acc-0.8945
Iter-91730, train loss-0.4417, acc-0.8600, valid loss-0.3596, acc-0.8978, test loss-0.3653, acc-0.8946
Iter-91740, train loss-0.5000, acc-0.8800, valid loss-0.3596, acc-0.8980, test loss-0.3652, acc-0.8945
Iter-91750, train loss-0.3471, acc-0.9200, valid loss-0.3596, acc-0.8980, test loss-0.3652, acc-0.8946
Iter-91760, train loss-0.1811, acc-0.9800, valid loss-0.3596, acc-0.8980, test loss-0.3652, acc-0.8945
Iter-91770, train loss-0.3925, acc-0.9200, valid loss-0.3596, acc-0.8978, test loss-0.3652, acc-0.8945
Iter-91780, train loss-0.4697, acc-0.8600, valid loss-0.3596, acc-0.8978, test loss-0.3652, acc-0.8947
Iter-91790, train loss-0.3496, acc-0.9000, valid loss-0.3595, acc-0.8978, test loss-0.3652, acc-0.8946
Iter-91800, train loss-0.4655, acc-0.8600, valid loss-0.3595, acc-0.8980, test loss-0.3652, acc-0.8946
Iter-91810, train loss-0.2339, acc-0.9600, valid loss-0.3595, acc-0.8980, test loss-0.3651, acc-0.8946
Iter-91820, train loss-0.4674, acc-0.8200, valid loss-0.3595, acc-0.8982, test loss-0.3651, acc-0.8946
Iter-91830, train loss-0.3216, acc-0.9400, valid loss-0.3595, acc-0.8980, test loss-0.3651, acc-0.8946
Iter-91840, train loss-0.5148, acc-0.8600, valid loss-0.3595, acc-0.8980, test loss-0.3651, acc-0.8946
Iter-91850, train loss-0.3075, acc-0.8400, valid loss-0.3595, acc-0.8980, test loss-0.3651, acc-0.8945
Iter-91860, train loss-0.2577, acc-0.9200, valid loss-0.3595, acc-0.8980, test loss-0.3651, acc-0.8945
Iter-91870, train loss-0.3375, acc-0.9000, valid loss-0.3595, acc-0.8980, test loss-0.3651, acc-0.8946
Iter-91880, train loss-0.3210, acc-0.9200, valid loss-0.3594, acc-0.8980, test loss-0.3651, acc-0.8945
Iter-91890, train loss-0.4187, acc-0.8800, valid loss-0.3594, acc-0.8980, test loss-0.3651, acc-0.8946
Iter-91900, train loss-0.2243, acc-0.9600, valid loss-0.3594, acc-0.8978, test loss-0.3650, acc-0.8947
Iter-91910, train loss-0.3836, acc-0.9200, valid loss-0.3594, acc-0.8978, test loss-0.3650, acc-0.8947
Iter-91920, train loss-0.4339, acc-0.8400, valid loss-0.3594, acc-0.8980, test loss-0.3650, acc-0.8947
Iter-91930, train loss-0.3003, acc-0.9200, valid loss-0.3594, acc-0.8980, test loss-0.3650, acc-0.8946
Iter-91940, train loss-0.3516, acc-0.9200, valid loss-0.3594, acc-0.8980, test loss-0.3650, acc-0.8945
Iter-91950, train loss-0.5286, acc-0.8800, valid loss-0.3594, acc-0.8980, test loss-0.3650, acc-0.8946
Iter-91960, train loss-0.4251, acc-0.8200, valid loss-0.3593, acc-0.8980, test loss-0.3650, acc-0.8946
Iter-91970, train loss-0.3764, acc-0.9000, valid loss-0.3593, acc-0.8982, test loss-0.3650, acc-0.8947
Iter-91980, train loss-0.3437, acc-0.9000, valid loss-0.3593, acc-0.8982, test loss-0.3649, acc-0.8947
Iter-91990, train loss-0.7801, acc-0.7200, valid loss-0.3593, acc-0.8984, test loss-0.3649, acc-0.8948
Iter-92000, train loss-0.3578, acc-0.9000, valid loss-0.3593, acc-0.8982, test loss-0.3649, acc-0.8947
Iter-92010, train loss-0.5375, acc-0.8800, valid loss-0.3593, acc-0.8982, test loss-0.3649, acc-0.8947
Iter-92020, train loss-0.3248, acc-0.9400, valid loss-0.3593, acc-0.8982, test loss-0.3649, acc-0.8948
Iter-92030, train loss-0.4925, acc-0.8800, valid loss-0.3593, acc-0.8982, test loss-0.3649, acc-0.8947
Iter-92040, train loss-0.3830, acc-0.8600, valid loss-0.3592, acc-0.8982, test loss-0.3649, acc-0.8948
Iter-92050, train loss-0.2526, acc-0.9400, valid loss-0.3592, acc-0.8982, test loss-0.3649, acc-0.8948
Iter-92060, train loss-0.3296, acc-0.9200, valid loss-0.3592, acc-0.8982, test loss-0.3648, acc-0.8948
Iter-92070, train loss-0.4317, acc-0.9000, valid loss-0.3592, acc-0.8980, test loss-0.3648, acc-0.8948
Iter-92080, train loss-0.5135, acc-0.8600, valid loss-0.3592, acc-0.8980, test loss-0.3648, acc-0.8948
Iter-92090, train loss-0.3833, acc-0.9000, valid loss-0.3592, acc-0.8980, test loss-0.3648, acc-0.8949
Iter-92100, train loss-0.4310, acc-0.8200, valid loss-0.3592, acc-0.8980, test loss-0.3648, acc-0.8950
Iter-92110, train loss-0.2589, acc-0.9600, valid loss-0.3591, acc-0.8980, test loss-0.3648, acc-0.8950
Iter-92120, train loss-0.4856, acc-0.8200, valid loss-0.3591, acc-0.8980, test loss-0.3648, acc-0.8948
Iter-92130, train loss-0.2481, acc-0.9200, valid loss-0.3591, acc-0.8982, test loss-0.3648, acc-0.8947
Iter-92140, train loss-0.3841, acc-0.9000, valid loss-0.3591, acc-0.8980, test loss-0.3648, acc-0.8947
Iter-92150, train loss-0.2286, acc-0.9600, valid loss-0.3591, acc-0.8980, test loss-0.3648, acc-0.8949
Iter-92160, train loss-0.5656, acc-0.8200, valid loss-0.3591, acc-0.8978, test loss-0.3647, acc-0.8949
Iter-92170, train loss-0.2905, acc-0.9200, valid loss-0.3591, acc-0.8978, test loss-0.3647, acc-0.8949
Iter-92180, train loss-0.2658, acc-0.9200, valid loss-0.3591, acc-0.8978, test loss-0.3647, acc-0.8949
Iter-92190, train loss-0.4546, acc-0.8600, valid loss-0.3591, acc-0.8978, test loss-0.3647, acc-0.8949
Iter-92200, train loss-0.3464, acc-0.9000, valid loss-0.3590, acc-0.8978, test loss-0.3647, acc-0.8949
Iter-92210, train loss-0.2979, acc-0.9000, valid loss-0.3590, acc-0.8978, test loss-0.3647, acc-0.8949
Iter-92220, train loss-0.6034, acc-0.8400, valid loss-0.3590, acc-0.8978, test loss-0.3647, acc-0.8949
Iter-92230, train loss-0.3319, acc-0.9200, valid loss-0.3590, acc-0.8978, test loss-0.3647, acc-0.8949
Iter-92240, train loss-0.3640, acc-0.8800, valid loss-0.3590, acc-0.8978, test loss-0.3646, acc-0.8949
Iter-92250, train loss-0.2947, acc-0.9000, valid loss-0.3590, acc-0.8976, test loss-0.3646, acc-0.8949
Iter-92260, train loss-0.2154, acc-0.9400, valid loss-0.3590, acc-0.8976, test loss-0.3646, acc-0.8950
Iter-92270, train loss-0.3362, acc-0.8600, valid loss-0.3590, acc-0.8976, test loss-0.3646, acc-0.8950
Iter-92280, train loss-0.5427, acc-0.8800, valid loss-0.3589, acc-0.8976, test loss-0.3646, acc-0.8950
Iter-92290, train loss-0.3132, acc-0.9000, valid loss-0.3589, acc-0.8980, test loss-0.3645, acc-0.8951
Iter-92300, train loss-0.3427, acc-0.9200, valid loss-0.3589, acc-0.8980, test loss-0.3645, acc-0.8951
Iter-92310, train loss-0.5209, acc-0.8600, valid loss-0.3589, acc-0.8980, test loss-0.3645, acc-0.8951
Iter-92320, train loss-0.2908, acc-0.9200, valid loss-0.3589, acc-0.8980, test loss-0.3645, acc-0.8951
Iter-92330, train loss-0.4436, acc-0.8400, valid loss-0.3589, acc-0.8980, test loss-0.3645, acc-0.8951
Iter-92340, train loss-0.6651, acc-0.8200, valid loss-0.3588, acc-0.8980, test loss-0.3645, acc-0.8950
Iter-92350, train loss-0.7284, acc-0.8000, valid loss-0.3588, acc-0.8980, test loss-0.3645, acc-0.8951
Iter-92360, train loss-0.3797, acc-0.9000, valid loss-0.3588, acc-0.8980, test loss-0.3645, acc-0.8951
Iter-92370, train loss-0.5947, acc-0.7800, valid loss-0.3588, acc-0.8978, test loss-0.3645, acc-0.8951
Iter-92380, train loss-0.3486, acc-0.8800, valid loss-0.3588, acc-0.8980, test loss-0.3645, acc-0.8951
Iter-92390, train loss-0.3135, acc-0.9400, valid loss-0.3588, acc-0.8980, test loss-0.3645, acc-0.8951
Iter-92400, train loss-0.3245, acc-0.9000, valid loss-0.3588, acc-0.8980, test loss-0.3645, acc-0.8950
Iter-92410, train loss-0.3631, acc-0.9000, valid loss-0.3588, acc-0.8982, test loss-0.3644, acc-0.8950
Iter-92420, train loss-0.3620, acc-0.8600, valid loss-0.3588, acc-0.8982, test loss-0.3644, acc-0.8951
Iter-92430, train loss-0.2773, acc-0.9200, valid loss-0.3587, acc-0.8980, test loss-0.3644, acc-0.8951
Iter-92440, train loss-0.1856, acc-0.9600, valid loss-0.3587, acc-0.8982, test loss-0.3644, acc-0.8951
Iter-92450, train loss-0.4491, acc-0.8600, valid loss-0.3587, acc-0.8980, test loss-0.3644, acc-0.8951
Iter-92460, train loss-0.3423, acc-0.8800, valid loss-0.3587, acc-0.8980, test loss-0.3644, acc-0.8950
Iter-92470, train loss-0.5029, acc-0.8800, valid loss-0.3587, acc-0.8980, test loss-0.3644, acc-0.8950
Iter-92480, train loss-0.1889, acc-0.9400, valid loss-0.3587, acc-0.8980, test loss-0.3644, acc-0.8950
Iter-92490, train loss-0.6707, acc-0.7800, valid loss-0.3586, acc-0.8982, test loss-0.3643, acc-0.8950
Iter-92500, train loss-0.5232, acc-0.8800, valid loss-0.3586, acc-0.8982, test loss-0.3643, acc-0.8950
Iter-92510, train loss-0.3011, acc-0.8800, valid loss-0.3586, acc-0.8980, test loss-0.3643, acc-0.8951
Iter-92520, train loss-0.2857, acc-0.9000, valid loss-0.3586, acc-0.8980, test loss-0.3643, acc-0.8952
Iter-92530, train loss-0.4946, acc-0.8800, valid loss-0.3586, acc-0.8978, test loss-0.3643, acc-0.8952
Iter-92540, train loss-0.3736, acc-0.9400, valid loss-0.3586, acc-0.8982, test loss-0.3643, acc-0.8951
Iter-92550, train loss-0.4982, acc-0.8000, valid loss-0.3586, acc-0.8978, test loss-0.3642, acc-0.8952
Iter-92560, train loss-0.6997, acc-0.8200, valid loss-0.3586, acc-0.8980, test loss-0.3642, acc-0.8951
Iter-92570, train loss-0.5707, acc-0.8200, valid loss-0.3585, acc-0.8980, test loss-0.3642, acc-0.8951
Iter-92580, train loss-0.5062, acc-0.8800, valid loss-0.3585, acc-0.8980, test loss-0.3642, acc-0.8952
Iter-92590, train loss-0.5773, acc-0.8400, valid loss-0.3585, acc-0.8980, test loss-0.3642, acc-0.8952
Iter-92600, train loss-0.2333, acc-0.9800, valid loss-0.3585, acc-0.8982, test loss-0.3642, acc-0.8952
Iter-92610, train loss-0.4063, acc-0.8800, valid loss-0.3585, acc-0.8980, test loss-0.3642, acc-0.8952
Iter-92620, train loss-0.2788, acc-0.9000, valid loss-0.3585, acc-0.8982, test loss-0.3641, acc-0.8952
Iter-92630, train loss-0.4631, acc-0.8600, valid loss-0.3585, acc-0.8982, test loss-0.3641, acc-0.8952
Iter-92640, train loss-0.4074, acc-0.8800, valid loss-0.3584, acc-0.8982, test loss-0.3641, acc-0.8952
Iter-92650, train loss-0.2969, acc-0.9400, valid loss-0.3584, acc-0.8982, test loss-0.3641, acc-0.8951
Iter-92660, train loss-0.3892, acc-0.8800, valid loss-0.3584, acc-0.8984, test loss-0.3641, acc-0.8951
Iter-92670, train loss-0.2600, acc-0.9400, valid loss-0.3584, acc-0.8982, test loss-0.3641, acc-0.8952
Iter-92680, train loss-0.3524, acc-0.9000, valid loss-0.3584, acc-0.8980, test loss-0.3641, acc-0.8953
Iter-92690, train loss-0.3889, acc-0.9000, valid loss-0.3584, acc-0.8980, test loss-0.3640, acc-0.8952
Iter-92700, train loss-0.3419, acc-0.9000, valid loss-0.3583, acc-0.8980, test loss-0.3640, acc-0.8953
Iter-92710, train loss-0.6133, acc-0.8600, valid loss-0.3583, acc-0.8982, test loss-0.3640, acc-0.8953
Iter-92720, train loss-0.2223, acc-0.9400, valid loss-0.3583, acc-0.8982, test loss-0.3640, acc-0.8952
Iter-92730, train loss-0.5281, acc-0.8800, valid loss-0.3583, acc-0.8980, test loss-0.3640, acc-0.8952
Iter-92740, train loss-0.4025, acc-0.8000, valid loss-0.3583, acc-0.8980, test loss-0.3640, acc-0.8951
Iter-92750, train loss-0.3009, acc-0.9200, valid loss-0.3583, acc-0.8980, test loss-0.3640, acc-0.8951
Iter-92760, train loss-0.3729, acc-0.9000, valid loss-0.3583, acc-0.8982, test loss-0.3640, acc-0.8952
Iter-92770, train loss-0.2298, acc-0.9400, valid loss-0.3583, acc-0.8984, test loss-0.3640, acc-0.8953
Iter-92780, train loss-0.4518, acc-0.8400, valid loss-0.3583, acc-0.8984, test loss-0.3639, acc-0.8953
Iter-92790, train loss-0.2993, acc-0.9200, valid loss-0.3582, acc-0.8984, test loss-0.3639, acc-0.8953
Iter-92800, train loss-0.3965, acc-0.8600, valid loss-0.3582, acc-0.8984, test loss-0.3639, acc-0.8953
Iter-92810, train loss-0.2715, acc-0.8800, valid loss-0.3582, acc-0.8984, test loss-0.3639, acc-0.8953
Iter-92820, train loss-0.4687, acc-0.8600, valid loss-0.3582, acc-0.8984, test loss-0.3639, acc-0.8952
Iter-92830, train loss-0.3362, acc-0.9000, valid loss-0.3582, acc-0.8984, test loss-0.3639, acc-0.8951
Iter-92840, train loss-0.4639, acc-0.8800, valid loss-0.3581, acc-0.8984, test loss-0.3639, acc-0.8951
Iter-92850, train loss-0.3999, acc-0.9000, valid loss-0.3581, acc-0.8984, test loss-0.3638, acc-0.8951
Iter-92860, train loss-0.3561, acc-0.9000, valid loss-0.3581, acc-0.8986, test loss-0.3638, acc-0.8951
Iter-92870, train loss-0.2562, acc-0.9400, valid loss-0.3581, acc-0.8986, test loss-0.3638, acc-0.8951
Iter-92880, train loss-0.5138, acc-0.8600, valid loss-0.3581, acc-0.8986, test loss-0.3638, acc-0.8950
Iter-92890, train loss-0.4340, acc-0.8400, valid loss-0.3581, acc-0.8984, test loss-0.3638, acc-0.8952
Iter-92900, train loss-0.3280, acc-0.9000, valid loss-0.3581, acc-0.8986, test loss-0.3638, acc-0.8951
Iter-92910, train loss-0.3242, acc-0.8800, valid loss-0.3581, acc-0.8984, test loss-0.3638, acc-0.8952
Iter-92920, train loss-0.5708, acc-0.8400, valid loss-0.3580, acc-0.8984, test loss-0.3638, acc-0.8952
Iter-92930, train loss-0.4767, acc-0.8800, valid loss-0.3580, acc-0.8984, test loss-0.3637, acc-0.8952
Iter-92940, train loss-0.3680, acc-0.8800, valid loss-0.3580, acc-0.8984, test loss-0.3637, acc-0.8951
Iter-92950, train loss-0.6030, acc-0.8200, valid loss-0.3580, acc-0.8984, test loss-0.3637, acc-0.8952
Iter-92960, train loss-0.4300, acc-0.8600, valid loss-0.3580, acc-0.8984, test loss-0.3637, acc-0.8952
Iter-92970, train loss-0.3504, acc-0.9000, valid loss-0.3580, acc-0.8984, test loss-0.3637, acc-0.8950
Iter-92980, train loss-0.3109, acc-0.9000, valid loss-0.3579, acc-0.8984, test loss-0.3637, acc-0.8952
Iter-92990, train loss-0.6092, acc-0.8400, valid loss-0.3579, acc-0.8984, test loss-0.3637, acc-0.8951
Iter-93000, train loss-0.4655, acc-0.8600, valid loss-0.3579, acc-0.8984, test loss-0.3637, acc-0.8952
Iter-93010, train loss-0.5699, acc-0.8200, valid loss-0.3579, acc-0.8984, test loss-0.3637, acc-0.8952
Iter-93020, train loss-0.2555, acc-0.9200, valid loss-0.3579, acc-0.8984, test loss-0.3637, acc-0.8951
Iter-93030, train loss-0.4886, acc-0.9200, valid loss-0.3579, acc-0.8984, test loss-0.3637, acc-0.8952
Iter-93040, train loss-0.6275, acc-0.8200, valid loss-0.3579, acc-0.8984, test loss-0.3637, acc-0.8952
Iter-93050, train loss-0.3014, acc-0.9000, valid loss-0.3579, acc-0.8986, test loss-0.3636, acc-0.8952
Iter-93060, train loss-0.4256, acc-0.9400, valid loss-0.3579, acc-0.8984, test loss-0.3636, acc-0.8950
Iter-93070, train loss-0.4719, acc-0.8400, valid loss-0.3579, acc-0.8982, test loss-0.3636, acc-0.8949
Iter-93080, train loss-0.4575, acc-0.8400, valid loss-0.3579, acc-0.8984, test loss-0.3636, acc-0.8948
Iter-93090, train loss-0.3285, acc-0.9000, valid loss-0.3578, acc-0.8984, test loss-0.3636, acc-0.8948
Iter-93100, train loss-0.2429, acc-0.9200, valid loss-0.3578, acc-0.8984, test loss-0.3636, acc-0.8948
Iter-93110, train loss-0.3258, acc-0.9000, valid loss-0.3578, acc-0.8984, test loss-0.3636, acc-0.8948
Iter-93120, train loss-0.4472, acc-0.8400, valid loss-0.3578, acc-0.8984, test loss-0.3636, acc-0.8948
Iter-93130, train loss-0.2642, acc-0.9000, valid loss-0.3578, acc-0.8984, test loss-0.3636, acc-0.8948
Iter-93140, train loss-0.3147, acc-0.9000, valid loss-0.3578, acc-0.8984, test loss-0.3635, acc-0.8948
Iter-93150, train loss-0.4755, acc-0.9000, valid loss-0.3578, acc-0.8986, test loss-0.3635, acc-0.8948
Iter-93160, train loss-0.4076, acc-0.8600, valid loss-0.3578, acc-0.8986, test loss-0.3635, acc-0.8948
Iter-93170, train loss-0.3707, acc-0.9000, valid loss-0.3577, acc-0.8984, test loss-0.3635, acc-0.8948
Iter-93180, train loss-0.5506, acc-0.8600, valid loss-0.3577, acc-0.8986, test loss-0.3635, acc-0.8948
Iter-93190, train loss-0.3799, acc-0.8600, valid loss-0.3577, acc-0.8986, test loss-0.3635, acc-0.8948
Iter-93200, train loss-0.4471, acc-0.8600, valid loss-0.3577, acc-0.8986, test loss-0.3634, acc-0.8948
Iter-93210, train loss-0.6353, acc-0.7800, valid loss-0.3577, acc-0.8986, test loss-0.3634, acc-0.8948
Iter-93220, train loss-0.3595, acc-0.9200, valid loss-0.3577, acc-0.8986, test loss-0.3634, acc-0.8948
Iter-93230, train loss-0.2769, acc-0.9200, valid loss-0.3577, acc-0.8986, test loss-0.3634, acc-0.8948
Iter-93240, train loss-0.4739, acc-0.8800, valid loss-0.3576, acc-0.8986, test loss-0.3634, acc-0.8949
Iter-93250, train loss-0.3764, acc-0.9000, valid loss-0.3576, acc-0.8986, test loss-0.3634, acc-0.8948
Iter-93260, train loss-0.5164, acc-0.8200, valid loss-0.3576, acc-0.8986, test loss-0.3634, acc-0.8948
Iter-93270, train loss-0.4879, acc-0.8600, valid loss-0.3576, acc-0.8986, test loss-0.3634, acc-0.8950
Iter-93280, train loss-0.3913, acc-0.9000, valid loss-0.3576, acc-0.8986, test loss-0.3634, acc-0.8949
Iter-93290, train loss-0.4813, acc-0.9200, valid loss-0.3576, acc-0.8986, test loss-0.3633, acc-0.8949
Iter-93300, train loss-0.3545, acc-0.9000, valid loss-0.3576, acc-0.8986, test loss-0.3633, acc-0.8949
Iter-93310, train loss-0.4248, acc-0.9000, valid loss-0.3576, acc-0.8986, test loss-0.3633, acc-0.8948
Iter-93320, train loss-0.5713, acc-0.8600, valid loss-0.3575, acc-0.8986, test loss-0.3633, acc-0.8950
Iter-93330, train loss-0.4891, acc-0.9000, valid loss-0.3575, acc-0.8986, test loss-0.3633, acc-0.8948
Iter-93340, train loss-0.4766, acc-0.9200, valid loss-0.3575, acc-0.8986, test loss-0.3633, acc-0.8948
Iter-93350, train loss-0.2898, acc-0.9400, valid loss-0.3575, acc-0.8986, test loss-0.3633, acc-0.8947
Iter-93360, train loss-0.2346, acc-0.9400, valid loss-0.3575, acc-0.8986, test loss-0.3633, acc-0.8949
Iter-93370, train loss-0.4359, acc-0.9000, valid loss-0.3575, acc-0.8986, test loss-0.3632, acc-0.8948
Iter-93380, train loss-0.3084, acc-0.9400, valid loss-0.3575, acc-0.8986, test loss-0.3632, acc-0.8948
Iter-93390, train loss-0.2694, acc-0.9200, valid loss-0.3574, acc-0.8988, test loss-0.3632, acc-0.8951
Iter-93400, train loss-0.4283, acc-0.8600, valid loss-0.3574, acc-0.8988, test loss-0.3632, acc-0.8951
Iter-93410, train loss-0.5205, acc-0.8800, valid loss-0.3574, acc-0.8988, test loss-0.3632, acc-0.8952
Iter-93420, train loss-0.3917, acc-0.9000, valid loss-0.3574, acc-0.8988, test loss-0.3632, acc-0.8952
Iter-93430, train loss-0.4672, acc-0.8600, valid loss-0.3574, acc-0.8988, test loss-0.3632, acc-0.8952
Iter-93440, train loss-0.2589, acc-0.9400, valid loss-0.3574, acc-0.8986, test loss-0.3631, acc-0.8952
Iter-93450, train loss-0.2177, acc-0.9600, valid loss-0.3574, acc-0.8988, test loss-0.3631, acc-0.8952
Iter-93460, train loss-0.4736, acc-0.8800, valid loss-0.3573, acc-0.8988, test loss-0.3631, acc-0.8953
Iter-93470, train loss-0.3150, acc-0.8800, valid loss-0.3573, acc-0.8988, test loss-0.3631, acc-0.8953
Iter-93480, train loss-0.4229, acc-0.8600, valid loss-0.3573, acc-0.8986, test loss-0.3631, acc-0.8953
Iter-93490, train loss-0.3314, acc-0.8800, valid loss-0.3573, acc-0.8986, test loss-0.3631, acc-0.8952
Iter-93500, train loss-0.2967, acc-0.9600, valid loss-0.3573, acc-0.8986, test loss-0.3631, acc-0.8953
Iter-93510, train loss-0.2495, acc-0.9600, valid loss-0.3573, acc-0.8986, test loss-0.3630, acc-0.8953
Iter-93520, train loss-0.3027, acc-0.9400, valid loss-0.3573, acc-0.8988, test loss-0.3630, acc-0.8952
Iter-93530, train loss-0.2507, acc-0.9600, valid loss-0.3573, acc-0.8986, test loss-0.3630, acc-0.8952
Iter-93540, train loss-0.4121, acc-0.9000, valid loss-0.3572, acc-0.8988, test loss-0.3630, acc-0.8952
Iter-93550, train loss-0.3550, acc-0.9000, valid loss-0.3572, acc-0.8988, test loss-0.3630, acc-0.8952
Iter-93560, train loss-0.3296, acc-0.8600, valid loss-0.3572, acc-0.8986, test loss-0.3630, acc-0.8952
Iter-93570, train loss-0.2079, acc-0.9400, valid loss-0.3572, acc-0.8988, test loss-0.3630, acc-0.8953
Iter-93580, train loss-0.2015, acc-0.9400, valid loss-0.3572, acc-0.8984, test loss-0.3629, acc-0.8953
Iter-93590, train loss-0.4702, acc-0.8600, valid loss-0.3572, acc-0.8984, test loss-0.3629, acc-0.8953
Iter-93600, train loss-0.2162, acc-0.9200, valid loss-0.3572, acc-0.8986, test loss-0.3629, acc-0.8953
Iter-93610, train loss-0.5606, acc-0.8000, valid loss-0.3572, acc-0.8986, test loss-0.3629, acc-0.8954
Iter-93620, train loss-0.3688, acc-0.9000, valid loss-0.3572, acc-0.8986, test loss-0.3629, acc-0.8954
Iter-93630, train loss-0.4740, acc-0.8600, valid loss-0.3571, acc-0.8986, test loss-0.3629, acc-0.8954
Iter-93640, train loss-0.3212, acc-0.9200, valid loss-0.3571, acc-0.8986, test loss-0.3629, acc-0.8953
Iter-93650, train loss-0.2475, acc-0.9400, valid loss-0.3571, acc-0.8986, test loss-0.3629, acc-0.8953
Iter-93660, train loss-0.2980, acc-0.9000, valid loss-0.3571, acc-0.8984, test loss-0.3628, acc-0.8953
Iter-93670, train loss-0.6767, acc-0.8400, valid loss-0.3571, acc-0.8984, test loss-0.3628, acc-0.8953
Iter-93680, train loss-0.3458, acc-0.9200, valid loss-0.3571, acc-0.8986, test loss-0.3628, acc-0.8951
Iter-93690, train loss-0.3413, acc-0.9600, valid loss-0.3571, acc-0.8984, test loss-0.3628, acc-0.8953
Iter-93700, train loss-0.2915, acc-0.9200, valid loss-0.3571, acc-0.8984, test loss-0.3628, acc-0.8952
Iter-93710, train loss-0.3319, acc-0.9200, valid loss-0.3571, acc-0.8986, test loss-0.3628, acc-0.8952
Iter-93720, train loss-0.3571, acc-0.8800, valid loss-0.3570, acc-0.8988, test loss-0.3628, acc-0.8952
Iter-93730, train loss-0.5338, acc-0.9000, valid loss-0.3570, acc-0.8984, test loss-0.3628, acc-0.8951
Iter-93740, train loss-0.4410, acc-0.8800, valid loss-0.3570, acc-0.8984, test loss-0.3627, acc-0.8951
Iter-93750, train loss-0.3137, acc-0.8600, valid loss-0.3570, acc-0.8988, test loss-0.3627, acc-0.8951
Iter-93760, train loss-0.3416, acc-0.9400, valid loss-0.3570, acc-0.8986, test loss-0.3627, acc-0.8953
Iter-93770, train loss-0.2168, acc-0.9600, valid loss-0.3570, acc-0.8986, test loss-0.3627, acc-0.8952
Iter-93780, train loss-0.5261, acc-0.8800, valid loss-0.3570, acc-0.8986, test loss-0.3627, acc-0.8952
Iter-93790, train loss-0.4784, acc-0.9000, valid loss-0.3570, acc-0.8986, test loss-0.3627, acc-0.8952
Iter-93800, train loss-0.2340, acc-0.9200, valid loss-0.3570, acc-0.8986, test loss-0.3627, acc-0.8953
Iter-93810, train loss-0.5367, acc-0.8600, valid loss-0.3569, acc-0.8986, test loss-0.3627, acc-0.8953
Iter-93820, train loss-0.3979, acc-0.9000, valid loss-0.3569, acc-0.8986, test loss-0.3627, acc-0.8954
Iter-93830, train loss-0.2503, acc-0.9200, valid loss-0.3569, acc-0.8986, test loss-0.3626, acc-0.8954
Iter-93840, train loss-0.3117, acc-0.9400, valid loss-0.3569, acc-0.8986, test loss-0.3626, acc-0.8954
Iter-93850, train loss-0.4407, acc-0.8600, valid loss-0.3569, acc-0.8986, test loss-0.3626, acc-0.8954
Iter-93860, train loss-0.5797, acc-0.8000, valid loss-0.3569, acc-0.8986, test loss-0.3626, acc-0.8953
Iter-93870, train loss-0.3045, acc-0.9400, valid loss-0.3569, acc-0.8986, test loss-0.3626, acc-0.8953
Iter-93880, train loss-0.4229, acc-0.8600, valid loss-0.3568, acc-0.8986, test loss-0.3626, acc-0.8955
Iter-93890, train loss-0.3733, acc-0.9000, valid loss-0.3568, acc-0.8986, test loss-0.3626, acc-0.8953
Iter-93900, train loss-0.4033, acc-0.8600, valid loss-0.3568, acc-0.8988, test loss-0.3625, acc-0.8954
Iter-93910, train loss-0.3080, acc-0.9400, valid loss-0.3568, acc-0.8986, test loss-0.3625, acc-0.8955
Iter-93920, train loss-0.3800, acc-0.9000, valid loss-0.3568, acc-0.8986, test loss-0.3625, acc-0.8954
Iter-93930, train loss-0.4553, acc-0.8600, valid loss-0.3568, acc-0.8986, test loss-0.3625, acc-0.8954
Iter-93940, train loss-0.4162, acc-0.9000, valid loss-0.3568, acc-0.8986, test loss-0.3625, acc-0.8954
Iter-93950, train loss-0.2663, acc-0.9200, valid loss-0.3567, acc-0.8986, test loss-0.3625, acc-0.8955
Iter-93960, train loss-0.4510, acc-0.8200, valid loss-0.3567, acc-0.8986, test loss-0.3625, acc-0.8955
Iter-93970, train loss-0.2414, acc-0.9400, valid loss-0.3567, acc-0.8986, test loss-0.3625, acc-0.8955
Iter-93980, train loss-0.4455, acc-0.8800, valid loss-0.3567, acc-0.8986, test loss-0.3624, acc-0.8955
Iter-93990, train loss-0.4368, acc-0.9000, valid loss-0.3567, acc-0.8986, test loss-0.3624, acc-0.8955
Iter-94000, train loss-0.4595, acc-0.8800, valid loss-0.3567, acc-0.8986, test loss-0.3624, acc-0.8955
Iter-94010, train loss-0.2776, acc-0.9200, valid loss-0.3567, acc-0.8986, test loss-0.3624, acc-0.8955
Iter-94020, train loss-0.3568, acc-0.9200, valid loss-0.3566, acc-0.8986, test loss-0.3624, acc-0.8955
Iter-94030, train loss-0.3005, acc-0.9200, valid loss-0.3566, acc-0.8988, test loss-0.3624, acc-0.8955
Iter-94040, train loss-0.3251, acc-0.8800, valid loss-0.3566, acc-0.8988, test loss-0.3623, acc-0.8955
Iter-94050, train loss-0.2748, acc-0.9400, valid loss-0.3566, acc-0.8988, test loss-0.3623, acc-0.8955
Iter-94060, train loss-0.6719, acc-0.7800, valid loss-0.3566, acc-0.8988, test loss-0.3623, acc-0.8955
Iter-94070, train loss-0.2664, acc-0.9200, valid loss-0.3566, acc-0.8988, test loss-0.3623, acc-0.8955
Iter-94080, train loss-0.3606, acc-0.8800, valid loss-0.3565, acc-0.8986, test loss-0.3623, acc-0.8955
Iter-94090, train loss-0.2817, acc-0.9000, valid loss-0.3565, acc-0.8986, test loss-0.3623, acc-0.8956
Iter-94100, train loss-0.4125, acc-0.8600, valid loss-0.3565, acc-0.8988, test loss-0.3623, acc-0.8956
Iter-94110, train loss-0.5832, acc-0.8400, valid loss-0.3565, acc-0.8988, test loss-0.3622, acc-0.8957
Iter-94120, train loss-0.2381, acc-0.9800, valid loss-0.3565, acc-0.8988, test loss-0.3622, acc-0.8958
Iter-94130, train loss-0.3559, acc-0.9200, valid loss-0.3565, acc-0.8988, test loss-0.3622, acc-0.8958
Iter-94140, train loss-0.6425, acc-0.7800, valid loss-0.3565, acc-0.8986, test loss-0.3622, acc-0.8958
Iter-94150, train loss-0.2795, acc-0.9000, valid loss-0.3564, acc-0.8986, test loss-0.3622, acc-0.8959
Iter-94160, train loss-0.3791, acc-0.8600, valid loss-0.3564, acc-0.8986, test loss-0.3622, acc-0.8959
Iter-94170, train loss-0.5751, acc-0.8400, valid loss-0.3564, acc-0.8986, test loss-0.3622, acc-0.8959
Iter-94180, train loss-0.3413, acc-0.9200, valid loss-0.3564, acc-0.8986, test loss-0.3622, acc-0.8959
Iter-94190, train loss-0.4228, acc-0.8800, valid loss-0.3564, acc-0.8986, test loss-0.3621, acc-0.8959
Iter-94200, train loss-0.3811, acc-0.9600, valid loss-0.3564, acc-0.8986, test loss-0.3621, acc-0.8959
Iter-94210, train loss-0.4583, acc-0.8800, valid loss-0.3563, acc-0.8986, test loss-0.3621, acc-0.8959
Iter-94220, train loss-0.5104, acc-0.8600, valid loss-0.3563, acc-0.8986, test loss-0.3621, acc-0.8958
Iter-94230, train loss-0.5199, acc-0.8600, valid loss-0.3563, acc-0.8988, test loss-0.3621, acc-0.8959
Iter-94240, train loss-0.8714, acc-0.7800, valid loss-0.3563, acc-0.8986, test loss-0.3621, acc-0.8959
Iter-94250, train loss-0.5160, acc-0.8600, valid loss-0.3563, acc-0.8986, test loss-0.3621, acc-0.8959
Iter-94260, train loss-0.5579, acc-0.7800, valid loss-0.3563, acc-0.8986, test loss-0.3620, acc-0.8959
Iter-94270, train loss-0.1617, acc-0.9800, valid loss-0.3563, acc-0.8986, test loss-0.3621, acc-0.8959
Iter-94280, train loss-0.2828, acc-0.9400, valid loss-0.3563, acc-0.8986, test loss-0.3620, acc-0.8959
Iter-94290, train loss-0.5091, acc-0.8600, valid loss-0.3562, acc-0.8986, test loss-0.3620, acc-0.8958
Iter-94300, train loss-0.5173, acc-0.8800, valid loss-0.3562, acc-0.8986, test loss-0.3620, acc-0.8958
Iter-94310, train loss-0.3539, acc-0.9000, valid loss-0.3562, acc-0.8986, test loss-0.3620, acc-0.8958
Iter-94320, train loss-0.2248, acc-0.9000, valid loss-0.3562, acc-0.8988, test loss-0.3620, acc-0.8958
Iter-94330, train loss-0.3210, acc-0.9000, valid loss-0.3562, acc-0.8988, test loss-0.3620, acc-0.8958
Iter-94340, train loss-0.5490, acc-0.8000, valid loss-0.3562, acc-0.8988, test loss-0.3619, acc-0.8958
Iter-94350, train loss-0.4202, acc-0.8600, valid loss-0.3562, acc-0.8988, test loss-0.3619, acc-0.8958
Iter-94360, train loss-0.2075, acc-0.9600, valid loss-0.3562, acc-0.8988, test loss-0.3619, acc-0.8958
Iter-94370, train loss-0.4138, acc-0.8600, valid loss-0.3561, acc-0.8988, test loss-0.3619, acc-0.8958
Iter-94380, train loss-0.3255, acc-0.9000, valid loss-0.3561, acc-0.8988, test loss-0.3619, acc-0.8958
Iter-94390, train loss-0.4005, acc-0.8800, valid loss-0.3561, acc-0.8988, test loss-0.3619, acc-0.8958
Iter-94400, train loss-0.2682, acc-0.9400, valid loss-0.3561, acc-0.8988, test loss-0.3619, acc-0.8958
Iter-94410, train loss-0.1773, acc-0.9600, valid loss-0.3561, acc-0.8988, test loss-0.3619, acc-0.8959
Iter-94420, train loss-0.5786, acc-0.8400, valid loss-0.3561, acc-0.8986, test loss-0.3618, acc-0.8959
Iter-94430, train loss-0.3720, acc-0.8800, valid loss-0.3561, acc-0.8988, test loss-0.3618, acc-0.8958
Iter-94440, train loss-0.4903, acc-0.8400, valid loss-0.3560, acc-0.8988, test loss-0.3618, acc-0.8958
Iter-94450, train loss-0.3192, acc-0.9400, valid loss-0.3560, acc-0.8988, test loss-0.3618, acc-0.8958
Iter-94460, train loss-0.4247, acc-0.9600, valid loss-0.3560, acc-0.8988, test loss-0.3618, acc-0.8958
Iter-94470, train loss-0.3263, acc-0.9000, valid loss-0.3560, acc-0.8990, test loss-0.3618, acc-0.8958
Iter-94480, train loss-0.3673, acc-0.9200, valid loss-0.3560, acc-0.8986, test loss-0.3618, acc-0.8958
Iter-94490, train loss-0.3476, acc-0.9400, valid loss-0.3560, acc-0.8986, test loss-0.3618, acc-0.8957
Iter-94500, train loss-0.4643, acc-0.9000, valid loss-0.3560, acc-0.8986, test loss-0.3618, acc-0.8958
Iter-94510, train loss-0.4002, acc-0.8600, valid loss-0.3560, acc-0.8986, test loss-0.3617, acc-0.8958
Iter-94520, train loss-0.5368, acc-0.8200, valid loss-0.3560, acc-0.8986, test loss-0.3617, acc-0.8958
Iter-94530, train loss-0.3753, acc-0.9200, valid loss-0.3559, acc-0.8988, test loss-0.3617, acc-0.8958
Iter-94540, train loss-0.3928, acc-0.8800, valid loss-0.3559, acc-0.8986, test loss-0.3617, acc-0.8958
Iter-94550, train loss-0.3613, acc-0.8600, valid loss-0.3559, acc-0.8986, test loss-0.3617, acc-0.8958
Iter-94560, train loss-0.3899, acc-0.9000, valid loss-0.3559, acc-0.8988, test loss-0.3617, acc-0.8958
Iter-94570, train loss-0.2515, acc-0.9400, valid loss-0.3559, acc-0.8988, test loss-0.3617, acc-0.8960
Iter-94580, train loss-0.2779, acc-0.9000, valid loss-0.3559, acc-0.8988, test loss-0.3617, acc-0.8959
Iter-94590, train loss-0.2692, acc-0.9200, valid loss-0.3559, acc-0.8988, test loss-0.3616, acc-0.8961
Iter-94600, train loss-0.2782, acc-0.9800, valid loss-0.3558, acc-0.8988, test loss-0.3616, acc-0.8961
Iter-94610, train loss-0.4115, acc-0.8600, valid loss-0.3558, acc-0.8988, test loss-0.3616, acc-0.8961
Iter-94620, train loss-0.2442, acc-0.9400, valid loss-0.3558, acc-0.8988, test loss-0.3616, acc-0.8960
Iter-94630, train loss-0.2736, acc-0.9200, valid loss-0.3558, acc-0.8988, test loss-0.3616, acc-0.8962
Iter-94640, train loss-0.5498, acc-0.8200, valid loss-0.3558, acc-0.8988, test loss-0.3616, acc-0.8962
Iter-94650, train loss-0.3371, acc-0.9400, valid loss-0.3558, acc-0.8988, test loss-0.3616, acc-0.8961
Iter-94660, train loss-0.2676, acc-0.9200, valid loss-0.3558, acc-0.8988, test loss-0.3616, acc-0.8963
Iter-94670, train loss-0.5207, acc-0.8200, valid loss-0.3558, acc-0.8988, test loss-0.3616, acc-0.8962
Iter-94680, train loss-0.6372, acc-0.8400, valid loss-0.3558, acc-0.8988, test loss-0.3616, acc-0.8962
Iter-94690, train loss-0.2126, acc-0.9200, valid loss-0.3557, acc-0.8988, test loss-0.3615, acc-0.8962
Iter-94700, train loss-0.2109, acc-0.9800, valid loss-0.3557, acc-0.8988, test loss-0.3615, acc-0.8962
Iter-94710, train loss-0.3276, acc-0.9200, valid loss-0.3557, acc-0.8988, test loss-0.3615, acc-0.8963
Iter-94720, train loss-0.2215, acc-0.9400, valid loss-0.3557, acc-0.8988, test loss-0.3615, acc-0.8963
Iter-94730, train loss-0.6065, acc-0.7800, valid loss-0.3557, acc-0.8988, test loss-0.3615, acc-0.8961
Iter-94740, train loss-0.2678, acc-0.9200, valid loss-0.3557, acc-0.8988, test loss-0.3615, acc-0.8961
Iter-94750, train loss-0.2915, acc-0.9000, valid loss-0.3557, acc-0.8988, test loss-0.3615, acc-0.8961
Iter-94760, train loss-0.2377, acc-0.9600, valid loss-0.3557, acc-0.8988, test loss-0.3614, acc-0.8962
Iter-94770, train loss-0.4329, acc-0.8200, valid loss-0.3557, acc-0.8988, test loss-0.3614, acc-0.8963
Iter-94780, train loss-0.2370, acc-0.9000, valid loss-0.3556, acc-0.8988, test loss-0.3614, acc-0.8962
Iter-94790, train loss-0.2641, acc-0.9000, valid loss-0.3556, acc-0.8988, test loss-0.3614, acc-0.8962
Iter-94800, train loss-0.4474, acc-0.8800, valid loss-0.3556, acc-0.8988, test loss-0.3614, acc-0.8961
Iter-94810, train loss-0.4081, acc-0.9400, valid loss-0.3556, acc-0.8988, test loss-0.3614, acc-0.8961
Iter-94820, train loss-0.4179, acc-0.8800, valid loss-0.3556, acc-0.8988, test loss-0.3614, acc-0.8962
Iter-94830, train loss-0.5536, acc-0.8600, valid loss-0.3556, acc-0.8988, test loss-0.3613, acc-0.8962
Iter-94840, train loss-0.3851, acc-0.9200, valid loss-0.3556, acc-0.8988, test loss-0.3613, acc-0.8962
Iter-94850, train loss-0.4057, acc-0.8600, valid loss-0.3556, acc-0.8988, test loss-0.3613, acc-0.8962
Iter-94860, train loss-0.4358, acc-0.8600, valid loss-0.3555, acc-0.8988, test loss-0.3613, acc-0.8962
Iter-94870, train loss-0.2472, acc-0.9200, valid loss-0.3555, acc-0.8990, test loss-0.3613, acc-0.8962
Iter-94880, train loss-0.3626, acc-0.9400, valid loss-0.3555, acc-0.8990, test loss-0.3613, acc-0.8962
Iter-94890, train loss-0.2431, acc-0.9200, valid loss-0.3555, acc-0.8990, test loss-0.3613, acc-0.8962
Iter-94900, train loss-0.1548, acc-0.9800, valid loss-0.3555, acc-0.8990, test loss-0.3613, acc-0.8962
Iter-94910, train loss-0.2784, acc-0.9200, valid loss-0.3555, acc-0.8990, test loss-0.3613, acc-0.8962
Iter-94920, train loss-0.3433, acc-0.9000, valid loss-0.3555, acc-0.8990, test loss-0.3612, acc-0.8962
Iter-94930, train loss-0.2944, acc-0.9200, valid loss-0.3555, acc-0.8990, test loss-0.3612, acc-0.8961
Iter-94940, train loss-0.6623, acc-0.8000, valid loss-0.3555, acc-0.8990, test loss-0.3612, acc-0.8961
Iter-94950, train loss-0.2186, acc-0.9600, valid loss-0.3555, acc-0.8990, test loss-0.3612, acc-0.8961
Iter-94960, train loss-0.4927, acc-0.8000, valid loss-0.3555, acc-0.8990, test loss-0.3612, acc-0.8962
Iter-94970, train loss-0.4542, acc-0.8400, valid loss-0.3554, acc-0.8990, test loss-0.3612, acc-0.8962
Iter-94980, train loss-0.2722, acc-0.9200, valid loss-0.3554, acc-0.8990, test loss-0.3612, acc-0.8961
Iter-94990, train loss-0.3007, acc-0.9000, valid loss-0.3554, acc-0.8990, test loss-0.3612, acc-0.8961
Iter-95000, train loss-0.2461, acc-0.9200, valid loss-0.3554, acc-0.8990, test loss-0.3612, acc-0.8961
Iter-95010, train loss-0.2927, acc-0.9000, valid loss-0.3554, acc-0.8990, test loss-0.3612, acc-0.8961
Iter-95020, train loss-0.2831, acc-0.9200, valid loss-0.3554, acc-0.8992, test loss-0.3612, acc-0.8961
Iter-95030, train loss-0.5095, acc-0.8000, valid loss-0.3554, acc-0.8992, test loss-0.3611, acc-0.8962
Iter-95040, train loss-0.3560, acc-0.9000, valid loss-0.3554, acc-0.8994, test loss-0.3611, acc-0.8961
Iter-95050, train loss-0.3484, acc-0.8800, valid loss-0.3554, acc-0.8994, test loss-0.3611, acc-0.8962
Iter-95060, train loss-0.2927, acc-0.9200, valid loss-0.3554, acc-0.8994, test loss-0.3611, acc-0.8962
Iter-95070, train loss-0.1685, acc-0.9600, valid loss-0.3554, acc-0.8992, test loss-0.3611, acc-0.8962
Iter-95080, train loss-0.4288, acc-0.8600, valid loss-0.3554, acc-0.8994, test loss-0.3611, acc-0.8962
Iter-95090, train loss-0.2697, acc-0.9400, valid loss-0.3553, acc-0.8994, test loss-0.3611, acc-0.8961
Iter-95100, train loss-0.5613, acc-0.8400, valid loss-0.3553, acc-0.8994, test loss-0.3611, acc-0.8961
Iter-95110, train loss-0.3584, acc-0.9200, valid loss-0.3553, acc-0.8994, test loss-0.3611, acc-0.8962
Iter-95120, train loss-0.3151, acc-0.9000, valid loss-0.3553, acc-0.8992, test loss-0.3611, acc-0.8961
Iter-95130, train loss-0.2621, acc-0.9000, valid loss-0.3553, acc-0.8992, test loss-0.3610, acc-0.8962
Iter-95140, train loss-0.3030, acc-0.9200, valid loss-0.3553, acc-0.8994, test loss-0.3610, acc-0.8962
Iter-95150, train loss-0.3975, acc-0.8400, valid loss-0.3553, acc-0.8992, test loss-0.3610, acc-0.8962
Iter-95160, train loss-0.4327, acc-0.8800, valid loss-0.3553, acc-0.8992, test loss-0.3610, acc-0.8961
Iter-95170, train loss-0.2257, acc-0.9600, valid loss-0.3553, acc-0.8992, test loss-0.3610, acc-0.8962
Iter-95180, train loss-0.2045, acc-0.9800, valid loss-0.3553, acc-0.8992, test loss-0.3610, acc-0.8962
Iter-95190, train loss-0.4189, acc-0.8200, valid loss-0.3553, acc-0.8992, test loss-0.3610, acc-0.8962
Iter-95200, train loss-0.5126, acc-0.8200, valid loss-0.3553, acc-0.8990, test loss-0.3610, acc-0.8962
Iter-95210, train loss-0.5190, acc-0.8400, valid loss-0.3552, acc-0.8990, test loss-0.3609, acc-0.8962
Iter-95220, train loss-0.3812, acc-0.9000, valid loss-0.3552, acc-0.8990, test loss-0.3609, acc-0.8962
Iter-95230, train loss-0.4113, acc-0.8800, valid loss-0.3552, acc-0.8990, test loss-0.3609, acc-0.8963
Iter-95240, train loss-0.4844, acc-0.8600, valid loss-0.3552, acc-0.8990, test loss-0.3609, acc-0.8963
Iter-95250, train loss-0.2827, acc-0.9200, valid loss-0.3552, acc-0.8990, test loss-0.3609, acc-0.8963
Iter-95260, train loss-0.3152, acc-0.9600, valid loss-0.3552, acc-0.8990, test loss-0.3609, acc-0.8964
Iter-95270, train loss-0.3506, acc-0.9000, valid loss-0.3552, acc-0.8990, test loss-0.3609, acc-0.8964
Iter-95280, train loss-0.2901, acc-0.9200, valid loss-0.3552, acc-0.8990, test loss-0.3609, acc-0.8964
Iter-95290, train loss-0.2788, acc-0.9400, valid loss-0.3551, acc-0.8990, test loss-0.3609, acc-0.8964
Iter-95300, train loss-0.3215, acc-0.9200, valid loss-0.3551, acc-0.8988, test loss-0.3608, acc-0.8964
Iter-95310, train loss-0.3381, acc-0.9000, valid loss-0.3551, acc-0.8990, test loss-0.3608, acc-0.8964
Iter-95320, train loss-0.3896, acc-0.9000, valid loss-0.3551, acc-0.8990, test loss-0.3608, acc-0.8964
Iter-95330, train loss-0.2755, acc-0.9200, valid loss-0.3551, acc-0.8990, test loss-0.3608, acc-0.8964
Iter-95340, train loss-0.2483, acc-0.9600, valid loss-0.3551, acc-0.8990, test loss-0.3608, acc-0.8964
Iter-95350, train loss-0.3259, acc-0.9000, valid loss-0.3550, acc-0.8990, test loss-0.3608, acc-0.8964
Iter-95360, train loss-0.5836, acc-0.8800, valid loss-0.3550, acc-0.8990, test loss-0.3608, acc-0.8964
Iter-95370, train loss-0.5117, acc-0.8200, valid loss-0.3550, acc-0.8990, test loss-0.3608, acc-0.8965
Iter-95380, train loss-0.4270, acc-0.9200, valid loss-0.3550, acc-0.8990, test loss-0.3608, acc-0.8964
Iter-95390, train loss-0.2621, acc-0.9600, valid loss-0.3550, acc-0.8988, test loss-0.3607, acc-0.8965
Iter-95400, train loss-0.3508, acc-0.9200, valid loss-0.3550, acc-0.8988, test loss-0.3607, acc-0.8965
Iter-95410, train loss-0.3693, acc-0.9200, valid loss-0.3550, acc-0.8988, test loss-0.3607, acc-0.8964
Iter-95420, train loss-0.3970, acc-0.8000, valid loss-0.3550, acc-0.8988, test loss-0.3607, acc-0.8964
Iter-95430, train loss-0.4088, acc-0.8600, valid loss-0.3550, acc-0.8988, test loss-0.3607, acc-0.8963
Iter-95440, train loss-0.4147, acc-0.9400, valid loss-0.3550, acc-0.8988, test loss-0.3607, acc-0.8964
Iter-95450, train loss-0.2826, acc-0.9200, valid loss-0.3550, acc-0.8988, test loss-0.3607, acc-0.8962
Iter-95460, train loss-0.2236, acc-0.9200, valid loss-0.3550, acc-0.8988, test loss-0.3606, acc-0.8961
Iter-95470, train loss-0.3793, acc-0.8800, valid loss-0.3550, acc-0.8988, test loss-0.3606, acc-0.8961
Iter-95480, train loss-0.4549, acc-0.9400, valid loss-0.3549, acc-0.8988, test loss-0.3606, acc-0.8962
Iter-95490, train loss-0.2694, acc-0.8800, valid loss-0.3549, acc-0.8988, test loss-0.3606, acc-0.8962
Iter-95500, train loss-0.4097, acc-0.8600, valid loss-0.3549, acc-0.8988, test loss-0.3606, acc-0.8960
Iter-95510, train loss-0.4509, acc-0.8000, valid loss-0.3549, acc-0.8988, test loss-0.3606, acc-0.8961
Iter-95520, train loss-0.4344, acc-0.9000, valid loss-0.3549, acc-0.8988, test loss-0.3606, acc-0.8961
Iter-95530, train loss-0.2746, acc-0.9600, valid loss-0.3548, acc-0.8988, test loss-0.3606, acc-0.8963
Iter-95540, train loss-0.3987, acc-0.8800, valid loss-0.3548, acc-0.8988, test loss-0.3605, acc-0.8961
Iter-95550, train loss-0.3271, acc-0.9000, valid loss-0.3548, acc-0.8988, test loss-0.3605, acc-0.8961
Iter-95560, train loss-0.4392, acc-0.7800, valid loss-0.3548, acc-0.8988, test loss-0.3605, acc-0.8962
Iter-95570, train loss-0.4687, acc-0.9000, valid loss-0.3548, acc-0.8988, test loss-0.3605, acc-0.8963
Iter-95580, train loss-0.2659, acc-0.9200, valid loss-0.3548, acc-0.8988, test loss-0.3605, acc-0.8964
Iter-95590, train loss-0.4014, acc-0.8600, valid loss-0.3548, acc-0.8988, test loss-0.3605, acc-0.8963
Iter-95600, train loss-0.4431, acc-0.9000, valid loss-0.3547, acc-0.8988, test loss-0.3605, acc-0.8963
Iter-95610, train loss-0.6076, acc-0.8000, valid loss-0.3547, acc-0.8988, test loss-0.3604, acc-0.8964
Iter-95620, train loss-0.3458, acc-0.9000, valid loss-0.3547, acc-0.8990, test loss-0.3604, acc-0.8963
Iter-95630, train loss-0.4527, acc-0.8400, valid loss-0.3547, acc-0.8988, test loss-0.3604, acc-0.8963
Iter-95640, train loss-0.3734, acc-0.9000, valid loss-0.3547, acc-0.8988, test loss-0.3604, acc-0.8963
Iter-95650, train loss-0.4160, acc-0.9000, valid loss-0.3547, acc-0.8988, test loss-0.3604, acc-0.8961
Iter-95660, train loss-0.5546, acc-0.8600, valid loss-0.3547, acc-0.8990, test loss-0.3604, acc-0.8962
Iter-95670, train loss-0.3637, acc-0.9000, valid loss-0.3547, acc-0.8988, test loss-0.3604, acc-0.8962
Iter-95680, train loss-0.3502, acc-0.8800, valid loss-0.3547, acc-0.8988, test loss-0.3604, acc-0.8963
Iter-95690, train loss-0.3747, acc-0.9000, valid loss-0.3546, acc-0.8988, test loss-0.3604, acc-0.8963
Iter-95700, train loss-0.4649, acc-0.8400, valid loss-0.3546, acc-0.8988, test loss-0.3603, acc-0.8963
Iter-95710, train loss-0.4425, acc-0.9000, valid loss-0.3546, acc-0.8988, test loss-0.3603, acc-0.8963
Iter-95720, train loss-0.2281, acc-0.9400, valid loss-0.3546, acc-0.8988, test loss-0.3603, acc-0.8963
Iter-95730, train loss-0.4200, acc-0.8400, valid loss-0.3546, acc-0.8988, test loss-0.3603, acc-0.8964
Iter-95740, train loss-0.3583, acc-0.9200, valid loss-0.3546, acc-0.8988, test loss-0.3603, acc-0.8964
Iter-95750, train loss-0.5304, acc-0.8600, valid loss-0.3545, acc-0.8988, test loss-0.3603, acc-0.8964
Iter-95760, train loss-0.3711, acc-0.9200, valid loss-0.3545, acc-0.8988, test loss-0.3603, acc-0.8963
Iter-95770, train loss-0.3701, acc-0.8800, valid loss-0.3545, acc-0.8988, test loss-0.3603, acc-0.8963
Iter-95780, train loss-0.4980, acc-0.9000, valid loss-0.3545, acc-0.8990, test loss-0.3602, acc-0.8963
Iter-95790, train loss-0.2730, acc-0.8600, valid loss-0.3545, acc-0.8990, test loss-0.3602, acc-0.8964
Iter-95800, train loss-0.4359, acc-0.8800, valid loss-0.3545, acc-0.8992, test loss-0.3602, acc-0.8963
Iter-95810, train loss-0.3276, acc-0.9000, valid loss-0.3545, acc-0.8992, test loss-0.3602, acc-0.8963
Iter-95820, train loss-0.4496, acc-0.8800, valid loss-0.3545, acc-0.8992, test loss-0.3602, acc-0.8963
Iter-95830, train loss-0.3761, acc-0.8800, valid loss-0.3544, acc-0.8992, test loss-0.3602, acc-0.8962
Iter-95840, train loss-0.4252, acc-0.9000, valid loss-0.3544, acc-0.8990, test loss-0.3602, acc-0.8962
Iter-95850, train loss-0.5139, acc-0.8800, valid loss-0.3544, acc-0.8992, test loss-0.3602, acc-0.8961
Iter-95860, train loss-0.5338, acc-0.8600, valid loss-0.3544, acc-0.8992, test loss-0.3602, acc-0.8960
Iter-95870, train loss-0.5510, acc-0.9000, valid loss-0.3544, acc-0.8992, test loss-0.3601, acc-0.8960
Iter-95880, train loss-0.5169, acc-0.8600, valid loss-0.3544, acc-0.8990, test loss-0.3601, acc-0.8959
Iter-95890, train loss-0.2862, acc-0.9200, valid loss-0.3544, acc-0.8990, test loss-0.3601, acc-0.8961
Iter-95900, train loss-0.2933, acc-0.9200, valid loss-0.3544, acc-0.8990, test loss-0.3601, acc-0.8961
Iter-95910, train loss-0.3140, acc-0.8800, valid loss-0.3544, acc-0.8990, test loss-0.3601, acc-0.8960
Iter-95920, train loss-0.4226, acc-0.8800, valid loss-0.3543, acc-0.8990, test loss-0.3601, acc-0.8961
Iter-95930, train loss-0.4255, acc-0.9200, valid loss-0.3543, acc-0.8990, test loss-0.3601, acc-0.8960
Iter-95940, train loss-0.3921, acc-0.9000, valid loss-0.3543, acc-0.8990, test loss-0.3601, acc-0.8961
Iter-95950, train loss-0.6528, acc-0.8000, valid loss-0.3543, acc-0.8990, test loss-0.3600, acc-0.8960
Iter-95960, train loss-0.2096, acc-0.9600, valid loss-0.3543, acc-0.8990, test loss-0.3600, acc-0.8961
Iter-95970, train loss-0.2924, acc-0.9200, valid loss-0.3543, acc-0.8990, test loss-0.3600, acc-0.8960
Iter-95980, train loss-0.4532, acc-0.8200, valid loss-0.3543, acc-0.8990, test loss-0.3600, acc-0.8960
Iter-95990, train loss-0.3360, acc-0.9000, valid loss-0.3542, acc-0.8990, test loss-0.3600, acc-0.8959
Iter-96000, train loss-0.3807, acc-0.9000, valid loss-0.3542, acc-0.8990, test loss-0.3600, acc-0.8960
Iter-96010, train loss-0.3792, acc-0.9000, valid loss-0.3542, acc-0.8990, test loss-0.3600, acc-0.8958
Iter-96020, train loss-0.2237, acc-0.9200, valid loss-0.3542, acc-0.8992, test loss-0.3600, acc-0.8959
Iter-96030, train loss-0.3776, acc-0.8800, valid loss-0.3542, acc-0.8994, test loss-0.3599, acc-0.8959
Iter-96040, train loss-0.1579, acc-0.9600, valid loss-0.3542, acc-0.8992, test loss-0.3599, acc-0.8959
Iter-96050, train loss-0.2414, acc-0.9800, valid loss-0.3542, acc-0.8994, test loss-0.3599, acc-0.8959
Iter-96060, train loss-0.5395, acc-0.8600, valid loss-0.3542, acc-0.8996, test loss-0.3599, acc-0.8959
Iter-96070, train loss-0.5915, acc-0.8200, valid loss-0.3542, acc-0.8994, test loss-0.3599, acc-0.8959
Iter-96080, train loss-0.4039, acc-0.8600, valid loss-0.3541, acc-0.8994, test loss-0.3599, acc-0.8959
Iter-96090, train loss-0.2559, acc-0.9200, valid loss-0.3541, acc-0.8994, test loss-0.3599, acc-0.8959
Iter-96100, train loss-0.6276, acc-0.8200, valid loss-0.3541, acc-0.8994, test loss-0.3599, acc-0.8959
Iter-96110, train loss-0.5877, acc-0.8200, valid loss-0.3541, acc-0.8994, test loss-0.3599, acc-0.8959
Iter-96120, train loss-0.3714, acc-0.8800, valid loss-0.3541, acc-0.8994, test loss-0.3598, acc-0.8959
Iter-96130, train loss-0.3341, acc-0.9000, valid loss-0.3541, acc-0.8994, test loss-0.3598, acc-0.8959
Iter-96140, train loss-0.2133, acc-0.9800, valid loss-0.3541, acc-0.8996, test loss-0.3598, acc-0.8960
Iter-96150, train loss-0.2602, acc-0.9400, valid loss-0.3541, acc-0.8996, test loss-0.3598, acc-0.8960
Iter-96160, train loss-0.5639, acc-0.8200, valid loss-0.3540, acc-0.8994, test loss-0.3598, acc-0.8960
Iter-96170, train loss-0.2963, acc-0.9200, valid loss-0.3540, acc-0.8996, test loss-0.3598, acc-0.8960
Iter-96180, train loss-0.6143, acc-0.8200, valid loss-0.3540, acc-0.8996, test loss-0.3598, acc-0.8959
Iter-96190, train loss-0.6818, acc-0.7400, valid loss-0.3540, acc-0.8996, test loss-0.3598, acc-0.8958
Iter-96200, train loss-0.3409, acc-0.8800, valid loss-0.3540, acc-0.8998, test loss-0.3598, acc-0.8959
Iter-96210, train loss-0.6692, acc-0.8400, valid loss-0.3540, acc-0.8994, test loss-0.3597, acc-0.8959
Iter-96220, train loss-0.4156, acc-0.9000, valid loss-0.3539, acc-0.8992, test loss-0.3597, acc-0.8959
Iter-96230, train loss-0.2887, acc-0.9200, valid loss-0.3540, acc-0.8992, test loss-0.3597, acc-0.8958
Iter-96240, train loss-0.3259, acc-0.8600, valid loss-0.3539, acc-0.8992, test loss-0.3597, acc-0.8960
Iter-96250, train loss-0.2991, acc-0.9400, valid loss-0.3539, acc-0.8994, test loss-0.3597, acc-0.8958
Iter-96260, train loss-0.1760, acc-0.9600, valid loss-0.3539, acc-0.8994, test loss-0.3597, acc-0.8961
Iter-96270, train loss-0.4934, acc-0.8200, valid loss-0.3539, acc-0.8994, test loss-0.3597, acc-0.8962
Iter-96280, train loss-0.4031, acc-0.8800, valid loss-0.3539, acc-0.8992, test loss-0.3597, acc-0.8959
Iter-96290, train loss-0.4272, acc-0.8800, valid loss-0.3539, acc-0.8992, test loss-0.3596, acc-0.8961
Iter-96300, train loss-0.3920, acc-0.8600, valid loss-0.3539, acc-0.8996, test loss-0.3596, acc-0.8959
Iter-96310, train loss-0.3662, acc-0.9200, valid loss-0.3538, acc-0.8996, test loss-0.3596, acc-0.8958
Iter-96320, train loss-0.3603, acc-0.8600, valid loss-0.3538, acc-0.8996, test loss-0.3596, acc-0.8958
Iter-96330, train loss-0.1562, acc-0.9800, valid loss-0.3538, acc-0.8994, test loss-0.3596, acc-0.8958
Iter-96340, train loss-0.3559, acc-0.9400, valid loss-0.3538, acc-0.8994, test loss-0.3596, acc-0.8958
Iter-96350, train loss-0.5940, acc-0.8600, valid loss-0.3538, acc-0.8994, test loss-0.3596, acc-0.8958
Iter-96360, train loss-0.2507, acc-0.9400, valid loss-0.3538, acc-0.8992, test loss-0.3596, acc-0.8959
Iter-96370, train loss-0.3881, acc-0.9200, valid loss-0.3538, acc-0.8992, test loss-0.3595, acc-0.8960
Iter-96380, train loss-0.5896, acc-0.8200, valid loss-0.3538, acc-0.8992, test loss-0.3595, acc-0.8960
Iter-96390, train loss-0.3348, acc-0.9000, valid loss-0.3537, acc-0.8992, test loss-0.3595, acc-0.8960
Iter-96400, train loss-0.3602, acc-0.9000, valid loss-0.3537, acc-0.8992, test loss-0.3595, acc-0.8959
Iter-96410, train loss-0.3935, acc-0.9200, valid loss-0.3537, acc-0.8994, test loss-0.3595, acc-0.8962
Iter-96420, train loss-0.5486, acc-0.8400, valid loss-0.3537, acc-0.8994, test loss-0.3595, acc-0.8961
Iter-96430, train loss-0.4443, acc-0.8400, valid loss-0.3537, acc-0.8996, test loss-0.3595, acc-0.8962
Iter-96440, train loss-0.2614, acc-0.9000, valid loss-0.3537, acc-0.8996, test loss-0.3595, acc-0.8961
Iter-96450, train loss-0.2679, acc-0.8600, valid loss-0.3537, acc-0.8996, test loss-0.3595, acc-0.8961
Iter-96460, train loss-0.3548, acc-0.9200, valid loss-0.3537, acc-0.8996, test loss-0.3595, acc-0.8961
Iter-96470, train loss-0.2688, acc-0.9200, valid loss-0.3536, acc-0.8994, test loss-0.3594, acc-0.8961
Iter-96480, train loss-0.2847, acc-0.9000, valid loss-0.3536, acc-0.8996, test loss-0.3594, acc-0.8961
Iter-96490, train loss-0.4581, acc-0.8800, valid loss-0.3536, acc-0.8996, test loss-0.3594, acc-0.8962
Iter-96500, train loss-0.6441, acc-0.8000, valid loss-0.3536, acc-0.8994, test loss-0.3594, acc-0.8961
Iter-96510, train loss-0.3945, acc-0.9000, valid loss-0.3536, acc-0.8994, test loss-0.3594, acc-0.8960
Iter-96520, train loss-0.4662, acc-0.9000, valid loss-0.3536, acc-0.8996, test loss-0.3594, acc-0.8960
Iter-96530, train loss-0.3787, acc-0.8800, valid loss-0.3536, acc-0.8996, test loss-0.3594, acc-0.8961
Iter-96540, train loss-0.2692, acc-0.9400, valid loss-0.3536, acc-0.8994, test loss-0.3593, acc-0.8962
Iter-96550, train loss-0.4186, acc-0.9200, valid loss-0.3536, acc-0.8996, test loss-0.3593, acc-0.8961
Iter-96560, train loss-0.3645, acc-0.8800, valid loss-0.3535, acc-0.8996, test loss-0.3593, acc-0.8961
Iter-96570, train loss-0.1631, acc-0.9600, valid loss-0.3535, acc-0.8996, test loss-0.3593, acc-0.8962
Iter-96580, train loss-0.5348, acc-0.8200, valid loss-0.3535, acc-0.8996, test loss-0.3593, acc-0.8962
Iter-96590, train loss-0.3637, acc-0.9000, valid loss-0.3535, acc-0.8996, test loss-0.3593, acc-0.8962
Iter-96600, train loss-0.2997, acc-0.9200, valid loss-0.3535, acc-0.8996, test loss-0.3593, acc-0.8962
Iter-96610, train loss-0.4968, acc-0.8600, valid loss-0.3535, acc-0.8996, test loss-0.3593, acc-0.8962
Iter-96620, train loss-0.5673, acc-0.9000, valid loss-0.3535, acc-0.8996, test loss-0.3593, acc-0.8962
Iter-96630, train loss-0.1490, acc-0.9400, valid loss-0.3535, acc-0.8996, test loss-0.3592, acc-0.8962
Iter-96640, train loss-0.3271, acc-0.9000, valid loss-0.3535, acc-0.8996, test loss-0.3592, acc-0.8962
Iter-96650, train loss-0.2200, acc-0.9400, valid loss-0.3535, acc-0.8996, test loss-0.3592, acc-0.8962
Iter-96660, train loss-0.4378, acc-0.8800, valid loss-0.3535, acc-0.8996, test loss-0.3592, acc-0.8962
Iter-96670, train loss-0.4374, acc-0.8400, valid loss-0.3535, acc-0.8996, test loss-0.3592, acc-0.8962
Iter-96680, train loss-0.6583, acc-0.8600, valid loss-0.3534, acc-0.8996, test loss-0.3592, acc-0.8962
Iter-96690, train loss-0.5166, acc-0.9200, valid loss-0.3534, acc-0.8996, test loss-0.3592, acc-0.8962
Iter-96700, train loss-0.4996, acc-0.8600, valid loss-0.3534, acc-0.8996, test loss-0.3592, acc-0.8962
Iter-96710, train loss-0.2405, acc-0.9600, valid loss-0.3534, acc-0.8996, test loss-0.3591, acc-0.8963
Iter-96720, train loss-0.3615, acc-0.8600, valid loss-0.3534, acc-0.8996, test loss-0.3591, acc-0.8963
Iter-96730, train loss-0.4972, acc-0.8600, valid loss-0.3534, acc-0.8996, test loss-0.3591, acc-0.8963
Iter-96740, train loss-0.3785, acc-0.8800, valid loss-0.3534, acc-0.8996, test loss-0.3591, acc-0.8963
Iter-96750, train loss-0.3775, acc-0.8600, valid loss-0.3534, acc-0.8996, test loss-0.3591, acc-0.8963
Iter-96760, train loss-0.3374, acc-0.8800, valid loss-0.3534, acc-0.8996, test loss-0.3591, acc-0.8963
Iter-96770, train loss-0.5913, acc-0.8000, valid loss-0.3533, acc-0.8996, test loss-0.3591, acc-0.8963
Iter-96780, train loss-0.2449, acc-0.9400, valid loss-0.3533, acc-0.8996, test loss-0.3591, acc-0.8963
Iter-96790, train loss-0.2219, acc-0.9200, valid loss-0.3533, acc-0.8996, test loss-0.3591, acc-0.8963
Iter-96800, train loss-0.2751, acc-0.8800, valid loss-0.3533, acc-0.8996, test loss-0.3591, acc-0.8963
Iter-96810, train loss-0.5264, acc-0.8600, valid loss-0.3533, acc-0.8998, test loss-0.3591, acc-0.8963
Iter-96820, train loss-0.4075, acc-0.9000, valid loss-0.3533, acc-0.8998, test loss-0.3591, acc-0.8963
Iter-96830, train loss-0.3493, acc-0.9400, valid loss-0.3533, acc-0.8998, test loss-0.3590, acc-0.8963
Iter-96840, train loss-0.3539, acc-0.9000, valid loss-0.3533, acc-0.8998, test loss-0.3590, acc-0.8963
Iter-96850, train loss-0.4904, acc-0.8800, valid loss-0.3533, acc-0.8998, test loss-0.3590, acc-0.8963
Iter-96860, train loss-0.2490, acc-0.9600, valid loss-0.3533, acc-0.8998, test loss-0.3590, acc-0.8963
Iter-96870, train loss-0.3083, acc-0.9200, valid loss-0.3533, acc-0.8998, test loss-0.3590, acc-0.8963
Iter-96880, train loss-0.4623, acc-0.9000, valid loss-0.3533, acc-0.9000, test loss-0.3590, acc-0.8963
Iter-96890, train loss-0.4565, acc-0.8600, valid loss-0.3532, acc-0.8998, test loss-0.3590, acc-0.8964
Iter-96900, train loss-0.3550, acc-0.8800, valid loss-0.3532, acc-0.8998, test loss-0.3590, acc-0.8964
Iter-96910, train loss-0.2253, acc-0.9400, valid loss-0.3532, acc-0.9000, test loss-0.3590, acc-0.8963
Iter-96920, train loss-0.2674, acc-0.9200, valid loss-0.3532, acc-0.9000, test loss-0.3590, acc-0.8963
Iter-96930, train loss-0.3241, acc-0.9200, valid loss-0.3532, acc-0.9002, test loss-0.3589, acc-0.8964
Iter-96940, train loss-0.5724, acc-0.8400, valid loss-0.3532, acc-0.9002, test loss-0.3589, acc-0.8963
Iter-96950, train loss-0.2929, acc-0.9400, valid loss-0.3532, acc-0.9002, test loss-0.3589, acc-0.8963
Iter-96960, train loss-0.3932, acc-0.9400, valid loss-0.3531, acc-0.9002, test loss-0.3589, acc-0.8963
Iter-96970, train loss-0.1667, acc-0.9800, valid loss-0.3531, acc-0.9002, test loss-0.3589, acc-0.8964
Iter-96980, train loss-0.3576, acc-0.9200, valid loss-0.3531, acc-0.9002, test loss-0.3589, acc-0.8964
Iter-96990, train loss-0.1873, acc-0.9800, valid loss-0.3531, acc-0.9002, test loss-0.3589, acc-0.8963
Iter-97000, train loss-0.2107, acc-0.9400, valid loss-0.3531, acc-0.9002, test loss-0.3589, acc-0.8964
Iter-97010, train loss-0.2287, acc-0.9400, valid loss-0.3531, acc-0.9002, test loss-0.3589, acc-0.8965
Iter-97020, train loss-0.4711, acc-0.8400, valid loss-0.3531, acc-0.9002, test loss-0.3588, acc-0.8964
Iter-97030, train loss-0.4120, acc-0.9000, valid loss-0.3531, acc-0.9002, test loss-0.3588, acc-0.8964
Iter-97040, train loss-0.2397, acc-0.9400, valid loss-0.3531, acc-0.9002, test loss-0.3588, acc-0.8964
Iter-97050, train loss-0.2613, acc-0.9000, valid loss-0.3531, acc-0.9002, test loss-0.3588, acc-0.8963
Iter-97060, train loss-0.3549, acc-0.8800, valid loss-0.3530, acc-0.9000, test loss-0.3588, acc-0.8963
Iter-97070, train loss-0.3427, acc-0.8800, valid loss-0.3530, acc-0.9000, test loss-0.3588, acc-0.8963
Iter-97080, train loss-0.3354, acc-0.9400, valid loss-0.3530, acc-0.9000, test loss-0.3588, acc-0.8964
Iter-97090, train loss-0.3301, acc-0.9000, valid loss-0.3530, acc-0.8998, test loss-0.3588, acc-0.8963
Iter-97100, train loss-0.4011, acc-0.9000, valid loss-0.3530, acc-0.8998, test loss-0.3588, acc-0.8963
Iter-97110, train loss-0.5453, acc-0.8600, valid loss-0.3530, acc-0.9000, test loss-0.3588, acc-0.8963
Iter-97120, train loss-0.3920, acc-0.8600, valid loss-0.3530, acc-0.8996, test loss-0.3587, acc-0.8963
Iter-97130, train loss-0.4395, acc-0.9200, valid loss-0.3530, acc-0.8998, test loss-0.3587, acc-0.8963
Iter-97140, train loss-0.4028, acc-0.8400, valid loss-0.3530, acc-0.8998, test loss-0.3587, acc-0.8963
Iter-97150, train loss-0.3595, acc-0.8800, valid loss-0.3530, acc-0.8998, test loss-0.3587, acc-0.8963
Iter-97160, train loss-0.5347, acc-0.8600, valid loss-0.3530, acc-0.9000, test loss-0.3587, acc-0.8963
Iter-97170, train loss-0.2261, acc-0.9600, valid loss-0.3529, acc-0.9000, test loss-0.3587, acc-0.8963
Iter-97180, train loss-0.3892, acc-0.9200, valid loss-0.3529, acc-0.8998, test loss-0.3587, acc-0.8963
Iter-97190, train loss-0.2518, acc-0.9400, valid loss-0.3529, acc-0.8996, test loss-0.3587, acc-0.8963
Iter-97200, train loss-0.4659, acc-0.8600, valid loss-0.3529, acc-0.8998, test loss-0.3587, acc-0.8963
Iter-97210, train loss-0.2294, acc-0.9400, valid loss-0.3529, acc-0.8998, test loss-0.3586, acc-0.8963
Iter-97220, train loss-0.2511, acc-0.9600, valid loss-0.3529, acc-0.8998, test loss-0.3586, acc-0.8963
Iter-97230, train loss-0.6898, acc-0.8000, valid loss-0.3529, acc-0.8998, test loss-0.3586, acc-0.8963
Iter-97240, train loss-0.2876, acc-0.9000, valid loss-0.3529, acc-0.9000, test loss-0.3586, acc-0.8963
Iter-97250, train loss-0.2316, acc-0.9200, valid loss-0.3528, acc-0.9000, test loss-0.3586, acc-0.8963
Iter-97260, train loss-0.2921, acc-0.9400, valid loss-0.3528, acc-0.9000, test loss-0.3586, acc-0.8963
Iter-97270, train loss-0.5621, acc-0.8400, valid loss-0.3528, acc-0.9000, test loss-0.3586, acc-0.8962
Iter-97280, train loss-0.5677, acc-0.8200, valid loss-0.3528, acc-0.9000, test loss-0.3586, acc-0.8962
Iter-97290, train loss-0.3122, acc-0.8400, valid loss-0.3528, acc-0.9002, test loss-0.3586, acc-0.8962
Iter-97300, train loss-0.3891, acc-0.9000, valid loss-0.3528, acc-0.9000, test loss-0.3585, acc-0.8963
Iter-97310, train loss-0.3886, acc-0.8800, valid loss-0.3528, acc-0.9000, test loss-0.3585, acc-0.8963
Iter-97320, train loss-0.4459, acc-0.8800, valid loss-0.3528, acc-0.9000, test loss-0.3585, acc-0.8963
Iter-97330, train loss-0.3560, acc-0.9000, valid loss-0.3528, acc-0.8998, test loss-0.3585, acc-0.8963
Iter-97340, train loss-0.5615, acc-0.8200, valid loss-0.3527, acc-0.9000, test loss-0.3585, acc-0.8963
Iter-97350, train loss-0.3447, acc-0.8600, valid loss-0.3527, acc-0.9000, test loss-0.3585, acc-0.8962
Iter-97360, train loss-0.3517, acc-0.9400, valid loss-0.3527, acc-0.9000, test loss-0.3585, acc-0.8961
Iter-97370, train loss-0.4953, acc-0.8800, valid loss-0.3527, acc-0.9000, test loss-0.3585, acc-0.8961
Iter-97380, train loss-0.6968, acc-0.7600, valid loss-0.3527, acc-0.9000, test loss-0.3584, acc-0.8961
Iter-97390, train loss-0.3149, acc-0.9200, valid loss-0.3527, acc-0.9000, test loss-0.3584, acc-0.8962
Iter-97400, train loss-0.3520, acc-0.9200, valid loss-0.3527, acc-0.9002, test loss-0.3584, acc-0.8962
Iter-97410, train loss-0.3449, acc-0.9200, valid loss-0.3527, acc-0.9002, test loss-0.3584, acc-0.8962
Iter-97420, train loss-0.2012, acc-0.9200, valid loss-0.3527, acc-0.9000, test loss-0.3584, acc-0.8963
Iter-97430, train loss-0.3837, acc-0.9200, valid loss-0.3526, acc-0.9000, test loss-0.3584, acc-0.8962
Iter-97440, train loss-0.4651, acc-0.8200, valid loss-0.3526, acc-0.9000, test loss-0.3584, acc-0.8962
Iter-97450, train loss-0.6770, acc-0.8000, valid loss-0.3526, acc-0.9000, test loss-0.3584, acc-0.8962
Iter-97460, train loss-0.2231, acc-0.9200, valid loss-0.3526, acc-0.9000, test loss-0.3584, acc-0.8962
Iter-97470, train loss-0.2957, acc-0.9200, valid loss-0.3526, acc-0.9000, test loss-0.3584, acc-0.8962
Iter-97480, train loss-0.4468, acc-0.8800, valid loss-0.3526, acc-0.8998, test loss-0.3583, acc-0.8962
Iter-97490, train loss-0.5084, acc-0.8800, valid loss-0.3526, acc-0.8998, test loss-0.3583, acc-0.8962
Iter-97500, train loss-0.5085, acc-0.8600, valid loss-0.3526, acc-0.8998, test loss-0.3583, acc-0.8963
Iter-97510, train loss-0.3177, acc-0.8800, valid loss-0.3525, acc-0.9000, test loss-0.3583, acc-0.8963
Iter-97520, train loss-0.5634, acc-0.7800, valid loss-0.3525, acc-0.9000, test loss-0.3583, acc-0.8963
Iter-97530, train loss-0.3006, acc-0.8800, valid loss-0.3525, acc-0.9000, test loss-0.3583, acc-0.8963
Iter-97540, train loss-0.4056, acc-0.8400, valid loss-0.3525, acc-0.9000, test loss-0.3583, acc-0.8963
Iter-97550, train loss-0.3213, acc-0.9400, valid loss-0.3525, acc-0.9000, test loss-0.3583, acc-0.8963
Iter-97560, train loss-0.2839, acc-0.9400, valid loss-0.3525, acc-0.9000, test loss-0.3582, acc-0.8963
Iter-97570, train loss-0.2455, acc-0.9600, valid loss-0.3525, acc-0.9002, test loss-0.3582, acc-0.8964
Iter-97580, train loss-0.4963, acc-0.8800, valid loss-0.3524, acc-0.9002, test loss-0.3582, acc-0.8963
Iter-97590, train loss-0.3821, acc-0.8800, valid loss-0.3524, acc-0.9000, test loss-0.3582, acc-0.8963
Iter-97600, train loss-0.4217, acc-0.8800, valid loss-0.3524, acc-0.9002, test loss-0.3582, acc-0.8964
Iter-97610, train loss-0.3561, acc-0.8600, valid loss-0.3524, acc-0.9002, test loss-0.3582, acc-0.8964
Iter-97620, train loss-0.4101, acc-0.8600, valid loss-0.3524, acc-0.9000, test loss-0.3582, acc-0.8963
Iter-97630, train loss-0.3453, acc-0.9200, valid loss-0.3524, acc-0.9000, test loss-0.3582, acc-0.8964
Iter-97640, train loss-0.4319, acc-0.9000, valid loss-0.3524, acc-0.8998, test loss-0.3581, acc-0.8964
Iter-97650, train loss-0.3045, acc-0.9400, valid loss-0.3524, acc-0.8996, test loss-0.3581, acc-0.8964
Iter-97660, train loss-0.5303, acc-0.8200, valid loss-0.3524, acc-0.9000, test loss-0.3581, acc-0.8964
Iter-97670, train loss-0.3283, acc-0.8800, valid loss-0.3524, acc-0.9000, test loss-0.3581, acc-0.8963
Iter-97680, train loss-0.4035, acc-0.9200, valid loss-0.3524, acc-0.8998, test loss-0.3581, acc-0.8963
Iter-97690, train loss-0.3441, acc-0.8800, valid loss-0.3523, acc-0.8998, test loss-0.3581, acc-0.8963
Iter-97700, train loss-0.4552, acc-0.9000, valid loss-0.3523, acc-0.8996, test loss-0.3581, acc-0.8963
Iter-97710, train loss-0.1881, acc-0.9600, valid loss-0.3523, acc-0.8998, test loss-0.3581, acc-0.8965
Iter-97720, train loss-0.1964, acc-0.9800, valid loss-0.3523, acc-0.8998, test loss-0.3581, acc-0.8965
Iter-97730, train loss-0.2816, acc-0.8600, valid loss-0.3523, acc-0.8996, test loss-0.3580, acc-0.8964
Iter-97740, train loss-0.5303, acc-0.8600, valid loss-0.3523, acc-0.8996, test loss-0.3580, acc-0.8963
Iter-97750, train loss-0.1919, acc-0.9600, valid loss-0.3523, acc-0.8996, test loss-0.3580, acc-0.8962
Iter-97760, train loss-0.2253, acc-0.9400, valid loss-0.3523, acc-0.8996, test loss-0.3580, acc-0.8963
Iter-97770, train loss-0.1746, acc-0.9400, valid loss-0.3522, acc-0.8996, test loss-0.3580, acc-0.8963
Iter-97780, train loss-0.4141, acc-0.8400, valid loss-0.3522, acc-0.8998, test loss-0.3580, acc-0.8963
Iter-97790, train loss-0.2197, acc-0.9600, valid loss-0.3522, acc-0.8998, test loss-0.3580, acc-0.8964
Iter-97800, train loss-0.1936, acc-1.0000, valid loss-0.3522, acc-0.8998, test loss-0.3580, acc-0.8965
Iter-97810, train loss-0.3312, acc-0.9000, valid loss-0.3522, acc-0.8998, test loss-0.3580, acc-0.8965
Iter-97820, train loss-0.4368, acc-0.8800, valid loss-0.3522, acc-0.9000, test loss-0.3580, acc-0.8964
Iter-97830, train loss-0.2294, acc-0.9200, valid loss-0.3522, acc-0.9002, test loss-0.3579, acc-0.8964
Iter-97840, train loss-0.3814, acc-0.8600, valid loss-0.3522, acc-0.9002, test loss-0.3579, acc-0.8964
Iter-97850, train loss-0.3006, acc-0.9000, valid loss-0.3521, acc-0.9002, test loss-0.3579, acc-0.8964
Iter-97860, train loss-0.4090, acc-0.8600, valid loss-0.3521, acc-0.9000, test loss-0.3579, acc-0.8964
Iter-97870, train loss-0.3224, acc-0.9600, valid loss-0.3521, acc-0.9000, test loss-0.3579, acc-0.8963
Iter-97880, train loss-0.3292, acc-0.9000, valid loss-0.3521, acc-0.9000, test loss-0.3579, acc-0.8964
Iter-97890, train loss-0.5803, acc-0.8200, valid loss-0.3521, acc-0.9000, test loss-0.3579, acc-0.8964
Iter-97900, train loss-0.4702, acc-0.8600, valid loss-0.3521, acc-0.8998, test loss-0.3579, acc-0.8964
Iter-97910, train loss-0.4718, acc-0.8600, valid loss-0.3521, acc-0.8998, test loss-0.3579, acc-0.8963
Iter-97920, train loss-0.1888, acc-0.9800, valid loss-0.3521, acc-0.9000, test loss-0.3579, acc-0.8963
Iter-97930, train loss-0.3368, acc-0.8800, valid loss-0.3521, acc-0.9000, test loss-0.3579, acc-0.8964
Iter-97940, train loss-0.2552, acc-0.9600, valid loss-0.3520, acc-0.9000, test loss-0.3578, acc-0.8964
Iter-97950, train loss-0.4250, acc-0.8600, valid loss-0.3520, acc-0.9000, test loss-0.3578, acc-0.8965
Iter-97960, train loss-0.3528, acc-0.9000, valid loss-0.3520, acc-0.8998, test loss-0.3578, acc-0.8965
Iter-97970, train loss-0.3135, acc-0.8800, valid loss-0.3520, acc-0.8998, test loss-0.3578, acc-0.8964
Iter-97980, train loss-0.4203, acc-0.8800, valid loss-0.3520, acc-0.8998, test loss-0.3578, acc-0.8965
Iter-97990, train loss-0.2749, acc-0.9400, valid loss-0.3520, acc-0.9000, test loss-0.3578, acc-0.8965
Iter-98000, train loss-0.3458, acc-0.9200, valid loss-0.3520, acc-0.8998, test loss-0.3578, acc-0.8965
Iter-98010, train loss-0.3568, acc-0.9200, valid loss-0.3520, acc-0.8998, test loss-0.3578, acc-0.8965
Iter-98020, train loss-0.2482, acc-0.9400, valid loss-0.3519, acc-0.9000, test loss-0.3578, acc-0.8964
Iter-98030, train loss-0.2559, acc-0.9000, valid loss-0.3519, acc-0.9000, test loss-0.3577, acc-0.8964
Iter-98040, train loss-0.2836, acc-0.9000, valid loss-0.3519, acc-0.9000, test loss-0.3577, acc-0.8965
Iter-98050, train loss-0.4653, acc-0.9000, valid loss-0.3519, acc-0.9000, test loss-0.3577, acc-0.8965
Iter-98060, train loss-0.3368, acc-0.9400, valid loss-0.3519, acc-0.8998, test loss-0.3577, acc-0.8964
Iter-98070, train loss-0.4542, acc-0.9000, valid loss-0.3519, acc-0.8998, test loss-0.3577, acc-0.8964
Iter-98080, train loss-0.3888, acc-0.8600, valid loss-0.3519, acc-0.8998, test loss-0.3577, acc-0.8965
Iter-98090, train loss-0.3261, acc-0.9400, valid loss-0.3518, acc-0.8998, test loss-0.3577, acc-0.8965
Iter-98100, train loss-0.2114, acc-0.9400, valid loss-0.3518, acc-0.9000, test loss-0.3577, acc-0.8965
Iter-98110, train loss-0.3122, acc-0.9000, valid loss-0.3518, acc-0.9000, test loss-0.3576, acc-0.8964
Iter-98120, train loss-0.4168, acc-0.9000, valid loss-0.3518, acc-0.9000, test loss-0.3576, acc-0.8965
Iter-98130, train loss-0.3802, acc-0.9200, valid loss-0.3518, acc-0.9002, test loss-0.3576, acc-0.8963
Iter-98140, train loss-0.3707, acc-0.8800, valid loss-0.3518, acc-0.9000, test loss-0.3576, acc-0.8964
Iter-98150, train loss-0.4293, acc-0.8800, valid loss-0.3518, acc-0.9000, test loss-0.3576, acc-0.8964
Iter-98160, train loss-0.3645, acc-0.9200, valid loss-0.3518, acc-0.9000, test loss-0.3576, acc-0.8964
Iter-98170, train loss-0.2582, acc-0.9400, valid loss-0.3518, acc-0.9000, test loss-0.3576, acc-0.8964
Iter-98180, train loss-0.3913, acc-0.9200, valid loss-0.3518, acc-0.9000, test loss-0.3576, acc-0.8964
Iter-98190, train loss-0.3857, acc-0.9200, valid loss-0.3518, acc-0.8998, test loss-0.3576, acc-0.8964
Iter-98200, train loss-0.4098, acc-0.8800, valid loss-0.3518, acc-0.9000, test loss-0.3576, acc-0.8963
Iter-98210, train loss-0.3603, acc-0.8600, valid loss-0.3517, acc-0.8998, test loss-0.3575, acc-0.8964
Iter-98220, train loss-0.3704, acc-0.8400, valid loss-0.3517, acc-0.8998, test loss-0.3575, acc-0.8964
Iter-98230, train loss-0.2373, acc-0.9400, valid loss-0.3517, acc-0.9000, test loss-0.3575, acc-0.8963
Iter-98240, train loss-0.4417, acc-0.8400, valid loss-0.3517, acc-0.9000, test loss-0.3575, acc-0.8963
Iter-98250, train loss-0.3122, acc-0.8800, valid loss-0.3517, acc-0.9000, test loss-0.3575, acc-0.8963
Iter-98260, train loss-0.2950, acc-0.9200, valid loss-0.3517, acc-0.9000, test loss-0.3575, acc-0.8964
Iter-98270, train loss-0.2479, acc-0.9600, valid loss-0.3517, acc-0.9000, test loss-0.3575, acc-0.8963
Iter-98280, train loss-0.3553, acc-0.9000, valid loss-0.3516, acc-0.9000, test loss-0.3574, acc-0.8963
Iter-98290, train loss-0.3816, acc-0.8400, valid loss-0.3516, acc-0.9000, test loss-0.3575, acc-0.8964
Iter-98300, train loss-0.1802, acc-0.9800, valid loss-0.3516, acc-0.9000, test loss-0.3574, acc-0.8964
Iter-98310, train loss-0.2854, acc-0.9400, valid loss-0.3516, acc-0.9002, test loss-0.3574, acc-0.8964
Iter-98320, train loss-0.2918, acc-0.9000, valid loss-0.3516, acc-0.9002, test loss-0.3574, acc-0.8963
Iter-98330, train loss-0.3434, acc-0.8800, valid loss-0.3516, acc-0.9002, test loss-0.3574, acc-0.8963
Iter-98340, train loss-0.2022, acc-0.9600, valid loss-0.3515, acc-0.9002, test loss-0.3574, acc-0.8963
Iter-98350, train loss-0.4870, acc-0.7600, valid loss-0.3515, acc-0.9002, test loss-0.3574, acc-0.8963
Iter-98360, train loss-0.2631, acc-0.9000, valid loss-0.3515, acc-0.9002, test loss-0.3574, acc-0.8963
Iter-98370, train loss-0.2747, acc-0.9400, valid loss-0.3515, acc-0.9002, test loss-0.3574, acc-0.8962
Iter-98380, train loss-0.7662, acc-0.8000, valid loss-0.3515, acc-0.9002, test loss-0.3573, acc-0.8962
Iter-98390, train loss-0.4902, acc-0.8400, valid loss-0.3515, acc-0.9002, test loss-0.3573, acc-0.8962
Iter-98400, train loss-0.3631, acc-0.9000, valid loss-0.3515, acc-0.9002, test loss-0.3573, acc-0.8962
Iter-98410, train loss-0.4058, acc-0.8800, valid loss-0.3515, acc-0.9000, test loss-0.3573, acc-0.8964
Iter-98420, train loss-0.4903, acc-0.8200, valid loss-0.3515, acc-0.9000, test loss-0.3573, acc-0.8964
Iter-98430, train loss-0.3629, acc-0.9400, valid loss-0.3514, acc-0.9000, test loss-0.3573, acc-0.8965
Iter-98440, train loss-0.2239, acc-0.9200, valid loss-0.3514, acc-0.9002, test loss-0.3573, acc-0.8963
Iter-98450, train loss-0.2421, acc-0.9400, valid loss-0.3514, acc-0.9000, test loss-0.3572, acc-0.8962
Iter-98460, train loss-0.4296, acc-0.8600, valid loss-0.3514, acc-0.9000, test loss-0.3572, acc-0.8963
Iter-98470, train loss-0.3865, acc-0.9200, valid loss-0.3514, acc-0.9000, test loss-0.3572, acc-0.8963
Iter-98480, train loss-0.4290, acc-0.8400, valid loss-0.3514, acc-0.9000, test loss-0.3572, acc-0.8963
Iter-98490, train loss-0.3085, acc-0.9200, valid loss-0.3514, acc-0.9002, test loss-0.3572, acc-0.8963
Iter-98500, train loss-0.3537, acc-0.9200, valid loss-0.3514, acc-0.9002, test loss-0.3572, acc-0.8965
Iter-98510, train loss-0.5008, acc-0.8600, valid loss-0.3514, acc-0.9002, test loss-0.3572, acc-0.8965
Iter-98520, train loss-0.3006, acc-0.9200, valid loss-0.3514, acc-0.9002, test loss-0.3572, acc-0.8964
Iter-98530, train loss-0.4161, acc-0.8400, valid loss-0.3514, acc-0.9002, test loss-0.3572, acc-0.8963
Iter-98540, train loss-0.5872, acc-0.8600, valid loss-0.3514, acc-0.9002, test loss-0.3571, acc-0.8963
Iter-98550, train loss-0.3542, acc-0.9200, valid loss-0.3513, acc-0.9000, test loss-0.3571, acc-0.8962
Iter-98560, train loss-0.3714, acc-0.9200, valid loss-0.3513, acc-0.9002, test loss-0.3571, acc-0.8963
Iter-98570, train loss-0.4056, acc-0.8600, valid loss-0.3513, acc-0.9002, test loss-0.3571, acc-0.8962
Iter-98580, train loss-0.4512, acc-0.8400, valid loss-0.3513, acc-0.9002, test loss-0.3571, acc-0.8962
Iter-98590, train loss-0.7216, acc-0.8200, valid loss-0.3513, acc-0.9002, test loss-0.3571, acc-0.8962
Iter-98600, train loss-0.2255, acc-0.9000, valid loss-0.3513, acc-0.9002, test loss-0.3571, acc-0.8963
Iter-98610, train loss-0.1510, acc-0.9800, valid loss-0.3513, acc-0.9002, test loss-0.3571, acc-0.8963
Iter-98620, train loss-0.4072, acc-0.8600, valid loss-0.3513, acc-0.9002, test loss-0.3571, acc-0.8963
Iter-98630, train loss-0.3766, acc-0.9200, valid loss-0.3513, acc-0.9002, test loss-0.3570, acc-0.8963
Iter-98640, train loss-0.5399, acc-0.8400, valid loss-0.3513, acc-0.9002, test loss-0.3570, acc-0.8963
Iter-98650, train loss-0.3660, acc-0.9000, valid loss-0.3513, acc-0.9002, test loss-0.3570, acc-0.8964
Iter-98660, train loss-0.3601, acc-0.9200, valid loss-0.3512, acc-0.9002, test loss-0.3570, acc-0.8962
Iter-98670, train loss-0.2593, acc-0.9400, valid loss-0.3512, acc-0.9002, test loss-0.3570, acc-0.8962
Iter-98680, train loss-0.3845, acc-0.9000, valid loss-0.3512, acc-0.9002, test loss-0.3570, acc-0.8962
Iter-98690, train loss-0.2025, acc-0.9600, valid loss-0.3512, acc-0.9002, test loss-0.3570, acc-0.8962
Iter-98700, train loss-0.3599, acc-0.8800, valid loss-0.3512, acc-0.9002, test loss-0.3570, acc-0.8962
Iter-98710, train loss-0.2719, acc-0.9600, valid loss-0.3512, acc-0.9002, test loss-0.3569, acc-0.8963
Iter-98720, train loss-0.2991, acc-0.9600, valid loss-0.3512, acc-0.9002, test loss-0.3569, acc-0.8964
Iter-98730, train loss-0.4579, acc-0.8600, valid loss-0.3511, acc-0.9002, test loss-0.3569, acc-0.8963
Iter-98740, train loss-0.2752, acc-0.9000, valid loss-0.3511, acc-0.9002, test loss-0.3569, acc-0.8964
Iter-98750, train loss-0.2799, acc-0.9200, valid loss-0.3511, acc-0.9002, test loss-0.3569, acc-0.8964
Iter-98760, train loss-0.2004, acc-0.9600, valid loss-0.3511, acc-0.9002, test loss-0.3569, acc-0.8964
Iter-98770, train loss-0.3023, acc-0.9000, valid loss-0.3511, acc-0.9002, test loss-0.3569, acc-0.8964
Iter-98780, train loss-0.4319, acc-0.8600, valid loss-0.3511, acc-0.9002, test loss-0.3569, acc-0.8964
Iter-98790, train loss-0.4224, acc-0.8800, valid loss-0.3511, acc-0.9002, test loss-0.3568, acc-0.8964
Iter-98800, train loss-0.2380, acc-0.9400, valid loss-0.3511, acc-0.9002, test loss-0.3568, acc-0.8963
Iter-98810, train loss-0.4929, acc-0.8600, valid loss-0.3510, acc-0.9002, test loss-0.3568, acc-0.8963
Iter-98820, train loss-0.3124, acc-0.8800, valid loss-0.3510, acc-0.9002, test loss-0.3568, acc-0.8963
Iter-98830, train loss-0.4882, acc-0.8800, valid loss-0.3510, acc-0.9002, test loss-0.3568, acc-0.8964
Iter-98840, train loss-0.4445, acc-0.8600, valid loss-0.3510, acc-0.9002, test loss-0.3568, acc-0.8963
Iter-98850, train loss-0.3111, acc-0.9000, valid loss-0.3510, acc-0.9002, test loss-0.3568, acc-0.8963
Iter-98860, train loss-0.3393, acc-0.9000, valid loss-0.3510, acc-0.9002, test loss-0.3568, acc-0.8963
Iter-98870, train loss-0.4744, acc-0.8800, valid loss-0.3510, acc-0.9002, test loss-0.3567, acc-0.8963
Iter-98880, train loss-0.4258, acc-0.8600, valid loss-0.3510, acc-0.9002, test loss-0.3567, acc-0.8964
Iter-98890, train loss-0.4110, acc-0.8600, valid loss-0.3510, acc-0.9002, test loss-0.3567, acc-0.8964
Iter-98900, train loss-0.4288, acc-0.8800, valid loss-0.3509, acc-0.9002, test loss-0.3567, acc-0.8964
Iter-98910, train loss-0.2424, acc-0.9800, valid loss-0.3509, acc-0.9002, test loss-0.3567, acc-0.8964
Iter-98920, train loss-0.4299, acc-0.9200, valid loss-0.3509, acc-0.9000, test loss-0.3567, acc-0.8964
Iter-98930, train loss-0.4516, acc-0.9000, valid loss-0.3509, acc-0.9002, test loss-0.3567, acc-0.8964
Iter-98940, train loss-0.4173, acc-0.8400, valid loss-0.3509, acc-0.9002, test loss-0.3566, acc-0.8964
Iter-98950, train loss-0.4661, acc-0.8800, valid loss-0.3509, acc-0.9002, test loss-0.3566, acc-0.8963
Iter-98960, train loss-0.4984, acc-0.8400, valid loss-0.3508, acc-0.9002, test loss-0.3566, acc-0.8964
Iter-98970, train loss-0.3707, acc-0.9200, valid loss-0.3508, acc-0.9002, test loss-0.3566, acc-0.8965
Iter-98980, train loss-0.2809, acc-0.9400, valid loss-0.3508, acc-0.9002, test loss-0.3566, acc-0.8966
Iter-98990, train loss-0.2917, acc-0.9000, valid loss-0.3508, acc-0.9002, test loss-0.3566, acc-0.8966
Iter-99000, train loss-0.1679, acc-0.9800, valid loss-0.3508, acc-0.9002, test loss-0.3566, acc-0.8965
Iter-99010, train loss-0.3398, acc-0.9200, valid loss-0.3508, acc-0.9002, test loss-0.3566, acc-0.8964
Iter-99020, train loss-0.6595, acc-0.7800, valid loss-0.3508, acc-0.9002, test loss-0.3566, acc-0.8964
Iter-99030, train loss-0.1062, acc-0.9800, valid loss-0.3508, acc-0.9002, test loss-0.3565, acc-0.8965
Iter-99040, train loss-0.2866, acc-0.9400, valid loss-0.3507, acc-0.9002, test loss-0.3565, acc-0.8965
Iter-99050, train loss-0.4262, acc-0.8400, valid loss-0.3507, acc-0.9002, test loss-0.3565, acc-0.8965
Iter-99060, train loss-0.3354, acc-0.9000, valid loss-0.3507, acc-0.9002, test loss-0.3565, acc-0.8966
Iter-99070, train loss-0.3341, acc-0.9200, valid loss-0.3507, acc-0.9002, test loss-0.3565, acc-0.8964
Iter-99080, train loss-0.2393, acc-0.9200, valid loss-0.3507, acc-0.9002, test loss-0.3565, acc-0.8964
Iter-99090, train loss-0.4450, acc-0.8800, valid loss-0.3507, acc-0.9002, test loss-0.3565, acc-0.8964
Iter-99100, train loss-0.3468, acc-0.8800, valid loss-0.3507, acc-0.9002, test loss-0.3565, acc-0.8964
Iter-99110, train loss-0.4833, acc-0.8600, valid loss-0.3507, acc-0.9002, test loss-0.3565, acc-0.8964
Iter-99120, train loss-0.5984, acc-0.7800, valid loss-0.3506, acc-0.9002, test loss-0.3564, acc-0.8964
Iter-99130, train loss-0.4335, acc-0.9000, valid loss-0.3506, acc-0.9002, test loss-0.3564, acc-0.8966
Iter-99140, train loss-0.1991, acc-0.9400, valid loss-0.3506, acc-0.9002, test loss-0.3564, acc-0.8968
Iter-99150, train loss-0.5844, acc-0.8000, valid loss-0.3506, acc-0.9002, test loss-0.3564, acc-0.8968
Iter-99160, train loss-0.3582, acc-0.9000, valid loss-0.3506, acc-0.9002, test loss-0.3564, acc-0.8967
Iter-99170, train loss-0.4495, acc-0.8800, valid loss-0.3506, acc-0.9002, test loss-0.3564, acc-0.8966
Iter-99180, train loss-0.2864, acc-0.9000, valid loss-0.3506, acc-0.9002, test loss-0.3564, acc-0.8966
Iter-99190, train loss-0.3163, acc-0.8600, valid loss-0.3506, acc-0.9002, test loss-0.3564, acc-0.8967
Iter-99200, train loss-0.5841, acc-0.8400, valid loss-0.3506, acc-0.9002, test loss-0.3564, acc-0.8966
Iter-99210, train loss-0.3003, acc-0.9200, valid loss-0.3506, acc-0.9002, test loss-0.3564, acc-0.8966
Iter-99220, train loss-0.3266, acc-0.8800, valid loss-0.3506, acc-0.9002, test loss-0.3564, acc-0.8967
Iter-99230, train loss-0.3311, acc-0.9000, valid loss-0.3506, acc-0.9002, test loss-0.3564, acc-0.8967
Iter-99240, train loss-0.3970, acc-0.8800, valid loss-0.3505, acc-0.9002, test loss-0.3563, acc-0.8967
Iter-99250, train loss-0.3056, acc-0.8800, valid loss-0.3505, acc-0.9002, test loss-0.3563, acc-0.8966
Iter-99260, train loss-0.2823, acc-0.8600, valid loss-0.3505, acc-0.9002, test loss-0.3563, acc-0.8966
Iter-99270, train loss-0.4354, acc-0.8600, valid loss-0.3505, acc-0.9002, test loss-0.3563, acc-0.8967
Iter-99280, train loss-0.4074, acc-0.8200, valid loss-0.3505, acc-0.9002, test loss-0.3563, acc-0.8966
Iter-99290, train loss-0.3484, acc-0.8800, valid loss-0.3505, acc-0.9002, test loss-0.3563, acc-0.8966
Iter-99300, train loss-0.3609, acc-0.8600, valid loss-0.3505, acc-0.9002, test loss-0.3563, acc-0.8966
Iter-99310, train loss-0.1850, acc-0.9400, valid loss-0.3505, acc-0.9002, test loss-0.3563, acc-0.8965
Iter-99320, train loss-0.4203, acc-0.8800, valid loss-0.3504, acc-0.9002, test loss-0.3563, acc-0.8965
Iter-99330, train loss-0.6454, acc-0.8400, valid loss-0.3504, acc-0.9002, test loss-0.3563, acc-0.8966
Iter-99340, train loss-0.5705, acc-0.7600, valid loss-0.3504, acc-0.9002, test loss-0.3562, acc-0.8967
Iter-99350, train loss-0.5967, acc-0.8000, valid loss-0.3504, acc-0.9002, test loss-0.3562, acc-0.8967
Iter-99360, train loss-0.3180, acc-0.9200, valid loss-0.3504, acc-0.9002, test loss-0.3562, acc-0.8966
Iter-99370, train loss-0.2450, acc-0.9800, valid loss-0.3504, acc-0.9002, test loss-0.3562, acc-0.8965
Iter-99380, train loss-0.2539, acc-0.9400, valid loss-0.3504, acc-0.9002, test loss-0.3562, acc-0.8966
Iter-99390, train loss-0.6408, acc-0.8400, valid loss-0.3504, acc-0.9002, test loss-0.3562, acc-0.8968
Iter-99400, train loss-0.2514, acc-0.9400, valid loss-0.3504, acc-0.9002, test loss-0.3562, acc-0.8968
Iter-99410, train loss-0.2675, acc-0.9200, valid loss-0.3503, acc-0.9002, test loss-0.3562, acc-0.8967
Iter-99420, train loss-0.2337, acc-0.9600, valid loss-0.3503, acc-0.9002, test loss-0.3562, acc-0.8965
Iter-99430, train loss-0.3471, acc-0.8400, valid loss-0.3503, acc-0.9002, test loss-0.3562, acc-0.8967
Iter-99440, train loss-0.7118, acc-0.7800, valid loss-0.3503, acc-0.9002, test loss-0.3562, acc-0.8966
Iter-99450, train loss-0.2767, acc-0.9000, valid loss-0.3503, acc-0.9002, test loss-0.3562, acc-0.8965
Iter-99460, train loss-0.3167, acc-0.9200, valid loss-0.3503, acc-0.9002, test loss-0.3561, acc-0.8967
Iter-99470, train loss-0.3238, acc-0.9000, valid loss-0.3503, acc-0.9002, test loss-0.3561, acc-0.8967
Iter-99480, train loss-0.4163, acc-0.8400, valid loss-0.3503, acc-0.9002, test loss-0.3561, acc-0.8967
Iter-99490, train loss-0.2325, acc-0.9600, valid loss-0.3502, acc-0.9002, test loss-0.3561, acc-0.8967
Iter-99500, train loss-0.3418, acc-0.9000, valid loss-0.3502, acc-0.9002, test loss-0.3561, acc-0.8967
Iter-99510, train loss-0.3676, acc-0.9200, valid loss-0.3502, acc-0.9002, test loss-0.3561, acc-0.8967
Iter-99520, train loss-0.3098, acc-0.9200, valid loss-0.3502, acc-0.9002, test loss-0.3561, acc-0.8968
Iter-99530, train loss-0.2398, acc-0.9400, valid loss-0.3502, acc-0.9002, test loss-0.3561, acc-0.8968
Iter-99540, train loss-0.2641, acc-0.9600, valid loss-0.3502, acc-0.9002, test loss-0.3561, acc-0.8968
Iter-99550, train loss-0.5983, acc-0.8400, valid loss-0.3502, acc-0.9002, test loss-0.3561, acc-0.8968
Iter-99560, train loss-0.4601, acc-0.8400, valid loss-0.3502, acc-0.9002, test loss-0.3561, acc-0.8967
Iter-99570, train loss-0.5737, acc-0.8200, valid loss-0.3502, acc-0.9002, test loss-0.3561, acc-0.8966
Iter-99580, train loss-0.2721, acc-0.9400, valid loss-0.3502, acc-0.9002, test loss-0.3560, acc-0.8967
Iter-99590, train loss-0.4114, acc-0.8800, valid loss-0.3501, acc-0.9002, test loss-0.3560, acc-0.8967
Iter-99600, train loss-0.4579, acc-0.8200, valid loss-0.3501, acc-0.9002, test loss-0.3560, acc-0.8967
Iter-99610, train loss-0.2935, acc-0.9200, valid loss-0.3501, acc-0.9002, test loss-0.3560, acc-0.8967
Iter-99620, train loss-0.2392, acc-0.9400, valid loss-0.3501, acc-0.9002, test loss-0.3560, acc-0.8967
Iter-99630, train loss-0.4914, acc-0.8400, valid loss-0.3501, acc-0.9002, test loss-0.3560, acc-0.8966
Iter-99640, train loss-0.3124, acc-0.8800, valid loss-0.3501, acc-0.9002, test loss-0.3560, acc-0.8967
Iter-99650, train loss-0.2174, acc-0.9800, valid loss-0.3501, acc-0.9002, test loss-0.3559, acc-0.8967
Iter-99660, train loss-0.5631, acc-0.8200, valid loss-0.3501, acc-0.9002, test loss-0.3559, acc-0.8969
Iter-99670, train loss-0.3824, acc-0.9000, valid loss-0.3501, acc-0.9002, test loss-0.3559, acc-0.8967
Iter-99680, train loss-0.4990, acc-0.8400, valid loss-0.3501, acc-0.9002, test loss-0.3559, acc-0.8966
Iter-99690, train loss-0.3714, acc-0.9400, valid loss-0.3501, acc-0.9002, test loss-0.3559, acc-0.8966
Iter-99700, train loss-0.6874, acc-0.8000, valid loss-0.3500, acc-0.9002, test loss-0.3559, acc-0.8967
Iter-99710, train loss-0.5275, acc-0.8800, valid loss-0.3500, acc-0.9002, test loss-0.3559, acc-0.8966
Iter-99720, train loss-0.4294, acc-0.8600, valid loss-0.3500, acc-0.9002, test loss-0.3559, acc-0.8967
Iter-99730, train loss-0.4400, acc-0.8400, valid loss-0.3500, acc-0.9002, test loss-0.3559, acc-0.8966
Iter-99740, train loss-0.4112, acc-0.9000, valid loss-0.3500, acc-0.9002, test loss-0.3559, acc-0.8966
Iter-99750, train loss-0.3605, acc-0.8800, valid loss-0.3500, acc-0.9002, test loss-0.3559, acc-0.8966
Iter-99760, train loss-0.2375, acc-0.9000, valid loss-0.3500, acc-0.9002, test loss-0.3559, acc-0.8966
Iter-99770, train loss-0.4706, acc-0.8400, valid loss-0.3500, acc-0.9002, test loss-0.3558, acc-0.8967
Iter-99780, train loss-0.2073, acc-0.9000, valid loss-0.3500, acc-0.9002, test loss-0.3558, acc-0.8967
Iter-99790, train loss-0.4395, acc-0.8600, valid loss-0.3500, acc-0.9002, test loss-0.3558, acc-0.8966
Iter-99800, train loss-0.3783, acc-0.8600, valid loss-0.3499, acc-0.9002, test loss-0.3558, acc-0.8966
Iter-99810, train loss-0.5849, acc-0.8400, valid loss-0.3499, acc-0.9002, test loss-0.3558, acc-0.8966
Iter-99820, train loss-0.2913, acc-0.9400, valid loss-0.3499, acc-0.9002, test loss-0.3558, acc-0.8966
Iter-99830, train loss-0.4572, acc-0.8800, valid loss-0.3499, acc-0.9002, test loss-0.3558, acc-0.8966
Iter-99840, train loss-0.2751, acc-0.9200, valid loss-0.3499, acc-0.9002, test loss-0.3558, acc-0.8966
Iter-99850, train loss-0.3420, acc-0.8800, valid loss-0.3499, acc-0.9002, test loss-0.3558, acc-0.8966
Iter-99860, train loss-0.5797, acc-0.8600, valid loss-0.3499, acc-0.9002, test loss-0.3557, acc-0.8967
Iter-99870, train loss-0.4759, acc-0.8800, valid loss-0.3499, acc-0.9002, test loss-0.3557, acc-0.8967
Iter-99880, train loss-0.4145, acc-0.8800, valid loss-0.3499, acc-0.9002, test loss-0.3557, acc-0.8967
Iter-99890, train loss-0.2671, acc-0.9200, valid loss-0.3499, acc-0.9002, test loss-0.3557, acc-0.8967
Iter-99900, train loss-0.3936, acc-0.8200, valid loss-0.3498, acc-0.9002, test loss-0.3557, acc-0.8966
Iter-99910, train loss-0.5475, acc-0.8800, valid loss-0.3498, acc-0.9002, test loss-0.3557, acc-0.8966
Iter-99920, train loss-0.4246, acc-0.9200, valid loss-0.3498, acc-0.9002, test loss-0.3557, acc-0.8966
Iter-99930, train loss-0.3098, acc-0.9400, valid loss-0.3498, acc-0.9002, test loss-0.3557, acc-0.8965
Iter-99940, train loss-0.3403, acc-0.9000, valid loss-0.3498, acc-0.9002, test loss-0.3557, acc-0.8967
Iter-99950, train loss-0.2819, acc-0.9400, valid loss-0.3498, acc-0.9002, test loss-0.3556, acc-0.8967
Iter-99960, train loss-0.4292, acc-0.9200, valid loss-0.3498, acc-0.9002, test loss-0.3556, acc-0.8968
Iter-99970, train loss-0.3666, acc-0.9000, valid loss-0.3498, acc-0.9002, test loss-0.3556, acc-0.8968
Iter-99980, train loss-0.2810, acc-0.9200, valid loss-0.3498, acc-0.9002, test loss-0.3556, acc-0.8968
Iter-99990, train loss-0.4207, acc-0.9200, valid loss-0.3498, acc-0.9002, test loss-0.3556, acc-0.8968
Iter-100000, train loss-0.5564, acc-0.7800, valid loss-0.3497, acc-0.9002, test loss-0.3556, acc-0.8969
In [6]:
# # Display the learning curve and losses for training, validation, and testing
# %matplotlib inline
# %config InlineBackend.figure_format = 'retina'
import matplotlib.pyplot as plt
plt.plot(nn.losses['train'], label='Train loss')
plt.plot(nn.losses['valid'], label='Valid loss')
plt.plot(nn.losses['test'], label='Test loss')
plt.legend()
plt.show()
In [7]:
plt.plot(nn.losses['train_acc'], label='Train accuracy')
plt.plot(nn.losses['valid_acc'], label='Valid accuracy')
plt.plot(nn.losses['test_acc'], label='Test accuracy')
plt.legend()
plt.show()
In [ ]:
In [ ]:
Content source: arasdar/DL
Similar notebooks: