Image Classification

In this project, you'll classify images from the CIFAR-10 dataset. The dataset consists of airplanes, dogs, cats, and other objects. You'll preprocess the images, then train a convolutional neural network on all the samples. The images need to be normalized and the labels need to be one-hot encoded. You'll get to apply what you learned and build a convolutional, max pooling, dropout, and fully connected layers. At the end, you'll get to see your neural network's predictions on the sample images.

Get the Data

Run the following cell to download the CIFAR-10 dataset for python.



In [1]:

    
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
from urllib.request import urlretrieve
from os.path import isfile, isdir
from tqdm import tqdm
import problem_unittests as tests
import tarfile

cifar10_dataset_folder_path = 'cifar-10-batches-py'

# Use Floyd's cifar-10 dataset if present
floyd_cifar10_location = '/input/cifar-10/python.tar.gz'
if isfile(floyd_cifar10_location):
    tar_gz_path = floyd_cifar10_location
else:
    tar_gz_path = 'cifar-10-python.tar.gz'

class DLProgress(tqdm):
    last_block = 0

    def hook(self, block_num=1, block_size=1, total_size=None):
        self.total = total_size
        self.update((block_num - self.last_block) * block_size)
        self.last_block = block_num

if not isfile(tar_gz_path):
    with DLProgress(unit='B', unit_scale=True, miniters=1, desc='CIFAR-10 Dataset') as pbar:
        urlretrieve(
            'https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz',
            tar_gz_path,
            pbar.hook)

if not isdir(cifar10_dataset_folder_path):
    with tarfile.open(tar_gz_path) as tar:
        tar.extractall()
        tar.close()


tests.test_folder_path(cifar10_dataset_folder_path)









    



CIFAR-10 Dataset: 171MB [01:08, 2.50MB/s]                              






    



All files found!

Explore the Data

The dataset is broken into batches to prevent your machine from running out of memory. The CIFAR-10 dataset consists of 5 batches, named data_batch_1, data_batch_2, etc.. Each batch contains the labels and images that are one of the following:

airplane
automobile
bird
cat
deer
dog
frog
horse
ship
truck

Understanding a dataset is part of making predictions on the data. Play around with the code cell below by changing the batch_id and sample_id. The batch_id is the id for a batch (1-5). The sample_id is the id for a image and label pair in the batch.

Ask yourself "What are all possible labels?", "What is the range of values for the image data?", "Are the labels in order or random?". Answers to questions like these will help you preprocess the data and end up with better predictions.



In [12]:

    
%matplotlib inline
%config InlineBackend.figure_format = 'retina'

import helper
import numpy as np

# Explore the dataset
batch_id = 1
sample_id = 5
helper.display_stats(cifar10_dataset_folder_path, batch_id, sample_id)









    



Stats of batch 1:
Samples: 10000
Label Counts: {0: 1005, 1: 974, 2: 1032, 3: 1016, 4: 999, 5: 937, 6: 1030, 7: 1001, 8: 1025, 9: 981}
First 20 Labels: [6, 9, 9, 4, 1, 1, 2, 7, 8, 3, 4, 7, 7, 2, 9, 9, 9, 3, 2, 6]

Example of Image 5:
Image - Min Value: 0 Max Value: 252
Image - Shape: (32, 32, 3)
Label - Label Id: 1 Name: automobile

Implement Preprocess Functions

Normalize

In the cell below, implement the normalize function to take in image data, x, and return it as a normalized Numpy array. The values should be in the range of 0 to 1, inclusive. The return object should be the same shape as x.



In [14]:

    
def normalize(x):
    """
    Normalize a list of sample image data in the range of 0 to 1
    : x: List of image data.  The image shape is (32, 32, 3)
    : return: Numpy array of normalize data
    """
    # TODO: Implement Function
    image_min = 0
    image_max = 255
    return (x - image_min)/(image_max - image_min ) 


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_normalize(normalize)









    



Tests Passed

One-hot encode

Just like the previous code cell, you'll be implementing a function for preprocessing. This time, you'll implement the one_hot_encode function. The input, x, are a list of labels. Implement the function to return the list of labels as One-Hot encoded Numpy array. The possible values for labels are 0 to 9. The one-hot encoding function should return the same encoding for each value between each call to one_hot_encode. Make sure to save the map of encodings outside the function.

Hint: Don't reinvent the wheel.



In [15]:

    
def one_hot_encode(x):
    """
    One hot encode a list of sample labels. Return a one-hot encoded vector for each label.
    : x: List of sample Labels
    : return: Numpy array of one-hot encoded labels
    """
    # TODO: Implement Function
    nb_classes = 10
    one_hot_targets = np.eye(nb_classes)[x]
    return one_hot_targets


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_one_hot_encode(one_hot_encode)









    



Tests Passed

Randomize Data

As you saw from exploring the data above, the order of the samples are randomized. It doesn't hurt to randomize it again, but you don't need to for this dataset.

Preprocess all the data and save it

Running the code cell below will preprocess all the CIFAR-10 data and save it to file. The code below also uses 10% of the training data for validation.



In [16]:

    
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
# Preprocess Training, Validation, and Testing Data
helper.preprocess_and_save_data(cifar10_dataset_folder_path, normalize, one_hot_encode)

Check Point

This is your first checkpoint. If you ever decide to come back to this notebook or have to restart the notebook, you can start from here. The preprocessed data has been saved to disk.



In [1]:

    
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import pickle
import problem_unittests as tests
import helper

# Load the Preprocessed Validation data
valid_features, valid_labels = pickle.load(open('preprocess_validation.p', mode='rb'))

Build the network

For the neural network, you'll build each layer into a function. Most of the code you've seen has been outside of functions. To test your code more thoroughly, we require that you put each layer in a function. This allows us to give you better feedback and test for simple mistakes using our unittests before you submit your project.

Note: If you're finding it hard to dedicate enough time for this course each week, we've provided a small shortcut to this part of the project. In the next couple of problems, you'll have the option to use classes from the TensorFlow Layers or TensorFlow Layers (contrib) packages to build each layer, except the layers you build in the "Convolutional and Max Pooling Layer" section. TF Layers is similar to Keras's and TFLearn's abstraction to layers, so it's easy to pickup.

However, if you would like to get the most out of this course, try to solve all the problems without using anything from the TF Layers packages. You can still use classes from other packages that happen to have the same name as ones you find in TF Layers! For example, instead of using the TF Layers version of the conv2d class, tf.layers.conv2d, you would want to use the TF Neural Network version of conv2d, tf.nn.conv2d.

Let's begin!

Input

The neural network needs to read the image data, one-hot encoded labels, and dropout keep probability. Implement the following functions

Implement neural_net_image_input
- Return a TF Placeholder
- Set the shape using image_shape with batch size set to None.
- Name the TensorFlow placeholder "x" using the TensorFlow name parameter in the TF Placeholder.
Implement neural_net_label_input
- Return a TF Placeholder
- Set the shape using n_classes with batch size set to None.
- Name the TensorFlow placeholder "y" using the TensorFlow name parameter in the TF Placeholder.
Implement neural_net_keep_prob_input
- Return a TF Placeholder for dropout keep probability.
- Name the TensorFlow placeholder "keep_prob" using the TensorFlow name parameter in the TF Placeholder.

These names will be used at the end of the project to load your saved model.

Note: None for shapes in TensorFlow allow for a dynamic size.



In [2]:

    
import tensorflow as tf

def neural_net_image_input(image_shape):
    """
    Return a Tensor for a batch of image input
    : image_shape: Shape of the images
    : return: Tensor for image input.
    """
    # TODO: Implement Function
    return tf.placeholder(tf.float32, [None, *image_shape], name="x")


def neural_net_label_input(n_classes):
    """
    Return a Tensor for a batch of label input
    : n_classes: Number of classes
    : return: Tensor for label input.
    """
    # TODO: Implement Function
    return tf.placeholder(tf.float32, [None, n_classes], name="y")


def neural_net_keep_prob_input():
    """
    Return a Tensor for keep probability
    : return: Tensor for keep probability.
    """
    # TODO: Implement Function
    return tf.placeholder(tf.float32, name="keep_prob")


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tf.reset_default_graph()
tests.test_nn_image_inputs(neural_net_image_input)
tests.test_nn_label_inputs(neural_net_label_input)
tests.test_nn_keep_prob_inputs(neural_net_keep_prob_input)









    



Image Input Tests Passed.
Label Input Tests Passed.
Keep Prob Tests Passed.

Convolution and Max Pooling Layer

Convolution layers have a lot of success with images. For this code cell, you should implement the function conv2d_maxpool to apply convolution then max pooling:

Create the weight and bias using conv_ksize, conv_num_outputs and the shape of x_tensor.
Apply a convolution to x_tensor using weight and conv_strides.
- We recommend you use same padding, but you're welcome to use any padding.
Add bias
Add a nonlinear activation to the convolution.
Apply Max Pooling using pool_ksize and pool_strides.
- We recommend you use same padding, but you're welcome to use any padding.

Note: You can't use TensorFlow Layers or TensorFlow Layers (contrib) for this layer, but you can still use TensorFlow's Neural Network package. You may still use the shortcut option for all the other layers.



In [85]:

    
def conv2d_maxpool(x_tensor, conv_num_outputs, conv_ksize, conv_strides, pool_ksize, pool_strides):
    """
    Apply convolution then max pooling to x_tensor
    :param x_tensor: TensorFlow Tensor
    :param conv_num_outputs: Number of outputs for the convolutional layer
    :param conv_ksize: kernal size 2-D Tuple for the convolutional layer
    :param conv_strides: Stride 2-D Tuple for convolution
    :param pool_ksize: kernal size 2-D Tuple for pool
    :param pool_strides: Stride 2-D Tuple for pool
    : return: A tensor that represents convolution and max pooling of x_tensor
    """
    # TODO: Implement Function
    # Filter (weights and bias)
    W_conv = tf.Variable(tf.truncated_normal((*conv_ksize, x_tensor.get_shape().as_list()[3], conv_num_outputs),
                                            stddev=0.1))
    b_conv = tf.Variable(tf.zeros(conv_num_outputs))
    padding = 'SAME'
    h_conv2d = tf.nn.relu(tf.nn.conv2d(x_tensor, W_conv, [1, *conv_strides, 1], padding) + b_conv)
    return tf.nn.max_pool(h_conv2d, [1, *pool_ksize, 1], [1, *pool_strides, 1], padding) 


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_con_pool(conv2d_maxpool)









    



Tests Passed

Flatten Layer

Implement the flatten function to change the dimension of x_tensor from a 4-D tensor to a 2-D tensor. The output should be the shape (Batch Size, Flattened Image Size). Shortcut option: you can use classes from the TensorFlow Layers or TensorFlow Layers (contrib) packages for this layer. For more of a challenge, only use other TensorFlow packages.



In [86]:

    
def flatten(x_tensor):
    """
    Flatten x_tensor to (Batch Size, Flattened Image Size)
    : x_tensor: A tensor of size (Batch Size, ...), where ... are the image dimensions.
    : return: A tensor of size (Batch Size, Flattened Image Size).
    """
    # TODO: Implement Function
    image_shape = x_tensor.get_shape().as_list()
    return tf.reshape(x_tensor, [-1, image_shape[1]*image_shape[2]*image_shape[3]])


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_flatten(flatten)









    



Tests Passed

Fully-Connected Layer

Implement the fully_conn function to apply a fully connected layer to x_tensor with the shape (Batch Size, num_outputs). Shortcut option: you can use classes from the TensorFlow Layers or TensorFlow Layers (contrib) packages for this layer. For more of a challenge, only use other TensorFlow packages.



In [87]:

    
def fully_conn(x_tensor, num_outputs):
    """
    Apply a fully connected layer to x_tensor using weight and bias
    : x_tensor: A 2-D tensor where the first dimension is batch size.
    : num_outputs: The number of output that the new tensor should be.
    : return: A 2-D tensor where the second dimension is num_outputs.
    """
    # TODO: Implement Function
    W_fc1 = tf.Variable(tf.truncated_normal([x_tensor.get_shape().as_list()[1], num_outputs], stddev=0.1))
    b_fc1 = tf.Variable(tf.zeros([num_outputs]))

    return tf.nn.relu(tf.matmul(x_tensor, W_fc1) + b_fc1)


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_fully_conn(fully_conn)









    



Tests Passed

Output Layer

Implement the output function to apply a fully connected layer to x_tensor with the shape (Batch Size, num_outputs). Shortcut option: you can use classes from the TensorFlow Layers or TensorFlow Layers (contrib) packages for this layer. For more of a challenge, only use other TensorFlow packages.

Note: Activation, softmax, or cross entropy should not be applied to this.



In [88]:

    
def output(x_tensor, num_outputs):
    """
    Apply a output layer to x_tensor using weight and bias
    : x_tensor: A 2-D tensor where the first dimension is batch size.
    : num_outputs: The number of output that the new tensor should be.
    : return: A 2-D tensor where the second dimension is num_outputs.
    """
    # TODO: Implement Function
    W_output = tf.Variable(tf.truncated_normal([x_tensor.get_shape().as_list()[1], num_outputs], stddev=0.1))
    b_output = tf.Variable(tf.zeros([num_outputs]))

    return tf.matmul(x_tensor, W_output) + b_output

"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_output(output)









    



Tests Passed

Create Convolutional Model

Implement the function conv_net to create a convolutional neural network model. The function takes in a batch of images, x, and outputs logits. Use the layers you created above to create this model:

Apply 1, 2, or 3 Convolution and Max Pool layers
Apply a Flatten Layer
Apply 1, 2, or 3 Fully Connected Layers
Apply an Output Layer
Return the output
Apply TensorFlow's Dropout to one or more layers in the model using keep_prob.



In [134]:

    
def conv_net(x, keep_prob):
    """
    Create a convolutional neural network model
    : x: Placeholder tensor that holds image data.
    : keep_prob: Placeholder tensor that hold dropout keep probability.
    : return: Tensor that represents logits
    """
    # TODO: Apply 1, 2, or 3 Convolution and Max Pool layers
    #    Play around with different number of outputs, kernel size and stride
    # Function Definition from Above:
    #    conv2d_maxpool(x_tensor, conv_num_outputs, conv_ksize, conv_strides, pool_ksize, pool_strides)
    h_conv_maxpool1 = conv2d_maxpool(x, 32, [5, 5], [1, 1], [2, 2], [2, 2])
    h_conv_drop1 = tf.nn.dropout(h_conv_maxpool1, keep_prob)
    h_conv_maxpool2 = conv2d_maxpool(h_conv_drop1, 64, [5, 5], [1, 1], [2, 2], [2, 2])
    h_conv_drop2 = tf.nn.dropout(h_conv_maxpool2, keep_prob)
    h_conv_maxpool3 = conv2d_maxpool(h_conv_drop2, 128, [5, 5], [1, 1], [2, 2], [2, 2])
    h_conv_drop3 = tf.nn.dropout(h_conv_maxpool3, keep_prob)

    # TODO: Apply a Flatten Layer
    # Function Definition from Above:
    #   flatten(x_tensor)
    h_flatten = flatten(h_conv_drop3)
    

    # TODO: Apply 1, 2, or 3 Fully Connected Layers
    #    Play around with different number of outputs
    # Function Definition from Above:
    #   fully_conn(x_tensor, num_outputs)
    h_fc1 = fully_conn(h_flatten, 1024)
    h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)
    h_fc2 = fully_conn(h_fc1_drop, 1024)
    h_fc2_drop = tf.nn.dropout(h_fc2, keep_prob)
    h_fc3 = fully_conn(h_fc2_drop, 1024)
    h_fc3_drop = tf.nn.dropout(h_fc3, keep_prob)

    
    # TODO: Apply an Output Layer
    #    Set this to the number of classes
    # Function Definition from Above:
    #   output(x_tensor, num_outputs)
    y_conv = output(h_fc3_drop, num_outputs=10)
    
    # TODO: return output
    return y_conv


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""

##############################
## Build the Neural Network ##
##############################

# Remove previous weights, bias, inputs, etc..
tf.reset_default_graph()

# Inputs
x = neural_net_image_input((32, 32, 3))
y = neural_net_label_input(10)
keep_prob = neural_net_keep_prob_input()

# Model
logits = conv_net(x, keep_prob)

# Name logits Tensor, so that is can be loaded from disk after training
logits = tf.identity(logits, name='logits')

# Loss and Optimizer
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=logits, labels=y))
optimizer = tf.train.AdamOptimizer().minimize(cost)

# Accuracy
correct_pred = tf.equal(tf.argmax(logits, 1), tf.argmax(y, 1))
accuracy = tf.reduce_mean(tf.cast(correct_pred, tf.float32), name='accuracy')

tests.test_conv_net(conv_net)









    



Neural Network Built!

Train the Neural Network

Single Optimization

Implement the function train_neural_network to do a single optimization. The optimization should use optimizer to optimize in session with a feed_dict of the following:

x for image input
y for labels
keep_prob for keep probability for dropout

This function will be called for each batch, so tf.global_variables_initializer() has already been called.

Note: Nothing needs to be returned. This function is only optimizing the neural network.



In [135]:

    
def train_neural_network(session, optimizer, keep_probability, feature_batch, label_batch):
    """
    Optimize the session on a batch of images and labels
    : session: Current TensorFlow session
    : optimizer: TensorFlow optimizer function
    : keep_probability: keep probability
    : feature_batch: Batch of Numpy image data
    : label_batch: Batch of Numpy label data
    """
    # TODO: Implement Function
    session.run(optimizer, feed_dict={x: feature_batch, y: label_batch, keep_prob: keep_probability})

"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_train_nn(train_neural_network)









    



Tests Passed

Show Stats

Implement the function print_stats to print loss and validation accuracy. Use the global variables valid_features and valid_labels to calculate validation accuracy. Use a keep probability of 1.0 to calculate the loss and validation accuracy.



In [136]:

    
def print_stats(session, feature_batch, label_batch, cost, accuracy):
    """
    Print information about loss and validation accuracy
    : session: Current TensorFlow session
    : feature_batch: Batch of Numpy image data
    : label_batch: Batch of Numpy label data
    : cost: TensorFlow cost function
    : accuracy: TensorFlow accuracy function
    """
    # TODO: Implement Function
    l, train_accuracy = session.run([cost, accuracy], feed_dict={x: feature_batch, y: label_batch, keep_prob: 1.0})
    valid_accuracy = session.run(accuracy, feed_dict={x: valid_features, y: valid_labels, keep_prob: 1.0})
    print('loss %f, train_accuracy %g, valid accuracy %g' % (l, train_accuracy, valid_accuracy))

Hyperparameters

Tune the following parameters:

Set epochs to the number of iterations until the network stops learning or start overfitting
Set batch_size to the highest number that your machine has memory for. Most people set them to common sizes of memory:
- 64
- 128
- 256
- ...
Set keep_probability to the probability of keeping a node using dropout



In [159]:

    
# TODO: Tune Parameters
epochs = 600
batch_size = 256
keep_probability = 0.6

Train on a Single CIFAR-10 Batch

Instead of training the neural network on all the CIFAR-10 batches of data, let's use a single batch. This should save time while you iterate on the model to get a better accuracy. Once the final validation accuracy is 50% or greater, run the model on all the data in the next section.



In [160]:

    
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
print('Checking the Training on a Single Batch...')
with tf.Session() as sess:
    # Initializing the variables
    sess.run(tf.global_variables_initializer())
    
    # Training cycle
    for epoch in range(epochs):
        batch_i = 1
        for batch_features, batch_labels in helper.load_preprocess_training_batch(batch_i, batch_size):
            train_neural_network(sess, optimizer, keep_probability, batch_features, batch_labels)
        print('Epoch {:>2}, CIFAR-10 Batch {}:  '.format(epoch + 1, batch_i), end='')
        print_stats(sess, batch_features, batch_labels, cost, accuracy)









    



Checking the Training on a Single Batch...
Epoch  1, CIFAR-10 Batch 1:  loss 2.312561, train_accuracy 0.15, valid accuracy 0.125
Epoch  2, CIFAR-10 Batch 1:  loss 2.300617, train_accuracy 0.075, valid accuracy 0.112
Epoch  3, CIFAR-10 Batch 1:  loss 2.305267, train_accuracy 0.05, valid accuracy 0.103
Epoch  4, CIFAR-10 Batch 1:  loss 2.314233, train_accuracy 0.1, valid accuracy 0.1016
Epoch  5, CIFAR-10 Batch 1:  loss 2.311855, train_accuracy 0.1, valid accuracy 0.097
Epoch  6, CIFAR-10 Batch 1:  loss 2.311517, train_accuracy 0.1, valid accuracy 0.095
Epoch  7, CIFAR-10 Batch 1:  loss 2.315464, train_accuracy 0.1, valid accuracy 0.0948
Epoch  8, CIFAR-10 Batch 1:  loss 2.324066, train_accuracy 0.1, valid accuracy 0.0946
Epoch  9, CIFAR-10 Batch 1:  loss 2.343038, train_accuracy 0.1, valid accuracy 0.0946
Epoch 10, CIFAR-10 Batch 1:  loss 2.362362, train_accuracy 0.1, valid accuracy 0.0946
Epoch 11, CIFAR-10 Batch 1:  loss 2.381929, train_accuracy 0.1, valid accuracy 0.0946
Epoch 12, CIFAR-10 Batch 1:  loss 2.403596, train_accuracy 0.1, valid accuracy 0.0946
Epoch 13, CIFAR-10 Batch 1:  loss 2.424383, train_accuracy 0.1, valid accuracy 0.0946
Epoch 14, CIFAR-10 Batch 1:  loss 2.456482, train_accuracy 0.1, valid accuracy 0.0946
Epoch 15, CIFAR-10 Batch 1:  loss 2.483245, train_accuracy 0.1, valid accuracy 0.0946
Epoch 16, CIFAR-10 Batch 1:  loss 2.512489, train_accuracy 0.1, valid accuracy 0.0946
Epoch 17, CIFAR-10 Batch 1:  loss 2.539088, train_accuracy 0.1, valid accuracy 0.0946
Epoch 18, CIFAR-10 Batch 1:  loss 2.561691, train_accuracy 0.1, valid accuracy 0.0946
Epoch 19, CIFAR-10 Batch 1:  loss 2.587651, train_accuracy 0.1, valid accuracy 0.0946
Epoch 20, CIFAR-10 Batch 1:  loss 2.635099, train_accuracy 0.1, valid accuracy 0.0944
Epoch 21, CIFAR-10 Batch 1:  loss 2.608403, train_accuracy 0.1, valid accuracy 0.0942
Epoch 22, CIFAR-10 Batch 1:  loss 2.586698, train_accuracy 0.1, valid accuracy 0.0944
Epoch 23, CIFAR-10 Batch 1:  loss 2.548783, train_accuracy 0.125, valid accuracy 0.0944
Epoch 24, CIFAR-10 Batch 1:  loss 2.535705, train_accuracy 0.125, valid accuracy 0.0966
Epoch 25, CIFAR-10 Batch 1:  loss 2.514002, train_accuracy 0.125, valid accuracy 0.102
Epoch 26, CIFAR-10 Batch 1:  loss 2.444414, train_accuracy 0.1, valid accuracy 0.115
Epoch 27, CIFAR-10 Batch 1:  loss 2.394633, train_accuracy 0.15, valid accuracy 0.1268
Epoch 28, CIFAR-10 Batch 1:  loss 2.354872, train_accuracy 0.2, valid accuracy 0.1436
Epoch 29, CIFAR-10 Batch 1:  loss 2.302407, train_accuracy 0.175, valid accuracy 0.1604
Epoch 30, CIFAR-10 Batch 1:  loss 2.360500, train_accuracy 0.125, valid accuracy 0.1602
Epoch 31, CIFAR-10 Batch 1:  loss 2.402143, train_accuracy 0.125, valid accuracy 0.157
Epoch 32, CIFAR-10 Batch 1:  loss 2.281110, train_accuracy 0.15, valid accuracy 0.1898
Epoch 33, CIFAR-10 Batch 1:  loss 2.357641, train_accuracy 0.125, valid accuracy 0.1666
Epoch 34, CIFAR-10 Batch 1:  loss 2.256577, train_accuracy 0.125, valid accuracy 0.1896
Epoch 35, CIFAR-10 Batch 1:  loss 2.271103, train_accuracy 0.15, valid accuracy 0.1962
Epoch 36, CIFAR-10 Batch 1:  loss 2.216105, train_accuracy 0.175, valid accuracy 0.202
Epoch 37, CIFAR-10 Batch 1:  loss 2.119258, train_accuracy 0.175, valid accuracy 0.2238
Epoch 38, CIFAR-10 Batch 1:  loss 2.158032, train_accuracy 0.175, valid accuracy 0.2308
Epoch 39, CIFAR-10 Batch 1:  loss 2.230690, train_accuracy 0.125, valid accuracy 0.2118
Epoch 40, CIFAR-10 Batch 1:  loss 2.182487, train_accuracy 0.15, valid accuracy 0.2202
Epoch 41, CIFAR-10 Batch 1:  loss 2.072435, train_accuracy 0.15, valid accuracy 0.2398
Epoch 42, CIFAR-10 Batch 1:  loss 2.213862, train_accuracy 0.175, valid accuracy 0.2126
Epoch 43, CIFAR-10 Batch 1:  loss 2.135815, train_accuracy 0.15, valid accuracy 0.2322
Epoch 44, CIFAR-10 Batch 1:  loss 2.212153, train_accuracy 0.15, valid accuracy 0.218
Epoch 45, CIFAR-10 Batch 1:  loss 2.156616, train_accuracy 0.175, valid accuracy 0.226
Epoch 46, CIFAR-10 Batch 1:  loss 2.055180, train_accuracy 0.25, valid accuracy 0.2406
Epoch 47, CIFAR-10 Batch 1:  loss 2.059034, train_accuracy 0.175, valid accuracy 0.2544
Epoch 48, CIFAR-10 Batch 1:  loss 2.059647, train_accuracy 0.175, valid accuracy 0.2496
Epoch 49, CIFAR-10 Batch 1:  loss 1.996524, train_accuracy 0.225, valid accuracy 0.246
Epoch 50, CIFAR-10 Batch 1:  loss 2.008526, train_accuracy 0.25, valid accuracy 0.262
Epoch 51, CIFAR-10 Batch 1:  loss 2.070482, train_accuracy 0.2, valid accuracy 0.2504
Epoch 52, CIFAR-10 Batch 1:  loss 1.996046, train_accuracy 0.225, valid accuracy 0.265
Epoch 53, CIFAR-10 Batch 1:  loss 2.090189, train_accuracy 0.3, valid accuracy 0.2542
Epoch 54, CIFAR-10 Batch 1:  loss 1.980262, train_accuracy 0.275, valid accuracy 0.2674
Epoch 55, CIFAR-10 Batch 1:  loss 1.993598, train_accuracy 0.3, valid accuracy 0.2688
Epoch 56, CIFAR-10 Batch 1:  loss 1.937981, train_accuracy 0.3, valid accuracy 0.2878
Epoch 57, CIFAR-10 Batch 1:  loss 1.902350, train_accuracy 0.35, valid accuracy 0.2764
Epoch 58, CIFAR-10 Batch 1:  loss 1.929422, train_accuracy 0.325, valid accuracy 0.2812
Epoch 59, CIFAR-10 Batch 1:  loss 1.971360, train_accuracy 0.325, valid accuracy 0.2722
Epoch 60, CIFAR-10 Batch 1:  loss 1.918163, train_accuracy 0.325, valid accuracy 0.2914
Epoch 61, CIFAR-10 Batch 1:  loss 2.035899, train_accuracy 0.3, valid accuracy 0.266
Epoch 62, CIFAR-10 Batch 1:  loss 2.055534, train_accuracy 0.3, valid accuracy 0.275
Epoch 63, CIFAR-10 Batch 1:  loss 2.031682, train_accuracy 0.325, valid accuracy 0.2848
Epoch 64, CIFAR-10 Batch 1:  loss 2.032642, train_accuracy 0.3, valid accuracy 0.272
Epoch 65, CIFAR-10 Batch 1:  loss 2.044261, train_accuracy 0.325, valid accuracy 0.282
Epoch 66, CIFAR-10 Batch 1:  loss 2.035256, train_accuracy 0.275, valid accuracy 0.2762
Epoch 67, CIFAR-10 Batch 1:  loss 1.957087, train_accuracy 0.375, valid accuracy 0.291
Epoch 68, CIFAR-10 Batch 1:  loss 1.952211, train_accuracy 0.35, valid accuracy 0.2852
Epoch 69, CIFAR-10 Batch 1:  loss 1.915447, train_accuracy 0.275, valid accuracy 0.3016
Epoch 70, CIFAR-10 Batch 1:  loss 1.894190, train_accuracy 0.325, valid accuracy 0.2924
Epoch 71, CIFAR-10 Batch 1:  loss 1.853135, train_accuracy 0.375, valid accuracy 0.3066
Epoch 72, CIFAR-10 Batch 1:  loss 1.772229, train_accuracy 0.35, valid accuracy 0.3194
Epoch 73, CIFAR-10 Batch 1:  loss 1.839634, train_accuracy 0.325, valid accuracy 0.2994
Epoch 74, CIFAR-10 Batch 1:  loss 1.939697, train_accuracy 0.325, valid accuracy 0.2812
Epoch 75, CIFAR-10 Batch 1:  loss 1.871938, train_accuracy 0.375, valid accuracy 0.3064
Epoch 76, CIFAR-10 Batch 1:  loss 1.717269, train_accuracy 0.4, valid accuracy 0.3288
Epoch 77, CIFAR-10 Batch 1:  loss 1.712000, train_accuracy 0.425, valid accuracy 0.3332
Epoch 78, CIFAR-10 Batch 1:  loss 1.740502, train_accuracy 0.45, valid accuracy 0.3074
Epoch 79, CIFAR-10 Batch 1:  loss 1.653355, train_accuracy 0.375, valid accuracy 0.3494
Epoch 80, CIFAR-10 Batch 1:  loss 1.693317, train_accuracy 0.4, valid accuracy 0.3446
Epoch 81, CIFAR-10 Batch 1:  loss 1.732082, train_accuracy 0.425, valid accuracy 0.3252
Epoch 82, CIFAR-10 Batch 1:  loss 1.738801, train_accuracy 0.4, valid accuracy 0.3128
Epoch 83, CIFAR-10 Batch 1:  loss 1.634805, train_accuracy 0.425, valid accuracy 0.3442
Epoch 84, CIFAR-10 Batch 1:  loss 1.644272, train_accuracy 0.475, valid accuracy 0.348
Epoch 85, CIFAR-10 Batch 1:  loss 1.598599, train_accuracy 0.4, valid accuracy 0.3438
Epoch 86, CIFAR-10 Batch 1:  loss 1.643878, train_accuracy 0.425, valid accuracy 0.3568
Epoch 87, CIFAR-10 Batch 1:  loss 1.554946, train_accuracy 0.425, valid accuracy 0.3548
Epoch 88, CIFAR-10 Batch 1:  loss 1.503479, train_accuracy 0.45, valid accuracy 0.3868
Epoch 89, CIFAR-10 Batch 1:  loss 1.608453, train_accuracy 0.375, valid accuracy 0.3598
Epoch 90, CIFAR-10 Batch 1:  loss 1.644360, train_accuracy 0.425, valid accuracy 0.3522
Epoch 91, CIFAR-10 Batch 1:  loss 1.506720, train_accuracy 0.45, valid accuracy 0.3856
Epoch 92, CIFAR-10 Batch 1:  loss 1.543811, train_accuracy 0.425, valid accuracy 0.3904
Epoch 93, CIFAR-10 Batch 1:  loss 1.456621, train_accuracy 0.45, valid accuracy 0.3842
Epoch 94, CIFAR-10 Batch 1:  loss 1.439984, train_accuracy 0.425, valid accuracy 0.3992
Epoch 95, CIFAR-10 Batch 1:  loss 1.479715, train_accuracy 0.4, valid accuracy 0.3972
Epoch 96, CIFAR-10 Batch 1:  loss 1.451247, train_accuracy 0.4, valid accuracy 0.4004
Epoch 97, CIFAR-10 Batch 1:  loss 1.491055, train_accuracy 0.45, valid accuracy 0.3944
Epoch 98, CIFAR-10 Batch 1:  loss 1.391663, train_accuracy 0.45, valid accuracy 0.4282
Epoch 99, CIFAR-10 Batch 1:  loss 1.477727, train_accuracy 0.4, valid accuracy 0.393
Epoch 100, CIFAR-10 Batch 1:  loss 1.366663, train_accuracy 0.475, valid accuracy 0.415
Epoch 101, CIFAR-10 Batch 1:  loss 1.347913, train_accuracy 0.5, valid accuracy 0.4298
Epoch 102, CIFAR-10 Batch 1:  loss 1.325369, train_accuracy 0.5, valid accuracy 0.431
Epoch 103, CIFAR-10 Batch 1:  loss 1.271630, train_accuracy 0.55, valid accuracy 0.4328
Epoch 104, CIFAR-10 Batch 1:  loss 1.309869, train_accuracy 0.5, valid accuracy 0.434
Epoch 105, CIFAR-10 Batch 1:  loss 1.348710, train_accuracy 0.5, valid accuracy 0.4072
Epoch 106, CIFAR-10 Batch 1:  loss 1.253899, train_accuracy 0.575, valid accuracy 0.434
Epoch 107, CIFAR-10 Batch 1:  loss 1.362493, train_accuracy 0.475, valid accuracy 0.4096
Epoch 108, CIFAR-10 Batch 1:  loss 1.330299, train_accuracy 0.525, valid accuracy 0.4208
Epoch 109, CIFAR-10 Batch 1:  loss 1.313998, train_accuracy 0.575, valid accuracy 0.4184
Epoch 110, CIFAR-10 Batch 1:  loss 1.316847, train_accuracy 0.6, valid accuracy 0.4362
Epoch 111, CIFAR-10 Batch 1:  loss 1.347208, train_accuracy 0.5, valid accuracy 0.419
Epoch 112, CIFAR-10 Batch 1:  loss 1.228617, train_accuracy 0.6, valid accuracy 0.4464
Epoch 113, CIFAR-10 Batch 1:  loss 1.273110, train_accuracy 0.525, valid accuracy 0.434
Epoch 114, CIFAR-10 Batch 1:  loss 1.264521, train_accuracy 0.525, valid accuracy 0.4444
Epoch 115, CIFAR-10 Batch 1:  loss 1.237081, train_accuracy 0.55, valid accuracy 0.4474
Epoch 116, CIFAR-10 Batch 1:  loss 1.232516, train_accuracy 0.6, valid accuracy 0.4626
Epoch 117, CIFAR-10 Batch 1:  loss 1.187383, train_accuracy 0.6, valid accuracy 0.4572
Epoch 118, CIFAR-10 Batch 1:  loss 1.237448, train_accuracy 0.55, valid accuracy 0.4374
Epoch 119, CIFAR-10 Batch 1:  loss 1.193815, train_accuracy 0.55, valid accuracy 0.463
Epoch 120, CIFAR-10 Batch 1:  loss 1.109348, train_accuracy 0.6, valid accuracy 0.4656
Epoch 121, CIFAR-10 Batch 1:  loss 1.212321, train_accuracy 0.5, valid accuracy 0.4562
Epoch 122, CIFAR-10 Batch 1:  loss 1.125972, train_accuracy 0.6, valid accuracy 0.4746
Epoch 123, CIFAR-10 Batch 1:  loss 1.154700, train_accuracy 0.575, valid accuracy 0.4812
Epoch 124, CIFAR-10 Batch 1:  loss 1.085715, train_accuracy 0.65, valid accuracy 0.4928
Epoch 125, CIFAR-10 Batch 1:  loss 1.078615, train_accuracy 0.65, valid accuracy 0.4956
Epoch 126, CIFAR-10 Batch 1:  loss 1.057190, train_accuracy 0.675, valid accuracy 0.4918
Epoch 127, CIFAR-10 Batch 1:  loss 1.123332, train_accuracy 0.6, valid accuracy 0.4744
Epoch 128, CIFAR-10 Batch 1:  loss 1.116494, train_accuracy 0.65, valid accuracy 0.4774
Epoch 129, CIFAR-10 Batch 1:  loss 1.083166, train_accuracy 0.65, valid accuracy 0.478
Epoch 130, CIFAR-10 Batch 1:  loss 1.116404, train_accuracy 0.6, valid accuracy 0.4724
Epoch 131, CIFAR-10 Batch 1:  loss 1.115869, train_accuracy 0.6, valid accuracy 0.4644
Epoch 132, CIFAR-10 Batch 1:  loss 1.127482, train_accuracy 0.575, valid accuracy 0.474
Epoch 133, CIFAR-10 Batch 1:  loss 1.029274, train_accuracy 0.625, valid accuracy 0.4826
Epoch 134, CIFAR-10 Batch 1:  loss 1.018227, train_accuracy 0.65, valid accuracy 0.486
Epoch 135, CIFAR-10 Batch 1:  loss 1.030870, train_accuracy 0.65, valid accuracy 0.492
Epoch 136, CIFAR-10 Batch 1:  loss 0.982496, train_accuracy 0.675, valid accuracy 0.502
Epoch 137, CIFAR-10 Batch 1:  loss 0.978893, train_accuracy 0.7, valid accuracy 0.4864
Epoch 138, CIFAR-10 Batch 1:  loss 1.041295, train_accuracy 0.65, valid accuracy 0.4914
Epoch 139, CIFAR-10 Batch 1:  loss 0.992990, train_accuracy 0.675, valid accuracy 0.4994
Epoch 140, CIFAR-10 Batch 1:  loss 0.956903, train_accuracy 0.725, valid accuracy 0.4886
Epoch 141, CIFAR-10 Batch 1:  loss 0.935323, train_accuracy 0.675, valid accuracy 0.5074
Epoch 142, CIFAR-10 Batch 1:  loss 0.949833, train_accuracy 0.725, valid accuracy 0.4856
Epoch 143, CIFAR-10 Batch 1:  loss 0.904306, train_accuracy 0.675, valid accuracy 0.5166
Epoch 144, CIFAR-10 Batch 1:  loss 0.926509, train_accuracy 0.7, valid accuracy 0.5168
Epoch 145, CIFAR-10 Batch 1:  loss 0.866621, train_accuracy 0.7, valid accuracy 0.5324
Epoch 146, CIFAR-10 Batch 1:  loss 0.935732, train_accuracy 0.675, valid accuracy 0.517
Epoch 147, CIFAR-10 Batch 1:  loss 0.866222, train_accuracy 0.75, valid accuracy 0.5268
Epoch 148, CIFAR-10 Batch 1:  loss 0.902047, train_accuracy 0.75, valid accuracy 0.512
Epoch 149, CIFAR-10 Batch 1:  loss 0.932491, train_accuracy 0.7, valid accuracy 0.5246
Epoch 150, CIFAR-10 Batch 1:  loss 0.850661, train_accuracy 0.725, valid accuracy 0.52
Epoch 151, CIFAR-10 Batch 1:  loss 0.933424, train_accuracy 0.675, valid accuracy 0.4962
Epoch 152, CIFAR-10 Batch 1:  loss 0.805882, train_accuracy 0.8, valid accuracy 0.5168
Epoch 153, CIFAR-10 Batch 1:  loss 0.889112, train_accuracy 0.75, valid accuracy 0.5034
Epoch 154, CIFAR-10 Batch 1:  loss 0.814018, train_accuracy 0.8, valid accuracy 0.5298
Epoch 155, CIFAR-10 Batch 1:  loss 0.828970, train_accuracy 0.775, valid accuracy 0.532
Epoch 156, CIFAR-10 Batch 1:  loss 0.785511, train_accuracy 0.75, valid accuracy 0.5274
Epoch 157, CIFAR-10 Batch 1:  loss 0.810792, train_accuracy 0.775, valid accuracy 0.5198
Epoch 158, CIFAR-10 Batch 1:  loss 0.770692, train_accuracy 0.725, valid accuracy 0.5426
Epoch 159, CIFAR-10 Batch 1:  loss 0.846093, train_accuracy 0.75, valid accuracy 0.5334
Epoch 160, CIFAR-10 Batch 1:  loss 0.800865, train_accuracy 0.75, valid accuracy 0.5174
Epoch 161, CIFAR-10 Batch 1:  loss 0.777110, train_accuracy 0.75, valid accuracy 0.5366
Epoch 162, CIFAR-10 Batch 1:  loss 0.839331, train_accuracy 0.725, valid accuracy 0.5336
Epoch 163, CIFAR-10 Batch 1:  loss 0.722487, train_accuracy 0.825, valid accuracy 0.5502
Epoch 164, CIFAR-10 Batch 1:  loss 0.732012, train_accuracy 0.8, valid accuracy 0.5364
Epoch 165, CIFAR-10 Batch 1:  loss 0.726750, train_accuracy 0.8, valid accuracy 0.5394
Epoch 166, CIFAR-10 Batch 1:  loss 0.731616, train_accuracy 0.85, valid accuracy 0.5494
Epoch 167, CIFAR-10 Batch 1:  loss 0.707911, train_accuracy 0.9, valid accuracy 0.5516
Epoch 168, CIFAR-10 Batch 1:  loss 0.692947, train_accuracy 0.85, valid accuracy 0.5398
Epoch 169, CIFAR-10 Batch 1:  loss 0.673117, train_accuracy 0.85, valid accuracy 0.5628
Epoch 170, CIFAR-10 Batch 1:  loss 0.682221, train_accuracy 0.85, valid accuracy 0.5602
Epoch 171, CIFAR-10 Batch 1:  loss 0.685058, train_accuracy 0.8, valid accuracy 0.5598
Epoch 172, CIFAR-10 Batch 1:  loss 0.693073, train_accuracy 0.825, valid accuracy 0.5422
Epoch 173, CIFAR-10 Batch 1:  loss 0.667445, train_accuracy 0.825, valid accuracy 0.5612
Epoch 174, CIFAR-10 Batch 1:  loss 0.680469, train_accuracy 0.85, valid accuracy 0.539
Epoch 175, CIFAR-10 Batch 1:  loss 0.645765, train_accuracy 0.9, valid accuracy 0.5506
Epoch 176, CIFAR-10 Batch 1:  loss 0.639620, train_accuracy 0.875, valid accuracy 0.5444
Epoch 177, CIFAR-10 Batch 1:  loss 0.649764, train_accuracy 0.8, valid accuracy 0.5534
Epoch 178, CIFAR-10 Batch 1:  loss 0.617328, train_accuracy 0.875, valid accuracy 0.5544
Epoch 179, CIFAR-10 Batch 1:  loss 0.587419, train_accuracy 0.875, valid accuracy 0.5702
Epoch 180, CIFAR-10 Batch 1:  loss 0.557072, train_accuracy 0.875, valid accuracy 0.5804
Epoch 181, CIFAR-10 Batch 1:  loss 0.544673, train_accuracy 0.9, valid accuracy 0.5742
Epoch 182, CIFAR-10 Batch 1:  loss 0.535526, train_accuracy 0.9, valid accuracy 0.5636
Epoch 183, CIFAR-10 Batch 1:  loss 0.555479, train_accuracy 0.875, valid accuracy 0.5768
Epoch 184, CIFAR-10 Batch 1:  loss 0.530559, train_accuracy 0.875, valid accuracy 0.5798
Epoch 185, CIFAR-10 Batch 1:  loss 0.517625, train_accuracy 0.925, valid accuracy 0.5896
Epoch 186, CIFAR-10 Batch 1:  loss 0.535573, train_accuracy 0.925, valid accuracy 0.576
Epoch 187, CIFAR-10 Batch 1:  loss 0.515925, train_accuracy 0.9, valid accuracy 0.5748
Epoch 188, CIFAR-10 Batch 1:  loss 0.604528, train_accuracy 0.875, valid accuracy 0.5462
Epoch 189, CIFAR-10 Batch 1:  loss 0.525599, train_accuracy 0.925, valid accuracy 0.5584
Epoch 190, CIFAR-10 Batch 1:  loss 0.495244, train_accuracy 0.9, valid accuracy 0.5684
Epoch 191, CIFAR-10 Batch 1:  loss 0.476542, train_accuracy 0.9, valid accuracy 0.585
Epoch 192, CIFAR-10 Batch 1:  loss 0.518196, train_accuracy 0.9, valid accuracy 0.5574
Epoch 193, CIFAR-10 Batch 1:  loss 0.449511, train_accuracy 0.925, valid accuracy 0.5768
Epoch 194, CIFAR-10 Batch 1:  loss 0.473994, train_accuracy 0.9, valid accuracy 0.5756
Epoch 195, CIFAR-10 Batch 1:  loss 0.537782, train_accuracy 0.875, valid accuracy 0.5438
Epoch 196, CIFAR-10 Batch 1:  loss 0.436037, train_accuracy 0.925, valid accuracy 0.5852
Epoch 197, CIFAR-10 Batch 1:  loss 0.446965, train_accuracy 0.925, valid accuracy 0.576
Epoch 198, CIFAR-10 Batch 1:  loss 0.454484, train_accuracy 0.9, valid accuracy 0.5624
Epoch 199, CIFAR-10 Batch 1:  loss 0.444205, train_accuracy 0.9, valid accuracy 0.5456
Epoch 200, CIFAR-10 Batch 1:  loss 0.434274, train_accuracy 0.95, valid accuracy 0.5624
Epoch 201, CIFAR-10 Batch 1:  loss 0.439374, train_accuracy 0.9, valid accuracy 0.5732
Epoch 202, CIFAR-10 Batch 1:  loss 0.447145, train_accuracy 0.875, valid accuracy 0.5904
Epoch 203, CIFAR-10 Batch 1:  loss 0.348606, train_accuracy 0.95, valid accuracy 0.5858
Epoch 204, CIFAR-10 Batch 1:  loss 0.391458, train_accuracy 0.925, valid accuracy 0.5612
Epoch 205, CIFAR-10 Batch 1:  loss 0.354685, train_accuracy 0.925, valid accuracy 0.5918
Epoch 206, CIFAR-10 Batch 1:  loss 0.363725, train_accuracy 0.95, valid accuracy 0.5938
Epoch 207, CIFAR-10 Batch 1:  loss 0.355905, train_accuracy 0.925, valid accuracy 0.5896
Epoch 208, CIFAR-10 Batch 1:  loss 0.356753, train_accuracy 0.925, valid accuracy 0.6008
Epoch 209, CIFAR-10 Batch 1:  loss 0.347290, train_accuracy 0.95, valid accuracy 0.6054
Epoch 210, CIFAR-10 Batch 1:  loss 0.357455, train_accuracy 0.95, valid accuracy 0.5888
Epoch 211, CIFAR-10 Batch 1:  loss 0.322762, train_accuracy 0.95, valid accuracy 0.5958
Epoch 212, CIFAR-10 Batch 1:  loss 0.321825, train_accuracy 0.95, valid accuracy 0.5922
Epoch 213, CIFAR-10 Batch 1:  loss 0.304644, train_accuracy 0.95, valid accuracy 0.6046
Epoch 214, CIFAR-10 Batch 1:  loss 0.268853, train_accuracy 0.95, valid accuracy 0.6068
Epoch 215, CIFAR-10 Batch 1:  loss 0.309829, train_accuracy 0.925, valid accuracy 0.5764
Epoch 216, CIFAR-10 Batch 1:  loss 0.254380, train_accuracy 0.975, valid accuracy 0.6034
Epoch 217, CIFAR-10 Batch 1:  loss 0.291961, train_accuracy 0.975, valid accuracy 0.5878
Epoch 218, CIFAR-10 Batch 1:  loss 0.263717, train_accuracy 0.975, valid accuracy 0.5872
Epoch 219, CIFAR-10 Batch 1:  loss 0.273017, train_accuracy 0.95, valid accuracy 0.5836
Epoch 220, CIFAR-10 Batch 1:  loss 0.278778, train_accuracy 0.95, valid accuracy 0.6134
Epoch 221, CIFAR-10 Batch 1:  loss 0.275739, train_accuracy 0.975, valid accuracy 0.59
Epoch 222, CIFAR-10 Batch 1:  loss 0.281888, train_accuracy 0.975, valid accuracy 0.5874
Epoch 223, CIFAR-10 Batch 1:  loss 0.258191, train_accuracy 0.975, valid accuracy 0.6016
Epoch 224, CIFAR-10 Batch 1:  loss 0.206432, train_accuracy 1, valid accuracy 0.6092
Epoch 225, CIFAR-10 Batch 1:  loss 0.228918, train_accuracy 1, valid accuracy 0.6106
Epoch 226, CIFAR-10 Batch 1:  loss 0.228459, train_accuracy 0.95, valid accuracy 0.607
Epoch 227, CIFAR-10 Batch 1:  loss 0.242706, train_accuracy 0.975, valid accuracy 0.5928
Epoch 228, CIFAR-10 Batch 1:  loss 0.211163, train_accuracy 1, valid accuracy 0.597
Epoch 229, CIFAR-10 Batch 1:  loss 0.213235, train_accuracy 1, valid accuracy 0.5856
Epoch 230, CIFAR-10 Batch 1:  loss 0.193634, train_accuracy 1, valid accuracy 0.6062
Epoch 231, CIFAR-10 Batch 1:  loss 0.184538, train_accuracy 1, valid accuracy 0.6116
Epoch 232, CIFAR-10 Batch 1:  loss 0.192917, train_accuracy 0.975, valid accuracy 0.6144
Epoch 233, CIFAR-10 Batch 1:  loss 0.190167, train_accuracy 0.975, valid accuracy 0.6086
Epoch 234, CIFAR-10 Batch 1:  loss 0.146949, train_accuracy 1, valid accuracy 0.6034
Epoch 235, CIFAR-10 Batch 1:  loss 0.136713, train_accuracy 1, valid accuracy 0.6238
Epoch 236, CIFAR-10 Batch 1:  loss 0.136979, train_accuracy 1, valid accuracy 0.6178
Epoch 237, CIFAR-10 Batch 1:  loss 0.161897, train_accuracy 1, valid accuracy 0.6104
Epoch 238, CIFAR-10 Batch 1:  loss 0.140757, train_accuracy 1, valid accuracy 0.6134
Epoch 239, CIFAR-10 Batch 1:  loss 0.157645, train_accuracy 1, valid accuracy 0.6116
Epoch 240, CIFAR-10 Batch 1:  loss 0.160723, train_accuracy 1, valid accuracy 0.5898
Epoch 241, CIFAR-10 Batch 1:  loss 0.122534, train_accuracy 1, valid accuracy 0.6186
Epoch 242, CIFAR-10 Batch 1:  loss 0.144422, train_accuracy 1, valid accuracy 0.5904
Epoch 243, CIFAR-10 Batch 1:  loss 0.144948, train_accuracy 0.975, valid accuracy 0.5992
Epoch 244, CIFAR-10 Batch 1:  loss 0.135945, train_accuracy 1, valid accuracy 0.6094
Epoch 245, CIFAR-10 Batch 1:  loss 0.115611, train_accuracy 1, valid accuracy 0.6212
Epoch 246, CIFAR-10 Batch 1:  loss 0.109809, train_accuracy 1, valid accuracy 0.6188
Epoch 247, CIFAR-10 Batch 1:  loss 0.130898, train_accuracy 1, valid accuracy 0.6116
Epoch 248, CIFAR-10 Batch 1:  loss 0.119019, train_accuracy 1, valid accuracy 0.6224
Epoch 249, CIFAR-10 Batch 1:  loss 0.095728, train_accuracy 1, valid accuracy 0.6334
Epoch 250, CIFAR-10 Batch 1:  loss 0.121556, train_accuracy 1, valid accuracy 0.614
Epoch 251, CIFAR-10 Batch 1:  loss 0.101990, train_accuracy 1, valid accuracy 0.6122
Epoch 252, CIFAR-10 Batch 1:  loss 0.122306, train_accuracy 1, valid accuracy 0.5924
Epoch 253, CIFAR-10 Batch 1:  loss 0.092119, train_accuracy 1, valid accuracy 0.6136
Epoch 254, CIFAR-10 Batch 1:  loss 0.101799, train_accuracy 1, valid accuracy 0.6172
Epoch 255, CIFAR-10 Batch 1:  loss 0.103523, train_accuracy 1, valid accuracy 0.6154
Epoch 256, CIFAR-10 Batch 1:  loss 0.082929, train_accuracy 1, valid accuracy 0.635
Epoch 257, CIFAR-10 Batch 1:  loss 0.072520, train_accuracy 1, valid accuracy 0.6168
Epoch 258, CIFAR-10 Batch 1:  loss 0.075378, train_accuracy 1, valid accuracy 0.6242
Epoch 259, CIFAR-10 Batch 1:  loss 0.070907, train_accuracy 1, valid accuracy 0.6164
Epoch 260, CIFAR-10 Batch 1:  loss 0.060429, train_accuracy 1, valid accuracy 0.6372
Epoch 261, CIFAR-10 Batch 1:  loss 0.072498, train_accuracy 1, valid accuracy 0.6298
Epoch 262, CIFAR-10 Batch 1:  loss 0.069513, train_accuracy 1, valid accuracy 0.6246
Epoch 263, CIFAR-10 Batch 1:  loss 0.059295, train_accuracy 1, valid accuracy 0.6358
Epoch 264, CIFAR-10 Batch 1:  loss 0.051020, train_accuracy 1, valid accuracy 0.626
Epoch 265, CIFAR-10 Batch 1:  loss 0.058887, train_accuracy 1, valid accuracy 0.6352
Epoch 266, CIFAR-10 Batch 1:  loss 0.069514, train_accuracy 1, valid accuracy 0.6072
Epoch 267, CIFAR-10 Batch 1:  loss 0.056659, train_accuracy 1, valid accuracy 0.6138
Epoch 268, CIFAR-10 Batch 1:  loss 0.054898, train_accuracy 1, valid accuracy 0.6366
Epoch 269, CIFAR-10 Batch 1:  loss 0.059951, train_accuracy 1, valid accuracy 0.634
Epoch 270, CIFAR-10 Batch 1:  loss 0.060350, train_accuracy 1, valid accuracy 0.6254
Epoch 271, CIFAR-10 Batch 1:  loss 0.049077, train_accuracy 1, valid accuracy 0.625
Epoch 272, CIFAR-10 Batch 1:  loss 0.048326, train_accuracy 1, valid accuracy 0.6376
Epoch 273, CIFAR-10 Batch 1:  loss 0.059308, train_accuracy 1, valid accuracy 0.6344
Epoch 274, CIFAR-10 Batch 1:  loss 0.040095, train_accuracy 1, valid accuracy 0.6428
Epoch 275, CIFAR-10 Batch 1:  loss 0.044424, train_accuracy 1, valid accuracy 0.6216
Epoch 276, CIFAR-10 Batch 1:  loss 0.045277, train_accuracy 1, valid accuracy 0.642
Epoch 277, CIFAR-10 Batch 1:  loss 0.037222, train_accuracy 1, valid accuracy 0.6424
Epoch 278, CIFAR-10 Batch 1:  loss 0.031479, train_accuracy 1, valid accuracy 0.6418
Epoch 279, CIFAR-10 Batch 1:  loss 0.027053, train_accuracy 1, valid accuracy 0.6422
Epoch 280, CIFAR-10 Batch 1:  loss 0.026720, train_accuracy 1, valid accuracy 0.6368
Epoch 281, CIFAR-10 Batch 1:  loss 0.026677, train_accuracy 1, valid accuracy 0.6446
Epoch 282, CIFAR-10 Batch 1:  loss 0.028722, train_accuracy 1, valid accuracy 0.6458
Epoch 283, CIFAR-10 Batch 1:  loss 0.029608, train_accuracy 1, valid accuracy 0.6388
Epoch 284, CIFAR-10 Batch 1:  loss 0.043596, train_accuracy 1, valid accuracy 0.6086
Epoch 285, CIFAR-10 Batch 1:  loss 0.028550, train_accuracy 1, valid accuracy 0.6426
Epoch 286, CIFAR-10 Batch 1:  loss 0.024009, train_accuracy 1, valid accuracy 0.6424
Epoch 287, CIFAR-10 Batch 1:  loss 0.036036, train_accuracy 1, valid accuracy 0.642
Epoch 288, CIFAR-10 Batch 1:  loss 0.037624, train_accuracy 1, valid accuracy 0.628
Epoch 289, CIFAR-10 Batch 1:  loss 0.035466, train_accuracy 1, valid accuracy 0.633
Epoch 290, CIFAR-10 Batch 1:  loss 0.032957, train_accuracy 1, valid accuracy 0.647
Epoch 291, CIFAR-10 Batch 1:  loss 0.035134, train_accuracy 1, valid accuracy 0.6428
Epoch 292, CIFAR-10 Batch 1:  loss 0.042568, train_accuracy 1, valid accuracy 0.6504
Epoch 293, CIFAR-10 Batch 1:  loss 0.036961, train_accuracy 1, valid accuracy 0.6412
Epoch 294, CIFAR-10 Batch 1:  loss 0.037030, train_accuracy 1, valid accuracy 0.6494
Epoch 295, CIFAR-10 Batch 1:  loss 0.031398, train_accuracy 1, valid accuracy 0.645
Epoch 296, CIFAR-10 Batch 1:  loss 0.033353, train_accuracy 1, valid accuracy 0.6286
Epoch 297, CIFAR-10 Batch 1:  loss 0.023536, train_accuracy 1, valid accuracy 0.6442
Epoch 298, CIFAR-10 Batch 1:  loss 0.026580, train_accuracy 1, valid accuracy 0.649
Epoch 299, CIFAR-10 Batch 1:  loss 0.026225, train_accuracy 1, valid accuracy 0.6362
Epoch 300, CIFAR-10 Batch 1:  loss 0.019279, train_accuracy 1, valid accuracy 0.6376
Epoch 301, CIFAR-10 Batch 1:  loss 0.029785, train_accuracy 1, valid accuracy 0.6522
Epoch 302, CIFAR-10 Batch 1:  loss 0.037614, train_accuracy 1, valid accuracy 0.6528
Epoch 303, CIFAR-10 Batch 1:  loss 0.023230, train_accuracy 1, valid accuracy 0.6568
Epoch 304, CIFAR-10 Batch 1:  loss 0.028250, train_accuracy 1, valid accuracy 0.6284
Epoch 305, CIFAR-10 Batch 1:  loss 0.025673, train_accuracy 1, valid accuracy 0.6522
Epoch 306, CIFAR-10 Batch 1:  loss 0.023065, train_accuracy 1, valid accuracy 0.6482
Epoch 307, CIFAR-10 Batch 1:  loss 0.028744, train_accuracy 1, valid accuracy 0.6436
Epoch 308, CIFAR-10 Batch 1:  loss 0.019150, train_accuracy 1, valid accuracy 0.647
Epoch 309, CIFAR-10 Batch 1:  loss 0.018525, train_accuracy 1, valid accuracy 0.6322
Epoch 310, CIFAR-10 Batch 1:  loss 0.018114, train_accuracy 1, valid accuracy 0.639
Epoch 311, CIFAR-10 Batch 1:  loss 0.032869, train_accuracy 1, valid accuracy 0.633
Epoch 312, CIFAR-10 Batch 1:  loss 0.018099, train_accuracy 1, valid accuracy 0.618
Epoch 313, CIFAR-10 Batch 1:  loss 0.020043, train_accuracy 1, valid accuracy 0.6404
Epoch 314, CIFAR-10 Batch 1:  loss 0.011243, train_accuracy 1, valid accuracy 0.6408
Epoch 315, CIFAR-10 Batch 1:  loss 0.020375, train_accuracy 1, valid accuracy 0.6554
Epoch 316, CIFAR-10 Batch 1:  loss 0.016493, train_accuracy 1, valid accuracy 0.6426
Epoch 317, CIFAR-10 Batch 1:  loss 0.011665, train_accuracy 1, valid accuracy 0.6456
Epoch 318, CIFAR-10 Batch 1:  loss 0.011098, train_accuracy 1, valid accuracy 0.643
Epoch 319, CIFAR-10 Batch 1:  loss 0.012649, train_accuracy 1, valid accuracy 0.6296
Epoch 320, CIFAR-10 Batch 1:  loss 0.024615, train_accuracy 1, valid accuracy 0.6384
Epoch 321, CIFAR-10 Batch 1:  loss 0.011446, train_accuracy 1, valid accuracy 0.6336
Epoch 322, CIFAR-10 Batch 1:  loss 0.007576, train_accuracy 1, valid accuracy 0.6386
Epoch 323, CIFAR-10 Batch 1:  loss 0.010586, train_accuracy 1, valid accuracy 0.6348
Epoch 324, CIFAR-10 Batch 1:  loss 0.011260, train_accuracy 1, valid accuracy 0.6336
Epoch 325, CIFAR-10 Batch 1:  loss 0.015735, train_accuracy 1, valid accuracy 0.6468
Epoch 326, CIFAR-10 Batch 1:  loss 0.007050, train_accuracy 1, valid accuracy 0.641
Epoch 327, CIFAR-10 Batch 1:  loss 0.013236, train_accuracy 1, valid accuracy 0.6344
Epoch 328, CIFAR-10 Batch 1:  loss 0.012683, train_accuracy 1, valid accuracy 0.6316
Epoch 329, CIFAR-10 Batch 1:  loss 0.014233, train_accuracy 1, valid accuracy 0.6254
Epoch 330, CIFAR-10 Batch 1:  loss 0.007825, train_accuracy 1, valid accuracy 0.6512
Epoch 331, CIFAR-10 Batch 1:  loss 0.012143, train_accuracy 1, valid accuracy 0.6456
Epoch 332, CIFAR-10 Batch 1:  loss 0.009054, train_accuracy 1, valid accuracy 0.6406
Epoch 333, CIFAR-10 Batch 1:  loss 0.007262, train_accuracy 1, valid accuracy 0.6452
Epoch 334, CIFAR-10 Batch 1:  loss 0.004519, train_accuracy 1, valid accuracy 0.6534
Epoch 335, CIFAR-10 Batch 1:  loss 0.009209, train_accuracy 1, valid accuracy 0.6436
Epoch 336, CIFAR-10 Batch 1:  loss 0.007079, train_accuracy 1, valid accuracy 0.6362
Epoch 337, CIFAR-10 Batch 1:  loss 0.013783, train_accuracy 1, valid accuracy 0.6482
Epoch 338, CIFAR-10 Batch 1:  loss 0.012017, train_accuracy 1, valid accuracy 0.6428
Epoch 339, CIFAR-10 Batch 1:  loss 0.008574, train_accuracy 1, valid accuracy 0.6464
Epoch 340, CIFAR-10 Batch 1:  loss 0.006926, train_accuracy 1, valid accuracy 0.6396
Epoch 341, CIFAR-10 Batch 1:  loss 0.009378, train_accuracy 1, valid accuracy 0.639
Epoch 342, CIFAR-10 Batch 1:  loss 0.005311, train_accuracy 1, valid accuracy 0.6606
Epoch 343, CIFAR-10 Batch 1:  loss 0.004814, train_accuracy 1, valid accuracy 0.6414
Epoch 344, CIFAR-10 Batch 1:  loss 0.007073, train_accuracy 1, valid accuracy 0.6344
Epoch 345, CIFAR-10 Batch 1:  loss 0.006275, train_accuracy 1, valid accuracy 0.6596
Epoch 346, CIFAR-10 Batch 1:  loss 0.005301, train_accuracy 1, valid accuracy 0.6548
Epoch 347, CIFAR-10 Batch 1:  loss 0.006724, train_accuracy 1, valid accuracy 0.6524
Epoch 348, CIFAR-10 Batch 1:  loss 0.004266, train_accuracy 1, valid accuracy 0.6592
Epoch 349, CIFAR-10 Batch 1:  loss 0.006233, train_accuracy 1, valid accuracy 0.6216
Epoch 350, CIFAR-10 Batch 1:  loss 0.002972, train_accuracy 1, valid accuracy 0.6486
Epoch 351, CIFAR-10 Batch 1:  loss 0.003394, train_accuracy 1, valid accuracy 0.653
Epoch 352, CIFAR-10 Batch 1:  loss 0.002449, train_accuracy 1, valid accuracy 0.6578
Epoch 353, CIFAR-10 Batch 1:  loss 0.002873, train_accuracy 1, valid accuracy 0.6606
Epoch 354, CIFAR-10 Batch 1:  loss 0.002272, train_accuracy 1, valid accuracy 0.6634
Epoch 355, CIFAR-10 Batch 1:  loss 0.006290, train_accuracy 1, valid accuracy 0.6566
Epoch 356, CIFAR-10 Batch 1:  loss 0.004444, train_accuracy 1, valid accuracy 0.6604
Epoch 357, CIFAR-10 Batch 1:  loss 0.003743, train_accuracy 1, valid accuracy 0.6578
Epoch 358, CIFAR-10 Batch 1:  loss 0.002922, train_accuracy 1, valid accuracy 0.6514
Epoch 359, CIFAR-10 Batch 1:  loss 0.003911, train_accuracy 1, valid accuracy 0.6588
Epoch 360, CIFAR-10 Batch 1:  loss 0.003237, train_accuracy 1, valid accuracy 0.6528
Epoch 361, CIFAR-10 Batch 1:  loss 0.002325, train_accuracy 1, valid accuracy 0.6552
Epoch 362, CIFAR-10 Batch 1:  loss 0.003607, train_accuracy 1, valid accuracy 0.6582
Epoch 363, CIFAR-10 Batch 1:  loss 0.004775, train_accuracy 1, valid accuracy 0.6674
Epoch 364, CIFAR-10 Batch 1:  loss 0.005296, train_accuracy 1, valid accuracy 0.6644
Epoch 365, CIFAR-10 Batch 1:  loss 0.003685, train_accuracy 1, valid accuracy 0.6496
Epoch 366, CIFAR-10 Batch 1:  loss 0.001370, train_accuracy 1, valid accuracy 0.6672
Epoch 367, CIFAR-10 Batch 1:  loss 0.004389, train_accuracy 1, valid accuracy 0.657
Epoch 368, CIFAR-10 Batch 1:  loss 0.004789, train_accuracy 1, valid accuracy 0.6584
Epoch 369, CIFAR-10 Batch 1:  loss 0.003065, train_accuracy 1, valid accuracy 0.6508
Epoch 370, CIFAR-10 Batch 1:  loss 0.006805, train_accuracy 1, valid accuracy 0.6542
Epoch 371, CIFAR-10 Batch 1:  loss 0.002211, train_accuracy 1, valid accuracy 0.6608
Epoch 372, CIFAR-10 Batch 1:  loss 0.002235, train_accuracy 1, valid accuracy 0.6544
Epoch 373, CIFAR-10 Batch 1:  loss 0.002832, train_accuracy 1, valid accuracy 0.6622
Epoch 374, CIFAR-10 Batch 1:  loss 0.004036, train_accuracy 1, valid accuracy 0.6514
Epoch 375, CIFAR-10 Batch 1:  loss 0.003402, train_accuracy 1, valid accuracy 0.6496
Epoch 376, CIFAR-10 Batch 1:  loss 0.003108, train_accuracy 1, valid accuracy 0.6534
Epoch 377, CIFAR-10 Batch 1:  loss 0.002487, train_accuracy 1, valid accuracy 0.6542
Epoch 378, CIFAR-10 Batch 1:  loss 0.002142, train_accuracy 1, valid accuracy 0.6474
Epoch 379, CIFAR-10 Batch 1:  loss 0.002397, train_accuracy 1, valid accuracy 0.6576
Epoch 380, CIFAR-10 Batch 1:  loss 0.002262, train_accuracy 1, valid accuracy 0.6626
Epoch 381, CIFAR-10 Batch 1:  loss 0.002426, train_accuracy 1, valid accuracy 0.6468
Epoch 382, CIFAR-10 Batch 1:  loss 0.002947, train_accuracy 1, valid accuracy 0.6566
Epoch 383, CIFAR-10 Batch 1:  loss 0.003225, train_accuracy 1, valid accuracy 0.6584
Epoch 384, CIFAR-10 Batch 1:  loss 0.001134, train_accuracy 1, valid accuracy 0.6422
Epoch 385, CIFAR-10 Batch 1:  loss 0.001690, train_accuracy 1, valid accuracy 0.6472
Epoch 386, CIFAR-10 Batch 1:  loss 0.003496, train_accuracy 1, valid accuracy 0.6486
Epoch 387, CIFAR-10 Batch 1:  loss 0.003133, train_accuracy 1, valid accuracy 0.6474
Epoch 388, CIFAR-10 Batch 1:  loss 0.002191, train_accuracy 1, valid accuracy 0.6542
Epoch 389, CIFAR-10 Batch 1:  loss 0.002814, train_accuracy 1, valid accuracy 0.6552
Epoch 390, CIFAR-10 Batch 1:  loss 0.002282, train_accuracy 1, valid accuracy 0.6558
Epoch 391, CIFAR-10 Batch 1:  loss 0.002976, train_accuracy 1, valid accuracy 0.6338
Epoch 392, CIFAR-10 Batch 1:  loss 0.002132, train_accuracy 1, valid accuracy 0.6274
Epoch 393, CIFAR-10 Batch 1:  loss 0.001677, train_accuracy 1, valid accuracy 0.6622
Epoch 394, CIFAR-10 Batch 1:  loss 0.002307, train_accuracy 1, valid accuracy 0.6532
Epoch 395, CIFAR-10 Batch 1:  loss 0.001082, train_accuracy 1, valid accuracy 0.6632
Epoch 396, CIFAR-10 Batch 1:  loss 0.001204, train_accuracy 1, valid accuracy 0.6604
Epoch 397, CIFAR-10 Batch 1:  loss 0.001356, train_accuracy 1, valid accuracy 0.6552
Epoch 398, CIFAR-10 Batch 1:  loss 0.002081, train_accuracy 1, valid accuracy 0.6574
Epoch 399, CIFAR-10 Batch 1:  loss 0.012555, train_accuracy 1, valid accuracy 0.6384
Epoch 400, CIFAR-10 Batch 1:  loss 0.001756, train_accuracy 1, valid accuracy 0.6646
Epoch 401, CIFAR-10 Batch 1:  loss 0.002124, train_accuracy 1, valid accuracy 0.664
Epoch 402, CIFAR-10 Batch 1:  loss 0.001039, train_accuracy 1, valid accuracy 0.6554
Epoch 403, CIFAR-10 Batch 1:  loss 0.001239, train_accuracy 1, valid accuracy 0.6358
Epoch 404, CIFAR-10 Batch 1:  loss 0.001334, train_accuracy 1, valid accuracy 0.6506
Epoch 405, CIFAR-10 Batch 1:  loss 0.001183, train_accuracy 1, valid accuracy 0.6662
Epoch 406, CIFAR-10 Batch 1:  loss 0.001791, train_accuracy 1, valid accuracy 0.6626
Epoch 407, CIFAR-10 Batch 1:  loss 0.000715, train_accuracy 1, valid accuracy 0.6578
Epoch 408, CIFAR-10 Batch 1:  loss 0.001437, train_accuracy 1, valid accuracy 0.6536
Epoch 409, CIFAR-10 Batch 1:  loss 0.002518, train_accuracy 1, valid accuracy 0.6588
Epoch 410, CIFAR-10 Batch 1:  loss 0.000433, train_accuracy 1, valid accuracy 0.6566
Epoch 411, CIFAR-10 Batch 1:  loss 0.003549, train_accuracy 1, valid accuracy 0.6556
Epoch 412, CIFAR-10 Batch 1:  loss 0.001365, train_accuracy 1, valid accuracy 0.655
Epoch 413, CIFAR-10 Batch 1:  loss 0.002182, train_accuracy 1, valid accuracy 0.6628
Epoch 414, CIFAR-10 Batch 1:  loss 0.001548, train_accuracy 1, valid accuracy 0.6548
Epoch 415, CIFAR-10 Batch 1:  loss 0.003143, train_accuracy 1, valid accuracy 0.6486
Epoch 416, CIFAR-10 Batch 1:  loss 0.006639, train_accuracy 1, valid accuracy 0.6648
Epoch 417, CIFAR-10 Batch 1:  loss 0.001568, train_accuracy 1, valid accuracy 0.6632
Epoch 418, CIFAR-10 Batch 1:  loss 0.001408, train_accuracy 1, valid accuracy 0.65
Epoch 419, CIFAR-10 Batch 1:  loss 0.001898, train_accuracy 1, valid accuracy 0.661
Epoch 420, CIFAR-10 Batch 1:  loss 0.002104, train_accuracy 1, valid accuracy 0.6504
Epoch 421, CIFAR-10 Batch 1:  loss 0.001042, train_accuracy 1, valid accuracy 0.6496
Epoch 422, CIFAR-10 Batch 1:  loss 0.002364, train_accuracy 1, valid accuracy 0.6534
Epoch 423, CIFAR-10 Batch 1:  loss 0.001750, train_accuracy 1, valid accuracy 0.6596
Epoch 424, CIFAR-10 Batch 1:  loss 0.001213, train_accuracy 1, valid accuracy 0.668
Epoch 425, CIFAR-10 Batch 1:  loss 0.001071, train_accuracy 1, valid accuracy 0.6528
Epoch 426, CIFAR-10 Batch 1:  loss 0.001179, train_accuracy 1, valid accuracy 0.6544
Epoch 427, CIFAR-10 Batch 1:  loss 0.000980, train_accuracy 1, valid accuracy 0.6548
Epoch 428, CIFAR-10 Batch 1:  loss 0.001532, train_accuracy 1, valid accuracy 0.6566
Epoch 429, CIFAR-10 Batch 1:  loss 0.001147, train_accuracy 1, valid accuracy 0.6584
Epoch 430, CIFAR-10 Batch 1:  loss 0.001223, train_accuracy 1, valid accuracy 0.668
Epoch 431, CIFAR-10 Batch 1:  loss 0.001667, train_accuracy 1, valid accuracy 0.6598
Epoch 432, CIFAR-10 Batch 1:  loss 0.001341, train_accuracy 1, valid accuracy 0.658
Epoch 433, CIFAR-10 Batch 1:  loss 0.000827, train_accuracy 1, valid accuracy 0.6642
Epoch 434, CIFAR-10 Batch 1:  loss 0.000632, train_accuracy 1, valid accuracy 0.6738
Epoch 435, CIFAR-10 Batch 1:  loss 0.000756, train_accuracy 1, valid accuracy 0.6676
Epoch 436, CIFAR-10 Batch 1:  loss 0.001162, train_accuracy 1, valid accuracy 0.6644
Epoch 437, CIFAR-10 Batch 1:  loss 0.000632, train_accuracy 1, valid accuracy 0.6604
Epoch 438, CIFAR-10 Batch 1:  loss 0.001025, train_accuracy 1, valid accuracy 0.6586
Epoch 439, CIFAR-10 Batch 1:  loss 0.001597, train_accuracy 1, valid accuracy 0.6598
Epoch 440, CIFAR-10 Batch 1:  loss 0.001458, train_accuracy 1, valid accuracy 0.6564
Epoch 441, CIFAR-10 Batch 1:  loss 0.001659, train_accuracy 1, valid accuracy 0.6598
Epoch 442, CIFAR-10 Batch 1:  loss 0.001052, train_accuracy 1, valid accuracy 0.6674
Epoch 443, CIFAR-10 Batch 1:  loss 0.000611, train_accuracy 1, valid accuracy 0.6658
Epoch 444, CIFAR-10 Batch 1:  loss 0.000758, train_accuracy 1, valid accuracy 0.6716
Epoch 445, CIFAR-10 Batch 1:  loss 0.002168, train_accuracy 1, valid accuracy 0.6652
Epoch 446, CIFAR-10 Batch 1:  loss 0.001309, train_accuracy 1, valid accuracy 0.6702
Epoch 447, CIFAR-10 Batch 1:  loss 0.000422, train_accuracy 1, valid accuracy 0.6608
Epoch 448, CIFAR-10 Batch 1:  loss 0.001125, train_accuracy 1, valid accuracy 0.6586
Epoch 449, CIFAR-10 Batch 1:  loss 0.000960, train_accuracy 1, valid accuracy 0.6672
Epoch 450, CIFAR-10 Batch 1:  loss 0.001193, train_accuracy 1, valid accuracy 0.6652
Epoch 451, CIFAR-10 Batch 1:  loss 0.001166, train_accuracy 1, valid accuracy 0.6608
Epoch 452, CIFAR-10 Batch 1:  loss 0.000572, train_accuracy 1, valid accuracy 0.6562
Epoch 453, CIFAR-10 Batch 1:  loss 0.000955, train_accuracy 1, valid accuracy 0.6612
Epoch 454, CIFAR-10 Batch 1:  loss 0.000519, train_accuracy 1, valid accuracy 0.6668
Epoch 455, CIFAR-10 Batch 1:  loss 0.000889, train_accuracy 1, valid accuracy 0.6682
Epoch 456, CIFAR-10 Batch 1:  loss 0.000944, train_accuracy 1, valid accuracy 0.6712
Epoch 457, CIFAR-10 Batch 1:  loss 0.000535, train_accuracy 1, valid accuracy 0.6638
Epoch 458, CIFAR-10 Batch 1:  loss 0.000458, train_accuracy 1, valid accuracy 0.6642
Epoch 459, CIFAR-10 Batch 1:  loss 0.000616, train_accuracy 1, valid accuracy 0.6522
Epoch 460, CIFAR-10 Batch 1:  loss 0.000553, train_accuracy 1, valid accuracy 0.6508
Epoch 461, CIFAR-10 Batch 1:  loss 0.000804, train_accuracy 1, valid accuracy 0.6578
Epoch 462, CIFAR-10 Batch 1:  loss 0.000373, train_accuracy 1, valid accuracy 0.662
Epoch 463, CIFAR-10 Batch 1:  loss 0.000789, train_accuracy 1, valid accuracy 0.6426
Epoch 464, CIFAR-10 Batch 1:  loss 0.000908, train_accuracy 1, valid accuracy 0.649
Epoch 465, CIFAR-10 Batch 1:  loss 0.000429, train_accuracy 1, valid accuracy 0.6598
Epoch 466, CIFAR-10 Batch 1:  loss 0.000359, train_accuracy 1, valid accuracy 0.6494
Epoch 467, CIFAR-10 Batch 1:  loss 0.000324, train_accuracy 1, valid accuracy 0.6542
Epoch 468, CIFAR-10 Batch 1:  loss 0.000493, train_accuracy 1, valid accuracy 0.6528
Epoch 469, CIFAR-10 Batch 1:  loss 0.000605, train_accuracy 1, valid accuracy 0.6616
Epoch 470, CIFAR-10 Batch 1:  loss 0.000629, train_accuracy 1, valid accuracy 0.6618
Epoch 471, CIFAR-10 Batch 1:  loss 0.000419, train_accuracy 1, valid accuracy 0.661
Epoch 472, CIFAR-10 Batch 1:  loss 0.000293, train_accuracy 1, valid accuracy 0.6592
Epoch 473, CIFAR-10 Batch 1:  loss 0.000297, train_accuracy 1, valid accuracy 0.6614
Epoch 474, CIFAR-10 Batch 1:  loss 0.000368, train_accuracy 1, valid accuracy 0.6672
Epoch 475, CIFAR-10 Batch 1:  loss 0.001155, train_accuracy 1, valid accuracy 0.6654
Epoch 476, CIFAR-10 Batch 1:  loss 0.001204, train_accuracy 1, valid accuracy 0.6586
Epoch 477, CIFAR-10 Batch 1:  loss 0.000281, train_accuracy 1, valid accuracy 0.65
Epoch 478, CIFAR-10 Batch 1:  loss 0.000983, train_accuracy 1, valid accuracy 0.6692
Epoch 479, CIFAR-10 Batch 1:  loss 0.000416, train_accuracy 1, valid accuracy 0.6618
Epoch 480, CIFAR-10 Batch 1:  loss 0.000666, train_accuracy 1, valid accuracy 0.6616
Epoch 481, CIFAR-10 Batch 1:  loss 0.001717, train_accuracy 1, valid accuracy 0.6466
Epoch 482, CIFAR-10 Batch 1:  loss 0.000954, train_accuracy 1, valid accuracy 0.6676
Epoch 483, CIFAR-10 Batch 1:  loss 0.000280, train_accuracy 1, valid accuracy 0.6648
Epoch 484, CIFAR-10 Batch 1:  loss 0.000315, train_accuracy 1, valid accuracy 0.6688
Epoch 485, CIFAR-10 Batch 1:  loss 0.000908, train_accuracy 1, valid accuracy 0.6622
Epoch 486, CIFAR-10 Batch 1:  loss 0.000757, train_accuracy 1, valid accuracy 0.6574
Epoch 487, CIFAR-10 Batch 1:  loss 0.000513, train_accuracy 1, valid accuracy 0.6618
Epoch 488, CIFAR-10 Batch 1:  loss 0.000408, train_accuracy 1, valid accuracy 0.6562
Epoch 489, CIFAR-10 Batch 1:  loss 0.000456, train_accuracy 1, valid accuracy 0.6596
Epoch 490, CIFAR-10 Batch 1:  loss 0.000498, train_accuracy 1, valid accuracy 0.6672
Epoch 491, CIFAR-10 Batch 1:  loss 0.000696, train_accuracy 1, valid accuracy 0.6584
Epoch 492, CIFAR-10 Batch 1:  loss 0.000470, train_accuracy 1, valid accuracy 0.6568
Epoch 493, CIFAR-10 Batch 1:  loss 0.000627, train_accuracy 1, valid accuracy 0.6422
Epoch 494, CIFAR-10 Batch 1:  loss 0.000269, train_accuracy 1, valid accuracy 0.668
Epoch 495, CIFAR-10 Batch 1:  loss 0.000415, train_accuracy 1, valid accuracy 0.6566
Epoch 496, CIFAR-10 Batch 1:  loss 0.000193, train_accuracy 1, valid accuracy 0.6596
Epoch 497, CIFAR-10 Batch 1:  loss 0.000229, train_accuracy 1, valid accuracy 0.6672
Epoch 498, CIFAR-10 Batch 1:  loss 0.000169, train_accuracy 1, valid accuracy 0.6612
Epoch 499, CIFAR-10 Batch 1:  loss 0.000310, train_accuracy 1, valid accuracy 0.6624
Epoch 500, CIFAR-10 Batch 1:  loss 0.000485, train_accuracy 1, valid accuracy 0.659
Epoch 501, CIFAR-10 Batch 1:  loss 0.000379, train_accuracy 1, valid accuracy 0.6594
Epoch 502, CIFAR-10 Batch 1:  loss 0.000168, train_accuracy 1, valid accuracy 0.6652
Epoch 503, CIFAR-10 Batch 1:  loss 0.000373, train_accuracy 1, valid accuracy 0.6598
Epoch 504, CIFAR-10 Batch 1:  loss 0.000186, train_accuracy 1, valid accuracy 0.6662
Epoch 505, CIFAR-10 Batch 1:  loss 0.000132, train_accuracy 1, valid accuracy 0.671
Epoch 506, CIFAR-10 Batch 1:  loss 0.000366, train_accuracy 1, valid accuracy 0.6532
Epoch 507, CIFAR-10 Batch 1:  loss 0.000359, train_accuracy 1, valid accuracy 0.6612
Epoch 508, CIFAR-10 Batch 1:  loss 0.000201, train_accuracy 1, valid accuracy 0.6676
Epoch 509, CIFAR-10 Batch 1:  loss 0.000209, train_accuracy 1, valid accuracy 0.665
Epoch 510, CIFAR-10 Batch 1:  loss 0.000655, train_accuracy 1, valid accuracy 0.664
Epoch 511, CIFAR-10 Batch 1:  loss 0.000388, train_accuracy 1, valid accuracy 0.6686
Epoch 512, CIFAR-10 Batch 1:  loss 0.000952, train_accuracy 1, valid accuracy 0.6674
Epoch 513, CIFAR-10 Batch 1:  loss 0.000184, train_accuracy 1, valid accuracy 0.667
Epoch 514, CIFAR-10 Batch 1:  loss 0.000127, train_accuracy 1, valid accuracy 0.67
Epoch 515, CIFAR-10 Batch 1:  loss 0.000190, train_accuracy 1, valid accuracy 0.6668
Epoch 516, CIFAR-10 Batch 1:  loss 0.000257, train_accuracy 1, valid accuracy 0.6558
Epoch 517, CIFAR-10 Batch 1:  loss 0.000261, train_accuracy 1, valid accuracy 0.663
Epoch 518, CIFAR-10 Batch 1:  loss 0.000113, train_accuracy 1, valid accuracy 0.6568
Epoch 519, CIFAR-10 Batch 1:  loss 0.000371, train_accuracy 1, valid accuracy 0.6648
Epoch 520, CIFAR-10 Batch 1:  loss 0.000338, train_accuracy 1, valid accuracy 0.6678
Epoch 521, CIFAR-10 Batch 1:  loss 0.000117, train_accuracy 1, valid accuracy 0.6762
Epoch 522, CIFAR-10 Batch 1:  loss 0.000803, train_accuracy 1, valid accuracy 0.6794
Epoch 523, CIFAR-10 Batch 1:  loss 0.000222, train_accuracy 1, valid accuracy 0.6724
Epoch 524, CIFAR-10 Batch 1:  loss 0.000253, train_accuracy 1, valid accuracy 0.673
Epoch 525, CIFAR-10 Batch 1:  loss 0.000327, train_accuracy 1, valid accuracy 0.6626
Epoch 526, CIFAR-10 Batch 1:  loss 0.000458, train_accuracy 1, valid accuracy 0.677
Epoch 527, CIFAR-10 Batch 1:  loss 0.000209, train_accuracy 1, valid accuracy 0.6676
Epoch 528, CIFAR-10 Batch 1:  loss 0.000435, train_accuracy 1, valid accuracy 0.6742
Epoch 529, CIFAR-10 Batch 1:  loss 0.000226, train_accuracy 1, valid accuracy 0.671
Epoch 530, CIFAR-10 Batch 1:  loss 0.000470, train_accuracy 1, valid accuracy 0.6678
Epoch 531, CIFAR-10 Batch 1:  loss 0.000199, train_accuracy 1, valid accuracy 0.6722
Epoch 532, CIFAR-10 Batch 1:  loss 0.000101, train_accuracy 1, valid accuracy 0.6754
Epoch 533, CIFAR-10 Batch 1:  loss 0.000056, train_accuracy 1, valid accuracy 0.6734
Epoch 534, CIFAR-10 Batch 1:  loss 0.000155, train_accuracy 1, valid accuracy 0.6714
Epoch 535, CIFAR-10 Batch 1:  loss 0.000315, train_accuracy 1, valid accuracy 0.6728
Epoch 536, CIFAR-10 Batch 1:  loss 0.000136, train_accuracy 1, valid accuracy 0.6658
Epoch 537, CIFAR-10 Batch 1:  loss 0.000200, train_accuracy 1, valid accuracy 0.6736
Epoch 538, CIFAR-10 Batch 1:  loss 0.000166, train_accuracy 1, valid accuracy 0.6668
Epoch 539, CIFAR-10 Batch 1:  loss 0.000352, train_accuracy 1, valid accuracy 0.6662
Epoch 540, CIFAR-10 Batch 1:  loss 0.000266, train_accuracy 1, valid accuracy 0.6676
Epoch 541, CIFAR-10 Batch 1:  loss 0.000167, train_accuracy 1, valid accuracy 0.649
Epoch 542, CIFAR-10 Batch 1:  loss 0.000235, train_accuracy 1, valid accuracy 0.663
Epoch 543, CIFAR-10 Batch 1:  loss 0.000418, train_accuracy 1, valid accuracy 0.6714
Epoch 544, CIFAR-10 Batch 1:  loss 0.000128, train_accuracy 1, valid accuracy 0.6658
Epoch 545, CIFAR-10 Batch 1:  loss 0.000699, train_accuracy 1, valid accuracy 0.6654
Epoch 546, CIFAR-10 Batch 1:  loss 0.000207, train_accuracy 1, valid accuracy 0.662
Epoch 547, CIFAR-10 Batch 1:  loss 0.000095, train_accuracy 1, valid accuracy 0.6712
Epoch 548, CIFAR-10 Batch 1:  loss 0.000106, train_accuracy 1, valid accuracy 0.6652
Epoch 549, CIFAR-10 Batch 1:  loss 0.000213, train_accuracy 1, valid accuracy 0.6608
Epoch 550, CIFAR-10 Batch 1:  loss 0.000131, train_accuracy 1, valid accuracy 0.6668
Epoch 551, CIFAR-10 Batch 1:  loss 0.000369, train_accuracy 1, valid accuracy 0.6662
Epoch 552, CIFAR-10 Batch 1:  loss 0.000797, train_accuracy 1, valid accuracy 0.6706
Epoch 553, CIFAR-10 Batch 1:  loss 0.000184, train_accuracy 1, valid accuracy 0.6656
Epoch 554, CIFAR-10 Batch 1:  loss 0.000325, train_accuracy 1, valid accuracy 0.669
Epoch 555, CIFAR-10 Batch 1:  loss 0.000118, train_accuracy 1, valid accuracy 0.6696
Epoch 556, CIFAR-10 Batch 1:  loss 0.000164, train_accuracy 1, valid accuracy 0.6708
Epoch 557, CIFAR-10 Batch 1:  loss 0.000115, train_accuracy 1, valid accuracy 0.6676
Epoch 558, CIFAR-10 Batch 1:  loss 0.001063, train_accuracy 1, valid accuracy 0.6646
Epoch 559, CIFAR-10 Batch 1:  loss 0.000054, train_accuracy 1, valid accuracy 0.6752
Epoch 560, CIFAR-10 Batch 1:  loss 0.000502, train_accuracy 1, valid accuracy 0.677
Epoch 561, CIFAR-10 Batch 1:  loss 0.000101, train_accuracy 1, valid accuracy 0.6686
Epoch 562, CIFAR-10 Batch 1:  loss 0.000069, train_accuracy 1, valid accuracy 0.677
Epoch 563, CIFAR-10 Batch 1:  loss 0.000403, train_accuracy 1, valid accuracy 0.6794
Epoch 564, CIFAR-10 Batch 1:  loss 0.000090, train_accuracy 1, valid accuracy 0.6756
Epoch 565, CIFAR-10 Batch 1:  loss 0.000108, train_accuracy 1, valid accuracy 0.6762
Epoch 566, CIFAR-10 Batch 1:  loss 0.001369, train_accuracy 1, valid accuracy 0.6708
Epoch 567, CIFAR-10 Batch 1:  loss 0.000265, train_accuracy 1, valid accuracy 0.6738
Epoch 568, CIFAR-10 Batch 1:  loss 0.000393, train_accuracy 1, valid accuracy 0.6762
Epoch 569, CIFAR-10 Batch 1:  loss 0.000223, train_accuracy 1, valid accuracy 0.6702
Epoch 570, CIFAR-10 Batch 1:  loss 0.000118, train_accuracy 1, valid accuracy 0.6714
Epoch 571, CIFAR-10 Batch 1:  loss 0.000046, train_accuracy 1, valid accuracy 0.6648
Epoch 572, CIFAR-10 Batch 1:  loss 0.000092, train_accuracy 1, valid accuracy 0.67
Epoch 573, CIFAR-10 Batch 1:  loss 0.000060, train_accuracy 1, valid accuracy 0.6706
Epoch 574, CIFAR-10 Batch 1:  loss 0.000274, train_accuracy 1, valid accuracy 0.6756
Epoch 575, CIFAR-10 Batch 1:  loss 0.000056, train_accuracy 1, valid accuracy 0.667
Epoch 576, CIFAR-10 Batch 1:  loss 0.000073, train_accuracy 1, valid accuracy 0.6678
Epoch 577, CIFAR-10 Batch 1:  loss 0.000187, train_accuracy 1, valid accuracy 0.6686
Epoch 578, CIFAR-10 Batch 1:  loss 0.000117, train_accuracy 1, valid accuracy 0.6614
Epoch 579, CIFAR-10 Batch 1:  loss 0.000073, train_accuracy 1, valid accuracy 0.6774
Epoch 580, CIFAR-10 Batch 1:  loss 0.000107, train_accuracy 1, valid accuracy 0.6726
Epoch 581, CIFAR-10 Batch 1:  loss 0.000164, train_accuracy 1, valid accuracy 0.6688
Epoch 582, CIFAR-10 Batch 1:  loss 0.000124, train_accuracy 1, valid accuracy 0.677
Epoch 583, CIFAR-10 Batch 1:  loss 0.000521, train_accuracy 1, valid accuracy 0.67
Epoch 584, CIFAR-10 Batch 1:  loss 0.000151, train_accuracy 1, valid accuracy 0.6742
Epoch 585, CIFAR-10 Batch 1:  loss 0.000101, train_accuracy 1, valid accuracy 0.664
Epoch 586, CIFAR-10 Batch 1:  loss 0.000046, train_accuracy 1, valid accuracy 0.6752
Epoch 587, CIFAR-10 Batch 1:  loss 0.000042, train_accuracy 1, valid accuracy 0.6744
Epoch 588, CIFAR-10 Batch 1:  loss 0.000044, train_accuracy 1, valid accuracy 0.6656
Epoch 589, CIFAR-10 Batch 1:  loss 0.000055, train_accuracy 1, valid accuracy 0.6716
Epoch 590, CIFAR-10 Batch 1:  loss 0.000066, train_accuracy 1, valid accuracy 0.6666
Epoch 591, CIFAR-10 Batch 1:  loss 0.000104, train_accuracy 1, valid accuracy 0.6718
Epoch 592, CIFAR-10 Batch 1:  loss 0.000039, train_accuracy 1, valid accuracy 0.6594
Epoch 593, CIFAR-10 Batch 1:  loss 0.000223, train_accuracy 1, valid accuracy 0.6738
Epoch 594, CIFAR-10 Batch 1:  loss 0.000131, train_accuracy 1, valid accuracy 0.6708
Epoch 595, CIFAR-10 Batch 1:  loss 0.000040, train_accuracy 1, valid accuracy 0.6594
Epoch 596, CIFAR-10 Batch 1:  loss 0.000243, train_accuracy 1, valid accuracy 0.6714
Epoch 597, CIFAR-10 Batch 1:  loss 0.000375, train_accuracy 1, valid accuracy 0.658
Epoch 598, CIFAR-10 Batch 1:  loss 0.000060, train_accuracy 1, valid accuracy 0.6758
Epoch 599, CIFAR-10 Batch 1:  loss 0.000651, train_accuracy 1, valid accuracy 0.6804
Epoch 600, CIFAR-10 Batch 1:  loss 0.000074, train_accuracy 1, valid accuracy 0.6738

Fully Train the Model

Now that you got a good accuracy with a single CIFAR-10 batch, try it with all five batches.



In [161]:

    
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
save_model_path = './image_classification'

print('Training...')
with tf.Session() as sess:
    # Initializing the variables
    sess.run(tf.global_variables_initializer())
    
    # Training cycle
    for epoch in range(epochs):
        # Loop over all batches
        n_batches = 5
        for batch_i in range(1, n_batches + 1):
            for batch_features, batch_labels in helper.load_preprocess_training_batch(batch_i, batch_size):
                train_neural_network(sess, optimizer, keep_probability, batch_features, batch_labels)
            print('Epoch {:>2}, CIFAR-10 Batch {}:  '.format(epoch + 1, batch_i), end='')
            print_stats(sess, batch_features, batch_labels, cost, accuracy)
            
    # Save Model
    saver = tf.train.Saver()
    save_path = saver.save(sess, save_model_path)









    



Training...
Epoch  1, CIFAR-10 Batch 1:  loss 2.271044, train_accuracy 0.175, valid accuracy 0.134
Epoch  1, CIFAR-10 Batch 2:  loss 2.334074, train_accuracy 0.05, valid accuracy 0.0946
Epoch  1, CIFAR-10 Batch 3:  loss 2.319691, train_accuracy 0.05, valid accuracy 0.0946
Epoch  1, CIFAR-10 Batch 4:  loss 2.285172, train_accuracy 0.15, valid accuracy 0.0946
Epoch  1, CIFAR-10 Batch 5:  loss 2.306232, train_accuracy 0.1, valid accuracy 0.0946
Epoch  2, CIFAR-10 Batch 1:  loss 2.307927, train_accuracy 0.1, valid accuracy 0.0946
Epoch  2, CIFAR-10 Batch 2:  loss 2.307360, train_accuracy 0.05, valid accuracy 0.0946
Epoch  2, CIFAR-10 Batch 3:  loss 2.310335, train_accuracy 0.05, valid accuracy 0.0946
Epoch  2, CIFAR-10 Batch 4:  loss 2.286575, train_accuracy 0.15, valid accuracy 0.0946
Epoch  2, CIFAR-10 Batch 5:  loss 2.313197, train_accuracy 0.1, valid accuracy 0.0946
Epoch  3, CIFAR-10 Batch 1:  loss 2.321627, train_accuracy 0.1, valid accuracy 0.0946
Epoch  3, CIFAR-10 Batch 2:  loss 2.318487, train_accuracy 0.05, valid accuracy 0.0946
Epoch  3, CIFAR-10 Batch 3:  loss 2.341111, train_accuracy 0.05, valid accuracy 0.0946
Epoch  3, CIFAR-10 Batch 4:  loss 2.281078, train_accuracy 0.15, valid accuracy 0.0946
Epoch  3, CIFAR-10 Batch 5:  loss 2.370485, train_accuracy 0.1, valid accuracy 0.0946
Epoch  4, CIFAR-10 Batch 1:  loss 2.385434, train_accuracy 0.1, valid accuracy 0.0946
Epoch  4, CIFAR-10 Batch 2:  loss 2.368228, train_accuracy 0.05, valid accuracy 0.0946
Epoch  4, CIFAR-10 Batch 3:  loss 2.389083, train_accuracy 0.05, valid accuracy 0.0946
Epoch  4, CIFAR-10 Batch 4:  loss 2.311312, train_accuracy 0.15, valid accuracy 0.0946
Epoch  4, CIFAR-10 Batch 5:  loss 2.502681, train_accuracy 0.1, valid accuracy 0.0946
Epoch  5, CIFAR-10 Batch 1:  loss 2.443245, train_accuracy 0.1, valid accuracy 0.0992
Epoch  5, CIFAR-10 Batch 2:  loss 2.379207, train_accuracy 0.075, valid accuracy 0.1008
Epoch  5, CIFAR-10 Batch 3:  loss 2.220908, train_accuracy 0.075, valid accuracy 0.1048
Epoch  5, CIFAR-10 Batch 4:  loss 2.261624, train_accuracy 0.15, valid accuracy 0.1144
Epoch  5, CIFAR-10 Batch 5:  loss 2.476570, train_accuracy 0.15, valid accuracy 0.1138
Epoch  6, CIFAR-10 Batch 1:  loss 2.501565, train_accuracy 0.1, valid accuracy 0.1142
Epoch  6, CIFAR-10 Batch 2:  loss 2.192663, train_accuracy 0.125, valid accuracy 0.1344
Epoch  6, CIFAR-10 Batch 3:  loss 2.125289, train_accuracy 0.125, valid accuracy 0.125
Epoch  6, CIFAR-10 Batch 4:  loss 2.102998, train_accuracy 0.2, valid accuracy 0.1438
Epoch  6, CIFAR-10 Batch 5:  loss 2.193203, train_accuracy 0.2, valid accuracy 0.16
Epoch  7, CIFAR-10 Batch 1:  loss 2.465976, train_accuracy 0.125, valid accuracy 0.156
Epoch  7, CIFAR-10 Batch 2:  loss 2.047404, train_accuracy 0.1, valid accuracy 0.1662
Epoch  7, CIFAR-10 Batch 3:  loss 2.017732, train_accuracy 0.15, valid accuracy 0.1562
Epoch  7, CIFAR-10 Batch 4:  loss 2.044284, train_accuracy 0.2, valid accuracy 0.151
Epoch  7, CIFAR-10 Batch 5:  loss 2.055947, train_accuracy 0.3, valid accuracy 0.2
Epoch  8, CIFAR-10 Batch 1:  loss 2.418849, train_accuracy 0.15, valid accuracy 0.1934
Epoch  8, CIFAR-10 Batch 2:  loss 2.070016, train_accuracy 0.1, valid accuracy 0.159
Epoch  8, CIFAR-10 Batch 3:  loss 1.891150, train_accuracy 0.175, valid accuracy 0.217
Epoch  8, CIFAR-10 Batch 4:  loss 1.920752, train_accuracy 0.225, valid accuracy 0.1978
Epoch  8, CIFAR-10 Batch 5:  loss 2.070249, train_accuracy 0.275, valid accuracy 0.2064
Epoch  9, CIFAR-10 Batch 1:  loss 2.481122, train_accuracy 0.125, valid accuracy 0.1798
Epoch  9, CIFAR-10 Batch 2:  loss 1.980615, train_accuracy 0.125, valid accuracy 0.1828
Epoch  9, CIFAR-10 Batch 3:  loss 1.919423, train_accuracy 0.15, valid accuracy 0.2076
Epoch  9, CIFAR-10 Batch 4:  loss 1.918592, train_accuracy 0.2, valid accuracy 0.2006
Epoch  9, CIFAR-10 Batch 5:  loss 2.085673, train_accuracy 0.3, valid accuracy 0.198
Epoch 10, CIFAR-10 Batch 1:  loss 2.339881, train_accuracy 0.175, valid accuracy 0.2148
Epoch 10, CIFAR-10 Batch 2:  loss 1.846202, train_accuracy 0.2, valid accuracy 0.2282
Epoch 10, CIFAR-10 Batch 3:  loss 1.947275, train_accuracy 0.15, valid accuracy 0.1998
Epoch 10, CIFAR-10 Batch 4:  loss 1.833250, train_accuracy 0.275, valid accuracy 0.2656
Epoch 10, CIFAR-10 Batch 5:  loss 2.078455, train_accuracy 0.275, valid accuracy 0.2048
Epoch 11, CIFAR-10 Batch 1:  loss 2.128017, train_accuracy 0.275, valid accuracy 0.2804
Epoch 11, CIFAR-10 Batch 2:  loss 1.970453, train_accuracy 0.175, valid accuracy 0.1824
Epoch 11, CIFAR-10 Batch 3:  loss 1.869625, train_accuracy 0.175, valid accuracy 0.221
Epoch 11, CIFAR-10 Batch 4:  loss 1.962877, train_accuracy 0.175, valid accuracy 0.2162
Epoch 11, CIFAR-10 Batch 5:  loss 1.893363, train_accuracy 0.275, valid accuracy 0.2574
Epoch 12, CIFAR-10 Batch 1:  loss 2.254345, train_accuracy 0.225, valid accuracy 0.2496
Epoch 12, CIFAR-10 Batch 2:  loss 1.946390, train_accuracy 0.175, valid accuracy 0.1932
Epoch 12, CIFAR-10 Batch 3:  loss 1.805210, train_accuracy 0.25, valid accuracy 0.2492
Epoch 12, CIFAR-10 Batch 4:  loss 1.899080, train_accuracy 0.25, valid accuracy 0.2502
Epoch 12, CIFAR-10 Batch 5:  loss 2.063457, train_accuracy 0.2, valid accuracy 0.221
Epoch 13, CIFAR-10 Batch 1:  loss 2.104213, train_accuracy 0.225, valid accuracy 0.2798
Epoch 13, CIFAR-10 Batch 2:  loss 2.042329, train_accuracy 0.2, valid accuracy 0.1964
Epoch 13, CIFAR-10 Batch 3:  loss 1.853367, train_accuracy 0.25, valid accuracy 0.2508
Epoch 13, CIFAR-10 Batch 4:  loss 1.905025, train_accuracy 0.275, valid accuracy 0.2384
Epoch 13, CIFAR-10 Batch 5:  loss 2.033168, train_accuracy 0.275, valid accuracy 0.2478
Epoch 14, CIFAR-10 Batch 1:  loss 2.131140, train_accuracy 0.3, valid accuracy 0.273
Epoch 14, CIFAR-10 Batch 2:  loss 1.948706, train_accuracy 0.175, valid accuracy 0.234
Epoch 14, CIFAR-10 Batch 3:  loss 1.781045, train_accuracy 0.375, valid accuracy 0.2864
Epoch 14, CIFAR-10 Batch 4:  loss 1.832924, train_accuracy 0.3, valid accuracy 0.2776
Epoch 14, CIFAR-10 Batch 5:  loss 1.867050, train_accuracy 0.35, valid accuracy 0.2902
Epoch 15, CIFAR-10 Batch 1:  loss 1.978346, train_accuracy 0.3, valid accuracy 0.3236
Epoch 15, CIFAR-10 Batch 2:  loss 1.877707, train_accuracy 0.225, valid accuracy 0.267
Epoch 15, CIFAR-10 Batch 3:  loss 1.877368, train_accuracy 0.375, valid accuracy 0.2662
Epoch 15, CIFAR-10 Batch 4:  loss 1.854818, train_accuracy 0.325, valid accuracy 0.2778
Epoch 15, CIFAR-10 Batch 5:  loss 1.991333, train_accuracy 0.375, valid accuracy 0.2736
Epoch 16, CIFAR-10 Batch 1:  loss 2.116701, train_accuracy 0.275, valid accuracy 0.2942
Epoch 16, CIFAR-10 Batch 2:  loss 1.846279, train_accuracy 0.225, valid accuracy 0.2826
Epoch 16, CIFAR-10 Batch 3:  loss 1.670655, train_accuracy 0.375, valid accuracy 0.3392
Epoch 16, CIFAR-10 Batch 4:  loss 1.822055, train_accuracy 0.325, valid accuracy 0.3122
Epoch 16, CIFAR-10 Batch 5:  loss 1.890085, train_accuracy 0.275, valid accuracy 0.3046
Epoch 17, CIFAR-10 Batch 1:  loss 2.092095, train_accuracy 0.35, valid accuracy 0.32
Epoch 17, CIFAR-10 Batch 2:  loss 1.924997, train_accuracy 0.225, valid accuracy 0.2596
Epoch 17, CIFAR-10 Batch 3:  loss 1.678016, train_accuracy 0.4, valid accuracy 0.3296
Epoch 17, CIFAR-10 Batch 4:  loss 1.846020, train_accuracy 0.325, valid accuracy 0.302
Epoch 17, CIFAR-10 Batch 5:  loss 1.837720, train_accuracy 0.35, valid accuracy 0.3238
Epoch 18, CIFAR-10 Batch 1:  loss 1.870719, train_accuracy 0.375, valid accuracy 0.3782
Epoch 18, CIFAR-10 Batch 2:  loss 1.829102, train_accuracy 0.2, valid accuracy 0.2938
Epoch 18, CIFAR-10 Batch 3:  loss 1.625607, train_accuracy 0.425, valid accuracy 0.3246
Epoch 18, CIFAR-10 Batch 4:  loss 1.793105, train_accuracy 0.35, valid accuracy 0.329
Epoch 18, CIFAR-10 Batch 5:  loss 1.873678, train_accuracy 0.35, valid accuracy 0.3198
Epoch 19, CIFAR-10 Batch 1:  loss 1.965288, train_accuracy 0.35, valid accuracy 0.3586
Epoch 19, CIFAR-10 Batch 2:  loss 1.878035, train_accuracy 0.25, valid accuracy 0.2962
Epoch 19, CIFAR-10 Batch 3:  loss 1.707242, train_accuracy 0.375, valid accuracy 0.3402
Epoch 19, CIFAR-10 Batch 4:  loss 1.716683, train_accuracy 0.35, valid accuracy 0.3554
Epoch 19, CIFAR-10 Batch 5:  loss 1.896444, train_accuracy 0.35, valid accuracy 0.3142
Epoch 20, CIFAR-10 Batch 1:  loss 1.876846, train_accuracy 0.35, valid accuracy 0.3732
Epoch 20, CIFAR-10 Batch 2:  loss 1.695452, train_accuracy 0.325, valid accuracy 0.3526
Epoch 20, CIFAR-10 Batch 3:  loss 1.681953, train_accuracy 0.45, valid accuracy 0.3548
Epoch 20, CIFAR-10 Batch 4:  loss 1.750260, train_accuracy 0.35, valid accuracy 0.344
Epoch 20, CIFAR-10 Batch 5:  loss 1.785180, train_accuracy 0.35, valid accuracy 0.3528
Epoch 21, CIFAR-10 Batch 1:  loss 1.802564, train_accuracy 0.375, valid accuracy 0.4086
Epoch 21, CIFAR-10 Batch 2:  loss 1.751834, train_accuracy 0.3, valid accuracy 0.3188
Epoch 21, CIFAR-10 Batch 3:  loss 1.567519, train_accuracy 0.375, valid accuracy 0.3978
Epoch 21, CIFAR-10 Batch 4:  loss 1.654115, train_accuracy 0.425, valid accuracy 0.3906
Epoch 21, CIFAR-10 Batch 5:  loss 1.763160, train_accuracy 0.475, valid accuracy 0.3396
Epoch 22, CIFAR-10 Batch 1:  loss 1.721428, train_accuracy 0.4, valid accuracy 0.4288
Epoch 22, CIFAR-10 Batch 2:  loss 1.662067, train_accuracy 0.35, valid accuracy 0.342
Epoch 22, CIFAR-10 Batch 3:  loss 1.489990, train_accuracy 0.45, valid accuracy 0.4216
Epoch 22, CIFAR-10 Batch 4:  loss 1.760961, train_accuracy 0.375, valid accuracy 0.3584
Epoch 22, CIFAR-10 Batch 5:  loss 1.926632, train_accuracy 0.35, valid accuracy 0.3078
Epoch 23, CIFAR-10 Batch 1:  loss 1.970618, train_accuracy 0.325, valid accuracy 0.381
Epoch 23, CIFAR-10 Batch 2:  loss 1.663108, train_accuracy 0.325, valid accuracy 0.3704
Epoch 23, CIFAR-10 Batch 3:  loss 1.655905, train_accuracy 0.375, valid accuracy 0.368
Epoch 23, CIFAR-10 Batch 4:  loss 1.705709, train_accuracy 0.45, valid accuracy 0.381
Epoch 23, CIFAR-10 Batch 5:  loss 1.624981, train_accuracy 0.425, valid accuracy 0.3942
Epoch 24, CIFAR-10 Batch 1:  loss 1.690188, train_accuracy 0.375, valid accuracy 0.4266
Epoch 24, CIFAR-10 Batch 2:  loss 1.580999, train_accuracy 0.375, valid accuracy 0.39
Epoch 24, CIFAR-10 Batch 3:  loss 1.524716, train_accuracy 0.375, valid accuracy 0.413
Epoch 24, CIFAR-10 Batch 4:  loss 1.641106, train_accuracy 0.425, valid accuracy 0.4108
Epoch 24, CIFAR-10 Batch 5:  loss 1.508768, train_accuracy 0.45, valid accuracy 0.4282
Epoch 25, CIFAR-10 Batch 1:  loss 1.849681, train_accuracy 0.35, valid accuracy 0.4168
Epoch 25, CIFAR-10 Batch 2:  loss 1.664115, train_accuracy 0.35, valid accuracy 0.39
Epoch 25, CIFAR-10 Batch 3:  loss 1.486332, train_accuracy 0.5, valid accuracy 0.4314
Epoch 25, CIFAR-10 Batch 4:  loss 1.507466, train_accuracy 0.45, valid accuracy 0.4344
Epoch 25, CIFAR-10 Batch 5:  loss 1.549412, train_accuracy 0.45, valid accuracy 0.3986
Epoch 26, CIFAR-10 Batch 1:  loss 1.529082, train_accuracy 0.425, valid accuracy 0.459
Epoch 26, CIFAR-10 Batch 2:  loss 1.675216, train_accuracy 0.3, valid accuracy 0.3722
Epoch 26, CIFAR-10 Batch 3:  loss 1.445127, train_accuracy 0.475, valid accuracy 0.4428
Epoch 26, CIFAR-10 Batch 4:  loss 1.510837, train_accuracy 0.525, valid accuracy 0.4476
Epoch 26, CIFAR-10 Batch 5:  loss 1.539917, train_accuracy 0.475, valid accuracy 0.4172
Epoch 27, CIFAR-10 Batch 1:  loss 1.842671, train_accuracy 0.35, valid accuracy 0.4212
Epoch 27, CIFAR-10 Batch 2:  loss 1.523134, train_accuracy 0.4, valid accuracy 0.4284
Epoch 27, CIFAR-10 Batch 3:  loss 1.448401, train_accuracy 0.45, valid accuracy 0.448
Epoch 27, CIFAR-10 Batch 4:  loss 1.523332, train_accuracy 0.475, valid accuracy 0.4366
Epoch 27, CIFAR-10 Batch 5:  loss 1.567614, train_accuracy 0.475, valid accuracy 0.4106
Epoch 28, CIFAR-10 Batch 1:  loss 1.599994, train_accuracy 0.4, valid accuracy 0.4684
Epoch 28, CIFAR-10 Batch 2:  loss 1.468679, train_accuracy 0.525, valid accuracy 0.43
Epoch 28, CIFAR-10 Batch 3:  loss 1.463580, train_accuracy 0.5, valid accuracy 0.4222
Epoch 28, CIFAR-10 Batch 4:  loss 1.471771, train_accuracy 0.525, valid accuracy 0.4458
Epoch 28, CIFAR-10 Batch 5:  loss 1.476007, train_accuracy 0.475, valid accuracy 0.4314
Epoch 29, CIFAR-10 Batch 1:  loss 1.498986, train_accuracy 0.475, valid accuracy 0.4804
Epoch 29, CIFAR-10 Batch 2:  loss 1.507118, train_accuracy 0.475, valid accuracy 0.431
Epoch 29, CIFAR-10 Batch 3:  loss 1.313117, train_accuracy 0.575, valid accuracy 0.4928
Epoch 29, CIFAR-10 Batch 4:  loss 1.472046, train_accuracy 0.55, valid accuracy 0.4566
Epoch 29, CIFAR-10 Batch 5:  loss 1.400502, train_accuracy 0.575, valid accuracy 0.4428
Epoch 30, CIFAR-10 Batch 1:  loss 1.525667, train_accuracy 0.4, valid accuracy 0.4634
Epoch 30, CIFAR-10 Batch 2:  loss 1.415250, train_accuracy 0.475, valid accuracy 0.4276
Epoch 30, CIFAR-10 Batch 3:  loss 1.439679, train_accuracy 0.45, valid accuracy 0.4472
Epoch 30, CIFAR-10 Batch 4:  loss 1.322776, train_accuracy 0.55, valid accuracy 0.4994
Epoch 30, CIFAR-10 Batch 5:  loss 1.370803, train_accuracy 0.475, valid accuracy 0.4542
Epoch 31, CIFAR-10 Batch 1:  loss 1.424599, train_accuracy 0.425, valid accuracy 0.4906
Epoch 31, CIFAR-10 Batch 2:  loss 1.349580, train_accuracy 0.425, valid accuracy 0.4816
Epoch 31, CIFAR-10 Batch 3:  loss 1.334344, train_accuracy 0.45, valid accuracy 0.4682
Epoch 31, CIFAR-10 Batch 4:  loss 1.320359, train_accuracy 0.6, valid accuracy 0.4802
Epoch 31, CIFAR-10 Batch 5:  loss 1.289434, train_accuracy 0.525, valid accuracy 0.4752
Epoch 32, CIFAR-10 Batch 1:  loss 1.372583, train_accuracy 0.475, valid accuracy 0.4862
Epoch 32, CIFAR-10 Batch 2:  loss 1.364829, train_accuracy 0.525, valid accuracy 0.4488
Epoch 32, CIFAR-10 Batch 3:  loss 1.377442, train_accuracy 0.475, valid accuracy 0.476
Epoch 32, CIFAR-10 Batch 4:  loss 1.523991, train_accuracy 0.45, valid accuracy 0.4472
Epoch 32, CIFAR-10 Batch 5:  loss 1.479313, train_accuracy 0.525, valid accuracy 0.4532
Epoch 33, CIFAR-10 Batch 1:  loss 1.601253, train_accuracy 0.45, valid accuracy 0.4756
Epoch 33, CIFAR-10 Batch 2:  loss 1.308216, train_accuracy 0.5, valid accuracy 0.5068
Epoch 33, CIFAR-10 Batch 3:  loss 1.262900, train_accuracy 0.525, valid accuracy 0.5202
Epoch 33, CIFAR-10 Batch 4:  loss 1.269783, train_accuracy 0.575, valid accuracy 0.5102
Epoch 33, CIFAR-10 Batch 5:  loss 1.194381, train_accuracy 0.625, valid accuracy 0.5244
Epoch 34, CIFAR-10 Batch 1:  loss 1.342451, train_accuracy 0.425, valid accuracy 0.5058
Epoch 34, CIFAR-10 Batch 2:  loss 1.296099, train_accuracy 0.475, valid accuracy 0.4968
Epoch 34, CIFAR-10 Batch 3:  loss 1.344078, train_accuracy 0.45, valid accuracy 0.486
Epoch 34, CIFAR-10 Batch 4:  loss 1.393783, train_accuracy 0.55, valid accuracy 0.5002
Epoch 34, CIFAR-10 Batch 5:  loss 1.247730, train_accuracy 0.575, valid accuracy 0.5072
Epoch 35, CIFAR-10 Batch 1:  loss 1.181515, train_accuracy 0.575, valid accuracy 0.5528
Epoch 35, CIFAR-10 Batch 2:  loss 1.297121, train_accuracy 0.525, valid accuracy 0.4924
Epoch 35, CIFAR-10 Batch 3:  loss 1.130408, train_accuracy 0.625, valid accuracy 0.5438
Epoch 35, CIFAR-10 Batch 4:  loss 1.220897, train_accuracy 0.6, valid accuracy 0.529
Epoch 35, CIFAR-10 Batch 5:  loss 1.360435, train_accuracy 0.5, valid accuracy 0.458
Epoch 36, CIFAR-10 Batch 1:  loss 1.278018, train_accuracy 0.475, valid accuracy 0.5312
Epoch 36, CIFAR-10 Batch 2:  loss 1.228923, train_accuracy 0.5, valid accuracy 0.518
Epoch 36, CIFAR-10 Batch 3:  loss 1.189265, train_accuracy 0.55, valid accuracy 0.5134
Epoch 36, CIFAR-10 Batch 4:  loss 1.131226, train_accuracy 0.55, valid accuracy 0.5498
Epoch 36, CIFAR-10 Batch 5:  loss 1.147809, train_accuracy 0.6, valid accuracy 0.514
Epoch 37, CIFAR-10 Batch 1:  loss 1.310145, train_accuracy 0.5, valid accuracy 0.5326
Epoch 37, CIFAR-10 Batch 2:  loss 1.228172, train_accuracy 0.5, valid accuracy 0.5278
Epoch 37, CIFAR-10 Batch 3:  loss 1.275494, train_accuracy 0.475, valid accuracy 0.503
Epoch 37, CIFAR-10 Batch 4:  loss 1.109810, train_accuracy 0.65, valid accuracy 0.5366
Epoch 37, CIFAR-10 Batch 5:  loss 1.202451, train_accuracy 0.625, valid accuracy 0.5168
Epoch 38, CIFAR-10 Batch 1:  loss 1.194128, train_accuracy 0.55, valid accuracy 0.554
Epoch 38, CIFAR-10 Batch 2:  loss 1.182513, train_accuracy 0.525, valid accuracy 0.538
Epoch 38, CIFAR-10 Batch 3:  loss 1.091033, train_accuracy 0.6, valid accuracy 0.554
Epoch 38, CIFAR-10 Batch 4:  loss 1.152747, train_accuracy 0.575, valid accuracy 0.5408
Epoch 38, CIFAR-10 Batch 5:  loss 1.076687, train_accuracy 0.6, valid accuracy 0.5486
Epoch 39, CIFAR-10 Batch 1:  loss 1.143540, train_accuracy 0.6, valid accuracy 0.5498
Epoch 39, CIFAR-10 Batch 2:  loss 1.178898, train_accuracy 0.525, valid accuracy 0.557
Epoch 39, CIFAR-10 Batch 3:  loss 1.217030, train_accuracy 0.5, valid accuracy 0.5298
Epoch 39, CIFAR-10 Batch 4:  loss 0.992664, train_accuracy 0.7, valid accuracy 0.5716
Epoch 39, CIFAR-10 Batch 5:  loss 1.162126, train_accuracy 0.6, valid accuracy 0.5086
Epoch 40, CIFAR-10 Batch 1:  loss 1.158829, train_accuracy 0.525, valid accuracy 0.5612
Epoch 40, CIFAR-10 Batch 2:  loss 1.247941, train_accuracy 0.5, valid accuracy 0.5496
Epoch 40, CIFAR-10 Batch 3:  loss 1.158469, train_accuracy 0.5, valid accuracy 0.5356
Epoch 40, CIFAR-10 Batch 4:  loss 0.947233, train_accuracy 0.7, valid accuracy 0.5898
Epoch 40, CIFAR-10 Batch 5:  loss 1.048439, train_accuracy 0.65, valid accuracy 0.544
Epoch 41, CIFAR-10 Batch 1:  loss 1.125670, train_accuracy 0.55, valid accuracy 0.5666
Epoch 41, CIFAR-10 Batch 2:  loss 1.152957, train_accuracy 0.525, valid accuracy 0.5598
Epoch 41, CIFAR-10 Batch 3:  loss 1.080931, train_accuracy 0.55, valid accuracy 0.5768
Epoch 41, CIFAR-10 Batch 4:  loss 1.115965, train_accuracy 0.575, valid accuracy 0.5396
Epoch 41, CIFAR-10 Batch 5:  loss 1.208271, train_accuracy 0.625, valid accuracy 0.5062
Epoch 42, CIFAR-10 Batch 1:  loss 1.086362, train_accuracy 0.6, valid accuracy 0.5828
Epoch 42, CIFAR-10 Batch 2:  loss 1.151170, train_accuracy 0.55, valid accuracy 0.57
Epoch 42, CIFAR-10 Batch 3:  loss 1.032405, train_accuracy 0.525, valid accuracy 0.5818
Epoch 42, CIFAR-10 Batch 4:  loss 1.030534, train_accuracy 0.7, valid accuracy 0.5698
Epoch 42, CIFAR-10 Batch 5:  loss 1.033735, train_accuracy 0.725, valid accuracy 0.5686
Epoch 43, CIFAR-10 Batch 1:  loss 1.121751, train_accuracy 0.575, valid accuracy 0.5722
Epoch 43, CIFAR-10 Batch 2:  loss 1.143200, train_accuracy 0.625, valid accuracy 0.5702
Epoch 43, CIFAR-10 Batch 3:  loss 0.953962, train_accuracy 0.675, valid accuracy 0.585
Epoch 43, CIFAR-10 Batch 4:  loss 0.982509, train_accuracy 0.675, valid accuracy 0.5744
Epoch 43, CIFAR-10 Batch 5:  loss 0.970620, train_accuracy 0.675, valid accuracy 0.559
Epoch 44, CIFAR-10 Batch 1:  loss 1.035159, train_accuracy 0.575, valid accuracy 0.5916
Epoch 44, CIFAR-10 Batch 2:  loss 1.162024, train_accuracy 0.575, valid accuracy 0.5742
Epoch 44, CIFAR-10 Batch 3:  loss 0.970094, train_accuracy 0.575, valid accuracy 0.591
Epoch 44, CIFAR-10 Batch 4:  loss 0.869121, train_accuracy 0.725, valid accuracy 0.5992
Epoch 44, CIFAR-10 Batch 5:  loss 0.927748, train_accuracy 0.775, valid accuracy 0.596
Epoch 45, CIFAR-10 Batch 1:  loss 0.953810, train_accuracy 0.625, valid accuracy 0.6096
Epoch 45, CIFAR-10 Batch 2:  loss 1.177540, train_accuracy 0.55, valid accuracy 0.5814
Epoch 45, CIFAR-10 Batch 3:  loss 1.136366, train_accuracy 0.575, valid accuracy 0.5378
Epoch 45, CIFAR-10 Batch 4:  loss 0.912520, train_accuracy 0.725, valid accuracy 0.5984
Epoch 45, CIFAR-10 Batch 5:  loss 0.933036, train_accuracy 0.775, valid accuracy 0.5786
Epoch 46, CIFAR-10 Batch 1:  loss 0.915990, train_accuracy 0.675, valid accuracy 0.6054
Epoch 46, CIFAR-10 Batch 2:  loss 1.092247, train_accuracy 0.625, valid accuracy 0.569
Epoch 46, CIFAR-10 Batch 3:  loss 1.031745, train_accuracy 0.55, valid accuracy 0.572
Epoch 46, CIFAR-10 Batch 4:  loss 0.880066, train_accuracy 0.775, valid accuracy 0.5838
Epoch 46, CIFAR-10 Batch 5:  loss 0.965285, train_accuracy 0.725, valid accuracy 0.59
Epoch 47, CIFAR-10 Batch 1:  loss 0.946281, train_accuracy 0.675, valid accuracy 0.6122
Epoch 47, CIFAR-10 Batch 2:  loss 0.971507, train_accuracy 0.6, valid accuracy 0.591
Epoch 47, CIFAR-10 Batch 3:  loss 0.836289, train_accuracy 0.65, valid accuracy 0.6318
Epoch 47, CIFAR-10 Batch 4:  loss 0.934505, train_accuracy 0.75, valid accuracy 0.5664
Epoch 47, CIFAR-10 Batch 5:  loss 0.891864, train_accuracy 0.7, valid accuracy 0.5884
Epoch 48, CIFAR-10 Batch 1:  loss 0.839569, train_accuracy 0.675, valid accuracy 0.6278
Epoch 48, CIFAR-10 Batch 2:  loss 1.060783, train_accuracy 0.625, valid accuracy 0.5846
Epoch 48, CIFAR-10 Batch 3:  loss 0.902010, train_accuracy 0.6, valid accuracy 0.6098
Epoch 48, CIFAR-10 Batch 4:  loss 0.791108, train_accuracy 0.775, valid accuracy 0.6268
Epoch 48, CIFAR-10 Batch 5:  loss 0.847665, train_accuracy 0.725, valid accuracy 0.6012
Epoch 49, CIFAR-10 Batch 1:  loss 0.934348, train_accuracy 0.675, valid accuracy 0.617
Epoch 49, CIFAR-10 Batch 2:  loss 0.980735, train_accuracy 0.7, valid accuracy 0.6162
Epoch 49, CIFAR-10 Batch 3:  loss 0.877590, train_accuracy 0.6, valid accuracy 0.6106
Epoch 49, CIFAR-10 Batch 4:  loss 0.662328, train_accuracy 0.875, valid accuracy 0.6424
Epoch 49, CIFAR-10 Batch 5:  loss 0.967010, train_accuracy 0.675, valid accuracy 0.578
Epoch 50, CIFAR-10 Batch 1:  loss 0.821405, train_accuracy 0.675, valid accuracy 0.6466
Epoch 50, CIFAR-10 Batch 2:  loss 0.898624, train_accuracy 0.625, valid accuracy 0.6192
Epoch 50, CIFAR-10 Batch 3:  loss 0.806297, train_accuracy 0.75, valid accuracy 0.6234
Epoch 50, CIFAR-10 Batch 4:  loss 0.693006, train_accuracy 0.825, valid accuracy 0.6578
Epoch 50, CIFAR-10 Batch 5:  loss 0.813661, train_accuracy 0.75, valid accuracy 0.6212
Epoch 51, CIFAR-10 Batch 1:  loss 0.819792, train_accuracy 0.725, valid accuracy 0.6482
Epoch 51, CIFAR-10 Batch 2:  loss 0.861473, train_accuracy 0.725, valid accuracy 0.624
Epoch 51, CIFAR-10 Batch 3:  loss 0.784527, train_accuracy 0.8, valid accuracy 0.616
Epoch 51, CIFAR-10 Batch 4:  loss 0.748757, train_accuracy 0.8, valid accuracy 0.6264
Epoch 51, CIFAR-10 Batch 5:  loss 0.903420, train_accuracy 0.7, valid accuracy 0.5836
Epoch 52, CIFAR-10 Batch 1:  loss 0.724446, train_accuracy 0.725, valid accuracy 0.6496
Epoch 52, CIFAR-10 Batch 2:  loss 0.792611, train_accuracy 0.775, valid accuracy 0.644
Epoch 52, CIFAR-10 Batch 3:  loss 0.808915, train_accuracy 0.675, valid accuracy 0.6266
Epoch 52, CIFAR-10 Batch 4:  loss 0.768101, train_accuracy 0.725, valid accuracy 0.6182
Epoch 52, CIFAR-10 Batch 5:  loss 0.787360, train_accuracy 0.725, valid accuracy 0.5986
Epoch 53, CIFAR-10 Batch 1:  loss 0.744170, train_accuracy 0.675, valid accuracy 0.6502
Epoch 53, CIFAR-10 Batch 2:  loss 0.877120, train_accuracy 0.7, valid accuracy 0.6312
Epoch 53, CIFAR-10 Batch 3:  loss 0.828575, train_accuracy 0.7, valid accuracy 0.6014
Epoch 53, CIFAR-10 Batch 4:  loss 0.741724, train_accuracy 0.775, valid accuracy 0.6136
Epoch 53, CIFAR-10 Batch 5:  loss 0.791263, train_accuracy 0.8, valid accuracy 0.613
Epoch 54, CIFAR-10 Batch 1:  loss 0.758548, train_accuracy 0.725, valid accuracy 0.643
Epoch 54, CIFAR-10 Batch 2:  loss 0.880433, train_accuracy 0.75, valid accuracy 0.6108
Epoch 54, CIFAR-10 Batch 3:  loss 0.693740, train_accuracy 0.8, valid accuracy 0.6236
Epoch 54, CIFAR-10 Batch 4:  loss 0.709744, train_accuracy 0.775, valid accuracy 0.628
Epoch 54, CIFAR-10 Batch 5:  loss 0.695770, train_accuracy 0.8, valid accuracy 0.6346
Epoch 55, CIFAR-10 Batch 1:  loss 0.715957, train_accuracy 0.675, valid accuracy 0.6536
Epoch 55, CIFAR-10 Batch 2:  loss 0.626039, train_accuracy 0.875, valid accuracy 0.6656
Epoch 55, CIFAR-10 Batch 3:  loss 0.744257, train_accuracy 0.725, valid accuracy 0.6456
Epoch 55, CIFAR-10 Batch 4:  loss 0.602788, train_accuracy 0.875, valid accuracy 0.6564
Epoch 55, CIFAR-10 Batch 5:  loss 0.655302, train_accuracy 0.8, valid accuracy 0.6504
Epoch 56, CIFAR-10 Batch 1:  loss 0.722425, train_accuracy 0.775, valid accuracy 0.66
Epoch 56, CIFAR-10 Batch 2:  loss 0.737733, train_accuracy 0.8, valid accuracy 0.6622
Epoch 56, CIFAR-10 Batch 3:  loss 0.765361, train_accuracy 0.75, valid accuracy 0.6284
Epoch 56, CIFAR-10 Batch 4:  loss 0.597862, train_accuracy 0.85, valid accuracy 0.6516
Epoch 56, CIFAR-10 Batch 5:  loss 0.712303, train_accuracy 0.8, valid accuracy 0.632
Epoch 57, CIFAR-10 Batch 1:  loss 0.711807, train_accuracy 0.7, valid accuracy 0.6652
Epoch 57, CIFAR-10 Batch 2:  loss 0.769213, train_accuracy 0.825, valid accuracy 0.6074
Epoch 57, CIFAR-10 Batch 3:  loss 0.658236, train_accuracy 0.775, valid accuracy 0.6466
Epoch 57, CIFAR-10 Batch 4:  loss 0.629447, train_accuracy 0.875, valid accuracy 0.6324
Epoch 57, CIFAR-10 Batch 5:  loss 0.616860, train_accuracy 0.8, valid accuracy 0.6442
Epoch 58, CIFAR-10 Batch 1:  loss 0.604290, train_accuracy 0.75, valid accuracy 0.6762
Epoch 58, CIFAR-10 Batch 2:  loss 0.641316, train_accuracy 0.825, valid accuracy 0.658
Epoch 58, CIFAR-10 Batch 3:  loss 0.684759, train_accuracy 0.75, valid accuracy 0.634
Epoch 58, CIFAR-10 Batch 4:  loss 0.631951, train_accuracy 0.825, valid accuracy 0.6372
Epoch 58, CIFAR-10 Batch 5:  loss 0.591599, train_accuracy 0.875, valid accuracy 0.6626
Epoch 59, CIFAR-10 Batch 1:  loss 0.645560, train_accuracy 0.75, valid accuracy 0.675
Epoch 59, CIFAR-10 Batch 2:  loss 0.634261, train_accuracy 0.825, valid accuracy 0.6468
Epoch 59, CIFAR-10 Batch 3:  loss 0.664659, train_accuracy 0.8, valid accuracy 0.6446
Epoch 59, CIFAR-10 Batch 4:  loss 0.537345, train_accuracy 0.85, valid accuracy 0.6682
Epoch 59, CIFAR-10 Batch 5:  loss 0.646416, train_accuracy 0.825, valid accuracy 0.6552
Epoch 60, CIFAR-10 Batch 1:  loss 0.597867, train_accuracy 0.825, valid accuracy 0.6788
Epoch 60, CIFAR-10 Batch 2:  loss 0.653040, train_accuracy 0.775, valid accuracy 0.6522
Epoch 60, CIFAR-10 Batch 3:  loss 0.621518, train_accuracy 0.9, valid accuracy 0.6546
Epoch 60, CIFAR-10 Batch 4:  loss 0.522278, train_accuracy 0.85, valid accuracy 0.6656
Epoch 60, CIFAR-10 Batch 5:  loss 0.574840, train_accuracy 0.875, valid accuracy 0.6578
Epoch 61, CIFAR-10 Batch 1:  loss 0.684382, train_accuracy 0.75, valid accuracy 0.6618
Epoch 61, CIFAR-10 Batch 2:  loss 0.575198, train_accuracy 0.875, valid accuracy 0.6832
Epoch 61, CIFAR-10 Batch 3:  loss 0.520367, train_accuracy 0.875, valid accuracy 0.6842
Epoch 61, CIFAR-10 Batch 4:  loss 0.483688, train_accuracy 0.925, valid accuracy 0.6848
Epoch 61, CIFAR-10 Batch 5:  loss 0.596578, train_accuracy 0.85, valid accuracy 0.6586
Epoch 62, CIFAR-10 Batch 1:  loss 0.553290, train_accuracy 0.8, valid accuracy 0.698
Epoch 62, CIFAR-10 Batch 2:  loss 0.519195, train_accuracy 0.9, valid accuracy 0.6812
Epoch 62, CIFAR-10 Batch 3:  loss 0.510083, train_accuracy 0.875, valid accuracy 0.6736
Epoch 62, CIFAR-10 Batch 4:  loss 0.492281, train_accuracy 0.875, valid accuracy 0.6812
Epoch 62, CIFAR-10 Batch 5:  loss 0.578490, train_accuracy 0.825, valid accuracy 0.6664
Epoch 63, CIFAR-10 Batch 1:  loss 0.556583, train_accuracy 0.8, valid accuracy 0.6938
Epoch 63, CIFAR-10 Batch 2:  loss 0.530466, train_accuracy 0.85, valid accuracy 0.6786
Epoch 63, CIFAR-10 Batch 3:  loss 0.505612, train_accuracy 0.9, valid accuracy 0.678
Epoch 63, CIFAR-10 Batch 4:  loss 0.522833, train_accuracy 0.875, valid accuracy 0.6536
Epoch 63, CIFAR-10 Batch 5:  loss 0.545556, train_accuracy 0.85, valid accuracy 0.6624
Epoch 64, CIFAR-10 Batch 1:  loss 0.556303, train_accuracy 0.775, valid accuracy 0.699
Epoch 64, CIFAR-10 Batch 2:  loss 0.493317, train_accuracy 0.9, valid accuracy 0.6862
Epoch 64, CIFAR-10 Batch 3:  loss 0.573200, train_accuracy 0.9, valid accuracy 0.686
Epoch 64, CIFAR-10 Batch 4:  loss 0.482270, train_accuracy 0.825, valid accuracy 0.6734
Epoch 64, CIFAR-10 Batch 5:  loss 0.529360, train_accuracy 0.875, valid accuracy 0.665
Epoch 65, CIFAR-10 Batch 1:  loss 0.503995, train_accuracy 0.875, valid accuracy 0.708
Epoch 65, CIFAR-10 Batch 2:  loss 0.555663, train_accuracy 0.85, valid accuracy 0.6796
Epoch 65, CIFAR-10 Batch 3:  loss 0.535766, train_accuracy 0.9, valid accuracy 0.6862
Epoch 65, CIFAR-10 Batch 4:  loss 0.442237, train_accuracy 0.9, valid accuracy 0.6778
Epoch 65, CIFAR-10 Batch 5:  loss 0.581082, train_accuracy 0.85, valid accuracy 0.672
Epoch 66, CIFAR-10 Batch 1:  loss 0.485018, train_accuracy 0.875, valid accuracy 0.7076
Epoch 66, CIFAR-10 Batch 2:  loss 0.490134, train_accuracy 0.9, valid accuracy 0.6894
Epoch 66, CIFAR-10 Batch 3:  loss 0.552040, train_accuracy 0.875, valid accuracy 0.666
Epoch 66, CIFAR-10 Batch 4:  loss 0.403948, train_accuracy 0.95, valid accuracy 0.6828
Epoch 66, CIFAR-10 Batch 5:  loss 0.494317, train_accuracy 0.875, valid accuracy 0.678
Epoch 67, CIFAR-10 Batch 1:  loss 0.464382, train_accuracy 0.875, valid accuracy 0.7038
Epoch 67, CIFAR-10 Batch 2:  loss 0.505741, train_accuracy 0.85, valid accuracy 0.6744
Epoch 67, CIFAR-10 Batch 3:  loss 0.512518, train_accuracy 0.9, valid accuracy 0.6776
Epoch 67, CIFAR-10 Batch 4:  loss 0.430421, train_accuracy 0.925, valid accuracy 0.6924
Epoch 67, CIFAR-10 Batch 5:  loss 0.488371, train_accuracy 0.875, valid accuracy 0.6882
Epoch 68, CIFAR-10 Batch 1:  loss 0.453979, train_accuracy 0.9, valid accuracy 0.7086
Epoch 68, CIFAR-10 Batch 2:  loss 0.437693, train_accuracy 0.9, valid accuracy 0.6994
Epoch 68, CIFAR-10 Batch 3:  loss 0.457800, train_accuracy 0.925, valid accuracy 0.7114
Epoch 68, CIFAR-10 Batch 4:  loss 0.444153, train_accuracy 0.9, valid accuracy 0.6948
Epoch 68, CIFAR-10 Batch 5:  loss 0.445373, train_accuracy 0.95, valid accuracy 0.7052
Epoch 69, CIFAR-10 Batch 1:  loss 0.528806, train_accuracy 0.85, valid accuracy 0.7126
Epoch 69, CIFAR-10 Batch 2:  loss 0.404905, train_accuracy 0.95, valid accuracy 0.695
Epoch 69, CIFAR-10 Batch 3:  loss 0.459547, train_accuracy 0.95, valid accuracy 0.7044
Epoch 69, CIFAR-10 Batch 4:  loss 0.482656, train_accuracy 0.875, valid accuracy 0.692
Epoch 69, CIFAR-10 Batch 5:  loss 0.396591, train_accuracy 0.925, valid accuracy 0.6974
Epoch 70, CIFAR-10 Batch 1:  loss 0.485558, train_accuracy 0.9, valid accuracy 0.7062
Epoch 70, CIFAR-10 Batch 2:  loss 0.418883, train_accuracy 0.975, valid accuracy 0.702
Epoch 70, CIFAR-10 Batch 3:  loss 0.463238, train_accuracy 0.875, valid accuracy 0.6684
Epoch 70, CIFAR-10 Batch 4:  loss 0.365099, train_accuracy 0.95, valid accuracy 0.7042
Epoch 70, CIFAR-10 Batch 5:  loss 0.431817, train_accuracy 0.85, valid accuracy 0.687
Epoch 71, CIFAR-10 Batch 1:  loss 0.445296, train_accuracy 0.9, valid accuracy 0.7066
Epoch 71, CIFAR-10 Batch 2:  loss 0.451146, train_accuracy 0.9, valid accuracy 0.6858
Epoch 71, CIFAR-10 Batch 3:  loss 0.433473, train_accuracy 0.9, valid accuracy 0.7064
Epoch 71, CIFAR-10 Batch 4:  loss 0.384506, train_accuracy 0.9, valid accuracy 0.6918
Epoch 71, CIFAR-10 Batch 5:  loss 0.387208, train_accuracy 0.875, valid accuracy 0.709
Epoch 72, CIFAR-10 Batch 1:  loss 0.593620, train_accuracy 0.85, valid accuracy 0.692
Epoch 72, CIFAR-10 Batch 2:  loss 0.419875, train_accuracy 0.9, valid accuracy 0.696
Epoch 72, CIFAR-10 Batch 3:  loss 0.428347, train_accuracy 0.975, valid accuracy 0.6886
Epoch 72, CIFAR-10 Batch 4:  loss 0.349258, train_accuracy 0.95, valid accuracy 0.7098
Epoch 72, CIFAR-10 Batch 5:  loss 0.361295, train_accuracy 0.925, valid accuracy 0.6992
Epoch 73, CIFAR-10 Batch 1:  loss 0.444569, train_accuracy 0.95, valid accuracy 0.7226
Epoch 73, CIFAR-10 Batch 2:  loss 0.348188, train_accuracy 0.975, valid accuracy 0.7072
Epoch 73, CIFAR-10 Batch 3:  loss 0.424960, train_accuracy 0.95, valid accuracy 0.6796
Epoch 73, CIFAR-10 Batch 4:  loss 0.409780, train_accuracy 0.9, valid accuracy 0.6938
Epoch 73, CIFAR-10 Batch 5:  loss 0.452432, train_accuracy 0.875, valid accuracy 0.684
Epoch 74, CIFAR-10 Batch 1:  loss 0.396286, train_accuracy 0.95, valid accuracy 0.7256
Epoch 74, CIFAR-10 Batch 2:  loss 0.408701, train_accuracy 0.95, valid accuracy 0.7036
Epoch 74, CIFAR-10 Batch 3:  loss 0.415674, train_accuracy 0.95, valid accuracy 0.698
Epoch 74, CIFAR-10 Batch 4:  loss 0.328526, train_accuracy 0.9, valid accuracy 0.6906
Epoch 74, CIFAR-10 Batch 5:  loss 0.373147, train_accuracy 0.9, valid accuracy 0.7108
Epoch 75, CIFAR-10 Batch 1:  loss 0.381045, train_accuracy 0.975, valid accuracy 0.735
Epoch 75, CIFAR-10 Batch 2:  loss 0.378748, train_accuracy 0.925, valid accuracy 0.6946
Epoch 75, CIFAR-10 Batch 3:  loss 0.415470, train_accuracy 0.95, valid accuracy 0.6948
Epoch 75, CIFAR-10 Batch 4:  loss 0.297050, train_accuracy 0.95, valid accuracy 0.721
Epoch 75, CIFAR-10 Batch 5:  loss 0.405161, train_accuracy 0.95, valid accuracy 0.697
Epoch 76, CIFAR-10 Batch 1:  loss 0.437699, train_accuracy 0.925, valid accuracy 0.7218
Epoch 76, CIFAR-10 Batch 2:  loss 0.333497, train_accuracy 0.95, valid accuracy 0.715
Epoch 76, CIFAR-10 Batch 3:  loss 0.401009, train_accuracy 0.975, valid accuracy 0.7006
Epoch 76, CIFAR-10 Batch 4:  loss 0.326549, train_accuracy 0.925, valid accuracy 0.7086
Epoch 76, CIFAR-10 Batch 5:  loss 0.325145, train_accuracy 0.975, valid accuracy 0.7216
Epoch 77, CIFAR-10 Batch 1:  loss 0.438713, train_accuracy 0.9, valid accuracy 0.7098
Epoch 77, CIFAR-10 Batch 2:  loss 0.353128, train_accuracy 0.95, valid accuracy 0.7086
Epoch 77, CIFAR-10 Batch 3:  loss 0.343943, train_accuracy 0.95, valid accuracy 0.7032
Epoch 77, CIFAR-10 Batch 4:  loss 0.349678, train_accuracy 0.925, valid accuracy 0.7022
Epoch 77, CIFAR-10 Batch 5:  loss 0.268707, train_accuracy 1, valid accuracy 0.7222
Epoch 78, CIFAR-10 Batch 1:  loss 0.389804, train_accuracy 0.9, valid accuracy 0.7198
Epoch 78, CIFAR-10 Batch 2:  loss 0.315393, train_accuracy 0.975, valid accuracy 0.7166
Epoch 78, CIFAR-10 Batch 3:  loss 0.331260, train_accuracy 0.95, valid accuracy 0.7106
Epoch 78, CIFAR-10 Batch 4:  loss 0.326001, train_accuracy 0.925, valid accuracy 0.7036
Epoch 78, CIFAR-10 Batch 5:  loss 0.273141, train_accuracy 0.975, valid accuracy 0.7074
Epoch 79, CIFAR-10 Batch 1:  loss 0.388199, train_accuracy 0.925, valid accuracy 0.7186
Epoch 79, CIFAR-10 Batch 2:  loss 0.319482, train_accuracy 0.975, valid accuracy 0.7142
Epoch 79, CIFAR-10 Batch 3:  loss 0.301571, train_accuracy 0.95, valid accuracy 0.7274
Epoch 79, CIFAR-10 Batch 4:  loss 0.266620, train_accuracy 0.975, valid accuracy 0.7274
Epoch 79, CIFAR-10 Batch 5:  loss 0.303517, train_accuracy 0.95, valid accuracy 0.711
Epoch 80, CIFAR-10 Batch 1:  loss 0.343106, train_accuracy 0.975, valid accuracy 0.743
Epoch 80, CIFAR-10 Batch 2:  loss 0.317904, train_accuracy 0.975, valid accuracy 0.7138
Epoch 80, CIFAR-10 Batch 3:  loss 0.286390, train_accuracy 0.95, valid accuracy 0.7064
Epoch 80, CIFAR-10 Batch 4:  loss 0.257363, train_accuracy 0.95, valid accuracy 0.7252
Epoch 80, CIFAR-10 Batch 5:  loss 0.288184, train_accuracy 0.975, valid accuracy 0.7114
Epoch 81, CIFAR-10 Batch 1:  loss 0.362915, train_accuracy 0.95, valid accuracy 0.735
Epoch 81, CIFAR-10 Batch 2:  loss 0.312612, train_accuracy 0.975, valid accuracy 0.7216
Epoch 81, CIFAR-10 Batch 3:  loss 0.361740, train_accuracy 0.95, valid accuracy 0.6998
Epoch 81, CIFAR-10 Batch 4:  loss 0.270288, train_accuracy 0.95, valid accuracy 0.7238
Epoch 81, CIFAR-10 Batch 5:  loss 0.242587, train_accuracy 1, valid accuracy 0.7298
Epoch 82, CIFAR-10 Batch 1:  loss 0.304943, train_accuracy 0.95, valid accuracy 0.7492
Epoch 82, CIFAR-10 Batch 2:  loss 0.290070, train_accuracy 0.975, valid accuracy 0.729
Epoch 82, CIFAR-10 Batch 3:  loss 0.307349, train_accuracy 0.975, valid accuracy 0.7162
Epoch 82, CIFAR-10 Batch 4:  loss 0.263745, train_accuracy 0.95, valid accuracy 0.7422
Epoch 82, CIFAR-10 Batch 5:  loss 0.314377, train_accuracy 0.95, valid accuracy 0.7138
Epoch 83, CIFAR-10 Batch 1:  loss 0.355767, train_accuracy 0.95, valid accuracy 0.713
Epoch 83, CIFAR-10 Batch 2:  loss 0.282227, train_accuracy 0.975, valid accuracy 0.723
Epoch 83, CIFAR-10 Batch 3:  loss 0.256077, train_accuracy 0.95, valid accuracy 0.7204
Epoch 83, CIFAR-10 Batch 4:  loss 0.244001, train_accuracy 0.975, valid accuracy 0.7402
Epoch 83, CIFAR-10 Batch 5:  loss 0.262094, train_accuracy 0.95, valid accuracy 0.7286
Epoch 84, CIFAR-10 Batch 1:  loss 0.313435, train_accuracy 0.95, valid accuracy 0.7454
Epoch 84, CIFAR-10 Batch 2:  loss 0.349619, train_accuracy 0.925, valid accuracy 0.7084
Epoch 84, CIFAR-10 Batch 3:  loss 0.262030, train_accuracy 0.975, valid accuracy 0.7196
Epoch 84, CIFAR-10 Batch 4:  loss 0.262100, train_accuracy 0.925, valid accuracy 0.728
Epoch 84, CIFAR-10 Batch 5:  loss 0.287046, train_accuracy 0.975, valid accuracy 0.7152
Epoch 85, CIFAR-10 Batch 1:  loss 0.299888, train_accuracy 0.95, valid accuracy 0.7456
Epoch 85, CIFAR-10 Batch 2:  loss 0.287766, train_accuracy 0.975, valid accuracy 0.7308
Epoch 85, CIFAR-10 Batch 3:  loss 0.275444, train_accuracy 1, valid accuracy 0.716
Epoch 85, CIFAR-10 Batch 4:  loss 0.270724, train_accuracy 0.925, valid accuracy 0.7246
Epoch 85, CIFAR-10 Batch 5:  loss 0.221934, train_accuracy 1, valid accuracy 0.7346
Epoch 86, CIFAR-10 Batch 1:  loss 0.324443, train_accuracy 0.9, valid accuracy 0.7392
Epoch 86, CIFAR-10 Batch 2:  loss 0.241802, train_accuracy 0.975, valid accuracy 0.7328
Epoch 86, CIFAR-10 Batch 3:  loss 0.390288, train_accuracy 0.925, valid accuracy 0.6768
Epoch 86, CIFAR-10 Batch 4:  loss 0.220279, train_accuracy 0.95, valid accuracy 0.7382
Epoch 86, CIFAR-10 Batch 5:  loss 0.210745, train_accuracy 1, valid accuracy 0.738
Epoch 87, CIFAR-10 Batch 1:  loss 0.297152, train_accuracy 0.95, valid accuracy 0.7468
Epoch 87, CIFAR-10 Batch 2:  loss 0.262388, train_accuracy 0.975, valid accuracy 0.7188
Epoch 87, CIFAR-10 Batch 3:  loss 0.266239, train_accuracy 0.975, valid accuracy 0.7214
Epoch 87, CIFAR-10 Batch 4:  loss 0.206904, train_accuracy 0.975, valid accuracy 0.7314
Epoch 87, CIFAR-10 Batch 5:  loss 0.247299, train_accuracy 1, valid accuracy 0.7114
Epoch 88, CIFAR-10 Batch 1:  loss 0.310059, train_accuracy 0.95, valid accuracy 0.7484
Epoch 88, CIFAR-10 Batch 2:  loss 0.285476, train_accuracy 0.95, valid accuracy 0.7214
Epoch 88, CIFAR-10 Batch 3:  loss 0.244184, train_accuracy 0.975, valid accuracy 0.7052
Epoch 88, CIFAR-10 Batch 4:  loss 0.205253, train_accuracy 0.975, valid accuracy 0.7384
Epoch 88, CIFAR-10 Batch 5:  loss 0.231289, train_accuracy 1, valid accuracy 0.7294
Epoch 89, CIFAR-10 Batch 1:  loss 0.280811, train_accuracy 0.925, valid accuracy 0.7526
Epoch 89, CIFAR-10 Batch 2:  loss 0.268638, train_accuracy 0.95, valid accuracy 0.7308
Epoch 89, CIFAR-10 Batch 3:  loss 0.252547, train_accuracy 0.95, valid accuracy 0.7326
Epoch 89, CIFAR-10 Batch 4:  loss 0.218241, train_accuracy 0.95, valid accuracy 0.7206
Epoch 89, CIFAR-10 Batch 5:  loss 0.330778, train_accuracy 0.975, valid accuracy 0.6786
Epoch 90, CIFAR-10 Batch 1:  loss 0.283055, train_accuracy 0.975, valid accuracy 0.7446
Epoch 90, CIFAR-10 Batch 2:  loss 0.294140, train_accuracy 0.975, valid accuracy 0.7316
Epoch 90, CIFAR-10 Batch 3:  loss 0.209385, train_accuracy 0.975, valid accuracy 0.726
Epoch 90, CIFAR-10 Batch 4:  loss 0.221696, train_accuracy 0.95, valid accuracy 0.736
Epoch 90, CIFAR-10 Batch 5:  loss 0.251832, train_accuracy 1, valid accuracy 0.7144
Epoch 91, CIFAR-10 Batch 1:  loss 0.278032, train_accuracy 0.975, valid accuracy 0.746
Epoch 91, CIFAR-10 Batch 2:  loss 0.203122, train_accuracy 1, valid accuracy 0.7532
Epoch 91, CIFAR-10 Batch 3:  loss 0.260541, train_accuracy 0.975, valid accuracy 0.6934
Epoch 91, CIFAR-10 Batch 4:  loss 0.209139, train_accuracy 0.975, valid accuracy 0.736
Epoch 91, CIFAR-10 Batch 5:  loss 0.189152, train_accuracy 1, valid accuracy 0.7294
Epoch 92, CIFAR-10 Batch 1:  loss 0.235726, train_accuracy 0.975, valid accuracy 0.7628
Epoch 92, CIFAR-10 Batch 2:  loss 0.215720, train_accuracy 1, valid accuracy 0.7358
Epoch 92, CIFAR-10 Batch 3:  loss 0.251879, train_accuracy 0.975, valid accuracy 0.703
Epoch 92, CIFAR-10 Batch 4:  loss 0.168907, train_accuracy 0.95, valid accuracy 0.7574
Epoch 92, CIFAR-10 Batch 5:  loss 0.195817, train_accuracy 0.975, valid accuracy 0.7314
Epoch 93, CIFAR-10 Batch 1:  loss 0.272657, train_accuracy 0.975, valid accuracy 0.7556
Epoch 93, CIFAR-10 Batch 2:  loss 0.202095, train_accuracy 0.975, valid accuracy 0.7556
Epoch 93, CIFAR-10 Batch 3:  loss 0.220301, train_accuracy 1, valid accuracy 0.7368
Epoch 93, CIFAR-10 Batch 4:  loss 0.196345, train_accuracy 0.95, valid accuracy 0.744
Epoch 93, CIFAR-10 Batch 5:  loss 0.187791, train_accuracy 1, valid accuracy 0.7322
Epoch 94, CIFAR-10 Batch 1:  loss 0.235554, train_accuracy 0.95, valid accuracy 0.7582
Epoch 94, CIFAR-10 Batch 2:  loss 0.205702, train_accuracy 0.975, valid accuracy 0.7362
Epoch 94, CIFAR-10 Batch 3:  loss 0.211005, train_accuracy 0.975, valid accuracy 0.715
Epoch 94, CIFAR-10 Batch 4:  loss 0.209714, train_accuracy 0.95, valid accuracy 0.7498
Epoch 94, CIFAR-10 Batch 5:  loss 0.188027, train_accuracy 1, valid accuracy 0.7348
Epoch 95, CIFAR-10 Batch 1:  loss 0.291226, train_accuracy 0.925, valid accuracy 0.738
Epoch 95, CIFAR-10 Batch 2:  loss 0.193775, train_accuracy 1, valid accuracy 0.7364
Epoch 95, CIFAR-10 Batch 3:  loss 0.218080, train_accuracy 1, valid accuracy 0.7116
Epoch 95, CIFAR-10 Batch 4:  loss 0.180257, train_accuracy 0.95, valid accuracy 0.7406
Epoch 95, CIFAR-10 Batch 5:  loss 0.173394, train_accuracy 1, valid accuracy 0.7354
Epoch 96, CIFAR-10 Batch 1:  loss 0.246748, train_accuracy 0.975, valid accuracy 0.7618
Epoch 96, CIFAR-10 Batch 2:  loss 0.187309, train_accuracy 0.975, valid accuracy 0.7356
Epoch 96, CIFAR-10 Batch 3:  loss 0.164537, train_accuracy 1, valid accuracy 0.7498
Epoch 96, CIFAR-10 Batch 4:  loss 0.220719, train_accuracy 0.95, valid accuracy 0.733
Epoch 96, CIFAR-10 Batch 5:  loss 0.158058, train_accuracy 1, valid accuracy 0.7576
Epoch 97, CIFAR-10 Batch 1:  loss 0.234221, train_accuracy 0.975, valid accuracy 0.7538
Epoch 97, CIFAR-10 Batch 2:  loss 0.210890, train_accuracy 0.975, valid accuracy 0.7484
Epoch 97, CIFAR-10 Batch 3:  loss 0.171037, train_accuracy 1, valid accuracy 0.7376
Epoch 97, CIFAR-10 Batch 4:  loss 0.210297, train_accuracy 0.975, valid accuracy 0.7412
Epoch 97, CIFAR-10 Batch 5:  loss 0.174516, train_accuracy 1, valid accuracy 0.7374
Epoch 98, CIFAR-10 Batch 1:  loss 0.215604, train_accuracy 0.95, valid accuracy 0.7512
Epoch 98, CIFAR-10 Batch 2:  loss 0.207287, train_accuracy 1, valid accuracy 0.7496
Epoch 98, CIFAR-10 Batch 3:  loss 0.177129, train_accuracy 1, valid accuracy 0.7286
Epoch 98, CIFAR-10 Batch 4:  loss 0.167291, train_accuracy 0.95, valid accuracy 0.7534
Epoch 98, CIFAR-10 Batch 5:  loss 0.190549, train_accuracy 0.975, valid accuracy 0.7184
Epoch 99, CIFAR-10 Batch 1:  loss 0.221690, train_accuracy 0.975, valid accuracy 0.7456
Epoch 99, CIFAR-10 Batch 2:  loss 0.182064, train_accuracy 1, valid accuracy 0.7498
Epoch 99, CIFAR-10 Batch 3:  loss 0.179396, train_accuracy 0.975, valid accuracy 0.729
Epoch 99, CIFAR-10 Batch 4:  loss 0.207154, train_accuracy 0.95, valid accuracy 0.7496
Epoch 99, CIFAR-10 Batch 5:  loss 0.117632, train_accuracy 1, valid accuracy 0.7648
Epoch 100, CIFAR-10 Batch 1:  loss 0.188507, train_accuracy 1, valid accuracy 0.7592
Epoch 100, CIFAR-10 Batch 2:  loss 0.207158, train_accuracy 1, valid accuracy 0.7374
Epoch 100, CIFAR-10 Batch 3:  loss 0.176509, train_accuracy 0.975, valid accuracy 0.7412
Epoch 100, CIFAR-10 Batch 4:  loss 0.150519, train_accuracy 0.975, valid accuracy 0.765
Epoch 100, CIFAR-10 Batch 5:  loss 0.165463, train_accuracy 1, valid accuracy 0.73
Epoch 101, CIFAR-10 Batch 1:  loss 0.200974, train_accuracy 0.975, valid accuracy 0.7578
Epoch 101, CIFAR-10 Batch 2:  loss 0.175899, train_accuracy 1, valid accuracy 0.7252
Epoch 101, CIFAR-10 Batch 3:  loss 0.168425, train_accuracy 0.975, valid accuracy 0.7282
Epoch 101, CIFAR-10 Batch 4:  loss 0.164477, train_accuracy 0.95, valid accuracy 0.757
Epoch 101, CIFAR-10 Batch 5:  loss 0.153765, train_accuracy 1, valid accuracy 0.7384
Epoch 102, CIFAR-10 Batch 1:  loss 0.182251, train_accuracy 1, valid accuracy 0.761
Epoch 102, CIFAR-10 Batch 2:  loss 0.173281, train_accuracy 1, valid accuracy 0.7344
Epoch 102, CIFAR-10 Batch 3:  loss 0.158348, train_accuracy 0.975, valid accuracy 0.7332
Epoch 102, CIFAR-10 Batch 4:  loss 0.177006, train_accuracy 0.95, valid accuracy 0.7422
Epoch 102, CIFAR-10 Batch 5:  loss 0.142608, train_accuracy 1, valid accuracy 0.7424
Epoch 103, CIFAR-10 Batch 1:  loss 0.189819, train_accuracy 1, valid accuracy 0.7594
Epoch 103, CIFAR-10 Batch 2:  loss 0.163389, train_accuracy 1, valid accuracy 0.7492
Epoch 103, CIFAR-10 Batch 3:  loss 0.200847, train_accuracy 0.95, valid accuracy 0.7356
Epoch 103, CIFAR-10 Batch 4:  loss 0.150733, train_accuracy 1, valid accuracy 0.758
Epoch 103, CIFAR-10 Batch 5:  loss 0.136260, train_accuracy 1, valid accuracy 0.7434
Epoch 104, CIFAR-10 Batch 1:  loss 0.214737, train_accuracy 1, valid accuracy 0.7548
Epoch 104, CIFAR-10 Batch 2:  loss 0.163329, train_accuracy 0.975, valid accuracy 0.7414
Epoch 104, CIFAR-10 Batch 3:  loss 0.186494, train_accuracy 1, valid accuracy 0.7444
Epoch 104, CIFAR-10 Batch 4:  loss 0.144177, train_accuracy 0.975, valid accuracy 0.751
Epoch 104, CIFAR-10 Batch 5:  loss 0.181900, train_accuracy 1, valid accuracy 0.732
Epoch 105, CIFAR-10 Batch 1:  loss 0.170262, train_accuracy 1, valid accuracy 0.7668
Epoch 105, CIFAR-10 Batch 2:  loss 0.180172, train_accuracy 1, valid accuracy 0.7502
Epoch 105, CIFAR-10 Batch 3:  loss 0.156582, train_accuracy 1, valid accuracy 0.738
Epoch 105, CIFAR-10 Batch 4:  loss 0.175381, train_accuracy 0.975, valid accuracy 0.7552
Epoch 105, CIFAR-10 Batch 5:  loss 0.112746, train_accuracy 1, valid accuracy 0.7514
Epoch 106, CIFAR-10 Batch 1:  loss 0.178845, train_accuracy 1, valid accuracy 0.754
Epoch 106, CIFAR-10 Batch 2:  loss 0.185258, train_accuracy 0.975, valid accuracy 0.7386
Epoch 106, CIFAR-10 Batch 3:  loss 0.149840, train_accuracy 1, valid accuracy 0.7488
Epoch 106, CIFAR-10 Batch 4:  loss 0.147479, train_accuracy 1, valid accuracy 0.76
Epoch 106, CIFAR-10 Batch 5:  loss 0.146041, train_accuracy 1, valid accuracy 0.752
Epoch 107, CIFAR-10 Batch 1:  loss 0.145794, train_accuracy 1, valid accuracy 0.7632
Epoch 107, CIFAR-10 Batch 2:  loss 0.155446, train_accuracy 1, valid accuracy 0.7356
Epoch 107, CIFAR-10 Batch 3:  loss 0.173740, train_accuracy 0.975, valid accuracy 0.737
Epoch 107, CIFAR-10 Batch 4:  loss 0.157232, train_accuracy 0.95, valid accuracy 0.7574
Epoch 107, CIFAR-10 Batch 5:  loss 0.103955, train_accuracy 1, valid accuracy 0.7432
Epoch 108, CIFAR-10 Batch 1:  loss 0.149351, train_accuracy 1, valid accuracy 0.7672
Epoch 108, CIFAR-10 Batch 2:  loss 0.141181, train_accuracy 1, valid accuracy 0.746
Epoch 108, CIFAR-10 Batch 3:  loss 0.128715, train_accuracy 1, valid accuracy 0.7554
Epoch 108, CIFAR-10 Batch 4:  loss 0.189682, train_accuracy 1, valid accuracy 0.7462
Epoch 108, CIFAR-10 Batch 5:  loss 0.107694, train_accuracy 1, valid accuracy 0.7492
Epoch 109, CIFAR-10 Batch 1:  loss 0.152399, train_accuracy 1, valid accuracy 0.7688
Epoch 109, CIFAR-10 Batch 2:  loss 0.106702, train_accuracy 1, valid accuracy 0.762
Epoch 109, CIFAR-10 Batch 3:  loss 0.146071, train_accuracy 1, valid accuracy 0.7634
Epoch 109, CIFAR-10 Batch 4:  loss 0.142716, train_accuracy 1, valid accuracy 0.7644
Epoch 109, CIFAR-10 Batch 5:  loss 0.115409, train_accuracy 1, valid accuracy 0.7468
Epoch 110, CIFAR-10 Batch 1:  loss 0.169844, train_accuracy 1, valid accuracy 0.7616
Epoch 110, CIFAR-10 Batch 2:  loss 0.142787, train_accuracy 1, valid accuracy 0.738
Epoch 110, CIFAR-10 Batch 3:  loss 0.147212, train_accuracy 0.975, valid accuracy 0.7684
Epoch 110, CIFAR-10 Batch 4:  loss 0.137188, train_accuracy 0.975, valid accuracy 0.7646
Epoch 110, CIFAR-10 Batch 5:  loss 0.096221, train_accuracy 1, valid accuracy 0.7542
Epoch 111, CIFAR-10 Batch 1:  loss 0.153003, train_accuracy 1, valid accuracy 0.7694
Epoch 111, CIFAR-10 Batch 2:  loss 0.149669, train_accuracy 1, valid accuracy 0.7404
Epoch 111, CIFAR-10 Batch 3:  loss 0.151975, train_accuracy 1, valid accuracy 0.7432
Epoch 111, CIFAR-10 Batch 4:  loss 0.126635, train_accuracy 1, valid accuracy 0.7614
Epoch 111, CIFAR-10 Batch 5:  loss 0.086850, train_accuracy 1, valid accuracy 0.7638
Epoch 112, CIFAR-10 Batch 1:  loss 0.127230, train_accuracy 1, valid accuracy 0.7708
Epoch 112, CIFAR-10 Batch 2:  loss 0.108316, train_accuracy 1, valid accuracy 0.7472
Epoch 112, CIFAR-10 Batch 3:  loss 0.134257, train_accuracy 1, valid accuracy 0.7526
Epoch 112, CIFAR-10 Batch 4:  loss 0.151566, train_accuracy 0.95, valid accuracy 0.7646
Epoch 112, CIFAR-10 Batch 5:  loss 0.087767, train_accuracy 1, valid accuracy 0.76
Epoch 113, CIFAR-10 Batch 1:  loss 0.114350, train_accuracy 1, valid accuracy 0.7602
Epoch 113, CIFAR-10 Batch 2:  loss 0.113712, train_accuracy 1, valid accuracy 0.7388
Epoch 113, CIFAR-10 Batch 3:  loss 0.146909, train_accuracy 1, valid accuracy 0.7448
Epoch 113, CIFAR-10 Batch 4:  loss 0.178822, train_accuracy 0.975, valid accuracy 0.7596
Epoch 113, CIFAR-10 Batch 5:  loss 0.106407, train_accuracy 1, valid accuracy 0.7472
Epoch 114, CIFAR-10 Batch 1:  loss 0.173264, train_accuracy 1, valid accuracy 0.7628
Epoch 114, CIFAR-10 Batch 2:  loss 0.124487, train_accuracy 1, valid accuracy 0.7508
Epoch 114, CIFAR-10 Batch 3:  loss 0.135689, train_accuracy 1, valid accuracy 0.7428
Epoch 114, CIFAR-10 Batch 4:  loss 0.144941, train_accuracy 1, valid accuracy 0.7588
Epoch 114, CIFAR-10 Batch 5:  loss 0.142013, train_accuracy 1, valid accuracy 0.7238
Epoch 115, CIFAR-10 Batch 1:  loss 0.139609, train_accuracy 1, valid accuracy 0.757
Epoch 115, CIFAR-10 Batch 2:  loss 0.139380, train_accuracy 1, valid accuracy 0.7262
Epoch 115, CIFAR-10 Batch 3:  loss 0.161087, train_accuracy 1, valid accuracy 0.72
Epoch 115, CIFAR-10 Batch 4:  loss 0.146779, train_accuracy 1, valid accuracy 0.765
Epoch 115, CIFAR-10 Batch 5:  loss 0.103279, train_accuracy 1, valid accuracy 0.7614
Epoch 116, CIFAR-10 Batch 1:  loss 0.124400, train_accuracy 1, valid accuracy 0.7604
Epoch 116, CIFAR-10 Batch 2:  loss 0.119178, train_accuracy 1, valid accuracy 0.7514
Epoch 116, CIFAR-10 Batch 3:  loss 0.144206, train_accuracy 1, valid accuracy 0.7552
Epoch 116, CIFAR-10 Batch 4:  loss 0.129640, train_accuracy 1, valid accuracy 0.757
Epoch 116, CIFAR-10 Batch 5:  loss 0.084232, train_accuracy 1, valid accuracy 0.7706
Epoch 117, CIFAR-10 Batch 1:  loss 0.119316, train_accuracy 1, valid accuracy 0.7608
Epoch 117, CIFAR-10 Batch 2:  loss 0.085422, train_accuracy 1, valid accuracy 0.7708
Epoch 117, CIFAR-10 Batch 3:  loss 0.130962, train_accuracy 1, valid accuracy 0.763
Epoch 117, CIFAR-10 Batch 4:  loss 0.143391, train_accuracy 1, valid accuracy 0.7608
Epoch 117, CIFAR-10 Batch 5:  loss 0.089652, train_accuracy 1, valid accuracy 0.753
Epoch 118, CIFAR-10 Batch 1:  loss 0.112740, train_accuracy 1, valid accuracy 0.7728
Epoch 118, CIFAR-10 Batch 2:  loss 0.145422, train_accuracy 0.975, valid accuracy 0.74
Epoch 118, CIFAR-10 Batch 3:  loss 0.131971, train_accuracy 1, valid accuracy 0.7464
Epoch 118, CIFAR-10 Batch 4:  loss 0.115012, train_accuracy 1, valid accuracy 0.7666
Epoch 118, CIFAR-10 Batch 5:  loss 0.079504, train_accuracy 1, valid accuracy 0.7638
Epoch 119, CIFAR-10 Batch 1:  loss 0.113977, train_accuracy 1, valid accuracy 0.7772
Epoch 119, CIFAR-10 Batch 2:  loss 0.108491, train_accuracy 1, valid accuracy 0.741
Epoch 119, CIFAR-10 Batch 3:  loss 0.117242, train_accuracy 1, valid accuracy 0.7442
Epoch 119, CIFAR-10 Batch 4:  loss 0.141286, train_accuracy 1, valid accuracy 0.7578
Epoch 119, CIFAR-10 Batch 5:  loss 0.095137, train_accuracy 1, valid accuracy 0.76
Epoch 120, CIFAR-10 Batch 1:  loss 0.116266, train_accuracy 1, valid accuracy 0.7734
Epoch 120, CIFAR-10 Batch 2:  loss 0.094064, train_accuracy 1, valid accuracy 0.7688
Epoch 120, CIFAR-10 Batch 3:  loss 0.128597, train_accuracy 1, valid accuracy 0.7486
Epoch 120, CIFAR-10 Batch 4:  loss 0.112146, train_accuracy 1, valid accuracy 0.7724
Epoch 120, CIFAR-10 Batch 5:  loss 0.059790, train_accuracy 1, valid accuracy 0.7716
Epoch 121, CIFAR-10 Batch 1:  loss 0.109376, train_accuracy 1, valid accuracy 0.7706
Epoch 121, CIFAR-10 Batch 2:  loss 0.130408, train_accuracy 1, valid accuracy 0.7264
Epoch 121, CIFAR-10 Batch 3:  loss 0.103876, train_accuracy 1, valid accuracy 0.7582
Epoch 121, CIFAR-10 Batch 4:  loss 0.109577, train_accuracy 1, valid accuracy 0.7668
Epoch 121, CIFAR-10 Batch 5:  loss 0.085516, train_accuracy 1, valid accuracy 0.7578
Epoch 122, CIFAR-10 Batch 1:  loss 0.114699, train_accuracy 1, valid accuracy 0.7678
Epoch 122, CIFAR-10 Batch 2:  loss 0.092775, train_accuracy 1, valid accuracy 0.757
Epoch 122, CIFAR-10 Batch 3:  loss 0.120725, train_accuracy 1, valid accuracy 0.745
Epoch 122, CIFAR-10 Batch 4:  loss 0.087679, train_accuracy 1, valid accuracy 0.7694
Epoch 122, CIFAR-10 Batch 5:  loss 0.103477, train_accuracy 1, valid accuracy 0.7512
Epoch 123, CIFAR-10 Batch 1:  loss 0.104650, train_accuracy 1, valid accuracy 0.7768
Epoch 123, CIFAR-10 Batch 2:  loss 0.084737, train_accuracy 1, valid accuracy 0.7536
Epoch 123, CIFAR-10 Batch 3:  loss 0.099333, train_accuracy 1, valid accuracy 0.7634
Epoch 123, CIFAR-10 Batch 4:  loss 0.080360, train_accuracy 1, valid accuracy 0.7704
Epoch 123, CIFAR-10 Batch 5:  loss 0.084027, train_accuracy 1, valid accuracy 0.773
Epoch 124, CIFAR-10 Batch 1:  loss 0.079694, train_accuracy 1, valid accuracy 0.777
Epoch 124, CIFAR-10 Batch 2:  loss 0.080198, train_accuracy 1, valid accuracy 0.7546
Epoch 124, CIFAR-10 Batch 3:  loss 0.082638, train_accuracy 1, valid accuracy 0.7544
Epoch 124, CIFAR-10 Batch 4:  loss 0.090729, train_accuracy 1, valid accuracy 0.7726
Epoch 124, CIFAR-10 Batch 5:  loss 0.090900, train_accuracy 1, valid accuracy 0.7526
Epoch 125, CIFAR-10 Batch 1:  loss 0.081832, train_accuracy 1, valid accuracy 0.7806
Epoch 125, CIFAR-10 Batch 2:  loss 0.090939, train_accuracy 1, valid accuracy 0.75
Epoch 125, CIFAR-10 Batch 3:  loss 0.072840, train_accuracy 1, valid accuracy 0.7618
Epoch 125, CIFAR-10 Batch 4:  loss 0.098424, train_accuracy 0.975, valid accuracy 0.7654
Epoch 125, CIFAR-10 Batch 5:  loss 0.071199, train_accuracy 1, valid accuracy 0.7562
Epoch 126, CIFAR-10 Batch 1:  loss 0.089007, train_accuracy 1, valid accuracy 0.7692
Epoch 126, CIFAR-10 Batch 2:  loss 0.071252, train_accuracy 1, valid accuracy 0.7458
Epoch 126, CIFAR-10 Batch 3:  loss 0.068289, train_accuracy 1, valid accuracy 0.7612
Epoch 126, CIFAR-10 Batch 4:  loss 0.163706, train_accuracy 0.975, valid accuracy 0.7508
Epoch 126, CIFAR-10 Batch 5:  loss 0.084356, train_accuracy 1, valid accuracy 0.739
Epoch 127, CIFAR-10 Batch 1:  loss 0.081390, train_accuracy 1, valid accuracy 0.7676
Epoch 127, CIFAR-10 Batch 2:  loss 0.115803, train_accuracy 1, valid accuracy 0.7378
Epoch 127, CIFAR-10 Batch 3:  loss 0.077971, train_accuracy 1, valid accuracy 0.7566
Epoch 127, CIFAR-10 Batch 4:  loss 0.100057, train_accuracy 1, valid accuracy 0.7624
Epoch 127, CIFAR-10 Batch 5:  loss 0.062108, train_accuracy 1, valid accuracy 0.761
Epoch 128, CIFAR-10 Batch 1:  loss 0.087679, train_accuracy 1, valid accuracy 0.7794
Epoch 128, CIFAR-10 Batch 2:  loss 0.059464, train_accuracy 1, valid accuracy 0.7612
Epoch 128, CIFAR-10 Batch 3:  loss 0.063561, train_accuracy 1, valid accuracy 0.7604
Epoch 128, CIFAR-10 Batch 4:  loss 0.101954, train_accuracy 1, valid accuracy 0.7556
Epoch 128, CIFAR-10 Batch 5:  loss 0.055586, train_accuracy 1, valid accuracy 0.7626
Epoch 129, CIFAR-10 Batch 1:  loss 0.094369, train_accuracy 1, valid accuracy 0.7764
Epoch 129, CIFAR-10 Batch 2:  loss 0.073989, train_accuracy 1, valid accuracy 0.7582
Epoch 129, CIFAR-10 Batch 3:  loss 0.073423, train_accuracy 1, valid accuracy 0.7588
Epoch 129, CIFAR-10 Batch 4:  loss 0.071902, train_accuracy 1, valid accuracy 0.7764
Epoch 129, CIFAR-10 Batch 5:  loss 0.052740, train_accuracy 1, valid accuracy 0.77
Epoch 130, CIFAR-10 Batch 1:  loss 0.085677, train_accuracy 1, valid accuracy 0.7684
Epoch 130, CIFAR-10 Batch 2:  loss 0.057832, train_accuracy 1, valid accuracy 0.7596
Epoch 130, CIFAR-10 Batch 3:  loss 0.069018, train_accuracy 1, valid accuracy 0.7616
Epoch 130, CIFAR-10 Batch 4:  loss 0.089614, train_accuracy 1, valid accuracy 0.7516
Epoch 130, CIFAR-10 Batch 5:  loss 0.106299, train_accuracy 1, valid accuracy 0.7628
Epoch 131, CIFAR-10 Batch 1:  loss 0.096902, train_accuracy 1, valid accuracy 0.7608
Epoch 131, CIFAR-10 Batch 2:  loss 0.089907, train_accuracy 1, valid accuracy 0.7372
Epoch 131, CIFAR-10 Batch 3:  loss 0.074626, train_accuracy 1, valid accuracy 0.7644
Epoch 131, CIFAR-10 Batch 4:  loss 0.071163, train_accuracy 1, valid accuracy 0.7632
Epoch 131, CIFAR-10 Batch 5:  loss 0.083832, train_accuracy 1, valid accuracy 0.7482
Epoch 132, CIFAR-10 Batch 1:  loss 0.087113, train_accuracy 1, valid accuracy 0.7742
Epoch 132, CIFAR-10 Batch 2:  loss 0.057124, train_accuracy 1, valid accuracy 0.7536
Epoch 132, CIFAR-10 Batch 3:  loss 0.060568, train_accuracy 1, valid accuracy 0.7666
Epoch 132, CIFAR-10 Batch 4:  loss 0.091620, train_accuracy 1, valid accuracy 0.7694
Epoch 132, CIFAR-10 Batch 5:  loss 0.075473, train_accuracy 1, valid accuracy 0.7634
Epoch 133, CIFAR-10 Batch 1:  loss 0.076120, train_accuracy 1, valid accuracy 0.7758
Epoch 133, CIFAR-10 Batch 2:  loss 0.049971, train_accuracy 1, valid accuracy 0.7606
Epoch 133, CIFAR-10 Batch 3:  loss 0.064961, train_accuracy 1, valid accuracy 0.7646
Epoch 133, CIFAR-10 Batch 4:  loss 0.067694, train_accuracy 1, valid accuracy 0.7582
Epoch 133, CIFAR-10 Batch 5:  loss 0.057637, train_accuracy 1, valid accuracy 0.7742
Epoch 134, CIFAR-10 Batch 1:  loss 0.071956, train_accuracy 1, valid accuracy 0.777
Epoch 134, CIFAR-10 Batch 2:  loss 0.041080, train_accuracy 1, valid accuracy 0.7696
Epoch 134, CIFAR-10 Batch 3:  loss 0.085768, train_accuracy 1, valid accuracy 0.7504
Epoch 134, CIFAR-10 Batch 4:  loss 0.093242, train_accuracy 1, valid accuracy 0.762
Epoch 134, CIFAR-10 Batch 5:  loss 0.049847, train_accuracy 1, valid accuracy 0.774
Epoch 135, CIFAR-10 Batch 1:  loss 0.082924, train_accuracy 1, valid accuracy 0.7654
Epoch 135, CIFAR-10 Batch 2:  loss 0.048733, train_accuracy 1, valid accuracy 0.7402
Epoch 135, CIFAR-10 Batch 3:  loss 0.061726, train_accuracy 1, valid accuracy 0.7638
Epoch 135, CIFAR-10 Batch 4:  loss 0.073138, train_accuracy 1, valid accuracy 0.7768
Epoch 135, CIFAR-10 Batch 5:  loss 0.071892, train_accuracy 1, valid accuracy 0.7544
Epoch 136, CIFAR-10 Batch 1:  loss 0.076867, train_accuracy 1, valid accuracy 0.771
Epoch 136, CIFAR-10 Batch 2:  loss 0.035729, train_accuracy 1, valid accuracy 0.7696
Epoch 136, CIFAR-10 Batch 3:  loss 0.076245, train_accuracy 1, valid accuracy 0.7654
Epoch 136, CIFAR-10 Batch 4:  loss 0.061655, train_accuracy 1, valid accuracy 0.7838
Epoch 136, CIFAR-10 Batch 5:  loss 0.077607, train_accuracy 1, valid accuracy 0.7464
Epoch 137, CIFAR-10 Batch 1:  loss 0.086502, train_accuracy 1, valid accuracy 0.772
Epoch 137, CIFAR-10 Batch 2:  loss 0.050846, train_accuracy 1, valid accuracy 0.7636
Epoch 137, CIFAR-10 Batch 3:  loss 0.067862, train_accuracy 1, valid accuracy 0.776
Epoch 137, CIFAR-10 Batch 4:  loss 0.050852, train_accuracy 1, valid accuracy 0.773
Epoch 137, CIFAR-10 Batch 5:  loss 0.103253, train_accuracy 1, valid accuracy 0.7356
Epoch 138, CIFAR-10 Batch 1:  loss 0.073699, train_accuracy 1, valid accuracy 0.7786
Epoch 138, CIFAR-10 Batch 2:  loss 0.072762, train_accuracy 1, valid accuracy 0.7362
Epoch 138, CIFAR-10 Batch 3:  loss 0.045855, train_accuracy 1, valid accuracy 0.7802
Epoch 138, CIFAR-10 Batch 4:  loss 0.052903, train_accuracy 1, valid accuracy 0.7688
Epoch 138, CIFAR-10 Batch 5:  loss 0.072709, train_accuracy 1, valid accuracy 0.7636
Epoch 139, CIFAR-10 Batch 1:  loss 0.065842, train_accuracy 1, valid accuracy 0.7754
Epoch 139, CIFAR-10 Batch 2:  loss 0.048679, train_accuracy 1, valid accuracy 0.762
Epoch 139, CIFAR-10 Batch 3:  loss 0.067138, train_accuracy 1, valid accuracy 0.7564
Epoch 139, CIFAR-10 Batch 4:  loss 0.061532, train_accuracy 1, valid accuracy 0.7696
Epoch 139, CIFAR-10 Batch 5:  loss 0.069727, train_accuracy 1, valid accuracy 0.7652
Epoch 140, CIFAR-10 Batch 1:  loss 0.087470, train_accuracy 1, valid accuracy 0.7746
Epoch 140, CIFAR-10 Batch 2:  loss 0.053916, train_accuracy 1, valid accuracy 0.758
Epoch 140, CIFAR-10 Batch 3:  loss 0.038920, train_accuracy 1, valid accuracy 0.7754
Epoch 140, CIFAR-10 Batch 4:  loss 0.058706, train_accuracy 1, valid accuracy 0.7696
Epoch 140, CIFAR-10 Batch 5:  loss 0.058903, train_accuracy 1, valid accuracy 0.767
Epoch 141, CIFAR-10 Batch 1:  loss 0.069731, train_accuracy 1, valid accuracy 0.7806
Epoch 141, CIFAR-10 Batch 2:  loss 0.061664, train_accuracy 1, valid accuracy 0.7648
Epoch 141, CIFAR-10 Batch 3:  loss 0.036632, train_accuracy 1, valid accuracy 0.7736
Epoch 141, CIFAR-10 Batch 4:  loss 0.059635, train_accuracy 1, valid accuracy 0.786
Epoch 141, CIFAR-10 Batch 5:  loss 0.047789, train_accuracy 1, valid accuracy 0.766
Epoch 142, CIFAR-10 Batch 1:  loss 0.051732, train_accuracy 1, valid accuracy 0.7724
Epoch 142, CIFAR-10 Batch 2:  loss 0.040149, train_accuracy 1, valid accuracy 0.7526
Epoch 142, CIFAR-10 Batch 3:  loss 0.058502, train_accuracy 1, valid accuracy 0.766
Epoch 142, CIFAR-10 Batch 4:  loss 0.059437, train_accuracy 1, valid accuracy 0.7764
Epoch 142, CIFAR-10 Batch 5:  loss 0.089044, train_accuracy 1, valid accuracy 0.7624
Epoch 143, CIFAR-10 Batch 1:  loss 0.060771, train_accuracy 1, valid accuracy 0.768
Epoch 143, CIFAR-10 Batch 2:  loss 0.042642, train_accuracy 1, valid accuracy 0.7536
Epoch 143, CIFAR-10 Batch 3:  loss 0.079303, train_accuracy 1, valid accuracy 0.76
Epoch 143, CIFAR-10 Batch 4:  loss 0.056654, train_accuracy 1, valid accuracy 0.779
Epoch 143, CIFAR-10 Batch 5:  loss 0.059908, train_accuracy 1, valid accuracy 0.7624
Epoch 144, CIFAR-10 Batch 1:  loss 0.053706, train_accuracy 1, valid accuracy 0.7804
Epoch 144, CIFAR-10 Batch 2:  loss 0.038228, train_accuracy 1, valid accuracy 0.7654
Epoch 144, CIFAR-10 Batch 3:  loss 0.058099, train_accuracy 1, valid accuracy 0.758
Epoch 144, CIFAR-10 Batch 4:  loss 0.066170, train_accuracy 0.975, valid accuracy 0.778
Epoch 144, CIFAR-10 Batch 5:  loss 0.049193, train_accuracy 1, valid accuracy 0.7758
Epoch 145, CIFAR-10 Batch 1:  loss 0.046952, train_accuracy 1, valid accuracy 0.7844
Epoch 145, CIFAR-10 Batch 2:  loss 0.033329, train_accuracy 1, valid accuracy 0.7772
Epoch 145, CIFAR-10 Batch 3:  loss 0.070127, train_accuracy 1, valid accuracy 0.7672
Epoch 145, CIFAR-10 Batch 4:  loss 0.063872, train_accuracy 1, valid accuracy 0.7728
Epoch 145, CIFAR-10 Batch 5:  loss 0.031907, train_accuracy 1, valid accuracy 0.7766
Epoch 146, CIFAR-10 Batch 1:  loss 0.051123, train_accuracy 1, valid accuracy 0.7804
Epoch 146, CIFAR-10 Batch 2:  loss 0.064065, train_accuracy 0.975, valid accuracy 0.7538
Epoch 146, CIFAR-10 Batch 3:  loss 0.067346, train_accuracy 1, valid accuracy 0.7646
Epoch 146, CIFAR-10 Batch 4:  loss 0.069656, train_accuracy 1, valid accuracy 0.7658
Epoch 146, CIFAR-10 Batch 5:  loss 0.066386, train_accuracy 1, valid accuracy 0.7426
Epoch 147, CIFAR-10 Batch 1:  loss 0.045137, train_accuracy 1, valid accuracy 0.7816
Epoch 147, CIFAR-10 Batch 2:  loss 0.038330, train_accuracy 1, valid accuracy 0.763
Epoch 147, CIFAR-10 Batch 3:  loss 0.044856, train_accuracy 1, valid accuracy 0.777
Epoch 147, CIFAR-10 Batch 4:  loss 0.078312, train_accuracy 0.975, valid accuracy 0.7612
Epoch 147, CIFAR-10 Batch 5:  loss 0.042709, train_accuracy 1, valid accuracy 0.7776
Epoch 148, CIFAR-10 Batch 1:  loss 0.040386, train_accuracy 1, valid accuracy 0.781
Epoch 148, CIFAR-10 Batch 2:  loss 0.032122, train_accuracy 1, valid accuracy 0.7558
Epoch 148, CIFAR-10 Batch 3:  loss 0.043818, train_accuracy 1, valid accuracy 0.7744
Epoch 148, CIFAR-10 Batch 4:  loss 0.045029, train_accuracy 1, valid accuracy 0.782
Epoch 148, CIFAR-10 Batch 5:  loss 0.047014, train_accuracy 1, valid accuracy 0.7666
Epoch 149, CIFAR-10 Batch 1:  loss 0.055759, train_accuracy 1, valid accuracy 0.7904
Epoch 149, CIFAR-10 Batch 2:  loss 0.040617, train_accuracy 1, valid accuracy 0.7692
Epoch 149, CIFAR-10 Batch 3:  loss 0.062168, train_accuracy 1, valid accuracy 0.77
Epoch 149, CIFAR-10 Batch 4:  loss 0.064705, train_accuracy 1, valid accuracy 0.7662
Epoch 149, CIFAR-10 Batch 5:  loss 0.043368, train_accuracy 1, valid accuracy 0.7656
Epoch 150, CIFAR-10 Batch 1:  loss 0.056404, train_accuracy 1, valid accuracy 0.7784
Epoch 150, CIFAR-10 Batch 2:  loss 0.042379, train_accuracy 1, valid accuracy 0.7816
Epoch 150, CIFAR-10 Batch 3:  loss 0.036468, train_accuracy 1, valid accuracy 0.7832
Epoch 150, CIFAR-10 Batch 4:  loss 0.066334, train_accuracy 0.975, valid accuracy 0.7688
Epoch 150, CIFAR-10 Batch 5:  loss 0.052403, train_accuracy 1, valid accuracy 0.7678
Epoch 151, CIFAR-10 Batch 1:  loss 0.059825, train_accuracy 1, valid accuracy 0.7854
Epoch 151, CIFAR-10 Batch 2:  loss 0.038964, train_accuracy 1, valid accuracy 0.7688
Epoch 151, CIFAR-10 Batch 3:  loss 0.043168, train_accuracy 1, valid accuracy 0.7758
Epoch 151, CIFAR-10 Batch 4:  loss 0.042927, train_accuracy 1, valid accuracy 0.7814
Epoch 151, CIFAR-10 Batch 5:  loss 0.039540, train_accuracy 1, valid accuracy 0.7686
Epoch 152, CIFAR-10 Batch 1:  loss 0.051347, train_accuracy 1, valid accuracy 0.7778
Epoch 152, CIFAR-10 Batch 2:  loss 0.046129, train_accuracy 1, valid accuracy 0.779
Epoch 152, CIFAR-10 Batch 3:  loss 0.053208, train_accuracy 1, valid accuracy 0.766
Epoch 152, CIFAR-10 Batch 4:  loss 0.059126, train_accuracy 0.975, valid accuracy 0.7738
Epoch 152, CIFAR-10 Batch 5:  loss 0.030179, train_accuracy 1, valid accuracy 0.7738
Epoch 153, CIFAR-10 Batch 1:  loss 0.038702, train_accuracy 1, valid accuracy 0.7842
Epoch 153, CIFAR-10 Batch 2:  loss 0.033709, train_accuracy 1, valid accuracy 0.781
Epoch 153, CIFAR-10 Batch 3:  loss 0.044980, train_accuracy 1, valid accuracy 0.7632
Epoch 153, CIFAR-10 Batch 4:  loss 0.030266, train_accuracy 1, valid accuracy 0.78
Epoch 153, CIFAR-10 Batch 5:  loss 0.047337, train_accuracy 1, valid accuracy 0.778
Epoch 154, CIFAR-10 Batch 1:  loss 0.034135, train_accuracy 1, valid accuracy 0.7822
Epoch 154, CIFAR-10 Batch 2:  loss 0.022310, train_accuracy 1, valid accuracy 0.7788
Epoch 154, CIFAR-10 Batch 3:  loss 0.034123, train_accuracy 1, valid accuracy 0.7724
Epoch 154, CIFAR-10 Batch 4:  loss 0.039459, train_accuracy 1, valid accuracy 0.7838
Epoch 154, CIFAR-10 Batch 5:  loss 0.045504, train_accuracy 1, valid accuracy 0.7688
Epoch 155, CIFAR-10 Batch 1:  loss 0.041743, train_accuracy 1, valid accuracy 0.7812
Epoch 155, CIFAR-10 Batch 2:  loss 0.042963, train_accuracy 1, valid accuracy 0.7664
Epoch 155, CIFAR-10 Batch 3:  loss 0.031257, train_accuracy 1, valid accuracy 0.7846
Epoch 155, CIFAR-10 Batch 4:  loss 0.034852, train_accuracy 1, valid accuracy 0.7772
Epoch 155, CIFAR-10 Batch 5:  loss 0.038767, train_accuracy 1, valid accuracy 0.7692
Epoch 156, CIFAR-10 Batch 1:  loss 0.038363, train_accuracy 1, valid accuracy 0.7826
Epoch 156, CIFAR-10 Batch 2:  loss 0.024972, train_accuracy 1, valid accuracy 0.7792
Epoch 156, CIFAR-10 Batch 3:  loss 0.035612, train_accuracy 1, valid accuracy 0.7744
Epoch 156, CIFAR-10 Batch 4:  loss 0.074001, train_accuracy 0.975, valid accuracy 0.7668
Epoch 156, CIFAR-10 Batch 5:  loss 0.041205, train_accuracy 1, valid accuracy 0.7604
Epoch 157, CIFAR-10 Batch 1:  loss 0.043662, train_accuracy 1, valid accuracy 0.7848
Epoch 157, CIFAR-10 Batch 2:  loss 0.026762, train_accuracy 1, valid accuracy 0.7696
Epoch 157, CIFAR-10 Batch 3:  loss 0.033257, train_accuracy 1, valid accuracy 0.7808
Epoch 157, CIFAR-10 Batch 4:  loss 0.057474, train_accuracy 1, valid accuracy 0.777
Epoch 157, CIFAR-10 Batch 5:  loss 0.034308, train_accuracy 1, valid accuracy 0.7766
Epoch 158, CIFAR-10 Batch 1:  loss 0.046933, train_accuracy 1, valid accuracy 0.7814
Epoch 158, CIFAR-10 Batch 2:  loss 0.020183, train_accuracy 1, valid accuracy 0.78
Epoch 158, CIFAR-10 Batch 3:  loss 0.029607, train_accuracy 1, valid accuracy 0.7744
Epoch 158, CIFAR-10 Batch 4:  loss 0.047996, train_accuracy 1, valid accuracy 0.7814
Epoch 158, CIFAR-10 Batch 5:  loss 0.041868, train_accuracy 1, valid accuracy 0.7752
Epoch 159, CIFAR-10 Batch 1:  loss 0.044711, train_accuracy 1, valid accuracy 0.782
Epoch 159, CIFAR-10 Batch 2:  loss 0.061200, train_accuracy 1, valid accuracy 0.7596
Epoch 159, CIFAR-10 Batch 3:  loss 0.034721, train_accuracy 1, valid accuracy 0.7656
Epoch 159, CIFAR-10 Batch 4:  loss 0.036171, train_accuracy 1, valid accuracy 0.7808
Epoch 159, CIFAR-10 Batch 5:  loss 0.067569, train_accuracy 1, valid accuracy 0.7494
Epoch 160, CIFAR-10 Batch 1:  loss 0.042177, train_accuracy 1, valid accuracy 0.7792
Epoch 160, CIFAR-10 Batch 2:  loss 0.029247, train_accuracy 1, valid accuracy 0.7814
Epoch 160, CIFAR-10 Batch 3:  loss 0.024640, train_accuracy 1, valid accuracy 0.775
Epoch 160, CIFAR-10 Batch 4:  loss 0.036433, train_accuracy 1, valid accuracy 0.7754
Epoch 160, CIFAR-10 Batch 5:  loss 0.050875, train_accuracy 1, valid accuracy 0.7656
Epoch 161, CIFAR-10 Batch 1:  loss 0.035379, train_accuracy 1, valid accuracy 0.7804
Epoch 161, CIFAR-10 Batch 2:  loss 0.017564, train_accuracy 1, valid accuracy 0.7782
Epoch 161, CIFAR-10 Batch 3:  loss 0.039190, train_accuracy 1, valid accuracy 0.7696
Epoch 161, CIFAR-10 Batch 4:  loss 0.045438, train_accuracy 1, valid accuracy 0.7808
Epoch 161, CIFAR-10 Batch 5:  loss 0.030178, train_accuracy 1, valid accuracy 0.7794
Epoch 162, CIFAR-10 Batch 1:  loss 0.034243, train_accuracy 1, valid accuracy 0.7858
Epoch 162, CIFAR-10 Batch 2:  loss 0.036548, train_accuracy 1, valid accuracy 0.7666
Epoch 162, CIFAR-10 Batch 3:  loss 0.029155, train_accuracy 1, valid accuracy 0.7706
Epoch 162, CIFAR-10 Batch 4:  loss 0.050178, train_accuracy 1, valid accuracy 0.7712
Epoch 162, CIFAR-10 Batch 5:  loss 0.028662, train_accuracy 1, valid accuracy 0.7702
Epoch 163, CIFAR-10 Batch 1:  loss 0.038500, train_accuracy 1, valid accuracy 0.7884
Epoch 163, CIFAR-10 Batch 2:  loss 0.030928, train_accuracy 1, valid accuracy 0.7754
Epoch 163, CIFAR-10 Batch 3:  loss 0.049153, train_accuracy 1, valid accuracy 0.7822
Epoch 163, CIFAR-10 Batch 4:  loss 0.067394, train_accuracy 1, valid accuracy 0.7634
Epoch 163, CIFAR-10 Batch 5:  loss 0.026409, train_accuracy 1, valid accuracy 0.784
Epoch 164, CIFAR-10 Batch 1:  loss 0.039543, train_accuracy 1, valid accuracy 0.7806
Epoch 164, CIFAR-10 Batch 2:  loss 0.031830, train_accuracy 1, valid accuracy 0.7744
Epoch 164, CIFAR-10 Batch 3:  loss 0.033778, train_accuracy 1, valid accuracy 0.7796
Epoch 164, CIFAR-10 Batch 4:  loss 0.029514, train_accuracy 1, valid accuracy 0.783
Epoch 164, CIFAR-10 Batch 5:  loss 0.026419, train_accuracy 1, valid accuracy 0.7776
Epoch 165, CIFAR-10 Batch 1:  loss 0.034781, train_accuracy 1, valid accuracy 0.7892
Epoch 165, CIFAR-10 Batch 2:  loss 0.026147, train_accuracy 1, valid accuracy 0.7884
Epoch 165, CIFAR-10 Batch 3:  loss 0.021372, train_accuracy 1, valid accuracy 0.7874
Epoch 165, CIFAR-10 Batch 4:  loss 0.018998, train_accuracy 1, valid accuracy 0.7868
Epoch 165, CIFAR-10 Batch 5:  loss 0.035600, train_accuracy 1, valid accuracy 0.7784
Epoch 166, CIFAR-10 Batch 1:  loss 0.026438, train_accuracy 1, valid accuracy 0.7818
Epoch 166, CIFAR-10 Batch 2:  loss 0.024313, train_accuracy 1, valid accuracy 0.78
Epoch 166, CIFAR-10 Batch 3:  loss 0.025620, train_accuracy 1, valid accuracy 0.7856
Epoch 166, CIFAR-10 Batch 4:  loss 0.047858, train_accuracy 1, valid accuracy 0.7812
Epoch 166, CIFAR-10 Batch 5:  loss 0.043412, train_accuracy 1, valid accuracy 0.767
Epoch 167, CIFAR-10 Batch 1:  loss 0.040185, train_accuracy 1, valid accuracy 0.7768
Epoch 167, CIFAR-10 Batch 2:  loss 0.028984, train_accuracy 1, valid accuracy 0.777
Epoch 167, CIFAR-10 Batch 3:  loss 0.034236, train_accuracy 1, valid accuracy 0.7752
Epoch 167, CIFAR-10 Batch 4:  loss 0.046434, train_accuracy 1, valid accuracy 0.7838
Epoch 167, CIFAR-10 Batch 5:  loss 0.035854, train_accuracy 1, valid accuracy 0.7778
Epoch 168, CIFAR-10 Batch 1:  loss 0.024937, train_accuracy 1, valid accuracy 0.7958
Epoch 168, CIFAR-10 Batch 2:  loss 0.039193, train_accuracy 1, valid accuracy 0.7618
Epoch 168, CIFAR-10 Batch 3:  loss 0.033515, train_accuracy 1, valid accuracy 0.7826
Epoch 168, CIFAR-10 Batch 4:  loss 0.043801, train_accuracy 0.975, valid accuracy 0.7846
Epoch 168, CIFAR-10 Batch 5:  loss 0.039815, train_accuracy 1, valid accuracy 0.7442
Epoch 169, CIFAR-10 Batch 1:  loss 0.027599, train_accuracy 1, valid accuracy 0.7828
Epoch 169, CIFAR-10 Batch 2:  loss 0.023173, train_accuracy 1, valid accuracy 0.7718
Epoch 169, CIFAR-10 Batch 3:  loss 0.032100, train_accuracy 1, valid accuracy 0.7682
Epoch 169, CIFAR-10 Batch 4:  loss 0.037516, train_accuracy 1, valid accuracy 0.7762
Epoch 169, CIFAR-10 Batch 5:  loss 0.039999, train_accuracy 1, valid accuracy 0.7688
Epoch 170, CIFAR-10 Batch 1:  loss 0.029894, train_accuracy 1, valid accuracy 0.7756
Epoch 170, CIFAR-10 Batch 2:  loss 0.017163, train_accuracy 1, valid accuracy 0.7822
Epoch 170, CIFAR-10 Batch 3:  loss 0.044608, train_accuracy 1, valid accuracy 0.7874
Epoch 170, CIFAR-10 Batch 4:  loss 0.044865, train_accuracy 1, valid accuracy 0.7866
Epoch 170, CIFAR-10 Batch 5:  loss 0.022886, train_accuracy 1, valid accuracy 0.7798
Epoch 171, CIFAR-10 Batch 1:  loss 0.023781, train_accuracy 1, valid accuracy 0.79
Epoch 171, CIFAR-10 Batch 2:  loss 0.021268, train_accuracy 1, valid accuracy 0.7872
Epoch 171, CIFAR-10 Batch 3:  loss 0.029420, train_accuracy 1, valid accuracy 0.781
Epoch 171, CIFAR-10 Batch 4:  loss 0.049719, train_accuracy 0.975, valid accuracy 0.7692
Epoch 171, CIFAR-10 Batch 5:  loss 0.033261, train_accuracy 1, valid accuracy 0.7722
Epoch 172, CIFAR-10 Batch 1:  loss 0.030982, train_accuracy 1, valid accuracy 0.7866
Epoch 172, CIFAR-10 Batch 2:  loss 0.029981, train_accuracy 1, valid accuracy 0.7868
Epoch 172, CIFAR-10 Batch 3:  loss 0.038683, train_accuracy 1, valid accuracy 0.7752
Epoch 172, CIFAR-10 Batch 4:  loss 0.031797, train_accuracy 1, valid accuracy 0.7848
Epoch 172, CIFAR-10 Batch 5:  loss 0.025738, train_accuracy 1, valid accuracy 0.7766
Epoch 173, CIFAR-10 Batch 1:  loss 0.026394, train_accuracy 1, valid accuracy 0.795
Epoch 173, CIFAR-10 Batch 2:  loss 0.029774, train_accuracy 1, valid accuracy 0.7668
Epoch 173, CIFAR-10 Batch 3:  loss 0.028213, train_accuracy 1, valid accuracy 0.7824
Epoch 173, CIFAR-10 Batch 4:  loss 0.044686, train_accuracy 1, valid accuracy 0.7618
Epoch 173, CIFAR-10 Batch 5:  loss 0.022713, train_accuracy 1, valid accuracy 0.7674
Epoch 174, CIFAR-10 Batch 1:  loss 0.033169, train_accuracy 1, valid accuracy 0.7946
Epoch 174, CIFAR-10 Batch 2:  loss 0.028017, train_accuracy 1, valid accuracy 0.7768
Epoch 174, CIFAR-10 Batch 3:  loss 0.018478, train_accuracy 1, valid accuracy 0.786
Epoch 174, CIFAR-10 Batch 4:  loss 0.034992, train_accuracy 1, valid accuracy 0.7888
Epoch 174, CIFAR-10 Batch 5:  loss 0.029810, train_accuracy 1, valid accuracy 0.7654
Epoch 175, CIFAR-10 Batch 1:  loss 0.110487, train_accuracy 0.975, valid accuracy 0.7692
Epoch 175, CIFAR-10 Batch 2:  loss 0.011297, train_accuracy 1, valid accuracy 0.7818
Epoch 175, CIFAR-10 Batch 3:  loss 0.042072, train_accuracy 1, valid accuracy 0.7712
Epoch 175, CIFAR-10 Batch 4:  loss 0.040785, train_accuracy 1, valid accuracy 0.786
Epoch 175, CIFAR-10 Batch 5:  loss 0.016942, train_accuracy 1, valid accuracy 0.7914
Epoch 176, CIFAR-10 Batch 1:  loss 0.028312, train_accuracy 1, valid accuracy 0.791
Epoch 176, CIFAR-10 Batch 2:  loss 0.023843, train_accuracy 1, valid accuracy 0.7758
Epoch 176, CIFAR-10 Batch 3:  loss 0.015783, train_accuracy 1, valid accuracy 0.7918
Epoch 176, CIFAR-10 Batch 4:  loss 0.033127, train_accuracy 1, valid accuracy 0.7894
Epoch 176, CIFAR-10 Batch 5:  loss 0.016139, train_accuracy 1, valid accuracy 0.7706
Epoch 177, CIFAR-10 Batch 1:  loss 0.024850, train_accuracy 1, valid accuracy 0.7974
Epoch 177, CIFAR-10 Batch 2:  loss 0.018498, train_accuracy 1, valid accuracy 0.7902
Epoch 177, CIFAR-10 Batch 3:  loss 0.024839, train_accuracy 1, valid accuracy 0.7814
Epoch 177, CIFAR-10 Batch 4:  loss 0.026677, train_accuracy 1, valid accuracy 0.7842
Epoch 177, CIFAR-10 Batch 5:  loss 0.015957, train_accuracy 1, valid accuracy 0.779
Epoch 178, CIFAR-10 Batch 1:  loss 0.027042, train_accuracy 1, valid accuracy 0.7926
Epoch 178, CIFAR-10 Batch 2:  loss 0.092561, train_accuracy 0.975, valid accuracy 0.75
Epoch 178, CIFAR-10 Batch 3:  loss 0.022071, train_accuracy 1, valid accuracy 0.7792
Epoch 178, CIFAR-10 Batch 4:  loss 0.029273, train_accuracy 1, valid accuracy 0.7826
Epoch 178, CIFAR-10 Batch 5:  loss 0.018090, train_accuracy 1, valid accuracy 0.7806
Epoch 179, CIFAR-10 Batch 1:  loss 0.019984, train_accuracy 1, valid accuracy 0.7922
Epoch 179, CIFAR-10 Batch 2:  loss 0.015480, train_accuracy 1, valid accuracy 0.7882
Epoch 179, CIFAR-10 Batch 3:  loss 0.030072, train_accuracy 1, valid accuracy 0.7808
Epoch 179, CIFAR-10 Batch 4:  loss 0.016110, train_accuracy 1, valid accuracy 0.7884
Epoch 179, CIFAR-10 Batch 5:  loss 0.028567, train_accuracy 1, valid accuracy 0.7704
Epoch 180, CIFAR-10 Batch 1:  loss 0.023080, train_accuracy 1, valid accuracy 0.7916
Epoch 180, CIFAR-10 Batch 2:  loss 0.016288, train_accuracy 1, valid accuracy 0.7816
Epoch 180, CIFAR-10 Batch 3:  loss 0.015960, train_accuracy 1, valid accuracy 0.787
Epoch 180, CIFAR-10 Batch 4:  loss 0.031681, train_accuracy 1, valid accuracy 0.7884
Epoch 180, CIFAR-10 Batch 5:  loss 0.027985, train_accuracy 1, valid accuracy 0.7934
Epoch 181, CIFAR-10 Batch 1:  loss 0.047364, train_accuracy 1, valid accuracy 0.7834
Epoch 181, CIFAR-10 Batch 2:  loss 0.028401, train_accuracy 1, valid accuracy 0.7736
Epoch 181, CIFAR-10 Batch 3:  loss 0.026866, train_accuracy 1, valid accuracy 0.7842
Epoch 181, CIFAR-10 Batch 4:  loss 0.016979, train_accuracy 1, valid accuracy 0.7832
Epoch 181, CIFAR-10 Batch 5:  loss 0.023727, train_accuracy 1, valid accuracy 0.7836
Epoch 182, CIFAR-10 Batch 1:  loss 0.026873, train_accuracy 1, valid accuracy 0.7996
Epoch 182, CIFAR-10 Batch 2:  loss 0.029161, train_accuracy 1, valid accuracy 0.7674
Epoch 182, CIFAR-10 Batch 3:  loss 0.029087, train_accuracy 1, valid accuracy 0.7868
Epoch 182, CIFAR-10 Batch 4:  loss 0.021366, train_accuracy 1, valid accuracy 0.7844
Epoch 182, CIFAR-10 Batch 5:  loss 0.024557, train_accuracy 1, valid accuracy 0.7898
Epoch 183, CIFAR-10 Batch 1:  loss 0.023892, train_accuracy 1, valid accuracy 0.7846
Epoch 183, CIFAR-10 Batch 2:  loss 0.021660, train_accuracy 1, valid accuracy 0.7844
Epoch 183, CIFAR-10 Batch 3:  loss 0.030171, train_accuracy 1, valid accuracy 0.7908
Epoch 183, CIFAR-10 Batch 4:  loss 0.047316, train_accuracy 0.975, valid accuracy 0.776
Epoch 183, CIFAR-10 Batch 5:  loss 0.021657, train_accuracy 1, valid accuracy 0.7766
Epoch 184, CIFAR-10 Batch 1:  loss 0.038461, train_accuracy 0.975, valid accuracy 0.7896
Epoch 184, CIFAR-10 Batch 2:  loss 0.020536, train_accuracy 1, valid accuracy 0.7938
Epoch 184, CIFAR-10 Batch 3:  loss 0.019527, train_accuracy 1, valid accuracy 0.7732
Epoch 184, CIFAR-10 Batch 4:  loss 0.058002, train_accuracy 0.975, valid accuracy 0.776
Epoch 184, CIFAR-10 Batch 5:  loss 0.013577, train_accuracy 1, valid accuracy 0.7844
Epoch 185, CIFAR-10 Batch 1:  loss 0.034596, train_accuracy 1, valid accuracy 0.7832
Epoch 185, CIFAR-10 Batch 2:  loss 0.010848, train_accuracy 1, valid accuracy 0.7838
Epoch 185, CIFAR-10 Batch 3:  loss 0.020603, train_accuracy 1, valid accuracy 0.7798
Epoch 185, CIFAR-10 Batch 4:  loss 0.020685, train_accuracy 1, valid accuracy 0.791
Epoch 185, CIFAR-10 Batch 5:  loss 0.012392, train_accuracy 1, valid accuracy 0.781
Epoch 186, CIFAR-10 Batch 1:  loss 0.025899, train_accuracy 1, valid accuracy 0.792
Epoch 186, CIFAR-10 Batch 2:  loss 0.051514, train_accuracy 1, valid accuracy 0.7706
Epoch 186, CIFAR-10 Batch 3:  loss 0.028268, train_accuracy 1, valid accuracy 0.7894
Epoch 186, CIFAR-10 Batch 4:  loss 0.028234, train_accuracy 1, valid accuracy 0.7808
Epoch 186, CIFAR-10 Batch 5:  loss 0.016188, train_accuracy 1, valid accuracy 0.7844
Epoch 187, CIFAR-10 Batch 1:  loss 0.026025, train_accuracy 1, valid accuracy 0.7862
Epoch 187, CIFAR-10 Batch 2:  loss 0.023210, train_accuracy 1, valid accuracy 0.7648
Epoch 187, CIFAR-10 Batch 3:  loss 0.024193, train_accuracy 1, valid accuracy 0.762
Epoch 187, CIFAR-10 Batch 4:  loss 0.016580, train_accuracy 1, valid accuracy 0.7848
Epoch 187, CIFAR-10 Batch 5:  loss 0.019221, train_accuracy 1, valid accuracy 0.7662
Epoch 188, CIFAR-10 Batch 1:  loss 0.037037, train_accuracy 1, valid accuracy 0.7918
Epoch 188, CIFAR-10 Batch 2:  loss 0.016413, train_accuracy 1, valid accuracy 0.7736
Epoch 188, CIFAR-10 Batch 3:  loss 0.021869, train_accuracy 1, valid accuracy 0.788
Epoch 188, CIFAR-10 Batch 4:  loss 0.045145, train_accuracy 0.975, valid accuracy 0.7876
Epoch 188, CIFAR-10 Batch 5:  loss 0.009676, train_accuracy 1, valid accuracy 0.7862
Epoch 189, CIFAR-10 Batch 1:  loss 0.021704, train_accuracy 1, valid accuracy 0.7896
Epoch 189, CIFAR-10 Batch 2:  loss 0.016012, train_accuracy 1, valid accuracy 0.7872
Epoch 189, CIFAR-10 Batch 3:  loss 0.027715, train_accuracy 1, valid accuracy 0.7698
Epoch 189, CIFAR-10 Batch 4:  loss 0.031915, train_accuracy 1, valid accuracy 0.7928
Epoch 189, CIFAR-10 Batch 5:  loss 0.027615, train_accuracy 1, valid accuracy 0.7696
Epoch 190, CIFAR-10 Batch 1:  loss 0.023925, train_accuracy 1, valid accuracy 0.7784
Epoch 190, CIFAR-10 Batch 2:  loss 0.014687, train_accuracy 1, valid accuracy 0.78
Epoch 190, CIFAR-10 Batch 3:  loss 0.037377, train_accuracy 1, valid accuracy 0.7784
Epoch 190, CIFAR-10 Batch 4:  loss 0.042561, train_accuracy 0.975, valid accuracy 0.787
Epoch 190, CIFAR-10 Batch 5:  loss 0.028876, train_accuracy 1, valid accuracy 0.762
Epoch 191, CIFAR-10 Batch 1:  loss 0.025444, train_accuracy 1, valid accuracy 0.7782
Epoch 191, CIFAR-10 Batch 2:  loss 0.018452, train_accuracy 1, valid accuracy 0.784
Epoch 191, CIFAR-10 Batch 3:  loss 0.017686, train_accuracy 1, valid accuracy 0.7864
Epoch 191, CIFAR-10 Batch 4:  loss 0.019041, train_accuracy 1, valid accuracy 0.7946
Epoch 191, CIFAR-10 Batch 5:  loss 0.012802, train_accuracy 1, valid accuracy 0.7804
Epoch 192, CIFAR-10 Batch 1:  loss 0.015216, train_accuracy 1, valid accuracy 0.7904
Epoch 192, CIFAR-10 Batch 2:  loss 0.017638, train_accuracy 1, valid accuracy 0.7936
Epoch 192, CIFAR-10 Batch 3:  loss 0.026981, train_accuracy 1, valid accuracy 0.7772
Epoch 192, CIFAR-10 Batch 4:  loss 0.031692, train_accuracy 1, valid accuracy 0.7806
Epoch 192, CIFAR-10 Batch 5:  loss 0.033762, train_accuracy 1, valid accuracy 0.761
Epoch 193, CIFAR-10 Batch 1:  loss 0.037994, train_accuracy 1, valid accuracy 0.7684
Epoch 193, CIFAR-10 Batch 2:  loss 0.017489, train_accuracy 1, valid accuracy 0.7828
Epoch 193, CIFAR-10 Batch 3:  loss 0.028277, train_accuracy 1, valid accuracy 0.7858
Epoch 193, CIFAR-10 Batch 4:  loss 0.041885, train_accuracy 0.975, valid accuracy 0.7764
Epoch 193, CIFAR-10 Batch 5:  loss 0.019729, train_accuracy 1, valid accuracy 0.7764
Epoch 194, CIFAR-10 Batch 1:  loss 0.022182, train_accuracy 1, valid accuracy 0.796
Epoch 194, CIFAR-10 Batch 2:  loss 0.031619, train_accuracy 1, valid accuracy 0.7724
Epoch 194, CIFAR-10 Batch 3:  loss 0.020820, train_accuracy 1, valid accuracy 0.7796
Epoch 194, CIFAR-10 Batch 4:  loss 0.027712, train_accuracy 1, valid accuracy 0.7898
Epoch 194, CIFAR-10 Batch 5:  loss 0.017687, train_accuracy 1, valid accuracy 0.7748
Epoch 195, CIFAR-10 Batch 1:  loss 0.010922, train_accuracy 1, valid accuracy 0.8
Epoch 195, CIFAR-10 Batch 2:  loss 0.010532, train_accuracy 1, valid accuracy 0.7966
Epoch 195, CIFAR-10 Batch 3:  loss 0.030230, train_accuracy 1, valid accuracy 0.7858
Epoch 195, CIFAR-10 Batch 4:  loss 0.053313, train_accuracy 0.975, valid accuracy 0.7852
Epoch 195, CIFAR-10 Batch 5:  loss 0.011896, train_accuracy 1, valid accuracy 0.7794
Epoch 196, CIFAR-10 Batch 1:  loss 0.020325, train_accuracy 1, valid accuracy 0.7898
Epoch 196, CIFAR-10 Batch 2:  loss 0.011549, train_accuracy 1, valid accuracy 0.7786
Epoch 196, CIFAR-10 Batch 3:  loss 0.027888, train_accuracy 1, valid accuracy 0.786
Epoch 196, CIFAR-10 Batch 4:  loss 0.023946, train_accuracy 1, valid accuracy 0.7954
Epoch 196, CIFAR-10 Batch 5:  loss 0.016125, train_accuracy 1, valid accuracy 0.7798
Epoch 197, CIFAR-10 Batch 1:  loss 0.020430, train_accuracy 1, valid accuracy 0.7948
Epoch 197, CIFAR-10 Batch 2:  loss 0.014944, train_accuracy 1, valid accuracy 0.7892
Epoch 197, CIFAR-10 Batch 3:  loss 0.019596, train_accuracy 1, valid accuracy 0.7888
Epoch 197, CIFAR-10 Batch 4:  loss 0.021267, train_accuracy 1, valid accuracy 0.7938
Epoch 197, CIFAR-10 Batch 5:  loss 0.013708, train_accuracy 1, valid accuracy 0.785
Epoch 198, CIFAR-10 Batch 1:  loss 0.023915, train_accuracy 1, valid accuracy 0.7926
Epoch 198, CIFAR-10 Batch 2:  loss 0.009832, train_accuracy 1, valid accuracy 0.79
Epoch 198, CIFAR-10 Batch 3:  loss 0.017356, train_accuracy 1, valid accuracy 0.7858
Epoch 198, CIFAR-10 Batch 4:  loss 0.016048, train_accuracy 1, valid accuracy 0.7918
Epoch 198, CIFAR-10 Batch 5:  loss 0.013858, train_accuracy 1, valid accuracy 0.7884
Epoch 199, CIFAR-10 Batch 1:  loss 0.021398, train_accuracy 1, valid accuracy 0.7904
Epoch 199, CIFAR-10 Batch 2:  loss 0.008405, train_accuracy 1, valid accuracy 0.7892
Epoch 199, CIFAR-10 Batch 3:  loss 0.019073, train_accuracy 1, valid accuracy 0.78
Epoch 199, CIFAR-10 Batch 4:  loss 0.033494, train_accuracy 1, valid accuracy 0.7828
Epoch 199, CIFAR-10 Batch 5:  loss 0.009217, train_accuracy 1, valid accuracy 0.7978
Epoch 200, CIFAR-10 Batch 1:  loss 0.023047, train_accuracy 1, valid accuracy 0.7934
Epoch 200, CIFAR-10 Batch 2:  loss 0.006376, train_accuracy 1, valid accuracy 0.7942
Epoch 200, CIFAR-10 Batch 3:  loss 0.022377, train_accuracy 1, valid accuracy 0.7816
Epoch 200, CIFAR-10 Batch 4:  loss 0.011276, train_accuracy 1, valid accuracy 0.7946
Epoch 200, CIFAR-10 Batch 5:  loss 0.021505, train_accuracy 1, valid accuracy 0.7754
Epoch 201, CIFAR-10 Batch 1:  loss 0.017343, train_accuracy 1, valid accuracy 0.7956
Epoch 201, CIFAR-10 Batch 2:  loss 0.015956, train_accuracy 1, valid accuracy 0.7882
Epoch 201, CIFAR-10 Batch 3:  loss 0.037355, train_accuracy 1, valid accuracy 0.7564
Epoch 201, CIFAR-10 Batch 4:  loss 0.018445, train_accuracy 1, valid accuracy 0.7878
Epoch 201, CIFAR-10 Batch 5:  loss 0.014702, train_accuracy 1, valid accuracy 0.7908
Epoch 202, CIFAR-10 Batch 1:  loss 0.015507, train_accuracy 1, valid accuracy 0.79
Epoch 202, CIFAR-10 Batch 2:  loss 0.008785, train_accuracy 1, valid accuracy 0.7862
Epoch 202, CIFAR-10 Batch 3:  loss 0.024725, train_accuracy 1, valid accuracy 0.7634
Epoch 202, CIFAR-10 Batch 4:  loss 0.017002, train_accuracy 1, valid accuracy 0.7936
Epoch 202, CIFAR-10 Batch 5:  loss 0.016980, train_accuracy 1, valid accuracy 0.7772
Epoch 203, CIFAR-10 Batch 1:  loss 0.012666, train_accuracy 1, valid accuracy 0.791
Epoch 203, CIFAR-10 Batch 2:  loss 0.012610, train_accuracy 1, valid accuracy 0.7804
Epoch 203, CIFAR-10 Batch 3:  loss 0.018706, train_accuracy 1, valid accuracy 0.7784
Epoch 203, CIFAR-10 Batch 4:  loss 0.012311, train_accuracy 1, valid accuracy 0.797
Epoch 203, CIFAR-10 Batch 5:  loss 0.022424, train_accuracy 1, valid accuracy 0.796
Epoch 204, CIFAR-10 Batch 1:  loss 0.016712, train_accuracy 1, valid accuracy 0.8076
Epoch 204, CIFAR-10 Batch 2:  loss 0.036430, train_accuracy 1, valid accuracy 0.7688
Epoch 204, CIFAR-10 Batch 3:  loss 0.016928, train_accuracy 1, valid accuracy 0.7824
Epoch 204, CIFAR-10 Batch 4:  loss 0.019275, train_accuracy 1, valid accuracy 0.7818
Epoch 204, CIFAR-10 Batch 5:  loss 0.013168, train_accuracy 1, valid accuracy 0.7862
Epoch 205, CIFAR-10 Batch 1:  loss 0.012212, train_accuracy 1, valid accuracy 0.7986
Epoch 205, CIFAR-10 Batch 2:  loss 0.007805, train_accuracy 1, valid accuracy 0.7872
Epoch 205, CIFAR-10 Batch 3:  loss 0.011740, train_accuracy 1, valid accuracy 0.787
Epoch 205, CIFAR-10 Batch 4:  loss 0.014641, train_accuracy 1, valid accuracy 0.7886
Epoch 205, CIFAR-10 Batch 5:  loss 0.011587, train_accuracy 1, valid accuracy 0.79
Epoch 206, CIFAR-10 Batch 1:  loss 0.016605, train_accuracy 1, valid accuracy 0.7864
Epoch 206, CIFAR-10 Batch 2:  loss 0.007706, train_accuracy 1, valid accuracy 0.7968
Epoch 206, CIFAR-10 Batch 3:  loss 0.011368, train_accuracy 1, valid accuracy 0.786
Epoch 206, CIFAR-10 Batch 4:  loss 0.025801, train_accuracy 1, valid accuracy 0.7934
Epoch 206, CIFAR-10 Batch 5:  loss 0.015560, train_accuracy 1, valid accuracy 0.781
Epoch 207, CIFAR-10 Batch 1:  loss 0.017779, train_accuracy 1, valid accuracy 0.7794
Epoch 207, CIFAR-10 Batch 2:  loss 0.008305, train_accuracy 1, valid accuracy 0.785
Epoch 207, CIFAR-10 Batch 3:  loss 0.013588, train_accuracy 1, valid accuracy 0.7692
Epoch 207, CIFAR-10 Batch 4:  loss 0.030694, train_accuracy 1, valid accuracy 0.7844
Epoch 207, CIFAR-10 Batch 5:  loss 0.007974, train_accuracy 1, valid accuracy 0.7782
Epoch 208, CIFAR-10 Batch 1:  loss 0.018181, train_accuracy 1, valid accuracy 0.7962
Epoch 208, CIFAR-10 Batch 2:  loss 0.010095, train_accuracy 1, valid accuracy 0.7868
Epoch 208, CIFAR-10 Batch 3:  loss 0.013439, train_accuracy 1, valid accuracy 0.7596
Epoch 208, CIFAR-10 Batch 4:  loss 0.012736, train_accuracy 1, valid accuracy 0.7856
Epoch 208, CIFAR-10 Batch 5:  loss 0.009724, train_accuracy 1, valid accuracy 0.7916
Epoch 209, CIFAR-10 Batch 1:  loss 0.018473, train_accuracy 1, valid accuracy 0.7888
Epoch 209, CIFAR-10 Batch 2:  loss 0.017094, train_accuracy 1, valid accuracy 0.7776
Epoch 209, CIFAR-10 Batch 3:  loss 0.016940, train_accuracy 1, valid accuracy 0.7772
Epoch 209, CIFAR-10 Batch 4:  loss 0.013741, train_accuracy 1, valid accuracy 0.799
Epoch 209, CIFAR-10 Batch 5:  loss 0.019128, train_accuracy 1, valid accuracy 0.7732
Epoch 210, CIFAR-10 Batch 1:  loss 0.010671, train_accuracy 1, valid accuracy 0.793
Epoch 210, CIFAR-10 Batch 2:  loss 0.012899, train_accuracy 1, valid accuracy 0.7916
Epoch 210, CIFAR-10 Batch 3:  loss 0.020090, train_accuracy 1, valid accuracy 0.7768
Epoch 210, CIFAR-10 Batch 4:  loss 0.026389, train_accuracy 1, valid accuracy 0.7796
Epoch 210, CIFAR-10 Batch 5:  loss 0.009829, train_accuracy 1, valid accuracy 0.789
Epoch 211, CIFAR-10 Batch 1:  loss 0.010259, train_accuracy 1, valid accuracy 0.7982
Epoch 211, CIFAR-10 Batch 2:  loss 0.008709, train_accuracy 1, valid accuracy 0.7822
Epoch 211, CIFAR-10 Batch 3:  loss 0.015747, train_accuracy 1, valid accuracy 0.7794
Epoch 211, CIFAR-10 Batch 4:  loss 0.012051, train_accuracy 1, valid accuracy 0.7978
Epoch 211, CIFAR-10 Batch 5:  loss 0.010362, train_accuracy 1, valid accuracy 0.7946
Epoch 212, CIFAR-10 Batch 1:  loss 0.018809, train_accuracy 1, valid accuracy 0.794
Epoch 212, CIFAR-10 Batch 2:  loss 0.010567, train_accuracy 1, valid accuracy 0.7946
Epoch 212, CIFAR-10 Batch 3:  loss 0.008539, train_accuracy 1, valid accuracy 0.7888
Epoch 212, CIFAR-10 Batch 4:  loss 0.013698, train_accuracy 1, valid accuracy 0.7972
Epoch 212, CIFAR-10 Batch 5:  loss 0.036382, train_accuracy 1, valid accuracy 0.759
Epoch 213, CIFAR-10 Batch 1:  loss 0.013316, train_accuracy 1, valid accuracy 0.79
Epoch 213, CIFAR-10 Batch 2:  loss 0.017202, train_accuracy 1, valid accuracy 0.7914
Epoch 213, CIFAR-10 Batch 3:  loss 0.013302, train_accuracy 1, valid accuracy 0.7788
Epoch 213, CIFAR-10 Batch 4:  loss 0.012021, train_accuracy 1, valid accuracy 0.797
Epoch 213, CIFAR-10 Batch 5:  loss 0.025058, train_accuracy 1, valid accuracy 0.7734
Epoch 214, CIFAR-10 Batch 1:  loss 0.016635, train_accuracy 1, valid accuracy 0.7898
Epoch 214, CIFAR-10 Batch 2:  loss 0.017989, train_accuracy 1, valid accuracy 0.78
Epoch 214, CIFAR-10 Batch 3:  loss 0.014254, train_accuracy 1, valid accuracy 0.773
Epoch 214, CIFAR-10 Batch 4:  loss 0.014185, train_accuracy 1, valid accuracy 0.7994
Epoch 214, CIFAR-10 Batch 5:  loss 0.020100, train_accuracy 1, valid accuracy 0.759
Epoch 215, CIFAR-10 Batch 1:  loss 0.017472, train_accuracy 1, valid accuracy 0.7796
Epoch 215, CIFAR-10 Batch 2:  loss 0.008781, train_accuracy 1, valid accuracy 0.785
Epoch 215, CIFAR-10 Batch 3:  loss 0.009478, train_accuracy 1, valid accuracy 0.7822
Epoch 215, CIFAR-10 Batch 4:  loss 0.030528, train_accuracy 1, valid accuracy 0.781
Epoch 215, CIFAR-10 Batch 5:  loss 0.017242, train_accuracy 1, valid accuracy 0.7818
Epoch 216, CIFAR-10 Batch 1:  loss 0.016529, train_accuracy 1, valid accuracy 0.7894
Epoch 216, CIFAR-10 Batch 2:  loss 0.018276, train_accuracy 1, valid accuracy 0.7832
Epoch 216, CIFAR-10 Batch 3:  loss 0.008096, train_accuracy 1, valid accuracy 0.7836
Epoch 216, CIFAR-10 Batch 4:  loss 0.007004, train_accuracy 1, valid accuracy 0.7842
Epoch 216, CIFAR-10 Batch 5:  loss 0.015671, train_accuracy 1, valid accuracy 0.7968
Epoch 217, CIFAR-10 Batch 1:  loss 0.016233, train_accuracy 1, valid accuracy 0.7892
Epoch 217, CIFAR-10 Batch 2:  loss 0.007890, train_accuracy 1, valid accuracy 0.7848
Epoch 217, CIFAR-10 Batch 3:  loss 0.012293, train_accuracy 1, valid accuracy 0.796
Epoch 217, CIFAR-10 Batch 4:  loss 0.022928, train_accuracy 1, valid accuracy 0.7772
Epoch 217, CIFAR-10 Batch 5:  loss 0.011427, train_accuracy 1, valid accuracy 0.7866
Epoch 218, CIFAR-10 Batch 1:  loss 0.011958, train_accuracy 1, valid accuracy 0.7996
Epoch 218, CIFAR-10 Batch 2:  loss 0.006884, train_accuracy 1, valid accuracy 0.7812
Epoch 218, CIFAR-10 Batch 3:  loss 0.010904, train_accuracy 1, valid accuracy 0.7858
Epoch 218, CIFAR-10 Batch 4:  loss 0.012114, train_accuracy 1, valid accuracy 0.7886
Epoch 218, CIFAR-10 Batch 5:  loss 0.010252, train_accuracy 1, valid accuracy 0.7992
Epoch 219, CIFAR-10 Batch 1:  loss 0.009002, train_accuracy 1, valid accuracy 0.7978
Epoch 219, CIFAR-10 Batch 2:  loss 0.010806, train_accuracy 1, valid accuracy 0.7916
Epoch 219, CIFAR-10 Batch 3:  loss 0.010085, train_accuracy 1, valid accuracy 0.7712
Epoch 219, CIFAR-10 Batch 4:  loss 0.013100, train_accuracy 1, valid accuracy 0.7856
Epoch 219, CIFAR-10 Batch 5:  loss 0.011592, train_accuracy 1, valid accuracy 0.788
Epoch 220, CIFAR-10 Batch 1:  loss 0.013366, train_accuracy 1, valid accuracy 0.802
Epoch 220, CIFAR-10 Batch 2:  loss 0.007778, train_accuracy 1, valid accuracy 0.7898
Epoch 220, CIFAR-10 Batch 3:  loss 0.022938, train_accuracy 1, valid accuracy 0.7834
Epoch 220, CIFAR-10 Batch 4:  loss 0.008887, train_accuracy 1, valid accuracy 0.8026
Epoch 220, CIFAR-10 Batch 5:  loss 0.007260, train_accuracy 1, valid accuracy 0.7964
Epoch 221, CIFAR-10 Batch 1:  loss 0.012810, train_accuracy 1, valid accuracy 0.7952
Epoch 221, CIFAR-10 Batch 2:  loss 0.009804, train_accuracy 1, valid accuracy 0.7788
Epoch 221, CIFAR-10 Batch 3:  loss 0.009673, train_accuracy 1, valid accuracy 0.793
Epoch 221, CIFAR-10 Batch 4:  loss 0.010440, train_accuracy 1, valid accuracy 0.7964
Epoch 221, CIFAR-10 Batch 5:  loss 0.008504, train_accuracy 1, valid accuracy 0.7874
Epoch 222, CIFAR-10 Batch 1:  loss 0.012226, train_accuracy 1, valid accuracy 0.7998
Epoch 222, CIFAR-10 Batch 2:  loss 0.004415, train_accuracy 1, valid accuracy 0.7926
Epoch 222, CIFAR-10 Batch 3:  loss 0.013573, train_accuracy 1, valid accuracy 0.7772
Epoch 222, CIFAR-10 Batch 4:  loss 0.014033, train_accuracy 1, valid accuracy 0.7906
Epoch 222, CIFAR-10 Batch 5:  loss 0.013950, train_accuracy 1, valid accuracy 0.7872
Epoch 223, CIFAR-10 Batch 1:  loss 0.013890, train_accuracy 1, valid accuracy 0.8022
Epoch 223, CIFAR-10 Batch 2:  loss 0.008250, train_accuracy 1, valid accuracy 0.7904
Epoch 223, CIFAR-10 Batch 3:  loss 0.008443, train_accuracy 1, valid accuracy 0.7864
Epoch 223, CIFAR-10 Batch 4:  loss 0.013319, train_accuracy 1, valid accuracy 0.7946
Epoch 223, CIFAR-10 Batch 5:  loss 0.019221, train_accuracy 1, valid accuracy 0.7862
Epoch 224, CIFAR-10 Batch 1:  loss 0.011528, train_accuracy 1, valid accuracy 0.7914
Epoch 224, CIFAR-10 Batch 2:  loss 0.006423, train_accuracy 1, valid accuracy 0.789
Epoch 224, CIFAR-10 Batch 3:  loss 0.007512, train_accuracy 1, valid accuracy 0.785
Epoch 224, CIFAR-10 Batch 4:  loss 0.023307, train_accuracy 1, valid accuracy 0.7936
Epoch 224, CIFAR-10 Batch 5:  loss 0.030842, train_accuracy 1, valid accuracy 0.7814
Epoch 225, CIFAR-10 Batch 1:  loss 0.011852, train_accuracy 1, valid accuracy 0.7966
Epoch 225, CIFAR-10 Batch 2:  loss 0.005585, train_accuracy 1, valid accuracy 0.7924
Epoch 225, CIFAR-10 Batch 3:  loss 0.006353, train_accuracy 1, valid accuracy 0.7812
Epoch 225, CIFAR-10 Batch 4:  loss 0.039229, train_accuracy 0.975, valid accuracy 0.7842
Epoch 225, CIFAR-10 Batch 5:  loss 0.006760, train_accuracy 1, valid accuracy 0.7976
Epoch 226, CIFAR-10 Batch 1:  loss 0.011248, train_accuracy 1, valid accuracy 0.791
Epoch 226, CIFAR-10 Batch 2:  loss 0.005107, train_accuracy 1, valid accuracy 0.791
Epoch 226, CIFAR-10 Batch 3:  loss 0.015669, train_accuracy 1, valid accuracy 0.7848
Epoch 226, CIFAR-10 Batch 4:  loss 0.008444, train_accuracy 1, valid accuracy 0.7896
Epoch 226, CIFAR-10 Batch 5:  loss 0.014511, train_accuracy 1, valid accuracy 0.782
Epoch 227, CIFAR-10 Batch 1:  loss 0.015052, train_accuracy 1, valid accuracy 0.799
Epoch 227, CIFAR-10 Batch 2:  loss 0.005982, train_accuracy 1, valid accuracy 0.7842
Epoch 227, CIFAR-10 Batch 3:  loss 0.010661, train_accuracy 1, valid accuracy 0.7848
Epoch 227, CIFAR-10 Batch 4:  loss 0.011114, train_accuracy 1, valid accuracy 0.7828
Epoch 227, CIFAR-10 Batch 5:  loss 0.007318, train_accuracy 1, valid accuracy 0.7944
Epoch 228, CIFAR-10 Batch 1:  loss 0.012025, train_accuracy 1, valid accuracy 0.7902
Epoch 228, CIFAR-10 Batch 2:  loss 0.013757, train_accuracy 1, valid accuracy 0.7896
Epoch 228, CIFAR-10 Batch 3:  loss 0.013932, train_accuracy 1, valid accuracy 0.7708
Epoch 228, CIFAR-10 Batch 4:  loss 0.009094, train_accuracy 1, valid accuracy 0.796
Epoch 228, CIFAR-10 Batch 5:  loss 0.017097, train_accuracy 1, valid accuracy 0.7936
Epoch 229, CIFAR-10 Batch 1:  loss 0.010690, train_accuracy 1, valid accuracy 0.799
Epoch 229, CIFAR-10 Batch 2:  loss 0.006436, train_accuracy 1, valid accuracy 0.7972
Epoch 229, CIFAR-10 Batch 3:  loss 0.005228, train_accuracy 1, valid accuracy 0.7864
Epoch 229, CIFAR-10 Batch 4:  loss 0.008009, train_accuracy 1, valid accuracy 0.7882
Epoch 229, CIFAR-10 Batch 5:  loss 0.018767, train_accuracy 1, valid accuracy 0.7592
Epoch 230, CIFAR-10 Batch 1:  loss 0.019588, train_accuracy 1, valid accuracy 0.788
Epoch 230, CIFAR-10 Batch 2:  loss 0.005067, train_accuracy 1, valid accuracy 0.7972
Epoch 230, CIFAR-10 Batch 3:  loss 0.012734, train_accuracy 1, valid accuracy 0.7824
Epoch 230, CIFAR-10 Batch 4:  loss 0.009848, train_accuracy 1, valid accuracy 0.79
Epoch 230, CIFAR-10 Batch 5:  loss 0.006088, train_accuracy 1, valid accuracy 0.795
Epoch 231, CIFAR-10 Batch 1:  loss 0.013645, train_accuracy 1, valid accuracy 0.797
Epoch 231, CIFAR-10 Batch 2:  loss 0.008911, train_accuracy 1, valid accuracy 0.7734
Epoch 231, CIFAR-10 Batch 3:  loss 0.013898, train_accuracy 1, valid accuracy 0.7908
Epoch 231, CIFAR-10 Batch 4:  loss 0.006219, train_accuracy 1, valid accuracy 0.799
Epoch 231, CIFAR-10 Batch 5:  loss 0.009377, train_accuracy 1, valid accuracy 0.7906
Epoch 232, CIFAR-10 Batch 1:  loss 0.012435, train_accuracy 1, valid accuracy 0.7902
Epoch 232, CIFAR-10 Batch 2:  loss 0.006399, train_accuracy 1, valid accuracy 0.7952
Epoch 232, CIFAR-10 Batch 3:  loss 0.005835, train_accuracy 1, valid accuracy 0.7916
Epoch 232, CIFAR-10 Batch 4:  loss 0.007505, train_accuracy 1, valid accuracy 0.8004
Epoch 232, CIFAR-10 Batch 5:  loss 0.016445, train_accuracy 1, valid accuracy 0.7762
Epoch 233, CIFAR-10 Batch 1:  loss 0.015440, train_accuracy 1, valid accuracy 0.7856
Epoch 233, CIFAR-10 Batch 2:  loss 0.004423, train_accuracy 1, valid accuracy 0.8056
Epoch 233, CIFAR-10 Batch 3:  loss 0.016805, train_accuracy 1, valid accuracy 0.7832
Epoch 233, CIFAR-10 Batch 4:  loss 0.008069, train_accuracy 1, valid accuracy 0.7954
Epoch 233, CIFAR-10 Batch 5:  loss 0.007734, train_accuracy 1, valid accuracy 0.7972
Epoch 234, CIFAR-10 Batch 1:  loss 0.008561, train_accuracy 1, valid accuracy 0.8022
Epoch 234, CIFAR-10 Batch 2:  loss 0.004427, train_accuracy 1, valid accuracy 0.8048
Epoch 234, CIFAR-10 Batch 3:  loss 0.009219, train_accuracy 1, valid accuracy 0.795
Epoch 234, CIFAR-10 Batch 4:  loss 0.005873, train_accuracy 1, valid accuracy 0.7984
Epoch 234, CIFAR-10 Batch 5:  loss 0.008918, train_accuracy 1, valid accuracy 0.7996
Epoch 235, CIFAR-10 Batch 1:  loss 0.009762, train_accuracy 1, valid accuracy 0.8002
Epoch 235, CIFAR-10 Batch 2:  loss 0.004219, train_accuracy 1, valid accuracy 0.7948
Epoch 235, CIFAR-10 Batch 3:  loss 0.012786, train_accuracy 1, valid accuracy 0.786
Epoch 235, CIFAR-10 Batch 4:  loss 0.007831, train_accuracy 1, valid accuracy 0.7956
Epoch 235, CIFAR-10 Batch 5:  loss 0.012944, train_accuracy 1, valid accuracy 0.7886
Epoch 236, CIFAR-10 Batch 1:  loss 0.007875, train_accuracy 1, valid accuracy 0.8008
Epoch 236, CIFAR-10 Batch 2:  loss 0.003279, train_accuracy 1, valid accuracy 0.7902
Epoch 236, CIFAR-10 Batch 3:  loss 0.017131, train_accuracy 1, valid accuracy 0.7844
Epoch 236, CIFAR-10 Batch 4:  loss 0.007557, train_accuracy 1, valid accuracy 0.8034
Epoch 236, CIFAR-10 Batch 5:  loss 0.008342, train_accuracy 1, valid accuracy 0.7904
Epoch 237, CIFAR-10 Batch 1:  loss 0.011877, train_accuracy 1, valid accuracy 0.8046
Epoch 237, CIFAR-10 Batch 2:  loss 0.005485, train_accuracy 1, valid accuracy 0.7888
Epoch 237, CIFAR-10 Batch 3:  loss 0.007856, train_accuracy 1, valid accuracy 0.7864
Epoch 237, CIFAR-10 Batch 4:  loss 0.004300, train_accuracy 1, valid accuracy 0.7894
Epoch 237, CIFAR-10 Batch 5:  loss 0.006127, train_accuracy 1, valid accuracy 0.7924
Epoch 238, CIFAR-10 Batch 1:  loss 0.009051, train_accuracy 1, valid accuracy 0.7922
Epoch 238, CIFAR-10 Batch 2:  loss 0.004912, train_accuracy 1, valid accuracy 0.7932
Epoch 238, CIFAR-10 Batch 3:  loss 0.008693, train_accuracy 1, valid accuracy 0.7884
Epoch 238, CIFAR-10 Batch 4:  loss 0.005811, train_accuracy 1, valid accuracy 0.7998
Epoch 238, CIFAR-10 Batch 5:  loss 0.004488, train_accuracy 1, valid accuracy 0.7984
Epoch 239, CIFAR-10 Batch 1:  loss 0.005831, train_accuracy 1, valid accuracy 0.796
Epoch 239, CIFAR-10 Batch 2:  loss 0.009850, train_accuracy 1, valid accuracy 0.7926
Epoch 239, CIFAR-10 Batch 3:  loss 0.014095, train_accuracy 1, valid accuracy 0.7866
Epoch 239, CIFAR-10 Batch 4:  loss 0.006683, train_accuracy 1, valid accuracy 0.796
Epoch 239, CIFAR-10 Batch 5:  loss 0.009608, train_accuracy 1, valid accuracy 0.7966
Epoch 240, CIFAR-10 Batch 1:  loss 0.002836, train_accuracy 1, valid accuracy 0.801
Epoch 240, CIFAR-10 Batch 2:  loss 0.003680, train_accuracy 1, valid accuracy 0.8008
Epoch 240, CIFAR-10 Batch 3:  loss 0.013345, train_accuracy 1, valid accuracy 0.7896
Epoch 240, CIFAR-10 Batch 4:  loss 0.006605, train_accuracy 1, valid accuracy 0.7944
Epoch 240, CIFAR-10 Batch 5:  loss 0.007937, train_accuracy 1, valid accuracy 0.7906
Epoch 241, CIFAR-10 Batch 1:  loss 0.007459, train_accuracy 1, valid accuracy 0.7976
Epoch 241, CIFAR-10 Batch 2:  loss 0.005688, train_accuracy 1, valid accuracy 0.8008
Epoch 241, CIFAR-10 Batch 3:  loss 0.007846, train_accuracy 1, valid accuracy 0.7828
Epoch 241, CIFAR-10 Batch 4:  loss 0.007755, train_accuracy 1, valid accuracy 0.7944
Epoch 241, CIFAR-10 Batch 5:  loss 0.025264, train_accuracy 1, valid accuracy 0.7696
Epoch 242, CIFAR-10 Batch 1:  loss 0.008215, train_accuracy 1, valid accuracy 0.7936
Epoch 242, CIFAR-10 Batch 2:  loss 0.003010, train_accuracy 1, valid accuracy 0.7968
Epoch 242, CIFAR-10 Batch 3:  loss 0.013690, train_accuracy 1, valid accuracy 0.7866
Epoch 242, CIFAR-10 Batch 4:  loss 0.003534, train_accuracy 1, valid accuracy 0.8
Epoch 242, CIFAR-10 Batch 5:  loss 0.017567, train_accuracy 1, valid accuracy 0.7822
Epoch 243, CIFAR-10 Batch 1:  loss 0.005637, train_accuracy 1, valid accuracy 0.7982
Epoch 243, CIFAR-10 Batch 2:  loss 0.018143, train_accuracy 1, valid accuracy 0.7926
Epoch 243, CIFAR-10 Batch 3:  loss 0.010276, train_accuracy 1, valid accuracy 0.7916
Epoch 243, CIFAR-10 Batch 4:  loss 0.002960, train_accuracy 1, valid accuracy 0.7996
Epoch 243, CIFAR-10 Batch 5:  loss 0.006071, train_accuracy 1, valid accuracy 0.7992
Epoch 244, CIFAR-10 Batch 1:  loss 0.003633, train_accuracy 1, valid accuracy 0.798
Epoch 244, CIFAR-10 Batch 2:  loss 0.002820, train_accuracy 1, valid accuracy 0.7932
Epoch 244, CIFAR-10 Batch 3:  loss 0.006547, train_accuracy 1, valid accuracy 0.798
Epoch 244, CIFAR-10 Batch 4:  loss 0.003222, train_accuracy 1, valid accuracy 0.7858
Epoch 244, CIFAR-10 Batch 5:  loss 0.004037, train_accuracy 1, valid accuracy 0.7968
Epoch 245, CIFAR-10 Batch 1:  loss 0.007323, train_accuracy 1, valid accuracy 0.799
Epoch 245, CIFAR-10 Batch 2:  loss 0.004845, train_accuracy 1, valid accuracy 0.7984
Epoch 245, CIFAR-10 Batch 3:  loss 0.010048, train_accuracy 1, valid accuracy 0.7768
Epoch 245, CIFAR-10 Batch 4:  loss 0.004402, train_accuracy 1, valid accuracy 0.8016
Epoch 245, CIFAR-10 Batch 5:  loss 0.005615, train_accuracy 1, valid accuracy 0.7862
Epoch 246, CIFAR-10 Batch 1:  loss 0.003429, train_accuracy 1, valid accuracy 0.7964
Epoch 246, CIFAR-10 Batch 2:  loss 0.005633, train_accuracy 1, valid accuracy 0.7988
Epoch 246, CIFAR-10 Batch 3:  loss 0.007697, train_accuracy 1, valid accuracy 0.796
Epoch 246, CIFAR-10 Batch 4:  loss 0.006220, train_accuracy 1, valid accuracy 0.7906
Epoch 246, CIFAR-10 Batch 5:  loss 0.004322, train_accuracy 1, valid accuracy 0.7784
Epoch 247, CIFAR-10 Batch 1:  loss 0.006324, train_accuracy 1, valid accuracy 0.7944
Epoch 247, CIFAR-10 Batch 2:  loss 0.002239, train_accuracy 1, valid accuracy 0.805
Epoch 247, CIFAR-10 Batch 3:  loss 0.010528, train_accuracy 1, valid accuracy 0.807
Epoch 247, CIFAR-10 Batch 4:  loss 0.008308, train_accuracy 1, valid accuracy 0.8046
Epoch 247, CIFAR-10 Batch 5:  loss 0.005701, train_accuracy 1, valid accuracy 0.8034
Epoch 248, CIFAR-10 Batch 1:  loss 0.011584, train_accuracy 1, valid accuracy 0.7894
Epoch 248, CIFAR-10 Batch 2:  loss 0.007129, train_accuracy 1, valid accuracy 0.7926
Epoch 248, CIFAR-10 Batch 3:  loss 0.010261, train_accuracy 1, valid accuracy 0.798
Epoch 248, CIFAR-10 Batch 4:  loss 0.002781, train_accuracy 1, valid accuracy 0.8014
Epoch 248, CIFAR-10 Batch 5:  loss 0.007034, train_accuracy 1, valid accuracy 0.7998
Epoch 249, CIFAR-10 Batch 1:  loss 0.003895, train_accuracy 1, valid accuracy 0.8076
Epoch 249, CIFAR-10 Batch 2:  loss 0.009963, train_accuracy 1, valid accuracy 0.7944
Epoch 249, CIFAR-10 Batch 3:  loss 0.008382, train_accuracy 1, valid accuracy 0.8012
Epoch 249, CIFAR-10 Batch 4:  loss 0.005027, train_accuracy 1, valid accuracy 0.7986
Epoch 249, CIFAR-10 Batch 5:  loss 0.002859, train_accuracy 1, valid accuracy 0.8024
Epoch 250, CIFAR-10 Batch 1:  loss 0.003276, train_accuracy 1, valid accuracy 0.7972
Epoch 250, CIFAR-10 Batch 2:  loss 0.005267, train_accuracy 1, valid accuracy 0.7962
Epoch 250, CIFAR-10 Batch 3:  loss 0.006534, train_accuracy 1, valid accuracy 0.7756
Epoch 250, CIFAR-10 Batch 4:  loss 0.003019, train_accuracy 1, valid accuracy 0.799
Epoch 250, CIFAR-10 Batch 5:  loss 0.003970, train_accuracy 1, valid accuracy 0.7932
Epoch 251, CIFAR-10 Batch 1:  loss 0.004527, train_accuracy 1, valid accuracy 0.8068
Epoch 251, CIFAR-10 Batch 2:  loss 0.002281, train_accuracy 1, valid accuracy 0.7946
Epoch 251, CIFAR-10 Batch 3:  loss 0.007804, train_accuracy 1, valid accuracy 0.7916
Epoch 251, CIFAR-10 Batch 4:  loss 0.011874, train_accuracy 1, valid accuracy 0.7986
Epoch 251, CIFAR-10 Batch 5:  loss 0.015506, train_accuracy 1, valid accuracy 0.781
Epoch 252, CIFAR-10 Batch 1:  loss 0.005943, train_accuracy 1, valid accuracy 0.7932
Epoch 252, CIFAR-10 Batch 2:  loss 0.008019, train_accuracy 1, valid accuracy 0.7934
Epoch 252, CIFAR-10 Batch 3:  loss 0.007399, train_accuracy 1, valid accuracy 0.7854
Epoch 252, CIFAR-10 Batch 4:  loss 0.002959, train_accuracy 1, valid accuracy 0.7926
Epoch 252, CIFAR-10 Batch 5:  loss 0.022994, train_accuracy 1, valid accuracy 0.7744
Epoch 253, CIFAR-10 Batch 1:  loss 0.005579, train_accuracy 1, valid accuracy 0.7976
Epoch 253, CIFAR-10 Batch 2:  loss 0.006365, train_accuracy 1, valid accuracy 0.7954
Epoch 253, CIFAR-10 Batch 3:  loss 0.006480, train_accuracy 1, valid accuracy 0.7904
Epoch 253, CIFAR-10 Batch 4:  loss 0.005236, train_accuracy 1, valid accuracy 0.7984
Epoch 253, CIFAR-10 Batch 5:  loss 0.010050, train_accuracy 1, valid accuracy 0.7856
Epoch 254, CIFAR-10 Batch 1:  loss 0.006715, train_accuracy 1, valid accuracy 0.7978
Epoch 254, CIFAR-10 Batch 2:  loss 0.005588, train_accuracy 1, valid accuracy 0.7934
Epoch 254, CIFAR-10 Batch 3:  loss 0.004706, train_accuracy 1, valid accuracy 0.7868
Epoch 254, CIFAR-10 Batch 4:  loss 0.006711, train_accuracy 1, valid accuracy 0.7914
Epoch 254, CIFAR-10 Batch 5:  loss 0.007918, train_accuracy 1, valid accuracy 0.7916
Epoch 255, CIFAR-10 Batch 1:  loss 0.004618, train_accuracy 1, valid accuracy 0.791
Epoch 255, CIFAR-10 Batch 2:  loss 0.302290, train_accuracy 0.95, valid accuracy 0.6868
Epoch 255, CIFAR-10 Batch 3:  loss 0.014686, train_accuracy 1, valid accuracy 0.7924
Epoch 255, CIFAR-10 Batch 4:  loss 0.018079, train_accuracy 1, valid accuracy 0.783
Epoch 255, CIFAR-10 Batch 5:  loss 0.019938, train_accuracy 1, valid accuracy 0.776
Epoch 256, CIFAR-10 Batch 1:  loss 0.004401, train_accuracy 1, valid accuracy 0.7922
Epoch 256, CIFAR-10 Batch 2:  loss 0.004885, train_accuracy 1, valid accuracy 0.799
Epoch 256, CIFAR-10 Batch 3:  loss 0.005276, train_accuracy 1, valid accuracy 0.7912
Epoch 256, CIFAR-10 Batch 4:  loss 0.001968, train_accuracy 1, valid accuracy 0.801
Epoch 256, CIFAR-10 Batch 5:  loss 0.013652, train_accuracy 1, valid accuracy 0.7836
Epoch 257, CIFAR-10 Batch 1:  loss 0.002710, train_accuracy 1, valid accuracy 0.7982
Epoch 257, CIFAR-10 Batch 2:  loss 0.003147, train_accuracy 1, valid accuracy 0.7908
Epoch 257, CIFAR-10 Batch 3:  loss 0.005942, train_accuracy 1, valid accuracy 0.7862
Epoch 257, CIFAR-10 Batch 4:  loss 0.004083, train_accuracy 1, valid accuracy 0.7922
Epoch 257, CIFAR-10 Batch 5:  loss 0.009339, train_accuracy 1, valid accuracy 0.7926
Epoch 258, CIFAR-10 Batch 1:  loss 0.002461, train_accuracy 1, valid accuracy 0.7914
Epoch 258, CIFAR-10 Batch 2:  loss 0.002954, train_accuracy 1, valid accuracy 0.7996
Epoch 258, CIFAR-10 Batch 3:  loss 0.007255, train_accuracy 1, valid accuracy 0.7928
Epoch 258, CIFAR-10 Batch 4:  loss 0.006713, train_accuracy 1, valid accuracy 0.788
Epoch 258, CIFAR-10 Batch 5:  loss 0.004564, train_accuracy 1, valid accuracy 0.7966
Epoch 259, CIFAR-10 Batch 1:  loss 0.004381, train_accuracy 1, valid accuracy 0.8032
Epoch 259, CIFAR-10 Batch 2:  loss 0.003492, train_accuracy 1, valid accuracy 0.8016
Epoch 259, CIFAR-10 Batch 3:  loss 0.012967, train_accuracy 1, valid accuracy 0.7866
Epoch 259, CIFAR-10 Batch 4:  loss 0.005202, train_accuracy 1, valid accuracy 0.7988
Epoch 259, CIFAR-10 Batch 5:  loss 0.010670, train_accuracy 1, valid accuracy 0.7916
Epoch 260, CIFAR-10 Batch 1:  loss 0.002863, train_accuracy 1, valid accuracy 0.795
Epoch 260, CIFAR-10 Batch 2:  loss 0.003065, train_accuracy 1, valid accuracy 0.7968
Epoch 260, CIFAR-10 Batch 3:  loss 0.006599, train_accuracy 1, valid accuracy 0.7854
Epoch 260, CIFAR-10 Batch 4:  loss 0.005129, train_accuracy 1, valid accuracy 0.796
Epoch 260, CIFAR-10 Batch 5:  loss 0.007810, train_accuracy 1, valid accuracy 0.799
Epoch 261, CIFAR-10 Batch 1:  loss 0.001754, train_accuracy 1, valid accuracy 0.803
Epoch 261, CIFAR-10 Batch 2:  loss 0.002262, train_accuracy 1, valid accuracy 0.81
Epoch 261, CIFAR-10 Batch 3:  loss 0.009517, train_accuracy 1, valid accuracy 0.7944
Epoch 261, CIFAR-10 Batch 4:  loss 0.004899, train_accuracy 1, valid accuracy 0.8018
Epoch 261, CIFAR-10 Batch 5:  loss 0.003041, train_accuracy 1, valid accuracy 0.8016
Epoch 262, CIFAR-10 Batch 1:  loss 0.003549, train_accuracy 1, valid accuracy 0.7998
Epoch 262, CIFAR-10 Batch 2:  loss 0.003627, train_accuracy 1, valid accuracy 0.7958
Epoch 262, CIFAR-10 Batch 3:  loss 0.008949, train_accuracy 1, valid accuracy 0.7962
Epoch 262, CIFAR-10 Batch 4:  loss 0.007378, train_accuracy 1, valid accuracy 0.7924
Epoch 262, CIFAR-10 Batch 5:  loss 0.008971, train_accuracy 1, valid accuracy 0.7812
Epoch 263, CIFAR-10 Batch 1:  loss 0.004232, train_accuracy 1, valid accuracy 0.7924
Epoch 263, CIFAR-10 Batch 2:  loss 0.003940, train_accuracy 1, valid accuracy 0.7958
Epoch 263, CIFAR-10 Batch 3:  loss 0.006807, train_accuracy 1, valid accuracy 0.7916
Epoch 263, CIFAR-10 Batch 4:  loss 0.015491, train_accuracy 1, valid accuracy 0.7956
Epoch 263, CIFAR-10 Batch 5:  loss 0.003343, train_accuracy 1, valid accuracy 0.7958
Epoch 264, CIFAR-10 Batch 1:  loss 0.003680, train_accuracy 1, valid accuracy 0.791
Epoch 264, CIFAR-10 Batch 2:  loss 0.006757, train_accuracy 1, valid accuracy 0.7912
Epoch 264, CIFAR-10 Batch 3:  loss 0.014176, train_accuracy 1, valid accuracy 0.7928
Epoch 264, CIFAR-10 Batch 4:  loss 0.009943, train_accuracy 1, valid accuracy 0.8012
Epoch 264, CIFAR-10 Batch 5:  loss 0.006472, train_accuracy 1, valid accuracy 0.7864
Epoch 265, CIFAR-10 Batch 1:  loss 0.004852, train_accuracy 1, valid accuracy 0.7986
Epoch 265, CIFAR-10 Batch 2:  loss 0.003113, train_accuracy 1, valid accuracy 0.7904
Epoch 265, CIFAR-10 Batch 3:  loss 0.013226, train_accuracy 1, valid accuracy 0.7952
Epoch 265, CIFAR-10 Batch 4:  loss 0.007391, train_accuracy 1, valid accuracy 0.7918
Epoch 265, CIFAR-10 Batch 5:  loss 0.002689, train_accuracy 1, valid accuracy 0.7922
Epoch 266, CIFAR-10 Batch 1:  loss 0.002954, train_accuracy 1, valid accuracy 0.7976
Epoch 266, CIFAR-10 Batch 2:  loss 0.004246, train_accuracy 1, valid accuracy 0.7988
Epoch 266, CIFAR-10 Batch 3:  loss 0.008772, train_accuracy 1, valid accuracy 0.7912
Epoch 266, CIFAR-10 Batch 4:  loss 0.012804, train_accuracy 1, valid accuracy 0.7848
Epoch 266, CIFAR-10 Batch 5:  loss 0.001736, train_accuracy 1, valid accuracy 0.7966
Epoch 267, CIFAR-10 Batch 1:  loss 0.002005, train_accuracy 1, valid accuracy 0.7988
Epoch 267, CIFAR-10 Batch 2:  loss 0.006315, train_accuracy 1, valid accuracy 0.8008
Epoch 267, CIFAR-10 Batch 3:  loss 0.002961, train_accuracy 1, valid accuracy 0.7998
Epoch 267, CIFAR-10 Batch 4:  loss 0.004228, train_accuracy 1, valid accuracy 0.7996
Epoch 267, CIFAR-10 Batch 5:  loss 0.002868, train_accuracy 1, valid accuracy 0.7934
Epoch 268, CIFAR-10 Batch 1:  loss 0.006895, train_accuracy 1, valid accuracy 0.8002
Epoch 268, CIFAR-10 Batch 2:  loss 0.005426, train_accuracy 1, valid accuracy 0.7782
Epoch 268, CIFAR-10 Batch 3:  loss 0.004754, train_accuracy 1, valid accuracy 0.797
Epoch 268, CIFAR-10 Batch 4:  loss 0.002349, train_accuracy 1, valid accuracy 0.797
Epoch 268, CIFAR-10 Batch 5:  loss 0.007663, train_accuracy 1, valid accuracy 0.7798
Epoch 269, CIFAR-10 Batch 1:  loss 0.007454, train_accuracy 1, valid accuracy 0.7992
Epoch 269, CIFAR-10 Batch 2:  loss 0.002827, train_accuracy 1, valid accuracy 0.7948
Epoch 269, CIFAR-10 Batch 3:  loss 0.009348, train_accuracy 1, valid accuracy 0.7868
Epoch 269, CIFAR-10 Batch 4:  loss 0.005343, train_accuracy 1, valid accuracy 0.7912
Epoch 269, CIFAR-10 Batch 5:  loss 0.001812, train_accuracy 1, valid accuracy 0.8012
Epoch 270, CIFAR-10 Batch 1:  loss 0.004295, train_accuracy 1, valid accuracy 0.7976
Epoch 270, CIFAR-10 Batch 2:  loss 0.005398, train_accuracy 1, valid accuracy 0.7918
Epoch 270, CIFAR-10 Batch 3:  loss 0.002119, train_accuracy 1, valid accuracy 0.7892
Epoch 270, CIFAR-10 Batch 4:  loss 0.004449, train_accuracy 1, valid accuracy 0.7992
Epoch 270, CIFAR-10 Batch 5:  loss 0.004162, train_accuracy 1, valid accuracy 0.7952
Epoch 271, CIFAR-10 Batch 1:  loss 0.006403, train_accuracy 1, valid accuracy 0.795
Epoch 271, CIFAR-10 Batch 2:  loss 0.004564, train_accuracy 1, valid accuracy 0.791
Epoch 271, CIFAR-10 Batch 3:  loss 0.005282, train_accuracy 1, valid accuracy 0.7942
Epoch 271, CIFAR-10 Batch 4:  loss 0.003873, train_accuracy 1, valid accuracy 0.7988
Epoch 271, CIFAR-10 Batch 5:  loss 0.003321, train_accuracy 1, valid accuracy 0.8016
Epoch 272, CIFAR-10 Batch 1:  loss 0.006888, train_accuracy 1, valid accuracy 0.7774
Epoch 272, CIFAR-10 Batch 2:  loss 0.004139, train_accuracy 1, valid accuracy 0.7962
Epoch 272, CIFAR-10 Batch 3:  loss 0.011524, train_accuracy 1, valid accuracy 0.7788
Epoch 272, CIFAR-10 Batch 4:  loss 0.005966, train_accuracy 1, valid accuracy 0.7884
Epoch 272, CIFAR-10 Batch 5:  loss 0.003953, train_accuracy 1, valid accuracy 0.787
Epoch 273, CIFAR-10 Batch 1:  loss 0.010871, train_accuracy 1, valid accuracy 0.7918
Epoch 273, CIFAR-10 Batch 2:  loss 0.002867, train_accuracy 1, valid accuracy 0.8072
Epoch 273, CIFAR-10 Batch 3:  loss 0.002937, train_accuracy 1, valid accuracy 0.791
Epoch 273, CIFAR-10 Batch 4:  loss 0.005027, train_accuracy 1, valid accuracy 0.7894
Epoch 273, CIFAR-10 Batch 5:  loss 0.004018, train_accuracy 1, valid accuracy 0.7866
Epoch 274, CIFAR-10 Batch 1:  loss 0.009636, train_accuracy 1, valid accuracy 0.7936
Epoch 274, CIFAR-10 Batch 2:  loss 0.010037, train_accuracy 1, valid accuracy 0.7922
Epoch 274, CIFAR-10 Batch 3:  loss 0.004246, train_accuracy 1, valid accuracy 0.7964
Epoch 274, CIFAR-10 Batch 4:  loss 0.007105, train_accuracy 1, valid accuracy 0.7968
Epoch 274, CIFAR-10 Batch 5:  loss 0.001970, train_accuracy 1, valid accuracy 0.7978
Epoch 275, CIFAR-10 Batch 1:  loss 0.003770, train_accuracy 1, valid accuracy 0.8014
Epoch 275, CIFAR-10 Batch 2:  loss 0.002059, train_accuracy 1, valid accuracy 0.7942
Epoch 275, CIFAR-10 Batch 3:  loss 0.008010, train_accuracy 1, valid accuracy 0.787
Epoch 275, CIFAR-10 Batch 4:  loss 0.003802, train_accuracy 1, valid accuracy 0.8
Epoch 275, CIFAR-10 Batch 5:  loss 0.009989, train_accuracy 1, valid accuracy 0.7934
Epoch 276, CIFAR-10 Batch 1:  loss 0.005759, train_accuracy 1, valid accuracy 0.7962
Epoch 276, CIFAR-10 Batch 2:  loss 0.002973, train_accuracy 1, valid accuracy 0.7974
Epoch 276, CIFAR-10 Batch 3:  loss 0.003792, train_accuracy 1, valid accuracy 0.7924
Epoch 276, CIFAR-10 Batch 4:  loss 0.008732, train_accuracy 1, valid accuracy 0.7988
Epoch 276, CIFAR-10 Batch 5:  loss 0.009361, train_accuracy 1, valid accuracy 0.785
Epoch 277, CIFAR-10 Batch 1:  loss 0.003298, train_accuracy 1, valid accuracy 0.807
Epoch 277, CIFAR-10 Batch 2:  loss 0.003368, train_accuracy 1, valid accuracy 0.8034
Epoch 277, CIFAR-10 Batch 3:  loss 0.003798, train_accuracy 1, valid accuracy 0.8026
Epoch 277, CIFAR-10 Batch 4:  loss 0.007120, train_accuracy 1, valid accuracy 0.7932
Epoch 277, CIFAR-10 Batch 5:  loss 0.006930, train_accuracy 1, valid accuracy 0.7822
Epoch 278, CIFAR-10 Batch 1:  loss 0.002398, train_accuracy 1, valid accuracy 0.8098
Epoch 278, CIFAR-10 Batch 2:  loss 0.002778, train_accuracy 1, valid accuracy 0.8022
Epoch 278, CIFAR-10 Batch 3:  loss 0.004464, train_accuracy 1, valid accuracy 0.794
Epoch 278, CIFAR-10 Batch 4:  loss 0.005731, train_accuracy 1, valid accuracy 0.8016
Epoch 278, CIFAR-10 Batch 5:  loss 0.003457, train_accuracy 1, valid accuracy 0.8004
Epoch 279, CIFAR-10 Batch 1:  loss 0.001664, train_accuracy 1, valid accuracy 0.8014
Epoch 279, CIFAR-10 Batch 2:  loss 0.001554, train_accuracy 1, valid accuracy 0.7964
Epoch 279, CIFAR-10 Batch 3:  loss 0.003241, train_accuracy 1, valid accuracy 0.7896
Epoch 279, CIFAR-10 Batch 4:  loss 0.005073, train_accuracy 1, valid accuracy 0.8022
Epoch 279, CIFAR-10 Batch 5:  loss 0.001312, train_accuracy 1, valid accuracy 0.7974
Epoch 280, CIFAR-10 Batch 1:  loss 0.005771, train_accuracy 1, valid accuracy 0.7924
Epoch 280, CIFAR-10 Batch 2:  loss 0.003304, train_accuracy 1, valid accuracy 0.7924
Epoch 280, CIFAR-10 Batch 3:  loss 0.011272, train_accuracy 1, valid accuracy 0.7966
Epoch 280, CIFAR-10 Batch 4:  loss 0.001947, train_accuracy 1, valid accuracy 0.8062
Epoch 280, CIFAR-10 Batch 5:  loss 0.002044, train_accuracy 1, valid accuracy 0.7922
Epoch 281, CIFAR-10 Batch 1:  loss 0.005790, train_accuracy 1, valid accuracy 0.7914
Epoch 281, CIFAR-10 Batch 2:  loss 0.005231, train_accuracy 1, valid accuracy 0.7924
Epoch 281, CIFAR-10 Batch 3:  loss 0.004874, train_accuracy 1, valid accuracy 0.793
Epoch 281, CIFAR-10 Batch 4:  loss 0.005496, train_accuracy 1, valid accuracy 0.7794
Epoch 281, CIFAR-10 Batch 5:  loss 0.005988, train_accuracy 1, valid accuracy 0.7942
Epoch 282, CIFAR-10 Batch 1:  loss 0.015373, train_accuracy 1, valid accuracy 0.7942
Epoch 282, CIFAR-10 Batch 2:  loss 0.002494, train_accuracy 1, valid accuracy 0.789
Epoch 282, CIFAR-10 Batch 3:  loss 0.002182, train_accuracy 1, valid accuracy 0.7982
Epoch 282, CIFAR-10 Batch 4:  loss 0.001840, train_accuracy 1, valid accuracy 0.805
Epoch 282, CIFAR-10 Batch 5:  loss 0.004875, train_accuracy 1, valid accuracy 0.799
Epoch 283, CIFAR-10 Batch 1:  loss 0.006490, train_accuracy 1, valid accuracy 0.8012
Epoch 283, CIFAR-10 Batch 2:  loss 0.001970, train_accuracy 1, valid accuracy 0.8056
Epoch 283, CIFAR-10 Batch 3:  loss 0.005088, train_accuracy 1, valid accuracy 0.788
Epoch 283, CIFAR-10 Batch 4:  loss 0.011316, train_accuracy 1, valid accuracy 0.8048
Epoch 283, CIFAR-10 Batch 5:  loss 0.011265, train_accuracy 1, valid accuracy 0.7814
Epoch 284, CIFAR-10 Batch 1:  loss 0.007790, train_accuracy 1, valid accuracy 0.8014
Epoch 284, CIFAR-10 Batch 2:  loss 0.009117, train_accuracy 1, valid accuracy 0.7954
Epoch 284, CIFAR-10 Batch 3:  loss 0.003025, train_accuracy 1, valid accuracy 0.7952
Epoch 284, CIFAR-10 Batch 4:  loss 0.010196, train_accuracy 1, valid accuracy 0.7868
Epoch 284, CIFAR-10 Batch 5:  loss 0.003321, train_accuracy 1, valid accuracy 0.7914
Epoch 285, CIFAR-10 Batch 1:  loss 0.003867, train_accuracy 1, valid accuracy 0.7994
Epoch 285, CIFAR-10 Batch 2:  loss 0.011906, train_accuracy 1, valid accuracy 0.787
Epoch 285, CIFAR-10 Batch 3:  loss 0.002520, train_accuracy 1, valid accuracy 0.8008
Epoch 285, CIFAR-10 Batch 4:  loss 0.005304, train_accuracy 1, valid accuracy 0.7976
Epoch 285, CIFAR-10 Batch 5:  loss 0.004500, train_accuracy 1, valid accuracy 0.7984
Epoch 286, CIFAR-10 Batch 1:  loss 0.001701, train_accuracy 1, valid accuracy 0.7998
Epoch 286, CIFAR-10 Batch 2:  loss 0.003953, train_accuracy 1, valid accuracy 0.8
Epoch 286, CIFAR-10 Batch 3:  loss 0.003138, train_accuracy 1, valid accuracy 0.7972
Epoch 286, CIFAR-10 Batch 4:  loss 0.009211, train_accuracy 1, valid accuracy 0.8002
Epoch 286, CIFAR-10 Batch 5:  loss 0.002492, train_accuracy 1, valid accuracy 0.7924
Epoch 287, CIFAR-10 Batch 1:  loss 0.004541, train_accuracy 1, valid accuracy 0.7974
Epoch 287, CIFAR-10 Batch 2:  loss 0.002554, train_accuracy 1, valid accuracy 0.8008
Epoch 287, CIFAR-10 Batch 3:  loss 0.003501, train_accuracy 1, valid accuracy 0.7922
Epoch 287, CIFAR-10 Batch 4:  loss 0.009469, train_accuracy 1, valid accuracy 0.7972
Epoch 287, CIFAR-10 Batch 5:  loss 0.001900, train_accuracy 1, valid accuracy 0.7958
Epoch 288, CIFAR-10 Batch 1:  loss 0.004049, train_accuracy 1, valid accuracy 0.7962
Epoch 288, CIFAR-10 Batch 2:  loss 0.003269, train_accuracy 1, valid accuracy 0.804
Epoch 288, CIFAR-10 Batch 3:  loss 0.002518, train_accuracy 1, valid accuracy 0.7924
Epoch 288, CIFAR-10 Batch 4:  loss 0.010463, train_accuracy 1, valid accuracy 0.7904
Epoch 288, CIFAR-10 Batch 5:  loss 0.004331, train_accuracy 1, valid accuracy 0.791
Epoch 289, CIFAR-10 Batch 1:  loss 0.004313, train_accuracy 1, valid accuracy 0.811
Epoch 289, CIFAR-10 Batch 2:  loss 0.002897, train_accuracy 1, valid accuracy 0.8002
Epoch 289, CIFAR-10 Batch 3:  loss 0.001747, train_accuracy 1, valid accuracy 0.7966
Epoch 289, CIFAR-10 Batch 4:  loss 0.004181, train_accuracy 1, valid accuracy 0.804
Epoch 289, CIFAR-10 Batch 5:  loss 0.002581, train_accuracy 1, valid accuracy 0.7898
Epoch 290, CIFAR-10 Batch 1:  loss 0.001601, train_accuracy 1, valid accuracy 0.7956
Epoch 290, CIFAR-10 Batch 2:  loss 0.004271, train_accuracy 1, valid accuracy 0.7988
Epoch 290, CIFAR-10 Batch 3:  loss 0.006244, train_accuracy 1, valid accuracy 0.7918
Epoch 290, CIFAR-10 Batch 4:  loss 0.004590, train_accuracy 1, valid accuracy 0.7946
Epoch 290, CIFAR-10 Batch 5:  loss 0.003973, train_accuracy 1, valid accuracy 0.7988
Epoch 291, CIFAR-10 Batch 1:  loss 0.002912, train_accuracy 1, valid accuracy 0.797
Epoch 291, CIFAR-10 Batch 2:  loss 0.003211, train_accuracy 1, valid accuracy 0.7966
Epoch 291, CIFAR-10 Batch 3:  loss 0.001866, train_accuracy 1, valid accuracy 0.7918
Epoch 291, CIFAR-10 Batch 4:  loss 0.005113, train_accuracy 1, valid accuracy 0.7938
Epoch 291, CIFAR-10 Batch 5:  loss 0.004462, train_accuracy 1, valid accuracy 0.7874
Epoch 292, CIFAR-10 Batch 1:  loss 0.002418, train_accuracy 1, valid accuracy 0.798
Epoch 292, CIFAR-10 Batch 2:  loss 0.001885, train_accuracy 1, valid accuracy 0.8014
Epoch 292, CIFAR-10 Batch 3:  loss 0.006024, train_accuracy 1, valid accuracy 0.7814
Epoch 292, CIFAR-10 Batch 4:  loss 0.003201, train_accuracy 1, valid accuracy 0.7996
Epoch 292, CIFAR-10 Batch 5:  loss 0.004094, train_accuracy 1, valid accuracy 0.7766
Epoch 293, CIFAR-10 Batch 1:  loss 0.002328, train_accuracy 1, valid accuracy 0.801
Epoch 293, CIFAR-10 Batch 2:  loss 0.002199, train_accuracy 1, valid accuracy 0.8002
Epoch 293, CIFAR-10 Batch 3:  loss 0.002776, train_accuracy 1, valid accuracy 0.7942
Epoch 293, CIFAR-10 Batch 4:  loss 0.007693, train_accuracy 1, valid accuracy 0.7938
Epoch 293, CIFAR-10 Batch 5:  loss 0.002206, train_accuracy 1, valid accuracy 0.7988
Epoch 294, CIFAR-10 Batch 1:  loss 0.006111, train_accuracy 1, valid accuracy 0.7974
Epoch 294, CIFAR-10 Batch 2:  loss 0.001826, train_accuracy 1, valid accuracy 0.803
Epoch 294, CIFAR-10 Batch 3:  loss 0.008373, train_accuracy 1, valid accuracy 0.7926
Epoch 294, CIFAR-10 Batch 4:  loss 0.006087, train_accuracy 1, valid accuracy 0.7982
Epoch 294, CIFAR-10 Batch 5:  loss 0.001998, train_accuracy 1, valid accuracy 0.7958
Epoch 295, CIFAR-10 Batch 1:  loss 0.002246, train_accuracy 1, valid accuracy 0.7942
Epoch 295, CIFAR-10 Batch 2:  loss 0.002790, train_accuracy 1, valid accuracy 0.7974
Epoch 295, CIFAR-10 Batch 3:  loss 0.001664, train_accuracy 1, valid accuracy 0.8004
Epoch 295, CIFAR-10 Batch 4:  loss 0.003861, train_accuracy 1, valid accuracy 0.8076
Epoch 295, CIFAR-10 Batch 5:  loss 0.005666, train_accuracy 1, valid accuracy 0.7898
Epoch 296, CIFAR-10 Batch 1:  loss 0.002371, train_accuracy 1, valid accuracy 0.8022
Epoch 296, CIFAR-10 Batch 2:  loss 0.002954, train_accuracy 1, valid accuracy 0.7962
Epoch 296, CIFAR-10 Batch 3:  loss 0.001293, train_accuracy 1, valid accuracy 0.7928
Epoch 296, CIFAR-10 Batch 4:  loss 0.003811, train_accuracy 1, valid accuracy 0.7978
Epoch 296, CIFAR-10 Batch 5:  loss 0.004888, train_accuracy 1, valid accuracy 0.7868
Epoch 297, CIFAR-10 Batch 1:  loss 0.001120, train_accuracy 1, valid accuracy 0.7972
Epoch 297, CIFAR-10 Batch 2:  loss 0.002058, train_accuracy 1, valid accuracy 0.8026
Epoch 297, CIFAR-10 Batch 3:  loss 0.002068, train_accuracy 1, valid accuracy 0.7974
Epoch 297, CIFAR-10 Batch 4:  loss 0.006388, train_accuracy 1, valid accuracy 0.7924
Epoch 297, CIFAR-10 Batch 5:  loss 0.005917, train_accuracy 1, valid accuracy 0.7896
Epoch 298, CIFAR-10 Batch 1:  loss 0.003974, train_accuracy 1, valid accuracy 0.8006
Epoch 298, CIFAR-10 Batch 2:  loss 0.001370, train_accuracy 1, valid accuracy 0.7994
Epoch 298, CIFAR-10 Batch 3:  loss 0.005301, train_accuracy 1, valid accuracy 0.7902
Epoch 298, CIFAR-10 Batch 4:  loss 0.002489, train_accuracy 1, valid accuracy 0.7958
Epoch 298, CIFAR-10 Batch 5:  loss 0.002900, train_accuracy 1, valid accuracy 0.7994
Epoch 299, CIFAR-10 Batch 1:  loss 0.001150, train_accuracy 1, valid accuracy 0.8048
Epoch 299, CIFAR-10 Batch 2:  loss 0.002958, train_accuracy 1, valid accuracy 0.797
Epoch 299, CIFAR-10 Batch 3:  loss 0.004397, train_accuracy 1, valid accuracy 0.7928
Epoch 299, CIFAR-10 Batch 4:  loss 0.003707, train_accuracy 1, valid accuracy 0.798
Epoch 299, CIFAR-10 Batch 5:  loss 0.002471, train_accuracy 1, valid accuracy 0.7954
Epoch 300, CIFAR-10 Batch 1:  loss 0.000769, train_accuracy 1, valid accuracy 0.8042
Epoch 300, CIFAR-10 Batch 2:  loss 0.002421, train_accuracy 1, valid accuracy 0.783
Epoch 300, CIFAR-10 Batch 3:  loss 0.018824, train_accuracy 1, valid accuracy 0.7798
Epoch 300, CIFAR-10 Batch 4:  loss 0.003055, train_accuracy 1, valid accuracy 0.7976
Epoch 300, CIFAR-10 Batch 5:  loss 0.002541, train_accuracy 1, valid accuracy 0.8008
Epoch 301, CIFAR-10 Batch 1:  loss 0.001717, train_accuracy 1, valid accuracy 0.7984
Epoch 301, CIFAR-10 Batch 2:  loss 0.001124, train_accuracy 1, valid accuracy 0.8034
Epoch 301, CIFAR-10 Batch 3:  loss 0.001477, train_accuracy 1, valid accuracy 0.7932
Epoch 301, CIFAR-10 Batch 4:  loss 0.002549, train_accuracy 1, valid accuracy 0.7952
Epoch 301, CIFAR-10 Batch 5:  loss 0.002175, train_accuracy 1, valid accuracy 0.7954
Epoch 302, CIFAR-10 Batch 1:  loss 0.000902, train_accuracy 1, valid accuracy 0.8006
Epoch 302, CIFAR-10 Batch 2:  loss 0.005063, train_accuracy 1, valid accuracy 0.784
Epoch 302, CIFAR-10 Batch 3:  loss 0.003988, train_accuracy 1, valid accuracy 0.8022
Epoch 302, CIFAR-10 Batch 4:  loss 0.009220, train_accuracy 1, valid accuracy 0.7994
Epoch 302, CIFAR-10 Batch 5:  loss 0.004951, train_accuracy 1, valid accuracy 0.795
Epoch 303, CIFAR-10 Batch 1:  loss 0.001324, train_accuracy 1, valid accuracy 0.8002
Epoch 303, CIFAR-10 Batch 2:  loss 0.004937, train_accuracy 1, valid accuracy 0.7972
Epoch 303, CIFAR-10 Batch 3:  loss 0.002791, train_accuracy 1, valid accuracy 0.7944
Epoch 303, CIFAR-10 Batch 4:  loss 0.002281, train_accuracy 1, valid accuracy 0.793
Epoch 303, CIFAR-10 Batch 5:  loss 0.002786, train_accuracy 1, valid accuracy 0.797
Epoch 304, CIFAR-10 Batch 1:  loss 0.004474, train_accuracy 1, valid accuracy 0.789
Epoch 304, CIFAR-10 Batch 2:  loss 0.002260, train_accuracy 1, valid accuracy 0.7988
Epoch 304, CIFAR-10 Batch 3:  loss 0.003370, train_accuracy 1, valid accuracy 0.7686
Epoch 304, CIFAR-10 Batch 4:  loss 0.003702, train_accuracy 1, valid accuracy 0.7846
Epoch 304, CIFAR-10 Batch 5:  loss 0.008709, train_accuracy 1, valid accuracy 0.788
Epoch 305, CIFAR-10 Batch 1:  loss 0.006407, train_accuracy 1, valid accuracy 0.7952
Epoch 305, CIFAR-10 Batch 2:  loss 0.005959, train_accuracy 1, valid accuracy 0.796
Epoch 305, CIFAR-10 Batch 3:  loss 0.001400, train_accuracy 1, valid accuracy 0.7752
Epoch 305, CIFAR-10 Batch 4:  loss 0.002213, train_accuracy 1, valid accuracy 0.7976
Epoch 305, CIFAR-10 Batch 5:  loss 0.002233, train_accuracy 1, valid accuracy 0.7998
Epoch 306, CIFAR-10 Batch 1:  loss 0.002985, train_accuracy 1, valid accuracy 0.7964
Epoch 306, CIFAR-10 Batch 2:  loss 0.004626, train_accuracy 1, valid accuracy 0.7952
Epoch 306, CIFAR-10 Batch 3:  loss 0.000857, train_accuracy 1, valid accuracy 0.7954
Epoch 306, CIFAR-10 Batch 4:  loss 0.002705, train_accuracy 1, valid accuracy 0.803
Epoch 306, CIFAR-10 Batch 5:  loss 0.000908, train_accuracy 1, valid accuracy 0.795
Epoch 307, CIFAR-10 Batch 1:  loss 0.005556, train_accuracy 1, valid accuracy 0.8038
Epoch 307, CIFAR-10 Batch 2:  loss 0.001809, train_accuracy 1, valid accuracy 0.8058
Epoch 307, CIFAR-10 Batch 3:  loss 0.000705, train_accuracy 1, valid accuracy 0.7962
Epoch 307, CIFAR-10 Batch 4:  loss 0.002483, train_accuracy 1, valid accuracy 0.802
Epoch 307, CIFAR-10 Batch 5:  loss 0.000464, train_accuracy 1, valid accuracy 0.8034
Epoch 308, CIFAR-10 Batch 1:  loss 0.004731, train_accuracy 1, valid accuracy 0.7998
Epoch 308, CIFAR-10 Batch 2:  loss 0.004362, train_accuracy 1, valid accuracy 0.8052
Epoch 308, CIFAR-10 Batch 3:  loss 0.002120, train_accuracy 1, valid accuracy 0.7914
Epoch 308, CIFAR-10 Batch 4:  loss 0.003373, train_accuracy 1, valid accuracy 0.806
Epoch 308, CIFAR-10 Batch 5:  loss 0.006828, train_accuracy 1, valid accuracy 0.7792
Epoch 309, CIFAR-10 Batch 1:  loss 0.002780, train_accuracy 1, valid accuracy 0.7962
Epoch 309, CIFAR-10 Batch 2:  loss 0.003379, train_accuracy 1, valid accuracy 0.7896
Epoch 309, CIFAR-10 Batch 3:  loss 0.002478, train_accuracy 1, valid accuracy 0.7958
Epoch 309, CIFAR-10 Batch 4:  loss 0.003771, train_accuracy 1, valid accuracy 0.7998
Epoch 309, CIFAR-10 Batch 5:  loss 0.005978, train_accuracy 1, valid accuracy 0.7846
Epoch 310, CIFAR-10 Batch 1:  loss 0.003129, train_accuracy 1, valid accuracy 0.7966
Epoch 310, CIFAR-10 Batch 2:  loss 0.002788, train_accuracy 1, valid accuracy 0.8008
Epoch 310, CIFAR-10 Batch 3:  loss 0.004070, train_accuracy 1, valid accuracy 0.7928
Epoch 310, CIFAR-10 Batch 4:  loss 0.008327, train_accuracy 1, valid accuracy 0.8016
Epoch 310, CIFAR-10 Batch 5:  loss 0.000772, train_accuracy 1, valid accuracy 0.8004
Epoch 311, CIFAR-10 Batch 1:  loss 0.001306, train_accuracy 1, valid accuracy 0.7966
Epoch 311, CIFAR-10 Batch 2:  loss 0.001475, train_accuracy 1, valid accuracy 0.8092
Epoch 311, CIFAR-10 Batch 3:  loss 0.001738, train_accuracy 1, valid accuracy 0.803
Epoch 311, CIFAR-10 Batch 4:  loss 0.003314, train_accuracy 1, valid accuracy 0.8058
Epoch 311, CIFAR-10 Batch 5:  loss 0.002312, train_accuracy 1, valid accuracy 0.7986
Epoch 312, CIFAR-10 Batch 1:  loss 0.002904, train_accuracy 1, valid accuracy 0.801
Epoch 312, CIFAR-10 Batch 2:  loss 0.000900, train_accuracy 1, valid accuracy 0.8066
Epoch 312, CIFAR-10 Batch 3:  loss 0.004399, train_accuracy 1, valid accuracy 0.79
Epoch 312, CIFAR-10 Batch 4:  loss 0.008625, train_accuracy 1, valid accuracy 0.7924
Epoch 312, CIFAR-10 Batch 5:  loss 0.001028, train_accuracy 1, valid accuracy 0.8
Epoch 313, CIFAR-10 Batch 1:  loss 0.000929, train_accuracy 1, valid accuracy 0.805
Epoch 313, CIFAR-10 Batch 2:  loss 0.002883, train_accuracy 1, valid accuracy 0.8046
Epoch 313, CIFAR-10 Batch 3:  loss 0.003360, train_accuracy 1, valid accuracy 0.7984
Epoch 313, CIFAR-10 Batch 4:  loss 0.014576, train_accuracy 1, valid accuracy 0.7874
Epoch 313, CIFAR-10 Batch 5:  loss 0.001416, train_accuracy 1, valid accuracy 0.8
Epoch 314, CIFAR-10 Batch 1:  loss 0.001658, train_accuracy 1, valid accuracy 0.8074
Epoch 314, CIFAR-10 Batch 2:  loss 0.001225, train_accuracy 1, valid accuracy 0.801
Epoch 314, CIFAR-10 Batch 3:  loss 0.001880, train_accuracy 1, valid accuracy 0.7976
Epoch 314, CIFAR-10 Batch 4:  loss 0.003454, train_accuracy 1, valid accuracy 0.798
Epoch 314, CIFAR-10 Batch 5:  loss 0.001558, train_accuracy 1, valid accuracy 0.805
Epoch 315, CIFAR-10 Batch 1:  loss 0.000892, train_accuracy 1, valid accuracy 0.8
Epoch 315, CIFAR-10 Batch 2:  loss 0.003571, train_accuracy 1, valid accuracy 0.797
Epoch 315, CIFAR-10 Batch 3:  loss 0.002853, train_accuracy 1, valid accuracy 0.8012
Epoch 315, CIFAR-10 Batch 4:  loss 0.011891, train_accuracy 1, valid accuracy 0.805
Epoch 315, CIFAR-10 Batch 5:  loss 0.001860, train_accuracy 1, valid accuracy 0.7986
Epoch 316, CIFAR-10 Batch 1:  loss 0.001502, train_accuracy 1, valid accuracy 0.8004
Epoch 316, CIFAR-10 Batch 2:  loss 0.021604, train_accuracy 1, valid accuracy 0.798
Epoch 316, CIFAR-10 Batch 3:  loss 0.002072, train_accuracy 1, valid accuracy 0.7982
Epoch 316, CIFAR-10 Batch 4:  loss 0.003948, train_accuracy 1, valid accuracy 0.7922
Epoch 316, CIFAR-10 Batch 5:  loss 0.001861, train_accuracy 1, valid accuracy 0.802
Epoch 317, CIFAR-10 Batch 1:  loss 0.001898, train_accuracy 1, valid accuracy 0.8064
Epoch 317, CIFAR-10 Batch 2:  loss 0.002490, train_accuracy 1, valid accuracy 0.8086
Epoch 317, CIFAR-10 Batch 3:  loss 0.002400, train_accuracy 1, valid accuracy 0.7908
Epoch 317, CIFAR-10 Batch 4:  loss 0.004469, train_accuracy 1, valid accuracy 0.7986
Epoch 317, CIFAR-10 Batch 5:  loss 0.000820, train_accuracy 1, valid accuracy 0.8028
Epoch 318, CIFAR-10 Batch 1:  loss 0.002156, train_accuracy 1, valid accuracy 0.8028
Epoch 318, CIFAR-10 Batch 2:  loss 0.001603, train_accuracy 1, valid accuracy 0.7974
Epoch 318, CIFAR-10 Batch 3:  loss 0.005135, train_accuracy 1, valid accuracy 0.8022
Epoch 318, CIFAR-10 Batch 4:  loss 0.003335, train_accuracy 1, valid accuracy 0.8012
Epoch 318, CIFAR-10 Batch 5:  loss 0.003186, train_accuracy 1, valid accuracy 0.803
Epoch 319, CIFAR-10 Batch 1:  loss 0.000938, train_accuracy 1, valid accuracy 0.8122
Epoch 319, CIFAR-10 Batch 2:  loss 0.001348, train_accuracy 1, valid accuracy 0.8008
Epoch 319, CIFAR-10 Batch 3:  loss 0.002774, train_accuracy 1, valid accuracy 0.7936
Epoch 319, CIFAR-10 Batch 4:  loss 0.012506, train_accuracy 1, valid accuracy 0.8044
Epoch 319, CIFAR-10 Batch 5:  loss 0.002831, train_accuracy 1, valid accuracy 0.814
Epoch 320, CIFAR-10 Batch 1:  loss 0.004123, train_accuracy 1, valid accuracy 0.8014
Epoch 320, CIFAR-10 Batch 2:  loss 0.001710, train_accuracy 1, valid accuracy 0.7994
Epoch 320, CIFAR-10 Batch 3:  loss 0.001970, train_accuracy 1, valid accuracy 0.7922
Epoch 320, CIFAR-10 Batch 4:  loss 0.004576, train_accuracy 1, valid accuracy 0.7988
Epoch 320, CIFAR-10 Batch 5:  loss 0.006923, train_accuracy 1, valid accuracy 0.8
Epoch 321, CIFAR-10 Batch 1:  loss 0.001557, train_accuracy 1, valid accuracy 0.8122
Epoch 321, CIFAR-10 Batch 2:  loss 0.004311, train_accuracy 1, valid accuracy 0.7968
Epoch 321, CIFAR-10 Batch 3:  loss 0.002549, train_accuracy 1, valid accuracy 0.8012
Epoch 321, CIFAR-10 Batch 4:  loss 0.004070, train_accuracy 1, valid accuracy 0.8004
Epoch 321, CIFAR-10 Batch 5:  loss 0.001635, train_accuracy 1, valid accuracy 0.7996
Epoch 322, CIFAR-10 Batch 1:  loss 0.002500, train_accuracy 1, valid accuracy 0.8066
Epoch 322, CIFAR-10 Batch 2:  loss 0.003780, train_accuracy 1, valid accuracy 0.8002
Epoch 322, CIFAR-10 Batch 3:  loss 0.003050, train_accuracy 1, valid accuracy 0.7914
Epoch 322, CIFAR-10 Batch 4:  loss 0.004891, train_accuracy 1, valid accuracy 0.8042
Epoch 322, CIFAR-10 Batch 5:  loss 0.002273, train_accuracy 1, valid accuracy 0.794
Epoch 323, CIFAR-10 Batch 1:  loss 0.004401, train_accuracy 1, valid accuracy 0.8046
Epoch 323, CIFAR-10 Batch 2:  loss 0.006961, train_accuracy 1, valid accuracy 0.7966
Epoch 323, CIFAR-10 Batch 3:  loss 0.002368, train_accuracy 1, valid accuracy 0.7978
Epoch 323, CIFAR-10 Batch 4:  loss 0.005093, train_accuracy 1, valid accuracy 0.7906
Epoch 323, CIFAR-10 Batch 5:  loss 0.000777, train_accuracy 1, valid accuracy 0.8086
Epoch 324, CIFAR-10 Batch 1:  loss 0.003193, train_accuracy 1, valid accuracy 0.8062
Epoch 324, CIFAR-10 Batch 2:  loss 0.002684, train_accuracy 1, valid accuracy 0.81
Epoch 324, CIFAR-10 Batch 3:  loss 0.003087, train_accuracy 1, valid accuracy 0.8008
Epoch 324, CIFAR-10 Batch 4:  loss 0.009938, train_accuracy 1, valid accuracy 0.8062
Epoch 324, CIFAR-10 Batch 5:  loss 0.001866, train_accuracy 1, valid accuracy 0.8082
Epoch 325, CIFAR-10 Batch 1:  loss 0.002587, train_accuracy 1, valid accuracy 0.8068
Epoch 325, CIFAR-10 Batch 2:  loss 0.002132, train_accuracy 1, valid accuracy 0.8066
Epoch 325, CIFAR-10 Batch 3:  loss 0.002315, train_accuracy 1, valid accuracy 0.8024
Epoch 325, CIFAR-10 Batch 4:  loss 0.006921, train_accuracy 1, valid accuracy 0.7918
Epoch 325, CIFAR-10 Batch 5:  loss 0.002135, train_accuracy 1, valid accuracy 0.7984
Epoch 326, CIFAR-10 Batch 1:  loss 0.002073, train_accuracy 1, valid accuracy 0.8028
Epoch 326, CIFAR-10 Batch 2:  loss 0.005975, train_accuracy 1, valid accuracy 0.785
Epoch 326, CIFAR-10 Batch 3:  loss 0.001362, train_accuracy 1, valid accuracy 0.8042
Epoch 326, CIFAR-10 Batch 4:  loss 0.001918, train_accuracy 1, valid accuracy 0.8004
Epoch 326, CIFAR-10 Batch 5:  loss 0.000912, train_accuracy 1, valid accuracy 0.804
Epoch 327, CIFAR-10 Batch 1:  loss 0.003977, train_accuracy 1, valid accuracy 0.801
Epoch 327, CIFAR-10 Batch 2:  loss 0.001289, train_accuracy 1, valid accuracy 0.803
Epoch 327, CIFAR-10 Batch 3:  loss 0.002310, train_accuracy 1, valid accuracy 0.8026
Epoch 327, CIFAR-10 Batch 4:  loss 0.002554, train_accuracy 1, valid accuracy 0.8028
Epoch 327, CIFAR-10 Batch 5:  loss 0.001557, train_accuracy 1, valid accuracy 0.8016
Epoch 328, CIFAR-10 Batch 1:  loss 0.002678, train_accuracy 1, valid accuracy 0.802
Epoch 328, CIFAR-10 Batch 2:  loss 0.007600, train_accuracy 1, valid accuracy 0.7976
Epoch 328, CIFAR-10 Batch 3:  loss 0.003808, train_accuracy 1, valid accuracy 0.8074
Epoch 328, CIFAR-10 Batch 4:  loss 0.002386, train_accuracy 1, valid accuracy 0.8036
Epoch 328, CIFAR-10 Batch 5:  loss 0.000988, train_accuracy 1, valid accuracy 0.8078
Epoch 329, CIFAR-10 Batch 1:  loss 0.011776, train_accuracy 1, valid accuracy 0.787
Epoch 329, CIFAR-10 Batch 2:  loss 0.001320, train_accuracy 1, valid accuracy 0.798
Epoch 329, CIFAR-10 Batch 3:  loss 0.002309, train_accuracy 1, valid accuracy 0.7964
Epoch 329, CIFAR-10 Batch 4:  loss 0.004171, train_accuracy 1, valid accuracy 0.7878
Epoch 329, CIFAR-10 Batch 5:  loss 0.002334, train_accuracy 1, valid accuracy 0.8024
Epoch 330, CIFAR-10 Batch 1:  loss 0.002157, train_accuracy 1, valid accuracy 0.7984
Epoch 330, CIFAR-10 Batch 2:  loss 0.006828, train_accuracy 1, valid accuracy 0.7882
Epoch 330, CIFAR-10 Batch 3:  loss 0.002724, train_accuracy 1, valid accuracy 0.7968
Epoch 330, CIFAR-10 Batch 4:  loss 0.002798, train_accuracy 1, valid accuracy 0.8008
Epoch 330, CIFAR-10 Batch 5:  loss 0.000712, train_accuracy 1, valid accuracy 0.7986
Epoch 331, CIFAR-10 Batch 1:  loss 0.001299, train_accuracy 1, valid accuracy 0.803
Epoch 331, CIFAR-10 Batch 2:  loss 0.002927, train_accuracy 1, valid accuracy 0.7992
Epoch 331, CIFAR-10 Batch 3:  loss 0.002149, train_accuracy 1, valid accuracy 0.797
Epoch 331, CIFAR-10 Batch 4:  loss 0.003133, train_accuracy 1, valid accuracy 0.8096
Epoch 331, CIFAR-10 Batch 5:  loss 0.002435, train_accuracy 1, valid accuracy 0.8036
Epoch 332, CIFAR-10 Batch 1:  loss 0.001262, train_accuracy 1, valid accuracy 0.8098
Epoch 332, CIFAR-10 Batch 2:  loss 0.003357, train_accuracy 1, valid accuracy 0.8004
Epoch 332, CIFAR-10 Batch 3:  loss 0.003459, train_accuracy 1, valid accuracy 0.7936
Epoch 332, CIFAR-10 Batch 4:  loss 0.003125, train_accuracy 1, valid accuracy 0.7942
Epoch 332, CIFAR-10 Batch 5:  loss 0.001341, train_accuracy 1, valid accuracy 0.8048
Epoch 333, CIFAR-10 Batch 1:  loss 0.001993, train_accuracy 1, valid accuracy 0.7986
Epoch 333, CIFAR-10 Batch 2:  loss 0.001768, train_accuracy 1, valid accuracy 0.7988
Epoch 333, CIFAR-10 Batch 3:  loss 0.002010, train_accuracy 1, valid accuracy 0.7974
Epoch 333, CIFAR-10 Batch 4:  loss 0.001774, train_accuracy 1, valid accuracy 0.7976
Epoch 333, CIFAR-10 Batch 5:  loss 0.002353, train_accuracy 1, valid accuracy 0.7948
Epoch 334, CIFAR-10 Batch 1:  loss 0.000871, train_accuracy 1, valid accuracy 0.8032
Epoch 334, CIFAR-10 Batch 2:  loss 0.003065, train_accuracy 1, valid accuracy 0.7962
Epoch 334, CIFAR-10 Batch 3:  loss 0.005678, train_accuracy 1, valid accuracy 0.7986
Epoch 334, CIFAR-10 Batch 4:  loss 0.002658, train_accuracy 1, valid accuracy 0.797
Epoch 334, CIFAR-10 Batch 5:  loss 0.000684, train_accuracy 1, valid accuracy 0.8036
Epoch 335, CIFAR-10 Batch 1:  loss 0.000933, train_accuracy 1, valid accuracy 0.8104
Epoch 335, CIFAR-10 Batch 2:  loss 0.001573, train_accuracy 1, valid accuracy 0.804
Epoch 335, CIFAR-10 Batch 3:  loss 0.003166, train_accuracy 1, valid accuracy 0.8006
Epoch 335, CIFAR-10 Batch 4:  loss 0.002764, train_accuracy 1, valid accuracy 0.806
Epoch 335, CIFAR-10 Batch 5:  loss 0.001558, train_accuracy 1, valid accuracy 0.8018
Epoch 336, CIFAR-10 Batch 1:  loss 0.001107, train_accuracy 1, valid accuracy 0.802
Epoch 336, CIFAR-10 Batch 2:  loss 0.000903, train_accuracy 1, valid accuracy 0.8112
Epoch 336, CIFAR-10 Batch 3:  loss 0.001483, train_accuracy 1, valid accuracy 0.7946
Epoch 336, CIFAR-10 Batch 4:  loss 0.003586, train_accuracy 1, valid accuracy 0.7848
Epoch 336, CIFAR-10 Batch 5:  loss 0.001149, train_accuracy 1, valid accuracy 0.8014
Epoch 337, CIFAR-10 Batch 1:  loss 0.007973, train_accuracy 1, valid accuracy 0.7938
Epoch 337, CIFAR-10 Batch 2:  loss 0.001259, train_accuracy 1, valid accuracy 0.8034
Epoch 337, CIFAR-10 Batch 3:  loss 0.003370, train_accuracy 1, valid accuracy 0.8054
Epoch 337, CIFAR-10 Batch 4:  loss 0.002597, train_accuracy 1, valid accuracy 0.8
Epoch 337, CIFAR-10 Batch 5:  loss 0.002618, train_accuracy 1, valid accuracy 0.789
Epoch 338, CIFAR-10 Batch 1:  loss 0.000819, train_accuracy 1, valid accuracy 0.8104
Epoch 338, CIFAR-10 Batch 2:  loss 0.003686, train_accuracy 1, valid accuracy 0.7938
Epoch 338, CIFAR-10 Batch 3:  loss 0.004608, train_accuracy 1, valid accuracy 0.8108
Epoch 338, CIFAR-10 Batch 4:  loss 0.001127, train_accuracy 1, valid accuracy 0.8006
Epoch 338, CIFAR-10 Batch 5:  loss 0.000987, train_accuracy 1, valid accuracy 0.7982
Epoch 339, CIFAR-10 Batch 1:  loss 0.001506, train_accuracy 1, valid accuracy 0.8036
Epoch 339, CIFAR-10 Batch 2:  loss 0.001064, train_accuracy 1, valid accuracy 0.8078
Epoch 339, CIFAR-10 Batch 3:  loss 0.002060, train_accuracy 1, valid accuracy 0.7976
Epoch 339, CIFAR-10 Batch 4:  loss 0.002212, train_accuracy 1, valid accuracy 0.8068
Epoch 339, CIFAR-10 Batch 5:  loss 0.000831, train_accuracy 1, valid accuracy 0.8026
Epoch 340, CIFAR-10 Batch 1:  loss 0.001616, train_accuracy 1, valid accuracy 0.804
Epoch 340, CIFAR-10 Batch 2:  loss 0.002675, train_accuracy 1, valid accuracy 0.8048
Epoch 340, CIFAR-10 Batch 3:  loss 0.003052, train_accuracy 1, valid accuracy 0.7876
Epoch 340, CIFAR-10 Batch 4:  loss 0.001966, train_accuracy 1, valid accuracy 0.8012
Epoch 340, CIFAR-10 Batch 5:  loss 0.000774, train_accuracy 1, valid accuracy 0.8098
Epoch 341, CIFAR-10 Batch 1:  loss 0.001832, train_accuracy 1, valid accuracy 0.803
Epoch 341, CIFAR-10 Batch 2:  loss 0.001835, train_accuracy 1, valid accuracy 0.8022
Epoch 341, CIFAR-10 Batch 3:  loss 0.003119, train_accuracy 1, valid accuracy 0.793
Epoch 341, CIFAR-10 Batch 4:  loss 0.000602, train_accuracy 1, valid accuracy 0.7994
Epoch 341, CIFAR-10 Batch 5:  loss 0.001197, train_accuracy 1, valid accuracy 0.798
Epoch 342, CIFAR-10 Batch 1:  loss 0.002540, train_accuracy 1, valid accuracy 0.8076
Epoch 342, CIFAR-10 Batch 2:  loss 0.001222, train_accuracy 1, valid accuracy 0.8062
Epoch 342, CIFAR-10 Batch 3:  loss 0.002542, train_accuracy 1, valid accuracy 0.7994
Epoch 342, CIFAR-10 Batch 4:  loss 0.001403, train_accuracy 1, valid accuracy 0.8054
Epoch 342, CIFAR-10 Batch 5:  loss 0.001160, train_accuracy 1, valid accuracy 0.7986
Epoch 343, CIFAR-10 Batch 1:  loss 0.001504, train_accuracy 1, valid accuracy 0.8092
Epoch 343, CIFAR-10 Batch 2:  loss 0.001702, train_accuracy 1, valid accuracy 0.808
Epoch 343, CIFAR-10 Batch 3:  loss 0.001308, train_accuracy 1, valid accuracy 0.8082
Epoch 343, CIFAR-10 Batch 4:  loss 0.001548, train_accuracy 1, valid accuracy 0.8052
Epoch 343, CIFAR-10 Batch 5:  loss 0.001879, train_accuracy 1, valid accuracy 0.8116
Epoch 344, CIFAR-10 Batch 1:  loss 0.002362, train_accuracy 1, valid accuracy 0.8132
Epoch 344, CIFAR-10 Batch 2:  loss 0.001179, train_accuracy 1, valid accuracy 0.8038
Epoch 344, CIFAR-10 Batch 3:  loss 0.002844, train_accuracy 1, valid accuracy 0.8036
Epoch 344, CIFAR-10 Batch 4:  loss 0.001992, train_accuracy 1, valid accuracy 0.81
Epoch 344, CIFAR-10 Batch 5:  loss 0.000424, train_accuracy 1, valid accuracy 0.806
Epoch 345, CIFAR-10 Batch 1:  loss 0.000847, train_accuracy 1, valid accuracy 0.8104
Epoch 345, CIFAR-10 Batch 2:  loss 0.000782, train_accuracy 1, valid accuracy 0.813
Epoch 345, CIFAR-10 Batch 3:  loss 0.001235, train_accuracy 1, valid accuracy 0.7998
Epoch 345, CIFAR-10 Batch 4:  loss 0.001603, train_accuracy 1, valid accuracy 0.801
Epoch 345, CIFAR-10 Batch 5:  loss 0.001218, train_accuracy 1, valid accuracy 0.8058
Epoch 346, CIFAR-10 Batch 1:  loss 0.001856, train_accuracy 1, valid accuracy 0.8064
Epoch 346, CIFAR-10 Batch 2:  loss 0.001273, train_accuracy 1, valid accuracy 0.7976
Epoch 346, CIFAR-10 Batch 3:  loss 0.002792, train_accuracy 1, valid accuracy 0.7982
Epoch 346, CIFAR-10 Batch 4:  loss 0.003140, train_accuracy 1, valid accuracy 0.7974
Epoch 346, CIFAR-10 Batch 5:  loss 0.001480, train_accuracy 1, valid accuracy 0.8102
Epoch 347, CIFAR-10 Batch 1:  loss 0.003129, train_accuracy 1, valid accuracy 0.8026
Epoch 347, CIFAR-10 Batch 2:  loss 0.003517, train_accuracy 1, valid accuracy 0.8078
Epoch 347, CIFAR-10 Batch 3:  loss 0.003017, train_accuracy 1, valid accuracy 0.8068
Epoch 347, CIFAR-10 Batch 4:  loss 0.003704, train_accuracy 1, valid accuracy 0.7914
Epoch 347, CIFAR-10 Batch 5:  loss 0.001823, train_accuracy 1, valid accuracy 0.8002
Epoch 348, CIFAR-10 Batch 1:  loss 0.000973, train_accuracy 1, valid accuracy 0.8034
Epoch 348, CIFAR-10 Batch 2:  loss 0.002540, train_accuracy 1, valid accuracy 0.8034
Epoch 348, CIFAR-10 Batch 3:  loss 0.001488, train_accuracy 1, valid accuracy 0.7988
Epoch 348, CIFAR-10 Batch 4:  loss 0.002983, train_accuracy 1, valid accuracy 0.8102
Epoch 348, CIFAR-10 Batch 5:  loss 0.002236, train_accuracy 1, valid accuracy 0.794
Epoch 349, CIFAR-10 Batch 1:  loss 0.002571, train_accuracy 1, valid accuracy 0.808
Epoch 349, CIFAR-10 Batch 2:  loss 0.002150, train_accuracy 1, valid accuracy 0.7798
Epoch 349, CIFAR-10 Batch 3:  loss 0.000958, train_accuracy 1, valid accuracy 0.7958
Epoch 349, CIFAR-10 Batch 4:  loss 0.004445, train_accuracy 1, valid accuracy 0.8078
Epoch 349, CIFAR-10 Batch 5:  loss 0.002456, train_accuracy 1, valid accuracy 0.797
Epoch 350, CIFAR-10 Batch 1:  loss 0.002417, train_accuracy 1, valid accuracy 0.8018
Epoch 350, CIFAR-10 Batch 2:  loss 0.008382, train_accuracy 1, valid accuracy 0.8128
Epoch 350, CIFAR-10 Batch 3:  loss 0.002422, train_accuracy 1, valid accuracy 0.8036
Epoch 350, CIFAR-10 Batch 4:  loss 0.002691, train_accuracy 1, valid accuracy 0.8006
Epoch 350, CIFAR-10 Batch 5:  loss 0.001066, train_accuracy 1, valid accuracy 0.8026
Epoch 351, CIFAR-10 Batch 1:  loss 0.000630, train_accuracy 1, valid accuracy 0.8062
Epoch 351, CIFAR-10 Batch 2:  loss 0.001718, train_accuracy 1, valid accuracy 0.8072
Epoch 351, CIFAR-10 Batch 3:  loss 0.001170, train_accuracy 1, valid accuracy 0.8024
Epoch 351, CIFAR-10 Batch 4:  loss 0.001168, train_accuracy 1, valid accuracy 0.8042
Epoch 351, CIFAR-10 Batch 5:  loss 0.003193, train_accuracy 1, valid accuracy 0.7916
Epoch 352, CIFAR-10 Batch 1:  loss 0.000683, train_accuracy 1, valid accuracy 0.8048
Epoch 352, CIFAR-10 Batch 2:  loss 0.001800, train_accuracy 1, valid accuracy 0.8048
Epoch 352, CIFAR-10 Batch 3:  loss 0.003525, train_accuracy 1, valid accuracy 0.7984
Epoch 352, CIFAR-10 Batch 4:  loss 0.002431, train_accuracy 1, valid accuracy 0.8054
Epoch 352, CIFAR-10 Batch 5:  loss 0.000595, train_accuracy 1, valid accuracy 0.8004
Epoch 353, CIFAR-10 Batch 1:  loss 0.001625, train_accuracy 1, valid accuracy 0.8046
Epoch 353, CIFAR-10 Batch 2:  loss 0.001055, train_accuracy 1, valid accuracy 0.8054
Epoch 353, CIFAR-10 Batch 3:  loss 0.004259, train_accuracy 1, valid accuracy 0.8026
Epoch 353, CIFAR-10 Batch 4:  loss 0.001809, train_accuracy 1, valid accuracy 0.7892
Epoch 353, CIFAR-10 Batch 5:  loss 0.001672, train_accuracy 1, valid accuracy 0.8038
Epoch 354, CIFAR-10 Batch 1:  loss 0.000926, train_accuracy 1, valid accuracy 0.8034
Epoch 354, CIFAR-10 Batch 2:  loss 0.000698, train_accuracy 1, valid accuracy 0.803
Epoch 354, CIFAR-10 Batch 3:  loss 0.000978, train_accuracy 1, valid accuracy 0.8064
Epoch 354, CIFAR-10 Batch 4:  loss 0.005532, train_accuracy 1, valid accuracy 0.7926
Epoch 354, CIFAR-10 Batch 5:  loss 0.001098, train_accuracy 1, valid accuracy 0.7918
Epoch 355, CIFAR-10 Batch 1:  loss 0.001643, train_accuracy 1, valid accuracy 0.8092
Epoch 355, CIFAR-10 Batch 2:  loss 0.000595, train_accuracy 1, valid accuracy 0.8038
Epoch 355, CIFAR-10 Batch 3:  loss 0.002083, train_accuracy 1, valid accuracy 0.8004
Epoch 355, CIFAR-10 Batch 4:  loss 0.003872, train_accuracy 1, valid accuracy 0.8026
Epoch 355, CIFAR-10 Batch 5:  loss 0.001630, train_accuracy 1, valid accuracy 0.7838
Epoch 356, CIFAR-10 Batch 1:  loss 0.001448, train_accuracy 1, valid accuracy 0.8052
Epoch 356, CIFAR-10 Batch 2:  loss 0.002180, train_accuracy 1, valid accuracy 0.7988
Epoch 356, CIFAR-10 Batch 3:  loss 0.001583, train_accuracy 1, valid accuracy 0.8036
Epoch 356, CIFAR-10 Batch 4:  loss 0.002404, train_accuracy 1, valid accuracy 0.8034
Epoch 356, CIFAR-10 Batch 5:  loss 0.005833, train_accuracy 1, valid accuracy 0.782
Epoch 357, CIFAR-10 Batch 1:  loss 0.001449, train_accuracy 1, valid accuracy 0.8108
Epoch 357, CIFAR-10 Batch 2:  loss 0.003120, train_accuracy 1, valid accuracy 0.8036
Epoch 357, CIFAR-10 Batch 3:  loss 0.001011, train_accuracy 1, valid accuracy 0.8044
Epoch 357, CIFAR-10 Batch 4:  loss 0.003328, train_accuracy 1, valid accuracy 0.802
Epoch 357, CIFAR-10 Batch 5:  loss 0.001300, train_accuracy 1, valid accuracy 0.7918
Epoch 358, CIFAR-10 Batch 1:  loss 0.000977, train_accuracy 1, valid accuracy 0.8074
Epoch 358, CIFAR-10 Batch 2:  loss 0.006350, train_accuracy 1, valid accuracy 0.7794
Epoch 358, CIFAR-10 Batch 3:  loss 0.001251, train_accuracy 1, valid accuracy 0.804
Epoch 358, CIFAR-10 Batch 4:  loss 0.002030, train_accuracy 1, valid accuracy 0.8044
Epoch 358, CIFAR-10 Batch 5:  loss 0.000526, train_accuracy 1, valid accuracy 0.8066
Epoch 359, CIFAR-10 Batch 1:  loss 0.001093, train_accuracy 1, valid accuracy 0.8052
Epoch 359, CIFAR-10 Batch 2:  loss 0.002689, train_accuracy 1, valid accuracy 0.7974
Epoch 359, CIFAR-10 Batch 3:  loss 0.015738, train_accuracy 1, valid accuracy 0.774
Epoch 359, CIFAR-10 Batch 4:  loss 0.001777, train_accuracy 1, valid accuracy 0.797
Epoch 359, CIFAR-10 Batch 5:  loss 0.001238, train_accuracy 1, valid accuracy 0.7946
Epoch 360, CIFAR-10 Batch 1:  loss 0.000590, train_accuracy 1, valid accuracy 0.8068
Epoch 360, CIFAR-10 Batch 2:  loss 0.002822, train_accuracy 1, valid accuracy 0.7944
Epoch 360, CIFAR-10 Batch 3:  loss 0.000755, train_accuracy 1, valid accuracy 0.8006
Epoch 360, CIFAR-10 Batch 4:  loss 0.000820, train_accuracy 1, valid accuracy 0.8152
Epoch 360, CIFAR-10 Batch 5:  loss 0.002141, train_accuracy 1, valid accuracy 0.8018
Epoch 361, CIFAR-10 Batch 1:  loss 0.001187, train_accuracy 1, valid accuracy 0.8104
Epoch 361, CIFAR-10 Batch 2:  loss 0.000881, train_accuracy 1, valid accuracy 0.796
Epoch 361, CIFAR-10 Batch 3:  loss 0.002196, train_accuracy 1, valid accuracy 0.7956
Epoch 361, CIFAR-10 Batch 4:  loss 0.000387, train_accuracy 1, valid accuracy 0.8018
Epoch 361, CIFAR-10 Batch 5:  loss 0.001003, train_accuracy 1, valid accuracy 0.793
Epoch 362, CIFAR-10 Batch 1:  loss 0.001249, train_accuracy 1, valid accuracy 0.8064
Epoch 362, CIFAR-10 Batch 2:  loss 0.001127, train_accuracy 1, valid accuracy 0.8098
Epoch 362, CIFAR-10 Batch 3:  loss 0.002722, train_accuracy 1, valid accuracy 0.7968
Epoch 362, CIFAR-10 Batch 4:  loss 0.002265, train_accuracy 1, valid accuracy 0.8038
Epoch 362, CIFAR-10 Batch 5:  loss 0.001424, train_accuracy 1, valid accuracy 0.8014
Epoch 363, CIFAR-10 Batch 1:  loss 0.001128, train_accuracy 1, valid accuracy 0.8032
Epoch 363, CIFAR-10 Batch 2:  loss 0.001198, train_accuracy 1, valid accuracy 0.7964
Epoch 363, CIFAR-10 Batch 3:  loss 0.001022, train_accuracy 1, valid accuracy 0.8032
Epoch 363, CIFAR-10 Batch 4:  loss 0.003821, train_accuracy 1, valid accuracy 0.7922
Epoch 363, CIFAR-10 Batch 5:  loss 0.001804, train_accuracy 1, valid accuracy 0.802
Epoch 364, CIFAR-10 Batch 1:  loss 0.000798, train_accuracy 1, valid accuracy 0.8012
Epoch 364, CIFAR-10 Batch 2:  loss 0.001194, train_accuracy 1, valid accuracy 0.802
Epoch 364, CIFAR-10 Batch 3:  loss 0.000576, train_accuracy 1, valid accuracy 0.7986
Epoch 364, CIFAR-10 Batch 4:  loss 0.001582, train_accuracy 1, valid accuracy 0.805
Epoch 364, CIFAR-10 Batch 5:  loss 0.004348, train_accuracy 1, valid accuracy 0.7956
Epoch 365, CIFAR-10 Batch 1:  loss 0.003890, train_accuracy 1, valid accuracy 0.7958
Epoch 365, CIFAR-10 Batch 2:  loss 0.003250, train_accuracy 1, valid accuracy 0.8012
Epoch 365, CIFAR-10 Batch 3:  loss 0.001596, train_accuracy 1, valid accuracy 0.8092
Epoch 365, CIFAR-10 Batch 4:  loss 0.001640, train_accuracy 1, valid accuracy 0.8058
Epoch 365, CIFAR-10 Batch 5:  loss 0.000931, train_accuracy 1, valid accuracy 0.8036
Epoch 366, CIFAR-10 Batch 1:  loss 0.001320, train_accuracy 1, valid accuracy 0.8032
Epoch 366, CIFAR-10 Batch 2:  loss 0.001849, train_accuracy 1, valid accuracy 0.7932
Epoch 366, CIFAR-10 Batch 3:  loss 0.001745, train_accuracy 1, valid accuracy 0.7786
Epoch 366, CIFAR-10 Batch 4:  loss 0.001647, train_accuracy 1, valid accuracy 0.8026
Epoch 366, CIFAR-10 Batch 5:  loss 0.000453, train_accuracy 1, valid accuracy 0.8078
Epoch 367, CIFAR-10 Batch 1:  loss 0.000647, train_accuracy 1, valid accuracy 0.8056
Epoch 367, CIFAR-10 Batch 2:  loss 0.004358, train_accuracy 1, valid accuracy 0.7864
Epoch 367, CIFAR-10 Batch 3:  loss 0.001196, train_accuracy 1, valid accuracy 0.8052
Epoch 367, CIFAR-10 Batch 4:  loss 0.003313, train_accuracy 1, valid accuracy 0.7992
Epoch 367, CIFAR-10 Batch 5:  loss 0.000537, train_accuracy 1, valid accuracy 0.8048
Epoch 368, CIFAR-10 Batch 1:  loss 0.000744, train_accuracy 1, valid accuracy 0.8064
Epoch 368, CIFAR-10 Batch 2:  loss 0.000599, train_accuracy 1, valid accuracy 0.805
Epoch 368, CIFAR-10 Batch 3:  loss 0.000892, train_accuracy 1, valid accuracy 0.809
Epoch 368, CIFAR-10 Batch 4:  loss 0.002171, train_accuracy 1, valid accuracy 0.798
Epoch 368, CIFAR-10 Batch 5:  loss 0.001365, train_accuracy 1, valid accuracy 0.8096
Epoch 369, CIFAR-10 Batch 1:  loss 0.000693, train_accuracy 1, valid accuracy 0.8112
Epoch 369, CIFAR-10 Batch 2:  loss 0.002661, train_accuracy 1, valid accuracy 0.807
Epoch 369, CIFAR-10 Batch 3:  loss 0.001380, train_accuracy 1, valid accuracy 0.8116
Epoch 369, CIFAR-10 Batch 4:  loss 0.002227, train_accuracy 1, valid accuracy 0.8034
Epoch 369, CIFAR-10 Batch 5:  loss 0.004325, train_accuracy 1, valid accuracy 0.7848
Epoch 370, CIFAR-10 Batch 1:  loss 0.000224, train_accuracy 1, valid accuracy 0.805
Epoch 370, CIFAR-10 Batch 2:  loss 0.001797, train_accuracy 1, valid accuracy 0.8036
Epoch 370, CIFAR-10 Batch 3:  loss 0.000579, train_accuracy 1, valid accuracy 0.8086
Epoch 370, CIFAR-10 Batch 4:  loss 0.000813, train_accuracy 1, valid accuracy 0.8054
Epoch 370, CIFAR-10 Batch 5:  loss 0.007627, train_accuracy 1, valid accuracy 0.7964
Epoch 371, CIFAR-10 Batch 1:  loss 0.000299, train_accuracy 1, valid accuracy 0.8166
Epoch 371, CIFAR-10 Batch 2:  loss 0.001739, train_accuracy 1, valid accuracy 0.7966
Epoch 371, CIFAR-10 Batch 3:  loss 0.000670, train_accuracy 1, valid accuracy 0.7988
Epoch 371, CIFAR-10 Batch 4:  loss 0.001229, train_accuracy 1, valid accuracy 0.8058
Epoch 371, CIFAR-10 Batch 5:  loss 0.000743, train_accuracy 1, valid accuracy 0.7934
Epoch 372, CIFAR-10 Batch 1:  loss 0.000480, train_accuracy 1, valid accuracy 0.806
Epoch 372, CIFAR-10 Batch 2:  loss 0.001994, train_accuracy 1, valid accuracy 0.8002
Epoch 372, CIFAR-10 Batch 3:  loss 0.001136, train_accuracy 1, valid accuracy 0.8042
Epoch 372, CIFAR-10 Batch 4:  loss 0.002272, train_accuracy 1, valid accuracy 0.8104
Epoch 372, CIFAR-10 Batch 5:  loss 0.001313, train_accuracy 1, valid accuracy 0.804
Epoch 373, CIFAR-10 Batch 1:  loss 0.000378, train_accuracy 1, valid accuracy 0.8116
Epoch 373, CIFAR-10 Batch 2:  loss 0.007029, train_accuracy 1, valid accuracy 0.7986
Epoch 373, CIFAR-10 Batch 3:  loss 0.002063, train_accuracy 1, valid accuracy 0.8076
Epoch 373, CIFAR-10 Batch 4:  loss 0.000945, train_accuracy 1, valid accuracy 0.8038
Epoch 373, CIFAR-10 Batch 5:  loss 0.000639, train_accuracy 1, valid accuracy 0.815
Epoch 374, CIFAR-10 Batch 1:  loss 0.001400, train_accuracy 1, valid accuracy 0.8098
Epoch 374, CIFAR-10 Batch 2:  loss 0.000739, train_accuracy 1, valid accuracy 0.7956
Epoch 374, CIFAR-10 Batch 3:  loss 0.002670, train_accuracy 1, valid accuracy 0.8144
Epoch 374, CIFAR-10 Batch 4:  loss 0.001582, train_accuracy 1, valid accuracy 0.8032
Epoch 374, CIFAR-10 Batch 5:  loss 0.000505, train_accuracy 1, valid accuracy 0.8012
Epoch 375, CIFAR-10 Batch 1:  loss 0.000730, train_accuracy 1, valid accuracy 0.8058
Epoch 375, CIFAR-10 Batch 2:  loss 0.001274, train_accuracy 1, valid accuracy 0.8026
Epoch 375, CIFAR-10 Batch 3:  loss 0.000909, train_accuracy 1, valid accuracy 0.806
Epoch 375, CIFAR-10 Batch 4:  loss 0.002208, train_accuracy 1, valid accuracy 0.8026
Epoch 375, CIFAR-10 Batch 5:  loss 0.000577, train_accuracy 1, valid accuracy 0.804
Epoch 376, CIFAR-10 Batch 1:  loss 0.001070, train_accuracy 1, valid accuracy 0.8084
Epoch 376, CIFAR-10 Batch 2:  loss 0.001355, train_accuracy 1, valid accuracy 0.8082
Epoch 376, CIFAR-10 Batch 3:  loss 0.001276, train_accuracy 1, valid accuracy 0.7974
Epoch 376, CIFAR-10 Batch 4:  loss 0.001064, train_accuracy 1, valid accuracy 0.7928
Epoch 376, CIFAR-10 Batch 5:  loss 0.003417, train_accuracy 1, valid accuracy 0.7902
Epoch 377, CIFAR-10 Batch 1:  loss 0.000651, train_accuracy 1, valid accuracy 0.803
Epoch 377, CIFAR-10 Batch 2:  loss 0.002309, train_accuracy 1, valid accuracy 0.799
Epoch 377, CIFAR-10 Batch 3:  loss 0.000260, train_accuracy 1, valid accuracy 0.8048
Epoch 377, CIFAR-10 Batch 4:  loss 0.000302, train_accuracy 1, valid accuracy 0.8072
Epoch 377, CIFAR-10 Batch 5:  loss 0.000425, train_accuracy 1, valid accuracy 0.8038
Epoch 378, CIFAR-10 Batch 1:  loss 0.000863, train_accuracy 1, valid accuracy 0.8054
Epoch 378, CIFAR-10 Batch 2:  loss 0.007873, train_accuracy 1, valid accuracy 0.7998
Epoch 378, CIFAR-10 Batch 3:  loss 0.000667, train_accuracy 1, valid accuracy 0.7984
Epoch 378, CIFAR-10 Batch 4:  loss 0.002745, train_accuracy 1, valid accuracy 0.8074
Epoch 378, CIFAR-10 Batch 5:  loss 0.000424, train_accuracy 1, valid accuracy 0.8026
Epoch 379, CIFAR-10 Batch 1:  loss 0.000598, train_accuracy 1, valid accuracy 0.798
Epoch 379, CIFAR-10 Batch 2:  loss 0.001031, train_accuracy 1, valid accuracy 0.8042
Epoch 379, CIFAR-10 Batch 3:  loss 0.002254, train_accuracy 1, valid accuracy 0.7926
Epoch 379, CIFAR-10 Batch 4:  loss 0.002511, train_accuracy 1, valid accuracy 0.8022
Epoch 379, CIFAR-10 Batch 5:  loss 0.000758, train_accuracy 1, valid accuracy 0.8062
Epoch 380, CIFAR-10 Batch 1:  loss 0.000795, train_accuracy 1, valid accuracy 0.8064
Epoch 380, CIFAR-10 Batch 2:  loss 0.004044, train_accuracy 1, valid accuracy 0.8006
Epoch 380, CIFAR-10 Batch 3:  loss 0.001328, train_accuracy 1, valid accuracy 0.7968
Epoch 380, CIFAR-10 Batch 4:  loss 0.002217, train_accuracy 1, valid accuracy 0.8038
Epoch 380, CIFAR-10 Batch 5:  loss 0.003779, train_accuracy 1, valid accuracy 0.7944
Epoch 381, CIFAR-10 Batch 1:  loss 0.001117, train_accuracy 1, valid accuracy 0.801
Epoch 381, CIFAR-10 Batch 2:  loss 0.001746, train_accuracy 1, valid accuracy 0.7876
Epoch 381, CIFAR-10 Batch 3:  loss 0.000897, train_accuracy 1, valid accuracy 0.8006
Epoch 381, CIFAR-10 Batch 4:  loss 0.001175, train_accuracy 1, valid accuracy 0.8058
Epoch 381, CIFAR-10 Batch 5:  loss 0.003854, train_accuracy 1, valid accuracy 0.7934
Epoch 382, CIFAR-10 Batch 1:  loss 0.001377, train_accuracy 1, valid accuracy 0.8012
Epoch 382, CIFAR-10 Batch 2:  loss 0.002545, train_accuracy 1, valid accuracy 0.797
Epoch 382, CIFAR-10 Batch 3:  loss 0.000334, train_accuracy 1, valid accuracy 0.8066
Epoch 382, CIFAR-10 Batch 4:  loss 0.001119, train_accuracy 1, valid accuracy 0.799
Epoch 382, CIFAR-10 Batch 5:  loss 0.000449, train_accuracy 1, valid accuracy 0.8054
Epoch 383, CIFAR-10 Batch 1:  loss 0.000342, train_accuracy 1, valid accuracy 0.804
Epoch 383, CIFAR-10 Batch 2:  loss 0.000771, train_accuracy 1, valid accuracy 0.8018
Epoch 383, CIFAR-10 Batch 3:  loss 0.000294, train_accuracy 1, valid accuracy 0.8032
Epoch 383, CIFAR-10 Batch 4:  loss 0.001296, train_accuracy 1, valid accuracy 0.7998
Epoch 383, CIFAR-10 Batch 5:  loss 0.001225, train_accuracy 1, valid accuracy 0.7976
Epoch 384, CIFAR-10 Batch 1:  loss 0.000649, train_accuracy 1, valid accuracy 0.8066
Epoch 384, CIFAR-10 Batch 2:  loss 0.001036, train_accuracy 1, valid accuracy 0.7952
Epoch 384, CIFAR-10 Batch 3:  loss 0.001162, train_accuracy 1, valid accuracy 0.8064
Epoch 384, CIFAR-10 Batch 4:  loss 0.001838, train_accuracy 1, valid accuracy 0.8098
Epoch 384, CIFAR-10 Batch 5:  loss 0.002108, train_accuracy 1, valid accuracy 0.7988
Epoch 385, CIFAR-10 Batch 1:  loss 0.000851, train_accuracy 1, valid accuracy 0.806
Epoch 385, CIFAR-10 Batch 2:  loss 0.002127, train_accuracy 1, valid accuracy 0.8014
Epoch 385, CIFAR-10 Batch 3:  loss 0.001148, train_accuracy 1, valid accuracy 0.8062
Epoch 385, CIFAR-10 Batch 4:  loss 0.001162, train_accuracy 1, valid accuracy 0.802
Epoch 385, CIFAR-10 Batch 5:  loss 0.018321, train_accuracy 1, valid accuracy 0.7728
Epoch 386, CIFAR-10 Batch 1:  loss 0.000481, train_accuracy 1, valid accuracy 0.8028
Epoch 386, CIFAR-10 Batch 2:  loss 0.000794, train_accuracy 1, valid accuracy 0.7968
Epoch 386, CIFAR-10 Batch 3:  loss 0.001132, train_accuracy 1, valid accuracy 0.7996
Epoch 386, CIFAR-10 Batch 4:  loss 0.001368, train_accuracy 1, valid accuracy 0.8044
Epoch 386, CIFAR-10 Batch 5:  loss 0.000721, train_accuracy 1, valid accuracy 0.7966
Epoch 387, CIFAR-10 Batch 1:  loss 0.001079, train_accuracy 1, valid accuracy 0.8044
Epoch 387, CIFAR-10 Batch 2:  loss 0.002011, train_accuracy 1, valid accuracy 0.7932
Epoch 387, CIFAR-10 Batch 3:  loss 0.000308, train_accuracy 1, valid accuracy 0.8042
Epoch 387, CIFAR-10 Batch 4:  loss 0.001377, train_accuracy 1, valid accuracy 0.8044
Epoch 387, CIFAR-10 Batch 5:  loss 0.001465, train_accuracy 1, valid accuracy 0.8004
Epoch 388, CIFAR-10 Batch 1:  loss 0.000723, train_accuracy 1, valid accuracy 0.8062
Epoch 388, CIFAR-10 Batch 2:  loss 0.000327, train_accuracy 1, valid accuracy 0.8062
Epoch 388, CIFAR-10 Batch 3:  loss 0.000330, train_accuracy 1, valid accuracy 0.7958
Epoch 388, CIFAR-10 Batch 4:  loss 0.000979, train_accuracy 1, valid accuracy 0.8094
Epoch 388, CIFAR-10 Batch 5:  loss 0.001558, train_accuracy 1, valid accuracy 0.8018
Epoch 389, CIFAR-10 Batch 1:  loss 0.001154, train_accuracy 1, valid accuracy 0.802
Epoch 389, CIFAR-10 Batch 2:  loss 0.000201, train_accuracy 1, valid accuracy 0.8074
Epoch 389, CIFAR-10 Batch 3:  loss 0.000205, train_accuracy 1, valid accuracy 0.806
Epoch 389, CIFAR-10 Batch 4:  loss 0.001734, train_accuracy 1, valid accuracy 0.805
Epoch 389, CIFAR-10 Batch 5:  loss 0.000769, train_accuracy 1, valid accuracy 0.7926
Epoch 390, CIFAR-10 Batch 1:  loss 0.000387, train_accuracy 1, valid accuracy 0.8062
Epoch 390, CIFAR-10 Batch 2:  loss 0.001827, train_accuracy 1, valid accuracy 0.7982
Epoch 390, CIFAR-10 Batch 3:  loss 0.000558, train_accuracy 1, valid accuracy 0.8006
Epoch 390, CIFAR-10 Batch 4:  loss 0.000979, train_accuracy 1, valid accuracy 0.8026
Epoch 390, CIFAR-10 Batch 5:  loss 0.001734, train_accuracy 1, valid accuracy 0.7966
Epoch 391, CIFAR-10 Batch 1:  loss 0.000402, train_accuracy 1, valid accuracy 0.8074
Epoch 391, CIFAR-10 Batch 2:  loss 0.000512, train_accuracy 1, valid accuracy 0.8058
Epoch 391, CIFAR-10 Batch 3:  loss 0.000420, train_accuracy 1, valid accuracy 0.8066
Epoch 391, CIFAR-10 Batch 4:  loss 0.001541, train_accuracy 1, valid accuracy 0.8078
Epoch 391, CIFAR-10 Batch 5:  loss 0.000411, train_accuracy 1, valid accuracy 0.8102
Epoch 392, CIFAR-10 Batch 1:  loss 0.000739, train_accuracy 1, valid accuracy 0.7952
Epoch 392, CIFAR-10 Batch 2:  loss 0.000697, train_accuracy 1, valid accuracy 0.804
Epoch 392, CIFAR-10 Batch 3:  loss 0.000829, train_accuracy 1, valid accuracy 0.8056
Epoch 392, CIFAR-10 Batch 4:  loss 0.001583, train_accuracy 1, valid accuracy 0.8034
Epoch 392, CIFAR-10 Batch 5:  loss 0.000353, train_accuracy 1, valid accuracy 0.8066
Epoch 393, CIFAR-10 Batch 1:  loss 0.000990, train_accuracy 1, valid accuracy 0.7992
Epoch 393, CIFAR-10 Batch 2:  loss 0.000881, train_accuracy 1, valid accuracy 0.7992
Epoch 393, CIFAR-10 Batch 3:  loss 0.000910, train_accuracy 1, valid accuracy 0.8018
Epoch 393, CIFAR-10 Batch 4:  loss 0.001681, train_accuracy 1, valid accuracy 0.8032
Epoch 393, CIFAR-10 Batch 5:  loss 0.000242, train_accuracy 1, valid accuracy 0.8056
Epoch 394, CIFAR-10 Batch 1:  loss 0.000524, train_accuracy 1, valid accuracy 0.8098
Epoch 394, CIFAR-10 Batch 2:  loss 0.001264, train_accuracy 1, valid accuracy 0.808
Epoch 394, CIFAR-10 Batch 3:  loss 0.001144, train_accuracy 1, valid accuracy 0.7992
Epoch 394, CIFAR-10 Batch 4:  loss 0.001961, train_accuracy 1, valid accuracy 0.7936
Epoch 394, CIFAR-10 Batch 5:  loss 0.002525, train_accuracy 1, valid accuracy 0.7972
Epoch 395, CIFAR-10 Batch 1:  loss 0.000679, train_accuracy 1, valid accuracy 0.8074
Epoch 395, CIFAR-10 Batch 2:  loss 0.004278, train_accuracy 1, valid accuracy 0.7924
Epoch 395, CIFAR-10 Batch 3:  loss 0.000528, train_accuracy 1, valid accuracy 0.7956
Epoch 395, CIFAR-10 Batch 4:  loss 0.002812, train_accuracy 1, valid accuracy 0.797
Epoch 395, CIFAR-10 Batch 5:  loss 0.000427, train_accuracy 1, valid accuracy 0.8004
Epoch 396, CIFAR-10 Batch 1:  loss 0.000291, train_accuracy 1, valid accuracy 0.8054
Epoch 396, CIFAR-10 Batch 2:  loss 0.000769, train_accuracy 1, valid accuracy 0.7912
Epoch 396, CIFAR-10 Batch 3:  loss 0.000581, train_accuracy 1, valid accuracy 0.7952
Epoch 396, CIFAR-10 Batch 4:  loss 0.000639, train_accuracy 1, valid accuracy 0.7968
Epoch 396, CIFAR-10 Batch 5:  loss 0.000705, train_accuracy 1, valid accuracy 0.793
Epoch 397, CIFAR-10 Batch 1:  loss 0.000650, train_accuracy 1, valid accuracy 0.808
Epoch 397, CIFAR-10 Batch 2:  loss 0.001522, train_accuracy 1, valid accuracy 0.8014
Epoch 397, CIFAR-10 Batch 3:  loss 0.000175, train_accuracy 1, valid accuracy 0.7962
Epoch 397, CIFAR-10 Batch 4:  loss 0.001365, train_accuracy 1, valid accuracy 0.7988
Epoch 397, CIFAR-10 Batch 5:  loss 0.002032, train_accuracy 1, valid accuracy 0.8082
Epoch 398, CIFAR-10 Batch 1:  loss 0.000624, train_accuracy 1, valid accuracy 0.809
Epoch 398, CIFAR-10 Batch 2:  loss 0.000414, train_accuracy 1, valid accuracy 0.7994
Epoch 398, CIFAR-10 Batch 3:  loss 0.000495, train_accuracy 1, valid accuracy 0.792
Epoch 398, CIFAR-10 Batch 4:  loss 0.000710, train_accuracy 1, valid accuracy 0.8044
Epoch 398, CIFAR-10 Batch 5:  loss 0.000595, train_accuracy 1, valid accuracy 0.8104
Epoch 399, CIFAR-10 Batch 1:  loss 0.001179, train_accuracy 1, valid accuracy 0.8096
Epoch 399, CIFAR-10 Batch 2:  loss 0.001976, train_accuracy 1, valid accuracy 0.7942
Epoch 399, CIFAR-10 Batch 3:  loss 0.000210, train_accuracy 1, valid accuracy 0.8054
Epoch 399, CIFAR-10 Batch 4:  loss 0.000944, train_accuracy 1, valid accuracy 0.7974
Epoch 399, CIFAR-10 Batch 5:  loss 0.001107, train_accuracy 1, valid accuracy 0.7986
Epoch 400, CIFAR-10 Batch 1:  loss 0.000529, train_accuracy 1, valid accuracy 0.8064
Epoch 400, CIFAR-10 Batch 2:  loss 0.000414, train_accuracy 1, valid accuracy 0.8088
Epoch 400, CIFAR-10 Batch 3:  loss 0.001068, train_accuracy 1, valid accuracy 0.8026
Epoch 400, CIFAR-10 Batch 4:  loss 0.001821, train_accuracy 1, valid accuracy 0.8004
Epoch 400, CIFAR-10 Batch 5:  loss 0.001128, train_accuracy 1, valid accuracy 0.8068
Epoch 401, CIFAR-10 Batch 1:  loss 0.000690, train_accuracy 1, valid accuracy 0.8068
Epoch 401, CIFAR-10 Batch 2:  loss 0.000609, train_accuracy 1, valid accuracy 0.807
Epoch 401, CIFAR-10 Batch 3:  loss 0.002669, train_accuracy 1, valid accuracy 0.7952
Epoch 401, CIFAR-10 Batch 4:  loss 0.002475, train_accuracy 1, valid accuracy 0.797
Epoch 401, CIFAR-10 Batch 5:  loss 0.001004, train_accuracy 1, valid accuracy 0.7938
Epoch 402, CIFAR-10 Batch 1:  loss 0.000697, train_accuracy 1, valid accuracy 0.8018
Epoch 402, CIFAR-10 Batch 2:  loss 0.002274, train_accuracy 1, valid accuracy 0.7964
Epoch 402, CIFAR-10 Batch 3:  loss 0.020076, train_accuracy 0.975, valid accuracy 0.7936
Epoch 402, CIFAR-10 Batch 4:  loss 0.005178, train_accuracy 1, valid accuracy 0.8038
Epoch 402, CIFAR-10 Batch 5:  loss 0.000736, train_accuracy 1, valid accuracy 0.802
Epoch 403, CIFAR-10 Batch 1:  loss 0.002028, train_accuracy 1, valid accuracy 0.8098
Epoch 403, CIFAR-10 Batch 2:  loss 0.000407, train_accuracy 1, valid accuracy 0.8026
Epoch 403, CIFAR-10 Batch 3:  loss 0.000595, train_accuracy 1, valid accuracy 0.8032
Epoch 403, CIFAR-10 Batch 4:  loss 0.001172, train_accuracy 1, valid accuracy 0.7952
Epoch 403, CIFAR-10 Batch 5:  loss 0.001435, train_accuracy 1, valid accuracy 0.7918
Epoch 404, CIFAR-10 Batch 1:  loss 0.001880, train_accuracy 1, valid accuracy 0.7932
Epoch 404, CIFAR-10 Batch 2:  loss 0.001239, train_accuracy 1, valid accuracy 0.8102
Epoch 404, CIFAR-10 Batch 3:  loss 0.000672, train_accuracy 1, valid accuracy 0.8004
Epoch 404, CIFAR-10 Batch 4:  loss 0.000648, train_accuracy 1, valid accuracy 0.8042
Epoch 404, CIFAR-10 Batch 5:  loss 0.000590, train_accuracy 1, valid accuracy 0.811
Epoch 405, CIFAR-10 Batch 1:  loss 0.002005, train_accuracy 1, valid accuracy 0.8024
Epoch 405, CIFAR-10 Batch 2:  loss 0.000959, train_accuracy 1, valid accuracy 0.7976
Epoch 405, CIFAR-10 Batch 3:  loss 0.000518, train_accuracy 1, valid accuracy 0.8052
Epoch 405, CIFAR-10 Batch 4:  loss 0.001151, train_accuracy 1, valid accuracy 0.8054
Epoch 405, CIFAR-10 Batch 5:  loss 0.002045, train_accuracy 1, valid accuracy 0.7986
Epoch 406, CIFAR-10 Batch 1:  loss 0.000651, train_accuracy 1, valid accuracy 0.807
Epoch 406, CIFAR-10 Batch 2:  loss 0.001643, train_accuracy 1, valid accuracy 0.7964
Epoch 406, CIFAR-10 Batch 3:  loss 0.000982, train_accuracy 1, valid accuracy 0.8018
Epoch 406, CIFAR-10 Batch 4:  loss 0.001279, train_accuracy 1, valid accuracy 0.7968
Epoch 406, CIFAR-10 Batch 5:  loss 0.000668, train_accuracy 1, valid accuracy 0.8044
Epoch 407, CIFAR-10 Batch 1:  loss 0.000817, train_accuracy 1, valid accuracy 0.8034
Epoch 407, CIFAR-10 Batch 2:  loss 0.001301, train_accuracy 1, valid accuracy 0.8042
Epoch 407, CIFAR-10 Batch 3:  loss 0.000295, train_accuracy 1, valid accuracy 0.81
Epoch 407, CIFAR-10 Batch 4:  loss 0.001787, train_accuracy 1, valid accuracy 0.8032
Epoch 407, CIFAR-10 Batch 5:  loss 0.000944, train_accuracy 1, valid accuracy 0.806
Epoch 408, CIFAR-10 Batch 1:  loss 0.000534, train_accuracy 1, valid accuracy 0.8112
Epoch 408, CIFAR-10 Batch 2:  loss 0.002888, train_accuracy 1, valid accuracy 0.792
Epoch 408, CIFAR-10 Batch 3:  loss 0.001842, train_accuracy 1, valid accuracy 0.7994
Epoch 408, CIFAR-10 Batch 4:  loss 0.000912, train_accuracy 1, valid accuracy 0.8112
Epoch 408, CIFAR-10 Batch 5:  loss 0.000363, train_accuracy 1, valid accuracy 0.8026
Epoch 409, CIFAR-10 Batch 1:  loss 0.000450, train_accuracy 1, valid accuracy 0.7972
Epoch 409, CIFAR-10 Batch 2:  loss 0.000517, train_accuracy 1, valid accuracy 0.8078
Epoch 409, CIFAR-10 Batch 3:  loss 0.001701, train_accuracy 1, valid accuracy 0.8038
Epoch 409, CIFAR-10 Batch 4:  loss 0.001052, train_accuracy 1, valid accuracy 0.795
Epoch 409, CIFAR-10 Batch 5:  loss 0.000747, train_accuracy 1, valid accuracy 0.7972
Epoch 410, CIFAR-10 Batch 1:  loss 0.000680, train_accuracy 1, valid accuracy 0.8058
Epoch 410, CIFAR-10 Batch 2:  loss 0.000467, train_accuracy 1, valid accuracy 0.8078
Epoch 410, CIFAR-10 Batch 3:  loss 0.002402, train_accuracy 1, valid accuracy 0.802
Epoch 410, CIFAR-10 Batch 4:  loss 0.001578, train_accuracy 1, valid accuracy 0.8106
Epoch 410, CIFAR-10 Batch 5:  loss 0.000671, train_accuracy 1, valid accuracy 0.7974
Epoch 411, CIFAR-10 Batch 1:  loss 0.000663, train_accuracy 1, valid accuracy 0.8128
Epoch 411, CIFAR-10 Batch 2:  loss 0.000626, train_accuracy 1, valid accuracy 0.806
Epoch 411, CIFAR-10 Batch 3:  loss 0.001102, train_accuracy 1, valid accuracy 0.8036
Epoch 411, CIFAR-10 Batch 4:  loss 0.001425, train_accuracy 1, valid accuracy 0.8042
Epoch 411, CIFAR-10 Batch 5:  loss 0.000235, train_accuracy 1, valid accuracy 0.8082
Epoch 412, CIFAR-10 Batch 1:  loss 0.001335, train_accuracy 1, valid accuracy 0.8036
Epoch 412, CIFAR-10 Batch 2:  loss 0.000216, train_accuracy 1, valid accuracy 0.8064
Epoch 412, CIFAR-10 Batch 3:  loss 0.001251, train_accuracy 1, valid accuracy 0.797
Epoch 412, CIFAR-10 Batch 4:  loss 0.001038, train_accuracy 1, valid accuracy 0.8086
Epoch 412, CIFAR-10 Batch 5:  loss 0.000401, train_accuracy 1, valid accuracy 0.7998
Epoch 413, CIFAR-10 Batch 1:  loss 0.000518, train_accuracy 1, valid accuracy 0.8068
Epoch 413, CIFAR-10 Batch 2:  loss 0.001029, train_accuracy 1, valid accuracy 0.7956
Epoch 413, CIFAR-10 Batch 3:  loss 0.001265, train_accuracy 1, valid accuracy 0.8
Epoch 413, CIFAR-10 Batch 4:  loss 0.002791, train_accuracy 1, valid accuracy 0.8078
Epoch 413, CIFAR-10 Batch 5:  loss 0.000481, train_accuracy 1, valid accuracy 0.8058
Epoch 414, CIFAR-10 Batch 1:  loss 0.000496, train_accuracy 1, valid accuracy 0.8058
Epoch 414, CIFAR-10 Batch 2:  loss 0.006153, train_accuracy 1, valid accuracy 0.791
Epoch 414, CIFAR-10 Batch 3:  loss 0.000356, train_accuracy 1, valid accuracy 0.8036
Epoch 414, CIFAR-10 Batch 4:  loss 0.000919, train_accuracy 1, valid accuracy 0.8052
Epoch 414, CIFAR-10 Batch 5:  loss 0.000388, train_accuracy 1, valid accuracy 0.8014
Epoch 415, CIFAR-10 Batch 1:  loss 0.000262, train_accuracy 1, valid accuracy 0.8038
Epoch 415, CIFAR-10 Batch 2:  loss 0.000392, train_accuracy 1, valid accuracy 0.8074
Epoch 415, CIFAR-10 Batch 3:  loss 0.001485, train_accuracy 1, valid accuracy 0.7914
Epoch 415, CIFAR-10 Batch 4:  loss 0.000895, train_accuracy 1, valid accuracy 0.8026
Epoch 415, CIFAR-10 Batch 5:  loss 0.002265, train_accuracy 1, valid accuracy 0.7992
Epoch 416, CIFAR-10 Batch 1:  loss 0.000873, train_accuracy 1, valid accuracy 0.805
Epoch 416, CIFAR-10 Batch 2:  loss 0.000542, train_accuracy 1, valid accuracy 0.8058
Epoch 416, CIFAR-10 Batch 3:  loss 0.000129, train_accuracy 1, valid accuracy 0.8058
Epoch 416, CIFAR-10 Batch 4:  loss 0.001432, train_accuracy 1, valid accuracy 0.7936
Epoch 416, CIFAR-10 Batch 5:  loss 0.000564, train_accuracy 1, valid accuracy 0.8066
Epoch 417, CIFAR-10 Batch 1:  loss 0.001010, train_accuracy 1, valid accuracy 0.8002
Epoch 417, CIFAR-10 Batch 2:  loss 0.001182, train_accuracy 1, valid accuracy 0.8014
Epoch 417, CIFAR-10 Batch 3:  loss 0.000298, train_accuracy 1, valid accuracy 0.812
Epoch 417, CIFAR-10 Batch 4:  loss 0.000782, train_accuracy 1, valid accuracy 0.8092
Epoch 417, CIFAR-10 Batch 5:  loss 0.000569, train_accuracy 1, valid accuracy 0.805
Epoch 418, CIFAR-10 Batch 1:  loss 0.001166, train_accuracy 1, valid accuracy 0.79
Epoch 418, CIFAR-10 Batch 2:  loss 0.000187, train_accuracy 1, valid accuracy 0.7984
Epoch 418, CIFAR-10 Batch 3:  loss 0.000500, train_accuracy 1, valid accuracy 0.8022
Epoch 418, CIFAR-10 Batch 4:  loss 0.000517, train_accuracy 1, valid accuracy 0.8052
Epoch 418, CIFAR-10 Batch 5:  loss 0.000215, train_accuracy 1, valid accuracy 0.8078
Epoch 419, CIFAR-10 Batch 1:  loss 0.000302, train_accuracy 1, valid accuracy 0.8108
Epoch 419, CIFAR-10 Batch 2:  loss 0.001472, train_accuracy 1, valid accuracy 0.7962
Epoch 419, CIFAR-10 Batch 3:  loss 0.000479, train_accuracy 1, valid accuracy 0.805
Epoch 419, CIFAR-10 Batch 4:  loss 0.004737, train_accuracy 1, valid accuracy 0.8048
Epoch 419, CIFAR-10 Batch 5:  loss 0.000091, train_accuracy 1, valid accuracy 0.8038
Epoch 420, CIFAR-10 Batch 1:  loss 0.001515, train_accuracy 1, valid accuracy 0.7996
Epoch 420, CIFAR-10 Batch 2:  loss 0.001321, train_accuracy 1, valid accuracy 0.797
Epoch 420, CIFAR-10 Batch 3:  loss 0.000608, train_accuracy 1, valid accuracy 0.7994
Epoch 420, CIFAR-10 Batch 4:  loss 0.000913, train_accuracy 1, valid accuracy 0.8082
Epoch 420, CIFAR-10 Batch 5:  loss 0.000452, train_accuracy 1, valid accuracy 0.799
Epoch 421, CIFAR-10 Batch 1:  loss 0.001024, train_accuracy 1, valid accuracy 0.8092
Epoch 421, CIFAR-10 Batch 2:  loss 0.000634, train_accuracy 1, valid accuracy 0.8064
Epoch 421, CIFAR-10 Batch 3:  loss 0.000551, train_accuracy 1, valid accuracy 0.8066
Epoch 421, CIFAR-10 Batch 4:  loss 0.001356, train_accuracy 1, valid accuracy 0.8032
Epoch 421, CIFAR-10 Batch 5:  loss 0.000428, train_accuracy 1, valid accuracy 0.8096
Epoch 422, CIFAR-10 Batch 1:  loss 0.000612, train_accuracy 1, valid accuracy 0.814
Epoch 422, CIFAR-10 Batch 2:  loss 0.001959, train_accuracy 1, valid accuracy 0.7858
Epoch 422, CIFAR-10 Batch 3:  loss 0.006481, train_accuracy 1, valid accuracy 0.8034
Epoch 422, CIFAR-10 Batch 4:  loss 0.011520, train_accuracy 1, valid accuracy 0.7934
Epoch 422, CIFAR-10 Batch 5:  loss 0.005725, train_accuracy 1, valid accuracy 0.7862
Epoch 423, CIFAR-10 Batch 1:  loss 0.000902, train_accuracy 1, valid accuracy 0.8026
Epoch 423, CIFAR-10 Batch 2:  loss 0.000601, train_accuracy 1, valid accuracy 0.7948
Epoch 423, CIFAR-10 Batch 3:  loss 0.001071, train_accuracy 1, valid accuracy 0.804
Epoch 423, CIFAR-10 Batch 4:  loss 0.001017, train_accuracy 1, valid accuracy 0.804
Epoch 423, CIFAR-10 Batch 5:  loss 0.000950, train_accuracy 1, valid accuracy 0.8088
Epoch 424, CIFAR-10 Batch 1:  loss 0.002431, train_accuracy 1, valid accuracy 0.8028
Epoch 424, CIFAR-10 Batch 2:  loss 0.000682, train_accuracy 1, valid accuracy 0.8046
Epoch 424, CIFAR-10 Batch 3:  loss 0.000233, train_accuracy 1, valid accuracy 0.806
Epoch 424, CIFAR-10 Batch 4:  loss 0.000695, train_accuracy 1, valid accuracy 0.8052
Epoch 424, CIFAR-10 Batch 5:  loss 0.000844, train_accuracy 1, valid accuracy 0.8012
Epoch 425, CIFAR-10 Batch 1:  loss 0.000592, train_accuracy 1, valid accuracy 0.806
Epoch 425, CIFAR-10 Batch 2:  loss 0.000721, train_accuracy 1, valid accuracy 0.8066
Epoch 425, CIFAR-10 Batch 3:  loss 0.000969, train_accuracy 1, valid accuracy 0.8002
Epoch 425, CIFAR-10 Batch 4:  loss 0.001084, train_accuracy 1, valid accuracy 0.8108
Epoch 425, CIFAR-10 Batch 5:  loss 0.000270, train_accuracy 1, valid accuracy 0.8074
Epoch 426, CIFAR-10 Batch 1:  loss 0.000417, train_accuracy 1, valid accuracy 0.8116
Epoch 426, CIFAR-10 Batch 2:  loss 0.000670, train_accuracy 1, valid accuracy 0.808
Epoch 426, CIFAR-10 Batch 3:  loss 0.000714, train_accuracy 1, valid accuracy 0.8026
Epoch 426, CIFAR-10 Batch 4:  loss 0.000838, train_accuracy 1, valid accuracy 0.8054
Epoch 426, CIFAR-10 Batch 5:  loss 0.000369, train_accuracy 1, valid accuracy 0.807
Epoch 427, CIFAR-10 Batch 1:  loss 0.000578, train_accuracy 1, valid accuracy 0.8098
Epoch 427, CIFAR-10 Batch 2:  loss 0.000499, train_accuracy 1, valid accuracy 0.8004
Epoch 427, CIFAR-10 Batch 3:  loss 0.000200, train_accuracy 1, valid accuracy 0.8052
Epoch 427, CIFAR-10 Batch 4:  loss 0.000679, train_accuracy 1, valid accuracy 0.8054
Epoch 427, CIFAR-10 Batch 5:  loss 0.000230, train_accuracy 1, valid accuracy 0.8048
Epoch 428, CIFAR-10 Batch 1:  loss 0.000866, train_accuracy 1, valid accuracy 0.8058
Epoch 428, CIFAR-10 Batch 2:  loss 0.000277, train_accuracy 1, valid accuracy 0.8034
Epoch 428, CIFAR-10 Batch 3:  loss 0.001010, train_accuracy 1, valid accuracy 0.796
Epoch 428, CIFAR-10 Batch 4:  loss 0.000355, train_accuracy 1, valid accuracy 0.8032
Epoch 428, CIFAR-10 Batch 5:  loss 0.000302, train_accuracy 1, valid accuracy 0.801
Epoch 429, CIFAR-10 Batch 1:  loss 0.001441, train_accuracy 1, valid accuracy 0.7938
Epoch 429, CIFAR-10 Batch 2:  loss 0.000559, train_accuracy 1, valid accuracy 0.8006
Epoch 429, CIFAR-10 Batch 3:  loss 0.000321, train_accuracy 1, valid accuracy 0.8008
Epoch 429, CIFAR-10 Batch 4:  loss 0.001180, train_accuracy 1, valid accuracy 0.807
Epoch 429, CIFAR-10 Batch 5:  loss 0.001507, train_accuracy 1, valid accuracy 0.8
Epoch 430, CIFAR-10 Batch 1:  loss 0.000835, train_accuracy 1, valid accuracy 0.806
Epoch 430, CIFAR-10 Batch 2:  loss 0.000811, train_accuracy 1, valid accuracy 0.7876
Epoch 430, CIFAR-10 Batch 3:  loss 0.000274, train_accuracy 1, valid accuracy 0.803
Epoch 430, CIFAR-10 Batch 4:  loss 0.000458, train_accuracy 1, valid accuracy 0.8042
Epoch 430, CIFAR-10 Batch 5:  loss 0.001870, train_accuracy 1, valid accuracy 0.7952
Epoch 431, CIFAR-10 Batch 1:  loss 0.000563, train_accuracy 1, valid accuracy 0.805
Epoch 431, CIFAR-10 Batch 2:  loss 0.000518, train_accuracy 1, valid accuracy 0.8058
Epoch 431, CIFAR-10 Batch 3:  loss 0.001143, train_accuracy 1, valid accuracy 0.812
Epoch 431, CIFAR-10 Batch 4:  loss 0.000427, train_accuracy 1, valid accuracy 0.815
Epoch 431, CIFAR-10 Batch 5:  loss 0.000575, train_accuracy 1, valid accuracy 0.7998
Epoch 432, CIFAR-10 Batch 1:  loss 0.000283, train_accuracy 1, valid accuracy 0.8
Epoch 432, CIFAR-10 Batch 2:  loss 0.002934, train_accuracy 1, valid accuracy 0.796
Epoch 432, CIFAR-10 Batch 3:  loss 0.000763, train_accuracy 1, valid accuracy 0.802
Epoch 432, CIFAR-10 Batch 4:  loss 0.001022, train_accuracy 1, valid accuracy 0.7978
Epoch 432, CIFAR-10 Batch 5:  loss 0.000150, train_accuracy 1, valid accuracy 0.8072
Epoch 433, CIFAR-10 Batch 1:  loss 0.000560, train_accuracy 1, valid accuracy 0.8054
Epoch 433, CIFAR-10 Batch 2:  loss 0.001241, train_accuracy 1, valid accuracy 0.8092
Epoch 433, CIFAR-10 Batch 3:  loss 0.000850, train_accuracy 1, valid accuracy 0.8056
Epoch 433, CIFAR-10 Batch 4:  loss 0.000570, train_accuracy 1, valid accuracy 0.8068
Epoch 433, CIFAR-10 Batch 5:  loss 0.000482, train_accuracy 1, valid accuracy 0.807
Epoch 434, CIFAR-10 Batch 1:  loss 0.000803, train_accuracy 1, valid accuracy 0.801
Epoch 434, CIFAR-10 Batch 2:  loss 0.002409, train_accuracy 1, valid accuracy 0.8016
Epoch 434, CIFAR-10 Batch 3:  loss 0.000615, train_accuracy 1, valid accuracy 0.8032
Epoch 434, CIFAR-10 Batch 4:  loss 0.000409, train_accuracy 1, valid accuracy 0.8116
Epoch 434, CIFAR-10 Batch 5:  loss 0.001604, train_accuracy 1, valid accuracy 0.8026
Epoch 435, CIFAR-10 Batch 1:  loss 0.000212, train_accuracy 1, valid accuracy 0.8104
Epoch 435, CIFAR-10 Batch 2:  loss 0.001303, train_accuracy 1, valid accuracy 0.7926
Epoch 435, CIFAR-10 Batch 3:  loss 0.000958, train_accuracy 1, valid accuracy 0.8014
Epoch 435, CIFAR-10 Batch 4:  loss 0.000406, train_accuracy 1, valid accuracy 0.8078
Epoch 435, CIFAR-10 Batch 5:  loss 0.001369, train_accuracy 1, valid accuracy 0.7806
Epoch 436, CIFAR-10 Batch 1:  loss 0.000935, train_accuracy 1, valid accuracy 0.8048
Epoch 436, CIFAR-10 Batch 2:  loss 0.000451, train_accuracy 1, valid accuracy 0.7982
Epoch 436, CIFAR-10 Batch 3:  loss 0.000682, train_accuracy 1, valid accuracy 0.793
Epoch 436, CIFAR-10 Batch 4:  loss 0.001037, train_accuracy 1, valid accuracy 0.7952
Epoch 436, CIFAR-10 Batch 5:  loss 0.000286, train_accuracy 1, valid accuracy 0.805
Epoch 437, CIFAR-10 Batch 1:  loss 0.000528, train_accuracy 1, valid accuracy 0.807
Epoch 437, CIFAR-10 Batch 2:  loss 0.001077, train_accuracy 1, valid accuracy 0.8012
Epoch 437, CIFAR-10 Batch 3:  loss 0.000644, train_accuracy 1, valid accuracy 0.7952
Epoch 437, CIFAR-10 Batch 4:  loss 0.000247, train_accuracy 1, valid accuracy 0.8046
Epoch 437, CIFAR-10 Batch 5:  loss 0.000509, train_accuracy 1, valid accuracy 0.7846
Epoch 438, CIFAR-10 Batch 1:  loss 0.000407, train_accuracy 1, valid accuracy 0.8012
Epoch 438, CIFAR-10 Batch 2:  loss 0.000647, train_accuracy 1, valid accuracy 0.8022
Epoch 438, CIFAR-10 Batch 3:  loss 0.000911, train_accuracy 1, valid accuracy 0.7946
Epoch 438, CIFAR-10 Batch 4:  loss 0.000393, train_accuracy 1, valid accuracy 0.8034
Epoch 438, CIFAR-10 Batch 5:  loss 0.001223, train_accuracy 1, valid accuracy 0.789
Epoch 439, CIFAR-10 Batch 1:  loss 0.000493, train_accuracy 1, valid accuracy 0.808
Epoch 439, CIFAR-10 Batch 2:  loss 0.000109, train_accuracy 1, valid accuracy 0.8048
Epoch 439, CIFAR-10 Batch 3:  loss 0.000231, train_accuracy 1, valid accuracy 0.8054
Epoch 439, CIFAR-10 Batch 4:  loss 0.000700, train_accuracy 1, valid accuracy 0.7998
Epoch 439, CIFAR-10 Batch 5:  loss 0.000326, train_accuracy 1, valid accuracy 0.8052
Epoch 440, CIFAR-10 Batch 1:  loss 0.000295, train_accuracy 1, valid accuracy 0.8012
Epoch 440, CIFAR-10 Batch 2:  loss 0.000328, train_accuracy 1, valid accuracy 0.8068
Epoch 440, CIFAR-10 Batch 3:  loss 0.000405, train_accuracy 1, valid accuracy 0.7944
Epoch 440, CIFAR-10 Batch 4:  loss 0.000843, train_accuracy 1, valid accuracy 0.7984
Epoch 440, CIFAR-10 Batch 5:  loss 0.000272, train_accuracy 1, valid accuracy 0.799
Epoch 441, CIFAR-10 Batch 1:  loss 0.000665, train_accuracy 1, valid accuracy 0.8014
Epoch 441, CIFAR-10 Batch 2:  loss 0.000445, train_accuracy 1, valid accuracy 0.7984
Epoch 441, CIFAR-10 Batch 3:  loss 0.001021, train_accuracy 1, valid accuracy 0.7962
Epoch 441, CIFAR-10 Batch 4:  loss 0.000421, train_accuracy 1, valid accuracy 0.8044
Epoch 441, CIFAR-10 Batch 5:  loss 0.000840, train_accuracy 1, valid accuracy 0.802
Epoch 442, CIFAR-10 Batch 1:  loss 0.000284, train_accuracy 1, valid accuracy 0.805
Epoch 442, CIFAR-10 Batch 2:  loss 0.000149, train_accuracy 1, valid accuracy 0.8044
Epoch 442, CIFAR-10 Batch 3:  loss 0.000307, train_accuracy 1, valid accuracy 0.8028
Epoch 442, CIFAR-10 Batch 4:  loss 0.000701, train_accuracy 1, valid accuracy 0.8056
Epoch 442, CIFAR-10 Batch 5:  loss 0.000416, train_accuracy 1, valid accuracy 0.8004
Epoch 443, CIFAR-10 Batch 1:  loss 0.000727, train_accuracy 1, valid accuracy 0.8
Epoch 443, CIFAR-10 Batch 2:  loss 0.000369, train_accuracy 1, valid accuracy 0.8038
Epoch 443, CIFAR-10 Batch 3:  loss 0.000548, train_accuracy 1, valid accuracy 0.799
Epoch 443, CIFAR-10 Batch 4:  loss 0.000371, train_accuracy 1, valid accuracy 0.8058
Epoch 443, CIFAR-10 Batch 5:  loss 0.000251, train_accuracy 1, valid accuracy 0.8004
Epoch 444, CIFAR-10 Batch 1:  loss 0.000341, train_accuracy 1, valid accuracy 0.811
Epoch 444, CIFAR-10 Batch 2:  loss 0.000624, train_accuracy 1, valid accuracy 0.8022
Epoch 444, CIFAR-10 Batch 3:  loss 0.003491, train_accuracy 1, valid accuracy 0.792
Epoch 444, CIFAR-10 Batch 4:  loss 0.007713, train_accuracy 1, valid accuracy 0.7872
Epoch 444, CIFAR-10 Batch 5:  loss 0.000401, train_accuracy 1, valid accuracy 0.8074
Epoch 445, CIFAR-10 Batch 1:  loss 0.001077, train_accuracy 1, valid accuracy 0.8004
Epoch 445, CIFAR-10 Batch 2:  loss 0.002994, train_accuracy 1, valid accuracy 0.7892
Epoch 445, CIFAR-10 Batch 3:  loss 0.000391, train_accuracy 1, valid accuracy 0.8
Epoch 445, CIFAR-10 Batch 4:  loss 0.000261, train_accuracy 1, valid accuracy 0.8032
Epoch 445, CIFAR-10 Batch 5:  loss 0.000259, train_accuracy 1, valid accuracy 0.7978
Epoch 446, CIFAR-10 Batch 1:  loss 0.000957, train_accuracy 1, valid accuracy 0.7974
Epoch 446, CIFAR-10 Batch 2:  loss 0.000297, train_accuracy 1, valid accuracy 0.7952
Epoch 446, CIFAR-10 Batch 3:  loss 0.001971, train_accuracy 1, valid accuracy 0.7944
Epoch 446, CIFAR-10 Batch 4:  loss 0.000337, train_accuracy 1, valid accuracy 0.802
Epoch 446, CIFAR-10 Batch 5:  loss 0.000328, train_accuracy 1, valid accuracy 0.8068
Epoch 447, CIFAR-10 Batch 1:  loss 0.000378, train_accuracy 1, valid accuracy 0.803
Epoch 447, CIFAR-10 Batch 2:  loss 0.001038, train_accuracy 1, valid accuracy 0.7938
Epoch 447, CIFAR-10 Batch 3:  loss 0.000287, train_accuracy 1, valid accuracy 0.801
Epoch 447, CIFAR-10 Batch 4:  loss 0.000388, train_accuracy 1, valid accuracy 0.8098
Epoch 447, CIFAR-10 Batch 5:  loss 0.000354, train_accuracy 1, valid accuracy 0.8114
Epoch 448, CIFAR-10 Batch 1:  loss 0.001350, train_accuracy 1, valid accuracy 0.7986
Epoch 448, CIFAR-10 Batch 2:  loss 0.000489, train_accuracy 1, valid accuracy 0.7976
Epoch 448, CIFAR-10 Batch 3:  loss 0.001375, train_accuracy 1, valid accuracy 0.8004
Epoch 448, CIFAR-10 Batch 4:  loss 0.000295, train_accuracy 1, valid accuracy 0.8104
Epoch 448, CIFAR-10 Batch 5:  loss 0.003021, train_accuracy 1, valid accuracy 0.8052
Epoch 449, CIFAR-10 Batch 1:  loss 0.000361, train_accuracy 1, valid accuracy 0.8086
Epoch 449, CIFAR-10 Batch 2:  loss 0.000103, train_accuracy 1, valid accuracy 0.8082
Epoch 449, CIFAR-10 Batch 3:  loss 0.000536, train_accuracy 1, valid accuracy 0.8052
Epoch 449, CIFAR-10 Batch 4:  loss 0.000138, train_accuracy 1, valid accuracy 0.804
Epoch 449, CIFAR-10 Batch 5:  loss 0.000315, train_accuracy 1, valid accuracy 0.81
Epoch 450, CIFAR-10 Batch 1:  loss 0.000275, train_accuracy 1, valid accuracy 0.8126
Epoch 450, CIFAR-10 Batch 2:  loss 0.000259, train_accuracy 1, valid accuracy 0.8022
Epoch 450, CIFAR-10 Batch 3:  loss 0.000435, train_accuracy 1, valid accuracy 0.808
Epoch 450, CIFAR-10 Batch 4:  loss 0.000181, train_accuracy 1, valid accuracy 0.8074
Epoch 450, CIFAR-10 Batch 5:  loss 0.000384, train_accuracy 1, valid accuracy 0.79
Epoch 451, CIFAR-10 Batch 1:  loss 0.000283, train_accuracy 1, valid accuracy 0.8068
Epoch 451, CIFAR-10 Batch 2:  loss 0.000290, train_accuracy 1, valid accuracy 0.8046
Epoch 451, CIFAR-10 Batch 3:  loss 0.000850, train_accuracy 1, valid accuracy 0.7992
Epoch 451, CIFAR-10 Batch 4:  loss 0.000537, train_accuracy 1, valid accuracy 0.789
Epoch 451, CIFAR-10 Batch 5:  loss 0.000369, train_accuracy 1, valid accuracy 0.7964
Epoch 452, CIFAR-10 Batch 1:  loss 0.000400, train_accuracy 1, valid accuracy 0.8056
Epoch 452, CIFAR-10 Batch 2:  loss 0.001063, train_accuracy 1, valid accuracy 0.7918
Epoch 452, CIFAR-10 Batch 3:  loss 0.002266, train_accuracy 1, valid accuracy 0.7964
Epoch 452, CIFAR-10 Batch 4:  loss 0.000667, train_accuracy 1, valid accuracy 0.7992
Epoch 452, CIFAR-10 Batch 5:  loss 0.000095, train_accuracy 1, valid accuracy 0.8056
Epoch 453, CIFAR-10 Batch 1:  loss 0.001625, train_accuracy 1, valid accuracy 0.8022
Epoch 453, CIFAR-10 Batch 2:  loss 0.001545, train_accuracy 1, valid accuracy 0.8034
Epoch 453, CIFAR-10 Batch 3:  loss 0.000446, train_accuracy 1, valid accuracy 0.8018
Epoch 453, CIFAR-10 Batch 4:  loss 0.000126, train_accuracy 1, valid accuracy 0.8018
Epoch 453, CIFAR-10 Batch 5:  loss 0.001247, train_accuracy 1, valid accuracy 0.7914
Epoch 454, CIFAR-10 Batch 1:  loss 0.000454, train_accuracy 1, valid accuracy 0.7956
Epoch 454, CIFAR-10 Batch 2:  loss 0.000342, train_accuracy 1, valid accuracy 0.798
Epoch 454, CIFAR-10 Batch 3:  loss 0.001678, train_accuracy 1, valid accuracy 0.7962
Epoch 454, CIFAR-10 Batch 4:  loss 0.000314, train_accuracy 1, valid accuracy 0.796
Epoch 454, CIFAR-10 Batch 5:  loss 0.002219, train_accuracy 1, valid accuracy 0.7936
Epoch 455, CIFAR-10 Batch 1:  loss 0.000653, train_accuracy 1, valid accuracy 0.796
Epoch 455, CIFAR-10 Batch 2:  loss 0.000451, train_accuracy 1, valid accuracy 0.7976
Epoch 455, CIFAR-10 Batch 3:  loss 0.002190, train_accuracy 1, valid accuracy 0.8008
Epoch 455, CIFAR-10 Batch 4:  loss 0.000425, train_accuracy 1, valid accuracy 0.7992
Epoch 455, CIFAR-10 Batch 5:  loss 0.001291, train_accuracy 1, valid accuracy 0.8006
Epoch 456, CIFAR-10 Batch 1:  loss 0.000491, train_accuracy 1, valid accuracy 0.8014
Epoch 456, CIFAR-10 Batch 2:  loss 0.041548, train_accuracy 0.975, valid accuracy 0.7724
Epoch 456, CIFAR-10 Batch 3:  loss 0.006766, train_accuracy 1, valid accuracy 0.7814
Epoch 456, CIFAR-10 Batch 4:  loss 0.002069, train_accuracy 1, valid accuracy 0.7858
Epoch 456, CIFAR-10 Batch 5:  loss 0.001423, train_accuracy 1, valid accuracy 0.8
Epoch 457, CIFAR-10 Batch 1:  loss 0.001198, train_accuracy 1, valid accuracy 0.7936
Epoch 457, CIFAR-10 Batch 2:  loss 0.002636, train_accuracy 1, valid accuracy 0.8066
Epoch 457, CIFAR-10 Batch 3:  loss 0.000836, train_accuracy 1, valid accuracy 0.799
Epoch 457, CIFAR-10 Batch 4:  loss 0.000389, train_accuracy 1, valid accuracy 0.7918
Epoch 457, CIFAR-10 Batch 5:  loss 0.000594, train_accuracy 1, valid accuracy 0.8034
Epoch 458, CIFAR-10 Batch 1:  loss 0.000233, train_accuracy 1, valid accuracy 0.8066
Epoch 458, CIFAR-10 Batch 2:  loss 0.000317, train_accuracy 1, valid accuracy 0.8012
Epoch 458, CIFAR-10 Batch 3:  loss 0.002126, train_accuracy 1, valid accuracy 0.7932
Epoch 458, CIFAR-10 Batch 4:  loss 0.000238, train_accuracy 1, valid accuracy 0.8002
Epoch 458, CIFAR-10 Batch 5:  loss 0.000457, train_accuracy 1, valid accuracy 0.8068
Epoch 459, CIFAR-10 Batch 1:  loss 0.000744, train_accuracy 1, valid accuracy 0.809
Epoch 459, CIFAR-10 Batch 2:  loss 0.001899, train_accuracy 1, valid accuracy 0.8018
Epoch 459, CIFAR-10 Batch 3:  loss 0.000277, train_accuracy 1, valid accuracy 0.8018
Epoch 459, CIFAR-10 Batch 4:  loss 0.000316, train_accuracy 1, valid accuracy 0.801
Epoch 459, CIFAR-10 Batch 5:  loss 0.000399, train_accuracy 1, valid accuracy 0.798
Epoch 460, CIFAR-10 Batch 1:  loss 0.000225, train_accuracy 1, valid accuracy 0.8044
Epoch 460, CIFAR-10 Batch 2:  loss 0.000412, train_accuracy 1, valid accuracy 0.7954
Epoch 460, CIFAR-10 Batch 3:  loss 0.000704, train_accuracy 1, valid accuracy 0.8
Epoch 460, CIFAR-10 Batch 4:  loss 0.000328, train_accuracy 1, valid accuracy 0.7934
Epoch 460, CIFAR-10 Batch 5:  loss 0.000908, train_accuracy 1, valid accuracy 0.7912
Epoch 461, CIFAR-10 Batch 1:  loss 0.000220, train_accuracy 1, valid accuracy 0.8076
Epoch 461, CIFAR-10 Batch 2:  loss 0.000765, train_accuracy 1, valid accuracy 0.7994
Epoch 461, CIFAR-10 Batch 3:  loss 0.000320, train_accuracy 1, valid accuracy 0.8032
Epoch 461, CIFAR-10 Batch 4:  loss 0.000933, train_accuracy 1, valid accuracy 0.801
Epoch 461, CIFAR-10 Batch 5:  loss 0.000302, train_accuracy 1, valid accuracy 0.8042
Epoch 462, CIFAR-10 Batch 1:  loss 0.000379, train_accuracy 1, valid accuracy 0.8032
Epoch 462, CIFAR-10 Batch 2:  loss 0.000073, train_accuracy 1, valid accuracy 0.8012
Epoch 462, CIFAR-10 Batch 3:  loss 0.000289, train_accuracy 1, valid accuracy 0.8012
Epoch 462, CIFAR-10 Batch 4:  loss 0.000204, train_accuracy 1, valid accuracy 0.8122
Epoch 462, CIFAR-10 Batch 5:  loss 0.000118, train_accuracy 1, valid accuracy 0.802
Epoch 463, CIFAR-10 Batch 1:  loss 0.000788, train_accuracy 1, valid accuracy 0.8072
Epoch 463, CIFAR-10 Batch 2:  loss 0.000096, train_accuracy 1, valid accuracy 0.8028
Epoch 463, CIFAR-10 Batch 3:  loss 0.001333, train_accuracy 1, valid accuracy 0.7936
Epoch 463, CIFAR-10 Batch 4:  loss 0.000194, train_accuracy 1, valid accuracy 0.8054
Epoch 463, CIFAR-10 Batch 5:  loss 0.000089, train_accuracy 1, valid accuracy 0.8022
Epoch 464, CIFAR-10 Batch 1:  loss 0.000254, train_accuracy 1, valid accuracy 0.8068
Epoch 464, CIFAR-10 Batch 2:  loss 0.000246, train_accuracy 1, valid accuracy 0.8088
Epoch 464, CIFAR-10 Batch 3:  loss 0.000437, train_accuracy 1, valid accuracy 0.8004
Epoch 464, CIFAR-10 Batch 4:  loss 0.000397, train_accuracy 1, valid accuracy 0.803
Epoch 464, CIFAR-10 Batch 5:  loss 0.000714, train_accuracy 1, valid accuracy 0.8084
Epoch 465, CIFAR-10 Batch 1:  loss 0.000317, train_accuracy 1, valid accuracy 0.8068
Epoch 465, CIFAR-10 Batch 2:  loss 0.000238, train_accuracy 1, valid accuracy 0.7992
Epoch 465, CIFAR-10 Batch 3:  loss 0.000432, train_accuracy 1, valid accuracy 0.7984
Epoch 465, CIFAR-10 Batch 4:  loss 0.000639, train_accuracy 1, valid accuracy 0.806
Epoch 465, CIFAR-10 Batch 5:  loss 0.000374, train_accuracy 1, valid accuracy 0.8026
Epoch 466, CIFAR-10 Batch 1:  loss 0.000807, train_accuracy 1, valid accuracy 0.8064
Epoch 466, CIFAR-10 Batch 2:  loss 0.000108, train_accuracy 1, valid accuracy 0.8016
Epoch 466, CIFAR-10 Batch 3:  loss 0.000161, train_accuracy 1, valid accuracy 0.801
Epoch 466, CIFAR-10 Batch 4:  loss 0.010699, train_accuracy 1, valid accuracy 0.791
Epoch 466, CIFAR-10 Batch 5:  loss 0.000288, train_accuracy 1, valid accuracy 0.8
Epoch 467, CIFAR-10 Batch 1:  loss 0.000105, train_accuracy 1, valid accuracy 0.8088
Epoch 467, CIFAR-10 Batch 2:  loss 0.000254, train_accuracy 1, valid accuracy 0.8002
Epoch 467, CIFAR-10 Batch 3:  loss 0.001017, train_accuracy 1, valid accuracy 0.8072
Epoch 467, CIFAR-10 Batch 4:  loss 0.000541, train_accuracy 1, valid accuracy 0.8024
Epoch 467, CIFAR-10 Batch 5:  loss 0.000874, train_accuracy 1, valid accuracy 0.7988
Epoch 468, CIFAR-10 Batch 1:  loss 0.001042, train_accuracy 1, valid accuracy 0.7994
Epoch 468, CIFAR-10 Batch 2:  loss 0.007858, train_accuracy 1, valid accuracy 0.7842
Epoch 468, CIFAR-10 Batch 3:  loss 0.000321, train_accuracy 1, valid accuracy 0.8
Epoch 468, CIFAR-10 Batch 4:  loss 0.000203, train_accuracy 1, valid accuracy 0.7954
Epoch 468, CIFAR-10 Batch 5:  loss 0.003228, train_accuracy 1, valid accuracy 0.7944
Epoch 469, CIFAR-10 Batch 1:  loss 0.000325, train_accuracy 1, valid accuracy 0.799
Epoch 469, CIFAR-10 Batch 2:  loss 0.000583, train_accuracy 1, valid accuracy 0.7882
Epoch 469, CIFAR-10 Batch 3:  loss 0.001213, train_accuracy 1, valid accuracy 0.796
Epoch 469, CIFAR-10 Batch 4:  loss 0.000199, train_accuracy 1, valid accuracy 0.8024
Epoch 469, CIFAR-10 Batch 5:  loss 0.000331, train_accuracy 1, valid accuracy 0.8036
Epoch 470, CIFAR-10 Batch 1:  loss 0.000440, train_accuracy 1, valid accuracy 0.7924
Epoch 470, CIFAR-10 Batch 2:  loss 0.000438, train_accuracy 1, valid accuracy 0.8
Epoch 470, CIFAR-10 Batch 3:  loss 0.000150, train_accuracy 1, valid accuracy 0.7982
Epoch 470, CIFAR-10 Batch 4:  loss 0.000741, train_accuracy 1, valid accuracy 0.8008
Epoch 470, CIFAR-10 Batch 5:  loss 0.000197, train_accuracy 1, valid accuracy 0.8036
Epoch 471, CIFAR-10 Batch 1:  loss 0.000428, train_accuracy 1, valid accuracy 0.7994
Epoch 471, CIFAR-10 Batch 2:  loss 0.000133, train_accuracy 1, valid accuracy 0.791
Epoch 471, CIFAR-10 Batch 3:  loss 0.000152, train_accuracy 1, valid accuracy 0.7998
Epoch 471, CIFAR-10 Batch 4:  loss 0.000712, train_accuracy 1, valid accuracy 0.7862
Epoch 471, CIFAR-10 Batch 5:  loss 0.000898, train_accuracy 1, valid accuracy 0.791
Epoch 472, CIFAR-10 Batch 1:  loss 0.000212, train_accuracy 1, valid accuracy 0.8052
Epoch 472, CIFAR-10 Batch 2:  loss 0.000231, train_accuracy 1, valid accuracy 0.8028
Epoch 472, CIFAR-10 Batch 3:  loss 0.000517, train_accuracy 1, valid accuracy 0.796
Epoch 472, CIFAR-10 Batch 4:  loss 0.000671, train_accuracy 1, valid accuracy 0.7978
Epoch 472, CIFAR-10 Batch 5:  loss 0.000652, train_accuracy 1, valid accuracy 0.805
Epoch 473, CIFAR-10 Batch 1:  loss 0.000177, train_accuracy 1, valid accuracy 0.7954
Epoch 473, CIFAR-10 Batch 2:  loss 0.000375, train_accuracy 1, valid accuracy 0.7952
Epoch 473, CIFAR-10 Batch 3:  loss 0.000480, train_accuracy 1, valid accuracy 0.7998
Epoch 473, CIFAR-10 Batch 4:  loss 0.000261, train_accuracy 1, valid accuracy 0.7966
Epoch 473, CIFAR-10 Batch 5:  loss 0.000672, train_accuracy 1, valid accuracy 0.801
Epoch 474, CIFAR-10 Batch 1:  loss 0.000291, train_accuracy 1, valid accuracy 0.8052
Epoch 474, CIFAR-10 Batch 2:  loss 0.000179, train_accuracy 1, valid accuracy 0.7998
Epoch 474, CIFAR-10 Batch 3:  loss 0.000644, train_accuracy 1, valid accuracy 0.8014
Epoch 474, CIFAR-10 Batch 4:  loss 0.000234, train_accuracy 1, valid accuracy 0.7978
Epoch 474, CIFAR-10 Batch 5:  loss 0.000541, train_accuracy 1, valid accuracy 0.8006
Epoch 475, CIFAR-10 Batch 1:  loss 0.000192, train_accuracy 1, valid accuracy 0.8026
Epoch 475, CIFAR-10 Batch 2:  loss 0.000322, train_accuracy 1, valid accuracy 0.8006
Epoch 475, CIFAR-10 Batch 3:  loss 0.001986, train_accuracy 1, valid accuracy 0.786
Epoch 475, CIFAR-10 Batch 4:  loss 0.000434, train_accuracy 1, valid accuracy 0.7938
Epoch 475, CIFAR-10 Batch 5:  loss 0.000415, train_accuracy 1, valid accuracy 0.8016
Epoch 476, CIFAR-10 Batch 1:  loss 0.000426, train_accuracy 1, valid accuracy 0.807
Epoch 476, CIFAR-10 Batch 2:  loss 0.001592, train_accuracy 1, valid accuracy 0.7952
Epoch 476, CIFAR-10 Batch 3:  loss 0.000590, train_accuracy 1, valid accuracy 0.798
Epoch 476, CIFAR-10 Batch 4:  loss 0.011954, train_accuracy 1, valid accuracy 0.7786
Epoch 476, CIFAR-10 Batch 5:  loss 0.001215, train_accuracy 1, valid accuracy 0.794
Epoch 477, CIFAR-10 Batch 1:  loss 0.000469, train_accuracy 1, valid accuracy 0.807
Epoch 477, CIFAR-10 Batch 2:  loss 0.001800, train_accuracy 1, valid accuracy 0.7924
Epoch 477, CIFAR-10 Batch 3:  loss 0.000359, train_accuracy 1, valid accuracy 0.8036
Epoch 477, CIFAR-10 Batch 4:  loss 0.000388, train_accuracy 1, valid accuracy 0.8046
Epoch 477, CIFAR-10 Batch 5:  loss 0.000176, train_accuracy 1, valid accuracy 0.7994
Epoch 478, CIFAR-10 Batch 1:  loss 0.000752, train_accuracy 1, valid accuracy 0.8
Epoch 478, CIFAR-10 Batch 2:  loss 0.001363, train_accuracy 1, valid accuracy 0.7894
Epoch 478, CIFAR-10 Batch 3:  loss 0.000278, train_accuracy 1, valid accuracy 0.7992
Epoch 478, CIFAR-10 Batch 4:  loss 0.000406, train_accuracy 1, valid accuracy 0.7954
Epoch 478, CIFAR-10 Batch 5:  loss 0.000709, train_accuracy 1, valid accuracy 0.8024
Epoch 479, CIFAR-10 Batch 1:  loss 0.000282, train_accuracy 1, valid accuracy 0.804
Epoch 479, CIFAR-10 Batch 2:  loss 0.000605, train_accuracy 1, valid accuracy 0.7924
Epoch 479, CIFAR-10 Batch 3:  loss 0.000576, train_accuracy 1, valid accuracy 0.8048
Epoch 479, CIFAR-10 Batch 4:  loss 0.000227, train_accuracy 1, valid accuracy 0.806
Epoch 479, CIFAR-10 Batch 5:  loss 0.000285, train_accuracy 1, valid accuracy 0.808
Epoch 480, CIFAR-10 Batch 1:  loss 0.000434, train_accuracy 1, valid accuracy 0.8052
Epoch 480, CIFAR-10 Batch 2:  loss 0.001070, train_accuracy 1, valid accuracy 0.8006
Epoch 480, CIFAR-10 Batch 3:  loss 0.000408, train_accuracy 1, valid accuracy 0.8014
Epoch 480, CIFAR-10 Batch 4:  loss 0.000744, train_accuracy 1, valid accuracy 0.8054
Epoch 480, CIFAR-10 Batch 5:  loss 0.001671, train_accuracy 1, valid accuracy 0.8002
Epoch 481, CIFAR-10 Batch 1:  loss 0.000201, train_accuracy 1, valid accuracy 0.8054
Epoch 481, CIFAR-10 Batch 2:  loss 0.002813, train_accuracy 1, valid accuracy 0.789
Epoch 481, CIFAR-10 Batch 3:  loss 0.000295, train_accuracy 1, valid accuracy 0.7984
Epoch 481, CIFAR-10 Batch 4:  loss 0.000257, train_accuracy 1, valid accuracy 0.7954
Epoch 481, CIFAR-10 Batch 5:  loss 0.000135, train_accuracy 1, valid accuracy 0.8064
Epoch 482, CIFAR-10 Batch 1:  loss 0.000822, train_accuracy 1, valid accuracy 0.79
Epoch 482, CIFAR-10 Batch 2:  loss 0.000684, train_accuracy 1, valid accuracy 0.7894
Epoch 482, CIFAR-10 Batch 3:  loss 0.001564, train_accuracy 1, valid accuracy 0.797
Epoch 482, CIFAR-10 Batch 4:  loss 0.000389, train_accuracy 1, valid accuracy 0.8028
Epoch 482, CIFAR-10 Batch 5:  loss 0.000873, train_accuracy 1, valid accuracy 0.804
Epoch 483, CIFAR-10 Batch 1:  loss 0.000195, train_accuracy 1, valid accuracy 0.801
Epoch 483, CIFAR-10 Batch 2:  loss 0.000476, train_accuracy 1, valid accuracy 0.789
Epoch 483, CIFAR-10 Batch 3:  loss 0.000121, train_accuracy 1, valid accuracy 0.8018
Epoch 483, CIFAR-10 Batch 4:  loss 0.000196, train_accuracy 1, valid accuracy 0.8058
Epoch 483, CIFAR-10 Batch 5:  loss 0.001458, train_accuracy 1, valid accuracy 0.795
Epoch 484, CIFAR-10 Batch 1:  loss 0.000197, train_accuracy 1, valid accuracy 0.806
Epoch 484, CIFAR-10 Batch 2:  loss 0.000329, train_accuracy 1, valid accuracy 0.8058
Epoch 484, CIFAR-10 Batch 3:  loss 0.001170, train_accuracy 1, valid accuracy 0.7962
Epoch 484, CIFAR-10 Batch 4:  loss 0.000174, train_accuracy 1, valid accuracy 0.801
Epoch 484, CIFAR-10 Batch 5:  loss 0.000799, train_accuracy 1, valid accuracy 0.7944
Epoch 485, CIFAR-10 Batch 1:  loss 0.000836, train_accuracy 1, valid accuracy 0.7906
Epoch 485, CIFAR-10 Batch 2:  loss 0.000109, train_accuracy 1, valid accuracy 0.8046
Epoch 485, CIFAR-10 Batch 3:  loss 0.000159, train_accuracy 1, valid accuracy 0.8026
Epoch 485, CIFAR-10 Batch 4:  loss 0.000587, train_accuracy 1, valid accuracy 0.7992
Epoch 485, CIFAR-10 Batch 5:  loss 0.000208, train_accuracy 1, valid accuracy 0.8004
Epoch 486, CIFAR-10 Batch 1:  loss 0.000354, train_accuracy 1, valid accuracy 0.8058
Epoch 486, CIFAR-10 Batch 2:  loss 0.000039, train_accuracy 1, valid accuracy 0.807
Epoch 486, CIFAR-10 Batch 3:  loss 0.000587, train_accuracy 1, valid accuracy 0.809
Epoch 486, CIFAR-10 Batch 4:  loss 0.000064, train_accuracy 1, valid accuracy 0.799
Epoch 486, CIFAR-10 Batch 5:  loss 0.001471, train_accuracy 1, valid accuracy 0.7938
Epoch 487, CIFAR-10 Batch 1:  loss 0.001044, train_accuracy 1, valid accuracy 0.791
Epoch 487, CIFAR-10 Batch 2:  loss 0.001432, train_accuracy 1, valid accuracy 0.784
Epoch 487, CIFAR-10 Batch 3:  loss 0.000557, train_accuracy 1, valid accuracy 0.7904
Epoch 487, CIFAR-10 Batch 4:  loss 0.000213, train_accuracy 1, valid accuracy 0.8024
Epoch 487, CIFAR-10 Batch 5:  loss 0.000327, train_accuracy 1, valid accuracy 0.8102
Epoch 488, CIFAR-10 Batch 1:  loss 0.000370, train_accuracy 1, valid accuracy 0.8086
Epoch 488, CIFAR-10 Batch 2:  loss 0.000363, train_accuracy 1, valid accuracy 0.8008
Epoch 488, CIFAR-10 Batch 3:  loss 0.004039, train_accuracy 1, valid accuracy 0.7852
Epoch 488, CIFAR-10 Batch 4:  loss 0.000295, train_accuracy 1, valid accuracy 0.8006
Epoch 488, CIFAR-10 Batch 5:  loss 0.000527, train_accuracy 1, valid accuracy 0.8096
Epoch 489, CIFAR-10 Batch 1:  loss 0.001133, train_accuracy 1, valid accuracy 0.8058
Epoch 489, CIFAR-10 Batch 2:  loss 0.001027, train_accuracy 1, valid accuracy 0.7968
Epoch 489, CIFAR-10 Batch 3:  loss 0.000240, train_accuracy 1, valid accuracy 0.8058
Epoch 489, CIFAR-10 Batch 4:  loss 0.000465, train_accuracy 1, valid accuracy 0.798
Epoch 489, CIFAR-10 Batch 5:  loss 0.001623, train_accuracy 1, valid accuracy 0.7822
Epoch 490, CIFAR-10 Batch 1:  loss 0.000693, train_accuracy 1, valid accuracy 0.8028
Epoch 490, CIFAR-10 Batch 2:  loss 0.000150, train_accuracy 1, valid accuracy 0.7998
Epoch 490, CIFAR-10 Batch 3:  loss 0.000172, train_accuracy 1, valid accuracy 0.7938
Epoch 490, CIFAR-10 Batch 4:  loss 0.000609, train_accuracy 1, valid accuracy 0.7928
Epoch 490, CIFAR-10 Batch 5:  loss 0.000224, train_accuracy 1, valid accuracy 0.7976
Epoch 491, CIFAR-10 Batch 1:  loss 0.001530, train_accuracy 1, valid accuracy 0.798
Epoch 491, CIFAR-10 Batch 2:  loss 0.003040, train_accuracy 1, valid accuracy 0.794
Epoch 491, CIFAR-10 Batch 3:  loss 0.000396, train_accuracy 1, valid accuracy 0.797
Epoch 491, CIFAR-10 Batch 4:  loss 0.000563, train_accuracy 1, valid accuracy 0.791
Epoch 491, CIFAR-10 Batch 5:  loss 0.000116, train_accuracy 1, valid accuracy 0.8106
Epoch 492, CIFAR-10 Batch 1:  loss 0.000230, train_accuracy 1, valid accuracy 0.8036
Epoch 492, CIFAR-10 Batch 2:  loss 0.000786, train_accuracy 1, valid accuracy 0.7942
Epoch 492, CIFAR-10 Batch 3:  loss 0.000376, train_accuracy 1, valid accuracy 0.8018
Epoch 492, CIFAR-10 Batch 4:  loss 0.000246, train_accuracy 1, valid accuracy 0.8046
Epoch 492, CIFAR-10 Batch 5:  loss 0.000473, train_accuracy 1, valid accuracy 0.7998
Epoch 493, CIFAR-10 Batch 1:  loss 0.000143, train_accuracy 1, valid accuracy 0.8072
Epoch 493, CIFAR-10 Batch 2:  loss 0.000051, train_accuracy 1, valid accuracy 0.8046
Epoch 493, CIFAR-10 Batch 3:  loss 0.000896, train_accuracy 1, valid accuracy 0.8074
Epoch 493, CIFAR-10 Batch 4:  loss 0.000060, train_accuracy 1, valid accuracy 0.807
Epoch 493, CIFAR-10 Batch 5:  loss 0.000508, train_accuracy 1, valid accuracy 0.797
Epoch 494, CIFAR-10 Batch 1:  loss 0.000419, train_accuracy 1, valid accuracy 0.8096
Epoch 494, CIFAR-10 Batch 2:  loss 0.000279, train_accuracy 1, valid accuracy 0.7996
Epoch 494, CIFAR-10 Batch 3:  loss 0.000408, train_accuracy 1, valid accuracy 0.8006
Epoch 494, CIFAR-10 Batch 4:  loss 0.000068, train_accuracy 1, valid accuracy 0.8008
Epoch 494, CIFAR-10 Batch 5:  loss 0.000439, train_accuracy 1, valid accuracy 0.8068
Epoch 495, CIFAR-10 Batch 1:  loss 0.000248, train_accuracy 1, valid accuracy 0.7938
Epoch 495, CIFAR-10 Batch 2:  loss 0.000620, train_accuracy 1, valid accuracy 0.804
Epoch 495, CIFAR-10 Batch 3:  loss 0.000935, train_accuracy 1, valid accuracy 0.8004
Epoch 495, CIFAR-10 Batch 4:  loss 0.000616, train_accuracy 1, valid accuracy 0.7932
Epoch 495, CIFAR-10 Batch 5:  loss 0.000224, train_accuracy 1, valid accuracy 0.8048
Epoch 496, CIFAR-10 Batch 1:  loss 0.000099, train_accuracy 1, valid accuracy 0.7996
Epoch 496, CIFAR-10 Batch 2:  loss 0.000278, train_accuracy 1, valid accuracy 0.7944
Epoch 496, CIFAR-10 Batch 3:  loss 0.000126, train_accuracy 1, valid accuracy 0.7974
Epoch 496, CIFAR-10 Batch 4:  loss 0.000559, train_accuracy 1, valid accuracy 0.7972
Epoch 496, CIFAR-10 Batch 5:  loss 0.000316, train_accuracy 1, valid accuracy 0.7882
Epoch 497, CIFAR-10 Batch 1:  loss 0.001704, train_accuracy 1, valid accuracy 0.7936
Epoch 497, CIFAR-10 Batch 2:  loss 0.000635, train_accuracy 1, valid accuracy 0.7836
Epoch 497, CIFAR-10 Batch 3:  loss 0.001936, train_accuracy 1, valid accuracy 0.7928
Epoch 497, CIFAR-10 Batch 4:  loss 0.001140, train_accuracy 1, valid accuracy 0.7864
Epoch 497, CIFAR-10 Batch 5:  loss 0.000284, train_accuracy 1, valid accuracy 0.8054
Epoch 498, CIFAR-10 Batch 1:  loss 0.001089, train_accuracy 1, valid accuracy 0.804
Epoch 498, CIFAR-10 Batch 2:  loss 0.000651, train_accuracy 1, valid accuracy 0.799
Epoch 498, CIFAR-10 Batch 3:  loss 0.000319, train_accuracy 1, valid accuracy 0.8082
Epoch 498, CIFAR-10 Batch 4:  loss 0.000632, train_accuracy 1, valid accuracy 0.8038
Epoch 498, CIFAR-10 Batch 5:  loss 0.000645, train_accuracy 1, valid accuracy 0.7976
Epoch 499, CIFAR-10 Batch 1:  loss 0.000607, train_accuracy 1, valid accuracy 0.8054
Epoch 499, CIFAR-10 Batch 2:  loss 0.000247, train_accuracy 1, valid accuracy 0.804
Epoch 499, CIFAR-10 Batch 3:  loss 0.000888, train_accuracy 1, valid accuracy 0.7994
Epoch 499, CIFAR-10 Batch 4:  loss 0.000103, train_accuracy 1, valid accuracy 0.8024
Epoch 499, CIFAR-10 Batch 5:  loss 0.000029, train_accuracy 1, valid accuracy 0.8072
Epoch 500, CIFAR-10 Batch 1:  loss 0.001546, train_accuracy 1, valid accuracy 0.7982
Epoch 500, CIFAR-10 Batch 2:  loss 0.000358, train_accuracy 1, valid accuracy 0.8126
Epoch 500, CIFAR-10 Batch 3:  loss 0.001190, train_accuracy 1, valid accuracy 0.8048
Epoch 500, CIFAR-10 Batch 4:  loss 0.000143, train_accuracy 1, valid accuracy 0.8012
Epoch 500, CIFAR-10 Batch 5:  loss 0.000428, train_accuracy 1, valid accuracy 0.8004
Epoch 501, CIFAR-10 Batch 1:  loss 0.001698, train_accuracy 1, valid accuracy 0.7978
Epoch 501, CIFAR-10 Batch 2:  loss 0.000526, train_accuracy 1, valid accuracy 0.7926
Epoch 501, CIFAR-10 Batch 3:  loss 0.000227, train_accuracy 1, valid accuracy 0.8012
Epoch 501, CIFAR-10 Batch 4:  loss 0.000337, train_accuracy 1, valid accuracy 0.7842
Epoch 501, CIFAR-10 Batch 5:  loss 0.000098, train_accuracy 1, valid accuracy 0.8024
Epoch 502, CIFAR-10 Batch 1:  loss 0.001311, train_accuracy 1, valid accuracy 0.7922
Epoch 502, CIFAR-10 Batch 2:  loss 0.000507, train_accuracy 1, valid accuracy 0.805
Epoch 502, CIFAR-10 Batch 3:  loss 0.002871, train_accuracy 1, valid accuracy 0.786
Epoch 502, CIFAR-10 Batch 4:  loss 0.000239, train_accuracy 1, valid accuracy 0.8082
Epoch 502, CIFAR-10 Batch 5:  loss 0.000175, train_accuracy 1, valid accuracy 0.8044
Epoch 503, CIFAR-10 Batch 1:  loss 0.000106, train_accuracy 1, valid accuracy 0.8044
Epoch 503, CIFAR-10 Batch 2:  loss 0.000298, train_accuracy 1, valid accuracy 0.8036
Epoch 503, CIFAR-10 Batch 3:  loss 0.000661, train_accuracy 1, valid accuracy 0.7954
Epoch 503, CIFAR-10 Batch 4:  loss 0.000342, train_accuracy 1, valid accuracy 0.8046
Epoch 503, CIFAR-10 Batch 5:  loss 0.000269, train_accuracy 1, valid accuracy 0.801
Epoch 504, CIFAR-10 Batch 1:  loss 0.000120, train_accuracy 1, valid accuracy 0.8084
Epoch 504, CIFAR-10 Batch 2:  loss 0.001110, train_accuracy 1, valid accuracy 0.791
Epoch 504, CIFAR-10 Batch 3:  loss 0.000286, train_accuracy 1, valid accuracy 0.7998
Epoch 504, CIFAR-10 Batch 4:  loss 0.000185, train_accuracy 1, valid accuracy 0.8018
Epoch 504, CIFAR-10 Batch 5:  loss 0.000425, train_accuracy 1, valid accuracy 0.7874
Epoch 505, CIFAR-10 Batch 1:  loss 0.001195, train_accuracy 1, valid accuracy 0.8016
Epoch 505, CIFAR-10 Batch 2:  loss 0.000194, train_accuracy 1, valid accuracy 0.8
Epoch 505, CIFAR-10 Batch 3:  loss 0.000660, train_accuracy 1, valid accuracy 0.8044
Epoch 505, CIFAR-10 Batch 4:  loss 0.000142, train_accuracy 1, valid accuracy 0.8044
Epoch 505, CIFAR-10 Batch 5:  loss 0.000131, train_accuracy 1, valid accuracy 0.8068
Epoch 506, CIFAR-10 Batch 1:  loss 0.000220, train_accuracy 1, valid accuracy 0.8088
Epoch 506, CIFAR-10 Batch 2:  loss 0.000208, train_accuracy 1, valid accuracy 0.8098
Epoch 506, CIFAR-10 Batch 3:  loss 0.000121, train_accuracy 1, valid accuracy 0.8096
Epoch 506, CIFAR-10 Batch 4:  loss 0.000643, train_accuracy 1, valid accuracy 0.8004
Epoch 506, CIFAR-10 Batch 5:  loss 0.000796, train_accuracy 1, valid accuracy 0.7958
Epoch 507, CIFAR-10 Batch 1:  loss 0.000282, train_accuracy 1, valid accuracy 0.804
Epoch 507, CIFAR-10 Batch 2:  loss 0.000822, train_accuracy 1, valid accuracy 0.7996
Epoch 507, CIFAR-10 Batch 3:  loss 0.000308, train_accuracy 1, valid accuracy 0.7994
Epoch 507, CIFAR-10 Batch 4:  loss 0.000233, train_accuracy 1, valid accuracy 0.8104
Epoch 507, CIFAR-10 Batch 5:  loss 0.000161, train_accuracy 1, valid accuracy 0.804
Epoch 508, CIFAR-10 Batch 1:  loss 0.000643, train_accuracy 1, valid accuracy 0.8002
Epoch 508, CIFAR-10 Batch 2:  loss 0.006860, train_accuracy 1, valid accuracy 0.782
Epoch 508, CIFAR-10 Batch 3:  loss 0.000261, train_accuracy 1, valid accuracy 0.8026
Epoch 508, CIFAR-10 Batch 4:  loss 0.000307, train_accuracy 1, valid accuracy 0.8084
Epoch 508, CIFAR-10 Batch 5:  loss 0.000431, train_accuracy 1, valid accuracy 0.8094
Epoch 509, CIFAR-10 Batch 1:  loss 0.000246, train_accuracy 1, valid accuracy 0.8094
Epoch 509, CIFAR-10 Batch 2:  loss 0.002037, train_accuracy 1, valid accuracy 0.7922
Epoch 509, CIFAR-10 Batch 3:  loss 0.000529, train_accuracy 1, valid accuracy 0.8006
Epoch 509, CIFAR-10 Batch 4:  loss 0.001157, train_accuracy 1, valid accuracy 0.801
Epoch 509, CIFAR-10 Batch 5:  loss 0.000403, train_accuracy 1, valid accuracy 0.8026
Epoch 510, CIFAR-10 Batch 1:  loss 0.000651, train_accuracy 1, valid accuracy 0.8094
Epoch 510, CIFAR-10 Batch 2:  loss 0.000529, train_accuracy 1, valid accuracy 0.8038
Epoch 510, CIFAR-10 Batch 3:  loss 0.003394, train_accuracy 1, valid accuracy 0.7888
Epoch 510, CIFAR-10 Batch 4:  loss 0.000165, train_accuracy 1, valid accuracy 0.8032
Epoch 510, CIFAR-10 Batch 5:  loss 0.000508, train_accuracy 1, valid accuracy 0.8014
Epoch 511, CIFAR-10 Batch 1:  loss 0.000384, train_accuracy 1, valid accuracy 0.8124
Epoch 511, CIFAR-10 Batch 2:  loss 0.000104, train_accuracy 1, valid accuracy 0.8118
Epoch 511, CIFAR-10 Batch 3:  loss 0.000221, train_accuracy 1, valid accuracy 0.8034
Epoch 511, CIFAR-10 Batch 4:  loss 0.017946, train_accuracy 1, valid accuracy 0.7844
Epoch 511, CIFAR-10 Batch 5:  loss 0.000434, train_accuracy 1, valid accuracy 0.8074
Epoch 512, CIFAR-10 Batch 1:  loss 0.000071, train_accuracy 1, valid accuracy 0.8094
Epoch 512, CIFAR-10 Batch 2:  loss 0.000485, train_accuracy 1, valid accuracy 0.785
Epoch 512, CIFAR-10 Batch 3:  loss 0.000158, train_accuracy 1, valid accuracy 0.7942
Epoch 512, CIFAR-10 Batch 4:  loss 0.000170, train_accuracy 1, valid accuracy 0.8042
Epoch 512, CIFAR-10 Batch 5:  loss 0.000122, train_accuracy 1, valid accuracy 0.8016
Epoch 513, CIFAR-10 Batch 1:  loss 0.000878, train_accuracy 1, valid accuracy 0.7984
Epoch 513, CIFAR-10 Batch 2:  loss 0.000188, train_accuracy 1, valid accuracy 0.8042
Epoch 513, CIFAR-10 Batch 3:  loss 0.000542, train_accuracy 1, valid accuracy 0.8082
Epoch 513, CIFAR-10 Batch 4:  loss 0.000316, train_accuracy 1, valid accuracy 0.8092
Epoch 513, CIFAR-10 Batch 5:  loss 0.000057, train_accuracy 1, valid accuracy 0.8062
Epoch 514, CIFAR-10 Batch 1:  loss 0.000402, train_accuracy 1, valid accuracy 0.8078
Epoch 514, CIFAR-10 Batch 2:  loss 0.006459, train_accuracy 1, valid accuracy 0.771
Epoch 514, CIFAR-10 Batch 3:  loss 0.000202, train_accuracy 1, valid accuracy 0.7962
Epoch 514, CIFAR-10 Batch 4:  loss 0.000365, train_accuracy 1, valid accuracy 0.803
Epoch 514, CIFAR-10 Batch 5:  loss 0.000261, train_accuracy 1, valid accuracy 0.8052
Epoch 515, CIFAR-10 Batch 1:  loss 0.000503, train_accuracy 1, valid accuracy 0.803
Epoch 515, CIFAR-10 Batch 2:  loss 0.000283, train_accuracy 1, valid accuracy 0.8002
Epoch 515, CIFAR-10 Batch 3:  loss 0.001964, train_accuracy 1, valid accuracy 0.8036
Epoch 515, CIFAR-10 Batch 4:  loss 0.000172, train_accuracy 1, valid accuracy 0.8036
Epoch 515, CIFAR-10 Batch 5:  loss 0.000515, train_accuracy 1, valid accuracy 0.8064
Epoch 516, CIFAR-10 Batch 1:  loss 0.000538, train_accuracy 1, valid accuracy 0.8064
Epoch 516, CIFAR-10 Batch 2:  loss 0.000326, train_accuracy 1, valid accuracy 0.8066
Epoch 516, CIFAR-10 Batch 3:  loss 0.000094, train_accuracy 1, valid accuracy 0.8032
Epoch 516, CIFAR-10 Batch 4:  loss 0.000299, train_accuracy 1, valid accuracy 0.8062
Epoch 516, CIFAR-10 Batch 5:  loss 0.000046, train_accuracy 1, valid accuracy 0.8082
Epoch 517, CIFAR-10 Batch 1:  loss 0.000567, train_accuracy 1, valid accuracy 0.7972
Epoch 517, CIFAR-10 Batch 2:  loss 0.000087, train_accuracy 1, valid accuracy 0.807
Epoch 517, CIFAR-10 Batch 3:  loss 0.000049, train_accuracy 1, valid accuracy 0.8052
Epoch 517, CIFAR-10 Batch 4:  loss 0.000567, train_accuracy 1, valid accuracy 0.7994
Epoch 517, CIFAR-10 Batch 5:  loss 0.000093, train_accuracy 1, valid accuracy 0.7968
Epoch 518, CIFAR-10 Batch 1:  loss 0.000111, train_accuracy 1, valid accuracy 0.8074
Epoch 518, CIFAR-10 Batch 2:  loss 0.000302, train_accuracy 1, valid accuracy 0.7942
Epoch 518, CIFAR-10 Batch 3:  loss 0.000466, train_accuracy 1, valid accuracy 0.793
Epoch 518, CIFAR-10 Batch 4:  loss 0.001230, train_accuracy 1, valid accuracy 0.8006
Epoch 518, CIFAR-10 Batch 5:  loss 0.000050, train_accuracy 1, valid accuracy 0.8012
Epoch 519, CIFAR-10 Batch 1:  loss 0.001685, train_accuracy 1, valid accuracy 0.7922
Epoch 519, CIFAR-10 Batch 2:  loss 0.001235, train_accuracy 1, valid accuracy 0.789
Epoch 519, CIFAR-10 Batch 3:  loss 0.000082, train_accuracy 1, valid accuracy 0.7992
Epoch 519, CIFAR-10 Batch 4:  loss 0.000245, train_accuracy 1, valid accuracy 0.8062
Epoch 519, CIFAR-10 Batch 5:  loss 0.002398, train_accuracy 1, valid accuracy 0.7928
Epoch 520, CIFAR-10 Batch 1:  loss 0.000079, train_accuracy 1, valid accuracy 0.8068
Epoch 520, CIFAR-10 Batch 2:  loss 0.000358, train_accuracy 1, valid accuracy 0.7924
Epoch 520, CIFAR-10 Batch 3:  loss 0.000170, train_accuracy 1, valid accuracy 0.8018
Epoch 520, CIFAR-10 Batch 4:  loss 0.000192, train_accuracy 1, valid accuracy 0.8058
Epoch 520, CIFAR-10 Batch 5:  loss 0.000777, train_accuracy 1, valid accuracy 0.8016
Epoch 521, CIFAR-10 Batch 1:  loss 0.000400, train_accuracy 1, valid accuracy 0.7914
Epoch 521, CIFAR-10 Batch 2:  loss 0.000951, train_accuracy 1, valid accuracy 0.7968
Epoch 521, CIFAR-10 Batch 3:  loss 0.000178, train_accuracy 1, valid accuracy 0.8016
Epoch 521, CIFAR-10 Batch 4:  loss 0.000841, train_accuracy 1, valid accuracy 0.8032
Epoch 521, CIFAR-10 Batch 5:  loss 0.000097, train_accuracy 1, valid accuracy 0.8048
Epoch 522, CIFAR-10 Batch 1:  loss 0.000825, train_accuracy 1, valid accuracy 0.7972
Epoch 522, CIFAR-10 Batch 2:  loss 0.000150, train_accuracy 1, valid accuracy 0.8094
Epoch 522, CIFAR-10 Batch 3:  loss 0.000224, train_accuracy 1, valid accuracy 0.805
Epoch 522, CIFAR-10 Batch 4:  loss 0.000088, train_accuracy 1, valid accuracy 0.8046
Epoch 522, CIFAR-10 Batch 5:  loss 0.000159, train_accuracy 1, valid accuracy 0.7954
Epoch 523, CIFAR-10 Batch 1:  loss 0.000157, train_accuracy 1, valid accuracy 0.8092
Epoch 523, CIFAR-10 Batch 2:  loss 0.000143, train_accuracy 1, valid accuracy 0.8074
Epoch 523, CIFAR-10 Batch 3:  loss 0.000242, train_accuracy 1, valid accuracy 0.7992
Epoch 523, CIFAR-10 Batch 4:  loss 0.000321, train_accuracy 1, valid accuracy 0.812
Epoch 523, CIFAR-10 Batch 5:  loss 0.000480, train_accuracy 1, valid accuracy 0.807
Epoch 524, CIFAR-10 Batch 1:  loss 0.000270, train_accuracy 1, valid accuracy 0.8078
Epoch 524, CIFAR-10 Batch 2:  loss 0.000472, train_accuracy 1, valid accuracy 0.8026
Epoch 524, CIFAR-10 Batch 3:  loss 0.000099, train_accuracy 1, valid accuracy 0.8038
Epoch 524, CIFAR-10 Batch 4:  loss 0.000544, train_accuracy 1, valid accuracy 0.7988
Epoch 524, CIFAR-10 Batch 5:  loss 0.000726, train_accuracy 1, valid accuracy 0.8032
Epoch 525, CIFAR-10 Batch 1:  loss 0.000309, train_accuracy 1, valid accuracy 0.8096
Epoch 525, CIFAR-10 Batch 2:  loss 0.000144, train_accuracy 1, valid accuracy 0.7992
Epoch 525, CIFAR-10 Batch 3:  loss 0.000099, train_accuracy 1, valid accuracy 0.807
Epoch 525, CIFAR-10 Batch 4:  loss 0.000170, train_accuracy 1, valid accuracy 0.8024
Epoch 525, CIFAR-10 Batch 5:  loss 0.000279, train_accuracy 1, valid accuracy 0.8108
Epoch 526, CIFAR-10 Batch 1:  loss 0.000290, train_accuracy 1, valid accuracy 0.8038
Epoch 526, CIFAR-10 Batch 2:  loss 0.000741, train_accuracy 1, valid accuracy 0.8048
Epoch 526, CIFAR-10 Batch 3:  loss 0.000157, train_accuracy 1, valid accuracy 0.8054
Epoch 526, CIFAR-10 Batch 4:  loss 0.000943, train_accuracy 1, valid accuracy 0.8078
Epoch 526, CIFAR-10 Batch 5:  loss 0.001103, train_accuracy 1, valid accuracy 0.8004
Epoch 527, CIFAR-10 Batch 1:  loss 0.000562, train_accuracy 1, valid accuracy 0.8102
Epoch 527, CIFAR-10 Batch 2:  loss 0.002187, train_accuracy 1, valid accuracy 0.8088
Epoch 527, CIFAR-10 Batch 3:  loss 0.000052, train_accuracy 1, valid accuracy 0.806
Epoch 527, CIFAR-10 Batch 4:  loss 0.003233, train_accuracy 1, valid accuracy 0.8008
Epoch 527, CIFAR-10 Batch 5:  loss 0.000239, train_accuracy 1, valid accuracy 0.8142
Epoch 528, CIFAR-10 Batch 1:  loss 0.001447, train_accuracy 1, valid accuracy 0.806
Epoch 528, CIFAR-10 Batch 2:  loss 0.000681, train_accuracy 1, valid accuracy 0.7966
Epoch 528, CIFAR-10 Batch 3:  loss 0.000515, train_accuracy 1, valid accuracy 0.8036
Epoch 528, CIFAR-10 Batch 4:  loss 0.001651, train_accuracy 1, valid accuracy 0.8008
Epoch 528, CIFAR-10 Batch 5:  loss 0.000561, train_accuracy 1, valid accuracy 0.8096
Epoch 529, CIFAR-10 Batch 1:  loss 0.002255, train_accuracy 1, valid accuracy 0.8078
Epoch 529, CIFAR-10 Batch 2:  loss 0.001049, train_accuracy 1, valid accuracy 0.8032
Epoch 529, CIFAR-10 Batch 3:  loss 0.000111, train_accuracy 1, valid accuracy 0.8094
Epoch 529, CIFAR-10 Batch 4:  loss 0.000760, train_accuracy 1, valid accuracy 0.7938
Epoch 529, CIFAR-10 Batch 5:  loss 0.000108, train_accuracy 1, valid accuracy 0.8098
Epoch 530, CIFAR-10 Batch 1:  loss 0.001608, train_accuracy 1, valid accuracy 0.8048
Epoch 530, CIFAR-10 Batch 2:  loss 0.002136, train_accuracy 1, valid accuracy 0.8088
Epoch 530, CIFAR-10 Batch 3:  loss 0.000422, train_accuracy 1, valid accuracy 0.7992
Epoch 530, CIFAR-10 Batch 4:  loss 0.000095, train_accuracy 1, valid accuracy 0.8124
Epoch 530, CIFAR-10 Batch 5:  loss 0.000144, train_accuracy 1, valid accuracy 0.8048
Epoch 531, CIFAR-10 Batch 1:  loss 0.000151, train_accuracy 1, valid accuracy 0.8092
Epoch 531, CIFAR-10 Batch 2:  loss 0.001307, train_accuracy 1, valid accuracy 0.796
Epoch 531, CIFAR-10 Batch 3:  loss 0.000097, train_accuracy 1, valid accuracy 0.806
Epoch 531, CIFAR-10 Batch 4:  loss 0.000116, train_accuracy 1, valid accuracy 0.7968
Epoch 531, CIFAR-10 Batch 5:  loss 0.000387, train_accuracy 1, valid accuracy 0.8048
Epoch 532, CIFAR-10 Batch 1:  loss 0.000659, train_accuracy 1, valid accuracy 0.802
Epoch 532, CIFAR-10 Batch 2:  loss 0.000395, train_accuracy 1, valid accuracy 0.808
Epoch 532, CIFAR-10 Batch 3:  loss 0.000089, train_accuracy 1, valid accuracy 0.8086
Epoch 532, CIFAR-10 Batch 4:  loss 0.000157, train_accuracy 1, valid accuracy 0.803
Epoch 532, CIFAR-10 Batch 5:  loss 0.000198, train_accuracy 1, valid accuracy 0.8098
Epoch 533, CIFAR-10 Batch 1:  loss 0.000331, train_accuracy 1, valid accuracy 0.8066
Epoch 533, CIFAR-10 Batch 2:  loss 0.000061, train_accuracy 1, valid accuracy 0.8094
Epoch 533, CIFAR-10 Batch 3:  loss 0.000034, train_accuracy 1, valid accuracy 0.8048
Epoch 533, CIFAR-10 Batch 4:  loss 0.000102, train_accuracy 1, valid accuracy 0.8112
Epoch 533, CIFAR-10 Batch 5:  loss 0.001347, train_accuracy 1, valid accuracy 0.794
Epoch 534, CIFAR-10 Batch 1:  loss 0.000472, train_accuracy 1, valid accuracy 0.8006
Epoch 534, CIFAR-10 Batch 2:  loss 0.000446, train_accuracy 1, valid accuracy 0.7954
Epoch 534, CIFAR-10 Batch 3:  loss 0.000117, train_accuracy 1, valid accuracy 0.8102
Epoch 534, CIFAR-10 Batch 4:  loss 0.000063, train_accuracy 1, valid accuracy 0.8096
Epoch 534, CIFAR-10 Batch 5:  loss 0.000715, train_accuracy 1, valid accuracy 0.8022
Epoch 535, CIFAR-10 Batch 1:  loss 0.000335, train_accuracy 1, valid accuracy 0.8074
Epoch 535, CIFAR-10 Batch 2:  loss 0.002026, train_accuracy 1, valid accuracy 0.7968
Epoch 535, CIFAR-10 Batch 3:  loss 0.000144, train_accuracy 1, valid accuracy 0.8074
Epoch 535, CIFAR-10 Batch 4:  loss 0.000215, train_accuracy 1, valid accuracy 0.8018
Epoch 535, CIFAR-10 Batch 5:  loss 0.000869, train_accuracy 1, valid accuracy 0.8052
Epoch 536, CIFAR-10 Batch 1:  loss 0.000292, train_accuracy 1, valid accuracy 0.8082
Epoch 536, CIFAR-10 Batch 2:  loss 0.001238, train_accuracy 1, valid accuracy 0.7952
Epoch 536, CIFAR-10 Batch 3:  loss 0.000126, train_accuracy 1, valid accuracy 0.803
Epoch 536, CIFAR-10 Batch 4:  loss 0.000200, train_accuracy 1, valid accuracy 0.8056
Epoch 536, CIFAR-10 Batch 5:  loss 0.000455, train_accuracy 1, valid accuracy 0.8078
Epoch 537, CIFAR-10 Batch 1:  loss 0.000177, train_accuracy 1, valid accuracy 0.804
Epoch 537, CIFAR-10 Batch 2:  loss 0.000083, train_accuracy 1, valid accuracy 0.8094
Epoch 537, CIFAR-10 Batch 3:  loss 0.000343, train_accuracy 1, valid accuracy 0.8052
Epoch 537, CIFAR-10 Batch 4:  loss 0.000430, train_accuracy 1, valid accuracy 0.8026
Epoch 537, CIFAR-10 Batch 5:  loss 0.000231, train_accuracy 1, valid accuracy 0.8012
Epoch 538, CIFAR-10 Batch 1:  loss 0.000139, train_accuracy 1, valid accuracy 0.8064
Epoch 538, CIFAR-10 Batch 2:  loss 0.000113, train_accuracy 1, valid accuracy 0.8076
Epoch 538, CIFAR-10 Batch 3:  loss 0.000143, train_accuracy 1, valid accuracy 0.805
Epoch 538, CIFAR-10 Batch 4:  loss 0.000166, train_accuracy 1, valid accuracy 0.8042
Epoch 538, CIFAR-10 Batch 5:  loss 0.001362, train_accuracy 1, valid accuracy 0.8034
Epoch 539, CIFAR-10 Batch 1:  loss 0.000426, train_accuracy 1, valid accuracy 0.8108
Epoch 539, CIFAR-10 Batch 2:  loss 0.000417, train_accuracy 1, valid accuracy 0.8072
Epoch 539, CIFAR-10 Batch 3:  loss 0.000509, train_accuracy 1, valid accuracy 0.8062
Epoch 539, CIFAR-10 Batch 4:  loss 0.000148, train_accuracy 1, valid accuracy 0.8044
Epoch 539, CIFAR-10 Batch 5:  loss 0.000221, train_accuracy 1, valid accuracy 0.7982
Epoch 540, CIFAR-10 Batch 1:  loss 0.000454, train_accuracy 1, valid accuracy 0.809
Epoch 540, CIFAR-10 Batch 2:  loss 0.000541, train_accuracy 1, valid accuracy 0.8032
Epoch 540, CIFAR-10 Batch 3:  loss 0.000184, train_accuracy 1, valid accuracy 0.8006
Epoch 540, CIFAR-10 Batch 4:  loss 0.000127, train_accuracy 1, valid accuracy 0.8126
Epoch 540, CIFAR-10 Batch 5:  loss 0.000285, train_accuracy 1, valid accuracy 0.809
Epoch 541, CIFAR-10 Batch 1:  loss 0.000155, train_accuracy 1, valid accuracy 0.8082
Epoch 541, CIFAR-10 Batch 2:  loss 0.000375, train_accuracy 1, valid accuracy 0.8012
Epoch 541, CIFAR-10 Batch 3:  loss 0.000177, train_accuracy 1, valid accuracy 0.8004
Epoch 541, CIFAR-10 Batch 4:  loss 0.000232, train_accuracy 1, valid accuracy 0.8114
Epoch 541, CIFAR-10 Batch 5:  loss 0.000550, train_accuracy 1, valid accuracy 0.8022
Epoch 542, CIFAR-10 Batch 1:  loss 0.000678, train_accuracy 1, valid accuracy 0.8008
Epoch 542, CIFAR-10 Batch 2:  loss 0.000231, train_accuracy 1, valid accuracy 0.8052
Epoch 542, CIFAR-10 Batch 3:  loss 0.000297, train_accuracy 1, valid accuracy 0.8038
Epoch 542, CIFAR-10 Batch 4:  loss 0.000212, train_accuracy 1, valid accuracy 0.7966
Epoch 542, CIFAR-10 Batch 5:  loss 0.000471, train_accuracy 1, valid accuracy 0.808
Epoch 543, CIFAR-10 Batch 1:  loss 0.000323, train_accuracy 1, valid accuracy 0.811
Epoch 543, CIFAR-10 Batch 2:  loss 0.001800, train_accuracy 1, valid accuracy 0.7936
Epoch 543, CIFAR-10 Batch 3:  loss 0.000270, train_accuracy 1, valid accuracy 0.8062
Epoch 543, CIFAR-10 Batch 4:  loss 0.000122, train_accuracy 1, valid accuracy 0.8022
Epoch 543, CIFAR-10 Batch 5:  loss 0.000793, train_accuracy 1, valid accuracy 0.7966
Epoch 544, CIFAR-10 Batch 1:  loss 0.000086, train_accuracy 1, valid accuracy 0.806
Epoch 544, CIFAR-10 Batch 2:  loss 0.017978, train_accuracy 1, valid accuracy 0.7974
Epoch 544, CIFAR-10 Batch 3:  loss 0.000290, train_accuracy 1, valid accuracy 0.8038
Epoch 544, CIFAR-10 Batch 4:  loss 0.500520, train_accuracy 0.875, valid accuracy 0.6604
Epoch 544, CIFAR-10 Batch 5:  loss 0.001530, train_accuracy 1, valid accuracy 0.7932
Epoch 545, CIFAR-10 Batch 1:  loss 0.000157, train_accuracy 1, valid accuracy 0.8034
Epoch 545, CIFAR-10 Batch 2:  loss 0.000682, train_accuracy 1, valid accuracy 0.7986
Epoch 545, CIFAR-10 Batch 3:  loss 0.000213, train_accuracy 1, valid accuracy 0.8036
Epoch 545, CIFAR-10 Batch 4:  loss 0.000301, train_accuracy 1, valid accuracy 0.8052
Epoch 545, CIFAR-10 Batch 5:  loss 0.000497, train_accuracy 1, valid accuracy 0.8078
Epoch 546, CIFAR-10 Batch 1:  loss 0.001270, train_accuracy 1, valid accuracy 0.8064
Epoch 546, CIFAR-10 Batch 2:  loss 0.007971, train_accuracy 1, valid accuracy 0.796
Epoch 546, CIFAR-10 Batch 3:  loss 0.000229, train_accuracy 1, valid accuracy 0.8056
Epoch 546, CIFAR-10 Batch 4:  loss 0.000033, train_accuracy 1, valid accuracy 0.813
Epoch 546, CIFAR-10 Batch 5:  loss 0.002391, train_accuracy 1, valid accuracy 0.8114
Epoch 547, CIFAR-10 Batch 1:  loss 0.000048, train_accuracy 1, valid accuracy 0.8144
Epoch 547, CIFAR-10 Batch 2:  loss 0.001799, train_accuracy 1, valid accuracy 0.7826
Epoch 547, CIFAR-10 Batch 3:  loss 0.000081, train_accuracy 1, valid accuracy 0.8126
Epoch 547, CIFAR-10 Batch 4:  loss 0.000221, train_accuracy 1, valid accuracy 0.8032
Epoch 547, CIFAR-10 Batch 5:  loss 0.001359, train_accuracy 1, valid accuracy 0.8074
Epoch 548, CIFAR-10 Batch 1:  loss 0.000200, train_accuracy 1, valid accuracy 0.81
Epoch 548, CIFAR-10 Batch 2:  loss 0.000084, train_accuracy 1, valid accuracy 0.8106
Epoch 548, CIFAR-10 Batch 3:  loss 0.000322, train_accuracy 1, valid accuracy 0.8018
Epoch 548, CIFAR-10 Batch 4:  loss 0.000261, train_accuracy 1, valid accuracy 0.802
Epoch 548, CIFAR-10 Batch 5:  loss 0.000370, train_accuracy 1, valid accuracy 0.8092
Epoch 549, CIFAR-10 Batch 1:  loss 0.000198, train_accuracy 1, valid accuracy 0.8008
Epoch 549, CIFAR-10 Batch 2:  loss 0.003122, train_accuracy 1, valid accuracy 0.7976
Epoch 549, CIFAR-10 Batch 3:  loss 0.000602, train_accuracy 1, valid accuracy 0.8076
Epoch 549, CIFAR-10 Batch 4:  loss 0.000099, train_accuracy 1, valid accuracy 0.8078
Epoch 549, CIFAR-10 Batch 5:  loss 0.000381, train_accuracy 1, valid accuracy 0.803
Epoch 550, CIFAR-10 Batch 1:  loss 0.000317, train_accuracy 1, valid accuracy 0.8018
Epoch 550, CIFAR-10 Batch 2:  loss 0.000614, train_accuracy 1, valid accuracy 0.8096
Epoch 550, CIFAR-10 Batch 3:  loss 0.000236, train_accuracy 1, valid accuracy 0.8044
Epoch 550, CIFAR-10 Batch 4:  loss 0.000067, train_accuracy 1, valid accuracy 0.8098
Epoch 550, CIFAR-10 Batch 5:  loss 0.000112, train_accuracy 1, valid accuracy 0.8094
Epoch 551, CIFAR-10 Batch 1:  loss 0.000550, train_accuracy 1, valid accuracy 0.8062
Epoch 551, CIFAR-10 Batch 2:  loss 0.000394, train_accuracy 1, valid accuracy 0.8124
Epoch 551, CIFAR-10 Batch 3:  loss 0.000114, train_accuracy 1, valid accuracy 0.807
Epoch 551, CIFAR-10 Batch 4:  loss 0.000165, train_accuracy 1, valid accuracy 0.7972
Epoch 551, CIFAR-10 Batch 5:  loss 0.000991, train_accuracy 1, valid accuracy 0.8046
Epoch 552, CIFAR-10 Batch 1:  loss 0.000270, train_accuracy 1, valid accuracy 0.8096
Epoch 552, CIFAR-10 Batch 2:  loss 0.002111, train_accuracy 1, valid accuracy 0.7952
Epoch 552, CIFAR-10 Batch 3:  loss 0.000676, train_accuracy 1, valid accuracy 0.8038
Epoch 552, CIFAR-10 Batch 4:  loss 0.000137, train_accuracy 1, valid accuracy 0.7962
Epoch 552, CIFAR-10 Batch 5:  loss 0.002367, train_accuracy 1, valid accuracy 0.7804
Epoch 553, CIFAR-10 Batch 1:  loss 0.000701, train_accuracy 1, valid accuracy 0.81
Epoch 553, CIFAR-10 Batch 2:  loss 0.000209, train_accuracy 1, valid accuracy 0.8076
Epoch 553, CIFAR-10 Batch 3:  loss 0.000464, train_accuracy 1, valid accuracy 0.8046
Epoch 553, CIFAR-10 Batch 4:  loss 0.000327, train_accuracy 1, valid accuracy 0.7986
Epoch 553, CIFAR-10 Batch 5:  loss 0.000800, train_accuracy 1, valid accuracy 0.7934
Epoch 554, CIFAR-10 Batch 1:  loss 0.000424, train_accuracy 1, valid accuracy 0.8008
Epoch 554, CIFAR-10 Batch 2:  loss 0.000510, train_accuracy 1, valid accuracy 0.7972
Epoch 554, CIFAR-10 Batch 3:  loss 0.000445, train_accuracy 1, valid accuracy 0.8004
Epoch 554, CIFAR-10 Batch 4:  loss 0.004401, train_accuracy 1, valid accuracy 0.7886
Epoch 554, CIFAR-10 Batch 5:  loss 0.000346, train_accuracy 1, valid accuracy 0.8116
Epoch 555, CIFAR-10 Batch 1:  loss 0.000770, train_accuracy 1, valid accuracy 0.8112
Epoch 555, CIFAR-10 Batch 2:  loss 0.002207, train_accuracy 1, valid accuracy 0.7994
Epoch 555, CIFAR-10 Batch 3:  loss 0.000111, train_accuracy 1, valid accuracy 0.813
Epoch 555, CIFAR-10 Batch 4:  loss 0.000049, train_accuracy 1, valid accuracy 0.8086
Epoch 555, CIFAR-10 Batch 5:  loss 0.000193, train_accuracy 1, valid accuracy 0.8076
Epoch 556, CIFAR-10 Batch 1:  loss 0.000509, train_accuracy 1, valid accuracy 0.811
Epoch 556, CIFAR-10 Batch 2:  loss 0.000117, train_accuracy 1, valid accuracy 0.7982
Epoch 556, CIFAR-10 Batch 3:  loss 0.000331, train_accuracy 1, valid accuracy 0.7992
Epoch 556, CIFAR-10 Batch 4:  loss 0.000263, train_accuracy 1, valid accuracy 0.81
Epoch 556, CIFAR-10 Batch 5:  loss 0.000139, train_accuracy 1, valid accuracy 0.814
Epoch 557, CIFAR-10 Batch 1:  loss 0.000436, train_accuracy 1, valid accuracy 0.8016
Epoch 557, CIFAR-10 Batch 2:  loss 0.000191, train_accuracy 1, valid accuracy 0.8016
Epoch 557, CIFAR-10 Batch 3:  loss 0.000343, train_accuracy 1, valid accuracy 0.803
Epoch 557, CIFAR-10 Batch 4:  loss 0.000518, train_accuracy 1, valid accuracy 0.7996
Epoch 557, CIFAR-10 Batch 5:  loss 0.002704, train_accuracy 1, valid accuracy 0.7958
Epoch 558, CIFAR-10 Batch 1:  loss 0.000787, train_accuracy 1, valid accuracy 0.798
Epoch 558, CIFAR-10 Batch 2:  loss 0.001183, train_accuracy 1, valid accuracy 0.7892
Epoch 558, CIFAR-10 Batch 3:  loss 0.000328, train_accuracy 1, valid accuracy 0.7932
Epoch 558, CIFAR-10 Batch 4:  loss 0.000191, train_accuracy 1, valid accuracy 0.8048
Epoch 558, CIFAR-10 Batch 5:  loss 0.000540, train_accuracy 1, valid accuracy 0.8032
Epoch 559, CIFAR-10 Batch 1:  loss 0.000758, train_accuracy 1, valid accuracy 0.8022
Epoch 559, CIFAR-10 Batch 2:  loss 0.000145, train_accuracy 1, valid accuracy 0.8046
Epoch 559, CIFAR-10 Batch 3:  loss 0.001799, train_accuracy 1, valid accuracy 0.7908
Epoch 559, CIFAR-10 Batch 4:  loss 0.000518, train_accuracy 1, valid accuracy 0.7972
Epoch 559, CIFAR-10 Batch 5:  loss 0.000639, train_accuracy 1, valid accuracy 0.8076
Epoch 560, CIFAR-10 Batch 1:  loss 0.000298, train_accuracy 1, valid accuracy 0.8034
Epoch 560, CIFAR-10 Batch 2:  loss 0.000402, train_accuracy 1, valid accuracy 0.7992
Epoch 560, CIFAR-10 Batch 3:  loss 0.000209, train_accuracy 1, valid accuracy 0.815
Epoch 560, CIFAR-10 Batch 4:  loss 0.000209, train_accuracy 1, valid accuracy 0.8048
Epoch 560, CIFAR-10 Batch 5:  loss 0.000425, train_accuracy 1, valid accuracy 0.794
Epoch 561, CIFAR-10 Batch 1:  loss 0.000169, train_accuracy 1, valid accuracy 0.8116
Epoch 561, CIFAR-10 Batch 2:  loss 0.000062, train_accuracy 1, valid accuracy 0.8044
Epoch 561, CIFAR-10 Batch 3:  loss 0.000134, train_accuracy 1, valid accuracy 0.8032
Epoch 561, CIFAR-10 Batch 4:  loss 0.000114, train_accuracy 1, valid accuracy 0.808
Epoch 561, CIFAR-10 Batch 5:  loss 0.000252, train_accuracy 1, valid accuracy 0.8054
Epoch 562, CIFAR-10 Batch 1:  loss 0.000153, train_accuracy 1, valid accuracy 0.7964
Epoch 562, CIFAR-10 Batch 2:  loss 0.000141, train_accuracy 1, valid accuracy 0.7972
Epoch 562, CIFAR-10 Batch 3:  loss 0.000288, train_accuracy 1, valid accuracy 0.8016
Epoch 562, CIFAR-10 Batch 4:  loss 0.000963, train_accuracy 1, valid accuracy 0.793
Epoch 562, CIFAR-10 Batch 5:  loss 0.000287, train_accuracy 1, valid accuracy 0.8086
Epoch 563, CIFAR-10 Batch 1:  loss 0.001019, train_accuracy 1, valid accuracy 0.8048
Epoch 563, CIFAR-10 Batch 2:  loss 0.000312, train_accuracy 1, valid accuracy 0.8114
Epoch 563, CIFAR-10 Batch 3:  loss 0.000478, train_accuracy 1, valid accuracy 0.8038
Epoch 563, CIFAR-10 Batch 4:  loss 0.000184, train_accuracy 1, valid accuracy 0.7978
Epoch 563, CIFAR-10 Batch 5:  loss 0.000076, train_accuracy 1, valid accuracy 0.8102
Epoch 564, CIFAR-10 Batch 1:  loss 0.000238, train_accuracy 1, valid accuracy 0.8032
Epoch 564, CIFAR-10 Batch 2:  loss 0.000212, train_accuracy 1, valid accuracy 0.8028
Epoch 564, CIFAR-10 Batch 3:  loss 0.000714, train_accuracy 1, valid accuracy 0.7974
Epoch 564, CIFAR-10 Batch 4:  loss 0.000566, train_accuracy 1, valid accuracy 0.7994
Epoch 564, CIFAR-10 Batch 5:  loss 0.000060, train_accuracy 1, valid accuracy 0.8084
Epoch 565, CIFAR-10 Batch 1:  loss 0.001603, train_accuracy 1, valid accuracy 0.7972
Epoch 565, CIFAR-10 Batch 2:  loss 0.000180, train_accuracy 1, valid accuracy 0.7946
Epoch 565, CIFAR-10 Batch 3:  loss 0.000338, train_accuracy 1, valid accuracy 0.8042
Epoch 565, CIFAR-10 Batch 4:  loss 0.000231, train_accuracy 1, valid accuracy 0.802
Epoch 565, CIFAR-10 Batch 5:  loss 0.000098, train_accuracy 1, valid accuracy 0.807
Epoch 566, CIFAR-10 Batch 1:  loss 0.000268, train_accuracy 1, valid accuracy 0.8092
Epoch 566, CIFAR-10 Batch 2:  loss 0.000065, train_accuracy 1, valid accuracy 0.8064
Epoch 566, CIFAR-10 Batch 3:  loss 0.000129, train_accuracy 1, valid accuracy 0.8094
Epoch 566, CIFAR-10 Batch 4:  loss 0.000065, train_accuracy 1, valid accuracy 0.8026
Epoch 566, CIFAR-10 Batch 5:  loss 0.000116, train_accuracy 1, valid accuracy 0.808
Epoch 567, CIFAR-10 Batch 1:  loss 0.000158, train_accuracy 1, valid accuracy 0.8018
Epoch 567, CIFAR-10 Batch 2:  loss 0.000083, train_accuracy 1, valid accuracy 0.8068
Epoch 567, CIFAR-10 Batch 3:  loss 0.000340, train_accuracy 1, valid accuracy 0.8108
Epoch 567, CIFAR-10 Batch 4:  loss 0.000189, train_accuracy 1, valid accuracy 0.8048
Epoch 567, CIFAR-10 Batch 5:  loss 0.000147, train_accuracy 1, valid accuracy 0.8048
Epoch 568, CIFAR-10 Batch 1:  loss 0.000148, train_accuracy 1, valid accuracy 0.8038
Epoch 568, CIFAR-10 Batch 2:  loss 0.001200, train_accuracy 1, valid accuracy 0.796
Epoch 568, CIFAR-10 Batch 3:  loss 0.000235, train_accuracy 1, valid accuracy 0.8036
Epoch 568, CIFAR-10 Batch 4:  loss 0.000696, train_accuracy 1, valid accuracy 0.8028
Epoch 568, CIFAR-10 Batch 5:  loss 0.003385, train_accuracy 1, valid accuracy 0.7932
Epoch 569, CIFAR-10 Batch 1:  loss 0.000190, train_accuracy 1, valid accuracy 0.8
Epoch 569, CIFAR-10 Batch 2:  loss 0.000694, train_accuracy 1, valid accuracy 0.8062
Epoch 569, CIFAR-10 Batch 3:  loss 0.000239, train_accuracy 1, valid accuracy 0.803
Epoch 569, CIFAR-10 Batch 4:  loss 0.000157, train_accuracy 1, valid accuracy 0.81
Epoch 569, CIFAR-10 Batch 5:  loss 0.001001, train_accuracy 1, valid accuracy 0.806
Epoch 570, CIFAR-10 Batch 1:  loss 0.002764, train_accuracy 1, valid accuracy 0.7882
Epoch 570, CIFAR-10 Batch 2:  loss 0.000536, train_accuracy 1, valid accuracy 0.8
Epoch 570, CIFAR-10 Batch 3:  loss 0.001791, train_accuracy 1, valid accuracy 0.7966
Epoch 570, CIFAR-10 Batch 4:  loss 0.000469, train_accuracy 1, valid accuracy 0.8028
Epoch 570, CIFAR-10 Batch 5:  loss 0.004230, train_accuracy 1, valid accuracy 0.787
Epoch 571, CIFAR-10 Batch 1:  loss 0.000392, train_accuracy 1, valid accuracy 0.8026
Epoch 571, CIFAR-10 Batch 2:  loss 0.000342, train_accuracy 1, valid accuracy 0.795
Epoch 571, CIFAR-10 Batch 3:  loss 0.000208, train_accuracy 1, valid accuracy 0.8068
Epoch 571, CIFAR-10 Batch 4:  loss 0.000329, train_accuracy 1, valid accuracy 0.7954
Epoch 571, CIFAR-10 Batch 5:  loss 0.000117, train_accuracy 1, valid accuracy 0.801
Epoch 572, CIFAR-10 Batch 1:  loss 0.000079, train_accuracy 1, valid accuracy 0.8108
Epoch 572, CIFAR-10 Batch 2:  loss 0.000100, train_accuracy 1, valid accuracy 0.8112
Epoch 572, CIFAR-10 Batch 3:  loss 0.000162, train_accuracy 1, valid accuracy 0.8114
Epoch 572, CIFAR-10 Batch 4:  loss 0.000135, train_accuracy 1, valid accuracy 0.8094
Epoch 572, CIFAR-10 Batch 5:  loss 0.000129, train_accuracy 1, valid accuracy 0.8066
Epoch 573, CIFAR-10 Batch 1:  loss 0.000085, train_accuracy 1, valid accuracy 0.8102
Epoch 573, CIFAR-10 Batch 2:  loss 0.001554, train_accuracy 1, valid accuracy 0.7952
Epoch 573, CIFAR-10 Batch 3:  loss 0.000070, train_accuracy 1, valid accuracy 0.8008
Epoch 573, CIFAR-10 Batch 4:  loss 0.001038, train_accuracy 1, valid accuracy 0.8
Epoch 573, CIFAR-10 Batch 5:  loss 0.000721, train_accuracy 1, valid accuracy 0.8028
Epoch 574, CIFAR-10 Batch 1:  loss 0.000305, train_accuracy 1, valid accuracy 0.8038
Epoch 574, CIFAR-10 Batch 2:  loss 0.000336, train_accuracy 1, valid accuracy 0.801
Epoch 574, CIFAR-10 Batch 3:  loss 0.001490, train_accuracy 1, valid accuracy 0.7934
Epoch 574, CIFAR-10 Batch 4:  loss 0.000844, train_accuracy 1, valid accuracy 0.801
Epoch 574, CIFAR-10 Batch 5:  loss 0.000396, train_accuracy 1, valid accuracy 0.8008
Epoch 575, CIFAR-10 Batch 1:  loss 0.000117, train_accuracy 1, valid accuracy 0.8156
Epoch 575, CIFAR-10 Batch 2:  loss 0.000365, train_accuracy 1, valid accuracy 0.7984
Epoch 575, CIFAR-10 Batch 3:  loss 0.008616, train_accuracy 1, valid accuracy 0.794
Epoch 575, CIFAR-10 Batch 4:  loss 0.002653, train_accuracy 1, valid accuracy 0.7982
Epoch 575, CIFAR-10 Batch 5:  loss 0.000202, train_accuracy 1, valid accuracy 0.7988
Epoch 576, CIFAR-10 Batch 1:  loss 0.000094, train_accuracy 1, valid accuracy 0.8076
Epoch 576, CIFAR-10 Batch 2:  loss 0.000071, train_accuracy 1, valid accuracy 0.7994
Epoch 576, CIFAR-10 Batch 3:  loss 0.000101, train_accuracy 1, valid accuracy 0.8038
Epoch 576, CIFAR-10 Batch 4:  loss 0.000243, train_accuracy 1, valid accuracy 0.803
Epoch 576, CIFAR-10 Batch 5:  loss 0.000511, train_accuracy 1, valid accuracy 0.7902
Epoch 577, CIFAR-10 Batch 1:  loss 0.000051, train_accuracy 1, valid accuracy 0.8096
Epoch 577, CIFAR-10 Batch 2:  loss 0.000175, train_accuracy 1, valid accuracy 0.8136
Epoch 577, CIFAR-10 Batch 3:  loss 0.000606, train_accuracy 1, valid accuracy 0.8006
Epoch 577, CIFAR-10 Batch 4:  loss 0.000374, train_accuracy 1, valid accuracy 0.809
Epoch 577, CIFAR-10 Batch 5:  loss 0.000540, train_accuracy 1, valid accuracy 0.799
Epoch 578, CIFAR-10 Batch 1:  loss 0.000114, train_accuracy 1, valid accuracy 0.8038
Epoch 578, CIFAR-10 Batch 2:  loss 0.000062, train_accuracy 1, valid accuracy 0.8132
Epoch 578, CIFAR-10 Batch 3:  loss 0.000136, train_accuracy 1, valid accuracy 0.804
Epoch 578, CIFAR-10 Batch 4:  loss 0.001308, train_accuracy 1, valid accuracy 0.8132
Epoch 578, CIFAR-10 Batch 5:  loss 0.000346, train_accuracy 1, valid accuracy 0.7996
Epoch 579, CIFAR-10 Batch 1:  loss 0.000215, train_accuracy 1, valid accuracy 0.8108
Epoch 579, CIFAR-10 Batch 2:  loss 0.000122, train_accuracy 1, valid accuracy 0.807
Epoch 579, CIFAR-10 Batch 3:  loss 0.000113, train_accuracy 1, valid accuracy 0.8112
Epoch 579, CIFAR-10 Batch 4:  loss 0.000138, train_accuracy 1, valid accuracy 0.8068
Epoch 579, CIFAR-10 Batch 5:  loss 0.000194, train_accuracy 1, valid accuracy 0.808
Epoch 580, CIFAR-10 Batch 1:  loss 0.000415, train_accuracy 1, valid accuracy 0.8014
Epoch 580, CIFAR-10 Batch 2:  loss 0.019873, train_accuracy 1, valid accuracy 0.7578
Epoch 580, CIFAR-10 Batch 3:  loss 0.000271, train_accuracy 1, valid accuracy 0.8028
Epoch 580, CIFAR-10 Batch 4:  loss 0.001503, train_accuracy 1, valid accuracy 0.8066
Epoch 580, CIFAR-10 Batch 5:  loss 0.000117, train_accuracy 1, valid accuracy 0.8072
Epoch 581, CIFAR-10 Batch 1:  loss 0.000516, train_accuracy 1, valid accuracy 0.8012
Epoch 581, CIFAR-10 Batch 2:  loss 0.000699, train_accuracy 1, valid accuracy 0.8114
Epoch 581, CIFAR-10 Batch 3:  loss 0.001734, train_accuracy 1, valid accuracy 0.8034
Epoch 581, CIFAR-10 Batch 4:  loss 0.000718, train_accuracy 1, valid accuracy 0.8088
Epoch 581, CIFAR-10 Batch 5:  loss 0.000722, train_accuracy 1, valid accuracy 0.8066
Epoch 582, CIFAR-10 Batch 1:  loss 0.000162, train_accuracy 1, valid accuracy 0.7994
Epoch 582, CIFAR-10 Batch 2:  loss 0.000147, train_accuracy 1, valid accuracy 0.8114
Epoch 582, CIFAR-10 Batch 3:  loss 0.000234, train_accuracy 1, valid accuracy 0.807
Epoch 582, CIFAR-10 Batch 4:  loss 0.000461, train_accuracy 1, valid accuracy 0.7928
Epoch 582, CIFAR-10 Batch 5:  loss 0.000102, train_accuracy 1, valid accuracy 0.807
Epoch 583, CIFAR-10 Batch 1:  loss 0.000034, train_accuracy 1, valid accuracy 0.8086
Epoch 583, CIFAR-10 Batch 2:  loss 0.000183, train_accuracy 1, valid accuracy 0.7916
Epoch 583, CIFAR-10 Batch 3:  loss 0.000951, train_accuracy 1, valid accuracy 0.806
Epoch 583, CIFAR-10 Batch 4:  loss 0.000530, train_accuracy 1, valid accuracy 0.8018
Epoch 583, CIFAR-10 Batch 5:  loss 0.000179, train_accuracy 1, valid accuracy 0.8032
Epoch 584, CIFAR-10 Batch 1:  loss 0.000098, train_accuracy 1, valid accuracy 0.803
Epoch 584, CIFAR-10 Batch 2:  loss 0.000160, train_accuracy 1, valid accuracy 0.8044
Epoch 584, CIFAR-10 Batch 3:  loss 0.000445, train_accuracy 1, valid accuracy 0.8106
Epoch 584, CIFAR-10 Batch 4:  loss 0.000096, train_accuracy 1, valid accuracy 0.8028
Epoch 584, CIFAR-10 Batch 5:  loss 0.000050, train_accuracy 1, valid accuracy 0.8118
Epoch 585, CIFAR-10 Batch 1:  loss 0.000130, train_accuracy 1, valid accuracy 0.8088
Epoch 585, CIFAR-10 Batch 2:  loss 0.000353, train_accuracy 1, valid accuracy 0.798
Epoch 585, CIFAR-10 Batch 3:  loss 0.000217, train_accuracy 1, valid accuracy 0.8082
Epoch 585, CIFAR-10 Batch 4:  loss 0.000173, train_accuracy 1, valid accuracy 0.8054
Epoch 585, CIFAR-10 Batch 5:  loss 0.000565, train_accuracy 1, valid accuracy 0.8016
Epoch 586, CIFAR-10 Batch 1:  loss 0.000150, train_accuracy 1, valid accuracy 0.8082
Epoch 586, CIFAR-10 Batch 2:  loss 0.002231, train_accuracy 1, valid accuracy 0.7952
Epoch 586, CIFAR-10 Batch 3:  loss 0.000056, train_accuracy 1, valid accuracy 0.8104
Epoch 586, CIFAR-10 Batch 4:  loss 0.000105, train_accuracy 1, valid accuracy 0.8102
Epoch 586, CIFAR-10 Batch 5:  loss 0.000133, train_accuracy 1, valid accuracy 0.7978
Epoch 587, CIFAR-10 Batch 1:  loss 0.000253, train_accuracy 1, valid accuracy 0.8048
Epoch 587, CIFAR-10 Batch 2:  loss 0.000103, train_accuracy 1, valid accuracy 0.806
Epoch 587, CIFAR-10 Batch 3:  loss 0.000803, train_accuracy 1, valid accuracy 0.7964
Epoch 587, CIFAR-10 Batch 4:  loss 0.001621, train_accuracy 1, valid accuracy 0.7994
Epoch 587, CIFAR-10 Batch 5:  loss 0.000124, train_accuracy 1, valid accuracy 0.8058
Epoch 588, CIFAR-10 Batch 1:  loss 0.000293, train_accuracy 1, valid accuracy 0.7958
Epoch 588, CIFAR-10 Batch 2:  loss 0.001063, train_accuracy 1, valid accuracy 0.8002
Epoch 588, CIFAR-10 Batch 3:  loss 0.000528, train_accuracy 1, valid accuracy 0.7968
Epoch 588, CIFAR-10 Batch 4:  loss 0.000492, train_accuracy 1, valid accuracy 0.8048
Epoch 588, CIFAR-10 Batch 5:  loss 0.006819, train_accuracy 1, valid accuracy 0.7882
Epoch 589, CIFAR-10 Batch 1:  loss 0.000164, train_accuracy 1, valid accuracy 0.7986
Epoch 589, CIFAR-10 Batch 2:  loss 0.001509, train_accuracy 1, valid accuracy 0.8006
Epoch 589, CIFAR-10 Batch 3:  loss 0.000252, train_accuracy 1, valid accuracy 0.8088
Epoch 589, CIFAR-10 Batch 4:  loss 0.000111, train_accuracy 1, valid accuracy 0.8016
Epoch 589, CIFAR-10 Batch 5:  loss 0.001602, train_accuracy 1, valid accuracy 0.7836
Epoch 590, CIFAR-10 Batch 1:  loss 0.000049, train_accuracy 1, valid accuracy 0.8048
Epoch 590, CIFAR-10 Batch 2:  loss 0.000283, train_accuracy 1, valid accuracy 0.8006
Epoch 590, CIFAR-10 Batch 3:  loss 0.000064, train_accuracy 1, valid accuracy 0.7992
Epoch 590, CIFAR-10 Batch 4:  loss 0.000208, train_accuracy 1, valid accuracy 0.8116
Epoch 590, CIFAR-10 Batch 5:  loss 0.000403, train_accuracy 1, valid accuracy 0.8052
Epoch 591, CIFAR-10 Batch 1:  loss 0.000239, train_accuracy 1, valid accuracy 0.8056
Epoch 591, CIFAR-10 Batch 2:  loss 0.000380, train_accuracy 1, valid accuracy 0.7994
Epoch 591, CIFAR-10 Batch 3:  loss 0.000322, train_accuracy 1, valid accuracy 0.802
Epoch 591, CIFAR-10 Batch 4:  loss 0.000230, train_accuracy 1, valid accuracy 0.811
Epoch 591, CIFAR-10 Batch 5:  loss 0.000316, train_accuracy 1, valid accuracy 0.8052
Epoch 592, CIFAR-10 Batch 1:  loss 0.000176, train_accuracy 1, valid accuracy 0.8076
Epoch 592, CIFAR-10 Batch 2:  loss 0.000059, train_accuracy 1, valid accuracy 0.8132
Epoch 592, CIFAR-10 Batch 3:  loss 0.000471, train_accuracy 1, valid accuracy 0.8028
Epoch 592, CIFAR-10 Batch 4:  loss 0.000215, train_accuracy 1, valid accuracy 0.8078
Epoch 592, CIFAR-10 Batch 5:  loss 0.000151, train_accuracy 1, valid accuracy 0.8118
Epoch 593, CIFAR-10 Batch 1:  loss 0.000077, train_accuracy 1, valid accuracy 0.8054
Epoch 593, CIFAR-10 Batch 2:  loss 0.000226, train_accuracy 1, valid accuracy 0.805
Epoch 593, CIFAR-10 Batch 3:  loss 0.000529, train_accuracy 1, valid accuracy 0.8094
Epoch 593, CIFAR-10 Batch 4:  loss 0.000306, train_accuracy 1, valid accuracy 0.8074
Epoch 593, CIFAR-10 Batch 5:  loss 0.000133, train_accuracy 1, valid accuracy 0.8074
Epoch 594, CIFAR-10 Batch 1:  loss 0.000086, train_accuracy 1, valid accuracy 0.8014
Epoch 594, CIFAR-10 Batch 2:  loss 0.000257, train_accuracy 1, valid accuracy 0.797
Epoch 594, CIFAR-10 Batch 3:  loss 0.000235, train_accuracy 1, valid accuracy 0.8084
Epoch 594, CIFAR-10 Batch 4:  loss 0.000223, train_accuracy 1, valid accuracy 0.8064
Epoch 594, CIFAR-10 Batch 5:  loss 0.001470, train_accuracy 1, valid accuracy 0.7962
Epoch 595, CIFAR-10 Batch 1:  loss 0.000135, train_accuracy 1, valid accuracy 0.808
Epoch 595, CIFAR-10 Batch 2:  loss 0.000279, train_accuracy 1, valid accuracy 0.7972
Epoch 595, CIFAR-10 Batch 3:  loss 0.000079, train_accuracy 1, valid accuracy 0.8078
Epoch 595, CIFAR-10 Batch 4:  loss 0.000344, train_accuracy 1, valid accuracy 0.7984
Epoch 595, CIFAR-10 Batch 5:  loss 0.002672, train_accuracy 1, valid accuracy 0.7806
Epoch 596, CIFAR-10 Batch 1:  loss 0.000265, train_accuracy 1, valid accuracy 0.802
Epoch 596, CIFAR-10 Batch 2:  loss 0.000213, train_accuracy 1, valid accuracy 0.806
Epoch 596, CIFAR-10 Batch 3:  loss 0.001343, train_accuracy 1, valid accuracy 0.8004
Epoch 596, CIFAR-10 Batch 4:  loss 0.000478, train_accuracy 1, valid accuracy 0.8098
Epoch 596, CIFAR-10 Batch 5:  loss 0.000104, train_accuracy 1, valid accuracy 0.8088
Epoch 597, CIFAR-10 Batch 1:  loss 0.000081, train_accuracy 1, valid accuracy 0.8068
Epoch 597, CIFAR-10 Batch 2:  loss 0.000357, train_accuracy 1, valid accuracy 0.7948
Epoch 597, CIFAR-10 Batch 3:  loss 0.000628, train_accuracy 1, valid accuracy 0.8058
Epoch 597, CIFAR-10 Batch 4:  loss 0.000362, train_accuracy 1, valid accuracy 0.7976
Epoch 597, CIFAR-10 Batch 5:  loss 0.000269, train_accuracy 1, valid accuracy 0.7948
Epoch 598, CIFAR-10 Batch 1:  loss 0.000070, train_accuracy 1, valid accuracy 0.8098
Epoch 598, CIFAR-10 Batch 2:  loss 0.000346, train_accuracy 1, valid accuracy 0.8082
Epoch 598, CIFAR-10 Batch 3:  loss 0.000077, train_accuracy 1, valid accuracy 0.8024
Epoch 598, CIFAR-10 Batch 4:  loss 0.000479, train_accuracy 1, valid accuracy 0.8
Epoch 598, CIFAR-10 Batch 5:  loss 0.000262, train_accuracy 1, valid accuracy 0.807
Epoch 599, CIFAR-10 Batch 1:  loss 0.000124, train_accuracy 1, valid accuracy 0.8046
Epoch 599, CIFAR-10 Batch 2:  loss 0.000108, train_accuracy 1, valid accuracy 0.81
Epoch 599, CIFAR-10 Batch 3:  loss 0.000094, train_accuracy 1, valid accuracy 0.7984
Epoch 599, CIFAR-10 Batch 4:  loss 0.000218, train_accuracy 1, valid accuracy 0.8108
Epoch 599, CIFAR-10 Batch 5:  loss 0.000106, train_accuracy 1, valid accuracy 0.7952
Epoch 600, CIFAR-10 Batch 1:  loss 0.000316, train_accuracy 1, valid accuracy 0.805
Epoch 600, CIFAR-10 Batch 2:  loss 0.000227, train_accuracy 1, valid accuracy 0.8076
Epoch 600, CIFAR-10 Batch 3:  loss 0.007094, train_accuracy 1, valid accuracy 0.8006
Epoch 600, CIFAR-10 Batch 4:  loss 0.000157, train_accuracy 1, valid accuracy 0.805
Epoch 600, CIFAR-10 Batch 5:  loss 0.000052, train_accuracy 1, valid accuracy 0.7994

Checkpoint

The model has been saved to disk.

Test Model

Test your model against the test dataset. This will be your final accuracy. You should have an accuracy greater than 50%. If you don't, keep tweaking the model architecture and parameters.



In [166]:

    
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
%matplotlib inline
%config InlineBackend.figure_format = 'retina'

import tensorflow as tf
import pickle
import helper
import random

# Set batch size if not already set
try:
    if batch_size:
        pass
except NameError:
    batch_size = 64

save_model_path = './image_classification'
n_samples = 4
top_n_predictions = 3

def test_model():
    """
    Test the saved model against the test dataset
    """

    test_features, test_labels = pickle.load(open('preprocess_test.p', mode='rb'))
    loaded_graph = tf.Graph()

    with tf.Session(graph=loaded_graph) as sess:
        # Load model
        loader = tf.train.import_meta_graph(save_model_path + '.meta')
        loader.restore(sess, save_model_path)

        # Get Tensors from loaded model
        loaded_x = loaded_graph.get_tensor_by_name('x:0')
        loaded_y = loaded_graph.get_tensor_by_name('y:0')
        loaded_keep_prob = loaded_graph.get_tensor_by_name('keep_prob:0')
        loaded_logits = loaded_graph.get_tensor_by_name('logits:0')
        loaded_acc = loaded_graph.get_tensor_by_name('accuracy:0')
        
        # Get accuracy in batches for memory limitations
        test_batch_acc_total = 0
        test_batch_count = 0
        
        for test_feature_batch, test_label_batch in helper.batch_features_labels(test_features, test_labels, batch_size):
            test_batch_acc_total += sess.run(
                loaded_acc,
                feed_dict={loaded_x: test_feature_batch, loaded_y: test_label_batch, loaded_keep_prob: 1.0})
            test_batch_count += 1

        print('Testing Accuracy: {}\n'.format(test_batch_acc_total/test_batch_count))

        # Print Random Samples
        random_test_features, random_test_labels = tuple(zip(*random.sample(list(zip(test_features, test_labels)), n_samples)))
        random_test_predictions = sess.run(
            tf.nn.top_k(tf.nn.softmax(loaded_logits), top_n_predictions),
            feed_dict={loaded_x: random_test_features, loaded_y: random_test_labels, loaded_keep_prob: 1.0})
        helper.display_image_predictions(random_test_features, random_test_labels, random_test_predictions)


test_model()









    



INFO:tensorflow:Restoring parameters from ./image_classification
Testing Accuracy: 0.79580078125

Why 50-80% Accuracy?

You might be wondering why you can't get an accuracy any higher. First things first, 50% isn't bad for a simple CNN. Pure guessing would get you 10% accuracy. However, you might notice people are getting scores well above 80%. That's because we haven't taught you all there is to know about neural networks. We still need to cover a few more techniques.

Submitting This Project

When submitting this project, make sure to run all the cells before saving the notebook. Save the notebook file as "dlnd_image_classification.ipynb" and save it as a HTML file under "File" -> "Download as". Include the "helper.py" and "problem_unittests.py" files in your submission.