Make necessary imports



In [1]:

    
import sys
sys.path.insert(0, '../')



In [2]:

    
import time
import numpy as np
np.set_printoptions(precision=3, linewidth=200, suppress=True)
import tensorflow as tf



In [3]:

    
from library.datasets.cifar10 import CIFAR10
from library.plot_tools import plot
from library.utils import file_utils
from library.tf.models.linear_classifier import LinearClassifier

Select Tensorflow device



In [4]:

    
# from tensorflow.python.client import device_lib
# local_device_protos = device_lib.list_local_devices()
# cpu_devices = [x.name for x in local_device_protos if x.device_type == 'CPU']
# gpu_devices = [x.name for x in local_device_protos if x.device_type == 'GPU']
# print('Available CPU Devices: ', end='')
# print(cpu_devices)
# print('Available GPU Devices: ', end='')
# print(gpu_devices)
# if len(gpu_devices) == 0:
#     if len(cpu_devices) > 0:
#         device_name = '/cpu:0'
#         print('Using CPU: %s' %device_name)
#     else:
#         print('No CPU present in the system!!!')
# else:
#     device_name = '/gpu:0'
#     print('Using GPU: %s' %device_name)



In [5]:

    
total_time = 0

Experiment



In [6]:

    
exp_no = 1
data_source = 'Website'
num_images_required = 1.0
file_no = 101

Parameters for experiment



In [7]:

    
device_name = '/gpu:0'
learn_rate = 0.01
train_epochs = 100
display_step = 1
reg_const = 0.01
train_val_split_data = None
train_val_split = 0.8
transform = True
transform_method = 'StandardScaler'
learn_rate_type = 'constant'
dataset = 'cifar10'
descent_methods = ['adam', 'gradient', 'adaprop', 'rmsprop', 'adadelta', 'momentum']

Log directories



In [8]:

    
log_dirs = []
log_files = []
model_files = []



In [9]:

    
for index, descent_method in enumerate(descent_methods):
    log_dir = '../logs/' + dataset + '/' + str(file_no).zfill(3) + '_tf_linear_raw_gradtest/exp_no_' + str(index+1).zfill(3) + '/'
    log_file = log_dir + 'linear_classifier.ckpt'
    model_file = log_dir + 'linear_classifier.pb'
    log_dirs.append(log_dir)
    log_files.append(log_file)
    model_files.append(model_file)
    print('Writing tensorboard logs to %s' % log_file)
    print('View logs by running tensorboard: ', end='')
    print('\"tensorboard --logdir=\'./%s/101_tf_linear_raw_gradtest/\' --port 61111\"' %dataset)









    



Writing tensorboard logs to ../logs/cifar10/101_tf_linear_raw_gradtest/exp_no_001/linear_classifier.ckpt
View logs by running tensorboard: "tensorboard --logdir='./cifar10/101_tf_linear_raw_gradtest/' --port 61111"
Writing tensorboard logs to ../logs/cifar10/101_tf_linear_raw_gradtest/exp_no_002/linear_classifier.ckpt
View logs by running tensorboard: "tensorboard --logdir='./cifar10/101_tf_linear_raw_gradtest/' --port 61111"
Writing tensorboard logs to ../logs/cifar10/101_tf_linear_raw_gradtest/exp_no_003/linear_classifier.ckpt
View logs by running tensorboard: "tensorboard --logdir='./cifar10/101_tf_linear_raw_gradtest/' --port 61111"
Writing tensorboard logs to ../logs/cifar10/101_tf_linear_raw_gradtest/exp_no_004/linear_classifier.ckpt
View logs by running tensorboard: "tensorboard --logdir='./cifar10/101_tf_linear_raw_gradtest/' --port 61111"
Writing tensorboard logs to ../logs/cifar10/101_tf_linear_raw_gradtest/exp_no_005/linear_classifier.ckpt
View logs by running tensorboard: "tensorboard --logdir='./cifar10/101_tf_linear_raw_gradtest/' --port 61111"
Writing tensorboard logs to ../logs/cifar10/101_tf_linear_raw_gradtest/exp_no_006/linear_classifier.ckpt
View logs by running tensorboard: "tensorboard --logdir='./cifar10/101_tf_linear_raw_gradtest/' --port 61111"

Step 1: Load CIFAR 10 Dataset



In [10]:

    
start = time.time()
one_hot = True
cifar10 = CIFAR10(one_hot_encode=one_hot, num_images=num_images_required, preprocess='StandardScaler',
                  train_validate_split=train_val_split_data, endian='little')
cifar10.load_data(train=True, test=True, data_directory='./datasets/cifar10/')
end = time.time()
print('[ Step 1] Loaded CIFAR 10 Dataset in %.4f ms' %((end-start)*1000))
total_time += (end-start)









    



Loading CIFAR 10 Dataset
Downloading and extracting CIFAR 10 file
MD5sum of the file: ./datasets/cifar10/cifar-10.tar.gz is verified
Loading 50000 train images
Loading CIFAR 10 Training Dataset
Reading unpicked data file: ./datasets/cifar10/cifar-10-batches/data_batch_1
Reading unpicked data file: ./datasets/cifar10/cifar-10-batches/data_batch_2
Reading unpicked data file: ./datasets/cifar10/cifar-10-batches/data_batch_3
Reading unpicked data file: ./datasets/cifar10/cifar-10-batches/data_batch_4
Reading unpicked data file: ./datasets/cifar10/cifar-10-batches/data_batch_5






    



/net/voxel03/misc/me/praneethas/Softwares/anaconda3/lib/python3.6/site-packages/sklearn/utils/validation.py:429: DataConversionWarning: Data with input dtype uint8 was converted to float64 by StandardScaler.
  warnings.warn(msg, _DataConversionWarning)






    



Loading 10000 test images
Loading CIFAR 10 Test Dataset
Unpickling test file: ./datasets/cifar10/cifar-10-batches/test_batch
Reading unpicked test file: ./datasets/cifar10/cifar-10-batches/test_batch
Loaded CIFAR 10 Dataset in 5.7875 seconds
[ Step 1] Loaded CIFAR 10 Dataset in 5788.6543 ms



In [11]:

    
print('Train data shape:', cifar10.train.data.shape)
if one_hot is True:
    print('Train labels shape:', cifar10.train.one_hot_labels.shape)
print('Train class labels shape:', cifar10.train.class_labels.shape)
if train_val_split_data is not None:
    print('Validate data shape:', cifar10.validate.data.shape)
    if one_hot is True:
        print('Validate labels shape:', cifar10.validate.one_hot_labels.shape)
    print('Validate class labels shape:', cifar10.validate.class_labels.shape)
print('Test data shape:', cifar10.test.data.shape)
if one_hot is True:
    print('Test labels shape:', cifar10.test.one_hot_labels.shape)
print('Test class labels shape:', cifar10.test.class_labels.shape)









    



Train data shape: (50000, 3072)
Train labels shape: (50000, 10)
Train class labels shape: (50000,)
Test data shape: (10000, 3072)
Test labels shape: (10000, 10)
Test class labels shape: (10000,)



In [12]:

    
print('Training images')
print(cifar10.train.data[:5])
if one_hot is True:
    print('Training labels')
    print(cifar10.train.one_hot_labels[:5])
print('Training classes')
print(cifar10.train.class_labels[:5])
print('Testing images')
print(cifar10.test.data[:5])
if one_hot is True:
    print('Testing labels')
    print(cifar10.test.one_hot_labels[:5])
print('Testing classes')
print(cifar10.test.class_labels[:5])









    



Training images
[[-0.977 -1.203 -1.122 ...,  0.403 -0.459 -0.641]
 [ 0.317 -0.057 -0.361 ...,  0.388  0.431  0.448]
 [ 1.693  1.696  1.688 ..., -0.475 -0.474 -0.46 ]
 [-1.399 -1.286 -1.288 ..., -1.323 -1.179 -1.035]
 [ 0.535  0.523  0.636 ..., -0.491 -0.551 -0.52 ]]
Training labels
[[ 0.  0.  0.  0.  0.  0.  1.  0.  0.  0.]
 [ 0.  0.  0.  0.  0.  0.  0.  0.  0.  1.]
 [ 0.  0.  0.  0.  0.  0.  0.  0.  0.  1.]
 [ 0.  0.  0.  0.  1.  0.  0.  0.  0.  0.]
 [ 0.  1.  0.  0.  0.  0.  0.  0.  0.  0.]]
Training classes
[6 9 9 4 1]
Testing images
[[ 0.375  0.401  0.468 ...,  0.142  0.219 -0.081]
 [ 1.425  1.395  1.395 ...,  0.98   1.174  1.274]
 [ 0.375  0.387  0.109 ..., -1.658 -1.722 -1.65 ]
 [ 0.334  0.511  0.621 ..., -1.006 -0.967 -0.995]
 [-0.894 -0.829 -1.15  ...,  0.328  0.481  0.025]]
Testing labels
[[ 0.  0.  0.  1.  0.  0.  0.  0.  0.  0.]
 [ 0.  0.  0.  0.  0.  0.  0.  0.  1.  0.]
 [ 0.  0.  0.  0.  0.  0.  0.  0.  1.  0.]
 [ 1.  0.  0.  0.  0.  0.  0.  0.  0.  0.]
 [ 0.  0.  0.  0.  0.  0.  1.  0.  0.  0.]]
Testing classes
[3 8 8 0 6]

Step 1.1 Load sample images



In [13]:

    
cifar10.plot_sample(plot_data=True, plot_test=True, fig_size=(7, 7))









    



Plotting CIFAR 10 Train Dataset






    












    



Plotting CIFAR 10 Test Dataset



In [14]:

    
cifar10.plot_images(cifar10.train.data[:50, :], cifar10.train.class_names[:50], 
                    nrows=5, ncols=10, fig_size=(20,50), fontsize=35, convert=True)









    












    Out[14]:





True

Step 2: Linear Regression

Step 2.1: Model linear regression y = Wx + b



In [15]:

    
num_features = cifar10.train.data.shape[1]
num_classes = 10



In [16]:

    
net_configs = []
for index, descent_method in enumerate(descent_methods):
    network_config = {'weight': {'type': 'zeros', 'name': 'Weight'},
                      'bias': {'type': 'ones', 'name': 'Bias'},
                      'activation_fn': 'relu',
                      'descent_method': descent_method}
    net_configs.append(network_config)



In [17]:

    
lcs = []
for index, descent_method in enumerate(descent_methods):
    lc = LinearClassifier(verbose=False, device=device_name, session_type='interactive', num_iter=train_epochs, 
                          err_tolerance=1e-7, train_validate_split=train_val_split, display_step=display_step,
                          learn_step=10, learn_rate_type=learn_rate_type, learn_rate=learn_rate, reg_const=reg_const, 
                          logs=True, log_dir=log_dirs[index], test_log=True, save_model=True, 
                          checkpoint_filename=log_files[index], 
                          save_checkpoint=True, restore=False, model_name=model_files[index], config=net_configs[index])
    lcs.append(lc)

Step 2.2: Create the tensorflow graph



In [18]:

    
start = time.time()
for index, descent_method in enumerate(descent_methods):
    lcs[index].create_graph(num_features=num_features, num_classes=num_classes)
end = time.time()
print('Generated the tensorflow graph in %.4f ms' %((end-start)*1000))
total_time = (end-start)









    



Session: <tensorflow.python.client.session.InteractiveSession object at 0x7f1012c7acf8>
Model has been trained for 0 iterations
Tensorflow graph created in 0.9352 seconds
Session: <tensorflow.python.client.session.InteractiveSession object at 0x7f100509de80>
Model has been trained for 0 iterations
Tensorflow graph created in 0.5989 seconds
Session: <tensorflow.python.client.session.InteractiveSession object at 0x7f100d0f56a0>
Model has been trained for 0 iterations
Tensorflow graph created in 0.4127 seconds
Session: <tensorflow.python.client.session.InteractiveSession object at 0x7f100d0d4a58>
Model has been trained for 0 iterations
Tensorflow graph created in 0.5035 seconds
Session: <tensorflow.python.client.session.InteractiveSession object at 0x7f1004e39860>
Model has been trained for 0 iterations
Tensorflow graph created in 0.6852 seconds
Session: <tensorflow.python.client.session.InteractiveSession object at 0x7f100c418cc0>
Model has been trained for 0 iterations
Tensorflow graph created in 0.8014 seconds
Generated the tensorflow graph in 3937.9349 ms



In [19]:

    
for index, descent_method in enumerate(descent_methods):
    lcs[index].print_parameters()









    



Parameters for Linear classifier
>> Input Parameters
Input                  : Tensor("Inputs/Data/X_input:0", shape=(?, 3072), dtype=float32, device=/device:GPU:0) 
True one hot labels    : Tensor("Inputs/Train_Labels/y_true_one_hot_label:0", shape=(?, 10), dtype=float32, device=/device:GPU:0) 
True class             : Tensor("Inputs/Train_Labels/y_true_class:0", shape=(?,), dtype=int64, device=/device:GPU:0) 
Predict one hot labels : Tensor("Predictions/one_hot:0", shape=(?, 10), dtype=float32, device=/device:GPU:0) 
Predict class          : Tensor("Predictions/predict_class:0", shape=(?,), dtype=int64, device=/device:GPU:0) 
>> Model params
last_successful_epoch/initial_value
last_successful_epoch
last_successful_epoch/Assign
last_successful_epoch/read
add/y
add
assign_updated_epoch
Inputs/Data/X_input
Inputs/Train_Labels/y_true_one_hot_label
Inputs/Train_Labels/y_true_class
Parameters/Weights/zeros
Parameters/Weights/Weight
Parameters/Weights/Weight/Assign
Parameters/Weights/Weight/read
Parameters/Bias/ones
Parameters/Bias/Bias
Parameters/Bias/Bias/Assign
Parameters/Bias/Bias/read
Predictions/mlp_layer/MatMul
Predictions/mlp_layer
Predictions/mlp_layer_1
Predictions/Softmax
Predictions/predict_class/dimension
Predictions/predict_class
Predictions/one_hot/Const
Predictions/one_hot/Const_1
Predictions/one_hot/depth
Predictions/one_hot/on_value
Predictions/one_hot/off_value
Predictions/one_hot
Cross_Entropy/Rank
Cross_Entropy/Shape
Cross_Entropy/Rank_1
Cross_Entropy/Shape_1
Cross_Entropy/Sub/y
Cross_Entropy/Sub
Cross_Entropy/Slice/begin
Cross_Entropy/Slice/size
Cross_Entropy/Slice
Cross_Entropy/concat/values_0
Cross_Entropy/concat/axis
Cross_Entropy/concat
Cross_Entropy/Reshape
Cross_Entropy/Rank_2
Cross_Entropy/Shape_2
Cross_Entropy/Sub_1/y
Cross_Entropy/Sub_1
Cross_Entropy/Slice_1/begin
Cross_Entropy/Slice_1/size
Cross_Entropy/Slice_1
Cross_Entropy/concat_1/values_0
Cross_Entropy/concat_1/axis
Cross_Entropy/concat_1
Cross_Entropy/Reshape_1
Cross_Entropy/SoftmaxCrossEntropyWithLogits
Cross_Entropy/Sub_2/y
Cross_Entropy/Sub_2
Cross_Entropy/Slice_2/begin
Cross_Entropy/Slice_2/size
Cross_Entropy/Slice_2
Cross_Entropy/Reshape_2
Loss_Function/Const
Loss_Function/Square
Loss_Function/Const_1
Loss_Function/Mean
Loss_Function/Const_2
Loss_Function/Mean_1
Loss_Function/Mul
Loss_Function/Add
Loss_Function/train_loss_list/initial_value
Loss_Function/train_loss_list
Loss_Function/train_loss_list/Assign
Loss_Function/train_loss_list/read
Loss_Function/Assign/value
Loss_Function/Assign
Loss_Function/train_error/tags
Loss_Function/train_error
Loss_Function/validate_loss_list/initial_value
Loss_Function/validate_loss_list
Loss_Function/validate_loss_list/Assign
Loss_Function/validate_loss_list/read
Loss_Function/Assign_1/value
Loss_Function/Assign_1
Loss_Function/validate_error/tags
Loss_Function/validate_error
Optimizer/learning_rate_progress/initial_value
Optimizer/learning_rate_progress
Optimizer/learning_rate_progress/Assign
Optimizer/learning_rate_progress/read
Optimizer/Assign/value
Optimizer/Assign
Optimizer/learning_rate/tags
Optimizer/learning_rate/values
Optimizer/learning_rate
Optimizer/gradients/Shape
Optimizer/gradients/Const
Optimizer/gradients/Fill
Optimizer/gradients/Loss_Function/Add_grad/Shape
Optimizer/gradients/Loss_Function/Add_grad/Shape_1
Optimizer/gradients/Loss_Function/Add_grad/BroadcastGradientArgs
Optimizer/gradients/Loss_Function/Add_grad/Sum
Optimizer/gradients/Loss_Function/Add_grad/Reshape
Optimizer/gradients/Loss_Function/Add_grad/Sum_1
Optimizer/gradients/Loss_Function/Add_grad/Reshape_1
Optimizer/gradients/Loss_Function/Add_grad/tuple/group_deps
Optimizer/gradients/Loss_Function/Add_grad/tuple/control_dependency
Optimizer/gradients/Loss_Function/Add_grad/tuple/control_dependency_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Reshape/shape
Optimizer/gradients/Loss_Function/Mean_1_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape
Optimizer/gradients/Loss_Function/Mean_1_grad/Tile
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape_2
Optimizer/gradients/Loss_Function/Mean_1_grad/Const
Optimizer/gradients/Loss_Function/Mean_1_grad/Prod
Optimizer/gradients/Loss_Function/Mean_1_grad/Const_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Prod_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Maximum/y
Optimizer/gradients/Loss_Function/Mean_1_grad/Maximum
Optimizer/gradients/Loss_Function/Mean_1_grad/floordiv
Optimizer/gradients/Loss_Function/Mean_1_grad/Cast
Optimizer/gradients/Loss_Function/Mean_1_grad/truediv
Optimizer/gradients/Loss_Function/Mul_grad/Shape
Optimizer/gradients/Loss_Function/Mul_grad/Shape_1
Optimizer/gradients/Loss_Function/Mul_grad/BroadcastGradientArgs
Optimizer/gradients/Loss_Function/Mul_grad/mul
Optimizer/gradients/Loss_Function/Mul_grad/Sum
Optimizer/gradients/Loss_Function/Mul_grad/Reshape
Optimizer/gradients/Loss_Function/Mul_grad/mul_1
Optimizer/gradients/Loss_Function/Mul_grad/Sum_1
Optimizer/gradients/Loss_Function/Mul_grad/Reshape_1
Optimizer/gradients/Loss_Function/Mul_grad/tuple/group_deps
Optimizer/gradients/Loss_Function/Mul_grad/tuple/control_dependency
Optimizer/gradients/Loss_Function/Mul_grad/tuple/control_dependency_1
Optimizer/gradients/Cross_Entropy/Reshape_2_grad/Shape
Optimizer/gradients/Cross_Entropy/Reshape_2_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_grad/Reshape/shape
Optimizer/gradients/Loss_Function/Mean_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_grad/Tile/multiples
Optimizer/gradients/Loss_Function/Mean_grad/Tile
Optimizer/gradients/Loss_Function/Mean_grad/Shape
Optimizer/gradients/Loss_Function/Mean_grad/Shape_1
Optimizer/gradients/Loss_Function/Mean_grad/Const
Optimizer/gradients/Loss_Function/Mean_grad/Prod
Optimizer/gradients/Loss_Function/Mean_grad/Const_1
Optimizer/gradients/Loss_Function/Mean_grad/Prod_1
Optimizer/gradients/Loss_Function/Mean_grad/Maximum/y
Optimizer/gradients/Loss_Function/Mean_grad/Maximum
Optimizer/gradients/Loss_Function/Mean_grad/floordiv
Optimizer/gradients/Loss_Function/Mean_grad/Cast
Optimizer/gradients/Loss_Function/Mean_grad/truediv
Optimizer/gradients/zeros_like
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer/gradients/Loss_Function/Square_grad/mul/x
Optimizer/gradients/Loss_Function/Square_grad/mul
Optimizer/gradients/Loss_Function/Square_grad/mul_1
Optimizer/gradients/Cross_Entropy/Reshape_grad/Shape
Optimizer/gradients/Cross_Entropy/Reshape_grad/Reshape
Optimizer/gradients/Predictions/Softmax_grad/mul
Optimizer/gradients/Predictions/Softmax_grad/Sum/reduction_indices
Optimizer/gradients/Predictions/Softmax_grad/Sum
Optimizer/gradients/Predictions/Softmax_grad/Reshape/shape
Optimizer/gradients/Predictions/Softmax_grad/Reshape
Optimizer/gradients/Predictions/Softmax_grad/sub
Optimizer/gradients/Predictions/Softmax_grad/mul_1
Optimizer/gradients/Predictions/mlp_layer_1_grad/ReluGrad
Optimizer/gradients/Predictions/mlp_layer_grad/BiasAddGrad
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/group_deps
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/control_dependency
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/control_dependency_1
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/MatMul
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/MatMul_1
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer/gradients/AddN
Optimizer/GradientDescent/learning_rate
Optimizer/GradientDescent/update_Parameters/Weights/Weight/ApplyGradientDescent
Optimizer/GradientDescent/update_Parameters/Bias/Bias/ApplyGradientDescent
Optimizer/GradientDescent
Equal
Accuracy/Cast
Accuracy/Const
Accuracy/Mean
Accuracy/train_accuracy_list/initial_value
Accuracy/train_accuracy_list
Accuracy/train_accuracy_list/Assign
Accuracy/train_accuracy_list/read
Accuracy/Assign/value
Accuracy/Assign
Accuracy/train_accuracy/tags
Accuracy/train_accuracy
Accuracy/Cast_1
Accuracy/Const_1
Accuracy/Mean_1
Accuracy/validate_accuracy_list/initial_value
Accuracy/validate_accuracy_list
Accuracy/validate_accuracy_list/Assign
Accuracy/validate_accuracy_list/read
Accuracy/Assign_1/value
Accuracy/Assign_1
Accuracy/validate_accuracy/tags
Accuracy/validate_accuracy
Accuracy/Cast_2
Accuracy/Const_2
Accuracy/Mean_2
Accuracy/test_accuracy_list/initial_value
Accuracy/test_accuracy_list
Accuracy/test_accuracy_list/Assign
Accuracy/test_accuracy_list/read
Accuracy/Assign_2/value
Accuracy/Assign_2
Accuracy/test_accuracy/tags
Accuracy/test_accuracy
init/NoOp
init/NoOp_1
init
Merge/MergeSummary
save/Const
save/SaveV2/tensor_names
save/SaveV2/shape_and_slices
save/SaveV2
save/control_dependency
save/RestoreV2/tensor_names
save/RestoreV2/shape_and_slices
save/RestoreV2
save/Assign
save/RestoreV2_1/tensor_names
save/RestoreV2_1/shape_and_slices
save/RestoreV2_1
save/Assign_1
save/RestoreV2_2/tensor_names
save/RestoreV2_2/shape_and_slices
save/RestoreV2_2
save/Assign_2
save/RestoreV2_3/tensor_names
save/RestoreV2_3/shape_and_slices
save/RestoreV2_3
save/Assign_3
save/RestoreV2_4/tensor_names
save/RestoreV2_4/shape_and_slices
save/RestoreV2_4
save/Assign_4
save/RestoreV2_5/tensor_names
save/RestoreV2_5/shape_and_slices
save/RestoreV2_5
save/Assign_5
save/RestoreV2_6/tensor_names
save/RestoreV2_6/shape_and_slices
save/RestoreV2_6
save/Assign_6
save/RestoreV2_7/tensor_names
save/RestoreV2_7/shape_and_slices
save/RestoreV2_7
save/Assign_7
save/RestoreV2_8/tensor_names
save/RestoreV2_8/shape_and_slices
save/RestoreV2_8
save/Assign_8
save/restore_all/NoOp
save/restore_all/NoOp_1
save/restore_all
last_successful_epoch_1/initial_value
last_successful_epoch_1
last_successful_epoch_1/Assign
last_successful_epoch_1/read
add_1/y
add_1
assign_updated_epoch_1
Inputs_1/Data/X_input
Inputs_1/Train_Labels/y_true_one_hot_label
Inputs_1/Train_Labels/y_true_class
Parameters_1/Weights/zeros
Parameters_1/Weights/Weight
Parameters_1/Weights/Weight/Assign
Parameters_1/Weights/Weight/read
Parameters_1/Bias/ones
Parameters_1/Bias/Bias
Parameters_1/Bias/Bias/Assign
Parameters_1/Bias/Bias/read
Predictions_1/mlp_layer/MatMul
Predictions_1/mlp_layer
Predictions_1/mlp_layer_1
Predictions_1/Softmax
Predictions_1/predict_class/dimension
Predictions_1/predict_class
Predictions_1/one_hot/Const
Predictions_1/one_hot/Const_1
Predictions_1/one_hot/depth
Predictions_1/one_hot/on_value
Predictions_1/one_hot/off_value
Predictions_1/one_hot
Cross_Entropy_1/Rank
Cross_Entropy_1/Shape
Cross_Entropy_1/Rank_1
Cross_Entropy_1/Shape_1
Cross_Entropy_1/Sub/y
Cross_Entropy_1/Sub
Cross_Entropy_1/Slice/begin
Cross_Entropy_1/Slice/size
Cross_Entropy_1/Slice
Cross_Entropy_1/concat/values_0
Cross_Entropy_1/concat/axis
Cross_Entropy_1/concat
Cross_Entropy_1/Reshape
Cross_Entropy_1/Rank_2
Cross_Entropy_1/Shape_2
Cross_Entropy_1/Sub_1/y
Cross_Entropy_1/Sub_1
Cross_Entropy_1/Slice_1/begin
Cross_Entropy_1/Slice_1/size
Cross_Entropy_1/Slice_1
Cross_Entropy_1/concat_1/values_0
Cross_Entropy_1/concat_1/axis
Cross_Entropy_1/concat_1
Cross_Entropy_1/Reshape_1
Cross_Entropy_1/SoftmaxCrossEntropyWithLogits
Cross_Entropy_1/Sub_2/y
Cross_Entropy_1/Sub_2
Cross_Entropy_1/Slice_2/begin
Cross_Entropy_1/Slice_2/size
Cross_Entropy_1/Slice_2
Cross_Entropy_1/Reshape_2
Loss_Function_1/Const
Loss_Function_1/Square
Loss_Function_1/Const_1
Loss_Function_1/Mean
Loss_Function_1/Const_2
Loss_Function_1/Mean_1
Loss_Function_1/Mul
Loss_Function_1/Add
Loss_Function_1/train_loss_list/initial_value
Loss_Function_1/train_loss_list
Loss_Function_1/train_loss_list/Assign
Loss_Function_1/train_loss_list/read
Loss_Function_1/Assign/value
Loss_Function_1/Assign
Loss_Function_1/train_error/tags
Loss_Function_1/train_error
Loss_Function_1/validate_loss_list/initial_value
Loss_Function_1/validate_loss_list
Loss_Function_1/validate_loss_list/Assign
Loss_Function_1/validate_loss_list/read
Loss_Function_1/Assign_1/value
Loss_Function_1/Assign_1
Loss_Function_1/validate_error/tags
Loss_Function_1/validate_error
Optimizer_1/learning_rate_progress/initial_value
Optimizer_1/learning_rate_progress
Optimizer_1/learning_rate_progress/Assign
Optimizer_1/learning_rate_progress/read
Optimizer_1/Assign/value
Optimizer_1/Assign
Optimizer_1/learning_rate/tags
Optimizer_1/learning_rate/values
Optimizer_1/learning_rate
Optimizer_1/gradients/Shape
Optimizer_1/gradients/Const
Optimizer_1/gradients/Fill
Optimizer_1/gradients/Loss_Function_1/Add_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Add_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/BroadcastGradientArgs
Optimizer_1/gradients/Loss_Function_1/Add_grad/Sum
Optimizer_1/gradients/Loss_Function_1/Add_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Add_grad/Sum_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/Reshape_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/group_deps
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/control_dependency
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/control_dependency_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Reshape/shape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Tile
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape_2
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Const
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Prod
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Const_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Prod_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Maximum/y
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Maximum
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/floordiv
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Cast
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/truediv
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/BroadcastGradientArgs
Optimizer_1/gradients/Loss_Function_1/Mul_grad/mul
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Sum
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mul_grad/mul_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Sum_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Reshape_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/group_deps
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/control_dependency
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/control_dependency_1
Optimizer_1/gradients/Cross_Entropy_1/Reshape_2_grad/Shape
Optimizer_1/gradients/Cross_Entropy_1/Reshape_2_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Reshape/shape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Tile/multiples
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Tile
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Const
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Prod
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Const_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Prod_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Maximum/y
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Maximum
Optimizer_1/gradients/Loss_Function_1/Mean_grad/floordiv
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Cast
Optimizer_1/gradients/Loss_Function_1/Mean_grad/truediv
Optimizer_1/gradients/zeros_like
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul/x
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul_1
Optimizer_1/gradients/Cross_Entropy_1/Reshape_grad/Shape
Optimizer_1/gradients/Cross_Entropy_1/Reshape_grad/Reshape
Optimizer_1/gradients/Predictions_1/Softmax_grad/mul
Optimizer_1/gradients/Predictions_1/Softmax_grad/Sum/reduction_indices
Optimizer_1/gradients/Predictions_1/Softmax_grad/Sum
Optimizer_1/gradients/Predictions_1/Softmax_grad/Reshape/shape
Optimizer_1/gradients/Predictions_1/Softmax_grad/Reshape
Optimizer_1/gradients/Predictions_1/Softmax_grad/sub
Optimizer_1/gradients/Predictions_1/Softmax_grad/mul_1
Optimizer_1/gradients/Predictions_1/mlp_layer_1_grad/ReluGrad
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/BiasAddGrad
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/group_deps
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/control_dependency
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/control_dependency_1
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/MatMul
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/MatMul_1
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_1/gradients/AddN
Optimizer_1/GradientDescent/learning_rate
Optimizer_1/GradientDescent/update_Parameters_1/Weights/Weight/ApplyGradientDescent
Optimizer_1/GradientDescent/update_Parameters_1/Bias/Bias/ApplyGradientDescent
Optimizer_1/GradientDescent
Equal_1
Accuracy_1/Cast
Accuracy_1/Const
Accuracy_1/Mean
Accuracy_1/train_accuracy_list/initial_value
Accuracy_1/train_accuracy_list
Accuracy_1/train_accuracy_list/Assign
Accuracy_1/train_accuracy_list/read
Accuracy_1/Assign/value
Accuracy_1/Assign
Accuracy_1/train_accuracy/tags
Accuracy_1/train_accuracy
Accuracy_1/Cast_1
Accuracy_1/Const_1
Accuracy_1/Mean_1
Accuracy_1/validate_accuracy_list/initial_value
Accuracy_1/validate_accuracy_list
Accuracy_1/validate_accuracy_list/Assign
Accuracy_1/validate_accuracy_list/read
Accuracy_1/Assign_1/value
Accuracy_1/Assign_1
Accuracy_1/validate_accuracy/tags
Accuracy_1/validate_accuracy
Accuracy_1/Cast_2
Accuracy_1/Const_2
Accuracy_1/Mean_2
Accuracy_1/test_accuracy_list/initial_value
Accuracy_1/test_accuracy_list
Accuracy_1/test_accuracy_list/Assign
Accuracy_1/test_accuracy_list/read
Accuracy_1/Assign_2/value
Accuracy_1/Assign_2
Accuracy_1/test_accuracy/tags
Accuracy_1/test_accuracy
init_1/NoOp
init_1/NoOp_1
init_1
Merge_1/MergeSummary
save_1/Const
save_1/SaveV2/tensor_names
save_1/SaveV2/shape_and_slices
save_1/SaveV2
save_1/control_dependency
save_1/RestoreV2/tensor_names
save_1/RestoreV2/shape_and_slices
save_1/RestoreV2
save_1/Assign
save_1/RestoreV2_1/tensor_names
save_1/RestoreV2_1/shape_and_slices
save_1/RestoreV2_1
save_1/Assign_1
save_1/RestoreV2_2/tensor_names
save_1/RestoreV2_2/shape_and_slices
save_1/RestoreV2_2
save_1/Assign_2
save_1/RestoreV2_3/tensor_names
save_1/RestoreV2_3/shape_and_slices
save_1/RestoreV2_3
save_1/Assign_3
save_1/RestoreV2_4/tensor_names
save_1/RestoreV2_4/shape_and_slices
save_1/RestoreV2_4
save_1/Assign_4
save_1/RestoreV2_5/tensor_names
save_1/RestoreV2_5/shape_and_slices
save_1/RestoreV2_5
save_1/Assign_5
save_1/RestoreV2_6/tensor_names
save_1/RestoreV2_6/shape_and_slices
save_1/RestoreV2_6
save_1/Assign_6
save_1/RestoreV2_7/tensor_names
save_1/RestoreV2_7/shape_and_slices
save_1/RestoreV2_7
save_1/Assign_7
save_1/RestoreV2_8/tensor_names
save_1/RestoreV2_8/shape_and_slices
save_1/RestoreV2_8
save_1/Assign_8
save_1/RestoreV2_9/tensor_names
save_1/RestoreV2_9/shape_and_slices
save_1/RestoreV2_9
save_1/Assign_9
save_1/RestoreV2_10/tensor_names
save_1/RestoreV2_10/shape_and_slices
save_1/RestoreV2_10
save_1/Assign_10
save_1/RestoreV2_11/tensor_names
save_1/RestoreV2_11/shape_and_slices
save_1/RestoreV2_11
save_1/Assign_11
save_1/RestoreV2_12/tensor_names
save_1/RestoreV2_12/shape_and_slices
save_1/RestoreV2_12
save_1/Assign_12
save_1/RestoreV2_13/tensor_names
save_1/RestoreV2_13/shape_and_slices
save_1/RestoreV2_13
save_1/Assign_13
save_1/RestoreV2_14/tensor_names
save_1/RestoreV2_14/shape_and_slices
save_1/RestoreV2_14
save_1/Assign_14
save_1/RestoreV2_15/tensor_names
save_1/RestoreV2_15/shape_and_slices
save_1/RestoreV2_15
save_1/Assign_15
save_1/RestoreV2_16/tensor_names
save_1/RestoreV2_16/shape_and_slices
save_1/RestoreV2_16
save_1/Assign_16
save_1/RestoreV2_17/tensor_names
save_1/RestoreV2_17/shape_and_slices
save_1/RestoreV2_17
save_1/Assign_17
save_1/restore_all/NoOp
save_1/restore_all/NoOp_1
save_1/restore_all
last_successful_epoch_2/initial_value
last_successful_epoch_2
last_successful_epoch_2/Assign
last_successful_epoch_2/read
add_2/y
add_2
assign_updated_epoch_2
Inputs_2/Data/X_input
Inputs_2/Train_Labels/y_true_one_hot_label
Inputs_2/Train_Labels/y_true_class
Parameters_2/Weights/zeros
Parameters_2/Weights/Weight
Parameters_2/Weights/Weight/Assign
Parameters_2/Weights/Weight/read
Parameters_2/Bias/ones
Parameters_2/Bias/Bias
Parameters_2/Bias/Bias/Assign
Parameters_2/Bias/Bias/read
Predictions_2/mlp_layer/MatMul
Predictions_2/mlp_layer
Predictions_2/mlp_layer_1
Predictions_2/Softmax
Predictions_2/predict_class/dimension
Predictions_2/predict_class
Predictions_2/one_hot/Const
Predictions_2/one_hot/Const_1
Predictions_2/one_hot/depth
Predictions_2/one_hot/on_value
Predictions_2/one_hot/off_value
Predictions_2/one_hot
Cross_Entropy_2/Rank
Cross_Entropy_2/Shape
Cross_Entropy_2/Rank_1
Cross_Entropy_2/Shape_1
Cross_Entropy_2/Sub/y
Cross_Entropy_2/Sub
Cross_Entropy_2/Slice/begin
Cross_Entropy_2/Slice/size
Cross_Entropy_2/Slice
Cross_Entropy_2/concat/values_0
Cross_Entropy_2/concat/axis
Cross_Entropy_2/concat
Cross_Entropy_2/Reshape
Cross_Entropy_2/Rank_2
Cross_Entropy_2/Shape_2
Cross_Entropy_2/Sub_1/y
Cross_Entropy_2/Sub_1
Cross_Entropy_2/Slice_1/begin
Cross_Entropy_2/Slice_1/size
Cross_Entropy_2/Slice_1
Cross_Entropy_2/concat_1/values_0
Cross_Entropy_2/concat_1/axis
Cross_Entropy_2/concat_1
Cross_Entropy_2/Reshape_1
Cross_Entropy_2/SoftmaxCrossEntropyWithLogits
Cross_Entropy_2/Sub_2/y
Cross_Entropy_2/Sub_2
Cross_Entropy_2/Slice_2/begin
Cross_Entropy_2/Slice_2/size
Cross_Entropy_2/Slice_2
Cross_Entropy_2/Reshape_2
Loss_Function_2/Const
Loss_Function_2/Square
Loss_Function_2/Const_1
Loss_Function_2/Mean
Loss_Function_2/Const_2
Loss_Function_2/Mean_1
Loss_Function_2/Mul
Loss_Function_2/Add
Loss_Function_2/train_loss_list/initial_value
Loss_Function_2/train_loss_list
Loss_Function_2/train_loss_list/Assign
Loss_Function_2/train_loss_list/read
Loss_Function_2/Assign/value
Loss_Function_2/Assign
Loss_Function_2/train_error/tags
Loss_Function_2/train_error
Loss_Function_2/validate_loss_list/initial_value
Loss_Function_2/validate_loss_list
Loss_Function_2/validate_loss_list/Assign
Loss_Function_2/validate_loss_list/read
Loss_Function_2/Assign_1/value
Loss_Function_2/Assign_1
Loss_Function_2/validate_error/tags
Loss_Function_2/validate_error
Optimizer_2/learning_rate_progress/initial_value
Optimizer_2/learning_rate_progress
Optimizer_2/learning_rate_progress/Assign
Optimizer_2/learning_rate_progress/read
Optimizer_2/Assign/value
Optimizer_2/Assign
Optimizer_2/learning_rate/tags
Optimizer_2/learning_rate/values
Optimizer_2/learning_rate
Optimizer_2/gradients/Shape
Optimizer_2/gradients/Const
Optimizer_2/gradients/Fill
Optimizer_2/gradients/Loss_Function_2/Add_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Add_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/BroadcastGradientArgs
Optimizer_2/gradients/Loss_Function_2/Add_grad/Sum
Optimizer_2/gradients/Loss_Function_2/Add_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Add_grad/Sum_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/Reshape_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/group_deps
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/control_dependency
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/control_dependency_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Reshape/shape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Tile
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape_2
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Const
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Prod
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Const_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Prod_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Maximum/y
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Maximum
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/floordiv
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Cast
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/truediv
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/BroadcastGradientArgs
Optimizer_2/gradients/Loss_Function_2/Mul_grad/mul
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Sum
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mul_grad/mul_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Sum_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Reshape_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/group_deps
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/control_dependency
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/control_dependency_1
Optimizer_2/gradients/Cross_Entropy_2/Reshape_2_grad/Shape
Optimizer_2/gradients/Cross_Entropy_2/Reshape_2_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Reshape/shape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Tile/multiples
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Tile
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Const
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Prod
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Const_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Prod_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Maximum/y
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Maximum
Optimizer_2/gradients/Loss_Function_2/Mean_grad/floordiv
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Cast
Optimizer_2/gradients/Loss_Function_2/Mean_grad/truediv
Optimizer_2/gradients/zeros_like
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul/x
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul_1
Optimizer_2/gradients/Cross_Entropy_2/Reshape_grad/Shape
Optimizer_2/gradients/Cross_Entropy_2/Reshape_grad/Reshape
Optimizer_2/gradients/Predictions_2/Softmax_grad/mul
Optimizer_2/gradients/Predictions_2/Softmax_grad/Sum/reduction_indices
Optimizer_2/gradients/Predictions_2/Softmax_grad/Sum
Optimizer_2/gradients/Predictions_2/Softmax_grad/Reshape/shape
Optimizer_2/gradients/Predictions_2/Softmax_grad/Reshape
Optimizer_2/gradients/Predictions_2/Softmax_grad/sub
Optimizer_2/gradients/Predictions_2/Softmax_grad/mul_1
Optimizer_2/gradients/Predictions_2/mlp_layer_1_grad/ReluGrad
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/BiasAddGrad
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/group_deps
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/control_dependency
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/control_dependency_1
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/MatMul
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/MatMul_1
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_2/gradients/AddN
Optimizer_2/GradientDescent/learning_rate
Optimizer_2/GradientDescent/update_Parameters_2/Weights/Weight/ApplyGradientDescent
Optimizer_2/GradientDescent/update_Parameters_2/Bias/Bias/ApplyGradientDescent
Optimizer_2/GradientDescent
Equal_2
Accuracy_2/Cast
Accuracy_2/Const
Accuracy_2/Mean
Accuracy_2/train_accuracy_list/initial_value
Accuracy_2/train_accuracy_list
Accuracy_2/train_accuracy_list/Assign
Accuracy_2/train_accuracy_list/read
Accuracy_2/Assign/value
Accuracy_2/Assign
Accuracy_2/train_accuracy/tags
Accuracy_2/train_accuracy
Accuracy_2/Cast_1
Accuracy_2/Const_1
Accuracy_2/Mean_1
Accuracy_2/validate_accuracy_list/initial_value
Accuracy_2/validate_accuracy_list
Accuracy_2/validate_accuracy_list/Assign
Accuracy_2/validate_accuracy_list/read
Accuracy_2/Assign_1/value
Accuracy_2/Assign_1
Accuracy_2/validate_accuracy/tags
Accuracy_2/validate_accuracy
Accuracy_2/Cast_2
Accuracy_2/Const_2
Accuracy_2/Mean_2
Accuracy_2/test_accuracy_list/initial_value
Accuracy_2/test_accuracy_list
Accuracy_2/test_accuracy_list/Assign
Accuracy_2/test_accuracy_list/read
Accuracy_2/Assign_2/value
Accuracy_2/Assign_2
Accuracy_2/test_accuracy/tags
Accuracy_2/test_accuracy
init_2/NoOp
init_2/NoOp_1
init_2
Merge_2/MergeSummary
save_2/Const
save_2/SaveV2/tensor_names
save_2/SaveV2/shape_and_slices
save_2/SaveV2
save_2/control_dependency
save_2/RestoreV2/tensor_names
save_2/RestoreV2/shape_and_slices
save_2/RestoreV2
save_2/Assign
save_2/RestoreV2_1/tensor_names
save_2/RestoreV2_1/shape_and_slices
save_2/RestoreV2_1
save_2/Assign_1
save_2/RestoreV2_2/tensor_names
save_2/RestoreV2_2/shape_and_slices
save_2/RestoreV2_2
save_2/Assign_2
save_2/RestoreV2_3/tensor_names
save_2/RestoreV2_3/shape_and_slices
save_2/RestoreV2_3
save_2/Assign_3
save_2/RestoreV2_4/tensor_names
save_2/RestoreV2_4/shape_and_slices
save_2/RestoreV2_4
save_2/Assign_4
save_2/RestoreV2_5/tensor_names
save_2/RestoreV2_5/shape_and_slices
save_2/RestoreV2_5
save_2/Assign_5
save_2/RestoreV2_6/tensor_names
save_2/RestoreV2_6/shape_and_slices
save_2/RestoreV2_6
save_2/Assign_6
save_2/RestoreV2_7/tensor_names
save_2/RestoreV2_7/shape_and_slices
save_2/RestoreV2_7
save_2/Assign_7
save_2/RestoreV2_8/tensor_names
save_2/RestoreV2_8/shape_and_slices
save_2/RestoreV2_8
save_2/Assign_8
save_2/RestoreV2_9/tensor_names
save_2/RestoreV2_9/shape_and_slices
save_2/RestoreV2_9
save_2/Assign_9
save_2/RestoreV2_10/tensor_names
save_2/RestoreV2_10/shape_and_slices
save_2/RestoreV2_10
save_2/Assign_10
save_2/RestoreV2_11/tensor_names
save_2/RestoreV2_11/shape_and_slices
save_2/RestoreV2_11
save_2/Assign_11
save_2/RestoreV2_12/tensor_names
save_2/RestoreV2_12/shape_and_slices
save_2/RestoreV2_12
save_2/Assign_12
save_2/RestoreV2_13/tensor_names
save_2/RestoreV2_13/shape_and_slices
save_2/RestoreV2_13
save_2/Assign_13
save_2/RestoreV2_14/tensor_names
save_2/RestoreV2_14/shape_and_slices
save_2/RestoreV2_14
save_2/Assign_14
save_2/RestoreV2_15/tensor_names
save_2/RestoreV2_15/shape_and_slices
save_2/RestoreV2_15
save_2/Assign_15
save_2/RestoreV2_16/tensor_names
save_2/RestoreV2_16/shape_and_slices
save_2/RestoreV2_16
save_2/Assign_16
save_2/RestoreV2_17/tensor_names
save_2/RestoreV2_17/shape_and_slices
save_2/RestoreV2_17
save_2/Assign_17
save_2/RestoreV2_18/tensor_names
save_2/RestoreV2_18/shape_and_slices
save_2/RestoreV2_18
save_2/Assign_18
save_2/RestoreV2_19/tensor_names
save_2/RestoreV2_19/shape_and_slices
save_2/RestoreV2_19
save_2/Assign_19
save_2/RestoreV2_20/tensor_names
save_2/RestoreV2_20/shape_and_slices
save_2/RestoreV2_20
save_2/Assign_20
save_2/RestoreV2_21/tensor_names
save_2/RestoreV2_21/shape_and_slices
save_2/RestoreV2_21
save_2/Assign_21
save_2/RestoreV2_22/tensor_names
save_2/RestoreV2_22/shape_and_slices
save_2/RestoreV2_22
save_2/Assign_22
save_2/RestoreV2_23/tensor_names
save_2/RestoreV2_23/shape_and_slices
save_2/RestoreV2_23
save_2/Assign_23
save_2/RestoreV2_24/tensor_names
save_2/RestoreV2_24/shape_and_slices
save_2/RestoreV2_24
save_2/Assign_24
save_2/RestoreV2_25/tensor_names
save_2/RestoreV2_25/shape_and_slices
save_2/RestoreV2_25
save_2/Assign_25
save_2/RestoreV2_26/tensor_names
save_2/RestoreV2_26/shape_and_slices
save_2/RestoreV2_26
save_2/Assign_26
save_2/restore_all/NoOp
save_2/restore_all/NoOp_1
save_2/restore_all
last_successful_epoch_3/initial_value
last_successful_epoch_3
last_successful_epoch_3/Assign
last_successful_epoch_3/read
add_3/y
add_3
assign_updated_epoch_3
Inputs_3/Data/X_input
Inputs_3/Train_Labels/y_true_one_hot_label
Inputs_3/Train_Labels/y_true_class
Parameters_3/Weights/zeros
Parameters_3/Weights/Weight
Parameters_3/Weights/Weight/Assign
Parameters_3/Weights/Weight/read
Parameters_3/Bias/ones
Parameters_3/Bias/Bias
Parameters_3/Bias/Bias/Assign
Parameters_3/Bias/Bias/read
Predictions_3/mlp_layer/MatMul
Predictions_3/mlp_layer
Predictions_3/mlp_layer_1
Predictions_3/Softmax
Predictions_3/predict_class/dimension
Predictions_3/predict_class
Predictions_3/one_hot/Const
Predictions_3/one_hot/Const_1
Predictions_3/one_hot/depth
Predictions_3/one_hot/on_value
Predictions_3/one_hot/off_value
Predictions_3/one_hot
Cross_Entropy_3/Rank
Cross_Entropy_3/Shape
Cross_Entropy_3/Rank_1
Cross_Entropy_3/Shape_1
Cross_Entropy_3/Sub/y
Cross_Entropy_3/Sub
Cross_Entropy_3/Slice/begin
Cross_Entropy_3/Slice/size
Cross_Entropy_3/Slice
Cross_Entropy_3/concat/values_0
Cross_Entropy_3/concat/axis
Cross_Entropy_3/concat
Cross_Entropy_3/Reshape
Cross_Entropy_3/Rank_2
Cross_Entropy_3/Shape_2
Cross_Entropy_3/Sub_1/y
Cross_Entropy_3/Sub_1
Cross_Entropy_3/Slice_1/begin
Cross_Entropy_3/Slice_1/size
Cross_Entropy_3/Slice_1
Cross_Entropy_3/concat_1/values_0
Cross_Entropy_3/concat_1/axis
Cross_Entropy_3/concat_1
Cross_Entropy_3/Reshape_1
Cross_Entropy_3/SoftmaxCrossEntropyWithLogits
Cross_Entropy_3/Sub_2/y
Cross_Entropy_3/Sub_2
Cross_Entropy_3/Slice_2/begin
Cross_Entropy_3/Slice_2/size
Cross_Entropy_3/Slice_2
Cross_Entropy_3/Reshape_2
Loss_Function_3/Const
Loss_Function_3/Square
Loss_Function_3/Const_1
Loss_Function_3/Mean
Loss_Function_3/Const_2
Loss_Function_3/Mean_1
Loss_Function_3/Mul
Loss_Function_3/Add
Loss_Function_3/train_loss_list/initial_value
Loss_Function_3/train_loss_list
Loss_Function_3/train_loss_list/Assign
Loss_Function_3/train_loss_list/read
Loss_Function_3/Assign/value
Loss_Function_3/Assign
Loss_Function_3/train_error/tags
Loss_Function_3/train_error
Loss_Function_3/validate_loss_list/initial_value
Loss_Function_3/validate_loss_list
Loss_Function_3/validate_loss_list/Assign
Loss_Function_3/validate_loss_list/read
Loss_Function_3/Assign_1/value
Loss_Function_3/Assign_1
Loss_Function_3/validate_error/tags
Loss_Function_3/validate_error
Optimizer_3/learning_rate_progress/initial_value
Optimizer_3/learning_rate_progress
Optimizer_3/learning_rate_progress/Assign
Optimizer_3/learning_rate_progress/read
Optimizer_3/Assign/value
Optimizer_3/Assign
Optimizer_3/learning_rate/tags
Optimizer_3/learning_rate/values
Optimizer_3/learning_rate
Optimizer_3/gradients/Shape
Optimizer_3/gradients/Const
Optimizer_3/gradients/Fill
Optimizer_3/gradients/Loss_Function_3/Add_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Add_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/BroadcastGradientArgs
Optimizer_3/gradients/Loss_Function_3/Add_grad/Sum
Optimizer_3/gradients/Loss_Function_3/Add_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Add_grad/Sum_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/Reshape_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/group_deps
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/control_dependency
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/control_dependency_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Reshape/shape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Tile
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape_2
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Const
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Prod
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Const_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Prod_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Maximum/y
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Maximum
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/floordiv
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Cast
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/truediv
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/BroadcastGradientArgs
Optimizer_3/gradients/Loss_Function_3/Mul_grad/mul
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Sum
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mul_grad/mul_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Sum_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Reshape_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/group_deps
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/control_dependency
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/control_dependency_1
Optimizer_3/gradients/Cross_Entropy_3/Reshape_2_grad/Shape
Optimizer_3/gradients/Cross_Entropy_3/Reshape_2_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Reshape/shape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Tile/multiples
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Tile
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Const
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Prod
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Const_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Prod_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Maximum/y
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Maximum
Optimizer_3/gradients/Loss_Function_3/Mean_grad/floordiv
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Cast
Optimizer_3/gradients/Loss_Function_3/Mean_grad/truediv
Optimizer_3/gradients/zeros_like
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul/x
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul_1
Optimizer_3/gradients/Cross_Entropy_3/Reshape_grad/Shape
Optimizer_3/gradients/Cross_Entropy_3/Reshape_grad/Reshape
Optimizer_3/gradients/Predictions_3/Softmax_grad/mul
Optimizer_3/gradients/Predictions_3/Softmax_grad/Sum/reduction_indices
Optimizer_3/gradients/Predictions_3/Softmax_grad/Sum
Optimizer_3/gradients/Predictions_3/Softmax_grad/Reshape/shape
Optimizer_3/gradients/Predictions_3/Softmax_grad/Reshape
Optimizer_3/gradients/Predictions_3/Softmax_grad/sub
Optimizer_3/gradients/Predictions_3/Softmax_grad/mul_1
Optimizer_3/gradients/Predictions_3/mlp_layer_1_grad/ReluGrad
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/BiasAddGrad
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/group_deps
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/control_dependency
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/control_dependency_1
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/MatMul
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/MatMul_1
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_3/gradients/AddN
Optimizer_3/GradientDescent/learning_rate
Optimizer_3/GradientDescent/update_Parameters_3/Weights/Weight/ApplyGradientDescent
Optimizer_3/GradientDescent/update_Parameters_3/Bias/Bias/ApplyGradientDescent
Optimizer_3/GradientDescent
Equal_3
Accuracy_3/Cast
Accuracy_3/Const
Accuracy_3/Mean
Accuracy_3/train_accuracy_list/initial_value
Accuracy_3/train_accuracy_list
Accuracy_3/train_accuracy_list/Assign
Accuracy_3/train_accuracy_list/read
Accuracy_3/Assign/value
Accuracy_3/Assign
Accuracy_3/train_accuracy/tags
Accuracy_3/train_accuracy
Accuracy_3/Cast_1
Accuracy_3/Const_1
Accuracy_3/Mean_1
Accuracy_3/validate_accuracy_list/initial_value
Accuracy_3/validate_accuracy_list
Accuracy_3/validate_accuracy_list/Assign
Accuracy_3/validate_accuracy_list/read
Accuracy_3/Assign_1/value
Accuracy_3/Assign_1
Accuracy_3/validate_accuracy/tags
Accuracy_3/validate_accuracy
Accuracy_3/Cast_2
Accuracy_3/Const_2
Accuracy_3/Mean_2
Accuracy_3/test_accuracy_list/initial_value
Accuracy_3/test_accuracy_list
Accuracy_3/test_accuracy_list/Assign
Accuracy_3/test_accuracy_list/read
Accuracy_3/Assign_2/value
Accuracy_3/Assign_2
Accuracy_3/test_accuracy/tags
Accuracy_3/test_accuracy
init_3/NoOp
init_3/NoOp_1
init_3
Merge_3/MergeSummary
save_3/Const
save_3/SaveV2/tensor_names
save_3/SaveV2/shape_and_slices
save_3/SaveV2
save_3/control_dependency
save_3/RestoreV2/tensor_names
save_3/RestoreV2/shape_and_slices
save_3/RestoreV2
save_3/Assign
save_3/RestoreV2_1/tensor_names
save_3/RestoreV2_1/shape_and_slices
save_3/RestoreV2_1
save_3/Assign_1
save_3/RestoreV2_2/tensor_names
save_3/RestoreV2_2/shape_and_slices
save_3/RestoreV2_2
save_3/Assign_2
save_3/RestoreV2_3/tensor_names
save_3/RestoreV2_3/shape_and_slices
save_3/RestoreV2_3
save_3/Assign_3
save_3/RestoreV2_4/tensor_names
save_3/RestoreV2_4/shape_and_slices
save_3/RestoreV2_4
save_3/Assign_4
save_3/RestoreV2_5/tensor_names
save_3/RestoreV2_5/shape_and_slices
save_3/RestoreV2_5
save_3/Assign_5
save_3/RestoreV2_6/tensor_names
save_3/RestoreV2_6/shape_and_slices
save_3/RestoreV2_6
save_3/Assign_6
save_3/RestoreV2_7/tensor_names
save_3/RestoreV2_7/shape_and_slices
save_3/RestoreV2_7
save_3/Assign_7
save_3/RestoreV2_8/tensor_names
save_3/RestoreV2_8/shape_and_slices
save_3/RestoreV2_8
save_3/Assign_8
save_3/RestoreV2_9/tensor_names
save_3/RestoreV2_9/shape_and_slices
save_3/RestoreV2_9
save_3/Assign_9
save_3/RestoreV2_10/tensor_names
save_3/RestoreV2_10/shape_and_slices
save_3/RestoreV2_10
save_3/Assign_10
save_3/RestoreV2_11/tensor_names
save_3/RestoreV2_11/shape_and_slices
save_3/RestoreV2_11
save_3/Assign_11
save_3/RestoreV2_12/tensor_names
save_3/RestoreV2_12/shape_and_slices
save_3/RestoreV2_12
save_3/Assign_12
save_3/RestoreV2_13/tensor_names
save_3/RestoreV2_13/shape_and_slices
save_3/RestoreV2_13
save_3/Assign_13
save_3/RestoreV2_14/tensor_names
save_3/RestoreV2_14/shape_and_slices
save_3/RestoreV2_14
save_3/Assign_14
save_3/RestoreV2_15/tensor_names
save_3/RestoreV2_15/shape_and_slices
save_3/RestoreV2_15
save_3/Assign_15
save_3/RestoreV2_16/tensor_names
save_3/RestoreV2_16/shape_and_slices
save_3/RestoreV2_16
save_3/Assign_16
save_3/RestoreV2_17/tensor_names
save_3/RestoreV2_17/shape_and_slices
save_3/RestoreV2_17
save_3/Assign_17
save_3/RestoreV2_18/tensor_names
save_3/RestoreV2_18/shape_and_slices
save_3/RestoreV2_18
save_3/Assign_18
save_3/RestoreV2_19/tensor_names
save_3/RestoreV2_19/shape_and_slices
save_3/RestoreV2_19
save_3/Assign_19
save_3/RestoreV2_20/tensor_names
save_3/RestoreV2_20/shape_and_slices
save_3/RestoreV2_20
save_3/Assign_20
save_3/RestoreV2_21/tensor_names
save_3/RestoreV2_21/shape_and_slices
save_3/RestoreV2_21
save_3/Assign_21
save_3/RestoreV2_22/tensor_names
save_3/RestoreV2_22/shape_and_slices
save_3/RestoreV2_22
save_3/Assign_22
save_3/RestoreV2_23/tensor_names
save_3/RestoreV2_23/shape_and_slices
save_3/RestoreV2_23
save_3/Assign_23
save_3/RestoreV2_24/tensor_names
save_3/RestoreV2_24/shape_and_slices
save_3/RestoreV2_24
save_3/Assign_24
save_3/RestoreV2_25/tensor_names
save_3/RestoreV2_25/shape_and_slices
save_3/RestoreV2_25
save_3/Assign_25
save_3/RestoreV2_26/tensor_names
save_3/RestoreV2_26/shape_and_slices
save_3/RestoreV2_26
save_3/Assign_26
save_3/RestoreV2_27/tensor_names
save_3/RestoreV2_27/shape_and_slices
save_3/RestoreV2_27
save_3/Assign_27
save_3/RestoreV2_28/tensor_names
save_3/RestoreV2_28/shape_and_slices
save_3/RestoreV2_28
save_3/Assign_28
save_3/RestoreV2_29/tensor_names
save_3/RestoreV2_29/shape_and_slices
save_3/RestoreV2_29
save_3/Assign_29
save_3/RestoreV2_30/tensor_names
save_3/RestoreV2_30/shape_and_slices
save_3/RestoreV2_30
save_3/Assign_30
save_3/RestoreV2_31/tensor_names
save_3/RestoreV2_31/shape_and_slices
save_3/RestoreV2_31
save_3/Assign_31
save_3/RestoreV2_32/tensor_names
save_3/RestoreV2_32/shape_and_slices
save_3/RestoreV2_32
save_3/Assign_32
save_3/RestoreV2_33/tensor_names
save_3/RestoreV2_33/shape_and_slices
save_3/RestoreV2_33
save_3/Assign_33
save_3/RestoreV2_34/tensor_names
save_3/RestoreV2_34/shape_and_slices
save_3/RestoreV2_34
save_3/Assign_34
save_3/RestoreV2_35/tensor_names
save_3/RestoreV2_35/shape_and_slices
save_3/RestoreV2_35
save_3/Assign_35
save_3/restore_all/NoOp
save_3/restore_all/NoOp_1
save_3/restore_all
last_successful_epoch_4/initial_value
last_successful_epoch_4
last_successful_epoch_4/Assign
last_successful_epoch_4/read
add_4/y
add_4
assign_updated_epoch_4
Inputs_4/Data/X_input
Inputs_4/Train_Labels/y_true_one_hot_label
Inputs_4/Train_Labels/y_true_class
Parameters_4/Weights/zeros
Parameters_4/Weights/Weight
Parameters_4/Weights/Weight/Assign
Parameters_4/Weights/Weight/read
Parameters_4/Bias/ones
Parameters_4/Bias/Bias
Parameters_4/Bias/Bias/Assign
Parameters_4/Bias/Bias/read
Predictions_4/mlp_layer/MatMul
Predictions_4/mlp_layer
Predictions_4/mlp_layer_1
Predictions_4/Softmax
Predictions_4/predict_class/dimension
Predictions_4/predict_class
Predictions_4/one_hot/Const
Predictions_4/one_hot/Const_1
Predictions_4/one_hot/depth
Predictions_4/one_hot/on_value
Predictions_4/one_hot/off_value
Predictions_4/one_hot
Cross_Entropy_4/Rank
Cross_Entropy_4/Shape
Cross_Entropy_4/Rank_1
Cross_Entropy_4/Shape_1
Cross_Entropy_4/Sub/y
Cross_Entropy_4/Sub
Cross_Entropy_4/Slice/begin
Cross_Entropy_4/Slice/size
Cross_Entropy_4/Slice
Cross_Entropy_4/concat/values_0
Cross_Entropy_4/concat/axis
Cross_Entropy_4/concat
Cross_Entropy_4/Reshape
Cross_Entropy_4/Rank_2
Cross_Entropy_4/Shape_2
Cross_Entropy_4/Sub_1/y
Cross_Entropy_4/Sub_1
Cross_Entropy_4/Slice_1/begin
Cross_Entropy_4/Slice_1/size
Cross_Entropy_4/Slice_1
Cross_Entropy_4/concat_1/values_0
Cross_Entropy_4/concat_1/axis
Cross_Entropy_4/concat_1
Cross_Entropy_4/Reshape_1
Cross_Entropy_4/SoftmaxCrossEntropyWithLogits
Cross_Entropy_4/Sub_2/y
Cross_Entropy_4/Sub_2
Cross_Entropy_4/Slice_2/begin
Cross_Entropy_4/Slice_2/size
Cross_Entropy_4/Slice_2
Cross_Entropy_4/Reshape_2
Loss_Function_4/Const
Loss_Function_4/Square
Loss_Function_4/Const_1
Loss_Function_4/Mean
Loss_Function_4/Const_2
Loss_Function_4/Mean_1
Loss_Function_4/Mul
Loss_Function_4/Add
Loss_Function_4/train_loss_list/initial_value
Loss_Function_4/train_loss_list
Loss_Function_4/train_loss_list/Assign
Loss_Function_4/train_loss_list/read
Loss_Function_4/Assign/value
Loss_Function_4/Assign
Loss_Function_4/train_error/tags
Loss_Function_4/train_error
Loss_Function_4/validate_loss_list/initial_value
Loss_Function_4/validate_loss_list
Loss_Function_4/validate_loss_list/Assign
Loss_Function_4/validate_loss_list/read
Loss_Function_4/Assign_1/value
Loss_Function_4/Assign_1
Loss_Function_4/validate_error/tags
Loss_Function_4/validate_error
Optimizer_4/learning_rate_progress/initial_value
Optimizer_4/learning_rate_progress
Optimizer_4/learning_rate_progress/Assign
Optimizer_4/learning_rate_progress/read
Optimizer_4/Assign/value
Optimizer_4/Assign
Optimizer_4/learning_rate/tags
Optimizer_4/learning_rate/values
Optimizer_4/learning_rate
Optimizer_4/gradients/Shape
Optimizer_4/gradients/Const
Optimizer_4/gradients/Fill
Optimizer_4/gradients/Loss_Function_4/Add_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Add_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/BroadcastGradientArgs
Optimizer_4/gradients/Loss_Function_4/Add_grad/Sum
Optimizer_4/gradients/Loss_Function_4/Add_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Add_grad/Sum_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/Reshape_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/group_deps
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/control_dependency
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/control_dependency_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Reshape/shape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Tile
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape_2
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Const
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Prod
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Const_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Prod_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Maximum/y
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Maximum
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/floordiv
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Cast
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/truediv
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/BroadcastGradientArgs
Optimizer_4/gradients/Loss_Function_4/Mul_grad/mul
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Sum
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mul_grad/mul_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Sum_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Reshape_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/group_deps
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/control_dependency
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/control_dependency_1
Optimizer_4/gradients/Cross_Entropy_4/Reshape_2_grad/Shape
Optimizer_4/gradients/Cross_Entropy_4/Reshape_2_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Reshape/shape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Tile/multiples
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Tile
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Const
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Prod
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Const_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Prod_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Maximum/y
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Maximum
Optimizer_4/gradients/Loss_Function_4/Mean_grad/floordiv
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Cast
Optimizer_4/gradients/Loss_Function_4/Mean_grad/truediv
Optimizer_4/gradients/zeros_like
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul/x
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul_1
Optimizer_4/gradients/Cross_Entropy_4/Reshape_grad/Shape
Optimizer_4/gradients/Cross_Entropy_4/Reshape_grad/Reshape
Optimizer_4/gradients/Predictions_4/Softmax_grad/mul
Optimizer_4/gradients/Predictions_4/Softmax_grad/Sum/reduction_indices
Optimizer_4/gradients/Predictions_4/Softmax_grad/Sum
Optimizer_4/gradients/Predictions_4/Softmax_grad/Reshape/shape
Optimizer_4/gradients/Predictions_4/Softmax_grad/Reshape
Optimizer_4/gradients/Predictions_4/Softmax_grad/sub
Optimizer_4/gradients/Predictions_4/Softmax_grad/mul_1
Optimizer_4/gradients/Predictions_4/mlp_layer_1_grad/ReluGrad
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/BiasAddGrad
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/group_deps
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/control_dependency
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/control_dependency_1
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/MatMul
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/MatMul_1
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_4/gradients/AddN
Optimizer_4/GradientDescent/learning_rate
Optimizer_4/GradientDescent/update_Parameters_4/Weights/Weight/ApplyGradientDescent
Optimizer_4/GradientDescent/update_Parameters_4/Bias/Bias/ApplyGradientDescent
Optimizer_4/GradientDescent
Equal_4
Accuracy_4/Cast
Accuracy_4/Const
Accuracy_4/Mean
Accuracy_4/train_accuracy_list/initial_value
Accuracy_4/train_accuracy_list
Accuracy_4/train_accuracy_list/Assign
Accuracy_4/train_accuracy_list/read
Accuracy_4/Assign/value
Accuracy_4/Assign
Accuracy_4/train_accuracy/tags
Accuracy_4/train_accuracy
Accuracy_4/Cast_1
Accuracy_4/Const_1
Accuracy_4/Mean_1
Accuracy_4/validate_accuracy_list/initial_value
Accuracy_4/validate_accuracy_list
Accuracy_4/validate_accuracy_list/Assign
Accuracy_4/validate_accuracy_list/read
Accuracy_4/Assign_1/value
Accuracy_4/Assign_1
Accuracy_4/validate_accuracy/tags
Accuracy_4/validate_accuracy
Accuracy_4/Cast_2
Accuracy_4/Const_2
Accuracy_4/Mean_2
Accuracy_4/test_accuracy_list/initial_value
Accuracy_4/test_accuracy_list
Accuracy_4/test_accuracy_list/Assign
Accuracy_4/test_accuracy_list/read
Accuracy_4/Assign_2/value
Accuracy_4/Assign_2
Accuracy_4/test_accuracy/tags
Accuracy_4/test_accuracy
init_4/NoOp
init_4/NoOp_1
init_4
Merge_4/MergeSummary
save_4/Const
save_4/SaveV2/tensor_names
save_4/SaveV2/shape_and_slices
save_4/SaveV2
save_4/control_dependency
save_4/RestoreV2/tensor_names
save_4/RestoreV2/shape_and_slices
save_4/RestoreV2
save_4/Assign
save_4/RestoreV2_1/tensor_names
save_4/RestoreV2_1/shape_and_slices
save_4/RestoreV2_1
save_4/Assign_1
save_4/RestoreV2_2/tensor_names
save_4/RestoreV2_2/shape_and_slices
save_4/RestoreV2_2
save_4/Assign_2
save_4/RestoreV2_3/tensor_names
save_4/RestoreV2_3/shape_and_slices
save_4/RestoreV2_3
save_4/Assign_3
save_4/RestoreV2_4/tensor_names
save_4/RestoreV2_4/shape_and_slices
save_4/RestoreV2_4
save_4/Assign_4
save_4/RestoreV2_5/tensor_names
save_4/RestoreV2_5/shape_and_slices
save_4/RestoreV2_5
save_4/Assign_5
save_4/RestoreV2_6/tensor_names
save_4/RestoreV2_6/shape_and_slices
save_4/RestoreV2_6
save_4/Assign_6
save_4/RestoreV2_7/tensor_names
save_4/RestoreV2_7/shape_and_slices
save_4/RestoreV2_7
save_4/Assign_7
save_4/RestoreV2_8/tensor_names
save_4/RestoreV2_8/shape_and_slices
save_4/RestoreV2_8
save_4/Assign_8
save_4/RestoreV2_9/tensor_names
save_4/RestoreV2_9/shape_and_slices
save_4/RestoreV2_9
save_4/Assign_9
save_4/RestoreV2_10/tensor_names
save_4/RestoreV2_10/shape_and_slices
save_4/RestoreV2_10
save_4/Assign_10
save_4/RestoreV2_11/tensor_names
save_4/RestoreV2_11/shape_and_slices
save_4/RestoreV2_11
save_4/Assign_11
save_4/RestoreV2_12/tensor_names
save_4/RestoreV2_12/shape_and_slices
save_4/RestoreV2_12
save_4/Assign_12
save_4/RestoreV2_13/tensor_names
save_4/RestoreV2_13/shape_and_slices
save_4/RestoreV2_13
save_4/Assign_13
save_4/RestoreV2_14/tensor_names
save_4/RestoreV2_14/shape_and_slices
save_4/RestoreV2_14
save_4/Assign_14
save_4/RestoreV2_15/tensor_names
save_4/RestoreV2_15/shape_and_slices
save_4/RestoreV2_15
save_4/Assign_15
save_4/RestoreV2_16/tensor_names
save_4/RestoreV2_16/shape_and_slices
save_4/RestoreV2_16
save_4/Assign_16
save_4/RestoreV2_17/tensor_names
save_4/RestoreV2_17/shape_and_slices
save_4/RestoreV2_17
save_4/Assign_17
save_4/RestoreV2_18/tensor_names
save_4/RestoreV2_18/shape_and_slices
save_4/RestoreV2_18
save_4/Assign_18
save_4/RestoreV2_19/tensor_names
save_4/RestoreV2_19/shape_and_slices
save_4/RestoreV2_19
save_4/Assign_19
save_4/RestoreV2_20/tensor_names
save_4/RestoreV2_20/shape_and_slices
save_4/RestoreV2_20
save_4/Assign_20
save_4/RestoreV2_21/tensor_names
save_4/RestoreV2_21/shape_and_slices
save_4/RestoreV2_21
save_4/Assign_21
save_4/RestoreV2_22/tensor_names
save_4/RestoreV2_22/shape_and_slices
save_4/RestoreV2_22
save_4/Assign_22
save_4/RestoreV2_23/tensor_names
save_4/RestoreV2_23/shape_and_slices
save_4/RestoreV2_23
save_4/Assign_23
save_4/RestoreV2_24/tensor_names
save_4/RestoreV2_24/shape_and_slices
save_4/RestoreV2_24
save_4/Assign_24
save_4/RestoreV2_25/tensor_names
save_4/RestoreV2_25/shape_and_slices
save_4/RestoreV2_25
save_4/Assign_25
save_4/RestoreV2_26/tensor_names
save_4/RestoreV2_26/shape_and_slices
save_4/RestoreV2_26
save_4/Assign_26
save_4/RestoreV2_27/tensor_names
save_4/RestoreV2_27/shape_and_slices
save_4/RestoreV2_27
save_4/Assign_27
save_4/RestoreV2_28/tensor_names
save_4/RestoreV2_28/shape_and_slices
save_4/RestoreV2_28
save_4/Assign_28
save_4/RestoreV2_29/tensor_names
save_4/RestoreV2_29/shape_and_slices
save_4/RestoreV2_29
save_4/Assign_29
save_4/RestoreV2_30/tensor_names
save_4/RestoreV2_30/shape_and_slices
save_4/RestoreV2_30
save_4/Assign_30
save_4/RestoreV2_31/tensor_names
save_4/RestoreV2_31/shape_and_slices
save_4/RestoreV2_31
save_4/Assign_31
save_4/RestoreV2_32/tensor_names
save_4/RestoreV2_32/shape_and_slices
save_4/RestoreV2_32
save_4/Assign_32
save_4/RestoreV2_33/tensor_names
save_4/RestoreV2_33/shape_and_slices
save_4/RestoreV2_33
save_4/Assign_33
save_4/RestoreV2_34/tensor_names
save_4/RestoreV2_34/shape_and_slices
save_4/RestoreV2_34
save_4/Assign_34
save_4/RestoreV2_35/tensor_names
save_4/RestoreV2_35/shape_and_slices
save_4/RestoreV2_35
save_4/Assign_35
save_4/RestoreV2_36/tensor_names
save_4/RestoreV2_36/shape_and_slices
save_4/RestoreV2_36
save_4/Assign_36
save_4/RestoreV2_37/tensor_names
save_4/RestoreV2_37/shape_and_slices
save_4/RestoreV2_37
save_4/Assign_37
save_4/RestoreV2_38/tensor_names
save_4/RestoreV2_38/shape_and_slices
save_4/RestoreV2_38
save_4/Assign_38
save_4/RestoreV2_39/tensor_names
save_4/RestoreV2_39/shape_and_slices
save_4/RestoreV2_39
save_4/Assign_39
save_4/RestoreV2_40/tensor_names
save_4/RestoreV2_40/shape_and_slices
save_4/RestoreV2_40
save_4/Assign_40
save_4/RestoreV2_41/tensor_names
save_4/RestoreV2_41/shape_and_slices
save_4/RestoreV2_41
save_4/Assign_41
save_4/RestoreV2_42/tensor_names
save_4/RestoreV2_42/shape_and_slices
save_4/RestoreV2_42
save_4/Assign_42
save_4/RestoreV2_43/tensor_names
save_4/RestoreV2_43/shape_and_slices
save_4/RestoreV2_43
save_4/Assign_43
save_4/RestoreV2_44/tensor_names
save_4/RestoreV2_44/shape_and_slices
save_4/RestoreV2_44
save_4/Assign_44
save_4/restore_all/NoOp
save_4/restore_all/NoOp_1
save_4/restore_all
last_successful_epoch_5/initial_value
last_successful_epoch_5
last_successful_epoch_5/Assign
last_successful_epoch_5/read
add_5/y
add_5
assign_updated_epoch_5
Inputs_5/Data/X_input
Inputs_5/Train_Labels/y_true_one_hot_label
Inputs_5/Train_Labels/y_true_class
Parameters_5/Weights/zeros
Parameters_5/Weights/Weight
Parameters_5/Weights/Weight/Assign
Parameters_5/Weights/Weight/read
Parameters_5/Bias/ones
Parameters_5/Bias/Bias
Parameters_5/Bias/Bias/Assign
Parameters_5/Bias/Bias/read
Predictions_5/mlp_layer/MatMul
Predictions_5/mlp_layer
Predictions_5/mlp_layer_1
Predictions_5/Softmax
Predictions_5/predict_class/dimension
Predictions_5/predict_class
Predictions_5/one_hot/Const
Predictions_5/one_hot/Const_1
Predictions_5/one_hot/depth
Predictions_5/one_hot/on_value
Predictions_5/one_hot/off_value
Predictions_5/one_hot
Cross_Entropy_5/Rank
Cross_Entropy_5/Shape
Cross_Entropy_5/Rank_1
Cross_Entropy_5/Shape_1
Cross_Entropy_5/Sub/y
Cross_Entropy_5/Sub
Cross_Entropy_5/Slice/begin
Cross_Entropy_5/Slice/size
Cross_Entropy_5/Slice
Cross_Entropy_5/concat/values_0
Cross_Entropy_5/concat/axis
Cross_Entropy_5/concat
Cross_Entropy_5/Reshape
Cross_Entropy_5/Rank_2
Cross_Entropy_5/Shape_2
Cross_Entropy_5/Sub_1/y
Cross_Entropy_5/Sub_1
Cross_Entropy_5/Slice_1/begin
Cross_Entropy_5/Slice_1/size
Cross_Entropy_5/Slice_1
Cross_Entropy_5/concat_1/values_0
Cross_Entropy_5/concat_1/axis
Cross_Entropy_5/concat_1
Cross_Entropy_5/Reshape_1
Cross_Entropy_5/SoftmaxCrossEntropyWithLogits
Cross_Entropy_5/Sub_2/y
Cross_Entropy_5/Sub_2
Cross_Entropy_5/Slice_2/begin
Cross_Entropy_5/Slice_2/size
Cross_Entropy_5/Slice_2
Cross_Entropy_5/Reshape_2
Loss_Function_5/Const
Loss_Function_5/Square
Loss_Function_5/Const_1
Loss_Function_5/Mean
Loss_Function_5/Const_2
Loss_Function_5/Mean_1
Loss_Function_5/Mul
Loss_Function_5/Add
Loss_Function_5/train_loss_list/initial_value
Loss_Function_5/train_loss_list
Loss_Function_5/train_loss_list/Assign
Loss_Function_5/train_loss_list/read
Loss_Function_5/Assign/value
Loss_Function_5/Assign
Loss_Function_5/train_error/tags
Loss_Function_5/train_error
Loss_Function_5/validate_loss_list/initial_value
Loss_Function_5/validate_loss_list
Loss_Function_5/validate_loss_list/Assign
Loss_Function_5/validate_loss_list/read
Loss_Function_5/Assign_1/value
Loss_Function_5/Assign_1
Loss_Function_5/validate_error/tags
Loss_Function_5/validate_error
Optimizer_5/learning_rate_progress/initial_value
Optimizer_5/learning_rate_progress
Optimizer_5/learning_rate_progress/Assign
Optimizer_5/learning_rate_progress/read
Optimizer_5/Assign/value
Optimizer_5/Assign
Optimizer_5/learning_rate/tags
Optimizer_5/learning_rate/values
Optimizer_5/learning_rate
Optimizer_5/gradients/Shape
Optimizer_5/gradients/Const
Optimizer_5/gradients/Fill
Optimizer_5/gradients/Loss_Function_5/Add_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Add_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/BroadcastGradientArgs
Optimizer_5/gradients/Loss_Function_5/Add_grad/Sum
Optimizer_5/gradients/Loss_Function_5/Add_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Add_grad/Sum_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/Reshape_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/group_deps
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/control_dependency
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/control_dependency_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Reshape/shape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Tile
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape_2
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Const
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Prod
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Const_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Prod_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Maximum/y
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Maximum
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/floordiv
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Cast
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/truediv
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/BroadcastGradientArgs
Optimizer_5/gradients/Loss_Function_5/Mul_grad/mul
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Sum
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mul_grad/mul_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Sum_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Reshape_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/group_deps
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/control_dependency
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/control_dependency_1
Optimizer_5/gradients/Cross_Entropy_5/Reshape_2_grad/Shape
Optimizer_5/gradients/Cross_Entropy_5/Reshape_2_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Reshape/shape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Tile/multiples
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Tile
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Const
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Prod
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Const_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Prod_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Maximum/y
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Maximum
Optimizer_5/gradients/Loss_Function_5/Mean_grad/floordiv
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Cast
Optimizer_5/gradients/Loss_Function_5/Mean_grad/truediv
Optimizer_5/gradients/zeros_like
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul/x
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul_1
Optimizer_5/gradients/Cross_Entropy_5/Reshape_grad/Shape
Optimizer_5/gradients/Cross_Entropy_5/Reshape_grad/Reshape
Optimizer_5/gradients/Predictions_5/Softmax_grad/mul
Optimizer_5/gradients/Predictions_5/Softmax_grad/Sum/reduction_indices
Optimizer_5/gradients/Predictions_5/Softmax_grad/Sum
Optimizer_5/gradients/Predictions_5/Softmax_grad/Reshape/shape
Optimizer_5/gradients/Predictions_5/Softmax_grad/Reshape
Optimizer_5/gradients/Predictions_5/Softmax_grad/sub
Optimizer_5/gradients/Predictions_5/Softmax_grad/mul_1
Optimizer_5/gradients/Predictions_5/mlp_layer_1_grad/ReluGrad
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/BiasAddGrad
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/group_deps
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/control_dependency
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/control_dependency_1
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/MatMul
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/MatMul_1
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_5/gradients/AddN
Optimizer_5/GradientDescent/learning_rate
Optimizer_5/GradientDescent/update_Parameters_5/Weights/Weight/ApplyGradientDescent
Optimizer_5/GradientDescent/update_Parameters_5/Bias/Bias/ApplyGradientDescent
Optimizer_5/GradientDescent
Equal_5
Accuracy_5/Cast
Accuracy_5/Const
Accuracy_5/Mean
Accuracy_5/train_accuracy_list/initial_value
Accuracy_5/train_accuracy_list
Accuracy_5/train_accuracy_list/Assign
Accuracy_5/train_accuracy_list/read
Accuracy_5/Assign/value
Accuracy_5/Assign
Accuracy_5/train_accuracy/tags
Accuracy_5/train_accuracy
Accuracy_5/Cast_1
Accuracy_5/Const_1
Accuracy_5/Mean_1
Accuracy_5/validate_accuracy_list/initial_value
Accuracy_5/validate_accuracy_list
Accuracy_5/validate_accuracy_list/Assign
Accuracy_5/validate_accuracy_list/read
Accuracy_5/Assign_1/value
Accuracy_5/Assign_1
Accuracy_5/validate_accuracy/tags
Accuracy_5/validate_accuracy
Accuracy_5/Cast_2
Accuracy_5/Const_2
Accuracy_5/Mean_2
Accuracy_5/test_accuracy_list/initial_value
Accuracy_5/test_accuracy_list
Accuracy_5/test_accuracy_list/Assign
Accuracy_5/test_accuracy_list/read
Accuracy_5/Assign_2/value
Accuracy_5/Assign_2
Accuracy_5/test_accuracy/tags
Accuracy_5/test_accuracy
init_5/NoOp
init_5/NoOp_1
init_5
Merge_5/MergeSummary
save_5/Const
save_5/SaveV2/tensor_names
save_5/SaveV2/shape_and_slices
save_5/SaveV2
save_5/control_dependency
save_5/RestoreV2/tensor_names
save_5/RestoreV2/shape_and_slices
save_5/RestoreV2
save_5/Assign
save_5/RestoreV2_1/tensor_names
save_5/RestoreV2_1/shape_and_slices
save_5/RestoreV2_1
save_5/Assign_1
save_5/RestoreV2_2/tensor_names
save_5/RestoreV2_2/shape_and_slices
save_5/RestoreV2_2
save_5/Assign_2
save_5/RestoreV2_3/tensor_names
save_5/RestoreV2_3/shape_and_slices
save_5/RestoreV2_3
save_5/Assign_3
save_5/RestoreV2_4/tensor_names
save_5/RestoreV2_4/shape_and_slices
save_5/RestoreV2_4
save_5/Assign_4
save_5/RestoreV2_5/tensor_names
save_5/RestoreV2_5/shape_and_slices
save_5/RestoreV2_5
save_5/Assign_5
save_5/RestoreV2_6/tensor_names
save_5/RestoreV2_6/shape_and_slices
save_5/RestoreV2_6
save_5/Assign_6
save_5/RestoreV2_7/tensor_names
save_5/RestoreV2_7/shape_and_slices
save_5/RestoreV2_7
save_5/Assign_7
save_5/RestoreV2_8/tensor_names
save_5/RestoreV2_8/shape_and_slices
save_5/RestoreV2_8
save_5/Assign_8
save_5/RestoreV2_9/tensor_names
save_5/RestoreV2_9/shape_and_slices
save_5/RestoreV2_9
save_5/Assign_9
save_5/RestoreV2_10/tensor_names
save_5/RestoreV2_10/shape_and_slices
save_5/RestoreV2_10
save_5/Assign_10
save_5/RestoreV2_11/tensor_names
save_5/RestoreV2_11/shape_and_slices
save_5/RestoreV2_11
save_5/Assign_11
save_5/RestoreV2_12/tensor_names
save_5/RestoreV2_12/shape_and_slices
save_5/RestoreV2_12
save_5/Assign_12
save_5/RestoreV2_13/tensor_names
save_5/RestoreV2_13/shape_and_slices
save_5/RestoreV2_13
save_5/Assign_13
save_5/RestoreV2_14/tensor_names
save_5/RestoreV2_14/shape_and_slices
save_5/RestoreV2_14
save_5/Assign_14
save_5/RestoreV2_15/tensor_names
save_5/RestoreV2_15/shape_and_slices
save_5/RestoreV2_15
save_5/Assign_15
save_5/RestoreV2_16/tensor_names
save_5/RestoreV2_16/shape_and_slices
save_5/RestoreV2_16
save_5/Assign_16
save_5/RestoreV2_17/tensor_names
save_5/RestoreV2_17/shape_and_slices
save_5/RestoreV2_17
save_5/Assign_17
save_5/RestoreV2_18/tensor_names
save_5/RestoreV2_18/shape_and_slices
save_5/RestoreV2_18
save_5/Assign_18
save_5/RestoreV2_19/tensor_names
save_5/RestoreV2_19/shape_and_slices
save_5/RestoreV2_19
save_5/Assign_19
save_5/RestoreV2_20/tensor_names
save_5/RestoreV2_20/shape_and_slices
save_5/RestoreV2_20
save_5/Assign_20
save_5/RestoreV2_21/tensor_names
save_5/RestoreV2_21/shape_and_slices
save_5/RestoreV2_21
save_5/Assign_21
save_5/RestoreV2_22/tensor_names
save_5/RestoreV2_22/shape_and_slices
save_5/RestoreV2_22
save_5/Assign_22
save_5/RestoreV2_23/tensor_names
save_5/RestoreV2_23/shape_and_slices
save_5/RestoreV2_23
save_5/Assign_23
save_5/RestoreV2_24/tensor_names
save_5/RestoreV2_24/shape_and_slices
save_5/RestoreV2_24
save_5/Assign_24
save_5/RestoreV2_25/tensor_names
save_5/RestoreV2_25/shape_and_slices
save_5/RestoreV2_25
save_5/Assign_25
save_5/RestoreV2_26/tensor_names
save_5/RestoreV2_26/shape_and_slices
save_5/RestoreV2_26
save_5/Assign_26
save_5/RestoreV2_27/tensor_names
save_5/RestoreV2_27/shape_and_slices
save_5/RestoreV2_27
save_5/Assign_27
save_5/RestoreV2_28/tensor_names
save_5/RestoreV2_28/shape_and_slices
save_5/RestoreV2_28
save_5/Assign_28
save_5/RestoreV2_29/tensor_names
save_5/RestoreV2_29/shape_and_slices
save_5/RestoreV2_29
save_5/Assign_29
save_5/RestoreV2_30/tensor_names
save_5/RestoreV2_30/shape_and_slices
save_5/RestoreV2_30
save_5/Assign_30
save_5/RestoreV2_31/tensor_names
save_5/RestoreV2_31/shape_and_slices
save_5/RestoreV2_31
save_5/Assign_31
save_5/RestoreV2_32/tensor_names
save_5/RestoreV2_32/shape_and_slices
save_5/RestoreV2_32
save_5/Assign_32
save_5/RestoreV2_33/tensor_names
save_5/RestoreV2_33/shape_and_slices
save_5/RestoreV2_33
save_5/Assign_33
save_5/RestoreV2_34/tensor_names
save_5/RestoreV2_34/shape_and_slices
save_5/RestoreV2_34
save_5/Assign_34
save_5/RestoreV2_35/tensor_names
save_5/RestoreV2_35/shape_and_slices
save_5/RestoreV2_35
save_5/Assign_35
save_5/RestoreV2_36/tensor_names
save_5/RestoreV2_36/shape_and_slices
save_5/RestoreV2_36
save_5/Assign_36
save_5/RestoreV2_37/tensor_names
save_5/RestoreV2_37/shape_and_slices
save_5/RestoreV2_37
save_5/Assign_37
save_5/RestoreV2_38/tensor_names
save_5/RestoreV2_38/shape_and_slices
save_5/RestoreV2_38
save_5/Assign_38
save_5/RestoreV2_39/tensor_names
save_5/RestoreV2_39/shape_and_slices
save_5/RestoreV2_39
save_5/Assign_39
save_5/RestoreV2_40/tensor_names
save_5/RestoreV2_40/shape_and_slices
save_5/RestoreV2_40
save_5/Assign_40
save_5/RestoreV2_41/tensor_names
save_5/RestoreV2_41/shape_and_slices
save_5/RestoreV2_41
save_5/Assign_41
save_5/RestoreV2_42/tensor_names
save_5/RestoreV2_42/shape_and_slices
save_5/RestoreV2_42
save_5/Assign_42
save_5/RestoreV2_43/tensor_names
save_5/RestoreV2_43/shape_and_slices
save_5/RestoreV2_43
save_5/Assign_43
save_5/RestoreV2_44/tensor_names
save_5/RestoreV2_44/shape_and_slices
save_5/RestoreV2_44
save_5/Assign_44
save_5/RestoreV2_45/tensor_names
save_5/RestoreV2_45/shape_and_slices
save_5/RestoreV2_45
save_5/Assign_45
save_5/RestoreV2_46/tensor_names
save_5/RestoreV2_46/shape_and_slices
save_5/RestoreV2_46
save_5/Assign_46
save_5/RestoreV2_47/tensor_names
save_5/RestoreV2_47/shape_and_slices
save_5/RestoreV2_47
save_5/Assign_47
save_5/RestoreV2_48/tensor_names
save_5/RestoreV2_48/shape_and_slices
save_5/RestoreV2_48
save_5/Assign_48
save_5/RestoreV2_49/tensor_names
save_5/RestoreV2_49/shape_and_slices
save_5/RestoreV2_49
save_5/Assign_49
save_5/RestoreV2_50/tensor_names
save_5/RestoreV2_50/shape_and_slices
save_5/RestoreV2_50
save_5/Assign_50
save_5/RestoreV2_51/tensor_names
save_5/RestoreV2_51/shape_and_slices
save_5/RestoreV2_51
save_5/Assign_51
save_5/RestoreV2_52/tensor_names
save_5/RestoreV2_52/shape_and_slices
save_5/RestoreV2_52
save_5/Assign_52
save_5/RestoreV2_53/tensor_names
save_5/RestoreV2_53/shape_and_slices
save_5/RestoreV2_53
save_5/Assign_53
save_5/restore_all/NoOp
save_5/restore_all/NoOp_1
save_5/restore_all
Output Logits  : Tensor("Predictions/Softmax:0", shape=(?, 10), dtype=float32, device=/device:GPU:0)
Entropy        : Tensor("Cross_Entropy/Reshape_2:0", shape=(?,), dtype=float32, device=/device:GPU:0)
Predictions    : Tensor("Equal:0", shape=(?,), dtype=bool, device=/device:GPU:0) 
Optimizer      : name: "Optimizer/GradientDescent"
op: "NoOp"
input: "^Optimizer/GradientDescent/update_Parameters/Weights/Weight/ApplyGradientDescent"
input: "^Optimizer/GradientDescent/update_Parameters/Bias/Bias/ApplyGradientDescent"
device: "/device:GPU:0"
 
>> Model info
Train loss     : Tensor("Loss_Function/Add:0", shape=(), dtype=float32, device=/device:GPU:0)
Train accuracy : Tensor("Accuracy/Mean:0", shape=(), dtype=float32, device=/device:GPU:0)
Val. loss      : Tensor("Loss_Function/Add:0", shape=(), dtype=float32, device=/device:GPU:0)
Val. accuracy  : Tensor("Accuracy/Mean_1:0", shape=(), dtype=float32, device=/device:GPU:0)
>> Summary params
Train loss     : Tensor("Loss_Function/train_error:0", shape=(), dtype=string, device=/device:GPU:0)
Train accuracy : Tensor("Accuracy/train_accuracy:0", shape=(), dtype=string, device=/device:GPU:0)
Val. loss      : Tensor("Loss_Function/validate_error:0", shape=(), dtype=string, device=/device:GPU:0)
Val. accuracy  : Tensor("Accuracy/validate_accuracy:0", shape=(), dtype=string, device=/device:GPU:0)
Parameters for Linear classifier
>> Input Parameters
Input                  : Tensor("Inputs_1/Data/X_input:0", shape=(?, 3072), dtype=float32, device=/device:GPU:0) 
True one hot labels    : Tensor("Inputs_1/Train_Labels/y_true_one_hot_label:0", shape=(?, 10), dtype=float32, device=/device:GPU:0) 
True class             : Tensor("Inputs_1/Train_Labels/y_true_class:0", shape=(?,), dtype=int64, device=/device:GPU:0) 
Predict one hot labels : Tensor("Predictions_1/one_hot:0", shape=(?, 10), dtype=float32, device=/device:GPU:0) 
Predict class          : Tensor("Predictions_1/predict_class:0", shape=(?,), dtype=int64, device=/device:GPU:0) 
>> Model params
last_successful_epoch/initial_value
last_successful_epoch
last_successful_epoch/Assign
last_successful_epoch/read
add/y
add
assign_updated_epoch
Inputs/Data/X_input
Inputs/Train_Labels/y_true_one_hot_label
Inputs/Train_Labels/y_true_class
Parameters/Weights/zeros
Parameters/Weights/Weight
Parameters/Weights/Weight/Assign
Parameters/Weights/Weight/read
Parameters/Bias/ones
Parameters/Bias/Bias
Parameters/Bias/Bias/Assign
Parameters/Bias/Bias/read
Predictions/mlp_layer/MatMul
Predictions/mlp_layer
Predictions/mlp_layer_1
Predictions/Softmax
Predictions/predict_class/dimension
Predictions/predict_class
Predictions/one_hot/Const
Predictions/one_hot/Const_1
Predictions/one_hot/depth
Predictions/one_hot/on_value
Predictions/one_hot/off_value
Predictions/one_hot
Cross_Entropy/Rank
Cross_Entropy/Shape
Cross_Entropy/Rank_1
Cross_Entropy/Shape_1
Cross_Entropy/Sub/y
Cross_Entropy/Sub
Cross_Entropy/Slice/begin
Cross_Entropy/Slice/size
Cross_Entropy/Slice
Cross_Entropy/concat/values_0
Cross_Entropy/concat/axis
Cross_Entropy/concat
Cross_Entropy/Reshape
Cross_Entropy/Rank_2
Cross_Entropy/Shape_2
Cross_Entropy/Sub_1/y
Cross_Entropy/Sub_1
Cross_Entropy/Slice_1/begin
Cross_Entropy/Slice_1/size
Cross_Entropy/Slice_1
Cross_Entropy/concat_1/values_0
Cross_Entropy/concat_1/axis
Cross_Entropy/concat_1
Cross_Entropy/Reshape_1
Cross_Entropy/SoftmaxCrossEntropyWithLogits
Cross_Entropy/Sub_2/y
Cross_Entropy/Sub_2
Cross_Entropy/Slice_2/begin
Cross_Entropy/Slice_2/size
Cross_Entropy/Slice_2
Cross_Entropy/Reshape_2
Loss_Function/Const
Loss_Function/Square
Loss_Function/Const_1
Loss_Function/Mean
Loss_Function/Const_2
Loss_Function/Mean_1
Loss_Function/Mul
Loss_Function/Add
Loss_Function/train_loss_list/initial_value
Loss_Function/train_loss_list
Loss_Function/train_loss_list/Assign
Loss_Function/train_loss_list/read
Loss_Function/Assign/value
Loss_Function/Assign
Loss_Function/train_error/tags
Loss_Function/train_error
Loss_Function/validate_loss_list/initial_value
Loss_Function/validate_loss_list
Loss_Function/validate_loss_list/Assign
Loss_Function/validate_loss_list/read
Loss_Function/Assign_1/value
Loss_Function/Assign_1
Loss_Function/validate_error/tags
Loss_Function/validate_error
Optimizer/learning_rate_progress/initial_value
Optimizer/learning_rate_progress
Optimizer/learning_rate_progress/Assign
Optimizer/learning_rate_progress/read
Optimizer/Assign/value
Optimizer/Assign
Optimizer/learning_rate/tags
Optimizer/learning_rate/values
Optimizer/learning_rate
Optimizer/gradients/Shape
Optimizer/gradients/Const
Optimizer/gradients/Fill
Optimizer/gradients/Loss_Function/Add_grad/Shape
Optimizer/gradients/Loss_Function/Add_grad/Shape_1
Optimizer/gradients/Loss_Function/Add_grad/BroadcastGradientArgs
Optimizer/gradients/Loss_Function/Add_grad/Sum
Optimizer/gradients/Loss_Function/Add_grad/Reshape
Optimizer/gradients/Loss_Function/Add_grad/Sum_1
Optimizer/gradients/Loss_Function/Add_grad/Reshape_1
Optimizer/gradients/Loss_Function/Add_grad/tuple/group_deps
Optimizer/gradients/Loss_Function/Add_grad/tuple/control_dependency
Optimizer/gradients/Loss_Function/Add_grad/tuple/control_dependency_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Reshape/shape
Optimizer/gradients/Loss_Function/Mean_1_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape
Optimizer/gradients/Loss_Function/Mean_1_grad/Tile
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape_2
Optimizer/gradients/Loss_Function/Mean_1_grad/Const
Optimizer/gradients/Loss_Function/Mean_1_grad/Prod
Optimizer/gradients/Loss_Function/Mean_1_grad/Const_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Prod_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Maximum/y
Optimizer/gradients/Loss_Function/Mean_1_grad/Maximum
Optimizer/gradients/Loss_Function/Mean_1_grad/floordiv
Optimizer/gradients/Loss_Function/Mean_1_grad/Cast
Optimizer/gradients/Loss_Function/Mean_1_grad/truediv
Optimizer/gradients/Loss_Function/Mul_grad/Shape
Optimizer/gradients/Loss_Function/Mul_grad/Shape_1
Optimizer/gradients/Loss_Function/Mul_grad/BroadcastGradientArgs
Optimizer/gradients/Loss_Function/Mul_grad/mul
Optimizer/gradients/Loss_Function/Mul_grad/Sum
Optimizer/gradients/Loss_Function/Mul_grad/Reshape
Optimizer/gradients/Loss_Function/Mul_grad/mul_1
Optimizer/gradients/Loss_Function/Mul_grad/Sum_1
Optimizer/gradients/Loss_Function/Mul_grad/Reshape_1
Optimizer/gradients/Loss_Function/Mul_grad/tuple/group_deps
Optimizer/gradients/Loss_Function/Mul_grad/tuple/control_dependency
Optimizer/gradients/Loss_Function/Mul_grad/tuple/control_dependency_1
Optimizer/gradients/Cross_Entropy/Reshape_2_grad/Shape
Optimizer/gradients/Cross_Entropy/Reshape_2_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_grad/Reshape/shape
Optimizer/gradients/Loss_Function/Mean_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_grad/Tile/multiples
Optimizer/gradients/Loss_Function/Mean_grad/Tile
Optimizer/gradients/Loss_Function/Mean_grad/Shape
Optimizer/gradients/Loss_Function/Mean_grad/Shape_1
Optimizer/gradients/Loss_Function/Mean_grad/Const
Optimizer/gradients/Loss_Function/Mean_grad/Prod
Optimizer/gradients/Loss_Function/Mean_grad/Const_1
Optimizer/gradients/Loss_Function/Mean_grad/Prod_1
Optimizer/gradients/Loss_Function/Mean_grad/Maximum/y
Optimizer/gradients/Loss_Function/Mean_grad/Maximum
Optimizer/gradients/Loss_Function/Mean_grad/floordiv
Optimizer/gradients/Loss_Function/Mean_grad/Cast
Optimizer/gradients/Loss_Function/Mean_grad/truediv
Optimizer/gradients/zeros_like
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer/gradients/Loss_Function/Square_grad/mul/x
Optimizer/gradients/Loss_Function/Square_grad/mul
Optimizer/gradients/Loss_Function/Square_grad/mul_1
Optimizer/gradients/Cross_Entropy/Reshape_grad/Shape
Optimizer/gradients/Cross_Entropy/Reshape_grad/Reshape
Optimizer/gradients/Predictions/Softmax_grad/mul
Optimizer/gradients/Predictions/Softmax_grad/Sum/reduction_indices
Optimizer/gradients/Predictions/Softmax_grad/Sum
Optimizer/gradients/Predictions/Softmax_grad/Reshape/shape
Optimizer/gradients/Predictions/Softmax_grad/Reshape
Optimizer/gradients/Predictions/Softmax_grad/sub
Optimizer/gradients/Predictions/Softmax_grad/mul_1
Optimizer/gradients/Predictions/mlp_layer_1_grad/ReluGrad
Optimizer/gradients/Predictions/mlp_layer_grad/BiasAddGrad
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/group_deps
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/control_dependency
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/control_dependency_1
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/MatMul
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/MatMul_1
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer/gradients/AddN
Optimizer/GradientDescent/learning_rate
Optimizer/GradientDescent/update_Parameters/Weights/Weight/ApplyGradientDescent
Optimizer/GradientDescent/update_Parameters/Bias/Bias/ApplyGradientDescent
Optimizer/GradientDescent
Equal
Accuracy/Cast
Accuracy/Const
Accuracy/Mean
Accuracy/train_accuracy_list/initial_value
Accuracy/train_accuracy_list
Accuracy/train_accuracy_list/Assign
Accuracy/train_accuracy_list/read
Accuracy/Assign/value
Accuracy/Assign
Accuracy/train_accuracy/tags
Accuracy/train_accuracy
Accuracy/Cast_1
Accuracy/Const_1
Accuracy/Mean_1
Accuracy/validate_accuracy_list/initial_value
Accuracy/validate_accuracy_list
Accuracy/validate_accuracy_list/Assign
Accuracy/validate_accuracy_list/read
Accuracy/Assign_1/value
Accuracy/Assign_1
Accuracy/validate_accuracy/tags
Accuracy/validate_accuracy
Accuracy/Cast_2
Accuracy/Const_2
Accuracy/Mean_2
Accuracy/test_accuracy_list/initial_value
Accuracy/test_accuracy_list
Accuracy/test_accuracy_list/Assign
Accuracy/test_accuracy_list/read
Accuracy/Assign_2/value
Accuracy/Assign_2
Accuracy/test_accuracy/tags
Accuracy/test_accuracy
init/NoOp
init/NoOp_1
init
Merge/MergeSummary
save/Const
save/SaveV2/tensor_names
save/SaveV2/shape_and_slices
save/SaveV2
save/control_dependency
save/RestoreV2/tensor_names
save/RestoreV2/shape_and_slices
save/RestoreV2
save/Assign
save/RestoreV2_1/tensor_names
save/RestoreV2_1/shape_and_slices
save/RestoreV2_1
save/Assign_1
save/RestoreV2_2/tensor_names
save/RestoreV2_2/shape_and_slices
save/RestoreV2_2
save/Assign_2
save/RestoreV2_3/tensor_names
save/RestoreV2_3/shape_and_slices
save/RestoreV2_3
save/Assign_3
save/RestoreV2_4/tensor_names
save/RestoreV2_4/shape_and_slices
save/RestoreV2_4
save/Assign_4
save/RestoreV2_5/tensor_names
save/RestoreV2_5/shape_and_slices
save/RestoreV2_5
save/Assign_5
save/RestoreV2_6/tensor_names
save/RestoreV2_6/shape_and_slices
save/RestoreV2_6
save/Assign_6
save/RestoreV2_7/tensor_names
save/RestoreV2_7/shape_and_slices
save/RestoreV2_7
save/Assign_7
save/RestoreV2_8/tensor_names
save/RestoreV2_8/shape_and_slices
save/RestoreV2_8
save/Assign_8
save/restore_all/NoOp
save/restore_all/NoOp_1
save/restore_all
last_successful_epoch_1/initial_value
last_successful_epoch_1
last_successful_epoch_1/Assign
last_successful_epoch_1/read
add_1/y
add_1
assign_updated_epoch_1
Inputs_1/Data/X_input
Inputs_1/Train_Labels/y_true_one_hot_label
Inputs_1/Train_Labels/y_true_class
Parameters_1/Weights/zeros
Parameters_1/Weights/Weight
Parameters_1/Weights/Weight/Assign
Parameters_1/Weights/Weight/read
Parameters_1/Bias/ones
Parameters_1/Bias/Bias
Parameters_1/Bias/Bias/Assign
Parameters_1/Bias/Bias/read
Predictions_1/mlp_layer/MatMul
Predictions_1/mlp_layer
Predictions_1/mlp_layer_1
Predictions_1/Softmax
Predictions_1/predict_class/dimension
Predictions_1/predict_class
Predictions_1/one_hot/Const
Predictions_1/one_hot/Const_1
Predictions_1/one_hot/depth
Predictions_1/one_hot/on_value
Predictions_1/one_hot/off_value
Predictions_1/one_hot
Cross_Entropy_1/Rank
Cross_Entropy_1/Shape
Cross_Entropy_1/Rank_1
Cross_Entropy_1/Shape_1
Cross_Entropy_1/Sub/y
Cross_Entropy_1/Sub
Cross_Entropy_1/Slice/begin
Cross_Entropy_1/Slice/size
Cross_Entropy_1/Slice
Cross_Entropy_1/concat/values_0
Cross_Entropy_1/concat/axis
Cross_Entropy_1/concat
Cross_Entropy_1/Reshape
Cross_Entropy_1/Rank_2
Cross_Entropy_1/Shape_2
Cross_Entropy_1/Sub_1/y
Cross_Entropy_1/Sub_1
Cross_Entropy_1/Slice_1/begin
Cross_Entropy_1/Slice_1/size
Cross_Entropy_1/Slice_1
Cross_Entropy_1/concat_1/values_0
Cross_Entropy_1/concat_1/axis
Cross_Entropy_1/concat_1
Cross_Entropy_1/Reshape_1
Cross_Entropy_1/SoftmaxCrossEntropyWithLogits
Cross_Entropy_1/Sub_2/y
Cross_Entropy_1/Sub_2
Cross_Entropy_1/Slice_2/begin
Cross_Entropy_1/Slice_2/size
Cross_Entropy_1/Slice_2
Cross_Entropy_1/Reshape_2
Loss_Function_1/Const
Loss_Function_1/Square
Loss_Function_1/Const_1
Loss_Function_1/Mean
Loss_Function_1/Const_2
Loss_Function_1/Mean_1
Loss_Function_1/Mul
Loss_Function_1/Add
Loss_Function_1/train_loss_list/initial_value
Loss_Function_1/train_loss_list
Loss_Function_1/train_loss_list/Assign
Loss_Function_1/train_loss_list/read
Loss_Function_1/Assign/value
Loss_Function_1/Assign
Loss_Function_1/train_error/tags
Loss_Function_1/train_error
Loss_Function_1/validate_loss_list/initial_value
Loss_Function_1/validate_loss_list
Loss_Function_1/validate_loss_list/Assign
Loss_Function_1/validate_loss_list/read
Loss_Function_1/Assign_1/value
Loss_Function_1/Assign_1
Loss_Function_1/validate_error/tags
Loss_Function_1/validate_error
Optimizer_1/learning_rate_progress/initial_value
Optimizer_1/learning_rate_progress
Optimizer_1/learning_rate_progress/Assign
Optimizer_1/learning_rate_progress/read
Optimizer_1/Assign/value
Optimizer_1/Assign
Optimizer_1/learning_rate/tags
Optimizer_1/learning_rate/values
Optimizer_1/learning_rate
Optimizer_1/gradients/Shape
Optimizer_1/gradients/Const
Optimizer_1/gradients/Fill
Optimizer_1/gradients/Loss_Function_1/Add_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Add_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/BroadcastGradientArgs
Optimizer_1/gradients/Loss_Function_1/Add_grad/Sum
Optimizer_1/gradients/Loss_Function_1/Add_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Add_grad/Sum_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/Reshape_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/group_deps
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/control_dependency
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/control_dependency_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Reshape/shape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Tile
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape_2
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Const
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Prod
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Const_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Prod_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Maximum/y
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Maximum
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/floordiv
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Cast
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/truediv
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/BroadcastGradientArgs
Optimizer_1/gradients/Loss_Function_1/Mul_grad/mul
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Sum
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mul_grad/mul_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Sum_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Reshape_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/group_deps
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/control_dependency
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/control_dependency_1
Optimizer_1/gradients/Cross_Entropy_1/Reshape_2_grad/Shape
Optimizer_1/gradients/Cross_Entropy_1/Reshape_2_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Reshape/shape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Tile/multiples
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Tile
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Const
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Prod
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Const_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Prod_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Maximum/y
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Maximum
Optimizer_1/gradients/Loss_Function_1/Mean_grad/floordiv
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Cast
Optimizer_1/gradients/Loss_Function_1/Mean_grad/truediv
Optimizer_1/gradients/zeros_like
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul/x
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul_1
Optimizer_1/gradients/Cross_Entropy_1/Reshape_grad/Shape
Optimizer_1/gradients/Cross_Entropy_1/Reshape_grad/Reshape
Optimizer_1/gradients/Predictions_1/Softmax_grad/mul
Optimizer_1/gradients/Predictions_1/Softmax_grad/Sum/reduction_indices
Optimizer_1/gradients/Predictions_1/Softmax_grad/Sum
Optimizer_1/gradients/Predictions_1/Softmax_grad/Reshape/shape
Optimizer_1/gradients/Predictions_1/Softmax_grad/Reshape
Optimizer_1/gradients/Predictions_1/Softmax_grad/sub
Optimizer_1/gradients/Predictions_1/Softmax_grad/mul_1
Optimizer_1/gradients/Predictions_1/mlp_layer_1_grad/ReluGrad
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/BiasAddGrad
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/group_deps
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/control_dependency
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/control_dependency_1
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/MatMul
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/MatMul_1
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_1/gradients/AddN
Optimizer_1/GradientDescent/learning_rate
Optimizer_1/GradientDescent/update_Parameters_1/Weights/Weight/ApplyGradientDescent
Optimizer_1/GradientDescent/update_Parameters_1/Bias/Bias/ApplyGradientDescent
Optimizer_1/GradientDescent
Equal_1
Accuracy_1/Cast
Accuracy_1/Const
Accuracy_1/Mean
Accuracy_1/train_accuracy_list/initial_value
Accuracy_1/train_accuracy_list
Accuracy_1/train_accuracy_list/Assign
Accuracy_1/train_accuracy_list/read
Accuracy_1/Assign/value
Accuracy_1/Assign
Accuracy_1/train_accuracy/tags
Accuracy_1/train_accuracy
Accuracy_1/Cast_1
Accuracy_1/Const_1
Accuracy_1/Mean_1
Accuracy_1/validate_accuracy_list/initial_value
Accuracy_1/validate_accuracy_list
Accuracy_1/validate_accuracy_list/Assign
Accuracy_1/validate_accuracy_list/read
Accuracy_1/Assign_1/value
Accuracy_1/Assign_1
Accuracy_1/validate_accuracy/tags
Accuracy_1/validate_accuracy
Accuracy_1/Cast_2
Accuracy_1/Const_2
Accuracy_1/Mean_2
Accuracy_1/test_accuracy_list/initial_value
Accuracy_1/test_accuracy_list
Accuracy_1/test_accuracy_list/Assign
Accuracy_1/test_accuracy_list/read
Accuracy_1/Assign_2/value
Accuracy_1/Assign_2
Accuracy_1/test_accuracy/tags
Accuracy_1/test_accuracy
init_1/NoOp
init_1/NoOp_1
init_1
Merge_1/MergeSummary
save_1/Const
save_1/SaveV2/tensor_names
save_1/SaveV2/shape_and_slices
save_1/SaveV2
save_1/control_dependency
save_1/RestoreV2/tensor_names
save_1/RestoreV2/shape_and_slices
save_1/RestoreV2
save_1/Assign
save_1/RestoreV2_1/tensor_names
save_1/RestoreV2_1/shape_and_slices
save_1/RestoreV2_1
save_1/Assign_1
save_1/RestoreV2_2/tensor_names
save_1/RestoreV2_2/shape_and_slices
save_1/RestoreV2_2
save_1/Assign_2
save_1/RestoreV2_3/tensor_names
save_1/RestoreV2_3/shape_and_slices
save_1/RestoreV2_3
save_1/Assign_3
save_1/RestoreV2_4/tensor_names
save_1/RestoreV2_4/shape_and_slices
save_1/RestoreV2_4
save_1/Assign_4
save_1/RestoreV2_5/tensor_names
save_1/RestoreV2_5/shape_and_slices
save_1/RestoreV2_5
save_1/Assign_5
save_1/RestoreV2_6/tensor_names
save_1/RestoreV2_6/shape_and_slices
save_1/RestoreV2_6
save_1/Assign_6
save_1/RestoreV2_7/tensor_names
save_1/RestoreV2_7/shape_and_slices
save_1/RestoreV2_7
save_1/Assign_7
save_1/RestoreV2_8/tensor_names
save_1/RestoreV2_8/shape_and_slices
save_1/RestoreV2_8
save_1/Assign_8
save_1/RestoreV2_9/tensor_names
save_1/RestoreV2_9/shape_and_slices
save_1/RestoreV2_9
save_1/Assign_9
save_1/RestoreV2_10/tensor_names
save_1/RestoreV2_10/shape_and_slices
save_1/RestoreV2_10
save_1/Assign_10
save_1/RestoreV2_11/tensor_names
save_1/RestoreV2_11/shape_and_slices
save_1/RestoreV2_11
save_1/Assign_11
save_1/RestoreV2_12/tensor_names
save_1/RestoreV2_12/shape_and_slices
save_1/RestoreV2_12
save_1/Assign_12
save_1/RestoreV2_13/tensor_names
save_1/RestoreV2_13/shape_and_slices
save_1/RestoreV2_13
save_1/Assign_13
save_1/RestoreV2_14/tensor_names
save_1/RestoreV2_14/shape_and_slices
save_1/RestoreV2_14
save_1/Assign_14
save_1/RestoreV2_15/tensor_names
save_1/RestoreV2_15/shape_and_slices
save_1/RestoreV2_15
save_1/Assign_15
save_1/RestoreV2_16/tensor_names
save_1/RestoreV2_16/shape_and_slices
save_1/RestoreV2_16
save_1/Assign_16
save_1/RestoreV2_17/tensor_names
save_1/RestoreV2_17/shape_and_slices
save_1/RestoreV2_17
save_1/Assign_17
save_1/restore_all/NoOp
save_1/restore_all/NoOp_1
save_1/restore_all
last_successful_epoch_2/initial_value
last_successful_epoch_2
last_successful_epoch_2/Assign
last_successful_epoch_2/read
add_2/y
add_2
assign_updated_epoch_2
Inputs_2/Data/X_input
Inputs_2/Train_Labels/y_true_one_hot_label
Inputs_2/Train_Labels/y_true_class
Parameters_2/Weights/zeros
Parameters_2/Weights/Weight
Parameters_2/Weights/Weight/Assign
Parameters_2/Weights/Weight/read
Parameters_2/Bias/ones
Parameters_2/Bias/Bias
Parameters_2/Bias/Bias/Assign
Parameters_2/Bias/Bias/read
Predictions_2/mlp_layer/MatMul
Predictions_2/mlp_layer
Predictions_2/mlp_layer_1
Predictions_2/Softmax
Predictions_2/predict_class/dimension
Predictions_2/predict_class
Predictions_2/one_hot/Const
Predictions_2/one_hot/Const_1
Predictions_2/one_hot/depth
Predictions_2/one_hot/on_value
Predictions_2/one_hot/off_value
Predictions_2/one_hot
Cross_Entropy_2/Rank
Cross_Entropy_2/Shape
Cross_Entropy_2/Rank_1
Cross_Entropy_2/Shape_1
Cross_Entropy_2/Sub/y
Cross_Entropy_2/Sub
Cross_Entropy_2/Slice/begin
Cross_Entropy_2/Slice/size
Cross_Entropy_2/Slice
Cross_Entropy_2/concat/values_0
Cross_Entropy_2/concat/axis
Cross_Entropy_2/concat
Cross_Entropy_2/Reshape
Cross_Entropy_2/Rank_2
Cross_Entropy_2/Shape_2
Cross_Entropy_2/Sub_1/y
Cross_Entropy_2/Sub_1
Cross_Entropy_2/Slice_1/begin
Cross_Entropy_2/Slice_1/size
Cross_Entropy_2/Slice_1
Cross_Entropy_2/concat_1/values_0
Cross_Entropy_2/concat_1/axis
Cross_Entropy_2/concat_1
Cross_Entropy_2/Reshape_1
Cross_Entropy_2/SoftmaxCrossEntropyWithLogits
Cross_Entropy_2/Sub_2/y
Cross_Entropy_2/Sub_2
Cross_Entropy_2/Slice_2/begin
Cross_Entropy_2/Slice_2/size
Cross_Entropy_2/Slice_2
Cross_Entropy_2/Reshape_2
Loss_Function_2/Const
Loss_Function_2/Square
Loss_Function_2/Const_1
Loss_Function_2/Mean
Loss_Function_2/Const_2
Loss_Function_2/Mean_1
Loss_Function_2/Mul
Loss_Function_2/Add
Loss_Function_2/train_loss_list/initial_value
Loss_Function_2/train_loss_list
Loss_Function_2/train_loss_list/Assign
Loss_Function_2/train_loss_list/read
Loss_Function_2/Assign/value
Loss_Function_2/Assign
Loss_Function_2/train_error/tags
Loss_Function_2/train_error
Loss_Function_2/validate_loss_list/initial_value
Loss_Function_2/validate_loss_list
Loss_Function_2/validate_loss_list/Assign
Loss_Function_2/validate_loss_list/read
Loss_Function_2/Assign_1/value
Loss_Function_2/Assign_1
Loss_Function_2/validate_error/tags
Loss_Function_2/validate_error
Optimizer_2/learning_rate_progress/initial_value
Optimizer_2/learning_rate_progress
Optimizer_2/learning_rate_progress/Assign
Optimizer_2/learning_rate_progress/read
Optimizer_2/Assign/value
Optimizer_2/Assign
Optimizer_2/learning_rate/tags
Optimizer_2/learning_rate/values
Optimizer_2/learning_rate
Optimizer_2/gradients/Shape
Optimizer_2/gradients/Const
Optimizer_2/gradients/Fill
Optimizer_2/gradients/Loss_Function_2/Add_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Add_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/BroadcastGradientArgs
Optimizer_2/gradients/Loss_Function_2/Add_grad/Sum
Optimizer_2/gradients/Loss_Function_2/Add_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Add_grad/Sum_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/Reshape_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/group_deps
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/control_dependency
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/control_dependency_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Reshape/shape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Tile
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape_2
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Const
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Prod
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Const_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Prod_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Maximum/y
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Maximum
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/floordiv
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Cast
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/truediv
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/BroadcastGradientArgs
Optimizer_2/gradients/Loss_Function_2/Mul_grad/mul
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Sum
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mul_grad/mul_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Sum_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Reshape_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/group_deps
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/control_dependency
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/control_dependency_1
Optimizer_2/gradients/Cross_Entropy_2/Reshape_2_grad/Shape
Optimizer_2/gradients/Cross_Entropy_2/Reshape_2_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Reshape/shape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Tile/multiples
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Tile
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Const
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Prod
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Const_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Prod_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Maximum/y
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Maximum
Optimizer_2/gradients/Loss_Function_2/Mean_grad/floordiv
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Cast
Optimizer_2/gradients/Loss_Function_2/Mean_grad/truediv
Optimizer_2/gradients/zeros_like
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul/x
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul_1
Optimizer_2/gradients/Cross_Entropy_2/Reshape_grad/Shape
Optimizer_2/gradients/Cross_Entropy_2/Reshape_grad/Reshape
Optimizer_2/gradients/Predictions_2/Softmax_grad/mul
Optimizer_2/gradients/Predictions_2/Softmax_grad/Sum/reduction_indices
Optimizer_2/gradients/Predictions_2/Softmax_grad/Sum
Optimizer_2/gradients/Predictions_2/Softmax_grad/Reshape/shape
Optimizer_2/gradients/Predictions_2/Softmax_grad/Reshape
Optimizer_2/gradients/Predictions_2/Softmax_grad/sub
Optimizer_2/gradients/Predictions_2/Softmax_grad/mul_1
Optimizer_2/gradients/Predictions_2/mlp_layer_1_grad/ReluGrad
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/BiasAddGrad
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/group_deps
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/control_dependency
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/control_dependency_1
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/MatMul
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/MatMul_1
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_2/gradients/AddN
Optimizer_2/GradientDescent/learning_rate
Optimizer_2/GradientDescent/update_Parameters_2/Weights/Weight/ApplyGradientDescent
Optimizer_2/GradientDescent/update_Parameters_2/Bias/Bias/ApplyGradientDescent
Optimizer_2/GradientDescent
Equal_2
Accuracy_2/Cast
Accuracy_2/Const
Accuracy_2/Mean
Accuracy_2/train_accuracy_list/initial_value
Accuracy_2/train_accuracy_list
Accuracy_2/train_accuracy_list/Assign
Accuracy_2/train_accuracy_list/read
Accuracy_2/Assign/value
Accuracy_2/Assign
Accuracy_2/train_accuracy/tags
Accuracy_2/train_accuracy
Accuracy_2/Cast_1
Accuracy_2/Const_1
Accuracy_2/Mean_1
Accuracy_2/validate_accuracy_list/initial_value
Accuracy_2/validate_accuracy_list
Accuracy_2/validate_accuracy_list/Assign
Accuracy_2/validate_accuracy_list/read
Accuracy_2/Assign_1/value
Accuracy_2/Assign_1
Accuracy_2/validate_accuracy/tags
Accuracy_2/validate_accuracy
Accuracy_2/Cast_2
Accuracy_2/Const_2
Accuracy_2/Mean_2
Accuracy_2/test_accuracy_list/initial_value
Accuracy_2/test_accuracy_list
Accuracy_2/test_accuracy_list/Assign
Accuracy_2/test_accuracy_list/read
Accuracy_2/Assign_2/value
Accuracy_2/Assign_2
Accuracy_2/test_accuracy/tags
Accuracy_2/test_accuracy
init_2/NoOp
init_2/NoOp_1
init_2
Merge_2/MergeSummary
save_2/Const
save_2/SaveV2/tensor_names
save_2/SaveV2/shape_and_slices
save_2/SaveV2
save_2/control_dependency
save_2/RestoreV2/tensor_names
save_2/RestoreV2/shape_and_slices
save_2/RestoreV2
save_2/Assign
save_2/RestoreV2_1/tensor_names
save_2/RestoreV2_1/shape_and_slices
save_2/RestoreV2_1
save_2/Assign_1
save_2/RestoreV2_2/tensor_names
save_2/RestoreV2_2/shape_and_slices
save_2/RestoreV2_2
save_2/Assign_2
save_2/RestoreV2_3/tensor_names
save_2/RestoreV2_3/shape_and_slices
save_2/RestoreV2_3
save_2/Assign_3
save_2/RestoreV2_4/tensor_names
save_2/RestoreV2_4/shape_and_slices
save_2/RestoreV2_4
save_2/Assign_4
save_2/RestoreV2_5/tensor_names
save_2/RestoreV2_5/shape_and_slices
save_2/RestoreV2_5
save_2/Assign_5
save_2/RestoreV2_6/tensor_names
save_2/RestoreV2_6/shape_and_slices
save_2/RestoreV2_6
save_2/Assign_6
save_2/RestoreV2_7/tensor_names
save_2/RestoreV2_7/shape_and_slices
save_2/RestoreV2_7
save_2/Assign_7
save_2/RestoreV2_8/tensor_names
save_2/RestoreV2_8/shape_and_slices
save_2/RestoreV2_8
save_2/Assign_8
save_2/RestoreV2_9/tensor_names
save_2/RestoreV2_9/shape_and_slices
save_2/RestoreV2_9
save_2/Assign_9
save_2/RestoreV2_10/tensor_names
save_2/RestoreV2_10/shape_and_slices
save_2/RestoreV2_10
save_2/Assign_10
save_2/RestoreV2_11/tensor_names
save_2/RestoreV2_11/shape_and_slices
save_2/RestoreV2_11
save_2/Assign_11
save_2/RestoreV2_12/tensor_names
save_2/RestoreV2_12/shape_and_slices
save_2/RestoreV2_12
save_2/Assign_12
save_2/RestoreV2_13/tensor_names
save_2/RestoreV2_13/shape_and_slices
save_2/RestoreV2_13
save_2/Assign_13
save_2/RestoreV2_14/tensor_names
save_2/RestoreV2_14/shape_and_slices
save_2/RestoreV2_14
save_2/Assign_14
save_2/RestoreV2_15/tensor_names
save_2/RestoreV2_15/shape_and_slices
save_2/RestoreV2_15
save_2/Assign_15
save_2/RestoreV2_16/tensor_names
save_2/RestoreV2_16/shape_and_slices
save_2/RestoreV2_16
save_2/Assign_16
save_2/RestoreV2_17/tensor_names
save_2/RestoreV2_17/shape_and_slices
save_2/RestoreV2_17
save_2/Assign_17
save_2/RestoreV2_18/tensor_names
save_2/RestoreV2_18/shape_and_slices
save_2/RestoreV2_18
save_2/Assign_18
save_2/RestoreV2_19/tensor_names
save_2/RestoreV2_19/shape_and_slices
save_2/RestoreV2_19
save_2/Assign_19
save_2/RestoreV2_20/tensor_names
save_2/RestoreV2_20/shape_and_slices
save_2/RestoreV2_20
save_2/Assign_20
save_2/RestoreV2_21/tensor_names
save_2/RestoreV2_21/shape_and_slices
save_2/RestoreV2_21
save_2/Assign_21
save_2/RestoreV2_22/tensor_names
save_2/RestoreV2_22/shape_and_slices
save_2/RestoreV2_22
save_2/Assign_22
save_2/RestoreV2_23/tensor_names
save_2/RestoreV2_23/shape_and_slices
save_2/RestoreV2_23
save_2/Assign_23
save_2/RestoreV2_24/tensor_names
save_2/RestoreV2_24/shape_and_slices
save_2/RestoreV2_24
save_2/Assign_24
save_2/RestoreV2_25/tensor_names
save_2/RestoreV2_25/shape_and_slices
save_2/RestoreV2_25
save_2/Assign_25
save_2/RestoreV2_26/tensor_names
save_2/RestoreV2_26/shape_and_slices
save_2/RestoreV2_26
save_2/Assign_26
save_2/restore_all/NoOp
save_2/restore_all/NoOp_1
save_2/restore_all
last_successful_epoch_3/initial_value
last_successful_epoch_3
last_successful_epoch_3/Assign
last_successful_epoch_3/read
add_3/y
add_3
assign_updated_epoch_3
Inputs_3/Data/X_input
Inputs_3/Train_Labels/y_true_one_hot_label
Inputs_3/Train_Labels/y_true_class
Parameters_3/Weights/zeros
Parameters_3/Weights/Weight
Parameters_3/Weights/Weight/Assign
Parameters_3/Weights/Weight/read
Parameters_3/Bias/ones
Parameters_3/Bias/Bias
Parameters_3/Bias/Bias/Assign
Parameters_3/Bias/Bias/read
Predictions_3/mlp_layer/MatMul
Predictions_3/mlp_layer
Predictions_3/mlp_layer_1
Predictions_3/Softmax
Predictions_3/predict_class/dimension
Predictions_3/predict_class
Predictions_3/one_hot/Const
Predictions_3/one_hot/Const_1
Predictions_3/one_hot/depth
Predictions_3/one_hot/on_value
Predictions_3/one_hot/off_value
Predictions_3/one_hot
Cross_Entropy_3/Rank
Cross_Entropy_3/Shape
Cross_Entropy_3/Rank_1
Cross_Entropy_3/Shape_1
Cross_Entropy_3/Sub/y
Cross_Entropy_3/Sub
Cross_Entropy_3/Slice/begin
Cross_Entropy_3/Slice/size
Cross_Entropy_3/Slice
Cross_Entropy_3/concat/values_0
Cross_Entropy_3/concat/axis
Cross_Entropy_3/concat
Cross_Entropy_3/Reshape
Cross_Entropy_3/Rank_2
Cross_Entropy_3/Shape_2
Cross_Entropy_3/Sub_1/y
Cross_Entropy_3/Sub_1
Cross_Entropy_3/Slice_1/begin
Cross_Entropy_3/Slice_1/size
Cross_Entropy_3/Slice_1
Cross_Entropy_3/concat_1/values_0
Cross_Entropy_3/concat_1/axis
Cross_Entropy_3/concat_1
Cross_Entropy_3/Reshape_1
Cross_Entropy_3/SoftmaxCrossEntropyWithLogits
Cross_Entropy_3/Sub_2/y
Cross_Entropy_3/Sub_2
Cross_Entropy_3/Slice_2/begin
Cross_Entropy_3/Slice_2/size
Cross_Entropy_3/Slice_2
Cross_Entropy_3/Reshape_2
Loss_Function_3/Const
Loss_Function_3/Square
Loss_Function_3/Const_1
Loss_Function_3/Mean
Loss_Function_3/Const_2
Loss_Function_3/Mean_1
Loss_Function_3/Mul
Loss_Function_3/Add
Loss_Function_3/train_loss_list/initial_value
Loss_Function_3/train_loss_list
Loss_Function_3/train_loss_list/Assign
Loss_Function_3/train_loss_list/read
Loss_Function_3/Assign/value
Loss_Function_3/Assign
Loss_Function_3/train_error/tags
Loss_Function_3/train_error
Loss_Function_3/validate_loss_list/initial_value
Loss_Function_3/validate_loss_list
Loss_Function_3/validate_loss_list/Assign
Loss_Function_3/validate_loss_list/read
Loss_Function_3/Assign_1/value
Loss_Function_3/Assign_1
Loss_Function_3/validate_error/tags
Loss_Function_3/validate_error
Optimizer_3/learning_rate_progress/initial_value
Optimizer_3/learning_rate_progress
Optimizer_3/learning_rate_progress/Assign
Optimizer_3/learning_rate_progress/read
Optimizer_3/Assign/value
Optimizer_3/Assign
Optimizer_3/learning_rate/tags
Optimizer_3/learning_rate/values
Optimizer_3/learning_rate
Optimizer_3/gradients/Shape
Optimizer_3/gradients/Const
Optimizer_3/gradients/Fill
Optimizer_3/gradients/Loss_Function_3/Add_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Add_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/BroadcastGradientArgs
Optimizer_3/gradients/Loss_Function_3/Add_grad/Sum
Optimizer_3/gradients/Loss_Function_3/Add_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Add_grad/Sum_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/Reshape_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/group_deps
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/control_dependency
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/control_dependency_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Reshape/shape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Tile
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape_2
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Const
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Prod
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Const_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Prod_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Maximum/y
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Maximum
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/floordiv
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Cast
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/truediv
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/BroadcastGradientArgs
Optimizer_3/gradients/Loss_Function_3/Mul_grad/mul
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Sum
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mul_grad/mul_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Sum_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Reshape_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/group_deps
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/control_dependency
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/control_dependency_1
Optimizer_3/gradients/Cross_Entropy_3/Reshape_2_grad/Shape
Optimizer_3/gradients/Cross_Entropy_3/Reshape_2_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Reshape/shape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Tile/multiples
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Tile
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Const
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Prod
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Const_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Prod_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Maximum/y
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Maximum
Optimizer_3/gradients/Loss_Function_3/Mean_grad/floordiv
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Cast
Optimizer_3/gradients/Loss_Function_3/Mean_grad/truediv
Optimizer_3/gradients/zeros_like
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul/x
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul_1
Optimizer_3/gradients/Cross_Entropy_3/Reshape_grad/Shape
Optimizer_3/gradients/Cross_Entropy_3/Reshape_grad/Reshape
Optimizer_3/gradients/Predictions_3/Softmax_grad/mul
Optimizer_3/gradients/Predictions_3/Softmax_grad/Sum/reduction_indices
Optimizer_3/gradients/Predictions_3/Softmax_grad/Sum
Optimizer_3/gradients/Predictions_3/Softmax_grad/Reshape/shape
Optimizer_3/gradients/Predictions_3/Softmax_grad/Reshape
Optimizer_3/gradients/Predictions_3/Softmax_grad/sub
Optimizer_3/gradients/Predictions_3/Softmax_grad/mul_1
Optimizer_3/gradients/Predictions_3/mlp_layer_1_grad/ReluGrad
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/BiasAddGrad
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/group_deps
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/control_dependency
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/control_dependency_1
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/MatMul
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/MatMul_1
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_3/gradients/AddN
Optimizer_3/GradientDescent/learning_rate
Optimizer_3/GradientDescent/update_Parameters_3/Weights/Weight/ApplyGradientDescent
Optimizer_3/GradientDescent/update_Parameters_3/Bias/Bias/ApplyGradientDescent
Optimizer_3/GradientDescent
Equal_3
Accuracy_3/Cast
Accuracy_3/Const
Accuracy_3/Mean
Accuracy_3/train_accuracy_list/initial_value
Accuracy_3/train_accuracy_list
Accuracy_3/train_accuracy_list/Assign
Accuracy_3/train_accuracy_list/read
Accuracy_3/Assign/value
Accuracy_3/Assign
Accuracy_3/train_accuracy/tags
Accuracy_3/train_accuracy
Accuracy_3/Cast_1
Accuracy_3/Const_1
Accuracy_3/Mean_1
Accuracy_3/validate_accuracy_list/initial_value
Accuracy_3/validate_accuracy_list
Accuracy_3/validate_accuracy_list/Assign
Accuracy_3/validate_accuracy_list/read
Accuracy_3/Assign_1/value
Accuracy_3/Assign_1
Accuracy_3/validate_accuracy/tags
Accuracy_3/validate_accuracy
Accuracy_3/Cast_2
Accuracy_3/Const_2
Accuracy_3/Mean_2
Accuracy_3/test_accuracy_list/initial_value
Accuracy_3/test_accuracy_list
Accuracy_3/test_accuracy_list/Assign
Accuracy_3/test_accuracy_list/read
Accuracy_3/Assign_2/value
Accuracy_3/Assign_2
Accuracy_3/test_accuracy/tags
Accuracy_3/test_accuracy
init_3/NoOp
init_3/NoOp_1
init_3
Merge_3/MergeSummary
save_3/Const
save_3/SaveV2/tensor_names
save_3/SaveV2/shape_and_slices
save_3/SaveV2
save_3/control_dependency
save_3/RestoreV2/tensor_names
save_3/RestoreV2/shape_and_slices
save_3/RestoreV2
save_3/Assign
save_3/RestoreV2_1/tensor_names
save_3/RestoreV2_1/shape_and_slices
save_3/RestoreV2_1
save_3/Assign_1
save_3/RestoreV2_2/tensor_names
save_3/RestoreV2_2/shape_and_slices
save_3/RestoreV2_2
save_3/Assign_2
save_3/RestoreV2_3/tensor_names
save_3/RestoreV2_3/shape_and_slices
save_3/RestoreV2_3
save_3/Assign_3
save_3/RestoreV2_4/tensor_names
save_3/RestoreV2_4/shape_and_slices
save_3/RestoreV2_4
save_3/Assign_4
save_3/RestoreV2_5/tensor_names
save_3/RestoreV2_5/shape_and_slices
save_3/RestoreV2_5
save_3/Assign_5
save_3/RestoreV2_6/tensor_names
save_3/RestoreV2_6/shape_and_slices
save_3/RestoreV2_6
save_3/Assign_6
save_3/RestoreV2_7/tensor_names
save_3/RestoreV2_7/shape_and_slices
save_3/RestoreV2_7
save_3/Assign_7
save_3/RestoreV2_8/tensor_names
save_3/RestoreV2_8/shape_and_slices
save_3/RestoreV2_8
save_3/Assign_8
save_3/RestoreV2_9/tensor_names
save_3/RestoreV2_9/shape_and_slices
save_3/RestoreV2_9
save_3/Assign_9
save_3/RestoreV2_10/tensor_names
save_3/RestoreV2_10/shape_and_slices
save_3/RestoreV2_10
save_3/Assign_10
save_3/RestoreV2_11/tensor_names
save_3/RestoreV2_11/shape_and_slices
save_3/RestoreV2_11
save_3/Assign_11
save_3/RestoreV2_12/tensor_names
save_3/RestoreV2_12/shape_and_slices
save_3/RestoreV2_12
save_3/Assign_12
save_3/RestoreV2_13/tensor_names
save_3/RestoreV2_13/shape_and_slices
save_3/RestoreV2_13
save_3/Assign_13
save_3/RestoreV2_14/tensor_names
save_3/RestoreV2_14/shape_and_slices
save_3/RestoreV2_14
save_3/Assign_14
save_3/RestoreV2_15/tensor_names
save_3/RestoreV2_15/shape_and_slices
save_3/RestoreV2_15
save_3/Assign_15
save_3/RestoreV2_16/tensor_names
save_3/RestoreV2_16/shape_and_slices
save_3/RestoreV2_16
save_3/Assign_16
save_3/RestoreV2_17/tensor_names
save_3/RestoreV2_17/shape_and_slices
save_3/RestoreV2_17
save_3/Assign_17
save_3/RestoreV2_18/tensor_names
save_3/RestoreV2_18/shape_and_slices
save_3/RestoreV2_18
save_3/Assign_18
save_3/RestoreV2_19/tensor_names
save_3/RestoreV2_19/shape_and_slices
save_3/RestoreV2_19
save_3/Assign_19
save_3/RestoreV2_20/tensor_names
save_3/RestoreV2_20/shape_and_slices
save_3/RestoreV2_20
save_3/Assign_20
save_3/RestoreV2_21/tensor_names
save_3/RestoreV2_21/shape_and_slices
save_3/RestoreV2_21
save_3/Assign_21
save_3/RestoreV2_22/tensor_names
save_3/RestoreV2_22/shape_and_slices
save_3/RestoreV2_22
save_3/Assign_22
save_3/RestoreV2_23/tensor_names
save_3/RestoreV2_23/shape_and_slices
save_3/RestoreV2_23
save_3/Assign_23
save_3/RestoreV2_24/tensor_names
save_3/RestoreV2_24/shape_and_slices
save_3/RestoreV2_24
save_3/Assign_24
save_3/RestoreV2_25/tensor_names
save_3/RestoreV2_25/shape_and_slices
save_3/RestoreV2_25
save_3/Assign_25
save_3/RestoreV2_26/tensor_names
save_3/RestoreV2_26/shape_and_slices
save_3/RestoreV2_26
save_3/Assign_26
save_3/RestoreV2_27/tensor_names
save_3/RestoreV2_27/shape_and_slices
save_3/RestoreV2_27
save_3/Assign_27
save_3/RestoreV2_28/tensor_names
save_3/RestoreV2_28/shape_and_slices
save_3/RestoreV2_28
save_3/Assign_28
save_3/RestoreV2_29/tensor_names
save_3/RestoreV2_29/shape_and_slices
save_3/RestoreV2_29
save_3/Assign_29
save_3/RestoreV2_30/tensor_names
save_3/RestoreV2_30/shape_and_slices
save_3/RestoreV2_30
save_3/Assign_30
save_3/RestoreV2_31/tensor_names
save_3/RestoreV2_31/shape_and_slices
save_3/RestoreV2_31
save_3/Assign_31
save_3/RestoreV2_32/tensor_names
save_3/RestoreV2_32/shape_and_slices
save_3/RestoreV2_32
save_3/Assign_32
save_3/RestoreV2_33/tensor_names
save_3/RestoreV2_33/shape_and_slices
save_3/RestoreV2_33
save_3/Assign_33
save_3/RestoreV2_34/tensor_names
save_3/RestoreV2_34/shape_and_slices
save_3/RestoreV2_34
save_3/Assign_34
save_3/RestoreV2_35/tensor_names
save_3/RestoreV2_35/shape_and_slices
save_3/RestoreV2_35
save_3/Assign_35
save_3/restore_all/NoOp
save_3/restore_all/NoOp_1
save_3/restore_all
last_successful_epoch_4/initial_value
last_successful_epoch_4
last_successful_epoch_4/Assign
last_successful_epoch_4/read
add_4/y
add_4
assign_updated_epoch_4
Inputs_4/Data/X_input
Inputs_4/Train_Labels/y_true_one_hot_label
Inputs_4/Train_Labels/y_true_class
Parameters_4/Weights/zeros
Parameters_4/Weights/Weight
Parameters_4/Weights/Weight/Assign
Parameters_4/Weights/Weight/read
Parameters_4/Bias/ones
Parameters_4/Bias/Bias
Parameters_4/Bias/Bias/Assign
Parameters_4/Bias/Bias/read
Predictions_4/mlp_layer/MatMul
Predictions_4/mlp_layer
Predictions_4/mlp_layer_1
Predictions_4/Softmax
Predictions_4/predict_class/dimension
Predictions_4/predict_class
Predictions_4/one_hot/Const
Predictions_4/one_hot/Const_1
Predictions_4/one_hot/depth
Predictions_4/one_hot/on_value
Predictions_4/one_hot/off_value
Predictions_4/one_hot
Cross_Entropy_4/Rank
Cross_Entropy_4/Shape
Cross_Entropy_4/Rank_1
Cross_Entropy_4/Shape_1
Cross_Entropy_4/Sub/y
Cross_Entropy_4/Sub
Cross_Entropy_4/Slice/begin
Cross_Entropy_4/Slice/size
Cross_Entropy_4/Slice
Cross_Entropy_4/concat/values_0
Cross_Entropy_4/concat/axis
Cross_Entropy_4/concat
Cross_Entropy_4/Reshape
Cross_Entropy_4/Rank_2
Cross_Entropy_4/Shape_2
Cross_Entropy_4/Sub_1/y
Cross_Entropy_4/Sub_1
Cross_Entropy_4/Slice_1/begin
Cross_Entropy_4/Slice_1/size
Cross_Entropy_4/Slice_1
Cross_Entropy_4/concat_1/values_0
Cross_Entropy_4/concat_1/axis
Cross_Entropy_4/concat_1
Cross_Entropy_4/Reshape_1
Cross_Entropy_4/SoftmaxCrossEntropyWithLogits
Cross_Entropy_4/Sub_2/y
Cross_Entropy_4/Sub_2
Cross_Entropy_4/Slice_2/begin
Cross_Entropy_4/Slice_2/size
Cross_Entropy_4/Slice_2
Cross_Entropy_4/Reshape_2
Loss_Function_4/Const
Loss_Function_4/Square
Loss_Function_4/Const_1
Loss_Function_4/Mean
Loss_Function_4/Const_2
Loss_Function_4/Mean_1
Loss_Function_4/Mul
Loss_Function_4/Add
Loss_Function_4/train_loss_list/initial_value
Loss_Function_4/train_loss_list
Loss_Function_4/train_loss_list/Assign
Loss_Function_4/train_loss_list/read
Loss_Function_4/Assign/value
Loss_Function_4/Assign
Loss_Function_4/train_error/tags
Loss_Function_4/train_error
Loss_Function_4/validate_loss_list/initial_value
Loss_Function_4/validate_loss_list
Loss_Function_4/validate_loss_list/Assign
Loss_Function_4/validate_loss_list/read
Loss_Function_4/Assign_1/value
Loss_Function_4/Assign_1
Loss_Function_4/validate_error/tags
Loss_Function_4/validate_error
Optimizer_4/learning_rate_progress/initial_value
Optimizer_4/learning_rate_progress
Optimizer_4/learning_rate_progress/Assign
Optimizer_4/learning_rate_progress/read
Optimizer_4/Assign/value
Optimizer_4/Assign
Optimizer_4/learning_rate/tags
Optimizer_4/learning_rate/values
Optimizer_4/learning_rate
Optimizer_4/gradients/Shape
Optimizer_4/gradients/Const
Optimizer_4/gradients/Fill
Optimizer_4/gradients/Loss_Function_4/Add_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Add_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/BroadcastGradientArgs
Optimizer_4/gradients/Loss_Function_4/Add_grad/Sum
Optimizer_4/gradients/Loss_Function_4/Add_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Add_grad/Sum_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/Reshape_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/group_deps
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/control_dependency
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/control_dependency_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Reshape/shape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Tile
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape_2
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Const
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Prod
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Const_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Prod_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Maximum/y
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Maximum
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/floordiv
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Cast
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/truediv
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/BroadcastGradientArgs
Optimizer_4/gradients/Loss_Function_4/Mul_grad/mul
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Sum
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mul_grad/mul_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Sum_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Reshape_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/group_deps
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/control_dependency
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/control_dependency_1
Optimizer_4/gradients/Cross_Entropy_4/Reshape_2_grad/Shape
Optimizer_4/gradients/Cross_Entropy_4/Reshape_2_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Reshape/shape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Tile/multiples
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Tile
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Const
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Prod
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Const_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Prod_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Maximum/y
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Maximum
Optimizer_4/gradients/Loss_Function_4/Mean_grad/floordiv
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Cast
Optimizer_4/gradients/Loss_Function_4/Mean_grad/truediv
Optimizer_4/gradients/zeros_like
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul/x
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul_1
Optimizer_4/gradients/Cross_Entropy_4/Reshape_grad/Shape
Optimizer_4/gradients/Cross_Entropy_4/Reshape_grad/Reshape
Optimizer_4/gradients/Predictions_4/Softmax_grad/mul
Optimizer_4/gradients/Predictions_4/Softmax_grad/Sum/reduction_indices
Optimizer_4/gradients/Predictions_4/Softmax_grad/Sum
Optimizer_4/gradients/Predictions_4/Softmax_grad/Reshape/shape
Optimizer_4/gradients/Predictions_4/Softmax_grad/Reshape
Optimizer_4/gradients/Predictions_4/Softmax_grad/sub
Optimizer_4/gradients/Predictions_4/Softmax_grad/mul_1
Optimizer_4/gradients/Predictions_4/mlp_layer_1_grad/ReluGrad
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/BiasAddGrad
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/group_deps
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/control_dependency
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/control_dependency_1
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/MatMul
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/MatMul_1
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_4/gradients/AddN
Optimizer_4/GradientDescent/learning_rate
Optimizer_4/GradientDescent/update_Parameters_4/Weights/Weight/ApplyGradientDescent
Optimizer_4/GradientDescent/update_Parameters_4/Bias/Bias/ApplyGradientDescent
Optimizer_4/GradientDescent
Equal_4
Accuracy_4/Cast
Accuracy_4/Const
Accuracy_4/Mean
Accuracy_4/train_accuracy_list/initial_value
Accuracy_4/train_accuracy_list
Accuracy_4/train_accuracy_list/Assign
Accuracy_4/train_accuracy_list/read
Accuracy_4/Assign/value
Accuracy_4/Assign
Accuracy_4/train_accuracy/tags
Accuracy_4/train_accuracy
Accuracy_4/Cast_1
Accuracy_4/Const_1
Accuracy_4/Mean_1
Accuracy_4/validate_accuracy_list/initial_value
Accuracy_4/validate_accuracy_list
Accuracy_4/validate_accuracy_list/Assign
Accuracy_4/validate_accuracy_list/read
Accuracy_4/Assign_1/value
Accuracy_4/Assign_1
Accuracy_4/validate_accuracy/tags
Accuracy_4/validate_accuracy
Accuracy_4/Cast_2
Accuracy_4/Const_2
Accuracy_4/Mean_2
Accuracy_4/test_accuracy_list/initial_value
Accuracy_4/test_accuracy_list
Accuracy_4/test_accuracy_list/Assign
Accuracy_4/test_accuracy_list/read
Accuracy_4/Assign_2/value
Accuracy_4/Assign_2
Accuracy_4/test_accuracy/tags
Accuracy_4/test_accuracy
init_4/NoOp
init_4/NoOp_1
init_4
Merge_4/MergeSummary
save_4/Const
save_4/SaveV2/tensor_names
save_4/SaveV2/shape_and_slices
save_4/SaveV2
save_4/control_dependency
save_4/RestoreV2/tensor_names
save_4/RestoreV2/shape_and_slices
save_4/RestoreV2
save_4/Assign
save_4/RestoreV2_1/tensor_names
save_4/RestoreV2_1/shape_and_slices
save_4/RestoreV2_1
save_4/Assign_1
save_4/RestoreV2_2/tensor_names
save_4/RestoreV2_2/shape_and_slices
save_4/RestoreV2_2
save_4/Assign_2
save_4/RestoreV2_3/tensor_names
save_4/RestoreV2_3/shape_and_slices
save_4/RestoreV2_3
save_4/Assign_3
save_4/RestoreV2_4/tensor_names
save_4/RestoreV2_4/shape_and_slices
save_4/RestoreV2_4
save_4/Assign_4
save_4/RestoreV2_5/tensor_names
save_4/RestoreV2_5/shape_and_slices
save_4/RestoreV2_5
save_4/Assign_5
save_4/RestoreV2_6/tensor_names
save_4/RestoreV2_6/shape_and_slices
save_4/RestoreV2_6
save_4/Assign_6
save_4/RestoreV2_7/tensor_names
save_4/RestoreV2_7/shape_and_slices
save_4/RestoreV2_7
save_4/Assign_7
save_4/RestoreV2_8/tensor_names
save_4/RestoreV2_8/shape_and_slices
save_4/RestoreV2_8
save_4/Assign_8
save_4/RestoreV2_9/tensor_names
save_4/RestoreV2_9/shape_and_slices
save_4/RestoreV2_9
save_4/Assign_9
save_4/RestoreV2_10/tensor_names
save_4/RestoreV2_10/shape_and_slices
save_4/RestoreV2_10
save_4/Assign_10
save_4/RestoreV2_11/tensor_names
save_4/RestoreV2_11/shape_and_slices
save_4/RestoreV2_11
save_4/Assign_11
save_4/RestoreV2_12/tensor_names
save_4/RestoreV2_12/shape_and_slices
save_4/RestoreV2_12
save_4/Assign_12
save_4/RestoreV2_13/tensor_names
save_4/RestoreV2_13/shape_and_slices
save_4/RestoreV2_13
save_4/Assign_13
save_4/RestoreV2_14/tensor_names
save_4/RestoreV2_14/shape_and_slices
save_4/RestoreV2_14
save_4/Assign_14
save_4/RestoreV2_15/tensor_names
save_4/RestoreV2_15/shape_and_slices
save_4/RestoreV2_15
save_4/Assign_15
save_4/RestoreV2_16/tensor_names
save_4/RestoreV2_16/shape_and_slices
save_4/RestoreV2_16
save_4/Assign_16
save_4/RestoreV2_17/tensor_names
save_4/RestoreV2_17/shape_and_slices
save_4/RestoreV2_17
save_4/Assign_17
save_4/RestoreV2_18/tensor_names
save_4/RestoreV2_18/shape_and_slices
save_4/RestoreV2_18
save_4/Assign_18
save_4/RestoreV2_19/tensor_names
save_4/RestoreV2_19/shape_and_slices
save_4/RestoreV2_19
save_4/Assign_19
save_4/RestoreV2_20/tensor_names
save_4/RestoreV2_20/shape_and_slices
save_4/RestoreV2_20
save_4/Assign_20
save_4/RestoreV2_21/tensor_names
save_4/RestoreV2_21/shape_and_slices
save_4/RestoreV2_21
save_4/Assign_21
save_4/RestoreV2_22/tensor_names
save_4/RestoreV2_22/shape_and_slices
save_4/RestoreV2_22
save_4/Assign_22
save_4/RestoreV2_23/tensor_names
save_4/RestoreV2_23/shape_and_slices
save_4/RestoreV2_23
save_4/Assign_23
save_4/RestoreV2_24/tensor_names
save_4/RestoreV2_24/shape_and_slices
save_4/RestoreV2_24
save_4/Assign_24
save_4/RestoreV2_25/tensor_names
save_4/RestoreV2_25/shape_and_slices
save_4/RestoreV2_25
save_4/Assign_25
save_4/RestoreV2_26/tensor_names
save_4/RestoreV2_26/shape_and_slices
save_4/RestoreV2_26
save_4/Assign_26
save_4/RestoreV2_27/tensor_names
save_4/RestoreV2_27/shape_and_slices
save_4/RestoreV2_27
save_4/Assign_27
save_4/RestoreV2_28/tensor_names
save_4/RestoreV2_28/shape_and_slices
save_4/RestoreV2_28
save_4/Assign_28
save_4/RestoreV2_29/tensor_names
save_4/RestoreV2_29/shape_and_slices
save_4/RestoreV2_29
save_4/Assign_29
save_4/RestoreV2_30/tensor_names
save_4/RestoreV2_30/shape_and_slices
save_4/RestoreV2_30
save_4/Assign_30
save_4/RestoreV2_31/tensor_names
save_4/RestoreV2_31/shape_and_slices
save_4/RestoreV2_31
save_4/Assign_31
save_4/RestoreV2_32/tensor_names
save_4/RestoreV2_32/shape_and_slices
save_4/RestoreV2_32
save_4/Assign_32
save_4/RestoreV2_33/tensor_names
save_4/RestoreV2_33/shape_and_slices
save_4/RestoreV2_33
save_4/Assign_33
save_4/RestoreV2_34/tensor_names
save_4/RestoreV2_34/shape_and_slices
save_4/RestoreV2_34
save_4/Assign_34
save_4/RestoreV2_35/tensor_names
save_4/RestoreV2_35/shape_and_slices
save_4/RestoreV2_35
save_4/Assign_35
save_4/RestoreV2_36/tensor_names
save_4/RestoreV2_36/shape_and_slices
save_4/RestoreV2_36
save_4/Assign_36
save_4/RestoreV2_37/tensor_names
save_4/RestoreV2_37/shape_and_slices
save_4/RestoreV2_37
save_4/Assign_37
save_4/RestoreV2_38/tensor_names
save_4/RestoreV2_38/shape_and_slices
save_4/RestoreV2_38
save_4/Assign_38
save_4/RestoreV2_39/tensor_names
save_4/RestoreV2_39/shape_and_slices
save_4/RestoreV2_39
save_4/Assign_39
save_4/RestoreV2_40/tensor_names
save_4/RestoreV2_40/shape_and_slices
save_4/RestoreV2_40
save_4/Assign_40
save_4/RestoreV2_41/tensor_names
save_4/RestoreV2_41/shape_and_slices
save_4/RestoreV2_41
save_4/Assign_41
save_4/RestoreV2_42/tensor_names
save_4/RestoreV2_42/shape_and_slices
save_4/RestoreV2_42
save_4/Assign_42
save_4/RestoreV2_43/tensor_names
save_4/RestoreV2_43/shape_and_slices
save_4/RestoreV2_43
save_4/Assign_43
save_4/RestoreV2_44/tensor_names
save_4/RestoreV2_44/shape_and_slices
save_4/RestoreV2_44
save_4/Assign_44
save_4/restore_all/NoOp
save_4/restore_all/NoOp_1
save_4/restore_all
last_successful_epoch_5/initial_value
last_successful_epoch_5
last_successful_epoch_5/Assign
last_successful_epoch_5/read
add_5/y
add_5
assign_updated_epoch_5
Inputs_5/Data/X_input
Inputs_5/Train_Labels/y_true_one_hot_label
Inputs_5/Train_Labels/y_true_class
Parameters_5/Weights/zeros
Parameters_5/Weights/Weight
Parameters_5/Weights/Weight/Assign
Parameters_5/Weights/Weight/read
Parameters_5/Bias/ones
Parameters_5/Bias/Bias
Parameters_5/Bias/Bias/Assign
Parameters_5/Bias/Bias/read
Predictions_5/mlp_layer/MatMul
Predictions_5/mlp_layer
Predictions_5/mlp_layer_1
Predictions_5/Softmax
Predictions_5/predict_class/dimension
Predictions_5/predict_class
Predictions_5/one_hot/Const
Predictions_5/one_hot/Const_1
Predictions_5/one_hot/depth
Predictions_5/one_hot/on_value
Predictions_5/one_hot/off_value
Predictions_5/one_hot
Cross_Entropy_5/Rank
Cross_Entropy_5/Shape
Cross_Entropy_5/Rank_1
Cross_Entropy_5/Shape_1
Cross_Entropy_5/Sub/y
Cross_Entropy_5/Sub
Cross_Entropy_5/Slice/begin
Cross_Entropy_5/Slice/size
Cross_Entropy_5/Slice
Cross_Entropy_5/concat/values_0
Cross_Entropy_5/concat/axis
Cross_Entropy_5/concat
Cross_Entropy_5/Reshape
Cross_Entropy_5/Rank_2
Cross_Entropy_5/Shape_2
Cross_Entropy_5/Sub_1/y
Cross_Entropy_5/Sub_1
Cross_Entropy_5/Slice_1/begin
Cross_Entropy_5/Slice_1/size
Cross_Entropy_5/Slice_1
Cross_Entropy_5/concat_1/values_0
Cross_Entropy_5/concat_1/axis
Cross_Entropy_5/concat_1
Cross_Entropy_5/Reshape_1
Cross_Entropy_5/SoftmaxCrossEntropyWithLogits
Cross_Entropy_5/Sub_2/y
Cross_Entropy_5/Sub_2
Cross_Entropy_5/Slice_2/begin
Cross_Entropy_5/Slice_2/size
Cross_Entropy_5/Slice_2
Cross_Entropy_5/Reshape_2
Loss_Function_5/Const
Loss_Function_5/Square
Loss_Function_5/Const_1
Loss_Function_5/Mean
Loss_Function_5/Const_2
Loss_Function_5/Mean_1
Loss_Function_5/Mul
Loss_Function_5/Add
Loss_Function_5/train_loss_list/initial_value
Loss_Function_5/train_loss_list
Loss_Function_5/train_loss_list/Assign
Loss_Function_5/train_loss_list/read
Loss_Function_5/Assign/value
Loss_Function_5/Assign
Loss_Function_5/train_error/tags
Loss_Function_5/train_error
Loss_Function_5/validate_loss_list/initial_value
Loss_Function_5/validate_loss_list
Loss_Function_5/validate_loss_list/Assign
Loss_Function_5/validate_loss_list/read
Loss_Function_5/Assign_1/value
Loss_Function_5/Assign_1
Loss_Function_5/validate_error/tags
Loss_Function_5/validate_error
Optimizer_5/learning_rate_progress/initial_value
Optimizer_5/learning_rate_progress
Optimizer_5/learning_rate_progress/Assign
Optimizer_5/learning_rate_progress/read
Optimizer_5/Assign/value
Optimizer_5/Assign
Optimizer_5/learning_rate/tags
Optimizer_5/learning_rate/values
Optimizer_5/learning_rate
Optimizer_5/gradients/Shape
Optimizer_5/gradients/Const
Optimizer_5/gradients/Fill
Optimizer_5/gradients/Loss_Function_5/Add_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Add_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/BroadcastGradientArgs
Optimizer_5/gradients/Loss_Function_5/Add_grad/Sum
Optimizer_5/gradients/Loss_Function_5/Add_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Add_grad/Sum_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/Reshape_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/group_deps
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/control_dependency
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/control_dependency_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Reshape/shape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Tile
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape_2
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Const
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Prod
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Const_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Prod_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Maximum/y
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Maximum
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/floordiv
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Cast
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/truediv
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/BroadcastGradientArgs
Optimizer_5/gradients/Loss_Function_5/Mul_grad/mul
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Sum
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mul_grad/mul_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Sum_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Reshape_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/group_deps
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/control_dependency
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/control_dependency_1
Optimizer_5/gradients/Cross_Entropy_5/Reshape_2_grad/Shape
Optimizer_5/gradients/Cross_Entropy_5/Reshape_2_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Reshape/shape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Tile/multiples
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Tile
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Const
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Prod
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Const_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Prod_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Maximum/y
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Maximum
Optimizer_5/gradients/Loss_Function_5/Mean_grad/floordiv
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Cast
Optimizer_5/gradients/Loss_Function_5/Mean_grad/truediv
Optimizer_5/gradients/zeros_like
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul/x
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul_1
Optimizer_5/gradients/Cross_Entropy_5/Reshape_grad/Shape
Optimizer_5/gradients/Cross_Entropy_5/Reshape_grad/Reshape
Optimizer_5/gradients/Predictions_5/Softmax_grad/mul
Optimizer_5/gradients/Predictions_5/Softmax_grad/Sum/reduction_indices
Optimizer_5/gradients/Predictions_5/Softmax_grad/Sum
Optimizer_5/gradients/Predictions_5/Softmax_grad/Reshape/shape
Optimizer_5/gradients/Predictions_5/Softmax_grad/Reshape
Optimizer_5/gradients/Predictions_5/Softmax_grad/sub
Optimizer_5/gradients/Predictions_5/Softmax_grad/mul_1
Optimizer_5/gradients/Predictions_5/mlp_layer_1_grad/ReluGrad
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/BiasAddGrad
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/group_deps
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/control_dependency
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/control_dependency_1
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/MatMul
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/MatMul_1
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_5/gradients/AddN
Optimizer_5/GradientDescent/learning_rate
Optimizer_5/GradientDescent/update_Parameters_5/Weights/Weight/ApplyGradientDescent
Optimizer_5/GradientDescent/update_Parameters_5/Bias/Bias/ApplyGradientDescent
Optimizer_5/GradientDescent
Equal_5
Accuracy_5/Cast
Accuracy_5/Const
Accuracy_5/Mean
Accuracy_5/train_accuracy_list/initial_value
Accuracy_5/train_accuracy_list
Accuracy_5/train_accuracy_list/Assign
Accuracy_5/train_accuracy_list/read
Accuracy_5/Assign/value
Accuracy_5/Assign
Accuracy_5/train_accuracy/tags
Accuracy_5/train_accuracy
Accuracy_5/Cast_1
Accuracy_5/Const_1
Accuracy_5/Mean_1
Accuracy_5/validate_accuracy_list/initial_value
Accuracy_5/validate_accuracy_list
Accuracy_5/validate_accuracy_list/Assign
Accuracy_5/validate_accuracy_list/read
Accuracy_5/Assign_1/value
Accuracy_5/Assign_1
Accuracy_5/validate_accuracy/tags
Accuracy_5/validate_accuracy
Accuracy_5/Cast_2
Accuracy_5/Const_2
Accuracy_5/Mean_2
Accuracy_5/test_accuracy_list/initial_value
Accuracy_5/test_accuracy_list
Accuracy_5/test_accuracy_list/Assign
Accuracy_5/test_accuracy_list/read
Accuracy_5/Assign_2/value
Accuracy_5/Assign_2
Accuracy_5/test_accuracy/tags
Accuracy_5/test_accuracy
init_5/NoOp
init_5/NoOp_1
init_5
Merge_5/MergeSummary
save_5/Const
save_5/SaveV2/tensor_names
save_5/SaveV2/shape_and_slices
save_5/SaveV2
save_5/control_dependency
save_5/RestoreV2/tensor_names
save_5/RestoreV2/shape_and_slices
save_5/RestoreV2
save_5/Assign
save_5/RestoreV2_1/tensor_names
save_5/RestoreV2_1/shape_and_slices
save_5/RestoreV2_1
save_5/Assign_1
save_5/RestoreV2_2/tensor_names
save_5/RestoreV2_2/shape_and_slices
save_5/RestoreV2_2
save_5/Assign_2
save_5/RestoreV2_3/tensor_names
save_5/RestoreV2_3/shape_and_slices
save_5/RestoreV2_3
save_5/Assign_3
save_5/RestoreV2_4/tensor_names
save_5/RestoreV2_4/shape_and_slices
save_5/RestoreV2_4
save_5/Assign_4
save_5/RestoreV2_5/tensor_names
save_5/RestoreV2_5/shape_and_slices
save_5/RestoreV2_5
save_5/Assign_5
save_5/RestoreV2_6/tensor_names
save_5/RestoreV2_6/shape_and_slices
save_5/RestoreV2_6
save_5/Assign_6
save_5/RestoreV2_7/tensor_names
save_5/RestoreV2_7/shape_and_slices
save_5/RestoreV2_7
save_5/Assign_7
save_5/RestoreV2_8/tensor_names
save_5/RestoreV2_8/shape_and_slices
save_5/RestoreV2_8
save_5/Assign_8
save_5/RestoreV2_9/tensor_names
save_5/RestoreV2_9/shape_and_slices
save_5/RestoreV2_9
save_5/Assign_9
save_5/RestoreV2_10/tensor_names
save_5/RestoreV2_10/shape_and_slices
save_5/RestoreV2_10
save_5/Assign_10
save_5/RestoreV2_11/tensor_names
save_5/RestoreV2_11/shape_and_slices
save_5/RestoreV2_11
save_5/Assign_11
save_5/RestoreV2_12/tensor_names
save_5/RestoreV2_12/shape_and_slices
save_5/RestoreV2_12
save_5/Assign_12
save_5/RestoreV2_13/tensor_names
save_5/RestoreV2_13/shape_and_slices
save_5/RestoreV2_13
save_5/Assign_13
save_5/RestoreV2_14/tensor_names
save_5/RestoreV2_14/shape_and_slices
save_5/RestoreV2_14
save_5/Assign_14
save_5/RestoreV2_15/tensor_names
save_5/RestoreV2_15/shape_and_slices
save_5/RestoreV2_15
save_5/Assign_15
save_5/RestoreV2_16/tensor_names
save_5/RestoreV2_16/shape_and_slices
save_5/RestoreV2_16
save_5/Assign_16
save_5/RestoreV2_17/tensor_names
save_5/RestoreV2_17/shape_and_slices
save_5/RestoreV2_17
save_5/Assign_17
save_5/RestoreV2_18/tensor_names
save_5/RestoreV2_18/shape_and_slices
save_5/RestoreV2_18
save_5/Assign_18
save_5/RestoreV2_19/tensor_names
save_5/RestoreV2_19/shape_and_slices
save_5/RestoreV2_19
save_5/Assign_19
save_5/RestoreV2_20/tensor_names
save_5/RestoreV2_20/shape_and_slices
save_5/RestoreV2_20
save_5/Assign_20
save_5/RestoreV2_21/tensor_names
save_5/RestoreV2_21/shape_and_slices
save_5/RestoreV2_21
save_5/Assign_21
save_5/RestoreV2_22/tensor_names
save_5/RestoreV2_22/shape_and_slices
save_5/RestoreV2_22
save_5/Assign_22
save_5/RestoreV2_23/tensor_names
save_5/RestoreV2_23/shape_and_slices
save_5/RestoreV2_23
save_5/Assign_23
save_5/RestoreV2_24/tensor_names
save_5/RestoreV2_24/shape_and_slices
save_5/RestoreV2_24
save_5/Assign_24
save_5/RestoreV2_25/tensor_names
save_5/RestoreV2_25/shape_and_slices
save_5/RestoreV2_25
save_5/Assign_25
save_5/RestoreV2_26/tensor_names
save_5/RestoreV2_26/shape_and_slices
save_5/RestoreV2_26
save_5/Assign_26
save_5/RestoreV2_27/tensor_names
save_5/RestoreV2_27/shape_and_slices
save_5/RestoreV2_27
save_5/Assign_27
save_5/RestoreV2_28/tensor_names
save_5/RestoreV2_28/shape_and_slices
save_5/RestoreV2_28
save_5/Assign_28
save_5/RestoreV2_29/tensor_names
save_5/RestoreV2_29/shape_and_slices
save_5/RestoreV2_29
save_5/Assign_29
save_5/RestoreV2_30/tensor_names
save_5/RestoreV2_30/shape_and_slices
save_5/RestoreV2_30
save_5/Assign_30
save_5/RestoreV2_31/tensor_names
save_5/RestoreV2_31/shape_and_slices
save_5/RestoreV2_31
save_5/Assign_31
save_5/RestoreV2_32/tensor_names
save_5/RestoreV2_32/shape_and_slices
save_5/RestoreV2_32
save_5/Assign_32
save_5/RestoreV2_33/tensor_names
save_5/RestoreV2_33/shape_and_slices
save_5/RestoreV2_33
save_5/Assign_33
save_5/RestoreV2_34/tensor_names
save_5/RestoreV2_34/shape_and_slices
save_5/RestoreV2_34
save_5/Assign_34
save_5/RestoreV2_35/tensor_names
save_5/RestoreV2_35/shape_and_slices
save_5/RestoreV2_35
save_5/Assign_35
save_5/RestoreV2_36/tensor_names
save_5/RestoreV2_36/shape_and_slices
save_5/RestoreV2_36
save_5/Assign_36
save_5/RestoreV2_37/tensor_names
save_5/RestoreV2_37/shape_and_slices
save_5/RestoreV2_37
save_5/Assign_37
save_5/RestoreV2_38/tensor_names
save_5/RestoreV2_38/shape_and_slices
save_5/RestoreV2_38
save_5/Assign_38
save_5/RestoreV2_39/tensor_names
save_5/RestoreV2_39/shape_and_slices
save_5/RestoreV2_39
save_5/Assign_39
save_5/RestoreV2_40/tensor_names
save_5/RestoreV2_40/shape_and_slices
save_5/RestoreV2_40
save_5/Assign_40
save_5/RestoreV2_41/tensor_names
save_5/RestoreV2_41/shape_and_slices
save_5/RestoreV2_41
save_5/Assign_41
save_5/RestoreV2_42/tensor_names
save_5/RestoreV2_42/shape_and_slices
save_5/RestoreV2_42
save_5/Assign_42
save_5/RestoreV2_43/tensor_names
save_5/RestoreV2_43/shape_and_slices
save_5/RestoreV2_43
save_5/Assign_43
save_5/RestoreV2_44/tensor_names
save_5/RestoreV2_44/shape_and_slices
save_5/RestoreV2_44
save_5/Assign_44
save_5/RestoreV2_45/tensor_names
save_5/RestoreV2_45/shape_and_slices
save_5/RestoreV2_45
save_5/Assign_45
save_5/RestoreV2_46/tensor_names
save_5/RestoreV2_46/shape_and_slices
save_5/RestoreV2_46
save_5/Assign_46
save_5/RestoreV2_47/tensor_names
save_5/RestoreV2_47/shape_and_slices
save_5/RestoreV2_47
save_5/Assign_47
save_5/RestoreV2_48/tensor_names
save_5/RestoreV2_48/shape_and_slices
save_5/RestoreV2_48
save_5/Assign_48
save_5/RestoreV2_49/tensor_names
save_5/RestoreV2_49/shape_and_slices
save_5/RestoreV2_49
save_5/Assign_49
save_5/RestoreV2_50/tensor_names
save_5/RestoreV2_50/shape_and_slices
save_5/RestoreV2_50
save_5/Assign_50
save_5/RestoreV2_51/tensor_names
save_5/RestoreV2_51/shape_and_slices
save_5/RestoreV2_51
save_5/Assign_51
save_5/RestoreV2_52/tensor_names
save_5/RestoreV2_52/shape_and_slices
save_5/RestoreV2_52
save_5/Assign_52
save_5/RestoreV2_53/tensor_names
save_5/RestoreV2_53/shape_and_slices
save_5/RestoreV2_53
save_5/Assign_53
save_5/restore_all/NoOp
save_5/restore_all/NoOp_1
save_5/restore_all
Output Logits  : Tensor("Predictions_1/Softmax:0", shape=(?, 10), dtype=float32, device=/device:GPU:0)
Entropy        : Tensor("Cross_Entropy_1/Reshape_2:0", shape=(?,), dtype=float32, device=/device:GPU:0)
Predictions    : Tensor("Equal_1:0", shape=(?,), dtype=bool, device=/device:GPU:0) 
Optimizer      : name: "Optimizer_1/GradientDescent"
op: "NoOp"
input: "^Optimizer_1/GradientDescent/update_Parameters_1/Weights/Weight/ApplyGradientDescent"
input: "^Optimizer_1/GradientDescent/update_Parameters_1/Bias/Bias/ApplyGradientDescent"
device: "/device:GPU:0"
 
>> Model info
Train loss     : Tensor("Loss_Function_1/Add:0", shape=(), dtype=float32, device=/device:GPU:0)
Train accuracy : Tensor("Accuracy_1/Mean:0", shape=(), dtype=float32, device=/device:GPU:0)
Val. loss      : Tensor("Loss_Function_1/Add:0", shape=(), dtype=float32, device=/device:GPU:0)
Val. accuracy  : Tensor("Accuracy_1/Mean_1:0", shape=(), dtype=float32, device=/device:GPU:0)
>> Summary params
Train loss     : Tensor("Loss_Function_1/train_error:0", shape=(), dtype=string, device=/device:GPU:0)
Train accuracy : Tensor("Accuracy_1/train_accuracy:0", shape=(), dtype=string, device=/device:GPU:0)
Val. loss      : Tensor("Loss_Function_1/validate_error:0", shape=(), dtype=string, device=/device:GPU:0)
Val. accuracy  : Tensor("Accuracy_1/validate_accuracy:0", shape=(), dtype=string, device=/device:GPU:0)
Parameters for Linear classifier
>> Input Parameters
Input                  : Tensor("Inputs_2/Data/X_input:0", shape=(?, 3072), dtype=float32, device=/device:GPU:0) 
True one hot labels    : Tensor("Inputs_2/Train_Labels/y_true_one_hot_label:0", shape=(?, 10), dtype=float32, device=/device:GPU:0) 
True class             : Tensor("Inputs_2/Train_Labels/y_true_class:0", shape=(?,), dtype=int64, device=/device:GPU:0) 
Predict one hot labels : Tensor("Predictions_2/one_hot:0", shape=(?, 10), dtype=float32, device=/device:GPU:0) 
Predict class          : Tensor("Predictions_2/predict_class:0", shape=(?,), dtype=int64, device=/device:GPU:0) 
>> Model params
last_successful_epoch/initial_value
last_successful_epoch
last_successful_epoch/Assign
last_successful_epoch/read
add/y
add
assign_updated_epoch
Inputs/Data/X_input
Inputs/Train_Labels/y_true_one_hot_label
Inputs/Train_Labels/y_true_class
Parameters/Weights/zeros
Parameters/Weights/Weight
Parameters/Weights/Weight/Assign
Parameters/Weights/Weight/read
Parameters/Bias/ones
Parameters/Bias/Bias
Parameters/Bias/Bias/Assign
Parameters/Bias/Bias/read
Predictions/mlp_layer/MatMul
Predictions/mlp_layer
Predictions/mlp_layer_1
Predictions/Softmax
Predictions/predict_class/dimension
Predictions/predict_class
Predictions/one_hot/Const
Predictions/one_hot/Const_1
Predictions/one_hot/depth
Predictions/one_hot/on_value
Predictions/one_hot/off_value
Predictions/one_hot
Cross_Entropy/Rank
Cross_Entropy/Shape
Cross_Entropy/Rank_1
Cross_Entropy/Shape_1
Cross_Entropy/Sub/y
Cross_Entropy/Sub
Cross_Entropy/Slice/begin
Cross_Entropy/Slice/size
Cross_Entropy/Slice
Cross_Entropy/concat/values_0
Cross_Entropy/concat/axis
Cross_Entropy/concat
Cross_Entropy/Reshape
Cross_Entropy/Rank_2
Cross_Entropy/Shape_2
Cross_Entropy/Sub_1/y
Cross_Entropy/Sub_1
Cross_Entropy/Slice_1/begin
Cross_Entropy/Slice_1/size
Cross_Entropy/Slice_1
Cross_Entropy/concat_1/values_0
Cross_Entropy/concat_1/axis
Cross_Entropy/concat_1
Cross_Entropy/Reshape_1
Cross_Entropy/SoftmaxCrossEntropyWithLogits
Cross_Entropy/Sub_2/y
Cross_Entropy/Sub_2
Cross_Entropy/Slice_2/begin
Cross_Entropy/Slice_2/size
Cross_Entropy/Slice_2
Cross_Entropy/Reshape_2
Loss_Function/Const
Loss_Function/Square
Loss_Function/Const_1
Loss_Function/Mean
Loss_Function/Const_2
Loss_Function/Mean_1
Loss_Function/Mul
Loss_Function/Add
Loss_Function/train_loss_list/initial_value
Loss_Function/train_loss_list
Loss_Function/train_loss_list/Assign
Loss_Function/train_loss_list/read
Loss_Function/Assign/value
Loss_Function/Assign
Loss_Function/train_error/tags
Loss_Function/train_error
Loss_Function/validate_loss_list/initial_value
Loss_Function/validate_loss_list
Loss_Function/validate_loss_list/Assign
Loss_Function/validate_loss_list/read
Loss_Function/Assign_1/value
Loss_Function/Assign_1
Loss_Function/validate_error/tags
Loss_Function/validate_error
Optimizer/learning_rate_progress/initial_value
Optimizer/learning_rate_progress
Optimizer/learning_rate_progress/Assign
Optimizer/learning_rate_progress/read
Optimizer/Assign/value
Optimizer/Assign
Optimizer/learning_rate/tags
Optimizer/learning_rate/values
Optimizer/learning_rate
Optimizer/gradients/Shape
Optimizer/gradients/Const
Optimizer/gradients/Fill
Optimizer/gradients/Loss_Function/Add_grad/Shape
Optimizer/gradients/Loss_Function/Add_grad/Shape_1
Optimizer/gradients/Loss_Function/Add_grad/BroadcastGradientArgs
Optimizer/gradients/Loss_Function/Add_grad/Sum
Optimizer/gradients/Loss_Function/Add_grad/Reshape
Optimizer/gradients/Loss_Function/Add_grad/Sum_1
Optimizer/gradients/Loss_Function/Add_grad/Reshape_1
Optimizer/gradients/Loss_Function/Add_grad/tuple/group_deps
Optimizer/gradients/Loss_Function/Add_grad/tuple/control_dependency
Optimizer/gradients/Loss_Function/Add_grad/tuple/control_dependency_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Reshape/shape
Optimizer/gradients/Loss_Function/Mean_1_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape
Optimizer/gradients/Loss_Function/Mean_1_grad/Tile
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape_2
Optimizer/gradients/Loss_Function/Mean_1_grad/Const
Optimizer/gradients/Loss_Function/Mean_1_grad/Prod
Optimizer/gradients/Loss_Function/Mean_1_grad/Const_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Prod_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Maximum/y
Optimizer/gradients/Loss_Function/Mean_1_grad/Maximum
Optimizer/gradients/Loss_Function/Mean_1_grad/floordiv
Optimizer/gradients/Loss_Function/Mean_1_grad/Cast
Optimizer/gradients/Loss_Function/Mean_1_grad/truediv
Optimizer/gradients/Loss_Function/Mul_grad/Shape
Optimizer/gradients/Loss_Function/Mul_grad/Shape_1
Optimizer/gradients/Loss_Function/Mul_grad/BroadcastGradientArgs
Optimizer/gradients/Loss_Function/Mul_grad/mul
Optimizer/gradients/Loss_Function/Mul_grad/Sum
Optimizer/gradients/Loss_Function/Mul_grad/Reshape
Optimizer/gradients/Loss_Function/Mul_grad/mul_1
Optimizer/gradients/Loss_Function/Mul_grad/Sum_1
Optimizer/gradients/Loss_Function/Mul_grad/Reshape_1
Optimizer/gradients/Loss_Function/Mul_grad/tuple/group_deps
Optimizer/gradients/Loss_Function/Mul_grad/tuple/control_dependency
Optimizer/gradients/Loss_Function/Mul_grad/tuple/control_dependency_1
Optimizer/gradients/Cross_Entropy/Reshape_2_grad/Shape
Optimizer/gradients/Cross_Entropy/Reshape_2_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_grad/Reshape/shape
Optimizer/gradients/Loss_Function/Mean_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_grad/Tile/multiples
Optimizer/gradients/Loss_Function/Mean_grad/Tile
Optimizer/gradients/Loss_Function/Mean_grad/Shape
Optimizer/gradients/Loss_Function/Mean_grad/Shape_1
Optimizer/gradients/Loss_Function/Mean_grad/Const
Optimizer/gradients/Loss_Function/Mean_grad/Prod
Optimizer/gradients/Loss_Function/Mean_grad/Const_1
Optimizer/gradients/Loss_Function/Mean_grad/Prod_1
Optimizer/gradients/Loss_Function/Mean_grad/Maximum/y
Optimizer/gradients/Loss_Function/Mean_grad/Maximum
Optimizer/gradients/Loss_Function/Mean_grad/floordiv
Optimizer/gradients/Loss_Function/Mean_grad/Cast
Optimizer/gradients/Loss_Function/Mean_grad/truediv
Optimizer/gradients/zeros_like
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer/gradients/Loss_Function/Square_grad/mul/x
Optimizer/gradients/Loss_Function/Square_grad/mul
Optimizer/gradients/Loss_Function/Square_grad/mul_1
Optimizer/gradients/Cross_Entropy/Reshape_grad/Shape
Optimizer/gradients/Cross_Entropy/Reshape_grad/Reshape
Optimizer/gradients/Predictions/Softmax_grad/mul
Optimizer/gradients/Predictions/Softmax_grad/Sum/reduction_indices
Optimizer/gradients/Predictions/Softmax_grad/Sum
Optimizer/gradients/Predictions/Softmax_grad/Reshape/shape
Optimizer/gradients/Predictions/Softmax_grad/Reshape
Optimizer/gradients/Predictions/Softmax_grad/sub
Optimizer/gradients/Predictions/Softmax_grad/mul_1
Optimizer/gradients/Predictions/mlp_layer_1_grad/ReluGrad
Optimizer/gradients/Predictions/mlp_layer_grad/BiasAddGrad
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/group_deps
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/control_dependency
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/control_dependency_1
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/MatMul
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/MatMul_1
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer/gradients/AddN
Optimizer/GradientDescent/learning_rate
Optimizer/GradientDescent/update_Parameters/Weights/Weight/ApplyGradientDescent
Optimizer/GradientDescent/update_Parameters/Bias/Bias/ApplyGradientDescent
Optimizer/GradientDescent
Equal
Accuracy/Cast
Accuracy/Const
Accuracy/Mean
Accuracy/train_accuracy_list/initial_value
Accuracy/train_accuracy_list
Accuracy/train_accuracy_list/Assign
Accuracy/train_accuracy_list/read
Accuracy/Assign/value
Accuracy/Assign
Accuracy/train_accuracy/tags
Accuracy/train_accuracy
Accuracy/Cast_1
Accuracy/Const_1
Accuracy/Mean_1
Accuracy/validate_accuracy_list/initial_value
Accuracy/validate_accuracy_list
Accuracy/validate_accuracy_list/Assign
Accuracy/validate_accuracy_list/read
Accuracy/Assign_1/value
Accuracy/Assign_1
Accuracy/validate_accuracy/tags
Accuracy/validate_accuracy
Accuracy/Cast_2
Accuracy/Const_2
Accuracy/Mean_2
Accuracy/test_accuracy_list/initial_value
Accuracy/test_accuracy_list
Accuracy/test_accuracy_list/Assign
Accuracy/test_accuracy_list/read
Accuracy/Assign_2/value
Accuracy/Assign_2
Accuracy/test_accuracy/tags
Accuracy/test_accuracy
init/NoOp
init/NoOp_1
init
Merge/MergeSummary
save/Const
save/SaveV2/tensor_names
save/SaveV2/shape_and_slices
save/SaveV2
save/control_dependency
save/RestoreV2/tensor_names
save/RestoreV2/shape_and_slices
save/RestoreV2
save/Assign
save/RestoreV2_1/tensor_names
save/RestoreV2_1/shape_and_slices
save/RestoreV2_1
save/Assign_1
save/RestoreV2_2/tensor_names
save/RestoreV2_2/shape_and_slices
save/RestoreV2_2
save/Assign_2
save/RestoreV2_3/tensor_names
save/RestoreV2_3/shape_and_slices
save/RestoreV2_3
save/Assign_3
save/RestoreV2_4/tensor_names
save/RestoreV2_4/shape_and_slices
save/RestoreV2_4
save/Assign_4
save/RestoreV2_5/tensor_names
save/RestoreV2_5/shape_and_slices
save/RestoreV2_5
save/Assign_5
save/RestoreV2_6/tensor_names
save/RestoreV2_6/shape_and_slices
save/RestoreV2_6
save/Assign_6
save/RestoreV2_7/tensor_names
save/RestoreV2_7/shape_and_slices
save/RestoreV2_7
save/Assign_7
save/RestoreV2_8/tensor_names
save/RestoreV2_8/shape_and_slices
save/RestoreV2_8
save/Assign_8
save/restore_all/NoOp
save/restore_all/NoOp_1
save/restore_all
last_successful_epoch_1/initial_value
last_successful_epoch_1
last_successful_epoch_1/Assign
last_successful_epoch_1/read
add_1/y
add_1
assign_updated_epoch_1
Inputs_1/Data/X_input
Inputs_1/Train_Labels/y_true_one_hot_label
Inputs_1/Train_Labels/y_true_class
Parameters_1/Weights/zeros
Parameters_1/Weights/Weight
Parameters_1/Weights/Weight/Assign
Parameters_1/Weights/Weight/read
Parameters_1/Bias/ones
Parameters_1/Bias/Bias
Parameters_1/Bias/Bias/Assign
Parameters_1/Bias/Bias/read
Predictions_1/mlp_layer/MatMul
Predictions_1/mlp_layer
Predictions_1/mlp_layer_1
Predictions_1/Softmax
Predictions_1/predict_class/dimension
Predictions_1/predict_class
Predictions_1/one_hot/Const
Predictions_1/one_hot/Const_1
Predictions_1/one_hot/depth
Predictions_1/one_hot/on_value
Predictions_1/one_hot/off_value
Predictions_1/one_hot
Cross_Entropy_1/Rank
Cross_Entropy_1/Shape
Cross_Entropy_1/Rank_1
Cross_Entropy_1/Shape_1
Cross_Entropy_1/Sub/y
Cross_Entropy_1/Sub
Cross_Entropy_1/Slice/begin
Cross_Entropy_1/Slice/size
Cross_Entropy_1/Slice
Cross_Entropy_1/concat/values_0
Cross_Entropy_1/concat/axis
Cross_Entropy_1/concat
Cross_Entropy_1/Reshape
Cross_Entropy_1/Rank_2
Cross_Entropy_1/Shape_2
Cross_Entropy_1/Sub_1/y
Cross_Entropy_1/Sub_1
Cross_Entropy_1/Slice_1/begin
Cross_Entropy_1/Slice_1/size
Cross_Entropy_1/Slice_1
Cross_Entropy_1/concat_1/values_0
Cross_Entropy_1/concat_1/axis
Cross_Entropy_1/concat_1
Cross_Entropy_1/Reshape_1
Cross_Entropy_1/SoftmaxCrossEntropyWithLogits
Cross_Entropy_1/Sub_2/y
Cross_Entropy_1/Sub_2
Cross_Entropy_1/Slice_2/begin
Cross_Entropy_1/Slice_2/size
Cross_Entropy_1/Slice_2
Cross_Entropy_1/Reshape_2
Loss_Function_1/Const
Loss_Function_1/Square
Loss_Function_1/Const_1
Loss_Function_1/Mean
Loss_Function_1/Const_2
Loss_Function_1/Mean_1
Loss_Function_1/Mul
Loss_Function_1/Add
Loss_Function_1/train_loss_list/initial_value
Loss_Function_1/train_loss_list
Loss_Function_1/train_loss_list/Assign
Loss_Function_1/train_loss_list/read
Loss_Function_1/Assign/value
Loss_Function_1/Assign
Loss_Function_1/train_error/tags
Loss_Function_1/train_error
Loss_Function_1/validate_loss_list/initial_value
Loss_Function_1/validate_loss_list
Loss_Function_1/validate_loss_list/Assign
Loss_Function_1/validate_loss_list/read
Loss_Function_1/Assign_1/value
Loss_Function_1/Assign_1
Loss_Function_1/validate_error/tags
Loss_Function_1/validate_error
Optimizer_1/learning_rate_progress/initial_value
Optimizer_1/learning_rate_progress
Optimizer_1/learning_rate_progress/Assign
Optimizer_1/learning_rate_progress/read
Optimizer_1/Assign/value
Optimizer_1/Assign
Optimizer_1/learning_rate/tags
Optimizer_1/learning_rate/values
Optimizer_1/learning_rate
Optimizer_1/gradients/Shape
Optimizer_1/gradients/Const
Optimizer_1/gradients/Fill
Optimizer_1/gradients/Loss_Function_1/Add_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Add_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/BroadcastGradientArgs
Optimizer_1/gradients/Loss_Function_1/Add_grad/Sum
Optimizer_1/gradients/Loss_Function_1/Add_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Add_grad/Sum_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/Reshape_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/group_deps
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/control_dependency
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/control_dependency_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Reshape/shape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Tile
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape_2
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Const
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Prod
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Const_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Prod_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Maximum/y
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Maximum
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/floordiv
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Cast
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/truediv
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/BroadcastGradientArgs
Optimizer_1/gradients/Loss_Function_1/Mul_grad/mul
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Sum
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mul_grad/mul_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Sum_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Reshape_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/group_deps
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/control_dependency
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/control_dependency_1
Optimizer_1/gradients/Cross_Entropy_1/Reshape_2_grad/Shape
Optimizer_1/gradients/Cross_Entropy_1/Reshape_2_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Reshape/shape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Tile/multiples
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Tile
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Const
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Prod
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Const_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Prod_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Maximum/y
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Maximum
Optimizer_1/gradients/Loss_Function_1/Mean_grad/floordiv
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Cast
Optimizer_1/gradients/Loss_Function_1/Mean_grad/truediv
Optimizer_1/gradients/zeros_like
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul/x
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul_1
Optimizer_1/gradients/Cross_Entropy_1/Reshape_grad/Shape
Optimizer_1/gradients/Cross_Entropy_1/Reshape_grad/Reshape
Optimizer_1/gradients/Predictions_1/Softmax_grad/mul
Optimizer_1/gradients/Predictions_1/Softmax_grad/Sum/reduction_indices
Optimizer_1/gradients/Predictions_1/Softmax_grad/Sum
Optimizer_1/gradients/Predictions_1/Softmax_grad/Reshape/shape
Optimizer_1/gradients/Predictions_1/Softmax_grad/Reshape
Optimizer_1/gradients/Predictions_1/Softmax_grad/sub
Optimizer_1/gradients/Predictions_1/Softmax_grad/mul_1
Optimizer_1/gradients/Predictions_1/mlp_layer_1_grad/ReluGrad
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/BiasAddGrad
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/group_deps
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/control_dependency
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/control_dependency_1
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/MatMul
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/MatMul_1
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_1/gradients/AddN
Optimizer_1/GradientDescent/learning_rate
Optimizer_1/GradientDescent/update_Parameters_1/Weights/Weight/ApplyGradientDescent
Optimizer_1/GradientDescent/update_Parameters_1/Bias/Bias/ApplyGradientDescent
Optimizer_1/GradientDescent
Equal_1
Accuracy_1/Cast
Accuracy_1/Const
Accuracy_1/Mean
Accuracy_1/train_accuracy_list/initial_value
Accuracy_1/train_accuracy_list
Accuracy_1/train_accuracy_list/Assign
Accuracy_1/train_accuracy_list/read
Accuracy_1/Assign/value
Accuracy_1/Assign
Accuracy_1/train_accuracy/tags
Accuracy_1/train_accuracy
Accuracy_1/Cast_1
Accuracy_1/Const_1
Accuracy_1/Mean_1
Accuracy_1/validate_accuracy_list/initial_value
Accuracy_1/validate_accuracy_list
Accuracy_1/validate_accuracy_list/Assign
Accuracy_1/validate_accuracy_list/read
Accuracy_1/Assign_1/value
Accuracy_1/Assign_1
Accuracy_1/validate_accuracy/tags
Accuracy_1/validate_accuracy
Accuracy_1/Cast_2
Accuracy_1/Const_2
Accuracy_1/Mean_2
Accuracy_1/test_accuracy_list/initial_value
Accuracy_1/test_accuracy_list
Accuracy_1/test_accuracy_list/Assign
Accuracy_1/test_accuracy_list/read
Accuracy_1/Assign_2/value
Accuracy_1/Assign_2
Accuracy_1/test_accuracy/tags
Accuracy_1/test_accuracy
init_1/NoOp
init_1/NoOp_1
init_1
Merge_1/MergeSummary
save_1/Const
save_1/SaveV2/tensor_names
save_1/SaveV2/shape_and_slices
save_1/SaveV2
save_1/control_dependency
save_1/RestoreV2/tensor_names
save_1/RestoreV2/shape_and_slices
save_1/RestoreV2
save_1/Assign
save_1/RestoreV2_1/tensor_names
save_1/RestoreV2_1/shape_and_slices
save_1/RestoreV2_1
save_1/Assign_1
save_1/RestoreV2_2/tensor_names
save_1/RestoreV2_2/shape_and_slices
save_1/RestoreV2_2
save_1/Assign_2
save_1/RestoreV2_3/tensor_names
save_1/RestoreV2_3/shape_and_slices
save_1/RestoreV2_3
save_1/Assign_3
save_1/RestoreV2_4/tensor_names
save_1/RestoreV2_4/shape_and_slices
save_1/RestoreV2_4
save_1/Assign_4
save_1/RestoreV2_5/tensor_names
save_1/RestoreV2_5/shape_and_slices
save_1/RestoreV2_5
save_1/Assign_5
save_1/RestoreV2_6/tensor_names
save_1/RestoreV2_6/shape_and_slices
save_1/RestoreV2_6
save_1/Assign_6
save_1/RestoreV2_7/tensor_names
save_1/RestoreV2_7/shape_and_slices
save_1/RestoreV2_7
save_1/Assign_7
save_1/RestoreV2_8/tensor_names
save_1/RestoreV2_8/shape_and_slices
save_1/RestoreV2_8
save_1/Assign_8
save_1/RestoreV2_9/tensor_names
save_1/RestoreV2_9/shape_and_slices
save_1/RestoreV2_9
save_1/Assign_9
save_1/RestoreV2_10/tensor_names
save_1/RestoreV2_10/shape_and_slices
save_1/RestoreV2_10
save_1/Assign_10
save_1/RestoreV2_11/tensor_names
save_1/RestoreV2_11/shape_and_slices
save_1/RestoreV2_11
save_1/Assign_11
save_1/RestoreV2_12/tensor_names
save_1/RestoreV2_12/shape_and_slices
save_1/RestoreV2_12
save_1/Assign_12
save_1/RestoreV2_13/tensor_names
save_1/RestoreV2_13/shape_and_slices
save_1/RestoreV2_13
save_1/Assign_13
save_1/RestoreV2_14/tensor_names
save_1/RestoreV2_14/shape_and_slices
save_1/RestoreV2_14
save_1/Assign_14
save_1/RestoreV2_15/tensor_names
save_1/RestoreV2_15/shape_and_slices
save_1/RestoreV2_15
save_1/Assign_15
save_1/RestoreV2_16/tensor_names
save_1/RestoreV2_16/shape_and_slices
save_1/RestoreV2_16
save_1/Assign_16
save_1/RestoreV2_17/tensor_names
save_1/RestoreV2_17/shape_and_slices
save_1/RestoreV2_17
save_1/Assign_17
save_1/restore_all/NoOp
save_1/restore_all/NoOp_1
save_1/restore_all
last_successful_epoch_2/initial_value
last_successful_epoch_2
last_successful_epoch_2/Assign
last_successful_epoch_2/read
add_2/y
add_2
assign_updated_epoch_2
Inputs_2/Data/X_input
Inputs_2/Train_Labels/y_true_one_hot_label
Inputs_2/Train_Labels/y_true_class
Parameters_2/Weights/zeros
Parameters_2/Weights/Weight
Parameters_2/Weights/Weight/Assign
Parameters_2/Weights/Weight/read
Parameters_2/Bias/ones
Parameters_2/Bias/Bias
Parameters_2/Bias/Bias/Assign
Parameters_2/Bias/Bias/read
Predictions_2/mlp_layer/MatMul
Predictions_2/mlp_layer
Predictions_2/mlp_layer_1
Predictions_2/Softmax
Predictions_2/predict_class/dimension
Predictions_2/predict_class
Predictions_2/one_hot/Const
Predictions_2/one_hot/Const_1
Predictions_2/one_hot/depth
Predictions_2/one_hot/on_value
Predictions_2/one_hot/off_value
Predictions_2/one_hot
Cross_Entropy_2/Rank
Cross_Entropy_2/Shape
Cross_Entropy_2/Rank_1
Cross_Entropy_2/Shape_1
Cross_Entropy_2/Sub/y
Cross_Entropy_2/Sub
Cross_Entropy_2/Slice/begin
Cross_Entropy_2/Slice/size
Cross_Entropy_2/Slice
Cross_Entropy_2/concat/values_0
Cross_Entropy_2/concat/axis
Cross_Entropy_2/concat
Cross_Entropy_2/Reshape
Cross_Entropy_2/Rank_2
Cross_Entropy_2/Shape_2
Cross_Entropy_2/Sub_1/y
Cross_Entropy_2/Sub_1
Cross_Entropy_2/Slice_1/begin
Cross_Entropy_2/Slice_1/size
Cross_Entropy_2/Slice_1
Cross_Entropy_2/concat_1/values_0
Cross_Entropy_2/concat_1/axis
Cross_Entropy_2/concat_1
Cross_Entropy_2/Reshape_1
Cross_Entropy_2/SoftmaxCrossEntropyWithLogits
Cross_Entropy_2/Sub_2/y
Cross_Entropy_2/Sub_2
Cross_Entropy_2/Slice_2/begin
Cross_Entropy_2/Slice_2/size
Cross_Entropy_2/Slice_2
Cross_Entropy_2/Reshape_2
Loss_Function_2/Const
Loss_Function_2/Square
Loss_Function_2/Const_1
Loss_Function_2/Mean
Loss_Function_2/Const_2
Loss_Function_2/Mean_1
Loss_Function_2/Mul
Loss_Function_2/Add
Loss_Function_2/train_loss_list/initial_value
Loss_Function_2/train_loss_list
Loss_Function_2/train_loss_list/Assign
Loss_Function_2/train_loss_list/read
Loss_Function_2/Assign/value
Loss_Function_2/Assign
Loss_Function_2/train_error/tags
Loss_Function_2/train_error
Loss_Function_2/validate_loss_list/initial_value
Loss_Function_2/validate_loss_list
Loss_Function_2/validate_loss_list/Assign
Loss_Function_2/validate_loss_list/read
Loss_Function_2/Assign_1/value
Loss_Function_2/Assign_1
Loss_Function_2/validate_error/tags
Loss_Function_2/validate_error
Optimizer_2/learning_rate_progress/initial_value
Optimizer_2/learning_rate_progress
Optimizer_2/learning_rate_progress/Assign
Optimizer_2/learning_rate_progress/read
Optimizer_2/Assign/value
Optimizer_2/Assign
Optimizer_2/learning_rate/tags
Optimizer_2/learning_rate/values
Optimizer_2/learning_rate
Optimizer_2/gradients/Shape
Optimizer_2/gradients/Const
Optimizer_2/gradients/Fill
Optimizer_2/gradients/Loss_Function_2/Add_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Add_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/BroadcastGradientArgs
Optimizer_2/gradients/Loss_Function_2/Add_grad/Sum
Optimizer_2/gradients/Loss_Function_2/Add_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Add_grad/Sum_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/Reshape_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/group_deps
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/control_dependency
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/control_dependency_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Reshape/shape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Tile
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape_2
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Const
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Prod
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Const_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Prod_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Maximum/y
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Maximum
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/floordiv
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Cast
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/truediv
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/BroadcastGradientArgs
Optimizer_2/gradients/Loss_Function_2/Mul_grad/mul
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Sum
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mul_grad/mul_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Sum_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Reshape_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/group_deps
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/control_dependency
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/control_dependency_1
Optimizer_2/gradients/Cross_Entropy_2/Reshape_2_grad/Shape
Optimizer_2/gradients/Cross_Entropy_2/Reshape_2_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Reshape/shape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Tile/multiples
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Tile
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Const
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Prod
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Const_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Prod_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Maximum/y
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Maximum
Optimizer_2/gradients/Loss_Function_2/Mean_grad/floordiv
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Cast
Optimizer_2/gradients/Loss_Function_2/Mean_grad/truediv
Optimizer_2/gradients/zeros_like
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul/x
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul_1
Optimizer_2/gradients/Cross_Entropy_2/Reshape_grad/Shape
Optimizer_2/gradients/Cross_Entropy_2/Reshape_grad/Reshape
Optimizer_2/gradients/Predictions_2/Softmax_grad/mul
Optimizer_2/gradients/Predictions_2/Softmax_grad/Sum/reduction_indices
Optimizer_2/gradients/Predictions_2/Softmax_grad/Sum
Optimizer_2/gradients/Predictions_2/Softmax_grad/Reshape/shape
Optimizer_2/gradients/Predictions_2/Softmax_grad/Reshape
Optimizer_2/gradients/Predictions_2/Softmax_grad/sub
Optimizer_2/gradients/Predictions_2/Softmax_grad/mul_1
Optimizer_2/gradients/Predictions_2/mlp_layer_1_grad/ReluGrad
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/BiasAddGrad
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/group_deps
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/control_dependency
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/control_dependency_1
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/MatMul
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/MatMul_1
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_2/gradients/AddN
Optimizer_2/GradientDescent/learning_rate
Optimizer_2/GradientDescent/update_Parameters_2/Weights/Weight/ApplyGradientDescent
Optimizer_2/GradientDescent/update_Parameters_2/Bias/Bias/ApplyGradientDescent
Optimizer_2/GradientDescent
Equal_2
Accuracy_2/Cast
Accuracy_2/Const
Accuracy_2/Mean
Accuracy_2/train_accuracy_list/initial_value
Accuracy_2/train_accuracy_list
Accuracy_2/train_accuracy_list/Assign
Accuracy_2/train_accuracy_list/read
Accuracy_2/Assign/value
Accuracy_2/Assign
Accuracy_2/train_accuracy/tags
Accuracy_2/train_accuracy
Accuracy_2/Cast_1
Accuracy_2/Const_1
Accuracy_2/Mean_1
Accuracy_2/validate_accuracy_list/initial_value
Accuracy_2/validate_accuracy_list
Accuracy_2/validate_accuracy_list/Assign
Accuracy_2/validate_accuracy_list/read
Accuracy_2/Assign_1/value
Accuracy_2/Assign_1
Accuracy_2/validate_accuracy/tags
Accuracy_2/validate_accuracy
Accuracy_2/Cast_2
Accuracy_2/Const_2
Accuracy_2/Mean_2
Accuracy_2/test_accuracy_list/initial_value
Accuracy_2/test_accuracy_list
Accuracy_2/test_accuracy_list/Assign
Accuracy_2/test_accuracy_list/read
Accuracy_2/Assign_2/value
Accuracy_2/Assign_2
Accuracy_2/test_accuracy/tags
Accuracy_2/test_accuracy
init_2/NoOp
init_2/NoOp_1
init_2
Merge_2/MergeSummary
save_2/Const
save_2/SaveV2/tensor_names
save_2/SaveV2/shape_and_slices
save_2/SaveV2
save_2/control_dependency
save_2/RestoreV2/tensor_names
save_2/RestoreV2/shape_and_slices
save_2/RestoreV2
save_2/Assign
save_2/RestoreV2_1/tensor_names
save_2/RestoreV2_1/shape_and_slices
save_2/RestoreV2_1
save_2/Assign_1
save_2/RestoreV2_2/tensor_names
save_2/RestoreV2_2/shape_and_slices
save_2/RestoreV2_2
save_2/Assign_2
save_2/RestoreV2_3/tensor_names
save_2/RestoreV2_3/shape_and_slices
save_2/RestoreV2_3
save_2/Assign_3
save_2/RestoreV2_4/tensor_names
save_2/RestoreV2_4/shape_and_slices
save_2/RestoreV2_4
save_2/Assign_4
save_2/RestoreV2_5/tensor_names
save_2/RestoreV2_5/shape_and_slices
save_2/RestoreV2_5
save_2/Assign_5
save_2/RestoreV2_6/tensor_names
save_2/RestoreV2_6/shape_and_slices
save_2/RestoreV2_6
save_2/Assign_6
save_2/RestoreV2_7/tensor_names
save_2/RestoreV2_7/shape_and_slices
save_2/RestoreV2_7
save_2/Assign_7
save_2/RestoreV2_8/tensor_names
save_2/RestoreV2_8/shape_and_slices
save_2/RestoreV2_8
save_2/Assign_8
save_2/RestoreV2_9/tensor_names
save_2/RestoreV2_9/shape_and_slices
save_2/RestoreV2_9
save_2/Assign_9
save_2/RestoreV2_10/tensor_names
save_2/RestoreV2_10/shape_and_slices
save_2/RestoreV2_10
save_2/Assign_10
save_2/RestoreV2_11/tensor_names
save_2/RestoreV2_11/shape_and_slices
save_2/RestoreV2_11
save_2/Assign_11
save_2/RestoreV2_12/tensor_names
save_2/RestoreV2_12/shape_and_slices
save_2/RestoreV2_12
save_2/Assign_12
save_2/RestoreV2_13/tensor_names
save_2/RestoreV2_13/shape_and_slices
save_2/RestoreV2_13
save_2/Assign_13
save_2/RestoreV2_14/tensor_names
save_2/RestoreV2_14/shape_and_slices
save_2/RestoreV2_14
save_2/Assign_14
save_2/RestoreV2_15/tensor_names
save_2/RestoreV2_15/shape_and_slices
save_2/RestoreV2_15
save_2/Assign_15
save_2/RestoreV2_16/tensor_names
save_2/RestoreV2_16/shape_and_slices
save_2/RestoreV2_16
save_2/Assign_16
save_2/RestoreV2_17/tensor_names
save_2/RestoreV2_17/shape_and_slices
save_2/RestoreV2_17
save_2/Assign_17
save_2/RestoreV2_18/tensor_names
save_2/RestoreV2_18/shape_and_slices
save_2/RestoreV2_18
save_2/Assign_18
save_2/RestoreV2_19/tensor_names
save_2/RestoreV2_19/shape_and_slices
save_2/RestoreV2_19
save_2/Assign_19
save_2/RestoreV2_20/tensor_names
save_2/RestoreV2_20/shape_and_slices
save_2/RestoreV2_20
save_2/Assign_20
save_2/RestoreV2_21/tensor_names
save_2/RestoreV2_21/shape_and_slices
save_2/RestoreV2_21
save_2/Assign_21
save_2/RestoreV2_22/tensor_names
save_2/RestoreV2_22/shape_and_slices
save_2/RestoreV2_22
save_2/Assign_22
save_2/RestoreV2_23/tensor_names
save_2/RestoreV2_23/shape_and_slices
save_2/RestoreV2_23
save_2/Assign_23
save_2/RestoreV2_24/tensor_names
save_2/RestoreV2_24/shape_and_slices
save_2/RestoreV2_24
save_2/Assign_24
save_2/RestoreV2_25/tensor_names
save_2/RestoreV2_25/shape_and_slices
save_2/RestoreV2_25
save_2/Assign_25
save_2/RestoreV2_26/tensor_names
save_2/RestoreV2_26/shape_and_slices
save_2/RestoreV2_26
save_2/Assign_26
save_2/restore_all/NoOp
save_2/restore_all/NoOp_1
save_2/restore_all
last_successful_epoch_3/initial_value
last_successful_epoch_3
last_successful_epoch_3/Assign
last_successful_epoch_3/read
add_3/y
add_3
assign_updated_epoch_3
Inputs_3/Data/X_input
Inputs_3/Train_Labels/y_true_one_hot_label
Inputs_3/Train_Labels/y_true_class
Parameters_3/Weights/zeros
Parameters_3/Weights/Weight
Parameters_3/Weights/Weight/Assign
Parameters_3/Weights/Weight/read
Parameters_3/Bias/ones
Parameters_3/Bias/Bias
Parameters_3/Bias/Bias/Assign
Parameters_3/Bias/Bias/read
Predictions_3/mlp_layer/MatMul
Predictions_3/mlp_layer
Predictions_3/mlp_layer_1
Predictions_3/Softmax
Predictions_3/predict_class/dimension
Predictions_3/predict_class
Predictions_3/one_hot/Const
Predictions_3/one_hot/Const_1
Predictions_3/one_hot/depth
Predictions_3/one_hot/on_value
Predictions_3/one_hot/off_value
Predictions_3/one_hot
Cross_Entropy_3/Rank
Cross_Entropy_3/Shape
Cross_Entropy_3/Rank_1
Cross_Entropy_3/Shape_1
Cross_Entropy_3/Sub/y
Cross_Entropy_3/Sub
Cross_Entropy_3/Slice/begin
Cross_Entropy_3/Slice/size
Cross_Entropy_3/Slice
Cross_Entropy_3/concat/values_0
Cross_Entropy_3/concat/axis
Cross_Entropy_3/concat
Cross_Entropy_3/Reshape
Cross_Entropy_3/Rank_2
Cross_Entropy_3/Shape_2
Cross_Entropy_3/Sub_1/y
Cross_Entropy_3/Sub_1
Cross_Entropy_3/Slice_1/begin
Cross_Entropy_3/Slice_1/size
Cross_Entropy_3/Slice_1
Cross_Entropy_3/concat_1/values_0
Cross_Entropy_3/concat_1/axis
Cross_Entropy_3/concat_1
Cross_Entropy_3/Reshape_1
Cross_Entropy_3/SoftmaxCrossEntropyWithLogits
Cross_Entropy_3/Sub_2/y
Cross_Entropy_3/Sub_2
Cross_Entropy_3/Slice_2/begin
Cross_Entropy_3/Slice_2/size
Cross_Entropy_3/Slice_2
Cross_Entropy_3/Reshape_2
Loss_Function_3/Const
Loss_Function_3/Square
Loss_Function_3/Const_1
Loss_Function_3/Mean
Loss_Function_3/Const_2
Loss_Function_3/Mean_1
Loss_Function_3/Mul
Loss_Function_3/Add
Loss_Function_3/train_loss_list/initial_value
Loss_Function_3/train_loss_list
Loss_Function_3/train_loss_list/Assign
Loss_Function_3/train_loss_list/read
Loss_Function_3/Assign/value
Loss_Function_3/Assign
Loss_Function_3/train_error/tags
Loss_Function_3/train_error
Loss_Function_3/validate_loss_list/initial_value
Loss_Function_3/validate_loss_list
Loss_Function_3/validate_loss_list/Assign
Loss_Function_3/validate_loss_list/read
Loss_Function_3/Assign_1/value
Loss_Function_3/Assign_1
Loss_Function_3/validate_error/tags
Loss_Function_3/validate_error
Optimizer_3/learning_rate_progress/initial_value
Optimizer_3/learning_rate_progress
Optimizer_3/learning_rate_progress/Assign
Optimizer_3/learning_rate_progress/read
Optimizer_3/Assign/value
Optimizer_3/Assign
Optimizer_3/learning_rate/tags
Optimizer_3/learning_rate/values
Optimizer_3/learning_rate
Optimizer_3/gradients/Shape
Optimizer_3/gradients/Const
Optimizer_3/gradients/Fill
Optimizer_3/gradients/Loss_Function_3/Add_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Add_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/BroadcastGradientArgs
Optimizer_3/gradients/Loss_Function_3/Add_grad/Sum
Optimizer_3/gradients/Loss_Function_3/Add_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Add_grad/Sum_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/Reshape_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/group_deps
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/control_dependency
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/control_dependency_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Reshape/shape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Tile
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape_2
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Const
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Prod
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Const_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Prod_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Maximum/y
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Maximum
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/floordiv
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Cast
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/truediv
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/BroadcastGradientArgs
Optimizer_3/gradients/Loss_Function_3/Mul_grad/mul
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Sum
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mul_grad/mul_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Sum_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Reshape_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/group_deps
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/control_dependency
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/control_dependency_1
Optimizer_3/gradients/Cross_Entropy_3/Reshape_2_grad/Shape
Optimizer_3/gradients/Cross_Entropy_3/Reshape_2_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Reshape/shape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Tile/multiples
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Tile
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Const
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Prod
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Const_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Prod_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Maximum/y
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Maximum
Optimizer_3/gradients/Loss_Function_3/Mean_grad/floordiv
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Cast
Optimizer_3/gradients/Loss_Function_3/Mean_grad/truediv
Optimizer_3/gradients/zeros_like
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul/x
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul_1
Optimizer_3/gradients/Cross_Entropy_3/Reshape_grad/Shape
Optimizer_3/gradients/Cross_Entropy_3/Reshape_grad/Reshape
Optimizer_3/gradients/Predictions_3/Softmax_grad/mul
Optimizer_3/gradients/Predictions_3/Softmax_grad/Sum/reduction_indices
Optimizer_3/gradients/Predictions_3/Softmax_grad/Sum
Optimizer_3/gradients/Predictions_3/Softmax_grad/Reshape/shape
Optimizer_3/gradients/Predictions_3/Softmax_grad/Reshape
Optimizer_3/gradients/Predictions_3/Softmax_grad/sub
Optimizer_3/gradients/Predictions_3/Softmax_grad/mul_1
Optimizer_3/gradients/Predictions_3/mlp_layer_1_grad/ReluGrad
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/BiasAddGrad
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/group_deps
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/control_dependency
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/control_dependency_1
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/MatMul
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/MatMul_1
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_3/gradients/AddN
Optimizer_3/GradientDescent/learning_rate
Optimizer_3/GradientDescent/update_Parameters_3/Weights/Weight/ApplyGradientDescent
Optimizer_3/GradientDescent/update_Parameters_3/Bias/Bias/ApplyGradientDescent
Optimizer_3/GradientDescent
Equal_3
Accuracy_3/Cast
Accuracy_3/Const
Accuracy_3/Mean
Accuracy_3/train_accuracy_list/initial_value
Accuracy_3/train_accuracy_list
Accuracy_3/train_accuracy_list/Assign
Accuracy_3/train_accuracy_list/read
Accuracy_3/Assign/value
Accuracy_3/Assign
Accuracy_3/train_accuracy/tags
Accuracy_3/train_accuracy
Accuracy_3/Cast_1
Accuracy_3/Const_1
Accuracy_3/Mean_1
Accuracy_3/validate_accuracy_list/initial_value
Accuracy_3/validate_accuracy_list
Accuracy_3/validate_accuracy_list/Assign
Accuracy_3/validate_accuracy_list/read
Accuracy_3/Assign_1/value
Accuracy_3/Assign_1
Accuracy_3/validate_accuracy/tags
Accuracy_3/validate_accuracy
Accuracy_3/Cast_2
Accuracy_3/Const_2
Accuracy_3/Mean_2
Accuracy_3/test_accuracy_list/initial_value
Accuracy_3/test_accuracy_list
Accuracy_3/test_accuracy_list/Assign
Accuracy_3/test_accuracy_list/read
Accuracy_3/Assign_2/value
Accuracy_3/Assign_2
Accuracy_3/test_accuracy/tags
Accuracy_3/test_accuracy
init_3/NoOp
init_3/NoOp_1
init_3
Merge_3/MergeSummary
save_3/Const
save_3/SaveV2/tensor_names
save_3/SaveV2/shape_and_slices
save_3/SaveV2
save_3/control_dependency
save_3/RestoreV2/tensor_names
save_3/RestoreV2/shape_and_slices
save_3/RestoreV2
save_3/Assign
save_3/RestoreV2_1/tensor_names
save_3/RestoreV2_1/shape_and_slices
save_3/RestoreV2_1
save_3/Assign_1
save_3/RestoreV2_2/tensor_names
save_3/RestoreV2_2/shape_and_slices
save_3/RestoreV2_2
save_3/Assign_2
save_3/RestoreV2_3/tensor_names
save_3/RestoreV2_3/shape_and_slices
save_3/RestoreV2_3
save_3/Assign_3
save_3/RestoreV2_4/tensor_names
save_3/RestoreV2_4/shape_and_slices
save_3/RestoreV2_4
save_3/Assign_4
save_3/RestoreV2_5/tensor_names
save_3/RestoreV2_5/shape_and_slices
save_3/RestoreV2_5
save_3/Assign_5
save_3/RestoreV2_6/tensor_names
save_3/RestoreV2_6/shape_and_slices
save_3/RestoreV2_6
save_3/Assign_6
save_3/RestoreV2_7/tensor_names
save_3/RestoreV2_7/shape_and_slices
save_3/RestoreV2_7
save_3/Assign_7
save_3/RestoreV2_8/tensor_names
save_3/RestoreV2_8/shape_and_slices
save_3/RestoreV2_8
save_3/Assign_8
save_3/RestoreV2_9/tensor_names
save_3/RestoreV2_9/shape_and_slices
save_3/RestoreV2_9
save_3/Assign_9
save_3/RestoreV2_10/tensor_names
save_3/RestoreV2_10/shape_and_slices
save_3/RestoreV2_10
save_3/Assign_10
save_3/RestoreV2_11/tensor_names
save_3/RestoreV2_11/shape_and_slices
save_3/RestoreV2_11
save_3/Assign_11
save_3/RestoreV2_12/tensor_names
save_3/RestoreV2_12/shape_and_slices
save_3/RestoreV2_12
save_3/Assign_12
save_3/RestoreV2_13/tensor_names
save_3/RestoreV2_13/shape_and_slices
save_3/RestoreV2_13
save_3/Assign_13
save_3/RestoreV2_14/tensor_names
save_3/RestoreV2_14/shape_and_slices
save_3/RestoreV2_14
save_3/Assign_14
save_3/RestoreV2_15/tensor_names
save_3/RestoreV2_15/shape_and_slices
save_3/RestoreV2_15
save_3/Assign_15
save_3/RestoreV2_16/tensor_names
save_3/RestoreV2_16/shape_and_slices
save_3/RestoreV2_16
save_3/Assign_16
save_3/RestoreV2_17/tensor_names
save_3/RestoreV2_17/shape_and_slices
save_3/RestoreV2_17
save_3/Assign_17
save_3/RestoreV2_18/tensor_names
save_3/RestoreV2_18/shape_and_slices
save_3/RestoreV2_18
save_3/Assign_18
save_3/RestoreV2_19/tensor_names
save_3/RestoreV2_19/shape_and_slices
save_3/RestoreV2_19
save_3/Assign_19
save_3/RestoreV2_20/tensor_names
save_3/RestoreV2_20/shape_and_slices
save_3/RestoreV2_20
save_3/Assign_20
save_3/RestoreV2_21/tensor_names
save_3/RestoreV2_21/shape_and_slices
save_3/RestoreV2_21
save_3/Assign_21
save_3/RestoreV2_22/tensor_names
save_3/RestoreV2_22/shape_and_slices
save_3/RestoreV2_22
save_3/Assign_22
save_3/RestoreV2_23/tensor_names
save_3/RestoreV2_23/shape_and_slices
save_3/RestoreV2_23
save_3/Assign_23
save_3/RestoreV2_24/tensor_names
save_3/RestoreV2_24/shape_and_slices
save_3/RestoreV2_24
save_3/Assign_24
save_3/RestoreV2_25/tensor_names
save_3/RestoreV2_25/shape_and_slices
save_3/RestoreV2_25
save_3/Assign_25
save_3/RestoreV2_26/tensor_names
save_3/RestoreV2_26/shape_and_slices
save_3/RestoreV2_26
save_3/Assign_26
save_3/RestoreV2_27/tensor_names
save_3/RestoreV2_27/shape_and_slices
save_3/RestoreV2_27
save_3/Assign_27
save_3/RestoreV2_28/tensor_names
save_3/RestoreV2_28/shape_and_slices
save_3/RestoreV2_28
save_3/Assign_28
save_3/RestoreV2_29/tensor_names
save_3/RestoreV2_29/shape_and_slices
save_3/RestoreV2_29
save_3/Assign_29
save_3/RestoreV2_30/tensor_names
save_3/RestoreV2_30/shape_and_slices
save_3/RestoreV2_30
save_3/Assign_30
save_3/RestoreV2_31/tensor_names
save_3/RestoreV2_31/shape_and_slices
save_3/RestoreV2_31
save_3/Assign_31
save_3/RestoreV2_32/tensor_names
save_3/RestoreV2_32/shape_and_slices
save_3/RestoreV2_32
save_3/Assign_32
save_3/RestoreV2_33/tensor_names
save_3/RestoreV2_33/shape_and_slices
save_3/RestoreV2_33
save_3/Assign_33
save_3/RestoreV2_34/tensor_names
save_3/RestoreV2_34/shape_and_slices
save_3/RestoreV2_34
save_3/Assign_34
save_3/RestoreV2_35/tensor_names
save_3/RestoreV2_35/shape_and_slices
save_3/RestoreV2_35
save_3/Assign_35
save_3/restore_all/NoOp
save_3/restore_all/NoOp_1
save_3/restore_all
last_successful_epoch_4/initial_value
last_successful_epoch_4
last_successful_epoch_4/Assign
last_successful_epoch_4/read
add_4/y
add_4
assign_updated_epoch_4
Inputs_4/Data/X_input
Inputs_4/Train_Labels/y_true_one_hot_label
Inputs_4/Train_Labels/y_true_class
Parameters_4/Weights/zeros
Parameters_4/Weights/Weight
Parameters_4/Weights/Weight/Assign
Parameters_4/Weights/Weight/read
Parameters_4/Bias/ones
Parameters_4/Bias/Bias
Parameters_4/Bias/Bias/Assign
Parameters_4/Bias/Bias/read
Predictions_4/mlp_layer/MatMul
Predictions_4/mlp_layer
Predictions_4/mlp_layer_1
Predictions_4/Softmax
Predictions_4/predict_class/dimension
Predictions_4/predict_class
Predictions_4/one_hot/Const
Predictions_4/one_hot/Const_1
Predictions_4/one_hot/depth
Predictions_4/one_hot/on_value
Predictions_4/one_hot/off_value
Predictions_4/one_hot
Cross_Entropy_4/Rank
Cross_Entropy_4/Shape
Cross_Entropy_4/Rank_1
Cross_Entropy_4/Shape_1
Cross_Entropy_4/Sub/y
Cross_Entropy_4/Sub
Cross_Entropy_4/Slice/begin
Cross_Entropy_4/Slice/size
Cross_Entropy_4/Slice
Cross_Entropy_4/concat/values_0
Cross_Entropy_4/concat/axis
Cross_Entropy_4/concat
Cross_Entropy_4/Reshape
Cross_Entropy_4/Rank_2
Cross_Entropy_4/Shape_2
Cross_Entropy_4/Sub_1/y
Cross_Entropy_4/Sub_1
Cross_Entropy_4/Slice_1/begin
Cross_Entropy_4/Slice_1/size
Cross_Entropy_4/Slice_1
Cross_Entropy_4/concat_1/values_0
Cross_Entropy_4/concat_1/axis
Cross_Entropy_4/concat_1
Cross_Entropy_4/Reshape_1
Cross_Entropy_4/SoftmaxCrossEntropyWithLogits
Cross_Entropy_4/Sub_2/y
Cross_Entropy_4/Sub_2
Cross_Entropy_4/Slice_2/begin
Cross_Entropy_4/Slice_2/size
Cross_Entropy_4/Slice_2
Cross_Entropy_4/Reshape_2
Loss_Function_4/Const
Loss_Function_4/Square
Loss_Function_4/Const_1
Loss_Function_4/Mean
Loss_Function_4/Const_2
Loss_Function_4/Mean_1
Loss_Function_4/Mul
Loss_Function_4/Add
Loss_Function_4/train_loss_list/initial_value
Loss_Function_4/train_loss_list
Loss_Function_4/train_loss_list/Assign
Loss_Function_4/train_loss_list/read
Loss_Function_4/Assign/value
Loss_Function_4/Assign
Loss_Function_4/train_error/tags
Loss_Function_4/train_error
Loss_Function_4/validate_loss_list/initial_value
Loss_Function_4/validate_loss_list
Loss_Function_4/validate_loss_list/Assign
Loss_Function_4/validate_loss_list/read
Loss_Function_4/Assign_1/value
Loss_Function_4/Assign_1
Loss_Function_4/validate_error/tags
Loss_Function_4/validate_error
Optimizer_4/learning_rate_progress/initial_value
Optimizer_4/learning_rate_progress
Optimizer_4/learning_rate_progress/Assign
Optimizer_4/learning_rate_progress/read
Optimizer_4/Assign/value
Optimizer_4/Assign
Optimizer_4/learning_rate/tags
Optimizer_4/learning_rate/values
Optimizer_4/learning_rate
Optimizer_4/gradients/Shape
Optimizer_4/gradients/Const
Optimizer_4/gradients/Fill
Optimizer_4/gradients/Loss_Function_4/Add_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Add_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/BroadcastGradientArgs
Optimizer_4/gradients/Loss_Function_4/Add_grad/Sum
Optimizer_4/gradients/Loss_Function_4/Add_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Add_grad/Sum_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/Reshape_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/group_deps
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/control_dependency
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/control_dependency_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Reshape/shape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Tile
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape_2
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Const
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Prod
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Const_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Prod_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Maximum/y
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Maximum
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/floordiv
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Cast
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/truediv
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/BroadcastGradientArgs
Optimizer_4/gradients/Loss_Function_4/Mul_grad/mul
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Sum
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mul_grad/mul_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Sum_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Reshape_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/group_deps
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/control_dependency
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/control_dependency_1
Optimizer_4/gradients/Cross_Entropy_4/Reshape_2_grad/Shape
Optimizer_4/gradients/Cross_Entropy_4/Reshape_2_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Reshape/shape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Tile/multiples
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Tile
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Const
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Prod
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Const_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Prod_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Maximum/y
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Maximum
Optimizer_4/gradients/Loss_Function_4/Mean_grad/floordiv
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Cast
Optimizer_4/gradients/Loss_Function_4/Mean_grad/truediv
Optimizer_4/gradients/zeros_like
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul/x
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul_1
Optimizer_4/gradients/Cross_Entropy_4/Reshape_grad/Shape
Optimizer_4/gradients/Cross_Entropy_4/Reshape_grad/Reshape
Optimizer_4/gradients/Predictions_4/Softmax_grad/mul
Optimizer_4/gradients/Predictions_4/Softmax_grad/Sum/reduction_indices
Optimizer_4/gradients/Predictions_4/Softmax_grad/Sum
Optimizer_4/gradients/Predictions_4/Softmax_grad/Reshape/shape
Optimizer_4/gradients/Predictions_4/Softmax_grad/Reshape
Optimizer_4/gradients/Predictions_4/Softmax_grad/sub
Optimizer_4/gradients/Predictions_4/Softmax_grad/mul_1
Optimizer_4/gradients/Predictions_4/mlp_layer_1_grad/ReluGrad
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/BiasAddGrad
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/group_deps
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/control_dependency
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/control_dependency_1
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/MatMul
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/MatMul_1
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_4/gradients/AddN
Optimizer_4/GradientDescent/learning_rate
Optimizer_4/GradientDescent/update_Parameters_4/Weights/Weight/ApplyGradientDescent
Optimizer_4/GradientDescent/update_Parameters_4/Bias/Bias/ApplyGradientDescent
Optimizer_4/GradientDescent
Equal_4
Accuracy_4/Cast
Accuracy_4/Const
Accuracy_4/Mean
Accuracy_4/train_accuracy_list/initial_value
Accuracy_4/train_accuracy_list
Accuracy_4/train_accuracy_list/Assign
Accuracy_4/train_accuracy_list/read
Accuracy_4/Assign/value
Accuracy_4/Assign
Accuracy_4/train_accuracy/tags
Accuracy_4/train_accuracy
Accuracy_4/Cast_1
Accuracy_4/Const_1
Accuracy_4/Mean_1
Accuracy_4/validate_accuracy_list/initial_value
Accuracy_4/validate_accuracy_list
Accuracy_4/validate_accuracy_list/Assign
Accuracy_4/validate_accuracy_list/read
Accuracy_4/Assign_1/value
Accuracy_4/Assign_1
Accuracy_4/validate_accuracy/tags
Accuracy_4/validate_accuracy
Accuracy_4/Cast_2
Accuracy_4/Const_2
Accuracy_4/Mean_2
Accuracy_4/test_accuracy_list/initial_value
Accuracy_4/test_accuracy_list
Accuracy_4/test_accuracy_list/Assign
Accuracy_4/test_accuracy_list/read
Accuracy_4/Assign_2/value
Accuracy_4/Assign_2
Accuracy_4/test_accuracy/tags
Accuracy_4/test_accuracy
init_4/NoOp
init_4/NoOp_1
init_4
Merge_4/MergeSummary
save_4/Const
save_4/SaveV2/tensor_names
save_4/SaveV2/shape_and_slices
save_4/SaveV2
save_4/control_dependency
save_4/RestoreV2/tensor_names
save_4/RestoreV2/shape_and_slices
save_4/RestoreV2
save_4/Assign
save_4/RestoreV2_1/tensor_names
save_4/RestoreV2_1/shape_and_slices
save_4/RestoreV2_1
save_4/Assign_1
save_4/RestoreV2_2/tensor_names
save_4/RestoreV2_2/shape_and_slices
save_4/RestoreV2_2
save_4/Assign_2
save_4/RestoreV2_3/tensor_names
save_4/RestoreV2_3/shape_and_slices
save_4/RestoreV2_3
save_4/Assign_3
save_4/RestoreV2_4/tensor_names
save_4/RestoreV2_4/shape_and_slices
save_4/RestoreV2_4
save_4/Assign_4
save_4/RestoreV2_5/tensor_names
save_4/RestoreV2_5/shape_and_slices
save_4/RestoreV2_5
save_4/Assign_5
save_4/RestoreV2_6/tensor_names
save_4/RestoreV2_6/shape_and_slices
save_4/RestoreV2_6
save_4/Assign_6
save_4/RestoreV2_7/tensor_names
save_4/RestoreV2_7/shape_and_slices
save_4/RestoreV2_7
save_4/Assign_7
save_4/RestoreV2_8/tensor_names
save_4/RestoreV2_8/shape_and_slices
save_4/RestoreV2_8
save_4/Assign_8
save_4/RestoreV2_9/tensor_names
save_4/RestoreV2_9/shape_and_slices
save_4/RestoreV2_9
save_4/Assign_9
save_4/RestoreV2_10/tensor_names
save_4/RestoreV2_10/shape_and_slices
save_4/RestoreV2_10
save_4/Assign_10
save_4/RestoreV2_11/tensor_names
save_4/RestoreV2_11/shape_and_slices
save_4/RestoreV2_11
save_4/Assign_11
save_4/RestoreV2_12/tensor_names
save_4/RestoreV2_12/shape_and_slices
save_4/RestoreV2_12
save_4/Assign_12
save_4/RestoreV2_13/tensor_names
save_4/RestoreV2_13/shape_and_slices
save_4/RestoreV2_13
save_4/Assign_13
save_4/RestoreV2_14/tensor_names
save_4/RestoreV2_14/shape_and_slices
save_4/RestoreV2_14
save_4/Assign_14
save_4/RestoreV2_15/tensor_names
save_4/RestoreV2_15/shape_and_slices
save_4/RestoreV2_15
save_4/Assign_15
save_4/RestoreV2_16/tensor_names
save_4/RestoreV2_16/shape_and_slices
save_4/RestoreV2_16
save_4/Assign_16
save_4/RestoreV2_17/tensor_names
save_4/RestoreV2_17/shape_and_slices
save_4/RestoreV2_17
save_4/Assign_17
save_4/RestoreV2_18/tensor_names
save_4/RestoreV2_18/shape_and_slices
save_4/RestoreV2_18
save_4/Assign_18
save_4/RestoreV2_19/tensor_names
save_4/RestoreV2_19/shape_and_slices
save_4/RestoreV2_19
save_4/Assign_19
save_4/RestoreV2_20/tensor_names
save_4/RestoreV2_20/shape_and_slices
save_4/RestoreV2_20
save_4/Assign_20
save_4/RestoreV2_21/tensor_names
save_4/RestoreV2_21/shape_and_slices
save_4/RestoreV2_21
save_4/Assign_21
save_4/RestoreV2_22/tensor_names
save_4/RestoreV2_22/shape_and_slices
save_4/RestoreV2_22
save_4/Assign_22
save_4/RestoreV2_23/tensor_names
save_4/RestoreV2_23/shape_and_slices
save_4/RestoreV2_23
save_4/Assign_23
save_4/RestoreV2_24/tensor_names
save_4/RestoreV2_24/shape_and_slices
save_4/RestoreV2_24
save_4/Assign_24
save_4/RestoreV2_25/tensor_names
save_4/RestoreV2_25/shape_and_slices
save_4/RestoreV2_25
save_4/Assign_25
save_4/RestoreV2_26/tensor_names
save_4/RestoreV2_26/shape_and_slices
save_4/RestoreV2_26
save_4/Assign_26
save_4/RestoreV2_27/tensor_names
save_4/RestoreV2_27/shape_and_slices
save_4/RestoreV2_27
save_4/Assign_27
save_4/RestoreV2_28/tensor_names
save_4/RestoreV2_28/shape_and_slices
save_4/RestoreV2_28
save_4/Assign_28
save_4/RestoreV2_29/tensor_names
save_4/RestoreV2_29/shape_and_slices
save_4/RestoreV2_29
save_4/Assign_29
save_4/RestoreV2_30/tensor_names
save_4/RestoreV2_30/shape_and_slices
save_4/RestoreV2_30
save_4/Assign_30
save_4/RestoreV2_31/tensor_names
save_4/RestoreV2_31/shape_and_slices
save_4/RestoreV2_31
save_4/Assign_31
save_4/RestoreV2_32/tensor_names
save_4/RestoreV2_32/shape_and_slices
save_4/RestoreV2_32
save_4/Assign_32
save_4/RestoreV2_33/tensor_names
save_4/RestoreV2_33/shape_and_slices
save_4/RestoreV2_33
save_4/Assign_33
save_4/RestoreV2_34/tensor_names
save_4/RestoreV2_34/shape_and_slices
save_4/RestoreV2_34
save_4/Assign_34
save_4/RestoreV2_35/tensor_names
save_4/RestoreV2_35/shape_and_slices
save_4/RestoreV2_35
save_4/Assign_35
save_4/RestoreV2_36/tensor_names
save_4/RestoreV2_36/shape_and_slices
save_4/RestoreV2_36
save_4/Assign_36
save_4/RestoreV2_37/tensor_names
save_4/RestoreV2_37/shape_and_slices
save_4/RestoreV2_37
save_4/Assign_37
save_4/RestoreV2_38/tensor_names
save_4/RestoreV2_38/shape_and_slices
save_4/RestoreV2_38
save_4/Assign_38
save_4/RestoreV2_39/tensor_names
save_4/RestoreV2_39/shape_and_slices
save_4/RestoreV2_39
save_4/Assign_39
save_4/RestoreV2_40/tensor_names
save_4/RestoreV2_40/shape_and_slices
save_4/RestoreV2_40
save_4/Assign_40
save_4/RestoreV2_41/tensor_names
save_4/RestoreV2_41/shape_and_slices
save_4/RestoreV2_41
save_4/Assign_41
save_4/RestoreV2_42/tensor_names
save_4/RestoreV2_42/shape_and_slices
save_4/RestoreV2_42
save_4/Assign_42
save_4/RestoreV2_43/tensor_names
save_4/RestoreV2_43/shape_and_slices
save_4/RestoreV2_43
save_4/Assign_43
save_4/RestoreV2_44/tensor_names
save_4/RestoreV2_44/shape_and_slices
save_4/RestoreV2_44
save_4/Assign_44
save_4/restore_all/NoOp
save_4/restore_all/NoOp_1
save_4/restore_all
last_successful_epoch_5/initial_value
last_successful_epoch_5
last_successful_epoch_5/Assign
last_successful_epoch_5/read
add_5/y
add_5
assign_updated_epoch_5
Inputs_5/Data/X_input
Inputs_5/Train_Labels/y_true_one_hot_label
Inputs_5/Train_Labels/y_true_class
Parameters_5/Weights/zeros
Parameters_5/Weights/Weight
Parameters_5/Weights/Weight/Assign
Parameters_5/Weights/Weight/read
Parameters_5/Bias/ones
Parameters_5/Bias/Bias
Parameters_5/Bias/Bias/Assign
Parameters_5/Bias/Bias/read
Predictions_5/mlp_layer/MatMul
Predictions_5/mlp_layer
Predictions_5/mlp_layer_1
Predictions_5/Softmax
Predictions_5/predict_class/dimension
Predictions_5/predict_class
Predictions_5/one_hot/Const
Predictions_5/one_hot/Const_1
Predictions_5/one_hot/depth
Predictions_5/one_hot/on_value
Predictions_5/one_hot/off_value
Predictions_5/one_hot
Cross_Entropy_5/Rank
Cross_Entropy_5/Shape
Cross_Entropy_5/Rank_1
Cross_Entropy_5/Shape_1
Cross_Entropy_5/Sub/y
Cross_Entropy_5/Sub
Cross_Entropy_5/Slice/begin
Cross_Entropy_5/Slice/size
Cross_Entropy_5/Slice
Cross_Entropy_5/concat/values_0
Cross_Entropy_5/concat/axis
Cross_Entropy_5/concat
Cross_Entropy_5/Reshape
Cross_Entropy_5/Rank_2
Cross_Entropy_5/Shape_2
Cross_Entropy_5/Sub_1/y
Cross_Entropy_5/Sub_1
Cross_Entropy_5/Slice_1/begin
Cross_Entropy_5/Slice_1/size
Cross_Entropy_5/Slice_1
Cross_Entropy_5/concat_1/values_0
Cross_Entropy_5/concat_1/axis
Cross_Entropy_5/concat_1
Cross_Entropy_5/Reshape_1
Cross_Entropy_5/SoftmaxCrossEntropyWithLogits
Cross_Entropy_5/Sub_2/y
Cross_Entropy_5/Sub_2
Cross_Entropy_5/Slice_2/begin
Cross_Entropy_5/Slice_2/size
Cross_Entropy_5/Slice_2
Cross_Entropy_5/Reshape_2
Loss_Function_5/Const
Loss_Function_5/Square
Loss_Function_5/Const_1
Loss_Function_5/Mean
Loss_Function_5/Const_2
Loss_Function_5/Mean_1
Loss_Function_5/Mul
Loss_Function_5/Add
Loss_Function_5/train_loss_list/initial_value
Loss_Function_5/train_loss_list
Loss_Function_5/train_loss_list/Assign
Loss_Function_5/train_loss_list/read
Loss_Function_5/Assign/value
Loss_Function_5/Assign
Loss_Function_5/train_error/tags
Loss_Function_5/train_error
Loss_Function_5/validate_loss_list/initial_value
Loss_Function_5/validate_loss_list
Loss_Function_5/validate_loss_list/Assign
Loss_Function_5/validate_loss_list/read
Loss_Function_5/Assign_1/value
Loss_Function_5/Assign_1
Loss_Function_5/validate_error/tags
Loss_Function_5/validate_error
Optimizer_5/learning_rate_progress/initial_value
Optimizer_5/learning_rate_progress
Optimizer_5/learning_rate_progress/Assign
Optimizer_5/learning_rate_progress/read
Optimizer_5/Assign/value
Optimizer_5/Assign
Optimizer_5/learning_rate/tags
Optimizer_5/learning_rate/values
Optimizer_5/learning_rate
Optimizer_5/gradients/Shape
Optimizer_5/gradients/Const
Optimizer_5/gradients/Fill
Optimizer_5/gradients/Loss_Function_5/Add_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Add_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/BroadcastGradientArgs
Optimizer_5/gradients/Loss_Function_5/Add_grad/Sum
Optimizer_5/gradients/Loss_Function_5/Add_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Add_grad/Sum_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/Reshape_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/group_deps
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/control_dependency
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/control_dependency_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Reshape/shape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Tile
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape_2
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Const
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Prod
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Const_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Prod_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Maximum/y
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Maximum
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/floordiv
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Cast
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/truediv
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/BroadcastGradientArgs
Optimizer_5/gradients/Loss_Function_5/Mul_grad/mul
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Sum
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mul_grad/mul_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Sum_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Reshape_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/group_deps
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/control_dependency
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/control_dependency_1
Optimizer_5/gradients/Cross_Entropy_5/Reshape_2_grad/Shape
Optimizer_5/gradients/Cross_Entropy_5/Reshape_2_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Reshape/shape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Tile/multiples
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Tile
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Const
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Prod
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Const_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Prod_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Maximum/y
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Maximum
Optimizer_5/gradients/Loss_Function_5/Mean_grad/floordiv
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Cast
Optimizer_5/gradients/Loss_Function_5/Mean_grad/truediv
Optimizer_5/gradients/zeros_like
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul/x
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul_1
Optimizer_5/gradients/Cross_Entropy_5/Reshape_grad/Shape
Optimizer_5/gradients/Cross_Entropy_5/Reshape_grad/Reshape
Optimizer_5/gradients/Predictions_5/Softmax_grad/mul
Optimizer_5/gradients/Predictions_5/Softmax_grad/Sum/reduction_indices
Optimizer_5/gradients/Predictions_5/Softmax_grad/Sum
Optimizer_5/gradients/Predictions_5/Softmax_grad/Reshape/shape
Optimizer_5/gradients/Predictions_5/Softmax_grad/Reshape
Optimizer_5/gradients/Predictions_5/Softmax_grad/sub
Optimizer_5/gradients/Predictions_5/Softmax_grad/mul_1
Optimizer_5/gradients/Predictions_5/mlp_layer_1_grad/ReluGrad
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/BiasAddGrad
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/group_deps
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/control_dependency
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/control_dependency_1
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/MatMul
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/MatMul_1
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_5/gradients/AddN
Optimizer_5/GradientDescent/learning_rate
Optimizer_5/GradientDescent/update_Parameters_5/Weights/Weight/ApplyGradientDescent
Optimizer_5/GradientDescent/update_Parameters_5/Bias/Bias/ApplyGradientDescent
Optimizer_5/GradientDescent
Equal_5
Accuracy_5/Cast
Accuracy_5/Const
Accuracy_5/Mean
Accuracy_5/train_accuracy_list/initial_value
Accuracy_5/train_accuracy_list
Accuracy_5/train_accuracy_list/Assign
Accuracy_5/train_accuracy_list/read
Accuracy_5/Assign/value
Accuracy_5/Assign
Accuracy_5/train_accuracy/tags
Accuracy_5/train_accuracy
Accuracy_5/Cast_1
Accuracy_5/Const_1
Accuracy_5/Mean_1
Accuracy_5/validate_accuracy_list/initial_value
Accuracy_5/validate_accuracy_list
Accuracy_5/validate_accuracy_list/Assign
Accuracy_5/validate_accuracy_list/read
Accuracy_5/Assign_1/value
Accuracy_5/Assign_1
Accuracy_5/validate_accuracy/tags
Accuracy_5/validate_accuracy
Accuracy_5/Cast_2
Accuracy_5/Const_2
Accuracy_5/Mean_2
Accuracy_5/test_accuracy_list/initial_value
Accuracy_5/test_accuracy_list
Accuracy_5/test_accuracy_list/Assign
Accuracy_5/test_accuracy_list/read
Accuracy_5/Assign_2/value
Accuracy_5/Assign_2
Accuracy_5/test_accuracy/tags
Accuracy_5/test_accuracy
init_5/NoOp
init_5/NoOp_1
init_5
Merge_5/MergeSummary
save_5/Const
save_5/SaveV2/tensor_names
save_5/SaveV2/shape_and_slices
save_5/SaveV2
save_5/control_dependency
save_5/RestoreV2/tensor_names
save_5/RestoreV2/shape_and_slices
save_5/RestoreV2
save_5/Assign
save_5/RestoreV2_1/tensor_names
save_5/RestoreV2_1/shape_and_slices
save_5/RestoreV2_1
save_5/Assign_1
save_5/RestoreV2_2/tensor_names
save_5/RestoreV2_2/shape_and_slices
save_5/RestoreV2_2
save_5/Assign_2
save_5/RestoreV2_3/tensor_names
save_5/RestoreV2_3/shape_and_slices
save_5/RestoreV2_3
save_5/Assign_3
save_5/RestoreV2_4/tensor_names
save_5/RestoreV2_4/shape_and_slices
save_5/RestoreV2_4
save_5/Assign_4
save_5/RestoreV2_5/tensor_names
save_5/RestoreV2_5/shape_and_slices
save_5/RestoreV2_5
save_5/Assign_5
save_5/RestoreV2_6/tensor_names
save_5/RestoreV2_6/shape_and_slices
save_5/RestoreV2_6
save_5/Assign_6
save_5/RestoreV2_7/tensor_names
save_5/RestoreV2_7/shape_and_slices
save_5/RestoreV2_7
save_5/Assign_7
save_5/RestoreV2_8/tensor_names
save_5/RestoreV2_8/shape_and_slices
save_5/RestoreV2_8
save_5/Assign_8
save_5/RestoreV2_9/tensor_names
save_5/RestoreV2_9/shape_and_slices
save_5/RestoreV2_9
save_5/Assign_9
save_5/RestoreV2_10/tensor_names
save_5/RestoreV2_10/shape_and_slices
save_5/RestoreV2_10
save_5/Assign_10
save_5/RestoreV2_11/tensor_names
save_5/RestoreV2_11/shape_and_slices
save_5/RestoreV2_11
save_5/Assign_11
save_5/RestoreV2_12/tensor_names
save_5/RestoreV2_12/shape_and_slices
save_5/RestoreV2_12
save_5/Assign_12
save_5/RestoreV2_13/tensor_names
save_5/RestoreV2_13/shape_and_slices
save_5/RestoreV2_13
save_5/Assign_13
save_5/RestoreV2_14/tensor_names
save_5/RestoreV2_14/shape_and_slices
save_5/RestoreV2_14
save_5/Assign_14
save_5/RestoreV2_15/tensor_names
save_5/RestoreV2_15/shape_and_slices
save_5/RestoreV2_15
save_5/Assign_15
save_5/RestoreV2_16/tensor_names
save_5/RestoreV2_16/shape_and_slices
save_5/RestoreV2_16
save_5/Assign_16
save_5/RestoreV2_17/tensor_names
save_5/RestoreV2_17/shape_and_slices
save_5/RestoreV2_17
save_5/Assign_17
save_5/RestoreV2_18/tensor_names
save_5/RestoreV2_18/shape_and_slices
save_5/RestoreV2_18
save_5/Assign_18
save_5/RestoreV2_19/tensor_names
save_5/RestoreV2_19/shape_and_slices
save_5/RestoreV2_19
save_5/Assign_19
save_5/RestoreV2_20/tensor_names
save_5/RestoreV2_20/shape_and_slices
save_5/RestoreV2_20
save_5/Assign_20
save_5/RestoreV2_21/tensor_names
save_5/RestoreV2_21/shape_and_slices
save_5/RestoreV2_21
save_5/Assign_21
save_5/RestoreV2_22/tensor_names
save_5/RestoreV2_22/shape_and_slices
save_5/RestoreV2_22
save_5/Assign_22
save_5/RestoreV2_23/tensor_names
save_5/RestoreV2_23/shape_and_slices
save_5/RestoreV2_23
save_5/Assign_23
save_5/RestoreV2_24/tensor_names
save_5/RestoreV2_24/shape_and_slices
save_5/RestoreV2_24
save_5/Assign_24
save_5/RestoreV2_25/tensor_names
save_5/RestoreV2_25/shape_and_slices
save_5/RestoreV2_25
save_5/Assign_25
save_5/RestoreV2_26/tensor_names
save_5/RestoreV2_26/shape_and_slices
save_5/RestoreV2_26
save_5/Assign_26
save_5/RestoreV2_27/tensor_names
save_5/RestoreV2_27/shape_and_slices
save_5/RestoreV2_27
save_5/Assign_27
save_5/RestoreV2_28/tensor_names
save_5/RestoreV2_28/shape_and_slices
save_5/RestoreV2_28
save_5/Assign_28
save_5/RestoreV2_29/tensor_names
save_5/RestoreV2_29/shape_and_slices
save_5/RestoreV2_29
save_5/Assign_29
save_5/RestoreV2_30/tensor_names
save_5/RestoreV2_30/shape_and_slices
save_5/RestoreV2_30
save_5/Assign_30
save_5/RestoreV2_31/tensor_names
save_5/RestoreV2_31/shape_and_slices
save_5/RestoreV2_31
save_5/Assign_31
save_5/RestoreV2_32/tensor_names
save_5/RestoreV2_32/shape_and_slices
save_5/RestoreV2_32
save_5/Assign_32
save_5/RestoreV2_33/tensor_names
save_5/RestoreV2_33/shape_and_slices
save_5/RestoreV2_33
save_5/Assign_33
save_5/RestoreV2_34/tensor_names
save_5/RestoreV2_34/shape_and_slices
save_5/RestoreV2_34
save_5/Assign_34
save_5/RestoreV2_35/tensor_names
save_5/RestoreV2_35/shape_and_slices
save_5/RestoreV2_35
save_5/Assign_35
save_5/RestoreV2_36/tensor_names
save_5/RestoreV2_36/shape_and_slices
save_5/RestoreV2_36
save_5/Assign_36
save_5/RestoreV2_37/tensor_names
save_5/RestoreV2_37/shape_and_slices
save_5/RestoreV2_37
save_5/Assign_37
save_5/RestoreV2_38/tensor_names
save_5/RestoreV2_38/shape_and_slices
save_5/RestoreV2_38
save_5/Assign_38
save_5/RestoreV2_39/tensor_names
save_5/RestoreV2_39/shape_and_slices
save_5/RestoreV2_39
save_5/Assign_39
save_5/RestoreV2_40/tensor_names
save_5/RestoreV2_40/shape_and_slices
save_5/RestoreV2_40
save_5/Assign_40
save_5/RestoreV2_41/tensor_names
save_5/RestoreV2_41/shape_and_slices
save_5/RestoreV2_41
save_5/Assign_41
save_5/RestoreV2_42/tensor_names
save_5/RestoreV2_42/shape_and_slices
save_5/RestoreV2_42
save_5/Assign_42
save_5/RestoreV2_43/tensor_names
save_5/RestoreV2_43/shape_and_slices
save_5/RestoreV2_43
save_5/Assign_43
save_5/RestoreV2_44/tensor_names
save_5/RestoreV2_44/shape_and_slices
save_5/RestoreV2_44
save_5/Assign_44
save_5/RestoreV2_45/tensor_names
save_5/RestoreV2_45/shape_and_slices
save_5/RestoreV2_45
save_5/Assign_45
save_5/RestoreV2_46/tensor_names
save_5/RestoreV2_46/shape_and_slices
save_5/RestoreV2_46
save_5/Assign_46
save_5/RestoreV2_47/tensor_names
save_5/RestoreV2_47/shape_and_slices
save_5/RestoreV2_47
save_5/Assign_47
save_5/RestoreV2_48/tensor_names
save_5/RestoreV2_48/shape_and_slices
save_5/RestoreV2_48
save_5/Assign_48
save_5/RestoreV2_49/tensor_names
save_5/RestoreV2_49/shape_and_slices
save_5/RestoreV2_49
save_5/Assign_49
save_5/RestoreV2_50/tensor_names
save_5/RestoreV2_50/shape_and_slices
save_5/RestoreV2_50
save_5/Assign_50
save_5/RestoreV2_51/tensor_names
save_5/RestoreV2_51/shape_and_slices
save_5/RestoreV2_51
save_5/Assign_51
save_5/RestoreV2_52/tensor_names
save_5/RestoreV2_52/shape_and_slices
save_5/RestoreV2_52
save_5/Assign_52
save_5/RestoreV2_53/tensor_names
save_5/RestoreV2_53/shape_and_slices
save_5/RestoreV2_53
save_5/Assign_53
save_5/restore_all/NoOp
save_5/restore_all/NoOp_1
save_5/restore_all
Output Logits  : Tensor("Predictions_2/Softmax:0", shape=(?, 10), dtype=float32, device=/device:GPU:0)
Entropy        : Tensor("Cross_Entropy_2/Reshape_2:0", shape=(?,), dtype=float32, device=/device:GPU:0)
Predictions    : Tensor("Equal_2:0", shape=(?,), dtype=bool, device=/device:GPU:0) 
Optimizer      : name: "Optimizer_2/GradientDescent"
op: "NoOp"
input: "^Optimizer_2/GradientDescent/update_Parameters_2/Weights/Weight/ApplyGradientDescent"
input: "^Optimizer_2/GradientDescent/update_Parameters_2/Bias/Bias/ApplyGradientDescent"
device: "/device:GPU:0"
 
>> Model info
Train loss     : Tensor("Loss_Function_2/Add:0", shape=(), dtype=float32, device=/device:GPU:0)
Train accuracy : Tensor("Accuracy_2/Mean:0", shape=(), dtype=float32, device=/device:GPU:0)
Val. loss      : Tensor("Loss_Function_2/Add:0", shape=(), dtype=float32, device=/device:GPU:0)
Val. accuracy  : Tensor("Accuracy_2/Mean_1:0", shape=(), dtype=float32, device=/device:GPU:0)
>> Summary params
Train loss     : Tensor("Loss_Function_2/train_error:0", shape=(), dtype=string, device=/device:GPU:0)
Train accuracy : Tensor("Accuracy_2/train_accuracy:0", shape=(), dtype=string, device=/device:GPU:0)
Val. loss      : Tensor("Loss_Function_2/validate_error:0", shape=(), dtype=string, device=/device:GPU:0)
Val. accuracy  : Tensor("Accuracy_2/validate_accuracy:0", shape=(), dtype=string, device=/device:GPU:0)
Parameters for Linear classifier
>> Input Parameters
Input                  : Tensor("Inputs_3/Data/X_input:0", shape=(?, 3072), dtype=float32, device=/device:GPU:0) 
True one hot labels    : Tensor("Inputs_3/Train_Labels/y_true_one_hot_label:0", shape=(?, 10), dtype=float32, device=/device:GPU:0) 
True class             : Tensor("Inputs_3/Train_Labels/y_true_class:0", shape=(?,), dtype=int64, device=/device:GPU:0) 
Predict one hot labels : Tensor("Predictions_3/one_hot:0", shape=(?, 10), dtype=float32, device=/device:GPU:0) 
Predict class          : Tensor("Predictions_3/predict_class:0", shape=(?,), dtype=int64, device=/device:GPU:0) 
>> Model params
last_successful_epoch/initial_value
last_successful_epoch
last_successful_epoch/Assign
last_successful_epoch/read
add/y
add
assign_updated_epoch
Inputs/Data/X_input
Inputs/Train_Labels/y_true_one_hot_label
Inputs/Train_Labels/y_true_class
Parameters/Weights/zeros
Parameters/Weights/Weight
Parameters/Weights/Weight/Assign
Parameters/Weights/Weight/read
Parameters/Bias/ones
Parameters/Bias/Bias
Parameters/Bias/Bias/Assign
Parameters/Bias/Bias/read
Predictions/mlp_layer/MatMul
Predictions/mlp_layer
Predictions/mlp_layer_1
Predictions/Softmax
Predictions/predict_class/dimension
Predictions/predict_class
Predictions/one_hot/Const
Predictions/one_hot/Const_1
Predictions/one_hot/depth
Predictions/one_hot/on_value
Predictions/one_hot/off_value
Predictions/one_hot
Cross_Entropy/Rank
Cross_Entropy/Shape
Cross_Entropy/Rank_1
Cross_Entropy/Shape_1
Cross_Entropy/Sub/y
Cross_Entropy/Sub
Cross_Entropy/Slice/begin
Cross_Entropy/Slice/size
Cross_Entropy/Slice
Cross_Entropy/concat/values_0
Cross_Entropy/concat/axis
Cross_Entropy/concat
Cross_Entropy/Reshape
Cross_Entropy/Rank_2
Cross_Entropy/Shape_2
Cross_Entropy/Sub_1/y
Cross_Entropy/Sub_1
Cross_Entropy/Slice_1/begin
Cross_Entropy/Slice_1/size
Cross_Entropy/Slice_1
Cross_Entropy/concat_1/values_0
Cross_Entropy/concat_1/axis
Cross_Entropy/concat_1
Cross_Entropy/Reshape_1
Cross_Entropy/SoftmaxCrossEntropyWithLogits
Cross_Entropy/Sub_2/y
Cross_Entropy/Sub_2
Cross_Entropy/Slice_2/begin
Cross_Entropy/Slice_2/size
Cross_Entropy/Slice_2
Cross_Entropy/Reshape_2
Loss_Function/Const
Loss_Function/Square
Loss_Function/Const_1
Loss_Function/Mean
Loss_Function/Const_2
Loss_Function/Mean_1
Loss_Function/Mul
Loss_Function/Add
Loss_Function/train_loss_list/initial_value
Loss_Function/train_loss_list
Loss_Function/train_loss_list/Assign
Loss_Function/train_loss_list/read
Loss_Function/Assign/value
Loss_Function/Assign
Loss_Function/train_error/tags
Loss_Function/train_error
Loss_Function/validate_loss_list/initial_value
Loss_Function/validate_loss_list
Loss_Function/validate_loss_list/Assign
Loss_Function/validate_loss_list/read
Loss_Function/Assign_1/value
Loss_Function/Assign_1
Loss_Function/validate_error/tags
Loss_Function/validate_error
Optimizer/learning_rate_progress/initial_value
Optimizer/learning_rate_progress
Optimizer/learning_rate_progress/Assign
Optimizer/learning_rate_progress/read
Optimizer/Assign/value
Optimizer/Assign
Optimizer/learning_rate/tags
Optimizer/learning_rate/values
Optimizer/learning_rate
Optimizer/gradients/Shape
Optimizer/gradients/Const
Optimizer/gradients/Fill
Optimizer/gradients/Loss_Function/Add_grad/Shape
Optimizer/gradients/Loss_Function/Add_grad/Shape_1
Optimizer/gradients/Loss_Function/Add_grad/BroadcastGradientArgs
Optimizer/gradients/Loss_Function/Add_grad/Sum
Optimizer/gradients/Loss_Function/Add_grad/Reshape
Optimizer/gradients/Loss_Function/Add_grad/Sum_1
Optimizer/gradients/Loss_Function/Add_grad/Reshape_1
Optimizer/gradients/Loss_Function/Add_grad/tuple/group_deps
Optimizer/gradients/Loss_Function/Add_grad/tuple/control_dependency
Optimizer/gradients/Loss_Function/Add_grad/tuple/control_dependency_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Reshape/shape
Optimizer/gradients/Loss_Function/Mean_1_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape
Optimizer/gradients/Loss_Function/Mean_1_grad/Tile
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape_2
Optimizer/gradients/Loss_Function/Mean_1_grad/Const
Optimizer/gradients/Loss_Function/Mean_1_grad/Prod
Optimizer/gradients/Loss_Function/Mean_1_grad/Const_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Prod_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Maximum/y
Optimizer/gradients/Loss_Function/Mean_1_grad/Maximum
Optimizer/gradients/Loss_Function/Mean_1_grad/floordiv
Optimizer/gradients/Loss_Function/Mean_1_grad/Cast
Optimizer/gradients/Loss_Function/Mean_1_grad/truediv
Optimizer/gradients/Loss_Function/Mul_grad/Shape
Optimizer/gradients/Loss_Function/Mul_grad/Shape_1
Optimizer/gradients/Loss_Function/Mul_grad/BroadcastGradientArgs
Optimizer/gradients/Loss_Function/Mul_grad/mul
Optimizer/gradients/Loss_Function/Mul_grad/Sum
Optimizer/gradients/Loss_Function/Mul_grad/Reshape
Optimizer/gradients/Loss_Function/Mul_grad/mul_1
Optimizer/gradients/Loss_Function/Mul_grad/Sum_1
Optimizer/gradients/Loss_Function/Mul_grad/Reshape_1
Optimizer/gradients/Loss_Function/Mul_grad/tuple/group_deps
Optimizer/gradients/Loss_Function/Mul_grad/tuple/control_dependency
Optimizer/gradients/Loss_Function/Mul_grad/tuple/control_dependency_1
Optimizer/gradients/Cross_Entropy/Reshape_2_grad/Shape
Optimizer/gradients/Cross_Entropy/Reshape_2_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_grad/Reshape/shape
Optimizer/gradients/Loss_Function/Mean_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_grad/Tile/multiples
Optimizer/gradients/Loss_Function/Mean_grad/Tile
Optimizer/gradients/Loss_Function/Mean_grad/Shape
Optimizer/gradients/Loss_Function/Mean_grad/Shape_1
Optimizer/gradients/Loss_Function/Mean_grad/Const
Optimizer/gradients/Loss_Function/Mean_grad/Prod
Optimizer/gradients/Loss_Function/Mean_grad/Const_1
Optimizer/gradients/Loss_Function/Mean_grad/Prod_1
Optimizer/gradients/Loss_Function/Mean_grad/Maximum/y
Optimizer/gradients/Loss_Function/Mean_grad/Maximum
Optimizer/gradients/Loss_Function/Mean_grad/floordiv
Optimizer/gradients/Loss_Function/Mean_grad/Cast
Optimizer/gradients/Loss_Function/Mean_grad/truediv
Optimizer/gradients/zeros_like
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer/gradients/Loss_Function/Square_grad/mul/x
Optimizer/gradients/Loss_Function/Square_grad/mul
Optimizer/gradients/Loss_Function/Square_grad/mul_1
Optimizer/gradients/Cross_Entropy/Reshape_grad/Shape
Optimizer/gradients/Cross_Entropy/Reshape_grad/Reshape
Optimizer/gradients/Predictions/Softmax_grad/mul
Optimizer/gradients/Predictions/Softmax_grad/Sum/reduction_indices
Optimizer/gradients/Predictions/Softmax_grad/Sum
Optimizer/gradients/Predictions/Softmax_grad/Reshape/shape
Optimizer/gradients/Predictions/Softmax_grad/Reshape
Optimizer/gradients/Predictions/Softmax_grad/sub
Optimizer/gradients/Predictions/Softmax_grad/mul_1
Optimizer/gradients/Predictions/mlp_layer_1_grad/ReluGrad
Optimizer/gradients/Predictions/mlp_layer_grad/BiasAddGrad
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/group_deps
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/control_dependency
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/control_dependency_1
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/MatMul
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/MatMul_1
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer/gradients/AddN
Optimizer/GradientDescent/learning_rate
Optimizer/GradientDescent/update_Parameters/Weights/Weight/ApplyGradientDescent
Optimizer/GradientDescent/update_Parameters/Bias/Bias/ApplyGradientDescent
Optimizer/GradientDescent
Equal
Accuracy/Cast
Accuracy/Const
Accuracy/Mean
Accuracy/train_accuracy_list/initial_value
Accuracy/train_accuracy_list
Accuracy/train_accuracy_list/Assign
Accuracy/train_accuracy_list/read
Accuracy/Assign/value
Accuracy/Assign
Accuracy/train_accuracy/tags
Accuracy/train_accuracy
Accuracy/Cast_1
Accuracy/Const_1
Accuracy/Mean_1
Accuracy/validate_accuracy_list/initial_value
Accuracy/validate_accuracy_list
Accuracy/validate_accuracy_list/Assign
Accuracy/validate_accuracy_list/read
Accuracy/Assign_1/value
Accuracy/Assign_1
Accuracy/validate_accuracy/tags
Accuracy/validate_accuracy
Accuracy/Cast_2
Accuracy/Const_2
Accuracy/Mean_2
Accuracy/test_accuracy_list/initial_value
Accuracy/test_accuracy_list
Accuracy/test_accuracy_list/Assign
Accuracy/test_accuracy_list/read
Accuracy/Assign_2/value
Accuracy/Assign_2
Accuracy/test_accuracy/tags
Accuracy/test_accuracy
init/NoOp
init/NoOp_1
init
Merge/MergeSummary
save/Const
save/SaveV2/tensor_names
save/SaveV2/shape_and_slices
save/SaveV2
save/control_dependency
save/RestoreV2/tensor_names
save/RestoreV2/shape_and_slices
save/RestoreV2
save/Assign
save/RestoreV2_1/tensor_names
save/RestoreV2_1/shape_and_slices
save/RestoreV2_1
save/Assign_1
save/RestoreV2_2/tensor_names
save/RestoreV2_2/shape_and_slices
save/RestoreV2_2
save/Assign_2
save/RestoreV2_3/tensor_names
save/RestoreV2_3/shape_and_slices
save/RestoreV2_3
save/Assign_3
save/RestoreV2_4/tensor_names
save/RestoreV2_4/shape_and_slices
save/RestoreV2_4
save/Assign_4
save/RestoreV2_5/tensor_names
save/RestoreV2_5/shape_and_slices
save/RestoreV2_5
save/Assign_5
save/RestoreV2_6/tensor_names
save/RestoreV2_6/shape_and_slices
save/RestoreV2_6
save/Assign_6
save/RestoreV2_7/tensor_names
save/RestoreV2_7/shape_and_slices
save/RestoreV2_7
save/Assign_7
save/RestoreV2_8/tensor_names
save/RestoreV2_8/shape_and_slices
save/RestoreV2_8
save/Assign_8
save/restore_all/NoOp
save/restore_all/NoOp_1
save/restore_all
last_successful_epoch_1/initial_value
last_successful_epoch_1
last_successful_epoch_1/Assign
last_successful_epoch_1/read
add_1/y
add_1
assign_updated_epoch_1
Inputs_1/Data/X_input
Inputs_1/Train_Labels/y_true_one_hot_label
Inputs_1/Train_Labels/y_true_class
Parameters_1/Weights/zeros
Parameters_1/Weights/Weight
Parameters_1/Weights/Weight/Assign
Parameters_1/Weights/Weight/read
Parameters_1/Bias/ones
Parameters_1/Bias/Bias
Parameters_1/Bias/Bias/Assign
Parameters_1/Bias/Bias/read
Predictions_1/mlp_layer/MatMul
Predictions_1/mlp_layer
Predictions_1/mlp_layer_1
Predictions_1/Softmax
Predictions_1/predict_class/dimension
Predictions_1/predict_class
Predictions_1/one_hot/Const
Predictions_1/one_hot/Const_1
Predictions_1/one_hot/depth
Predictions_1/one_hot/on_value
Predictions_1/one_hot/off_value
Predictions_1/one_hot
Cross_Entropy_1/Rank
Cross_Entropy_1/Shape
Cross_Entropy_1/Rank_1
Cross_Entropy_1/Shape_1
Cross_Entropy_1/Sub/y
Cross_Entropy_1/Sub
Cross_Entropy_1/Slice/begin
Cross_Entropy_1/Slice/size
Cross_Entropy_1/Slice
Cross_Entropy_1/concat/values_0
Cross_Entropy_1/concat/axis
Cross_Entropy_1/concat
Cross_Entropy_1/Reshape
Cross_Entropy_1/Rank_2
Cross_Entropy_1/Shape_2
Cross_Entropy_1/Sub_1/y
Cross_Entropy_1/Sub_1
Cross_Entropy_1/Slice_1/begin
Cross_Entropy_1/Slice_1/size
Cross_Entropy_1/Slice_1
Cross_Entropy_1/concat_1/values_0
Cross_Entropy_1/concat_1/axis
Cross_Entropy_1/concat_1
Cross_Entropy_1/Reshape_1
Cross_Entropy_1/SoftmaxCrossEntropyWithLogits
Cross_Entropy_1/Sub_2/y
Cross_Entropy_1/Sub_2
Cross_Entropy_1/Slice_2/begin
Cross_Entropy_1/Slice_2/size
Cross_Entropy_1/Slice_2
Cross_Entropy_1/Reshape_2
Loss_Function_1/Const
Loss_Function_1/Square
Loss_Function_1/Const_1
Loss_Function_1/Mean
Loss_Function_1/Const_2
Loss_Function_1/Mean_1
Loss_Function_1/Mul
Loss_Function_1/Add
Loss_Function_1/train_loss_list/initial_value
Loss_Function_1/train_loss_list
Loss_Function_1/train_loss_list/Assign
Loss_Function_1/train_loss_list/read
Loss_Function_1/Assign/value
Loss_Function_1/Assign
Loss_Function_1/train_error/tags
Loss_Function_1/train_error
Loss_Function_1/validate_loss_list/initial_value
Loss_Function_1/validate_loss_list
Loss_Function_1/validate_loss_list/Assign
Loss_Function_1/validate_loss_list/read
Loss_Function_1/Assign_1/value
Loss_Function_1/Assign_1
Loss_Function_1/validate_error/tags
Loss_Function_1/validate_error
Optimizer_1/learning_rate_progress/initial_value
Optimizer_1/learning_rate_progress
Optimizer_1/learning_rate_progress/Assign
Optimizer_1/learning_rate_progress/read
Optimizer_1/Assign/value
Optimizer_1/Assign
Optimizer_1/learning_rate/tags
Optimizer_1/learning_rate/values
Optimizer_1/learning_rate
Optimizer_1/gradients/Shape
Optimizer_1/gradients/Const
Optimizer_1/gradients/Fill
Optimizer_1/gradients/Loss_Function_1/Add_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Add_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/BroadcastGradientArgs
Optimizer_1/gradients/Loss_Function_1/Add_grad/Sum
Optimizer_1/gradients/Loss_Function_1/Add_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Add_grad/Sum_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/Reshape_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/group_deps
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/control_dependency
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/control_dependency_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Reshape/shape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Tile
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape_2
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Const
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Prod
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Const_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Prod_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Maximum/y
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Maximum
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/floordiv
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Cast
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/truediv
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/BroadcastGradientArgs
Optimizer_1/gradients/Loss_Function_1/Mul_grad/mul
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Sum
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mul_grad/mul_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Sum_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Reshape_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/group_deps
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/control_dependency
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/control_dependency_1
Optimizer_1/gradients/Cross_Entropy_1/Reshape_2_grad/Shape
Optimizer_1/gradients/Cross_Entropy_1/Reshape_2_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Reshape/shape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Tile/multiples
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Tile
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Const
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Prod
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Const_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Prod_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Maximum/y
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Maximum
Optimizer_1/gradients/Loss_Function_1/Mean_grad/floordiv
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Cast
Optimizer_1/gradients/Loss_Function_1/Mean_grad/truediv
Optimizer_1/gradients/zeros_like
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul/x
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul_1
Optimizer_1/gradients/Cross_Entropy_1/Reshape_grad/Shape
Optimizer_1/gradients/Cross_Entropy_1/Reshape_grad/Reshape
Optimizer_1/gradients/Predictions_1/Softmax_grad/mul
Optimizer_1/gradients/Predictions_1/Softmax_grad/Sum/reduction_indices
Optimizer_1/gradients/Predictions_1/Softmax_grad/Sum
Optimizer_1/gradients/Predictions_1/Softmax_grad/Reshape/shape
Optimizer_1/gradients/Predictions_1/Softmax_grad/Reshape
Optimizer_1/gradients/Predictions_1/Softmax_grad/sub
Optimizer_1/gradients/Predictions_1/Softmax_grad/mul_1
Optimizer_1/gradients/Predictions_1/mlp_layer_1_grad/ReluGrad
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/BiasAddGrad
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/group_deps
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/control_dependency
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/control_dependency_1
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/MatMul
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/MatMul_1
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_1/gradients/AddN
Optimizer_1/GradientDescent/learning_rate
Optimizer_1/GradientDescent/update_Parameters_1/Weights/Weight/ApplyGradientDescent
Optimizer_1/GradientDescent/update_Parameters_1/Bias/Bias/ApplyGradientDescent
Optimizer_1/GradientDescent
Equal_1
Accuracy_1/Cast
Accuracy_1/Const
Accuracy_1/Mean
Accuracy_1/train_accuracy_list/initial_value
Accuracy_1/train_accuracy_list
Accuracy_1/train_accuracy_list/Assign
Accuracy_1/train_accuracy_list/read
Accuracy_1/Assign/value
Accuracy_1/Assign
Accuracy_1/train_accuracy/tags
Accuracy_1/train_accuracy
Accuracy_1/Cast_1
Accuracy_1/Const_1
Accuracy_1/Mean_1
Accuracy_1/validate_accuracy_list/initial_value
Accuracy_1/validate_accuracy_list
Accuracy_1/validate_accuracy_list/Assign
Accuracy_1/validate_accuracy_list/read
Accuracy_1/Assign_1/value
Accuracy_1/Assign_1
Accuracy_1/validate_accuracy/tags
Accuracy_1/validate_accuracy
Accuracy_1/Cast_2
Accuracy_1/Const_2
Accuracy_1/Mean_2
Accuracy_1/test_accuracy_list/initial_value
Accuracy_1/test_accuracy_list
Accuracy_1/test_accuracy_list/Assign
Accuracy_1/test_accuracy_list/read
Accuracy_1/Assign_2/value
Accuracy_1/Assign_2
Accuracy_1/test_accuracy/tags
Accuracy_1/test_accuracy
init_1/NoOp
init_1/NoOp_1
init_1
Merge_1/MergeSummary
save_1/Const
save_1/SaveV2/tensor_names
save_1/SaveV2/shape_and_slices
save_1/SaveV2
save_1/control_dependency
save_1/RestoreV2/tensor_names
save_1/RestoreV2/shape_and_slices
save_1/RestoreV2
save_1/Assign
save_1/RestoreV2_1/tensor_names
save_1/RestoreV2_1/shape_and_slices
save_1/RestoreV2_1
save_1/Assign_1
save_1/RestoreV2_2/tensor_names
save_1/RestoreV2_2/shape_and_slices
save_1/RestoreV2_2
save_1/Assign_2
save_1/RestoreV2_3/tensor_names
save_1/RestoreV2_3/shape_and_slices
save_1/RestoreV2_3
save_1/Assign_3
save_1/RestoreV2_4/tensor_names
save_1/RestoreV2_4/shape_and_slices
save_1/RestoreV2_4
save_1/Assign_4
save_1/RestoreV2_5/tensor_names
save_1/RestoreV2_5/shape_and_slices
save_1/RestoreV2_5
save_1/Assign_5
save_1/RestoreV2_6/tensor_names
save_1/RestoreV2_6/shape_and_slices
save_1/RestoreV2_6
save_1/Assign_6
save_1/RestoreV2_7/tensor_names
save_1/RestoreV2_7/shape_and_slices
save_1/RestoreV2_7
save_1/Assign_7
save_1/RestoreV2_8/tensor_names
save_1/RestoreV2_8/shape_and_slices
save_1/RestoreV2_8
save_1/Assign_8
save_1/RestoreV2_9/tensor_names
save_1/RestoreV2_9/shape_and_slices
save_1/RestoreV2_9
save_1/Assign_9
save_1/RestoreV2_10/tensor_names
save_1/RestoreV2_10/shape_and_slices
save_1/RestoreV2_10
save_1/Assign_10
save_1/RestoreV2_11/tensor_names
save_1/RestoreV2_11/shape_and_slices
save_1/RestoreV2_11
save_1/Assign_11
save_1/RestoreV2_12/tensor_names
save_1/RestoreV2_12/shape_and_slices
save_1/RestoreV2_12
save_1/Assign_12
save_1/RestoreV2_13/tensor_names
save_1/RestoreV2_13/shape_and_slices
save_1/RestoreV2_13
save_1/Assign_13
save_1/RestoreV2_14/tensor_names
save_1/RestoreV2_14/shape_and_slices
save_1/RestoreV2_14
save_1/Assign_14
save_1/RestoreV2_15/tensor_names
save_1/RestoreV2_15/shape_and_slices
save_1/RestoreV2_15
save_1/Assign_15
save_1/RestoreV2_16/tensor_names
save_1/RestoreV2_16/shape_and_slices
save_1/RestoreV2_16
save_1/Assign_16
save_1/RestoreV2_17/tensor_names
save_1/RestoreV2_17/shape_and_slices
save_1/RestoreV2_17
save_1/Assign_17
save_1/restore_all/NoOp
save_1/restore_all/NoOp_1
save_1/restore_all
last_successful_epoch_2/initial_value
last_successful_epoch_2
last_successful_epoch_2/Assign
last_successful_epoch_2/read
add_2/y
add_2
assign_updated_epoch_2
Inputs_2/Data/X_input
Inputs_2/Train_Labels/y_true_one_hot_label
Inputs_2/Train_Labels/y_true_class
Parameters_2/Weights/zeros
Parameters_2/Weights/Weight
Parameters_2/Weights/Weight/Assign
Parameters_2/Weights/Weight/read
Parameters_2/Bias/ones
Parameters_2/Bias/Bias
Parameters_2/Bias/Bias/Assign
Parameters_2/Bias/Bias/read
Predictions_2/mlp_layer/MatMul
Predictions_2/mlp_layer
Predictions_2/mlp_layer_1
Predictions_2/Softmax
Predictions_2/predict_class/dimension
Predictions_2/predict_class
Predictions_2/one_hot/Const
Predictions_2/one_hot/Const_1
Predictions_2/one_hot/depth
Predictions_2/one_hot/on_value
Predictions_2/one_hot/off_value
Predictions_2/one_hot
Cross_Entropy_2/Rank
Cross_Entropy_2/Shape
Cross_Entropy_2/Rank_1
Cross_Entropy_2/Shape_1
Cross_Entropy_2/Sub/y
Cross_Entropy_2/Sub
Cross_Entropy_2/Slice/begin
Cross_Entropy_2/Slice/size
Cross_Entropy_2/Slice
Cross_Entropy_2/concat/values_0
Cross_Entropy_2/concat/axis
Cross_Entropy_2/concat
Cross_Entropy_2/Reshape
Cross_Entropy_2/Rank_2
Cross_Entropy_2/Shape_2
Cross_Entropy_2/Sub_1/y
Cross_Entropy_2/Sub_1
Cross_Entropy_2/Slice_1/begin
Cross_Entropy_2/Slice_1/size
Cross_Entropy_2/Slice_1
Cross_Entropy_2/concat_1/values_0
Cross_Entropy_2/concat_1/axis
Cross_Entropy_2/concat_1
Cross_Entropy_2/Reshape_1
Cross_Entropy_2/SoftmaxCrossEntropyWithLogits
Cross_Entropy_2/Sub_2/y
Cross_Entropy_2/Sub_2
Cross_Entropy_2/Slice_2/begin
Cross_Entropy_2/Slice_2/size
Cross_Entropy_2/Slice_2
Cross_Entropy_2/Reshape_2
Loss_Function_2/Const
Loss_Function_2/Square
Loss_Function_2/Const_1
Loss_Function_2/Mean
Loss_Function_2/Const_2
Loss_Function_2/Mean_1
Loss_Function_2/Mul
Loss_Function_2/Add
Loss_Function_2/train_loss_list/initial_value
Loss_Function_2/train_loss_list
Loss_Function_2/train_loss_list/Assign
Loss_Function_2/train_loss_list/read
Loss_Function_2/Assign/value
Loss_Function_2/Assign
Loss_Function_2/train_error/tags
Loss_Function_2/train_error
Loss_Function_2/validate_loss_list/initial_value
Loss_Function_2/validate_loss_list
Loss_Function_2/validate_loss_list/Assign
Loss_Function_2/validate_loss_list/read
Loss_Function_2/Assign_1/value
Loss_Function_2/Assign_1
Loss_Function_2/validate_error/tags
Loss_Function_2/validate_error
Optimizer_2/learning_rate_progress/initial_value
Optimizer_2/learning_rate_progress
Optimizer_2/learning_rate_progress/Assign
Optimizer_2/learning_rate_progress/read
Optimizer_2/Assign/value
Optimizer_2/Assign
Optimizer_2/learning_rate/tags
Optimizer_2/learning_rate/values
Optimizer_2/learning_rate
Optimizer_2/gradients/Shape
Optimizer_2/gradients/Const
Optimizer_2/gradients/Fill
Optimizer_2/gradients/Loss_Function_2/Add_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Add_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/BroadcastGradientArgs
Optimizer_2/gradients/Loss_Function_2/Add_grad/Sum
Optimizer_2/gradients/Loss_Function_2/Add_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Add_grad/Sum_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/Reshape_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/group_deps
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/control_dependency
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/control_dependency_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Reshape/shape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Tile
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape_2
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Const
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Prod
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Const_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Prod_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Maximum/y
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Maximum
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/floordiv
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Cast
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/truediv
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/BroadcastGradientArgs
Optimizer_2/gradients/Loss_Function_2/Mul_grad/mul
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Sum
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mul_grad/mul_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Sum_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Reshape_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/group_deps
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/control_dependency
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/control_dependency_1
Optimizer_2/gradients/Cross_Entropy_2/Reshape_2_grad/Shape
Optimizer_2/gradients/Cross_Entropy_2/Reshape_2_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Reshape/shape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Tile/multiples
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Tile
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Const
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Prod
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Const_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Prod_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Maximum/y
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Maximum
Optimizer_2/gradients/Loss_Function_2/Mean_grad/floordiv
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Cast
Optimizer_2/gradients/Loss_Function_2/Mean_grad/truediv
Optimizer_2/gradients/zeros_like
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul/x
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul_1
Optimizer_2/gradients/Cross_Entropy_2/Reshape_grad/Shape
Optimizer_2/gradients/Cross_Entropy_2/Reshape_grad/Reshape
Optimizer_2/gradients/Predictions_2/Softmax_grad/mul
Optimizer_2/gradients/Predictions_2/Softmax_grad/Sum/reduction_indices
Optimizer_2/gradients/Predictions_2/Softmax_grad/Sum
Optimizer_2/gradients/Predictions_2/Softmax_grad/Reshape/shape
Optimizer_2/gradients/Predictions_2/Softmax_grad/Reshape
Optimizer_2/gradients/Predictions_2/Softmax_grad/sub
Optimizer_2/gradients/Predictions_2/Softmax_grad/mul_1
Optimizer_2/gradients/Predictions_2/mlp_layer_1_grad/ReluGrad
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/BiasAddGrad
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/group_deps
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/control_dependency
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/control_dependency_1
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/MatMul
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/MatMul_1
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_2/gradients/AddN
Optimizer_2/GradientDescent/learning_rate
Optimizer_2/GradientDescent/update_Parameters_2/Weights/Weight/ApplyGradientDescent
Optimizer_2/GradientDescent/update_Parameters_2/Bias/Bias/ApplyGradientDescent
Optimizer_2/GradientDescent
Equal_2
Accuracy_2/Cast
Accuracy_2/Const
Accuracy_2/Mean
Accuracy_2/train_accuracy_list/initial_value
Accuracy_2/train_accuracy_list
Accuracy_2/train_accuracy_list/Assign
Accuracy_2/train_accuracy_list/read
Accuracy_2/Assign/value
Accuracy_2/Assign
Accuracy_2/train_accuracy/tags
Accuracy_2/train_accuracy
Accuracy_2/Cast_1
Accuracy_2/Const_1
Accuracy_2/Mean_1
Accuracy_2/validate_accuracy_list/initial_value
Accuracy_2/validate_accuracy_list
Accuracy_2/validate_accuracy_list/Assign
Accuracy_2/validate_accuracy_list/read
Accuracy_2/Assign_1/value
Accuracy_2/Assign_1
Accuracy_2/validate_accuracy/tags
Accuracy_2/validate_accuracy
Accuracy_2/Cast_2
Accuracy_2/Const_2
Accuracy_2/Mean_2
Accuracy_2/test_accuracy_list/initial_value
Accuracy_2/test_accuracy_list
Accuracy_2/test_accuracy_list/Assign
Accuracy_2/test_accuracy_list/read
Accuracy_2/Assign_2/value
Accuracy_2/Assign_2
Accuracy_2/test_accuracy/tags
Accuracy_2/test_accuracy
init_2/NoOp
init_2/NoOp_1
init_2
Merge_2/MergeSummary
save_2/Const
save_2/SaveV2/tensor_names
save_2/SaveV2/shape_and_slices
save_2/SaveV2
save_2/control_dependency
save_2/RestoreV2/tensor_names
save_2/RestoreV2/shape_and_slices
save_2/RestoreV2
save_2/Assign
save_2/RestoreV2_1/tensor_names
save_2/RestoreV2_1/shape_and_slices
save_2/RestoreV2_1
save_2/Assign_1
save_2/RestoreV2_2/tensor_names
save_2/RestoreV2_2/shape_and_slices
save_2/RestoreV2_2
save_2/Assign_2
save_2/RestoreV2_3/tensor_names
save_2/RestoreV2_3/shape_and_slices
save_2/RestoreV2_3
save_2/Assign_3
save_2/RestoreV2_4/tensor_names
save_2/RestoreV2_4/shape_and_slices
save_2/RestoreV2_4
save_2/Assign_4
save_2/RestoreV2_5/tensor_names
save_2/RestoreV2_5/shape_and_slices
save_2/RestoreV2_5
save_2/Assign_5
save_2/RestoreV2_6/tensor_names
save_2/RestoreV2_6/shape_and_slices
save_2/RestoreV2_6
save_2/Assign_6
save_2/RestoreV2_7/tensor_names
save_2/RestoreV2_7/shape_and_slices
save_2/RestoreV2_7
save_2/Assign_7
save_2/RestoreV2_8/tensor_names
save_2/RestoreV2_8/shape_and_slices
save_2/RestoreV2_8
save_2/Assign_8
save_2/RestoreV2_9/tensor_names
save_2/RestoreV2_9/shape_and_slices
save_2/RestoreV2_9
save_2/Assign_9
save_2/RestoreV2_10/tensor_names
save_2/RestoreV2_10/shape_and_slices
save_2/RestoreV2_10
save_2/Assign_10
save_2/RestoreV2_11/tensor_names
save_2/RestoreV2_11/shape_and_slices
save_2/RestoreV2_11
save_2/Assign_11
save_2/RestoreV2_12/tensor_names
save_2/RestoreV2_12/shape_and_slices
save_2/RestoreV2_12
save_2/Assign_12
save_2/RestoreV2_13/tensor_names
save_2/RestoreV2_13/shape_and_slices
save_2/RestoreV2_13
save_2/Assign_13
save_2/RestoreV2_14/tensor_names
save_2/RestoreV2_14/shape_and_slices
save_2/RestoreV2_14
save_2/Assign_14
save_2/RestoreV2_15/tensor_names
save_2/RestoreV2_15/shape_and_slices
save_2/RestoreV2_15
save_2/Assign_15
save_2/RestoreV2_16/tensor_names
save_2/RestoreV2_16/shape_and_slices
save_2/RestoreV2_16
save_2/Assign_16
save_2/RestoreV2_17/tensor_names
save_2/RestoreV2_17/shape_and_slices
save_2/RestoreV2_17
save_2/Assign_17
save_2/RestoreV2_18/tensor_names
save_2/RestoreV2_18/shape_and_slices
save_2/RestoreV2_18
save_2/Assign_18
save_2/RestoreV2_19/tensor_names
save_2/RestoreV2_19/shape_and_slices
save_2/RestoreV2_19
save_2/Assign_19
save_2/RestoreV2_20/tensor_names
save_2/RestoreV2_20/shape_and_slices
save_2/RestoreV2_20
save_2/Assign_20
save_2/RestoreV2_21/tensor_names
save_2/RestoreV2_21/shape_and_slices
save_2/RestoreV2_21
save_2/Assign_21
save_2/RestoreV2_22/tensor_names
save_2/RestoreV2_22/shape_and_slices
save_2/RestoreV2_22
save_2/Assign_22
save_2/RestoreV2_23/tensor_names
save_2/RestoreV2_23/shape_and_slices
save_2/RestoreV2_23
save_2/Assign_23
save_2/RestoreV2_24/tensor_names
save_2/RestoreV2_24/shape_and_slices
save_2/RestoreV2_24
save_2/Assign_24
save_2/RestoreV2_25/tensor_names
save_2/RestoreV2_25/shape_and_slices
save_2/RestoreV2_25
save_2/Assign_25
save_2/RestoreV2_26/tensor_names
save_2/RestoreV2_26/shape_and_slices
save_2/RestoreV2_26
save_2/Assign_26
save_2/restore_all/NoOp
save_2/restore_all/NoOp_1
save_2/restore_all
last_successful_epoch_3/initial_value
last_successful_epoch_3
last_successful_epoch_3/Assign
last_successful_epoch_3/read
add_3/y
add_3
assign_updated_epoch_3
Inputs_3/Data/X_input
Inputs_3/Train_Labels/y_true_one_hot_label
Inputs_3/Train_Labels/y_true_class
Parameters_3/Weights/zeros
Parameters_3/Weights/Weight
Parameters_3/Weights/Weight/Assign
Parameters_3/Weights/Weight/read
Parameters_3/Bias/ones
Parameters_3/Bias/Bias
Parameters_3/Bias/Bias/Assign
Parameters_3/Bias/Bias/read
Predictions_3/mlp_layer/MatMul
Predictions_3/mlp_layer
Predictions_3/mlp_layer_1
Predictions_3/Softmax
Predictions_3/predict_class/dimension
Predictions_3/predict_class
Predictions_3/one_hot/Const
Predictions_3/one_hot/Const_1
Predictions_3/one_hot/depth
Predictions_3/one_hot/on_value
Predictions_3/one_hot/off_value
Predictions_3/one_hot
Cross_Entropy_3/Rank
Cross_Entropy_3/Shape
Cross_Entropy_3/Rank_1
Cross_Entropy_3/Shape_1
Cross_Entropy_3/Sub/y
Cross_Entropy_3/Sub
Cross_Entropy_3/Slice/begin
Cross_Entropy_3/Slice/size
Cross_Entropy_3/Slice
Cross_Entropy_3/concat/values_0
Cross_Entropy_3/concat/axis
Cross_Entropy_3/concat
Cross_Entropy_3/Reshape
Cross_Entropy_3/Rank_2
Cross_Entropy_3/Shape_2
Cross_Entropy_3/Sub_1/y
Cross_Entropy_3/Sub_1
Cross_Entropy_3/Slice_1/begin
Cross_Entropy_3/Slice_1/size
Cross_Entropy_3/Slice_1
Cross_Entropy_3/concat_1/values_0
Cross_Entropy_3/concat_1/axis
Cross_Entropy_3/concat_1
Cross_Entropy_3/Reshape_1
Cross_Entropy_3/SoftmaxCrossEntropyWithLogits
Cross_Entropy_3/Sub_2/y
Cross_Entropy_3/Sub_2
Cross_Entropy_3/Slice_2/begin
Cross_Entropy_3/Slice_2/size
Cross_Entropy_3/Slice_2
Cross_Entropy_3/Reshape_2
Loss_Function_3/Const
Loss_Function_3/Square
Loss_Function_3/Const_1
Loss_Function_3/Mean
Loss_Function_3/Const_2
Loss_Function_3/Mean_1
Loss_Function_3/Mul
Loss_Function_3/Add
Loss_Function_3/train_loss_list/initial_value
Loss_Function_3/train_loss_list
Loss_Function_3/train_loss_list/Assign
Loss_Function_3/train_loss_list/read
Loss_Function_3/Assign/value
Loss_Function_3/Assign
Loss_Function_3/train_error/tags
Loss_Function_3/train_error
Loss_Function_3/validate_loss_list/initial_value
Loss_Function_3/validate_loss_list
Loss_Function_3/validate_loss_list/Assign
Loss_Function_3/validate_loss_list/read
Loss_Function_3/Assign_1/value
Loss_Function_3/Assign_1
Loss_Function_3/validate_error/tags
Loss_Function_3/validate_error
Optimizer_3/learning_rate_progress/initial_value
Optimizer_3/learning_rate_progress
Optimizer_3/learning_rate_progress/Assign
Optimizer_3/learning_rate_progress/read
Optimizer_3/Assign/value
Optimizer_3/Assign
Optimizer_3/learning_rate/tags
Optimizer_3/learning_rate/values
Optimizer_3/learning_rate
Optimizer_3/gradients/Shape
Optimizer_3/gradients/Const
Optimizer_3/gradients/Fill
Optimizer_3/gradients/Loss_Function_3/Add_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Add_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/BroadcastGradientArgs
Optimizer_3/gradients/Loss_Function_3/Add_grad/Sum
Optimizer_3/gradients/Loss_Function_3/Add_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Add_grad/Sum_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/Reshape_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/group_deps
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/control_dependency
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/control_dependency_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Reshape/shape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Tile
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape_2
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Const
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Prod
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Const_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Prod_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Maximum/y
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Maximum
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/floordiv
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Cast
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/truediv
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/BroadcastGradientArgs
Optimizer_3/gradients/Loss_Function_3/Mul_grad/mul
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Sum
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mul_grad/mul_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Sum_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Reshape_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/group_deps
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/control_dependency
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/control_dependency_1
Optimizer_3/gradients/Cross_Entropy_3/Reshape_2_grad/Shape
Optimizer_3/gradients/Cross_Entropy_3/Reshape_2_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Reshape/shape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Tile/multiples
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Tile
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Const
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Prod
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Const_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Prod_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Maximum/y
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Maximum
Optimizer_3/gradients/Loss_Function_3/Mean_grad/floordiv
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Cast
Optimizer_3/gradients/Loss_Function_3/Mean_grad/truediv
Optimizer_3/gradients/zeros_like
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul/x
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul_1
Optimizer_3/gradients/Cross_Entropy_3/Reshape_grad/Shape
Optimizer_3/gradients/Cross_Entropy_3/Reshape_grad/Reshape
Optimizer_3/gradients/Predictions_3/Softmax_grad/mul
Optimizer_3/gradients/Predictions_3/Softmax_grad/Sum/reduction_indices
Optimizer_3/gradients/Predictions_3/Softmax_grad/Sum
Optimizer_3/gradients/Predictions_3/Softmax_grad/Reshape/shape
Optimizer_3/gradients/Predictions_3/Softmax_grad/Reshape
Optimizer_3/gradients/Predictions_3/Softmax_grad/sub
Optimizer_3/gradients/Predictions_3/Softmax_grad/mul_1
Optimizer_3/gradients/Predictions_3/mlp_layer_1_grad/ReluGrad
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/BiasAddGrad
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/group_deps
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/control_dependency
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/control_dependency_1
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/MatMul
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/MatMul_1
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_3/gradients/AddN
Optimizer_3/GradientDescent/learning_rate
Optimizer_3/GradientDescent/update_Parameters_3/Weights/Weight/ApplyGradientDescent
Optimizer_3/GradientDescent/update_Parameters_3/Bias/Bias/ApplyGradientDescent
Optimizer_3/GradientDescent
Equal_3
Accuracy_3/Cast
Accuracy_3/Const
Accuracy_3/Mean
Accuracy_3/train_accuracy_list/initial_value
Accuracy_3/train_accuracy_list
Accuracy_3/train_accuracy_list/Assign
Accuracy_3/train_accuracy_list/read
Accuracy_3/Assign/value
Accuracy_3/Assign
Accuracy_3/train_accuracy/tags
Accuracy_3/train_accuracy
Accuracy_3/Cast_1
Accuracy_3/Const_1
Accuracy_3/Mean_1
Accuracy_3/validate_accuracy_list/initial_value
Accuracy_3/validate_accuracy_list
Accuracy_3/validate_accuracy_list/Assign
Accuracy_3/validate_accuracy_list/read
Accuracy_3/Assign_1/value
Accuracy_3/Assign_1
Accuracy_3/validate_accuracy/tags
Accuracy_3/validate_accuracy
Accuracy_3/Cast_2
Accuracy_3/Const_2
Accuracy_3/Mean_2
Accuracy_3/test_accuracy_list/initial_value
Accuracy_3/test_accuracy_list
Accuracy_3/test_accuracy_list/Assign
Accuracy_3/test_accuracy_list/read
Accuracy_3/Assign_2/value
Accuracy_3/Assign_2
Accuracy_3/test_accuracy/tags
Accuracy_3/test_accuracy
init_3/NoOp
init_3/NoOp_1
init_3
Merge_3/MergeSummary
save_3/Const
save_3/SaveV2/tensor_names
save_3/SaveV2/shape_and_slices
save_3/SaveV2
save_3/control_dependency
save_3/RestoreV2/tensor_names
save_3/RestoreV2/shape_and_slices
save_3/RestoreV2
save_3/Assign
save_3/RestoreV2_1/tensor_names
save_3/RestoreV2_1/shape_and_slices
save_3/RestoreV2_1
save_3/Assign_1
save_3/RestoreV2_2/tensor_names
save_3/RestoreV2_2/shape_and_slices
save_3/RestoreV2_2
save_3/Assign_2
save_3/RestoreV2_3/tensor_names
save_3/RestoreV2_3/shape_and_slices
save_3/RestoreV2_3
save_3/Assign_3
save_3/RestoreV2_4/tensor_names
save_3/RestoreV2_4/shape_and_slices
save_3/RestoreV2_4
save_3/Assign_4
save_3/RestoreV2_5/tensor_names
save_3/RestoreV2_5/shape_and_slices
save_3/RestoreV2_5
save_3/Assign_5
save_3/RestoreV2_6/tensor_names
save_3/RestoreV2_6/shape_and_slices
save_3/RestoreV2_6
save_3/Assign_6
save_3/RestoreV2_7/tensor_names
save_3/RestoreV2_7/shape_and_slices
save_3/RestoreV2_7
save_3/Assign_7
save_3/RestoreV2_8/tensor_names
save_3/RestoreV2_8/shape_and_slices
save_3/RestoreV2_8
save_3/Assign_8
save_3/RestoreV2_9/tensor_names
save_3/RestoreV2_9/shape_and_slices
save_3/RestoreV2_9
save_3/Assign_9
save_3/RestoreV2_10/tensor_names
save_3/RestoreV2_10/shape_and_slices
save_3/RestoreV2_10
save_3/Assign_10
save_3/RestoreV2_11/tensor_names
save_3/RestoreV2_11/shape_and_slices
save_3/RestoreV2_11
save_3/Assign_11
save_3/RestoreV2_12/tensor_names
save_3/RestoreV2_12/shape_and_slices
save_3/RestoreV2_12
save_3/Assign_12
save_3/RestoreV2_13/tensor_names
save_3/RestoreV2_13/shape_and_slices
save_3/RestoreV2_13
save_3/Assign_13
save_3/RestoreV2_14/tensor_names
save_3/RestoreV2_14/shape_and_slices
save_3/RestoreV2_14
save_3/Assign_14
save_3/RestoreV2_15/tensor_names
save_3/RestoreV2_15/shape_and_slices
save_3/RestoreV2_15
save_3/Assign_15
save_3/RestoreV2_16/tensor_names
save_3/RestoreV2_16/shape_and_slices
save_3/RestoreV2_16
save_3/Assign_16
save_3/RestoreV2_17/tensor_names
save_3/RestoreV2_17/shape_and_slices
save_3/RestoreV2_17
save_3/Assign_17
save_3/RestoreV2_18/tensor_names
save_3/RestoreV2_18/shape_and_slices
save_3/RestoreV2_18
save_3/Assign_18
save_3/RestoreV2_19/tensor_names
save_3/RestoreV2_19/shape_and_slices
save_3/RestoreV2_19
save_3/Assign_19
save_3/RestoreV2_20/tensor_names
save_3/RestoreV2_20/shape_and_slices
save_3/RestoreV2_20
save_3/Assign_20
save_3/RestoreV2_21/tensor_names
save_3/RestoreV2_21/shape_and_slices
save_3/RestoreV2_21
save_3/Assign_21
save_3/RestoreV2_22/tensor_names
save_3/RestoreV2_22/shape_and_slices
save_3/RestoreV2_22
save_3/Assign_22
save_3/RestoreV2_23/tensor_names
save_3/RestoreV2_23/shape_and_slices
save_3/RestoreV2_23
save_3/Assign_23
save_3/RestoreV2_24/tensor_names
save_3/RestoreV2_24/shape_and_slices
save_3/RestoreV2_24
save_3/Assign_24
save_3/RestoreV2_25/tensor_names
save_3/RestoreV2_25/shape_and_slices
save_3/RestoreV2_25
save_3/Assign_25
save_3/RestoreV2_26/tensor_names
save_3/RestoreV2_26/shape_and_slices
save_3/RestoreV2_26
save_3/Assign_26
save_3/RestoreV2_27/tensor_names
save_3/RestoreV2_27/shape_and_slices
save_3/RestoreV2_27
save_3/Assign_27
save_3/RestoreV2_28/tensor_names
save_3/RestoreV2_28/shape_and_slices
save_3/RestoreV2_28
save_3/Assign_28
save_3/RestoreV2_29/tensor_names
save_3/RestoreV2_29/shape_and_slices
save_3/RestoreV2_29
save_3/Assign_29
save_3/RestoreV2_30/tensor_names
save_3/RestoreV2_30/shape_and_slices
save_3/RestoreV2_30
save_3/Assign_30
save_3/RestoreV2_31/tensor_names
save_3/RestoreV2_31/shape_and_slices
save_3/RestoreV2_31
save_3/Assign_31
save_3/RestoreV2_32/tensor_names
save_3/RestoreV2_32/shape_and_slices
save_3/RestoreV2_32
save_3/Assign_32
save_3/RestoreV2_33/tensor_names
save_3/RestoreV2_33/shape_and_slices
save_3/RestoreV2_33
save_3/Assign_33
save_3/RestoreV2_34/tensor_names
save_3/RestoreV2_34/shape_and_slices
save_3/RestoreV2_34
save_3/Assign_34
save_3/RestoreV2_35/tensor_names
save_3/RestoreV2_35/shape_and_slices
save_3/RestoreV2_35
save_3/Assign_35
save_3/restore_all/NoOp
save_3/restore_all/NoOp_1
save_3/restore_all
last_successful_epoch_4/initial_value
last_successful_epoch_4
last_successful_epoch_4/Assign
last_successful_epoch_4/read
add_4/y
add_4
assign_updated_epoch_4
Inputs_4/Data/X_input
Inputs_4/Train_Labels/y_true_one_hot_label
Inputs_4/Train_Labels/y_true_class
Parameters_4/Weights/zeros
Parameters_4/Weights/Weight
Parameters_4/Weights/Weight/Assign
Parameters_4/Weights/Weight/read
Parameters_4/Bias/ones
Parameters_4/Bias/Bias
Parameters_4/Bias/Bias/Assign
Parameters_4/Bias/Bias/read
Predictions_4/mlp_layer/MatMul
Predictions_4/mlp_layer
Predictions_4/mlp_layer_1
Predictions_4/Softmax
Predictions_4/predict_class/dimension
Predictions_4/predict_class
Predictions_4/one_hot/Const
Predictions_4/one_hot/Const_1
Predictions_4/one_hot/depth
Predictions_4/one_hot/on_value
Predictions_4/one_hot/off_value
Predictions_4/one_hot
Cross_Entropy_4/Rank
Cross_Entropy_4/Shape
Cross_Entropy_4/Rank_1
Cross_Entropy_4/Shape_1
Cross_Entropy_4/Sub/y
Cross_Entropy_4/Sub
Cross_Entropy_4/Slice/begin
Cross_Entropy_4/Slice/size
Cross_Entropy_4/Slice
Cross_Entropy_4/concat/values_0
Cross_Entropy_4/concat/axis
Cross_Entropy_4/concat
Cross_Entropy_4/Reshape
Cross_Entropy_4/Rank_2
Cross_Entropy_4/Shape_2
Cross_Entropy_4/Sub_1/y
Cross_Entropy_4/Sub_1
Cross_Entropy_4/Slice_1/begin
Cross_Entropy_4/Slice_1/size
Cross_Entropy_4/Slice_1
Cross_Entropy_4/concat_1/values_0
Cross_Entropy_4/concat_1/axis
Cross_Entropy_4/concat_1
Cross_Entropy_4/Reshape_1
Cross_Entropy_4/SoftmaxCrossEntropyWithLogits
Cross_Entropy_4/Sub_2/y
Cross_Entropy_4/Sub_2
Cross_Entropy_4/Slice_2/begin
Cross_Entropy_4/Slice_2/size
Cross_Entropy_4/Slice_2
Cross_Entropy_4/Reshape_2
Loss_Function_4/Const
Loss_Function_4/Square
Loss_Function_4/Const_1
Loss_Function_4/Mean
Loss_Function_4/Const_2
Loss_Function_4/Mean_1
Loss_Function_4/Mul
Loss_Function_4/Add
Loss_Function_4/train_loss_list/initial_value
Loss_Function_4/train_loss_list
Loss_Function_4/train_loss_list/Assign
Loss_Function_4/train_loss_list/read
Loss_Function_4/Assign/value
Loss_Function_4/Assign
Loss_Function_4/train_error/tags
Loss_Function_4/train_error
Loss_Function_4/validate_loss_list/initial_value
Loss_Function_4/validate_loss_list
Loss_Function_4/validate_loss_list/Assign
Loss_Function_4/validate_loss_list/read
Loss_Function_4/Assign_1/value
Loss_Function_4/Assign_1
Loss_Function_4/validate_error/tags
Loss_Function_4/validate_error
Optimizer_4/learning_rate_progress/initial_value
Optimizer_4/learning_rate_progress
Optimizer_4/learning_rate_progress/Assign
Optimizer_4/learning_rate_progress/read
Optimizer_4/Assign/value
Optimizer_4/Assign
Optimizer_4/learning_rate/tags
Optimizer_4/learning_rate/values
Optimizer_4/learning_rate
Optimizer_4/gradients/Shape
Optimizer_4/gradients/Const
Optimizer_4/gradients/Fill
Optimizer_4/gradients/Loss_Function_4/Add_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Add_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/BroadcastGradientArgs
Optimizer_4/gradients/Loss_Function_4/Add_grad/Sum
Optimizer_4/gradients/Loss_Function_4/Add_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Add_grad/Sum_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/Reshape_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/group_deps
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/control_dependency
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/control_dependency_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Reshape/shape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Tile
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape_2
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Const
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Prod
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Const_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Prod_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Maximum/y
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Maximum
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/floordiv
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Cast
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/truediv
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/BroadcastGradientArgs
Optimizer_4/gradients/Loss_Function_4/Mul_grad/mul
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Sum
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mul_grad/mul_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Sum_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Reshape_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/group_deps
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/control_dependency
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/control_dependency_1
Optimizer_4/gradients/Cross_Entropy_4/Reshape_2_grad/Shape
Optimizer_4/gradients/Cross_Entropy_4/Reshape_2_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Reshape/shape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Tile/multiples
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Tile
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Const
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Prod
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Const_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Prod_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Maximum/y
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Maximum
Optimizer_4/gradients/Loss_Function_4/Mean_grad/floordiv
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Cast
Optimizer_4/gradients/Loss_Function_4/Mean_grad/truediv
Optimizer_4/gradients/zeros_like
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul/x
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul_1
Optimizer_4/gradients/Cross_Entropy_4/Reshape_grad/Shape
Optimizer_4/gradients/Cross_Entropy_4/Reshape_grad/Reshape
Optimizer_4/gradients/Predictions_4/Softmax_grad/mul
Optimizer_4/gradients/Predictions_4/Softmax_grad/Sum/reduction_indices
Optimizer_4/gradients/Predictions_4/Softmax_grad/Sum
Optimizer_4/gradients/Predictions_4/Softmax_grad/Reshape/shape
Optimizer_4/gradients/Predictions_4/Softmax_grad/Reshape
Optimizer_4/gradients/Predictions_4/Softmax_grad/sub
Optimizer_4/gradients/Predictions_4/Softmax_grad/mul_1
Optimizer_4/gradients/Predictions_4/mlp_layer_1_grad/ReluGrad
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/BiasAddGrad
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/group_deps
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/control_dependency
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/control_dependency_1
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/MatMul
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/MatMul_1
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_4/gradients/AddN
Optimizer_4/GradientDescent/learning_rate
Optimizer_4/GradientDescent/update_Parameters_4/Weights/Weight/ApplyGradientDescent
Optimizer_4/GradientDescent/update_Parameters_4/Bias/Bias/ApplyGradientDescent
Optimizer_4/GradientDescent
Equal_4
Accuracy_4/Cast
Accuracy_4/Const
Accuracy_4/Mean
Accuracy_4/train_accuracy_list/initial_value
Accuracy_4/train_accuracy_list
Accuracy_4/train_accuracy_list/Assign
Accuracy_4/train_accuracy_list/read
Accuracy_4/Assign/value
Accuracy_4/Assign
Accuracy_4/train_accuracy/tags
Accuracy_4/train_accuracy
Accuracy_4/Cast_1
Accuracy_4/Const_1
Accuracy_4/Mean_1
Accuracy_4/validate_accuracy_list/initial_value
Accuracy_4/validate_accuracy_list
Accuracy_4/validate_accuracy_list/Assign
Accuracy_4/validate_accuracy_list/read
Accuracy_4/Assign_1/value
Accuracy_4/Assign_1
Accuracy_4/validate_accuracy/tags
Accuracy_4/validate_accuracy
Accuracy_4/Cast_2
Accuracy_4/Const_2
Accuracy_4/Mean_2
Accuracy_4/test_accuracy_list/initial_value
Accuracy_4/test_accuracy_list
Accuracy_4/test_accuracy_list/Assign
Accuracy_4/test_accuracy_list/read
Accuracy_4/Assign_2/value
Accuracy_4/Assign_2
Accuracy_4/test_accuracy/tags
Accuracy_4/test_accuracy
init_4/NoOp
init_4/NoOp_1
init_4
Merge_4/MergeSummary
save_4/Const
save_4/SaveV2/tensor_names
save_4/SaveV2/shape_and_slices
save_4/SaveV2
save_4/control_dependency
save_4/RestoreV2/tensor_names
save_4/RestoreV2/shape_and_slices
save_4/RestoreV2
save_4/Assign
save_4/RestoreV2_1/tensor_names
save_4/RestoreV2_1/shape_and_slices
save_4/RestoreV2_1
save_4/Assign_1
save_4/RestoreV2_2/tensor_names
save_4/RestoreV2_2/shape_and_slices
save_4/RestoreV2_2
save_4/Assign_2
save_4/RestoreV2_3/tensor_names
save_4/RestoreV2_3/shape_and_slices
save_4/RestoreV2_3
save_4/Assign_3
save_4/RestoreV2_4/tensor_names
save_4/RestoreV2_4/shape_and_slices
save_4/RestoreV2_4
save_4/Assign_4
save_4/RestoreV2_5/tensor_names
save_4/RestoreV2_5/shape_and_slices
save_4/RestoreV2_5
save_4/Assign_5
save_4/RestoreV2_6/tensor_names
save_4/RestoreV2_6/shape_and_slices
save_4/RestoreV2_6
save_4/Assign_6
save_4/RestoreV2_7/tensor_names
save_4/RestoreV2_7/shape_and_slices
save_4/RestoreV2_7
save_4/Assign_7
save_4/RestoreV2_8/tensor_names
save_4/RestoreV2_8/shape_and_slices
save_4/RestoreV2_8
save_4/Assign_8
save_4/RestoreV2_9/tensor_names
save_4/RestoreV2_9/shape_and_slices
save_4/RestoreV2_9
save_4/Assign_9
save_4/RestoreV2_10/tensor_names
save_4/RestoreV2_10/shape_and_slices
save_4/RestoreV2_10
save_4/Assign_10
save_4/RestoreV2_11/tensor_names
save_4/RestoreV2_11/shape_and_slices
save_4/RestoreV2_11
save_4/Assign_11
save_4/RestoreV2_12/tensor_names
save_4/RestoreV2_12/shape_and_slices
save_4/RestoreV2_12
save_4/Assign_12
save_4/RestoreV2_13/tensor_names
save_4/RestoreV2_13/shape_and_slices
save_4/RestoreV2_13
save_4/Assign_13
save_4/RestoreV2_14/tensor_names
save_4/RestoreV2_14/shape_and_slices
save_4/RestoreV2_14
save_4/Assign_14
save_4/RestoreV2_15/tensor_names
save_4/RestoreV2_15/shape_and_slices
save_4/RestoreV2_15
save_4/Assign_15
save_4/RestoreV2_16/tensor_names
save_4/RestoreV2_16/shape_and_slices
save_4/RestoreV2_16
save_4/Assign_16
save_4/RestoreV2_17/tensor_names
save_4/RestoreV2_17/shape_and_slices
save_4/RestoreV2_17
save_4/Assign_17
save_4/RestoreV2_18/tensor_names
save_4/RestoreV2_18/shape_and_slices
save_4/RestoreV2_18
save_4/Assign_18
save_4/RestoreV2_19/tensor_names
save_4/RestoreV2_19/shape_and_slices
save_4/RestoreV2_19
save_4/Assign_19
save_4/RestoreV2_20/tensor_names
save_4/RestoreV2_20/shape_and_slices
save_4/RestoreV2_20
save_4/Assign_20
save_4/RestoreV2_21/tensor_names
save_4/RestoreV2_21/shape_and_slices
save_4/RestoreV2_21
save_4/Assign_21
save_4/RestoreV2_22/tensor_names
save_4/RestoreV2_22/shape_and_slices
save_4/RestoreV2_22
save_4/Assign_22
save_4/RestoreV2_23/tensor_names
save_4/RestoreV2_23/shape_and_slices
save_4/RestoreV2_23
save_4/Assign_23
save_4/RestoreV2_24/tensor_names
save_4/RestoreV2_24/shape_and_slices
save_4/RestoreV2_24
save_4/Assign_24
save_4/RestoreV2_25/tensor_names
save_4/RestoreV2_25/shape_and_slices
save_4/RestoreV2_25
save_4/Assign_25
save_4/RestoreV2_26/tensor_names
save_4/RestoreV2_26/shape_and_slices
save_4/RestoreV2_26
save_4/Assign_26
save_4/RestoreV2_27/tensor_names
save_4/RestoreV2_27/shape_and_slices
save_4/RestoreV2_27
save_4/Assign_27
save_4/RestoreV2_28/tensor_names
save_4/RestoreV2_28/shape_and_slices
save_4/RestoreV2_28
save_4/Assign_28
save_4/RestoreV2_29/tensor_names
save_4/RestoreV2_29/shape_and_slices
save_4/RestoreV2_29
save_4/Assign_29
save_4/RestoreV2_30/tensor_names
save_4/RestoreV2_30/shape_and_slices
save_4/RestoreV2_30
save_4/Assign_30
save_4/RestoreV2_31/tensor_names
save_4/RestoreV2_31/shape_and_slices
save_4/RestoreV2_31
save_4/Assign_31
save_4/RestoreV2_32/tensor_names
save_4/RestoreV2_32/shape_and_slices
save_4/RestoreV2_32
save_4/Assign_32
save_4/RestoreV2_33/tensor_names
save_4/RestoreV2_33/shape_and_slices
save_4/RestoreV2_33
save_4/Assign_33
save_4/RestoreV2_34/tensor_names
save_4/RestoreV2_34/shape_and_slices
save_4/RestoreV2_34
save_4/Assign_34
save_4/RestoreV2_35/tensor_names
save_4/RestoreV2_35/shape_and_slices
save_4/RestoreV2_35
save_4/Assign_35
save_4/RestoreV2_36/tensor_names
save_4/RestoreV2_36/shape_and_slices
save_4/RestoreV2_36
save_4/Assign_36
save_4/RestoreV2_37/tensor_names
save_4/RestoreV2_37/shape_and_slices
save_4/RestoreV2_37
save_4/Assign_37
save_4/RestoreV2_38/tensor_names
save_4/RestoreV2_38/shape_and_slices
save_4/RestoreV2_38
save_4/Assign_38
save_4/RestoreV2_39/tensor_names
save_4/RestoreV2_39/shape_and_slices
save_4/RestoreV2_39
save_4/Assign_39
save_4/RestoreV2_40/tensor_names
save_4/RestoreV2_40/shape_and_slices
save_4/RestoreV2_40
save_4/Assign_40
save_4/RestoreV2_41/tensor_names
save_4/RestoreV2_41/shape_and_slices
save_4/RestoreV2_41
save_4/Assign_41
save_4/RestoreV2_42/tensor_names
save_4/RestoreV2_42/shape_and_slices
save_4/RestoreV2_42
save_4/Assign_42
save_4/RestoreV2_43/tensor_names
save_4/RestoreV2_43/shape_and_slices
save_4/RestoreV2_43
save_4/Assign_43
save_4/RestoreV2_44/tensor_names
save_4/RestoreV2_44/shape_and_slices
save_4/RestoreV2_44
save_4/Assign_44
save_4/restore_all/NoOp
save_4/restore_all/NoOp_1
save_4/restore_all
last_successful_epoch_5/initial_value
last_successful_epoch_5
last_successful_epoch_5/Assign
last_successful_epoch_5/read
add_5/y
add_5
assign_updated_epoch_5
Inputs_5/Data/X_input
Inputs_5/Train_Labels/y_true_one_hot_label
Inputs_5/Train_Labels/y_true_class
Parameters_5/Weights/zeros
Parameters_5/Weights/Weight
Parameters_5/Weights/Weight/Assign
Parameters_5/Weights/Weight/read
Parameters_5/Bias/ones
Parameters_5/Bias/Bias
Parameters_5/Bias/Bias/Assign
Parameters_5/Bias/Bias/read
Predictions_5/mlp_layer/MatMul
Predictions_5/mlp_layer
Predictions_5/mlp_layer_1
Predictions_5/Softmax
Predictions_5/predict_class/dimension
Predictions_5/predict_class
Predictions_5/one_hot/Const
Predictions_5/one_hot/Const_1
Predictions_5/one_hot/depth
Predictions_5/one_hot/on_value
Predictions_5/one_hot/off_value
Predictions_5/one_hot
Cross_Entropy_5/Rank
Cross_Entropy_5/Shape
Cross_Entropy_5/Rank_1
Cross_Entropy_5/Shape_1
Cross_Entropy_5/Sub/y
Cross_Entropy_5/Sub
Cross_Entropy_5/Slice/begin
Cross_Entropy_5/Slice/size
Cross_Entropy_5/Slice
Cross_Entropy_5/concat/values_0
Cross_Entropy_5/concat/axis
Cross_Entropy_5/concat
Cross_Entropy_5/Reshape
Cross_Entropy_5/Rank_2
Cross_Entropy_5/Shape_2
Cross_Entropy_5/Sub_1/y
Cross_Entropy_5/Sub_1
Cross_Entropy_5/Slice_1/begin
Cross_Entropy_5/Slice_1/size
Cross_Entropy_5/Slice_1
Cross_Entropy_5/concat_1/values_0
Cross_Entropy_5/concat_1/axis
Cross_Entropy_5/concat_1
Cross_Entropy_5/Reshape_1
Cross_Entropy_5/SoftmaxCrossEntropyWithLogits
Cross_Entropy_5/Sub_2/y
Cross_Entropy_5/Sub_2
Cross_Entropy_5/Slice_2/begin
Cross_Entropy_5/Slice_2/size
Cross_Entropy_5/Slice_2
Cross_Entropy_5/Reshape_2
Loss_Function_5/Const
Loss_Function_5/Square
Loss_Function_5/Const_1
Loss_Function_5/Mean
Loss_Function_5/Const_2
Loss_Function_5/Mean_1
Loss_Function_5/Mul
Loss_Function_5/Add
Loss_Function_5/train_loss_list/initial_value
Loss_Function_5/train_loss_list
Loss_Function_5/train_loss_list/Assign
Loss_Function_5/train_loss_list/read
Loss_Function_5/Assign/value
Loss_Function_5/Assign
Loss_Function_5/train_error/tags
Loss_Function_5/train_error
Loss_Function_5/validate_loss_list/initial_value
Loss_Function_5/validate_loss_list
Loss_Function_5/validate_loss_list/Assign
Loss_Function_5/validate_loss_list/read
Loss_Function_5/Assign_1/value
Loss_Function_5/Assign_1
Loss_Function_5/validate_error/tags
Loss_Function_5/validate_error
Optimizer_5/learning_rate_progress/initial_value
Optimizer_5/learning_rate_progress
Optimizer_5/learning_rate_progress/Assign
Optimizer_5/learning_rate_progress/read
Optimizer_5/Assign/value
Optimizer_5/Assign
Optimizer_5/learning_rate/tags
Optimizer_5/learning_rate/values
Optimizer_5/learning_rate
Optimizer_5/gradients/Shape
Optimizer_5/gradients/Const
Optimizer_5/gradients/Fill
Optimizer_5/gradients/Loss_Function_5/Add_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Add_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/BroadcastGradientArgs
Optimizer_5/gradients/Loss_Function_5/Add_grad/Sum
Optimizer_5/gradients/Loss_Function_5/Add_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Add_grad/Sum_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/Reshape_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/group_deps
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/control_dependency
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/control_dependency_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Reshape/shape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Tile
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape_2
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Const
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Prod
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Const_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Prod_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Maximum/y
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Maximum
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/floordiv
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Cast
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/truediv
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/BroadcastGradientArgs
Optimizer_5/gradients/Loss_Function_5/Mul_grad/mul
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Sum
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mul_grad/mul_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Sum_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Reshape_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/group_deps
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/control_dependency
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/control_dependency_1
Optimizer_5/gradients/Cross_Entropy_5/Reshape_2_grad/Shape
Optimizer_5/gradients/Cross_Entropy_5/Reshape_2_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Reshape/shape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Tile/multiples
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Tile
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Const
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Prod
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Const_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Prod_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Maximum/y
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Maximum
Optimizer_5/gradients/Loss_Function_5/Mean_grad/floordiv
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Cast
Optimizer_5/gradients/Loss_Function_5/Mean_grad/truediv
Optimizer_5/gradients/zeros_like
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul/x
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul_1
Optimizer_5/gradients/Cross_Entropy_5/Reshape_grad/Shape
Optimizer_5/gradients/Cross_Entropy_5/Reshape_grad/Reshape
Optimizer_5/gradients/Predictions_5/Softmax_grad/mul
Optimizer_5/gradients/Predictions_5/Softmax_grad/Sum/reduction_indices
Optimizer_5/gradients/Predictions_5/Softmax_grad/Sum
Optimizer_5/gradients/Predictions_5/Softmax_grad/Reshape/shape
Optimizer_5/gradients/Predictions_5/Softmax_grad/Reshape
Optimizer_5/gradients/Predictions_5/Softmax_grad/sub
Optimizer_5/gradients/Predictions_5/Softmax_grad/mul_1
Optimizer_5/gradients/Predictions_5/mlp_layer_1_grad/ReluGrad
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/BiasAddGrad
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/group_deps
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/control_dependency
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/control_dependency_1
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/MatMul
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/MatMul_1
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_5/gradients/AddN
Optimizer_5/GradientDescent/learning_rate
Optimizer_5/GradientDescent/update_Parameters_5/Weights/Weight/ApplyGradientDescent
Optimizer_5/GradientDescent/update_Parameters_5/Bias/Bias/ApplyGradientDescent
Optimizer_5/GradientDescent
Equal_5
Accuracy_5/Cast
Accuracy_5/Const
Accuracy_5/Mean
Accuracy_5/train_accuracy_list/initial_value
Accuracy_5/train_accuracy_list
Accuracy_5/train_accuracy_list/Assign
Accuracy_5/train_accuracy_list/read
Accuracy_5/Assign/value
Accuracy_5/Assign
Accuracy_5/train_accuracy/tags
Accuracy_5/train_accuracy
Accuracy_5/Cast_1
Accuracy_5/Const_1
Accuracy_5/Mean_1
Accuracy_5/validate_accuracy_list/initial_value
Accuracy_5/validate_accuracy_list
Accuracy_5/validate_accuracy_list/Assign
Accuracy_5/validate_accuracy_list/read
Accuracy_5/Assign_1/value
Accuracy_5/Assign_1
Accuracy_5/validate_accuracy/tags
Accuracy_5/validate_accuracy
Accuracy_5/Cast_2
Accuracy_5/Const_2
Accuracy_5/Mean_2
Accuracy_5/test_accuracy_list/initial_value
Accuracy_5/test_accuracy_list
Accuracy_5/test_accuracy_list/Assign
Accuracy_5/test_accuracy_list/read
Accuracy_5/Assign_2/value
Accuracy_5/Assign_2
Accuracy_5/test_accuracy/tags
Accuracy_5/test_accuracy
init_5/NoOp
init_5/NoOp_1
init_5
Merge_5/MergeSummary
save_5/Const
save_5/SaveV2/tensor_names
save_5/SaveV2/shape_and_slices
save_5/SaveV2
save_5/control_dependency
save_5/RestoreV2/tensor_names
save_5/RestoreV2/shape_and_slices
save_5/RestoreV2
save_5/Assign
save_5/RestoreV2_1/tensor_names
save_5/RestoreV2_1/shape_and_slices
save_5/RestoreV2_1
save_5/Assign_1
save_5/RestoreV2_2/tensor_names
save_5/RestoreV2_2/shape_and_slices
save_5/RestoreV2_2
save_5/Assign_2
save_5/RestoreV2_3/tensor_names
save_5/RestoreV2_3/shape_and_slices
save_5/RestoreV2_3
save_5/Assign_3
save_5/RestoreV2_4/tensor_names
save_5/RestoreV2_4/shape_and_slices
save_5/RestoreV2_4
save_5/Assign_4
save_5/RestoreV2_5/tensor_names
save_5/RestoreV2_5/shape_and_slices
save_5/RestoreV2_5
save_5/Assign_5
save_5/RestoreV2_6/tensor_names
save_5/RestoreV2_6/shape_and_slices
save_5/RestoreV2_6
save_5/Assign_6
save_5/RestoreV2_7/tensor_names
save_5/RestoreV2_7/shape_and_slices
save_5/RestoreV2_7
save_5/Assign_7
save_5/RestoreV2_8/tensor_names
save_5/RestoreV2_8/shape_and_slices
save_5/RestoreV2_8
save_5/Assign_8
save_5/RestoreV2_9/tensor_names
save_5/RestoreV2_9/shape_and_slices
save_5/RestoreV2_9
save_5/Assign_9
save_5/RestoreV2_10/tensor_names
save_5/RestoreV2_10/shape_and_slices
save_5/RestoreV2_10
save_5/Assign_10
save_5/RestoreV2_11/tensor_names
save_5/RestoreV2_11/shape_and_slices
save_5/RestoreV2_11
save_5/Assign_11
save_5/RestoreV2_12/tensor_names
save_5/RestoreV2_12/shape_and_slices
save_5/RestoreV2_12
save_5/Assign_12
save_5/RestoreV2_13/tensor_names
save_5/RestoreV2_13/shape_and_slices
save_5/RestoreV2_13
save_5/Assign_13
save_5/RestoreV2_14/tensor_names
save_5/RestoreV2_14/shape_and_slices
save_5/RestoreV2_14
save_5/Assign_14
save_5/RestoreV2_15/tensor_names
save_5/RestoreV2_15/shape_and_slices
save_5/RestoreV2_15
save_5/Assign_15
save_5/RestoreV2_16/tensor_names
save_5/RestoreV2_16/shape_and_slices
save_5/RestoreV2_16
save_5/Assign_16
save_5/RestoreV2_17/tensor_names
save_5/RestoreV2_17/shape_and_slices
save_5/RestoreV2_17
save_5/Assign_17
save_5/RestoreV2_18/tensor_names
save_5/RestoreV2_18/shape_and_slices
save_5/RestoreV2_18
save_5/Assign_18
save_5/RestoreV2_19/tensor_names
save_5/RestoreV2_19/shape_and_slices
save_5/RestoreV2_19
save_5/Assign_19
save_5/RestoreV2_20/tensor_names
save_5/RestoreV2_20/shape_and_slices
save_5/RestoreV2_20
save_5/Assign_20
save_5/RestoreV2_21/tensor_names
save_5/RestoreV2_21/shape_and_slices
save_5/RestoreV2_21
save_5/Assign_21
save_5/RestoreV2_22/tensor_names
save_5/RestoreV2_22/shape_and_slices
save_5/RestoreV2_22
save_5/Assign_22
save_5/RestoreV2_23/tensor_names
save_5/RestoreV2_23/shape_and_slices
save_5/RestoreV2_23
save_5/Assign_23
save_5/RestoreV2_24/tensor_names
save_5/RestoreV2_24/shape_and_slices
save_5/RestoreV2_24
save_5/Assign_24
save_5/RestoreV2_25/tensor_names
save_5/RestoreV2_25/shape_and_slices
save_5/RestoreV2_25
save_5/Assign_25
save_5/RestoreV2_26/tensor_names
save_5/RestoreV2_26/shape_and_slices
save_5/RestoreV2_26
save_5/Assign_26
save_5/RestoreV2_27/tensor_names
save_5/RestoreV2_27/shape_and_slices
save_5/RestoreV2_27
save_5/Assign_27
save_5/RestoreV2_28/tensor_names
save_5/RestoreV2_28/shape_and_slices
save_5/RestoreV2_28
save_5/Assign_28
save_5/RestoreV2_29/tensor_names
save_5/RestoreV2_29/shape_and_slices
save_5/RestoreV2_29
save_5/Assign_29
save_5/RestoreV2_30/tensor_names
save_5/RestoreV2_30/shape_and_slices
save_5/RestoreV2_30
save_5/Assign_30
save_5/RestoreV2_31/tensor_names
save_5/RestoreV2_31/shape_and_slices
save_5/RestoreV2_31
save_5/Assign_31
save_5/RestoreV2_32/tensor_names
save_5/RestoreV2_32/shape_and_slices
save_5/RestoreV2_32
save_5/Assign_32
save_5/RestoreV2_33/tensor_names
save_5/RestoreV2_33/shape_and_slices
save_5/RestoreV2_33
save_5/Assign_33
save_5/RestoreV2_34/tensor_names
save_5/RestoreV2_34/shape_and_slices
save_5/RestoreV2_34
save_5/Assign_34
save_5/RestoreV2_35/tensor_names
save_5/RestoreV2_35/shape_and_slices
save_5/RestoreV2_35
save_5/Assign_35
save_5/RestoreV2_36/tensor_names
save_5/RestoreV2_36/shape_and_slices
save_5/RestoreV2_36
save_5/Assign_36
save_5/RestoreV2_37/tensor_names
save_5/RestoreV2_37/shape_and_slices
save_5/RestoreV2_37
save_5/Assign_37
save_5/RestoreV2_38/tensor_names
save_5/RestoreV2_38/shape_and_slices
save_5/RestoreV2_38
save_5/Assign_38
save_5/RestoreV2_39/tensor_names
save_5/RestoreV2_39/shape_and_slices
save_5/RestoreV2_39
save_5/Assign_39
save_5/RestoreV2_40/tensor_names
save_5/RestoreV2_40/shape_and_slices
save_5/RestoreV2_40
save_5/Assign_40
save_5/RestoreV2_41/tensor_names
save_5/RestoreV2_41/shape_and_slices
save_5/RestoreV2_41
save_5/Assign_41
save_5/RestoreV2_42/tensor_names
save_5/RestoreV2_42/shape_and_slices
save_5/RestoreV2_42
save_5/Assign_42
save_5/RestoreV2_43/tensor_names
save_5/RestoreV2_43/shape_and_slices
save_5/RestoreV2_43
save_5/Assign_43
save_5/RestoreV2_44/tensor_names
save_5/RestoreV2_44/shape_and_slices
save_5/RestoreV2_44
save_5/Assign_44
save_5/RestoreV2_45/tensor_names
save_5/RestoreV2_45/shape_and_slices
save_5/RestoreV2_45
save_5/Assign_45
save_5/RestoreV2_46/tensor_names
save_5/RestoreV2_46/shape_and_slices
save_5/RestoreV2_46
save_5/Assign_46
save_5/RestoreV2_47/tensor_names
save_5/RestoreV2_47/shape_and_slices
save_5/RestoreV2_47
save_5/Assign_47
save_5/RestoreV2_48/tensor_names
save_5/RestoreV2_48/shape_and_slices
save_5/RestoreV2_48
save_5/Assign_48
save_5/RestoreV2_49/tensor_names
save_5/RestoreV2_49/shape_and_slices
save_5/RestoreV2_49
save_5/Assign_49
save_5/RestoreV2_50/tensor_names
save_5/RestoreV2_50/shape_and_slices
save_5/RestoreV2_50
save_5/Assign_50
save_5/RestoreV2_51/tensor_names
save_5/RestoreV2_51/shape_and_slices
save_5/RestoreV2_51
save_5/Assign_51
save_5/RestoreV2_52/tensor_names
save_5/RestoreV2_52/shape_and_slices
save_5/RestoreV2_52
save_5/Assign_52
save_5/RestoreV2_53/tensor_names
save_5/RestoreV2_53/shape_and_slices
save_5/RestoreV2_53
save_5/Assign_53
save_5/restore_all/NoOp
save_5/restore_all/NoOp_1
save_5/restore_all
Output Logits  : Tensor("Predictions_3/Softmax:0", shape=(?, 10), dtype=float32, device=/device:GPU:0)
Entropy        : Tensor("Cross_Entropy_3/Reshape_2:0", shape=(?,), dtype=float32, device=/device:GPU:0)
Predictions    : Tensor("Equal_3:0", shape=(?,), dtype=bool, device=/device:GPU:0) 
Optimizer      : name: "Optimizer_3/GradientDescent"
op: "NoOp"
input: "^Optimizer_3/GradientDescent/update_Parameters_3/Weights/Weight/ApplyGradientDescent"
input: "^Optimizer_3/GradientDescent/update_Parameters_3/Bias/Bias/ApplyGradientDescent"
device: "/device:GPU:0"
 
>> Model info
Train loss     : Tensor("Loss_Function_3/Add:0", shape=(), dtype=float32, device=/device:GPU:0)
Train accuracy : Tensor("Accuracy_3/Mean:0", shape=(), dtype=float32, device=/device:GPU:0)
Val. loss      : Tensor("Loss_Function_3/Add:0", shape=(), dtype=float32, device=/device:GPU:0)
Val. accuracy  : Tensor("Accuracy_3/Mean_1:0", shape=(), dtype=float32, device=/device:GPU:0)
>> Summary params
Train loss     : Tensor("Loss_Function_3/train_error:0", shape=(), dtype=string, device=/device:GPU:0)
Train accuracy : Tensor("Accuracy_3/train_accuracy:0", shape=(), dtype=string, device=/device:GPU:0)
Val. loss      : Tensor("Loss_Function_3/validate_error:0", shape=(), dtype=string, device=/device:GPU:0)
Val. accuracy  : Tensor("Accuracy_3/validate_accuracy:0", shape=(), dtype=string, device=/device:GPU:0)
Parameters for Linear classifier
>> Input Parameters
Input                  : Tensor("Inputs_4/Data/X_input:0", shape=(?, 3072), dtype=float32, device=/device:GPU:0) 
True one hot labels    : Tensor("Inputs_4/Train_Labels/y_true_one_hot_label:0", shape=(?, 10), dtype=float32, device=/device:GPU:0) 
True class             : Tensor("Inputs_4/Train_Labels/y_true_class:0", shape=(?,), dtype=int64, device=/device:GPU:0) 
Predict one hot labels : Tensor("Predictions_4/one_hot:0", shape=(?, 10), dtype=float32, device=/device:GPU:0) 
Predict class          : Tensor("Predictions_4/predict_class:0", shape=(?,), dtype=int64, device=/device:GPU:0) 
>> Model params
last_successful_epoch/initial_value
last_successful_epoch
last_successful_epoch/Assign
last_successful_epoch/read
add/y
add
assign_updated_epoch
Inputs/Data/X_input
Inputs/Train_Labels/y_true_one_hot_label
Inputs/Train_Labels/y_true_class
Parameters/Weights/zeros
Parameters/Weights/Weight
Parameters/Weights/Weight/Assign
Parameters/Weights/Weight/read
Parameters/Bias/ones
Parameters/Bias/Bias
Parameters/Bias/Bias/Assign
Parameters/Bias/Bias/read
Predictions/mlp_layer/MatMul
Predictions/mlp_layer
Predictions/mlp_layer_1
Predictions/Softmax
Predictions/predict_class/dimension
Predictions/predict_class
Predictions/one_hot/Const
Predictions/one_hot/Const_1
Predictions/one_hot/depth
Predictions/one_hot/on_value
Predictions/one_hot/off_value
Predictions/one_hot
Cross_Entropy/Rank
Cross_Entropy/Shape
Cross_Entropy/Rank_1
Cross_Entropy/Shape_1
Cross_Entropy/Sub/y
Cross_Entropy/Sub
Cross_Entropy/Slice/begin
Cross_Entropy/Slice/size
Cross_Entropy/Slice
Cross_Entropy/concat/values_0
Cross_Entropy/concat/axis
Cross_Entropy/concat
Cross_Entropy/Reshape
Cross_Entropy/Rank_2
Cross_Entropy/Shape_2
Cross_Entropy/Sub_1/y
Cross_Entropy/Sub_1
Cross_Entropy/Slice_1/begin
Cross_Entropy/Slice_1/size
Cross_Entropy/Slice_1
Cross_Entropy/concat_1/values_0
Cross_Entropy/concat_1/axis
Cross_Entropy/concat_1
Cross_Entropy/Reshape_1
Cross_Entropy/SoftmaxCrossEntropyWithLogits
Cross_Entropy/Sub_2/y
Cross_Entropy/Sub_2
Cross_Entropy/Slice_2/begin
Cross_Entropy/Slice_2/size
Cross_Entropy/Slice_2
Cross_Entropy/Reshape_2
Loss_Function/Const
Loss_Function/Square
Loss_Function/Const_1
Loss_Function/Mean
Loss_Function/Const_2
Loss_Function/Mean_1
Loss_Function/Mul
Loss_Function/Add
Loss_Function/train_loss_list/initial_value
Loss_Function/train_loss_list
Loss_Function/train_loss_list/Assign
Loss_Function/train_loss_list/read
Loss_Function/Assign/value
Loss_Function/Assign
Loss_Function/train_error/tags
Loss_Function/train_error
Loss_Function/validate_loss_list/initial_value
Loss_Function/validate_loss_list
Loss_Function/validate_loss_list/Assign
Loss_Function/validate_loss_list/read
Loss_Function/Assign_1/value
Loss_Function/Assign_1
Loss_Function/validate_error/tags
Loss_Function/validate_error
Optimizer/learning_rate_progress/initial_value
Optimizer/learning_rate_progress
Optimizer/learning_rate_progress/Assign
Optimizer/learning_rate_progress/read
Optimizer/Assign/value
Optimizer/Assign
Optimizer/learning_rate/tags
Optimizer/learning_rate/values
Optimizer/learning_rate
Optimizer/gradients/Shape
Optimizer/gradients/Const
Optimizer/gradients/Fill
Optimizer/gradients/Loss_Function/Add_grad/Shape
Optimizer/gradients/Loss_Function/Add_grad/Shape_1
Optimizer/gradients/Loss_Function/Add_grad/BroadcastGradientArgs
Optimizer/gradients/Loss_Function/Add_grad/Sum
Optimizer/gradients/Loss_Function/Add_grad/Reshape
Optimizer/gradients/Loss_Function/Add_grad/Sum_1
Optimizer/gradients/Loss_Function/Add_grad/Reshape_1
Optimizer/gradients/Loss_Function/Add_grad/tuple/group_deps
Optimizer/gradients/Loss_Function/Add_grad/tuple/control_dependency
Optimizer/gradients/Loss_Function/Add_grad/tuple/control_dependency_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Reshape/shape
Optimizer/gradients/Loss_Function/Mean_1_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape
Optimizer/gradients/Loss_Function/Mean_1_grad/Tile
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape_2
Optimizer/gradients/Loss_Function/Mean_1_grad/Const
Optimizer/gradients/Loss_Function/Mean_1_grad/Prod
Optimizer/gradients/Loss_Function/Mean_1_grad/Const_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Prod_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Maximum/y
Optimizer/gradients/Loss_Function/Mean_1_grad/Maximum
Optimizer/gradients/Loss_Function/Mean_1_grad/floordiv
Optimizer/gradients/Loss_Function/Mean_1_grad/Cast
Optimizer/gradients/Loss_Function/Mean_1_grad/truediv
Optimizer/gradients/Loss_Function/Mul_grad/Shape
Optimizer/gradients/Loss_Function/Mul_grad/Shape_1
Optimizer/gradients/Loss_Function/Mul_grad/BroadcastGradientArgs
Optimizer/gradients/Loss_Function/Mul_grad/mul
Optimizer/gradients/Loss_Function/Mul_grad/Sum
Optimizer/gradients/Loss_Function/Mul_grad/Reshape
Optimizer/gradients/Loss_Function/Mul_grad/mul_1
Optimizer/gradients/Loss_Function/Mul_grad/Sum_1
Optimizer/gradients/Loss_Function/Mul_grad/Reshape_1
Optimizer/gradients/Loss_Function/Mul_grad/tuple/group_deps
Optimizer/gradients/Loss_Function/Mul_grad/tuple/control_dependency
Optimizer/gradients/Loss_Function/Mul_grad/tuple/control_dependency_1
Optimizer/gradients/Cross_Entropy/Reshape_2_grad/Shape
Optimizer/gradients/Cross_Entropy/Reshape_2_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_grad/Reshape/shape
Optimizer/gradients/Loss_Function/Mean_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_grad/Tile/multiples
Optimizer/gradients/Loss_Function/Mean_grad/Tile
Optimizer/gradients/Loss_Function/Mean_grad/Shape
Optimizer/gradients/Loss_Function/Mean_grad/Shape_1
Optimizer/gradients/Loss_Function/Mean_grad/Const
Optimizer/gradients/Loss_Function/Mean_grad/Prod
Optimizer/gradients/Loss_Function/Mean_grad/Const_1
Optimizer/gradients/Loss_Function/Mean_grad/Prod_1
Optimizer/gradients/Loss_Function/Mean_grad/Maximum/y
Optimizer/gradients/Loss_Function/Mean_grad/Maximum
Optimizer/gradients/Loss_Function/Mean_grad/floordiv
Optimizer/gradients/Loss_Function/Mean_grad/Cast
Optimizer/gradients/Loss_Function/Mean_grad/truediv
Optimizer/gradients/zeros_like
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer/gradients/Loss_Function/Square_grad/mul/x
Optimizer/gradients/Loss_Function/Square_grad/mul
Optimizer/gradients/Loss_Function/Square_grad/mul_1
Optimizer/gradients/Cross_Entropy/Reshape_grad/Shape
Optimizer/gradients/Cross_Entropy/Reshape_grad/Reshape
Optimizer/gradients/Predictions/Softmax_grad/mul
Optimizer/gradients/Predictions/Softmax_grad/Sum/reduction_indices
Optimizer/gradients/Predictions/Softmax_grad/Sum
Optimizer/gradients/Predictions/Softmax_grad/Reshape/shape
Optimizer/gradients/Predictions/Softmax_grad/Reshape
Optimizer/gradients/Predictions/Softmax_grad/sub
Optimizer/gradients/Predictions/Softmax_grad/mul_1
Optimizer/gradients/Predictions/mlp_layer_1_grad/ReluGrad
Optimizer/gradients/Predictions/mlp_layer_grad/BiasAddGrad
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/group_deps
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/control_dependency
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/control_dependency_1
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/MatMul
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/MatMul_1
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer/gradients/AddN
Optimizer/GradientDescent/learning_rate
Optimizer/GradientDescent/update_Parameters/Weights/Weight/ApplyGradientDescent
Optimizer/GradientDescent/update_Parameters/Bias/Bias/ApplyGradientDescent
Optimizer/GradientDescent
Equal
Accuracy/Cast
Accuracy/Const
Accuracy/Mean
Accuracy/train_accuracy_list/initial_value
Accuracy/train_accuracy_list
Accuracy/train_accuracy_list/Assign
Accuracy/train_accuracy_list/read
Accuracy/Assign/value
Accuracy/Assign
Accuracy/train_accuracy/tags
Accuracy/train_accuracy
Accuracy/Cast_1
Accuracy/Const_1
Accuracy/Mean_1
Accuracy/validate_accuracy_list/initial_value
Accuracy/validate_accuracy_list
Accuracy/validate_accuracy_list/Assign
Accuracy/validate_accuracy_list/read
Accuracy/Assign_1/value
Accuracy/Assign_1
Accuracy/validate_accuracy/tags
Accuracy/validate_accuracy
Accuracy/Cast_2
Accuracy/Const_2
Accuracy/Mean_2
Accuracy/test_accuracy_list/initial_value
Accuracy/test_accuracy_list
Accuracy/test_accuracy_list/Assign
Accuracy/test_accuracy_list/read
Accuracy/Assign_2/value
Accuracy/Assign_2
Accuracy/test_accuracy/tags
Accuracy/test_accuracy
init/NoOp
init/NoOp_1
init
Merge/MergeSummary
save/Const
save/SaveV2/tensor_names
save/SaveV2/shape_and_slices
save/SaveV2
save/control_dependency
save/RestoreV2/tensor_names
save/RestoreV2/shape_and_slices
save/RestoreV2
save/Assign
save/RestoreV2_1/tensor_names
save/RestoreV2_1/shape_and_slices
save/RestoreV2_1
save/Assign_1
save/RestoreV2_2/tensor_names
save/RestoreV2_2/shape_and_slices
save/RestoreV2_2
save/Assign_2
save/RestoreV2_3/tensor_names
save/RestoreV2_3/shape_and_slices
save/RestoreV2_3
save/Assign_3
save/RestoreV2_4/tensor_names
save/RestoreV2_4/shape_and_slices
save/RestoreV2_4
save/Assign_4
save/RestoreV2_5/tensor_names
save/RestoreV2_5/shape_and_slices
save/RestoreV2_5
save/Assign_5
save/RestoreV2_6/tensor_names
save/RestoreV2_6/shape_and_slices
save/RestoreV2_6
save/Assign_6
save/RestoreV2_7/tensor_names
save/RestoreV2_7/shape_and_slices
save/RestoreV2_7
save/Assign_7
save/RestoreV2_8/tensor_names
save/RestoreV2_8/shape_and_slices
save/RestoreV2_8
save/Assign_8
save/restore_all/NoOp
save/restore_all/NoOp_1
save/restore_all
last_successful_epoch_1/initial_value
last_successful_epoch_1
last_successful_epoch_1/Assign
last_successful_epoch_1/read
add_1/y
add_1
assign_updated_epoch_1
Inputs_1/Data/X_input
Inputs_1/Train_Labels/y_true_one_hot_label
Inputs_1/Train_Labels/y_true_class
Parameters_1/Weights/zeros
Parameters_1/Weights/Weight
Parameters_1/Weights/Weight/Assign
Parameters_1/Weights/Weight/read
Parameters_1/Bias/ones
Parameters_1/Bias/Bias
Parameters_1/Bias/Bias/Assign
Parameters_1/Bias/Bias/read
Predictions_1/mlp_layer/MatMul
Predictions_1/mlp_layer
Predictions_1/mlp_layer_1
Predictions_1/Softmax
Predictions_1/predict_class/dimension
Predictions_1/predict_class
Predictions_1/one_hot/Const
Predictions_1/one_hot/Const_1
Predictions_1/one_hot/depth
Predictions_1/one_hot/on_value
Predictions_1/one_hot/off_value
Predictions_1/one_hot
Cross_Entropy_1/Rank
Cross_Entropy_1/Shape
Cross_Entropy_1/Rank_1
Cross_Entropy_1/Shape_1
Cross_Entropy_1/Sub/y
Cross_Entropy_1/Sub
Cross_Entropy_1/Slice/begin
Cross_Entropy_1/Slice/size
Cross_Entropy_1/Slice
Cross_Entropy_1/concat/values_0
Cross_Entropy_1/concat/axis
Cross_Entropy_1/concat
Cross_Entropy_1/Reshape
Cross_Entropy_1/Rank_2
Cross_Entropy_1/Shape_2
Cross_Entropy_1/Sub_1/y
Cross_Entropy_1/Sub_1
Cross_Entropy_1/Slice_1/begin
Cross_Entropy_1/Slice_1/size
Cross_Entropy_1/Slice_1
Cross_Entropy_1/concat_1/values_0
Cross_Entropy_1/concat_1/axis
Cross_Entropy_1/concat_1
Cross_Entropy_1/Reshape_1
Cross_Entropy_1/SoftmaxCrossEntropyWithLogits
Cross_Entropy_1/Sub_2/y
Cross_Entropy_1/Sub_2
Cross_Entropy_1/Slice_2/begin
Cross_Entropy_1/Slice_2/size
Cross_Entropy_1/Slice_2
Cross_Entropy_1/Reshape_2
Loss_Function_1/Const
Loss_Function_1/Square
Loss_Function_1/Const_1
Loss_Function_1/Mean
Loss_Function_1/Const_2
Loss_Function_1/Mean_1
Loss_Function_1/Mul
Loss_Function_1/Add
Loss_Function_1/train_loss_list/initial_value
Loss_Function_1/train_loss_list
Loss_Function_1/train_loss_list/Assign
Loss_Function_1/train_loss_list/read
Loss_Function_1/Assign/value
Loss_Function_1/Assign
Loss_Function_1/train_error/tags
Loss_Function_1/train_error
Loss_Function_1/validate_loss_list/initial_value
Loss_Function_1/validate_loss_list
Loss_Function_1/validate_loss_list/Assign
Loss_Function_1/validate_loss_list/read
Loss_Function_1/Assign_1/value
Loss_Function_1/Assign_1
Loss_Function_1/validate_error/tags
Loss_Function_1/validate_error
Optimizer_1/learning_rate_progress/initial_value
Optimizer_1/learning_rate_progress
Optimizer_1/learning_rate_progress/Assign
Optimizer_1/learning_rate_progress/read
Optimizer_1/Assign/value
Optimizer_1/Assign
Optimizer_1/learning_rate/tags
Optimizer_1/learning_rate/values
Optimizer_1/learning_rate
Optimizer_1/gradients/Shape
Optimizer_1/gradients/Const
Optimizer_1/gradients/Fill
Optimizer_1/gradients/Loss_Function_1/Add_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Add_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/BroadcastGradientArgs
Optimizer_1/gradients/Loss_Function_1/Add_grad/Sum
Optimizer_1/gradients/Loss_Function_1/Add_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Add_grad/Sum_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/Reshape_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/group_deps
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/control_dependency
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/control_dependency_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Reshape/shape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Tile
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape_2
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Const
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Prod
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Const_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Prod_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Maximum/y
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Maximum
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/floordiv
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Cast
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/truediv
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/BroadcastGradientArgs
Optimizer_1/gradients/Loss_Function_1/Mul_grad/mul
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Sum
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mul_grad/mul_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Sum_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Reshape_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/group_deps
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/control_dependency
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/control_dependency_1
Optimizer_1/gradients/Cross_Entropy_1/Reshape_2_grad/Shape
Optimizer_1/gradients/Cross_Entropy_1/Reshape_2_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Reshape/shape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Tile/multiples
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Tile
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Const
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Prod
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Const_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Prod_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Maximum/y
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Maximum
Optimizer_1/gradients/Loss_Function_1/Mean_grad/floordiv
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Cast
Optimizer_1/gradients/Loss_Function_1/Mean_grad/truediv
Optimizer_1/gradients/zeros_like
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul/x
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul_1
Optimizer_1/gradients/Cross_Entropy_1/Reshape_grad/Shape
Optimizer_1/gradients/Cross_Entropy_1/Reshape_grad/Reshape
Optimizer_1/gradients/Predictions_1/Softmax_grad/mul
Optimizer_1/gradients/Predictions_1/Softmax_grad/Sum/reduction_indices
Optimizer_1/gradients/Predictions_1/Softmax_grad/Sum
Optimizer_1/gradients/Predictions_1/Softmax_grad/Reshape/shape
Optimizer_1/gradients/Predictions_1/Softmax_grad/Reshape
Optimizer_1/gradients/Predictions_1/Softmax_grad/sub
Optimizer_1/gradients/Predictions_1/Softmax_grad/mul_1
Optimizer_1/gradients/Predictions_1/mlp_layer_1_grad/ReluGrad
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/BiasAddGrad
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/group_deps
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/control_dependency
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/control_dependency_1
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/MatMul
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/MatMul_1
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_1/gradients/AddN
Optimizer_1/GradientDescent/learning_rate
Optimizer_1/GradientDescent/update_Parameters_1/Weights/Weight/ApplyGradientDescent
Optimizer_1/GradientDescent/update_Parameters_1/Bias/Bias/ApplyGradientDescent
Optimizer_1/GradientDescent
Equal_1
Accuracy_1/Cast
Accuracy_1/Const
Accuracy_1/Mean
Accuracy_1/train_accuracy_list/initial_value
Accuracy_1/train_accuracy_list
Accuracy_1/train_accuracy_list/Assign
Accuracy_1/train_accuracy_list/read
Accuracy_1/Assign/value
Accuracy_1/Assign
Accuracy_1/train_accuracy/tags
Accuracy_1/train_accuracy
Accuracy_1/Cast_1
Accuracy_1/Const_1
Accuracy_1/Mean_1
Accuracy_1/validate_accuracy_list/initial_value
Accuracy_1/validate_accuracy_list
Accuracy_1/validate_accuracy_list/Assign
Accuracy_1/validate_accuracy_list/read
Accuracy_1/Assign_1/value
Accuracy_1/Assign_1
Accuracy_1/validate_accuracy/tags
Accuracy_1/validate_accuracy
Accuracy_1/Cast_2
Accuracy_1/Const_2
Accuracy_1/Mean_2
Accuracy_1/test_accuracy_list/initial_value
Accuracy_1/test_accuracy_list
Accuracy_1/test_accuracy_list/Assign
Accuracy_1/test_accuracy_list/read
Accuracy_1/Assign_2/value
Accuracy_1/Assign_2
Accuracy_1/test_accuracy/tags
Accuracy_1/test_accuracy
init_1/NoOp
init_1/NoOp_1
init_1
Merge_1/MergeSummary
save_1/Const
save_1/SaveV2/tensor_names
save_1/SaveV2/shape_and_slices
save_1/SaveV2
save_1/control_dependency
save_1/RestoreV2/tensor_names
save_1/RestoreV2/shape_and_slices
save_1/RestoreV2
save_1/Assign
save_1/RestoreV2_1/tensor_names
save_1/RestoreV2_1/shape_and_slices
save_1/RestoreV2_1
save_1/Assign_1
save_1/RestoreV2_2/tensor_names
save_1/RestoreV2_2/shape_and_slices
save_1/RestoreV2_2
save_1/Assign_2
save_1/RestoreV2_3/tensor_names
save_1/RestoreV2_3/shape_and_slices
save_1/RestoreV2_3
save_1/Assign_3
save_1/RestoreV2_4/tensor_names
save_1/RestoreV2_4/shape_and_slices
save_1/RestoreV2_4
save_1/Assign_4
save_1/RestoreV2_5/tensor_names
save_1/RestoreV2_5/shape_and_slices
save_1/RestoreV2_5
save_1/Assign_5
save_1/RestoreV2_6/tensor_names
save_1/RestoreV2_6/shape_and_slices
save_1/RestoreV2_6
save_1/Assign_6
save_1/RestoreV2_7/tensor_names
save_1/RestoreV2_7/shape_and_slices
save_1/RestoreV2_7
save_1/Assign_7
save_1/RestoreV2_8/tensor_names
save_1/RestoreV2_8/shape_and_slices
save_1/RestoreV2_8
save_1/Assign_8
save_1/RestoreV2_9/tensor_names
save_1/RestoreV2_9/shape_and_slices
save_1/RestoreV2_9
save_1/Assign_9
save_1/RestoreV2_10/tensor_names
save_1/RestoreV2_10/shape_and_slices
save_1/RestoreV2_10
save_1/Assign_10
save_1/RestoreV2_11/tensor_names
save_1/RestoreV2_11/shape_and_slices
save_1/RestoreV2_11
save_1/Assign_11
save_1/RestoreV2_12/tensor_names
save_1/RestoreV2_12/shape_and_slices
save_1/RestoreV2_12
save_1/Assign_12
save_1/RestoreV2_13/tensor_names
save_1/RestoreV2_13/shape_and_slices
save_1/RestoreV2_13
save_1/Assign_13
save_1/RestoreV2_14/tensor_names
save_1/RestoreV2_14/shape_and_slices
save_1/RestoreV2_14
save_1/Assign_14
save_1/RestoreV2_15/tensor_names
save_1/RestoreV2_15/shape_and_slices
save_1/RestoreV2_15
save_1/Assign_15
save_1/RestoreV2_16/tensor_names
save_1/RestoreV2_16/shape_and_slices
save_1/RestoreV2_16
save_1/Assign_16
save_1/RestoreV2_17/tensor_names
save_1/RestoreV2_17/shape_and_slices
save_1/RestoreV2_17
save_1/Assign_17
save_1/restore_all/NoOp
save_1/restore_all/NoOp_1
save_1/restore_all
last_successful_epoch_2/initial_value
last_successful_epoch_2
last_successful_epoch_2/Assign
last_successful_epoch_2/read
add_2/y
add_2
assign_updated_epoch_2
Inputs_2/Data/X_input
Inputs_2/Train_Labels/y_true_one_hot_label
Inputs_2/Train_Labels/y_true_class
Parameters_2/Weights/zeros
Parameters_2/Weights/Weight
Parameters_2/Weights/Weight/Assign
Parameters_2/Weights/Weight/read
Parameters_2/Bias/ones
Parameters_2/Bias/Bias
Parameters_2/Bias/Bias/Assign
Parameters_2/Bias/Bias/read
Predictions_2/mlp_layer/MatMul
Predictions_2/mlp_layer
Predictions_2/mlp_layer_1
Predictions_2/Softmax
Predictions_2/predict_class/dimension
Predictions_2/predict_class
Predictions_2/one_hot/Const
Predictions_2/one_hot/Const_1
Predictions_2/one_hot/depth
Predictions_2/one_hot/on_value
Predictions_2/one_hot/off_value
Predictions_2/one_hot
Cross_Entropy_2/Rank
Cross_Entropy_2/Shape
Cross_Entropy_2/Rank_1
Cross_Entropy_2/Shape_1
Cross_Entropy_2/Sub/y
Cross_Entropy_2/Sub
Cross_Entropy_2/Slice/begin
Cross_Entropy_2/Slice/size
Cross_Entropy_2/Slice
Cross_Entropy_2/concat/values_0
Cross_Entropy_2/concat/axis
Cross_Entropy_2/concat
Cross_Entropy_2/Reshape
Cross_Entropy_2/Rank_2
Cross_Entropy_2/Shape_2
Cross_Entropy_2/Sub_1/y
Cross_Entropy_2/Sub_1
Cross_Entropy_2/Slice_1/begin
Cross_Entropy_2/Slice_1/size
Cross_Entropy_2/Slice_1
Cross_Entropy_2/concat_1/values_0
Cross_Entropy_2/concat_1/axis
Cross_Entropy_2/concat_1
Cross_Entropy_2/Reshape_1
Cross_Entropy_2/SoftmaxCrossEntropyWithLogits
Cross_Entropy_2/Sub_2/y
Cross_Entropy_2/Sub_2
Cross_Entropy_2/Slice_2/begin
Cross_Entropy_2/Slice_2/size
Cross_Entropy_2/Slice_2
Cross_Entropy_2/Reshape_2
Loss_Function_2/Const
Loss_Function_2/Square
Loss_Function_2/Const_1
Loss_Function_2/Mean
Loss_Function_2/Const_2
Loss_Function_2/Mean_1
Loss_Function_2/Mul
Loss_Function_2/Add
Loss_Function_2/train_loss_list/initial_value
Loss_Function_2/train_loss_list
Loss_Function_2/train_loss_list/Assign
Loss_Function_2/train_loss_list/read
Loss_Function_2/Assign/value
Loss_Function_2/Assign
Loss_Function_2/train_error/tags
Loss_Function_2/train_error
Loss_Function_2/validate_loss_list/initial_value
Loss_Function_2/validate_loss_list
Loss_Function_2/validate_loss_list/Assign
Loss_Function_2/validate_loss_list/read
Loss_Function_2/Assign_1/value
Loss_Function_2/Assign_1
Loss_Function_2/validate_error/tags
Loss_Function_2/validate_error
Optimizer_2/learning_rate_progress/initial_value
Optimizer_2/learning_rate_progress
Optimizer_2/learning_rate_progress/Assign
Optimizer_2/learning_rate_progress/read
Optimizer_2/Assign/value
Optimizer_2/Assign
Optimizer_2/learning_rate/tags
Optimizer_2/learning_rate/values
Optimizer_2/learning_rate
Optimizer_2/gradients/Shape
Optimizer_2/gradients/Const
Optimizer_2/gradients/Fill
Optimizer_2/gradients/Loss_Function_2/Add_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Add_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/BroadcastGradientArgs
Optimizer_2/gradients/Loss_Function_2/Add_grad/Sum
Optimizer_2/gradients/Loss_Function_2/Add_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Add_grad/Sum_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/Reshape_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/group_deps
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/control_dependency
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/control_dependency_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Reshape/shape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Tile
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape_2
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Const
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Prod
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Const_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Prod_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Maximum/y
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Maximum
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/floordiv
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Cast
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/truediv
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/BroadcastGradientArgs
Optimizer_2/gradients/Loss_Function_2/Mul_grad/mul
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Sum
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mul_grad/mul_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Sum_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Reshape_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/group_deps
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/control_dependency
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/control_dependency_1
Optimizer_2/gradients/Cross_Entropy_2/Reshape_2_grad/Shape
Optimizer_2/gradients/Cross_Entropy_2/Reshape_2_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Reshape/shape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Tile/multiples
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Tile
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Const
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Prod
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Const_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Prod_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Maximum/y
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Maximum
Optimizer_2/gradients/Loss_Function_2/Mean_grad/floordiv
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Cast
Optimizer_2/gradients/Loss_Function_2/Mean_grad/truediv
Optimizer_2/gradients/zeros_like
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul/x
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul_1
Optimizer_2/gradients/Cross_Entropy_2/Reshape_grad/Shape
Optimizer_2/gradients/Cross_Entropy_2/Reshape_grad/Reshape
Optimizer_2/gradients/Predictions_2/Softmax_grad/mul
Optimizer_2/gradients/Predictions_2/Softmax_grad/Sum/reduction_indices
Optimizer_2/gradients/Predictions_2/Softmax_grad/Sum
Optimizer_2/gradients/Predictions_2/Softmax_grad/Reshape/shape
Optimizer_2/gradients/Predictions_2/Softmax_grad/Reshape
Optimizer_2/gradients/Predictions_2/Softmax_grad/sub
Optimizer_2/gradients/Predictions_2/Softmax_grad/mul_1
Optimizer_2/gradients/Predictions_2/mlp_layer_1_grad/ReluGrad
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/BiasAddGrad
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/group_deps
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/control_dependency
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/control_dependency_1
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/MatMul
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/MatMul_1
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_2/gradients/AddN
Optimizer_2/GradientDescent/learning_rate
Optimizer_2/GradientDescent/update_Parameters_2/Weights/Weight/ApplyGradientDescent
Optimizer_2/GradientDescent/update_Parameters_2/Bias/Bias/ApplyGradientDescent
Optimizer_2/GradientDescent
Equal_2
Accuracy_2/Cast
Accuracy_2/Const
Accuracy_2/Mean
Accuracy_2/train_accuracy_list/initial_value
Accuracy_2/train_accuracy_list
Accuracy_2/train_accuracy_list/Assign
Accuracy_2/train_accuracy_list/read
Accuracy_2/Assign/value
Accuracy_2/Assign
Accuracy_2/train_accuracy/tags
Accuracy_2/train_accuracy
Accuracy_2/Cast_1
Accuracy_2/Const_1
Accuracy_2/Mean_1
Accuracy_2/validate_accuracy_list/initial_value
Accuracy_2/validate_accuracy_list
Accuracy_2/validate_accuracy_list/Assign
Accuracy_2/validate_accuracy_list/read
Accuracy_2/Assign_1/value
Accuracy_2/Assign_1
Accuracy_2/validate_accuracy/tags
Accuracy_2/validate_accuracy
Accuracy_2/Cast_2
Accuracy_2/Const_2
Accuracy_2/Mean_2
Accuracy_2/test_accuracy_list/initial_value
Accuracy_2/test_accuracy_list
Accuracy_2/test_accuracy_list/Assign
Accuracy_2/test_accuracy_list/read
Accuracy_2/Assign_2/value
Accuracy_2/Assign_2
Accuracy_2/test_accuracy/tags
Accuracy_2/test_accuracy
init_2/NoOp
init_2/NoOp_1
init_2
Merge_2/MergeSummary
save_2/Const
save_2/SaveV2/tensor_names
save_2/SaveV2/shape_and_slices
save_2/SaveV2
save_2/control_dependency
save_2/RestoreV2/tensor_names
save_2/RestoreV2/shape_and_slices
save_2/RestoreV2
save_2/Assign
save_2/RestoreV2_1/tensor_names
save_2/RestoreV2_1/shape_and_slices
save_2/RestoreV2_1
save_2/Assign_1
save_2/RestoreV2_2/tensor_names
save_2/RestoreV2_2/shape_and_slices
save_2/RestoreV2_2
save_2/Assign_2
save_2/RestoreV2_3/tensor_names
save_2/RestoreV2_3/shape_and_slices
save_2/RestoreV2_3
save_2/Assign_3
save_2/RestoreV2_4/tensor_names
save_2/RestoreV2_4/shape_and_slices
save_2/RestoreV2_4
save_2/Assign_4
save_2/RestoreV2_5/tensor_names
save_2/RestoreV2_5/shape_and_slices
save_2/RestoreV2_5
save_2/Assign_5
save_2/RestoreV2_6/tensor_names
save_2/RestoreV2_6/shape_and_slices
save_2/RestoreV2_6
save_2/Assign_6
save_2/RestoreV2_7/tensor_names
save_2/RestoreV2_7/shape_and_slices
save_2/RestoreV2_7
save_2/Assign_7
save_2/RestoreV2_8/tensor_names
save_2/RestoreV2_8/shape_and_slices
save_2/RestoreV2_8
save_2/Assign_8
save_2/RestoreV2_9/tensor_names
save_2/RestoreV2_9/shape_and_slices
save_2/RestoreV2_9
save_2/Assign_9
save_2/RestoreV2_10/tensor_names
save_2/RestoreV2_10/shape_and_slices
save_2/RestoreV2_10
save_2/Assign_10
save_2/RestoreV2_11/tensor_names
save_2/RestoreV2_11/shape_and_slices
save_2/RestoreV2_11
save_2/Assign_11
save_2/RestoreV2_12/tensor_names
save_2/RestoreV2_12/shape_and_slices
save_2/RestoreV2_12
save_2/Assign_12
save_2/RestoreV2_13/tensor_names
save_2/RestoreV2_13/shape_and_slices
save_2/RestoreV2_13
save_2/Assign_13
save_2/RestoreV2_14/tensor_names
save_2/RestoreV2_14/shape_and_slices
save_2/RestoreV2_14
save_2/Assign_14
save_2/RestoreV2_15/tensor_names
save_2/RestoreV2_15/shape_and_slices
save_2/RestoreV2_15
save_2/Assign_15
save_2/RestoreV2_16/tensor_names
save_2/RestoreV2_16/shape_and_slices
save_2/RestoreV2_16
save_2/Assign_16
save_2/RestoreV2_17/tensor_names
save_2/RestoreV2_17/shape_and_slices
save_2/RestoreV2_17
save_2/Assign_17
save_2/RestoreV2_18/tensor_names
save_2/RestoreV2_18/shape_and_slices
save_2/RestoreV2_18
save_2/Assign_18
save_2/RestoreV2_19/tensor_names
save_2/RestoreV2_19/shape_and_slices
save_2/RestoreV2_19
save_2/Assign_19
save_2/RestoreV2_20/tensor_names
save_2/RestoreV2_20/shape_and_slices
save_2/RestoreV2_20
save_2/Assign_20
save_2/RestoreV2_21/tensor_names
save_2/RestoreV2_21/shape_and_slices
save_2/RestoreV2_21
save_2/Assign_21
save_2/RestoreV2_22/tensor_names
save_2/RestoreV2_22/shape_and_slices
save_2/RestoreV2_22
save_2/Assign_22
save_2/RestoreV2_23/tensor_names
save_2/RestoreV2_23/shape_and_slices
save_2/RestoreV2_23
save_2/Assign_23
save_2/RestoreV2_24/tensor_names
save_2/RestoreV2_24/shape_and_slices
save_2/RestoreV2_24
save_2/Assign_24
save_2/RestoreV2_25/tensor_names
save_2/RestoreV2_25/shape_and_slices
save_2/RestoreV2_25
save_2/Assign_25
save_2/RestoreV2_26/tensor_names
save_2/RestoreV2_26/shape_and_slices
save_2/RestoreV2_26
save_2/Assign_26
save_2/restore_all/NoOp
save_2/restore_all/NoOp_1
save_2/restore_all
last_successful_epoch_3/initial_value
last_successful_epoch_3
last_successful_epoch_3/Assign
last_successful_epoch_3/read
add_3/y
add_3
assign_updated_epoch_3
Inputs_3/Data/X_input
Inputs_3/Train_Labels/y_true_one_hot_label
Inputs_3/Train_Labels/y_true_class
Parameters_3/Weights/zeros
Parameters_3/Weights/Weight
Parameters_3/Weights/Weight/Assign
Parameters_3/Weights/Weight/read
Parameters_3/Bias/ones
Parameters_3/Bias/Bias
Parameters_3/Bias/Bias/Assign
Parameters_3/Bias/Bias/read
Predictions_3/mlp_layer/MatMul
Predictions_3/mlp_layer
Predictions_3/mlp_layer_1
Predictions_3/Softmax
Predictions_3/predict_class/dimension
Predictions_3/predict_class
Predictions_3/one_hot/Const
Predictions_3/one_hot/Const_1
Predictions_3/one_hot/depth
Predictions_3/one_hot/on_value
Predictions_3/one_hot/off_value
Predictions_3/one_hot
Cross_Entropy_3/Rank
Cross_Entropy_3/Shape
Cross_Entropy_3/Rank_1
Cross_Entropy_3/Shape_1
Cross_Entropy_3/Sub/y
Cross_Entropy_3/Sub
Cross_Entropy_3/Slice/begin
Cross_Entropy_3/Slice/size
Cross_Entropy_3/Slice
Cross_Entropy_3/concat/values_0
Cross_Entropy_3/concat/axis
Cross_Entropy_3/concat
Cross_Entropy_3/Reshape
Cross_Entropy_3/Rank_2
Cross_Entropy_3/Shape_2
Cross_Entropy_3/Sub_1/y
Cross_Entropy_3/Sub_1
Cross_Entropy_3/Slice_1/begin
Cross_Entropy_3/Slice_1/size
Cross_Entropy_3/Slice_1
Cross_Entropy_3/concat_1/values_0
Cross_Entropy_3/concat_1/axis
Cross_Entropy_3/concat_1
Cross_Entropy_3/Reshape_1
Cross_Entropy_3/SoftmaxCrossEntropyWithLogits
Cross_Entropy_3/Sub_2/y
Cross_Entropy_3/Sub_2
Cross_Entropy_3/Slice_2/begin
Cross_Entropy_3/Slice_2/size
Cross_Entropy_3/Slice_2
Cross_Entropy_3/Reshape_2
Loss_Function_3/Const
Loss_Function_3/Square
Loss_Function_3/Const_1
Loss_Function_3/Mean
Loss_Function_3/Const_2
Loss_Function_3/Mean_1
Loss_Function_3/Mul
Loss_Function_3/Add
Loss_Function_3/train_loss_list/initial_value
Loss_Function_3/train_loss_list
Loss_Function_3/train_loss_list/Assign
Loss_Function_3/train_loss_list/read
Loss_Function_3/Assign/value
Loss_Function_3/Assign
Loss_Function_3/train_error/tags
Loss_Function_3/train_error
Loss_Function_3/validate_loss_list/initial_value
Loss_Function_3/validate_loss_list
Loss_Function_3/validate_loss_list/Assign
Loss_Function_3/validate_loss_list/read
Loss_Function_3/Assign_1/value
Loss_Function_3/Assign_1
Loss_Function_3/validate_error/tags
Loss_Function_3/validate_error
Optimizer_3/learning_rate_progress/initial_value
Optimizer_3/learning_rate_progress
Optimizer_3/learning_rate_progress/Assign
Optimizer_3/learning_rate_progress/read
Optimizer_3/Assign/value
Optimizer_3/Assign
Optimizer_3/learning_rate/tags
Optimizer_3/learning_rate/values
Optimizer_3/learning_rate
Optimizer_3/gradients/Shape
Optimizer_3/gradients/Const
Optimizer_3/gradients/Fill
Optimizer_3/gradients/Loss_Function_3/Add_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Add_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/BroadcastGradientArgs
Optimizer_3/gradients/Loss_Function_3/Add_grad/Sum
Optimizer_3/gradients/Loss_Function_3/Add_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Add_grad/Sum_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/Reshape_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/group_deps
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/control_dependency
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/control_dependency_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Reshape/shape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Tile
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape_2
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Const
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Prod
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Const_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Prod_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Maximum/y
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Maximum
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/floordiv
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Cast
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/truediv
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/BroadcastGradientArgs
Optimizer_3/gradients/Loss_Function_3/Mul_grad/mul
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Sum
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mul_grad/mul_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Sum_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Reshape_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/group_deps
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/control_dependency
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/control_dependency_1
Optimizer_3/gradients/Cross_Entropy_3/Reshape_2_grad/Shape
Optimizer_3/gradients/Cross_Entropy_3/Reshape_2_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Reshape/shape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Tile/multiples
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Tile
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Const
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Prod
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Const_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Prod_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Maximum/y
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Maximum
Optimizer_3/gradients/Loss_Function_3/Mean_grad/floordiv
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Cast
Optimizer_3/gradients/Loss_Function_3/Mean_grad/truediv
Optimizer_3/gradients/zeros_like
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul/x
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul_1
Optimizer_3/gradients/Cross_Entropy_3/Reshape_grad/Shape
Optimizer_3/gradients/Cross_Entropy_3/Reshape_grad/Reshape
Optimizer_3/gradients/Predictions_3/Softmax_grad/mul
Optimizer_3/gradients/Predictions_3/Softmax_grad/Sum/reduction_indices
Optimizer_3/gradients/Predictions_3/Softmax_grad/Sum
Optimizer_3/gradients/Predictions_3/Softmax_grad/Reshape/shape
Optimizer_3/gradients/Predictions_3/Softmax_grad/Reshape
Optimizer_3/gradients/Predictions_3/Softmax_grad/sub
Optimizer_3/gradients/Predictions_3/Softmax_grad/mul_1
Optimizer_3/gradients/Predictions_3/mlp_layer_1_grad/ReluGrad
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/BiasAddGrad
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/group_deps
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/control_dependency
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/control_dependency_1
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/MatMul
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/MatMul_1
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_3/gradients/AddN
Optimizer_3/GradientDescent/learning_rate
Optimizer_3/GradientDescent/update_Parameters_3/Weights/Weight/ApplyGradientDescent
Optimizer_3/GradientDescent/update_Parameters_3/Bias/Bias/ApplyGradientDescent
Optimizer_3/GradientDescent
Equal_3
Accuracy_3/Cast
Accuracy_3/Const
Accuracy_3/Mean
Accuracy_3/train_accuracy_list/initial_value
Accuracy_3/train_accuracy_list
Accuracy_3/train_accuracy_list/Assign
Accuracy_3/train_accuracy_list/read
Accuracy_3/Assign/value
Accuracy_3/Assign
Accuracy_3/train_accuracy/tags
Accuracy_3/train_accuracy
Accuracy_3/Cast_1
Accuracy_3/Const_1
Accuracy_3/Mean_1
Accuracy_3/validate_accuracy_list/initial_value
Accuracy_3/validate_accuracy_list
Accuracy_3/validate_accuracy_list/Assign
Accuracy_3/validate_accuracy_list/read
Accuracy_3/Assign_1/value
Accuracy_3/Assign_1
Accuracy_3/validate_accuracy/tags
Accuracy_3/validate_accuracy
Accuracy_3/Cast_2
Accuracy_3/Const_2
Accuracy_3/Mean_2
Accuracy_3/test_accuracy_list/initial_value
Accuracy_3/test_accuracy_list
Accuracy_3/test_accuracy_list/Assign
Accuracy_3/test_accuracy_list/read
Accuracy_3/Assign_2/value
Accuracy_3/Assign_2
Accuracy_3/test_accuracy/tags
Accuracy_3/test_accuracy
init_3/NoOp
init_3/NoOp_1
init_3
Merge_3/MergeSummary
save_3/Const
save_3/SaveV2/tensor_names
save_3/SaveV2/shape_and_slices
save_3/SaveV2
save_3/control_dependency
save_3/RestoreV2/tensor_names
save_3/RestoreV2/shape_and_slices
save_3/RestoreV2
save_3/Assign
save_3/RestoreV2_1/tensor_names
save_3/RestoreV2_1/shape_and_slices
save_3/RestoreV2_1
save_3/Assign_1
save_3/RestoreV2_2/tensor_names
save_3/RestoreV2_2/shape_and_slices
save_3/RestoreV2_2
save_3/Assign_2
save_3/RestoreV2_3/tensor_names
save_3/RestoreV2_3/shape_and_slices
save_3/RestoreV2_3
save_3/Assign_3
save_3/RestoreV2_4/tensor_names
save_3/RestoreV2_4/shape_and_slices
save_3/RestoreV2_4
save_3/Assign_4
save_3/RestoreV2_5/tensor_names
save_3/RestoreV2_5/shape_and_slices
save_3/RestoreV2_5
save_3/Assign_5
save_3/RestoreV2_6/tensor_names
save_3/RestoreV2_6/shape_and_slices
save_3/RestoreV2_6
save_3/Assign_6
save_3/RestoreV2_7/tensor_names
save_3/RestoreV2_7/shape_and_slices
save_3/RestoreV2_7
save_3/Assign_7
save_3/RestoreV2_8/tensor_names
save_3/RestoreV2_8/shape_and_slices
save_3/RestoreV2_8
save_3/Assign_8
save_3/RestoreV2_9/tensor_names
save_3/RestoreV2_9/shape_and_slices
save_3/RestoreV2_9
save_3/Assign_9
save_3/RestoreV2_10/tensor_names
save_3/RestoreV2_10/shape_and_slices
save_3/RestoreV2_10
save_3/Assign_10
save_3/RestoreV2_11/tensor_names
save_3/RestoreV2_11/shape_and_slices
save_3/RestoreV2_11
save_3/Assign_11
save_3/RestoreV2_12/tensor_names
save_3/RestoreV2_12/shape_and_slices
save_3/RestoreV2_12
save_3/Assign_12
save_3/RestoreV2_13/tensor_names
save_3/RestoreV2_13/shape_and_slices
save_3/RestoreV2_13
save_3/Assign_13
save_3/RestoreV2_14/tensor_names
save_3/RestoreV2_14/shape_and_slices
save_3/RestoreV2_14
save_3/Assign_14
save_3/RestoreV2_15/tensor_names
save_3/RestoreV2_15/shape_and_slices
save_3/RestoreV2_15
save_3/Assign_15
save_3/RestoreV2_16/tensor_names
save_3/RestoreV2_16/shape_and_slices
save_3/RestoreV2_16
save_3/Assign_16
save_3/RestoreV2_17/tensor_names
save_3/RestoreV2_17/shape_and_slices
save_3/RestoreV2_17
save_3/Assign_17
save_3/RestoreV2_18/tensor_names
save_3/RestoreV2_18/shape_and_slices
save_3/RestoreV2_18
save_3/Assign_18
save_3/RestoreV2_19/tensor_names
save_3/RestoreV2_19/shape_and_slices
save_3/RestoreV2_19
save_3/Assign_19
save_3/RestoreV2_20/tensor_names
save_3/RestoreV2_20/shape_and_slices
save_3/RestoreV2_20
save_3/Assign_20
save_3/RestoreV2_21/tensor_names
save_3/RestoreV2_21/shape_and_slices
save_3/RestoreV2_21
save_3/Assign_21
save_3/RestoreV2_22/tensor_names
save_3/RestoreV2_22/shape_and_slices
save_3/RestoreV2_22
save_3/Assign_22
save_3/RestoreV2_23/tensor_names
save_3/RestoreV2_23/shape_and_slices
save_3/RestoreV2_23
save_3/Assign_23
save_3/RestoreV2_24/tensor_names
save_3/RestoreV2_24/shape_and_slices
save_3/RestoreV2_24
save_3/Assign_24
save_3/RestoreV2_25/tensor_names
save_3/RestoreV2_25/shape_and_slices
save_3/RestoreV2_25
save_3/Assign_25
save_3/RestoreV2_26/tensor_names
save_3/RestoreV2_26/shape_and_slices
save_3/RestoreV2_26
save_3/Assign_26
save_3/RestoreV2_27/tensor_names
save_3/RestoreV2_27/shape_and_slices
save_3/RestoreV2_27
save_3/Assign_27
save_3/RestoreV2_28/tensor_names
save_3/RestoreV2_28/shape_and_slices
save_3/RestoreV2_28
save_3/Assign_28
save_3/RestoreV2_29/tensor_names
save_3/RestoreV2_29/shape_and_slices
save_3/RestoreV2_29
save_3/Assign_29
save_3/RestoreV2_30/tensor_names
save_3/RestoreV2_30/shape_and_slices
save_3/RestoreV2_30
save_3/Assign_30
save_3/RestoreV2_31/tensor_names
save_3/RestoreV2_31/shape_and_slices
save_3/RestoreV2_31
save_3/Assign_31
save_3/RestoreV2_32/tensor_names
save_3/RestoreV2_32/shape_and_slices
save_3/RestoreV2_32
save_3/Assign_32
save_3/RestoreV2_33/tensor_names
save_3/RestoreV2_33/shape_and_slices
save_3/RestoreV2_33
save_3/Assign_33
save_3/RestoreV2_34/tensor_names
save_3/RestoreV2_34/shape_and_slices
save_3/RestoreV2_34
save_3/Assign_34
save_3/RestoreV2_35/tensor_names
save_3/RestoreV2_35/shape_and_slices
save_3/RestoreV2_35
save_3/Assign_35
save_3/restore_all/NoOp
save_3/restore_all/NoOp_1
save_3/restore_all
last_successful_epoch_4/initial_value
last_successful_epoch_4
last_successful_epoch_4/Assign
last_successful_epoch_4/read
add_4/y
add_4
assign_updated_epoch_4
Inputs_4/Data/X_input
Inputs_4/Train_Labels/y_true_one_hot_label
Inputs_4/Train_Labels/y_true_class
Parameters_4/Weights/zeros
Parameters_4/Weights/Weight
Parameters_4/Weights/Weight/Assign
Parameters_4/Weights/Weight/read
Parameters_4/Bias/ones
Parameters_4/Bias/Bias
Parameters_4/Bias/Bias/Assign
Parameters_4/Bias/Bias/read
Predictions_4/mlp_layer/MatMul
Predictions_4/mlp_layer
Predictions_4/mlp_layer_1
Predictions_4/Softmax
Predictions_4/predict_class/dimension
Predictions_4/predict_class
Predictions_4/one_hot/Const
Predictions_4/one_hot/Const_1
Predictions_4/one_hot/depth
Predictions_4/one_hot/on_value
Predictions_4/one_hot/off_value
Predictions_4/one_hot
Cross_Entropy_4/Rank
Cross_Entropy_4/Shape
Cross_Entropy_4/Rank_1
Cross_Entropy_4/Shape_1
Cross_Entropy_4/Sub/y
Cross_Entropy_4/Sub
Cross_Entropy_4/Slice/begin
Cross_Entropy_4/Slice/size
Cross_Entropy_4/Slice
Cross_Entropy_4/concat/values_0
Cross_Entropy_4/concat/axis
Cross_Entropy_4/concat
Cross_Entropy_4/Reshape
Cross_Entropy_4/Rank_2
Cross_Entropy_4/Shape_2
Cross_Entropy_4/Sub_1/y
Cross_Entropy_4/Sub_1
Cross_Entropy_4/Slice_1/begin
Cross_Entropy_4/Slice_1/size
Cross_Entropy_4/Slice_1
Cross_Entropy_4/concat_1/values_0
Cross_Entropy_4/concat_1/axis
Cross_Entropy_4/concat_1
Cross_Entropy_4/Reshape_1
Cross_Entropy_4/SoftmaxCrossEntropyWithLogits
Cross_Entropy_4/Sub_2/y
Cross_Entropy_4/Sub_2
Cross_Entropy_4/Slice_2/begin
Cross_Entropy_4/Slice_2/size
Cross_Entropy_4/Slice_2
Cross_Entropy_4/Reshape_2
Loss_Function_4/Const
Loss_Function_4/Square
Loss_Function_4/Const_1
Loss_Function_4/Mean
Loss_Function_4/Const_2
Loss_Function_4/Mean_1
Loss_Function_4/Mul
Loss_Function_4/Add
Loss_Function_4/train_loss_list/initial_value
Loss_Function_4/train_loss_list
Loss_Function_4/train_loss_list/Assign
Loss_Function_4/train_loss_list/read
Loss_Function_4/Assign/value
Loss_Function_4/Assign
Loss_Function_4/train_error/tags
Loss_Function_4/train_error
Loss_Function_4/validate_loss_list/initial_value
Loss_Function_4/validate_loss_list
Loss_Function_4/validate_loss_list/Assign
Loss_Function_4/validate_loss_list/read
Loss_Function_4/Assign_1/value
Loss_Function_4/Assign_1
Loss_Function_4/validate_error/tags
Loss_Function_4/validate_error
Optimizer_4/learning_rate_progress/initial_value
Optimizer_4/learning_rate_progress
Optimizer_4/learning_rate_progress/Assign
Optimizer_4/learning_rate_progress/read
Optimizer_4/Assign/value
Optimizer_4/Assign
Optimizer_4/learning_rate/tags
Optimizer_4/learning_rate/values
Optimizer_4/learning_rate
Optimizer_4/gradients/Shape
Optimizer_4/gradients/Const
Optimizer_4/gradients/Fill
Optimizer_4/gradients/Loss_Function_4/Add_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Add_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/BroadcastGradientArgs
Optimizer_4/gradients/Loss_Function_4/Add_grad/Sum
Optimizer_4/gradients/Loss_Function_4/Add_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Add_grad/Sum_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/Reshape_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/group_deps
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/control_dependency
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/control_dependency_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Reshape/shape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Tile
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape_2
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Const
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Prod
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Const_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Prod_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Maximum/y
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Maximum
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/floordiv
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Cast
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/truediv
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/BroadcastGradientArgs
Optimizer_4/gradients/Loss_Function_4/Mul_grad/mul
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Sum
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mul_grad/mul_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Sum_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Reshape_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/group_deps
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/control_dependency
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/control_dependency_1
Optimizer_4/gradients/Cross_Entropy_4/Reshape_2_grad/Shape
Optimizer_4/gradients/Cross_Entropy_4/Reshape_2_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Reshape/shape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Tile/multiples
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Tile
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Const
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Prod
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Const_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Prod_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Maximum/y
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Maximum
Optimizer_4/gradients/Loss_Function_4/Mean_grad/floordiv
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Cast
Optimizer_4/gradients/Loss_Function_4/Mean_grad/truediv
Optimizer_4/gradients/zeros_like
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul/x
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul_1
Optimizer_4/gradients/Cross_Entropy_4/Reshape_grad/Shape
Optimizer_4/gradients/Cross_Entropy_4/Reshape_grad/Reshape
Optimizer_4/gradients/Predictions_4/Softmax_grad/mul
Optimizer_4/gradients/Predictions_4/Softmax_grad/Sum/reduction_indices
Optimizer_4/gradients/Predictions_4/Softmax_grad/Sum
Optimizer_4/gradients/Predictions_4/Softmax_grad/Reshape/shape
Optimizer_4/gradients/Predictions_4/Softmax_grad/Reshape
Optimizer_4/gradients/Predictions_4/Softmax_grad/sub
Optimizer_4/gradients/Predictions_4/Softmax_grad/mul_1
Optimizer_4/gradients/Predictions_4/mlp_layer_1_grad/ReluGrad
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/BiasAddGrad
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/group_deps
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/control_dependency
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/control_dependency_1
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/MatMul
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/MatMul_1
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_4/gradients/AddN
Optimizer_4/GradientDescent/learning_rate
Optimizer_4/GradientDescent/update_Parameters_4/Weights/Weight/ApplyGradientDescent
Optimizer_4/GradientDescent/update_Parameters_4/Bias/Bias/ApplyGradientDescent
Optimizer_4/GradientDescent
Equal_4
Accuracy_4/Cast
Accuracy_4/Const
Accuracy_4/Mean
Accuracy_4/train_accuracy_list/initial_value
Accuracy_4/train_accuracy_list
Accuracy_4/train_accuracy_list/Assign
Accuracy_4/train_accuracy_list/read
Accuracy_4/Assign/value
Accuracy_4/Assign
Accuracy_4/train_accuracy/tags
Accuracy_4/train_accuracy
Accuracy_4/Cast_1
Accuracy_4/Const_1
Accuracy_4/Mean_1
Accuracy_4/validate_accuracy_list/initial_value
Accuracy_4/validate_accuracy_list
Accuracy_4/validate_accuracy_list/Assign
Accuracy_4/validate_accuracy_list/read
Accuracy_4/Assign_1/value
Accuracy_4/Assign_1
Accuracy_4/validate_accuracy/tags
Accuracy_4/validate_accuracy
Accuracy_4/Cast_2
Accuracy_4/Const_2
Accuracy_4/Mean_2
Accuracy_4/test_accuracy_list/initial_value
Accuracy_4/test_accuracy_list
Accuracy_4/test_accuracy_list/Assign
Accuracy_4/test_accuracy_list/read
Accuracy_4/Assign_2/value
Accuracy_4/Assign_2
Accuracy_4/test_accuracy/tags
Accuracy_4/test_accuracy
init_4/NoOp
init_4/NoOp_1
init_4
Merge_4/MergeSummary
save_4/Const
save_4/SaveV2/tensor_names
save_4/SaveV2/shape_and_slices
save_4/SaveV2
save_4/control_dependency
save_4/RestoreV2/tensor_names
save_4/RestoreV2/shape_and_slices
save_4/RestoreV2
save_4/Assign
save_4/RestoreV2_1/tensor_names
save_4/RestoreV2_1/shape_and_slices
save_4/RestoreV2_1
save_4/Assign_1
save_4/RestoreV2_2/tensor_names
save_4/RestoreV2_2/shape_and_slices
save_4/RestoreV2_2
save_4/Assign_2
save_4/RestoreV2_3/tensor_names
save_4/RestoreV2_3/shape_and_slices
save_4/RestoreV2_3
save_4/Assign_3
save_4/RestoreV2_4/tensor_names
save_4/RestoreV2_4/shape_and_slices
save_4/RestoreV2_4
save_4/Assign_4
save_4/RestoreV2_5/tensor_names
save_4/RestoreV2_5/shape_and_slices
save_4/RestoreV2_5
save_4/Assign_5
save_4/RestoreV2_6/tensor_names
save_4/RestoreV2_6/shape_and_slices
save_4/RestoreV2_6
save_4/Assign_6
save_4/RestoreV2_7/tensor_names
save_4/RestoreV2_7/shape_and_slices
save_4/RestoreV2_7
save_4/Assign_7
save_4/RestoreV2_8/tensor_names
save_4/RestoreV2_8/shape_and_slices
save_4/RestoreV2_8
save_4/Assign_8
save_4/RestoreV2_9/tensor_names
save_4/RestoreV2_9/shape_and_slices
save_4/RestoreV2_9
save_4/Assign_9
save_4/RestoreV2_10/tensor_names
save_4/RestoreV2_10/shape_and_slices
save_4/RestoreV2_10
save_4/Assign_10
save_4/RestoreV2_11/tensor_names
save_4/RestoreV2_11/shape_and_slices
save_4/RestoreV2_11
save_4/Assign_11
save_4/RestoreV2_12/tensor_names
save_4/RestoreV2_12/shape_and_slices
save_4/RestoreV2_12
save_4/Assign_12
save_4/RestoreV2_13/tensor_names
save_4/RestoreV2_13/shape_and_slices
save_4/RestoreV2_13
save_4/Assign_13
save_4/RestoreV2_14/tensor_names
save_4/RestoreV2_14/shape_and_slices
save_4/RestoreV2_14
save_4/Assign_14
save_4/RestoreV2_15/tensor_names
save_4/RestoreV2_15/shape_and_slices
save_4/RestoreV2_15
save_4/Assign_15
save_4/RestoreV2_16/tensor_names
save_4/RestoreV2_16/shape_and_slices
save_4/RestoreV2_16
save_4/Assign_16
save_4/RestoreV2_17/tensor_names
save_4/RestoreV2_17/shape_and_slices
save_4/RestoreV2_17
save_4/Assign_17
save_4/RestoreV2_18/tensor_names
save_4/RestoreV2_18/shape_and_slices
save_4/RestoreV2_18
save_4/Assign_18
save_4/RestoreV2_19/tensor_names
save_4/RestoreV2_19/shape_and_slices
save_4/RestoreV2_19
save_4/Assign_19
save_4/RestoreV2_20/tensor_names
save_4/RestoreV2_20/shape_and_slices
save_4/RestoreV2_20
save_4/Assign_20
save_4/RestoreV2_21/tensor_names
save_4/RestoreV2_21/shape_and_slices
save_4/RestoreV2_21
save_4/Assign_21
save_4/RestoreV2_22/tensor_names
save_4/RestoreV2_22/shape_and_slices
save_4/RestoreV2_22
save_4/Assign_22
save_4/RestoreV2_23/tensor_names
save_4/RestoreV2_23/shape_and_slices
save_4/RestoreV2_23
save_4/Assign_23
save_4/RestoreV2_24/tensor_names
save_4/RestoreV2_24/shape_and_slices
save_4/RestoreV2_24
save_4/Assign_24
save_4/RestoreV2_25/tensor_names
save_4/RestoreV2_25/shape_and_slices
save_4/RestoreV2_25
save_4/Assign_25
save_4/RestoreV2_26/tensor_names
save_4/RestoreV2_26/shape_and_slices
save_4/RestoreV2_26
save_4/Assign_26
save_4/RestoreV2_27/tensor_names
save_4/RestoreV2_27/shape_and_slices
save_4/RestoreV2_27
save_4/Assign_27
save_4/RestoreV2_28/tensor_names
save_4/RestoreV2_28/shape_and_slices
save_4/RestoreV2_28
save_4/Assign_28
save_4/RestoreV2_29/tensor_names
save_4/RestoreV2_29/shape_and_slices
save_4/RestoreV2_29
save_4/Assign_29
save_4/RestoreV2_30/tensor_names
save_4/RestoreV2_30/shape_and_slices
save_4/RestoreV2_30
save_4/Assign_30
save_4/RestoreV2_31/tensor_names
save_4/RestoreV2_31/shape_and_slices
save_4/RestoreV2_31
save_4/Assign_31
save_4/RestoreV2_32/tensor_names
save_4/RestoreV2_32/shape_and_slices
save_4/RestoreV2_32
save_4/Assign_32
save_4/RestoreV2_33/tensor_names
save_4/RestoreV2_33/shape_and_slices
save_4/RestoreV2_33
save_4/Assign_33
save_4/RestoreV2_34/tensor_names
save_4/RestoreV2_34/shape_and_slices
save_4/RestoreV2_34
save_4/Assign_34
save_4/RestoreV2_35/tensor_names
save_4/RestoreV2_35/shape_and_slices
save_4/RestoreV2_35
save_4/Assign_35
save_4/RestoreV2_36/tensor_names
save_4/RestoreV2_36/shape_and_slices
save_4/RestoreV2_36
save_4/Assign_36
save_4/RestoreV2_37/tensor_names
save_4/RestoreV2_37/shape_and_slices
save_4/RestoreV2_37
save_4/Assign_37
save_4/RestoreV2_38/tensor_names
save_4/RestoreV2_38/shape_and_slices
save_4/RestoreV2_38
save_4/Assign_38
save_4/RestoreV2_39/tensor_names
save_4/RestoreV2_39/shape_and_slices
save_4/RestoreV2_39
save_4/Assign_39
save_4/RestoreV2_40/tensor_names
save_4/RestoreV2_40/shape_and_slices
save_4/RestoreV2_40
save_4/Assign_40
save_4/RestoreV2_41/tensor_names
save_4/RestoreV2_41/shape_and_slices
save_4/RestoreV2_41
save_4/Assign_41
save_4/RestoreV2_42/tensor_names
save_4/RestoreV2_42/shape_and_slices
save_4/RestoreV2_42
save_4/Assign_42
save_4/RestoreV2_43/tensor_names
save_4/RestoreV2_43/shape_and_slices
save_4/RestoreV2_43
save_4/Assign_43
save_4/RestoreV2_44/tensor_names
save_4/RestoreV2_44/shape_and_slices
save_4/RestoreV2_44
save_4/Assign_44
save_4/restore_all/NoOp
save_4/restore_all/NoOp_1
save_4/restore_all
last_successful_epoch_5/initial_value
last_successful_epoch_5
last_successful_epoch_5/Assign
last_successful_epoch_5/read
add_5/y
add_5
assign_updated_epoch_5
Inputs_5/Data/X_input
Inputs_5/Train_Labels/y_true_one_hot_label
Inputs_5/Train_Labels/y_true_class
Parameters_5/Weights/zeros
Parameters_5/Weights/Weight
Parameters_5/Weights/Weight/Assign
Parameters_5/Weights/Weight/read
Parameters_5/Bias/ones
Parameters_5/Bias/Bias
Parameters_5/Bias/Bias/Assign
Parameters_5/Bias/Bias/read
Predictions_5/mlp_layer/MatMul
Predictions_5/mlp_layer
Predictions_5/mlp_layer_1
Predictions_5/Softmax
Predictions_5/predict_class/dimension
Predictions_5/predict_class
Predictions_5/one_hot/Const
Predictions_5/one_hot/Const_1
Predictions_5/one_hot/depth
Predictions_5/one_hot/on_value
Predictions_5/one_hot/off_value
Predictions_5/one_hot
Cross_Entropy_5/Rank
Cross_Entropy_5/Shape
Cross_Entropy_5/Rank_1
Cross_Entropy_5/Shape_1
Cross_Entropy_5/Sub/y
Cross_Entropy_5/Sub
Cross_Entropy_5/Slice/begin
Cross_Entropy_5/Slice/size
Cross_Entropy_5/Slice
Cross_Entropy_5/concat/values_0
Cross_Entropy_5/concat/axis
Cross_Entropy_5/concat
Cross_Entropy_5/Reshape
Cross_Entropy_5/Rank_2
Cross_Entropy_5/Shape_2
Cross_Entropy_5/Sub_1/y
Cross_Entropy_5/Sub_1
Cross_Entropy_5/Slice_1/begin
Cross_Entropy_5/Slice_1/size
Cross_Entropy_5/Slice_1
Cross_Entropy_5/concat_1/values_0
Cross_Entropy_5/concat_1/axis
Cross_Entropy_5/concat_1
Cross_Entropy_5/Reshape_1
Cross_Entropy_5/SoftmaxCrossEntropyWithLogits
Cross_Entropy_5/Sub_2/y
Cross_Entropy_5/Sub_2
Cross_Entropy_5/Slice_2/begin
Cross_Entropy_5/Slice_2/size
Cross_Entropy_5/Slice_2
Cross_Entropy_5/Reshape_2
Loss_Function_5/Const
Loss_Function_5/Square
Loss_Function_5/Const_1
Loss_Function_5/Mean
Loss_Function_5/Const_2
Loss_Function_5/Mean_1
Loss_Function_5/Mul
Loss_Function_5/Add
Loss_Function_5/train_loss_list/initial_value
Loss_Function_5/train_loss_list
Loss_Function_5/train_loss_list/Assign
Loss_Function_5/train_loss_list/read
Loss_Function_5/Assign/value
Loss_Function_5/Assign
Loss_Function_5/train_error/tags
Loss_Function_5/train_error
Loss_Function_5/validate_loss_list/initial_value
Loss_Function_5/validate_loss_list
Loss_Function_5/validate_loss_list/Assign
Loss_Function_5/validate_loss_list/read
Loss_Function_5/Assign_1/value
Loss_Function_5/Assign_1
Loss_Function_5/validate_error/tags
Loss_Function_5/validate_error
Optimizer_5/learning_rate_progress/initial_value
Optimizer_5/learning_rate_progress
Optimizer_5/learning_rate_progress/Assign
Optimizer_5/learning_rate_progress/read
Optimizer_5/Assign/value
Optimizer_5/Assign
Optimizer_5/learning_rate/tags
Optimizer_5/learning_rate/values
Optimizer_5/learning_rate
Optimizer_5/gradients/Shape
Optimizer_5/gradients/Const
Optimizer_5/gradients/Fill
Optimizer_5/gradients/Loss_Function_5/Add_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Add_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/BroadcastGradientArgs
Optimizer_5/gradients/Loss_Function_5/Add_grad/Sum
Optimizer_5/gradients/Loss_Function_5/Add_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Add_grad/Sum_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/Reshape_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/group_deps
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/control_dependency
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/control_dependency_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Reshape/shape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Tile
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape_2
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Const
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Prod
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Const_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Prod_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Maximum/y
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Maximum
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/floordiv
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Cast
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/truediv
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/BroadcastGradientArgs
Optimizer_5/gradients/Loss_Function_5/Mul_grad/mul
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Sum
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mul_grad/mul_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Sum_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Reshape_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/group_deps
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/control_dependency
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/control_dependency_1
Optimizer_5/gradients/Cross_Entropy_5/Reshape_2_grad/Shape
Optimizer_5/gradients/Cross_Entropy_5/Reshape_2_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Reshape/shape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Tile/multiples
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Tile
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Const
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Prod
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Const_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Prod_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Maximum/y
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Maximum
Optimizer_5/gradients/Loss_Function_5/Mean_grad/floordiv
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Cast
Optimizer_5/gradients/Loss_Function_5/Mean_grad/truediv
Optimizer_5/gradients/zeros_like
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul/x
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul_1
Optimizer_5/gradients/Cross_Entropy_5/Reshape_grad/Shape
Optimizer_5/gradients/Cross_Entropy_5/Reshape_grad/Reshape
Optimizer_5/gradients/Predictions_5/Softmax_grad/mul
Optimizer_5/gradients/Predictions_5/Softmax_grad/Sum/reduction_indices
Optimizer_5/gradients/Predictions_5/Softmax_grad/Sum
Optimizer_5/gradients/Predictions_5/Softmax_grad/Reshape/shape
Optimizer_5/gradients/Predictions_5/Softmax_grad/Reshape
Optimizer_5/gradients/Predictions_5/Softmax_grad/sub
Optimizer_5/gradients/Predictions_5/Softmax_grad/mul_1
Optimizer_5/gradients/Predictions_5/mlp_layer_1_grad/ReluGrad
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/BiasAddGrad
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/group_deps
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/control_dependency
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/control_dependency_1
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/MatMul
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/MatMul_1
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_5/gradients/AddN
Optimizer_5/GradientDescent/learning_rate
Optimizer_5/GradientDescent/update_Parameters_5/Weights/Weight/ApplyGradientDescent
Optimizer_5/GradientDescent/update_Parameters_5/Bias/Bias/ApplyGradientDescent
Optimizer_5/GradientDescent
Equal_5
Accuracy_5/Cast
Accuracy_5/Const
Accuracy_5/Mean
Accuracy_5/train_accuracy_list/initial_value
Accuracy_5/train_accuracy_list
Accuracy_5/train_accuracy_list/Assign
Accuracy_5/train_accuracy_list/read
Accuracy_5/Assign/value
Accuracy_5/Assign
Accuracy_5/train_accuracy/tags
Accuracy_5/train_accuracy
Accuracy_5/Cast_1
Accuracy_5/Const_1
Accuracy_5/Mean_1
Accuracy_5/validate_accuracy_list/initial_value
Accuracy_5/validate_accuracy_list
Accuracy_5/validate_accuracy_list/Assign
Accuracy_5/validate_accuracy_list/read
Accuracy_5/Assign_1/value
Accuracy_5/Assign_1
Accuracy_5/validate_accuracy/tags
Accuracy_5/validate_accuracy
Accuracy_5/Cast_2
Accuracy_5/Const_2
Accuracy_5/Mean_2
Accuracy_5/test_accuracy_list/initial_value
Accuracy_5/test_accuracy_list
Accuracy_5/test_accuracy_list/Assign
Accuracy_5/test_accuracy_list/read
Accuracy_5/Assign_2/value
Accuracy_5/Assign_2
Accuracy_5/test_accuracy/tags
Accuracy_5/test_accuracy
init_5/NoOp
init_5/NoOp_1
init_5
Merge_5/MergeSummary
save_5/Const
save_5/SaveV2/tensor_names
save_5/SaveV2/shape_and_slices
save_5/SaveV2
save_5/control_dependency
save_5/RestoreV2/tensor_names
save_5/RestoreV2/shape_and_slices
save_5/RestoreV2
save_5/Assign
save_5/RestoreV2_1/tensor_names
save_5/RestoreV2_1/shape_and_slices
save_5/RestoreV2_1
save_5/Assign_1
save_5/RestoreV2_2/tensor_names
save_5/RestoreV2_2/shape_and_slices
save_5/RestoreV2_2
save_5/Assign_2
save_5/RestoreV2_3/tensor_names
save_5/RestoreV2_3/shape_and_slices
save_5/RestoreV2_3
save_5/Assign_3
save_5/RestoreV2_4/tensor_names
save_5/RestoreV2_4/shape_and_slices
save_5/RestoreV2_4
save_5/Assign_4
save_5/RestoreV2_5/tensor_names
save_5/RestoreV2_5/shape_and_slices
save_5/RestoreV2_5
save_5/Assign_5
save_5/RestoreV2_6/tensor_names
save_5/RestoreV2_6/shape_and_slices
save_5/RestoreV2_6
save_5/Assign_6
save_5/RestoreV2_7/tensor_names
save_5/RestoreV2_7/shape_and_slices
save_5/RestoreV2_7
save_5/Assign_7
save_5/RestoreV2_8/tensor_names
save_5/RestoreV2_8/shape_and_slices
save_5/RestoreV2_8
save_5/Assign_8
save_5/RestoreV2_9/tensor_names
save_5/RestoreV2_9/shape_and_slices
save_5/RestoreV2_9
save_5/Assign_9
save_5/RestoreV2_10/tensor_names
save_5/RestoreV2_10/shape_and_slices
save_5/RestoreV2_10
save_5/Assign_10
save_5/RestoreV2_11/tensor_names
save_5/RestoreV2_11/shape_and_slices
save_5/RestoreV2_11
save_5/Assign_11
save_5/RestoreV2_12/tensor_names
save_5/RestoreV2_12/shape_and_slices
save_5/RestoreV2_12
save_5/Assign_12
save_5/RestoreV2_13/tensor_names
save_5/RestoreV2_13/shape_and_slices
save_5/RestoreV2_13
save_5/Assign_13
save_5/RestoreV2_14/tensor_names
save_5/RestoreV2_14/shape_and_slices
save_5/RestoreV2_14
save_5/Assign_14
save_5/RestoreV2_15/tensor_names
save_5/RestoreV2_15/shape_and_slices
save_5/RestoreV2_15
save_5/Assign_15
save_5/RestoreV2_16/tensor_names
save_5/RestoreV2_16/shape_and_slices
save_5/RestoreV2_16
save_5/Assign_16
save_5/RestoreV2_17/tensor_names
save_5/RestoreV2_17/shape_and_slices
save_5/RestoreV2_17
save_5/Assign_17
save_5/RestoreV2_18/tensor_names
save_5/RestoreV2_18/shape_and_slices
save_5/RestoreV2_18
save_5/Assign_18
save_5/RestoreV2_19/tensor_names
save_5/RestoreV2_19/shape_and_slices
save_5/RestoreV2_19
save_5/Assign_19
save_5/RestoreV2_20/tensor_names
save_5/RestoreV2_20/shape_and_slices
save_5/RestoreV2_20
save_5/Assign_20
save_5/RestoreV2_21/tensor_names
save_5/RestoreV2_21/shape_and_slices
save_5/RestoreV2_21
save_5/Assign_21
save_5/RestoreV2_22/tensor_names
save_5/RestoreV2_22/shape_and_slices
save_5/RestoreV2_22
save_5/Assign_22
save_5/RestoreV2_23/tensor_names
save_5/RestoreV2_23/shape_and_slices
save_5/RestoreV2_23
save_5/Assign_23
save_5/RestoreV2_24/tensor_names
save_5/RestoreV2_24/shape_and_slices
save_5/RestoreV2_24
save_5/Assign_24
save_5/RestoreV2_25/tensor_names
save_5/RestoreV2_25/shape_and_slices
save_5/RestoreV2_25
save_5/Assign_25
save_5/RestoreV2_26/tensor_names
save_5/RestoreV2_26/shape_and_slices
save_5/RestoreV2_26
save_5/Assign_26
save_5/RestoreV2_27/tensor_names
save_5/RestoreV2_27/shape_and_slices
save_5/RestoreV2_27
save_5/Assign_27
save_5/RestoreV2_28/tensor_names
save_5/RestoreV2_28/shape_and_slices
save_5/RestoreV2_28
save_5/Assign_28
save_5/RestoreV2_29/tensor_names
save_5/RestoreV2_29/shape_and_slices
save_5/RestoreV2_29
save_5/Assign_29
save_5/RestoreV2_30/tensor_names
save_5/RestoreV2_30/shape_and_slices
save_5/RestoreV2_30
save_5/Assign_30
save_5/RestoreV2_31/tensor_names
save_5/RestoreV2_31/shape_and_slices
save_5/RestoreV2_31
save_5/Assign_31
save_5/RestoreV2_32/tensor_names
save_5/RestoreV2_32/shape_and_slices
save_5/RestoreV2_32
save_5/Assign_32
save_5/RestoreV2_33/tensor_names
save_5/RestoreV2_33/shape_and_slices
save_5/RestoreV2_33
save_5/Assign_33
save_5/RestoreV2_34/tensor_names
save_5/RestoreV2_34/shape_and_slices
save_5/RestoreV2_34
save_5/Assign_34
save_5/RestoreV2_35/tensor_names
save_5/RestoreV2_35/shape_and_slices
save_5/RestoreV2_35
save_5/Assign_35
save_5/RestoreV2_36/tensor_names
save_5/RestoreV2_36/shape_and_slices
save_5/RestoreV2_36
save_5/Assign_36
save_5/RestoreV2_37/tensor_names
save_5/RestoreV2_37/shape_and_slices
save_5/RestoreV2_37
save_5/Assign_37
save_5/RestoreV2_38/tensor_names
save_5/RestoreV2_38/shape_and_slices
save_5/RestoreV2_38
save_5/Assign_38
save_5/RestoreV2_39/tensor_names
save_5/RestoreV2_39/shape_and_slices
save_5/RestoreV2_39
save_5/Assign_39
save_5/RestoreV2_40/tensor_names
save_5/RestoreV2_40/shape_and_slices
save_5/RestoreV2_40
save_5/Assign_40
save_5/RestoreV2_41/tensor_names
save_5/RestoreV2_41/shape_and_slices
save_5/RestoreV2_41
save_5/Assign_41
save_5/RestoreV2_42/tensor_names
save_5/RestoreV2_42/shape_and_slices
save_5/RestoreV2_42
save_5/Assign_42
save_5/RestoreV2_43/tensor_names
save_5/RestoreV2_43/shape_and_slices
save_5/RestoreV2_43
save_5/Assign_43
save_5/RestoreV2_44/tensor_names
save_5/RestoreV2_44/shape_and_slices
save_5/RestoreV2_44
save_5/Assign_44
save_5/RestoreV2_45/tensor_names
save_5/RestoreV2_45/shape_and_slices
save_5/RestoreV2_45
save_5/Assign_45
save_5/RestoreV2_46/tensor_names
save_5/RestoreV2_46/shape_and_slices
save_5/RestoreV2_46
save_5/Assign_46
save_5/RestoreV2_47/tensor_names
save_5/RestoreV2_47/shape_and_slices
save_5/RestoreV2_47
save_5/Assign_47
save_5/RestoreV2_48/tensor_names
save_5/RestoreV2_48/shape_and_slices
save_5/RestoreV2_48
save_5/Assign_48
save_5/RestoreV2_49/tensor_names
save_5/RestoreV2_49/shape_and_slices
save_5/RestoreV2_49
save_5/Assign_49
save_5/RestoreV2_50/tensor_names
save_5/RestoreV2_50/shape_and_slices
save_5/RestoreV2_50
save_5/Assign_50
save_5/RestoreV2_51/tensor_names
save_5/RestoreV2_51/shape_and_slices
save_5/RestoreV2_51
save_5/Assign_51
save_5/RestoreV2_52/tensor_names
save_5/RestoreV2_52/shape_and_slices
save_5/RestoreV2_52
save_5/Assign_52
save_5/RestoreV2_53/tensor_names
save_5/RestoreV2_53/shape_and_slices
save_5/RestoreV2_53
save_5/Assign_53
save_5/restore_all/NoOp
save_5/restore_all/NoOp_1
save_5/restore_all
Output Logits  : Tensor("Predictions_4/Softmax:0", shape=(?, 10), dtype=float32, device=/device:GPU:0)
Entropy        : Tensor("Cross_Entropy_4/Reshape_2:0", shape=(?,), dtype=float32, device=/device:GPU:0)
Predictions    : Tensor("Equal_4:0", shape=(?,), dtype=bool, device=/device:GPU:0) 
Optimizer      : name: "Optimizer_4/GradientDescent"
op: "NoOp"
input: "^Optimizer_4/GradientDescent/update_Parameters_4/Weights/Weight/ApplyGradientDescent"
input: "^Optimizer_4/GradientDescent/update_Parameters_4/Bias/Bias/ApplyGradientDescent"
device: "/device:GPU:0"
 
>> Model info
Train loss     : Tensor("Loss_Function_4/Add:0", shape=(), dtype=float32, device=/device:GPU:0)
Train accuracy : Tensor("Accuracy_4/Mean:0", shape=(), dtype=float32, device=/device:GPU:0)
Val. loss      : Tensor("Loss_Function_4/Add:0", shape=(), dtype=float32, device=/device:GPU:0)
Val. accuracy  : Tensor("Accuracy_4/Mean_1:0", shape=(), dtype=float32, device=/device:GPU:0)
>> Summary params
Train loss     : Tensor("Loss_Function_4/train_error:0", shape=(), dtype=string, device=/device:GPU:0)
Train accuracy : Tensor("Accuracy_4/train_accuracy:0", shape=(), dtype=string, device=/device:GPU:0)
Val. loss      : Tensor("Loss_Function_4/validate_error:0", shape=(), dtype=string, device=/device:GPU:0)
Val. accuracy  : Tensor("Accuracy_4/validate_accuracy:0", shape=(), dtype=string, device=/device:GPU:0)
Parameters for Linear classifier
>> Input Parameters
Input                  : Tensor("Inputs_5/Data/X_input:0", shape=(?, 3072), dtype=float32, device=/device:GPU:0) 
True one hot labels    : Tensor("Inputs_5/Train_Labels/y_true_one_hot_label:0", shape=(?, 10), dtype=float32, device=/device:GPU:0) 
True class             : Tensor("Inputs_5/Train_Labels/y_true_class:0", shape=(?,), dtype=int64, device=/device:GPU:0) 
Predict one hot labels : Tensor("Predictions_5/one_hot:0", shape=(?, 10), dtype=float32, device=/device:GPU:0) 
Predict class          : Tensor("Predictions_5/predict_class:0", shape=(?,), dtype=int64, device=/device:GPU:0) 
>> Model params
last_successful_epoch/initial_value
last_successful_epoch
last_successful_epoch/Assign
last_successful_epoch/read
add/y
add
assign_updated_epoch
Inputs/Data/X_input
Inputs/Train_Labels/y_true_one_hot_label
Inputs/Train_Labels/y_true_class
Parameters/Weights/zeros
Parameters/Weights/Weight
Parameters/Weights/Weight/Assign
Parameters/Weights/Weight/read
Parameters/Bias/ones
Parameters/Bias/Bias
Parameters/Bias/Bias/Assign
Parameters/Bias/Bias/read
Predictions/mlp_layer/MatMul
Predictions/mlp_layer
Predictions/mlp_layer_1
Predictions/Softmax
Predictions/predict_class/dimension
Predictions/predict_class
Predictions/one_hot/Const
Predictions/one_hot/Const_1
Predictions/one_hot/depth
Predictions/one_hot/on_value
Predictions/one_hot/off_value
Predictions/one_hot
Cross_Entropy/Rank
Cross_Entropy/Shape
Cross_Entropy/Rank_1
Cross_Entropy/Shape_1
Cross_Entropy/Sub/y
Cross_Entropy/Sub
Cross_Entropy/Slice/begin
Cross_Entropy/Slice/size
Cross_Entropy/Slice
Cross_Entropy/concat/values_0
Cross_Entropy/concat/axis
Cross_Entropy/concat
Cross_Entropy/Reshape
Cross_Entropy/Rank_2
Cross_Entropy/Shape_2
Cross_Entropy/Sub_1/y
Cross_Entropy/Sub_1
Cross_Entropy/Slice_1/begin
Cross_Entropy/Slice_1/size
Cross_Entropy/Slice_1
Cross_Entropy/concat_1/values_0
Cross_Entropy/concat_1/axis
Cross_Entropy/concat_1
Cross_Entropy/Reshape_1
Cross_Entropy/SoftmaxCrossEntropyWithLogits
Cross_Entropy/Sub_2/y
Cross_Entropy/Sub_2
Cross_Entropy/Slice_2/begin
Cross_Entropy/Slice_2/size
Cross_Entropy/Slice_2
Cross_Entropy/Reshape_2
Loss_Function/Const
Loss_Function/Square
Loss_Function/Const_1
Loss_Function/Mean
Loss_Function/Const_2
Loss_Function/Mean_1
Loss_Function/Mul
Loss_Function/Add
Loss_Function/train_loss_list/initial_value
Loss_Function/train_loss_list
Loss_Function/train_loss_list/Assign
Loss_Function/train_loss_list/read
Loss_Function/Assign/value
Loss_Function/Assign
Loss_Function/train_error/tags
Loss_Function/train_error
Loss_Function/validate_loss_list/initial_value
Loss_Function/validate_loss_list
Loss_Function/validate_loss_list/Assign
Loss_Function/validate_loss_list/read
Loss_Function/Assign_1/value
Loss_Function/Assign_1
Loss_Function/validate_error/tags
Loss_Function/validate_error
Optimizer/learning_rate_progress/initial_value
Optimizer/learning_rate_progress
Optimizer/learning_rate_progress/Assign
Optimizer/learning_rate_progress/read
Optimizer/Assign/value
Optimizer/Assign
Optimizer/learning_rate/tags
Optimizer/learning_rate/values
Optimizer/learning_rate
Optimizer/gradients/Shape
Optimizer/gradients/Const
Optimizer/gradients/Fill
Optimizer/gradients/Loss_Function/Add_grad/Shape
Optimizer/gradients/Loss_Function/Add_grad/Shape_1
Optimizer/gradients/Loss_Function/Add_grad/BroadcastGradientArgs
Optimizer/gradients/Loss_Function/Add_grad/Sum
Optimizer/gradients/Loss_Function/Add_grad/Reshape
Optimizer/gradients/Loss_Function/Add_grad/Sum_1
Optimizer/gradients/Loss_Function/Add_grad/Reshape_1
Optimizer/gradients/Loss_Function/Add_grad/tuple/group_deps
Optimizer/gradients/Loss_Function/Add_grad/tuple/control_dependency
Optimizer/gradients/Loss_Function/Add_grad/tuple/control_dependency_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Reshape/shape
Optimizer/gradients/Loss_Function/Mean_1_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape
Optimizer/gradients/Loss_Function/Mean_1_grad/Tile
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Shape_2
Optimizer/gradients/Loss_Function/Mean_1_grad/Const
Optimizer/gradients/Loss_Function/Mean_1_grad/Prod
Optimizer/gradients/Loss_Function/Mean_1_grad/Const_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Prod_1
Optimizer/gradients/Loss_Function/Mean_1_grad/Maximum/y
Optimizer/gradients/Loss_Function/Mean_1_grad/Maximum
Optimizer/gradients/Loss_Function/Mean_1_grad/floordiv
Optimizer/gradients/Loss_Function/Mean_1_grad/Cast
Optimizer/gradients/Loss_Function/Mean_1_grad/truediv
Optimizer/gradients/Loss_Function/Mul_grad/Shape
Optimizer/gradients/Loss_Function/Mul_grad/Shape_1
Optimizer/gradients/Loss_Function/Mul_grad/BroadcastGradientArgs
Optimizer/gradients/Loss_Function/Mul_grad/mul
Optimizer/gradients/Loss_Function/Mul_grad/Sum
Optimizer/gradients/Loss_Function/Mul_grad/Reshape
Optimizer/gradients/Loss_Function/Mul_grad/mul_1
Optimizer/gradients/Loss_Function/Mul_grad/Sum_1
Optimizer/gradients/Loss_Function/Mul_grad/Reshape_1
Optimizer/gradients/Loss_Function/Mul_grad/tuple/group_deps
Optimizer/gradients/Loss_Function/Mul_grad/tuple/control_dependency
Optimizer/gradients/Loss_Function/Mul_grad/tuple/control_dependency_1
Optimizer/gradients/Cross_Entropy/Reshape_2_grad/Shape
Optimizer/gradients/Cross_Entropy/Reshape_2_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_grad/Reshape/shape
Optimizer/gradients/Loss_Function/Mean_grad/Reshape
Optimizer/gradients/Loss_Function/Mean_grad/Tile/multiples
Optimizer/gradients/Loss_Function/Mean_grad/Tile
Optimizer/gradients/Loss_Function/Mean_grad/Shape
Optimizer/gradients/Loss_Function/Mean_grad/Shape_1
Optimizer/gradients/Loss_Function/Mean_grad/Const
Optimizer/gradients/Loss_Function/Mean_grad/Prod
Optimizer/gradients/Loss_Function/Mean_grad/Const_1
Optimizer/gradients/Loss_Function/Mean_grad/Prod_1
Optimizer/gradients/Loss_Function/Mean_grad/Maximum/y
Optimizer/gradients/Loss_Function/Mean_grad/Maximum
Optimizer/gradients/Loss_Function/Mean_grad/floordiv
Optimizer/gradients/Loss_Function/Mean_grad/Cast
Optimizer/gradients/Loss_Function/Mean_grad/truediv
Optimizer/gradients/zeros_like
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer/gradients/Cross_Entropy/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer/gradients/Loss_Function/Square_grad/mul/x
Optimizer/gradients/Loss_Function/Square_grad/mul
Optimizer/gradients/Loss_Function/Square_grad/mul_1
Optimizer/gradients/Cross_Entropy/Reshape_grad/Shape
Optimizer/gradients/Cross_Entropy/Reshape_grad/Reshape
Optimizer/gradients/Predictions/Softmax_grad/mul
Optimizer/gradients/Predictions/Softmax_grad/Sum/reduction_indices
Optimizer/gradients/Predictions/Softmax_grad/Sum
Optimizer/gradients/Predictions/Softmax_grad/Reshape/shape
Optimizer/gradients/Predictions/Softmax_grad/Reshape
Optimizer/gradients/Predictions/Softmax_grad/sub
Optimizer/gradients/Predictions/Softmax_grad/mul_1
Optimizer/gradients/Predictions/mlp_layer_1_grad/ReluGrad
Optimizer/gradients/Predictions/mlp_layer_grad/BiasAddGrad
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/group_deps
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/control_dependency
Optimizer/gradients/Predictions/mlp_layer_grad/tuple/control_dependency_1
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/MatMul
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/MatMul_1
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer/gradients/Predictions/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer/gradients/AddN
Optimizer/GradientDescent/learning_rate
Optimizer/GradientDescent/update_Parameters/Weights/Weight/ApplyGradientDescent
Optimizer/GradientDescent/update_Parameters/Bias/Bias/ApplyGradientDescent
Optimizer/GradientDescent
Equal
Accuracy/Cast
Accuracy/Const
Accuracy/Mean
Accuracy/train_accuracy_list/initial_value
Accuracy/train_accuracy_list
Accuracy/train_accuracy_list/Assign
Accuracy/train_accuracy_list/read
Accuracy/Assign/value
Accuracy/Assign
Accuracy/train_accuracy/tags
Accuracy/train_accuracy
Accuracy/Cast_1
Accuracy/Const_1
Accuracy/Mean_1
Accuracy/validate_accuracy_list/initial_value
Accuracy/validate_accuracy_list
Accuracy/validate_accuracy_list/Assign
Accuracy/validate_accuracy_list/read
Accuracy/Assign_1/value
Accuracy/Assign_1
Accuracy/validate_accuracy/tags
Accuracy/validate_accuracy
Accuracy/Cast_2
Accuracy/Const_2
Accuracy/Mean_2
Accuracy/test_accuracy_list/initial_value
Accuracy/test_accuracy_list
Accuracy/test_accuracy_list/Assign
Accuracy/test_accuracy_list/read
Accuracy/Assign_2/value
Accuracy/Assign_2
Accuracy/test_accuracy/tags
Accuracy/test_accuracy
init/NoOp
init/NoOp_1
init
Merge/MergeSummary
save/Const
save/SaveV2/tensor_names
save/SaveV2/shape_and_slices
save/SaveV2
save/control_dependency
save/RestoreV2/tensor_names
save/RestoreV2/shape_and_slices
save/RestoreV2
save/Assign
save/RestoreV2_1/tensor_names
save/RestoreV2_1/shape_and_slices
save/RestoreV2_1
save/Assign_1
save/RestoreV2_2/tensor_names
save/RestoreV2_2/shape_and_slices
save/RestoreV2_2
save/Assign_2
save/RestoreV2_3/tensor_names
save/RestoreV2_3/shape_and_slices
save/RestoreV2_3
save/Assign_3
save/RestoreV2_4/tensor_names
save/RestoreV2_4/shape_and_slices
save/RestoreV2_4
save/Assign_4
save/RestoreV2_5/tensor_names
save/RestoreV2_5/shape_and_slices
save/RestoreV2_5
save/Assign_5
save/RestoreV2_6/tensor_names
save/RestoreV2_6/shape_and_slices
save/RestoreV2_6
save/Assign_6
save/RestoreV2_7/tensor_names
save/RestoreV2_7/shape_and_slices
save/RestoreV2_7
save/Assign_7
save/RestoreV2_8/tensor_names
save/RestoreV2_8/shape_and_slices
save/RestoreV2_8
save/Assign_8
save/restore_all/NoOp
save/restore_all/NoOp_1
save/restore_all
last_successful_epoch_1/initial_value
last_successful_epoch_1
last_successful_epoch_1/Assign
last_successful_epoch_1/read
add_1/y
add_1
assign_updated_epoch_1
Inputs_1/Data/X_input
Inputs_1/Train_Labels/y_true_one_hot_label
Inputs_1/Train_Labels/y_true_class
Parameters_1/Weights/zeros
Parameters_1/Weights/Weight
Parameters_1/Weights/Weight/Assign
Parameters_1/Weights/Weight/read
Parameters_1/Bias/ones
Parameters_1/Bias/Bias
Parameters_1/Bias/Bias/Assign
Parameters_1/Bias/Bias/read
Predictions_1/mlp_layer/MatMul
Predictions_1/mlp_layer
Predictions_1/mlp_layer_1
Predictions_1/Softmax
Predictions_1/predict_class/dimension
Predictions_1/predict_class
Predictions_1/one_hot/Const
Predictions_1/one_hot/Const_1
Predictions_1/one_hot/depth
Predictions_1/one_hot/on_value
Predictions_1/one_hot/off_value
Predictions_1/one_hot
Cross_Entropy_1/Rank
Cross_Entropy_1/Shape
Cross_Entropy_1/Rank_1
Cross_Entropy_1/Shape_1
Cross_Entropy_1/Sub/y
Cross_Entropy_1/Sub
Cross_Entropy_1/Slice/begin
Cross_Entropy_1/Slice/size
Cross_Entropy_1/Slice
Cross_Entropy_1/concat/values_0
Cross_Entropy_1/concat/axis
Cross_Entropy_1/concat
Cross_Entropy_1/Reshape
Cross_Entropy_1/Rank_2
Cross_Entropy_1/Shape_2
Cross_Entropy_1/Sub_1/y
Cross_Entropy_1/Sub_1
Cross_Entropy_1/Slice_1/begin
Cross_Entropy_1/Slice_1/size
Cross_Entropy_1/Slice_1
Cross_Entropy_1/concat_1/values_0
Cross_Entropy_1/concat_1/axis
Cross_Entropy_1/concat_1
Cross_Entropy_1/Reshape_1
Cross_Entropy_1/SoftmaxCrossEntropyWithLogits
Cross_Entropy_1/Sub_2/y
Cross_Entropy_1/Sub_2
Cross_Entropy_1/Slice_2/begin
Cross_Entropy_1/Slice_2/size
Cross_Entropy_1/Slice_2
Cross_Entropy_1/Reshape_2
Loss_Function_1/Const
Loss_Function_1/Square
Loss_Function_1/Const_1
Loss_Function_1/Mean
Loss_Function_1/Const_2
Loss_Function_1/Mean_1
Loss_Function_1/Mul
Loss_Function_1/Add
Loss_Function_1/train_loss_list/initial_value
Loss_Function_1/train_loss_list
Loss_Function_1/train_loss_list/Assign
Loss_Function_1/train_loss_list/read
Loss_Function_1/Assign/value
Loss_Function_1/Assign
Loss_Function_1/train_error/tags
Loss_Function_1/train_error
Loss_Function_1/validate_loss_list/initial_value
Loss_Function_1/validate_loss_list
Loss_Function_1/validate_loss_list/Assign
Loss_Function_1/validate_loss_list/read
Loss_Function_1/Assign_1/value
Loss_Function_1/Assign_1
Loss_Function_1/validate_error/tags
Loss_Function_1/validate_error
Optimizer_1/learning_rate_progress/initial_value
Optimizer_1/learning_rate_progress
Optimizer_1/learning_rate_progress/Assign
Optimizer_1/learning_rate_progress/read
Optimizer_1/Assign/value
Optimizer_1/Assign
Optimizer_1/learning_rate/tags
Optimizer_1/learning_rate/values
Optimizer_1/learning_rate
Optimizer_1/gradients/Shape
Optimizer_1/gradients/Const
Optimizer_1/gradients/Fill
Optimizer_1/gradients/Loss_Function_1/Add_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Add_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/BroadcastGradientArgs
Optimizer_1/gradients/Loss_Function_1/Add_grad/Sum
Optimizer_1/gradients/Loss_Function_1/Add_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Add_grad/Sum_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/Reshape_1
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/group_deps
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/control_dependency
Optimizer_1/gradients/Loss_Function_1/Add_grad/tuple/control_dependency_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Reshape/shape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Tile
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Shape_2
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Const
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Prod
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Const_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Prod_1
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Maximum/y
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Maximum
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/floordiv
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/Cast
Optimizer_1/gradients/Loss_Function_1/Mean_1_grad/truediv
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/BroadcastGradientArgs
Optimizer_1/gradients/Loss_Function_1/Mul_grad/mul
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Sum
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mul_grad/mul_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Sum_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/Reshape_1
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/group_deps
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/control_dependency
Optimizer_1/gradients/Loss_Function_1/Mul_grad/tuple/control_dependency_1
Optimizer_1/gradients/Cross_Entropy_1/Reshape_2_grad/Shape
Optimizer_1/gradients/Cross_Entropy_1/Reshape_2_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Reshape/shape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Reshape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Tile/multiples
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Tile
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Shape
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Shape_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Const
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Prod
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Const_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Prod_1
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Maximum/y
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Maximum
Optimizer_1/gradients/Loss_Function_1/Mean_grad/floordiv
Optimizer_1/gradients/Loss_Function_1/Mean_grad/Cast
Optimizer_1/gradients/Loss_Function_1/Mean_grad/truediv
Optimizer_1/gradients/zeros_like
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_1/gradients/Cross_Entropy_1/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul/x
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul
Optimizer_1/gradients/Loss_Function_1/Square_grad/mul_1
Optimizer_1/gradients/Cross_Entropy_1/Reshape_grad/Shape
Optimizer_1/gradients/Cross_Entropy_1/Reshape_grad/Reshape
Optimizer_1/gradients/Predictions_1/Softmax_grad/mul
Optimizer_1/gradients/Predictions_1/Softmax_grad/Sum/reduction_indices
Optimizer_1/gradients/Predictions_1/Softmax_grad/Sum
Optimizer_1/gradients/Predictions_1/Softmax_grad/Reshape/shape
Optimizer_1/gradients/Predictions_1/Softmax_grad/Reshape
Optimizer_1/gradients/Predictions_1/Softmax_grad/sub
Optimizer_1/gradients/Predictions_1/Softmax_grad/mul_1
Optimizer_1/gradients/Predictions_1/mlp_layer_1_grad/ReluGrad
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/BiasAddGrad
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/group_deps
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/control_dependency
Optimizer_1/gradients/Predictions_1/mlp_layer_grad/tuple/control_dependency_1
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/MatMul
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/MatMul_1
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_1/gradients/Predictions_1/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_1/gradients/AddN
Optimizer_1/GradientDescent/learning_rate
Optimizer_1/GradientDescent/update_Parameters_1/Weights/Weight/ApplyGradientDescent
Optimizer_1/GradientDescent/update_Parameters_1/Bias/Bias/ApplyGradientDescent
Optimizer_1/GradientDescent
Equal_1
Accuracy_1/Cast
Accuracy_1/Const
Accuracy_1/Mean
Accuracy_1/train_accuracy_list/initial_value
Accuracy_1/train_accuracy_list
Accuracy_1/train_accuracy_list/Assign
Accuracy_1/train_accuracy_list/read
Accuracy_1/Assign/value
Accuracy_1/Assign
Accuracy_1/train_accuracy/tags
Accuracy_1/train_accuracy
Accuracy_1/Cast_1
Accuracy_1/Const_1
Accuracy_1/Mean_1
Accuracy_1/validate_accuracy_list/initial_value
Accuracy_1/validate_accuracy_list
Accuracy_1/validate_accuracy_list/Assign
Accuracy_1/validate_accuracy_list/read
Accuracy_1/Assign_1/value
Accuracy_1/Assign_1
Accuracy_1/validate_accuracy/tags
Accuracy_1/validate_accuracy
Accuracy_1/Cast_2
Accuracy_1/Const_2
Accuracy_1/Mean_2
Accuracy_1/test_accuracy_list/initial_value
Accuracy_1/test_accuracy_list
Accuracy_1/test_accuracy_list/Assign
Accuracy_1/test_accuracy_list/read
Accuracy_1/Assign_2/value
Accuracy_1/Assign_2
Accuracy_1/test_accuracy/tags
Accuracy_1/test_accuracy
init_1/NoOp
init_1/NoOp_1
init_1
Merge_1/MergeSummary
save_1/Const
save_1/SaveV2/tensor_names
save_1/SaveV2/shape_and_slices
save_1/SaveV2
save_1/control_dependency
save_1/RestoreV2/tensor_names
save_1/RestoreV2/shape_and_slices
save_1/RestoreV2
save_1/Assign
save_1/RestoreV2_1/tensor_names
save_1/RestoreV2_1/shape_and_slices
save_1/RestoreV2_1
save_1/Assign_1
save_1/RestoreV2_2/tensor_names
save_1/RestoreV2_2/shape_and_slices
save_1/RestoreV2_2
save_1/Assign_2
save_1/RestoreV2_3/tensor_names
save_1/RestoreV2_3/shape_and_slices
save_1/RestoreV2_3
save_1/Assign_3
save_1/RestoreV2_4/tensor_names
save_1/RestoreV2_4/shape_and_slices
save_1/RestoreV2_4
save_1/Assign_4
save_1/RestoreV2_5/tensor_names
save_1/RestoreV2_5/shape_and_slices
save_1/RestoreV2_5
save_1/Assign_5
save_1/RestoreV2_6/tensor_names
save_1/RestoreV2_6/shape_and_slices
save_1/RestoreV2_6
save_1/Assign_6
save_1/RestoreV2_7/tensor_names
save_1/RestoreV2_7/shape_and_slices
save_1/RestoreV2_7
save_1/Assign_7
save_1/RestoreV2_8/tensor_names
save_1/RestoreV2_8/shape_and_slices
save_1/RestoreV2_8
save_1/Assign_8
save_1/RestoreV2_9/tensor_names
save_1/RestoreV2_9/shape_and_slices
save_1/RestoreV2_9
save_1/Assign_9
save_1/RestoreV2_10/tensor_names
save_1/RestoreV2_10/shape_and_slices
save_1/RestoreV2_10
save_1/Assign_10
save_1/RestoreV2_11/tensor_names
save_1/RestoreV2_11/shape_and_slices
save_1/RestoreV2_11
save_1/Assign_11
save_1/RestoreV2_12/tensor_names
save_1/RestoreV2_12/shape_and_slices
save_1/RestoreV2_12
save_1/Assign_12
save_1/RestoreV2_13/tensor_names
save_1/RestoreV2_13/shape_and_slices
save_1/RestoreV2_13
save_1/Assign_13
save_1/RestoreV2_14/tensor_names
save_1/RestoreV2_14/shape_and_slices
save_1/RestoreV2_14
save_1/Assign_14
save_1/RestoreV2_15/tensor_names
save_1/RestoreV2_15/shape_and_slices
save_1/RestoreV2_15
save_1/Assign_15
save_1/RestoreV2_16/tensor_names
save_1/RestoreV2_16/shape_and_slices
save_1/RestoreV2_16
save_1/Assign_16
save_1/RestoreV2_17/tensor_names
save_1/RestoreV2_17/shape_and_slices
save_1/RestoreV2_17
save_1/Assign_17
save_1/restore_all/NoOp
save_1/restore_all/NoOp_1
save_1/restore_all
last_successful_epoch_2/initial_value
last_successful_epoch_2
last_successful_epoch_2/Assign
last_successful_epoch_2/read
add_2/y
add_2
assign_updated_epoch_2
Inputs_2/Data/X_input
Inputs_2/Train_Labels/y_true_one_hot_label
Inputs_2/Train_Labels/y_true_class
Parameters_2/Weights/zeros
Parameters_2/Weights/Weight
Parameters_2/Weights/Weight/Assign
Parameters_2/Weights/Weight/read
Parameters_2/Bias/ones
Parameters_2/Bias/Bias
Parameters_2/Bias/Bias/Assign
Parameters_2/Bias/Bias/read
Predictions_2/mlp_layer/MatMul
Predictions_2/mlp_layer
Predictions_2/mlp_layer_1
Predictions_2/Softmax
Predictions_2/predict_class/dimension
Predictions_2/predict_class
Predictions_2/one_hot/Const
Predictions_2/one_hot/Const_1
Predictions_2/one_hot/depth
Predictions_2/one_hot/on_value
Predictions_2/one_hot/off_value
Predictions_2/one_hot
Cross_Entropy_2/Rank
Cross_Entropy_2/Shape
Cross_Entropy_2/Rank_1
Cross_Entropy_2/Shape_1
Cross_Entropy_2/Sub/y
Cross_Entropy_2/Sub
Cross_Entropy_2/Slice/begin
Cross_Entropy_2/Slice/size
Cross_Entropy_2/Slice
Cross_Entropy_2/concat/values_0
Cross_Entropy_2/concat/axis
Cross_Entropy_2/concat
Cross_Entropy_2/Reshape
Cross_Entropy_2/Rank_2
Cross_Entropy_2/Shape_2
Cross_Entropy_2/Sub_1/y
Cross_Entropy_2/Sub_1
Cross_Entropy_2/Slice_1/begin
Cross_Entropy_2/Slice_1/size
Cross_Entropy_2/Slice_1
Cross_Entropy_2/concat_1/values_0
Cross_Entropy_2/concat_1/axis
Cross_Entropy_2/concat_1
Cross_Entropy_2/Reshape_1
Cross_Entropy_2/SoftmaxCrossEntropyWithLogits
Cross_Entropy_2/Sub_2/y
Cross_Entropy_2/Sub_2
Cross_Entropy_2/Slice_2/begin
Cross_Entropy_2/Slice_2/size
Cross_Entropy_2/Slice_2
Cross_Entropy_2/Reshape_2
Loss_Function_2/Const
Loss_Function_2/Square
Loss_Function_2/Const_1
Loss_Function_2/Mean
Loss_Function_2/Const_2
Loss_Function_2/Mean_1
Loss_Function_2/Mul
Loss_Function_2/Add
Loss_Function_2/train_loss_list/initial_value
Loss_Function_2/train_loss_list
Loss_Function_2/train_loss_list/Assign
Loss_Function_2/train_loss_list/read
Loss_Function_2/Assign/value
Loss_Function_2/Assign
Loss_Function_2/train_error/tags
Loss_Function_2/train_error
Loss_Function_2/validate_loss_list/initial_value
Loss_Function_2/validate_loss_list
Loss_Function_2/validate_loss_list/Assign
Loss_Function_2/validate_loss_list/read
Loss_Function_2/Assign_1/value
Loss_Function_2/Assign_1
Loss_Function_2/validate_error/tags
Loss_Function_2/validate_error
Optimizer_2/learning_rate_progress/initial_value
Optimizer_2/learning_rate_progress
Optimizer_2/learning_rate_progress/Assign
Optimizer_2/learning_rate_progress/read
Optimizer_2/Assign/value
Optimizer_2/Assign
Optimizer_2/learning_rate/tags
Optimizer_2/learning_rate/values
Optimizer_2/learning_rate
Optimizer_2/gradients/Shape
Optimizer_2/gradients/Const
Optimizer_2/gradients/Fill
Optimizer_2/gradients/Loss_Function_2/Add_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Add_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/BroadcastGradientArgs
Optimizer_2/gradients/Loss_Function_2/Add_grad/Sum
Optimizer_2/gradients/Loss_Function_2/Add_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Add_grad/Sum_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/Reshape_1
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/group_deps
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/control_dependency
Optimizer_2/gradients/Loss_Function_2/Add_grad/tuple/control_dependency_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Reshape/shape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Tile
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Shape_2
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Const
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Prod
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Const_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Prod_1
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Maximum/y
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Maximum
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/floordiv
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/Cast
Optimizer_2/gradients/Loss_Function_2/Mean_1_grad/truediv
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/BroadcastGradientArgs
Optimizer_2/gradients/Loss_Function_2/Mul_grad/mul
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Sum
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mul_grad/mul_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Sum_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/Reshape_1
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/group_deps
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/control_dependency
Optimizer_2/gradients/Loss_Function_2/Mul_grad/tuple/control_dependency_1
Optimizer_2/gradients/Cross_Entropy_2/Reshape_2_grad/Shape
Optimizer_2/gradients/Cross_Entropy_2/Reshape_2_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Reshape/shape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Reshape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Tile/multiples
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Tile
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Shape
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Shape_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Const
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Prod
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Const_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Prod_1
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Maximum/y
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Maximum
Optimizer_2/gradients/Loss_Function_2/Mean_grad/floordiv
Optimizer_2/gradients/Loss_Function_2/Mean_grad/Cast
Optimizer_2/gradients/Loss_Function_2/Mean_grad/truediv
Optimizer_2/gradients/zeros_like
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_2/gradients/Cross_Entropy_2/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul/x
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul
Optimizer_2/gradients/Loss_Function_2/Square_grad/mul_1
Optimizer_2/gradients/Cross_Entropy_2/Reshape_grad/Shape
Optimizer_2/gradients/Cross_Entropy_2/Reshape_grad/Reshape
Optimizer_2/gradients/Predictions_2/Softmax_grad/mul
Optimizer_2/gradients/Predictions_2/Softmax_grad/Sum/reduction_indices
Optimizer_2/gradients/Predictions_2/Softmax_grad/Sum
Optimizer_2/gradients/Predictions_2/Softmax_grad/Reshape/shape
Optimizer_2/gradients/Predictions_2/Softmax_grad/Reshape
Optimizer_2/gradients/Predictions_2/Softmax_grad/sub
Optimizer_2/gradients/Predictions_2/Softmax_grad/mul_1
Optimizer_2/gradients/Predictions_2/mlp_layer_1_grad/ReluGrad
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/BiasAddGrad
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/group_deps
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/control_dependency
Optimizer_2/gradients/Predictions_2/mlp_layer_grad/tuple/control_dependency_1
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/MatMul
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/MatMul_1
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_2/gradients/Predictions_2/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_2/gradients/AddN
Optimizer_2/GradientDescent/learning_rate
Optimizer_2/GradientDescent/update_Parameters_2/Weights/Weight/ApplyGradientDescent
Optimizer_2/GradientDescent/update_Parameters_2/Bias/Bias/ApplyGradientDescent
Optimizer_2/GradientDescent
Equal_2
Accuracy_2/Cast
Accuracy_2/Const
Accuracy_2/Mean
Accuracy_2/train_accuracy_list/initial_value
Accuracy_2/train_accuracy_list
Accuracy_2/train_accuracy_list/Assign
Accuracy_2/train_accuracy_list/read
Accuracy_2/Assign/value
Accuracy_2/Assign
Accuracy_2/train_accuracy/tags
Accuracy_2/train_accuracy
Accuracy_2/Cast_1
Accuracy_2/Const_1
Accuracy_2/Mean_1
Accuracy_2/validate_accuracy_list/initial_value
Accuracy_2/validate_accuracy_list
Accuracy_2/validate_accuracy_list/Assign
Accuracy_2/validate_accuracy_list/read
Accuracy_2/Assign_1/value
Accuracy_2/Assign_1
Accuracy_2/validate_accuracy/tags
Accuracy_2/validate_accuracy
Accuracy_2/Cast_2
Accuracy_2/Const_2
Accuracy_2/Mean_2
Accuracy_2/test_accuracy_list/initial_value
Accuracy_2/test_accuracy_list
Accuracy_2/test_accuracy_list/Assign
Accuracy_2/test_accuracy_list/read
Accuracy_2/Assign_2/value
Accuracy_2/Assign_2
Accuracy_2/test_accuracy/tags
Accuracy_2/test_accuracy
init_2/NoOp
init_2/NoOp_1
init_2
Merge_2/MergeSummary
save_2/Const
save_2/SaveV2/tensor_names
save_2/SaveV2/shape_and_slices
save_2/SaveV2
save_2/control_dependency
save_2/RestoreV2/tensor_names
save_2/RestoreV2/shape_and_slices
save_2/RestoreV2
save_2/Assign
save_2/RestoreV2_1/tensor_names
save_2/RestoreV2_1/shape_and_slices
save_2/RestoreV2_1
save_2/Assign_1
save_2/RestoreV2_2/tensor_names
save_2/RestoreV2_2/shape_and_slices
save_2/RestoreV2_2
save_2/Assign_2
save_2/RestoreV2_3/tensor_names
save_2/RestoreV2_3/shape_and_slices
save_2/RestoreV2_3
save_2/Assign_3
save_2/RestoreV2_4/tensor_names
save_2/RestoreV2_4/shape_and_slices
save_2/RestoreV2_4
save_2/Assign_4
save_2/RestoreV2_5/tensor_names
save_2/RestoreV2_5/shape_and_slices
save_2/RestoreV2_5
save_2/Assign_5
save_2/RestoreV2_6/tensor_names
save_2/RestoreV2_6/shape_and_slices
save_2/RestoreV2_6
save_2/Assign_6
save_2/RestoreV2_7/tensor_names
save_2/RestoreV2_7/shape_and_slices
save_2/RestoreV2_7
save_2/Assign_7
save_2/RestoreV2_8/tensor_names
save_2/RestoreV2_8/shape_and_slices
save_2/RestoreV2_8
save_2/Assign_8
save_2/RestoreV2_9/tensor_names
save_2/RestoreV2_9/shape_and_slices
save_2/RestoreV2_9
save_2/Assign_9
save_2/RestoreV2_10/tensor_names
save_2/RestoreV2_10/shape_and_slices
save_2/RestoreV2_10
save_2/Assign_10
save_2/RestoreV2_11/tensor_names
save_2/RestoreV2_11/shape_and_slices
save_2/RestoreV2_11
save_2/Assign_11
save_2/RestoreV2_12/tensor_names
save_2/RestoreV2_12/shape_and_slices
save_2/RestoreV2_12
save_2/Assign_12
save_2/RestoreV2_13/tensor_names
save_2/RestoreV2_13/shape_and_slices
save_2/RestoreV2_13
save_2/Assign_13
save_2/RestoreV2_14/tensor_names
save_2/RestoreV2_14/shape_and_slices
save_2/RestoreV2_14
save_2/Assign_14
save_2/RestoreV2_15/tensor_names
save_2/RestoreV2_15/shape_and_slices
save_2/RestoreV2_15
save_2/Assign_15
save_2/RestoreV2_16/tensor_names
save_2/RestoreV2_16/shape_and_slices
save_2/RestoreV2_16
save_2/Assign_16
save_2/RestoreV2_17/tensor_names
save_2/RestoreV2_17/shape_and_slices
save_2/RestoreV2_17
save_2/Assign_17
save_2/RestoreV2_18/tensor_names
save_2/RestoreV2_18/shape_and_slices
save_2/RestoreV2_18
save_2/Assign_18
save_2/RestoreV2_19/tensor_names
save_2/RestoreV2_19/shape_and_slices
save_2/RestoreV2_19
save_2/Assign_19
save_2/RestoreV2_20/tensor_names
save_2/RestoreV2_20/shape_and_slices
save_2/RestoreV2_20
save_2/Assign_20
save_2/RestoreV2_21/tensor_names
save_2/RestoreV2_21/shape_and_slices
save_2/RestoreV2_21
save_2/Assign_21
save_2/RestoreV2_22/tensor_names
save_2/RestoreV2_22/shape_and_slices
save_2/RestoreV2_22
save_2/Assign_22
save_2/RestoreV2_23/tensor_names
save_2/RestoreV2_23/shape_and_slices
save_2/RestoreV2_23
save_2/Assign_23
save_2/RestoreV2_24/tensor_names
save_2/RestoreV2_24/shape_and_slices
save_2/RestoreV2_24
save_2/Assign_24
save_2/RestoreV2_25/tensor_names
save_2/RestoreV2_25/shape_and_slices
save_2/RestoreV2_25
save_2/Assign_25
save_2/RestoreV2_26/tensor_names
save_2/RestoreV2_26/shape_and_slices
save_2/RestoreV2_26
save_2/Assign_26
save_2/restore_all/NoOp
save_2/restore_all/NoOp_1
save_2/restore_all
last_successful_epoch_3/initial_value
last_successful_epoch_3
last_successful_epoch_3/Assign
last_successful_epoch_3/read
add_3/y
add_3
assign_updated_epoch_3
Inputs_3/Data/X_input
Inputs_3/Train_Labels/y_true_one_hot_label
Inputs_3/Train_Labels/y_true_class
Parameters_3/Weights/zeros
Parameters_3/Weights/Weight
Parameters_3/Weights/Weight/Assign
Parameters_3/Weights/Weight/read
Parameters_3/Bias/ones
Parameters_3/Bias/Bias
Parameters_3/Bias/Bias/Assign
Parameters_3/Bias/Bias/read
Predictions_3/mlp_layer/MatMul
Predictions_3/mlp_layer
Predictions_3/mlp_layer_1
Predictions_3/Softmax
Predictions_3/predict_class/dimension
Predictions_3/predict_class
Predictions_3/one_hot/Const
Predictions_3/one_hot/Const_1
Predictions_3/one_hot/depth
Predictions_3/one_hot/on_value
Predictions_3/one_hot/off_value
Predictions_3/one_hot
Cross_Entropy_3/Rank
Cross_Entropy_3/Shape
Cross_Entropy_3/Rank_1
Cross_Entropy_3/Shape_1
Cross_Entropy_3/Sub/y
Cross_Entropy_3/Sub
Cross_Entropy_3/Slice/begin
Cross_Entropy_3/Slice/size
Cross_Entropy_3/Slice
Cross_Entropy_3/concat/values_0
Cross_Entropy_3/concat/axis
Cross_Entropy_3/concat
Cross_Entropy_3/Reshape
Cross_Entropy_3/Rank_2
Cross_Entropy_3/Shape_2
Cross_Entropy_3/Sub_1/y
Cross_Entropy_3/Sub_1
Cross_Entropy_3/Slice_1/begin
Cross_Entropy_3/Slice_1/size
Cross_Entropy_3/Slice_1
Cross_Entropy_3/concat_1/values_0
Cross_Entropy_3/concat_1/axis
Cross_Entropy_3/concat_1
Cross_Entropy_3/Reshape_1
Cross_Entropy_3/SoftmaxCrossEntropyWithLogits
Cross_Entropy_3/Sub_2/y
Cross_Entropy_3/Sub_2
Cross_Entropy_3/Slice_2/begin
Cross_Entropy_3/Slice_2/size
Cross_Entropy_3/Slice_2
Cross_Entropy_3/Reshape_2
Loss_Function_3/Const
Loss_Function_3/Square
Loss_Function_3/Const_1
Loss_Function_3/Mean
Loss_Function_3/Const_2
Loss_Function_3/Mean_1
Loss_Function_3/Mul
Loss_Function_3/Add
Loss_Function_3/train_loss_list/initial_value
Loss_Function_3/train_loss_list
Loss_Function_3/train_loss_list/Assign
Loss_Function_3/train_loss_list/read
Loss_Function_3/Assign/value
Loss_Function_3/Assign
Loss_Function_3/train_error/tags
Loss_Function_3/train_error
Loss_Function_3/validate_loss_list/initial_value
Loss_Function_3/validate_loss_list
Loss_Function_3/validate_loss_list/Assign
Loss_Function_3/validate_loss_list/read
Loss_Function_3/Assign_1/value
Loss_Function_3/Assign_1
Loss_Function_3/validate_error/tags
Loss_Function_3/validate_error
Optimizer_3/learning_rate_progress/initial_value
Optimizer_3/learning_rate_progress
Optimizer_3/learning_rate_progress/Assign
Optimizer_3/learning_rate_progress/read
Optimizer_3/Assign/value
Optimizer_3/Assign
Optimizer_3/learning_rate/tags
Optimizer_3/learning_rate/values
Optimizer_3/learning_rate
Optimizer_3/gradients/Shape
Optimizer_3/gradients/Const
Optimizer_3/gradients/Fill
Optimizer_3/gradients/Loss_Function_3/Add_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Add_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/BroadcastGradientArgs
Optimizer_3/gradients/Loss_Function_3/Add_grad/Sum
Optimizer_3/gradients/Loss_Function_3/Add_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Add_grad/Sum_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/Reshape_1
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/group_deps
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/control_dependency
Optimizer_3/gradients/Loss_Function_3/Add_grad/tuple/control_dependency_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Reshape/shape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Tile
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Shape_2
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Const
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Prod
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Const_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Prod_1
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Maximum/y
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Maximum
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/floordiv
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/Cast
Optimizer_3/gradients/Loss_Function_3/Mean_1_grad/truediv
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/BroadcastGradientArgs
Optimizer_3/gradients/Loss_Function_3/Mul_grad/mul
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Sum
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mul_grad/mul_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Sum_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/Reshape_1
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/group_deps
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/control_dependency
Optimizer_3/gradients/Loss_Function_3/Mul_grad/tuple/control_dependency_1
Optimizer_3/gradients/Cross_Entropy_3/Reshape_2_grad/Shape
Optimizer_3/gradients/Cross_Entropy_3/Reshape_2_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Reshape/shape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Reshape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Tile/multiples
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Tile
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Shape
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Shape_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Const
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Prod
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Const_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Prod_1
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Maximum/y
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Maximum
Optimizer_3/gradients/Loss_Function_3/Mean_grad/floordiv
Optimizer_3/gradients/Loss_Function_3/Mean_grad/Cast
Optimizer_3/gradients/Loss_Function_3/Mean_grad/truediv
Optimizer_3/gradients/zeros_like
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_3/gradients/Cross_Entropy_3/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul/x
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul
Optimizer_3/gradients/Loss_Function_3/Square_grad/mul_1
Optimizer_3/gradients/Cross_Entropy_3/Reshape_grad/Shape
Optimizer_3/gradients/Cross_Entropy_3/Reshape_grad/Reshape
Optimizer_3/gradients/Predictions_3/Softmax_grad/mul
Optimizer_3/gradients/Predictions_3/Softmax_grad/Sum/reduction_indices
Optimizer_3/gradients/Predictions_3/Softmax_grad/Sum
Optimizer_3/gradients/Predictions_3/Softmax_grad/Reshape/shape
Optimizer_3/gradients/Predictions_3/Softmax_grad/Reshape
Optimizer_3/gradients/Predictions_3/Softmax_grad/sub
Optimizer_3/gradients/Predictions_3/Softmax_grad/mul_1
Optimizer_3/gradients/Predictions_3/mlp_layer_1_grad/ReluGrad
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/BiasAddGrad
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/group_deps
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/control_dependency
Optimizer_3/gradients/Predictions_3/mlp_layer_grad/tuple/control_dependency_1
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/MatMul
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/MatMul_1
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_3/gradients/Predictions_3/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_3/gradients/AddN
Optimizer_3/GradientDescent/learning_rate
Optimizer_3/GradientDescent/update_Parameters_3/Weights/Weight/ApplyGradientDescent
Optimizer_3/GradientDescent/update_Parameters_3/Bias/Bias/ApplyGradientDescent
Optimizer_3/GradientDescent
Equal_3
Accuracy_3/Cast
Accuracy_3/Const
Accuracy_3/Mean
Accuracy_3/train_accuracy_list/initial_value
Accuracy_3/train_accuracy_list
Accuracy_3/train_accuracy_list/Assign
Accuracy_3/train_accuracy_list/read
Accuracy_3/Assign/value
Accuracy_3/Assign
Accuracy_3/train_accuracy/tags
Accuracy_3/train_accuracy
Accuracy_3/Cast_1
Accuracy_3/Const_1
Accuracy_3/Mean_1
Accuracy_3/validate_accuracy_list/initial_value
Accuracy_3/validate_accuracy_list
Accuracy_3/validate_accuracy_list/Assign
Accuracy_3/validate_accuracy_list/read
Accuracy_3/Assign_1/value
Accuracy_3/Assign_1
Accuracy_3/validate_accuracy/tags
Accuracy_3/validate_accuracy
Accuracy_3/Cast_2
Accuracy_3/Const_2
Accuracy_3/Mean_2
Accuracy_3/test_accuracy_list/initial_value
Accuracy_3/test_accuracy_list
Accuracy_3/test_accuracy_list/Assign
Accuracy_3/test_accuracy_list/read
Accuracy_3/Assign_2/value
Accuracy_3/Assign_2
Accuracy_3/test_accuracy/tags
Accuracy_3/test_accuracy
init_3/NoOp
init_3/NoOp_1
init_3
Merge_3/MergeSummary
save_3/Const
save_3/SaveV2/tensor_names
save_3/SaveV2/shape_and_slices
save_3/SaveV2
save_3/control_dependency
save_3/RestoreV2/tensor_names
save_3/RestoreV2/shape_and_slices
save_3/RestoreV2
save_3/Assign
save_3/RestoreV2_1/tensor_names
save_3/RestoreV2_1/shape_and_slices
save_3/RestoreV2_1
save_3/Assign_1
save_3/RestoreV2_2/tensor_names
save_3/RestoreV2_2/shape_and_slices
save_3/RestoreV2_2
save_3/Assign_2
save_3/RestoreV2_3/tensor_names
save_3/RestoreV2_3/shape_and_slices
save_3/RestoreV2_3
save_3/Assign_3
save_3/RestoreV2_4/tensor_names
save_3/RestoreV2_4/shape_and_slices
save_3/RestoreV2_4
save_3/Assign_4
save_3/RestoreV2_5/tensor_names
save_3/RestoreV2_5/shape_and_slices
save_3/RestoreV2_5
save_3/Assign_5
save_3/RestoreV2_6/tensor_names
save_3/RestoreV2_6/shape_and_slices
save_3/RestoreV2_6
save_3/Assign_6
save_3/RestoreV2_7/tensor_names
save_3/RestoreV2_7/shape_and_slices
save_3/RestoreV2_7
save_3/Assign_7
save_3/RestoreV2_8/tensor_names
save_3/RestoreV2_8/shape_and_slices
save_3/RestoreV2_8
save_3/Assign_8
save_3/RestoreV2_9/tensor_names
save_3/RestoreV2_9/shape_and_slices
save_3/RestoreV2_9
save_3/Assign_9
save_3/RestoreV2_10/tensor_names
save_3/RestoreV2_10/shape_and_slices
save_3/RestoreV2_10
save_3/Assign_10
save_3/RestoreV2_11/tensor_names
save_3/RestoreV2_11/shape_and_slices
save_3/RestoreV2_11
save_3/Assign_11
save_3/RestoreV2_12/tensor_names
save_3/RestoreV2_12/shape_and_slices
save_3/RestoreV2_12
save_3/Assign_12
save_3/RestoreV2_13/tensor_names
save_3/RestoreV2_13/shape_and_slices
save_3/RestoreV2_13
save_3/Assign_13
save_3/RestoreV2_14/tensor_names
save_3/RestoreV2_14/shape_and_slices
save_3/RestoreV2_14
save_3/Assign_14
save_3/RestoreV2_15/tensor_names
save_3/RestoreV2_15/shape_and_slices
save_3/RestoreV2_15
save_3/Assign_15
save_3/RestoreV2_16/tensor_names
save_3/RestoreV2_16/shape_and_slices
save_3/RestoreV2_16
save_3/Assign_16
save_3/RestoreV2_17/tensor_names
save_3/RestoreV2_17/shape_and_slices
save_3/RestoreV2_17
save_3/Assign_17
save_3/RestoreV2_18/tensor_names
save_3/RestoreV2_18/shape_and_slices
save_3/RestoreV2_18
save_3/Assign_18
save_3/RestoreV2_19/tensor_names
save_3/RestoreV2_19/shape_and_slices
save_3/RestoreV2_19
save_3/Assign_19
save_3/RestoreV2_20/tensor_names
save_3/RestoreV2_20/shape_and_slices
save_3/RestoreV2_20
save_3/Assign_20
save_3/RestoreV2_21/tensor_names
save_3/RestoreV2_21/shape_and_slices
save_3/RestoreV2_21
save_3/Assign_21
save_3/RestoreV2_22/tensor_names
save_3/RestoreV2_22/shape_and_slices
save_3/RestoreV2_22
save_3/Assign_22
save_3/RestoreV2_23/tensor_names
save_3/RestoreV2_23/shape_and_slices
save_3/RestoreV2_23
save_3/Assign_23
save_3/RestoreV2_24/tensor_names
save_3/RestoreV2_24/shape_and_slices
save_3/RestoreV2_24
save_3/Assign_24
save_3/RestoreV2_25/tensor_names
save_3/RestoreV2_25/shape_and_slices
save_3/RestoreV2_25
save_3/Assign_25
save_3/RestoreV2_26/tensor_names
save_3/RestoreV2_26/shape_and_slices
save_3/RestoreV2_26
save_3/Assign_26
save_3/RestoreV2_27/tensor_names
save_3/RestoreV2_27/shape_and_slices
save_3/RestoreV2_27
save_3/Assign_27
save_3/RestoreV2_28/tensor_names
save_3/RestoreV2_28/shape_and_slices
save_3/RestoreV2_28
save_3/Assign_28
save_3/RestoreV2_29/tensor_names
save_3/RestoreV2_29/shape_and_slices
save_3/RestoreV2_29
save_3/Assign_29
save_3/RestoreV2_30/tensor_names
save_3/RestoreV2_30/shape_and_slices
save_3/RestoreV2_30
save_3/Assign_30
save_3/RestoreV2_31/tensor_names
save_3/RestoreV2_31/shape_and_slices
save_3/RestoreV2_31
save_3/Assign_31
save_3/RestoreV2_32/tensor_names
save_3/RestoreV2_32/shape_and_slices
save_3/RestoreV2_32
save_3/Assign_32
save_3/RestoreV2_33/tensor_names
save_3/RestoreV2_33/shape_and_slices
save_3/RestoreV2_33
save_3/Assign_33
save_3/RestoreV2_34/tensor_names
save_3/RestoreV2_34/shape_and_slices
save_3/RestoreV2_34
save_3/Assign_34
save_3/RestoreV2_35/tensor_names
save_3/RestoreV2_35/shape_and_slices
save_3/RestoreV2_35
save_3/Assign_35
save_3/restore_all/NoOp
save_3/restore_all/NoOp_1
save_3/restore_all
last_successful_epoch_4/initial_value
last_successful_epoch_4
last_successful_epoch_4/Assign
last_successful_epoch_4/read
add_4/y
add_4
assign_updated_epoch_4
Inputs_4/Data/X_input
Inputs_4/Train_Labels/y_true_one_hot_label
Inputs_4/Train_Labels/y_true_class
Parameters_4/Weights/zeros
Parameters_4/Weights/Weight
Parameters_4/Weights/Weight/Assign
Parameters_4/Weights/Weight/read
Parameters_4/Bias/ones
Parameters_4/Bias/Bias
Parameters_4/Bias/Bias/Assign
Parameters_4/Bias/Bias/read
Predictions_4/mlp_layer/MatMul
Predictions_4/mlp_layer
Predictions_4/mlp_layer_1
Predictions_4/Softmax
Predictions_4/predict_class/dimension
Predictions_4/predict_class
Predictions_4/one_hot/Const
Predictions_4/one_hot/Const_1
Predictions_4/one_hot/depth
Predictions_4/one_hot/on_value
Predictions_4/one_hot/off_value
Predictions_4/one_hot
Cross_Entropy_4/Rank
Cross_Entropy_4/Shape
Cross_Entropy_4/Rank_1
Cross_Entropy_4/Shape_1
Cross_Entropy_4/Sub/y
Cross_Entropy_4/Sub
Cross_Entropy_4/Slice/begin
Cross_Entropy_4/Slice/size
Cross_Entropy_4/Slice
Cross_Entropy_4/concat/values_0
Cross_Entropy_4/concat/axis
Cross_Entropy_4/concat
Cross_Entropy_4/Reshape
Cross_Entropy_4/Rank_2
Cross_Entropy_4/Shape_2
Cross_Entropy_4/Sub_1/y
Cross_Entropy_4/Sub_1
Cross_Entropy_4/Slice_1/begin
Cross_Entropy_4/Slice_1/size
Cross_Entropy_4/Slice_1
Cross_Entropy_4/concat_1/values_0
Cross_Entropy_4/concat_1/axis
Cross_Entropy_4/concat_1
Cross_Entropy_4/Reshape_1
Cross_Entropy_4/SoftmaxCrossEntropyWithLogits
Cross_Entropy_4/Sub_2/y
Cross_Entropy_4/Sub_2
Cross_Entropy_4/Slice_2/begin
Cross_Entropy_4/Slice_2/size
Cross_Entropy_4/Slice_2
Cross_Entropy_4/Reshape_2
Loss_Function_4/Const
Loss_Function_4/Square
Loss_Function_4/Const_1
Loss_Function_4/Mean
Loss_Function_4/Const_2
Loss_Function_4/Mean_1
Loss_Function_4/Mul
Loss_Function_4/Add
Loss_Function_4/train_loss_list/initial_value
Loss_Function_4/train_loss_list
Loss_Function_4/train_loss_list/Assign
Loss_Function_4/train_loss_list/read
Loss_Function_4/Assign/value
Loss_Function_4/Assign
Loss_Function_4/train_error/tags
Loss_Function_4/train_error
Loss_Function_4/validate_loss_list/initial_value
Loss_Function_4/validate_loss_list
Loss_Function_4/validate_loss_list/Assign
Loss_Function_4/validate_loss_list/read
Loss_Function_4/Assign_1/value
Loss_Function_4/Assign_1
Loss_Function_4/validate_error/tags
Loss_Function_4/validate_error
Optimizer_4/learning_rate_progress/initial_value
Optimizer_4/learning_rate_progress
Optimizer_4/learning_rate_progress/Assign
Optimizer_4/learning_rate_progress/read
Optimizer_4/Assign/value
Optimizer_4/Assign
Optimizer_4/learning_rate/tags
Optimizer_4/learning_rate/values
Optimizer_4/learning_rate
Optimizer_4/gradients/Shape
Optimizer_4/gradients/Const
Optimizer_4/gradients/Fill
Optimizer_4/gradients/Loss_Function_4/Add_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Add_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/BroadcastGradientArgs
Optimizer_4/gradients/Loss_Function_4/Add_grad/Sum
Optimizer_4/gradients/Loss_Function_4/Add_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Add_grad/Sum_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/Reshape_1
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/group_deps
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/control_dependency
Optimizer_4/gradients/Loss_Function_4/Add_grad/tuple/control_dependency_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Reshape/shape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Tile
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Shape_2
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Const
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Prod
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Const_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Prod_1
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Maximum/y
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Maximum
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/floordiv
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/Cast
Optimizer_4/gradients/Loss_Function_4/Mean_1_grad/truediv
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/BroadcastGradientArgs
Optimizer_4/gradients/Loss_Function_4/Mul_grad/mul
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Sum
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mul_grad/mul_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Sum_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/Reshape_1
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/group_deps
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/control_dependency
Optimizer_4/gradients/Loss_Function_4/Mul_grad/tuple/control_dependency_1
Optimizer_4/gradients/Cross_Entropy_4/Reshape_2_grad/Shape
Optimizer_4/gradients/Cross_Entropy_4/Reshape_2_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Reshape/shape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Reshape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Tile/multiples
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Tile
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Shape
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Shape_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Const
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Prod
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Const_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Prod_1
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Maximum/y
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Maximum
Optimizer_4/gradients/Loss_Function_4/Mean_grad/floordiv
Optimizer_4/gradients/Loss_Function_4/Mean_grad/Cast
Optimizer_4/gradients/Loss_Function_4/Mean_grad/truediv
Optimizer_4/gradients/zeros_like
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_4/gradients/Cross_Entropy_4/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul/x
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul
Optimizer_4/gradients/Loss_Function_4/Square_grad/mul_1
Optimizer_4/gradients/Cross_Entropy_4/Reshape_grad/Shape
Optimizer_4/gradients/Cross_Entropy_4/Reshape_grad/Reshape
Optimizer_4/gradients/Predictions_4/Softmax_grad/mul
Optimizer_4/gradients/Predictions_4/Softmax_grad/Sum/reduction_indices
Optimizer_4/gradients/Predictions_4/Softmax_grad/Sum
Optimizer_4/gradients/Predictions_4/Softmax_grad/Reshape/shape
Optimizer_4/gradients/Predictions_4/Softmax_grad/Reshape
Optimizer_4/gradients/Predictions_4/Softmax_grad/sub
Optimizer_4/gradients/Predictions_4/Softmax_grad/mul_1
Optimizer_4/gradients/Predictions_4/mlp_layer_1_grad/ReluGrad
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/BiasAddGrad
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/group_deps
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/control_dependency
Optimizer_4/gradients/Predictions_4/mlp_layer_grad/tuple/control_dependency_1
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/MatMul
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/MatMul_1
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_4/gradients/Predictions_4/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_4/gradients/AddN
Optimizer_4/GradientDescent/learning_rate
Optimizer_4/GradientDescent/update_Parameters_4/Weights/Weight/ApplyGradientDescent
Optimizer_4/GradientDescent/update_Parameters_4/Bias/Bias/ApplyGradientDescent
Optimizer_4/GradientDescent
Equal_4
Accuracy_4/Cast
Accuracy_4/Const
Accuracy_4/Mean
Accuracy_4/train_accuracy_list/initial_value
Accuracy_4/train_accuracy_list
Accuracy_4/train_accuracy_list/Assign
Accuracy_4/train_accuracy_list/read
Accuracy_4/Assign/value
Accuracy_4/Assign
Accuracy_4/train_accuracy/tags
Accuracy_4/train_accuracy
Accuracy_4/Cast_1
Accuracy_4/Const_1
Accuracy_4/Mean_1
Accuracy_4/validate_accuracy_list/initial_value
Accuracy_4/validate_accuracy_list
Accuracy_4/validate_accuracy_list/Assign
Accuracy_4/validate_accuracy_list/read
Accuracy_4/Assign_1/value
Accuracy_4/Assign_1
Accuracy_4/validate_accuracy/tags
Accuracy_4/validate_accuracy
Accuracy_4/Cast_2
Accuracy_4/Const_2
Accuracy_4/Mean_2
Accuracy_4/test_accuracy_list/initial_value
Accuracy_4/test_accuracy_list
Accuracy_4/test_accuracy_list/Assign
Accuracy_4/test_accuracy_list/read
Accuracy_4/Assign_2/value
Accuracy_4/Assign_2
Accuracy_4/test_accuracy/tags
Accuracy_4/test_accuracy
init_4/NoOp
init_4/NoOp_1
init_4
Merge_4/MergeSummary
save_4/Const
save_4/SaveV2/tensor_names
save_4/SaveV2/shape_and_slices
save_4/SaveV2
save_4/control_dependency
save_4/RestoreV2/tensor_names
save_4/RestoreV2/shape_and_slices
save_4/RestoreV2
save_4/Assign
save_4/RestoreV2_1/tensor_names
save_4/RestoreV2_1/shape_and_slices
save_4/RestoreV2_1
save_4/Assign_1
save_4/RestoreV2_2/tensor_names
save_4/RestoreV2_2/shape_and_slices
save_4/RestoreV2_2
save_4/Assign_2
save_4/RestoreV2_3/tensor_names
save_4/RestoreV2_3/shape_and_slices
save_4/RestoreV2_3
save_4/Assign_3
save_4/RestoreV2_4/tensor_names
save_4/RestoreV2_4/shape_and_slices
save_4/RestoreV2_4
save_4/Assign_4
save_4/RestoreV2_5/tensor_names
save_4/RestoreV2_5/shape_and_slices
save_4/RestoreV2_5
save_4/Assign_5
save_4/RestoreV2_6/tensor_names
save_4/RestoreV2_6/shape_and_slices
save_4/RestoreV2_6
save_4/Assign_6
save_4/RestoreV2_7/tensor_names
save_4/RestoreV2_7/shape_and_slices
save_4/RestoreV2_7
save_4/Assign_7
save_4/RestoreV2_8/tensor_names
save_4/RestoreV2_8/shape_and_slices
save_4/RestoreV2_8
save_4/Assign_8
save_4/RestoreV2_9/tensor_names
save_4/RestoreV2_9/shape_and_slices
save_4/RestoreV2_9
save_4/Assign_9
save_4/RestoreV2_10/tensor_names
save_4/RestoreV2_10/shape_and_slices
save_4/RestoreV2_10
save_4/Assign_10
save_4/RestoreV2_11/tensor_names
save_4/RestoreV2_11/shape_and_slices
save_4/RestoreV2_11
save_4/Assign_11
save_4/RestoreV2_12/tensor_names
save_4/RestoreV2_12/shape_and_slices
save_4/RestoreV2_12
save_4/Assign_12
save_4/RestoreV2_13/tensor_names
save_4/RestoreV2_13/shape_and_slices
save_4/RestoreV2_13
save_4/Assign_13
save_4/RestoreV2_14/tensor_names
save_4/RestoreV2_14/shape_and_slices
save_4/RestoreV2_14
save_4/Assign_14
save_4/RestoreV2_15/tensor_names
save_4/RestoreV2_15/shape_and_slices
save_4/RestoreV2_15
save_4/Assign_15
save_4/RestoreV2_16/tensor_names
save_4/RestoreV2_16/shape_and_slices
save_4/RestoreV2_16
save_4/Assign_16
save_4/RestoreV2_17/tensor_names
save_4/RestoreV2_17/shape_and_slices
save_4/RestoreV2_17
save_4/Assign_17
save_4/RestoreV2_18/tensor_names
save_4/RestoreV2_18/shape_and_slices
save_4/RestoreV2_18
save_4/Assign_18
save_4/RestoreV2_19/tensor_names
save_4/RestoreV2_19/shape_and_slices
save_4/RestoreV2_19
save_4/Assign_19
save_4/RestoreV2_20/tensor_names
save_4/RestoreV2_20/shape_and_slices
save_4/RestoreV2_20
save_4/Assign_20
save_4/RestoreV2_21/tensor_names
save_4/RestoreV2_21/shape_and_slices
save_4/RestoreV2_21
save_4/Assign_21
save_4/RestoreV2_22/tensor_names
save_4/RestoreV2_22/shape_and_slices
save_4/RestoreV2_22
save_4/Assign_22
save_4/RestoreV2_23/tensor_names
save_4/RestoreV2_23/shape_and_slices
save_4/RestoreV2_23
save_4/Assign_23
save_4/RestoreV2_24/tensor_names
save_4/RestoreV2_24/shape_and_slices
save_4/RestoreV2_24
save_4/Assign_24
save_4/RestoreV2_25/tensor_names
save_4/RestoreV2_25/shape_and_slices
save_4/RestoreV2_25
save_4/Assign_25
save_4/RestoreV2_26/tensor_names
save_4/RestoreV2_26/shape_and_slices
save_4/RestoreV2_26
save_4/Assign_26
save_4/RestoreV2_27/tensor_names
save_4/RestoreV2_27/shape_and_slices
save_4/RestoreV2_27
save_4/Assign_27
save_4/RestoreV2_28/tensor_names
save_4/RestoreV2_28/shape_and_slices
save_4/RestoreV2_28
save_4/Assign_28
save_4/RestoreV2_29/tensor_names
save_4/RestoreV2_29/shape_and_slices
save_4/RestoreV2_29
save_4/Assign_29
save_4/RestoreV2_30/tensor_names
save_4/RestoreV2_30/shape_and_slices
save_4/RestoreV2_30
save_4/Assign_30
save_4/RestoreV2_31/tensor_names
save_4/RestoreV2_31/shape_and_slices
save_4/RestoreV2_31
save_4/Assign_31
save_4/RestoreV2_32/tensor_names
save_4/RestoreV2_32/shape_and_slices
save_4/RestoreV2_32
save_4/Assign_32
save_4/RestoreV2_33/tensor_names
save_4/RestoreV2_33/shape_and_slices
save_4/RestoreV2_33
save_4/Assign_33
save_4/RestoreV2_34/tensor_names
save_4/RestoreV2_34/shape_and_slices
save_4/RestoreV2_34
save_4/Assign_34
save_4/RestoreV2_35/tensor_names
save_4/RestoreV2_35/shape_and_slices
save_4/RestoreV2_35
save_4/Assign_35
save_4/RestoreV2_36/tensor_names
save_4/RestoreV2_36/shape_and_slices
save_4/RestoreV2_36
save_4/Assign_36
save_4/RestoreV2_37/tensor_names
save_4/RestoreV2_37/shape_and_slices
save_4/RestoreV2_37
save_4/Assign_37
save_4/RestoreV2_38/tensor_names
save_4/RestoreV2_38/shape_and_slices
save_4/RestoreV2_38
save_4/Assign_38
save_4/RestoreV2_39/tensor_names
save_4/RestoreV2_39/shape_and_slices
save_4/RestoreV2_39
save_4/Assign_39
save_4/RestoreV2_40/tensor_names
save_4/RestoreV2_40/shape_and_slices
save_4/RestoreV2_40
save_4/Assign_40
save_4/RestoreV2_41/tensor_names
save_4/RestoreV2_41/shape_and_slices
save_4/RestoreV2_41
save_4/Assign_41
save_4/RestoreV2_42/tensor_names
save_4/RestoreV2_42/shape_and_slices
save_4/RestoreV2_42
save_4/Assign_42
save_4/RestoreV2_43/tensor_names
save_4/RestoreV2_43/shape_and_slices
save_4/RestoreV2_43
save_4/Assign_43
save_4/RestoreV2_44/tensor_names
save_4/RestoreV2_44/shape_and_slices
save_4/RestoreV2_44
save_4/Assign_44
save_4/restore_all/NoOp
save_4/restore_all/NoOp_1
save_4/restore_all
last_successful_epoch_5/initial_value
last_successful_epoch_5
last_successful_epoch_5/Assign
last_successful_epoch_5/read
add_5/y
add_5
assign_updated_epoch_5
Inputs_5/Data/X_input
Inputs_5/Train_Labels/y_true_one_hot_label
Inputs_5/Train_Labels/y_true_class
Parameters_5/Weights/zeros
Parameters_5/Weights/Weight
Parameters_5/Weights/Weight/Assign
Parameters_5/Weights/Weight/read
Parameters_5/Bias/ones
Parameters_5/Bias/Bias
Parameters_5/Bias/Bias/Assign
Parameters_5/Bias/Bias/read
Predictions_5/mlp_layer/MatMul
Predictions_5/mlp_layer
Predictions_5/mlp_layer_1
Predictions_5/Softmax
Predictions_5/predict_class/dimension
Predictions_5/predict_class
Predictions_5/one_hot/Const
Predictions_5/one_hot/Const_1
Predictions_5/one_hot/depth
Predictions_5/one_hot/on_value
Predictions_5/one_hot/off_value
Predictions_5/one_hot
Cross_Entropy_5/Rank
Cross_Entropy_5/Shape
Cross_Entropy_5/Rank_1
Cross_Entropy_5/Shape_1
Cross_Entropy_5/Sub/y
Cross_Entropy_5/Sub
Cross_Entropy_5/Slice/begin
Cross_Entropy_5/Slice/size
Cross_Entropy_5/Slice
Cross_Entropy_5/concat/values_0
Cross_Entropy_5/concat/axis
Cross_Entropy_5/concat
Cross_Entropy_5/Reshape
Cross_Entropy_5/Rank_2
Cross_Entropy_5/Shape_2
Cross_Entropy_5/Sub_1/y
Cross_Entropy_5/Sub_1
Cross_Entropy_5/Slice_1/begin
Cross_Entropy_5/Slice_1/size
Cross_Entropy_5/Slice_1
Cross_Entropy_5/concat_1/values_0
Cross_Entropy_5/concat_1/axis
Cross_Entropy_5/concat_1
Cross_Entropy_5/Reshape_1
Cross_Entropy_5/SoftmaxCrossEntropyWithLogits
Cross_Entropy_5/Sub_2/y
Cross_Entropy_5/Sub_2
Cross_Entropy_5/Slice_2/begin
Cross_Entropy_5/Slice_2/size
Cross_Entropy_5/Slice_2
Cross_Entropy_5/Reshape_2
Loss_Function_5/Const
Loss_Function_5/Square
Loss_Function_5/Const_1
Loss_Function_5/Mean
Loss_Function_5/Const_2
Loss_Function_5/Mean_1
Loss_Function_5/Mul
Loss_Function_5/Add
Loss_Function_5/train_loss_list/initial_value
Loss_Function_5/train_loss_list
Loss_Function_5/train_loss_list/Assign
Loss_Function_5/train_loss_list/read
Loss_Function_5/Assign/value
Loss_Function_5/Assign
Loss_Function_5/train_error/tags
Loss_Function_5/train_error
Loss_Function_5/validate_loss_list/initial_value
Loss_Function_5/validate_loss_list
Loss_Function_5/validate_loss_list/Assign
Loss_Function_5/validate_loss_list/read
Loss_Function_5/Assign_1/value
Loss_Function_5/Assign_1
Loss_Function_5/validate_error/tags
Loss_Function_5/validate_error
Optimizer_5/learning_rate_progress/initial_value
Optimizer_5/learning_rate_progress
Optimizer_5/learning_rate_progress/Assign
Optimizer_5/learning_rate_progress/read
Optimizer_5/Assign/value
Optimizer_5/Assign
Optimizer_5/learning_rate/tags
Optimizer_5/learning_rate/values
Optimizer_5/learning_rate
Optimizer_5/gradients/Shape
Optimizer_5/gradients/Const
Optimizer_5/gradients/Fill
Optimizer_5/gradients/Loss_Function_5/Add_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Add_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/BroadcastGradientArgs
Optimizer_5/gradients/Loss_Function_5/Add_grad/Sum
Optimizer_5/gradients/Loss_Function_5/Add_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Add_grad/Sum_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/Reshape_1
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/group_deps
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/control_dependency
Optimizer_5/gradients/Loss_Function_5/Add_grad/tuple/control_dependency_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Reshape/shape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Tile
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Shape_2
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Const
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Prod
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Const_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Prod_1
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Maximum/y
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Maximum
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/floordiv
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/Cast
Optimizer_5/gradients/Loss_Function_5/Mean_1_grad/truediv
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/BroadcastGradientArgs
Optimizer_5/gradients/Loss_Function_5/Mul_grad/mul
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Sum
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mul_grad/mul_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Sum_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/Reshape_1
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/group_deps
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/control_dependency
Optimizer_5/gradients/Loss_Function_5/Mul_grad/tuple/control_dependency_1
Optimizer_5/gradients/Cross_Entropy_5/Reshape_2_grad/Shape
Optimizer_5/gradients/Cross_Entropy_5/Reshape_2_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Reshape/shape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Reshape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Tile/multiples
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Tile
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Shape
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Shape_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Const
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Prod
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Const_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Prod_1
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Maximum/y
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Maximum
Optimizer_5/gradients/Loss_Function_5/Mean_grad/floordiv
Optimizer_5/gradients/Loss_Function_5/Mean_grad/Cast
Optimizer_5/gradients/Loss_Function_5/Mean_grad/truediv
Optimizer_5/gradients/zeros_like
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/PreventGradient
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/ExpandDims/dim
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/ExpandDims
Optimizer_5/gradients/Cross_Entropy_5/SoftmaxCrossEntropyWithLogits_grad/mul
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul/x
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul
Optimizer_5/gradients/Loss_Function_5/Square_grad/mul_1
Optimizer_5/gradients/Cross_Entropy_5/Reshape_grad/Shape
Optimizer_5/gradients/Cross_Entropy_5/Reshape_grad/Reshape
Optimizer_5/gradients/Predictions_5/Softmax_grad/mul
Optimizer_5/gradients/Predictions_5/Softmax_grad/Sum/reduction_indices
Optimizer_5/gradients/Predictions_5/Softmax_grad/Sum
Optimizer_5/gradients/Predictions_5/Softmax_grad/Reshape/shape
Optimizer_5/gradients/Predictions_5/Softmax_grad/Reshape
Optimizer_5/gradients/Predictions_5/Softmax_grad/sub
Optimizer_5/gradients/Predictions_5/Softmax_grad/mul_1
Optimizer_5/gradients/Predictions_5/mlp_layer_1_grad/ReluGrad
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/BiasAddGrad
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/group_deps
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/control_dependency
Optimizer_5/gradients/Predictions_5/mlp_layer_grad/tuple/control_dependency_1
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/MatMul
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/MatMul_1
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/group_deps
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/control_dependency
Optimizer_5/gradients/Predictions_5/mlp_layer/MatMul_grad/tuple/control_dependency_1
Optimizer_5/gradients/AddN
Optimizer_5/GradientDescent/learning_rate
Optimizer_5/GradientDescent/update_Parameters_5/Weights/Weight/ApplyGradientDescent
Optimizer_5/GradientDescent/update_Parameters_5/Bias/Bias/ApplyGradientDescent
Optimizer_5/GradientDescent
Equal_5
Accuracy_5/Cast
Accuracy_5/Const
Accuracy_5/Mean
Accuracy_5/train_accuracy_list/initial_value
Accuracy_5/train_accuracy_list
Accuracy_5/train_accuracy_list/Assign
Accuracy_5/train_accuracy_list/read
Accuracy_5/Assign/value
Accuracy_5/Assign
Accuracy_5/train_accuracy/tags
Accuracy_5/train_accuracy
Accuracy_5/Cast_1
Accuracy_5/Const_1
Accuracy_5/Mean_1
Accuracy_5/validate_accuracy_list/initial_value
Accuracy_5/validate_accuracy_list
Accuracy_5/validate_accuracy_list/Assign
Accuracy_5/validate_accuracy_list/read
Accuracy_5/Assign_1/value
Accuracy_5/Assign_1
Accuracy_5/validate_accuracy/tags
Accuracy_5/validate_accuracy
Accuracy_5/Cast_2
Accuracy_5/Const_2
Accuracy_5/Mean_2
Accuracy_5/test_accuracy_list/initial_value
Accuracy_5/test_accuracy_list
Accuracy_5/test_accuracy_list/Assign
Accuracy_5/test_accuracy_list/read
Accuracy_5/Assign_2/value
Accuracy_5/Assign_2
Accuracy_5/test_accuracy/tags
Accuracy_5/test_accuracy
init_5/NoOp
init_5/NoOp_1
init_5
Merge_5/MergeSummary
save_5/Const
save_5/SaveV2/tensor_names
save_5/SaveV2/shape_and_slices
save_5/SaveV2
save_5/control_dependency
save_5/RestoreV2/tensor_names
save_5/RestoreV2/shape_and_slices
save_5/RestoreV2
save_5/Assign
save_5/RestoreV2_1/tensor_names
save_5/RestoreV2_1/shape_and_slices
save_5/RestoreV2_1
save_5/Assign_1
save_5/RestoreV2_2/tensor_names
save_5/RestoreV2_2/shape_and_slices
save_5/RestoreV2_2
save_5/Assign_2
save_5/RestoreV2_3/tensor_names
save_5/RestoreV2_3/shape_and_slices
save_5/RestoreV2_3
save_5/Assign_3
save_5/RestoreV2_4/tensor_names
save_5/RestoreV2_4/shape_and_slices
save_5/RestoreV2_4
save_5/Assign_4
save_5/RestoreV2_5/tensor_names
save_5/RestoreV2_5/shape_and_slices
save_5/RestoreV2_5
save_5/Assign_5
save_5/RestoreV2_6/tensor_names
save_5/RestoreV2_6/shape_and_slices
save_5/RestoreV2_6
save_5/Assign_6
save_5/RestoreV2_7/tensor_names
save_5/RestoreV2_7/shape_and_slices
save_5/RestoreV2_7
save_5/Assign_7
save_5/RestoreV2_8/tensor_names
save_5/RestoreV2_8/shape_and_slices
save_5/RestoreV2_8
save_5/Assign_8
save_5/RestoreV2_9/tensor_names
save_5/RestoreV2_9/shape_and_slices
save_5/RestoreV2_9
save_5/Assign_9
save_5/RestoreV2_10/tensor_names
save_5/RestoreV2_10/shape_and_slices
save_5/RestoreV2_10
save_5/Assign_10
save_5/RestoreV2_11/tensor_names
save_5/RestoreV2_11/shape_and_slices
save_5/RestoreV2_11
save_5/Assign_11
save_5/RestoreV2_12/tensor_names
save_5/RestoreV2_12/shape_and_slices
save_5/RestoreV2_12
save_5/Assign_12
save_5/RestoreV2_13/tensor_names
save_5/RestoreV2_13/shape_and_slices
save_5/RestoreV2_13
save_5/Assign_13
save_5/RestoreV2_14/tensor_names
save_5/RestoreV2_14/shape_and_slices
save_5/RestoreV2_14
save_5/Assign_14
save_5/RestoreV2_15/tensor_names
save_5/RestoreV2_15/shape_and_slices
save_5/RestoreV2_15
save_5/Assign_15
save_5/RestoreV2_16/tensor_names
save_5/RestoreV2_16/shape_and_slices
save_5/RestoreV2_16
save_5/Assign_16
save_5/RestoreV2_17/tensor_names
save_5/RestoreV2_17/shape_and_slices
save_5/RestoreV2_17
save_5/Assign_17
save_5/RestoreV2_18/tensor_names
save_5/RestoreV2_18/shape_and_slices
save_5/RestoreV2_18
save_5/Assign_18
save_5/RestoreV2_19/tensor_names
save_5/RestoreV2_19/shape_and_slices
save_5/RestoreV2_19
save_5/Assign_19
save_5/RestoreV2_20/tensor_names
save_5/RestoreV2_20/shape_and_slices
save_5/RestoreV2_20
save_5/Assign_20
save_5/RestoreV2_21/tensor_names
save_5/RestoreV2_21/shape_and_slices
save_5/RestoreV2_21
save_5/Assign_21
save_5/RestoreV2_22/tensor_names
save_5/RestoreV2_22/shape_and_slices
save_5/RestoreV2_22
save_5/Assign_22
save_5/RestoreV2_23/tensor_names
save_5/RestoreV2_23/shape_and_slices
save_5/RestoreV2_23
save_5/Assign_23
save_5/RestoreV2_24/tensor_names
save_5/RestoreV2_24/shape_and_slices
save_5/RestoreV2_24
save_5/Assign_24
save_5/RestoreV2_25/tensor_names
save_5/RestoreV2_25/shape_and_slices
save_5/RestoreV2_25
save_5/Assign_25
save_5/RestoreV2_26/tensor_names
save_5/RestoreV2_26/shape_and_slices
save_5/RestoreV2_26
save_5/Assign_26
save_5/RestoreV2_27/tensor_names
save_5/RestoreV2_27/shape_and_slices
save_5/RestoreV2_27
save_5/Assign_27
save_5/RestoreV2_28/tensor_names
save_5/RestoreV2_28/shape_and_slices
save_5/RestoreV2_28
save_5/Assign_28
save_5/RestoreV2_29/tensor_names
save_5/RestoreV2_29/shape_and_slices
save_5/RestoreV2_29
save_5/Assign_29
save_5/RestoreV2_30/tensor_names
save_5/RestoreV2_30/shape_and_slices
save_5/RestoreV2_30
save_5/Assign_30
save_5/RestoreV2_31/tensor_names
save_5/RestoreV2_31/shape_and_slices
save_5/RestoreV2_31
save_5/Assign_31
save_5/RestoreV2_32/tensor_names
save_5/RestoreV2_32/shape_and_slices
save_5/RestoreV2_32
save_5/Assign_32
save_5/RestoreV2_33/tensor_names
save_5/RestoreV2_33/shape_and_slices
save_5/RestoreV2_33
save_5/Assign_33
save_5/RestoreV2_34/tensor_names
save_5/RestoreV2_34/shape_and_slices
save_5/RestoreV2_34
save_5/Assign_34
save_5/RestoreV2_35/tensor_names
save_5/RestoreV2_35/shape_and_slices
save_5/RestoreV2_35
save_5/Assign_35
save_5/RestoreV2_36/tensor_names
save_5/RestoreV2_36/shape_and_slices
save_5/RestoreV2_36
save_5/Assign_36
save_5/RestoreV2_37/tensor_names
save_5/RestoreV2_37/shape_and_slices
save_5/RestoreV2_37
save_5/Assign_37
save_5/RestoreV2_38/tensor_names
save_5/RestoreV2_38/shape_and_slices
save_5/RestoreV2_38
save_5/Assign_38
save_5/RestoreV2_39/tensor_names
save_5/RestoreV2_39/shape_and_slices
save_5/RestoreV2_39
save_5/Assign_39
save_5/RestoreV2_40/tensor_names
save_5/RestoreV2_40/shape_and_slices
save_5/RestoreV2_40
save_5/Assign_40
save_5/RestoreV2_41/tensor_names
save_5/RestoreV2_41/shape_and_slices
save_5/RestoreV2_41
save_5/Assign_41
save_5/RestoreV2_42/tensor_names
save_5/RestoreV2_42/shape_and_slices
save_5/RestoreV2_42
save_5/Assign_42
save_5/RestoreV2_43/tensor_names
save_5/RestoreV2_43/shape_and_slices
save_5/RestoreV2_43
save_5/Assign_43
save_5/RestoreV2_44/tensor_names
save_5/RestoreV2_44/shape_and_slices
save_5/RestoreV2_44
save_5/Assign_44
save_5/RestoreV2_45/tensor_names
save_5/RestoreV2_45/shape_and_slices
save_5/RestoreV2_45
save_5/Assign_45
save_5/RestoreV2_46/tensor_names
save_5/RestoreV2_46/shape_and_slices
save_5/RestoreV2_46
save_5/Assign_46
save_5/RestoreV2_47/tensor_names
save_5/RestoreV2_47/shape_and_slices
save_5/RestoreV2_47
save_5/Assign_47
save_5/RestoreV2_48/tensor_names
save_5/RestoreV2_48/shape_and_slices
save_5/RestoreV2_48
save_5/Assign_48
save_5/RestoreV2_49/tensor_names
save_5/RestoreV2_49/shape_and_slices
save_5/RestoreV2_49
save_5/Assign_49
save_5/RestoreV2_50/tensor_names
save_5/RestoreV2_50/shape_and_slices
save_5/RestoreV2_50
save_5/Assign_50
save_5/RestoreV2_51/tensor_names
save_5/RestoreV2_51/shape_and_slices
save_5/RestoreV2_51
save_5/Assign_51
save_5/RestoreV2_52/tensor_names
save_5/RestoreV2_52/shape_and_slices
save_5/RestoreV2_52
save_5/Assign_52
save_5/RestoreV2_53/tensor_names
save_5/RestoreV2_53/shape_and_slices
save_5/RestoreV2_53
save_5/Assign_53
save_5/restore_all/NoOp
save_5/restore_all/NoOp_1
save_5/restore_all
Output Logits  : Tensor("Predictions_5/Softmax:0", shape=(?, 10), dtype=float32, device=/device:GPU:0)
Entropy        : Tensor("Cross_Entropy_5/Reshape_2:0", shape=(?,), dtype=float32, device=/device:GPU:0)
Predictions    : Tensor("Equal_5:0", shape=(?,), dtype=bool, device=/device:GPU:0) 
Optimizer      : name: "Optimizer_5/GradientDescent"
op: "NoOp"
input: "^Optimizer_5/GradientDescent/update_Parameters_5/Weights/Weight/ApplyGradientDescent"
input: "^Optimizer_5/GradientDescent/update_Parameters_5/Bias/Bias/ApplyGradientDescent"
device: "/device:GPU:0"
 
>> Model info
Train loss     : Tensor("Loss_Function_5/Add:0", shape=(), dtype=float32, device=/device:GPU:0)
Train accuracy : Tensor("Accuracy_5/Mean:0", shape=(), dtype=float32, device=/device:GPU:0)
Val. loss      : Tensor("Loss_Function_5/Add:0", shape=(), dtype=float32, device=/device:GPU:0)
Val. accuracy  : Tensor("Accuracy_5/Mean_1:0", shape=(), dtype=float32, device=/device:GPU:0)
>> Summary params
Train loss     : Tensor("Loss_Function_5/train_error:0", shape=(), dtype=string, device=/device:GPU:0)
Train accuracy : Tensor("Accuracy_5/train_accuracy:0", shape=(), dtype=string, device=/device:GPU:0)
Val. loss      : Tensor("Loss_Function_5/validate_error:0", shape=(), dtype=string, device=/device:GPU:0)
Val. accuracy  : Tensor("Accuracy_5/validate_accuracy:0", shape=(), dtype=string, device=/device:GPU:0)

Step 2.3: Fit the model/training



In [20]:

    
for index, descent_method in enumerate(descent_methods):
    lcs[index].fit(cifar10.train.data, cifar10.train.one_hot_labels, cifar10.train.class_labels,
                   test_data=cifar10.test.data, test_labels=cifar10.test.one_hot_labels, 
                   test_classes=cifar10.test.class_labels)









    



Length of train loss          : 0
Length of train accuracy      : 0
Length of validate loss       : 0
Length of validate accuracy   : 0
Length of test accuracy       : 0
Restoring training from epoch : 0
>>> Epoch [  0/100]
train_loss: 2.3026 | train_acc: 0.1004 | val_loss: 2.3007 | val_acc: 0.2434 | test_acc: 0.2571 | Time: 1.3915 s
>>> Epoch [  1/100]
train_loss: 2.3006 | train_acc: 0.2498 | val_loss: 2.2987 | val_acc: 0.2433 | test_acc: 0.2565 | Time: 1.0060 s
>>> Epoch [  2/100]
train_loss: 2.2985 | train_acc: 0.2492 | val_loss: 2.2966 | val_acc: 0.2429 | test_acc: 0.2561 | Time: 1.0030 s
>>> Epoch [  3/100]
train_loss: 2.2964 | train_acc: 0.2483 | val_loss: 2.2944 | val_acc: 0.2425 | test_acc: 0.2559 | Time: 1.1103 s
>>> Epoch [  4/100]
train_loss: 2.2942 | train_acc: 0.2473 | val_loss: 2.2921 | val_acc: 0.2423 | test_acc: 0.2538 | Time: 0.9826 s
>>> Epoch [  5/100]
train_loss: 2.2918 | train_acc: 0.2465 | val_loss: 2.2897 | val_acc: 0.2416 | test_acc: 0.2516 | Time: 0.9726 s
>>> Epoch [  6/100]
train_loss: 2.2894 | train_acc: 0.2455 | val_loss: 2.2873 | val_acc: 0.2412 | test_acc: 0.2501 | Time: 1.0899 s
>>> Epoch [  7/100]
train_loss: 2.2869 | train_acc: 0.2443 | val_loss: 2.2847 | val_acc: 0.2413 | test_acc: 0.2492 | Time: 0.8820 s
>>> Epoch [  8/100]
train_loss: 2.2843 | train_acc: 0.2436 | val_loss: 2.2821 | val_acc: 0.2410 | test_acc: 0.2477 | Time: 1.0349 s
>>> Epoch [  9/100]
train_loss: 2.2817 | train_acc: 0.2432 | val_loss: 2.2795 | val_acc: 0.2408 | test_acc: 0.2470 | Time: 1.0423 s
>>> Epoch [ 10/100]
train_loss: 2.2790 | train_acc: 0.2424 | val_loss: 2.2769 | val_acc: 0.2403 | test_acc: 0.2462 | Time: 0.8490 s
>>> Epoch [ 11/100]
train_loss: 2.2764 | train_acc: 0.2423 | val_loss: 2.2743 | val_acc: 0.2397 | test_acc: 0.2463 | Time: 1.0436 s
>>> Epoch [ 12/100]
train_loss: 2.2738 | train_acc: 0.2423 | val_loss: 2.2718 | val_acc: 0.2395 | test_acc: 0.2461 | Time: 1.0275 s
>>> Epoch [ 13/100]
train_loss: 2.2713 | train_acc: 0.2424 | val_loss: 2.2694 | val_acc: 0.2396 | test_acc: 0.2469 | Time: 1.1160 s
>>> Epoch [ 14/100]
train_loss: 2.2689 | train_acc: 0.2426 | val_loss: 2.2671 | val_acc: 0.2403 | test_acc: 0.2472 | Time: 1.0137 s
>>> Epoch [ 15/100]
train_loss: 2.2666 | train_acc: 0.2428 | val_loss: 2.2649 | val_acc: 0.2412 | test_acc: 0.2487 | Time: 0.9927 s
>>> Epoch [ 16/100]
train_loss: 2.2643 | train_acc: 0.2434 | val_loss: 2.2628 | val_acc: 0.2420 | test_acc: 0.2503 | Time: 1.1302 s
>>> Epoch [ 17/100]
train_loss: 2.2621 | train_acc: 0.2443 | val_loss: 2.2607 | val_acc: 0.2427 | test_acc: 0.2518 | Time: 1.0599 s
>>> Epoch [ 18/100]
train_loss: 2.2600 | train_acc: 0.2448 | val_loss: 2.2587 | val_acc: 0.2427 | test_acc: 0.2533 | Time: 1.0555 s
>>> Epoch [ 19/100]
train_loss: 2.2580 | train_acc: 0.2458 | val_loss: 2.2568 | val_acc: 0.2429 | test_acc: 0.2539 | Time: 1.1236 s
>>> Epoch [ 20/100]
train_loss: 2.2560 | train_acc: 0.2466 | val_loss: 2.2549 | val_acc: 0.2440 | test_acc: 0.2542 | Time: 1.0431 s
>>> Epoch [ 21/100]
train_loss: 2.2541 | train_acc: 0.2477 | val_loss: 2.2531 | val_acc: 0.2447 | test_acc: 0.2550 | Time: 1.0181 s
>>> Epoch [ 22/100]
train_loss: 2.2522 | train_acc: 0.2488 | val_loss: 2.2513 | val_acc: 0.2453 | test_acc: 0.2561 | Time: 1.0633 s
>>> Epoch [ 23/100]
train_loss: 2.2503 | train_acc: 0.2497 | val_loss: 2.2495 | val_acc: 0.2471 | test_acc: 0.2572 | Time: 1.0756 s
>>> Epoch [ 24/100]
train_loss: 2.2485 | train_acc: 0.2506 | val_loss: 2.2478 | val_acc: 0.2479 | test_acc: 0.2582 | Time: 1.0129 s
>>> Epoch [ 25/100]
train_loss: 2.2467 | train_acc: 0.2519 | val_loss: 2.2461 | val_acc: 0.2491 | test_acc: 0.2591 | Time: 1.1029 s
>>> Epoch [ 26/100]
train_loss: 2.2450 | train_acc: 0.2533 | val_loss: 2.2445 | val_acc: 0.2494 | test_acc: 0.2603 | Time: 1.0199 s
>>> Epoch [ 27/100]
train_loss: 2.2433 | train_acc: 0.2546 | val_loss: 2.2429 | val_acc: 0.2501 | test_acc: 0.2607 | Time: 1.0421 s
>>> Epoch [ 28/100]
train_loss: 2.2416 | train_acc: 0.2549 | val_loss: 2.2413 | val_acc: 0.2508 | test_acc: 0.2616 | Time: 1.0902 s
>>> Epoch [ 29/100]
train_loss: 2.2400 | train_acc: 0.2554 | val_loss: 2.2398 | val_acc: 0.2516 | test_acc: 0.2625 | Time: 1.1327 s
>>> Epoch [ 30/100]
train_loss: 2.2384 | train_acc: 0.2563 | val_loss: 2.2383 | val_acc: 0.2523 | test_acc: 0.2635 | Time: 1.0578 s
>>> Epoch [ 31/100]
train_loss: 2.2368 | train_acc: 0.2574 | val_loss: 2.2368 | val_acc: 0.2538 | test_acc: 0.2638 | Time: 1.0445 s
>>> Epoch [ 32/100]
train_loss: 2.2353 | train_acc: 0.2584 | val_loss: 2.2354 | val_acc: 0.2550 | test_acc: 0.2645 | Time: 1.0625 s
>>> Epoch [ 33/100]
train_loss: 2.2338 | train_acc: 0.2591 | val_loss: 2.2340 | val_acc: 0.2549 | test_acc: 0.2645 | Time: 1.0274 s
>>> Epoch [ 34/100]
train_loss: 2.2323 | train_acc: 0.2599 | val_loss: 2.2326 | val_acc: 0.2560 | test_acc: 0.2656 | Time: 1.0614 s
>>> Epoch [ 35/100]
train_loss: 2.2309 | train_acc: 0.2608 | val_loss: 2.2313 | val_acc: 0.2562 | test_acc: 0.2659 | Time: 1.1766 s
>>> Epoch [ 36/100]
train_loss: 2.2295 | train_acc: 0.2613 | val_loss: 2.2300 | val_acc: 0.2571 | test_acc: 0.2670 | Time: 1.1343 s
>>> Epoch [ 37/100]
train_loss: 2.2281 | train_acc: 0.2620 | val_loss: 2.2287 | val_acc: 0.2581 | test_acc: 0.2678 | Time: 1.1239 s
>>> Epoch [ 38/100]
train_loss: 2.2268 | train_acc: 0.2627 | val_loss: 2.2275 | val_acc: 0.2590 | test_acc: 0.2691 | Time: 1.1907 s
>>> Epoch [ 39/100]
train_loss: 2.2255 | train_acc: 0.2632 | val_loss: 2.2263 | val_acc: 0.2597 | test_acc: 0.2700 | Time: 1.0508 s
>>> Epoch [ 40/100]
train_loss: 2.2242 | train_acc: 0.2636 | val_loss: 2.2251 | val_acc: 0.2615 | test_acc: 0.2709 | Time: 1.0951 s
>>> Epoch [ 41/100]
train_loss: 2.2230 | train_acc: 0.2643 | val_loss: 2.2239 | val_acc: 0.2614 | test_acc: 0.2713 | Time: 1.0064 s
>>> Epoch [ 42/100]
train_loss: 2.2218 | train_acc: 0.2653 | val_loss: 2.2228 | val_acc: 0.2624 | test_acc: 0.2726 | Time: 1.0853 s
>>> Epoch [ 43/100]
train_loss: 2.2206 | train_acc: 0.2660 | val_loss: 2.2217 | val_acc: 0.2630 | test_acc: 0.2729 | Time: 1.1437 s
>>> Epoch [ 44/100]
train_loss: 2.2195 | train_acc: 0.2668 | val_loss: 2.2206 | val_acc: 0.2639 | test_acc: 0.2734 | Time: 1.1692 s
>>> Epoch [ 45/100]
train_loss: 2.2183 | train_acc: 0.2675 | val_loss: 2.2195 | val_acc: 0.2646 | test_acc: 0.2743 | Time: 1.2463 s
>>> Epoch [ 46/100]
train_loss: 2.2172 | train_acc: 0.2686 | val_loss: 2.2185 | val_acc: 0.2653 | test_acc: 0.2749 | Time: 0.9890 s
>>> Epoch [ 47/100]
train_loss: 2.2161 | train_acc: 0.2691 | val_loss: 2.2175 | val_acc: 0.2658 | test_acc: 0.2753 | Time: 1.0790 s
>>> Epoch [ 48/100]
train_loss: 2.2151 | train_acc: 0.2700 | val_loss: 2.2165 | val_acc: 0.2670 | test_acc: 0.2764 | Time: 1.1050 s
>>> Epoch [ 49/100]
train_loss: 2.2140 | train_acc: 0.2710 | val_loss: 2.2155 | val_acc: 0.2675 | test_acc: 0.2781 | Time: 1.1416 s
>>> Epoch [ 50/100]
train_loss: 2.2130 | train_acc: 0.2718 | val_loss: 2.2146 | val_acc: 0.2679 | test_acc: 0.2794 | Time: 1.2574 s
>>> Epoch [ 51/100]
train_loss: 2.2120 | train_acc: 0.2721 | val_loss: 2.2137 | val_acc: 0.2685 | test_acc: 0.2802 | Time: 1.1058 s
>>> Epoch [ 52/100]
train_loss: 2.2111 | train_acc: 0.2730 | val_loss: 2.2127 | val_acc: 0.2688 | test_acc: 0.2812 | Time: 1.1560 s
>>> Epoch [ 53/100]
train_loss: 2.2101 | train_acc: 0.2739 | val_loss: 2.2118 | val_acc: 0.2699 | test_acc: 0.2824 | Time: 1.0288 s
>>> Epoch [ 54/100]
train_loss: 2.2092 | train_acc: 0.2746 | val_loss: 2.2110 | val_acc: 0.2708 | test_acc: 0.2836 | Time: 1.1457 s
>>> Epoch [ 55/100]
train_loss: 2.2082 | train_acc: 0.2752 | val_loss: 2.2101 | val_acc: 0.2723 | test_acc: 0.2841 | Time: 1.1459 s
>>> Epoch [ 56/100]
train_loss: 2.2073 | train_acc: 0.2759 | val_loss: 2.2093 | val_acc: 0.2731 | test_acc: 0.2849 | Time: 1.1210 s
>>> Epoch [ 57/100]
train_loss: 2.2064 | train_acc: 0.2770 | val_loss: 2.2084 | val_acc: 0.2736 | test_acc: 0.2862 | Time: 1.2285 s
>>> Epoch [ 58/100]
train_loss: 2.2056 | train_acc: 0.2782 | val_loss: 2.2076 | val_acc: 0.2748 | test_acc: 0.2869 | Time: 1.1777 s
>>> Epoch [ 59/100]
train_loss: 2.2047 | train_acc: 0.2790 | val_loss: 2.2068 | val_acc: 0.2754 | test_acc: 0.2872 | Time: 1.1839 s
>>> Epoch [ 60/100]
train_loss: 2.2039 | train_acc: 0.2796 | val_loss: 2.2060 | val_acc: 0.2762 | test_acc: 0.2880 | Time: 1.1522 s
>>> Epoch [ 61/100]
train_loss: 2.2030 | train_acc: 0.2804 | val_loss: 2.2053 | val_acc: 0.2766 | test_acc: 0.2890 | Time: 1.2377 s
>>> Epoch [ 62/100]
train_loss: 2.2022 | train_acc: 0.2815 | val_loss: 2.2045 | val_acc: 0.2772 | test_acc: 0.2896 | Time: 0.9801 s
>>> Epoch [ 63/100]
train_loss: 2.2014 | train_acc: 0.2823 | val_loss: 2.2037 | val_acc: 0.2782 | test_acc: 0.2902 | Time: 1.1718 s
>>> Epoch [ 64/100]
train_loss: 2.2006 | train_acc: 0.2830 | val_loss: 2.2030 | val_acc: 0.2791 | test_acc: 0.2909 | Time: 1.1680 s
>>> Epoch [ 65/100]
train_loss: 2.1998 | train_acc: 0.2838 | val_loss: 2.2023 | val_acc: 0.2794 | test_acc: 0.2926 | Time: 1.1149 s
>>> Epoch [ 66/100]
train_loss: 2.1991 | train_acc: 0.2847 | val_loss: 2.2016 | val_acc: 0.2808 | test_acc: 0.2926 | Time: 1.1672 s
>>> Epoch [ 67/100]
train_loss: 2.1983 | train_acc: 0.2859 | val_loss: 2.2009 | val_acc: 0.2820 | test_acc: 0.2940 | Time: 1.0830 s
>>> Epoch [ 68/100]
train_loss: 2.1976 | train_acc: 0.2864 | val_loss: 2.2002 | val_acc: 0.2833 | test_acc: 0.2945 | Time: 1.2651 s
>>> Epoch [ 69/100]
train_loss: 2.1968 | train_acc: 0.2873 | val_loss: 2.1995 | val_acc: 0.2845 | test_acc: 0.2953 | Time: 1.1535 s
>>> Epoch [ 70/100]
train_loss: 2.1961 | train_acc: 0.2879 | val_loss: 2.1988 | val_acc: 0.2848 | test_acc: 0.2959 | Time: 1.1822 s
>>> Epoch [ 71/100]
train_loss: 2.1954 | train_acc: 0.2887 | val_loss: 2.1981 | val_acc: 0.2856 | test_acc: 0.2967 | Time: 1.0507 s
>>> Epoch [ 72/100]
train_loss: 2.1946 | train_acc: 0.2894 | val_loss: 2.1975 | val_acc: 0.2867 | test_acc: 0.2975 | Time: 1.1600 s
>>> Epoch [ 73/100]
train_loss: 2.1939 | train_acc: 0.2902 | val_loss: 2.1968 | val_acc: 0.2870 | test_acc: 0.2986 | Time: 1.2506 s
>>> Epoch [ 74/100]
train_loss: 2.1932 | train_acc: 0.2912 | val_loss: 2.1962 | val_acc: 0.2875 | test_acc: 0.2991 | Time: 1.2234 s
>>> Epoch [ 75/100]
train_loss: 2.1926 | train_acc: 0.2919 | val_loss: 2.1956 | val_acc: 0.2879 | test_acc: 0.2997 | Time: 1.1646 s
>>> Epoch [ 76/100]
train_loss: 2.1919 | train_acc: 0.2931 | val_loss: 2.1949 | val_acc: 0.2891 | test_acc: 0.2998 | Time: 1.1760 s
>>> Epoch [ 77/100]
train_loss: 2.1912 | train_acc: 0.2940 | val_loss: 2.1943 | val_acc: 0.2903 | test_acc: 0.3004 | Time: 1.1614 s
>>> Epoch [ 78/100]
train_loss: 2.1905 | train_acc: 0.2950 | val_loss: 2.1937 | val_acc: 0.2907 | test_acc: 0.3014 | Time: 1.2051 s
>>> Epoch [ 79/100]
train_loss: 2.1899 | train_acc: 0.2962 | val_loss: 2.1931 | val_acc: 0.2914 | test_acc: 0.3021 | Time: 1.2181 s
>>> Epoch [ 80/100]
train_loss: 2.1892 | train_acc: 0.2968 | val_loss: 2.1925 | val_acc: 0.2923 | test_acc: 0.3028 | Time: 1.0881 s
>>> Epoch [ 81/100]
train_loss: 2.1886 | train_acc: 0.2976 | val_loss: 2.1919 | val_acc: 0.2931 | test_acc: 0.3033 | Time: 1.2188 s
>>> Epoch [ 82/100]
train_loss: 2.1879 | train_acc: 0.2982 | val_loss: 2.1913 | val_acc: 0.2932 | test_acc: 0.3038 | Time: 1.1832 s
>>> Epoch [ 83/100]
train_loss: 2.1873 | train_acc: 0.2991 | val_loss: 2.1908 | val_acc: 0.2939 | test_acc: 0.3046 | Time: 1.2089 s
>>> Epoch [ 84/100]
train_loss: 2.1867 | train_acc: 0.3000 | val_loss: 2.1902 | val_acc: 0.2945 | test_acc: 0.3049 | Time: 1.1841 s
>>> Epoch [ 85/100]
train_loss: 2.1861 | train_acc: 0.3004 | val_loss: 2.1896 | val_acc: 0.2952 | test_acc: 0.3056 | Time: 1.1517 s
>>> Epoch [ 86/100]
train_loss: 2.1855 | train_acc: 0.3007 | val_loss: 2.1891 | val_acc: 0.2960 | test_acc: 0.3060 | Time: 1.2523 s
>>> Epoch [ 87/100]
train_loss: 2.1849 | train_acc: 0.3014 | val_loss: 2.1885 | val_acc: 0.2969 | test_acc: 0.3064 | Time: 1.2562 s
>>> Epoch [ 88/100]
train_loss: 2.1843 | train_acc: 0.3022 | val_loss: 2.1880 | val_acc: 0.2978 | test_acc: 0.3078 | Time: 1.2230 s
>>> Epoch [ 89/100]
train_loss: 2.1837 | train_acc: 0.3028 | val_loss: 2.1875 | val_acc: 0.2977 | test_acc: 0.3081 | Time: 1.0554 s
>>> Epoch [ 90/100]
train_loss: 2.1831 | train_acc: 0.3036 | val_loss: 2.1869 | val_acc: 0.2982 | test_acc: 0.3087 | Time: 1.1358 s
>>> Epoch [ 91/100]
train_loss: 2.1825 | train_acc: 0.3040 | val_loss: 2.1864 | val_acc: 0.2990 | test_acc: 0.3092 | Time: 1.1976 s
>>> Epoch [ 92/100]
train_loss: 2.1819 | train_acc: 0.3047 | val_loss: 2.1859 | val_acc: 0.2999 | test_acc: 0.3095 | Time: 1.3097 s
>>> Epoch [ 93/100]
train_loss: 2.1814 | train_acc: 0.3052 | val_loss: 2.1854 | val_acc: 0.3008 | test_acc: 0.3104 | Time: 1.2280 s
>>> Epoch [ 94/100]
train_loss: 2.1808 | train_acc: 0.3058 | val_loss: 2.1849 | val_acc: 0.3011 | test_acc: 0.3109 | Time: 1.1215 s
>>> Epoch [ 95/100]
train_loss: 2.1802 | train_acc: 0.3064 | val_loss: 2.1844 | val_acc: 0.3012 | test_acc: 0.3116 | Time: 1.1530 s
>>> Epoch [ 96/100]
train_loss: 2.1797 | train_acc: 0.3069 | val_loss: 2.1839 | val_acc: 0.3017 | test_acc: 0.3117 | Time: 1.2460 s
>>> Epoch [ 97/100]
train_loss: 2.1791 | train_acc: 0.3072 | val_loss: 2.1834 | val_acc: 0.3023 | test_acc: 0.3120 | Time: 1.2508 s
>>> Epoch [ 98/100]
train_loss: 2.1786 | train_acc: 0.3081 | val_loss: 2.1829 | val_acc: 0.3031 | test_acc: 0.3118 | Time: 1.2506 s
>>> Epoch [ 99/100]
train_loss: 2.1781 | train_acc: 0.3087 | val_loss: 2.1824 | val_acc: 0.3039 | test_acc: 0.3126 | Time: 1.2872 s
Fit completed in 1.9200 seconds
Saving the graph to ../logs/cifar10/101_tf_linear_raw_gradtest/exp_no_001/linear_classifier.pb
INFO:tensorflow:Froze 2 variables.
Converted 2 variables to const ops.
11 ops in the final graph.
Length of train loss          : 0
Length of train accuracy      : 0
Length of validate loss       : 0
Length of validate accuracy   : 0
Length of test accuracy       : 0
Restoring training from epoch : 0
>>> Epoch [  0/100]
train_loss: 2.3026 | train_acc: 0.1001 | val_loss: 2.3006 | val_acc: 0.2475 | test_acc: 0.2555 | Time: 2.8077 s
>>> Epoch [  1/100]
train_loss: 2.3006 | train_acc: 0.2477 | val_loss: 2.2985 | val_acc: 0.2464 | test_acc: 0.2551 | Time: 1.9459 s
>>> Epoch [  2/100]
train_loss: 2.2986 | train_acc: 0.2469 | val_loss: 2.2964 | val_acc: 0.2454 | test_acc: 0.2538 | Time: 1.7694 s
>>> Epoch [  3/100]
train_loss: 2.2965 | train_acc: 0.2464 | val_loss: 2.2941 | val_acc: 0.2455 | test_acc: 0.2532 | Time: 2.0732 s
>>> Epoch [  4/100]
train_loss: 2.2943 | train_acc: 0.2455 | val_loss: 2.2918 | val_acc: 0.2442 | test_acc: 0.2523 | Time: 2.1660 s
>>> Epoch [  5/100]
train_loss: 2.2920 | train_acc: 0.2445 | val_loss: 2.2894 | val_acc: 0.2443 | test_acc: 0.2501 | Time: 2.9523 s
>>> Epoch [  6/100]
train_loss: 2.2897 | train_acc: 0.2441 | val_loss: 2.2868 | val_acc: 0.2431 | test_acc: 0.2493 | Time: 2.8660 s
>>> Epoch [  7/100]
train_loss: 2.2872 | train_acc: 0.2435 | val_loss: 2.2842 | val_acc: 0.2428 | test_acc: 0.2482 | Time: 3.2576 s
>>> Epoch [  8/100]
train_loss: 2.2847 | train_acc: 0.2425 | val_loss: 2.2815 | val_acc: 0.2414 | test_acc: 0.2479 | Time: 2.6699 s
>>> Epoch [  9/100]
train_loss: 2.2821 | train_acc: 0.2420 | val_loss: 2.2788 | val_acc: 0.2409 | test_acc: 0.2465 | Time: 2.6433 s
>>> Epoch [ 10/100]
train_loss: 2.2795 | train_acc: 0.2412 | val_loss: 2.2762 | val_acc: 0.2405 | test_acc: 0.2463 | Time: 2.8738 s
>>> Epoch [ 11/100]
train_loss: 2.2769 | train_acc: 0.2412 | val_loss: 2.2735 | val_acc: 0.2401 | test_acc: 0.2462 | Time: 2.6856 s
>>> Epoch [ 12/100]
train_loss: 2.2743 | train_acc: 0.2408 | val_loss: 2.2710 | val_acc: 0.2404 | test_acc: 0.2465 | Time: 1.6913 s
>>> Epoch [ 13/100]
train_loss: 2.2718 | train_acc: 0.2408 | val_loss: 2.2685 | val_acc: 0.2402 | test_acc: 0.2465 | Time: 1.6979 s
>>> Epoch [ 14/100]
train_loss: 2.2694 | train_acc: 0.2415 | val_loss: 2.2662 | val_acc: 0.2410 | test_acc: 0.2478 | Time: 1.7263 s
>>> Epoch [ 15/100]
train_loss: 2.2670 | train_acc: 0.2418 | val_loss: 2.2639 | val_acc: 0.2407 | test_acc: 0.2483 | Time: 1.7434 s
>>> Epoch [ 16/100]
train_loss: 2.2647 | train_acc: 0.2425 | val_loss: 2.2617 | val_acc: 0.2411 | test_acc: 0.2494 | Time: 1.8301 s
>>> Epoch [ 17/100]
train_loss: 2.2625 | train_acc: 0.2431 | val_loss: 2.2596 | val_acc: 0.2416 | test_acc: 0.2503 | Time: 1.8077 s
>>> Epoch [ 18/100]
train_loss: 2.2604 | train_acc: 0.2438 | val_loss: 2.2576 | val_acc: 0.2431 | test_acc: 0.2511 | Time: 1.7891 s
>>> Epoch [ 19/100]
train_loss: 2.2584 | train_acc: 0.2444 | val_loss: 2.2557 | val_acc: 0.2439 | test_acc: 0.2522 | Time: 1.9053 s
>>> Epoch [ 20/100]
train_loss: 2.2564 | train_acc: 0.2450 | val_loss: 2.2538 | val_acc: 0.2452 | test_acc: 0.2520 | Time: 1.8254 s
>>> Epoch [ 21/100]
train_loss: 2.2544 | train_acc: 0.2461 | val_loss: 2.2519 | val_acc: 0.2455 | test_acc: 0.2536 | Time: 1.8306 s
>>> Epoch [ 22/100]
train_loss: 2.2525 | train_acc: 0.2473 | val_loss: 2.2501 | val_acc: 0.2461 | test_acc: 0.2547 | Time: 1.8310 s
>>> Epoch [ 23/100]
train_loss: 2.2506 | train_acc: 0.2484 | val_loss: 2.2484 | val_acc: 0.2469 | test_acc: 0.2547 | Time: 1.8598 s
>>> Epoch [ 24/100]
train_loss: 2.2488 | train_acc: 0.2496 | val_loss: 2.2467 | val_acc: 0.2475 | test_acc: 0.2545 | Time: 1.9616 s
>>> Epoch [ 25/100]
train_loss: 2.2470 | train_acc: 0.2504 | val_loss: 2.2450 | val_acc: 0.2486 | test_acc: 0.2555 | Time: 1.8955 s
>>> Epoch [ 26/100]
train_loss: 2.2452 | train_acc: 0.2514 | val_loss: 2.2434 | val_acc: 0.2493 | test_acc: 0.2567 | Time: 1.8321 s
>>> Epoch [ 27/100]
train_loss: 2.2435 | train_acc: 0.2525 | val_loss: 2.2418 | val_acc: 0.2499 | test_acc: 0.2575 | Time: 1.8117 s
>>> Epoch [ 28/100]
train_loss: 2.2418 | train_acc: 0.2532 | val_loss: 2.2402 | val_acc: 0.2502 | test_acc: 0.2581 | Time: 1.8038 s
>>> Epoch [ 29/100]
train_loss: 2.2402 | train_acc: 0.2538 | val_loss: 2.2387 | val_acc: 0.2510 | test_acc: 0.2587 | Time: 1.7913 s
>>> Epoch [ 30/100]
train_loss: 2.2386 | train_acc: 0.2548 | val_loss: 2.2372 | val_acc: 0.2522 | test_acc: 0.2595 | Time: 1.9766 s
>>> Epoch [ 31/100]
train_loss: 2.2370 | train_acc: 0.2559 | val_loss: 2.2358 | val_acc: 0.2527 | test_acc: 0.2599 | Time: 1.8921 s
>>> Epoch [ 32/100]
train_loss: 2.2355 | train_acc: 0.2565 | val_loss: 2.2343 | val_acc: 0.2538 | test_acc: 0.2607 | Time: 1.8523 s
>>> Epoch [ 33/100]
train_loss: 2.2340 | train_acc: 0.2574 | val_loss: 2.2330 | val_acc: 0.2549 | test_acc: 0.2614 | Time: 1.9054 s
>>> Epoch [ 34/100]
train_loss: 2.2325 | train_acc: 0.2580 | val_loss: 2.2316 | val_acc: 0.2556 | test_acc: 0.2627 | Time: 1.7956 s
>>> Epoch [ 35/100]
train_loss: 2.2311 | train_acc: 0.2590 | val_loss: 2.2303 | val_acc: 0.2564 | test_acc: 0.2628 | Time: 1.8282 s
>>> Epoch [ 36/100]
train_loss: 2.2297 | train_acc: 0.2594 | val_loss: 2.2291 | val_acc: 0.2569 | test_acc: 0.2631 | Time: 1.9723 s
>>> Epoch [ 37/100]
train_loss: 2.2283 | train_acc: 0.2603 | val_loss: 2.2278 | val_acc: 0.2571 | test_acc: 0.2645 | Time: 1.9148 s
>>> Epoch [ 38/100]
train_loss: 2.2270 | train_acc: 0.2609 | val_loss: 2.2266 | val_acc: 0.2581 | test_acc: 0.2655 | Time: 1.8245 s
>>> Epoch [ 39/100]
train_loss: 2.2257 | train_acc: 0.2615 | val_loss: 2.2254 | val_acc: 0.2591 | test_acc: 0.2664 | Time: 1.8235 s
>>> Epoch [ 40/100]
train_loss: 2.2244 | train_acc: 0.2622 | val_loss: 2.2243 | val_acc: 0.2606 | test_acc: 0.2671 | Time: 1.8826 s
>>> Epoch [ 41/100]
train_loss: 2.2231 | train_acc: 0.2629 | val_loss: 2.2232 | val_acc: 0.2611 | test_acc: 0.2680 | Time: 1.9476 s
>>> Epoch [ 42/100]
train_loss: 2.2219 | train_acc: 0.2634 | val_loss: 2.2221 | val_acc: 0.2616 | test_acc: 0.2688 | Time: 1.8969 s
>>> Epoch [ 43/100]
train_loss: 2.2208 | train_acc: 0.2640 | val_loss: 2.2210 | val_acc: 0.2621 | test_acc: 0.2699 | Time: 1.9116 s
>>> Epoch [ 44/100]
train_loss: 2.2196 | train_acc: 0.2647 | val_loss: 2.2200 | val_acc: 0.2629 | test_acc: 0.2704 | Time: 1.8891 s
>>> Epoch [ 45/100]
train_loss: 2.2185 | train_acc: 0.2655 | val_loss: 2.2189 | val_acc: 0.2636 | test_acc: 0.2707 | Time: 1.9812 s
>>> Epoch [ 46/100]
train_loss: 2.2174 | train_acc: 0.2659 | val_loss: 2.2179 | val_acc: 0.2641 | test_acc: 0.2719 | Time: 1.9784 s
>>> Epoch [ 47/100]
train_loss: 2.2163 | train_acc: 0.2664 | val_loss: 2.2170 | val_acc: 0.2646 | test_acc: 0.2725 | Time: 1.9448 s
>>> Epoch [ 48/100]
train_loss: 2.2152 | train_acc: 0.2669 | val_loss: 2.2160 | val_acc: 0.2660 | test_acc: 0.2740 | Time: 3.6994 s
>>> Epoch [ 49/100]
train_loss: 2.2142 | train_acc: 0.2681 | val_loss: 2.2151 | val_acc: 0.2667 | test_acc: 0.2756 | Time: 3.5693 s
>>> Epoch [ 50/100]
train_loss: 2.2132 | train_acc: 0.2689 | val_loss: 2.2142 | val_acc: 0.2670 | test_acc: 0.2760 | Time: 3.3911 s
>>> Epoch [ 51/100]
train_loss: 2.2122 | train_acc: 0.2695 | val_loss: 2.2133 | val_acc: 0.2672 | test_acc: 0.2768 | Time: 3.2983 s
>>> Epoch [ 52/100]
train_loss: 2.2113 | train_acc: 0.2703 | val_loss: 2.2124 | val_acc: 0.2670 | test_acc: 0.2774 | Time: 3.2567 s
>>> Epoch [ 53/100]
train_loss: 2.2103 | train_acc: 0.2710 | val_loss: 2.2115 | val_acc: 0.2675 | test_acc: 0.2779 | Time: 2.9803 s
>>> Epoch [ 54/100]
train_loss: 2.2094 | train_acc: 0.2714 | val_loss: 2.2107 | val_acc: 0.2680 | test_acc: 0.2784 | Time: 1.7539 s
>>> Epoch [ 55/100]
train_loss: 2.2085 | train_acc: 0.2723 | val_loss: 2.2099 | val_acc: 0.2686 | test_acc: 0.2786 | Time: 1.7167 s
>>> Epoch [ 56/100]
train_loss: 2.2076 | train_acc: 0.2731 | val_loss: 2.2091 | val_acc: 0.2690 | test_acc: 0.2797 | Time: 1.7098 s
>>> Epoch [ 57/100]
train_loss: 2.2067 | train_acc: 0.2738 | val_loss: 2.2083 | val_acc: 0.2702 | test_acc: 0.2798 | Time: 1.9008 s
>>> Epoch [ 58/100]
train_loss: 2.2059 | train_acc: 0.2745 | val_loss: 2.2075 | val_acc: 0.2708 | test_acc: 0.2807 | Time: 1.9523 s
>>> Epoch [ 59/100]
train_loss: 2.2050 | train_acc: 0.2752 | val_loss: 2.2067 | val_acc: 0.2717 | test_acc: 0.2815 | Time: 1.9458 s
>>> Epoch [ 60/100]
train_loss: 2.2042 | train_acc: 0.2759 | val_loss: 2.2060 | val_acc: 0.2725 | test_acc: 0.2821 | Time: 1.9581 s
>>> Epoch [ 61/100]
train_loss: 2.2034 | train_acc: 0.2766 | val_loss: 2.2052 | val_acc: 0.2734 | test_acc: 0.2828 | Time: 1.9468 s
>>> Epoch [ 62/100]
train_loss: 2.2026 | train_acc: 0.2772 | val_loss: 2.2045 | val_acc: 0.2736 | test_acc: 0.2836 | Time: 1.9977 s
>>> Epoch [ 63/100]
train_loss: 2.2018 | train_acc: 0.2777 | val_loss: 2.2038 | val_acc: 0.2744 | test_acc: 0.2843 | Time: 1.9173 s
>>> Epoch [ 64/100]
train_loss: 2.2010 | train_acc: 0.2783 | val_loss: 2.2031 | val_acc: 0.2752 | test_acc: 0.2850 | Time: 2.0327 s
>>> Epoch [ 65/100]
train_loss: 2.2003 | train_acc: 0.2792 | val_loss: 2.2024 | val_acc: 0.2762 | test_acc: 0.2853 | Time: 2.0495 s
>>> Epoch [ 66/100]
train_loss: 2.1995 | train_acc: 0.2801 | val_loss: 2.2017 | val_acc: 0.2778 | test_acc: 0.2858 | Time: 1.8817 s
>>> Epoch [ 67/100]
train_loss: 2.1988 | train_acc: 0.2812 | val_loss: 2.2010 | val_acc: 0.2789 | test_acc: 0.2860 | Time: 2.0028 s
>>> Epoch [ 68/100]
train_loss: 2.1981 | train_acc: 0.2821 | val_loss: 2.2004 | val_acc: 0.2798 | test_acc: 0.2870 | Time: 2.0314 s
>>> Epoch [ 69/100]
train_loss: 2.1974 | train_acc: 0.2825 | val_loss: 2.1997 | val_acc: 0.2803 | test_acc: 0.2879 | Time: 1.9839 s
>>> Epoch [ 70/100]
train_loss: 2.1967 | train_acc: 0.2834 | val_loss: 2.1990 | val_acc: 0.2812 | test_acc: 0.2893 | Time: 1.9674 s
>>> Epoch [ 71/100]
train_loss: 2.1960 | train_acc: 0.2844 | val_loss: 2.1984 | val_acc: 0.2821 | test_acc: 0.2899 | Time: 1.9894 s
>>> Epoch [ 72/100]
train_loss: 2.1953 | train_acc: 0.2852 | val_loss: 2.1978 | val_acc: 0.2831 | test_acc: 0.2905 | Time: 2.1153 s
>>> Epoch [ 73/100]
train_loss: 2.1946 | train_acc: 0.2859 | val_loss: 2.1971 | val_acc: 0.2838 | test_acc: 0.2906 | Time: 2.0805 s
>>> Epoch [ 74/100]
train_loss: 2.1939 | train_acc: 0.2870 | val_loss: 2.1965 | val_acc: 0.2849 | test_acc: 0.2916 | Time: 1.9522 s
>>> Epoch [ 75/100]
train_loss: 2.1933 | train_acc: 0.2871 | val_loss: 2.1959 | val_acc: 0.2854 | test_acc: 0.2920 | Time: 1.9563 s
>>> Epoch [ 76/100]
train_loss: 2.1926 | train_acc: 0.2876 | val_loss: 2.1953 | val_acc: 0.2866 | test_acc: 0.2931 | Time: 2.0605 s
>>> Epoch [ 77/100]
train_loss: 2.1920 | train_acc: 0.2883 | val_loss: 2.1947 | val_acc: 0.2873 | test_acc: 0.2945 | Time: 1.9883 s
>>> Epoch [ 78/100]
train_loss: 2.1914 | train_acc: 0.2893 | val_loss: 2.1941 | val_acc: 0.2885 | test_acc: 0.2951 | Time: 1.9232 s
>>> Epoch [ 79/100]
train_loss: 2.1907 | train_acc: 0.2901 | val_loss: 2.1935 | val_acc: 0.2892 | test_acc: 0.2955 | Time: 2.2328 s
>>> Epoch [ 80/100]
train_loss: 2.1901 | train_acc: 0.2911 | val_loss: 2.1929 | val_acc: 0.2899 | test_acc: 0.2958 | Time: 2.2023 s
>>> Epoch [ 81/100]
train_loss: 2.1895 | train_acc: 0.2920 | val_loss: 2.1923 | val_acc: 0.2904 | test_acc: 0.2967 | Time: 2.2211 s
>>> Epoch [ 82/100]
train_loss: 2.1889 | train_acc: 0.2927 | val_loss: 2.1917 | val_acc: 0.2911 | test_acc: 0.2979 | Time: 2.1493 s
>>> Epoch [ 83/100]
train_loss: 2.1883 | train_acc: 0.2937 | val_loss: 2.1911 | val_acc: 0.2915 | test_acc: 0.2984 | Time: 2.2481 s
>>> Epoch [ 84/100]
train_loss: 2.1877 | train_acc: 0.2945 | val_loss: 2.1906 | val_acc: 0.2923 | test_acc: 0.2995 | Time: 2.1932 s
>>> Epoch [ 85/100]
train_loss: 2.1871 | train_acc: 0.2955 | val_loss: 2.1900 | val_acc: 0.2933 | test_acc: 0.3005 | Time: 2.2107 s
>>> Epoch [ 86/100]
train_loss: 2.1865 | train_acc: 0.2962 | val_loss: 2.1895 | val_acc: 0.2939 | test_acc: 0.3010 | Time: 2.1728 s
>>> Epoch [ 87/100]
train_loss: 2.1860 | train_acc: 0.2967 | val_loss: 2.1889 | val_acc: 0.2945 | test_acc: 0.3014 | Time: 2.2469 s
>>> Epoch [ 88/100]
train_loss: 2.1854 | train_acc: 0.2974 | val_loss: 2.1884 | val_acc: 0.2946 | test_acc: 0.3019 | Time: 2.0795 s
>>> Epoch [ 89/100]
train_loss: 2.1848 | train_acc: 0.2981 | val_loss: 2.1878 | val_acc: 0.2947 | test_acc: 0.3029 | Time: 2.1670 s
>>> Epoch [ 90/100]
train_loss: 2.1843 | train_acc: 0.2991 | val_loss: 2.1873 | val_acc: 0.2951 | test_acc: 0.3036 | Time: 2.2085 s
>>> Epoch [ 91/100]
train_loss: 2.1837 | train_acc: 0.2995 | val_loss: 2.1867 | val_acc: 0.2956 | test_acc: 0.3042 | Time: 2.3149 s
>>> Epoch [ 92/100]
train_loss: 2.1832 | train_acc: 0.3002 | val_loss: 2.1862 | val_acc: 0.2965 | test_acc: 0.3052 | Time: 2.0705 s
>>> Epoch [ 93/100]
train_loss: 2.1826 | train_acc: 0.3009 | val_loss: 2.1857 | val_acc: 0.2967 | test_acc: 0.3059 | Time: 2.2567 s
>>> Epoch [ 94/100]
train_loss: 2.1821 | train_acc: 0.3018 | val_loss: 2.1851 | val_acc: 0.2970 | test_acc: 0.3064 | Time: 2.3393 s
>>> Epoch [ 95/100]
train_loss: 2.1815 | train_acc: 0.3026 | val_loss: 2.1846 | val_acc: 0.2982 | test_acc: 0.3069 | Time: 2.2888 s
>>> Epoch [ 96/100]
train_loss: 2.1810 | train_acc: 0.3030 | val_loss: 2.1841 | val_acc: 0.2987 | test_acc: 0.3077 | Time: 2.3250 s
>>> Epoch [ 97/100]
train_loss: 2.1805 | train_acc: 0.3036 | val_loss: 2.1836 | val_acc: 0.2998 | test_acc: 0.3088 | Time: 2.2905 s
>>> Epoch [ 98/100]
train_loss: 2.1800 | train_acc: 0.3042 | val_loss: 2.1831 | val_acc: 0.2998 | test_acc: 0.3092 | Time: 2.2234 s
>>> Epoch [ 99/100]
train_loss: 2.1795 | train_acc: 0.3049 | val_loss: 2.1826 | val_acc: 0.3004 | test_acc: 0.3094 | Time: 2.3749 s
Fit completed in 3.2939 seconds
Saving the graph to ../logs/cifar10/101_tf_linear_raw_gradtest/exp_no_002/linear_classifier.pb
INFO:tensorflow:Froze 2 variables.
Converted 2 variables to const ops.
11 ops in the final graph.
Length of train loss          : 0
Length of train accuracy      : 0
Length of validate loss       : 0
Length of validate accuracy   : 0
Length of test accuracy       : 0
Restoring training from epoch : 0
>>> Epoch [  0/100]
train_loss: 2.3026 | train_acc: 0.0977 | val_loss: 2.3006 | val_acc: 0.2520 | test_acc: 0.2568 | Time: 6.0613 s
>>> Epoch [  1/100]
train_loss: 2.3006 | train_acc: 0.2488 | val_loss: 2.2986 | val_acc: 0.2512 | test_acc: 0.2566 | Time: 4.2678 s
>>> Epoch [  2/100]
train_loss: 2.2986 | train_acc: 0.2478 | val_loss: 2.2965 | val_acc: 0.2507 | test_acc: 0.2562 | Time: 4.2463 s
>>> Epoch [  3/100]
train_loss: 2.2965 | train_acc: 0.2469 | val_loss: 2.2943 | val_acc: 0.2510 | test_acc: 0.2553 | Time: 4.2823 s
>>> Epoch [  4/100]
train_loss: 2.2942 | train_acc: 0.2463 | val_loss: 2.2920 | val_acc: 0.2495 | test_acc: 0.2544 | Time: 4.3966 s
>>> Epoch [  5/100]
train_loss: 2.2919 | train_acc: 0.2456 | val_loss: 2.2895 | val_acc: 0.2490 | test_acc: 0.2528 | Time: 4.3405 s
>>> Epoch [  6/100]
train_loss: 2.2895 | train_acc: 0.2450 | val_loss: 2.2870 | val_acc: 0.2479 | test_acc: 0.2499 | Time: 4.4152 s
>>> Epoch [  7/100]
train_loss: 2.2870 | train_acc: 0.2439 | val_loss: 2.2845 | val_acc: 0.2470 | test_acc: 0.2492 | Time: 4.1549 s
>>> Epoch [  8/100]
train_loss: 2.2845 | train_acc: 0.2428 | val_loss: 2.2818 | val_acc: 0.2466 | test_acc: 0.2477 | Time: 4.3872 s
>>> Epoch [  9/100]
train_loss: 2.2819 | train_acc: 0.2420 | val_loss: 2.2792 | val_acc: 0.2460 | test_acc: 0.2473 | Time: 4.2587 s
>>> Epoch [ 10/100]
train_loss: 2.2792 | train_acc: 0.2417 | val_loss: 2.2765 | val_acc: 0.2456 | test_acc: 0.2467 | Time: 4.2014 s
>>> Epoch [ 11/100]
train_loss: 2.2766 | train_acc: 0.2415 | val_loss: 2.2739 | val_acc: 0.2451 | test_acc: 0.2460 | Time: 4.3761 s
>>> Epoch [ 12/100]
train_loss: 2.2741 | train_acc: 0.2413 | val_loss: 2.2714 | val_acc: 0.2456 | test_acc: 0.2459 | Time: 4.3703 s
>>> Epoch [ 13/100]
train_loss: 2.2716 | train_acc: 0.2413 | val_loss: 2.2690 | val_acc: 0.2454 | test_acc: 0.2460 | Time: 4.4042 s
>>> Epoch [ 14/100]
train_loss: 2.2692 | train_acc: 0.2417 | val_loss: 2.2666 | val_acc: 0.2456 | test_acc: 0.2479 | Time: 4.3114 s
>>> Epoch [ 15/100]
train_loss: 2.2669 | train_acc: 0.2426 | val_loss: 2.2644 | val_acc: 0.2465 | test_acc: 0.2492 | Time: 4.1508 s
>>> Epoch [ 16/100]
train_loss: 2.2646 | train_acc: 0.2432 | val_loss: 2.2622 | val_acc: 0.2465 | test_acc: 0.2507 | Time: 4.3895 s
>>> Epoch [ 17/100]
train_loss: 2.2625 | train_acc: 0.2440 | val_loss: 2.2601 | val_acc: 0.2472 | test_acc: 0.2513 | Time: 4.2951 s
>>> Epoch [ 18/100]
train_loss: 2.2604 | train_acc: 0.2448 | val_loss: 2.2581 | val_acc: 0.2477 | test_acc: 0.2525 | Time: 4.4428 s
>>> Epoch [ 19/100]
train_loss: 2.2583 | train_acc: 0.2458 | val_loss: 2.2561 | val_acc: 0.2478 | test_acc: 0.2530 | Time: 4.5678 s
>>> Epoch [ 20/100]
train_loss: 2.2563 | train_acc: 0.2466 | val_loss: 2.2542 | val_acc: 0.2484 | test_acc: 0.2538 | Time: 3.9561 s
>>> Epoch [ 21/100]
train_loss: 2.2544 | train_acc: 0.2477 | val_loss: 2.2524 | val_acc: 0.2487 | test_acc: 0.2547 | Time: 4.6491 s
>>> Epoch [ 22/100]
train_loss: 2.2525 | train_acc: 0.2482 | val_loss: 2.2506 | val_acc: 0.2496 | test_acc: 0.2550 | Time: 4.4170 s
>>> Epoch [ 23/100]
train_loss: 2.2507 | train_acc: 0.2495 | val_loss: 2.2488 | val_acc: 0.2505 | test_acc: 0.2561 | Time: 4.2157 s
>>> Epoch [ 24/100]
train_loss: 2.2489 | train_acc: 0.2506 | val_loss: 2.2471 | val_acc: 0.2509 | test_acc: 0.2563 | Time: 4.5238 s
>>> Epoch [ 25/100]
train_loss: 2.2471 | train_acc: 0.2516 | val_loss: 2.2454 | val_acc: 0.2519 | test_acc: 0.2573 | Time: 5.1642 s
>>> Epoch [ 26/100]
train_loss: 2.2453 | train_acc: 0.2525 | val_loss: 2.2437 | val_acc: 0.2522 | test_acc: 0.2586 | Time: 7.2032 s
>>> Epoch [ 27/100]
train_loss: 2.2436 | train_acc: 0.2533 | val_loss: 2.2421 | val_acc: 0.2528 | test_acc: 0.2598 | Time: 6.8526 s
>>> Epoch [ 28/100]
train_loss: 2.2420 | train_acc: 0.2542 | val_loss: 2.2405 | val_acc: 0.2539 | test_acc: 0.2599 | Time: 6.4149 s
>>> Epoch [ 29/100]
train_loss: 2.2404 | train_acc: 0.2548 | val_loss: 2.2390 | val_acc: 0.2541 | test_acc: 0.2611 | Time: 7.2802 s
>>> Epoch [ 30/100]
train_loss: 2.2388 | train_acc: 0.2557 | val_loss: 2.2375 | val_acc: 0.2541 | test_acc: 0.2608 | Time: 6.9571 s
>>> Epoch [ 31/100]
train_loss: 2.2372 | train_acc: 0.2569 | val_loss: 2.2360 | val_acc: 0.2545 | test_acc: 0.2618 | Time: 4.2662 s
>>> Epoch [ 32/100]
train_loss: 2.2357 | train_acc: 0.2578 | val_loss: 2.2346 | val_acc: 0.2551 | test_acc: 0.2618 | Time: 3.6256 s
>>> Epoch [ 33/100]
train_loss: 2.2342 | train_acc: 0.2583 | val_loss: 2.2332 | val_acc: 0.2557 | test_acc: 0.2626 | Time: 3.6466 s
>>> Epoch [ 34/100]
train_loss: 2.2328 | train_acc: 0.2587 | val_loss: 2.2318 | val_acc: 0.2560 | test_acc: 0.2640 | Time: 4.1530 s
>>> Epoch [ 35/100]
train_loss: 2.2313 | train_acc: 0.2596 | val_loss: 2.2305 | val_acc: 0.2575 | test_acc: 0.2647 | Time: 4.0267 s
>>> Epoch [ 36/100]
train_loss: 2.2300 | train_acc: 0.2607 | val_loss: 2.2292 | val_acc: 0.2579 | test_acc: 0.2656 | Time: 4.0698 s
>>> Epoch [ 37/100]
train_loss: 2.2286 | train_acc: 0.2616 | val_loss: 2.2279 | val_acc: 0.2583 | test_acc: 0.2663 | Time: 4.5524 s
>>> Epoch [ 38/100]
train_loss: 2.2273 | train_acc: 0.2623 | val_loss: 2.2266 | val_acc: 0.2592 | test_acc: 0.2670 | Time: 4.1494 s
>>> Epoch [ 39/100]
train_loss: 2.2260 | train_acc: 0.2630 | val_loss: 2.2254 | val_acc: 0.2598 | test_acc: 0.2678 | Time: 4.3762 s
>>> Epoch [ 40/100]
train_loss: 2.2247 | train_acc: 0.2638 | val_loss: 2.2242 | val_acc: 0.2601 | test_acc: 0.2687 | Time: 4.5450 s
>>> Epoch [ 41/100]
train_loss: 2.2235 | train_acc: 0.2648 | val_loss: 2.2230 | val_acc: 0.2602 | test_acc: 0.2691 | Time: 4.2998 s
>>> Epoch [ 42/100]
train_loss: 2.2223 | train_acc: 0.2656 | val_loss: 2.2219 | val_acc: 0.2615 | test_acc: 0.2702 | Time: 4.2752 s
>>> Epoch [ 43/100]
train_loss: 2.2212 | train_acc: 0.2662 | val_loss: 2.2208 | val_acc: 0.2620 | test_acc: 0.2716 | Time: 4.3378 s
>>> Epoch [ 44/100]
train_loss: 2.2200 | train_acc: 0.2668 | val_loss: 2.2197 | val_acc: 0.2626 | test_acc: 0.2721 | Time: 4.2676 s
>>> Epoch [ 45/100]
train_loss: 2.2189 | train_acc: 0.2673 | val_loss: 2.2186 | val_acc: 0.2635 | test_acc: 0.2728 | Time: 4.4133 s
>>> Epoch [ 46/100]
train_loss: 2.2178 | train_acc: 0.2682 | val_loss: 2.2176 | val_acc: 0.2645 | test_acc: 0.2732 | Time: 4.4138 s
>>> Epoch [ 47/100]
train_loss: 2.2167 | train_acc: 0.2689 | val_loss: 2.2165 | val_acc: 0.2654 | test_acc: 0.2738 | Time: 4.0733 s
>>> Epoch [ 48/100]
train_loss: 2.2157 | train_acc: 0.2697 | val_loss: 2.2155 | val_acc: 0.2666 | test_acc: 0.2751 | Time: 4.5186 s
>>> Epoch [ 49/100]
train_loss: 2.2146 | train_acc: 0.2704 | val_loss: 2.2146 | val_acc: 0.2673 | test_acc: 0.2758 | Time: 4.5769 s
>>> Epoch [ 50/100]
train_loss: 2.2136 | train_acc: 0.2711 | val_loss: 2.2136 | val_acc: 0.2681 | test_acc: 0.2775 | Time: 4.2931 s
>>> Epoch [ 51/100]
train_loss: 2.2126 | train_acc: 0.2719 | val_loss: 2.2127 | val_acc: 0.2689 | test_acc: 0.2785 | Time: 4.4485 s
>>> Epoch [ 52/100]
train_loss: 2.2117 | train_acc: 0.2727 | val_loss: 2.2117 | val_acc: 0.2692 | test_acc: 0.2793 | Time: 4.4780 s
>>> Epoch [ 53/100]
train_loss: 2.2107 | train_acc: 0.2736 | val_loss: 2.2108 | val_acc: 0.2700 | test_acc: 0.2799 | Time: 4.3159 s
>>> Epoch [ 54/100]
train_loss: 2.2098 | train_acc: 0.2747 | val_loss: 2.2099 | val_acc: 0.2707 | test_acc: 0.2811 | Time: 4.2722 s
>>> Epoch [ 55/100]
train_loss: 2.2089 | train_acc: 0.2752 | val_loss: 2.2091 | val_acc: 0.2713 | test_acc: 0.2818 | Time: 4.6038 s
>>> Epoch [ 56/100]
train_loss: 2.2080 | train_acc: 0.2759 | val_loss: 2.2082 | val_acc: 0.2724 | test_acc: 0.2833 | Time: 4.1776 s
>>> Epoch [ 57/100]
train_loss: 2.2071 | train_acc: 0.2768 | val_loss: 2.2074 | val_acc: 0.2730 | test_acc: 0.2838 | Time: 4.4703 s
>>> Epoch [ 58/100]
train_loss: 2.2063 | train_acc: 0.2776 | val_loss: 2.2066 | val_acc: 0.2742 | test_acc: 0.2848 | Time: 4.4454 s
>>> Epoch [ 59/100]
train_loss: 2.2054 | train_acc: 0.2784 | val_loss: 2.2057 | val_acc: 0.2753 | test_acc: 0.2854 | Time: 4.3839 s
>>> Epoch [ 60/100]
train_loss: 2.2046 | train_acc: 0.2792 | val_loss: 2.2049 | val_acc: 0.2767 | test_acc: 0.2859 | Time: 4.4482 s
>>> Epoch [ 61/100]
train_loss: 2.2038 | train_acc: 0.2798 | val_loss: 2.2042 | val_acc: 0.2776 | test_acc: 0.2858 | Time: 4.4261 s
>>> Epoch [ 62/100]
train_loss: 2.2029 | train_acc: 0.2805 | val_loss: 2.2034 | val_acc: 0.2777 | test_acc: 0.2860 | Time: 4.3511 s
>>> Epoch [ 63/100]
train_loss: 2.2022 | train_acc: 0.2812 | val_loss: 2.2026 | val_acc: 0.2787 | test_acc: 0.2865 | Time: 4.4779 s
>>> Epoch [ 64/100]
train_loss: 2.2014 | train_acc: 0.2820 | val_loss: 2.2019 | val_acc: 0.2798 | test_acc: 0.2877 | Time: 4.3172 s
>>> Epoch [ 65/100]
train_loss: 2.2006 | train_acc: 0.2830 | val_loss: 2.2012 | val_acc: 0.2805 | test_acc: 0.2890 | Time: 4.3955 s
>>> Epoch [ 66/100]
train_loss: 2.1998 | train_acc: 0.2836 | val_loss: 2.2004 | val_acc: 0.2814 | test_acc: 0.2891 | Time: 4.5959 s
>>> Epoch [ 67/100]
train_loss: 2.1991 | train_acc: 0.2845 | val_loss: 2.1997 | val_acc: 0.2818 | test_acc: 0.2903 | Time: 4.5409 s
>>> Epoch [ 68/100]
train_loss: 2.1983 | train_acc: 0.2853 | val_loss: 2.1990 | val_acc: 0.2823 | test_acc: 0.2919 | Time: 4.3900 s
>>> Epoch [ 69/100]
train_loss: 2.1976 | train_acc: 0.2862 | val_loss: 2.1983 | val_acc: 0.2827 | test_acc: 0.2926 | Time: 4.5475 s
>>> Epoch [ 70/100]
train_loss: 2.1969 | train_acc: 0.2874 | val_loss: 2.1976 | val_acc: 0.2830 | test_acc: 0.2935 | Time: 4.4365 s
>>> Epoch [ 71/100]
train_loss: 2.1962 | train_acc: 0.2886 | val_loss: 2.1970 | val_acc: 0.2835 | test_acc: 0.2942 | Time: 4.4542 s
>>> Epoch [ 72/100]
train_loss: 2.1955 | train_acc: 0.2892 | val_loss: 2.1963 | val_acc: 0.2846 | test_acc: 0.2939 | Time: 4.3876 s
>>> Epoch [ 73/100]
train_loss: 2.1948 | train_acc: 0.2897 | val_loss: 2.1956 | val_acc: 0.2855 | test_acc: 0.2946 | Time: 4.3262 s
>>> Epoch [ 74/100]
train_loss: 2.1941 | train_acc: 0.2906 | val_loss: 2.1950 | val_acc: 0.2859 | test_acc: 0.2956 | Time: 4.2692 s
>>> Epoch [ 75/100]
train_loss: 2.1934 | train_acc: 0.2916 | val_loss: 2.1944 | val_acc: 0.2869 | test_acc: 0.2969 | Time: 3.6717 s
>>> Epoch [ 76/100]
train_loss: 2.1928 | train_acc: 0.2921 | val_loss: 2.1937 | val_acc: 0.2877 | test_acc: 0.2973 | Time: 3.4885 s
>>> Epoch [ 77/100]
train_loss: 2.1921 | train_acc: 0.2925 | val_loss: 2.1931 | val_acc: 0.2883 | test_acc: 0.2987 | Time: 3.4131 s
>>> Epoch [ 78/100]
train_loss: 2.1915 | train_acc: 0.2934 | val_loss: 2.1925 | val_acc: 0.2890 | test_acc: 0.2991 | Time: 3.4244 s
>>> Epoch [ 79/100]
train_loss: 2.1908 | train_acc: 0.2940 | val_loss: 2.1919 | val_acc: 0.2900 | test_acc: 0.2999 | Time: 3.5770 s
>>> Epoch [ 80/100]
train_loss: 2.1902 | train_acc: 0.2948 | val_loss: 2.1913 | val_acc: 0.2907 | test_acc: 0.3008 | Time: 3.3572 s
>>> Epoch [ 81/100]
train_loss: 2.1895 | train_acc: 0.2953 | val_loss: 2.1907 | val_acc: 0.2913 | test_acc: 0.3019 | Time: 3.4910 s
>>> Epoch [ 82/100]
train_loss: 2.1889 | train_acc: 0.2961 | val_loss: 2.1901 | val_acc: 0.2922 | test_acc: 0.3023 | Time: 3.3962 s
>>> Epoch [ 83/100]
train_loss: 2.1883 | train_acc: 0.2969 | val_loss: 2.1895 | val_acc: 0.2930 | test_acc: 0.3033 | Time: 3.4428 s
>>> Epoch [ 84/100]
train_loss: 2.1877 | train_acc: 0.2978 | val_loss: 2.1889 | val_acc: 0.2945 | test_acc: 0.3037 | Time: 3.5479 s
>>> Epoch [ 85/100]
train_loss: 2.1871 | train_acc: 0.2987 | val_loss: 2.1883 | val_acc: 0.2955 | test_acc: 0.3047 | Time: 3.6919 s
>>> Epoch [ 86/100]
train_loss: 2.1865 | train_acc: 0.2996 | val_loss: 2.1878 | val_acc: 0.2960 | test_acc: 0.3055 | Time: 3.4427 s
>>> Epoch [ 87/100]
train_loss: 2.1859 | train_acc: 0.3002 | val_loss: 2.1872 | val_acc: 0.2967 | test_acc: 0.3064 | Time: 3.5330 s
>>> Epoch [ 88/100]
train_loss: 2.1853 | train_acc: 0.3008 | val_loss: 2.1867 | val_acc: 0.2972 | test_acc: 0.3068 | Time: 3.6359 s
>>> Epoch [ 89/100]
train_loss: 2.1847 | train_acc: 0.3016 | val_loss: 2.1861 | val_acc: 0.2981 | test_acc: 0.3081 | Time: 3.4342 s
>>> Epoch [ 90/100]
train_loss: 2.1841 | train_acc: 0.3020 | val_loss: 2.1856 | val_acc: 0.2986 | test_acc: 0.3082 | Time: 3.4270 s
>>> Epoch [ 91/100]
train_loss: 2.1836 | train_acc: 0.3029 | val_loss: 2.1850 | val_acc: 0.2995 | test_acc: 0.3091 | Time: 3.5657 s
>>> Epoch [ 92/100]
train_loss: 2.1830 | train_acc: 0.3037 | val_loss: 2.1845 | val_acc: 0.3009 | test_acc: 0.3096 | Time: 3.3959 s
>>> Epoch [ 93/100]
train_loss: 2.1824 | train_acc: 0.3045 | val_loss: 2.1840 | val_acc: 0.3010 | test_acc: 0.3104 | Time: 3.6795 s
>>> Epoch [ 94/100]
train_loss: 2.1819 | train_acc: 0.3052 | val_loss: 2.1834 | val_acc: 0.3017 | test_acc: 0.3108 | Time: 3.4570 s
>>> Epoch [ 95/100]
train_loss: 2.1813 | train_acc: 0.3057 | val_loss: 2.1829 | val_acc: 0.3020 | test_acc: 0.3113 | Time: 3.2938 s
>>> Epoch [ 96/100]
train_loss: 2.1808 | train_acc: 0.3065 | val_loss: 2.1824 | val_acc: 0.3025 | test_acc: 0.3118 | Time: 3.5145 s
>>> Epoch [ 97/100]
train_loss: 2.1802 | train_acc: 0.3073 | val_loss: 2.1819 | val_acc: 0.3025 | test_acc: 0.3120 | Time: 3.5800 s
>>> Epoch [ 98/100]
train_loss: 2.1797 | train_acc: 0.3079 | val_loss: 2.1814 | val_acc: 0.3028 | test_acc: 0.3123 | Time: 3.3216 s
>>> Epoch [ 99/100]
train_loss: 2.1792 | train_acc: 0.3086 | val_loss: 2.1809 | val_acc: 0.3030 | test_acc: 0.3127 | Time: 3.3848 s
Fit completed in 5.1932 seconds
Saving the graph to ../logs/cifar10/101_tf_linear_raw_gradtest/exp_no_003/linear_classifier.pb
INFO:tensorflow:Froze 2 variables.
Converted 2 variables to const ops.
11 ops in the final graph.
Length of train loss          : 0
Length of train accuracy      : 0
Length of validate loss       : 0
Length of validate accuracy   : 0
Length of test accuracy       : 0
Restoring training from epoch : 0
>>> Epoch [  0/100]
train_loss: 2.3026 | train_acc: 0.0995 | val_loss: 2.3006 | val_acc: 0.2467 | test_acc: 0.2550 | Time: 10.8867 s
>>> Epoch [  1/100]
train_loss: 2.3006 | train_acc: 0.2472 | val_loss: 2.2986 | val_acc: 0.2456 | test_acc: 0.2541 | Time: 7.6436 s
>>> Epoch [  2/100]
train_loss: 2.2986 | train_acc: 0.2466 | val_loss: 2.2964 | val_acc: 0.2449 | test_acc: 0.2539 | Time: 7.1591 s
>>> Epoch [  3/100]
train_loss: 2.2965 | train_acc: 0.2460 | val_loss: 2.2942 | val_acc: 0.2442 | test_acc: 0.2527 | Time: 7.3559 s
>>> Epoch [  4/100]
train_loss: 2.2943 | train_acc: 0.2450 | val_loss: 2.2919 | val_acc: 0.2433 | test_acc: 0.2499 | Time: 7.5107 s
>>> Epoch [  5/100]
train_loss: 2.2920 | train_acc: 0.2439 | val_loss: 2.2895 | val_acc: 0.2420 | test_acc: 0.2497 | Time: 7.1848 s
>>> Epoch [  6/100]
train_loss: 2.2896 | train_acc: 0.2432 | val_loss: 2.2870 | val_acc: 0.2408 | test_acc: 0.2481 | Time: 7.4129 s
>>> Epoch [  7/100]
train_loss: 2.2871 | train_acc: 0.2424 | val_loss: 2.2844 | val_acc: 0.2403 | test_acc: 0.2461 | Time: 7.4663 s
>>> Epoch [  8/100]
train_loss: 2.2845 | train_acc: 0.2416 | val_loss: 2.2817 | val_acc: 0.2393 | test_acc: 0.2447 | Time: 6.9846 s
>>> Epoch [  9/100]
train_loss: 2.2819 | train_acc: 0.2407 | val_loss: 2.2791 | val_acc: 0.2383 | test_acc: 0.2445 | Time: 7.5409 s
>>> Epoch [ 10/100]
train_loss: 2.2793 | train_acc: 0.2398 | val_loss: 2.2764 | val_acc: 0.2381 | test_acc: 0.2436 | Time: 15.5943 s
>>> Epoch [ 11/100]
train_loss: 2.2767 | train_acc: 0.2401 | val_loss: 2.2738 | val_acc: 0.2381 | test_acc: 0.2432 | Time: 14.7639 s
>>> Epoch [ 12/100]
train_loss: 2.2741 | train_acc: 0.2399 | val_loss: 2.2713 | val_acc: 0.2380 | test_acc: 0.2431 | Time: 15.2833 s
>>> Epoch [ 13/100]
train_loss: 2.2716 | train_acc: 0.2403 | val_loss: 2.2689 | val_acc: 0.2379 | test_acc: 0.2437 | Time: 9.1389 s
>>> Epoch [ 14/100]
train_loss: 2.2692 | train_acc: 0.2407 | val_loss: 2.2666 | val_acc: 0.2389 | test_acc: 0.2453 | Time: 7.7719 s
>>> Epoch [ 15/100]
train_loss: 2.2669 | train_acc: 0.2413 | val_loss: 2.2643 | val_acc: 0.2404 | test_acc: 0.2464 | Time: 9.6279 s
>>> Epoch [ 16/100]
train_loss: 2.2647 | train_acc: 0.2419 | val_loss: 2.2622 | val_acc: 0.2412 | test_acc: 0.2476 | Time: 15.1316 s
>>> Epoch [ 17/100]
train_loss: 2.2625 | train_acc: 0.2424 | val_loss: 2.2601 | val_acc: 0.2415 | test_acc: 0.2491 | Time: 10.6771 s
>>> Epoch [ 18/100]
train_loss: 2.2604 | train_acc: 0.2433 | val_loss: 2.2581 | val_acc: 0.2422 | test_acc: 0.2509 | Time: 9.5241 s
>>> Epoch [ 19/100]
train_loss: 2.2584 | train_acc: 0.2444 | val_loss: 2.2561 | val_acc: 0.2431 | test_acc: 0.2523 | Time: 9.9431 s
>>> Epoch [ 20/100]
train_loss: 2.2565 | train_acc: 0.2456 | val_loss: 2.2542 | val_acc: 0.2428 | test_acc: 0.2529 | Time: 10.7739 s
>>> Epoch [ 21/100]
train_loss: 2.2545 | train_acc: 0.2466 | val_loss: 2.2523 | val_acc: 0.2440 | test_acc: 0.2535 | Time: 9.3695 s
>>> Epoch [ 22/100]
train_loss: 2.2526 | train_acc: 0.2477 | val_loss: 2.2505 | val_acc: 0.2450 | test_acc: 0.2538 | Time: 9.5061 s
>>> Epoch [ 23/100]
train_loss: 2.2508 | train_acc: 0.2485 | val_loss: 2.2487 | val_acc: 0.2458 | test_acc: 0.2548 | Time: 8.7486 s
>>> Epoch [ 24/100]
train_loss: 2.2490 | train_acc: 0.2494 | val_loss: 2.2470 | val_acc: 0.2464 | test_acc: 0.2562 | Time: 9.4724 s
>>> Epoch [ 25/100]
train_loss: 2.2472 | train_acc: 0.2503 | val_loss: 2.2453 | val_acc: 0.2470 | test_acc: 0.2567 | Time: 9.4229 s
>>> Epoch [ 26/100]
train_loss: 2.2455 | train_acc: 0.2512 | val_loss: 2.2436 | val_acc: 0.2479 | test_acc: 0.2575 | Time: 9.0480 s
>>> Epoch [ 27/100]
train_loss: 2.2438 | train_acc: 0.2524 | val_loss: 2.2420 | val_acc: 0.2503 | test_acc: 0.2581 | Time: 9.5685 s
>>> Epoch [ 28/100]
train_loss: 2.2421 | train_acc: 0.2532 | val_loss: 2.2403 | val_acc: 0.2517 | test_acc: 0.2590 | Time: 9.3456 s
>>> Epoch [ 29/100]
train_loss: 2.2405 | train_acc: 0.2542 | val_loss: 2.2388 | val_acc: 0.2516 | test_acc: 0.2599 | Time: 9.0190 s
>>> Epoch [ 30/100]
train_loss: 2.2389 | train_acc: 0.2555 | val_loss: 2.2372 | val_acc: 0.2520 | test_acc: 0.2606 | Time: 9.6995 s
>>> Epoch [ 31/100]
train_loss: 2.2373 | train_acc: 0.2567 | val_loss: 2.2358 | val_acc: 0.2541 | test_acc: 0.2611 | Time: 9.2696 s
>>> Epoch [ 32/100]
train_loss: 2.2358 | train_acc: 0.2570 | val_loss: 2.2343 | val_acc: 0.2541 | test_acc: 0.2616 | Time: 8.7376 s
>>> Epoch [ 33/100]
train_loss: 2.2343 | train_acc: 0.2581 | val_loss: 2.2329 | val_acc: 0.2545 | test_acc: 0.2624 | Time: 9.6262 s
>>> Epoch [ 34/100]
train_loss: 2.2328 | train_acc: 0.2592 | val_loss: 2.2315 | val_acc: 0.2554 | test_acc: 0.2630 | Time: 9.6028 s
>>> Epoch [ 35/100]
train_loss: 2.2314 | train_acc: 0.2599 | val_loss: 2.2301 | val_acc: 0.2559 | test_acc: 0.2638 | Time: 8.8708 s
>>> Epoch [ 36/100]
train_loss: 2.2300 | train_acc: 0.2606 | val_loss: 2.2288 | val_acc: 0.2563 | test_acc: 0.2649 | Time: 9.5517 s
>>> Epoch [ 37/100]
train_loss: 2.2286 | train_acc: 0.2615 | val_loss: 2.2275 | val_acc: 0.2566 | test_acc: 0.2659 | Time: 9.1427 s
>>> Epoch [ 38/100]
train_loss: 2.2272 | train_acc: 0.2624 | val_loss: 2.2263 | val_acc: 0.2574 | test_acc: 0.2670 | Time: 9.1413 s
>>> Epoch [ 39/100]
train_loss: 2.2259 | train_acc: 0.2630 | val_loss: 2.2251 | val_acc: 0.2586 | test_acc: 0.2681 | Time: 15.5623 s
>>> Epoch [ 40/100]
train_loss: 2.2247 | train_acc: 0.2634 | val_loss: 2.2239 | val_acc: 0.2587 | test_acc: 0.2690 | Time: 12.9665 s
>>> Epoch [ 41/100]
train_loss: 2.2234 | train_acc: 0.2642 | val_loss: 2.2227 | val_acc: 0.2593 | test_acc: 0.2697 | Time: 9.1781 s
>>> Epoch [ 42/100]
train_loss: 2.2222 | train_acc: 0.2646 | val_loss: 2.2216 | val_acc: 0.2599 | test_acc: 0.2705 | Time: 9.5831 s
>>> Epoch [ 43/100]
train_loss: 2.2210 | train_acc: 0.2655 | val_loss: 2.2205 | val_acc: 0.2607 | test_acc: 0.2712 | Time: 9.6689 s
>>> Epoch [ 44/100]
train_loss: 2.2199 | train_acc: 0.2665 | val_loss: 2.2194 | val_acc: 0.2613 | test_acc: 0.2715 | Time: 9.2308 s
>>> Epoch [ 45/100]
train_loss: 2.2187 | train_acc: 0.2673 | val_loss: 2.2183 | val_acc: 0.2623 | test_acc: 0.2727 | Time: 9.6654 s
>>> Epoch [ 46/100]
train_loss: 2.2176 | train_acc: 0.2680 | val_loss: 2.2173 | val_acc: 0.2630 | test_acc: 0.2730 | Time: 9.5278 s
>>> Epoch [ 47/100]
train_loss: 2.2165 | train_acc: 0.2688 | val_loss: 2.2163 | val_acc: 0.2640 | test_acc: 0.2734 | Time: 8.9556 s
>>> Epoch [ 48/100]
train_loss: 2.2155 | train_acc: 0.2691 | val_loss: 2.2153 | val_acc: 0.2643 | test_acc: 0.2737 | Time: 9.3950 s
>>> Epoch [ 49/100]
train_loss: 2.2144 | train_acc: 0.2701 | val_loss: 2.2144 | val_acc: 0.2651 | test_acc: 0.2740 | Time: 8.8741 s
>>> Epoch [ 50/100]
train_loss: 2.2134 | train_acc: 0.2707 | val_loss: 2.2134 | val_acc: 0.2656 | test_acc: 0.2753 | Time: 8.5482 s
>>> Epoch [ 51/100]
train_loss: 2.2124 | train_acc: 0.2716 | val_loss: 2.2125 | val_acc: 0.2665 | test_acc: 0.2770 | Time: 8.4630 s
>>> Epoch [ 52/100]
train_loss: 2.2114 | train_acc: 0.2723 | val_loss: 2.2116 | val_acc: 0.2672 | test_acc: 0.2777 | Time: 16.1473 s
>>> Epoch [ 53/100]
train_loss: 2.2105 | train_acc: 0.2731 | val_loss: 2.2107 | val_acc: 0.2677 | test_acc: 0.2787 | Time: 8.9300 s
>>> Epoch [ 54/100]
train_loss: 2.2095 | train_acc: 0.2738 | val_loss: 2.2099 | val_acc: 0.2683 | test_acc: 0.2798 | Time: 9.6099 s
>>> Epoch [ 55/100]
train_loss: 2.2086 | train_acc: 0.2747 | val_loss: 2.2090 | val_acc: 0.2686 | test_acc: 0.2802 | Time: 9.4493 s
>>> Epoch [ 56/100]
train_loss: 2.2077 | train_acc: 0.2754 | val_loss: 2.2082 | val_acc: 0.2691 | test_acc: 0.2814 | Time: 8.9818 s
>>> Epoch [ 57/100]
train_loss: 2.2068 | train_acc: 0.2760 | val_loss: 2.2074 | val_acc: 0.2698 | test_acc: 0.2827 | Time: 9.6569 s
>>> Epoch [ 58/100]
train_loss: 2.2059 | train_acc: 0.2769 | val_loss: 2.2066 | val_acc: 0.2705 | test_acc: 0.2832 | Time: 9.2712 s
>>> Epoch [ 59/100]
train_loss: 2.2051 | train_acc: 0.2779 | val_loss: 2.2058 | val_acc: 0.2712 | test_acc: 0.2834 | Time: 8.9418 s
>>> Epoch [ 60/100]
train_loss: 2.2042 | train_acc: 0.2785 | val_loss: 2.2050 | val_acc: 0.2725 | test_acc: 0.2843 | Time: 9.3747 s
>>> Epoch [ 61/100]
train_loss: 2.2034 | train_acc: 0.2794 | val_loss: 2.2042 | val_acc: 0.2733 | test_acc: 0.2852 | Time: 9.2412 s
>>> Epoch [ 62/100]
train_loss: 2.2026 | train_acc: 0.2805 | val_loss: 2.2035 | val_acc: 0.2742 | test_acc: 0.2858 | Time: 8.8515 s
>>> Epoch [ 63/100]
train_loss: 2.2018 | train_acc: 0.2815 | val_loss: 2.2027 | val_acc: 0.2752 | test_acc: 0.2872 | Time: 8.4139 s
>>> Epoch [ 64/100]
train_loss: 2.2010 | train_acc: 0.2822 | val_loss: 2.2020 | val_acc: 0.2762 | test_acc: 0.2880 | Time: 7.9518 s
>>> Epoch [ 65/100]
train_loss: 2.2002 | train_acc: 0.2832 | val_loss: 2.2013 | val_acc: 0.2775 | test_acc: 0.2890 | Time: 8.1374 s
>>> Epoch [ 66/100]
train_loss: 2.1995 | train_acc: 0.2838 | val_loss: 2.2006 | val_acc: 0.2782 | test_acc: 0.2903 | Time: 8.7624 s
>>> Epoch [ 67/100]
train_loss: 2.1987 | train_acc: 0.2849 | val_loss: 2.1999 | val_acc: 0.2792 | test_acc: 0.2907 | Time: 8.6875 s
>>> Epoch [ 68/100]
train_loss: 2.1980 | train_acc: 0.2861 | val_loss: 2.1992 | val_acc: 0.2799 | test_acc: 0.2918 | Time: 8.2303 s
>>> Epoch [ 69/100]
train_loss: 2.1972 | train_acc: 0.2867 | val_loss: 2.1985 | val_acc: 0.2801 | test_acc: 0.2925 | Time: 8.8104 s
>>> Epoch [ 70/100]
train_loss: 2.1965 | train_acc: 0.2874 | val_loss: 2.1979 | val_acc: 0.2799 | test_acc: 0.2931 | Time: 9.0723 s
>>> Epoch [ 71/100]
train_loss: 2.1958 | train_acc: 0.2878 | val_loss: 2.1972 | val_acc: 0.2809 | test_acc: 0.2936 | Time: 9.2458 s
>>> Epoch [ 72/100]
train_loss: 2.1951 | train_acc: 0.2889 | val_loss: 2.1966 | val_acc: 0.2816 | test_acc: 0.2945 | Time: 9.2745 s
>>> Epoch [ 73/100]
train_loss: 2.1944 | train_acc: 0.2895 | val_loss: 2.1959 | val_acc: 0.2822 | test_acc: 0.2953 | Time: 9.3299 s
>>> Epoch [ 74/100]
train_loss: 2.1937 | train_acc: 0.2904 | val_loss: 2.1953 | val_acc: 0.2834 | test_acc: 0.2965 | Time: 9.1039 s
>>> Epoch [ 75/100]
train_loss: 2.1931 | train_acc: 0.2913 | val_loss: 2.1947 | val_acc: 0.2838 | test_acc: 0.2975 | Time: 9.6242 s
>>> Epoch [ 76/100]
train_loss: 2.1924 | train_acc: 0.2920 | val_loss: 2.1940 | val_acc: 0.2842 | test_acc: 0.2988 | Time: 9.2367 s
>>> Epoch [ 77/100]
train_loss: 2.1917 | train_acc: 0.2931 | val_loss: 2.1934 | val_acc: 0.2847 | test_acc: 0.2997 | Time: 8.9602 s
>>> Epoch [ 78/100]
train_loss: 2.1911 | train_acc: 0.2940 | val_loss: 2.1928 | val_acc: 0.2856 | test_acc: 0.3000 | Time: 9.4754 s
>>> Epoch [ 79/100]
train_loss: 2.1904 | train_acc: 0.2947 | val_loss: 2.1922 | val_acc: 0.2858 | test_acc: 0.3010 | Time: 9.4487 s
>>> Epoch [ 80/100]
train_loss: 2.1898 | train_acc: 0.2957 | val_loss: 2.1916 | val_acc: 0.2867 | test_acc: 0.3017 | Time: 8.8751 s
>>> Epoch [ 81/100]
train_loss: 2.1892 | train_acc: 0.2965 | val_loss: 2.1910 | val_acc: 0.2877 | test_acc: 0.3022 | Time: 9.5300 s
>>> Epoch [ 82/100]
train_loss: 2.1885 | train_acc: 0.2970 | val_loss: 2.1904 | val_acc: 0.2883 | test_acc: 0.3025 | Time: 9.3077 s
>>> Epoch [ 83/100]
train_loss: 2.1879 | train_acc: 0.2975 | val_loss: 2.1899 | val_acc: 0.2891 | test_acc: 0.3029 | Time: 9.0118 s
>>> Epoch [ 84/100]
train_loss: 2.1873 | train_acc: 0.2983 | val_loss: 2.1893 | val_acc: 0.2895 | test_acc: 0.3036 | Time: 9.3547 s
>>> Epoch [ 85/100]
train_loss: 2.1867 | train_acc: 0.2990 | val_loss: 2.1887 | val_acc: 0.2904 | test_acc: 0.3044 | Time: 9.2677 s
>>> Epoch [ 86/100]
train_loss: 2.1861 | train_acc: 0.2996 | val_loss: 2.1882 | val_acc: 0.2911 | test_acc: 0.3051 | Time: 8.8588 s
>>> Epoch [ 87/100]
train_loss: 2.1855 | train_acc: 0.3002 | val_loss: 2.1876 | val_acc: 0.2926 | test_acc: 0.3048 | Time: 9.5662 s
>>> Epoch [ 88/100]
train_loss: 2.1849 | train_acc: 0.3007 | val_loss: 2.1871 | val_acc: 0.2930 | test_acc: 0.3060 | Time: 9.4782 s
>>> Epoch [ 89/100]
train_loss: 2.1843 | train_acc: 0.3016 | val_loss: 2.1865 | val_acc: 0.2935 | test_acc: 0.3067 | Time: 15.3021 s
>>> Epoch [ 90/100]
train_loss: 2.1838 | train_acc: 0.3023 | val_loss: 2.1860 | val_acc: 0.2942 | test_acc: 0.3071 | Time: 10.9671 s
>>> Epoch [ 91/100]
train_loss: 2.1832 | train_acc: 0.3032 | val_loss: 2.1855 | val_acc: 0.2945 | test_acc: 0.3076 | Time: 8.8303 s
>>> Epoch [ 92/100]
train_loss: 2.1826 | train_acc: 0.3036 | val_loss: 2.1849 | val_acc: 0.2954 | test_acc: 0.3081 | Time: 8.5819 s
>>> Epoch [ 93/100]
train_loss: 2.1821 | train_acc: 0.3040 | val_loss: 2.1844 | val_acc: 0.2961 | test_acc: 0.3082 | Time: 8.9317 s
>>> Epoch [ 94/100]
train_loss: 2.1815 | train_acc: 0.3045 | val_loss: 2.1839 | val_acc: 0.2973 | test_acc: 0.3093 | Time: 8.9301 s
>>> Epoch [ 95/100]
train_loss: 2.1810 | train_acc: 0.3053 | val_loss: 2.1834 | val_acc: 0.2978 | test_acc: 0.3107 | Time: 8.6195 s
>>> Epoch [ 96/100]
train_loss: 2.1804 | train_acc: 0.3059 | val_loss: 2.1829 | val_acc: 0.2984 | test_acc: 0.3115 | Time: 9.0760 s
>>> Epoch [ 97/100]
train_loss: 2.1799 | train_acc: 0.3064 | val_loss: 2.1824 | val_acc: 0.2992 | test_acc: 0.3119 | Time: 8.4404 s
>>> Epoch [ 98/100]
train_loss: 2.1794 | train_acc: 0.3071 | val_loss: 2.1819 | val_acc: 0.3001 | test_acc: 0.3121 | Time: 7.7686 s
>>> Epoch [ 99/100]
train_loss: 2.1788 | train_acc: 0.3076 | val_loss: 2.1814 | val_acc: 0.3011 | test_acc: 0.3126 | Time: 15.9470 s
Fit completed in 20.4869 seconds
Saving the graph to ../logs/cifar10/101_tf_linear_raw_gradtest/exp_no_004/linear_classifier.pb
INFO:tensorflow:Froze 2 variables.
Converted 2 variables to const ops.
11 ops in the final graph.
Length of train loss          : 0
Length of train accuracy      : 0
Length of validate loss       : 0
Length of validate accuracy   : 0
Length of test accuracy       : 0
Restoring training from epoch : 0
>>> Epoch [  0/100]
train_loss: 2.3026 | train_acc: 0.0998 | val_loss: 2.3007 | val_acc: 0.2421 | test_acc: 0.2547 | Time: 26.1312 s
>>> Epoch [  1/100]
train_loss: 2.3006 | train_acc: 0.2482 | val_loss: 2.2986 | val_acc: 0.2422 | test_acc: 0.2536 | Time: 17.2307 s
>>> Epoch [  2/100]
train_loss: 2.2986 | train_acc: 0.2477 | val_loss: 2.2965 | val_acc: 0.2415 | test_acc: 0.2533 | Time: 16.4591 s
>>> Epoch [  3/100]
train_loss: 2.2964 | train_acc: 0.2471 | val_loss: 2.2944 | val_acc: 0.2404 | test_acc: 0.2517 | Time: 17.4649 s
>>> Epoch [  4/100]
train_loss: 2.2942 | train_acc: 0.2460 | val_loss: 2.2921 | val_acc: 0.2392 | test_acc: 0.2507 | Time: 18.1867 s
>>> Epoch [  5/100]
train_loss: 2.2918 | train_acc: 0.2447 | val_loss: 2.2897 | val_acc: 0.2383 | test_acc: 0.2489 | Time: 17.6210 s
>>> Epoch [  6/100]
train_loss: 2.2894 | train_acc: 0.2439 | val_loss: 2.2873 | val_acc: 0.2372 | test_acc: 0.2470 | Time: 20.3843 s
>>> Epoch [  7/100]
train_loss: 2.2868 | train_acc: 0.2427 | val_loss: 2.2848 | val_acc: 0.2359 | test_acc: 0.2461 | Time: 20.4894 s
>>> Epoch [  8/100]
train_loss: 2.2842 | train_acc: 0.2418 | val_loss: 2.2822 | val_acc: 0.2341 | test_acc: 0.2445 | Time: 19.1399 s
>>> Epoch [  9/100]
train_loss: 2.2815 | train_acc: 0.2409 | val_loss: 2.2796 | val_acc: 0.2340 | test_acc: 0.2439 | Time: 20.2579 s
>>> Epoch [ 10/100]
train_loss: 2.2788 | train_acc: 0.2407 | val_loss: 2.2771 | val_acc: 0.2341 | test_acc: 0.2434 | Time: 20.0232 s
>>> Epoch [ 11/100]
train_loss: 2.2762 | train_acc: 0.2403 | val_loss: 2.2746 | val_acc: 0.2342 | test_acc: 0.2435 | Time: 19.5546 s
>>> Epoch [ 12/100]
train_loss: 2.2736 | train_acc: 0.2402 | val_loss: 2.2722 | val_acc: 0.2342 | test_acc: 0.2440 | Time: 20.4233 s
>>> Epoch [ 13/100]
train_loss: 2.2710 | train_acc: 0.2407 | val_loss: 2.2699 | val_acc: 0.2345 | test_acc: 0.2445 | Time: 20.3532 s
>>> Epoch [ 14/100]
train_loss: 2.2686 | train_acc: 0.2407 | val_loss: 2.2677 | val_acc: 0.2351 | test_acc: 0.2449 | Time: 19.2748 s
>>> Epoch [ 15/100]
train_loss: 2.2663 | train_acc: 0.2415 | val_loss: 2.2656 | val_acc: 0.2355 | test_acc: 0.2460 | Time: 33.8137 s
>>> Epoch [ 16/100]
train_loss: 2.2641 | train_acc: 0.2419 | val_loss: 2.2635 | val_acc: 0.2369 | test_acc: 0.2473 | Time: 19.1217 s
>>> Epoch [ 17/100]
train_loss: 2.2619 | train_acc: 0.2423 | val_loss: 2.2615 | val_acc: 0.2377 | test_acc: 0.2493 | Time: 18.3198 s
>>> Epoch [ 18/100]
train_loss: 2.2598 | train_acc: 0.2436 | val_loss: 2.2596 | val_acc: 0.2375 | test_acc: 0.2510 | Time: 20.9091 s
>>> Epoch [ 19/100]
train_loss: 2.2578 | train_acc: 0.2447 | val_loss: 2.2577 | val_acc: 0.2384 | test_acc: 0.2526 | Time: 20.1739 s
>>> Epoch [ 20/100]
train_loss: 2.2558 | train_acc: 0.2456 | val_loss: 2.2559 | val_acc: 0.2390 | test_acc: 0.2530 | Time: 18.9394 s
>>> Epoch [ 21/100]
train_loss: 2.2539 | train_acc: 0.2468 | val_loss: 2.2541 | val_acc: 0.2394 | test_acc: 0.2538 | Time: 20.6857 s
>>> Epoch [ 22/100]
train_loss: 2.2520 | train_acc: 0.2478 | val_loss: 2.2523 | val_acc: 0.2403 | test_acc: 0.2543 | Time: 20.2280 s
>>> Epoch [ 23/100]
train_loss: 2.2502 | train_acc: 0.2486 | val_loss: 2.2506 | val_acc: 0.2415 | test_acc: 0.2554 | Time: 19.5895 s
>>> Epoch [ 24/100]
train_loss: 2.2484 | train_acc: 0.2498 | val_loss: 2.2489 | val_acc: 0.2426 | test_acc: 0.2564 | Time: 20.4381 s
>>> Epoch [ 25/100]
train_loss: 2.2466 | train_acc: 0.2503 | val_loss: 2.2473 | val_acc: 0.2441 | test_acc: 0.2568 | Time: 20.3423 s
>>> Epoch [ 26/100]
train_loss: 2.2449 | train_acc: 0.2517 | val_loss: 2.2456 | val_acc: 0.2453 | test_acc: 0.2574 | Time: 21.6719 s
>>> Epoch [ 27/100]
train_loss: 2.2432 | train_acc: 0.2522 | val_loss: 2.2440 | val_acc: 0.2468 | test_acc: 0.2579 | Time: 33.9885 s
>>> Epoch [ 28/100]
train_loss: 2.2416 | train_acc: 0.2532 | val_loss: 2.2425 | val_acc: 0.2478 | test_acc: 0.2584 | Time: 18.4694 s
>>> Epoch [ 29/100]
train_loss: 2.2399 | train_acc: 0.2537 | val_loss: 2.2409 | val_acc: 0.2487 | test_acc: 0.2588 | Time: 19.1716 s
>>> Epoch [ 30/100]
train_loss: 2.2384 | train_acc: 0.2547 | val_loss: 2.2395 | val_acc: 0.2490 | test_acc: 0.2599 | Time: 20.4829 s
>>> Epoch [ 31/100]
train_loss: 2.2368 | train_acc: 0.2555 | val_loss: 2.2380 | val_acc: 0.2503 | test_acc: 0.2610 | Time: 20.1972 s
>>> Epoch [ 32/100]
train_loss: 2.2353 | train_acc: 0.2563 | val_loss: 2.2366 | val_acc: 0.2516 | test_acc: 0.2627 | Time: 19.2760 s
>>> Epoch [ 33/100]
train_loss: 2.2338 | train_acc: 0.2572 | val_loss: 2.2352 | val_acc: 0.2527 | test_acc: 0.2632 | Time: 20.7201 s
>>> Epoch [ 34/100]
train_loss: 2.2324 | train_acc: 0.2585 | val_loss: 2.2338 | val_acc: 0.2527 | test_acc: 0.2633 | Time: 19.9841 s
>>> Epoch [ 35/100]
train_loss: 2.2310 | train_acc: 0.2593 | val_loss: 2.2325 | val_acc: 0.2530 | test_acc: 0.2641 | Time: 19.0371 s
>>> Epoch [ 36/100]
train_loss: 2.2296 | train_acc: 0.2599 | val_loss: 2.2312 | val_acc: 0.2539 | test_acc: 0.2650 | Time: 20.3308 s
>>> Epoch [ 37/100]
train_loss: 2.2282 | train_acc: 0.2605 | val_loss: 2.2299 | val_acc: 0.2542 | test_acc: 0.2652 | Time: 20.6775 s
>>> Epoch [ 38/100]
train_loss: 2.2269 | train_acc: 0.2611 | val_loss: 2.2287 | val_acc: 0.2552 | test_acc: 0.2653 | Time: 16.1752 s
>>> Epoch [ 39/100]
train_loss: 2.2257 | train_acc: 0.2616 | val_loss: 2.2275 | val_acc: 0.2561 | test_acc: 0.2662 | Time: 26.7978 s
>>> Epoch [ 40/100]
train_loss: 2.2244 | train_acc: 0.2623 | val_loss: 2.2263 | val_acc: 0.2572 | test_acc: 0.2669 | Time: 20.4785 s
>>> Epoch [ 41/100]
train_loss: 2.2232 | train_acc: 0.2632 | val_loss: 2.2251 | val_acc: 0.2588 | test_acc: 0.2674 | Time: 20.4494 s
>>> Epoch [ 42/100]
train_loss: 2.2220 | train_acc: 0.2640 | val_loss: 2.2240 | val_acc: 0.2594 | test_acc: 0.2681 | Time: 21.1007 s
>>> Epoch [ 43/100]
train_loss: 2.2208 | train_acc: 0.2648 | val_loss: 2.2229 | val_acc: 0.2599 | test_acc: 0.2693 | Time: 20.0882 s
>>> Epoch [ 44/100]
train_loss: 2.2197 | train_acc: 0.2658 | val_loss: 2.2218 | val_acc: 0.2598 | test_acc: 0.2702 | Time: 18.8261 s
>>> Epoch [ 45/100]
train_loss: 2.2186 | train_acc: 0.2666 | val_loss: 2.2207 | val_acc: 0.2608 | test_acc: 0.2710 | Time: 21.2350 s
>>> Epoch [ 46/100]
train_loss: 2.2175 | train_acc: 0.2674 | val_loss: 2.2197 | val_acc: 0.2615 | test_acc: 0.2721 | Time: 20.4055 s
>>> Epoch [ 47/100]
train_loss: 2.2165 | train_acc: 0.2679 | val_loss: 2.2187 | val_acc: 0.2626 | test_acc: 0.2738 | Time: 19.0431 s
>>> Epoch [ 48/100]
train_loss: 2.2154 | train_acc: 0.2685 | val_loss: 2.2177 | val_acc: 0.2638 | test_acc: 0.2742 | Time: 21.0905 s
>>> Epoch [ 49/100]
train_loss: 2.2144 | train_acc: 0.2691 | val_loss: 2.2167 | val_acc: 0.2645 | test_acc: 0.2743 | Time: 20.9982 s
>>> Epoch [ 50/100]
train_loss: 2.2134 | train_acc: 0.2699 | val_loss: 2.2157 | val_acc: 0.2649 | test_acc: 0.2750 | Time: 27.5131 s
>>> Epoch [ 51/100]
train_loss: 2.2124 | train_acc: 0.2706 | val_loss: 2.2148 | val_acc: 0.2658 | test_acc: 0.2758 | Time: 19.0550 s
>>> Epoch [ 52/100]
train_loss: 2.2115 | train_acc: 0.2715 | val_loss: 2.2138 | val_acc: 0.2666 | test_acc: 0.2773 | Time: 18.7299 s
>>> Epoch [ 53/100]
train_loss: 2.2105 | train_acc: 0.2723 | val_loss: 2.2129 | val_acc: 0.2672 | test_acc: 0.2783 | Time: 17.5199 s
>>> Epoch [ 54/100]
train_loss: 2.2096 | train_acc: 0.2730 | val_loss: 2.2120 | val_acc: 0.2685 | test_acc: 0.2792 | Time: 19.6639 s
>>> Epoch [ 55/100]
train_loss: 2.2087 | train_acc: 0.2738 | val_loss: 2.2112 | val_acc: 0.2695 | test_acc: 0.2803 | Time: 17.7062 s
>>> Epoch [ 56/100]
train_loss: 2.2078 | train_acc: 0.2745 | val_loss: 2.2103 | val_acc: 0.2698 | test_acc: 0.2817 | Time: 16.7005 s
>>> Epoch [ 57/100]
train_loss: 2.2069 | train_acc: 0.2753 | val_loss: 2.2095 | val_acc: 0.2701 | test_acc: 0.2822 | Time: 17.8245 s
>>> Epoch [ 58/100]
train_loss: 2.2061 | train_acc: 0.2759 | val_loss: 2.2086 | val_acc: 0.2713 | test_acc: 0.2833 | Time: 17.2981 s
>>> Epoch [ 59/100]
train_loss: 2.2052 | train_acc: 0.2768 | val_loss: 2.2078 | val_acc: 0.2724 | test_acc: 0.2846 | Time: 16.1581 s
>>> Epoch [ 60/100]
train_loss: 2.2044 | train_acc: 0.2775 | val_loss: 2.2070 | val_acc: 0.2733 | test_acc: 0.2853 | Time: 17.5855 s
>>> Epoch [ 61/100]
train_loss: 2.2036 | train_acc: 0.2784 | val_loss: 2.2062 | val_acc: 0.2738 | test_acc: 0.2861 | Time: 29.9645 s
>>> Epoch [ 62/100]
train_loss: 2.2028 | train_acc: 0.2792 | val_loss: 2.2054 | val_acc: 0.2750 | test_acc: 0.2867 | Time: 19.9409 s
>>> Epoch [ 63/100]
train_loss: 2.2020 | train_acc: 0.2801 | val_loss: 2.2047 | val_acc: 0.2768 | test_acc: 0.2872 | Time: 21.0544 s
>>> Epoch [ 64/100]
train_loss: 2.2012 | train_acc: 0.2810 | val_loss: 2.2039 | val_acc: 0.2767 | test_acc: 0.2879 | Time: 20.9142 s
>>> Epoch [ 65/100]
train_loss: 2.2005 | train_acc: 0.2817 | val_loss: 2.2032 | val_acc: 0.2779 | test_acc: 0.2890 | Time: 17.9596 s
>>> Epoch [ 66/100]
train_loss: 2.1997 | train_acc: 0.2829 | val_loss: 2.2024 | val_acc: 0.2791 | test_acc: 0.2893 | Time: 17.4759 s
>>> Epoch [ 67/100]
train_loss: 2.1990 | train_acc: 0.2835 | val_loss: 2.2017 | val_acc: 0.2796 | test_acc: 0.2899 | Time: 19.1091 s
>>> Epoch [ 68/100]
train_loss: 2.1982 | train_acc: 0.2844 | val_loss: 2.2010 | val_acc: 0.2809 | test_acc: 0.2909 | Time: 17.7529 s
>>> Epoch [ 69/100]
train_loss: 2.1975 | train_acc: 0.2849 | val_loss: 2.2003 | val_acc: 0.2817 | test_acc: 0.2913 | Time: 19.5649 s
>>> Epoch [ 70/100]
train_loss: 2.1968 | train_acc: 0.2859 | val_loss: 2.1996 | val_acc: 0.2822 | test_acc: 0.2917 | Time: 19.9505 s
>>> Epoch [ 71/100]
train_loss: 2.1961 | train_acc: 0.2866 | val_loss: 2.1989 | val_acc: 0.2832 | test_acc: 0.2920 | Time: 19.5119 s
>>> Epoch [ 72/100]
train_loss: 2.1954 | train_acc: 0.2875 | val_loss: 2.1982 | val_acc: 0.2842 | test_acc: 0.2928 | Time: 20.3579 s
>>> Epoch [ 73/100]
train_loss: 2.1947 | train_acc: 0.2884 | val_loss: 2.1975 | val_acc: 0.2855 | test_acc: 0.2940 | Time: 19.9338 s
>>> Epoch [ 74/100]
train_loss: 2.1940 | train_acc: 0.2891 | val_loss: 2.1969 | val_acc: 0.2863 | test_acc: 0.2953 | Time: 18.9778 s
>>> Epoch [ 75/100]
train_loss: 2.1934 | train_acc: 0.2900 | val_loss: 2.1962 | val_acc: 0.2872 | test_acc: 0.2958 | Time: 20.6830 s
>>> Epoch [ 76/100]
train_loss: 2.1927 | train_acc: 0.2908 | val_loss: 2.1956 | val_acc: 0.2889 | test_acc: 0.2963 | Time: 20.2692 s
>>> Epoch [ 77/100]
train_loss: 2.1920 | train_acc: 0.2915 | val_loss: 2.1949 | val_acc: 0.2892 | test_acc: 0.2976 | Time: 18.8524 s
>>> Epoch [ 78/100]
train_loss: 2.1914 | train_acc: 0.2924 | val_loss: 2.1943 | val_acc: 0.2901 | test_acc: 0.2981 | Time: 24.7632 s
>>> Epoch [ 79/100]
train_loss: 2.1908 | train_acc: 0.2932 | val_loss: 2.1937 | val_acc: 0.2909 | test_acc: 0.2991 | Time: 32.0051 s
>>> Epoch [ 80/100]
train_loss: 2.1901 | train_acc: 0.2938 | val_loss: 2.1931 | val_acc: 0.2913 | test_acc: 0.2995 | Time: 16.6634 s
>>> Epoch [ 81/100]
train_loss: 2.1895 | train_acc: 0.2947 | val_loss: 2.1925 | val_acc: 0.2922 | test_acc: 0.2998 | Time: 17.3788 s
>>> Epoch [ 82/100]
train_loss: 2.1889 | train_acc: 0.2953 | val_loss: 2.1919 | val_acc: 0.2928 | test_acc: 0.3006 | Time: 17.2083 s
>>> Epoch [ 83/100]
train_loss: 2.1883 | train_acc: 0.2961 | val_loss: 2.1913 | val_acc: 0.2946 | test_acc: 0.3014 | Time: 16.5485 s
>>> Epoch [ 84/100]
train_loss: 2.1877 | train_acc: 0.2966 | val_loss: 2.1907 | val_acc: 0.2953 | test_acc: 0.3025 | Time: 17.1133 s
>>> Epoch [ 85/100]
train_loss: 2.1871 | train_acc: 0.2973 | val_loss: 2.1901 | val_acc: 0.2957 | test_acc: 0.3033 | Time: 18.5844 s
>>> Epoch [ 86/100]
train_loss: 2.1865 | train_acc: 0.2981 | val_loss: 2.1895 | val_acc: 0.2962 | test_acc: 0.3046 | Time: 17.7679 s
>>> Epoch [ 87/100]
train_loss: 2.1859 | train_acc: 0.2988 | val_loss: 2.1889 | val_acc: 0.2970 | test_acc: 0.3054 | Time: 21.0904 s
>>> Epoch [ 88/100]
train_loss: 2.1853 | train_acc: 0.2995 | val_loss: 2.1884 | val_acc: 0.2978 | test_acc: 0.3062 | Time: 20.1782 s
>>> Epoch [ 89/100]
train_loss: 2.1847 | train_acc: 0.3003 | val_loss: 2.1878 | val_acc: 0.2979 | test_acc: 0.3073 | Time: 19.1267 s
>>> Epoch [ 90/100]
train_loss: 2.1842 | train_acc: 0.3010 | val_loss: 2.1872 | val_acc: 0.2987 | test_acc: 0.3082 | Time: 21.0929 s
>>> Epoch [ 91/100]
train_loss: 2.1836 | train_acc: 0.3016 | val_loss: 2.1867 | val_acc: 0.3000 | test_acc: 0.3086 | Time: 19.9188 s
>>> Epoch [ 92/100]
train_loss: 2.1830 | train_acc: 0.3021 | val_loss: 2.1861 | val_acc: 0.3008 | test_acc: 0.3094 | Time: 19.5111 s
>>> Epoch [ 93/100]
train_loss: 2.1825 | train_acc: 0.3027 | val_loss: 2.1856 | val_acc: 0.3014 | test_acc: 0.3097 | Time: 20.0992 s
>>> Epoch [ 94/100]
train_loss: 2.1819 | train_acc: 0.3030 | val_loss: 2.1851 | val_acc: 0.3022 | test_acc: 0.3102 | Time: 20.0431 s
>>> Epoch [ 95/100]
train_loss: 2.1814 | train_acc: 0.3037 | val_loss: 2.1845 | val_acc: 0.3031 | test_acc: 0.3105 | Time: 19.5783 s
>>> Epoch [ 96/100]
train_loss: 2.1809 | train_acc: 0.3040 | val_loss: 2.1840 | val_acc: 0.3040 | test_acc: 0.3115 | Time: 33.3424 s
>>> Epoch [ 97/100]
train_loss: 2.1803 | train_acc: 0.3047 | val_loss: 2.1835 | val_acc: 0.3041 | test_acc: 0.3118 | Time: 19.2803 s
>>> Epoch [ 98/100]
train_loss: 2.1798 | train_acc: 0.3050 | val_loss: 2.1830 | val_acc: 0.3048 | test_acc: 0.3119 | Time: 18.8429 s
>>> Epoch [ 99/100]
train_loss: 2.1793 | train_acc: 0.3056 | val_loss: 2.1825 | val_acc: 0.3053 | test_acc: 0.3124 | Time: 20.5580 s
Fit completed in 28.4273 seconds
Saving the graph to ../logs/cifar10/101_tf_linear_raw_gradtest/exp_no_005/linear_classifier.pb
INFO:tensorflow:Froze 2 variables.
Converted 2 variables to const ops.
11 ops in the final graph.
Length of train loss          : 0
Length of train accuracy      : 0
Length of validate loss       : 0
Length of validate accuracy   : 0
Length of test accuracy       : 0
Restoring training from epoch : 0
>>> Epoch [  0/100]
train_loss: 2.3026 | train_acc: 0.0979 | val_loss: 2.3006 | val_acc: 0.2505 | test_acc: 0.2550 | Time: 60.6848 s
>>> Epoch [  1/100]
train_loss: 2.3006 | train_acc: 0.2473 | val_loss: 2.2986 | val_acc: 0.2508 | test_acc: 0.2548 | Time: 43.5739 s
>>> Epoch [  2/100]
train_loss: 2.2986 | train_acc: 0.2467 | val_loss: 2.2964 | val_acc: 0.2499 | test_acc: 0.2539 | Time: 53.7449 s
>>> Epoch [  3/100]
train_loss: 2.2965 | train_acc: 0.2459 | val_loss: 2.2942 | val_acc: 0.2496 | test_acc: 0.2536 | Time: 45.1331 s
>>> Epoch [  4/100]
train_loss: 2.2943 | train_acc: 0.2453 | val_loss: 2.2919 | val_acc: 0.2496 | test_acc: 0.2526 | Time: 43.3296 s
>>> Epoch [  5/100]
train_loss: 2.2920 | train_acc: 0.2443 | val_loss: 2.2894 | val_acc: 0.2489 | test_acc: 0.2508 | Time: 44.1418 s
>>> Epoch [  6/100]
train_loss: 2.2896 | train_acc: 0.2432 | val_loss: 2.2869 | val_acc: 0.2484 | test_acc: 0.2497 | Time: 43.4014 s
>>> Epoch [  7/100]
train_loss: 2.2871 | train_acc: 0.2422 | val_loss: 2.2843 | val_acc: 0.2472 | test_acc: 0.2484 | Time: 45.3544 s
>>> Epoch [  8/100]
train_loss: 2.2846 | train_acc: 0.2408 | val_loss: 2.2817 | val_acc: 0.2464 | test_acc: 0.2469 | Time: 51.2347 s
>>> Epoch [  9/100]
train_loss: 2.2820 | train_acc: 0.2402 | val_loss: 2.2790 | val_acc: 0.2461 | test_acc: 0.2463 | Time: 48.6211 s
>>> Epoch [ 10/100]
train_loss: 2.2794 | train_acc: 0.2400 | val_loss: 2.2763 | val_acc: 0.2458 | test_acc: 0.2458 | Time: 39.1773 s
>>> Epoch [ 11/100]
train_loss: 2.2768 | train_acc: 0.2395 | val_loss: 2.2737 | val_acc: 0.2455 | test_acc: 0.2453 | Time: 37.4357 s
>>> Epoch [ 12/100]
train_loss: 2.2743 | train_acc: 0.2394 | val_loss: 2.2711 | val_acc: 0.2459 | test_acc: 0.2462 | Time: 39.6958 s
>>> Epoch [ 13/100]
train_loss: 2.2718 | train_acc: 0.2396 | val_loss: 2.2687 | val_acc: 0.2471 | test_acc: 0.2462 | Time: 49.8958 s
>>> Epoch [ 14/100]
train_loss: 2.2694 | train_acc: 0.2401 | val_loss: 2.2663 | val_acc: 0.2473 | test_acc: 0.2470 | Time: 49.1373 s
>>> Epoch [ 15/100]
train_loss: 2.2671 | train_acc: 0.2405 | val_loss: 2.2640 | val_acc: 0.2474 | test_acc: 0.2478 | Time: 36.7279 s
>>> Epoch [ 16/100]
train_loss: 2.2648 | train_acc: 0.2415 | val_loss: 2.2618 | val_acc: 0.2484 | test_acc: 0.2490 | Time: 40.6517 s
>>> Epoch [ 17/100]
train_loss: 2.2627 | train_acc: 0.2422 | val_loss: 2.2597 | val_acc: 0.2489 | test_acc: 0.2503 | Time: 44.7314 s
>>> Epoch [ 18/100]
train_loss: 2.2606 | train_acc: 0.2433 | val_loss: 2.2576 | val_acc: 0.2497 | test_acc: 0.2521 | Time: 44.7096 s
>>> Epoch [ 19/100]
train_loss: 2.2585 | train_acc: 0.2445 | val_loss: 2.2557 | val_acc: 0.2505 | test_acc: 0.2529 | Time: 43.7965 s
>>> Epoch [ 20/100]
train_loss: 2.2566 | train_acc: 0.2452 | val_loss: 2.2537 | val_acc: 0.2508 | test_acc: 0.2536 | Time: 43.2809 s
>>> Epoch [ 21/100]
train_loss: 2.2546 | train_acc: 0.2461 | val_loss: 2.2518 | val_acc: 0.2517 | test_acc: 0.2550 | Time: 55.7572 s
>>> Epoch [ 22/100]
train_loss: 2.2528 | train_acc: 0.2473 | val_loss: 2.2500 | val_acc: 0.2524 | test_acc: 0.2558 | Time: 40.3727 s
>>> Epoch [ 23/100]
train_loss: 2.2509 | train_acc: 0.2483 | val_loss: 2.2482 | val_acc: 0.2527 | test_acc: 0.2564 | Time: 42.7284 s
>>> Epoch [ 24/100]
train_loss: 2.2491 | train_acc: 0.2493 | val_loss: 2.2465 | val_acc: 0.2535 | test_acc: 0.2569 | Time: 44.4733 s
>>> Epoch [ 25/100]
train_loss: 2.2473 | train_acc: 0.2501 | val_loss: 2.2447 | val_acc: 0.2539 | test_acc: 0.2574 | Time: 43.3370 s
>>> Epoch [ 26/100]
train_loss: 2.2456 | train_acc: 0.2515 | val_loss: 2.2431 | val_acc: 0.2549 | test_acc: 0.2580 | Time: 44.1629 s
>>> Epoch [ 27/100]
train_loss: 2.2439 | train_acc: 0.2526 | val_loss: 2.2414 | val_acc: 0.2552 | test_acc: 0.2593 | Time: 43.2746 s
>>> Epoch [ 28/100]
train_loss: 2.2422 | train_acc: 0.2533 | val_loss: 2.2398 | val_acc: 0.2569 | test_acc: 0.2605 | Time: 60.9598 s
>>> Epoch [ 29/100]
train_loss: 2.2406 | train_acc: 0.2539 | val_loss: 2.2383 | val_acc: 0.2576 | test_acc: 0.2616 | Time: 42.1379 s
>>> Epoch [ 30/100]
train_loss: 2.2390 | train_acc: 0.2549 | val_loss: 2.2367 | val_acc: 0.2587 | test_acc: 0.2626 | Time: 44.1436 s
>>> Epoch [ 31/100]
train_loss: 2.2374 | train_acc: 0.2557 | val_loss: 2.2352 | val_acc: 0.2597 | test_acc: 0.2624 | Time: 43.9115 s
>>> Epoch [ 32/100]
train_loss: 2.2359 | train_acc: 0.2565 | val_loss: 2.2338 | val_acc: 0.2607 | test_acc: 0.2634 | Time: 43.8862 s
>>> Epoch [ 33/100]
train_loss: 2.2344 | train_acc: 0.2573 | val_loss: 2.2324 | val_acc: 0.2605 | test_acc: 0.2640 | Time: 43.3607 s
>>> Epoch [ 34/100]
train_loss: 2.2330 | train_acc: 0.2580 | val_loss: 2.2310 | val_acc: 0.2606 | test_acc: 0.2648 | Time: 61.9608 s
>>> Epoch [ 35/100]
train_loss: 2.2315 | train_acc: 0.2587 | val_loss: 2.2296 | val_acc: 0.2616 | test_acc: 0.2654 | Time: 41.9852 s
>>> Epoch [ 36/100]
train_loss: 2.2301 | train_acc: 0.2594 | val_loss: 2.2283 | val_acc: 0.2622 | test_acc: 0.2662 | Time: 43.5383 s
>>> Epoch [ 37/100]
train_loss: 2.2288 | train_acc: 0.2601 | val_loss: 2.2270 | val_acc: 0.2626 | test_acc: 0.2665 | Time: 43.8252 s
>>> Epoch [ 38/100]
train_loss: 2.2274 | train_acc: 0.2608 | val_loss: 2.2257 | val_acc: 0.2633 | test_acc: 0.2678 | Time: 43.1059 s
>>> Epoch [ 39/100]
train_loss: 2.2262 | train_acc: 0.2620 | val_loss: 2.2245 | val_acc: 0.2641 | test_acc: 0.2685 | Time: 44.5239 s
>>> Epoch [ 40/100]
train_loss: 2.2249 | train_acc: 0.2627 | val_loss: 2.2233 | val_acc: 0.2643 | test_acc: 0.2696 | Time: 33.9660 s
>>> Epoch [ 41/100]
train_loss: 2.2237 | train_acc: 0.2634 | val_loss: 2.2221 | val_acc: 0.2650 | test_acc: 0.2702 | Time: 33.5877 s
>>> Epoch [ 42/100]
train_loss: 2.2224 | train_acc: 0.2644 | val_loss: 2.2210 | val_acc: 0.2656 | test_acc: 0.2710 | Time: 33.2732 s
>>> Epoch [ 43/100]
train_loss: 2.2213 | train_acc: 0.2649 | val_loss: 2.2198 | val_acc: 0.2666 | test_acc: 0.2715 | Time: 33.7399 s
>>> Epoch [ 44/100]
train_loss: 2.2201 | train_acc: 0.2657 | val_loss: 2.2187 | val_acc: 0.2677 | test_acc: 0.2723 | Time: 33.5864 s
>>> Epoch [ 45/100]
train_loss: 2.2190 | train_acc: 0.2667 | val_loss: 2.2177 | val_acc: 0.2682 | test_acc: 0.2735 | Time: 33.3847 s
>>> Epoch [ 46/100]
train_loss: 2.2179 | train_acc: 0.2675 | val_loss: 2.2166 | val_acc: 0.2686 | test_acc: 0.2739 | Time: 74.3793 s
>>> Epoch [ 47/100]
train_loss: 2.2168 | train_acc: 0.2681 | val_loss: 2.2156 | val_acc: 0.2688 | test_acc: 0.2748 | Time: 51.4575 s
>>> Epoch [ 48/100]
train_loss: 2.2158 | train_acc: 0.2689 | val_loss: 2.2146 | val_acc: 0.2697 | test_acc: 0.2753 | Time: 44.6544 s
>>> Epoch [ 49/100]
train_loss: 2.2147 | train_acc: 0.2696 | val_loss: 2.2136 | val_acc: 0.2700 | test_acc: 0.2764 | Time: 43.5269 s
>>> Epoch [ 50/100]
train_loss: 2.2137 | train_acc: 0.2704 | val_loss: 2.2126 | val_acc: 0.2700 | test_acc: 0.2778 | Time: 42.6391 s
>>> Epoch [ 51/100]
train_loss: 2.2127 | train_acc: 0.2711 | val_loss: 2.2117 | val_acc: 0.2706 | test_acc: 0.2791 | Time: 43.7526 s
>>> Epoch [ 52/100]
train_loss: 2.2118 | train_acc: 0.2720 | val_loss: 2.2108 | val_acc: 0.2709 | test_acc: 0.2799 | Time: 43.7113 s
>>> Epoch [ 53/100]
train_loss: 2.2108 | train_acc: 0.2729 | val_loss: 2.2099 | val_acc: 0.2721 | test_acc: 0.2804 | Time: 57.6321 s
>>> Epoch [ 54/100]
train_loss: 2.2099 | train_acc: 0.2738 | val_loss: 2.2090 | val_acc: 0.2725 | test_acc: 0.2814 | Time: 49.3133 s
>>> Epoch [ 55/100]
train_loss: 2.2090 | train_acc: 0.2745 | val_loss: 2.2081 | val_acc: 0.2735 | test_acc: 0.2824 | Time: 46.1145 s
>>> Epoch [ 56/100]
train_loss: 2.2081 | train_acc: 0.2751 | val_loss: 2.2073 | val_acc: 0.2742 | test_acc: 0.2834 | Time: 50.3338 s
>>> Epoch [ 57/100]
train_loss: 2.2072 | train_acc: 0.2760 | val_loss: 2.2064 | val_acc: 0.2750 | test_acc: 0.2845 | Time: 38.9842 s
>>> Epoch [ 58/100]
train_loss: 2.2063 | train_acc: 0.2770 | val_loss: 2.2056 | val_acc: 0.2760 | test_acc: 0.2860 | Time: 38.8326 s
>>> Epoch [ 59/100]
train_loss: 2.2055 | train_acc: 0.2777 | val_loss: 2.2048 | val_acc: 0.2774 | test_acc: 0.2860 | Time: 40.6640 s
>>> Epoch [ 60/100]
train_loss: 2.2046 | train_acc: 0.2785 | val_loss: 2.2040 | val_acc: 0.2779 | test_acc: 0.2866 | Time: 43.6274 s
>>> Epoch [ 61/100]
train_loss: 2.2038 | train_acc: 0.2791 | val_loss: 2.2032 | val_acc: 0.2783 | test_acc: 0.2876 | Time: 44.4511 s
>>> Epoch [ 62/100]
train_loss: 2.2030 | train_acc: 0.2797 | val_loss: 2.2024 | val_acc: 0.2792 | test_acc: 0.2882 | Time: 44.1959 s
>>> Epoch [ 63/100]
train_loss: 2.2022 | train_acc: 0.2808 | val_loss: 2.2017 | val_acc: 0.2805 | test_acc: 0.2895 | Time: 43.9087 s
>>> Epoch [ 64/100]
train_loss: 2.2014 | train_acc: 0.2815 | val_loss: 2.2009 | val_acc: 0.2812 | test_acc: 0.2896 | Time: 45.0876 s
>>> Epoch [ 65/100]
train_loss: 2.2007 | train_acc: 0.2824 | val_loss: 2.2002 | val_acc: 0.2817 | test_acc: 0.2909 | Time: 57.7906 s
>>> Epoch [ 66/100]
train_loss: 2.1999 | train_acc: 0.2833 | val_loss: 2.1995 | val_acc: 0.2826 | test_acc: 0.2918 | Time: 37.2683 s
>>> Epoch [ 67/100]
train_loss: 2.1992 | train_acc: 0.2845 | val_loss: 2.1987 | val_acc: 0.2832 | test_acc: 0.2928 | Time: 37.4493 s
>>> Epoch [ 68/100]
train_loss: 2.1984 | train_acc: 0.2857 | val_loss: 2.1980 | val_acc: 0.2840 | test_acc: 0.2932 | Time: 37.6576 s
>>> Epoch [ 69/100]
train_loss: 2.1977 | train_acc: 0.2863 | val_loss: 2.1973 | val_acc: 0.2847 | test_acc: 0.2938 | Time: 39.6372 s
>>> Epoch [ 70/100]
train_loss: 2.1970 | train_acc: 0.2872 | val_loss: 2.1966 | val_acc: 0.2853 | test_acc: 0.2944 | Time: 42.3736 s
>>> Epoch [ 71/100]
train_loss: 2.1963 | train_acc: 0.2879 | val_loss: 2.1960 | val_acc: 0.2861 | test_acc: 0.2950 | Time: 43.5457 s
>>> Epoch [ 72/100]
train_loss: 2.1956 | train_acc: 0.2884 | val_loss: 2.1953 | val_acc: 0.2868 | test_acc: 0.2964 | Time: 43.1660 s
>>> Epoch [ 73/100]
train_loss: 2.1949 | train_acc: 0.2895 | val_loss: 2.1946 | val_acc: 0.2879 | test_acc: 0.2974 | Time: 43.9395 s
>>> Epoch [ 74/100]
train_loss: 2.1942 | train_acc: 0.2903 | val_loss: 2.1940 | val_acc: 0.2893 | test_acc: 0.2976 | Time: 44.3771 s
>>> Epoch [ 75/100]
train_loss: 2.1935 | train_acc: 0.2911 | val_loss: 2.1933 | val_acc: 0.2900 | test_acc: 0.2988 | Time: 57.9009 s
>>> Epoch [ 76/100]
train_loss: 2.1928 | train_acc: 0.2920 | val_loss: 2.1927 | val_acc: 0.2911 | test_acc: 0.2996 | Time: 39.5990 s
>>> Epoch [ 77/100]
train_loss: 2.1922 | train_acc: 0.2927 | val_loss: 2.1921 | val_acc: 0.2912 | test_acc: 0.2996 | Time: 43.2914 s
>>> Epoch [ 78/100]
train_loss: 2.1915 | train_acc: 0.2936 | val_loss: 2.1914 | val_acc: 0.2915 | test_acc: 0.3004 | Time: 43.2620 s
>>> Epoch [ 79/100]
train_loss: 2.1909 | train_acc: 0.2941 | val_loss: 2.1908 | val_acc: 0.2915 | test_acc: 0.3003 | Time: 43.9812 s
>>> Epoch [ 80/100]
train_loss: 2.1902 | train_acc: 0.2948 | val_loss: 2.1902 | val_acc: 0.2925 | test_acc: 0.3010 | Time: 43.5930 s
>>> Epoch [ 81/100]
train_loss: 2.1896 | train_acc: 0.2957 | val_loss: 2.1896 | val_acc: 0.2932 | test_acc: 0.3019 | Time: 44.2097 s
>>> Epoch [ 82/100]
train_loss: 2.1890 | train_acc: 0.2967 | val_loss: 2.1890 | val_acc: 0.2936 | test_acc: 0.3035 | Time: 58.4575 s
>>> Epoch [ 83/100]
train_loss: 2.1884 | train_acc: 0.2977 | val_loss: 2.1884 | val_acc: 0.2945 | test_acc: 0.3047 | Time: 39.1737 s
>>> Epoch [ 84/100]
train_loss: 2.1878 | train_acc: 0.2984 | val_loss: 2.1878 | val_acc: 0.2951 | test_acc: 0.3051 | Time: 39.2096 s
>>> Epoch [ 85/100]
train_loss: 2.1872 | train_acc: 0.2994 | val_loss: 2.1873 | val_acc: 0.2957 | test_acc: 0.3052 | Time: 41.4314 s
>>> Epoch [ 86/100]
train_loss: 2.1866 | train_acc: 0.3003 | val_loss: 2.1867 | val_acc: 0.2964 | test_acc: 0.3062 | Time: 42.9344 s
>>> Epoch [ 87/100]
train_loss: 2.1860 | train_acc: 0.3007 | val_loss: 2.1861 | val_acc: 0.2975 | test_acc: 0.3067 | Time: 43.6837 s
>>> Epoch [ 88/100]
train_loss: 2.1854 | train_acc: 0.3013 | val_loss: 2.1856 | val_acc: 0.2986 | test_acc: 0.3074 | Time: 43.5536 s
>>> Epoch [ 89/100]
train_loss: 2.1848 | train_acc: 0.3018 | val_loss: 2.1850 | val_acc: 0.2988 | test_acc: 0.3086 | Time: 43.8496 s
>>> Epoch [ 90/100]
train_loss: 2.1842 | train_acc: 0.3025 | val_loss: 2.1845 | val_acc: 0.2994 | test_acc: 0.3089 | Time: 48.6570 s
>>> Epoch [ 91/100]
train_loss: 2.1836 | train_acc: 0.3032 | val_loss: 2.1839 | val_acc: 0.3002 | test_acc: 0.3089 | Time: 50.0362 s
>>> Epoch [ 92/100]
train_loss: 2.1831 | train_acc: 0.3038 | val_loss: 2.1834 | val_acc: 0.3013 | test_acc: 0.3100 | Time: 44.0780 s
>>> Epoch [ 93/100]
train_loss: 2.1825 | train_acc: 0.3042 | val_loss: 2.1828 | val_acc: 0.3023 | test_acc: 0.3104 | Time: 44.1627 s
>>> Epoch [ 94/100]
train_loss: 2.1820 | train_acc: 0.3050 | val_loss: 2.1823 | val_acc: 0.3029 | test_acc: 0.3106 | Time: 43.2783 s
>>> Epoch [ 95/100]
train_loss: 2.1814 | train_acc: 0.3058 | val_loss: 2.1818 | val_acc: 0.3031 | test_acc: 0.3109 | Time: 44.2145 s
>>> Epoch [ 96/100]
train_loss: 2.1809 | train_acc: 0.3060 | val_loss: 2.1813 | val_acc: 0.3034 | test_acc: 0.3112 | Time: 43.3772 s
>>> Epoch [ 97/100]
train_loss: 2.1803 | train_acc: 0.3068 | val_loss: 2.1808 | val_acc: 0.3040 | test_acc: 0.3117 | Time: 44.4778 s
>>> Epoch [ 98/100]
train_loss: 2.1798 | train_acc: 0.3071 | val_loss: 2.1803 | val_acc: 0.3043 | test_acc: 0.3127 | Time: 45.2539 s
>>> Epoch [ 99/100]
train_loss: 2.1793 | train_acc: 0.3073 | val_loss: 2.1798 | val_acc: 0.3061 | test_acc: 0.3131 | Time: 43.1074 s
Fit completed in 56.8843 seconds
Saving the graph to ../logs/cifar10/101_tf_linear_raw_gradtest/exp_no_006/linear_classifier.pb
INFO:tensorflow:Froze 2 variables.
Converted 2 variables to const ops.
11 ops in the final graph.

Step 2.4: Make the predictions



In [21]:

    
# prediction_numbers = lc.predict(cifar10.test.data)
# prediction_classes = []
# num_test_images = cifar10.test.data.shape[0]
# for i in range(num_test_images):
#     prediction_classes.append(cifar10.classes[int(prediction_numbers[i])])



In [22]:

    
# cifar10.plot_images(cifar10.test.data[:50], cifar10.test.class_names[:50], cls_pred=prediction_classes[:50], 
#                     nrows=5, ncols=10, fig_size=(20,50), fontsize=30, convert=True)

Step 2.5: Print the results



In [23]:

    
for index, descent_method in enumerate(descent_methods):
    test_accuracy = lcs[index].score(cifar10.test.data, cifar10.test.class_labels)
    print('Accuracy of the linear classifier on test dataset using %s: %.4f' % (descent_method, test_accuracy))









    



Accuracy of the linear classifier on test dataset using adam: 0.3126
Accuracy of the linear classifier on test dataset using gradient: 0.3094
Accuracy of the linear classifier on test dataset using adaprop: 0.3127
Accuracy of the linear classifier on test dataset using rmsprop: 0.3126
Accuracy of the linear classifier on test dataset using adadelta: 0.3124
Accuracy of the linear classifier on test dataset using momentum: 0.3131



In [24]:

    
# lc.print_classification_results(cifar10.test.data, cifar10.test.one_hot_labels, cifar10.test.class_labels,
#                                 test_class_names=cifar10.classes, normalize=True)

Step 2.6: Plot results



In [25]:

    
for index, descent_method in enumerate(descent_methods):
    if index == 0:
        train_losses = lcs[index].out_params['list_train_loss']
        val_losses = lcs[index].out_params['list_val_loss']
        train_acc = lcs[index].out_params['list_train_acc']
        val_acc = lcs[index].out_params['list_val_acc']
    else:
        train_losses = np.vstack((train_losses, lcs[index].out_params['list_train_loss']))
        val_losses = np.vstack((val_losses, lcs[index].out_params['list_val_loss']))
        train_acc = np.vstack((train_acc, lcs[index].out_params['list_train_acc']))
        val_acc = np.vstack((val_acc, lcs[index].out_params['list_val_acc']))



In [26]:

    
colors = []
plot.plot_scores(train_losses, legend=descent_methods, colors=colors,
                 plot_title='Plot of train loss in various gradient descent methods',
                 plot_xlabel='No. of iterations', plot_ylabel='Train Loss')



In [27]:

    
colors = []
plot.plot_scores(val_losses, legend=descent_methods, colors=colors,
                 plot_title='Plot of val. loss in various gradient descent methods',
                 plot_xlabel='No. of iterations', plot_ylabel='Val. Loss')



In [33]:

    
colors = []
plot.plot_scores(train_acc, legend=descent_methods, colors=colors,
                 plot_title='Plot of train accuracy in various gradient descent methods',
                 plot_xlabel='No. of iterations', plot_ylabel='Train Accuracy')



In [34]:

    
colors = []
plot.plot_scores(val_acc, legend=descent_methods, colors=colors,
                 plot_title='Plot of val. accuracy in various gradient descent methods',
                 plot_xlabel='No. of iterations', plot_ylabel='Val. Accuracy')



In [31]:

    
for index, descent_method in enumerate(descent_methods):
    print('Plot for descent method: %s' % descent_method)
    lcs[index].plot_weights(classes=cifar10.classes)









    



Plot for descent method: adam






    












    



Plot for descent method: gradient






    












    



Plot for descent method: adaprop






    












    



Plot for descent method: rmsprop






    












    



Plot for descent method: adadelta






    












    



Plot for descent method: momentum

Step 2.7: Close the session



In [32]:

    
for index, descent_method in enumerate(descent_methods):
    lcs[index].close()









    



---------------------------------------------------------------------------
AssertionError                            Traceback (most recent call last)
<ipython-input-32-d5dd7f292f72> in <module>()
      1 for index, descent_method in enumerate(descent_methods):
----> 2     lcs[index].close()

/net/voxel03/misc/me/praneethas/PycharmProjects/object-recognition/library/tf/models/linear_classifier.py in close(self)
    535 
    536     def close(self):
--> 537         self.session.close()
    538 
    539     def __exit__(self, exc_type, exc_val, exc_tb):

/net/voxel03/misc/me/praneethas/Softwares/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py in close(self)
   1314     if self._explicit_graph is not None:
   1315       self._default_graph.__exit__(None, None, None)
-> 1316     self._default_session.__exit__(None, None, None)

/net/voxel03/misc/me/praneethas/Softwares/anaconda3/lib/python3.6/contextlib.py in __exit__(self, type, value, traceback)
     87         if type is None:
     88             try:
---> 89                 next(self.gen)
     90             except StopIteration:
     91                 return

/net/voxel03/misc/me/praneethas/Softwares/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/ops.py in get_controller(self, default)
   3617           raise AssertionError(
   3618               "Nesting violated for default stack of %s objects"
-> 3619               % type(default))
   3620         self.stack.pop()
   3621       else:

AssertionError: Nesting violated for default stack of <class 'tensorflow.python.client.session.InteractiveSession'> objects

Step 3: Write to file



In [ ]:

    
def output_HTML(read_file, output_file):
    from nbconvert import HTMLExporter
    import codecs
    import nbformat
    exporter = HTMLExporter()
    output_notebook = nbformat.read(read_file, as_version=4)
    print()
    output, resources = exporter.from_notebook_node(output_notebook)
    codecs.open(output_file, 'w', encoding='utf-8').write(output)



In [ ]:

    
%%javascript
var notebook = IPython.notebook
notebook.save_notebook()



In [ ]:

    
%%javascript
var kernel = IPython.notebook.kernel;
var thename = window.document.getElementById("notebook_name").innerHTML;
var command = "theNotebook = " + "'"+thename+"'";
kernel.execute(command);



In [ ]:

    
current_file = './' + theNotebook + '.ipynb'
for index, descent_method in enumerate(descent_methods):
    output_file = log_dirs[index] + str(file_no).zfill(3) + '_exp_no_' + str(index+1).zfill(3) + '_' + theNotebook + '.html'
    print('Current file: ' + str(current_file))
    print('Output file: ' + str(output_file))
    file_utils.mkdir_p(log_dirs[index]) 
    output_HTML(current_file, output_file)



In [ ]: