notebook.community

Edit and run

Language Translation

In this project, you’re going to take a peek into the realm of neural network machine translation. You’ll be training a sequence to sequence model on a dataset of English and French sentences that can translate new sentences from English to French.

Get the Data

Since translating the whole language of English to French will take lots of time to train, we have provided you with a small portion of the English corpus.





In [2]:



    


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

source_path = 'data/small_vocab_en'
target_path = 'data/small_vocab_fr'
source_text = helper.load_data(source_path)
target_text = helper.load_data(target_path)

Explore the Data

Play around with view_sentence_range to view different parts of the data.





In [3]:



    


view_sentence_range = (0, 10)

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import numpy as np

print('Dataset Stats')
print('Roughly the number of unique words: {}'.format(len({word: None for word in source_text.split()})))

sentences = source_text.split('\n')
word_counts = [len(sentence.split()) for sentence in sentences]
print('Number of sentences: {}'.format(len(sentences)))
print('Average number of words in a sentence: {}'.format(np.average(word_counts)))

print()
print('English sentences {} to {}:'.format(*view_sentence_range))
print('\n'.join(source_text.split('\n')[view_sentence_range[0]:view_sentence_range[1]]))
print()
print('French sentences {} to {}:'.format(*view_sentence_range))
print('\n'.join(target_text.split('\n')[view_sentence_range[0]:view_sentence_range[1]]))



    


















    






Dataset Stats
Roughly the number of unique words: 227
Number of sentences: 137861
Average number of words in a sentence: 13.225277634719028

English sentences 0 to 10:
new jersey is sometimes quiet during autumn , and it is snowy in april .
the united states is usually chilly during july , and it is usually freezing in november .
california is usually quiet during march , and it is usually hot in june .
the united states is sometimes mild during june , and it is cold in september .
your least liked fruit is the grape , but my least liked is the apple .
his favorite fruit is the orange , but my favorite is the grape .
paris is relaxing during december , but it is usually chilly in july .
new jersey is busy during spring , and it is never hot in march .
our least liked fruit is the lemon , but my least liked is the grape .
the united states is sometimes busy during january , and it is sometimes warm in november .

French sentences 0 to 10:
new jersey est parfois calme pendant l' automne , et il est neigeux en avril .
les états-unis est généralement froid en juillet , et il gèle habituellement en novembre .
california est généralement calme en mars , et il est généralement chaud en juin .
les états-unis est parfois légère en juin , et il fait froid en septembre .
votre moins aimé fruit est le raisin , mais mon moins aimé est la pomme .
son fruit préféré est l'orange , mais mon préféré est le raisin .
paris est relaxant en décembre , mais il est généralement froid en juillet .
new jersey est occupé au printemps , et il est jamais chaude en mars .
notre fruit est moins aimé le citron , mais mon moins aimé est le raisin .
les états-unis est parfois occupé en janvier , et il est parfois chaud en novembre .

Implement Preprocessing Function

Text to Word Ids

As you did with other RNNs, you must turn the text into a number so the computer can understand it. In the function text_to_ids(), you'll turn source_text and target_text from words to ids. However, you need to add the <EOS> word id at the end of each sentence from target_text. This will help the neural network predict when the sentence should end.

You can get the <EOS> word id by doing:

target_vocab_to_int['<EOS>']

You can get other word ids using source_vocab_to_int and target_vocab_to_int.





In [4]:



    


def text_to_ids(source_text, target_text, source_vocab_to_int, target_vocab_to_int):
    """
    Convert source and target text to proper word ids
    :param source_text: String that contains all the source text.
    :param target_text: String that contains all the target text.
    :param source_vocab_to_int: Dictionary to go from the source words to an id
    :param target_vocab_to_int: Dictionary to go from the target words to an id
    :return: A tuple of lists (source_id_text, target_id_text)
    """
    split = source_text.split('\n')
    sentences = [sentence + ' <EOS>' for sentence in target_text.split('\n')]
    sources = list(map(lambda x: [source_vocab_to_int[word] for word in x.split()], split))
    targets = list(map(lambda x: [target_vocab_to_int[word] for word in x.split()], sentences))
    return sources, targets

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



    


















    






Tests Passed

Preprocess all the data and save it

Running the code cell below will preprocess all the data and save it to file.





In [5]:



    


"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
helper.preprocess_and_save_data(source_path, target_path, text_to_ids)

Check Point

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





In [6]:



    


"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import numpy as np
import helper

(source_int_text, target_int_text), (source_vocab_to_int, target_vocab_to_int), _ = helper.load_preprocess()

Check the Version of TensorFlow and Access to GPU

This will check to make sure you have the correct version of TensorFlow and access to a GPU





In [7]:



    


"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
from distutils.version import LooseVersion
import warnings
import tensorflow as tf

# Check TensorFlow Version
assert LooseVersion(tf.__version__) in [LooseVersion('1.0.0'), LooseVersion('1.0.1')], 'This project requires TensorFlow version 1.0  You are using {}'.format(tf.__version__)
print('TensorFlow Version: {}'.format(tf.__version__))

# Check for a GPU
if not tf.test.gpu_device_name():
    warnings.warn('No GPU found. Please use a GPU to train your neural network.')
else:
    print('Default GPU Device: {}'.format(tf.test.gpu_device_name()))



    


















    






TensorFlow Version: 1.0.0
Default GPU Device: /gpu:0

Build the Neural Network

You'll build the components necessary to build a Sequence-to-Sequence model by implementing the following functions below:

  • model_inputs
  • process_decoding_input
  • encoding_layer
  • decoding_layer_train
  • decoding_layer_infer
  • decoding_layer
  • seq2seq_model

Input

Implement the model_inputs() function to create TF Placeholders for the Neural Network. It should create the following placeholders:

  • Input text placeholder named "input" using the TF Placeholder name parameter with rank 2.
  • Targets placeholder with rank 2.
  • Learning rate placeholder with rank 0.
  • Keep probability placeholder named "keep_prob" using the TF Placeholder name parameter with rank 0.

Return the placeholders in the following the tuple (Input, Targets, Learing Rate, Keep Probability)





In [8]:



    


def model_inputs():
    """
    Create TF Placeholders for input, targets, and learning rate.
    :return: Tuple (input, targets, learning rate, keep probability)
    """
    inputs = tf.placeholder(tf.int32, [None, None], name='input')
    targets = tf.placeholder(tf.int32, [None, None], name='targets')
    lr = tf.placeholder(tf.float32, name='lr')    
    keep_probability = tf.placeholder(tf.float32, name='keep_prob')
    return inputs, targets, lr, keep_probability

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



    


















    






Tests Passed

Process Decoding Input

Implement process_decoding_input using TensorFlow to remove the last word id from each batch in target_data and concat the GO ID to the beginning of each batch.





In [9]:



    


def process_decoding_input(target_data, target_vocab_to_int, batch_size):
    """
    Preprocess target data for decoding
    :param target_data: Target Placeholder
    :param target_vocab_to_int: Dictionary to go from the target words to an id
    :param batch_size: Batch Size
    :return: Preprocessed target data
    """
    #print(target_data.shape)
    length = target_data.shape[0]
    
    slices = tf.strided_slice(target_data, [0,0], [batch_size, -1], [1,1])
    #print(slices.shape)
    
    concat_id = target_vocab_to_int['<GO>']
    preprocessed = tf.concat([tf.fill([batch_size, 1], concat_id), slices], 1)
    #print(preprocessed.shape) #Must be same shape as target_data.shape
    
    return preprocessed

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



    


















    






Tests Passed

Encoding

Implement encoding_layer() to create a Encoder RNN layer using tf.nn.dynamic_rnn().





In [10]:



    


def encoding_layer(rnn_inputs, rnn_size, num_layers, keep_prob):
    """
    Create encoding layer
    :param rnn_inputs: Inputs for the RNN
    :param rnn_size: RNN Size
    :param num_layers: Number of layers
    :param keep_prob: Dropout keep probability
    :return: RNN state
    """
    lstm = tf.contrib.rnn.BasicLSTMCell(num_units=rnn_size)
    cell = tf.contrib.rnn.MultiRNNCell([lstm] * num_layers)
    outputs, state = tf.nn.dynamic_rnn(cell, rnn_inputs, dtype=tf.float32)
    return state
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_encoding_layer(encoding_layer)



    


















    






Tests Passed

Decoding - Training

Create training logits using tf.contrib.seq2seq.simple_decoder_fn_train() and tf.contrib.seq2seq.dynamic_rnn_decoder(). Apply the output_fn to the tf.contrib.seq2seq.dynamic_rnn_decoder() outputs.





In [11]:



    


def decoding_layer_train(encoder_state, dec_cell, dec_embed_input, sequence_length, decoding_scope,
                         output_fn, keep_prob):
    """
    Create a decoding layer for training
    :param encoder_state: Encoder State
    :param dec_cell: Decoder RNN Cell
    :param dec_embed_input: Decoder embedded input
    :param sequence_length: Sequence Length
    :param decoding_scope: TenorFlow Variable Scope for decoding
    :param output_fn: Function to apply the output layer
    :param keep_prob: Dropout keep probability
    :return: Train Logits
    """
    decoder_fn = tf.contrib.seq2seq.simple_decoder_fn_train(encoder_state, name='decoder_fn_train')
    dropout = tf.contrib.rnn.DropoutWrapper(dec_cell, keep_prob)
    outputs, final_state, final_context_state = tf.contrib.seq2seq.dynamic_rnn_decoder(
        dropout, 
        decoder_fn, 
        inputs=dec_embed_input,
        sequence_length=sequence_length,
        scope=decoding_scope
    )
    return output_fn(outputs)

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



    


















    






Tests Passed

Decoding - Inference

Create inference logits using tf.contrib.seq2seq.simple_decoder_fn_inference() and tf.contrib.seq2seq.dynamic_rnn_decoder().





In [12]:



    


def decoding_layer_infer(encoder_state, dec_cell, dec_embeddings, start_of_sequence_id, end_of_sequence_id,
                         maximum_length, vocab_size, decoding_scope, output_fn, keep_prob):
    """
    Create a decoding layer for inference
    :param encoder_state: Encoder state
    :param dec_cell: Decoder RNN Cell
    :param dec_embeddings: Decoder embeddings
    :param start_of_sequence_id: GO ID
    :param end_of_sequence_id: EOS Id
    :param maximum_length: The maximum allowed time steps to decode
    :param vocab_size: Size of vocabulary
    :param decoding_scope: TensorFlow Variable Scope for decoding
    :param output_fn: Function to apply the output layer
    :param keep_prob: Dropout keep probability
    :return: Inference Logits
    """
    decoder_fn = tf.contrib.seq2seq.simple_decoder_fn_inference(
        output_fn=output_fn,
        encoder_state=encoder_state,
        embeddings=dec_embeddings,
        start_of_sequence_id=start_of_sequence_id,
        end_of_sequence_id=end_of_sequence_id,
        maximum_length=maximum_length,
        num_decoder_symbols=vocab_size,
        name='inference_decoder_fn'
    )
    
    outputs, final_state, final_context_state = tf.contrib.seq2seq.dynamic_rnn_decoder(
        cell=dec_cell,
        decoder_fn=decoder_fn,
        scope=decoding_scope,
        name='inference_decoder_rnn'
    )
    return outputs

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



    


















    






Tests Passed

Build the Decoding Layer

Implement decoding_layer() to create a Decoder RNN layer.

  • Create RNN cell for decoding using rnn_size and num_layers.
  • Create the output fuction using lambda to transform it's input, logits, to class logits.
  • Use the your decoding_layer_train(encoder_state, dec_cell, dec_embed_input, sequence_length, decoding_scope, output_fn, keep_prob) function to get the training logits.
  • Use your decoding_layer_infer(encoder_state, dec_cell, dec_embeddings, start_of_sequence_id, end_of_sequence_id, maximum_length, vocab_size, decoding_scope, output_fn, keep_prob) function to get the inference logits.

Note: You'll need to use tf.variable_scope to share variables between training and inference.





In [13]:



    


def decoding_layer(dec_embed_input, dec_embeddings, encoder_state, vocab_size, sequence_length, rnn_size,
                   num_layers, target_vocab_to_int, keep_prob):
    """
    Create decoding layer
    :param dec_embed_input: Decoder embedded input
    :param dec_embeddings: Decoder embeddings
    :param encoder_state: The encoded state
    :param vocab_size: Size of vocabulary
    :param sequence_length: Sequence Length
    :param rnn_size: RNN Size
    :param num_layers: Number of layers
    :param target_vocab_to_int: Dictionary to go from the target words to an id
    :param keep_prob: Dropout keep probability
    :return: Tuple of (Training Logits, Inference Logits)
    """
    lstm = tf.contrib.rnn.BasicLSTMCell(rnn_size)
    cell = tf.contrib.rnn.MultiRNNCell([lstm] * num_layers)
        
    with tf.variable_scope('training') as dec_scope:        
        output_fn = lambda x: tf.contrib.layers.fully_connected(x, vocab_size, None, scope=dec_scope)
        
        train_logits = decoding_layer_train(
            encoder_state,
            cell,
            dec_embed_input,
            sequence_length,
            dec_scope,
            output_fn,
            keep_prob
        )
        
        begin = target_vocab_to_int['<GO>']
        end = target_vocab_to_int['<EOS>']
        dec_scope.reuse_variables()
        inf_logits = decoding_layer_infer(
            encoder_state,
            cell,
            dec_embeddings,
            begin,
            end,
            sequence_length,
            vocab_size,
            dec_scope,
            output_fn,
            keep_prob
        )
        
        
        return train_logits, inf_logits

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



    


















    






Tests Passed

Build the Neural Network

Apply the functions you implemented above to:

  • Apply embedding to the input data for the encoder.
  • Encode the input using your encoding_layer(rnn_inputs, rnn_size, num_layers, keep_prob).
  • Process target data using your process_decoding_input(target_data, target_vocab_to_int, batch_size) function.
  • Apply embedding to the target data for the decoder.
  • Decode the encoded input using your decoding_layer(dec_embed_input, dec_embeddings, encoder_state, vocab_size, sequence_length, rnn_size, num_layers, target_vocab_to_int, keep_prob).




In [14]:



    


def seq2seq_model(input_data, target_data, keep_prob, batch_size, sequence_length, source_vocab_size, target_vocab_size,
                  enc_embedding_size, dec_embedding_size, rnn_size, num_layers, target_vocab_to_int):
    """
    Build the Sequence-to-Sequence part of the neural network
    :param input_data: Input placeholder
    :param target_data: Target placeholder
    :param keep_prob: Dropout keep probability placeholder
    :param batch_size: Batch Size
    :param sequence_length: Sequence Length
    :param source_vocab_size: Source vocabulary size
    :param target_vocab_size: Target vocabulary size
    :param enc_embedding_size: Decoder embedding size
    :param dec_embedding_size: Encoder embedding size
    :param rnn_size: RNN Size
    :param num_layers: Number of layers
    :param target_vocab_to_int: Dictionary to go from the target words to an id
    :return: Tuple of (Training Logits, Inference Logits)
    """
    encoder_input = tf.contrib.layers.embed_sequence(input_data, source_vocab_size, enc_embedding_size)
    encoder_layer = encoding_layer(encoder_input, rnn_size, num_layers, keep_prob)
    
    decoder_input = process_decoding_input(target_data, target_vocab_to_int, batch_size)
    
    decoder_embedding = tf.Variable(tf.random_uniform([target_vocab_size, dec_embedding_size], minval=0))
    target_embedding = tf.nn.embedding_lookup(decoder_embedding, decoder_input)
    
    train_logits, inf_logits = decoding_layer(
        target_embedding,
        decoder_embedding,
        encoder_layer,
        target_vocab_size,
        sequence_length,
        rnn_size,
        num_layers,
        target_vocab_to_int,
        keep_prob
    )
    
    return train_logits, inf_logits

"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_seq2seq_model(seq2seq_model)



    


















    






Tests Passed

Neural Network Training

Hyperparameters

Tune the following parameters:

  • Set epochs to the number of epochs.
  • Set batch_size to the batch size.
  • Set rnn_size to the size of the RNNs.
  • Set num_layers to the number of layers.
  • Set encoding_embedding_size to the size of the embedding for the encoder.
  • Set decoding_embedding_size to the size of the embedding for the decoder.
  • Set learning_rate to the learning rate.
  • Set keep_probability to the Dropout keep probability




In [15]:



    


# Number of Epochs
epochs = 10
# Batch Size
batch_size = 1024
# RNN Size
rnn_size = 256
# Number of Layers
num_layers = 2
# Embedding Size
encoding_embedding_size = 64
decoding_embedding_size = 64
# Learning Rate
learning_rate = 0.01
# Dropout Keep Probability
keep_probability = 0.5

Build the Graph

Build the graph using the neural network you implemented.





In [16]:



    


"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
save_path = 'checkpoints/dev'
(source_int_text, target_int_text), (source_vocab_to_int, target_vocab_to_int), _ = helper.load_preprocess()
max_source_sentence_length = max([len(sentence) for sentence in source_int_text])

train_graph = tf.Graph()
with train_graph.as_default():
    input_data, targets, lr, keep_prob = model_inputs()
    sequence_length = tf.placeholder_with_default(max_source_sentence_length, None, name='sequence_length')
    input_shape = tf.shape(input_data)
    
    train_logits, inference_logits = seq2seq_model(
        tf.reverse(input_data, [-1]), targets, keep_prob, batch_size, sequence_length, len(source_vocab_to_int), len(target_vocab_to_int),
        encoding_embedding_size, decoding_embedding_size, rnn_size, num_layers, target_vocab_to_int)

    tf.identity(inference_logits, 'logits')
    with tf.name_scope("optimization"):
        # Loss function
        cost = tf.contrib.seq2seq.sequence_loss(
            train_logits,
            targets,
            tf.ones([input_shape[0], sequence_length]))

        # Optimizer
        optimizer = tf.train.AdamOptimizer(lr)

        # Gradient Clipping
        gradients = optimizer.compute_gradients(cost)
        capped_gradients = [(tf.clip_by_value(grad, -1., 1.), var) for grad, var in gradients if grad is not None]
        train_op = optimizer.apply_gradients(capped_gradients)

Train

Train the neural network on the preprocessed data. If you have a hard time getting a good loss, check the forums to see if anyone is having the same problem.





In [17]:



    


"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import time

def get_accuracy(target, logits):
    """
    Calculate accuracy
    """
    max_seq = max(target.shape[1], logits.shape[1])
    if max_seq - target.shape[1]:
        target = np.pad(
            target,
            [(0,0),(0,max_seq - target.shape[1])],
            'constant')
    if max_seq - logits.shape[1]:
        logits = np.pad(
            logits,
            [(0,0),(0,max_seq - logits.shape[1]), (0,0)],
            'constant')

    return np.mean(np.equal(target, np.argmax(logits, 2)))

train_source = source_int_text[batch_size:]
train_target = target_int_text[batch_size:]

valid_source = helper.pad_sentence_batch(source_int_text[:batch_size])
valid_target = helper.pad_sentence_batch(target_int_text[:batch_size])

with tf.Session(graph=train_graph) as sess:
    sess.run(tf.global_variables_initializer())

    for epoch_i in range(epochs):
        for batch_i, (source_batch, target_batch) in enumerate(
                helper.batch_data(train_source, train_target, batch_size)):
            start_time = time.time()
            
            _, loss = sess.run(
                [train_op, cost],
                {input_data: source_batch,
                 targets: target_batch,
                 lr: learning_rate,
                 sequence_length: target_batch.shape[1],
                 keep_prob: keep_probability})
            
            batch_train_logits = sess.run(
                inference_logits,
                {input_data: source_batch, keep_prob: 1.0})
            batch_valid_logits = sess.run(
                inference_logits,
                {input_data: valid_source, keep_prob: 1.0})
                
            train_acc = get_accuracy(target_batch, batch_train_logits)
            valid_acc = get_accuracy(np.array(valid_target), batch_valid_logits)
            end_time = time.time()
            print('Epoch {:>3} Batch {:>4}/{} - Train Accuracy: {:>6.3f}, Validation Accuracy: {:>6.3f}, Loss: {:>6.3f}'
                  .format(epoch_i, batch_i, len(source_int_text) // batch_size, train_acc, valid_acc, loss))

    # Save Model
    saver = tf.train.Saver()
    saver.save(sess, save_path)
    print('Model Trained and Saved')



    


















    






Epoch   0 Batch    0/134 - Train Accuracy:  0.267, Validation Accuracy:  0.310, Loss:  5.898
Epoch   0 Batch    1/134 - Train Accuracy:  0.274, Validation Accuracy:  0.342, Loss:  4.464
Epoch   0 Batch    2/134 - Train Accuracy:  0.309, Validation Accuracy:  0.343, Loss:  3.658
Epoch   0 Batch    3/134 - Train Accuracy:  0.308, Validation Accuracy:  0.340, Loss:  3.516
Epoch   0 Batch    4/134 - Train Accuracy:  0.306, Validation Accuracy:  0.344, Loss:  3.396
Epoch   0 Batch    5/134 - Train Accuracy:  0.311, Validation Accuracy:  0.344, Loss:  3.623
Epoch   0 Batch    6/134 - Train Accuracy:  0.341, Validation Accuracy:  0.344, Loss:  3.216
Epoch   0 Batch    7/134 - Train Accuracy:  0.319, Validation Accuracy:  0.359, Loss:  3.351
Epoch   0 Batch    8/134 - Train Accuracy:  0.318, Validation Accuracy:  0.356, Loss:  3.242
Epoch   0 Batch    9/134 - Train Accuracy:  0.348, Validation Accuracy:  0.350, Loss:  3.115
Epoch   0 Batch   10/134 - Train Accuracy:  0.320, Validation Accuracy:  0.350, Loss:  3.182
Epoch   0 Batch   11/134 - Train Accuracy:  0.337, Validation Accuracy:  0.364, Loss:  3.126
Epoch   0 Batch   12/134 - Train Accuracy:  0.393, Validation Accuracy:  0.394, Loss:  2.936
Epoch   0 Batch   13/134 - Train Accuracy:  0.346, Validation Accuracy:  0.382, Loss:  3.070
Epoch   0 Batch   14/134 - Train Accuracy:  0.345, Validation Accuracy:  0.379, Loss:  2.968
Epoch   0 Batch   15/134 - Train Accuracy:  0.358, Validation Accuracy:  0.391, Loss:  3.501
Epoch   0 Batch   16/134 - Train Accuracy:  0.378, Validation Accuracy:  0.407, Loss:  2.869
Epoch   0 Batch   17/134 - Train Accuracy:  0.357, Validation Accuracy:  0.386, Loss:  3.012
Epoch   0 Batch   18/134 - Train Accuracy:  0.351, Validation Accuracy:  0.380, Loss:  2.937
Epoch   0 Batch   19/134 - Train Accuracy:  0.347, Validation Accuracy:  0.377, Loss:  2.838
Epoch   0 Batch   20/134 - Train Accuracy:  0.389, Validation Accuracy:  0.394, Loss:  2.774
Epoch   0 Batch   21/134 - Train Accuracy:  0.366, Validation Accuracy:  0.389, Loss:  2.831
Epoch   0 Batch   22/134 - Train Accuracy:  0.332, Validation Accuracy:  0.394, Loss:  2.918
Epoch   0 Batch   23/134 - Train Accuracy:  0.402, Validation Accuracy:  0.403, Loss:  2.627
Epoch   0 Batch   24/134 - Train Accuracy:  0.350, Validation Accuracy:  0.404, Loss:  2.822
Epoch   0 Batch   25/134 - Train Accuracy:  0.387, Validation Accuracy:  0.416, Loss:  2.731
Epoch   0 Batch   26/134 - Train Accuracy:  0.361, Validation Accuracy:  0.416, Loss:  2.833
Epoch   0 Batch   27/134 - Train Accuracy:  0.387, Validation Accuracy:  0.415, Loss:  2.662
Epoch   0 Batch   28/134 - Train Accuracy:  0.394, Validation Accuracy:  0.421, Loss:  2.660
Epoch   0 Batch   29/134 - Train Accuracy:  0.428, Validation Accuracy:  0.436, Loss:  2.509
Epoch   0 Batch   30/134 - Train Accuracy:  0.440, Validation Accuracy:  0.444, Loss:  2.495
Epoch   0 Batch   31/134 - Train Accuracy:  0.415, Validation Accuracy:  0.444, Loss:  2.575
Epoch   0 Batch   32/134 - Train Accuracy:  0.431, Validation Accuracy:  0.437, Loss:  2.451
Epoch   0 Batch   33/134 - Train Accuracy:  0.412, Validation Accuracy:  0.441, Loss:  2.547
Epoch   0 Batch   34/134 - Train Accuracy:  0.427, Validation Accuracy:  0.451, Loss:  2.525
Epoch   0 Batch   35/134 - Train Accuracy:  0.446, Validation Accuracy:  0.449, Loss:  2.395
Epoch   0 Batch   36/134 - Train Accuracy:  0.412, Validation Accuracy:  0.438, Loss:  2.459
Epoch   0 Batch   37/134 - Train Accuracy:  0.423, Validation Accuracy:  0.458, Loss:  2.489
Epoch   0 Batch   38/134 - Train Accuracy:  0.434, Validation Accuracy:  0.464, Loss:  2.424
Epoch   0 Batch   39/134 - Train Accuracy:  0.442, Validation Accuracy:  0.468, Loss:  2.378
Epoch   0 Batch   40/134 - Train Accuracy:  0.447, Validation Accuracy:  0.470, Loss:  2.365
Epoch   0 Batch   41/134 - Train Accuracy:  0.448, Validation Accuracy:  0.472, Loss:  2.323
Epoch   0 Batch   42/134 - Train Accuracy:  0.435, Validation Accuracy:  0.465, Loss:  2.358
Epoch   0 Batch   43/134 - Train Accuracy:  0.452, Validation Accuracy:  0.480, Loss:  2.351
Epoch   0 Batch   44/134 - Train Accuracy:  0.449, Validation Accuracy:  0.478, Loss:  2.283
Epoch   0 Batch   45/134 - Train Accuracy:  0.457, Validation Accuracy:  0.487, Loss:  2.281
Epoch   0 Batch   46/134 - Train Accuracy:  0.482, Validation Accuracy:  0.489, Loss:  2.168
Epoch   0 Batch   47/134 - Train Accuracy:  0.475, Validation Accuracy:  0.499, Loss:  2.231
Epoch   0 Batch   48/134 - Train Accuracy:  0.472, Validation Accuracy:  0.494, Loss:  2.183
Epoch   0 Batch   49/134 - Train Accuracy:  0.483, Validation Accuracy:  0.502, Loss:  2.167
Epoch   0 Batch   50/134 - Train Accuracy:  0.461, Validation Accuracy:  0.489, Loss:  2.185
Epoch   0 Batch   51/134 - Train Accuracy:  0.468, Validation Accuracy:  0.497, Loss:  2.170
Epoch   0 Batch   52/134 - Train Accuracy:  0.474, Validation Accuracy:  0.507, Loss:  2.160
Epoch   0 Batch   53/134 - Train Accuracy:  0.456, Validation Accuracy:  0.490, Loss:  2.128
Epoch   0 Batch   54/134 - Train Accuracy:  0.463, Validation Accuracy:  0.498, Loss:  2.117
Epoch   0 Batch   55/134 - Train Accuracy:  0.474, Validation Accuracy:  0.502, Loss:  2.093
Epoch   0 Batch   56/134 - Train Accuracy:  0.506, Validation Accuracy:  0.514, Loss:  1.985
Epoch   0 Batch   57/134 - Train Accuracy:  0.478, Validation Accuracy:  0.501, Loss:  2.071
Epoch   0 Batch   58/134 - Train Accuracy:  0.460, Validation Accuracy:  0.479, Loss:  2.008
Epoch   0 Batch   59/134 - Train Accuracy:  0.465, Validation Accuracy:  0.511, Loss:  2.126
Epoch   0 Batch   60/134 - Train Accuracy:  0.491, Validation Accuracy:  0.514, Loss:  1.986
Epoch   0 Batch   61/134 - Train Accuracy:  0.459, Validation Accuracy:  0.489, Loss:  1.987
Epoch   0 Batch   62/134 - Train Accuracy:  0.481, Validation Accuracy:  0.503, Loss:  1.944
Epoch   0 Batch   63/134 - Train Accuracy:  0.505, Validation Accuracy:  0.520, Loss:  1.934
Epoch   0 Batch   64/134 - Train Accuracy:  0.476, Validation Accuracy:  0.508, Loss:  1.993
Epoch   0 Batch   65/134 - Train Accuracy:  0.433, Validation Accuracy:  0.460, Loss:  1.952
Epoch   0 Batch   66/134 - Train Accuracy:  0.446, Validation Accuracy:  0.467, Loss:  2.021
Epoch   0 Batch   67/134 - Train Accuracy:  0.433, Validation Accuracy:  0.496, Loss:  2.148
Epoch   0 Batch   68/134 - Train Accuracy:  0.453, Validation Accuracy:  0.499, Loss:  2.012
Epoch   0 Batch   69/134 - Train Accuracy:  0.476, Validation Accuracy:  0.498, Loss:  1.887
Epoch   0 Batch   70/134 - Train Accuracy:  0.488, Validation Accuracy:  0.508, Loss:  1.900
Epoch   0 Batch   71/134 - Train Accuracy:  0.467, Validation Accuracy:  0.497, Loss:  1.868
Epoch   0 Batch   72/134 - Train Accuracy:  0.502, Validation Accuracy:  0.529, Loss:  1.896
Epoch   0 Batch   73/134 - Train Accuracy:  0.515, Validation Accuracy:  0.522, Loss:  1.797
Epoch   0 Batch   74/134 - Train Accuracy:  0.481, Validation Accuracy:  0.505, Loss:  1.851
Epoch   0 Batch   75/134 - Train Accuracy:  0.526, Validation Accuracy:  0.526, Loss:  1.754
Epoch   0 Batch   76/134 - Train Accuracy:  0.508, Validation Accuracy:  0.531, Loss:  1.829
Epoch   0 Batch   77/134 - Train Accuracy:  0.538, Validation Accuracy:  0.533, Loss:  1.710
Epoch   0 Batch   78/134 - Train Accuracy:  0.474, Validation Accuracy:  0.501, Loss:  1.791
Epoch   0 Batch   79/134 - Train Accuracy:  0.496, Validation Accuracy:  0.520, Loss:  1.819
Epoch   0 Batch   80/134 - Train Accuracy:  0.502, Validation Accuracy:  0.530, Loss:  1.787
Epoch   0 Batch   81/134 - Train Accuracy:  0.504, Validation Accuracy:  0.525, Loss:  1.773
Epoch   0 Batch   82/134 - Train Accuracy:  0.528, Validation Accuracy:  0.527, Loss:  1.656
Epoch   0 Batch   83/134 - Train Accuracy:  0.501, Validation Accuracy:  0.526, Loss:  1.745
Epoch   0 Batch   84/134 - Train Accuracy:  0.500, Validation Accuracy:  0.528, Loss:  1.738
Epoch   0 Batch   85/134 - Train Accuracy:  0.522, Validation Accuracy:  0.542, Loss:  1.708
Epoch   0 Batch   86/134 - Train Accuracy:  0.505, Validation Accuracy:  0.528, Loss:  1.715
Epoch   0 Batch   87/134 - Train Accuracy:  0.505, Validation Accuracy:  0.528, Loss:  1.712
Epoch   0 Batch   88/134 - Train Accuracy:  0.546, Validation Accuracy:  0.546, Loss:  1.618
Epoch   0 Batch   89/134 - Train Accuracy:  0.521, Validation Accuracy:  0.542, Loss:  1.657
Epoch   0 Batch   90/134 - Train Accuracy:  0.486, Validation Accuracy:  0.512, Loss:  1.658
Epoch   0 Batch   91/134 - Train Accuracy:  0.578, Validation Accuracy:  0.533, Loss:  1.439
Epoch   0 Batch   92/134 - Train Accuracy:  0.507, Validation Accuracy:  0.529, Loss:  1.616
Epoch   0 Batch   93/134 - Train Accuracy:  0.522, Validation Accuracy:  0.535, Loss:  1.566
Epoch   0 Batch   94/134 - Train Accuracy:  0.522, Validation Accuracy:  0.545, Loss:  1.586
Epoch   0 Batch   95/134 - Train Accuracy:  0.514, Validation Accuracy:  0.534, Loss:  1.603
Epoch   0 Batch   96/134 - Train Accuracy:  0.522, Validation Accuracy:  0.543, Loss:  1.592
Epoch   0 Batch   97/134 - Train Accuracy:  0.486, Validation Accuracy:  0.522, Loss:  1.592
Epoch   0 Batch   98/134 - Train Accuracy:  0.509, Validation Accuracy:  0.550, Loss:  1.664
Epoch   0 Batch   99/134 - Train Accuracy:  0.503, Validation Accuracy:  0.531, Loss:  1.565
Epoch   0 Batch  100/134 - Train Accuracy:  0.528, Validation Accuracy:  0.550, Loss:  1.543
Epoch   0 Batch  101/134 - Train Accuracy:  0.510, Validation Accuracy:  0.554, Loss:  1.570
Epoch   0 Batch  102/134 - Train Accuracy:  0.488, Validation Accuracy:  0.516, Loss:  1.516
Epoch   0 Batch  103/134 - Train Accuracy:  0.553, Validation Accuracy:  0.555, Loss:  1.456
Epoch   0 Batch  104/134 - Train Accuracy:  0.530, Validation Accuracy:  0.548, Loss:  1.468
Epoch   0 Batch  105/134 - Train Accuracy:  0.517, Validation Accuracy:  0.531, Loss:  1.419
Epoch   0 Batch  106/134 - Train Accuracy:  0.536, Validation Accuracy:  0.547, Loss:  1.425
Epoch   0 Batch  107/134 - Train Accuracy:  0.536, Validation Accuracy:  0.540, Loss:  1.372
Epoch   0 Batch  108/134 - Train Accuracy:  0.495, Validation Accuracy:  0.542, Loss:  1.463
Epoch   0 Batch  109/134 - Train Accuracy:  0.550, Validation Accuracy:  0.542, Loss:  1.325
Epoch   0 Batch  110/134 - Train Accuracy:  0.542, Validation Accuracy:  0.555, Loss:  1.350
Epoch   0 Batch  111/134 - Train Accuracy:  0.520, Validation Accuracy:  0.537, Loss:  1.326
Epoch   0 Batch  112/134 - Train Accuracy:  0.534, Validation Accuracy:  0.552, Loss:  1.326
Epoch   0 Batch  113/134 - Train Accuracy:  0.575, Validation Accuracy:  0.537, Loss:  1.158
Epoch   0 Batch  114/134 - Train Accuracy:  0.503, Validation Accuracy:  0.527, Loss:  1.302
Epoch   0 Batch  115/134 - Train Accuracy:  0.486, Validation Accuracy:  0.530, Loss:  1.343
Epoch   0 Batch  116/134 - Train Accuracy:  0.521, Validation Accuracy:  0.532, Loss:  1.270
Epoch   0 Batch  117/134 - Train Accuracy:  0.522, Validation Accuracy:  0.546, Loss:  1.252
Epoch   0 Batch  118/134 - Train Accuracy:  0.520, Validation Accuracy:  0.541, Loss:  1.222
Epoch   0 Batch  119/134 - Train Accuracy:  0.560, Validation Accuracy:  0.543, Loss:  1.119
Epoch   0 Batch  120/134 - Train Accuracy:  0.537, Validation Accuracy:  0.555, Loss:  1.196
Epoch   0 Batch  121/134 - Train Accuracy:  0.529, Validation Accuracy:  0.545, Loss:  1.163
Epoch   0 Batch  122/134 - Train Accuracy:  0.533, Validation Accuracy:  0.552, Loss:  1.172
Epoch   0 Batch  123/134 - Train Accuracy:  0.538, Validation Accuracy:  0.553, Loss:  1.143
Epoch   0 Batch  124/134 - Train Accuracy:  0.567, Validation Accuracy:  0.560, Loss:  1.079
Epoch   0 Batch  125/134 - Train Accuracy:  0.545, Validation Accuracy:  0.558, Loss:  1.129
Epoch   0 Batch  126/134 - Train Accuracy:  0.550, Validation Accuracy:  0.570, Loss:  1.110
Epoch   0 Batch  127/134 - Train Accuracy:  0.564, Validation Accuracy:  0.570, Loss:  1.062
Epoch   0 Batch  128/134 - Train Accuracy:  0.531, Validation Accuracy:  0.559, Loss:  1.095
Epoch   0 Batch  129/134 - Train Accuracy:  0.549, Validation Accuracy:  0.559, Loss:  1.073
Epoch   0 Batch  130/134 - Train Accuracy:  0.535, Validation Accuracy:  0.568, Loss:  1.080
Epoch   0 Batch  131/134 - Train Accuracy:  0.538, Validation Accuracy:  0.578, Loss:  1.106
Epoch   0 Batch  132/134 - Train Accuracy:  0.569, Validation Accuracy:  0.585, Loss:  1.030
Epoch   1 Batch    0/134 - Train Accuracy:  0.549, Validation Accuracy:  0.574, Loss:  1.030
Epoch   1 Batch    1/134 - Train Accuracy:  0.533, Validation Accuracy:  0.571, Loss:  1.057
Epoch   1 Batch    2/134 - Train Accuracy:  0.564, Validation Accuracy:  0.579, Loss:  0.986
Epoch   1 Batch    3/134 - Train Accuracy:  0.555, Validation Accuracy:  0.575, Loss:  0.991
Epoch   1 Batch    4/134 - Train Accuracy:  0.547, Validation Accuracy:  0.569, Loss:  0.983
Epoch   1 Batch    5/134 - Train Accuracy:  0.539, Validation Accuracy:  0.567, Loss:  0.980
Epoch   1 Batch    6/134 - Train Accuracy:  0.570, Validation Accuracy:  0.562, Loss:  0.911
Epoch   1 Batch    7/134 - Train Accuracy:  0.561, Validation Accuracy:  0.577, Loss:  0.956
Epoch   1 Batch    8/134 - Train Accuracy:  0.566, Validation Accuracy:  0.578, Loss:  0.929
Epoch   1 Batch    9/134 - Train Accuracy:  0.593, Validation Accuracy:  0.581, Loss:  0.885
Epoch   1 Batch   10/134 - Train Accuracy:  0.567, Validation Accuracy:  0.578, Loss:  0.932
Epoch   1 Batch   11/134 - Train Accuracy:  0.557, Validation Accuracy:  0.570, Loss:  0.927
Epoch   1 Batch   12/134 - Train Accuracy:  0.579, Validation Accuracy:  0.571, Loss:  0.868
Epoch   1 Batch   13/134 - Train Accuracy:  0.548, Validation Accuracy:  0.582, Loss:  0.909
Epoch   1 Batch   14/134 - Train Accuracy:  0.553, Validation Accuracy:  0.567, Loss:  0.880
Epoch   1 Batch   15/134 - Train Accuracy:  0.563, Validation Accuracy:  0.578, Loss:  0.879
Epoch   1 Batch   16/134 - Train Accuracy:  0.581, Validation Accuracy:  0.591, Loss:  0.860
Epoch   1 Batch   17/134 - Train Accuracy:  0.553, Validation Accuracy:  0.561, Loss:  0.877
Epoch   1 Batch   18/134 - Train Accuracy:  0.576, Validation Accuracy:  0.596, Loss:  0.870
Epoch   1 Batch   19/134 - Train Accuracy:  0.580, Validation Accuracy:  0.597, Loss:  0.844
Epoch   1 Batch   20/134 - Train Accuracy:  0.582, Validation Accuracy:  0.585, Loss:  0.815
Epoch   1 Batch   21/134 - Train Accuracy:  0.570, Validation Accuracy:  0.586, Loss:  0.836
Epoch   1 Batch   22/134 - Train Accuracy:  0.551, Validation Accuracy:  0.589, Loss:  0.857
Epoch   1 Batch   23/134 - Train Accuracy:  0.595, Validation Accuracy:  0.592, Loss:  0.766
Epoch   1 Batch   24/134 - Train Accuracy:  0.569, Validation Accuracy:  0.599, Loss:  0.836
Epoch   1 Batch   25/134 - Train Accuracy:  0.558, Validation Accuracy:  0.593, Loss:  0.789
Epoch   1 Batch   26/134 - Train Accuracy:  0.551, Validation Accuracy:  0.591, Loss:  0.826
Epoch   1 Batch   27/134 - Train Accuracy:  0.598, Validation Accuracy:  0.607, Loss:  0.785
Epoch   1 Batch   28/134 - Train Accuracy:  0.597, Validation Accuracy:  0.616, Loss:  0.777
Epoch   1 Batch   29/134 - Train Accuracy:  0.612, Validation Accuracy:  0.612, Loss:  0.728
Epoch   1 Batch   30/134 - Train Accuracy:  0.612, Validation Accuracy:  0.617, Loss:  0.729
Epoch   1 Batch   31/134 - Train Accuracy:  0.586, Validation Accuracy:  0.614, Loss:  0.751
Epoch   1 Batch   32/134 - Train Accuracy:  0.612, Validation Accuracy:  0.613, Loss:  0.721
Epoch   1 Batch   33/134 - Train Accuracy:  0.586, Validation Accuracy:  0.612, Loss:  0.752
Epoch   1 Batch   34/134 - Train Accuracy:  0.599, Validation Accuracy:  0.618, Loss:  0.759
Epoch   1 Batch   35/134 - Train Accuracy:  0.625, Validation Accuracy:  0.621, Loss:  0.707
Epoch   1 Batch   36/134 - Train Accuracy:  0.616, Validation Accuracy:  0.624, Loss:  0.722
Epoch   1 Batch   37/134 - Train Accuracy:  0.602, Validation Accuracy:  0.627, Loss:  0.723
Epoch   1 Batch   38/134 - Train Accuracy:  0.623, Validation Accuracy:  0.631, Loss:  0.709
Epoch   1 Batch   39/134 - Train Accuracy:  0.610, Validation Accuracy:  0.627, Loss:  0.698
Epoch   1 Batch   40/134 - Train Accuracy:  0.617, Validation Accuracy:  0.627, Loss:  0.697
Epoch   1 Batch   41/134 - Train Accuracy:  0.618, Validation Accuracy:  0.631, Loss:  0.693
Epoch   1 Batch   42/134 - Train Accuracy:  0.616, Validation Accuracy:  0.629, Loss:  0.691
Epoch   1 Batch   43/134 - Train Accuracy:  0.601, Validation Accuracy:  0.621, Loss:  0.697
Epoch   1 Batch   44/134 - Train Accuracy:  0.613, Validation Accuracy:  0.628, Loss:  0.685
Epoch   1 Batch   45/134 - Train Accuracy:  0.634, Validation Accuracy:  0.634, Loss:  0.665
Epoch   1 Batch   46/134 - Train Accuracy:  0.645, Validation Accuracy:  0.635, Loss:  0.645
Epoch   1 Batch   47/134 - Train Accuracy:  0.632, Validation Accuracy:  0.648, Loss:  0.669
Epoch   1 Batch   48/134 - Train Accuracy:  0.635, Validation Accuracy:  0.647, Loss:  0.660
Epoch   1 Batch   49/134 - Train Accuracy:  0.640, Validation Accuracy:  0.646, Loss:  0.653
Epoch   1 Batch   50/134 - Train Accuracy:  0.636, Validation Accuracy:  0.647, Loss:  0.658
Epoch   1 Batch   51/134 - Train Accuracy:  0.623, Validation Accuracy:  0.638, Loss:  0.653
Epoch   1 Batch   52/134 - Train Accuracy:  0.616, Validation Accuracy:  0.635, Loss:  0.656
Epoch   1 Batch   53/134 - Train Accuracy:  0.633, Validation Accuracy:  0.649, Loss:  0.647
Epoch   1 Batch   54/134 - Train Accuracy:  0.600, Validation Accuracy:  0.622, Loss:  0.648
Epoch   1 Batch   55/134 - Train Accuracy:  0.630, Validation Accuracy:  0.646, Loss:  0.648
Epoch   1 Batch   56/134 - Train Accuracy:  0.649, Validation Accuracy:  0.646, Loss:  0.607
Epoch   1 Batch   57/134 - Train Accuracy:  0.620, Validation Accuracy:  0.623, Loss:  0.634
Epoch   1 Batch   58/134 - Train Accuracy:  0.648, Validation Accuracy:  0.651, Loss:  0.627
Epoch   1 Batch   59/134 - Train Accuracy:  0.625, Validation Accuracy:  0.650, Loss:  0.654
Epoch   1 Batch   60/134 - Train Accuracy:  0.644, Validation Accuracy:  0.646, Loss:  0.620
Epoch   1 Batch   61/134 - Train Accuracy:  0.632, Validation Accuracy:  0.629, Loss:  0.616
Epoch   1 Batch   62/134 - Train Accuracy:  0.648, Validation Accuracy:  0.655, Loss:  0.614
Epoch   1 Batch   63/134 - Train Accuracy:  0.655, Validation Accuracy:  0.657, Loss:  0.609
Epoch   1 Batch   64/134 - Train Accuracy:  0.635, Validation Accuracy:  0.649, Loss:  0.616
Epoch   1 Batch   65/134 - Train Accuracy:  0.630, Validation Accuracy:  0.655, Loss:  0.615
Epoch   1 Batch   66/134 - Train Accuracy:  0.642, Validation Accuracy:  0.652, Loss:  0.597
Epoch   1 Batch   67/134 - Train Accuracy:  0.613, Validation Accuracy:  0.653, Loss:  0.658
Epoch   1 Batch   68/134 - Train Accuracy:  0.628, Validation Accuracy:  0.656, Loss:  0.629
Epoch   1 Batch   69/134 - Train Accuracy:  0.647, Validation Accuracy:  0.654, Loss:  0.595
Epoch   1 Batch   70/134 - Train Accuracy:  0.645, Validation Accuracy:  0.655, Loss:  0.591
Epoch   1 Batch   71/134 - Train Accuracy:  0.644, Validation Accuracy:  0.660, Loss:  0.582
Epoch   1 Batch   72/134 - Train Accuracy:  0.649, Validation Accuracy:  0.657, Loss:  0.584
Epoch   1 Batch   73/134 - Train Accuracy:  0.653, Validation Accuracy:  0.663, Loss:  0.565
Epoch   1 Batch   74/134 - Train Accuracy:  0.642, Validation Accuracy:  0.657, Loss:  0.593
Epoch   1 Batch   75/134 - Train Accuracy:  0.670, Validation Accuracy:  0.662, Loss:  0.563
Epoch   1 Batch   76/134 - Train Accuracy:  0.661, Validation Accuracy:  0.662, Loss:  0.573
Epoch   1 Batch   77/134 - Train Accuracy:  0.661, Validation Accuracy:  0.666, Loss:  0.554
Epoch   1 Batch   78/134 - Train Accuracy:  0.642, Validation Accuracy:  0.664, Loss:  0.579
Epoch   1 Batch   79/134 - Train Accuracy:  0.653, Validation Accuracy:  0.665, Loss:  0.576
Epoch   1 Batch   80/134 - Train Accuracy:  0.644, Validation Accuracy:  0.661, Loss:  0.576
Epoch   1 Batch   81/134 - Train Accuracy:  0.659, Validation Accuracy:  0.662, Loss:  0.567
Epoch   1 Batch   82/134 - Train Accuracy:  0.660, Validation Accuracy:  0.664, Loss:  0.538
Epoch   1 Batch   83/134 - Train Accuracy:  0.646, Validation Accuracy:  0.662, Loss:  0.571
Epoch   1 Batch   84/134 - Train Accuracy:  0.666, Validation Accuracy:  0.668, Loss:  0.563
Epoch   1 Batch   85/134 - Train Accuracy:  0.676, Validation Accuracy:  0.669, Loss:  0.570
Epoch   1 Batch   86/134 - Train Accuracy:  0.666, Validation Accuracy:  0.675, Loss:  0.554
Epoch   1 Batch   87/134 - Train Accuracy:  0.653, Validation Accuracy:  0.672, Loss:  0.569
Epoch   1 Batch   88/134 - Train Accuracy:  0.679, Validation Accuracy:  0.672, Loss:  0.529
Epoch   1 Batch   89/134 - Train Accuracy:  0.664, Validation Accuracy:  0.669, Loss:  0.552
Epoch   1 Batch   90/134 - Train Accuracy:  0.669, Validation Accuracy:  0.677, Loss:  0.547
Epoch   1 Batch   91/134 - Train Accuracy:  0.710, Validation Accuracy:  0.674, Loss:  0.478
Epoch   1 Batch   92/134 - Train Accuracy:  0.667, Validation Accuracy:  0.676, Loss:  0.537
Epoch   1 Batch   93/134 - Train Accuracy:  0.672, Validation Accuracy:  0.681, Loss:  0.533
Epoch   1 Batch   94/134 - Train Accuracy:  0.673, Validation Accuracy:  0.682, Loss:  0.528
Epoch   1 Batch   95/134 - Train Accuracy:  0.675, Validation Accuracy:  0.683, Loss:  0.541
Epoch   1 Batch   96/134 - Train Accuracy:  0.666, Validation Accuracy:  0.665, Loss:  0.542
Epoch   1 Batch   97/134 - Train Accuracy:  0.643, Validation Accuracy:  0.670, Loss:  0.542
Epoch   1 Batch   98/134 - Train Accuracy:  0.645, Validation Accuracy:  0.684, Loss:  0.568
Epoch   1 Batch   99/134 - Train Accuracy:  0.650, Validation Accuracy:  0.657, Loss:  0.549
Epoch   1 Batch  100/134 - Train Accuracy:  0.662, Validation Accuracy:  0.673, Loss:  0.550
Epoch   1 Batch  101/134 - Train Accuracy:  0.646, Validation Accuracy:  0.677, Loss:  0.569
Epoch   1 Batch  102/134 - Train Accuracy:  0.660, Validation Accuracy:  0.673, Loss:  0.527
Epoch   1 Batch  103/134 - Train Accuracy:  0.671, Validation Accuracy:  0.669, Loss:  0.514
Epoch   1 Batch  104/134 - Train Accuracy:  0.686, Validation Accuracy:  0.681, Loss:  0.519
Epoch   1 Batch  105/134 - Train Accuracy:  0.670, Validation Accuracy:  0.689, Loss:  0.527
Epoch   1 Batch  106/134 - Train Accuracy:  0.666, Validation Accuracy:  0.674, Loss:  0.527
Epoch   1 Batch  107/134 - Train Accuracy:  0.671, Validation Accuracy:  0.668, Loss:  0.496
Epoch   1 Batch  108/134 - Train Accuracy:  0.641, Validation Accuracy:  0.683, Loss:  0.545
Epoch   1 Batch  109/134 - Train Accuracy:  0.696, Validation Accuracy:  0.690, Loss:  0.492
Epoch   1 Batch  110/134 - Train Accuracy:  0.697, Validation Accuracy:  0.683, Loss:  0.511
Epoch   1 Batch  111/134 - Train Accuracy:  0.681, Validation Accuracy:  0.682, Loss:  0.515
Epoch   1 Batch  112/134 - Train Accuracy:  0.679, Validation Accuracy:  0.691, Loss:  0.504
Epoch   1 Batch  113/134 - Train Accuracy:  0.721, Validation Accuracy:  0.687, Loss:  0.445
Epoch   1 Batch  114/134 - Train Accuracy:  0.678, Validation Accuracy:  0.690, Loss:  0.504
Epoch   1 Batch  115/134 - Train Accuracy:  0.649, Validation Accuracy:  0.686, Loss:  0.534
Epoch   1 Batch  116/134 - Train Accuracy:  0.689, Validation Accuracy:  0.692, Loss:  0.505
Epoch   1 Batch  117/134 - Train Accuracy:  0.681, Validation Accuracy:  0.689, Loss:  0.491
Epoch   1 Batch  118/134 - Train Accuracy:  0.692, Validation Accuracy:  0.693, Loss:  0.496
Epoch   1 Batch  119/134 - Train Accuracy:  0.701, Validation Accuracy:  0.682, Loss:  0.449
Epoch   1 Batch  120/134 - Train Accuracy:  0.678, Validation Accuracy:  0.685, Loss:  0.498
Epoch   1 Batch  121/134 - Train Accuracy:  0.693, Validation Accuracy:  0.699, Loss:  0.488
Epoch   1 Batch  122/134 - Train Accuracy:  0.680, Validation Accuracy:  0.694, Loss:  0.492
Epoch   1 Batch  123/134 - Train Accuracy:  0.681, Validation Accuracy:  0.699, Loss:  0.483
Epoch   1 Batch  124/134 - Train Accuracy:  0.707, Validation Accuracy:  0.693, Loss:  0.451
Epoch   1 Batch  125/134 - Train Accuracy:  0.695, Validation Accuracy:  0.689, Loss:  0.479
Epoch   1 Batch  126/134 - Train Accuracy:  0.685, Validation Accuracy:  0.697, Loss:  0.485
Epoch   1 Batch  127/134 - Train Accuracy:  0.691, Validation Accuracy:  0.691, Loss:  0.462
Epoch   1 Batch  128/134 - Train Accuracy:  0.678, Validation Accuracy:  0.687, Loss:  0.479
Epoch   1 Batch  129/134 - Train Accuracy:  0.685, Validation Accuracy:  0.694, Loss:  0.471
Epoch   1 Batch  130/134 - Train Accuracy:  0.698, Validation Accuracy:  0.707, Loss:  0.469
Epoch   1 Batch  131/134 - Train Accuracy:  0.683, Validation Accuracy:  0.706, Loss:  0.488
Epoch   1 Batch  132/134 - Train Accuracy:  0.702, Validation Accuracy:  0.699, Loss:  0.457
Epoch   2 Batch    0/134 - Train Accuracy:  0.698, Validation Accuracy:  0.709, Loss:  0.461
Epoch   2 Batch    1/134 - Train Accuracy:  0.680, Validation Accuracy:  0.707, Loss:  0.484
Epoch   2 Batch    2/134 - Train Accuracy:  0.692, Validation Accuracy:  0.698, Loss:  0.455
Epoch   2 Batch    3/134 - Train Accuracy:  0.698, Validation Accuracy:  0.706, Loss:  0.456
Epoch   2 Batch    4/134 - Train Accuracy:  0.708, Validation Accuracy:  0.711, Loss:  0.463
Epoch   2 Batch    5/134 - Train Accuracy:  0.709, Validation Accuracy:  0.715, Loss:  0.460
Epoch   2 Batch    6/134 - Train Accuracy:  0.715, Validation Accuracy:  0.715, Loss:  0.429
Epoch   2 Batch    7/134 - Train Accuracy:  0.701, Validation Accuracy:  0.712, Loss:  0.445
Epoch   2 Batch    8/134 - Train Accuracy:  0.719, Validation Accuracy:  0.712, Loss:  0.440
Epoch   2 Batch    9/134 - Train Accuracy:  0.721, Validation Accuracy:  0.709, Loss:  0.418
Epoch   2 Batch   10/134 - Train Accuracy:  0.694, Validation Accuracy:  0.715, Loss:  0.445
Epoch   2 Batch   11/134 - Train Accuracy:  0.712, Validation Accuracy:  0.722, Loss:  0.441
Epoch   2 Batch   12/134 - Train Accuracy:  0.718, Validation Accuracy:  0.716, Loss:  0.417
Epoch   2 Batch   13/134 - Train Accuracy:  0.694, Validation Accuracy:  0.719, Loss:  0.443
Epoch   2 Batch   14/134 - Train Accuracy:  0.723, Validation Accuracy:  0.720, Loss:  0.434
Epoch   2 Batch   15/134 - Train Accuracy:  0.712, Validation Accuracy:  0.715, Loss:  0.426
Epoch   2 Batch   16/134 - Train Accuracy:  0.715, Validation Accuracy:  0.720, Loss:  0.427
Epoch   2 Batch   17/134 - Train Accuracy:  0.711, Validation Accuracy:  0.725, Loss:  0.438
Epoch   2 Batch   18/134 - Train Accuracy:  0.713, Validation Accuracy:  0.724, Loss:  0.423
Epoch   2 Batch   19/134 - Train Accuracy:  0.714, Validation Accuracy:  0.722, Loss:  0.426
Epoch   2 Batch   20/134 - Train Accuracy:  0.719, Validation Accuracy:  0.717, Loss:  0.415
Epoch   2 Batch   21/134 - Train Accuracy:  0.719, Validation Accuracy:  0.723, Loss:  0.419
Epoch   2 Batch   22/134 - Train Accuracy:  0.694, Validation Accuracy:  0.719, Loss:  0.439
Epoch   2 Batch   23/134 - Train Accuracy:  0.726, Validation Accuracy:  0.712, Loss:  0.399
Epoch   2 Batch   24/134 - Train Accuracy:  0.695, Validation Accuracy:  0.714, Loss:  0.434
Epoch   2 Batch   25/134 - Train Accuracy:  0.722, Validation Accuracy:  0.728, Loss:  0.409
Epoch   2 Batch   26/134 - Train Accuracy:  0.711, Validation Accuracy:  0.732, Loss:  0.432
Epoch   2 Batch   27/134 - Train Accuracy:  0.731, Validation Accuracy:  0.733, Loss:  0.413
Epoch   2 Batch   28/134 - Train Accuracy:  0.715, Validation Accuracy:  0.722, Loss:  0.414
Epoch   2 Batch   29/134 - Train Accuracy:  0.721, Validation Accuracy:  0.718, Loss:  0.388
Epoch   2 Batch   30/134 - Train Accuracy:  0.737, Validation Accuracy:  0.731, Loss:  0.391
Epoch   2 Batch   31/134 - Train Accuracy:  0.721, Validation Accuracy:  0.732, Loss:  0.406
Epoch   2 Batch   32/134 - Train Accuracy:  0.717, Validation Accuracy:  0.720, Loss:  0.388
Epoch   2 Batch   33/134 - Train Accuracy:  0.712, Validation Accuracy:  0.728, Loss:  0.407
Epoch   2 Batch   34/134 - Train Accuracy:  0.720, Validation Accuracy:  0.731, Loss:  0.415
Epoch   2 Batch   35/134 - Train Accuracy:  0.739, Validation Accuracy:  0.732, Loss:  0.390
Epoch   2 Batch   36/134 - Train Accuracy:  0.731, Validation Accuracy:  0.732, Loss:  0.391
Epoch   2 Batch   37/134 - Train Accuracy:  0.722, Validation Accuracy:  0.729, Loss:  0.397
Epoch   2 Batch   38/134 - Train Accuracy:  0.729, Validation Accuracy:  0.736, Loss:  0.390
Epoch   2 Batch   39/134 - Train Accuracy:  0.735, Validation Accuracy:  0.735, Loss:  0.390
Epoch   2 Batch   40/134 - Train Accuracy:  0.740, Validation Accuracy:  0.734, Loss:  0.387
Epoch   2 Batch   41/134 - Train Accuracy:  0.730, Validation Accuracy:  0.735, Loss:  0.394
Epoch   2 Batch   42/134 - Train Accuracy:  0.721, Validation Accuracy:  0.724, Loss:  0.383
Epoch   2 Batch   43/134 - Train Accuracy:  0.726, Validation Accuracy:  0.739, Loss:  0.390
Epoch   2 Batch   44/134 - Train Accuracy:  0.743, Validation Accuracy:  0.745, Loss:  0.383
Epoch   2 Batch   45/134 - Train Accuracy:  0.739, Validation Accuracy:  0.742, Loss:  0.373
Epoch   2 Batch   46/134 - Train Accuracy:  0.742, Validation Accuracy:  0.740, Loss:  0.366
Epoch   2 Batch   47/134 - Train Accuracy:  0.740, Validation Accuracy:  0.744, Loss:  0.377
Epoch   2 Batch   48/134 - Train Accuracy:  0.753, Validation Accuracy:  0.751, Loss:  0.371
Epoch   2 Batch   49/134 - Train Accuracy:  0.751, Validation Accuracy:  0.754, Loss:  0.364
Epoch   2 Batch   50/134 - Train Accuracy:  0.735, Validation Accuracy:  0.748, Loss:  0.374
Epoch   2 Batch   51/134 - Train Accuracy:  0.745, Validation Accuracy:  0.748, Loss:  0.368
Epoch   2 Batch   52/134 - Train Accuracy:  0.739, Validation Accuracy:  0.749, Loss:  0.369
Epoch   2 Batch   53/134 - Train Accuracy:  0.755, Validation Accuracy:  0.755, Loss:  0.365
Epoch   2 Batch   54/134 - Train Accuracy:  0.743, Validation Accuracy:  0.746, Loss:  0.361
Epoch   2 Batch   55/134 - Train Accuracy:  0.750, Validation Accuracy:  0.753, Loss:  0.366
Epoch   2 Batch   56/134 - Train Accuracy:  0.763, Validation Accuracy:  0.755, Loss:  0.342
Epoch   2 Batch   57/134 - Train Accuracy:  0.766, Validation Accuracy:  0.758, Loss:  0.363
Epoch   2 Batch   58/134 - Train Accuracy:  0.752, Validation Accuracy:  0.747, Loss:  0.353
Epoch   2 Batch   59/134 - Train Accuracy:  0.745, Validation Accuracy:  0.755, Loss:  0.370
Epoch   2 Batch   60/134 - Train Accuracy:  0.756, Validation Accuracy:  0.757, Loss:  0.349
Epoch   2 Batch   61/134 - Train Accuracy:  0.755, Validation Accuracy:  0.756, Loss:  0.350
Epoch   2 Batch   62/134 - Train Accuracy:  0.758, Validation Accuracy:  0.759, Loss:  0.350
Epoch   2 Batch   63/134 - Train Accuracy:  0.767, Validation Accuracy:  0.764, Loss:  0.344
Epoch   2 Batch   64/134 - Train Accuracy:  0.762, Validation Accuracy:  0.772, Loss:  0.346
Epoch   2 Batch   65/134 - Train Accuracy:  0.752, Validation Accuracy:  0.765, Loss:  0.347
Epoch   2 Batch   66/134 - Train Accuracy:  0.765, Validation Accuracy:  0.761, Loss:  0.341
Epoch   2 Batch   67/134 - Train Accuracy:  0.742, Validation Accuracy:  0.772, Loss:  0.369
Epoch   2 Batch   68/134 - Train Accuracy:  0.756, Validation Accuracy:  0.763, Loss:  0.354
Epoch   2 Batch   69/134 - Train Accuracy:  0.772, Validation Accuracy:  0.774, Loss:  0.340
Epoch   2 Batch   70/134 - Train Accuracy:  0.768, Validation Accuracy:  0.773, Loss:  0.336
Epoch   2 Batch   71/134 - Train Accuracy:  0.763, Validation Accuracy:  0.765, Loss:  0.330
Epoch   2 Batch   72/134 - Train Accuracy:  0.760, Validation Accuracy:  0.769, Loss:  0.327
Epoch   2 Batch   73/134 - Train Accuracy:  0.778, Validation Accuracy:  0.765, Loss:  0.322
Epoch   2 Batch   74/134 - Train Accuracy:  0.772, Validation Accuracy:  0.777, Loss:  0.333
Epoch   2 Batch   75/134 - Train Accuracy:  0.786, Validation Accuracy:  0.780, Loss:  0.318
Epoch   2 Batch   76/134 - Train Accuracy:  0.780, Validation Accuracy:  0.770, Loss:  0.329
Epoch   2 Batch   77/134 - Train Accuracy:  0.779, Validation Accuracy:  0.771, Loss:  0.322
Epoch   2 Batch   78/134 - Train Accuracy:  0.767, Validation Accuracy:  0.778, Loss:  0.326
Epoch   2 Batch   79/134 - Train Accuracy:  0.762, Validation Accuracy:  0.772, Loss:  0.331
Epoch   2 Batch   80/134 - Train Accuracy:  0.773, Validation Accuracy:  0.783, Loss:  0.335
Epoch   2 Batch   81/134 - Train Accuracy:  0.773, Validation Accuracy:  0.774, Loss:  0.338
Epoch   2 Batch   82/134 - Train Accuracy:  0.782, Validation Accuracy:  0.776, Loss:  0.314
Epoch   2 Batch   83/134 - Train Accuracy:  0.762, Validation Accuracy:  0.769, Loss:  0.329
Epoch   2 Batch   84/134 - Train Accuracy:  0.767, Validation Accuracy:  0.767, Loss:  0.322
Epoch   2 Batch   85/134 - Train Accuracy:  0.761, Validation Accuracy:  0.765, Loss:  0.333
Epoch   2 Batch   86/134 - Train Accuracy:  0.769, Validation Accuracy:  0.773, Loss:  0.322
Epoch   2 Batch   87/134 - Train Accuracy:  0.767, Validation Accuracy:  0.780, Loss:  0.332
Epoch   2 Batch   88/134 - Train Accuracy:  0.790, Validation Accuracy:  0.780, Loss:  0.302
Epoch   2 Batch   89/134 - Train Accuracy:  0.775, Validation Accuracy:  0.787, Loss:  0.324
Epoch   2 Batch   90/134 - Train Accuracy:  0.777, Validation Accuracy:  0.777, Loss:  0.316
Epoch   2 Batch   91/134 - Train Accuracy:  0.810, Validation Accuracy:  0.780, Loss:  0.276
Epoch   2 Batch   92/134 - Train Accuracy:  0.784, Validation Accuracy:  0.782, Loss:  0.307
Epoch   2 Batch   93/134 - Train Accuracy:  0.786, Validation Accuracy:  0.776, Loss:  0.300
Epoch   2 Batch   94/134 - Train Accuracy:  0.787, Validation Accuracy:  0.791, Loss:  0.302
Epoch   2 Batch   95/134 - Train Accuracy:  0.775, Validation Accuracy:  0.784, Loss:  0.302
Epoch   2 Batch   96/134 - Train Accuracy:  0.794, Validation Accuracy:  0.788, Loss:  0.308
Epoch   2 Batch   97/134 - Train Accuracy:  0.775, Validation Accuracy:  0.784, Loss:  0.301
Epoch   2 Batch   98/134 - Train Accuracy:  0.772, Validation Accuracy:  0.783, Loss:  0.321
Epoch   2 Batch   99/134 - Train Accuracy:  0.784, Validation Accuracy:  0.790, Loss:  0.303
Epoch   2 Batch  100/134 - Train Accuracy:  0.795, Validation Accuracy:  0.789, Loss:  0.292
Epoch   2 Batch  101/134 - Train Accuracy:  0.781, Validation Accuracy:  0.788, Loss:  0.309
Epoch   2 Batch  102/134 - Train Accuracy:  0.795, Validation Accuracy:  0.793, Loss:  0.292
Epoch   2 Batch  103/134 - Train Accuracy:  0.807, Validation Accuracy:  0.794, Loss:  0.276
Epoch   2 Batch  104/134 - Train Accuracy:  0.798, Validation Accuracy:  0.796, Loss:  0.279
Epoch   2 Batch  105/134 - Train Accuracy:  0.800, Validation Accuracy:  0.788, Loss:  0.289
Epoch   2 Batch  106/134 - Train Accuracy:  0.795, Validation Accuracy:  0.785, Loss:  0.282
Epoch   2 Batch  107/134 - Train Accuracy:  0.809, Validation Accuracy:  0.796, Loss:  0.274
Epoch   2 Batch  108/134 - Train Accuracy:  0.792, Validation Accuracy:  0.797, Loss:  0.293
Epoch   2 Batch  109/134 - Train Accuracy:  0.809, Validation Accuracy:  0.800, Loss:  0.269
Epoch   2 Batch  110/134 - Train Accuracy:  0.812, Validation Accuracy:  0.804, Loss:  0.277
Epoch   2 Batch  111/134 - Train Accuracy:  0.802, Validation Accuracy:  0.806, Loss:  0.276
Epoch   2 Batch  112/134 - Train Accuracy:  0.808, Validation Accuracy:  0.808, Loss:  0.272
Epoch   2 Batch  113/134 - Train Accuracy:  0.819, Validation Accuracy:  0.804, Loss:  0.246
Epoch   2 Batch  114/134 - Train Accuracy:  0.792, Validation Accuracy:  0.809, Loss:  0.272
Epoch   2 Batch  115/134 - Train Accuracy:  0.796, Validation Accuracy:  0.807, Loss:  0.283
Epoch   2 Batch  116/134 - Train Accuracy:  0.809, Validation Accuracy:  0.805, Loss:  0.270
Epoch   2 Batch  117/134 - Train Accuracy:  0.803, Validation Accuracy:  0.810, Loss:  0.265
Epoch   2 Batch  118/134 - Train Accuracy:  0.812, Validation Accuracy:  0.809, Loss:  0.261
Epoch   2 Batch  119/134 - Train Accuracy:  0.820, Validation Accuracy:  0.805, Loss:  0.238
Epoch   2 Batch  120/134 - Train Accuracy:  0.800, Validation Accuracy:  0.803, Loss:  0.270
Epoch   2 Batch  121/134 - Train Accuracy:  0.806, Validation Accuracy:  0.812, Loss:  0.264
Epoch   2 Batch  122/134 - Train Accuracy:  0.809, Validation Accuracy:  0.814, Loss:  0.266
Epoch   2 Batch  123/134 - Train Accuracy:  0.819, Validation Accuracy:  0.815, Loss:  0.251
Epoch   2 Batch  124/134 - Train Accuracy:  0.826, Validation Accuracy:  0.811, Loss:  0.238
Epoch   2 Batch  125/134 - Train Accuracy:  0.827, Validation Accuracy:  0.818, Loss:  0.255
Epoch   2 Batch  126/134 - Train Accuracy:  0.799, Validation Accuracy:  0.824, Loss:  0.259
Epoch   2 Batch  127/134 - Train Accuracy:  0.813, Validation Accuracy:  0.820, Loss:  0.241
Epoch   2 Batch  128/134 - Train Accuracy:  0.809, Validation Accuracy:  0.812, Loss:  0.253
Epoch   2 Batch  129/134 - Train Accuracy:  0.816, Validation Accuracy:  0.817, Loss:  0.246
Epoch   2 Batch  130/134 - Train Accuracy:  0.829, Validation Accuracy:  0.822, Loss:  0.243
Epoch   2 Batch  131/134 - Train Accuracy:  0.807, Validation Accuracy:  0.832, Loss:  0.252
Epoch   2 Batch  132/134 - Train Accuracy:  0.833, Validation Accuracy:  0.828, Loss:  0.237
Epoch   3 Batch    0/134 - Train Accuracy:  0.828, Validation Accuracy:  0.821, Loss:  0.239
Epoch   3 Batch    1/134 - Train Accuracy:  0.808, Validation Accuracy:  0.820, Loss:  0.251
Epoch   3 Batch    2/134 - Train Accuracy:  0.822, Validation Accuracy:  0.817, Loss:  0.230
Epoch   3 Batch    3/134 - Train Accuracy:  0.821, Validation Accuracy:  0.823, Loss:  0.235
Epoch   3 Batch    4/134 - Train Accuracy:  0.826, Validation Accuracy:  0.828, Loss:  0.245
Epoch   3 Batch    5/134 - Train Accuracy:  0.815, Validation Accuracy:  0.817, Loss:  0.257
Epoch   3 Batch    6/134 - Train Accuracy:  0.813, Validation Accuracy:  0.817, Loss:  0.242
Epoch   3 Batch    7/134 - Train Accuracy:  0.800, Validation Accuracy:  0.824, Loss:  0.245
Epoch   3 Batch    8/134 - Train Accuracy:  0.807, Validation Accuracy:  0.809, Loss:  0.248
Epoch   3 Batch    9/134 - Train Accuracy:  0.829, Validation Accuracy:  0.818, Loss:  0.247
Epoch   3 Batch   10/134 - Train Accuracy:  0.822, Validation Accuracy:  0.819, Loss:  0.245
Epoch   3 Batch   11/134 - Train Accuracy:  0.820, Validation Accuracy:  0.826, Loss:  0.237
Epoch   3 Batch   12/134 - Train Accuracy:  0.836, Validation Accuracy:  0.835, Loss:  0.228
Epoch   3 Batch   13/134 - Train Accuracy:  0.813, Validation Accuracy:  0.831, Loss:  0.235
Epoch   3 Batch   14/134 - Train Accuracy:  0.830, Validation Accuracy:  0.828, Loss:  0.232
Epoch   3 Batch   15/134 - Train Accuracy:  0.821, Validation Accuracy:  0.827, Loss:  0.223
Epoch   3 Batch   16/134 - Train Accuracy:  0.824, Validation Accuracy:  0.829, Loss:  0.219
Epoch   3 Batch   17/134 - Train Accuracy:  0.820, Validation Accuracy:  0.828, Loss:  0.234
Epoch   3 Batch   18/134 - Train Accuracy:  0.830, Validation Accuracy:  0.836, Loss:  0.223
Epoch   3 Batch   19/134 - Train Accuracy:  0.828, Validation Accuracy:  0.842, Loss:  0.224
Epoch   3 Batch   20/134 - Train Accuracy:  0.838, Validation Accuracy:  0.839, Loss:  0.213
Epoch   3 Batch   21/134 - Train Accuracy:  0.832, Validation Accuracy:  0.833, Loss:  0.216
Epoch   3 Batch   22/134 - Train Accuracy:  0.825, Validation Accuracy:  0.842, Loss:  0.224
Epoch   3 Batch   23/134 - Train Accuracy:  0.846, Validation Accuracy:  0.837, Loss:  0.198
Epoch   3 Batch   24/134 - Train Accuracy:  0.827, Validation Accuracy:  0.839, Loss:  0.222
Epoch   3 Batch   25/134 - Train Accuracy:  0.837, Validation Accuracy:  0.840, Loss:  0.204
Epoch   3 Batch   26/134 - Train Accuracy:  0.834, Validation Accuracy:  0.843, Loss:  0.219
Epoch   3 Batch   27/134 - Train Accuracy:  0.844, Validation Accuracy:  0.846, Loss:  0.208
Epoch   3 Batch   28/134 - Train Accuracy:  0.837, Validation Accuracy:  0.848, Loss:  0.206
Epoch   3 Batch   29/134 - Train Accuracy:  0.854, Validation Accuracy:  0.849, Loss:  0.190
Epoch   3 Batch   30/134 - Train Accuracy:  0.856, Validation Accuracy:  0.846, Loss:  0.194
Epoch   3 Batch   31/134 - Train Accuracy:  0.849, Validation Accuracy:  0.845, Loss:  0.201
Epoch   3 Batch   32/134 - Train Accuracy:  0.856, Validation Accuracy:  0.852, Loss:  0.191
Epoch   3 Batch   33/134 - Train Accuracy:  0.839, Validation Accuracy:  0.854, Loss:  0.205
Epoch   3 Batch   34/134 - Train Accuracy:  0.841, Validation Accuracy:  0.855, Loss:  0.209
Epoch   3 Batch   35/134 - Train Accuracy:  0.852, Validation Accuracy:  0.849, Loss:  0.199
Epoch   3 Batch   36/134 - Train Accuracy:  0.851, Validation Accuracy:  0.854, Loss:  0.195
Epoch   3 Batch   37/134 - Train Accuracy:  0.851, Validation Accuracy:  0.858, Loss:  0.200
Epoch   3 Batch   38/134 - Train Accuracy:  0.849, Validation Accuracy:  0.853, Loss:  0.193
Epoch   3 Batch   39/134 - Train Accuracy:  0.846, Validation Accuracy:  0.854, Loss:  0.194
Epoch   3 Batch   40/134 - Train Accuracy:  0.858, Validation Accuracy:  0.856, Loss:  0.191
Epoch   3 Batch   41/134 - Train Accuracy:  0.846, Validation Accuracy:  0.859, Loss:  0.195
Epoch   3 Batch   42/134 - Train Accuracy:  0.847, Validation Accuracy:  0.859, Loss:  0.188
Epoch   3 Batch   43/134 - Train Accuracy:  0.835, Validation Accuracy:  0.857, Loss:  0.193
Epoch   3 Batch   44/134 - Train Accuracy:  0.851, Validation Accuracy:  0.859, Loss:  0.193
Epoch   3 Batch   45/134 - Train Accuracy:  0.863, Validation Accuracy:  0.860, Loss:  0.187
Epoch   3 Batch   46/134 - Train Accuracy:  0.858, Validation Accuracy:  0.859, Loss:  0.187
Epoch   3 Batch   47/134 - Train Accuracy:  0.850, Validation Accuracy:  0.866, Loss:  0.188
Epoch   3 Batch   48/134 - Train Accuracy:  0.856, Validation Accuracy:  0.861, Loss:  0.184
Epoch   3 Batch   49/134 - Train Accuracy:  0.863, Validation Accuracy:  0.867, Loss:  0.186
Epoch   3 Batch   50/134 - Train Accuracy:  0.845, Validation Accuracy:  0.860, Loss:  0.185
Epoch   3 Batch   51/134 - Train Accuracy:  0.855, Validation Accuracy:  0.863, Loss:  0.183
Epoch   3 Batch   52/134 - Train Accuracy:  0.852, Validation Accuracy:  0.868, Loss:  0.183
Epoch   3 Batch   53/134 - Train Accuracy:  0.861, Validation Accuracy:  0.861, Loss:  0.188
Epoch   3 Batch   54/134 - Train Accuracy:  0.854, Validation Accuracy:  0.869, Loss:  0.181
Epoch   3 Batch   55/134 - Train Accuracy:  0.859, Validation Accuracy:  0.864, Loss:  0.181
Epoch   3 Batch   56/134 - Train Accuracy:  0.864, Validation Accuracy:  0.865, Loss:  0.169
Epoch   3 Batch   57/134 - Train Accuracy:  0.864, Validation Accuracy:  0.865, Loss:  0.179
Epoch   3 Batch   58/134 - Train Accuracy:  0.870, Validation Accuracy:  0.865, Loss:  0.171
Epoch   3 Batch   59/134 - Train Accuracy:  0.853, Validation Accuracy:  0.862, Loss:  0.180
Epoch   3 Batch   60/134 - Train Accuracy:  0.859, Validation Accuracy:  0.873, Loss:  0.172
Epoch   3 Batch   61/134 - Train Accuracy:  0.861, Validation Accuracy:  0.870, Loss:  0.168
Epoch   3 Batch   62/134 - Train Accuracy:  0.859, Validation Accuracy:  0.875, Loss:  0.171
Epoch   3 Batch   63/134 - Train Accuracy:  0.864, Validation Accuracy:  0.871, Loss:  0.165
Epoch   3 Batch   64/134 - Train Accuracy:  0.868, Validation Accuracy:  0.872, Loss:  0.166
Epoch   3 Batch   65/134 - Train Accuracy:  0.855, Validation Accuracy:  0.878, Loss:  0.170
Epoch   3 Batch   66/134 - Train Accuracy:  0.866, Validation Accuracy:  0.876, Loss:  0.166
Epoch   3 Batch   67/134 - Train Accuracy:  0.859, Validation Accuracy:  0.884, Loss:  0.179
Epoch   3 Batch   68/134 - Train Accuracy:  0.864, Validation Accuracy:  0.878, Loss:  0.172
Epoch   3 Batch   69/134 - Train Accuracy:  0.863, Validation Accuracy:  0.869, Loss:  0.164
Epoch   3 Batch   70/134 - Train Accuracy:  0.868, Validation Accuracy:  0.876, Loss:  0.162
Epoch   3 Batch   71/134 - Train Accuracy:  0.875, Validation Accuracy:  0.880, Loss:  0.155
Epoch   3 Batch   72/134 - Train Accuracy:  0.873, Validation Accuracy:  0.882, Loss:  0.159
Epoch   3 Batch   73/134 - Train Accuracy:  0.875, Validation Accuracy:  0.880, Loss:  0.156
Epoch   3 Batch   74/134 - Train Accuracy:  0.869, Validation Accuracy:  0.880, Loss:  0.166
Epoch   3 Batch   75/134 - Train Accuracy:  0.872, Validation Accuracy:  0.880, Loss:  0.160
Epoch   3 Batch   76/134 - Train Accuracy:  0.877, Validation Accuracy:  0.876, Loss:  0.166
Epoch   3 Batch   77/134 - Train Accuracy:  0.876, Validation Accuracy:  0.879, Loss:  0.158
Epoch   3 Batch   78/134 - Train Accuracy:  0.860, Validation Accuracy:  0.871, Loss:  0.159
Epoch   3 Batch   79/134 - Train Accuracy:  0.870, Validation Accuracy:  0.880, Loss:  0.158
Epoch   3 Batch   80/134 - Train Accuracy:  0.876, Validation Accuracy:  0.885, Loss:  0.158
Epoch   3 Batch   81/134 - Train Accuracy:  0.871, Validation Accuracy:  0.880, Loss:  0.158
Epoch   3 Batch   82/134 - Train Accuracy:  0.886, Validation Accuracy:  0.886, Loss:  0.147
Epoch   3 Batch   83/134 - Train Accuracy:  0.880, Validation Accuracy:  0.886, Loss:  0.154
Epoch   3 Batch   84/134 - Train Accuracy:  0.881, Validation Accuracy:  0.895, Loss:  0.148
Epoch   3 Batch   85/134 - Train Accuracy:  0.888, Validation Accuracy:  0.893, Loss:  0.152
Epoch   3 Batch   86/134 - Train Accuracy:  0.892, Validation Accuracy:  0.889, Loss:  0.145
Epoch   3 Batch   87/134 - Train Accuracy:  0.865, Validation Accuracy:  0.890, Loss:  0.155
Epoch   3 Batch   88/134 - Train Accuracy:  0.881, Validation Accuracy:  0.889, Loss:  0.140
Epoch   3 Batch   89/134 - Train Accuracy:  0.879, Validation Accuracy:  0.890, Loss:  0.148
Epoch   3 Batch   90/134 - Train Accuracy:  0.881, Validation Accuracy:  0.887, Loss:  0.149
Epoch   3 Batch   91/134 - Train Accuracy:  0.900, Validation Accuracy:  0.886, Loss:  0.124
Epoch   3 Batch   92/134 - Train Accuracy:  0.889, Validation Accuracy:  0.885, Loss:  0.137
Epoch   3 Batch   93/134 - Train Accuracy:  0.888, Validation Accuracy:  0.888, Loss:  0.136
Epoch   3 Batch   94/134 - Train Accuracy:  0.887, Validation Accuracy:  0.888, Loss:  0.135
Epoch   3 Batch   95/134 - Train Accuracy:  0.882, Validation Accuracy:  0.894, Loss:  0.137
Epoch   3 Batch   96/134 - Train Accuracy:  0.886, Validation Accuracy:  0.894, Loss:  0.141
Epoch   3 Batch   97/134 - Train Accuracy:  0.881, Validation Accuracy:  0.894, Loss:  0.138
Epoch   3 Batch   98/134 - Train Accuracy:  0.880, Validation Accuracy:  0.895, Loss:  0.142
Epoch   3 Batch   99/134 - Train Accuracy:  0.882, Validation Accuracy:  0.893, Loss:  0.140
Epoch   3 Batch  100/134 - Train Accuracy:  0.884, Validation Accuracy:  0.890, Loss:  0.135
Epoch   3 Batch  101/134 - Train Accuracy:  0.877, Validation Accuracy:  0.884, Loss:  0.141
Epoch   3 Batch  102/134 - Train Accuracy:  0.877, Validation Accuracy:  0.886, Loss:  0.134
Epoch   3 Batch  103/134 - Train Accuracy:  0.891, Validation Accuracy:  0.892, Loss:  0.130
Epoch   3 Batch  104/134 - Train Accuracy:  0.892, Validation Accuracy:  0.897, Loss:  0.127
Epoch   3 Batch  105/134 - Train Accuracy:  0.895, Validation Accuracy:  0.899, Loss:  0.135
Epoch   3 Batch  106/134 - Train Accuracy:  0.886, Validation Accuracy:  0.893, Loss:  0.132
Epoch   3 Batch  107/134 - Train Accuracy:  0.889, Validation Accuracy:  0.886, Loss:  0.133
Epoch   3 Batch  108/134 - Train Accuracy:  0.887, Validation Accuracy:  0.886, Loss:  0.131
Epoch   3 Batch  109/134 - Train Accuracy:  0.900, Validation Accuracy:  0.895, Loss:  0.127
Epoch   3 Batch  110/134 - Train Accuracy:  0.907, Validation Accuracy:  0.896, Loss:  0.122
Epoch   3 Batch  111/134 - Train Accuracy:  0.889, Validation Accuracy:  0.896, Loss:  0.131
Epoch   3 Batch  112/134 - Train Accuracy:  0.885, Validation Accuracy:  0.901, Loss:  0.126
Epoch   3 Batch  113/134 - Train Accuracy:  0.900, Validation Accuracy:  0.896, Loss:  0.114
Epoch   3 Batch  114/134 - Train Accuracy:  0.887, Validation Accuracy:  0.896, Loss:  0.126
Epoch   3 Batch  115/134 - Train Accuracy:  0.880, Validation Accuracy:  0.898, Loss:  0.133
Epoch   3 Batch  116/134 - Train Accuracy:  0.895, Validation Accuracy:  0.896, Loss:  0.123
Epoch   3 Batch  117/134 - Train Accuracy:  0.900, Validation Accuracy:  0.903, Loss:  0.120
Epoch   3 Batch  118/134 - Train Accuracy:  0.900, Validation Accuracy:  0.905, Loss:  0.119
Epoch   3 Batch  119/134 - Train Accuracy:  0.898, Validation Accuracy:  0.902, Loss:  0.111
Epoch   3 Batch  120/134 - Train Accuracy:  0.884, Validation Accuracy:  0.899, Loss:  0.126
Epoch   3 Batch  121/134 - Train Accuracy:  0.892, Validation Accuracy:  0.896, Loss:  0.121
Epoch   3 Batch  122/134 - Train Accuracy:  0.877, Validation Accuracy:  0.896, Loss:  0.119
Epoch   3 Batch  123/134 - Train Accuracy:  0.896, Validation Accuracy:  0.898, Loss:  0.114
Epoch   3 Batch  124/134 - Train Accuracy:  0.907, Validation Accuracy:  0.903, Loss:  0.110
Epoch   3 Batch  125/134 - Train Accuracy:  0.906, Validation Accuracy:  0.907, Loss:  0.115
Epoch   3 Batch  126/134 - Train Accuracy:  0.879, Validation Accuracy:  0.902, Loss:  0.120
Epoch   3 Batch  127/134 - Train Accuracy:  0.890, Validation Accuracy:  0.902, Loss:  0.113
Epoch   3 Batch  128/134 - Train Accuracy:  0.890, Validation Accuracy:  0.907, Loss:  0.119
Epoch   3 Batch  129/134 - Train Accuracy:  0.897, Validation Accuracy:  0.909, Loss:  0.114
Epoch   3 Batch  130/134 - Train Accuracy:  0.903, Validation Accuracy:  0.908, Loss:  0.109
Epoch   3 Batch  131/134 - Train Accuracy:  0.883, Validation Accuracy:  0.905, Loss:  0.117
Epoch   3 Batch  132/134 - Train Accuracy:  0.903, Validation Accuracy:  0.906, Loss:  0.108
Epoch   4 Batch    0/134 - Train Accuracy:  0.899, Validation Accuracy:  0.904, Loss:  0.108
Epoch   4 Batch    1/134 - Train Accuracy:  0.890, Validation Accuracy:  0.902, Loss:  0.112
Epoch   4 Batch    2/134 - Train Accuracy:  0.909, Validation Accuracy:  0.909, Loss:  0.107
Epoch   4 Batch    3/134 - Train Accuracy:  0.910, Validation Accuracy:  0.909, Loss:  0.109
Epoch   4 Batch    4/134 - Train Accuracy:  0.896, Validation Accuracy:  0.905, Loss:  0.112
Epoch   4 Batch    5/134 - Train Accuracy:  0.900, Validation Accuracy:  0.907, Loss:  0.115
Epoch   4 Batch    6/134 - Train Accuracy:  0.896, Validation Accuracy:  0.901, Loss:  0.097
Epoch   4 Batch    7/134 - Train Accuracy:  0.902, Validation Accuracy:  0.906, Loss:  0.106
Epoch   4 Batch    8/134 - Train Accuracy:  0.906, Validation Accuracy:  0.911, Loss:  0.101
Epoch   4 Batch    9/134 - Train Accuracy:  0.918, Validation Accuracy:  0.914, Loss:  0.094
Epoch   4 Batch   10/134 - Train Accuracy:  0.906, Validation Accuracy:  0.917, Loss:  0.106
Epoch   4 Batch   11/134 - Train Accuracy:  0.897, Validation Accuracy:  0.916, Loss:  0.102
Epoch   4 Batch   12/134 - Train Accuracy:  0.905, Validation Accuracy:  0.917, Loss:  0.097
Epoch   4 Batch   13/134 - Train Accuracy:  0.895, Validation Accuracy:  0.912, Loss:  0.099
Epoch   4 Batch   14/134 - Train Accuracy:  0.903, Validation Accuracy:  0.911, Loss:  0.097
Epoch   4 Batch   15/134 - Train Accuracy:  0.900, Validation Accuracy:  0.910, Loss:  0.096
Epoch   4 Batch   16/134 - Train Accuracy:  0.898, Validation Accuracy:  0.902, Loss:  0.094
Epoch   4 Batch   17/134 - Train Accuracy:  0.902, Validation Accuracy:  0.910, Loss:  0.107
Epoch   4 Batch   18/134 - Train Accuracy:  0.902, Validation Accuracy:  0.913, Loss:  0.099
Epoch   4 Batch   19/134 - Train Accuracy:  0.911, Validation Accuracy:  0.912, Loss:  0.099
Epoch   4 Batch   20/134 - Train Accuracy:  0.910, Validation Accuracy:  0.912, Loss:  0.093
Epoch   4 Batch   21/134 - Train Accuracy:  0.912, Validation Accuracy:  0.919, Loss:  0.096
Epoch   4 Batch   22/134 - Train Accuracy:  0.903, Validation Accuracy:  0.918, Loss:  0.095
Epoch   4 Batch   23/134 - Train Accuracy:  0.923, Validation Accuracy:  0.917, Loss:  0.084
Epoch   4 Batch   24/134 - Train Accuracy:  0.901, Validation Accuracy:  0.917, Loss:  0.098
Epoch   4 Batch   25/134 - Train Accuracy:  0.910, Validation Accuracy:  0.917, Loss:  0.088
Epoch   4 Batch   26/134 - Train Accuracy:  0.913, Validation Accuracy:  0.920, Loss:  0.096
Epoch   4 Batch   27/134 - Train Accuracy:  0.906, Validation Accuracy:  0.923, Loss:  0.092
Epoch   4 Batch   28/134 - Train Accuracy:  0.908, Validation Accuracy:  0.922, Loss:  0.095
Epoch   4 Batch   29/134 - Train Accuracy:  0.926, Validation Accuracy:  0.925, Loss:  0.083
Epoch   4 Batch   30/134 - Train Accuracy:  0.919, Validation Accuracy:  0.924, Loss:  0.087
Epoch   4 Batch   31/134 - Train Accuracy:  0.913, Validation Accuracy:  0.924, Loss:  0.089
Epoch   4 Batch   32/134 - Train Accuracy:  0.918, Validation Accuracy:  0.923, Loss:  0.084
Epoch   4 Batch   33/134 - Train Accuracy:  0.906, Validation Accuracy:  0.922, Loss:  0.093
Epoch   4 Batch   34/134 - Train Accuracy:  0.907, Validation Accuracy:  0.919, Loss:  0.096
Epoch   4 Batch   35/134 - Train Accuracy:  0.911, Validation Accuracy:  0.920, Loss:  0.092
Epoch   4 Batch   36/134 - Train Accuracy:  0.917, Validation Accuracy:  0.921, Loss:  0.084
Epoch   4 Batch   37/134 - Train Accuracy:  0.909, Validation Accuracy:  0.921, Loss:  0.091
Epoch   4 Batch   38/134 - Train Accuracy:  0.918, Validation Accuracy:  0.927, Loss:  0.084
Epoch   4 Batch   39/134 - Train Accuracy:  0.915, Validation Accuracy:  0.927, Loss:  0.083
Epoch   4 Batch   40/134 - Train Accuracy:  0.926, Validation Accuracy:  0.922, Loss:  0.084
Epoch   4 Batch   41/134 - Train Accuracy:  0.905, Validation Accuracy:  0.920, Loss:  0.086
Epoch   4 Batch   42/134 - Train Accuracy:  0.910, Validation Accuracy:  0.921, Loss:  0.079
Epoch   4 Batch   43/134 - Train Accuracy:  0.912, Validation Accuracy:  0.918, Loss:  0.085
Epoch   4 Batch   44/134 - Train Accuracy:  0.920, Validation Accuracy:  0.923, Loss:  0.082
Epoch   4 Batch   45/134 - Train Accuracy:  0.927, Validation Accuracy:  0.923, Loss:  0.075
Epoch   4 Batch   46/134 - Train Accuracy:  0.920, Validation Accuracy:  0.926, Loss:  0.084
Epoch   4 Batch   47/134 - Train Accuracy:  0.918, Validation Accuracy:  0.928, Loss:  0.081
Epoch   4 Batch   48/134 - Train Accuracy:  0.925, Validation Accuracy:  0.929, Loss:  0.079
Epoch   4 Batch   49/134 - Train Accuracy:  0.920, Validation Accuracy:  0.931, Loss:  0.076
Epoch   4 Batch   50/134 - Train Accuracy:  0.919, Validation Accuracy:  0.927, Loss:  0.079
Epoch   4 Batch   51/134 - Train Accuracy:  0.919, Validation Accuracy:  0.926, Loss:  0.081
Epoch   4 Batch   52/134 - Train Accuracy:  0.918, Validation Accuracy:  0.932, Loss:  0.075
Epoch   4 Batch   53/134 - Train Accuracy:  0.932, Validation Accuracy:  0.930, Loss:  0.074
Epoch   4 Batch   54/134 - Train Accuracy:  0.913, Validation Accuracy:  0.925, Loss:  0.078
Epoch   4 Batch   55/134 - Train Accuracy:  0.914, Validation Accuracy:  0.930, Loss:  0.080
Epoch   4 Batch   56/134 - Train Accuracy:  0.919, Validation Accuracy:  0.931, Loss:  0.074
Epoch   4 Batch   57/134 - Train Accuracy:  0.922, Validation Accuracy:  0.927, Loss:  0.075
Epoch   4 Batch   58/134 - Train Accuracy:  0.922, Validation Accuracy:  0.926, Loss:  0.071
Epoch   4 Batch   59/134 - Train Accuracy:  0.911, Validation Accuracy:  0.929, Loss:  0.080
Epoch   4 Batch   60/134 - Train Accuracy:  0.918, Validation Accuracy:  0.921, Loss:  0.070
Epoch   4 Batch   61/134 - Train Accuracy:  0.923, Validation Accuracy:  0.929, Loss:  0.071
Epoch   4 Batch   62/134 - Train Accuracy:  0.912, Validation Accuracy:  0.920, Loss:  0.072
Epoch   4 Batch   63/134 - Train Accuracy:  0.913, Validation Accuracy:  0.918, Loss:  0.072
Epoch   4 Batch   64/134 - Train Accuracy:  0.919, Validation Accuracy:  0.924, Loss:  0.072
Epoch   4 Batch   65/134 - Train Accuracy:  0.900, Validation Accuracy:  0.928, Loss:  0.075
Epoch   4 Batch   66/134 - Train Accuracy:  0.919, Validation Accuracy:  0.929, Loss:  0.067
Epoch   4 Batch   67/134 - Train Accuracy:  0.911, Validation Accuracy:  0.928, Loss:  0.074
Epoch   4 Batch   68/134 - Train Accuracy:  0.908, Validation Accuracy:  0.928, Loss:  0.073
Epoch   4 Batch   69/134 - Train Accuracy:  0.923, Validation Accuracy:  0.928, Loss:  0.069
Epoch   4 Batch   70/134 - Train Accuracy:  0.924, Validation Accuracy:  0.930, Loss:  0.070
Epoch   4 Batch   71/134 - Train Accuracy:  0.930, Validation Accuracy:  0.929, Loss:  0.062
Epoch   4 Batch   72/134 - Train Accuracy:  0.928, Validation Accuracy:  0.927, Loss:  0.066
Epoch   4 Batch   73/134 - Train Accuracy:  0.923, Validation Accuracy:  0.934, Loss:  0.068
Epoch   4 Batch   74/134 - Train Accuracy:  0.928, Validation Accuracy:  0.938, Loss:  0.073
Epoch   4 Batch   75/134 - Train Accuracy:  0.937, Validation Accuracy:  0.942, Loss:  0.064
Epoch   4 Batch   76/134 - Train Accuracy:  0.932, Validation Accuracy:  0.937, Loss:  0.063
Epoch   4 Batch   77/134 - Train Accuracy:  0.926, Validation Accuracy:  0.937, Loss:  0.063
Epoch   4 Batch   78/134 - Train Accuracy:  0.920, Validation Accuracy:  0.937, Loss:  0.065
Epoch   4 Batch   79/134 - Train Accuracy:  0.922, Validation Accuracy:  0.932, Loss:  0.065
Epoch   4 Batch   80/134 - Train Accuracy:  0.935, Validation Accuracy:  0.936, Loss:  0.063
Epoch   4 Batch   81/134 - Train Accuracy:  0.941, Validation Accuracy:  0.937, Loss:  0.064
Epoch   4 Batch   82/134 - Train Accuracy:  0.925, Validation Accuracy:  0.940, Loss:  0.059
Epoch   4 Batch   83/134 - Train Accuracy:  0.926, Validation Accuracy:  0.940, Loss:  0.065
Epoch   4 Batch   84/134 - Train Accuracy:  0.929, Validation Accuracy:  0.943, Loss:  0.059
Epoch   4 Batch   85/134 - Train Accuracy:  0.927, Validation Accuracy:  0.938, Loss:  0.063
Epoch   4 Batch   86/134 - Train Accuracy:  0.932, Validation Accuracy:  0.936, Loss:  0.061
Epoch   4 Batch   87/134 - Train Accuracy:  0.916, Validation Accuracy:  0.936, Loss:  0.069
Epoch   4 Batch   88/134 - Train Accuracy:  0.938, Validation Accuracy:  0.936, Loss:  0.056
Epoch   4 Batch   89/134 - Train Accuracy:  0.930, Validation Accuracy:  0.938, Loss:  0.059
Epoch   4 Batch   90/134 - Train Accuracy:  0.919, Validation Accuracy:  0.937, Loss:  0.064
Epoch   4 Batch   91/134 - Train Accuracy:  0.941, Validation Accuracy:  0.941, Loss:  0.049
Epoch   4 Batch   92/134 - Train Accuracy:  0.934, Validation Accuracy:  0.940, Loss:  0.056
Epoch   4 Batch   93/134 - Train Accuracy:  0.934, Validation Accuracy:  0.939, Loss:  0.059
Epoch   4 Batch   94/134 - Train Accuracy:  0.930, Validation Accuracy:  0.939, Loss:  0.056
Epoch   4 Batch   95/134 - Train Accuracy:  0.925, Validation Accuracy:  0.938, Loss:  0.060
Epoch   4 Batch   96/134 - Train Accuracy:  0.933, Validation Accuracy:  0.937, Loss:  0.061
Epoch   4 Batch   97/134 - Train Accuracy:  0.923, Validation Accuracy:  0.940, Loss:  0.059
Epoch   4 Batch   98/134 - Train Accuracy:  0.931, Validation Accuracy:  0.943, Loss:  0.060
Epoch   4 Batch   99/134 - Train Accuracy:  0.930, Validation Accuracy:  0.945, Loss:  0.059
Epoch   4 Batch  100/134 - Train Accuracy:  0.933, Validation Accuracy:  0.944, Loss:  0.057
Epoch   4 Batch  101/134 - Train Accuracy:  0.929, Validation Accuracy:  0.936, Loss:  0.059
Epoch   4 Batch  102/134 - Train Accuracy:  0.932, Validation Accuracy:  0.941, Loss:  0.055
Epoch   4 Batch  103/134 - Train Accuracy:  0.937, Validation Accuracy:  0.945, Loss:  0.057
Epoch   4 Batch  104/134 - Train Accuracy:  0.939, Validation Accuracy:  0.943, Loss:  0.053
Epoch   4 Batch  105/134 - Train Accuracy:  0.938, Validation Accuracy:  0.944, Loss:  0.060
Epoch   4 Batch  106/134 - Train Accuracy:  0.931, Validation Accuracy:  0.941, Loss:  0.056
Epoch   4 Batch  107/134 - Train Accuracy:  0.934, Validation Accuracy:  0.937, Loss:  0.060
Epoch   4 Batch  108/134 - Train Accuracy:  0.928, Validation Accuracy:  0.936, Loss:  0.057
Epoch   4 Batch  109/134 - Train Accuracy:  0.935, Validation Accuracy:  0.938, Loss:  0.055
Epoch   4 Batch  110/134 - Train Accuracy:  0.940, Validation Accuracy:  0.940, Loss:  0.054
Epoch   4 Batch  111/134 - Train Accuracy:  0.928, Validation Accuracy:  0.944, Loss:  0.056
Epoch   4 Batch  112/134 - Train Accuracy:  0.930, Validation Accuracy:  0.944, Loss:  0.057
Epoch   4 Batch  113/134 - Train Accuracy:  0.938, Validation Accuracy:  0.943, Loss:  0.053
Epoch   4 Batch  114/134 - Train Accuracy:  0.936, Validation Accuracy:  0.948, Loss:  0.057
Epoch   4 Batch  115/134 - Train Accuracy:  0.927, Validation Accuracy:  0.939, Loss:  0.056
Epoch   4 Batch  116/134 - Train Accuracy:  0.938, Validation Accuracy:  0.938, Loss:  0.056
Epoch   4 Batch  117/134 - Train Accuracy:  0.948, Validation Accuracy:  0.943, Loss:  0.051
Epoch   4 Batch  118/134 - Train Accuracy:  0.941, Validation Accuracy:  0.939, Loss:  0.053
Epoch   4 Batch  119/134 - Train Accuracy:  0.944, Validation Accuracy:  0.949, Loss:  0.048
Epoch   4 Batch  120/134 - Train Accuracy:  0.936, Validation Accuracy:  0.948, Loss:  0.057
Epoch   4 Batch  121/134 - Train Accuracy:  0.940, Validation Accuracy:  0.944, Loss:  0.052
Epoch   4 Batch  122/134 - Train Accuracy:  0.934, Validation Accuracy:  0.942, Loss:  0.055
Epoch   4 Batch  123/134 - Train Accuracy:  0.939, Validation Accuracy:  0.941, Loss:  0.049
Epoch   4 Batch  124/134 - Train Accuracy:  0.939, Validation Accuracy:  0.940, Loss:  0.049
Epoch   4 Batch  125/134 - Train Accuracy:  0.951, Validation Accuracy:  0.945, Loss:  0.048
Epoch   4 Batch  126/134 - Train Accuracy:  0.920, Validation Accuracy:  0.946, Loss:  0.056
Epoch   4 Batch  127/134 - Train Accuracy:  0.936, Validation Accuracy:  0.949, Loss:  0.052
Epoch   4 Batch  128/134 - Train Accuracy:  0.931, Validation Accuracy:  0.940, Loss:  0.056
Epoch   4 Batch  129/134 - Train Accuracy:  0.934, Validation Accuracy:  0.939, Loss:  0.054
Epoch   4 Batch  130/134 - Train Accuracy:  0.937, Validation Accuracy:  0.945, Loss:  0.051
Epoch   4 Batch  131/134 - Train Accuracy:  0.929, Validation Accuracy:  0.942, Loss:  0.055
Epoch   4 Batch  132/134 - Train Accuracy:  0.933, Validation Accuracy:  0.944, Loss:  0.050
Epoch   5 Batch    0/134 - Train Accuracy:  0.931, Validation Accuracy:  0.935, Loss:  0.051
Epoch   5 Batch    1/134 - Train Accuracy:  0.934, Validation Accuracy:  0.942, Loss:  0.052
Epoch   5 Batch    2/134 - Train Accuracy:  0.943, Validation Accuracy:  0.945, Loss:  0.047
Epoch   5 Batch    3/134 - Train Accuracy:  0.936, Validation Accuracy:  0.940, Loss:  0.052
Epoch   5 Batch    4/134 - Train Accuracy:  0.930, Validation Accuracy:  0.944, Loss:  0.051
Epoch   5 Batch    5/134 - Train Accuracy:  0.932, Validation Accuracy:  0.942, Loss:  0.057
Epoch   5 Batch    6/134 - Train Accuracy:  0.935, Validation Accuracy:  0.943, Loss:  0.043
Epoch   5 Batch    7/134 - Train Accuracy:  0.938, Validation Accuracy:  0.947, Loss:  0.049
Epoch   5 Batch    8/134 - Train Accuracy:  0.937, Validation Accuracy:  0.951, Loss:  0.045
Epoch   5 Batch    9/134 - Train Accuracy:  0.943, Validation Accuracy:  0.953, Loss:  0.043
Epoch   5 Batch   10/134 - Train Accuracy:  0.934, Validation Accuracy:  0.949, Loss:  0.051
Epoch   5 Batch   11/134 - Train Accuracy:  0.933, Validation Accuracy:  0.953, Loss:  0.051
Epoch   5 Batch   12/134 - Train Accuracy:  0.936, Validation Accuracy:  0.945, Loss:  0.047
Epoch   5 Batch   13/134 - Train Accuracy:  0.935, Validation Accuracy:  0.948, Loss:  0.046
Epoch   5 Batch   14/134 - Train Accuracy:  0.939, Validation Accuracy:  0.940, Loss:  0.047
Epoch   5 Batch   15/134 - Train Accuracy:  0.945, Validation Accuracy:  0.945, Loss:  0.048
Epoch   5 Batch   16/134 - Train Accuracy:  0.942, Validation Accuracy:  0.945, Loss:  0.045
Epoch   5 Batch   17/134 - Train Accuracy:  0.934, Validation Accuracy:  0.950, Loss:  0.055
Epoch   5 Batch   18/134 - Train Accuracy:  0.938, Validation Accuracy:  0.947, Loss:  0.048
Epoch   5 Batch   19/134 - Train Accuracy:  0.941, Validation Accuracy:  0.947, Loss:  0.049
Epoch   5 Batch   20/134 - Train Accuracy:  0.946, Validation Accuracy:  0.944, Loss:  0.046
Epoch   5 Batch   21/134 - Train Accuracy:  0.945, Validation Accuracy:  0.949, Loss:  0.048
Epoch   5 Batch   22/134 - Train Accuracy:  0.942, Validation Accuracy:  0.950, Loss:  0.045
Epoch   5 Batch   23/134 - Train Accuracy:  0.953, Validation Accuracy:  0.951, Loss:  0.039
Epoch   5 Batch   24/134 - Train Accuracy:  0.938, Validation Accuracy:  0.948, Loss:  0.050
Epoch   5 Batch   25/134 - Train Accuracy:  0.941, Validation Accuracy:  0.950, Loss:  0.041
Epoch   5 Batch   26/134 - Train Accuracy:  0.942, Validation Accuracy:  0.952, Loss:  0.046
Epoch   5 Batch   27/134 - Train Accuracy:  0.939, Validation Accuracy:  0.954, Loss:  0.045
Epoch   5 Batch   28/134 - Train Accuracy:  0.937, Validation Accuracy:  0.953, Loss:  0.048
Epoch   5 Batch   29/134 - Train Accuracy:  0.952, Validation Accuracy:  0.953, Loss:  0.040
Epoch   5 Batch   30/134 - Train Accuracy:  0.938, Validation Accuracy:  0.955, Loss:  0.047
Epoch   5 Batch   31/134 - Train Accuracy:  0.936, Validation Accuracy:  0.955, Loss:  0.046
Epoch   5 Batch   32/134 - Train Accuracy:  0.947, Validation Accuracy:  0.955, Loss:  0.043
Epoch   5 Batch   33/134 - Train Accuracy:  0.938, Validation Accuracy:  0.948, Loss:  0.053
Epoch   5 Batch   34/134 - Train Accuracy:  0.936, Validation Accuracy:  0.951, Loss:  0.052
Epoch   5 Batch   35/134 - Train Accuracy:  0.943, Validation Accuracy:  0.952, Loss:  0.050
Epoch   5 Batch   36/134 - Train Accuracy:  0.943, Validation Accuracy:  0.951, Loss:  0.047
Epoch   5 Batch   37/134 - Train Accuracy:  0.933, Validation Accuracy:  0.949, Loss:  0.048
Epoch   5 Batch   38/134 - Train Accuracy:  0.945, Validation Accuracy:  0.949, Loss:  0.045
Epoch   5 Batch   39/134 - Train Accuracy:  0.944, Validation Accuracy:  0.949, Loss:  0.045
Epoch   5 Batch   40/134 - Train Accuracy:  0.946, Validation Accuracy:  0.949, Loss:  0.046
Epoch   5 Batch   41/134 - Train Accuracy:  0.935, Validation Accuracy:  0.948, Loss:  0.048
Epoch   5 Batch   42/134 - Train Accuracy:  0.941, Validation Accuracy:  0.949, Loss:  0.043
Epoch   5 Batch   43/134 - Train Accuracy:  0.937, Validation Accuracy:  0.947, Loss:  0.046
Epoch   5 Batch   44/134 - Train Accuracy:  0.942, Validation Accuracy:  0.952, Loss:  0.044
Epoch   5 Batch   45/134 - Train Accuracy:  0.955, Validation Accuracy:  0.947, Loss:  0.040
Epoch   5 Batch   46/134 - Train Accuracy:  0.945, Validation Accuracy:  0.951, Loss:  0.047
Epoch   5 Batch   47/134 - Train Accuracy:  0.940, Validation Accuracy:  0.950, Loss:  0.044
Epoch   5 Batch   48/134 - Train Accuracy:  0.940, Validation Accuracy:  0.951, Loss:  0.045
Epoch   5 Batch   49/134 - Train Accuracy:  0.949, Validation Accuracy:  0.953, Loss:  0.042
Epoch   5 Batch   50/134 - Train Accuracy:  0.948, Validation Accuracy:  0.953, Loss:  0.046
Epoch   5 Batch   51/134 - Train Accuracy:  0.944, Validation Accuracy:  0.952, Loss:  0.046
Epoch   5 Batch   52/134 - Train Accuracy:  0.948, Validation Accuracy:  0.949, Loss:  0.041
Epoch   5 Batch   53/134 - Train Accuracy:  0.950, Validation Accuracy:  0.949, Loss:  0.043
Epoch   5 Batch   54/134 - Train Accuracy:  0.940, Validation Accuracy:  0.953, Loss:  0.045
Epoch   5 Batch   55/134 - Train Accuracy:  0.942, Validation Accuracy:  0.952, Loss:  0.048
Epoch   5 Batch   56/134 - Train Accuracy:  0.941, Validation Accuracy:  0.955, Loss:  0.043
Epoch   5 Batch   57/134 - Train Accuracy:  0.950, Validation Accuracy:  0.952, Loss:  0.043
Epoch   5 Batch   58/134 - Train Accuracy:  0.946, Validation Accuracy:  0.952, Loss:  0.039
Epoch   5 Batch   59/134 - Train Accuracy:  0.948, Validation Accuracy:  0.954, Loss:  0.046
Epoch   5 Batch   60/134 - Train Accuracy:  0.950, Validation Accuracy:  0.958, Loss:  0.041
Epoch   5 Batch   61/134 - Train Accuracy:  0.949, Validation Accuracy:  0.956, Loss:  0.040
Epoch   5 Batch   62/134 - Train Accuracy:  0.944, Validation Accuracy:  0.947, Loss:  0.041
Epoch   5 Batch   63/134 - Train Accuracy:  0.935, Validation Accuracy:  0.944, Loss:  0.044
Epoch   5 Batch   64/134 - Train Accuracy:  0.948, Validation Accuracy:  0.952, Loss:  0.044
Epoch   5 Batch   65/134 - Train Accuracy:  0.935, Validation Accuracy:  0.951, Loss:  0.046
Epoch   5 Batch   66/134 - Train Accuracy:  0.939, Validation Accuracy:  0.946, Loss:  0.039
Epoch   5 Batch   67/134 - Train Accuracy:  0.944, Validation Accuracy:  0.950, Loss:  0.045
Epoch   5 Batch   68/134 - Train Accuracy:  0.934, Validation Accuracy:  0.952, Loss:  0.045
Epoch   5 Batch   69/134 - Train Accuracy:  0.944, Validation Accuracy:  0.949, Loss:  0.042
Epoch   5 Batch   70/134 - Train Accuracy:  0.943, Validation Accuracy:  0.949, Loss:  0.042
Epoch   5 Batch   71/134 - Train Accuracy:  0.955, Validation Accuracy:  0.953, Loss:  0.038
Epoch   5 Batch   72/134 - Train Accuracy:  0.951, Validation Accuracy:  0.957, Loss:  0.039
Epoch   5 Batch   73/134 - Train Accuracy:  0.945, Validation Accuracy:  0.956, Loss:  0.043
Epoch   5 Batch   74/134 - Train Accuracy:  0.942, Validation Accuracy:  0.954, Loss:  0.045
Epoch   5 Batch   75/134 - Train Accuracy:  0.951, Validation Accuracy:  0.953, Loss:  0.040
Epoch   5 Batch   76/134 - Train Accuracy:  0.954, Validation Accuracy:  0.954, Loss:  0.039
Epoch   5 Batch   77/134 - Train Accuracy:  0.950, Validation Accuracy:  0.952, Loss:  0.039
Epoch   5 Batch   78/134 - Train Accuracy:  0.940, Validation Accuracy:  0.950, Loss:  0.041
Epoch   5 Batch   79/134 - Train Accuracy:  0.950, Validation Accuracy:  0.949, Loss:  0.040
Epoch   5 Batch   80/134 - Train Accuracy:  0.957, Validation Accuracy:  0.951, Loss:  0.039
Epoch   5 Batch   81/134 - Train Accuracy:  0.959, Validation Accuracy:  0.953, Loss:  0.039
Epoch   5 Batch   82/134 - Train Accuracy:  0.943, Validation Accuracy:  0.955, Loss:  0.037
Epoch   5 Batch   83/134 - Train Accuracy:  0.951, Validation Accuracy:  0.957, Loss:  0.040
Epoch   5 Batch   84/134 - Train Accuracy:  0.951, Validation Accuracy:  0.957, Loss:  0.037
Epoch   5 Batch   85/134 - Train Accuracy:  0.950, Validation Accuracy:  0.958, Loss:  0.041
Epoch   5 Batch   86/134 - Train Accuracy:  0.953, Validation Accuracy:  0.958, Loss:  0.038
Epoch   5 Batch   87/134 - Train Accuracy:  0.935, Validation Accuracy:  0.951, Loss:  0.046
Epoch   5 Batch   88/134 - Train Accuracy:  0.953, Validation Accuracy:  0.951, Loss:  0.035
Epoch   5 Batch   89/134 - Train Accuracy:  0.949, Validation Accuracy:  0.956, Loss:  0.038
Epoch   5 Batch   90/134 - Train Accuracy:  0.943, Validation Accuracy:  0.956, Loss:  0.041
Epoch   5 Batch   91/134 - Train Accuracy:  0.955, Validation Accuracy:  0.957, Loss:  0.031
Epoch   5 Batch   92/134 - Train Accuracy:  0.954, Validation Accuracy:  0.956, Loss:  0.036
Epoch   5 Batch   93/134 - Train Accuracy:  0.950, Validation Accuracy:  0.952, Loss:  0.037
Epoch   5 Batch   94/134 - Train Accuracy:  0.949, Validation Accuracy:  0.955, Loss:  0.038
Epoch   5 Batch   95/134 - Train Accuracy:  0.947, Validation Accuracy:  0.952, Loss:  0.037
Epoch   5 Batch   96/134 - Train Accuracy:  0.941, Validation Accuracy:  0.957, Loss:  0.042
Epoch   5 Batch   97/134 - Train Accuracy:  0.942, Validation Accuracy:  0.952, Loss:  0.038
Epoch   5 Batch   98/134 - Train Accuracy:  0.945, Validation Accuracy:  0.953, Loss:  0.039
Epoch   5 Batch   99/134 - Train Accuracy:  0.956, Validation Accuracy:  0.954, Loss:  0.036
Epoch   5 Batch  100/134 - Train Accuracy:  0.954, Validation Accuracy:  0.956, Loss:  0.038
Epoch   5 Batch  101/134 - Train Accuracy:  0.953, Validation Accuracy:  0.958, Loss:  0.039
Epoch   5 Batch  102/134 - Train Accuracy:  0.951, Validation Accuracy:  0.959, Loss:  0.036
Epoch   5 Batch  103/134 - Train Accuracy:  0.955, Validation Accuracy:  0.957, Loss:  0.036
Epoch   5 Batch  104/134 - Train Accuracy:  0.951, Validation Accuracy:  0.955, Loss:  0.035
Epoch   5 Batch  105/134 - Train Accuracy:  0.951, Validation Accuracy:  0.961, Loss:  0.042
Epoch   5 Batch  106/134 - Train Accuracy:  0.952, Validation Accuracy:  0.961, Loss:  0.036
Epoch   5 Batch  107/134 - Train Accuracy:  0.950, Validation Accuracy:  0.960, Loss:  0.041
Epoch   5 Batch  108/134 - Train Accuracy:  0.950, Validation Accuracy:  0.959, Loss:  0.037
Epoch   5 Batch  109/134 - Train Accuracy:  0.949, Validation Accuracy:  0.961, Loss:  0.038
Epoch   5 Batch  110/134 - Train Accuracy:  0.952, Validation Accuracy:  0.960, Loss:  0.034
Epoch   5 Batch  111/134 - Train Accuracy:  0.949, Validation Accuracy:  0.958, Loss:  0.038
Epoch   5 Batch  112/134 - Train Accuracy:  0.948, Validation Accuracy:  0.955, Loss:  0.039
Epoch   5 Batch  113/134 - Train Accuracy:  0.948, Validation Accuracy:  0.951, Loss:  0.037
Epoch   5 Batch  114/134 - Train Accuracy:  0.951, Validation Accuracy:  0.955, Loss:  0.037
Epoch   5 Batch  115/134 - Train Accuracy:  0.945, Validation Accuracy:  0.950, Loss:  0.036
Epoch   5 Batch  116/134 - Train Accuracy:  0.956, Validation Accuracy:  0.952, Loss:  0.037
Epoch   5 Batch  117/134 - Train Accuracy:  0.965, Validation Accuracy:  0.956, Loss:  0.032
Epoch   5 Batch  118/134 - Train Accuracy:  0.955, Validation Accuracy:  0.954, Loss:  0.035
Epoch   5 Batch  119/134 - Train Accuracy:  0.961, Validation Accuracy:  0.958, Loss:  0.032
Epoch   5 Batch  120/134 - Train Accuracy:  0.953, Validation Accuracy:  0.959, Loss:  0.039
Epoch   5 Batch  121/134 - Train Accuracy:  0.955, Validation Accuracy:  0.959, Loss:  0.034
Epoch   5 Batch  122/134 - Train Accuracy:  0.953, Validation Accuracy:  0.958, Loss:  0.037
Epoch   5 Batch  123/134 - Train Accuracy:  0.957, Validation Accuracy:  0.958, Loss:  0.032
Epoch   5 Batch  124/134 - Train Accuracy:  0.954, Validation Accuracy:  0.959, Loss:  0.033
Epoch   5 Batch  125/134 - Train Accuracy:  0.960, Validation Accuracy:  0.958, Loss:  0.033
Epoch   5 Batch  126/134 - Train Accuracy:  0.947, Validation Accuracy:  0.960, Loss:  0.038
Epoch   5 Batch  127/134 - Train Accuracy:  0.955, Validation Accuracy:  0.958, Loss:  0.037
Epoch   5 Batch  128/134 - Train Accuracy:  0.940, Validation Accuracy:  0.951, Loss:  0.038
Epoch   5 Batch  129/134 - Train Accuracy:  0.951, Validation Accuracy:  0.954, Loss:  0.037
Epoch   5 Batch  130/134 - Train Accuracy:  0.958, Validation Accuracy:  0.961, Loss:  0.035
Epoch   5 Batch  131/134 - Train Accuracy:  0.942, Validation Accuracy:  0.953, Loss:  0.038
Epoch   5 Batch  132/134 - Train Accuracy:  0.945, Validation Accuracy:  0.954, Loss:  0.034
Epoch   6 Batch    0/134 - Train Accuracy:  0.954, Validation Accuracy:  0.958, Loss:  0.036
Epoch   6 Batch    1/134 - Train Accuracy:  0.945, Validation Accuracy:  0.952, Loss:  0.034
Epoch   6 Batch    2/134 - Train Accuracy:  0.956, Validation Accuracy:  0.957, Loss:  0.033
Epoch   6 Batch    3/134 - Train Accuracy:  0.956, Validation Accuracy:  0.956, Loss:  0.037
Epoch   6 Batch    4/134 - Train Accuracy:  0.953, Validation Accuracy:  0.960, Loss:  0.036
Epoch   6 Batch    5/134 - Train Accuracy:  0.952, Validation Accuracy:  0.963, Loss:  0.041
Epoch   6 Batch    6/134 - Train Accuracy:  0.954, Validation Accuracy:  0.955, Loss:  0.030
Epoch   6 Batch    7/134 - Train Accuracy:  0.954, Validation Accuracy:  0.963, Loss:  0.035
Epoch   6 Batch    8/134 - Train Accuracy:  0.955, Validation Accuracy:  0.962, Loss:  0.032
Epoch   6 Batch    9/134 - Train Accuracy:  0.957, Validation Accuracy:  0.960, Loss:  0.032
Epoch   6 Batch   10/134 - Train Accuracy:  0.947, Validation Accuracy:  0.957, Loss:  0.035
Epoch   6 Batch   11/134 - Train Accuracy:  0.943, Validation Accuracy:  0.962, Loss:  0.038
Epoch   6 Batch   12/134 - Train Accuracy:  0.951, Validation Accuracy:  0.961, Loss:  0.034
Epoch   6 Batch   13/134 - Train Accuracy:  0.953, Validation Accuracy:  0.959, Loss:  0.032
Epoch   6 Batch   14/134 - Train Accuracy:  0.949, Validation Accuracy:  0.962, Loss:  0.034
Epoch   6 Batch   15/134 - Train Accuracy:  0.952, Validation Accuracy:  0.959, Loss:  0.033
Epoch   6 Batch   16/134 - Train Accuracy:  0.952, Validation Accuracy:  0.961, Loss:  0.032
Epoch   6 Batch   17/134 - Train Accuracy:  0.952, Validation Accuracy:  0.965, Loss:  0.040
Epoch   6 Batch   18/134 - Train Accuracy:  0.955, Validation Accuracy:  0.964, Loss:  0.035
Epoch   6 Batch   19/134 - Train Accuracy:  0.955, Validation Accuracy:  0.962, Loss:  0.036
Epoch   6 Batch   20/134 - Train Accuracy:  0.956, Validation Accuracy:  0.959, Loss:  0.032
Epoch   6 Batch   21/134 - Train Accuracy:  0.957, Validation Accuracy:  0.962, Loss:  0.035
Epoch   6 Batch   22/134 - Train Accuracy:  0.961, Validation Accuracy:  0.965, Loss:  0.033
Epoch   6 Batch   23/134 - Train Accuracy:  0.959, Validation Accuracy:  0.961, Loss:  0.027
Epoch   6 Batch   24/134 - Train Accuracy:  0.953, Validation Accuracy:  0.962, Loss:  0.035
Epoch   6 Batch   25/134 - Train Accuracy:  0.954, Validation Accuracy:  0.961, Loss:  0.030
Epoch   6 Batch   26/134 - Train Accuracy:  0.954, Validation Accuracy:  0.960, Loss:  0.032
Epoch   6 Batch   27/134 - Train Accuracy:  0.960, Validation Accuracy:  0.959, Loss:  0.033
Epoch   6 Batch   28/134 - Train Accuracy:  0.957, Validation Accuracy:  0.961, Loss:  0.035
Epoch   6 Batch   29/134 - Train Accuracy:  0.962, Validation Accuracy:  0.957, Loss:  0.028
Epoch   6 Batch   30/134 - Train Accuracy:  0.948, Validation Accuracy:  0.959, Loss:  0.035
Epoch   6 Batch   31/134 - Train Accuracy:  0.950, Validation Accuracy:  0.959, Loss:  0.033
Epoch   6 Batch   32/134 - Train Accuracy:  0.952, Validation Accuracy:  0.960, Loss:  0.032
Epoch   6 Batch   33/134 - Train Accuracy:  0.949, Validation Accuracy:  0.962, Loss:  0.039
Epoch   6 Batch   34/134 - Train Accuracy:  0.951, Validation Accuracy:  0.959, Loss:  0.037
Epoch   6 Batch   35/134 - Train Accuracy:  0.955, Validation Accuracy:  0.960, Loss:  0.037
Epoch   6 Batch   36/134 - Train Accuracy:  0.953, Validation Accuracy:  0.956, Loss:  0.034
Epoch   6 Batch   37/134 - Train Accuracy:  0.944, Validation Accuracy:  0.951, Loss:  0.035
Epoch   6 Batch   38/134 - Train Accuracy:  0.955, Validation Accuracy:  0.959, Loss:  0.033
Epoch   6 Batch   39/134 - Train Accuracy:  0.948, Validation Accuracy:  0.956, Loss:  0.035
Epoch   6 Batch   40/134 - Train Accuracy:  0.961, Validation Accuracy:  0.961, Loss:  0.033
Epoch   6 Batch   41/134 - Train Accuracy:  0.952, Validation Accuracy:  0.959, Loss:  0.036
Epoch   6 Batch   42/134 - Train Accuracy:  0.958, Validation Accuracy:  0.958, Loss:  0.030
Epoch   6 Batch   43/134 - Train Accuracy:  0.948, Validation Accuracy:  0.959, Loss:  0.034
Epoch   6 Batch   44/134 - Train Accuracy:  0.955, Validation Accuracy:  0.964, Loss:  0.032
Epoch   6 Batch   45/134 - Train Accuracy:  0.965, Validation Accuracy:  0.965, Loss:  0.028
Epoch   6 Batch   46/134 - Train Accuracy:  0.957, Validation Accuracy:  0.964, Loss:  0.034
Epoch   6 Batch   47/134 - Train Accuracy:  0.958, Validation Accuracy:  0.958, Loss:  0.034
Epoch   6 Batch   48/134 - Train Accuracy:  0.958, Validation Accuracy:  0.958, Loss:  0.033
Epoch   6 Batch   49/134 - Train Accuracy:  0.960, Validation Accuracy:  0.958, Loss:  0.031
Epoch   6 Batch   50/134 - Train Accuracy:  0.949, Validation Accuracy:  0.960, Loss:  0.034
Epoch   6 Batch   51/134 - Train Accuracy:  0.954, Validation Accuracy:  0.961, Loss:  0.033
Epoch   6 Batch   52/134 - Train Accuracy:  0.962, Validation Accuracy:  0.961, Loss:  0.029
Epoch   6 Batch   53/134 - Train Accuracy:  0.964, Validation Accuracy:  0.962, Loss:  0.032
Epoch   6 Batch   54/134 - Train Accuracy:  0.950, Validation Accuracy:  0.961, Loss:  0.033
Epoch   6 Batch   55/134 - Train Accuracy:  0.956, Validation Accuracy:  0.960, Loss:  0.033
Epoch   6 Batch   56/134 - Train Accuracy:  0.955, Validation Accuracy:  0.963, Loss:  0.031
Epoch   6 Batch   57/134 - Train Accuracy:  0.959, Validation Accuracy:  0.963, Loss:  0.032
Epoch   6 Batch   58/134 - Train Accuracy:  0.960, Validation Accuracy:  0.963, Loss:  0.028
Epoch   6 Batch   59/134 - Train Accuracy:  0.952, Validation Accuracy:  0.963, Loss:  0.034
Epoch   6 Batch   60/134 - Train Accuracy:  0.957, Validation Accuracy:  0.961, Loss:  0.031
Epoch   6 Batch   61/134 - Train Accuracy:  0.962, Validation Accuracy:  0.963, Loss:  0.029
Epoch   6 Batch   62/134 - Train Accuracy:  0.956, Validation Accuracy:  0.957, Loss:  0.029
Epoch   6 Batch   63/134 - Train Accuracy:  0.954, Validation Accuracy:  0.958, Loss:  0.031
Epoch   6 Batch   64/134 - Train Accuracy:  0.950, Validation Accuracy:  0.959, Loss:  0.032
Epoch   6 Batch   65/134 - Train Accuracy:  0.944, Validation Accuracy:  0.964, Loss:  0.034
Epoch   6 Batch   66/134 - Train Accuracy:  0.955, Validation Accuracy:  0.964, Loss:  0.028
Epoch   6 Batch   67/134 - Train Accuracy:  0.955, Validation Accuracy:  0.961, Loss:  0.031
Epoch   6 Batch   68/134 - Train Accuracy:  0.950, Validation Accuracy:  0.959, Loss:  0.032
Epoch   6 Batch   69/134 - Train Accuracy:  0.957, Validation Accuracy:  0.960, Loss:  0.031
Epoch   6 Batch   70/134 - Train Accuracy:  0.955, Validation Accuracy:  0.958, Loss:  0.031
Epoch   6 Batch   71/134 - Train Accuracy:  0.966, Validation Accuracy:  0.959, Loss:  0.026
Epoch   6 Batch   72/134 - Train Accuracy:  0.963, Validation Accuracy:  0.961, Loss:  0.029
Epoch   6 Batch   73/134 - Train Accuracy:  0.960, Validation Accuracy:  0.963, Loss:  0.033
Epoch   6 Batch   74/134 - Train Accuracy:  0.961, Validation Accuracy:  0.964, Loss:  0.034
Epoch   6 Batch   75/134 - Train Accuracy:  0.963, Validation Accuracy:  0.960, Loss:  0.030
Epoch   6 Batch   76/134 - Train Accuracy:  0.966, Validation Accuracy:  0.963, Loss:  0.029
Epoch   6 Batch   77/134 - Train Accuracy:  0.962, Validation Accuracy:  0.961, Loss:  0.028
Epoch   6 Batch   78/134 - Train Accuracy:  0.955, Validation Accuracy:  0.962, Loss:  0.029
Epoch   6 Batch   79/134 - Train Accuracy:  0.961, Validation Accuracy:  0.961, Loss:  0.030
Epoch   6 Batch   80/134 - Train Accuracy:  0.962, Validation Accuracy:  0.965, Loss:  0.028
Epoch   6 Batch   81/134 - Train Accuracy:  0.967, Validation Accuracy:  0.965, Loss:  0.029
Epoch   6 Batch   82/134 - Train Accuracy:  0.960, Validation Accuracy:  0.964, Loss:  0.027
Epoch   6 Batch   83/134 - Train Accuracy:  0.962, Validation Accuracy:  0.966, Loss:  0.029
Epoch   6 Batch   84/134 - Train Accuracy:  0.963, Validation Accuracy:  0.967, Loss:  0.028
Epoch   6 Batch   85/134 - Train Accuracy:  0.962, Validation Accuracy:  0.965, Loss:  0.030
Epoch   6 Batch   86/134 - Train Accuracy:  0.961, Validation Accuracy:  0.967, Loss:  0.027
Epoch   6 Batch   87/134 - Train Accuracy:  0.949, Validation Accuracy:  0.966, Loss:  0.034
Epoch   6 Batch   88/134 - Train Accuracy:  0.967, Validation Accuracy:  0.965, Loss:  0.025
Epoch   6 Batch   89/134 - Train Accuracy:  0.964, Validation Accuracy:  0.968, Loss:  0.028
Epoch   6 Batch   90/134 - Train Accuracy:  0.955, Validation Accuracy:  0.965, Loss:  0.032
Epoch   6 Batch   91/134 - Train Accuracy:  0.959, Validation Accuracy:  0.963, Loss:  0.024
Epoch   6 Batch   92/134 - Train Accuracy:  0.966, Validation Accuracy:  0.964, Loss:  0.026
Epoch   6 Batch   93/134 - Train Accuracy:  0.962, Validation Accuracy:  0.963, Loss:  0.027
Epoch   6 Batch   94/134 - Train Accuracy:  0.958, Validation Accuracy:  0.961, Loss:  0.029
Epoch   6 Batch   95/134 - Train Accuracy:  0.967, Validation Accuracy:  0.963, Loss:  0.027
Epoch   6 Batch   96/134 - Train Accuracy:  0.953, Validation Accuracy:  0.964, Loss:  0.031
Epoch   6 Batch   97/134 - Train Accuracy:  0.959, Validation Accuracy:  0.964, Loss:  0.027
Epoch   6 Batch   98/134 - Train Accuracy:  0.962, Validation Accuracy:  0.966, Loss:  0.027
Epoch   6 Batch   99/134 - Train Accuracy:  0.968, Validation Accuracy:  0.963, Loss:  0.027
Epoch   6 Batch  100/134 - Train Accuracy:  0.966, Validation Accuracy:  0.961, Loss:  0.029
Epoch   6 Batch  101/134 - Train Accuracy:  0.965, Validation Accuracy:  0.965, Loss:  0.028
Epoch   6 Batch  102/134 - Train Accuracy:  0.961, Validation Accuracy:  0.967, Loss:  0.027
Epoch   6 Batch  103/134 - Train Accuracy:  0.964, Validation Accuracy:  0.966, Loss:  0.027
Epoch   6 Batch  104/134 - Train Accuracy:  0.968, Validation Accuracy:  0.968, Loss:  0.026
Epoch   6 Batch  105/134 - Train Accuracy:  0.960, Validation Accuracy:  0.966, Loss:  0.032
Epoch   6 Batch  106/134 - Train Accuracy:  0.962, Validation Accuracy:  0.964, Loss:  0.026
Epoch   6 Batch  107/134 - Train Accuracy:  0.961, Validation Accuracy:  0.964, Loss:  0.032
Epoch   6 Batch  108/134 - Train Accuracy:  0.957, Validation Accuracy:  0.964, Loss:  0.028
Epoch   6 Batch  109/134 - Train Accuracy:  0.957, Validation Accuracy:  0.966, Loss:  0.029
Epoch   6 Batch  110/134 - Train Accuracy:  0.961, Validation Accuracy:  0.967, Loss:  0.026
Epoch   6 Batch  111/134 - Train Accuracy:  0.959, Validation Accuracy:  0.967, Loss:  0.028
Epoch   6 Batch  112/134 - Train Accuracy:  0.958, Validation Accuracy:  0.968, Loss:  0.029
Epoch   6 Batch  113/134 - Train Accuracy:  0.957, Validation Accuracy:  0.965, Loss:  0.029
Epoch   6 Batch  114/134 - Train Accuracy:  0.957, Validation Accuracy:  0.967, Loss:  0.027
Epoch   6 Batch  115/134 - Train Accuracy:  0.959, Validation Accuracy:  0.966, Loss:  0.027
Epoch   6 Batch  116/134 - Train Accuracy:  0.966, Validation Accuracy:  0.965, Loss:  0.029
Epoch   6 Batch  117/134 - Train Accuracy:  0.974, Validation Accuracy:  0.963, Loss:  0.025
Epoch   6 Batch  118/134 - Train Accuracy:  0.966, Validation Accuracy:  0.962, Loss:  0.027
Epoch   6 Batch  119/134 - Train Accuracy:  0.965, Validation Accuracy:  0.962, Loss:  0.024
Epoch   6 Batch  120/134 - Train Accuracy:  0.952, Validation Accuracy:  0.962, Loss:  0.030
Epoch   6 Batch  121/134 - Train Accuracy:  0.965, Validation Accuracy:  0.961, Loss:  0.026
Epoch   6 Batch  122/134 - Train Accuracy:  0.957, Validation Accuracy:  0.964, Loss:  0.028
Epoch   6 Batch  123/134 - Train Accuracy:  0.969, Validation Accuracy:  0.966, Loss:  0.025
Epoch   6 Batch  124/134 - Train Accuracy:  0.964, Validation Accuracy:  0.965, Loss:  0.026
Epoch   6 Batch  125/134 - Train Accuracy:  0.968, Validation Accuracy:  0.964, Loss:  0.025
Epoch   6 Batch  126/134 - Train Accuracy:  0.962, Validation Accuracy:  0.965, Loss:  0.030
Epoch   6 Batch  127/134 - Train Accuracy:  0.962, Validation Accuracy:  0.967, Loss:  0.028
Epoch   6 Batch  128/134 - Train Accuracy:  0.951, Validation Accuracy:  0.963, Loss:  0.032
Epoch   6 Batch  129/134 - Train Accuracy:  0.965, Validation Accuracy:  0.962, Loss:  0.029
Epoch   6 Batch  130/134 - Train Accuracy:  0.964, Validation Accuracy:  0.963, Loss:  0.026
Epoch   6 Batch  131/134 - Train Accuracy:  0.960, Validation Accuracy:  0.960, Loss:  0.029
Epoch   6 Batch  132/134 - Train Accuracy:  0.955, Validation Accuracy:  0.960, Loss:  0.025
Epoch   7 Batch    0/134 - Train Accuracy:  0.966, Validation Accuracy:  0.962, Loss:  0.027
Epoch   7 Batch    1/134 - Train Accuracy:  0.956, Validation Accuracy:  0.957, Loss:  0.026
Epoch   7 Batch    2/134 - Train Accuracy:  0.963, Validation Accuracy:  0.963, Loss:  0.026
Epoch   7 Batch    3/134 - Train Accuracy:  0.960, Validation Accuracy:  0.960, Loss:  0.030
Epoch   7 Batch    4/134 - Train Accuracy:  0.965, Validation Accuracy:  0.963, Loss:  0.026
Epoch   7 Batch    5/134 - Train Accuracy:  0.965, Validation Accuracy:  0.964, Loss:  0.031
Epoch   7 Batch    6/134 - Train Accuracy:  0.968, Validation Accuracy:  0.961, Loss:  0.023
Epoch   7 Batch    7/134 - Train Accuracy:  0.964, Validation Accuracy:  0.964, Loss:  0.027
Epoch   7 Batch    8/134 - Train Accuracy:  0.963, Validation Accuracy:  0.967, Loss:  0.025
Epoch   7 Batch    9/134 - Train Accuracy:  0.965, Validation Accuracy:  0.966, Loss:  0.023
Epoch   7 Batch   10/134 - Train Accuracy:  0.958, Validation Accuracy:  0.969, Loss:  0.028
Epoch   7 Batch   11/134 - Train Accuracy:  0.958, Validation Accuracy:  0.967, Loss:  0.029
Epoch   7 Batch   12/134 - Train Accuracy:  0.958, Validation Accuracy:  0.957, Loss:  0.025
Epoch   7 Batch   13/134 - Train Accuracy:  0.963, Validation Accuracy:  0.967, Loss:  0.025
Epoch   7 Batch   14/134 - Train Accuracy:  0.959, Validation Accuracy:  0.961, Loss:  0.027
Epoch   7 Batch   15/134 - Train Accuracy:  0.964, Validation Accuracy:  0.962, Loss:  0.026
Epoch   7 Batch   16/134 - Train Accuracy:  0.957, Validation Accuracy:  0.961, Loss:  0.025
Epoch   7 Batch   17/134 - Train Accuracy:  0.960, Validation Accuracy:  0.962, Loss:  0.031
Epoch   7 Batch   18/134 - Train Accuracy:  0.963, Validation Accuracy:  0.964, Loss:  0.028
Epoch   7 Batch   19/134 - Train Accuracy:  0.961, Validation Accuracy:  0.966, Loss:  0.028
Epoch   7 Batch   20/134 - Train Accuracy:  0.962, Validation Accuracy:  0.969, Loss:  0.025
Epoch   7 Batch   21/134 - Train Accuracy:  0.967, Validation Accuracy:  0.969, Loss:  0.028
Epoch   7 Batch   22/134 - Train Accuracy:  0.965, Validation Accuracy:  0.966, Loss:  0.024
Epoch   7 Batch   23/134 - Train Accuracy:  0.971, Validation Accuracy:  0.967, Loss:  0.022
Epoch   7 Batch   24/134 - Train Accuracy:  0.954, Validation Accuracy:  0.966, Loss:  0.028
Epoch   7 Batch   25/134 - Train Accuracy:  0.965, Validation Accuracy:  0.966, Loss:  0.022
Epoch   7 Batch   26/134 - Train Accuracy:  0.968, Validation Accuracy:  0.970, Loss:  0.025
Epoch   7 Batch   27/134 - Train Accuracy:  0.964, Validation Accuracy:  0.967, Loss:  0.025
Epoch   7 Batch   28/134 - Train Accuracy:  0.966, Validation Accuracy:  0.962, Loss:  0.028
Epoch   7 Batch   29/134 - Train Accuracy:  0.966, Validation Accuracy:  0.960, Loss:  0.021
Epoch   7 Batch   30/134 - Train Accuracy:  0.961, Validation Accuracy:  0.962, Loss:  0.027
Epoch   7 Batch   31/134 - Train Accuracy:  0.964, Validation Accuracy:  0.967, Loss:  0.026
Epoch   7 Batch   32/134 - Train Accuracy:  0.962, Validation Accuracy:  0.967, Loss:  0.024
Epoch   7 Batch   33/134 - Train Accuracy:  0.956, Validation Accuracy:  0.964, Loss:  0.030
Epoch   7 Batch   34/134 - Train Accuracy:  0.956, Validation Accuracy:  0.963, Loss:  0.029
Epoch   7 Batch   35/134 - Train Accuracy:  0.966, Validation Accuracy:  0.966, Loss:  0.029
Epoch   7 Batch   36/134 - Train Accuracy:  0.966, Validation Accuracy:  0.967, Loss:  0.026
Epoch   7 Batch   37/134 - Train Accuracy:  0.959, Validation Accuracy:  0.968, Loss:  0.026
Epoch   7 Batch   38/134 - Train Accuracy:  0.969, Validation Accuracy:  0.964, Loss:  0.024
Epoch   7 Batch   39/134 - Train Accuracy:  0.964, Validation Accuracy:  0.967, Loss:  0.028
Epoch   7 Batch   40/134 - Train Accuracy:  0.964, Validation Accuracy:  0.968, Loss:  0.026
Epoch   7 Batch   41/134 - Train Accuracy:  0.964, Validation Accuracy:  0.970, Loss:  0.028
Epoch   7 Batch   42/134 - Train Accuracy:  0.966, Validation Accuracy:  0.964, Loss:  0.023
Epoch   7 Batch   43/134 - Train Accuracy:  0.960, Validation Accuracy:  0.961, Loss:  0.025
Epoch   7 Batch   44/134 - Train Accuracy:  0.964, Validation Accuracy:  0.965, Loss:  0.025
Epoch   7 Batch   45/134 - Train Accuracy:  0.970, Validation Accuracy:  0.968, Loss:  0.022
Epoch   7 Batch   46/134 - Train Accuracy:  0.967, Validation Accuracy:  0.965, Loss:  0.025
Epoch   7 Batch   47/134 - Train Accuracy:  0.961, Validation Accuracy:  0.963, Loss:  0.025
Epoch   7 Batch   48/134 - Train Accuracy:  0.966, Validation Accuracy:  0.965, Loss:  0.027
Epoch   7 Batch   49/134 - Train Accuracy:  0.962, Validation Accuracy:  0.967, Loss:  0.024
Epoch   7 Batch   50/134 - Train Accuracy:  0.959, Validation Accuracy:  0.966, Loss:  0.026
Epoch   7 Batch   51/134 - Train Accuracy:  0.961, Validation Accuracy:  0.965, Loss:  0.026
Epoch   7 Batch   52/134 - Train Accuracy:  0.972, Validation Accuracy:  0.965, Loss:  0.023
Epoch   7 Batch   53/134 - Train Accuracy:  0.968, Validation Accuracy:  0.966, Loss:  0.026
Epoch   7 Batch   54/134 - Train Accuracy:  0.960, Validation Accuracy:  0.966, Loss:  0.025
Epoch   7 Batch   55/134 - Train Accuracy:  0.965, Validation Accuracy:  0.965, Loss:  0.026
Epoch   7 Batch   56/134 - Train Accuracy:  0.959, Validation Accuracy:  0.966, Loss:  0.024
Epoch   7 Batch   57/134 - Train Accuracy:  0.962, Validation Accuracy:  0.962, Loss:  0.024
Epoch   7 Batch   58/134 - Train Accuracy:  0.966, Validation Accuracy:  0.964, Loss:  0.021
Epoch   7 Batch   59/134 - Train Accuracy:  0.964, Validation Accuracy:  0.967, Loss:  0.026
Epoch   7 Batch   60/134 - Train Accuracy:  0.960, Validation Accuracy:  0.962, Loss:  0.024
Epoch   7 Batch   61/134 - Train Accuracy:  0.959, Validation Accuracy:  0.961, Loss:  0.025
Epoch   7 Batch   62/134 - Train Accuracy:  0.955, Validation Accuracy:  0.961, Loss:  0.026
Epoch   7 Batch   63/134 - Train Accuracy:  0.970, Validation Accuracy:  0.964, Loss:  0.024
Epoch   7 Batch   64/134 - Train Accuracy:  0.957, Validation Accuracy:  0.960, Loss:  0.025
Epoch   7 Batch   65/134 - Train Accuracy:  0.958, Validation Accuracy:  0.964, Loss:  0.027
Epoch   7 Batch   66/134 - Train Accuracy:  0.963, Validation Accuracy:  0.967, Loss:  0.022
Epoch   7 Batch   67/134 - Train Accuracy:  0.961, Validation Accuracy:  0.965, Loss:  0.024
Epoch   7 Batch   68/134 - Train Accuracy:  0.961, Validation Accuracy:  0.966, Loss:  0.026
Epoch   7 Batch   69/134 - Train Accuracy:  0.966, Validation Accuracy:  0.965, Loss:  0.025
Epoch   7 Batch   70/134 - Train Accuracy:  0.964, Validation Accuracy:  0.961, Loss:  0.024
Epoch   7 Batch   71/134 - Train Accuracy:  0.965, Validation Accuracy:  0.965, Loss:  0.020
Epoch   7 Batch   72/134 - Train Accuracy:  0.969, Validation Accuracy:  0.964, Loss:  0.023
Epoch   7 Batch   73/134 - Train Accuracy:  0.968, Validation Accuracy:  0.965, Loss:  0.025
Epoch   7 Batch   74/134 - Train Accuracy:  0.965, Validation Accuracy:  0.965, Loss:  0.028
Epoch   7 Batch   75/134 - Train Accuracy:  0.966, Validation Accuracy:  0.968, Loss:  0.025
Epoch   7 Batch   76/134 - Train Accuracy:  0.968, Validation Accuracy:  0.968, Loss:  0.025
Epoch   7 Batch   77/134 - Train Accuracy:  0.968, Validation Accuracy:  0.969, Loss:  0.023
Epoch   7 Batch   78/134 - Train Accuracy:  0.966, Validation Accuracy:  0.969, Loss:  0.024
Epoch   7 Batch   79/134 - Train Accuracy:  0.966, Validation Accuracy:  0.963, Loss:  0.023
Epoch   7 Batch   80/134 - Train Accuracy:  0.967, Validation Accuracy:  0.960, Loss:  0.022
Epoch   7 Batch   81/134 - Train Accuracy:  0.970, Validation Accuracy:  0.963, Loss:  0.024
Epoch   7 Batch   82/134 - Train Accuracy:  0.967, Validation Accuracy:  0.967, Loss:  0.022
Epoch   7 Batch   83/134 - Train Accuracy:  0.970, Validation Accuracy:  0.966, Loss:  0.028
Epoch   7 Batch   84/134 - Train Accuracy:  0.965, Validation Accuracy:  0.964, Loss:  0.022
Epoch   7 Batch   85/134 - Train Accuracy:  0.971, Validation Accuracy:  0.962, Loss:  0.023
Epoch   7 Batch   86/134 - Train Accuracy:  0.964, Validation Accuracy:  0.964, Loss:  0.024
Epoch   7 Batch   87/134 - Train Accuracy:  0.955, Validation Accuracy:  0.968, Loss:  0.028
Epoch   7 Batch   88/134 - Train Accuracy:  0.973, Validation Accuracy:  0.968, Loss:  0.022
Epoch   7 Batch   89/134 - Train Accuracy:  0.972, Validation Accuracy:  0.968, Loss:  0.023
Epoch   7 Batch   90/134 - Train Accuracy:  0.961, Validation Accuracy:  0.968, Loss:  0.027
Epoch   7 Batch   91/134 - Train Accuracy:  0.968, Validation Accuracy:  0.965, Loss:  0.020
Epoch   7 Batch   92/134 - Train Accuracy:  0.968, Validation Accuracy:  0.963, Loss:  0.022
Epoch   7 Batch   93/134 - Train Accuracy:  0.966, Validation Accuracy:  0.962, Loss:  0.023
Epoch   7 Batch   94/134 - Train Accuracy:  0.967, Validation Accuracy:  0.966, Loss:  0.023
Epoch   7 Batch   95/134 - Train Accuracy:  0.970, Validation Accuracy:  0.965, Loss:  0.023
Epoch   7 Batch   96/134 - Train Accuracy:  0.962, Validation Accuracy:  0.967, Loss:  0.025
Epoch   7 Batch   97/134 - Train Accuracy:  0.964, Validation Accuracy:  0.968, Loss:  0.024
Epoch   7 Batch   98/134 - Train Accuracy:  0.967, Validation Accuracy:  0.969, Loss:  0.022
Epoch   7 Batch   99/134 - Train Accuracy:  0.970, Validation Accuracy:  0.963, Loss:  0.022
Epoch   7 Batch  100/134 - Train Accuracy:  0.965, Validation Accuracy:  0.965, Loss:  0.024
Epoch   7 Batch  101/134 - Train Accuracy:  0.968, Validation Accuracy:  0.968, Loss:  0.024
Epoch   7 Batch  102/134 - Train Accuracy:  0.966, Validation Accuracy:  0.970, Loss:  0.022
Epoch   7 Batch  103/134 - Train Accuracy:  0.966, Validation Accuracy:  0.972, Loss:  0.023
Epoch   7 Batch  104/134 - Train Accuracy:  0.961, Validation Accuracy:  0.967, Loss:  0.022
Epoch   7 Batch  105/134 - Train Accuracy:  0.962, Validation Accuracy:  0.970, Loss:  0.028
Epoch   7 Batch  106/134 - Train Accuracy:  0.965, Validation Accuracy:  0.965, Loss:  0.021
Epoch   7 Batch  107/134 - Train Accuracy:  0.968, Validation Accuracy:  0.964, Loss:  0.027
Epoch   7 Batch  108/134 - Train Accuracy:  0.964, Validation Accuracy:  0.962, Loss:  0.023
Epoch   7 Batch  109/134 - Train Accuracy:  0.962, Validation Accuracy:  0.959, Loss:  0.023
Epoch   7 Batch  110/134 - Train Accuracy:  0.966, Validation Accuracy:  0.966, Loss:  0.022
Epoch   7 Batch  111/134 - Train Accuracy:  0.963, Validation Accuracy:  0.969, Loss:  0.024
Epoch   7 Batch  112/134 - Train Accuracy:  0.964, Validation Accuracy:  0.969, Loss:  0.025
Epoch   7 Batch  113/134 - Train Accuracy:  0.969, Validation Accuracy:  0.971, Loss:  0.025
Epoch   7 Batch  114/134 - Train Accuracy:  0.969, Validation Accuracy:  0.969, Loss:  0.021
Epoch   7 Batch  115/134 - Train Accuracy:  0.965, Validation Accuracy:  0.968, Loss:  0.023
Epoch   7 Batch  116/134 - Train Accuracy:  0.970, Validation Accuracy:  0.966, Loss:  0.025
Epoch   7 Batch  117/134 - Train Accuracy:  0.970, Validation Accuracy:  0.967, Loss:  0.021
Epoch   7 Batch  118/134 - Train Accuracy:  0.977, Validation Accuracy:  0.969, Loss:  0.022
Epoch   7 Batch  119/134 - Train Accuracy:  0.975, Validation Accuracy:  0.966, Loss:  0.020
Epoch   7 Batch  120/134 - Train Accuracy:  0.965, Validation Accuracy:  0.966, Loss:  0.024
Epoch   7 Batch  121/134 - Train Accuracy:  0.967, Validation Accuracy:  0.966, Loss:  0.021
Epoch   7 Batch  122/134 - Train Accuracy:  0.963, Validation Accuracy:  0.966, Loss:  0.023
Epoch   7 Batch  123/134 - Train Accuracy:  0.970, Validation Accuracy:  0.969, Loss:  0.020
Epoch   7 Batch  124/134 - Train Accuracy:  0.970, Validation Accuracy:  0.968, Loss:  0.021
Epoch   7 Batch  125/134 - Train Accuracy:  0.971, Validation Accuracy:  0.969, Loss:  0.019
Epoch   7 Batch  126/134 - Train Accuracy:  0.966, Validation Accuracy:  0.969, Loss:  0.024
Epoch   7 Batch  127/134 - Train Accuracy:  0.968, Validation Accuracy:  0.969, Loss:  0.023
Epoch   7 Batch  128/134 - Train Accuracy:  0.958, Validation Accuracy:  0.968, Loss:  0.025
Epoch   7 Batch  129/134 - Train Accuracy:  0.970, Validation Accuracy:  0.966, Loss:  0.023
Epoch   7 Batch  130/134 - Train Accuracy:  0.969, Validation Accuracy:  0.967, Loss:  0.021
Epoch   7 Batch  131/134 - Train Accuracy:  0.960, Validation Accuracy:  0.966, Loss:  0.024
Epoch   7 Batch  132/134 - Train Accuracy:  0.963, Validation Accuracy:  0.966, Loss:  0.021
Epoch   8 Batch    0/134 - Train Accuracy:  0.970, Validation Accuracy:  0.967, Loss:  0.021
Epoch   8 Batch    1/134 - Train Accuracy:  0.961, Validation Accuracy:  0.968, Loss:  0.020
Epoch   8 Batch    2/134 - Train Accuracy:  0.971, Validation Accuracy:  0.971, Loss:  0.021
Epoch   8 Batch    3/134 - Train Accuracy:  0.968, Validation Accuracy:  0.969, Loss:  0.024
Epoch   8 Batch    4/134 - Train Accuracy:  0.971, Validation Accuracy:  0.970, Loss:  0.020
Epoch   8 Batch    5/134 - Train Accuracy:  0.968, Validation Accuracy:  0.969, Loss:  0.025
Epoch   8 Batch    6/134 - Train Accuracy:  0.969, Validation Accuracy:  0.971, Loss:  0.019
Epoch   8 Batch    7/134 - Train Accuracy:  0.969, Validation Accuracy:  0.972, Loss:  0.022
Epoch   8 Batch    8/134 - Train Accuracy:  0.966, Validation Accuracy:  0.969, Loss:  0.020
Epoch   8 Batch    9/134 - Train Accuracy:  0.970, Validation Accuracy:  0.971, Loss:  0.018
Epoch   8 Batch   10/134 - Train Accuracy:  0.968, Validation Accuracy:  0.971, Loss:  0.020
Epoch   8 Batch   11/134 - Train Accuracy:  0.963, Validation Accuracy:  0.972, Loss:  0.021
Epoch   8 Batch   12/134 - Train Accuracy:  0.969, Validation Accuracy:  0.969, Loss:  0.020
Epoch   8 Batch   13/134 - Train Accuracy:  0.968, Validation Accuracy:  0.967, Loss:  0.019
Epoch   8 Batch   14/134 - Train Accuracy:  0.967, Validation Accuracy:  0.971, Loss:  0.022
Epoch   8 Batch   15/134 - Train Accuracy:  0.968, Validation Accuracy:  0.971, Loss:  0.020
Epoch   8 Batch   16/134 - Train Accuracy:  0.968, Validation Accuracy:  0.970, Loss:  0.018
Epoch   8 Batch   17/134 - Train Accuracy:  0.965, Validation Accuracy:  0.972, Loss:  0.025
Epoch   8 Batch   18/134 - Train Accuracy:  0.964, Validation Accuracy:  0.972, Loss:  0.021
Epoch   8 Batch   19/134 - Train Accuracy:  0.969, Validation Accuracy:  0.973, Loss:  0.022
Epoch   8 Batch   20/134 - Train Accuracy:  0.972, Validation Accuracy:  0.970, Loss:  0.020
Epoch   8 Batch   21/134 - Train Accuracy:  0.967, Validation Accuracy:  0.968, Loss:  0.023
Epoch   8 Batch   22/134 - Train Accuracy:  0.968, Validation Accuracy:  0.970, Loss:  0.019
Epoch   8 Batch   23/134 - Train Accuracy:  0.973, Validation Accuracy:  0.971, Loss:  0.017
Epoch   8 Batch   24/134 - Train Accuracy:  0.969, Validation Accuracy:  0.971, Loss:  0.021
Epoch   8 Batch   25/134 - Train Accuracy:  0.966, Validation Accuracy:  0.968, Loss:  0.018
Epoch   8 Batch   26/134 - Train Accuracy:  0.967, Validation Accuracy:  0.968, Loss:  0.020
Epoch   8 Batch   27/134 - Train Accuracy:  0.967, Validation Accuracy:  0.971, Loss:  0.020
Epoch   8 Batch   28/134 - Train Accuracy:  0.974, Validation Accuracy:  0.969, Loss:  0.022
Epoch   8 Batch   29/134 - Train Accuracy:  0.969, Validation Accuracy:  0.968, Loss:  0.017
Epoch   8 Batch   30/134 - Train Accuracy:  0.966, Validation Accuracy:  0.968, Loss:  0.024
Epoch   8 Batch   31/134 - Train Accuracy:  0.971, Validation Accuracy:  0.969, Loss:  0.020
Epoch   8 Batch   32/134 - Train Accuracy:  0.970, Validation Accuracy:  0.967, Loss:  0.020
Epoch   8 Batch   33/134 - Train Accuracy:  0.965, Validation Accuracy:  0.967, Loss:  0.024
Epoch   8 Batch   34/134 - Train Accuracy:  0.966, Validation Accuracy:  0.969, Loss:  0.025
Epoch   8 Batch   35/134 - Train Accuracy:  0.971, Validation Accuracy:  0.970, Loss:  0.023
Epoch   8 Batch   36/134 - Train Accuracy:  0.969, Validation Accuracy:  0.971, Loss:  0.020
Epoch   8 Batch   37/134 - Train Accuracy:  0.967, Validation Accuracy:  0.968, Loss:  0.022
Epoch   8 Batch   38/134 - Train Accuracy:  0.974, Validation Accuracy:  0.967, Loss:  0.020
Epoch   8 Batch   39/134 - Train Accuracy:  0.966, Validation Accuracy:  0.973, Loss:  0.022
Epoch   8 Batch   40/134 - Train Accuracy:  0.966, Validation Accuracy:  0.973, Loss:  0.021
Epoch   8 Batch   41/134 - Train Accuracy:  0.968, Validation Accuracy:  0.974, Loss:  0.024
Epoch   8 Batch   42/134 - Train Accuracy:  0.969, Validation Accuracy:  0.970, Loss:  0.019
Epoch   8 Batch   43/134 - Train Accuracy:  0.964, Validation Accuracy:  0.963, Loss:  0.021
Epoch   8 Batch   44/134 - Train Accuracy:  0.968, Validation Accuracy:  0.968, Loss:  0.022
Epoch   8 Batch   45/134 - Train Accuracy:  0.975, Validation Accuracy:  0.969, Loss:  0.018
Epoch   8 Batch   46/134 - Train Accuracy:  0.971, Validation Accuracy:  0.969, Loss:  0.019
Epoch   8 Batch   47/134 - Train Accuracy:  0.970, Validation Accuracy:  0.970, Loss:  0.020
Epoch   8 Batch   48/134 - Train Accuracy:  0.967, Validation Accuracy:  0.970, Loss:  0.022
Epoch   8 Batch   49/134 - Train Accuracy:  0.969, Validation Accuracy:  0.970, Loss:  0.020
Epoch   8 Batch   50/134 - Train Accuracy:  0.968, Validation Accuracy:  0.968, Loss:  0.022
Epoch   8 Batch   51/134 - Train Accuracy:  0.967, Validation Accuracy:  0.967, Loss:  0.022
Epoch   8 Batch   52/134 - Train Accuracy:  0.974, Validation Accuracy:  0.970, Loss:  0.018
Epoch   8 Batch   53/134 - Train Accuracy:  0.969, Validation Accuracy:  0.968, Loss:  0.020
Epoch   8 Batch   54/134 - Train Accuracy:  0.965, Validation Accuracy:  0.967, Loss:  0.021
Epoch   8 Batch   55/134 - Train Accuracy:  0.968, Validation Accuracy:  0.968, Loss:  0.020
Epoch   8 Batch   56/134 - Train Accuracy:  0.965, Validation Accuracy:  0.971, Loss:  0.020
Epoch   8 Batch   57/134 - Train Accuracy:  0.975, Validation Accuracy:  0.970, Loss:  0.019
Epoch   8 Batch   58/134 - Train Accuracy:  0.970, Validation Accuracy:  0.971, Loss:  0.018
Epoch   8 Batch   59/134 - Train Accuracy:  0.971, Validation Accuracy:  0.973, Loss:  0.021
Epoch   8 Batch   60/134 - Train Accuracy:  0.966, Validation Accuracy:  0.973, Loss:  0.020
Epoch   8 Batch   61/134 - Train Accuracy:  0.972, Validation Accuracy:  0.972, Loss:  0.020
Epoch   8 Batch   62/134 - Train Accuracy:  0.967, Validation Accuracy:  0.966, Loss:  0.018
Epoch   8 Batch   63/134 - Train Accuracy:  0.969, Validation Accuracy:  0.967, Loss:  0.019
Epoch   8 Batch   64/134 - Train Accuracy:  0.966, Validation Accuracy:  0.971, Loss:  0.021
Epoch   8 Batch   65/134 - Train Accuracy:  0.959, Validation Accuracy:  0.971, Loss:  0.021
Epoch   8 Batch   66/134 - Train Accuracy:  0.968, Validation Accuracy:  0.972, Loss:  0.019
Epoch   8 Batch   67/134 - Train Accuracy:  0.970, Validation Accuracy:  0.971, Loss:  0.019
Epoch   8 Batch   68/134 - Train Accuracy:  0.969, Validation Accuracy:  0.969, Loss:  0.021
Epoch   8 Batch   69/134 - Train Accuracy:  0.970, Validation Accuracy:  0.971, Loss:  0.020
Epoch   8 Batch   70/134 - Train Accuracy:  0.970, Validation Accuracy:  0.972, Loss:  0.020
Epoch   8 Batch   71/134 - Train Accuracy:  0.974, Validation Accuracy:  0.972, Loss:  0.015
Epoch   8 Batch   72/134 - Train Accuracy:  0.969, Validation Accuracy:  0.968, Loss:  0.017
Epoch   8 Batch   73/134 - Train Accuracy:  0.970, Validation Accuracy:  0.969, Loss:  0.022
Epoch   8 Batch   74/134 - Train Accuracy:  0.967, Validation Accuracy:  0.973, Loss:  0.023
Epoch   8 Batch   75/134 - Train Accuracy:  0.967, Validation Accuracy:  0.974, Loss:  0.019
Epoch   8 Batch   76/134 - Train Accuracy:  0.975, Validation Accuracy:  0.971, Loss:  0.020
Epoch   8 Batch   77/134 - Train Accuracy:  0.974, Validation Accuracy:  0.973, Loss:  0.018
Epoch   8 Batch   78/134 - Train Accuracy:  0.966, Validation Accuracy:  0.975, Loss:  0.020
Epoch   8 Batch   79/134 - Train Accuracy:  0.970, Validation Accuracy:  0.974, Loss:  0.020
Epoch   8 Batch   80/134 - Train Accuracy:  0.975, Validation Accuracy:  0.971, Loss:  0.019
Epoch   8 Batch   81/134 - Train Accuracy:  0.972, Validation Accuracy:  0.968, Loss:  0.017
Epoch   8 Batch   82/134 - Train Accuracy:  0.974, Validation Accuracy:  0.969, Loss:  0.017
Epoch   8 Batch   83/134 - Train Accuracy:  0.972, Validation Accuracy:  0.970, Loss:  0.020
Epoch   8 Batch   84/134 - Train Accuracy:  0.966, Validation Accuracy:  0.969, Loss:  0.018
Epoch   8 Batch   85/134 - Train Accuracy:  0.973, Validation Accuracy:  0.971, Loss:  0.018
Epoch   8 Batch   86/134 - Train Accuracy:  0.968, Validation Accuracy:  0.970, Loss:  0.018
Epoch   8 Batch   87/134 - Train Accuracy:  0.968, Validation Accuracy:  0.967, Loss:  0.022
Epoch   8 Batch   88/134 - Train Accuracy:  0.976, Validation Accuracy:  0.967, Loss:  0.016
Epoch   8 Batch   89/134 - Train Accuracy:  0.971, Validation Accuracy:  0.970, Loss:  0.019
Epoch   8 Batch   90/134 - Train Accuracy:  0.966, Validation Accuracy:  0.972, Loss:  0.021
Epoch   8 Batch   91/134 - Train Accuracy:  0.972, Validation Accuracy:  0.974, Loss:  0.016
Epoch   8 Batch   92/134 - Train Accuracy:  0.977, Validation Accuracy:  0.972, Loss:  0.017
Epoch   8 Batch   93/134 - Train Accuracy:  0.968, Validation Accuracy:  0.969, Loss:  0.017
Epoch   8 Batch   94/134 - Train Accuracy:  0.970, Validation Accuracy:  0.970, Loss:  0.018
Epoch   8 Batch   95/134 - Train Accuracy:  0.974, Validation Accuracy:  0.972, Loss:  0.018
Epoch   8 Batch   96/134 - Train Accuracy:  0.963, Validation Accuracy:  0.970, Loss:  0.021
Epoch   8 Batch   97/134 - Train Accuracy:  0.969, Validation Accuracy:  0.975, Loss:  0.019
Epoch   8 Batch   98/134 - Train Accuracy:  0.973, Validation Accuracy:  0.971, Loss:  0.017
Epoch   8 Batch   99/134 - Train Accuracy:  0.977, Validation Accuracy:  0.970, Loss:  0.018
Epoch   8 Batch  100/134 - Train Accuracy:  0.973, Validation Accuracy:  0.974, Loss:  0.020
Epoch   8 Batch  101/134 - Train Accuracy:  0.973, Validation Accuracy:  0.968, Loss:  0.019
Epoch   8 Batch  102/134 - Train Accuracy:  0.970, Validation Accuracy:  0.970, Loss:  0.018
Epoch   8 Batch  103/134 - Train Accuracy:  0.968, Validation Accuracy:  0.977, Loss:  0.019
Epoch   8 Batch  104/134 - Train Accuracy:  0.970, Validation Accuracy:  0.971, Loss:  0.017
Epoch   8 Batch  105/134 - Train Accuracy:  0.969, Validation Accuracy:  0.975, Loss:  0.024
Epoch   8 Batch  106/134 - Train Accuracy:  0.970, Validation Accuracy:  0.970, Loss:  0.018
Epoch   8 Batch  107/134 - Train Accuracy:  0.964, Validation Accuracy:  0.972, Loss:  0.023
Epoch   8 Batch  108/134 - Train Accuracy:  0.972, Validation Accuracy:  0.976, Loss:  0.019
Epoch   8 Batch  109/134 - Train Accuracy:  0.962, Validation Accuracy:  0.968, Loss:  0.019
Epoch   8 Batch  110/134 - Train Accuracy:  0.971, Validation Accuracy:  0.969, Loss:  0.018
Epoch   8 Batch  111/134 - Train Accuracy:  0.966, Validation Accuracy:  0.973, Loss:  0.020
Epoch   8 Batch  112/134 - Train Accuracy:  0.965, Validation Accuracy:  0.970, Loss:  0.020
Epoch   8 Batch  113/134 - Train Accuracy:  0.969, Validation Accuracy:  0.973, Loss:  0.022
Epoch   8 Batch  114/134 - Train Accuracy:  0.970, Validation Accuracy:  0.972, Loss:  0.018
Epoch   8 Batch  115/134 - Train Accuracy:  0.968, Validation Accuracy:  0.970, Loss:  0.018
Epoch   8 Batch  116/134 - Train Accuracy:  0.974, Validation Accuracy:  0.963, Loss:  0.020
Epoch   8 Batch  117/134 - Train Accuracy:  0.977, Validation Accuracy:  0.967, Loss:  0.016
Epoch   8 Batch  118/134 - Train Accuracy:  0.977, Validation Accuracy:  0.971, Loss:  0.018
Epoch   8 Batch  119/134 - Train Accuracy:  0.972, Validation Accuracy:  0.973, Loss:  0.017
Epoch   8 Batch  120/134 - Train Accuracy:  0.970, Validation Accuracy:  0.970, Loss:  0.019
Epoch   8 Batch  121/134 - Train Accuracy:  0.971, Validation Accuracy:  0.972, Loss:  0.017
Epoch   8 Batch  122/134 - Train Accuracy:  0.967, Validation Accuracy:  0.968, Loss:  0.019
Epoch   8 Batch  123/134 - Train Accuracy:  0.969, Validation Accuracy:  0.968, Loss:  0.017
Epoch   8 Batch  124/134 - Train Accuracy:  0.968, Validation Accuracy:  0.972, Loss:  0.019
Epoch   8 Batch  125/134 - Train Accuracy:  0.980, Validation Accuracy:  0.973, Loss:  0.016
Epoch   8 Batch  126/134 - Train Accuracy:  0.965, Validation Accuracy:  0.973, Loss:  0.020
Epoch   8 Batch  127/134 - Train Accuracy:  0.970, Validation Accuracy:  0.973, Loss:  0.019
Epoch   8 Batch  128/134 - Train Accuracy:  0.967, Validation Accuracy:  0.975, Loss:  0.021
Epoch   8 Batch  129/134 - Train Accuracy:  0.972, Validation Accuracy:  0.977, Loss:  0.018
Epoch   8 Batch  130/134 - Train Accuracy:  0.972, Validation Accuracy:  0.971, Loss:  0.019
Epoch   8 Batch  131/134 - Train Accuracy:  0.969, Validation Accuracy:  0.969, Loss:  0.019
Epoch   8 Batch  132/134 - Train Accuracy:  0.969, Validation Accuracy:  0.971, Loss:  0.016
Epoch   9 Batch    0/134 - Train Accuracy:  0.977, Validation Accuracy:  0.971, Loss:  0.018
Epoch   9 Batch    1/134 - Train Accuracy:  0.967, Validation Accuracy:  0.969, Loss:  0.017
Epoch   9 Batch    2/134 - Train Accuracy:  0.973, Validation Accuracy:  0.967, Loss:  0.016
Epoch   9 Batch    3/134 - Train Accuracy:  0.972, Validation Accuracy:  0.970, Loss:  0.021
Epoch   9 Batch    4/134 - Train Accuracy:  0.974, Validation Accuracy:  0.972, Loss:  0.017
Epoch   9 Batch    5/134 - Train Accuracy:  0.967, Validation Accuracy:  0.970, Loss:  0.022
Epoch   9 Batch    6/134 - Train Accuracy:  0.975, Validation Accuracy:  0.971, Loss:  0.015
Epoch   9 Batch    7/134 - Train Accuracy:  0.971, Validation Accuracy:  0.970, Loss:  0.019
Epoch   9 Batch    8/134 - Train Accuracy:  0.973, Validation Accuracy:  0.972, Loss:  0.017
Epoch   9 Batch    9/134 - Train Accuracy:  0.974, Validation Accuracy:  0.976, Loss:  0.016
Epoch   9 Batch   10/134 - Train Accuracy:  0.969, Validation Accuracy:  0.975, Loss:  0.016
Epoch   9 Batch   11/134 - Train Accuracy:  0.963, Validation Accuracy:  0.972, Loss:  0.018
Epoch   9 Batch   12/134 - Train Accuracy:  0.970, Validation Accuracy:  0.968, Loss:  0.018
Epoch   9 Batch   13/134 - Train Accuracy:  0.973, Validation Accuracy:  0.970, Loss:  0.017
Epoch   9 Batch   14/134 - Train Accuracy:  0.971, Validation Accuracy:  0.974, Loss:  0.019
Epoch   9 Batch   15/134 - Train Accuracy:  0.970, Validation Accuracy:  0.970, Loss:  0.017
Epoch   9 Batch   16/134 - Train Accuracy:  0.973, Validation Accuracy:  0.967, Loss:  0.016
Epoch   9 Batch   17/134 - Train Accuracy:  0.969, Validation Accuracy:  0.974, Loss:  0.022
Epoch   9 Batch   18/134 - Train Accuracy:  0.973, Validation Accuracy:  0.973, Loss:  0.018
Epoch   9 Batch   19/134 - Train Accuracy:  0.969, Validation Accuracy:  0.974, Loss:  0.020
Epoch   9 Batch   20/134 - Train Accuracy:  0.972, Validation Accuracy:  0.974, Loss:  0.016
Epoch   9 Batch   21/134 - Train Accuracy:  0.970, Validation Accuracy:  0.975, Loss:  0.021
Epoch   9 Batch   22/134 - Train Accuracy:  0.968, Validation Accuracy:  0.976, Loss:  0.018
Epoch   9 Batch   23/134 - Train Accuracy:  0.973, Validation Accuracy:  0.974, Loss:  0.016
Epoch   9 Batch   24/134 - Train Accuracy:  0.963, Validation Accuracy:  0.973, Loss:  0.019
Epoch   9 Batch   25/134 - Train Accuracy:  0.966, Validation Accuracy:  0.972, Loss:  0.016
Epoch   9 Batch   26/134 - Train Accuracy:  0.969, Validation Accuracy:  0.973, Loss:  0.018
Epoch   9 Batch   27/134 - Train Accuracy:  0.976, Validation Accuracy:  0.972, Loss:  0.017
Epoch   9 Batch   28/134 - Train Accuracy:  0.973, Validation Accuracy:  0.973, Loss:  0.020
Epoch   9 Batch   29/134 - Train Accuracy:  0.975, Validation Accuracy:  0.970, Loss:  0.016
Epoch   9 Batch   30/134 - Train Accuracy:  0.967, Validation Accuracy:  0.973, Loss:  0.019
Epoch   9 Batch   31/134 - Train Accuracy:  0.970, Validation Accuracy:  0.976, Loss:  0.017
Epoch   9 Batch   32/134 - Train Accuracy:  0.968, Validation Accuracy:  0.975, Loss:  0.019
Epoch   9 Batch   33/134 - Train Accuracy:  0.969, Validation Accuracy:  0.970, Loss:  0.020
Epoch   9 Batch   34/134 - Train Accuracy:  0.974, Validation Accuracy:  0.968, Loss:  0.020
Epoch   9 Batch   35/134 - Train Accuracy:  0.968, Validation Accuracy:  0.974, Loss:  0.021
Epoch   9 Batch   36/134 - Train Accuracy:  0.972, Validation Accuracy:  0.974, Loss:  0.017
Epoch   9 Batch   37/134 - Train Accuracy:  0.969, Validation Accuracy:  0.973, Loss:  0.019
Epoch   9 Batch   38/134 - Train Accuracy:  0.976, Validation Accuracy:  0.974, Loss:  0.017
Epoch   9 Batch   39/134 - Train Accuracy:  0.965, Validation Accuracy:  0.976, Loss:  0.019
Epoch   9 Batch   40/134 - Train Accuracy:  0.975, Validation Accuracy:  0.976, Loss:  0.018
Epoch   9 Batch   41/134 - Train Accuracy:  0.971, Validation Accuracy:  0.977, Loss:  0.021
Epoch   9 Batch   42/134 - Train Accuracy:  0.971, Validation Accuracy:  0.974, Loss:  0.016
Epoch   9 Batch   43/134 - Train Accuracy:  0.969, Validation Accuracy:  0.967, Loss:  0.016
Epoch   9 Batch   44/134 - Train Accuracy:  0.972, Validation Accuracy:  0.971, Loss:  0.018
Epoch   9 Batch   45/134 - Train Accuracy:  0.978, Validation Accuracy:  0.973, Loss:  0.015
Epoch   9 Batch   46/134 - Train Accuracy:  0.976, Validation Accuracy:  0.974, Loss:  0.017
Epoch   9 Batch   47/134 - Train Accuracy:  0.973, Validation Accuracy:  0.974, Loss:  0.019
Epoch   9 Batch   48/134 - Train Accuracy:  0.968, Validation Accuracy:  0.970, Loss:  0.018
Epoch   9 Batch   49/134 - Train Accuracy:  0.972, Validation Accuracy:  0.971, Loss:  0.016
Epoch   9 Batch   50/134 - Train Accuracy:  0.972, Validation Accuracy:  0.973, Loss:  0.019
Epoch   9 Batch   51/134 - Train Accuracy:  0.973, Validation Accuracy:  0.971, Loss:  0.019
Epoch   9 Batch   52/134 - Train Accuracy:  0.975, Validation Accuracy:  0.972, Loss:  0.016
Epoch   9 Batch   53/134 - Train Accuracy:  0.976, Validation Accuracy:  0.973, Loss:  0.018
Epoch   9 Batch   54/134 - Train Accuracy:  0.971, Validation Accuracy:  0.973, Loss:  0.018
Epoch   9 Batch   55/134 - Train Accuracy:  0.970, Validation Accuracy:  0.974, Loss:  0.018
Epoch   9 Batch   56/134 - Train Accuracy:  0.973, Validation Accuracy:  0.974, Loss:  0.016
Epoch   9 Batch   57/134 - Train Accuracy:  0.976, Validation Accuracy:  0.974, Loss:  0.016
Epoch   9 Batch   58/134 - Train Accuracy:  0.972, Validation Accuracy:  0.973, Loss:  0.016
Epoch   9 Batch   59/134 - Train Accuracy:  0.966, Validation Accuracy:  0.974, Loss:  0.018
Epoch   9 Batch   60/134 - Train Accuracy:  0.971, Validation Accuracy:  0.976, Loss:  0.018
Epoch   9 Batch   61/134 - Train Accuracy:  0.973, Validation Accuracy:  0.973, Loss:  0.017
Epoch   9 Batch   62/134 - Train Accuracy:  0.969, Validation Accuracy:  0.965, Loss:  0.016
Epoch   9 Batch   63/134 - Train Accuracy:  0.977, Validation Accuracy:  0.970, Loss:  0.016
Epoch   9 Batch   64/134 - Train Accuracy:  0.967, Validation Accuracy:  0.973, Loss:  0.020
Epoch   9 Batch   65/134 - Train Accuracy:  0.964, Validation Accuracy:  0.970, Loss:  0.019
Epoch   9 Batch   66/134 - Train Accuracy:  0.972, Validation Accuracy:  0.972, Loss:  0.016
Epoch   9 Batch   67/134 - Train Accuracy:  0.967, Validation Accuracy:  0.971, Loss:  0.017
Epoch   9 Batch   68/134 - Train Accuracy:  0.969, Validation Accuracy:  0.972, Loss:  0.018
Epoch   9 Batch   69/134 - Train Accuracy:  0.969, Validation Accuracy:  0.972, Loss:  0.017
Epoch   9 Batch   70/134 - Train Accuracy:  0.973, Validation Accuracy:  0.972, Loss:  0.018
Epoch   9 Batch   71/134 - Train Accuracy:  0.976, Validation Accuracy:  0.969, Loss:  0.014
Epoch   9 Batch   72/134 - Train Accuracy:  0.977, Validation Accuracy:  0.971, Loss:  0.015
Epoch   9 Batch   73/134 - Train Accuracy:  0.973, Validation Accuracy:  0.975, Loss:  0.019
Epoch   9 Batch   74/134 - Train Accuracy:  0.970, Validation Accuracy:  0.973, Loss:  0.020
Epoch   9 Batch   75/134 - Train Accuracy:  0.975, Validation Accuracy:  0.973, Loss:  0.016
Epoch   9 Batch   76/134 - Train Accuracy:  0.977, Validation Accuracy:  0.972, Loss:  0.017
Epoch   9 Batch   77/134 - Train Accuracy:  0.975, Validation Accuracy:  0.970, Loss:  0.016
Epoch   9 Batch   78/134 - Train Accuracy:  0.968, Validation Accuracy:  0.974, Loss:  0.017
Epoch   9 Batch   79/134 - Train Accuracy:  0.973, Validation Accuracy:  0.974, Loss:  0.017
Epoch   9 Batch   80/134 - Train Accuracy:  0.983, Validation Accuracy:  0.973, Loss:  0.016
Epoch   9 Batch   81/134 - Train Accuracy:  0.979, Validation Accuracy:  0.968, Loss:  0.015
Epoch   9 Batch   82/134 - Train Accuracy:  0.978, Validation Accuracy:  0.970, Loss:  0.015
Epoch   9 Batch   83/134 - Train Accuracy:  0.975, Validation Accuracy:  0.972, Loss:  0.017
Epoch   9 Batch   84/134 - Train Accuracy:  0.974, Validation Accuracy:  0.973, Loss:  0.016
Epoch   9 Batch   85/134 - Train Accuracy:  0.975, Validation Accuracy:  0.975, Loss:  0.016
Epoch   9 Batch   86/134 - Train Accuracy:  0.974, Validation Accuracy:  0.974, Loss:  0.016
Epoch   9 Batch   87/134 - Train Accuracy:  0.970, Validation Accuracy:  0.974, Loss:  0.019
Epoch   9 Batch   88/134 - Train Accuracy:  0.981, Validation Accuracy:  0.973, Loss:  0.014
Epoch   9 Batch   89/134 - Train Accuracy:  0.977, Validation Accuracy:  0.974, Loss:  0.015
Epoch   9 Batch   90/134 - Train Accuracy:  0.971, Validation Accuracy:  0.974, Loss:  0.018
Epoch   9 Batch   91/134 - Train Accuracy:  0.977, Validation Accuracy:  0.972, Loss:  0.013
Epoch   9 Batch   92/134 - Train Accuracy:  0.982, Validation Accuracy:  0.972, Loss:  0.014
Epoch   9 Batch   93/134 - Train Accuracy:  0.975, Validation Accuracy:  0.970, Loss:  0.015
Epoch   9 Batch   94/134 - Train Accuracy:  0.978, Validation Accuracy:  0.974, Loss:  0.015
Epoch   9 Batch   95/134 - Train Accuracy:  0.978, Validation Accuracy:  0.973, Loss:  0.014
Epoch   9 Batch   96/134 - Train Accuracy:  0.971, Validation Accuracy:  0.972, Loss:  0.017
Epoch   9 Batch   97/134 - Train Accuracy:  0.972, Validation Accuracy:  0.975, Loss:  0.017
Epoch   9 Batch   98/134 - Train Accuracy:  0.973, Validation Accuracy:  0.972, Loss:  0.015
Epoch   9 Batch   99/134 - Train Accuracy:  0.978, Validation Accuracy:  0.970, Loss:  0.015
Epoch   9 Batch  100/134 - Train Accuracy:  0.975, Validation Accuracy:  0.972, Loss:  0.017
Epoch   9 Batch  101/134 - Train Accuracy:  0.973, Validation Accuracy:  0.971, Loss:  0.016
Epoch   9 Batch  102/134 - Train Accuracy:  0.972, Validation Accuracy:  0.972, Loss:  0.016
Epoch   9 Batch  103/134 - Train Accuracy:  0.975, Validation Accuracy:  0.978, Loss:  0.016
Epoch   9 Batch  104/134 - Train Accuracy:  0.974, Validation Accuracy:  0.972, Loss:  0.014
Epoch   9 Batch  105/134 - Train Accuracy:  0.970, Validation Accuracy:  0.976, Loss:  0.020
Epoch   9 Batch  106/134 - Train Accuracy:  0.971, Validation Accuracy:  0.971, Loss:  0.015
Epoch   9 Batch  107/134 - Train Accuracy:  0.960, Validation Accuracy:  0.968, Loss:  0.020
Epoch   9 Batch  108/134 - Train Accuracy:  0.976, Validation Accuracy:  0.974, Loss:  0.017
Epoch   9 Batch  109/134 - Train Accuracy:  0.968, Validation Accuracy:  0.971, Loss:  0.015
Epoch   9 Batch  110/134 - Train Accuracy:  0.973, Validation Accuracy:  0.970, Loss:  0.015
Epoch   9 Batch  111/134 - Train Accuracy:  0.973, Validation Accuracy:  0.972, Loss:  0.018
Epoch   9 Batch  112/134 - Train Accuracy:  0.975, Validation Accuracy:  0.975, Loss:  0.017
Epoch   9 Batch  113/134 - Train Accuracy:  0.974, Validation Accuracy:  0.974, Loss:  0.019
Epoch   9 Batch  114/134 - Train Accuracy:  0.976, Validation Accuracy:  0.973, Loss:  0.015
Epoch   9 Batch  115/134 - Train Accuracy:  0.973, Validation Accuracy:  0.970, Loss:  0.016
Epoch   9 Batch  116/134 - Train Accuracy:  0.974, Validation Accuracy:  0.969, Loss:  0.018
Epoch   9 Batch  117/134 - Train Accuracy:  0.979, Validation Accuracy:  0.968, Loss:  0.013
Epoch   9 Batch  118/134 - Train Accuracy:  0.976, Validation Accuracy:  0.971, Loss:  0.015
Epoch   9 Batch  119/134 - Train Accuracy:  0.981, Validation Accuracy:  0.972, Loss:  0.014
Epoch   9 Batch  120/134 - Train Accuracy:  0.975, Validation Accuracy:  0.974, Loss:  0.016
Epoch   9 Batch  121/134 - Train Accuracy:  0.977, Validation Accuracy:  0.974, Loss:  0.015
Epoch   9 Batch  122/134 - Train Accuracy:  0.971, Validation Accuracy:  0.974, Loss:  0.017
Epoch   9 Batch  123/134 - Train Accuracy:  0.978, Validation Accuracy:  0.972, Loss:  0.015
Epoch   9 Batch  124/134 - Train Accuracy:  0.973, Validation Accuracy:  0.974, Loss:  0.015
Epoch   9 Batch  125/134 - Train Accuracy:  0.975, Validation Accuracy:  0.973, Loss:  0.014
Epoch   9 Batch  126/134 - Train Accuracy:  0.972, Validation Accuracy:  0.973, Loss:  0.016
Epoch   9 Batch  127/134 - Train Accuracy:  0.973, Validation Accuracy:  0.976, Loss:  0.017
Epoch   9 Batch  128/134 - Train Accuracy:  0.968, Validation Accuracy:  0.973, Loss:  0.018
Epoch   9 Batch  129/134 - Train Accuracy:  0.975, Validation Accuracy:  0.973, Loss:  0.016
Epoch   9 Batch  130/134 - Train Accuracy:  0.975, Validation Accuracy:  0.974, Loss:  0.015
Epoch   9 Batch  131/134 - Train Accuracy:  0.972, Validation Accuracy:  0.975, Loss:  0.016
Epoch   9 Batch  132/134 - Train Accuracy:  0.973, Validation Accuracy:  0.976, Loss:  0.014
Model Trained and Saved

Save Parameters

Save the batch_size and save_path parameters for inference.





In [18]:



    


"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
# Save parameters for checkpoint
helper.save_params(save_path)

Checkpoint





In [19]:



    


"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import tensorflow as tf
import numpy as np
import helper
import problem_unittests as tests

_, (source_vocab_to_int, target_vocab_to_int), (source_int_to_vocab, target_int_to_vocab) = helper.load_preprocess()
load_path = helper.load_params()

Sentence to Sequence

To feed a sentence into the model for translation, you first need to preprocess it. Implement the function sentence_to_seq() to preprocess new sentences.

  • Convert the sentence to lowercase
  • Convert words into ids using vocab_to_int
  • Convert words not in the vocabulary, to the <UNK> word id.




In [20]:



    


def sentence_to_seq(sentence, vocab_to_int):
    """
    Convert a sentence to a sequence of ids
    :param sentence: String
    :param vocab_to_int: Dictionary to go from the words to an id
    :return: List of word ids
    """
    return [vocab_to_int.get(word, vocab_to_int['<UNK>']) for word in sentence.split()]
    #ids = []
    
    #for w in sentence.lower().split():
    #    try:
    #        ids.append(vocab_to_int[w])
    #    except KeyError:
    #        ids.append(vocab_to_int['<UNK>'])
    #return ids


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



    


















    






Tests Passed

Translate

This will translate translate_sentence from English to French.





In [21]:



    


translate_sentence = 'he saw a old yellow truck .'


"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
translate_sentence = sentence_to_seq(translate_sentence, source_vocab_to_int)

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

    input_data = loaded_graph.get_tensor_by_name('input:0')
    logits = loaded_graph.get_tensor_by_name('logits:0')
    keep_prob = loaded_graph.get_tensor_by_name('keep_prob:0')

    translate_logits = sess.run(logits, {input_data: [translate_sentence], keep_prob: 1.0})[0]

print('Input')
print('  Word Ids:      {}'.format([i for i in translate_sentence]))
print('  English Words: {}'.format([source_int_to_vocab[i] for i in translate_sentence]))

print('\nPrediction')
print('  Word Ids:      {}'.format([i for i in np.argmax(translate_logits, 1)]))
print('  French Words: {}'.format([target_int_to_vocab[i] for i in np.argmax(translate_logits, 1)]))



    


















    






Input
  Word Ids:      [222, 121, 148, 122, 34, 42, 33]
  English Words: ['he', 'saw', 'a', 'old', 'yellow', 'truck', '.']

Prediction
  Word Ids:      [218, 150, 141, 93, 143, 280, 1]
  French Words: ['il', 'a', 'un', 'vieux', 'camion', '.', '<EOS>']

Imperfect Translation

You might notice that some sentences translate better than others. Since the dataset you're using only has a vocabulary of 227 English words of the thousands that you use, you're only going to see good results using these words. Additionally, the translations in this data set were made by Google translate, so the translations themselves aren't particularly good. (We apologize to the French speakers out there!) Thankfully, for this project, you don't need a perfect translation. However, if you want to create a better translation model, you'll need better data.

You can train on the WMT10 French-English corpus. This dataset has more vocabulary and richer in topics discussed. However, this will take you days to train, so make sure you've a GPU and the neural network is performing well on dataset we provided. Just make sure you play with the WMT10 corpus after you've submitted this project.

Submitting This Project

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





In [ ]:

Content source: strandbygaard/deep-learning

Similar notebooks:

notebook.community | gallery | about