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 [1]:



    


"""
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 [2]:



    


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 [21]:



    


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)
    """
#     def lookup_ids(sentences, dictionary, is_target = False):    
#         ids_to_text = []
#         for sentence in sentences:
#             words = sentence.split(' ')
#             words = filter(None, words) # this is for the preprocess function in helper code being called below
#             ids = [dictionary[word] for word in words]
#             if(is_target):
#                 ids.append(dictionary['<EOS>'])
#             ids_to_text.append(ids)
#         return ids_to_text
   
#     source_id_text = lookup_ids(source_text.split('\n'), source_vocab_to_int, is_target = False)
#     target_id_text = lookup_ids(target_text.split('\n'), target_vocab_to_int, is_target = True) 
#     print(len(source_id_text), len(target_id_text))

# The more pythonic way to convert text to ids. Notice EOS is also added as a list. 
    source_id_text = [
    [source_vocab_to_int[word] for word in sentence.split()] for sentence in source_text.split('\n')]
    target_id_text = [
    [target_vocab_to_int[word] for word in sentence.split()] + [target_vocab_to_int['<EOS>']]
    for sentence in target_text.split('\n')]
    return source_id_text, target_id_text

"""
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 [4]:



    


"""
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 [5]:



    


"""
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 [6]:



    


"""
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: 1.0.1
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 [7]:



    


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, shape=(None, None), name='input')
    targets = tf.placeholder(tf.int32, shape=(None, None), name='targets')
    learning_rate = tf.placeholder(tf.float32, shape=None)
    keep_prob = tf.placeholder(tf.float32, name='keep_prob')
    return inputs, targets, learning_rate, keep_prob

"""
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 [8]:



    


def process_decoding_input(target_data, target_vocab_to_int, batch_size):
    """
    Preprocess target data for dencoding
    :param target_data: Target Placehoder
    :param target_vocab_to_int: Dictionary to go from the target words to an id
    :param batch_size: Batch Size
    :return: Preprocessed target data
    """
    
    '''This function is from http://www.alivelearn.net/wp-content/uploads/2017/04/dlnd_language_translation.html'''
    ending = tf.strided_slice(target_data, [0, 0], [batch_size, -1], [1, 1])
    target_data = tf.concat([tf.fill([batch_size, 1], target_vocab_to_int['<GO>']), ending], 1)
    return target_data

"""
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 [9]:



    


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
    """
    rnn = tf.contrib.rnn.BasicLSTMCell(rnn_size)
    cell = tf.contrib.rnn.MultiRNNCell([rnn] * num_layers)
    layer = tf.contrib.rnn.DropoutWrapper(cell, output_keep_prob=keep_prob)
    encoding_layer, rnn_state = tf.nn.dynamic_rnn(layer, rnn_inputs, dtype=tf.float32)
    return rnn_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 [10]:



    


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
    """
    dynamic_fn_train = tf.contrib.seq2seq.simple_decoder_fn_train(encoder_state)
    rnn_outputs, final_state, final_context_state = tf.contrib.seq2seq.dynamic_rnn_decoder(
        decoder_fn=dynamic_fn_train,
        cell=dec_cell,
        inputs=dec_embed_input,
        sequence_length=sequence_length,
        scope=decoding_scope,
    )
    
    train_logits =  output_fn(rnn_outputs)
    train_logits = tf.nn.dropout(train_logits, keep_prob)
    return train_logits


"""
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 [22]:



    


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
    """
    dynamic_fn_inference = tf.contrib.seq2seq.simple_decoder_fn_inference(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,
                                                                          output_fn = output_fn, 
                                                                          num_decoder_symbols = vocab_size,
                                                                          dtype=tf.int32, name=None)
    inference_logits, final_state, final_context_state = tf.contrib.seq2seq.dynamic_rnn_decoder(decoder_fn=dynamic_fn_inference, cell=dec_cell, scope=decoding_scope)
    return inference_logits

"""
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 function 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 [12]:



    


# ?decoding_layer_infer



    











In [23]:



    


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)
    """
    rnn = tf.contrib.rnn.BasicLSTMCell(rnn_size)
    cell = tf.contrib.rnn.MultiRNNCell([rnn] * num_layers)
    with tf.variable_scope('decoding_scope') as decoding_scope:
        output_fn = lambda x : tf.contrib.layers.fully_connected(x, vocab_size, None, scope=decoding_scope)
        train_logits = decoding_layer_train(encoder_state, cell, dec_embed_input, sequence_length, 
                                            decoding_scope, output_fn, keep_prob)
    
        decoding_scope.reuse_variables()
        inference_logits = decoding_layer_infer(encoder_state, cell, dec_embeddings, 
                                                target_vocab_to_int['<GO>'], target_vocab_to_int['<EOS>'],
                                               sequence_length, vocab_size, decoding_scope, output_fn, keep_prob)
    return train_logits, inference_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]:



    


# ?decoding_layer



    











In [13]:



    


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)
    """
    
    input_embedding = tf.contrib.layers.embed_sequence(input_data, source_vocab_size, enc_embedding_size)
    encoder_state = encoding_layer(keep_prob=keep_prob, num_layers=num_layers, rnn_inputs=input_embedding, rnn_size=rnn_size)
    processed_target_data = process_decoding_input(target_data, target_vocab_to_int, batch_size)
#     target_embedding = tf.contrib.layers.embed_sequence(target_data, target_vocab_size, dec_embedding_size)
    dec_embeddings = tf.Variable(tf.random_uniform([target_vocab_size, dec_embedding_size]))
    target_embedding = tf.nn.embedding_lookup(dec_embeddings, processed_target_data)
    layer = decoding_layer(target_embedding, dec_embeddings, encoder_state, target_vocab_size, sequence_length, rnn_size, \
                   num_layers, target_vocab_to_int, keep_prob)
    return layer


"""
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 [29]:



    


# Number of Epochs
epochs = 3
# Batch Size
batch_size = 256
# RNN Size
rnn_size = 512
# Number of Layers
num_layers = 2
# Embedding Size
encoding_embedding_size = 256
decoding_embedding_size = 256
# Learning Rate
learning_rate = 0.0011
# Dropout Keep Probability
keep_probability = 0.5
# Network parameters are from in www.alivelearn.net/wp-content/uploads/2017/04/dlnd_language_translation.html

Build the Graph

Build the graph using the neural network you implemented.





In [30]:



    


"""
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 forms to see if anyone is having the same problem.





In [31]:



    


"""
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    0/538 - Train Accuracy:  0.234, Validation Accuracy:  0.316, Loss:  5.927
Epoch   0 Batch    1/538 - Train Accuracy:  0.234, Validation Accuracy:  0.305, Loss:  5.384
Epoch   0 Batch    2/538 - Train Accuracy:  0.282, Validation Accuracy:  0.346, Loss:  5.596
Epoch   0 Batch    3/538 - Train Accuracy:  0.264, Validation Accuracy:  0.346, Loss:  5.180
Epoch   0 Batch    4/538 - Train Accuracy:  0.278, Validation Accuracy:  0.353, Loss:  4.910
Epoch   0 Batch    5/538 - Train Accuracy:  0.322, Validation Accuracy:  0.374, Loss:  4.788
Epoch   0 Batch    6/538 - Train Accuracy:  0.332, Validation Accuracy:  0.370, Loss:  4.632
Epoch   0 Batch    7/538 - Train Accuracy:  0.327, Validation Accuracy:  0.389, Loss:  4.727
Epoch   0 Batch    8/538 - Train Accuracy:  0.307, Validation Accuracy:  0.374, Loss:  4.630
Epoch   0 Batch    9/538 - Train Accuracy:  0.319, Validation Accuracy:  0.382, Loss:  4.611
Epoch   0 Batch   10/538 - Train Accuracy:  0.314, Validation Accuracy:  0.396, Loss:  4.549
Epoch   0 Batch   11/538 - Train Accuracy:  0.328, Validation Accuracy:  0.398, Loss:  4.498
Epoch   0 Batch   12/538 - Train Accuracy:  0.328, Validation Accuracy:  0.401, Loss:  4.493
Epoch   0 Batch   13/538 - Train Accuracy:  0.379, Validation Accuracy:  0.403, Loss:  4.354
Epoch   0 Batch   14/538 - Train Accuracy:  0.355, Validation Accuracy:  0.420, Loss:  4.427
Epoch   0 Batch   15/538 - Train Accuracy:  0.405, Validation Accuracy:  0.432, Loss:  4.250
Epoch   0 Batch   16/538 - Train Accuracy:  0.389, Validation Accuracy:  0.427, Loss:  4.244
Epoch   0 Batch   17/538 - Train Accuracy:  0.401, Validation Accuracy:  0.454, Loss:  4.319
Epoch   0 Batch   18/538 - Train Accuracy:  0.380, Validation Accuracy:  0.443, Loss:  4.248
Epoch   0 Batch   19/538 - Train Accuracy:  0.369, Validation Accuracy:  0.437, Loss:  4.273
Epoch   0 Batch   20/538 - Train Accuracy:  0.411, Validation Accuracy:  0.452, Loss:  4.172
Epoch   0 Batch   21/538 - Train Accuracy:  0.371, Validation Accuracy:  0.460, Loss:  4.282
Epoch   0 Batch   22/538 - Train Accuracy:  0.397, Validation Accuracy:  0.456, Loss:  4.202
Epoch   0 Batch   23/538 - Train Accuracy:  0.402, Validation Accuracy:  0.455, Loss:  4.142
Epoch   0 Batch   24/538 - Train Accuracy:  0.400, Validation Accuracy:  0.444, Loss:  4.178
Epoch   0 Batch   25/538 - Train Accuracy:  0.393, Validation Accuracy:  0.450, Loss:  4.171
Epoch   0 Batch   26/538 - Train Accuracy:  0.417, Validation Accuracy:  0.471, Loss:  4.181
Epoch   0 Batch   27/538 - Train Accuracy:  0.415, Validation Accuracy:  0.465, Loss:  4.068
Epoch   0 Batch   28/538 - Train Accuracy:  0.460, Validation Accuracy:  0.462, Loss:  3.969
Epoch   0 Batch   29/538 - Train Accuracy:  0.444, Validation Accuracy:  0.482, Loss:  4.045
Epoch   0 Batch   30/538 - Train Accuracy:  0.394, Validation Accuracy:  0.443, Loss:  4.074
Epoch   0 Batch   31/538 - Train Accuracy:  0.435, Validation Accuracy:  0.477, Loss:  3.990
Epoch   0 Batch   32/538 - Train Accuracy:  0.418, Validation Accuracy:  0.476, Loss:  4.009
Epoch   0 Batch   33/538 - Train Accuracy:  0.440, Validation Accuracy:  0.477, Loss:  3.995
Epoch   0 Batch   34/538 - Train Accuracy:  0.429, Validation Accuracy:  0.483, Loss:  4.036
Epoch   0 Batch   35/538 - Train Accuracy:  0.417, Validation Accuracy:  0.491, Loss:  4.010
Epoch   0 Batch   36/538 - Train Accuracy:  0.446, Validation Accuracy:  0.491, Loss:  3.954
Epoch   0 Batch   37/538 - Train Accuracy:  0.453, Validation Accuracy:  0.494, Loss:  4.021
Epoch   0 Batch   38/538 - Train Accuracy:  0.411, Validation Accuracy:  0.488, Loss:  3.971
Epoch   0 Batch   39/538 - Train Accuracy:  0.432, Validation Accuracy:  0.490, Loss:  3.960
Epoch   0 Batch   40/538 - Train Accuracy:  0.483, Validation Accuracy:  0.488, Loss:  3.847
Epoch   0 Batch   41/538 - Train Accuracy:  0.438, Validation Accuracy:  0.495, Loss:  3.861
Epoch   0 Batch   42/538 - Train Accuracy:  0.439, Validation Accuracy:  0.490, Loss:  3.917
Epoch   0 Batch   43/538 - Train Accuracy:  0.447, Validation Accuracy:  0.503, Loss:  3.894
Epoch   0 Batch   44/538 - Train Accuracy:  0.415, Validation Accuracy:  0.482, Loss:  3.972
Epoch   0 Batch   45/538 - Train Accuracy:  0.467, Validation Accuracy:  0.501, Loss:  3.789
Epoch   0 Batch   46/538 - Train Accuracy:  0.407, Validation Accuracy:  0.471, Loss:  3.899
Epoch   0 Batch   47/538 - Train Accuracy:  0.466, Validation Accuracy:  0.493, Loss:  3.859
Epoch   0 Batch   48/538 - Train Accuracy:  0.464, Validation Accuracy:  0.491, Loss:  3.833
Epoch   0 Batch   49/538 - Train Accuracy:  0.402, Validation Accuracy:  0.473, Loss:  3.850
Epoch   0 Batch   50/538 - Train Accuracy:  0.456, Validation Accuracy:  0.498, Loss:  3.855
Epoch   0 Batch   51/538 - Train Accuracy:  0.348, Validation Accuracy:  0.466, Loss:  4.008
Epoch   0 Batch   52/538 - Train Accuracy:  0.428, Validation Accuracy:  0.467, Loss:  3.842
Epoch   0 Batch   53/538 - Train Accuracy:  0.491, Validation Accuracy:  0.499, Loss:  3.698
Epoch   0 Batch   54/538 - Train Accuracy:  0.432, Validation Accuracy:  0.480, Loss:  3.898
Epoch   0 Batch   55/538 - Train Accuracy:  0.398, Validation Accuracy:  0.470, Loss:  3.754
Epoch   0 Batch   56/538 - Train Accuracy:  0.485, Validation Accuracy:  0.515, Loss:  3.756
Epoch   0 Batch   57/538 - Train Accuracy:  0.445, Validation Accuracy:  0.517, Loss:  3.761
Epoch   0 Batch   58/538 - Train Accuracy:  0.412, Validation Accuracy:  0.489, Loss:  3.747
Epoch   0 Batch   59/538 - Train Accuracy:  0.436, Validation Accuracy:  0.493, Loss:  3.731
Epoch   0 Batch   60/538 - Train Accuracy:  0.463, Validation Accuracy:  0.511, Loss:  3.712
Epoch   0 Batch   61/538 - Train Accuracy:  0.442, Validation Accuracy:  0.495, Loss:  3.660
Epoch   0 Batch   62/538 - Train Accuracy:  0.453, Validation Accuracy:  0.493, Loss:  3.669
Epoch   0 Batch   63/538 - Train Accuracy:  0.481, Validation Accuracy:  0.491, Loss:  3.584
Epoch   0 Batch   64/538 - Train Accuracy:  0.483, Validation Accuracy:  0.508, Loss:  3.586
Epoch   0 Batch   65/538 - Train Accuracy:  0.412, Validation Accuracy:  0.483, Loss:  3.645
Epoch   0 Batch   66/538 - Train Accuracy:  0.463, Validation Accuracy:  0.496, Loss:  3.524
Epoch   0 Batch   67/538 - Train Accuracy:  0.444, Validation Accuracy:  0.497, Loss:  3.612
Epoch   0 Batch   68/538 - Train Accuracy:  0.457, Validation Accuracy:  0.489, Loss:  3.585
Epoch   0 Batch   69/538 - Train Accuracy:  0.388, Validation Accuracy:  0.459, Loss:  3.595
Epoch   0 Batch   70/538 - Train Accuracy:  0.488, Validation Accuracy:  0.516, Loss:  3.609
Epoch   0 Batch   71/538 - Train Accuracy:  0.382, Validation Accuracy:  0.438, Loss:  3.606
Epoch   0 Batch   72/538 - Train Accuracy:  0.501, Validation Accuracy:  0.512, Loss:  3.539
Epoch   0 Batch   73/538 - Train Accuracy:  0.415, Validation Accuracy:  0.478, Loss:  3.611
Epoch   0 Batch   74/538 - Train Accuracy:  0.418, Validation Accuracy:  0.463, Loss:  3.487
Epoch   0 Batch   75/538 - Train Accuracy:  0.485, Validation Accuracy:  0.509, Loss:  3.546
Epoch   0 Batch   76/538 - Train Accuracy:  0.436, Validation Accuracy:  0.500, Loss:  3.553
Epoch   0 Batch   77/538 - Train Accuracy:  0.418, Validation Accuracy:  0.479, Loss:  3.528
Epoch   0 Batch   78/538 - Train Accuracy:  0.480, Validation Accuracy:  0.512, Loss:  3.463
Epoch   0 Batch   79/538 - Train Accuracy:  0.491, Validation Accuracy:  0.514, Loss:  3.421
Epoch   0 Batch   80/538 - Train Accuracy:  0.402, Validation Accuracy:  0.471, Loss:  3.548
Epoch   0 Batch   81/538 - Train Accuracy:  0.440, Validation Accuracy:  0.495, Loss:  3.463
Epoch   0 Batch   82/538 - Train Accuracy:  0.456, Validation Accuracy:  0.500, Loss:  3.486
Epoch   0 Batch   83/538 - Train Accuracy:  0.436, Validation Accuracy:  0.495, Loss:  3.449
Epoch   0 Batch   84/538 - Train Accuracy:  0.475, Validation Accuracy:  0.495, Loss:  3.382
Epoch   0 Batch   85/538 - Train Accuracy:  0.493, Validation Accuracy:  0.506, Loss:  3.406
Epoch   0 Batch   86/538 - Train Accuracy:  0.449, Validation Accuracy:  0.502, Loss:  3.454
Epoch   0 Batch   87/538 - Train Accuracy:  0.427, Validation Accuracy:  0.498, Loss:  3.378
Epoch   0 Batch   88/538 - Train Accuracy:  0.434, Validation Accuracy:  0.481, Loss:  3.455
Epoch   0 Batch   89/538 - Train Accuracy:  0.454, Validation Accuracy:  0.492, Loss:  3.452
Epoch   0 Batch   90/538 - Train Accuracy:  0.478, Validation Accuracy:  0.491, Loss:  3.421
Epoch   0 Batch   91/538 - Train Accuracy:  0.440, Validation Accuracy:  0.498, Loss:  3.480
Epoch   0 Batch   92/538 - Train Accuracy:  0.464, Validation Accuracy:  0.511, Loss:  3.395
Epoch   0 Batch   93/538 - Train Accuracy:  0.469, Validation Accuracy:  0.523, Loss:  3.452
Epoch   0 Batch   94/538 - Train Accuracy:  0.459, Validation Accuracy:  0.510, Loss:  3.385
Epoch   0 Batch   95/538 - Train Accuracy:  0.524, Validation Accuracy:  0.525, Loss:  3.325
Epoch   0 Batch   96/538 - Train Accuracy:  0.509, Validation Accuracy:  0.518, Loss:  3.363
Epoch   0 Batch   97/538 - Train Accuracy:  0.472, Validation Accuracy:  0.522, Loss:  3.366
Epoch   0 Batch   98/538 - Train Accuracy:  0.501, Validation Accuracy:  0.509, Loss:  3.288
Epoch   0 Batch   99/538 - Train Accuracy:  0.466, Validation Accuracy:  0.507, Loss:  3.397
Epoch   0 Batch  100/538 - Train Accuracy:  0.499, Validation Accuracy:  0.535, Loss:  3.323
Epoch   0 Batch  101/538 - Train Accuracy:  0.458, Validation Accuracy:  0.506, Loss:  3.300
Epoch   0 Batch  102/538 - Train Accuracy:  0.494, Validation Accuracy:  0.522, Loss:  3.344
Epoch   0 Batch  103/538 - Train Accuracy:  0.493, Validation Accuracy:  0.528, Loss:  3.353
Epoch   0 Batch  104/538 - Train Accuracy:  0.441, Validation Accuracy:  0.468, Loss:  3.291
Epoch   0 Batch  105/538 - Train Accuracy:  0.482, Validation Accuracy:  0.511, Loss:  3.315
Epoch   0 Batch  106/538 - Train Accuracy:  0.447, Validation Accuracy:  0.492, Loss:  3.310
Epoch   0 Batch  107/538 - Train Accuracy:  0.447, Validation Accuracy:  0.497, Loss:  3.365
Epoch   0 Batch  108/538 - Train Accuracy:  0.499, Validation Accuracy:  0.522, Loss:  3.232
Epoch   0 Batch  109/538 - Train Accuracy:  0.484, Validation Accuracy:  0.514, Loss:  3.297
Epoch   0 Batch  110/538 - Train Accuracy:  0.467, Validation Accuracy:  0.516, Loss:  3.381
Epoch   0 Batch  111/538 - Train Accuracy:  0.508, Validation Accuracy:  0.533, Loss:  3.267
Epoch   0 Batch  112/538 - Train Accuracy:  0.460, Validation Accuracy:  0.521, Loss:  3.250
Epoch   0 Batch  113/538 - Train Accuracy:  0.472, Validation Accuracy:  0.504, Loss:  3.370
Epoch   0 Batch  114/538 - Train Accuracy:  0.547, Validation Accuracy:  0.534, Loss:  3.270
Epoch   0 Batch  115/538 - Train Accuracy:  0.478, Validation Accuracy:  0.516, Loss:  3.257
Epoch   0 Batch  116/538 - Train Accuracy:  0.498, Validation Accuracy:  0.525, Loss:  3.239
Epoch   0 Batch  117/538 - Train Accuracy:  0.522, Validation Accuracy:  0.515, Loss:  3.256
Epoch   0 Batch  118/538 - Train Accuracy:  0.509, Validation Accuracy:  0.518, Loss:  3.241
Epoch   0 Batch  119/538 - Train Accuracy:  0.502, Validation Accuracy:  0.520, Loss:  3.189
Epoch   0 Batch  120/538 - Train Accuracy:  0.487, Validation Accuracy:  0.515, Loss:  3.308
Epoch   0 Batch  121/538 - Train Accuracy:  0.531, Validation Accuracy:  0.544, Loss:  3.270
Epoch   0 Batch  122/538 - Train Accuracy:  0.517, Validation Accuracy:  0.534, Loss:  3.240
Epoch   0 Batch  123/538 - Train Accuracy:  0.531, Validation Accuracy:  0.536, Loss:  3.200
Epoch   0 Batch  124/538 - Train Accuracy:  0.514, Validation Accuracy:  0.538, Loss:  3.164
Epoch   0 Batch  125/538 - Train Accuracy:  0.534, Validation Accuracy:  0.530, Loss:  3.205
Epoch   0 Batch  126/538 - Train Accuracy:  0.522, Validation Accuracy:  0.519, Loss:  3.257
Epoch   0 Batch  127/538 - Train Accuracy:  0.470, Validation Accuracy:  0.523, Loss:  3.223
Epoch   0 Batch  128/538 - Train Accuracy:  0.516, Validation Accuracy:  0.526, Loss:  3.223
Epoch   0 Batch  129/538 - Train Accuracy:  0.504, Validation Accuracy:  0.531, Loss:  3.200
Epoch   0 Batch  130/538 - Train Accuracy:  0.521, Validation Accuracy:  0.527, Loss:  3.192
Epoch   0 Batch  131/538 - Train Accuracy:  0.486, Validation Accuracy:  0.534, Loss:  3.229
Epoch   0 Batch  132/538 - Train Accuracy:  0.513, Validation Accuracy:  0.522, Loss:  3.238
Epoch   0 Batch  133/538 - Train Accuracy:  0.532, Validation Accuracy:  0.541, Loss:  3.154
Epoch   0 Batch  134/538 - Train Accuracy:  0.496, Validation Accuracy:  0.547, Loss:  3.247
Epoch   0 Batch  135/538 - Train Accuracy:  0.520, Validation Accuracy:  0.540, Loss:  3.216
Epoch   0 Batch  136/538 - Train Accuracy:  0.509, Validation Accuracy:  0.540, Loss:  3.185
Epoch   0 Batch  137/538 - Train Accuracy:  0.499, Validation Accuracy:  0.536, Loss:  3.199
Epoch   0 Batch  138/538 - Train Accuracy:  0.505, Validation Accuracy:  0.530, Loss:  3.193
Epoch   0 Batch  139/538 - Train Accuracy:  0.495, Validation Accuracy:  0.542, Loss:  3.229
Epoch   0 Batch  140/538 - Train Accuracy:  0.504, Validation Accuracy:  0.546, Loss:  3.237
Epoch   0 Batch  141/538 - Train Accuracy:  0.501, Validation Accuracy:  0.532, Loss:  3.244
Epoch   0 Batch  142/538 - Train Accuracy:  0.540, Validation Accuracy:  0.527, Loss:  3.150
Epoch   0 Batch  143/538 - Train Accuracy:  0.519, Validation Accuracy:  0.554, Loss:  3.101
Epoch   0 Batch  144/538 - Train Accuracy:  0.539, Validation Accuracy:  0.553, Loss:  3.223
Epoch   0 Batch  145/538 - Train Accuracy:  0.536, Validation Accuracy:  0.548, Loss:  3.156
Epoch   0 Batch  146/538 - Train Accuracy:  0.530, Validation Accuracy:  0.537, Loss:  3.182
Epoch   0 Batch  147/538 - Train Accuracy:  0.512, Validation Accuracy:  0.509, Loss:  3.153
Epoch   0 Batch  148/538 - Train Accuracy:  0.497, Validation Accuracy:  0.536, Loss:  3.165
Epoch   0 Batch  149/538 - Train Accuracy:  0.512, Validation Accuracy:  0.545, Loss:  3.162
Epoch   0 Batch  150/538 - Train Accuracy:  0.571, Validation Accuracy:  0.563, Loss:  3.169
Epoch   0 Batch  151/538 - Train Accuracy:  0.538, Validation Accuracy:  0.557, Loss:  3.151
Epoch   0 Batch  152/538 - Train Accuracy:  0.551, Validation Accuracy:  0.553, Loss:  3.185
Epoch   0 Batch  153/538 - Train Accuracy:  0.530, Validation Accuracy:  0.554, Loss:  3.188
Epoch   0 Batch  154/538 - Train Accuracy:  0.541, Validation Accuracy:  0.556, Loss:  3.093
Epoch   0 Batch  155/538 - Train Accuracy:  0.579, Validation Accuracy:  0.561, Loss:  3.124
Epoch   0 Batch  156/538 - Train Accuracy:  0.541, Validation Accuracy:  0.556, Loss:  3.136
Epoch   0 Batch  157/538 - Train Accuracy:  0.539, Validation Accuracy:  0.558, Loss:  3.110
Epoch   0 Batch  158/538 - Train Accuracy:  0.539, Validation Accuracy:  0.553, Loss:  3.226
Epoch   0 Batch  159/538 - Train Accuracy:  0.538, Validation Accuracy:  0.556, Loss:  3.130
Epoch   0 Batch  160/538 - Train Accuracy:  0.547, Validation Accuracy:  0.560, Loss:  3.085
Epoch   0 Batch  161/538 - Train Accuracy:  0.531, Validation Accuracy:  0.551, Loss:  3.087
Epoch   0 Batch  162/538 - Train Accuracy:  0.573, Validation Accuracy:  0.552, Loss:  3.068
Epoch   0 Batch  163/538 - Train Accuracy:  0.545, Validation Accuracy:  0.562, Loss:  3.118
Epoch   0 Batch  164/538 - Train Accuracy:  0.542, Validation Accuracy:  0.569, Loss:  3.137
Epoch   0 Batch  165/538 - Train Accuracy:  0.540, Validation Accuracy:  0.563, Loss:  3.109
Epoch   0 Batch  166/538 - Train Accuracy:  0.542, Validation Accuracy:  0.561, Loss:  3.162
Epoch   0 Batch  167/538 - Train Accuracy:  0.563, Validation Accuracy:  0.563, Loss:  3.089
Epoch   0 Batch  168/538 - Train Accuracy:  0.531, Validation Accuracy:  0.566, Loss:  3.126
Epoch   0 Batch  169/538 - Train Accuracy:  0.542, Validation Accuracy:  0.563, Loss:  3.126
Epoch   0 Batch  170/538 - Train Accuracy:  0.568, Validation Accuracy:  0.564, Loss:  3.065
Epoch   0 Batch  171/538 - Train Accuracy:  0.529, Validation Accuracy:  0.561, Loss:  3.124
Epoch   0 Batch  172/538 - Train Accuracy:  0.557, Validation Accuracy:  0.561, Loss:  3.074
Epoch   0 Batch  173/538 - Train Accuracy:  0.552, Validation Accuracy:  0.552, Loss:  3.101
Epoch   0 Batch  174/538 - Train Accuracy:  0.515, Validation Accuracy:  0.561, Loss:  3.083
Epoch   0 Batch  175/538 - Train Accuracy:  0.510, Validation Accuracy:  0.554, Loss:  3.167
Epoch   0 Batch  176/538 - Train Accuracy:  0.532, Validation Accuracy:  0.555, Loss:  3.180
Epoch   0 Batch  177/538 - Train Accuracy:  0.544, Validation Accuracy:  0.552, Loss:  3.119
Epoch   0 Batch  178/538 - Train Accuracy:  0.564, Validation Accuracy:  0.561, Loss:  3.045
Epoch   0 Batch  179/538 - Train Accuracy:  0.555, Validation Accuracy:  0.570, Loss:  3.099
Epoch   0 Batch  180/538 - Train Accuracy:  0.578, Validation Accuracy:  0.566, Loss:  3.080
Epoch   0 Batch  181/538 - Train Accuracy:  0.527, Validation Accuracy:  0.570, Loss:  3.115
Epoch   0 Batch  182/538 - Train Accuracy:  0.530, Validation Accuracy:  0.567, Loss:  3.087
Epoch   0 Batch  183/538 - Train Accuracy:  0.570, Validation Accuracy:  0.566, Loss:  3.035
Epoch   0 Batch  184/538 - Train Accuracy:  0.536, Validation Accuracy:  0.565, Loss:  3.063
Epoch   0 Batch  185/538 - Train Accuracy:  0.519, Validation Accuracy:  0.555, Loss:  3.105
Epoch   0 Batch  186/538 - Train Accuracy:  0.543, Validation Accuracy:  0.555, Loss:  3.128
Epoch   0 Batch  187/538 - Train Accuracy:  0.570, Validation Accuracy:  0.555, Loss:  3.061
Epoch   0 Batch  188/538 - Train Accuracy:  0.518, Validation Accuracy:  0.567, Loss:  3.020
Epoch   0 Batch  189/538 - Train Accuracy:  0.569, Validation Accuracy:  0.571, Loss:  3.098
Epoch   0 Batch  190/538 - Train Accuracy:  0.545, Validation Accuracy:  0.563, Loss:  3.076
Epoch   0 Batch  191/538 - Train Accuracy:  0.560, Validation Accuracy:  0.556, Loss:  3.060
Epoch   0 Batch  192/538 - Train Accuracy:  0.534, Validation Accuracy:  0.560, Loss:  3.077
Epoch   0 Batch  193/538 - Train Accuracy:  0.578, Validation Accuracy:  0.566, Loss:  3.041
Epoch   0 Batch  194/538 - Train Accuracy:  0.535, Validation Accuracy:  0.576, Loss:  3.047
Epoch   0 Batch  195/538 - Train Accuracy:  0.548, Validation Accuracy:  0.577, Loss:  3.041
Epoch   0 Batch  196/538 - Train Accuracy:  0.556, Validation Accuracy:  0.570, Loss:  3.074
Epoch   0 Batch  197/538 - Train Accuracy:  0.575, Validation Accuracy:  0.583, Loss:  2.987
Epoch   0 Batch  198/538 - Train Accuracy:  0.596, Validation Accuracy:  0.574, Loss:  3.098
Epoch   0 Batch  199/538 - Train Accuracy:  0.548, Validation Accuracy:  0.572, Loss:  3.091
Epoch   0 Batch  200/538 - Train Accuracy:  0.570, Validation Accuracy:  0.572, Loss:  3.043
Epoch   0 Batch  201/538 - Train Accuracy:  0.571, Validation Accuracy:  0.578, Loss:  3.119
Epoch   0 Batch  202/538 - Train Accuracy:  0.540, Validation Accuracy:  0.576, Loss:  3.082
Epoch   0 Batch  203/538 - Train Accuracy:  0.542, Validation Accuracy:  0.568, Loss:  3.111
Epoch   0 Batch  204/538 - Train Accuracy:  0.522, Validation Accuracy:  0.570, Loss:  3.082
Epoch   0 Batch  205/538 - Train Accuracy:  0.568, Validation Accuracy:  0.577, Loss:  3.043
Epoch   0 Batch  206/538 - Train Accuracy:  0.507, Validation Accuracy:  0.534, Loss:  3.092
Epoch   0 Batch  207/538 - Train Accuracy:  0.552, Validation Accuracy:  0.562, Loss:  3.110
Epoch   0 Batch  208/538 - Train Accuracy:  0.538, Validation Accuracy:  0.576, Loss:  3.089
Epoch   0 Batch  209/538 - Train Accuracy:  0.519, Validation Accuracy:  0.570, Loss:  3.082
Epoch   0 Batch  210/538 - Train Accuracy:  0.546, Validation Accuracy:  0.571, Loss:  3.172
Epoch   0 Batch  211/538 - Train Accuracy:  0.515, Validation Accuracy:  0.570, Loss:  3.105
Epoch   0 Batch  212/538 - Train Accuracy:  0.578, Validation Accuracy:  0.578, Loss:  3.089
Epoch   0 Batch  213/538 - Train Accuracy:  0.565, Validation Accuracy:  0.570, Loss:  3.066
Epoch   0 Batch  214/538 - Train Accuracy:  0.560, Validation Accuracy:  0.575, Loss:  3.127
Epoch   0 Batch  215/538 - Train Accuracy:  0.522, Validation Accuracy:  0.568, Loss:  3.115
Epoch   0 Batch  216/538 - Train Accuracy:  0.528, Validation Accuracy:  0.561, Loss:  3.072
Epoch   0 Batch  217/538 - Train Accuracy:  0.585, Validation Accuracy:  0.576, Loss:  3.061
Epoch   0 Batch  218/538 - Train Accuracy:  0.539, Validation Accuracy:  0.577, Loss:  3.072
Epoch   0 Batch  219/538 - Train Accuracy:  0.559, Validation Accuracy:  0.579, Loss:  3.052
Epoch   0 Batch  220/538 - Train Accuracy:  0.548, Validation Accuracy:  0.596, Loss:  3.104
Epoch   0 Batch  221/538 - Train Accuracy:  0.571, Validation Accuracy:  0.582, Loss:  3.045
Epoch   0 Batch  222/538 - Train Accuracy:  0.561, Validation Accuracy:  0.559, Loss:  3.059
Epoch   0 Batch  223/538 - Train Accuracy:  0.546, Validation Accuracy:  0.564, Loss:  3.089
Epoch   0 Batch  224/538 - Train Accuracy:  0.516, Validation Accuracy:  0.574, Loss:  3.036
Epoch   0 Batch  225/538 - Train Accuracy:  0.596, Validation Accuracy:  0.571, Loss:  3.021
Epoch   0 Batch  226/538 - Train Accuracy:  0.580, Validation Accuracy:  0.573, Loss:  3.044
Epoch   0 Batch  227/538 - Train Accuracy:  0.582, Validation Accuracy:  0.581, Loss:  2.982
Epoch   0 Batch  228/538 - Train Accuracy:  0.561, Validation Accuracy:  0.594, Loss:  3.030
Epoch   0 Batch  229/538 - Train Accuracy:  0.566, Validation Accuracy:  0.591, Loss:  2.998
Epoch   0 Batch  230/538 - Train Accuracy:  0.544, Validation Accuracy:  0.592, Loss:  2.978
Epoch   0 Batch  231/538 - Train Accuracy:  0.563, Validation Accuracy:  0.587, Loss:  3.032
Epoch   0 Batch  232/538 - Train Accuracy:  0.585, Validation Accuracy:  0.595, Loss:  2.995
Epoch   0 Batch  233/538 - Train Accuracy:  0.609, Validation Accuracy:  0.594, Loss:  3.043
Epoch   0 Batch  234/538 - Train Accuracy:  0.548, Validation Accuracy:  0.597, Loss:  3.140
Epoch   0 Batch  235/538 - Train Accuracy:  0.588, Validation Accuracy:  0.595, Loss:  3.018
Epoch   0 Batch  236/538 - Train Accuracy:  0.573, Validation Accuracy:  0.606, Loss:  3.057
Epoch   0 Batch  237/538 - Train Accuracy:  0.595, Validation Accuracy:  0.601, Loss:  2.988
Epoch   0 Batch  238/538 - Train Accuracy:  0.631, Validation Accuracy:  0.601, Loss:  3.008
Epoch   0 Batch  239/538 - Train Accuracy:  0.594, Validation Accuracy:  0.605, Loss:  3.063
Epoch   0 Batch  240/538 - Train Accuracy:  0.576, Validation Accuracy:  0.598, Loss:  3.063
Epoch   0 Batch  241/538 - Train Accuracy:  0.553, Validation Accuracy:  0.597, Loss:  3.082
Epoch   0 Batch  242/538 - Train Accuracy:  0.588, Validation Accuracy:  0.573, Loss:  3.056
Epoch   0 Batch  243/538 - Train Accuracy:  0.544, Validation Accuracy:  0.597, Loss:  3.067
Epoch   0 Batch  244/538 - Train Accuracy:  0.587, Validation Accuracy:  0.597, Loss:  3.078
Epoch   0 Batch  245/538 - Train Accuracy:  0.585, Validation Accuracy:  0.610, Loss:  3.101
Epoch   0 Batch  246/538 - Train Accuracy:  0.607, Validation Accuracy:  0.610, Loss:  3.022
Epoch   0 Batch  247/538 - Train Accuracy:  0.582, Validation Accuracy:  0.613, Loss:  3.083
Epoch   0 Batch  248/538 - Train Accuracy:  0.586, Validation Accuracy:  0.612, Loss:  3.007
Epoch   0 Batch  249/538 - Train Accuracy:  0.593, Validation Accuracy:  0.615, Loss:  3.004
Epoch   0 Batch  250/538 - Train Accuracy:  0.590, Validation Accuracy:  0.609, Loss:  3.031
Epoch   0 Batch  251/538 - Train Accuracy:  0.574, Validation Accuracy:  0.599, Loss:  3.127
Epoch   0 Batch  252/538 - Train Accuracy:  0.595, Validation Accuracy:  0.607, Loss:  3.099
Epoch   0 Batch  253/538 - Train Accuracy:  0.584, Validation Accuracy:  0.603, Loss:  3.014
Epoch   0 Batch  254/538 - Train Accuracy:  0.580, Validation Accuracy:  0.602, Loss:  3.011
Epoch   0 Batch  255/538 - Train Accuracy:  0.597, Validation Accuracy:  0.601, Loss:  2.989
Epoch   0 Batch  256/538 - Train Accuracy:  0.581, Validation Accuracy:  0.604, Loss:  3.041
Epoch   0 Batch  257/538 - Train Accuracy:  0.610, Validation Accuracy:  0.598, Loss:  2.974
Epoch   0 Batch  258/538 - Train Accuracy:  0.607, Validation Accuracy:  0.607, Loss:  3.095
Epoch   0 Batch  259/538 - Train Accuracy:  0.597, Validation Accuracy:  0.599, Loss:  3.007
Epoch   0 Batch  260/538 - Train Accuracy:  0.578, Validation Accuracy:  0.606, Loss:  3.031
Epoch   0 Batch  261/538 - Train Accuracy:  0.578, Validation Accuracy:  0.606, Loss:  3.019
Epoch   0 Batch  262/538 - Train Accuracy:  0.578, Validation Accuracy:  0.610, Loss:  3.038
Epoch   0 Batch  263/538 - Train Accuracy:  0.594, Validation Accuracy:  0.608, Loss:  3.039
Epoch   0 Batch  264/538 - Train Accuracy:  0.576, Validation Accuracy:  0.595, Loss:  3.060
Epoch   0 Batch  265/538 - Train Accuracy:  0.552, Validation Accuracy:  0.589, Loss:  3.003
Epoch   0 Batch  266/538 - Train Accuracy:  0.619, Validation Accuracy:  0.598, Loss:  3.079
Epoch   0 Batch  267/538 - Train Accuracy:  0.576, Validation Accuracy:  0.600, Loss:  3.097
Epoch   0 Batch  268/538 - Train Accuracy:  0.611, Validation Accuracy:  0.597, Loss:  2.980
Epoch   0 Batch  269/538 - Train Accuracy:  0.585, Validation Accuracy:  0.592, Loss:  3.074
Epoch   0 Batch  270/538 - Train Accuracy:  0.590, Validation Accuracy:  0.600, Loss:  2.982
Epoch   0 Batch  271/538 - Train Accuracy:  0.581, Validation Accuracy:  0.605, Loss:  3.001
Epoch   0 Batch  272/538 - Train Accuracy:  0.561, Validation Accuracy:  0.594, Loss:  3.023
Epoch   0 Batch  273/538 - Train Accuracy:  0.604, Validation Accuracy:  0.604, Loss:  3.061
Epoch   0 Batch  274/538 - Train Accuracy:  0.570, Validation Accuracy:  0.625, Loss:  3.085
Epoch   0 Batch  275/538 - Train Accuracy:  0.584, Validation Accuracy:  0.609, Loss:  3.077
Epoch   0 Batch  276/538 - Train Accuracy:  0.579, Validation Accuracy:  0.592, Loss:  3.038
Epoch   0 Batch  277/538 - Train Accuracy:  0.573, Validation Accuracy:  0.598, Loss:  3.008
Epoch   0 Batch  278/538 - Train Accuracy:  0.607, Validation Accuracy:  0.622, Loss:  2.990
Epoch   0 Batch  279/538 - Train Accuracy:  0.600, Validation Accuracy:  0.630, Loss:  2.989
Epoch   0 Batch  280/538 - Train Accuracy:  0.625, Validation Accuracy:  0.624, Loss:  2.989
Epoch   0 Batch  281/538 - Train Accuracy:  0.594, Validation Accuracy:  0.607, Loss:  2.991
Epoch   0 Batch  282/538 - Train Accuracy:  0.601, Validation Accuracy:  0.604, Loss:  3.028
Epoch   0 Batch  283/538 - Train Accuracy:  0.600, Validation Accuracy:  0.618, Loss:  3.068
Epoch   0 Batch  284/538 - Train Accuracy:  0.620, Validation Accuracy:  0.615, Loss:  2.951
Epoch   0 Batch  285/538 - Train Accuracy:  0.611, Validation Accuracy:  0.607, Loss:  3.014
Epoch   0 Batch  286/538 - Train Accuracy:  0.584, Validation Accuracy:  0.604, Loss:  3.038
Epoch   0 Batch  287/538 - Train Accuracy:  0.622, Validation Accuracy:  0.608, Loss:  3.062
Epoch   0 Batch  288/538 - Train Accuracy:  0.555, Validation Accuracy:  0.606, Loss:  3.091
Epoch   0 Batch  289/538 - Train Accuracy:  0.614, Validation Accuracy:  0.609, Loss:  3.006
Epoch   0 Batch  290/538 - Train Accuracy:  0.584, Validation Accuracy:  0.608, Loss:  3.096
Epoch   0 Batch  291/538 - Train Accuracy:  0.600, Validation Accuracy:  0.615, Loss:  3.026
Epoch   0 Batch  292/538 - Train Accuracy:  0.618, Validation Accuracy:  0.614, Loss:  2.970
Epoch   0 Batch  293/538 - Train Accuracy:  0.598, Validation Accuracy:  0.610, Loss:  2.968
Epoch   0 Batch  294/538 - Train Accuracy:  0.573, Validation Accuracy:  0.606, Loss:  3.048
Epoch   0 Batch  295/538 - Train Accuracy:  0.613, Validation Accuracy:  0.594, Loss:  3.012
Epoch   0 Batch  296/538 - Train Accuracy:  0.592, Validation Accuracy:  0.595, Loss:  3.019
Epoch   0 Batch  297/538 - Train Accuracy:  0.579, Validation Accuracy:  0.608, Loss:  2.975
Epoch   0 Batch  298/538 - Train Accuracy:  0.602, Validation Accuracy:  0.608, Loss:  3.036
Epoch   0 Batch  299/538 - Train Accuracy:  0.612, Validation Accuracy:  0.607, Loss:  2.994
Epoch   0 Batch  300/538 - Train Accuracy:  0.585, Validation Accuracy:  0.607, Loss:  2.964
Epoch   0 Batch  301/538 - Train Accuracy:  0.572, Validation Accuracy:  0.599, Loss:  3.047
Epoch   0 Batch  302/538 - Train Accuracy:  0.624, Validation Accuracy:  0.617, Loss:  2.944
Epoch   0 Batch  303/538 - Train Accuracy:  0.642, Validation Accuracy:  0.613, Loss:  3.055
Epoch   0 Batch  304/538 - Train Accuracy:  0.599, Validation Accuracy:  0.620, Loss:  3.000
Epoch   0 Batch  305/538 - Train Accuracy:  0.594, Validation Accuracy:  0.616, Loss:  2.995
Epoch   0 Batch  306/538 - Train Accuracy:  0.609, Validation Accuracy:  0.619, Loss:  2.987
Epoch   0 Batch  307/538 - Train Accuracy:  0.597, Validation Accuracy:  0.617, Loss:  3.006
Epoch   0 Batch  308/538 - Train Accuracy:  0.617, Validation Accuracy:  0.616, Loss:  2.988
Epoch   0 Batch  309/538 - Train Accuracy:  0.593, Validation Accuracy:  0.616, Loss:  3.015
Epoch   0 Batch  310/538 - Train Accuracy:  0.605, Validation Accuracy:  0.624, Loss:  3.001
Epoch   0 Batch  311/538 - Train Accuracy:  0.586, Validation Accuracy:  0.598, Loss:  3.034
Epoch   0 Batch  312/538 - Train Accuracy:  0.619, Validation Accuracy:  0.599, Loss:  2.901
Epoch   0 Batch  313/538 - Train Accuracy:  0.599, Validation Accuracy:  0.616, Loss:  2.969
Epoch   0 Batch  314/538 - Train Accuracy:  0.609, Validation Accuracy:  0.619, Loss:  3.043
Epoch   0 Batch  315/538 - Train Accuracy:  0.600, Validation Accuracy:  0.622, Loss:  3.008
Epoch   0 Batch  316/538 - Train Accuracy:  0.600, Validation Accuracy:  0.618, Loss:  3.018
Epoch   0 Batch  317/538 - Train Accuracy:  0.607, Validation Accuracy:  0.620, Loss:  2.979
Epoch   0 Batch  318/538 - Train Accuracy:  0.600, Validation Accuracy:  0.623, Loss:  2.996
Epoch   0 Batch  319/538 - Train Accuracy:  0.616, Validation Accuracy:  0.624, Loss:  2.977
Epoch   0 Batch  320/538 - Train Accuracy:  0.606, Validation Accuracy:  0.621, Loss:  2.983
Epoch   0 Batch  321/538 - Train Accuracy:  0.619, Validation Accuracy:  0.621, Loss:  2.986
Epoch   0 Batch  322/538 - Train Accuracy:  0.633, Validation Accuracy:  0.611, Loss:  2.985
Epoch   0 Batch  323/538 - Train Accuracy:  0.620, Validation Accuracy:  0.605, Loss:  2.984
Epoch   0 Batch  324/538 - Train Accuracy:  0.578, Validation Accuracy:  0.623, Loss:  3.020
Epoch   0 Batch  325/538 - Train Accuracy:  0.621, Validation Accuracy:  0.624, Loss:  2.983
Epoch   0 Batch  326/538 - Train Accuracy:  0.608, Validation Accuracy:  0.638, Loss:  2.978
Epoch   0 Batch  327/538 - Train Accuracy:  0.605, Validation Accuracy:  0.635, Loss:  3.022
Epoch   0 Batch  328/538 - Train Accuracy:  0.634, Validation Accuracy:  0.620, Loss:  3.075
Epoch   0 Batch  329/538 - Train Accuracy:  0.626, Validation Accuracy:  0.619, Loss:  2.980
Epoch   0 Batch  330/538 - Train Accuracy:  0.621, Validation Accuracy:  0.625, Loss:  2.975
Epoch   0 Batch  331/538 - Train Accuracy:  0.603, Validation Accuracy:  0.622, Loss:  2.999
Epoch   0 Batch  332/538 - Train Accuracy:  0.597, Validation Accuracy:  0.622, Loss:  2.972
Epoch   0 Batch  333/538 - Train Accuracy:  0.599, Validation Accuracy:  0.624, Loss:  2.988
Epoch   0 Batch  334/538 - Train Accuracy:  0.644, Validation Accuracy:  0.623, Loss:  2.938
Epoch   0 Batch  335/538 - Train Accuracy:  0.617, Validation Accuracy:  0.629, Loss:  2.961
Epoch   0 Batch  336/538 - Train Accuracy:  0.620, Validation Accuracy:  0.624, Loss:  2.967
Epoch   0 Batch  337/538 - Train Accuracy:  0.617, Validation Accuracy:  0.616, Loss:  2.988
Epoch   0 Batch  338/538 - Train Accuracy:  0.586, Validation Accuracy:  0.617, Loss:  2.938
Epoch   0 Batch  339/538 - Train Accuracy:  0.617, Validation Accuracy:  0.618, Loss:  3.019
Epoch   0 Batch  340/538 - Train Accuracy:  0.583, Validation Accuracy:  0.600, Loss:  3.022
Epoch   0 Batch  341/538 - Train Accuracy:  0.609, Validation Accuracy:  0.613, Loss:  2.990
Epoch   0 Batch  342/538 - Train Accuracy:  0.601, Validation Accuracy:  0.609, Loss:  2.951
Epoch   0 Batch  343/538 - Train Accuracy:  0.595, Validation Accuracy:  0.614, Loss:  2.979
Epoch   0 Batch  344/538 - Train Accuracy:  0.598, Validation Accuracy:  0.622, Loss:  3.024
Epoch   0 Batch  345/538 - Train Accuracy:  0.628, Validation Accuracy:  0.634, Loss:  2.985
Epoch   0 Batch  346/538 - Train Accuracy:  0.606, Validation Accuracy:  0.628, Loss:  3.003
Epoch   0 Batch  347/538 - Train Accuracy:  0.602, Validation Accuracy:  0.624, Loss:  2.979
Epoch   0 Batch  348/538 - Train Accuracy:  0.617, Validation Accuracy:  0.624, Loss:  2.931
Epoch   0 Batch  349/538 - Train Accuracy:  0.584, Validation Accuracy:  0.629, Loss:  3.000
Epoch   0 Batch  350/538 - Train Accuracy:  0.609, Validation Accuracy:  0.624, Loss:  2.988
Epoch   0 Batch  351/538 - Train Accuracy:  0.591, Validation Accuracy:  0.616, Loss:  3.044
Epoch   0 Batch  352/538 - Train Accuracy:  0.624, Validation Accuracy:  0.619, Loss:  3.043
Epoch   0 Batch  353/538 - Train Accuracy:  0.620, Validation Accuracy:  0.629, Loss:  3.032
Epoch   0 Batch  354/538 - Train Accuracy:  0.596, Validation Accuracy:  0.627, Loss:  2.996
Epoch   0 Batch  355/538 - Train Accuracy:  0.602, Validation Accuracy:  0.631, Loss:  3.038
Epoch   0 Batch  356/538 - Train Accuracy:  0.632, Validation Accuracy:  0.627, Loss:  2.945
Epoch   0 Batch  357/538 - Train Accuracy:  0.640, Validation Accuracy:  0.639, Loss:  3.020
Epoch   0 Batch  358/538 - Train Accuracy:  0.604, Validation Accuracy:  0.620, Loss:  3.034
Epoch   0 Batch  359/538 - Train Accuracy:  0.636, Validation Accuracy:  0.638, Loss:  3.016
Epoch   0 Batch  360/538 - Train Accuracy:  0.622, Validation Accuracy:  0.637, Loss:  2.980
Epoch   0 Batch  361/538 - Train Accuracy:  0.633, Validation Accuracy:  0.634, Loss:  2.994
Epoch   0 Batch  362/538 - Train Accuracy:  0.642, Validation Accuracy:  0.626, Loss:  3.001
Epoch   0 Batch  363/538 - Train Accuracy:  0.633, Validation Accuracy:  0.627, Loss:  2.954
Epoch   0 Batch  364/538 - Train Accuracy:  0.593, Validation Accuracy:  0.624, Loss:  2.999
Epoch   0 Batch  365/538 - Train Accuracy:  0.608, Validation Accuracy:  0.607, Loss:  2.951
Epoch   0 Batch  366/538 - Train Accuracy:  0.610, Validation Accuracy:  0.612, Loss:  3.022
Epoch   0 Batch  367/538 - Train Accuracy:  0.635, Validation Accuracy:  0.615, Loss:  2.991
Epoch   0 Batch  368/538 - Train Accuracy:  0.675, Validation Accuracy:  0.632, Loss:  2.951
Epoch   0 Batch  369/538 - Train Accuracy:  0.632, Validation Accuracy:  0.627, Loss:  2.930
Epoch   0 Batch  370/538 - Train Accuracy:  0.617, Validation Accuracy:  0.627, Loss:  2.987
Epoch   0 Batch  371/538 - Train Accuracy:  0.624, Validation Accuracy:  0.639, Loss:  2.938
Epoch   0 Batch  372/538 - Train Accuracy:  0.641, Validation Accuracy:  0.639, Loss:  2.961
Epoch   0 Batch  373/538 - Train Accuracy:  0.610, Validation Accuracy:  0.638, Loss:  2.975
Epoch   0 Batch  374/538 - Train Accuracy:  0.605, Validation Accuracy:  0.629, Loss:  2.952
Epoch   0 Batch  375/538 - Train Accuracy:  0.615, Validation Accuracy:  0.619, Loss:  2.885
Epoch   0 Batch  376/538 - Train Accuracy:  0.603, Validation Accuracy:  0.618, Loss:  2.993
Epoch   0 Batch  377/538 - Train Accuracy:  0.564, Validation Accuracy:  0.592, Loss:  2.971
Epoch   0 Batch  378/538 - Train Accuracy:  0.617, Validation Accuracy:  0.617, Loss:  2.924
Epoch   0 Batch  379/538 - Train Accuracy:  0.646, Validation Accuracy:  0.626, Loss:  2.955
Epoch   0 Batch  380/538 - Train Accuracy:  0.621, Validation Accuracy:  0.621, Loss:  2.972
Epoch   0 Batch  381/538 - Train Accuracy:  0.629, Validation Accuracy:  0.635, Loss:  2.973
Epoch   0 Batch  382/538 - Train Accuracy:  0.625, Validation Accuracy:  0.646, Loss:  2.971
Epoch   0 Batch  383/538 - Train Accuracy:  0.615, Validation Accuracy:  0.653, Loss:  2.978
Epoch   0 Batch  384/538 - Train Accuracy:  0.655, Validation Accuracy:  0.648, Loss:  2.937
Epoch   0 Batch  385/538 - Train Accuracy:  0.644, Validation Accuracy:  0.643, Loss:  3.003
Epoch   0 Batch  386/538 - Train Accuracy:  0.613, Validation Accuracy:  0.642, Loss:  2.910
Epoch   0 Batch  387/538 - Train Accuracy:  0.624, Validation Accuracy:  0.646, Loss:  2.978
Epoch   0 Batch  388/538 - Train Accuracy:  0.622, Validation Accuracy:  0.640, Loss:  2.988
Epoch   0 Batch  389/538 - Train Accuracy:  0.613, Validation Accuracy:  0.637, Loss:  3.008
Epoch   0 Batch  390/538 - Train Accuracy:  0.651, Validation Accuracy:  0.639, Loss:  2.951
Epoch   0 Batch  391/538 - Train Accuracy:  0.654, Validation Accuracy:  0.649, Loss:  2.962
Epoch   0 Batch  392/538 - Train Accuracy:  0.628, Validation Accuracy:  0.643, Loss:  2.953
Epoch   0 Batch  393/538 - Train Accuracy:  0.640, Validation Accuracy:  0.641, Loss:  2.959
Epoch   0 Batch  394/538 - Train Accuracy:  0.574, Validation Accuracy:  0.638, Loss:  2.944
Epoch   0 Batch  395/538 - Train Accuracy:  0.612, Validation Accuracy:  0.623, Loss:  2.983
Epoch   0 Batch  396/538 - Train Accuracy:  0.596, Validation Accuracy:  0.613, Loss:  2.958
Epoch   0 Batch  397/538 - Train Accuracy:  0.619, Validation Accuracy:  0.610, Loss:  2.906
Epoch   0 Batch  398/538 - Train Accuracy:  0.633, Validation Accuracy:  0.634, Loss:  3.016
Epoch   0 Batch  399/538 - Train Accuracy:  0.594, Validation Accuracy:  0.641, Loss:  3.019
Epoch   0 Batch  400/538 - Train Accuracy:  0.628, Validation Accuracy:  0.638, Loss:  2.936
Epoch   0 Batch  401/538 - Train Accuracy:  0.619, Validation Accuracy:  0.639, Loss:  3.039
Epoch   0 Batch  402/538 - Train Accuracy:  0.636, Validation Accuracy:  0.633, Loss:  2.961
Epoch   0 Batch  403/538 - Train Accuracy:  0.632, Validation Accuracy:  0.628, Loss:  2.932
Epoch   0 Batch  404/538 - Train Accuracy:  0.627, Validation Accuracy:  0.614, Loss:  2.936
Epoch   0 Batch  405/538 - Train Accuracy:  0.621, Validation Accuracy:  0.621, Loss:  2.984
Epoch   0 Batch  406/538 - Train Accuracy:  0.613, Validation Accuracy:  0.622, Loss:  2.985
Epoch   0 Batch  407/538 - Train Accuracy:  0.644, Validation Accuracy:  0.645, Loss:  2.944
Epoch   0 Batch  408/538 - Train Accuracy:  0.604, Validation Accuracy:  0.649, Loss:  2.986
Epoch   0 Batch  409/538 - Train Accuracy:  0.624, Validation Accuracy:  0.650, Loss:  2.973
Epoch   0 Batch  410/538 - Train Accuracy:  0.637, Validation Accuracy:  0.646, Loss:  2.995
Epoch   0 Batch  411/538 - Train Accuracy:  0.621, Validation Accuracy:  0.639, Loss:  2.976
Epoch   0 Batch  412/538 - Train Accuracy:  0.626, Validation Accuracy:  0.643, Loss:  2.971
Epoch   0 Batch  413/538 - Train Accuracy:  0.647, Validation Accuracy:  0.635, Loss:  2.936
Epoch   0 Batch  414/538 - Train Accuracy:  0.620, Validation Accuracy:  0.640, Loss:  2.955
Epoch   0 Batch  415/538 - Train Accuracy:  0.589, Validation Accuracy:  0.632, Loss:  2.982
Epoch   0 Batch  416/538 - Train Accuracy:  0.659, Validation Accuracy:  0.637, Loss:  2.934
Epoch   0 Batch  417/538 - Train Accuracy:  0.621, Validation Accuracy:  0.641, Loss:  2.925
Epoch   0 Batch  418/538 - Train Accuracy:  0.609, Validation Accuracy:  0.632, Loss:  2.985
Epoch   0 Batch  419/538 - Train Accuracy:  0.631, Validation Accuracy:  0.648, Loss:  2.958
Epoch   0 Batch  420/538 - Train Accuracy:  0.646, Validation Accuracy:  0.653, Loss:  2.942
Epoch   0 Batch  421/538 - Train Accuracy:  0.651, Validation Accuracy:  0.645, Loss:  2.882
Epoch   0 Batch  422/538 - Train Accuracy:  0.635, Validation Accuracy:  0.653, Loss:  2.923
Epoch   0 Batch  423/538 - Train Accuracy:  0.647, Validation Accuracy:  0.652, Loss:  2.925
Epoch   0 Batch  424/538 - Train Accuracy:  0.629, Validation Accuracy:  0.648, Loss:  2.957
Epoch   0 Batch  425/538 - Train Accuracy:  0.641, Validation Accuracy:  0.645, Loss:  3.005
Epoch   0 Batch  426/538 - Train Accuracy:  0.658, Validation Accuracy:  0.651, Loss:  2.915
Epoch   0 Batch  427/538 - Train Accuracy:  0.618, Validation Accuracy:  0.650, Loss:  2.966
Epoch   0 Batch  428/538 - Train Accuracy:  0.667, Validation Accuracy:  0.647, Loss:  2.959
Epoch   0 Batch  429/538 - Train Accuracy:  0.641, Validation Accuracy:  0.636, Loss:  2.919
Epoch   0 Batch  430/538 - Train Accuracy:  0.649, Validation Accuracy:  0.644, Loss:  2.936
Epoch   0 Batch  431/538 - Train Accuracy:  0.665, Validation Accuracy:  0.665, Loss:  2.898
Epoch   0 Batch  432/538 - Train Accuracy:  0.675, Validation Accuracy:  0.655, Loss:  2.939
Epoch   0 Batch  433/538 - Train Accuracy:  0.602, Validation Accuracy:  0.656, Loss:  2.979
Epoch   0 Batch  434/538 - Train Accuracy:  0.622, Validation Accuracy:  0.658, Loss:  2.979
Epoch   0 Batch  435/538 - Train Accuracy:  0.614, Validation Accuracy:  0.654, Loss:  3.016
Epoch   0 Batch  436/538 - Train Accuracy:  0.606, Validation Accuracy:  0.643, Loss:  3.005
Epoch   0 Batch  437/538 - Train Accuracy:  0.608, Validation Accuracy:  0.648, Loss:  2.971
Epoch   0 Batch  438/538 - Train Accuracy:  0.657, Validation Accuracy:  0.657, Loss:  3.018
Epoch   0 Batch  439/538 - Train Accuracy:  0.646, Validation Accuracy:  0.656, Loss:  2.893
Epoch   0 Batch  440/538 - Train Accuracy:  0.614, Validation Accuracy:  0.655, Loss:  2.932
Epoch   0 Batch  441/538 - Train Accuracy:  0.607, Validation Accuracy:  0.665, Loss:  2.962
Epoch   0 Batch  442/538 - Train Accuracy:  0.658, Validation Accuracy:  0.666, Loss:  2.898
Epoch   0 Batch  443/538 - Train Accuracy:  0.661, Validation Accuracy:  0.663, Loss:  2.946
Epoch   0 Batch  444/538 - Train Accuracy:  0.686, Validation Accuracy:  0.656, Loss:  2.849
Epoch   0 Batch  445/538 - Train Accuracy:  0.660, Validation Accuracy:  0.659, Loss:  2.939
Epoch   0 Batch  446/538 - Train Accuracy:  0.658, Validation Accuracy:  0.653, Loss:  2.924
Epoch   0 Batch  447/538 - Train Accuracy:  0.623, Validation Accuracy:  0.657, Loss:  2.992
Epoch   0 Batch  448/538 - Train Accuracy:  0.648, Validation Accuracy:  0.644, Loss:  2.906
Epoch   0 Batch  449/538 - Train Accuracy:  0.626, Validation Accuracy:  0.651, Loss:  2.949
Epoch   0 Batch  450/538 - Train Accuracy:  0.673, Validation Accuracy:  0.654, Loss:  2.905
Epoch   0 Batch  451/538 - Train Accuracy:  0.639, Validation Accuracy:  0.657, Loss:  2.930
Epoch   0 Batch  452/538 - Train Accuracy:  0.657, Validation Accuracy:  0.659, Loss:  2.952
Epoch   0 Batch  453/538 - Train Accuracy:  0.625, Validation Accuracy:  0.641, Loss:  2.947
Epoch   0 Batch  454/538 - Train Accuracy:  0.640, Validation Accuracy:  0.645, Loss:  2.951
Epoch   0 Batch  455/538 - Train Accuracy:  0.653, Validation Accuracy:  0.653, Loss:  2.942
Epoch   0 Batch  456/538 - Train Accuracy:  0.708, Validation Accuracy:  0.657, Loss:  2.928
Epoch   0 Batch  457/538 - Train Accuracy:  0.654, Validation Accuracy:  0.663, Loss:  2.933
Epoch   0 Batch  458/538 - Train Accuracy:  0.653, Validation Accuracy:  0.656, Loss:  2.897
Epoch   0 Batch  459/538 - Train Accuracy:  0.669, Validation Accuracy:  0.656, Loss:  2.904
Epoch   0 Batch  460/538 - Train Accuracy:  0.632, Validation Accuracy:  0.660, Loss:  2.957
Epoch   0 Batch  461/538 - Train Accuracy:  0.609, Validation Accuracy:  0.648, Loss:  2.943
Epoch   0 Batch  462/538 - Train Accuracy:  0.617, Validation Accuracy:  0.629, Loss:  2.952
Epoch   0 Batch  463/538 - Train Accuracy:  0.612, Validation Accuracy:  0.648, Loss:  2.966
Epoch   0 Batch  464/538 - Train Accuracy:  0.636, Validation Accuracy:  0.651, Loss:  2.957
Epoch   0 Batch  465/538 - Train Accuracy:  0.611, Validation Accuracy:  0.651, Loss:  2.932
Epoch   0 Batch  466/538 - Train Accuracy:  0.624, Validation Accuracy:  0.647, Loss:  2.911
Epoch   0 Batch  467/538 - Train Accuracy:  0.636, Validation Accuracy:  0.648, Loss:  2.914
Epoch   0 Batch  468/538 - Train Accuracy:  0.674, Validation Accuracy:  0.658, Loss:  2.951
Epoch   0 Batch  469/538 - Train Accuracy:  0.633, Validation Accuracy:  0.656, Loss:  2.899
Epoch   0 Batch  470/538 - Train Accuracy:  0.645, Validation Accuracy:  0.651, Loss:  2.908
Epoch   0 Batch  471/538 - Train Accuracy:  0.639, Validation Accuracy:  0.653, Loss:  2.918
Epoch   0 Batch  472/538 - Train Accuracy:  0.666, Validation Accuracy:  0.658, Loss:  2.953
Epoch   0 Batch  473/538 - Train Accuracy:  0.640, Validation Accuracy:  0.658, Loss:  2.946
Epoch   0 Batch  474/538 - Train Accuracy:  0.655, Validation Accuracy:  0.663, Loss:  2.955
Epoch   0 Batch  475/538 - Train Accuracy:  0.653, Validation Accuracy:  0.666, Loss:  2.878
Epoch   0 Batch  476/538 - Train Accuracy:  0.626, Validation Accuracy:  0.664, Loss:  3.007
Epoch   0 Batch  477/538 - Train Accuracy:  0.661, Validation Accuracy:  0.663, Loss:  2.943
Epoch   0 Batch  478/538 - Train Accuracy:  0.672, Validation Accuracy:  0.670, Loss:  2.914
Epoch   0 Batch  479/538 - Train Accuracy:  0.679, Validation Accuracy:  0.672, Loss:  2.926
Epoch   0 Batch  480/538 - Train Accuracy:  0.626, Validation Accuracy:  0.634, Loss:  2.965
Epoch   0 Batch  481/538 - Train Accuracy:  0.677, Validation Accuracy:  0.652, Loss:  2.927
Epoch   0 Batch  482/538 - Train Accuracy:  0.677, Validation Accuracy:  0.667, Loss:  2.902
Epoch   0 Batch  483/538 - Train Accuracy:  0.619, Validation Accuracy:  0.655, Loss:  2.962
Epoch   0 Batch  484/538 - Train Accuracy:  0.674, Validation Accuracy:  0.654, Loss:  2.909
Epoch   0 Batch  485/538 - Train Accuracy:  0.625, Validation Accuracy:  0.639, Loss:  2.899
Epoch   0 Batch  486/538 - Train Accuracy:  0.652, Validation Accuracy:  0.647, Loss:  2.880
Epoch   0 Batch  487/538 - Train Accuracy:  0.665, Validation Accuracy:  0.650, Loss:  2.933
Epoch   0 Batch  488/538 - Train Accuracy:  0.677, Validation Accuracy:  0.659, Loss:  2.946
Epoch   0 Batch  489/538 - Train Accuracy:  0.647, Validation Accuracy:  0.665, Loss:  2.885
Epoch   0 Batch  490/538 - Train Accuracy:  0.643, Validation Accuracy:  0.659, Loss:  2.872
Epoch   0 Batch  491/538 - Train Accuracy:  0.630, Validation Accuracy:  0.654, Loss:  2.931
Epoch   0 Batch  492/538 - Train Accuracy:  0.644, Validation Accuracy:  0.664, Loss:  2.963
Epoch   0 Batch  493/538 - Train Accuracy:  0.565, Validation Accuracy:  0.617, Loss:  2.964
Epoch   0 Batch  494/538 - Train Accuracy:  0.647, Validation Accuracy:  0.661, Loss:  2.939
Epoch   0 Batch  495/538 - Train Accuracy:  0.660, Validation Accuracy:  0.663, Loss:  2.867
Epoch   0 Batch  496/538 - Train Accuracy:  0.596, Validation Accuracy:  0.593, Loss:  2.999
Epoch   0 Batch  497/538 - Train Accuracy:  0.619, Validation Accuracy:  0.626, Loss:  2.889
Epoch   0 Batch  498/538 - Train Accuracy:  0.618, Validation Accuracy:  0.663, Loss:  2.894
Epoch   0 Batch  499/538 - Train Accuracy:  0.665, Validation Accuracy:  0.663, Loss:  2.907
Epoch   0 Batch  500/538 - Train Accuracy:  0.694, Validation Accuracy:  0.654, Loss:  2.856
Epoch   0 Batch  501/538 - Train Accuracy:  0.677, Validation Accuracy:  0.659, Loss:  2.917
Epoch   0 Batch  502/538 - Train Accuracy:  0.633, Validation Accuracy:  0.655, Loss:  2.910
Epoch   0 Batch  503/538 - Train Accuracy:  0.692, Validation Accuracy:  0.654, Loss:  2.934
Epoch   0 Batch  504/538 - Train Accuracy:  0.662, Validation Accuracy:  0.665, Loss:  2.840
Epoch   0 Batch  505/538 - Train Accuracy:  0.645, Validation Accuracy:  0.642, Loss:  2.831
Epoch   0 Batch  506/538 - Train Accuracy:  0.645, Validation Accuracy:  0.660, Loss:  2.916
Epoch   0 Batch  507/538 - Train Accuracy:  0.634, Validation Accuracy:  0.660, Loss:  2.928
Epoch   0 Batch  508/538 - Train Accuracy:  0.655, Validation Accuracy:  0.652, Loss:  2.874
Epoch   0 Batch  509/538 - Train Accuracy:  0.637, Validation Accuracy:  0.667, Loss:  2.938
Epoch   0 Batch  510/538 - Train Accuracy:  0.675, Validation Accuracy:  0.666, Loss:  2.870
Epoch   0 Batch  511/538 - Train Accuracy:  0.680, Validation Accuracy:  0.667, Loss:  2.855
Epoch   0 Batch  512/538 - Train Accuracy:  0.682, Validation Accuracy:  0.653, Loss:  2.847
Epoch   0 Batch  513/538 - Train Accuracy:  0.624, Validation Accuracy:  0.625, Loss:  2.887
Epoch   0 Batch  514/538 - Train Accuracy:  0.648, Validation Accuracy:  0.635, Loss:  2.925
Epoch   0 Batch  515/538 - Train Accuracy:  0.652, Validation Accuracy:  0.632, Loss:  2.904
Epoch   0 Batch  516/538 - Train Accuracy:  0.631, Validation Accuracy:  0.643, Loss:  2.927
Epoch   0 Batch  517/538 - Train Accuracy:  0.674, Validation Accuracy:  0.641, Loss:  2.943
Epoch   0 Batch  518/538 - Train Accuracy:  0.636, Validation Accuracy:  0.650, Loss:  2.900
Epoch   0 Batch  519/538 - Train Accuracy:  0.695, Validation Accuracy:  0.670, Loss:  2.874
Epoch   0 Batch  520/538 - Train Accuracy:  0.663, Validation Accuracy:  0.664, Loss:  2.894
Epoch   0 Batch  521/538 - Train Accuracy:  0.613, Validation Accuracy:  0.645, Loss:  2.923
Epoch   0 Batch  522/538 - Train Accuracy:  0.626, Validation Accuracy:  0.662, Loss:  2.916
Epoch   0 Batch  523/538 - Train Accuracy:  0.648, Validation Accuracy:  0.665, Loss:  2.963
Epoch   0 Batch  524/538 - Train Accuracy:  0.578, Validation Accuracy:  0.618, Loss:  2.923
Epoch   0 Batch  525/538 - Train Accuracy:  0.648, Validation Accuracy:  0.638, Loss:  2.942
Epoch   0 Batch  526/538 - Train Accuracy:  0.681, Validation Accuracy:  0.670, Loss:  2.850
Epoch   0 Batch  527/538 - Train Accuracy:  0.684, Validation Accuracy:  0.670, Loss:  2.863
Epoch   0 Batch  528/538 - Train Accuracy:  0.660, Validation Accuracy:  0.664, Loss:  2.950
Epoch   0 Batch  529/538 - Train Accuracy:  0.643, Validation Accuracy:  0.653, Loss:  2.869
Epoch   0 Batch  530/538 - Train Accuracy:  0.625, Validation Accuracy:  0.625, Loss:  2.943
Epoch   0 Batch  531/538 - Train Accuracy:  0.602, Validation Accuracy:  0.635, Loss:  2.919
Epoch   0 Batch  532/538 - Train Accuracy:  0.644, Validation Accuracy:  0.640, Loss:  2.902
Epoch   0 Batch  533/538 - Train Accuracy:  0.682, Validation Accuracy:  0.661, Loss:  2.936
Epoch   0 Batch  534/538 - Train Accuracy:  0.664, Validation Accuracy:  0.659, Loss:  2.839
Epoch   0 Batch  535/538 - Train Accuracy:  0.676, Validation Accuracy:  0.667, Loss:  2.867
Epoch   0 Batch  536/538 - Train Accuracy:  0.668, Validation Accuracy:  0.680, Loss:  2.924
Epoch   1 Batch    0/538 - Train Accuracy:  0.644, Validation Accuracy:  0.675, Loss:  2.985
Epoch   1 Batch    1/538 - Train Accuracy:  0.674, Validation Accuracy:  0.665, Loss:  2.874
Epoch   1 Batch    2/538 - Train Accuracy:  0.656, Validation Accuracy:  0.669, Loss:  2.888
Epoch   1 Batch    3/538 - Train Accuracy:  0.635, Validation Accuracy:  0.669, Loss:  2.896
Epoch   1 Batch    4/538 - Train Accuracy:  0.699, Validation Accuracy:  0.668, Loss:  2.884
Epoch   1 Batch    5/538 - Train Accuracy:  0.672, Validation Accuracy:  0.662, Loss:  2.858
Epoch   1 Batch    6/538 - Train Accuracy:  0.691, Validation Accuracy:  0.659, Loss:  2.837
Epoch   1 Batch    7/538 - Train Accuracy:  0.663, Validation Accuracy:  0.664, Loss:  2.864
Epoch   1 Batch    8/538 - Train Accuracy:  0.696, Validation Accuracy:  0.660, Loss:  2.887
Epoch   1 Batch    9/538 - Train Accuracy:  0.673, Validation Accuracy:  0.669, Loss:  2.864
Epoch   1 Batch   10/538 - Train Accuracy:  0.634, Validation Accuracy:  0.672, Loss:  2.930
Epoch   1 Batch   11/538 - Train Accuracy:  0.687, Validation Accuracy:  0.688, Loss:  2.866
Epoch   1 Batch   12/538 - Train Accuracy:  0.660, Validation Accuracy:  0.675, Loss:  2.909
Epoch   1 Batch   13/538 - Train Accuracy:  0.705, Validation Accuracy:  0.665, Loss:  2.880
Epoch   1 Batch   14/538 - Train Accuracy:  0.660, Validation Accuracy:  0.671, Loss:  2.848
Epoch   1 Batch   15/538 - Train Accuracy:  0.703, Validation Accuracy:  0.664, Loss:  2.819
Epoch   1 Batch   16/538 - Train Accuracy:  0.686, Validation Accuracy:  0.660, Loss:  2.872
Epoch   1 Batch   17/538 - Train Accuracy:  0.679, Validation Accuracy:  0.665, Loss:  2.861
Epoch   1 Batch   18/538 - Train Accuracy:  0.650, Validation Accuracy:  0.673, Loss:  2.913
Epoch   1 Batch   19/538 - Train Accuracy:  0.651, Validation Accuracy:  0.681, Loss:  2.975
Epoch   1 Batch   20/538 - Train Accuracy:  0.691, Validation Accuracy:  0.695, Loss:  2.895
Epoch   1 Batch   21/538 - Train Accuracy:  0.693, Validation Accuracy:  0.698, Loss:  2.900
Epoch   1 Batch   22/538 - Train Accuracy:  0.671, Validation Accuracy:  0.710, Loss:  2.851
Epoch   1 Batch   23/538 - Train Accuracy:  0.706, Validation Accuracy:  0.688, Loss:  2.906
Epoch   1 Batch   24/538 - Train Accuracy:  0.672, Validation Accuracy:  0.684, Loss:  2.878
Epoch   1 Batch   25/538 - Train Accuracy:  0.684, Validation Accuracy:  0.688, Loss:  2.884
Epoch   1 Batch   26/538 - Train Accuracy:  0.677, Validation Accuracy:  0.695, Loss:  2.907
Epoch   1 Batch   27/538 - Train Accuracy:  0.692, Validation Accuracy:  0.689, Loss:  2.883
Epoch   1 Batch   28/538 - Train Accuracy:  0.695, Validation Accuracy:  0.667, Loss:  2.881
Epoch   1 Batch   29/538 - Train Accuracy:  0.678, Validation Accuracy:  0.693, Loss:  2.954
Epoch   1 Batch   30/538 - Train Accuracy:  0.658, Validation Accuracy:  0.674, Loss:  2.923
Epoch   1 Batch   31/538 - Train Accuracy:  0.685, Validation Accuracy:  0.686, Loss:  2.819
Epoch   1 Batch   32/538 - Train Accuracy:  0.687, Validation Accuracy:  0.683, Loss:  2.876
Epoch   1 Batch   33/538 - Train Accuracy:  0.714, Validation Accuracy:  0.691, Loss:  2.861
Epoch   1 Batch   34/538 - Train Accuracy:  0.706, Validation Accuracy:  0.688, Loss:  2.903
Epoch   1 Batch   35/538 - Train Accuracy:  0.709, Validation Accuracy:  0.686, Loss:  2.835
Epoch   1 Batch   36/538 - Train Accuracy:  0.708, Validation Accuracy:  0.695, Loss:  2.860
Epoch   1 Batch   37/538 - Train Accuracy:  0.718, Validation Accuracy:  0.693, Loss:  2.907
Epoch   1 Batch   38/538 - Train Accuracy:  0.677, Validation Accuracy:  0.677, Loss:  2.826
Epoch   1 Batch   39/538 - Train Accuracy:  0.683, Validation Accuracy:  0.676, Loss:  2.867
Epoch   1 Batch   40/538 - Train Accuracy:  0.714, Validation Accuracy:  0.698, Loss:  2.864
Epoch   1 Batch   41/538 - Train Accuracy:  0.679, Validation Accuracy:  0.698, Loss:  2.816
Epoch   1 Batch   42/538 - Train Accuracy:  0.721, Validation Accuracy:  0.698, Loss:  2.854
Epoch   1 Batch   43/538 - Train Accuracy:  0.697, Validation Accuracy:  0.694, Loss:  2.962
Epoch   1 Batch   44/538 - Train Accuracy:  0.684, Validation Accuracy:  0.699, Loss:  2.877
Epoch   1 Batch   45/538 - Train Accuracy:  0.716, Validation Accuracy:  0.699, Loss:  2.857
Epoch   1 Batch   46/538 - Train Accuracy:  0.723, Validation Accuracy:  0.722, Loss:  2.873
Epoch   1 Batch   47/538 - Train Accuracy:  0.701, Validation Accuracy:  0.720, Loss:  2.889
Epoch   1 Batch   48/538 - Train Accuracy:  0.730, Validation Accuracy:  0.708, Loss:  2.861
Epoch   1 Batch   49/538 - Train Accuracy:  0.706, Validation Accuracy:  0.709, Loss:  2.821
Epoch   1 Batch   50/538 - Train Accuracy:  0.708, Validation Accuracy:  0.710, Loss:  2.867
Epoch   1 Batch   51/538 - Train Accuracy:  0.674, Validation Accuracy:  0.710, Loss:  2.868
Epoch   1 Batch   52/538 - Train Accuracy:  0.705, Validation Accuracy:  0.702, Loss:  2.849
Epoch   1 Batch   53/538 - Train Accuracy:  0.714, Validation Accuracy:  0.697, Loss:  2.823
Epoch   1 Batch   54/538 - Train Accuracy:  0.722, Validation Accuracy:  0.697, Loss:  2.846
Epoch   1 Batch   55/538 - Train Accuracy:  0.708, Validation Accuracy:  0.702, Loss:  2.818
Epoch   1 Batch   56/538 - Train Accuracy:  0.714, Validation Accuracy:  0.700, Loss:  2.920
Epoch   1 Batch   57/538 - Train Accuracy:  0.673, Validation Accuracy:  0.717, Loss:  2.897
Epoch   1 Batch   58/538 - Train Accuracy:  0.700, Validation Accuracy:  0.722, Loss:  2.855
Epoch   1 Batch   59/538 - Train Accuracy:  0.672, Validation Accuracy:  0.688, Loss:  2.918
Epoch   1 Batch   60/538 - Train Accuracy:  0.742, Validation Accuracy:  0.713, Loss:  2.885
Epoch   1 Batch   61/538 - Train Accuracy:  0.727, Validation Accuracy:  0.723, Loss:  2.886
Epoch   1 Batch   62/538 - Train Accuracy:  0.718, Validation Accuracy:  0.708, Loss:  2.825
Epoch   1 Batch   63/538 - Train Accuracy:  0.719, Validation Accuracy:  0.700, Loss:  2.869
Epoch   1 Batch   64/538 - Train Accuracy:  0.722, Validation Accuracy:  0.702, Loss:  2.833
Epoch   1 Batch   65/538 - Train Accuracy:  0.706, Validation Accuracy:  0.699, Loss:  2.855
Epoch   1 Batch   66/538 - Train Accuracy:  0.724, Validation Accuracy:  0.697, Loss:  2.851
Epoch   1 Batch   67/538 - Train Accuracy:  0.730, Validation Accuracy:  0.708, Loss:  2.807
Epoch   1 Batch   68/538 - Train Accuracy:  0.722, Validation Accuracy:  0.706, Loss:  2.849
Epoch   1 Batch   69/538 - Train Accuracy:  0.700, Validation Accuracy:  0.708, Loss:  2.920
Epoch   1 Batch   70/538 - Train Accuracy:  0.727, Validation Accuracy:  0.717, Loss:  2.861
Epoch   1 Batch   71/538 - Train Accuracy:  0.673, Validation Accuracy:  0.706, Loss:  2.821
Epoch   1 Batch   72/538 - Train Accuracy:  0.738, Validation Accuracy:  0.699, Loss:  2.849
Epoch   1 Batch   73/538 - Train Accuracy:  0.679, Validation Accuracy:  0.728, Loss:  2.844
Epoch   1 Batch   74/538 - Train Accuracy:  0.746, Validation Accuracy:  0.736, Loss:  2.797
Epoch   1 Batch   75/538 - Train Accuracy:  0.741, Validation Accuracy:  0.718, Loss:  2.760
Epoch   1 Batch   76/538 - Train Accuracy:  0.718, Validation Accuracy:  0.719, Loss:  2.876
Epoch   1 Batch   77/538 - Train Accuracy:  0.717, Validation Accuracy:  0.721, Loss:  2.865
Epoch   1 Batch   78/538 - Train Accuracy:  0.716, Validation Accuracy:  0.708, Loss:  2.908
Epoch   1 Batch   79/538 - Train Accuracy:  0.727, Validation Accuracy:  0.712, Loss:  2.805
Epoch   1 Batch   80/538 - Train Accuracy:  0.716, Validation Accuracy:  0.713, Loss:  2.811
Epoch   1 Batch   81/538 - Train Accuracy:  0.711, Validation Accuracy:  0.713, Loss:  2.815
Epoch   1 Batch   82/538 - Train Accuracy:  0.709, Validation Accuracy:  0.719, Loss:  2.863
Epoch   1 Batch   83/538 - Train Accuracy:  0.719, Validation Accuracy:  0.736, Loss:  2.904
Epoch   1 Batch   84/538 - Train Accuracy:  0.685, Validation Accuracy:  0.709, Loss:  2.848
Epoch   1 Batch   85/538 - Train Accuracy:  0.736, Validation Accuracy:  0.715, Loss:  2.830
Epoch   1 Batch   86/538 - Train Accuracy:  0.728, Validation Accuracy:  0.719, Loss:  2.883
Epoch   1 Batch   87/538 - Train Accuracy:  0.741, Validation Accuracy:  0.724, Loss:  2.844
Epoch   1 Batch   88/538 - Train Accuracy:  0.733, Validation Accuracy:  0.717, Loss:  2.878
Epoch   1 Batch   89/538 - Train Accuracy:  0.753, Validation Accuracy:  0.738, Loss:  2.817
Epoch   1 Batch   90/538 - Train Accuracy:  0.723, Validation Accuracy:  0.716, Loss:  2.824
Epoch   1 Batch   91/538 - Train Accuracy:  0.703, Validation Accuracy:  0.722, Loss:  2.868
Epoch   1 Batch   92/538 - Train Accuracy:  0.724, Validation Accuracy:  0.726, Loss:  2.807
Epoch   1 Batch   93/538 - Train Accuracy:  0.702, Validation Accuracy:  0.735, Loss:  2.840
Epoch   1 Batch   94/538 - Train Accuracy:  0.736, Validation Accuracy:  0.732, Loss:  2.852
Epoch   1 Batch   95/538 - Train Accuracy:  0.769, Validation Accuracy:  0.756, Loss:  2.838
Epoch   1 Batch   96/538 - Train Accuracy:  0.763, Validation Accuracy:  0.751, Loss:  2.835
Epoch   1 Batch   97/538 - Train Accuracy:  0.733, Validation Accuracy:  0.742, Loss:  2.896
Epoch   1 Batch   98/538 - Train Accuracy:  0.760, Validation Accuracy:  0.728, Loss:  2.842
Epoch   1 Batch   99/538 - Train Accuracy:  0.719, Validation Accuracy:  0.743, Loss:  2.795
Epoch   1 Batch  100/538 - Train Accuracy:  0.741, Validation Accuracy:  0.741, Loss:  2.802
Epoch   1 Batch  101/538 - Train Accuracy:  0.692, Validation Accuracy:  0.725, Loss:  2.801
Epoch   1 Batch  102/538 - Train Accuracy:  0.731, Validation Accuracy:  0.740, Loss:  2.786
Epoch   1 Batch  103/538 - Train Accuracy:  0.734, Validation Accuracy:  0.726, Loss:  2.828
Epoch   1 Batch  104/538 - Train Accuracy:  0.729, Validation Accuracy:  0.722, Loss:  2.769
Epoch   1 Batch  105/538 - Train Accuracy:  0.744, Validation Accuracy:  0.716, Loss:  2.898
Epoch   1 Batch  106/538 - Train Accuracy:  0.725, Validation Accuracy:  0.727, Loss:  2.810
Epoch   1 Batch  107/538 - Train Accuracy:  0.709, Validation Accuracy:  0.716, Loss:  2.806
Epoch   1 Batch  108/538 - Train Accuracy:  0.745, Validation Accuracy:  0.724, Loss:  2.782
Epoch   1 Batch  109/538 - Train Accuracy:  0.758, Validation Accuracy:  0.741, Loss:  2.757
Epoch   1 Batch  110/538 - Train Accuracy:  0.751, Validation Accuracy:  0.734, Loss:  2.789
Epoch   1 Batch  111/538 - Train Accuracy:  0.756, Validation Accuracy:  0.742, Loss:  2.816
Epoch   1 Batch  112/538 - Train Accuracy:  0.748, Validation Accuracy:  0.756, Loss:  2.825
Epoch   1 Batch  113/538 - Train Accuracy:  0.765, Validation Accuracy:  0.759, Loss:  2.854
Epoch   1 Batch  114/538 - Train Accuracy:  0.781, Validation Accuracy:  0.762, Loss:  2.824
Epoch   1 Batch  115/538 - Train Accuracy:  0.754, Validation Accuracy:  0.747, Loss:  2.849
Epoch   1 Batch  116/538 - Train Accuracy:  0.749, Validation Accuracy:  0.759, Loss:  2.831
Epoch   1 Batch  117/538 - Train Accuracy:  0.748, Validation Accuracy:  0.735, Loss:  2.739
Epoch   1 Batch  118/538 - Train Accuracy:  0.764, Validation Accuracy:  0.725, Loss:  2.809
Epoch   1 Batch  119/538 - Train Accuracy:  0.776, Validation Accuracy:  0.724, Loss:  2.769
Epoch   1 Batch  120/538 - Train Accuracy:  0.753, Validation Accuracy:  0.732, Loss:  2.810
Epoch   1 Batch  121/538 - Train Accuracy:  0.741, Validation Accuracy:  0.742, Loss:  2.835
Epoch   1 Batch  122/538 - Train Accuracy:  0.750, Validation Accuracy:  0.728, Loss:  2.814
Epoch   1 Batch  123/538 - Train Accuracy:  0.750, Validation Accuracy:  0.736, Loss:  2.822
Epoch   1 Batch  124/538 - Train Accuracy:  0.759, Validation Accuracy:  0.753, Loss:  2.782
Epoch   1 Batch  125/538 - Train Accuracy:  0.770, Validation Accuracy:  0.759, Loss:  2.849
Epoch   1 Batch  126/538 - Train Accuracy:  0.771, Validation Accuracy:  0.760, Loss:  2.786
Epoch   1 Batch  127/538 - Train Accuracy:  0.751, Validation Accuracy:  0.755, Loss:  2.849
Epoch   1 Batch  128/538 - Train Accuracy:  0.760, Validation Accuracy:  0.757, Loss:  2.809
Epoch   1 Batch  129/538 - Train Accuracy:  0.738, Validation Accuracy:  0.731, Loss:  2.789
Epoch   1 Batch  130/538 - Train Accuracy:  0.755, Validation Accuracy:  0.740, Loss:  2.834
Epoch   1 Batch  131/538 - Train Accuracy:  0.733, Validation Accuracy:  0.741, Loss:  2.787
Epoch   1 Batch  132/538 - Train Accuracy:  0.718, Validation Accuracy:  0.747, Loss:  2.791
Epoch   1 Batch  133/538 - Train Accuracy:  0.760, Validation Accuracy:  0.743, Loss:  2.805
Epoch   1 Batch  134/538 - Train Accuracy:  0.714, Validation Accuracy:  0.752, Loss:  2.801
Epoch   1 Batch  135/538 - Train Accuracy:  0.769, Validation Accuracy:  0.763, Loss:  2.855
Epoch   1 Batch  136/538 - Train Accuracy:  0.751, Validation Accuracy:  0.763, Loss:  2.824
Epoch   1 Batch  137/538 - Train Accuracy:  0.752, Validation Accuracy:  0.774, Loss:  2.782
Epoch   1 Batch  138/538 - Train Accuracy:  0.768, Validation Accuracy:  0.778, Loss:  2.793
Epoch   1 Batch  139/538 - Train Accuracy:  0.723, Validation Accuracy:  0.783, Loss:  2.769
Epoch   1 Batch  140/538 - Train Accuracy:  0.746, Validation Accuracy:  0.784, Loss:  2.820
Epoch   1 Batch  141/538 - Train Accuracy:  0.780, Validation Accuracy:  0.783, Loss:  2.767
Epoch   1 Batch  142/538 - Train Accuracy:  0.800, Validation Accuracy:  0.779, Loss:  2.756
Epoch   1 Batch  143/538 - Train Accuracy:  0.748, Validation Accuracy:  0.763, Loss:  2.819
Epoch   1 Batch  144/538 - Train Accuracy:  0.767, Validation Accuracy:  0.763, Loss:  2.757
Epoch   1 Batch  145/538 - Train Accuracy:  0.749, Validation Accuracy:  0.757, Loss:  2.840
Epoch   1 Batch  146/538 - Train Accuracy:  0.766, Validation Accuracy:  0.765, Loss:  2.819
Epoch   1 Batch  147/538 - Train Accuracy:  0.768, Validation Accuracy:  0.769, Loss:  2.779
Epoch   1 Batch  148/538 - Train Accuracy:  0.755, Validation Accuracy:  0.782, Loss:  2.832
Epoch   1 Batch  149/538 - Train Accuracy:  0.771, Validation Accuracy:  0.773, Loss:  2.731
Epoch   1 Batch  150/538 - Train Accuracy:  0.780, Validation Accuracy:  0.794, Loss:  2.778
Epoch   1 Batch  151/538 - Train Accuracy:  0.778, Validation Accuracy:  0.777, Loss:  2.761
Epoch   1 Batch  152/538 - Train Accuracy:  0.778, Validation Accuracy:  0.773, Loss:  2.804
Epoch   1 Batch  153/538 - Train Accuracy:  0.775, Validation Accuracy:  0.789, Loss:  2.792
Epoch   1 Batch  154/538 - Train Accuracy:  0.753, Validation Accuracy:  0.759, Loss:  2.757
Epoch   1 Batch  155/538 - Train Accuracy:  0.763, Validation Accuracy:  0.766, Loss:  2.884
Epoch   1 Batch  156/538 - Train Accuracy:  0.773, Validation Accuracy:  0.787, Loss:  2.784
Epoch   1 Batch  157/538 - Train Accuracy:  0.776, Validation Accuracy:  0.760, Loss:  2.723
Epoch   1 Batch  158/538 - Train Accuracy:  0.787, Validation Accuracy:  0.773, Loss:  2.810
Epoch   1 Batch  159/538 - Train Accuracy:  0.774, Validation Accuracy:  0.777, Loss:  2.797
Epoch   1 Batch  160/538 - Train Accuracy:  0.751, Validation Accuracy:  0.758, Loss:  2.798
Epoch   1 Batch  161/538 - Train Accuracy:  0.764, Validation Accuracy:  0.781, Loss:  2.763
Epoch   1 Batch  162/538 - Train Accuracy:  0.784, Validation Accuracy:  0.787, Loss:  2.801
Epoch   1 Batch  163/538 - Train Accuracy:  0.795, Validation Accuracy:  0.789, Loss:  2.781
Epoch   1 Batch  164/538 - Train Accuracy:  0.741, Validation Accuracy:  0.755, Loss:  2.764
Epoch   1 Batch  165/538 - Train Accuracy:  0.764, Validation Accuracy:  0.748, Loss:  2.783
Epoch   1 Batch  166/538 - Train Accuracy:  0.790, Validation Accuracy:  0.748, Loss:  2.820
Epoch   1 Batch  167/538 - Train Accuracy:  0.797, Validation Accuracy:  0.766, Loss:  2.827
Epoch   1 Batch  168/538 - Train Accuracy:  0.727, Validation Accuracy:  0.765, Loss:  2.811
Epoch   1 Batch  169/538 - Train Accuracy:  0.810, Validation Accuracy:  0.773, Loss:  2.799
Epoch   1 Batch  170/538 - Train Accuracy:  0.759, Validation Accuracy:  0.779, Loss:  2.757
Epoch   1 Batch  171/538 - Train Accuracy:  0.765, Validation Accuracy:  0.765, Loss:  2.799
Epoch   1 Batch  172/538 - Train Accuracy:  0.772, Validation Accuracy:  0.758, Loss:  2.696
Epoch   1 Batch  173/538 - Train Accuracy:  0.783, Validation Accuracy:  0.778, Loss:  2.876
Epoch   1 Batch  174/538 - Train Accuracy:  0.738, Validation Accuracy:  0.763, Loss:  2.810
Epoch   1 Batch  175/538 - Train Accuracy:  0.778, Validation Accuracy:  0.784, Loss:  2.727
Epoch   1 Batch  176/538 - Train Accuracy:  0.752, Validation Accuracy:  0.754, Loss:  2.767
Epoch   1 Batch  177/538 - Train Accuracy:  0.807, Validation Accuracy:  0.770, Loss:  2.795
Epoch   1 Batch  178/538 - Train Accuracy:  0.782, Validation Accuracy:  0.797, Loss:  2.768
Epoch   1 Batch  179/538 - Train Accuracy:  0.771, Validation Accuracy:  0.784, Loss:  2.762
Epoch   1 Batch  180/538 - Train Accuracy:  0.806, Validation Accuracy:  0.790, Loss:  2.777
Epoch   1 Batch  181/538 - Train Accuracy:  0.755, Validation Accuracy:  0.794, Loss:  2.785
Epoch   1 Batch  182/538 - Train Accuracy:  0.773, Validation Accuracy:  0.798, Loss:  2.735
Epoch   1 Batch  183/538 - Train Accuracy:  0.812, Validation Accuracy:  0.795, Loss:  2.776
Epoch   1 Batch  184/538 - Train Accuracy:  0.790, Validation Accuracy:  0.795, Loss:  2.843
Epoch   1 Batch  185/538 - Train Accuracy:  0.804, Validation Accuracy:  0.797, Loss:  2.825
Epoch   1 Batch  186/538 - Train Accuracy:  0.808, Validation Accuracy:  0.794, Loss:  2.806
Epoch   1 Batch  187/538 - Train Accuracy:  0.804, Validation Accuracy:  0.782, Loss:  2.763
Epoch   1 Batch  188/538 - Train Accuracy:  0.802, Validation Accuracy:  0.797, Loss:  2.803
Epoch   1 Batch  189/538 - Train Accuracy:  0.796, Validation Accuracy:  0.800, Loss:  2.763
Epoch   1 Batch  190/538 - Train Accuracy:  0.783, Validation Accuracy:  0.792, Loss:  2.828
Epoch   1 Batch  191/538 - Train Accuracy:  0.816, Validation Accuracy:  0.805, Loss:  2.787
Epoch   1 Batch  192/538 - Train Accuracy:  0.780, Validation Accuracy:  0.782, Loss:  2.749
Epoch   1 Batch  193/538 - Train Accuracy:  0.765, Validation Accuracy:  0.778, Loss:  2.780
Epoch   1 Batch  194/538 - Train Accuracy:  0.767, Validation Accuracy:  0.807, Loss:  2.790
Epoch   1 Batch  195/538 - Train Accuracy:  0.813, Validation Accuracy:  0.799, Loss:  2.763
Epoch   1 Batch  196/538 - Train Accuracy:  0.802, Validation Accuracy:  0.799, Loss:  2.806
Epoch   1 Batch  197/538 - Train Accuracy:  0.799, Validation Accuracy:  0.797, Loss:  2.739
Epoch   1 Batch  198/538 - Train Accuracy:  0.812, Validation Accuracy:  0.791, Loss:  2.767
Epoch   1 Batch  199/538 - Train Accuracy:  0.794, Validation Accuracy:  0.796, Loss:  2.811
Epoch   1 Batch  200/538 - Train Accuracy:  0.813, Validation Accuracy:  0.806, Loss:  2.729
Epoch   1 Batch  201/538 - Train Accuracy:  0.800, Validation Accuracy:  0.797, Loss:  2.736
Epoch   1 Batch  202/538 - Train Accuracy:  0.824, Validation Accuracy:  0.788, Loss:  2.782
Epoch   1 Batch  203/538 - Train Accuracy:  0.761, Validation Accuracy:  0.789, Loss:  2.819
Epoch   1 Batch  204/538 - Train Accuracy:  0.798, Validation Accuracy:  0.806, Loss:  2.823
Epoch   1 Batch  205/538 - Train Accuracy:  0.812, Validation Accuracy:  0.786, Loss:  2.779
Epoch   1 Batch  206/538 - Train Accuracy:  0.784, Validation Accuracy:  0.805, Loss:  2.769
Epoch   1 Batch  207/538 - Train Accuracy:  0.833, Validation Accuracy:  0.815, Loss:  2.810
Epoch   1 Batch  208/538 - Train Accuracy:  0.783, Validation Accuracy:  0.758, Loss:  2.806
Epoch   1 Batch  209/538 - Train Accuracy:  0.815, Validation Accuracy:  0.787, Loss:  2.763
Epoch   1 Batch  210/538 - Train Accuracy:  0.789, Validation Accuracy:  0.804, Loss:  2.757
Epoch   1 Batch  211/538 - Train Accuracy:  0.803, Validation Accuracy:  0.796, Loss:  2.781
Epoch   1 Batch  212/538 - Train Accuracy:  0.807, Validation Accuracy:  0.809, Loss:  2.790
Epoch   1 Batch  213/538 - Train Accuracy:  0.785, Validation Accuracy:  0.807, Loss:  2.812
Epoch   1 Batch  214/538 - Train Accuracy:  0.805, Validation Accuracy:  0.812, Loss:  2.845
Epoch   1 Batch  215/538 - Train Accuracy:  0.825, Validation Accuracy:  0.812, Loss:  2.710
Epoch   1 Batch  216/538 - Train Accuracy:  0.822, Validation Accuracy:  0.814, Loss:  2.706
Epoch   1 Batch  217/538 - Train Accuracy:  0.838, Validation Accuracy:  0.819, Loss:  2.830
Epoch   1 Batch  218/538 - Train Accuracy:  0.798, Validation Accuracy:  0.822, Loss:  2.739
Epoch   1 Batch  219/538 - Train Accuracy:  0.783, Validation Accuracy:  0.801, Loss:  2.832
Epoch   1 Batch  220/538 - Train Accuracy:  0.792, Validation Accuracy:  0.809, Loss:  2.810
Epoch   1 Batch  221/538 - Train Accuracy:  0.817, Validation Accuracy:  0.794, Loss:  2.787
Epoch   1 Batch  222/538 - Train Accuracy:  0.821, Validation Accuracy:  0.801, Loss:  2.729
Epoch   1 Batch  223/538 - Train Accuracy:  0.799, Validation Accuracy:  0.808, Loss:  2.761
Epoch   1 Batch  224/538 - Train Accuracy:  0.801, Validation Accuracy:  0.809, Loss:  2.762
Epoch   1 Batch  225/538 - Train Accuracy:  0.819, Validation Accuracy:  0.813, Loss:  2.767
Epoch   1 Batch  226/538 - Train Accuracy:  0.839, Validation Accuracy:  0.800, Loss:  2.745
Epoch   1 Batch  227/538 - Train Accuracy:  0.820, Validation Accuracy:  0.805, Loss:  2.733
Epoch   1 Batch  228/538 - Train Accuracy:  0.819, Validation Accuracy:  0.811, Loss:  2.788
Epoch   1 Batch  229/538 - Train Accuracy:  0.821, Validation Accuracy:  0.809, Loss:  2.747
Epoch   1 Batch  230/538 - Train Accuracy:  0.822, Validation Accuracy:  0.817, Loss:  2.783
Epoch   1 Batch  231/538 - Train Accuracy:  0.825, Validation Accuracy:  0.826, Loss:  2.792
Epoch   1 Batch  232/538 - Train Accuracy:  0.825, Validation Accuracy:  0.830, Loss:  2.705
Epoch   1 Batch  233/538 - Train Accuracy:  0.854, Validation Accuracy:  0.835, Loss:  2.765
Epoch   1 Batch  234/538 - Train Accuracy:  0.823, Validation Accuracy:  0.846, Loss:  2.784
Epoch   1 Batch  235/538 - Train Accuracy:  0.841, Validation Accuracy:  0.839, Loss:  2.778
Epoch   1 Batch  236/538 - Train Accuracy:  0.812, Validation Accuracy:  0.829, Loss:  2.731
Epoch   1 Batch  237/538 - Train Accuracy:  0.842, Validation Accuracy:  0.835, Loss:  2.731
Epoch   1 Batch  238/538 - Train Accuracy:  0.863, Validation Accuracy:  0.846, Loss:  2.720
Epoch   1 Batch  239/538 - Train Accuracy:  0.839, Validation Accuracy:  0.839, Loss:  2.745
Epoch   1 Batch  240/538 - Train Accuracy:  0.827, Validation Accuracy:  0.840, Loss:  2.768
Epoch   1 Batch  241/538 - Train Accuracy:  0.839, Validation Accuracy:  0.836, Loss:  2.738
Epoch   1 Batch  242/538 - Train Accuracy:  0.838, Validation Accuracy:  0.834, Loss:  2.745
Epoch   1 Batch  243/538 - Train Accuracy:  0.828, Validation Accuracy:  0.822, Loss:  2.739
Epoch   1 Batch  244/538 - Train Accuracy:  0.823, Validation Accuracy:  0.828, Loss:  2.734
Epoch   1 Batch  245/538 - Train Accuracy:  0.839, Validation Accuracy:  0.832, Loss:  2.690
Epoch   1 Batch  246/538 - Train Accuracy:  0.851, Validation Accuracy:  0.829, Loss:  2.723
Epoch   1 Batch  247/538 - Train Accuracy:  0.843, Validation Accuracy:  0.827, Loss:  2.681
Epoch   1 Batch  248/538 - Train Accuracy:  0.871, Validation Accuracy:  0.826, Loss:  2.708
Epoch   1 Batch  249/538 - Train Accuracy:  0.838, Validation Accuracy:  0.840, Loss:  2.831
Epoch   1 Batch  250/538 - Train Accuracy:  0.850, Validation Accuracy:  0.822, Loss:  2.721
Epoch   1 Batch  251/538 - Train Accuracy:  0.836, Validation Accuracy:  0.819, Loss:  2.787
Epoch   1 Batch  252/538 - Train Accuracy:  0.851, Validation Accuracy:  0.829, Loss:  2.696
Epoch   1 Batch  253/538 - Train Accuracy:  0.829, Validation Accuracy:  0.819, Loss:  2.769
Epoch   1 Batch  254/538 - Train Accuracy:  0.823, Validation Accuracy:  0.829, Loss:  2.770
Epoch   1 Batch  255/538 - Train Accuracy:  0.831, Validation Accuracy:  0.831, Loss:  2.768
Epoch   1 Batch  256/538 - Train Accuracy:  0.804, Validation Accuracy:  0.830, Loss:  2.781
Epoch   1 Batch  257/538 - Train Accuracy:  0.831, Validation Accuracy:  0.822, Loss:  2.763
Epoch   1 Batch  258/538 - Train Accuracy:  0.828, Validation Accuracy:  0.836, Loss:  2.742
Epoch   1 Batch  259/538 - Train Accuracy:  0.865, Validation Accuracy:  0.828, Loss:  2.744
Epoch   1 Batch  260/538 - Train Accuracy:  0.800, Validation Accuracy:  0.826, Loss:  2.737
Epoch   1 Batch  261/538 - Train Accuracy:  0.831, Validation Accuracy:  0.826, Loss:  2.777
Epoch   1 Batch  262/538 - Train Accuracy:  0.869, Validation Accuracy:  0.828, Loss:  2.740
Epoch   1 Batch  263/538 - Train Accuracy:  0.830, Validation Accuracy:  0.818, Loss:  2.792
Epoch   1 Batch  264/538 - Train Accuracy:  0.836, Validation Accuracy:  0.826, Loss:  2.713
Epoch   1 Batch  265/538 - Train Accuracy:  0.802, Validation Accuracy:  0.821, Loss:  2.730
Epoch   1 Batch  266/538 - Train Accuracy:  0.839, Validation Accuracy:  0.836, Loss:  2.762
Epoch   1 Batch  267/538 - Train Accuracy:  0.858, Validation Accuracy:  0.833, Loss:  2.804
Epoch   1 Batch  268/538 - Train Accuracy:  0.880, Validation Accuracy:  0.836, Loss:  2.718
Epoch   1 Batch  269/538 - Train Accuracy:  0.835, Validation Accuracy:  0.841, Loss:  2.724
Epoch   1 Batch  270/538 - Train Accuracy:  0.841, Validation Accuracy:  0.832, Loss:  2.737
Epoch   1 Batch  271/538 - Train Accuracy:  0.847, Validation Accuracy:  0.841, Loss:  2.737
Epoch   1 Batch  272/538 - Train Accuracy:  0.856, Validation Accuracy:  0.847, Loss:  2.703
Epoch   1 Batch  273/538 - Train Accuracy:  0.876, Validation Accuracy:  0.838, Loss:  2.772
Epoch   1 Batch  274/538 - Train Accuracy:  0.817, Validation Accuracy:  0.824, Loss:  2.756
Epoch   1 Batch  275/538 - Train Accuracy:  0.848, Validation Accuracy:  0.835, Loss:  2.745
Epoch   1 Batch  276/538 - Train Accuracy:  0.833, Validation Accuracy:  0.834, Loss:  2.683
Epoch   1 Batch  277/538 - Train Accuracy:  0.866, Validation Accuracy:  0.822, Loss:  2.736
Epoch   1 Batch  278/538 - Train Accuracy:  0.857, Validation Accuracy:  0.832, Loss:  2.668
Epoch   1 Batch  279/538 - Train Accuracy:  0.848, Validation Accuracy:  0.831, Loss:  2.767
Epoch   1 Batch  280/538 - Train Accuracy:  0.859, Validation Accuracy:  0.818, Loss:  2.764
Epoch   1 Batch  281/538 - Train Accuracy:  0.856, Validation Accuracy:  0.827, Loss:  2.702
Epoch   1 Batch  282/538 - Train Accuracy:  0.862, Validation Accuracy:  0.840, Loss:  2.734
Epoch   1 Batch  283/538 - Train Accuracy:  0.842, Validation Accuracy:  0.842, Loss:  2.753
Epoch   1 Batch  284/538 - Train Accuracy:  0.846, Validation Accuracy:  0.839, Loss:  2.772
Epoch   1 Batch  285/538 - Train Accuracy:  0.836, Validation Accuracy:  0.830, Loss:  2.710
Epoch   1 Batch  286/538 - Train Accuracy:  0.830, Validation Accuracy:  0.839, Loss:  2.772
Epoch   1 Batch  287/538 - Train Accuracy:  0.872, Validation Accuracy:  0.845, Loss:  2.745
Epoch   1 Batch  288/538 - Train Accuracy:  0.862, Validation Accuracy:  0.857, Loss:  2.762
Epoch   1 Batch  289/538 - Train Accuracy:  0.865, Validation Accuracy:  0.844, Loss:  2.737
Epoch   1 Batch  290/538 - Train Accuracy:  0.859, Validation Accuracy:  0.862, Loss:  2.738
Epoch   1 Batch  291/538 - Train Accuracy:  0.855, Validation Accuracy:  0.865, Loss:  2.693
Epoch   1 Batch  292/538 - Train Accuracy:  0.856, Validation Accuracy:  0.848, Loss:  2.740
Epoch   1 Batch  293/538 - Train Accuracy:  0.852, Validation Accuracy:  0.851, Loss:  2.751
Epoch   1 Batch  294/538 - Train Accuracy:  0.868, Validation Accuracy:  0.851, Loss:  2.750
Epoch   1 Batch  295/538 - Train Accuracy:  0.876, Validation Accuracy:  0.852, Loss:  2.755
Epoch   1 Batch  296/538 - Train Accuracy:  0.863, Validation Accuracy:  0.856, Loss:  2.729
Epoch   1 Batch  297/538 - Train Accuracy:  0.884, Validation Accuracy:  0.859, Loss:  2.693
Epoch   1 Batch  298/538 - Train Accuracy:  0.841, Validation Accuracy:  0.852, Loss:  2.718
Epoch   1 Batch  299/538 - Train Accuracy:  0.846, Validation Accuracy:  0.856, Loss:  2.771
Epoch   1 Batch  300/538 - Train Accuracy:  0.855, Validation Accuracy:  0.833, Loss:  2.713
Epoch   1 Batch  301/538 - Train Accuracy:  0.836, Validation Accuracy:  0.850, Loss:  2.698
Epoch   1 Batch  302/538 - Train Accuracy:  0.858, Validation Accuracy:  0.853, Loss:  2.768
Epoch   1 Batch  303/538 - Train Accuracy:  0.875, Validation Accuracy:  0.853, Loss:  2.731
Epoch   1 Batch  304/538 - Train Accuracy:  0.879, Validation Accuracy:  0.836, Loss:  2.720
Epoch   1 Batch  305/538 - Train Accuracy:  0.881, Validation Accuracy:  0.843, Loss:  2.675
Epoch   1 Batch  306/538 - Train Accuracy:  0.862, Validation Accuracy:  0.845, Loss:  2.677
Epoch   1 Batch  307/538 - Train Accuracy:  0.870, Validation Accuracy:  0.854, Loss:  2.736
Epoch   1 Batch  308/538 - Train Accuracy:  0.852, Validation Accuracy:  0.860, Loss:  2.700
Epoch   1 Batch  309/538 - Train Accuracy:  0.874, Validation Accuracy:  0.857, Loss:  2.729
Epoch   1 Batch  310/538 - Train Accuracy:  0.893, Validation Accuracy:  0.827, Loss:  2.645
Epoch   1 Batch  311/538 - Train Accuracy:  0.840, Validation Accuracy:  0.839, Loss:  2.727
Epoch   1 Batch  312/538 - Train Accuracy:  0.868, Validation Accuracy:  0.850, Loss:  2.710
Epoch   1 Batch  313/538 - Train Accuracy:  0.845, Validation Accuracy:  0.854, Loss:  2.667
Epoch   1 Batch  314/538 - Train Accuracy:  0.887, Validation Accuracy:  0.871, Loss:  2.701
Epoch   1 Batch  315/538 - Train Accuracy:  0.862, Validation Accuracy:  0.859, Loss:  2.770
Epoch   1 Batch  316/538 - Train Accuracy:  0.859, Validation Accuracy:  0.868, Loss:  2.734
Epoch   1 Batch  317/538 - Train Accuracy:  0.877, Validation Accuracy:  0.874, Loss:  2.691
Epoch   1 Batch  318/538 - Train Accuracy:  0.864, Validation Accuracy:  0.871, Loss:  2.785
Epoch   1 Batch  319/538 - Train Accuracy:  0.876, Validation Accuracy:  0.853, Loss:  2.708
Epoch   1 Batch  320/538 - Train Accuracy:  0.850, Validation Accuracy:  0.854, Loss:  2.690
Epoch   1 Batch  321/538 - Train Accuracy:  0.877, Validation Accuracy:  0.862, Loss:  2.691
Epoch   1 Batch  322/538 - Train Accuracy:  0.885, Validation Accuracy:  0.876, Loss:  2.650
Epoch   1 Batch  323/538 - Train Accuracy:  0.884, Validation Accuracy:  0.872, Loss:  2.741
Epoch   1 Batch  324/538 - Train Accuracy:  0.857, Validation Accuracy:  0.872, Loss:  2.704
Epoch   1 Batch  325/538 - Train Accuracy:  0.883, Validation Accuracy:  0.867, Loss:  2.722
Epoch   1 Batch  326/538 - Train Accuracy:  0.873, Validation Accuracy:  0.851, Loss:  2.698
Epoch   1 Batch  327/538 - Train Accuracy:  0.878, Validation Accuracy:  0.877, Loss:  2.677
Epoch   1 Batch  328/538 - Train Accuracy:  0.894, Validation Accuracy:  0.858, Loss:  2.695
Epoch   1 Batch  329/538 - Train Accuracy:  0.884, Validation Accuracy:  0.841, Loss:  2.653
Epoch   1 Batch  330/538 - Train Accuracy:  0.878, Validation Accuracy:  0.843, Loss:  2.658
Epoch   1 Batch  331/538 - Train Accuracy:  0.867, Validation Accuracy:  0.855, Loss:  2.751
Epoch   1 Batch  332/538 - Train Accuracy:  0.859, Validation Accuracy:  0.848, Loss:  2.711
Epoch   1 Batch  333/538 - Train Accuracy:  0.883, Validation Accuracy:  0.852, Loss:  2.708
Epoch   1 Batch  334/538 - Train Accuracy:  0.892, Validation Accuracy:  0.854, Loss:  2.711
Epoch   1 Batch  335/538 - Train Accuracy:  0.857, Validation Accuracy:  0.864, Loss:  2.699
Epoch   1 Batch  336/538 - Train Accuracy:  0.877, Validation Accuracy:  0.859, Loss:  2.758
Epoch   1 Batch  337/538 - Train Accuracy:  0.892, Validation Accuracy:  0.851, Loss:  2.683
Epoch   1 Batch  338/538 - Train Accuracy:  0.876, Validation Accuracy:  0.851, Loss:  2.746
Epoch   1 Batch  339/538 - Train Accuracy:  0.877, Validation Accuracy:  0.867, Loss:  2.720
Epoch   1 Batch  340/538 - Train Accuracy:  0.846, Validation Accuracy:  0.857, Loss:  2.692
Epoch   1 Batch  341/538 - Train Accuracy:  0.883, Validation Accuracy:  0.858, Loss:  2.752
Epoch   1 Batch  342/538 - Train Accuracy:  0.864, Validation Accuracy:  0.859, Loss:  2.755
Epoch   1 Batch  343/538 - Train Accuracy:  0.881, Validation Accuracy:  0.861, Loss:  2.722
Epoch   1 Batch  344/538 - Train Accuracy:  0.870, Validation Accuracy:  0.861, Loss:  2.677
Epoch   1 Batch  345/538 - Train Accuracy:  0.865, Validation Accuracy:  0.867, Loss:  2.696
Epoch   1 Batch  346/538 - Train Accuracy:  0.859, Validation Accuracy:  0.869, Loss:  2.638
Epoch   1 Batch  347/538 - Train Accuracy:  0.871, Validation Accuracy:  0.855, Loss:  2.710
Epoch   1 Batch  348/538 - Train Accuracy:  0.888, Validation Accuracy:  0.871, Loss:  2.692
Epoch   1 Batch  349/538 - Train Accuracy:  0.891, Validation Accuracy:  0.867, Loss:  2.733
Epoch   1 Batch  350/538 - Train Accuracy:  0.860, Validation Accuracy:  0.864, Loss:  2.733
Epoch   1 Batch  351/538 - Train Accuracy:  0.861, Validation Accuracy:  0.852, Loss:  2.692
Epoch   1 Batch  352/538 - Train Accuracy:  0.882, Validation Accuracy:  0.850, Loss:  2.711
Epoch   1 Batch  353/538 - Train Accuracy:  0.867, Validation Accuracy:  0.857, Loss:  2.701
Epoch   1 Batch  354/538 - Train Accuracy:  0.876, Validation Accuracy:  0.854, Loss:  2.718
Epoch   1 Batch  355/538 - Train Accuracy:  0.883, Validation Accuracy:  0.863, Loss:  2.787
Epoch   1 Batch  356/538 - Train Accuracy:  0.881, Validation Accuracy:  0.862, Loss:  2.753
Epoch   1 Batch  357/538 - Train Accuracy:  0.899, Validation Accuracy:  0.884, Loss:  2.759
Epoch   1 Batch  358/538 - Train Accuracy:  0.878, Validation Accuracy:  0.875, Loss:  2.690
Epoch   1 Batch  359/538 - Train Accuracy:  0.871, Validation Accuracy:  0.874, Loss:  2.758
Epoch   1 Batch  360/538 - Train Accuracy:  0.863, Validation Accuracy:  0.870, Loss:  2.722
Epoch   1 Batch  361/538 - Train Accuracy:  0.900, Validation Accuracy:  0.867, Loss:  2.691
Epoch   1 Batch  362/538 - Train Accuracy:  0.902, Validation Accuracy:  0.863, Loss:  2.624
Epoch   1 Batch  363/538 - Train Accuracy:  0.882, Validation Accuracy:  0.872, Loss:  2.670
Epoch   1 Batch  364/538 - Train Accuracy:  0.861, Validation Accuracy:  0.861, Loss:  2.711
Epoch   1 Batch  365/538 - Train Accuracy:  0.881, Validation Accuracy:  0.865, Loss:  2.659
Epoch   1 Batch  366/538 - Train Accuracy:  0.911, Validation Accuracy:  0.868, Loss:  2.687
Epoch   1 Batch  367/538 - Train Accuracy:  0.905, Validation Accuracy:  0.871, Loss:  2.700
Epoch   1 Batch  368/538 - Train Accuracy:  0.892, Validation Accuracy:  0.858, Loss:  2.663
Epoch   1 Batch  369/538 - Train Accuracy:  0.887, Validation Accuracy:  0.867, Loss:  2.708
Epoch   1 Batch  370/538 - Train Accuracy:  0.900, Validation Accuracy:  0.874, Loss:  2.702
Epoch   1 Batch  371/538 - Train Accuracy:  0.911, Validation Accuracy:  0.872, Loss:  2.647
Epoch   1 Batch  372/538 - Train Accuracy:  0.901, Validation Accuracy:  0.875, Loss:  2.684
Epoch   1 Batch  373/538 - Train Accuracy:  0.902, Validation Accuracy:  0.870, Loss:  2.665
Epoch   1 Batch  374/538 - Train Accuracy:  0.899, Validation Accuracy:  0.872, Loss:  2.715
Epoch   1 Batch  375/538 - Train Accuracy:  0.899, Validation Accuracy:  0.877, Loss:  2.668
Epoch   1 Batch  376/538 - Train Accuracy:  0.889, Validation Accuracy:  0.881, Loss:  2.712
Epoch   1 Batch  377/538 - Train Accuracy:  0.892, Validation Accuracy:  0.867, Loss:  2.715
Epoch   1 Batch  378/538 - Train Accuracy:  0.913, Validation Accuracy:  0.866, Loss:  2.704
Epoch   1 Batch  379/538 - Train Accuracy:  0.904, Validation Accuracy:  0.890, Loss:  2.686
Epoch   1 Batch  380/538 - Train Accuracy:  0.893, Validation Accuracy:  0.879, Loss:  2.625
Epoch   1 Batch  381/538 - Train Accuracy:  0.911, Validation Accuracy:  0.879, Loss:  2.756
Epoch   1 Batch  382/538 - Train Accuracy:  0.906, Validation Accuracy:  0.891, Loss:  2.615
Epoch   1 Batch  383/538 - Train Accuracy:  0.882, Validation Accuracy:  0.906, Loss:  2.715
Epoch   1 Batch  384/538 - Train Accuracy:  0.891, Validation Accuracy:  0.901, Loss:  2.714
Epoch   1 Batch  385/538 - Train Accuracy:  0.915, Validation Accuracy:  0.894, Loss:  2.634
Epoch   1 Batch  386/538 - Train Accuracy:  0.897, Validation Accuracy:  0.900, Loss:  2.696
Epoch   1 Batch  387/538 - Train Accuracy:  0.889, Validation Accuracy:  0.894, Loss:  2.715
Epoch   1 Batch  388/538 - Train Accuracy:  0.914, Validation Accuracy:  0.893, Loss:  2.702
Epoch   1 Batch  389/538 - Train Accuracy:  0.880, Validation Accuracy:  0.881, Loss:  2.727
Epoch   1 Batch  390/538 - Train Accuracy:  0.905, Validation Accuracy:  0.867, Loss:  2.665
Epoch   1 Batch  391/538 - Train Accuracy:  0.882, Validation Accuracy:  0.870, Loss:  2.706
Epoch   1 Batch  392/538 - Train Accuracy:  0.885, Validation Accuracy:  0.867, Loss:  2.723
Epoch   1 Batch  393/538 - Train Accuracy:  0.903, Validation Accuracy:  0.867, Loss:  2.658
Epoch   1 Batch  394/538 - Train Accuracy:  0.856, Validation Accuracy:  0.866, Loss:  2.685
Epoch   1 Batch  395/538 - Train Accuracy:  0.885, Validation Accuracy:  0.865, Loss:  2.794
Epoch   1 Batch  396/538 - Train Accuracy:  0.890, Validation Accuracy:  0.867, Loss:  2.798
Epoch   1 Batch  397/538 - Train Accuracy:  0.894, Validation Accuracy:  0.866, Loss:  2.678
Epoch   1 Batch  398/538 - Train Accuracy:  0.884, Validation Accuracy:  0.884, Loss:  2.722
Epoch   1 Batch  399/538 - Train Accuracy:  0.877, Validation Accuracy:  0.890, Loss:  2.704
Epoch   1 Batch  400/538 - Train Accuracy:  0.901, Validation Accuracy:  0.895, Loss:  2.649
Epoch   1 Batch  401/538 - Train Accuracy:  0.912, Validation Accuracy:  0.898, Loss:  2.688
Epoch   1 Batch  402/538 - Train Accuracy:  0.892, Validation Accuracy:  0.902, Loss:  2.695
Epoch   1 Batch  403/538 - Train Accuracy:  0.900, Validation Accuracy:  0.896, Loss:  2.679
Epoch   1 Batch  404/538 - Train Accuracy:  0.879, Validation Accuracy:  0.890, Loss:  2.652
Epoch   1 Batch  405/538 - Train Accuracy:  0.891, Validation Accuracy:  0.874, Loss:  2.719
Epoch   1 Batch  406/538 - Train Accuracy:  0.891, Validation Accuracy:  0.878, Loss:  2.670
Epoch   1 Batch  407/538 - Train Accuracy:  0.916, Validation Accuracy:  0.892, Loss:  2.709
Epoch   1 Batch  408/538 - Train Accuracy:  0.894, Validation Accuracy:  0.882, Loss:  2.656
Epoch   1 Batch  409/538 - Train Accuracy:  0.873, Validation Accuracy:  0.883, Loss:  2.707
Epoch   1 Batch  410/538 - Train Accuracy:  0.897, Validation Accuracy:  0.886, Loss:  2.765
Epoch   1 Batch  411/538 - Train Accuracy:  0.913, Validation Accuracy:  0.874, Loss:  2.633
Epoch   1 Batch  412/538 - Train Accuracy:  0.895, Validation Accuracy:  0.874, Loss:  2.702
Epoch   1 Batch  413/538 - Train Accuracy:  0.888, Validation Accuracy:  0.878, Loss:  2.645
Epoch   1 Batch  414/538 - Train Accuracy:  0.871, Validation Accuracy:  0.879, Loss:  2.703
Epoch   1 Batch  415/538 - Train Accuracy:  0.877, Validation Accuracy:  0.885, Loss:  2.635
Epoch   1 Batch  416/538 - Train Accuracy:  0.897, Validation Accuracy:  0.876, Loss:  2.656
Epoch   1 Batch  417/538 - Train Accuracy:  0.909, Validation Accuracy:  0.874, Loss:  2.716
Epoch   1 Batch  418/538 - Train Accuracy:  0.886, Validation Accuracy:  0.885, Loss:  2.711
Epoch   1 Batch  419/538 - Train Accuracy:  0.909, Validation Accuracy:  0.892, Loss:  2.648
Epoch   1 Batch  420/538 - Train Accuracy:  0.926, Validation Accuracy:  0.896, Loss:  2.737
Epoch   1 Batch  421/538 - Train Accuracy:  0.915, Validation Accuracy:  0.887, Loss:  2.602
Epoch   1 Batch  422/538 - Train Accuracy:  0.901, Validation Accuracy:  0.886, Loss:  2.662
Epoch   1 Batch  423/538 - Train Accuracy:  0.912, Validation Accuracy:  0.873, Loss:  2.696
Epoch   1 Batch  424/538 - Train Accuracy:  0.903, Validation Accuracy:  0.875, Loss:  2.686
Epoch   1 Batch  425/538 - Train Accuracy:  0.892, Validation Accuracy:  0.879, Loss:  2.661
Epoch   1 Batch  426/538 - Train Accuracy:  0.888, Validation Accuracy:  0.879, Loss:  2.620
Epoch   1 Batch  427/538 - Train Accuracy:  0.871, Validation Accuracy:  0.876, Loss:  2.590
Epoch   1 Batch  428/538 - Train Accuracy:  0.901, Validation Accuracy:  0.870, Loss:  2.626
Epoch   1 Batch  429/538 - Train Accuracy:  0.901, Validation Accuracy:  0.879, Loss:  2.640
Epoch   1 Batch  430/538 - Train Accuracy:  0.906, Validation Accuracy:  0.883, Loss:  2.615
Epoch   1 Batch  431/538 - Train Accuracy:  0.902, Validation Accuracy:  0.870, Loss:  2.700
Epoch   1 Batch  432/538 - Train Accuracy:  0.903, Validation Accuracy:  0.871, Loss:  2.616
Epoch   1 Batch  433/538 - Train Accuracy:  0.903, Validation Accuracy:  0.888, Loss:  2.686
Epoch   1 Batch  434/538 - Train Accuracy:  0.891, Validation Accuracy:  0.900, Loss:  2.673
Epoch   1 Batch  435/538 - Train Accuracy:  0.898, Validation Accuracy:  0.898, Loss:  2.678
Epoch   1 Batch  436/538 - Train Accuracy:  0.901, Validation Accuracy:  0.905, Loss:  2.654
Epoch   1 Batch  437/538 - Train Accuracy:  0.916, Validation Accuracy:  0.887, Loss:  2.635
Epoch   1 Batch  438/538 - Train Accuracy:  0.899, Validation Accuracy:  0.881, Loss:  2.685
Epoch   1 Batch  439/538 - Train Accuracy:  0.920, Validation Accuracy:  0.888, Loss:  2.652
Epoch   1 Batch  440/538 - Train Accuracy:  0.903, Validation Accuracy:  0.906, Loss:  2.666
Epoch   1 Batch  441/538 - Train Accuracy:  0.889, Validation Accuracy:  0.905, Loss:  2.689
Epoch   1 Batch  442/538 - Train Accuracy:  0.919, Validation Accuracy:  0.901, Loss:  2.633
Epoch   1 Batch  443/538 - Train Accuracy:  0.913, Validation Accuracy:  0.901, Loss:  2.663
Epoch   1 Batch  444/538 - Train Accuracy:  0.900, Validation Accuracy:  0.896, Loss:  2.696
Epoch   1 Batch  445/538 - Train Accuracy:  0.916, Validation Accuracy:  0.887, Loss:  2.616
Epoch   1 Batch  446/538 - Train Accuracy:  0.910, Validation Accuracy:  0.894, Loss:  2.683
Epoch   1 Batch  447/538 - Train Accuracy:  0.892, Validation Accuracy:  0.903, Loss:  2.720
Epoch   1 Batch  448/538 - Train Accuracy:  0.890, Validation Accuracy:  0.897, Loss:  2.685
Epoch   1 Batch  449/538 - Train Accuracy:  0.925, Validation Accuracy:  0.898, Loss:  2.632
Epoch   1 Batch  450/538 - Train Accuracy:  0.907, Validation Accuracy:  0.905, Loss:  2.661
Epoch   1 Batch  451/538 - Train Accuracy:  0.888, Validation Accuracy:  0.886, Loss:  2.636
Epoch   1 Batch  452/538 - Train Accuracy:  0.913, Validation Accuracy:  0.894, Loss:  2.651
Epoch   1 Batch  453/538 - Train Accuracy:  0.916, Validation Accuracy:  0.885, Loss:  2.663
Epoch   1 Batch  454/538 - Train Accuracy:  0.895, Validation Accuracy:  0.893, Loss:  2.680
Epoch   1 Batch  455/538 - Train Accuracy:  0.910, Validation Accuracy:  0.900, Loss:  2.652
Epoch   1 Batch  456/538 - Train Accuracy:  0.918, Validation Accuracy:  0.897, Loss:  2.624
Epoch   1 Batch  457/538 - Train Accuracy:  0.909, Validation Accuracy:  0.901, Loss:  2.660
Epoch   1 Batch  458/538 - Train Accuracy:  0.901, Validation Accuracy:  0.898, Loss:  2.669
Epoch   1 Batch  459/538 - Train Accuracy:  0.899, Validation Accuracy:  0.898, Loss:  2.669
Epoch   1 Batch  460/538 - Train Accuracy:  0.888, Validation Accuracy:  0.896, Loss:  2.709
Epoch   1 Batch  461/538 - Train Accuracy:  0.903, Validation Accuracy:  0.890, Loss:  2.657
Epoch   1 Batch  462/538 - Train Accuracy:  0.902, Validation Accuracy:  0.895, Loss:  2.649
Epoch   1 Batch  463/538 - Train Accuracy:  0.890, Validation Accuracy:  0.898, Loss:  2.632
Epoch   1 Batch  464/538 - Train Accuracy:  0.885, Validation Accuracy:  0.904, Loss:  2.680
Epoch   1 Batch  465/538 - Train Accuracy:  0.901, Validation Accuracy:  0.902, Loss:  2.669
Epoch   1 Batch  466/538 - Train Accuracy:  0.913, Validation Accuracy:  0.899, Loss:  2.695
Epoch   1 Batch  467/538 - Train Accuracy:  0.917, Validation Accuracy:  0.893, Loss:  2.740
Epoch   1 Batch  468/538 - Train Accuracy:  0.913, Validation Accuracy:  0.899, Loss:  2.720
Epoch   1 Batch  469/538 - Train Accuracy:  0.899, Validation Accuracy:  0.896, Loss:  2.718
Epoch   1 Batch  470/538 - Train Accuracy:  0.910, Validation Accuracy:  0.907, Loss:  2.637
Epoch   1 Batch  471/538 - Train Accuracy:  0.906, Validation Accuracy:  0.898, Loss:  2.684
Epoch   1 Batch  472/538 - Train Accuracy:  0.945, Validation Accuracy:  0.898, Loss:  2.609
Epoch   1 Batch  473/538 - Train Accuracy:  0.892, Validation Accuracy:  0.890, Loss:  2.667
Epoch   1 Batch  474/538 - Train Accuracy:  0.926, Validation Accuracy:  0.892, Loss:  2.629
Epoch   1 Batch  475/538 - Train Accuracy:  0.918, Validation Accuracy:  0.893, Loss:  2.676
Epoch   1 Batch  476/538 - Train Accuracy:  0.911, Validation Accuracy:  0.901, Loss:  2.586
Epoch   1 Batch  477/538 - Train Accuracy:  0.912, Validation Accuracy:  0.898, Loss:  2.770
Epoch   1 Batch  478/538 - Train Accuracy:  0.929, Validation Accuracy:  0.902, Loss:  2.648
Epoch   1 Batch  479/538 - Train Accuracy:  0.920, Validation Accuracy:  0.904, Loss:  2.668
Epoch   1 Batch  480/538 - Train Accuracy:  0.924, Validation Accuracy:  0.903, Loss:  2.665
Epoch   1 Batch  481/538 - Train Accuracy:  0.922, Validation Accuracy:  0.898, Loss:  2.708
Epoch   1 Batch  482/538 - Train Accuracy:  0.919, Validation Accuracy:  0.891, Loss:  2.603
Epoch   1 Batch  483/538 - Train Accuracy:  0.867, Validation Accuracy:  0.896, Loss:  2.661
Epoch   1 Batch  484/538 - Train Accuracy:  0.915, Validation Accuracy:  0.900, Loss:  2.601
Epoch   1 Batch  485/538 - Train Accuracy:  0.927, Validation Accuracy:  0.896, Loss:  2.640
Epoch   1 Batch  486/538 - Train Accuracy:  0.917, Validation Accuracy:  0.894, Loss:  2.634
Epoch   1 Batch  487/538 - Train Accuracy:  0.911, Validation Accuracy:  0.883, Loss:  2.598
Epoch   1 Batch  488/538 - Train Accuracy:  0.915, Validation Accuracy:  0.896, Loss:  2.674
Epoch   1 Batch  489/538 - Train Accuracy:  0.897, Validation Accuracy:  0.890, Loss:  2.668
Epoch   1 Batch  490/538 - Train Accuracy:  0.900, Validation Accuracy:  0.894, Loss:  2.620
Epoch   1 Batch  491/538 - Train Accuracy:  0.897, Validation Accuracy:  0.896, Loss:  2.631
Epoch   1 Batch  492/538 - Train Accuracy:  0.903, Validation Accuracy:  0.906, Loss:  2.682
Epoch   1 Batch  493/538 - Train Accuracy:  0.908, Validation Accuracy:  0.898, Loss:  2.577
Epoch   1 Batch  494/538 - Train Accuracy:  0.904, Validation Accuracy:  0.888, Loss:  2.633
Epoch   1 Batch  495/538 - Train Accuracy:  0.927, Validation Accuracy:  0.899, Loss:  2.619
Epoch   1 Batch  496/538 - Train Accuracy:  0.934, Validation Accuracy:  0.900, Loss:  2.692
Epoch   1 Batch  497/538 - Train Accuracy:  0.919, Validation Accuracy:  0.886, Loss:  2.636
Epoch   1 Batch  498/538 - Train Accuracy:  0.900, Validation Accuracy:  0.893, Loss:  2.662
Epoch   1 Batch  499/538 - Train Accuracy:  0.930, Validation Accuracy:  0.895, Loss:  2.655
Epoch   1 Batch  500/538 - Train Accuracy:  0.912, Validation Accuracy:  0.892, Loss:  2.615
Epoch   1 Batch  501/538 - Train Accuracy:  0.921, Validation Accuracy:  0.904, Loss:  2.690
Epoch   1 Batch  502/538 - Train Accuracy:  0.916, Validation Accuracy:  0.902, Loss:  2.630
Epoch   1 Batch  503/538 - Train Accuracy:  0.914, Validation Accuracy:  0.892, Loss:  2.691
Epoch   1 Batch  504/538 - Train Accuracy:  0.943, Validation Accuracy:  0.898, Loss:  2.662
Epoch   1 Batch  505/538 - Train Accuracy:  0.920, Validation Accuracy:  0.904, Loss:  2.586
Epoch   1 Batch  506/538 - Train Accuracy:  0.919, Validation Accuracy:  0.907, Loss:  2.674
Epoch   1 Batch  507/538 - Train Accuracy:  0.894, Validation Accuracy:  0.909, Loss:  2.700
Epoch   1 Batch  508/538 - Train Accuracy:  0.918, Validation Accuracy:  0.918, Loss:  2.618
Epoch   1 Batch  509/538 - Train Accuracy:  0.918, Validation Accuracy:  0.911, Loss:  2.693
Epoch   1 Batch  510/538 - Train Accuracy:  0.916, Validation Accuracy:  0.910, Loss:  2.672
Epoch   1 Batch  511/538 - Train Accuracy:  0.891, Validation Accuracy:  0.909, Loss:  2.614
Epoch   1 Batch  512/538 - Train Accuracy:  0.932, Validation Accuracy:  0.904, Loss:  2.691
Epoch   1 Batch  513/538 - Train Accuracy:  0.895, Validation Accuracy:  0.901, Loss:  2.632
Epoch   1 Batch  514/538 - Train Accuracy:  0.915, Validation Accuracy:  0.902, Loss:  2.633
Epoch   1 Batch  515/538 - Train Accuracy:  0.889, Validation Accuracy:  0.901, Loss:  2.630
Epoch   1 Batch  516/538 - Train Accuracy:  0.863, Validation Accuracy:  0.905, Loss:  2.607
Epoch   1 Batch  517/538 - Train Accuracy:  0.926, Validation Accuracy:  0.901, Loss:  2.600
Epoch   1 Batch  518/538 - Train Accuracy:  0.900, Validation Accuracy:  0.904, Loss:  2.676
Epoch   1 Batch  519/538 - Train Accuracy:  0.912, Validation Accuracy:  0.894, Loss:  2.616
Epoch   1 Batch  520/538 - Train Accuracy:  0.910, Validation Accuracy:  0.890, Loss:  2.612
Epoch   1 Batch  521/538 - Train Accuracy:  0.896, Validation Accuracy:  0.895, Loss:  2.727
Epoch   1 Batch  522/538 - Train Accuracy:  0.912, Validation Accuracy:  0.892, Loss:  2.654
Epoch   1 Batch  523/538 - Train Accuracy:  0.926, Validation Accuracy:  0.892, Loss:  2.719
Epoch   1 Batch  524/538 - Train Accuracy:  0.925, Validation Accuracy:  0.897, Loss:  2.615
Epoch   1 Batch  525/538 - Train Accuracy:  0.917, Validation Accuracy:  0.903, Loss:  2.625
Epoch   1 Batch  526/538 - Train Accuracy:  0.919, Validation Accuracy:  0.901, Loss:  2.655
Epoch   1 Batch  527/538 - Train Accuracy:  0.911, Validation Accuracy:  0.901, Loss:  2.630
Epoch   1 Batch  528/538 - Train Accuracy:  0.916, Validation Accuracy:  0.909, Loss:  2.629
Epoch   1 Batch  529/538 - Train Accuracy:  0.904, Validation Accuracy:  0.914, Loss:  2.633
Epoch   1 Batch  530/538 - Train Accuracy:  0.905, Validation Accuracy:  0.918, Loss:  2.686
Epoch   1 Batch  531/538 - Train Accuracy:  0.918, Validation Accuracy:  0.913, Loss:  2.716
Epoch   1 Batch  532/538 - Train Accuracy:  0.923, Validation Accuracy:  0.910, Loss:  2.612
Epoch   1 Batch  533/538 - Train Accuracy:  0.928, Validation Accuracy:  0.906, Loss:  2.643
Epoch   1 Batch  534/538 - Train Accuracy:  0.912, Validation Accuracy:  0.904, Loss:  2.634
Epoch   1 Batch  535/538 - Train Accuracy:  0.910, Validation Accuracy:  0.904, Loss:  2.685
Epoch   1 Batch  536/538 - Train Accuracy:  0.931, Validation Accuracy:  0.909, Loss:  2.672
Epoch   2 Batch    0/538 - Train Accuracy:  0.916, Validation Accuracy:  0.904, Loss:  2.625
Epoch   2 Batch    1/538 - Train Accuracy:  0.933, Validation Accuracy:  0.898, Loss:  2.742
Epoch   2 Batch    2/538 - Train Accuracy:  0.909, Validation Accuracy:  0.903, Loss:  2.662
Epoch   2 Batch    3/538 - Train Accuracy:  0.919, Validation Accuracy:  0.909, Loss:  2.647
Epoch   2 Batch    4/538 - Train Accuracy:  0.908, Validation Accuracy:  0.908, Loss:  2.662
Epoch   2 Batch    5/538 - Train Accuracy:  0.918, Validation Accuracy:  0.899, Loss:  2.629
Epoch   2 Batch    6/538 - Train Accuracy:  0.924, Validation Accuracy:  0.899, Loss:  2.615
Epoch   2 Batch    7/538 - Train Accuracy:  0.934, Validation Accuracy:  0.897, Loss:  2.639
Epoch   2 Batch    8/538 - Train Accuracy:  0.903, Validation Accuracy:  0.892, Loss:  2.644
Epoch   2 Batch    9/538 - Train Accuracy:  0.913, Validation Accuracy:  0.888, Loss:  2.662
Epoch   2 Batch   10/538 - Train Accuracy:  0.927, Validation Accuracy:  0.899, Loss:  2.646
Epoch   2 Batch   11/538 - Train Accuracy:  0.929, Validation Accuracy:  0.906, Loss:  2.632
Epoch   2 Batch   12/538 - Train Accuracy:  0.932, Validation Accuracy:  0.911, Loss:  2.709
Epoch   2 Batch   13/538 - Train Accuracy:  0.926, Validation Accuracy:  0.905, Loss:  2.651
Epoch   2 Batch   14/538 - Train Accuracy:  0.921, Validation Accuracy:  0.898, Loss:  2.628
Epoch   2 Batch   15/538 - Train Accuracy:  0.921, Validation Accuracy:  0.896, Loss:  2.668
Epoch   2 Batch   16/538 - Train Accuracy:  0.916, Validation Accuracy:  0.909, Loss:  2.740
Epoch   2 Batch   17/538 - Train Accuracy:  0.930, Validation Accuracy:  0.914, Loss:  2.654
Epoch   2 Batch   18/538 - Train Accuracy:  0.936, Validation Accuracy:  0.922, Loss:  2.670
Epoch   2 Batch   19/538 - Train Accuracy:  0.896, Validation Accuracy:  0.919, Loss:  2.605
Epoch   2 Batch   20/538 - Train Accuracy:  0.922, Validation Accuracy:  0.922, Loss:  2.653
Epoch   2 Batch   21/538 - Train Accuracy:  0.923, Validation Accuracy:  0.923, Loss:  2.649
Epoch   2 Batch   22/538 - Train Accuracy:  0.915, Validation Accuracy:  0.906, Loss:  2.625
Epoch   2 Batch   23/538 - Train Accuracy:  0.912, Validation Accuracy:  0.910, Loss:  2.718
Epoch   2 Batch   24/538 - Train Accuracy:  0.937, Validation Accuracy:  0.908, Loss:  2.658
Epoch   2 Batch   25/538 - Train Accuracy:  0.925, Validation Accuracy:  0.909, Loss:  2.627
Epoch   2 Batch   26/538 - Train Accuracy:  0.914, Validation Accuracy:  0.901, Loss:  2.668
Epoch   2 Batch   27/538 - Train Accuracy:  0.914, Validation Accuracy:  0.903, Loss:  2.638
Epoch   2 Batch   28/538 - Train Accuracy:  0.922, Validation Accuracy:  0.905, Loss:  2.618
Epoch   2 Batch   29/538 - Train Accuracy:  0.928, Validation Accuracy:  0.909, Loss:  2.635
Epoch   2 Batch   30/538 - Train Accuracy:  0.915, Validation Accuracy:  0.913, Loss:  2.629
Epoch   2 Batch   31/538 - Train Accuracy:  0.937, Validation Accuracy:  0.914, Loss:  2.645
Epoch   2 Batch   32/538 - Train Accuracy:  0.923, Validation Accuracy:  0.918, Loss:  2.673
Epoch   2 Batch   33/538 - Train Accuracy:  0.917, Validation Accuracy:  0.900, Loss:  2.652
Epoch   2 Batch   34/538 - Train Accuracy:  0.899, Validation Accuracy:  0.907, Loss:  2.628
Epoch   2 Batch   35/538 - Train Accuracy:  0.915, Validation Accuracy:  0.905, Loss:  2.647
Epoch   2 Batch   36/538 - Train Accuracy:  0.911, Validation Accuracy:  0.916, Loss:  2.639
Epoch   2 Batch   37/538 - Train Accuracy:  0.909, Validation Accuracy:  0.923, Loss:  2.583
Epoch   2 Batch   38/538 - Train Accuracy:  0.914, Validation Accuracy:  0.918, Loss:  2.605
Epoch   2 Batch   39/538 - Train Accuracy:  0.920, Validation Accuracy:  0.914, Loss:  2.682
Epoch   2 Batch   40/538 - Train Accuracy:  0.927, Validation Accuracy:  0.913, Loss:  2.650
Epoch   2 Batch   41/538 - Train Accuracy:  0.916, Validation Accuracy:  0.920, Loss:  2.614
Epoch   2 Batch   42/538 - Train Accuracy:  0.912, Validation Accuracy:  0.915, Loss:  2.621
Epoch   2 Batch   43/538 - Train Accuracy:  0.895, Validation Accuracy:  0.920, Loss:  2.647
Epoch   2 Batch   44/538 - Train Accuracy:  0.899, Validation Accuracy:  0.921, Loss:  2.594
Epoch   2 Batch   45/538 - Train Accuracy:  0.914, Validation Accuracy:  0.914, Loss:  2.630
Epoch   2 Batch   46/538 - Train Accuracy:  0.936, Validation Accuracy:  0.904, Loss:  2.575
Epoch   2 Batch   47/538 - Train Accuracy:  0.906, Validation Accuracy:  0.910, Loss:  2.667
Epoch   2 Batch   48/538 - Train Accuracy:  0.905, Validation Accuracy:  0.909, Loss:  2.576
Epoch   2 Batch   49/538 - Train Accuracy:  0.937, Validation Accuracy:  0.905, Loss:  2.634
Epoch   2 Batch   50/538 - Train Accuracy:  0.917, Validation Accuracy:  0.905, Loss:  2.639
Epoch   2 Batch   51/538 - Train Accuracy:  0.898, Validation Accuracy:  0.909, Loss:  2.654
Epoch   2 Batch   52/538 - Train Accuracy:  0.891, Validation Accuracy:  0.908, Loss:  2.687
Epoch   2 Batch   53/538 - Train Accuracy:  0.916, Validation Accuracy:  0.919, Loss:  2.636
Epoch   2 Batch   54/538 - Train Accuracy:  0.914, Validation Accuracy:  0.917, Loss:  2.673
Epoch   2 Batch   55/538 - Train Accuracy:  0.911, Validation Accuracy:  0.924, Loss:  2.661
Epoch   2 Batch   56/538 - Train Accuracy:  0.923, Validation Accuracy:  0.928, Loss:  2.661
Epoch   2 Batch   57/538 - Train Accuracy:  0.886, Validation Accuracy:  0.920, Loss:  2.683
Epoch   2 Batch   58/538 - Train Accuracy:  0.914, Validation Accuracy:  0.917, Loss:  2.654
Epoch   2 Batch   59/538 - Train Accuracy:  0.906, Validation Accuracy:  0.909, Loss:  2.716
Epoch   2 Batch   60/538 - Train Accuracy:  0.929, Validation Accuracy:  0.910, Loss:  2.703
Epoch   2 Batch   61/538 - Train Accuracy:  0.928, Validation Accuracy:  0.919, Loss:  2.661
Epoch   2 Batch   62/538 - Train Accuracy:  0.937, Validation Accuracy:  0.921, Loss:  2.579
Epoch   2 Batch   63/538 - Train Accuracy:  0.938, Validation Accuracy:  0.915, Loss:  2.625
Epoch   2 Batch   64/538 - Train Accuracy:  0.922, Validation Accuracy:  0.908, Loss:  2.695
Epoch   2 Batch   65/538 - Train Accuracy:  0.930, Validation Accuracy:  0.908, Loss:  2.632
Epoch   2 Batch   66/538 - Train Accuracy:  0.936, Validation Accuracy:  0.906, Loss:  2.615
Epoch   2 Batch   67/538 - Train Accuracy:  0.931, Validation Accuracy:  0.910, Loss:  2.626
Epoch   2 Batch   68/538 - Train Accuracy:  0.902, Validation Accuracy:  0.906, Loss:  2.658
Epoch   2 Batch   69/538 - Train Accuracy:  0.921, Validation Accuracy:  0.911, Loss:  2.649
Epoch   2 Batch   70/538 - Train Accuracy:  0.908, Validation Accuracy:  0.916, Loss:  2.639
Epoch   2 Batch   71/538 - Train Accuracy:  0.911, Validation Accuracy:  0.928, Loss:  2.625
Epoch   2 Batch   72/538 - Train Accuracy:  0.932, Validation Accuracy:  0.924, Loss:  2.650
Epoch   2 Batch   73/538 - Train Accuracy:  0.915, Validation Accuracy:  0.908, Loss:  2.629
Epoch   2 Batch   74/538 - Train Accuracy:  0.919, Validation Accuracy:  0.905, Loss:  2.652
Epoch   2 Batch   75/538 - Train Accuracy:  0.912, Validation Accuracy:  0.904, Loss:  2.703
Epoch   2 Batch   76/538 - Train Accuracy:  0.929, Validation Accuracy:  0.914, Loss:  2.679
Epoch   2 Batch   77/538 - Train Accuracy:  0.925, Validation Accuracy:  0.916, Loss:  2.623
Epoch   2 Batch   78/538 - Train Accuracy:  0.918, Validation Accuracy:  0.922, Loss:  2.692
Epoch   2 Batch   79/538 - Train Accuracy:  0.937, Validation Accuracy:  0.920, Loss:  2.626
Epoch   2 Batch   80/538 - Train Accuracy:  0.927, Validation Accuracy:  0.912, Loss:  2.646
Epoch   2 Batch   81/538 - Train Accuracy:  0.940, Validation Accuracy:  0.914, Loss:  2.610
Epoch   2 Batch   82/538 - Train Accuracy:  0.906, Validation Accuracy:  0.916, Loss:  2.660
Epoch   2 Batch   83/538 - Train Accuracy:  0.917, Validation Accuracy:  0.912, Loss:  2.646
Epoch   2 Batch   84/538 - Train Accuracy:  0.914, Validation Accuracy:  0.909, Loss:  2.707
Epoch   2 Batch   85/538 - Train Accuracy:  0.936, Validation Accuracy:  0.917, Loss:  2.552
Epoch   2 Batch   86/538 - Train Accuracy:  0.939, Validation Accuracy:  0.914, Loss:  2.558
Epoch   2 Batch   87/538 - Train Accuracy:  0.918, Validation Accuracy:  0.908, Loss:  2.611
Epoch   2 Batch   88/538 - Train Accuracy:  0.916, Validation Accuracy:  0.903, Loss:  2.599
Epoch   2 Batch   89/538 - Train Accuracy:  0.930, Validation Accuracy:  0.904, Loss:  2.619
Epoch   2 Batch   90/538 - Train Accuracy:  0.919, Validation Accuracy:  0.902, Loss:  2.655
Epoch   2 Batch   91/538 - Train Accuracy:  0.915, Validation Accuracy:  0.903, Loss:  2.686
Epoch   2 Batch   92/538 - Train Accuracy:  0.937, Validation Accuracy:  0.917, Loss:  2.647
Epoch   2 Batch   93/538 - Train Accuracy:  0.919, Validation Accuracy:  0.912, Loss:  2.636
Epoch   2 Batch   94/538 - Train Accuracy:  0.921, Validation Accuracy:  0.909, Loss:  2.596
Epoch   2 Batch   95/538 - Train Accuracy:  0.919, Validation Accuracy:  0.915, Loss:  2.640
Epoch   2 Batch   96/538 - Train Accuracy:  0.940, Validation Accuracy:  0.917, Loss:  2.625
Epoch   2 Batch   97/538 - Train Accuracy:  0.939, Validation Accuracy:  0.917, Loss:  2.641
Epoch   2 Batch   98/538 - Train Accuracy:  0.932, Validation Accuracy:  0.914, Loss:  2.643
Epoch   2 Batch   99/538 - Train Accuracy:  0.929, Validation Accuracy:  0.913, Loss:  2.629
Epoch   2 Batch  100/538 - Train Accuracy:  0.919, Validation Accuracy:  0.921, Loss:  2.605
Epoch   2 Batch  101/538 - Train Accuracy:  0.901, Validation Accuracy:  0.920, Loss:  2.653
Epoch   2 Batch  102/538 - Train Accuracy:  0.913, Validation Accuracy:  0.914, Loss:  2.674
Epoch   2 Batch  103/538 - Train Accuracy:  0.922, Validation Accuracy:  0.916, Loss:  2.626
Epoch   2 Batch  104/538 - Train Accuracy:  0.927, Validation Accuracy:  0.915, Loss:  2.620
Epoch   2 Batch  105/538 - Train Accuracy:  0.934, Validation Accuracy:  0.914, Loss:  2.628
Epoch   2 Batch  106/538 - Train Accuracy:  0.905, Validation Accuracy:  0.909, Loss:  2.663
Epoch   2 Batch  107/538 - Train Accuracy:  0.908, Validation Accuracy:  0.920, Loss:  2.636
Epoch   2 Batch  108/538 - Train Accuracy:  0.946, Validation Accuracy:  0.926, Loss:  2.584
Epoch   2 Batch  109/538 - Train Accuracy:  0.938, Validation Accuracy:  0.928, Loss:  2.652
Epoch   2 Batch  110/538 - Train Accuracy:  0.925, Validation Accuracy:  0.926, Loss:  2.668
Epoch   2 Batch  111/538 - Train Accuracy:  0.931, Validation Accuracy:  0.920, Loss:  2.611
Epoch   2 Batch  112/538 - Train Accuracy:  0.935, Validation Accuracy:  0.930, Loss:  2.615
Epoch   2 Batch  113/538 - Train Accuracy:  0.904, Validation Accuracy:  0.922, Loss:  2.659
Epoch   2 Batch  114/538 - Train Accuracy:  0.911, Validation Accuracy:  0.906, Loss:  2.609
Epoch   2 Batch  115/538 - Train Accuracy:  0.928, Validation Accuracy:  0.907, Loss:  2.587
Epoch   2 Batch  116/538 - Train Accuracy:  0.915, Validation Accuracy:  0.912, Loss:  2.638
Epoch   2 Batch  117/538 - Train Accuracy:  0.910, Validation Accuracy:  0.908, Loss:  2.684
Epoch   2 Batch  118/538 - Train Accuracy:  0.944, Validation Accuracy:  0.914, Loss:  2.676
Epoch   2 Batch  119/538 - Train Accuracy:  0.937, Validation Accuracy:  0.909, Loss:  2.640
Epoch   2 Batch  120/538 - Train Accuracy:  0.942, Validation Accuracy:  0.917, Loss:  2.599
Epoch   2 Batch  121/538 - Train Accuracy:  0.922, Validation Accuracy:  0.915, Loss:  2.610
Epoch   2 Batch  122/538 - Train Accuracy:  0.926, Validation Accuracy:  0.911, Loss:  2.600
Epoch   2 Batch  123/538 - Train Accuracy:  0.916, Validation Accuracy:  0.911, Loss:  2.625
Epoch   2 Batch  124/538 - Train Accuracy:  0.939, Validation Accuracy:  0.916, Loss:  2.587
Epoch   2 Batch  125/538 - Train Accuracy:  0.930, Validation Accuracy:  0.916, Loss:  2.656
Epoch   2 Batch  126/538 - Train Accuracy:  0.919, Validation Accuracy:  0.909, Loss:  2.668
Epoch   2 Batch  127/538 - Train Accuracy:  0.915, Validation Accuracy:  0.899, Loss:  2.646
Epoch   2 Batch  128/538 - Train Accuracy:  0.930, Validation Accuracy:  0.906, Loss:  2.601
Epoch   2 Batch  129/538 - Train Accuracy:  0.921, Validation Accuracy:  0.916, Loss:  2.603
Epoch   2 Batch  130/538 - Train Accuracy:  0.921, Validation Accuracy:  0.914, Loss:  2.637
Epoch   2 Batch  131/538 - Train Accuracy:  0.933, Validation Accuracy:  0.917, Loss:  2.655
Epoch   2 Batch  132/538 - Train Accuracy:  0.914, Validation Accuracy:  0.920, Loss:  2.737
Epoch   2 Batch  133/538 - Train Accuracy:  0.917, Validation Accuracy:  0.920, Loss:  2.711
Epoch   2 Batch  134/538 - Train Accuracy:  0.918, Validation Accuracy:  0.917, Loss:  2.682
Epoch   2 Batch  135/538 - Train Accuracy:  0.921, Validation Accuracy:  0.927, Loss:  2.620
Epoch   2 Batch  136/538 - Train Accuracy:  0.918, Validation Accuracy:  0.919, Loss:  2.681
Epoch   2 Batch  137/538 - Train Accuracy:  0.902, Validation Accuracy:  0.918, Loss:  2.690
Epoch   2 Batch  138/538 - Train Accuracy:  0.916, Validation Accuracy:  0.914, Loss:  2.607
Epoch   2 Batch  139/538 - Train Accuracy:  0.922, Validation Accuracy:  0.915, Loss:  2.617
Epoch   2 Batch  140/538 - Train Accuracy:  0.897, Validation Accuracy:  0.927, Loss:  2.632
Epoch   2 Batch  141/538 - Train Accuracy:  0.918, Validation Accuracy:  0.918, Loss:  2.635
Epoch   2 Batch  142/538 - Train Accuracy:  0.933, Validation Accuracy:  0.928, Loss:  2.625
Epoch   2 Batch  143/538 - Train Accuracy:  0.913, Validation Accuracy:  0.931, Loss:  2.633
Epoch   2 Batch  144/538 - Train Accuracy:  0.929, Validation Accuracy:  0.927, Loss:  2.624
Epoch   2 Batch  145/538 - Train Accuracy:  0.909, Validation Accuracy:  0.912, Loss:  2.640
Epoch   2 Batch  146/538 - Train Accuracy:  0.917, Validation Accuracy:  0.923, Loss:  2.632
Epoch   2 Batch  147/538 - Train Accuracy:  0.921, Validation Accuracy:  0.924, Loss:  2.610
Epoch   2 Batch  148/538 - Train Accuracy:  0.903, Validation Accuracy:  0.926, Loss:  2.628
Epoch   2 Batch  149/538 - Train Accuracy:  0.936, Validation Accuracy:  0.931, Loss:  2.658
Epoch   2 Batch  150/538 - Train Accuracy:  0.926, Validation Accuracy:  0.927, Loss:  2.633
Epoch   2 Batch  151/538 - Train Accuracy:  0.939, Validation Accuracy:  0.924, Loss:  2.684
Epoch   2 Batch  152/538 - Train Accuracy:  0.937, Validation Accuracy:  0.920, Loss:  2.602
Epoch   2 Batch  153/538 - Train Accuracy:  0.918, Validation Accuracy:  0.925, Loss:  2.644
Epoch   2 Batch  154/538 - Train Accuracy:  0.927, Validation Accuracy:  0.923, Loss:  2.602
Epoch   2 Batch  155/538 - Train Accuracy:  0.918, Validation Accuracy:  0.918, Loss:  2.644
Epoch   2 Batch  156/538 - Train Accuracy:  0.935, Validation Accuracy:  0.927, Loss:  2.641
Epoch   2 Batch  157/538 - Train Accuracy:  0.924, Validation Accuracy:  0.926, Loss:  2.563
Epoch   2 Batch  158/538 - Train Accuracy:  0.918, Validation Accuracy:  0.927, Loss:  2.633
Epoch   2 Batch  159/538 - Train Accuracy:  0.936, Validation Accuracy:  0.928, Loss:  2.665
Epoch   2 Batch  160/538 - Train Accuracy:  0.915, Validation Accuracy:  0.925, Loss:  2.702
Epoch   2 Batch  161/538 - Train Accuracy:  0.931, Validation Accuracy:  0.933, Loss:  2.630
Epoch   2 Batch  162/538 - Train Accuracy:  0.923, Validation Accuracy:  0.926, Loss:  2.666
Epoch   2 Batch  163/538 - Train Accuracy:  0.926, Validation Accuracy:  0.921, Loss:  2.640
Epoch   2 Batch  164/538 - Train Accuracy:  0.929, Validation Accuracy:  0.921, Loss:  2.715
Epoch   2 Batch  165/538 - Train Accuracy:  0.924, Validation Accuracy:  0.925, Loss:  2.650
Epoch   2 Batch  166/538 - Train Accuracy:  0.944, Validation Accuracy:  0.923, Loss:  2.677
Epoch   2 Batch  167/538 - Train Accuracy:  0.927, Validation Accuracy:  0.924, Loss:  2.654
Epoch   2 Batch  168/538 - Train Accuracy:  0.899, Validation Accuracy:  0.926, Loss:  2.703
Epoch   2 Batch  169/538 - Train Accuracy:  0.936, Validation Accuracy:  0.920, Loss:  2.605
Epoch   2 Batch  170/538 - Train Accuracy:  0.923, Validation Accuracy:  0.924, Loss:  2.608
Epoch   2 Batch  171/538 - Train Accuracy:  0.943, Validation Accuracy:  0.921, Loss:  2.594
Epoch   2 Batch  172/538 - Train Accuracy:  0.931, Validation Accuracy:  0.915, Loss:  2.598
Epoch   2 Batch  173/538 - Train Accuracy:  0.929, Validation Accuracy:  0.914, Loss:  2.595
Epoch   2 Batch  174/538 - Train Accuracy:  0.925, Validation Accuracy:  0.916, Loss:  2.590
Epoch   2 Batch  175/538 - Train Accuracy:  0.938, Validation Accuracy:  0.910, Loss:  2.622
Epoch   2 Batch  176/538 - Train Accuracy:  0.914, Validation Accuracy:  0.918, Loss:  2.561
Epoch   2 Batch  177/538 - Train Accuracy:  0.935, Validation Accuracy:  0.924, Loss:  2.626
Epoch   2 Batch  178/538 - Train Accuracy:  0.917, Validation Accuracy:  0.924, Loss:  2.663
Epoch   2 Batch  179/538 - Train Accuracy:  0.946, Validation Accuracy:  0.922, Loss:  2.580
Epoch   2 Batch  180/538 - Train Accuracy:  0.923, Validation Accuracy:  0.930, Loss:  2.686
Epoch   2 Batch  181/538 - Train Accuracy:  0.917, Validation Accuracy:  0.924, Loss:  2.617
Epoch   2 Batch  182/538 - Train Accuracy:  0.936, Validation Accuracy:  0.922, Loss:  2.631
Epoch   2 Batch  183/538 - Train Accuracy:  0.949, Validation Accuracy:  0.922, Loss:  2.618
Epoch   2 Batch  184/538 - Train Accuracy:  0.930, Validation Accuracy:  0.923, Loss:  2.717
Epoch   2 Batch  185/538 - Train Accuracy:  0.936, Validation Accuracy:  0.924, Loss:  2.609
Epoch   2 Batch  186/538 - Train Accuracy:  0.933, Validation Accuracy:  0.926, Loss:  2.581
Epoch   2 Batch  187/538 - Train Accuracy:  0.934, Validation Accuracy:  0.935, Loss:  2.633
Epoch   2 Batch  188/538 - Train Accuracy:  0.938, Validation Accuracy:  0.939, Loss:  2.659
Epoch   2 Batch  189/538 - Train Accuracy:  0.945, Validation Accuracy:  0.934, Loss:  2.632
Epoch   2 Batch  190/538 - Train Accuracy:  0.915, Validation Accuracy:  0.934, Loss:  2.683
Epoch   2 Batch  191/538 - Train Accuracy:  0.943, Validation Accuracy:  0.923, Loss:  2.541
Epoch   2 Batch  192/538 - Train Accuracy:  0.935, Validation Accuracy:  0.936, Loss:  2.621
Epoch   2 Batch  193/538 - Train Accuracy:  0.927, Validation Accuracy:  0.940, Loss:  2.634
Epoch   2 Batch  194/538 - Train Accuracy:  0.903, Validation Accuracy:  0.939, Loss:  2.587
Epoch   2 Batch  195/538 - Train Accuracy:  0.940, Validation Accuracy:  0.931, Loss:  2.698
Epoch   2 Batch  196/538 - Train Accuracy:  0.919, Validation Accuracy:  0.921, Loss:  2.563
Epoch   2 Batch  197/538 - Train Accuracy:  0.933, Validation Accuracy:  0.917, Loss:  2.645
Epoch   2 Batch  198/538 - Train Accuracy:  0.943, Validation Accuracy:  0.922, Loss:  2.631
Epoch   2 Batch  199/538 - Train Accuracy:  0.908, Validation Accuracy:  0.928, Loss:  2.602
Epoch   2 Batch  200/538 - Train Accuracy:  0.951, Validation Accuracy:  0.925, Loss:  2.648
Epoch   2 Batch  201/538 - Train Accuracy:  0.921, Validation Accuracy:  0.925, Loss:  2.573
Epoch   2 Batch  202/538 - Train Accuracy:  0.937, Validation Accuracy:  0.924, Loss:  2.649
Epoch   2 Batch  203/538 - Train Accuracy:  0.923, Validation Accuracy:  0.922, Loss:  2.606
Epoch   2 Batch  204/538 - Train Accuracy:  0.925, Validation Accuracy:  0.926, Loss:  2.650
Epoch   2 Batch  205/538 - Train Accuracy:  0.930, Validation Accuracy:  0.919, Loss:  2.658
Epoch   2 Batch  206/538 - Train Accuracy:  0.923, Validation Accuracy:  0.920, Loss:  2.644
Epoch   2 Batch  207/538 - Train Accuracy:  0.942, Validation Accuracy:  0.923, Loss:  2.582
Epoch   2 Batch  208/538 - Train Accuracy:  0.925, Validation Accuracy:  0.924, Loss:  2.676
Epoch   2 Batch  209/538 - Train Accuracy:  0.935, Validation Accuracy:  0.922, Loss:  2.572
Epoch   2 Batch  210/538 - Train Accuracy:  0.912, Validation Accuracy:  0.933, Loss:  2.671
Epoch   2 Batch  211/538 - Train Accuracy:  0.928, Validation Accuracy:  0.933, Loss:  2.606
Epoch   2 Batch  212/538 - Train Accuracy:  0.921, Validation Accuracy:  0.931, Loss:  2.572
Epoch   2 Batch  213/538 - Train Accuracy:  0.933, Validation Accuracy:  0.937, Loss:  2.651
Epoch   2 Batch  214/538 - Train Accuracy:  0.941, Validation Accuracy:  0.942, Loss:  2.603
Epoch   2 Batch  215/538 - Train Accuracy:  0.941, Validation Accuracy:  0.933, Loss:  2.621
Epoch   2 Batch  216/538 - Train Accuracy:  0.956, Validation Accuracy:  0.927, Loss:  2.672
Epoch   2 Batch  217/538 - Train Accuracy:  0.940, Validation Accuracy:  0.923, Loss:  2.655
Epoch   2 Batch  218/538 - Train Accuracy:  0.928, Validation Accuracy:  0.927, Loss:  2.653
Epoch   2 Batch  219/538 - Train Accuracy:  0.922, Validation Accuracy:  0.931, Loss:  2.612
Epoch   2 Batch  220/538 - Train Accuracy:  0.926, Validation Accuracy:  0.932, Loss:  2.595
Epoch   2 Batch  221/538 - Train Accuracy:  0.940, Validation Accuracy:  0.929, Loss:  2.676
Epoch   2 Batch  222/538 - Train Accuracy:  0.902, Validation Accuracy:  0.927, Loss:  2.673
Epoch   2 Batch  223/538 - Train Accuracy:  0.931, Validation Accuracy:  0.916, Loss:  2.654
Epoch   2 Batch  224/538 - Train Accuracy:  0.928, Validation Accuracy:  0.915, Loss:  2.703
Epoch   2 Batch  225/538 - Train Accuracy:  0.936, Validation Accuracy:  0.916, Loss:  2.651
Epoch   2 Batch  226/538 - Train Accuracy:  0.921, Validation Accuracy:  0.917, Loss:  2.663
Epoch   2 Batch  227/538 - Train Accuracy:  0.936, Validation Accuracy:  0.924, Loss:  2.616
Epoch   2 Batch  228/538 - Train Accuracy:  0.907, Validation Accuracy:  0.934, Loss:  2.604
Epoch   2 Batch  229/538 - Train Accuracy:  0.940, Validation Accuracy:  0.937, Loss:  2.624
Epoch   2 Batch  230/538 - Train Accuracy:  0.914, Validation Accuracy:  0.928, Loss:  2.625
Epoch   2 Batch  231/538 - Train Accuracy:  0.933, Validation Accuracy:  0.927, Loss:  2.606
Epoch   2 Batch  232/538 - Train Accuracy:  0.932, Validation Accuracy:  0.935, Loss:  2.672
Epoch   2 Batch  233/538 - Train Accuracy:  0.932, Validation Accuracy:  0.933, Loss:  2.639
Epoch   2 Batch  234/538 - Train Accuracy:  0.947, Validation Accuracy:  0.930, Loss:  2.648
Epoch   2 Batch  235/538 - Train Accuracy:  0.938, Validation Accuracy:  0.927, Loss:  2.585
Epoch   2 Batch  236/538 - Train Accuracy:  0.933, Validation Accuracy:  0.925, Loss:  2.712
Epoch   2 Batch  237/538 - Train Accuracy:  0.930, Validation Accuracy:  0.919, Loss:  2.602
Epoch   2 Batch  238/538 - Train Accuracy:  0.945, Validation Accuracy:  0.933, Loss:  2.646
Epoch   2 Batch  239/538 - Train Accuracy:  0.938, Validation Accuracy:  0.939, Loss:  2.675
Epoch   2 Batch  240/538 - Train Accuracy:  0.927, Validation Accuracy:  0.933, Loss:  2.646
Epoch   2 Batch  241/538 - Train Accuracy:  0.940, Validation Accuracy:  0.934, Loss:  2.634
Epoch   2 Batch  242/538 - Train Accuracy:  0.954, Validation Accuracy:  0.934, Loss:  2.657
Epoch   2 Batch  243/538 - Train Accuracy:  0.938, Validation Accuracy:  0.934, Loss:  2.656
Epoch   2 Batch  244/538 - Train Accuracy:  0.923, Validation Accuracy:  0.937, Loss:  2.649
Epoch   2 Batch  245/538 - Train Accuracy:  0.924, Validation Accuracy:  0.939, Loss:  2.662
Epoch   2 Batch  246/538 - Train Accuracy:  0.940, Validation Accuracy:  0.941, Loss:  2.630
Epoch   2 Batch  247/538 - Train Accuracy:  0.913, Validation Accuracy:  0.940, Loss:  2.637
Epoch   2 Batch  248/538 - Train Accuracy:  0.941, Validation Accuracy:  0.937, Loss:  2.693
Epoch   2 Batch  249/538 - Train Accuracy:  0.935, Validation Accuracy:  0.938, Loss:  2.659
Epoch   2 Batch  250/538 - Train Accuracy:  0.936, Validation Accuracy:  0.947, Loss:  2.640
Epoch   2 Batch  251/538 - Train Accuracy:  0.944, Validation Accuracy:  0.940, Loss:  2.637
Epoch   2 Batch  252/538 - Train Accuracy:  0.946, Validation Accuracy:  0.940, Loss:  2.629
Epoch   2 Batch  253/538 - Train Accuracy:  0.921, Validation Accuracy:  0.939, Loss:  2.616
Epoch   2 Batch  254/538 - Train Accuracy:  0.928, Validation Accuracy:  0.943, Loss:  2.677
Epoch   2 Batch  255/538 - Train Accuracy:  0.953, Validation Accuracy:  0.940, Loss:  2.616
Epoch   2 Batch  256/538 - Train Accuracy:  0.924, Validation Accuracy:  0.934, Loss:  2.660
Epoch   2 Batch  257/538 - Train Accuracy:  0.936, Validation Accuracy:  0.946, Loss:  2.566
Epoch   2 Batch  258/538 - Train Accuracy:  0.939, Validation Accuracy:  0.943, Loss:  2.694
Epoch   2 Batch  259/538 - Train Accuracy:  0.953, Validation Accuracy:  0.936, Loss:  2.666
Epoch   2 Batch  260/538 - Train Accuracy:  0.912, Validation Accuracy:  0.926, Loss:  2.712
Epoch   2 Batch  261/538 - Train Accuracy:  0.927, Validation Accuracy:  0.929, Loss:  2.633
Epoch   2 Batch  262/538 - Train Accuracy:  0.961, Validation Accuracy:  0.931, Loss:  2.666
Epoch   2 Batch  263/538 - Train Accuracy:  0.928, Validation Accuracy:  0.938, Loss:  2.611
Epoch   2 Batch  264/538 - Train Accuracy:  0.913, Validation Accuracy:  0.923, Loss:  2.617
Epoch   2 Batch  265/538 - Train Accuracy:  0.927, Validation Accuracy:  0.923, Loss:  2.637
Epoch   2 Batch  266/538 - Train Accuracy:  0.928, Validation Accuracy:  0.921, Loss:  2.587
Epoch   2 Batch  267/538 - Train Accuracy:  0.927, Validation Accuracy:  0.932, Loss:  2.630
Epoch   2 Batch  268/538 - Train Accuracy:  0.959, Validation Accuracy:  0.929, Loss:  2.669
Epoch   2 Batch  269/538 - Train Accuracy:  0.937, Validation Accuracy:  0.935, Loss:  2.632
Epoch   2 Batch  270/538 - Train Accuracy:  0.945, Validation Accuracy:  0.930, Loss:  2.624
Epoch   2 Batch  271/538 - Train Accuracy:  0.923, Validation Accuracy:  0.928, Loss:  2.638
Epoch   2 Batch  272/538 - Train Accuracy:  0.935, Validation Accuracy:  0.930, Loss:  2.661
Epoch   2 Batch  273/538 - Train Accuracy:  0.941, Validation Accuracy:  0.924, Loss:  2.676
Epoch   2 Batch  274/538 - Train Accuracy:  0.904, Validation Accuracy:  0.930, Loss:  2.656
Epoch   2 Batch  275/538 - Train Accuracy:  0.926, Validation Accuracy:  0.932, Loss:  2.639
Epoch   2 Batch  276/538 - Train Accuracy:  0.923, Validation Accuracy:  0.923, Loss:  2.568
Epoch   2 Batch  277/538 - Train Accuracy:  0.934, Validation Accuracy:  0.926, Loss:  2.635
Epoch   2 Batch  278/538 - Train Accuracy:  0.937, Validation Accuracy:  0.933, Loss:  2.592
Epoch   2 Batch  279/538 - Train Accuracy:  0.933, Validation Accuracy:  0.939, Loss:  2.586
Epoch   2 Batch  280/538 - Train Accuracy:  0.950, Validation Accuracy:  0.939, Loss:  2.684
Epoch   2 Batch  281/538 - Train Accuracy:  0.922, Validation Accuracy:  0.938, Loss:  2.680
Epoch   2 Batch  282/538 - Train Accuracy:  0.943, Validation Accuracy:  0.938, Loss:  2.630
Epoch   2 Batch  283/538 - Train Accuracy:  0.947, Validation Accuracy:  0.935, Loss:  2.576
Epoch   2 Batch  284/538 - Train Accuracy:  0.929, Validation Accuracy:  0.939, Loss:  2.643
Epoch   2 Batch  285/538 - Train Accuracy:  0.943, Validation Accuracy:  0.934, Loss:  2.667
Epoch   2 Batch  286/538 - Train Accuracy:  0.928, Validation Accuracy:  0.931, Loss:  2.617
Epoch   2 Batch  287/538 - Train Accuracy:  0.954, Validation Accuracy:  0.930, Loss:  2.605
Epoch   2 Batch  288/538 - Train Accuracy:  0.943, Validation Accuracy:  0.933, Loss:  2.700
Epoch   2 Batch  289/538 - Train Accuracy:  0.945, Validation Accuracy:  0.938, Loss:  2.580
Epoch   2 Batch  290/538 - Train Accuracy:  0.958, Validation Accuracy:  0.938, Loss:  2.599
Epoch   2 Batch  291/538 - Train Accuracy:  0.940, Validation Accuracy:  0.944, Loss:  2.633
Epoch   2 Batch  292/538 - Train Accuracy:  0.953, Validation Accuracy:  0.933, Loss:  2.582
Epoch   2 Batch  293/538 - Train Accuracy:  0.942, Validation Accuracy:  0.930, Loss:  2.608
Epoch   2 Batch  294/538 - Train Accuracy:  0.940, Validation Accuracy:  0.931, Loss:  2.633
Epoch   2 Batch  295/538 - Train Accuracy:  0.945, Validation Accuracy:  0.932, Loss:  2.662
Epoch   2 Batch  296/538 - Train Accuracy:  0.939, Validation Accuracy:  0.933, Loss:  2.696
Epoch   2 Batch  297/538 - Train Accuracy:  0.953, Validation Accuracy:  0.936, Loss:  2.719
Epoch   2 Batch  298/538 - Train Accuracy:  0.934, Validation Accuracy:  0.933, Loss:  2.644
Epoch   2 Batch  299/538 - Train Accuracy:  0.924, Validation Accuracy:  0.933, Loss:  2.683
Epoch   2 Batch  300/538 - Train Accuracy:  0.926, Validation Accuracy:  0.936, Loss:  2.576
Epoch   2 Batch  301/538 - Train Accuracy:  0.924, Validation Accuracy:  0.934, Loss:  2.632
Epoch   2 Batch  302/538 - Train Accuracy:  0.949, Validation Accuracy:  0.936, Loss:  2.615
Epoch   2 Batch  303/538 - Train Accuracy:  0.944, Validation Accuracy:  0.939, Loss:  2.632
Epoch   2 Batch  304/538 - Train Accuracy:  0.936, Validation Accuracy:  0.937, Loss:  2.619
Epoch   2 Batch  305/538 - Train Accuracy:  0.942, Validation Accuracy:  0.935, Loss:  2.628
Epoch   2 Batch  306/538 - Train Accuracy:  0.921, Validation Accuracy:  0.933, Loss:  2.648
Epoch   2 Batch  307/538 - Train Accuracy:  0.942, Validation Accuracy:  0.935, Loss:  2.647
Epoch   2 Batch  308/538 - Train Accuracy:  0.940, Validation Accuracy:  0.935, Loss:  2.630
Epoch   2 Batch  309/538 - Train Accuracy:  0.949, Validation Accuracy:  0.943, Loss:  2.671
Epoch   2 Batch  310/538 - Train Accuracy:  0.943, Validation Accuracy:  0.944, Loss:  2.640
Epoch   2 Batch  311/538 - Train Accuracy:  0.926, Validation Accuracy:  0.942, Loss:  2.584
Epoch   2 Batch  312/538 - Train Accuracy:  0.952, Validation Accuracy:  0.939, Loss:  2.609
Epoch   2 Batch  313/538 - Train Accuracy:  0.936, Validation Accuracy:  0.940, Loss:  2.607
Epoch   2 Batch  314/538 - Train Accuracy:  0.937, Validation Accuracy:  0.943, Loss:  2.585
Epoch   2 Batch  315/538 - Train Accuracy:  0.927, Validation Accuracy:  0.941, Loss:  2.659
Epoch   2 Batch  316/538 - Train Accuracy:  0.927, Validation Accuracy:  0.940, Loss:  2.593
Epoch   2 Batch  317/538 - Train Accuracy:  0.931, Validation Accuracy:  0.934, Loss:  2.629
Epoch   2 Batch  318/538 - Train Accuracy:  0.928, Validation Accuracy:  0.925, Loss:  2.571
Epoch   2 Batch  319/538 - Train Accuracy:  0.943, Validation Accuracy:  0.924, Loss:  2.597
Epoch   2 Batch  320/538 - Train Accuracy:  0.932, Validation Accuracy:  0.933, Loss:  2.640
Epoch   2 Batch  321/538 - Train Accuracy:  0.940, Validation Accuracy:  0.935, Loss:  2.697
Epoch   2 Batch  322/538 - Train Accuracy:  0.941, Validation Accuracy:  0.939, Loss:  2.609
Epoch   2 Batch  323/538 - Train Accuracy:  0.954, Validation Accuracy:  0.937, Loss:  2.595
Epoch   2 Batch  324/538 - Train Accuracy:  0.946, Validation Accuracy:  0.936, Loss:  2.653
Epoch   2 Batch  325/538 - Train Accuracy:  0.947, Validation Accuracy:  0.936, Loss:  2.626
Epoch   2 Batch  326/538 - Train Accuracy:  0.949, Validation Accuracy:  0.933, Loss:  2.548
Epoch   2 Batch  327/538 - Train Accuracy:  0.928, Validation Accuracy:  0.925, Loss:  2.630
Epoch   2 Batch  328/538 - Train Accuracy:  0.951, Validation Accuracy:  0.926, Loss:  2.626
Epoch   2 Batch  329/538 - Train Accuracy:  0.945, Validation Accuracy:  0.929, Loss:  2.683
Epoch   2 Batch  330/538 - Train Accuracy:  0.959, Validation Accuracy:  0.940, Loss:  2.644
Epoch   2 Batch  331/538 - Train Accuracy:  0.948, Validation Accuracy:  0.928, Loss:  2.632
Epoch   2 Batch  332/538 - Train Accuracy:  0.944, Validation Accuracy:  0.922, Loss:  2.606
Epoch   2 Batch  333/538 - Train Accuracy:  0.948, Validation Accuracy:  0.930, Loss:  2.575
Epoch   2 Batch  334/538 - Train Accuracy:  0.947, Validation Accuracy:  0.925, Loss:  2.586
Epoch   2 Batch  335/538 - Train Accuracy:  0.925, Validation Accuracy:  0.929, Loss:  2.667
Epoch   2 Batch  336/538 - Train Accuracy:  0.933, Validation Accuracy:  0.942, Loss:  2.658
Epoch   2 Batch  337/538 - Train Accuracy:  0.945, Validation Accuracy:  0.951, Loss:  2.600
Epoch   2 Batch  338/538 - Train Accuracy:  0.940, Validation Accuracy:  0.948, Loss:  2.645
Epoch   2 Batch  339/538 - Train Accuracy:  0.953, Validation Accuracy:  0.957, Loss:  2.620
Epoch   2 Batch  340/538 - Train Accuracy:  0.932, Validation Accuracy:  0.943, Loss:  2.623
Epoch   2 Batch  341/538 - Train Accuracy:  0.935, Validation Accuracy:  0.938, Loss:  2.658
Epoch   2 Batch  342/538 - Train Accuracy:  0.932, Validation Accuracy:  0.942, Loss:  2.660
Epoch   2 Batch  343/538 - Train Accuracy:  0.942, Validation Accuracy:  0.944, Loss:  2.654
Epoch   2 Batch  344/538 - Train Accuracy:  0.931, Validation Accuracy:  0.948, Loss:  2.611
Epoch   2 Batch  345/538 - Train Accuracy:  0.939, Validation Accuracy:  0.942, Loss:  2.650
Epoch   2 Batch  346/538 - Train Accuracy:  0.920, Validation Accuracy:  0.941, Loss:  2.595
Epoch   2 Batch  347/538 - Train Accuracy:  0.945, Validation Accuracy:  0.938, Loss:  2.636
Epoch   2 Batch  348/538 - Train Accuracy:  0.926, Validation Accuracy:  0.941, Loss:  2.640
Epoch   2 Batch  349/538 - Train Accuracy:  0.953, Validation Accuracy:  0.942, Loss:  2.641
Epoch   2 Batch  350/538 - Train Accuracy:  0.941, Validation Accuracy:  0.941, Loss:  2.668
Epoch   2 Batch  351/538 - Train Accuracy:  0.928, Validation Accuracy:  0.952, Loss:  2.624
Epoch   2 Batch  352/538 - Train Accuracy:  0.931, Validation Accuracy:  0.952, Loss:  2.699
Epoch   2 Batch  353/538 - Train Accuracy:  0.929, Validation Accuracy:  0.944, Loss:  2.628
Epoch   2 Batch  354/538 - Train Accuracy:  0.941, Validation Accuracy:  0.950, Loss:  2.576
Epoch   2 Batch  355/538 - Train Accuracy:  0.935, Validation Accuracy:  0.942, Loss:  2.667
Epoch   2 Batch  356/538 - Train Accuracy:  0.937, Validation Accuracy:  0.942, Loss:  2.596
Epoch   2 Batch  357/538 - Train Accuracy:  0.941, Validation Accuracy:  0.940, Loss:  2.666
Epoch   2 Batch  358/538 - Train Accuracy:  0.954, Validation Accuracy:  0.929, Loss:  2.626
Epoch   2 Batch  359/538 - Train Accuracy:  0.939, Validation Accuracy:  0.936, Loss:  2.597
Epoch   2 Batch  360/538 - Train Accuracy:  0.930, Validation Accuracy:  0.935, Loss:  2.604
Epoch   2 Batch  361/538 - Train Accuracy:  0.958, Validation Accuracy:  0.940, Loss:  2.633
Epoch   2 Batch  362/538 - Train Accuracy:  0.948, Validation Accuracy:  0.942, Loss:  2.612
Epoch   2 Batch  363/538 - Train Accuracy:  0.937, Validation Accuracy:  0.941, Loss:  2.654
Epoch   2 Batch  364/538 - Train Accuracy:  0.923, Validation Accuracy:  0.944, Loss:  2.603
Epoch   2 Batch  365/538 - Train Accuracy:  0.931, Validation Accuracy:  0.942, Loss:  2.641
Epoch   2 Batch  366/538 - Train Accuracy:  0.948, Validation Accuracy:  0.939, Loss:  2.625
Epoch   2 Batch  367/538 - Train Accuracy:  0.953, Validation Accuracy:  0.938, Loss:  2.628
Epoch   2 Batch  368/538 - Train Accuracy:  0.959, Validation Accuracy:  0.942, Loss:  2.652
Epoch   2 Batch  369/538 - Train Accuracy:  0.937, Validation Accuracy:  0.939, Loss:  2.595
Epoch   2 Batch  370/538 - Train Accuracy:  0.933, Validation Accuracy:  0.927, Loss:  2.665
Epoch   2 Batch  371/538 - Train Accuracy:  0.951, Validation Accuracy:  0.939, Loss:  2.563
Epoch   2 Batch  372/538 - Train Accuracy:  0.954, Validation Accuracy:  0.947, Loss:  2.658
Epoch   2 Batch  373/538 - Train Accuracy:  0.938, Validation Accuracy:  0.950, Loss:  2.632
Epoch   2 Batch  374/538 - Train Accuracy:  0.953, Validation Accuracy:  0.948, Loss:  2.639
Epoch   2 Batch  375/538 - Train Accuracy:  0.945, Validation Accuracy:  0.930, Loss:  2.572
Epoch   2 Batch  376/538 - Train Accuracy:  0.935, Validation Accuracy:  0.930, Loss:  2.598
Epoch   2 Batch  377/538 - Train Accuracy:  0.940, Validation Accuracy:  0.937, Loss:  2.658
Epoch   2 Batch  378/538 - Train Accuracy:  0.960, Validation Accuracy:  0.937, Loss:  2.676
Epoch   2 Batch  379/538 - Train Accuracy:  0.949, Validation Accuracy:  0.937, Loss:  2.594
Epoch   2 Batch  380/538 - Train Accuracy:  0.940, Validation Accuracy:  0.935, Loss:  2.668
Epoch   2 Batch  381/538 - Train Accuracy:  0.960, Validation Accuracy:  0.937, Loss:  2.654
Epoch   2 Batch  382/538 - Train Accuracy:  0.929, Validation Accuracy:  0.959, Loss:  2.591
Epoch   2 Batch  383/538 - Train Accuracy:  0.946, Validation Accuracy:  0.950, Loss:  2.655
Epoch   2 Batch  384/538 - Train Accuracy:  0.936, Validation Accuracy:  0.947, Loss:  2.617
Epoch   2 Batch  385/538 - Train Accuracy:  0.940, Validation Accuracy:  0.953, Loss:  2.637
Epoch   2 Batch  386/538 - Train Accuracy:  0.952, Validation Accuracy:  0.955, Loss:  2.583
Epoch   2 Batch  387/538 - Train Accuracy:  0.939, Validation Accuracy:  0.950, Loss:  2.589
Epoch   2 Batch  388/538 - Train Accuracy:  0.949, Validation Accuracy:  0.957, Loss:  2.595
Epoch   2 Batch  389/538 - Train Accuracy:  0.931, Validation Accuracy:  0.946, Loss:  2.654
Epoch   2 Batch  390/538 - Train Accuracy:  0.945, Validation Accuracy:  0.934, Loss:  2.665
Epoch   2 Batch  391/538 - Train Accuracy:  0.930, Validation Accuracy:  0.927, Loss:  2.639
Epoch   2 Batch  392/538 - Train Accuracy:  0.942, Validation Accuracy:  0.937, Loss:  2.582
Epoch   2 Batch  393/538 - Train Accuracy:  0.948, Validation Accuracy:  0.944, Loss:  2.621
Epoch   2 Batch  394/538 - Train Accuracy:  0.921, Validation Accuracy:  0.937, Loss:  2.615
Epoch   2 Batch  395/538 - Train Accuracy:  0.925, Validation Accuracy:  0.935, Loss:  2.604
Epoch   2 Batch  396/538 - Train Accuracy:  0.937, Validation Accuracy:  0.945, Loss:  2.624
Epoch   2 Batch  397/538 - Train Accuracy:  0.942, Validation Accuracy:  0.941, Loss:  2.608
Epoch   2 Batch  398/538 - Train Accuracy:  0.929, Validation Accuracy:  0.937, Loss:  2.650
Epoch   2 Batch  399/538 - Train Accuracy:  0.930, Validation Accuracy:  0.940, Loss:  2.667
Epoch   2 Batch  400/538 - Train Accuracy:  0.954, Validation Accuracy:  0.936, Loss:  2.600
Epoch   2 Batch  401/538 - Train Accuracy:  0.942, Validation Accuracy:  0.937, Loss:  2.658
Epoch   2 Batch  402/538 - Train Accuracy:  0.955, Validation Accuracy:  0.944, Loss:  2.625
Epoch   2 Batch  403/538 - Train Accuracy:  0.946, Validation Accuracy:  0.946, Loss:  2.621
Epoch   2 Batch  404/538 - Train Accuracy:  0.934, Validation Accuracy:  0.938, Loss:  2.582
Epoch   2 Batch  405/538 - Train Accuracy:  0.944, Validation Accuracy:  0.941, Loss:  2.635
Epoch   2 Batch  406/538 - Train Accuracy:  0.938, Validation Accuracy:  0.942, Loss:  2.588
Epoch   2 Batch  407/538 - Train Accuracy:  0.950, Validation Accuracy:  0.945, Loss:  2.603
Epoch   2 Batch  408/538 - Train Accuracy:  0.941, Validation Accuracy:  0.953, Loss:  2.654
Epoch   2 Batch  409/538 - Train Accuracy:  0.935, Validation Accuracy:  0.951, Loss:  2.602
Epoch   2 Batch  410/538 - Train Accuracy:  0.940, Validation Accuracy:  0.951, Loss:  2.580
Epoch   2 Batch  411/538 - Train Accuracy:  0.963, Validation Accuracy:  0.948, Loss:  2.604
Epoch   2 Batch  412/538 - Train Accuracy:  0.951, Validation Accuracy:  0.944, Loss:  2.705
Epoch   2 Batch  413/538 - Train Accuracy:  0.941, Validation Accuracy:  0.940, Loss:  2.632
Epoch   2 Batch  414/538 - Train Accuracy:  0.926, Validation Accuracy:  0.942, Loss:  2.612
Epoch   2 Batch  415/538 - Train Accuracy:  0.931, Validation Accuracy:  0.946, Loss:  2.669
Epoch   2 Batch  416/538 - Train Accuracy:  0.938, Validation Accuracy:  0.944, Loss:  2.635
Epoch   2 Batch  417/538 - Train Accuracy:  0.945, Validation Accuracy:  0.941, Loss:  2.705
Epoch   2 Batch  418/538 - Train Accuracy:  0.945, Validation Accuracy:  0.939, Loss:  2.690
Epoch   2 Batch  419/538 - Train Accuracy:  0.954, Validation Accuracy:  0.944, Loss:  2.635
Epoch   2 Batch  420/538 - Train Accuracy:  0.958, Validation Accuracy:  0.941, Loss:  2.600
Epoch   2 Batch  421/538 - Train Accuracy:  0.941, Validation Accuracy:  0.942, Loss:  2.623
Epoch   2 Batch  422/538 - Train Accuracy:  0.944, Validation Accuracy:  0.945, Loss:  2.663
Epoch   2 Batch  423/538 - Train Accuracy:  0.946, Validation Accuracy:  0.949, Loss:  2.660
Epoch   2 Batch  424/538 - Train Accuracy:  0.938, Validation Accuracy:  0.938, Loss:  2.554
Epoch   2 Batch  425/538 - Train Accuracy:  0.932, Validation Accuracy:  0.938, Loss:  2.639
Epoch   2 Batch  426/538 - Train Accuracy:  0.941, Validation Accuracy:  0.938, Loss:  2.669
Epoch   2 Batch  427/538 - Train Accuracy:  0.944, Validation Accuracy:  0.935, Loss:  2.667
Epoch   2 Batch  428/538 - Train Accuracy:  0.947, Validation Accuracy:  0.940, Loss:  2.589
Epoch   2 Batch  429/538 - Train Accuracy:  0.951, Validation Accuracy:  0.944, Loss:  2.642
Epoch   2 Batch  430/538 - Train Accuracy:  0.938, Validation Accuracy:  0.940, Loss:  2.647
Epoch   2 Batch  431/538 - Train Accuracy:  0.942, Validation Accuracy:  0.938, Loss:  2.572
Epoch   2 Batch  432/538 - Train Accuracy:  0.945, Validation Accuracy:  0.941, Loss:  2.609
Epoch   2 Batch  433/538 - Train Accuracy:  0.941, Validation Accuracy:  0.940, Loss:  2.678
Epoch   2 Batch  434/538 - Train Accuracy:  0.936, Validation Accuracy:  0.945, Loss:  2.609
Epoch   2 Batch  435/538 - Train Accuracy:  0.938, Validation Accuracy:  0.947, Loss:  2.631
Epoch   2 Batch  436/538 - Train Accuracy:  0.936, Validation Accuracy:  0.944, Loss:  2.636
Epoch   2 Batch  437/538 - Train Accuracy:  0.943, Validation Accuracy:  0.950, Loss:  2.614
Epoch   2 Batch  438/538 - Train Accuracy:  0.946, Validation Accuracy:  0.949, Loss:  2.580
Epoch   2 Batch  439/538 - Train Accuracy:  0.946, Validation Accuracy:  0.947, Loss:  2.633
Epoch   2 Batch  440/538 - Train Accuracy:  0.934, Validation Accuracy:  0.945, Loss:  2.619
Epoch   2 Batch  441/538 - Train Accuracy:  0.930, Validation Accuracy:  0.943, Loss:  2.666
Epoch   2 Batch  442/538 - Train Accuracy:  0.945, Validation Accuracy:  0.939, Loss:  2.591
Epoch   2 Batch  443/538 - Train Accuracy:  0.953, Validation Accuracy:  0.938, Loss:  2.660
Epoch   2 Batch  444/538 - Train Accuracy:  0.945, Validation Accuracy:  0.941, Loss:  2.657
Epoch   2 Batch  445/538 - Train Accuracy:  0.953, Validation Accuracy:  0.937, Loss:  2.639
Epoch   2 Batch  446/538 - Train Accuracy:  0.936, Validation Accuracy:  0.935, Loss:  2.641
Epoch   2 Batch  447/538 - Train Accuracy:  0.941, Validation Accuracy:  0.941, Loss:  2.660
Epoch   2 Batch  448/538 - Train Accuracy:  0.945, Validation Accuracy:  0.940, Loss:  2.670
Epoch   2 Batch  449/538 - Train Accuracy:  0.947, Validation Accuracy:  0.941, Loss:  2.693
Epoch   2 Batch  450/538 - Train Accuracy:  0.933, Validation Accuracy:  0.936, Loss:  2.602
Epoch   2 Batch  451/538 - Train Accuracy:  0.925, Validation Accuracy:  0.934, Loss:  2.605
Epoch   2 Batch  452/538 - Train Accuracy:  0.951, Validation Accuracy:  0.946, Loss:  2.670
Epoch   2 Batch  453/538 - Train Accuracy:  0.940, Validation Accuracy:  0.949, Loss:  2.608
Epoch   2 Batch  454/538 - Train Accuracy:  0.948, Validation Accuracy:  0.947, Loss:  2.646
Epoch   2 Batch  455/538 - Train Accuracy:  0.942, Validation Accuracy:  0.945, Loss:  2.624
Epoch   2 Batch  456/538 - Train Accuracy:  0.943, Validation Accuracy:  0.950, Loss:  2.555
Epoch   2 Batch  457/538 - Train Accuracy:  0.952, Validation Accuracy:  0.956, Loss:  2.614
Epoch   2 Batch  458/538 - Train Accuracy:  0.944, Validation Accuracy:  0.950, Loss:  2.644
Epoch   2 Batch  459/538 - Train Accuracy:  0.965, Validation Accuracy:  0.951, Loss:  2.578
Epoch   2 Batch  460/538 - Train Accuracy:  0.938, Validation Accuracy:  0.949, Loss:  2.630
Epoch   2 Batch  461/538 - Train Accuracy:  0.954, Validation Accuracy:  0.950, Loss:  2.691
Epoch   2 Batch  462/538 - Train Accuracy:  0.945, Validation Accuracy:  0.949, Loss:  2.604
Epoch   2 Batch  463/538 - Train Accuracy:  0.924, Validation Accuracy:  0.948, Loss:  2.664
Epoch   2 Batch  464/538 - Train Accuracy:  0.961, Validation Accuracy:  0.949, Loss:  2.612
Epoch   2 Batch  465/538 - Train Accuracy:  0.956, Validation Accuracy:  0.950, Loss:  2.621
Epoch   2 Batch  466/538 - Train Accuracy:  0.946, Validation Accuracy:  0.953, Loss:  2.646
Epoch   2 Batch  467/538 - Train Accuracy:  0.946, Validation Accuracy:  0.951, Loss:  2.583
Epoch   2 Batch  468/538 - Train Accuracy:  0.969, Validation Accuracy:  0.952, Loss:  2.597
Epoch   2 Batch  469/538 - Train Accuracy:  0.949, Validation Accuracy:  0.948, Loss:  2.590
Epoch   2 Batch  470/538 - Train Accuracy:  0.949, Validation Accuracy:  0.950, Loss:  2.599
Epoch   2 Batch  471/538 - Train Accuracy:  0.960, Validation Accuracy:  0.946, Loss:  2.617
Epoch   2 Batch  472/538 - Train Accuracy:  0.977, Validation Accuracy:  0.941, Loss:  2.657
Epoch   2 Batch  473/538 - Train Accuracy:  0.938, Validation Accuracy:  0.939, Loss:  2.662
Epoch   2 Batch  474/538 - Train Accuracy:  0.965, Validation Accuracy:  0.942, Loss:  2.562
Epoch   2 Batch  475/538 - Train Accuracy:  0.953, Validation Accuracy:  0.945, Loss:  2.609
Epoch   2 Batch  476/538 - Train Accuracy:  0.949, Validation Accuracy:  0.943, Loss:  2.614
Epoch   2 Batch  477/538 - Train Accuracy:  0.941, Validation Accuracy:  0.943, Loss:  2.602
Epoch   2 Batch  478/538 - Train Accuracy:  0.948, Validation Accuracy:  0.941, Loss:  2.601
Epoch   2 Batch  479/538 - Train Accuracy:  0.947, Validation Accuracy:  0.949, Loss:  2.610
Epoch   2 Batch  480/538 - Train Accuracy:  0.954, Validation Accuracy:  0.951, Loss:  2.604
Epoch   2 Batch  481/538 - Train Accuracy:  0.945, Validation Accuracy:  0.941, Loss:  2.622
Epoch   2 Batch  482/538 - Train Accuracy:  0.949, Validation Accuracy:  0.942, Loss:  2.619
Epoch   2 Batch  483/538 - Train Accuracy:  0.927, Validation Accuracy:  0.942, Loss:  2.579
Epoch   2 Batch  484/538 - Train Accuracy:  0.933, Validation Accuracy:  0.948, Loss:  2.612
Epoch   2 Batch  485/538 - Train Accuracy:  0.957, Validation Accuracy:  0.947, Loss:  2.642
Epoch   2 Batch  486/538 - Train Accuracy:  0.958, Validation Accuracy:  0.948, Loss:  2.636
Epoch   2 Batch  487/538 - Train Accuracy:  0.939, Validation Accuracy:  0.939, Loss:  2.625
Epoch   2 Batch  488/538 - Train Accuracy:  0.954, Validation Accuracy:  0.931, Loss:  2.598
Epoch   2 Batch  489/538 - Train Accuracy:  0.940, Validation Accuracy:  0.930, Loss:  2.648
Epoch   2 Batch  490/538 - Train Accuracy:  0.945, Validation Accuracy:  0.936, Loss:  2.683
Epoch   2 Batch  491/538 - Train Accuracy:  0.920, Validation Accuracy:  0.948, Loss:  2.675
Epoch   2 Batch  492/538 - Train Accuracy:  0.945, Validation Accuracy:  0.951, Loss:  2.592
Epoch   2 Batch  493/538 - Train Accuracy:  0.949, Validation Accuracy:  0.944, Loss:  2.604
Epoch   2 Batch  494/538 - Train Accuracy:  0.945, Validation Accuracy:  0.938, Loss:  2.630
Epoch   2 Batch  495/538 - Train Accuracy:  0.943, Validation Accuracy:  0.935, Loss:  2.616
Epoch   2 Batch  496/538 - Train Accuracy:  0.948, Validation Accuracy:  0.938, Loss:  2.598
Epoch   2 Batch  497/538 - Train Accuracy:  0.954, Validation Accuracy:  0.940, Loss:  2.622
Epoch   2 Batch  498/538 - Train Accuracy:  0.940, Validation Accuracy:  0.952, Loss:  2.689
Epoch   2 Batch  499/538 - Train Accuracy:  0.935, Validation Accuracy:  0.941, Loss:  2.675
Epoch   2 Batch  500/538 - Train Accuracy:  0.961, Validation Accuracy:  0.931, Loss:  2.603
Epoch   2 Batch  501/538 - Train Accuracy:  0.946, Validation Accuracy:  0.934, Loss:  2.602
Epoch   2 Batch  502/538 - Train Accuracy:  0.946, Validation Accuracy:  0.931, Loss:  2.639
Epoch   2 Batch  503/538 - Train Accuracy:  0.960, Validation Accuracy:  0.935, Loss:  2.609
Epoch   2 Batch  504/538 - Train Accuracy:  0.971, Validation Accuracy:  0.941, Loss:  2.561
Epoch   2 Batch  505/538 - Train Accuracy:  0.958, Validation Accuracy:  0.937, Loss:  2.655
Epoch   2 Batch  506/538 - Train Accuracy:  0.960, Validation Accuracy:  0.934, Loss:  2.621
Epoch   2 Batch  507/538 - Train Accuracy:  0.929, Validation Accuracy:  0.942, Loss:  2.631
Epoch   2 Batch  508/538 - Train Accuracy:  0.943, Validation Accuracy:  0.941, Loss:  2.618
Epoch   2 Batch  509/538 - Train Accuracy:  0.950, Validation Accuracy:  0.945, Loss:  2.604
Epoch   2 Batch  510/538 - Train Accuracy:  0.942, Validation Accuracy:  0.931, Loss:  2.606
Epoch   2 Batch  511/538 - Train Accuracy:  0.937, Validation Accuracy:  0.941, Loss:  2.561
Epoch   2 Batch  512/538 - Train Accuracy:  0.959, Validation Accuracy:  0.941, Loss:  2.652
Epoch   2 Batch  513/538 - Train Accuracy:  0.938, Validation Accuracy:  0.945, Loss:  2.615
Epoch   2 Batch  514/538 - Train Accuracy:  0.947, Validation Accuracy:  0.949, Loss:  2.669
Epoch   2 Batch  515/538 - Train Accuracy:  0.934, Validation Accuracy:  0.949, Loss:  2.607
Epoch   2 Batch  516/538 - Train Accuracy:  0.939, Validation Accuracy:  0.931, Loss:  2.602
Epoch   2 Batch  517/538 - Train Accuracy:  0.950, Validation Accuracy:  0.920, Loss:  2.570
Epoch   2 Batch  518/538 - Train Accuracy:  0.948, Validation Accuracy:  0.928, Loss:  2.635
Epoch   2 Batch  519/538 - Train Accuracy:  0.955, Validation Accuracy:  0.934, Loss:  2.596
Epoch   2 Batch  520/538 - Train Accuracy:  0.960, Validation Accuracy:  0.937, Loss:  2.683
Epoch   2 Batch  521/538 - Train Accuracy:  0.944, Validation Accuracy:  0.938, Loss:  2.611
Epoch   2 Batch  522/538 - Train Accuracy:  0.953, Validation Accuracy:  0.942, Loss:  2.618
Epoch   2 Batch  523/538 - Train Accuracy:  0.958, Validation Accuracy:  0.935, Loss:  2.616
Epoch   2 Batch  524/538 - Train Accuracy:  0.951, Validation Accuracy:  0.938, Loss:  2.634
Epoch   2 Batch  525/538 - Train Accuracy:  0.951, Validation Accuracy:  0.938, Loss:  2.580
Epoch   2 Batch  526/538 - Train Accuracy:  0.938, Validation Accuracy:  0.941, Loss:  2.637
Epoch   2 Batch  527/538 - Train Accuracy:  0.954, Validation Accuracy:  0.941, Loss:  2.587
Epoch   2 Batch  528/538 - Train Accuracy:  0.948, Validation Accuracy:  0.947, Loss:  2.654
Epoch   2 Batch  529/538 - Train Accuracy:  0.930, Validation Accuracy:  0.943, Loss:  2.632
Epoch   2 Batch  530/538 - Train Accuracy:  0.930, Validation Accuracy:  0.940, Loss:  2.570
Epoch   2 Batch  531/538 - Train Accuracy:  0.935, Validation Accuracy:  0.940, Loss:  2.692
Epoch   2 Batch  532/538 - Train Accuracy:  0.944, Validation Accuracy:  0.930, Loss:  2.664
Epoch   2 Batch  533/538 - Train Accuracy:  0.952, Validation Accuracy:  0.930, Loss:  2.607
Epoch   2 Batch  534/538 - Train Accuracy:  0.948, Validation Accuracy:  0.926, Loss:  2.571
Epoch   2 Batch  535/538 - Train Accuracy:  0.940, Validation Accuracy:  0.931, Loss:  2.703
Epoch   2 Batch  536/538 - Train Accuracy:  0.958, Validation Accuracy:  0.941, Loss:  2.662
Model Trained and Saved

Save Parameters

Save the batch_size and save_path parameters for inference.





In [32]:



    


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

Checkpoint





In [33]:



    


"""
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 [34]:



    


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
    """
    sentence = sentence.lower()
    words = sentence.split(' ')
    ids = []
    for w in words:
        ids.append(vocab_to_int.get(w, vocab_to_int.get('<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 [35]:



    


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:      [65, 42, 59, 85, 202, 169, 64]
  English Words: ['he', 'saw', 'a', 'old', 'yellow', 'truck', '.']

Prediction
  Word Ids:      [234, 26, 291, 356, 68, 71, 50, 205, 1]
  French Words: ['il', 'a', 'vu', 'un', 'vieux', 'camion', 'jaune', '.', '<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. 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.

Content source: NirantK/deep-learning-practice

Similar notebooks:

notebook.community | gallery | about