Merge branch 'master' into master

tutorials/embedding/word2vec.py

@@ -42,7 +42,7 @@ from six.moves import xrange  # pylint: disable=redefined-builtin
 import numpy as np
 import tensorflow as tf
 
-from tensorflow.models.embedding import gen_word2vec as word2vec
+word2vec = tf.load_op_library(os.path.join(os.path.dirname(os.path.realpath(__file__)), 'word2vec_ops.so'))
 
 flags = tf.app.flags
 
@@ -263,9 +263,9 @@ class Word2Vec(object):
     # cross-entropy(logits, labels)
     opts = self._options
     true_xent = tf.nn.sigmoid_cross_entropy_with_logits(
-        true_logits, tf.ones_like(true_logits))
+        labels=tf.ones_like(true_logits), logits=true_logits)
     sampled_xent = tf.nn.sigmoid_cross_entropy_with_logits(
-        sampled_logits, tf.zeros_like(sampled_logits))
+        labels=tf.zeros_like(sampled_logits), logits=sampled_logits)
 
     # NCE-loss is the sum of the true and noise (sampled words)
     # contributions, averaged over the batch.

tutorials/embedding/word2vec_optimized.py

@@ -41,7 +41,7 @@ from six.moves import xrange  # pylint: disable=redefined-builtin
 import numpy as np
 import tensorflow as tf
 
-from tensorflow.models.embedding import gen_word2vec as word2vec
+word2vec = tf.load_op_library(os.path.join(os.path.dirname(os.path.realpath(__file__)), 'word2vec_ops.so'))
 
 flags = tf.app.flags
 

tutorials/embedding/word2vec_optimized_test.py

@@ -23,7 +23,7 @@ import os
 
 import tensorflow as tf
 
-from tensorflow.models.embedding import word2vec_optimized
+import word2vec_optimized
 
 flags = tf.app.flags
 

tutorials/embedding/word2vec_test.py

@@ -23,7 +23,7 @@ import os
 
 import tensorflow as tf
 
-from tensorflow.models.embedding import word2vec
+import word2vec
 
 flags = tf.app.flags
 

tutorials/image/alexnet/alexnet_benchmark.py

@@ -17,7 +17,7 @@
 
 To run, use:
   bazel run -c opt --config=cuda \
-      third_party/tensorflow/models/image/alexnet:alexnet_benchmark
+      models/tutorials/image/alexnet:alexnet_benchmark
 
 Across 100 steps on batch size = 128.
 

tutorials/image/cifar10/__init__.py

@@ -18,5 +18,5 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
-from tensorflow.models.image.cifar10 import cifar10
-from tensorflow.models.image.cifar10 import cifar10_input
+import cifar10
+import cifar10_input

tutorials/image/cifar10/cifar10.py

@@ -35,7 +35,6 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
-import gzip
 import os
 import re
 import sys
@@ -44,7 +43,7 @@ import tarfile
 from six.moves import urllib
 import tensorflow as tf
 
-from tensorflow.models.image.cifar10 import cifar10_input
+import cifar10_input
 
 FLAGS = tf.app.flags.FLAGS
 
@@ -287,7 +286,7 @@ def loss(logits, labels):
   # Calculate the average cross entropy loss across the batch.
   labels = tf.cast(labels, tf.int64)
   cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(
-      logits, labels, name='cross_entropy_per_example')
+      labels=labels, logits=logits, name='cross_entropy_per_example')
   cross_entropy_mean = tf.reduce_mean(cross_entropy, name='cross_entropy')
   tf.add_to_collection('losses', cross_entropy_mean)
 

tutorials/image/cifar10/cifar10_eval.py

@@ -41,7 +41,7 @@ import time
 import numpy as np
 import tensorflow as tf
 
-from tensorflow.models.image.cifar10 import cifar10
+import cifar10
 
 FLAGS = tf.app.flags.FLAGS
 

tutorials/image/cifar10/cifar10_input.py

@@ -242,6 +242,10 @@ def inputs(eval_data, data_dir, batch_size):
   # Subtract off the mean and divide by the variance of the pixels.
   float_image = tf.image.per_image_standardization(resized_image)
 
+  # Set the shapes of tensors.
+  float_image.set_shape([height, width, 3])
+  read_input.label.set_shape([1])
+
   # Ensure that the random shuffling has good mixing properties.
   min_fraction_of_examples_in_queue = 0.4
   min_queue_examples = int(num_examples_per_epoch *

tutorials/image/cifar10/cifar10_input_test.py

@@ -23,7 +23,7 @@ import os
 
 import tensorflow as tf
 
-from tensorflow.models.image.cifar10 import cifar10_input
+import cifar10_input
 
 
 class CIFAR10InputTest(tf.test.TestCase):

tutorials/image/cifar10/cifar10_multi_gpu_train.py

@@ -47,7 +47,7 @@ import time
 import numpy as np
 from six.moves import xrange  # pylint: disable=redefined-builtin
 import tensorflow as tf
-from tensorflow.models.image.cifar10 import cifar10
+import cifar10
 
 FLAGS = tf.app.flags.FLAGS
 

tutorials/image/cifar10/cifar10_train.py

@@ -41,7 +41,7 @@ import time
 
 import tensorflow as tf
 
-from tensorflow.models.image.cifar10 import cifar10
+import cifar10
 
 FLAGS = tf.app.flags.FLAGS
 

tutorials/image/mnist/convolutional.py

@@ -228,7 +228,7 @@ def main(_):
   # Training computation: logits + cross-entropy loss.
   logits = model(train_data_node, True)
   loss = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(
-      logits, train_labels_node))
+      labels=train_labels_node, logits=logits))
 
   # L2 regularization for the fully connected parameters.
   regularizers = (tf.nn.l2_loss(fc1_weights) + tf.nn.l2_loss(fc1_biases) +

tutorials/rnn/README.md

@@ -2,8 +2,8 @@ This directory contains functions for creating recurrent neural networks
 and sequence-to-sequence models. Detailed instructions on how to get started
 and use them are available in the tutorials.
 
-* [RNN Tutorial](http://tensorflow.org/tutorials/recurrent/index.md)
-* [Sequence-to-Sequence Tutorial](http://tensorflow.org/tutorials/seq2seq/index.md)
+* [RNN Tutorial](http://tensorflow.org/tutorials/recurrent/)
+* [Sequence-to-Sequence Tutorial](http://tensorflow.org/tutorials/seq2seq/)
 
 Here is a short overview of what is in this directory.
 

tutorials/rnn/ptb/__init__.py

@@ -18,4 +18,4 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
-from tensorflow.models.rnn.ptb import reader
+import reader

tutorials/rnn/ptb/ptb_word_lm.py

@@ -61,7 +61,7 @@ import time
 import numpy as np
 import tensorflow as tf
 
-from tensorflow.models.rnn.ptb import reader
+import reader
 
 flags = tf.flags
 logging = tf.logging
@@ -126,7 +126,7 @@ class PTBModel(object):
     if is_training and config.keep_prob < 1:
       inputs = tf.nn.dropout(inputs, config.keep_prob)
 
-    # Simplified version of tensorflow.models.rnn.rnn.py's rnn().
+    # Simplified version of models/tutorials/rnn/rnn.py's rnn().
     # This builds an unrolled LSTM for tutorial purposes only.
     # In general, use the rnn() or state_saving_rnn() from rnn.py.
     #

tutorials/rnn/ptb/reader_test.py

@@ -13,7 +13,7 @@
 # limitations under the License.
 # ==============================================================================
 
-"""Tests for tensorflow.models.ptb_lstm.ptb_reader."""
+"""Tests for models.tutorials.rnn.ptb.reader."""
 
 from __future__ import absolute_import
 from __future__ import division
@@ -23,7 +23,7 @@ import os.path
 
 import tensorflow as tf
 
-from tensorflow.models.rnn.ptb import reader
+import reader
 
 
 class PtbReaderTest(tf.test.TestCase):

tutorials/rnn/translate/__init__.py

@@ -18,5 +18,5 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
-from tensorflow.models.rnn.translate import data_utils
-from tensorflow.models.rnn.translate import seq2seq_model
+import data_utils
+import seq2seq_model

tutorials/rnn/translate/data_utils.py

@@ -177,7 +177,7 @@ def initialize_vocabulary(vocabulary_path):
     rev_vocab = []
     with gfile.GFile(vocabulary_path, mode="rb") as f:
       rev_vocab.extend(f.readlines())
-    rev_vocab = [line.strip() for line in rev_vocab]
+    rev_vocab = [tf.compat.as_bytes(line.strip()) for line in rev_vocab]
     vocab = dict([(x, y) for (y, x) in enumerate(rev_vocab)])
     return vocab, rev_vocab
   else:

tutorials/rnn/translate/seq2seq_model.py

@@ -25,7 +25,7 @@ import numpy as np
 from six.moves import xrange  # pylint: disable=redefined-builtin
 import tensorflow as tf
 
-from tensorflow.models.rnn.translate import data_utils
+import data_utils
 
 
 class Seq2SeqModel(object):