Merge pull request #1063 from tensorflow/update-models-1.0

Converted the models repo to TF 1.0 using the upgrade script

swivel/swivel.py

@@ -135,8 +135,8 @@ def count_matrix_input(filenames, submatrix_rows, submatrix_cols):
   sparse_local_col = features['sparse_local_col'].values
   sparse_count = features['sparse_value'].values
 
-  sparse_indices = tf.concat([tf.expand_dims(sparse_local_row, 1),
-                              tf.expand_dims(sparse_local_col, 1)], 1)
+  sparse_indices = tf.concat(axis=1, values=[tf.expand_dims(sparse_local_row, 1),
+                                             tf.expand_dims(sparse_local_col, 1)])
   count = tf.sparse_to_dense(sparse_indices, [submatrix_rows, submatrix_cols],
                              sparse_count)
 

syntaxnet/syntaxnet/graph_builder.py

@@ -69,7 +69,7 @@ def EmbeddingLookupFeatures(params, sparse_features, allow_weights):
 
   if allow_weights:
     # Multiply by weights, reshaping to allow broadcast.
-    broadcast_weights_shape = tf.concat([tf.shape(weights), [1]], 0)
+    broadcast_weights_shape = tf.concat(axis=0, values=[tf.shape(weights), [1]])
     embeddings *= tf.reshape(weights, broadcast_weights_shape)
 
   # Sum embeddings by index.
@@ -330,7 +330,7 @@ class GreedyParser(object):
                                            i,
                                            return_average=return_average))
 
-    last_layer = tf.concat(embeddings, 1)
+    last_layer = tf.concat(axis=1, values=embeddings)
     last_layer_size = self.embedding_size
 
     # Create ReLU layers.

textsum/seq2seq_attention.py

@@ -86,7 +86,7 @@ def _Train(model, data_batcher):
     saver = tf.train.Saver()
     # Train dir is different from log_root to avoid summary directory
     # conflict with Supervisor.
-    summary_writer = tf.train.SummaryWriter(FLAGS.train_dir)
+    summary_writer = tf.summary.FileWriter(FLAGS.train_dir)
     sv = tf.train.Supervisor(logdir=FLAGS.log_root,
                              is_chief=True,
                              saver=saver,
@@ -119,7 +119,7 @@ def _Eval(model, data_batcher, vocab=None):
   """Runs model eval."""
   model.build_graph()
   saver = tf.train.Saver()
-  summary_writer = tf.train.SummaryWriter(FLAGS.eval_dir)
+  summary_writer = tf.summary.FileWriter(FLAGS.eval_dir)
   sess = tf.Session(config=tf.ConfigProto(allow_soft_placement=True))
   running_avg_loss = 0
   step = 0

textsum/seq2seq_attention_model.py

@@ -139,10 +139,10 @@ class Seq2SeqAttentionModel(object):
     vsize = self._vocab.NumIds()
 
     with tf.variable_scope('seq2seq'):
-      encoder_inputs = tf.unpack(tf.transpose(self._articles))
-      decoder_inputs = tf.unpack(tf.transpose(self._abstracts))
-      targets = tf.unpack(tf.transpose(self._targets))
-      loss_weights = tf.unpack(tf.transpose(self._loss_weights))
+      encoder_inputs = tf.unstack(tf.transpose(self._articles))
+      decoder_inputs = tf.unstack(tf.transpose(self._abstracts))
+      targets = tf.unstack(tf.transpose(self._targets))
+      loss_weights = tf.unstack(tf.transpose(self._loss_weights))
       article_lens = self._article_lens
 
       # Embedding shared by the input and outputs.
@@ -195,7 +195,7 @@ class Seq2SeqAttentionModel(object):
 
         encoder_outputs = [tf.reshape(x, [hps.batch_size, 1, 2*hps.num_hidden])
                            for x in encoder_outputs]
-        self._enc_top_states = tf.concat(1, encoder_outputs)
+        self._enc_top_states = tf.concat(axis=1, values=encoder_outputs)
         self._dec_in_state = fw_state
         # During decoding, follow up _dec_in_state are fed from beam_search.
         # dec_out_state are stored by beam_search for next step feeding.
@@ -218,7 +218,7 @@ class Seq2SeqAttentionModel(object):
           best_outputs = [tf.argmax(x, 1) for x in model_outputs]
           tf.logging.info('best_outputs%s', best_outputs[0].get_shape())
           self._outputs = tf.concat(
-              1, [tf.reshape(x, [hps.batch_size, 1]) for x in best_outputs])
+              axis=1, values=[tf.reshape(x, [hps.batch_size, 1]) for x in best_outputs])
 
           self._topk_log_probs, self._topk_ids = tf.nn.top_k(
               tf.log(tf.nn.softmax(model_outputs[-1])), hps.batch_size*2)
@@ -236,7 +236,7 @@ class Seq2SeqAttentionModel(object):
         else:
           self._loss = tf.nn.seq2seq.sequence_loss(
               model_outputs, targets, loss_weights)
-        tf.scalar_summary('loss', tf.minimum(12.0, self._loss))
+        tf.summary.scalar('loss', tf.minimum(12.0, self._loss))
 
   def _add_train_op(self):
     """Sets self._train_op, op to run for training."""
@@ -250,9 +250,9 @@ class Seq2SeqAttentionModel(object):
     with tf.device(self._get_gpu(self._num_gpus-1)):
       grads, global_norm = tf.clip_by_global_norm(
           tf.gradients(self._loss, tvars), hps.max_grad_norm)
-    tf.scalar_summary('global_norm', global_norm)
+    tf.summary.scalar('global_norm', global_norm)
     optimizer = tf.train.GradientDescentOptimizer(self._lr_rate)
-    tf.scalar_summary('learning rate', self._lr_rate)
+    tf.summary.scalar('learning rate', self._lr_rate)
     self._train_op = optimizer.apply_gradients(
         zip(grads, tvars), global_step=self.global_step, name='train_step')
 
@@ -296,4 +296,4 @@ class Seq2SeqAttentionModel(object):
     self.global_step = tf.Variable(0, name='global_step', trainable=False)
     if self._hps.mode == 'train':
       self._add_train_op()
-    self._summaries = tf.merge_all_summaries()
+    self._summaries = tf.summary.merge_all()

textsum/seq2seq_lib.py

@@ -127,7 +127,7 @@ def linear(args, output_size, bias, bias_start=0.0, scope=None):
     if len(args) == 1:
       res = tf.matmul(args[0], matrix)
     else:
-      res = tf.matmul(tf.concat(1, args), matrix)
+      res = tf.matmul(tf.concat(axis=1, values=args), matrix)
     if not bias:
       return res
     bias_term = tf.get_variable(

transformer/spatial_transformer.py

@@ -53,7 +53,7 @@ def transformer(U, theta, out_size, name='SpatialTransformer', **kwargs):
     def _repeat(x, n_repeats):
         with tf.variable_scope('_repeat'):
             rep = tf.transpose(
-                tf.expand_dims(tf.ones(shape=tf.pack([n_repeats, ])), 1), [1, 0])
+                tf.expand_dims(tf.ones(shape=tf.stack([n_repeats, ])), 1), [1, 0])
             rep = tf.cast(rep, 'int32')
             x = tf.matmul(tf.reshape(x, (-1, 1)), rep)
             return tf.reshape(x, [-1])
@@ -102,7 +102,7 @@ def transformer(U, theta, out_size, name='SpatialTransformer', **kwargs):
 
             # use indices to lookup pixels in the flat image and restore
             # channels dim
-            im_flat = tf.reshape(im, tf.pack([-1, channels]))
+            im_flat = tf.reshape(im, tf.stack([-1, channels]))
             im_flat = tf.cast(im_flat, 'float32')
             Ia = tf.gather(im_flat, idx_a)
             Ib = tf.gather(im_flat, idx_b)
@@ -128,16 +128,16 @@ def transformer(U, theta, out_size, name='SpatialTransformer', **kwargs):
             #                         np.linspace(-1, 1, height))
             #  ones = np.ones(np.prod(x_t.shape))
             #  grid = np.vstack([x_t.flatten(), y_t.flatten(), ones])
-            x_t = tf.matmul(tf.ones(shape=tf.pack([height, 1])),
+            x_t = tf.matmul(tf.ones(shape=tf.stack([height, 1])),
                             tf.transpose(tf.expand_dims(tf.linspace(-1.0, 1.0, width), 1), [1, 0]))
             y_t = tf.matmul(tf.expand_dims(tf.linspace(-1.0, 1.0, height), 1),
-                            tf.ones(shape=tf.pack([1, width])))
+                            tf.ones(shape=tf.stack([1, width])))
 
             x_t_flat = tf.reshape(x_t, (1, -1))
             y_t_flat = tf.reshape(y_t, (1, -1))
 
             ones = tf.ones_like(x_t_flat)
-            grid = tf.concat(0, [x_t_flat, y_t_flat, ones])
+            grid = tf.concat(axis=0, values=[x_t_flat, y_t_flat, ones])
             return grid
 
     def _transform(theta, input_dim, out_size):
@@ -157,11 +157,11 @@ def transformer(U, theta, out_size, name='SpatialTransformer', **kwargs):
             grid = _meshgrid(out_height, out_width)
             grid = tf.expand_dims(grid, 0)
             grid = tf.reshape(grid, [-1])
-            grid = tf.tile(grid, tf.pack([num_batch]))
-            grid = tf.reshape(grid, tf.pack([num_batch, 3, -1]))
+            grid = tf.tile(grid, tf.stack([num_batch]))
+            grid = tf.reshape(grid, tf.stack([num_batch, 3, -1]))
 
             # Transform A x (x_t, y_t, 1)^T -> (x_s, y_s)
-            T_g = tf.batch_matmul(theta, grid)
+            T_g = tf.matmul(theta, grid)
             x_s = tf.slice(T_g, [0, 0, 0], [-1, 1, -1])
             y_s = tf.slice(T_g, [0, 1, 0], [-1, 1, -1])
             x_s_flat = tf.reshape(x_s, [-1])
@@ -172,7 +172,7 @@ def transformer(U, theta, out_size, name='SpatialTransformer', **kwargs):
                 out_size)
 
             output = tf.reshape(
-                input_transformed, tf.pack([num_batch, out_height, out_width, num_channels]))
+                input_transformed, tf.stack([num_batch, out_height, out_width, num_channels]))
             return output
 
     with tf.variable_scope(name):

tutorials/embedding/word2vec.py

@@ -246,7 +246,7 @@ class Word2Vec(object):
     sampled_b = tf.nn.embedding_lookup(sm_b, sampled_ids)
 
     # True logits: [batch_size, 1]
-    true_logits = tf.reduce_sum(tf.mul(example_emb, true_w), 1) + true_b
+    true_logits = tf.reduce_sum(tf.multiply(example_emb, true_w), 1) + true_b
 
     # Sampled logits: [batch_size, num_sampled]
     # We replicate sampled noise labels for all examples in the batch

tutorials/image/cifar10/cifar10_multi_gpu_train.py

@@ -124,7 +124,7 @@ def average_gradients(tower_grads):
       grads.append(expanded_g)
 
     # Average over the 'tower' dimension.
-    grad = tf.concat(grads, 0)
+    grad = tf.concat(axis=0, values=grads)
     grad = tf.reduce_mean(grad, 0)
 
     # Keep in mind that the Variables are redundant because they are shared

tutorials/rnn/ptb/ptb_word_lm.py

@@ -146,7 +146,7 @@ class PTBModel(object):
         (cell_output, state) = cell(inputs[:, time_step, :], state)
         outputs.append(cell_output)
 
-    output = tf.reshape(tf.concat(outputs, 1), [-1, size])
+    output = tf.reshape(tf.concat(axis=1, values=outputs), [-1, size])
     softmax_w = tf.get_variable(
         "softmax_w", [size, vocab_size], dtype=data_type())
     softmax_b = tf.get_variable("softmax_b", [vocab_size], dtype=data_type())

video_prediction/lstm_ops.py

@@ -23,7 +23,7 @@ from tensorflow.contrib.slim import layers
 
 def init_state(inputs,
                state_shape,
-               state_initializer=tf.zeros_initializer,
+               state_initializer=tf.zeros_initializer(),
                dtype=tf.float32):
   """Helper function to create an initial state given inputs.
 
@@ -45,7 +45,7 @@ def init_state(inputs,
     batch_size = 0
 
   initial_state = state_initializer(
-      tf.pack([batch_size] + state_shape),
+      tf.stack([batch_size] + state_shape),
       dtype=dtype)
   initial_state.set_shape([inferred_batch_size] + state_shape)
 
@@ -89,8 +89,8 @@ def basic_conv_lstm_cell(inputs,
                          reuse=reuse):
     inputs.get_shape().assert_has_rank(4)
     state.get_shape().assert_has_rank(4)
-    c, h = tf.split(3, 2, state)
-    inputs_h = tf.concat(3, [inputs, h])
+    c, h = tf.split(axis=3, num_or_size_splits=2, value=state)
+    inputs_h = tf.concat(axis=3, values=[inputs, h])
     # Parameters of gates are concatenated into one conv for efficiency.
     i_j_f_o = layers.conv2d(inputs_h,
                             4 * num_channels, [filter_size, filter_size],
@@ -99,12 +99,12 @@ def basic_conv_lstm_cell(inputs,
                             scope='Gates')
 
     # i = input_gate, j = new_input, f = forget_gate, o = output_gate
-    i, j, f, o = tf.split(3, 4, i_j_f_o)
+    i, j, f, o = tf.split(axis=3, num_or_size_splits=4, value=i_j_f_o)
 
     new_c = c * tf.sigmoid(f + forget_bias) + tf.sigmoid(i) * tf.tanh(j)
     new_h = tf.tanh(new_c) * tf.sigmoid(o)
 
-    return new_h, tf.concat(3, [new_c, new_h])
+    return new_h, tf.concat(axis=3, values=[new_c, new_h])
 
 
 

video_prediction/prediction_input.py

@@ -97,11 +97,11 @@ def build_tfrecord_input(training=True):
       action = tf.reshape(features[action_name], shape=[1, STATE_DIM])
       action_seq.append(action)
 
-  image_seq = tf.concat(0, image_seq)
+  image_seq = tf.concat(axis=0, values=image_seq)
 
   if FLAGS.use_state:
-    state_seq = tf.concat(0, state_seq)
-    action_seq = tf.concat(0, action_seq)
+    state_seq = tf.concat(axis=0, values=state_seq)
+    action_seq = tf.concat(axis=0, values=action_seq)
     [image_batch, action_batch, state_batch] = tf.train.batch(
         [image_seq, action_seq, state_seq],
         FLAGS.batch_size,

video_prediction/prediction_model.py

@@ -109,7 +109,7 @@ def construct_model(images,
         prev_image = image
 
       # Predicted state is always fed back in
-      state_action = tf.concat(1, [action, current_state])
+      state_action = tf.concat(axis=1, values=[action, current_state])
 
       enc0 = slim.layers.conv2d(
           prev_image,
@@ -144,7 +144,7 @@ def construct_model(images,
       smear = tf.tile(
           smear, [1, int(enc2.get_shape()[1]), int(enc2.get_shape()[2]), 1])
       if use_state:
-        enc2 = tf.concat(3, [enc2, smear])
+        enc2 = tf.concat(axis=3, values=[enc2, smear])
       enc3 = slim.layers.conv2d(
           enc2, hidden4.get_shape()[3], [1, 1], stride=1, scope='conv4')
 
@@ -158,7 +158,7 @@ def construct_model(images,
           enc4, lstm_state6, lstm_size[5], scope='state6')  # 16x16
       hidden6 = tf_layers.layer_norm(hidden6, scope='layer_norm7')
       # Skip connection.
-      hidden6 = tf.concat(3, [hidden6, enc1])  # both 16x16
+      hidden6 = tf.concat(axis=3, values=[hidden6, enc1])  # both 16x16
 
       enc5 = slim.layers.conv2d_transpose(
           hidden6, hidden6.get_shape()[3], 3, stride=2, scope='convt2')
@@ -167,7 +167,7 @@ def construct_model(images,
       hidden7 = tf_layers.layer_norm(hidden7, scope='layer_norm8')
 
       # Skip connection.
-      hidden7 = tf.concat(3, [hidden7, enc0])  # both 32x32
+      hidden7 = tf.concat(axis=3, values=[hidden7, enc0])  # both 32x32
 
       enc6 = slim.layers.conv2d_transpose(
           hidden7,
@@ -207,7 +207,7 @@ def construct_model(images,
       masks = tf.reshape(
           tf.nn.softmax(tf.reshape(masks, [-1, num_masks + 1])),
           [int(batch_size), int(img_height), int(img_width), num_masks + 1])
-      mask_list = tf.split(3, num_masks + 1, masks)
+      mask_list = tf.split(axis=3, num_or_size_splits=num_masks + 1, value=masks)
       output = mask_list[0] * prev_image
       for layer, mask in zip(transformed, mask_list[1:]):
         output += layer * mask
@@ -277,8 +277,8 @@ def cdna_transformation(prev_image, cdna_input, num_masks, color_channels):
   cdna_kerns /= norm_factor
 
   cdna_kerns = tf.tile(cdna_kerns, [1, 1, 1, color_channels, 1])
-  cdna_kerns = tf.split(0, batch_size, cdna_kerns)
-  prev_images = tf.split(0, batch_size, prev_image)
+  cdna_kerns = tf.split(axis=0, num_or_size_splits=batch_size, value=cdna_kerns)
+  prev_images = tf.split(axis=0, num_or_size_splits=batch_size, value=prev_image)
 
   # Transform image.
   transformed = []
@@ -288,8 +288,8 @@ def cdna_transformation(prev_image, cdna_input, num_masks, color_channels):
       kernel = tf.expand_dims(kernel, -1)
     transformed.append(
         tf.nn.depthwise_conv2d(preimg, kernel, [1, 1, 1, 1], 'SAME'))
-  transformed = tf.concat(0, transformed)
-  transformed = tf.split(3, num_masks, transformed)
+  transformed = tf.concat(axis=0, values=transformed)
+  transformed = tf.split(axis=3, num_or_size_splits=num_masks, value=transformed)
   return transformed
 
 
@@ -314,7 +314,7 @@ def dna_transformation(prev_image, dna_input):
           tf.expand_dims(
               tf.slice(prev_image_pad, [0, xkern, ykern, 0],
                        [-1, image_height, image_width, -1]), [3]))
-  inputs = tf.concat(3, inputs)
+  inputs = tf.concat(axis=3, values=inputs)
 
   # Normalize channels to 1.
   kernel = tf.nn.relu(dna_input - RELU_SHIFT) + RELU_SHIFT

video_prediction/prediction_train.py

@@ -113,11 +113,11 @@ class Model(object):
     summaries = []
 
     # Split into timesteps.
-    actions = tf.split(1, actions.get_shape()[1], actions)
+    actions = tf.split(axis=1, num_or_size_splits=actions.get_shape()[1], value=actions)
     actions = [tf.squeeze(act) for act in actions]
-    states = tf.split(1, states.get_shape()[1], states)
+    states = tf.split(axis=1, num_or_size_splits=states.get_shape()[1], value=states)
     states = [tf.squeeze(st) for st in states]
-    images = tf.split(1, images.get_shape()[1], images)
+    images = tf.split(axis=1, num_or_size_splits=images.get_shape()[1], value=images)
     images = [tf.squeeze(img) for img in images]
 
     if reuse_scope is None:
@@ -157,8 +157,8 @@ class Model(object):
       psnr_i = peak_signal_to_noise_ratio(x, gx)
       psnr_all += psnr_i
       summaries.append(
-          tf.scalar_summary(prefix + '_recon_cost' + str(i), recon_cost))
-      summaries.append(tf.scalar_summary(prefix + '_psnr' + str(i), psnr_i))
+          tf.summary.scalar(prefix + '_recon_cost' + str(i), recon_cost))
+      summaries.append(tf.summary.scalar(prefix + '_psnr' + str(i), psnr_i))
       loss += recon_cost
 
     for i, state, gen_state in zip(
@@ -166,19 +166,19 @@ class Model(object):
         gen_states[FLAGS.context_frames - 1:]):
       state_cost = mean_squared_error(state, gen_state) * 1e-4
       summaries.append(
-          tf.scalar_summary(prefix + '_state_cost' + str(i), state_cost))
+          tf.summary.scalar(prefix + '_state_cost' + str(i), state_cost))
       loss += state_cost
-    summaries.append(tf.scalar_summary(prefix + '_psnr_all', psnr_all))
+    summaries.append(tf.summary.scalar(prefix + '_psnr_all', psnr_all))
     self.psnr_all = psnr_all
 
     self.loss = loss = loss / np.float32(len(images) - FLAGS.context_frames)
 
-    summaries.append(tf.scalar_summary(prefix + '_loss', loss))
+    summaries.append(tf.summary.scalar(prefix + '_loss', loss))
 
     self.lr = tf.placeholder_with_default(FLAGS.learning_rate, ())
 
     self.train_op = tf.train.AdamOptimizer(self.lr).minimize(loss)
-    self.summ_op = tf.merge_summary(summaries)
+    self.summ_op = tf.summary.merge(summaries)
 
 
 def main(unused_argv):
@@ -200,7 +200,7 @@ def main(unused_argv):
 
   # Make training session.
   sess = tf.InteractiveSession()
-  summary_writer = tf.train.SummaryWriter(
+  summary_writer = tf.summary.FileWriter(
       FLAGS.event_log_dir, graph=sess.graph, flush_secs=10)
 
   if FLAGS.pretrained_model: