Merge pull request #2329 from mozilla/tf_upgrade_v2

Run tf_upgrade_v2 to ease eventual transition to TF 2.0

DeepSpeech.py

@@ -44,7 +44,7 @@ def variable_on_cpu(name, shape, initializer):
 
 
 def create_overlapping_windows(batch_x):
-    batch_size = tf.shape(batch_x)[0]
+    batch_size = tf.shape(input=batch_x)[0]
     window_width = 2 * Config.n_context + 1
     num_channels = Config.n_input
 
@@ -55,7 +55,7 @@ def create_overlapping_windows(batch_x):
                                .reshape(window_width, num_channels, window_width * num_channels), tf.float32) # pylint: disable=bad-continuation
 
     # Create overlapping windows
-    batch_x = tf.nn.conv1d(batch_x, eye_filter, stride=1, padding='SAME')
+    batch_x = tf.nn.conv1d(input=batch_x, filters=eye_filter, stride=1, padding='SAME')
 
     # Remove dummy depth dimension and reshape into [batch_size, n_windows, window_width, n_input]
     batch_x = tf.reshape(batch_x, [batch_size, -1, window_width, num_channels])
@@ -65,8 +65,8 @@ def create_overlapping_windows(batch_x):
 
 def dense(name, x, units, dropout_rate=None, relu=True):
     with tfv1.variable_scope(name):
-        bias = variable_on_cpu('bias', [units], tf.zeros_initializer())
+        bias = variable_on_cpu('bias', [units], tfv1.zeros_initializer())
-        weights = variable_on_cpu('weights', [x.shape[-1], units], tf.contrib.layers.xavier_initializer())
+        weights = variable_on_cpu('weights', [x.shape[-1], units], tfv1.keras.initializers.VarianceScaling(scale=1.0, mode="fan_avg", distribution="uniform"))
 
     output = tf.nn.bias_add(tf.matmul(x, weights), bias)
 
@@ -147,7 +147,7 @@ def create_model(batch_x, seq_length, dropout, reuse=False, batch_size=None, pre
 
     # Input shape: [batch_size, n_steps, n_input + 2*n_input*n_context]
     if not batch_size:
-        batch_size = tf.shape(batch_x)[0]
+        batch_size = tf.shape(input=batch_x)[0]
 
     # Create overlapping feature windows if needed
     if overlap:
@@ -157,7 +157,7 @@ def create_model(batch_x, seq_length, dropout, reuse=False, batch_size=None, pre
     # This is done to prepare the batch for input into the first layer which expects a tensor of rank `2`.
 
     # Permute n_steps and batch_size
-    batch_x = tf.transpose(batch_x, [1, 0, 2, 3])
+    batch_x = tf.transpose(a=batch_x, perm=[1, 0, 2, 3])
     # Reshape to prepare input for first layer
     batch_x = tf.reshape(batch_x, [-1, Config.n_input + 2*Config.n_input*Config.n_context]) # (n_steps*batch_size, n_input + 2*n_input*n_context)
     layers['input_reshaped'] = batch_x
@@ -232,7 +232,7 @@ def calculate_mean_edit_distance_and_loss(iterator, dropout, reuse):
     non_finite_files = tf.gather(batch_filenames, tfv1.where(~tf.math.is_finite(total_loss)))
 
     # Calculate the average loss across the batch
-    avg_loss = tf.reduce_mean(total_loss)
+    avg_loss = tf.reduce_mean(input_tensor=total_loss)
 
     # Finally we return the average loss
     return avg_loss, non_finite_files
@@ -312,7 +312,7 @@ def get_tower_results(iterator, optimizer, dropout_rates):
 
                     tower_non_finite_files.append(non_finite_files)
 
-    avg_loss_across_towers = tf.reduce_mean(tower_avg_losses, 0)
+    avg_loss_across_towers = tf.reduce_mean(input_tensor=tower_avg_losses, axis=0)
     tfv1.summary.scalar(name='step_loss', tensor=avg_loss_across_towers, collections=['step_summaries'])
 
     all_non_finite_files = tf.concat(tower_non_finite_files, axis=0)
@@ -346,7 +346,7 @@ def average_gradients(tower_gradients):
 
             # Average over the 'tower' dimension
             grad = tf.concat(grads, 0)
-            grad = tf.reduce_mean(grad, 0)
+            grad = tf.reduce_mean(input_tensor=grad, axis=0)
 
             # Create a gradient/variable tuple for the current variable with its average gradient
             grad_and_var = (grad, grad_and_vars[0][1])
@@ -369,11 +369,11 @@ def log_variable(variable, gradient=None):
     Furthermore it logs a histogram of its state and (if given) of an optimization gradient.
     '''
     name = variable.name.replace(':', '_')
-    mean = tf.reduce_mean(variable)
+    mean = tf.reduce_mean(input_tensor=variable)
     tfv1.summary.scalar(name='%s/mean'   % name, tensor=mean)
-    tfv1.summary.scalar(name='%s/sttdev' % name, tensor=tf.sqrt(tf.reduce_mean(tf.square(variable - mean))))
+    tfv1.summary.scalar(name='%s/sttdev' % name, tensor=tf.sqrt(tf.reduce_mean(input_tensor=tf.square(variable - mean))))
-    tfv1.summary.scalar(name='%s/max'    % name, tensor=tf.reduce_max(variable))
+    tfv1.summary.scalar(name='%s/max'    % name, tensor=tf.reduce_max(input_tensor=variable))
-    tfv1.summary.scalar(name='%s/min'    % name, tensor=tf.reduce_min(variable))
+    tfv1.summary.scalar(name='%s/min'    % name, tensor=tf.reduce_min(input_tensor=variable))
     tfv1.summary.histogram(name=name, values=variable)
     if gradient is not None:
         if isinstance(gradient, tf.IndexedSlices):
@@ -667,7 +667,7 @@ def create_inference_graph(batch_size=1, n_steps=16, tflite=False):
         previous_state_c = tfv1.placeholder(tf.float32, [batch_size, Config.n_cell_dim], name='previous_state_c')
         previous_state_h = tfv1.placeholder(tf.float32, [batch_size, Config.n_cell_dim], name='previous_state_h')
 
-        previous_state = tf.contrib.rnn.LSTMStateTuple(previous_state_c, previous_state_h)
+        previous_state = tf.nn.rnn_cell.LSTMStateTuple(previous_state_c, previous_state_h)
 
     # One rate per layer
     no_dropout = [None] * 6

bin/graphdef_binary_to_text.py

@@ -1,14 +1,20 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 
-import tensorflow as tf
+import tensorflow.compat.v1 as tfv1
 import sys
 
-# Load and export as string
+from google.protobuf import text_format
-with tf.gfile.FastGFile(sys.argv[1], 'rb') as fin:
-    graph_def = tf.GraphDef()
-    graph_def.ParseFromString(fin.read())
 
-    with tf.gfile.FastGFile(sys.argv[1] + 'txt', 'w') as fout:
+
-        from google.protobuf import text_format
+def main():
-        fout.write(text_format.MessageToString(graph_def))
+    # Load and export as string
+    with tfv1.gfile.FastGFile(sys.argv[1], 'rb') as fin:
+        graph_def = tfv1.GraphDef()
+        graph_def.ParseFromString(fin.read())
+
+        with tfv1.gfile.FastGFile(sys.argv[1] + 'txt', 'w') as fout:
+            fout.write(text_format.MessageToString(graph_def))
+
+if __name__ == '__main__':
+    main()

bin/ops_in_graph.py

@@ -1,11 +1,15 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 
-import tensorflow as tf
+import tensorflow.compat.v1 as tfv1
 import sys
 
-with tf.gfile.FastGFile(sys.argv[1], 'rb') as fin:
+def main():
-    graph_def = tf.GraphDef()
+    with tfv1.gfile.FastGFile(sys.argv[1], 'rb') as fin:
-    graph_def.ParseFromString(fin.read())
+        graph_def = tfv1.GraphDef()
+        graph_def.ParseFromString(fin.read())
 
-    print('\n'.join(sorted(set(n.op for n in graph_def.node))))
+        print('\n'.join(sorted(set(n.op for n in graph_def.node))))
+
+if __name__ == '__main__':
+    main()

evaluate.py

@@ -63,7 +63,7 @@ def evaluate(test_csvs, create_model, try_loading):
                              dropout=no_dropout)
 
     # Transpose to batch major and apply softmax for decoder
-    transposed = tf.nn.softmax(tf.transpose(logits, [1, 0, 2]))
+    transposed = tf.nn.softmax(tf.transpose(a=logits, perm=[1, 0, 2]))
 
     loss = tfv1.nn.ctc_loss(labels=batch_y,
                           inputs=logits,

util/config.py

@@ -2,6 +2,7 @@ from __future__ import absolute_import, division, print_function
 
 import os
 import tensorflow as tf
+import tensorflow.compat.v1 as tfv1
 
 from attrdict import AttrDict
 from xdg import BaseDirectory as xdg
@@ -57,9 +58,9 @@ def initialize_globals():
         FLAGS.summary_dir = xdg.save_data_path(os.path.join('deepspeech', 'summaries'))
 
     # Standard session configuration that'll be used for all new sessions.
-    c.session_config = tf.ConfigProto(allow_soft_placement=True, log_device_placement=FLAGS.log_placement,
+    c.session_config = tfv1.ConfigProto(allow_soft_placement=True, log_device_placement=FLAGS.log_placement,
-                                      inter_op_parallelism_threads=FLAGS.inter_op_parallelism_threads,
+                                        inter_op_parallelism_threads=FLAGS.inter_op_parallelism_threads,
-                                      intra_op_parallelism_threads=FLAGS.intra_op_parallelism_threads)
+                                        intra_op_parallelism_threads=FLAGS.intra_op_parallelism_threads)
 
     c.alphabet = Alphabet(os.path.abspath(FLAGS.alphabet_config_path))
 

util/feeding.py

@@ -10,7 +10,7 @@ import pandas
 import tensorflow as tf
 import datetime
 
-from tensorflow.contrib.framework.python.ops import audio_ops as contrib_audio
+from tensorflow.python.ops import gen_audio_ops as contrib_audio
 
 from util.config import Config
 from util.logging import log_error
@@ -39,7 +39,7 @@ def samples_to_mfccs(samples, sample_rate):
     mfccs = contrib_audio.mfcc(spectrogram, sample_rate, dct_coefficient_count=Config.n_input)
     mfccs = tf.reshape(mfccs, [-1, Config.n_input])
 
-    return mfccs, tf.shape(mfccs)[0]
+    return mfccs, tf.shape(input=mfccs)[0]
 
 
 def audiofile_to_features(wav_filename):