Update keras to use collections.abc instead of collections, which will be deprecated in py3.9

PiperOrigin-RevId: 339588609
Change-Id: I30553aa31227ff8e56a15509d0261ec64de3240c

tensorflow/python/keras/callbacks.py

@@ -21,6 +21,7 @@ from __future__ import division
 from __future__ import print_function
 
 import collections
+import collections.abc as collections_abc
 import copy
 import csv
 import io
@@ -2608,7 +2609,7 @@ class CSVLogger(Callback):
       is_zero_dim_ndarray = isinstance(k, np.ndarray) and k.ndim == 0
       if isinstance(k, six.string_types):
         return k
-      elif isinstance(k, collections.Iterable) and not is_zero_dim_ndarray:
+      elif isinstance(k, collections_abc.Iterable) and not is_zero_dim_ndarray:
         return '"[%s]"' % (', '.join(map(str, k)))
       else:
         return k

tensorflow/python/keras/engine/training.py

@@ -968,7 +968,7 @@ class Model(base_layer.Layer, version_utils.ModelVersionSelector):
             form of datasets, generators, or `keras.utils.Sequence` instances
             (since they generate batches).
         validation_freq: Only relevant if validation data is provided. Integer
-            or `collections.Container` instance (e.g. list, tuple, etc.).
+            or `collections.abc.Container` instance (e.g. list, tuple, etc.).
             If an integer, specifies how many training epochs to run before a
             new validation run is performed, e.g. `validation_freq=2` runs
             validation every 2 epochs. If a Container, specifies the epochs on

tensorflow/python/keras/engine/training_arrays_v1.py

@@ -94,7 +94,7 @@ def model_iteration(model,
         validation from data tensors). Ignored with the default value of
         `None`.
       validation_freq: Only relevant if validation data is provided. Integer or
-        `collections.Container` instance (e.g. list, tuple, etc.). If an
+        `collections.abc.Container` instance (e.g. list, tuple, etc.). If an
         integer, specifies how many training epochs to run before a new
         validation run is performed, e.g. `validation_freq=2` runs
         validation every 2 epochs. If a Container, specifies the epochs on

tensorflow/python/keras/engine/training_utils_v1.py

@@ -20,6 +20,7 @@ from __future__ import print_function
 import abc
 import atexit
 import collections
+import collections.abc as collections_abc
 import functools
 import multiprocessing.pool
 import threading
@@ -584,7 +585,7 @@ def standardize_sample_or_class_weights(x_weight, output_names, weight_type):
                        'You should provide one `' + weight_type + '`'
                        'array per model output.')
     return x_weight
-  if isinstance(x_weight, collections.Mapping):
+  if isinstance(x_weight, collections_abc.Mapping):
     generic_utils.check_for_unexpected_keys(weight_type, x_weight, output_names)
     x_weights = []
     for name in output_names:
@@ -771,7 +772,7 @@ def collect_per_output_metric_info(metrics,
               [metrics_module.clone_metric(m) for m in metrics])
       else:
         nested_metrics = [metrics]
-  elif isinstance(metrics, collections.Mapping):
+  elif isinstance(metrics, collections_abc.Mapping):
     generic_utils.check_for_unexpected_keys('metrics', metrics, output_names)
     nested_metrics = []
     for name in output_names:
@@ -1087,7 +1088,7 @@ def get_loss_function(loss):
         'before passing them to Model.compile.'.format(loss))
 
   # Deserialize loss configuration, if needed.
-  if isinstance(loss, collections.Mapping):
+  if isinstance(loss, collections_abc.Mapping):
     loss = losses.get(loss)
 
   # Custom callable class.
@@ -1304,7 +1305,7 @@ def prepare_sample_weight_modes(training_endpoints, sample_weight_mode):
     ValueError: In case of invalid `sample_weight_mode` input.
   """
 
-  if isinstance(sample_weight_mode, collections.Mapping):
+  if isinstance(sample_weight_mode, collections_abc.Mapping):
     generic_utils.check_for_unexpected_keys(
         'sample_weight_mode', sample_weight_mode,
         [e.output_name for e in training_endpoints])
@@ -1351,7 +1352,7 @@ def prepare_loss_functions(loss, output_names):
       ValueError: If loss is a dict with keys not in model output names,
           or if loss is a list with len not equal to model outputs.
   """
-  if isinstance(loss, collections.Mapping):
+  if isinstance(loss, collections_abc.Mapping):
     generic_utils.check_for_unexpected_keys('loss', loss, output_names)
     loss_functions = []
     for name in output_names:
@@ -1363,7 +1364,7 @@ def prepare_loss_functions(loss, output_names):
       loss_functions.append(get_loss_function(loss.get(name, None)))
   elif isinstance(loss, six.string_types):
     loss_functions = [get_loss_function(loss) for _ in output_names]
-  elif isinstance(loss, collections.Sequence):
+  elif isinstance(loss, collections_abc.Sequence):
     if len(loss) != len(output_names):
       raise ValueError('When passing a list as loss, it should have one entry '
                        'per model outputs. The model has {} outputs, but you '
@@ -1397,7 +1398,7 @@ def prepare_loss_weights(training_endpoints, loss_weights=None):
   if loss_weights is None:
     for e in training_endpoints:
       e.loss_weight = 1.
-  elif isinstance(loss_weights, collections.Mapping):
+  elif isinstance(loss_weights, collections_abc.Mapping):
     generic_utils.check_for_unexpected_keys(
         'loss_weights', loss_weights,
         [e.output_name for e in training_endpoints])
@@ -1704,9 +1705,9 @@ def should_run_validation(validation_freq, epoch):
       raise ValueError('`validation_freq` can not be less than 1.')
     return one_indexed_epoch % validation_freq == 0
 
-  if not isinstance(validation_freq, collections.Container):
+  if not isinstance(validation_freq, collections_abc.Container):
     raise ValueError('`validation_freq` must be an Integer or '
-                     '`collections.Container` (e.g. list, tuple, etc.)')
+                     '`collections.abc.Container` (e.g. list, tuple, etc.)')
   return one_indexed_epoch in validation_freq
 
 

tensorflow/python/keras/engine/training_v1.py

@@ -18,6 +18,7 @@ from __future__ import division
 from __future__ import print_function
 
 import collections
+import collections.abc as collections_abc
 import warnings
 
 import numpy as np
@@ -745,7 +746,7 @@ class Model(training_lib.Model):
             the dataset at each epoch. This ensures that the same validation
             samples are used every time.
         validation_freq: Only relevant if validation data is provided. Integer
-            or `collections.Container` instance (e.g. list, tuple, etc.).
+            or `collections.abc.Container` instance (e.g. list, tuple, etc.).
             If an integer, specifies how many training epochs to run before a
             new validation run is performed, e.g. `validation_freq=2` runs
             validation every 2 epochs. If a Container, specifies the epochs on
@@ -1207,7 +1208,7 @@ class Model(training_lib.Model):
     # at this point.
     if self.run_eagerly or self._distribution_strategy:
       inputs = training_utils_v1.cast_if_floating_dtype(inputs)
-      if isinstance(inputs, collections.Sequence):
+      if isinstance(inputs, collections_abc.Sequence):
         # Unwrap lists with only one input, as we do when training on batch
         if len(inputs) == 1:
           inputs = inputs[0]

tensorflow/python/keras/keras_parameterized.py

@@ -18,7 +18,7 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
-import collections
+import collections.abc as collections_abc
 import functools
 import itertools
 import unittest
@@ -473,7 +473,7 @@ def _test_or_class_decorator(test_or_class, single_method_decorator):
     The decorated result.
   """
   def _decorate_test_or_class(obj):
-    if isinstance(obj, collections.Iterable):
+    if isinstance(obj, collections_abc.Iterable):
       return itertools.chain.from_iterable(
           single_method_decorator(method) for method in obj)
     if isinstance(obj, type):

tensorflow/python/keras/layers/preprocessing/preprocessing_test_utils.py

@@ -18,7 +18,7 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
-import collections
+import collections.abc as collections_abc
 import numpy as np
 
 from tensorflow.python.platform import test
@@ -37,7 +37,7 @@ class PreprocessingLayerTest(test.TestCase):
       self.assertEqual(len(a), len(b))
       for a_value, b_value in zip(a, b):
         self.assertAllCloseOrEqual(a_value, b_value, msg=msg)
-    elif isinstance(a, collections.Mapping):
+    elif isinstance(a, collections_abc.Mapping):
       self.assertEqual(len(a), len(b))
       for key, a_value in a.items():
         b_value = b[key]

tensorflow/python/keras/layers/recurrent.py

@@ -19,7 +19,7 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
-import collections
+import collections.abc as collections_abc
 import warnings
 
 import numpy as np
@@ -47,6 +47,7 @@ from tensorflow.python.platform import tf_logging as logging
 from tensorflow.python.training.tracking import base as trackable
 from tensorflow.python.training.tracking import data_structures
 from tensorflow.python.util import nest
+from tensorflow.python.util.compat import collections_abc
 from tensorflow.python.util.tf_export import keras_export
 from tensorflow.tools.docs import doc_controls
 
@@ -828,7 +829,7 @@ class RNN(Layer):
     # input shape: `(samples, time (padded with zeros), input_dim)`
     # note that the .build() method of subclasses MUST define
     # self.input_spec and self.state_spec with complete input shapes.
-    if (isinstance(inputs, collections.Sequence)
+    if (isinstance(inputs, collections_abc.Sequence)
         and not isinstance(inputs, tuple)):
       # get initial_state from full input spec
       # as they could be copied to multiple GPU.

tensorflow/python/keras/saving/saved_model/json_utils.py

@@ -25,7 +25,7 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
-import collections
+import collections.abc as collections_abc
 import json
 import numpy as np
 import wrapt
@@ -117,7 +117,7 @@ def get_json_type(obj):
   if isinstance(obj, dtypes.DType):
     return obj.name
 
-  if isinstance(obj, collections.Mapping):
+  if isinstance(obj, collections_abc.Mapping):
     return dict(obj)
 
   if obj is Ellipsis:

tensorflow/python/keras/saving/saving_utils.py

@@ -17,7 +17,7 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
-import collections
+import collections.abc as collections_abc
 import copy
 import os
 import six
@@ -82,7 +82,7 @@ def model_input_signature(model, keep_original_batch_size=False):
   input_specs = _enforce_names_consistency(input_specs)
   # Return a list with a single element as the model's input signature.
   if isinstance(input_specs,
-                collections.Sequence) and len(input_specs) == 1:
+                collections_abc.Sequence) and len(input_specs) == 1:
     # Note that the isinstance check filters out single-element dictionaries,
     # which should also be wrapped as a single-element list.
     return input_specs