# ==============================================================================
"""Tests for losses_utils."""

import tensorflow.compat.v2 as tf
from keras import combinations
from keras.utils import losses_utils


@combinations.generate(combinations.combine(mode=['graph', 'eager']))

            

Reported by Pylint.

              
@combinations.generate(combinations.combine(mode=['graph', 'eager']))
class RemoveSqueezableTest(tf.test.TestCase):
  """Test remove_squeezable_dimensions"""

  def test_ragged_3d_same_shape(self):
    """ shape (2, (sequence={1, 2}), 3)"""
    x = tf.ragged.constant([[[1, 2, 3]], [[4, 5, 6], [7, 8, 9]]])
    rank = x.shape.ndims

            

Reported by Pylint.

              class RemoveSqueezableTest(tf.test.TestCase):
  """Test remove_squeezable_dimensions"""

  def test_ragged_3d_same_shape(self):
    """ shape (2, (sequence={1, 2}), 3)"""
    x = tf.ragged.constant([[[1, 2, 3]], [[4, 5, 6], [7, 8, 9]]])
    rank = x.shape.ndims
    x_p, _ = losses_utils.remove_squeezable_dimensions(x, x)
    self.assertEqual(x_p.shape.ndims, rank)

            

Reported by Pylint.

                """Test remove_squeezable_dimensions"""

  def test_ragged_3d_same_shape(self):
    """ shape (2, (sequence={1, 2}), 3)"""
    x = tf.ragged.constant([[[1, 2, 3]], [[4, 5, 6], [7, 8, 9]]])
    rank = x.shape.ndims
    x_p, _ = losses_utils.remove_squeezable_dimensions(x, x)
    self.assertEqual(x_p.shape.ndims, rank)


            

Reported by Pylint.

              
  def test_ragged_3d_same_shape(self):
    """ shape (2, (sequence={1, 2}), 3)"""
    x = tf.ragged.constant([[[1, 2, 3]], [[4, 5, 6], [7, 8, 9]]])
    rank = x.shape.ndims
    x_p, _ = losses_utils.remove_squeezable_dimensions(x, x)
    self.assertEqual(x_p.shape.ndims, rank)

  def test_ragged_3d_4d_squeezable(self):

            

Reported by Pylint.

              
  def test_ragged_3d_same_shape(self):
    """ shape (2, (sequence={1, 2}), 3)"""
    x = tf.ragged.constant([[[1, 2, 3]], [[4, 5, 6], [7, 8, 9]]])
    rank = x.shape.ndims
    x_p, _ = losses_utils.remove_squeezable_dimensions(x, x)
    self.assertEqual(x_p.shape.ndims, rank)

  def test_ragged_3d_4d_squeezable(self):

            

Reported by Pylint.

                def test_ragged_3d_same_shape(self):
    """ shape (2, (sequence={1, 2}), 3)"""
    x = tf.ragged.constant([[[1, 2, 3]], [[4, 5, 6], [7, 8, 9]]])
    rank = x.shape.ndims
    x_p, _ = losses_utils.remove_squeezable_dimensions(x, x)
    self.assertEqual(x_p.shape.ndims, rank)

  def test_ragged_3d_4d_squeezable(self):
    """ shapes:

            

Reported by Pylint.

                  """ shape (2, (sequence={1, 2}), 3)"""
    x = tf.ragged.constant([[[1, 2, 3]], [[4, 5, 6], [7, 8, 9]]])
    rank = x.shape.ndims
    x_p, _ = losses_utils.remove_squeezable_dimensions(x, x)
    self.assertEqual(x_p.shape.ndims, rank)

  def test_ragged_3d_4d_squeezable(self):
    """ shapes:


            

Reported by Pylint.

                  x = tf.ragged.constant([[[1, 2, 3]], [[4, 5, 6], [7, 8, 9]]])
    rank = x.shape.ndims
    x_p, _ = losses_utils.remove_squeezable_dimensions(x, x)
    self.assertEqual(x_p.shape.ndims, rank)

  def test_ragged_3d_4d_squeezable(self):
    """ shapes:

        x: (2, (sequence={1, 2}), 3)

            

Reported by Pylint.

                  x_p, _ = losses_utils.remove_squeezable_dimensions(x, x)
    self.assertEqual(x_p.shape.ndims, rank)

  def test_ragged_3d_4d_squeezable(self):
    """ shapes:

        x: (2, (sequence={1, 2}), 3)
        y: (2, (sequence={1, 2}), 3, 1)
    """

            

Reported by Pylint.

              This file is copied from tensorflow/python/ops/control_flow_util.py.
"""

import tensorflow.compat.v2 as tf


def InXlaContext(graph):
  ctxt = graph._get_control_flow_context()  # pylint: disable=protected-access
  return GetContainingXLAContext(ctxt) is not None

            

Reported by Pylint.

              import tensorflow.compat.v2 as tf


def InXlaContext(graph):
  ctxt = graph._get_control_flow_context()  # pylint: disable=protected-access
  return GetContainingXLAContext(ctxt) is not None


def GraphOrParentsInXlaContext(graph):

            

Reported by Pylint.

              import tensorflow.compat.v2 as tf


def InXlaContext(graph):
  ctxt = graph._get_control_flow_context()  # pylint: disable=protected-access
  return GetContainingXLAContext(ctxt) is not None


def GraphOrParentsInXlaContext(graph):

            

Reported by Pylint.

              

def InXlaContext(graph):
  ctxt = graph._get_control_flow_context()  # pylint: disable=protected-access
  return GetContainingXLAContext(ctxt) is not None


def GraphOrParentsInXlaContext(graph):
  while True:

            

Reported by Pylint.

              
def InXlaContext(graph):
  ctxt = graph._get_control_flow_context()  # pylint: disable=protected-access
  return GetContainingXLAContext(ctxt) is not None


def GraphOrParentsInXlaContext(graph):
  while True:
    if InXlaContext(graph): return True

            

Reported by Pylint.

                return GetContainingXLAContext(ctxt) is not None


def GraphOrParentsInXlaContext(graph):
  while True:
    if InXlaContext(graph): return True
    try:
      graph = graph.outer_graph
    except AttributeError:

            

Reported by Pylint.

                return GetContainingXLAContext(ctxt) is not None


def GraphOrParentsInXlaContext(graph):
  while True:
    if InXlaContext(graph): return True
    try:
      graph = graph.outer_graph
    except AttributeError:

            

Reported by Pylint.

              

def GraphOrParentsInXlaContext(graph):
  while True:
    if InXlaContext(graph): return True
    try:
      graph = graph.outer_graph
    except AttributeError:
      return False

            

Reported by Pylint.

              
def GraphOrParentsInXlaContext(graph):
  while True:
    if InXlaContext(graph): return True
    try:
      graph = graph.outer_graph
    except AttributeError:
      return False


            

Reported by Pylint.

              
def GraphOrParentsInXlaContext(graph):
  while True:
    if InXlaContext(graph): return True
    try:
      graph = graph.outer_graph
    except AttributeError:
      return False


            

Reported by Pylint.

              from __future__ import division
from __future__ import print_function

import tensorflow as tf

import numpy as np

from keras.benchmarks import benchmark_util


            

Reported by Pylint.

              
import numpy as np

from keras.benchmarks import benchmark_util


class ConvMnistBenchmark(tf.test.Benchmark):
  """Benchmarks for Convnet using `tf.test.Benchmark`."""


            

Reported by Pylint.

              

class ConvMnistBenchmark(tf.test.Benchmark):
  """Benchmarks for Convnet using `tf.test.Benchmark`."""

  def __init__(self):
    super(ConvMnistBenchmark, self).__init__()
    self.num_classes = 10
    self.input_shape = (28, 28, 1)

            

Reported by Pylint.

              class ConvMnistBenchmark(tf.test.Benchmark):
  """Benchmarks for Convnet using `tf.test.Benchmark`."""

  def __init__(self):
    super(ConvMnistBenchmark, self).__init__()
    self.num_classes = 10
    self.input_shape = (28, 28, 1)
    (self.x_train, self.y_train), _ = tf.keras.datasets.mnist.load_data()
    self.x_train = self.x_train.astype('float32') / 255

            

Reported by Pylint.

                """Benchmarks for Convnet using `tf.test.Benchmark`."""

  def __init__(self):
    super(ConvMnistBenchmark, self).__init__()
    self.num_classes = 10
    self.input_shape = (28, 28, 1)
    (self.x_train, self.y_train), _ = tf.keras.datasets.mnist.load_data()
    self.x_train = self.x_train.astype('float32') / 255
    self.x_train = np.expand_dims(self.x_train, -1)

            

Reported by Pylint.

                """Benchmarks for Convnet using `tf.test.Benchmark`."""

  def __init__(self):
    super(ConvMnistBenchmark, self).__init__()
    self.num_classes = 10
    self.input_shape = (28, 28, 1)
    (self.x_train, self.y_train), _ = tf.keras.datasets.mnist.load_data()
    self.x_train = self.x_train.astype('float32') / 255
    self.x_train = np.expand_dims(self.x_train, -1)

            

Reported by Pylint.

              
  def __init__(self):
    super(ConvMnistBenchmark, self).__init__()
    self.num_classes = 10
    self.input_shape = (28, 28, 1)
    (self.x_train, self.y_train), _ = tf.keras.datasets.mnist.load_data()
    self.x_train = self.x_train.astype('float32') / 255
    self.x_train = np.expand_dims(self.x_train, -1)
    self.y_train = tf.keras.utils.to_categorical(self.y_train, self.num_classes)

            

Reported by Pylint.

                def __init__(self):
    super(ConvMnistBenchmark, self).__init__()
    self.num_classes = 10
    self.input_shape = (28, 28, 1)
    (self.x_train, self.y_train), _ = tf.keras.datasets.mnist.load_data()
    self.x_train = self.x_train.astype('float32') / 255
    self.x_train = np.expand_dims(self.x_train, -1)
    self.y_train = tf.keras.utils.to_categorical(self.y_train, self.num_classes)
    self.epochs = 15

            

Reported by Pylint.

                  super(ConvMnistBenchmark, self).__init__()
    self.num_classes = 10
    self.input_shape = (28, 28, 1)
    (self.x_train, self.y_train), _ = tf.keras.datasets.mnist.load_data()
    self.x_train = self.x_train.astype('float32') / 255
    self.x_train = np.expand_dims(self.x_train, -1)
    self.y_train = tf.keras.utils.to_categorical(self.y_train, self.num_classes)
    self.epochs = 15


            

Reported by Pylint.

                  self.num_classes = 10
    self.input_shape = (28, 28, 1)
    (self.x_train, self.y_train), _ = tf.keras.datasets.mnist.load_data()
    self.x_train = self.x_train.astype('float32') / 255
    self.x_train = np.expand_dims(self.x_train, -1)
    self.y_train = tf.keras.utils.to_categorical(self.y_train, self.num_classes)
    self.epochs = 15

  def _build_model(self):

            

Reported by Pylint.

              # ==============================================================================
"""Utilities related to distributed training."""

from absl import flags
from keras import backend

import tensorflow.compat.v2 as tf

FLAGS = flags.FLAGS

            

Reported by Pylint.

              from absl import flags
from keras import backend

import tensorflow.compat.v2 as tf

FLAGS = flags.FLAGS


# TODO(b/118776054): Currently we support global batch size for TPUStrategy and

            

Reported by Pylint.

              FLAGS = flags.FLAGS


# TODO(b/118776054): Currently we support global batch size for TPUStrategy and
# core MirroredStrategy only. Remove this check when contrib MirroredStrategy is
# no longer needed.
def global_batch_size_supported(distribution_strategy):
  return distribution_strategy.extended._global_batch_size  # pylint: disable=protected-access


            

Reported by Pylint.

                Returns:
    The result of calling `fn`.
  """
  # TODO(b/132666209): Remove this function when we support assign_*
  # for replica-local variables.
  strategy = None
  if 'strategy' in kwargs:
    strategy = kwargs.pop('strategy')
  else:

            

Reported by Pylint.

                  if tf.distribute.has_strategy():
      strategy = tf.distribute.get_strategy()

  # TODO(b/120571621): TPUStrategy does not implement replica-local variables.
  is_tpu = backend.is_tpu_strategy(strategy)
  if ((not is_tpu) and strategy and tf.distribute.in_cross_replica_context()):
    with strategy.scope():
      return strategy.extended.call_for_each_replica(fn, args, kwargs)
  return fn(*args, **kwargs)

            

Reported by Pylint.

              # TODO(b/118776054): Currently we support global batch size for TPUStrategy and
# core MirroredStrategy only. Remove this check when contrib MirroredStrategy is
# no longer needed.
def global_batch_size_supported(distribution_strategy):
  return distribution_strategy.extended._global_batch_size  # pylint: disable=protected-access


def call_replica_local_fn(fn, *args, **kwargs):
  """Call a function that uses replica-local variables.

            

Reported by Pylint.

              # core MirroredStrategy only. Remove this check when contrib MirroredStrategy is
# no longer needed.
def global_batch_size_supported(distribution_strategy):
  return distribution_strategy.extended._global_batch_size  # pylint: disable=protected-access


def call_replica_local_fn(fn, *args, **kwargs):
  """Call a function that uses replica-local variables.


            

Reported by Pylint.

                return distribution_strategy.extended._global_batch_size  # pylint: disable=protected-access


def call_replica_local_fn(fn, *args, **kwargs):
  """Call a function that uses replica-local variables.

  This function correctly handles calling `fn` in a cross-replica
  context.


            

Reported by Pylint.

              

def call_replica_local_fn(fn, *args, **kwargs):
  """Call a function that uses replica-local variables.

  This function correctly handles calling `fn` in a cross-replica
  context.

  Args:

            

Reported by Pylint.

                """
  # TODO(b/132666209): Remove this function when we support assign_*
  # for replica-local variables.
  strategy = None
  if 'strategy' in kwargs:
    strategy = kwargs.pop('strategy')
  else:
    if tf.distribute.has_strategy():
      strategy = tf.distribute.get_strategy()

            

Reported by Pylint.

              # ==============================================================================
"""Keras SavedModel serialization."""

import tensorflow.compat.v2 as tf

import os

from keras import backend as K
from keras.protobuf import saved_metadata_pb2

            

Reported by Pylint.

              
import os

from keras import backend as K
from keras.protobuf import saved_metadata_pb2
from keras.protobuf import versions_pb2
from keras.saving import saving_utils
from keras.saving.saved_model import constants
from keras.saving.saved_model import save_impl

            

Reported by Pylint.

              import os

from keras import backend as K
from keras.protobuf import saved_metadata_pb2
from keras.protobuf import versions_pb2
from keras.saving import saving_utils
from keras.saving.saved_model import constants
from keras.saving.saved_model import save_impl
from keras.saving.saved_model import utils

            

Reported by Pylint.

              
from keras import backend as K
from keras.protobuf import saved_metadata_pb2
from keras.protobuf import versions_pb2
from keras.saving import saving_utils
from keras.saving.saved_model import constants
from keras.saving.saved_model import save_impl
from keras.saving.saved_model import utils
from keras.utils.generic_utils import LazyLoader

            

Reported by Pylint.

              from keras import backend as K
from keras.protobuf import saved_metadata_pb2
from keras.protobuf import versions_pb2
from keras.saving import saving_utils
from keras.saving.saved_model import constants
from keras.saving.saved_model import save_impl
from keras.saving.saved_model import utils
from keras.utils.generic_utils import LazyLoader
from keras.utils.io_utils import ask_to_proceed_with_overwrite

            

Reported by Pylint.

              from keras.protobuf import saved_metadata_pb2
from keras.protobuf import versions_pb2
from keras.saving import saving_utils
from keras.saving.saved_model import constants
from keras.saving.saved_model import save_impl
from keras.saving.saved_model import utils
from keras.utils.generic_utils import LazyLoader
from keras.utils.io_utils import ask_to_proceed_with_overwrite
from tensorflow.python.saved_model import save as save_lib

            

Reported by Pylint.

              from keras.protobuf import versions_pb2
from keras.saving import saving_utils
from keras.saving.saved_model import constants
from keras.saving.saved_model import save_impl
from keras.saving.saved_model import utils
from keras.utils.generic_utils import LazyLoader
from keras.utils.io_utils import ask_to_proceed_with_overwrite
from tensorflow.python.saved_model import save as save_lib


            

Reported by Pylint.

              from keras.saving import saving_utils
from keras.saving.saved_model import constants
from keras.saving.saved_model import save_impl
from keras.saving.saved_model import utils
from keras.utils.generic_utils import LazyLoader
from keras.utils.io_utils import ask_to_proceed_with_overwrite
from tensorflow.python.saved_model import save as save_lib

# To avoid circular dependencies between keras/engine and keras/saving,

            

Reported by Pylint.

              from keras.saving.saved_model import constants
from keras.saving.saved_model import save_impl
from keras.saving.saved_model import utils
from keras.utils.generic_utils import LazyLoader
from keras.utils.io_utils import ask_to_proceed_with_overwrite
from tensorflow.python.saved_model import save as save_lib

# To avoid circular dependencies between keras/engine and keras/saving,
# code in keras/saving must delay imports.

            

Reported by Pylint.

              from keras.saving.saved_model import save_impl
from keras.saving.saved_model import utils
from keras.utils.generic_utils import LazyLoader
from keras.utils.io_utils import ask_to_proceed_with_overwrite
from tensorflow.python.saved_model import save as save_lib

# To avoid circular dependencies between keras/engine and keras/saving,
# code in keras/saving must delay imports.


            

Reported by Pylint.

              # limitations under the License.
# ==============================================================================
"""Saving utilities to support Python's Pickle protocol."""
# pylint: disable=g-bad-import-order
import tensorflow.compat.v2 as tf

import os
import tarfile
import io

            

Reported by Pylint.

              # ==============================================================================
"""Saving utilities to support Python's Pickle protocol."""
# pylint: disable=g-bad-import-order
import tensorflow.compat.v2 as tf

import os
import tarfile
import io
import uuid

            

Reported by Pylint.

              # pylint: disable=g-bad-import-order
import tensorflow.compat.v2 as tf

import os
import tarfile
import io
import uuid
import numpy


            

Reported by Pylint.

              import tensorflow.compat.v2 as tf

import os
import tarfile
import io
import uuid
import numpy

from keras.saving import save as save_module

            

Reported by Pylint.

              
import os
import tarfile
import io
import uuid
import numpy

from keras.saving import save as save_module


            

Reported by Pylint.

              import os
import tarfile
import io
import uuid
import numpy

from keras.saving import save as save_module



            

Reported by Pylint.

              

def deserialize_model_from_bytecode(serialized_model):
  """Reconstruct a Model from the output of `serialize_model_as_bytecode`.

  Args:
      serialized_model: (np.array) return value from
        `serialize_model_as_bytecode`.


            

Reported by Pylint.

                Returns:
      keras.Model: Keras Model instance.
  """
  temp_dir = f"ram://{uuid.uuid4()}"
  b = io.BytesIO(serialized_model)
  with tarfile.open(fileobj=b, mode="r") as archive:
    for name in archive.getnames():
      dest_path = os.path.join(temp_dir, name)
      member = archive.getmember(name)

            

Reported by Pylint.

                    keras.Model: Keras Model instance.
  """
  temp_dir = f"ram://{uuid.uuid4()}"
  b = io.BytesIO(serialized_model)
  with tarfile.open(fileobj=b, mode="r") as archive:
    for name in archive.getnames():
      dest_path = os.path.join(temp_dir, name)
      member = archive.getmember(name)
      tf.io.gfile.makedirs(os.path.dirname(dest_path))

            

Reported by Pylint.

                    keras.Model: Keras Model instance.
  """
  temp_dir = f"ram://{uuid.uuid4()}"
  b = io.BytesIO(serialized_model)
  with tarfile.open(fileobj=b, mode="r") as archive:
    for name in archive.getnames():
      dest_path = os.path.join(temp_dir, name)
      member = archive.getmember(name)
      tf.io.gfile.makedirs(os.path.dirname(dest_path))

            

Reported by Pylint.

              # ==============================================================================
"""InputSpec tests."""

import tensorflow.compat.v2 as tf
from keras import layers
from keras.engine import keras_tensor


class KerasTensorTest(tf.test.TestCase):

            

Reported by Pylint.

              from keras.engine import keras_tensor


class KerasTensorTest(tf.test.TestCase):

  def test_repr_and_string(self):
    kt = keras_tensor.KerasTensor(
        type_spec=tf.TensorSpec(shape=(1, 2, 3), dtype=tf.float32))
    expected_str = ("KerasTensor(type_spec=TensorSpec(shape=(1, 2, 3), "

            

Reported by Pylint.

              from keras.engine import keras_tensor


class KerasTensorTest(tf.test.TestCase):

  def test_repr_and_string(self):
    kt = keras_tensor.KerasTensor(
        type_spec=tf.TensorSpec(shape=(1, 2, 3), dtype=tf.float32))
    expected_str = ("KerasTensor(type_spec=TensorSpec(shape=(1, 2, 3), "

            

Reported by Pylint.

              
class KerasTensorTest(tf.test.TestCase):

  def test_repr_and_string(self):
    kt = keras_tensor.KerasTensor(
        type_spec=tf.TensorSpec(shape=(1, 2, 3), dtype=tf.float32))
    expected_str = ("KerasTensor(type_spec=TensorSpec(shape=(1, 2, 3), "
                    "dtype=tf.float32, name=None))")
    expected_repr = "<KerasTensor: shape=(1, 2, 3) dtype=float32>"

            

Reported by Pylint.

              
class KerasTensorTest(tf.test.TestCase):

  def test_repr_and_string(self):
    kt = keras_tensor.KerasTensor(
        type_spec=tf.TensorSpec(shape=(1, 2, 3), dtype=tf.float32))
    expected_str = ("KerasTensor(type_spec=TensorSpec(shape=(1, 2, 3), "
                    "dtype=tf.float32, name=None))")
    expected_repr = "<KerasTensor: shape=(1, 2, 3) dtype=float32>"

            

Reported by Pylint.

              class KerasTensorTest(tf.test.TestCase):

  def test_repr_and_string(self):
    kt = keras_tensor.KerasTensor(
        type_spec=tf.TensorSpec(shape=(1, 2, 3), dtype=tf.float32))
    expected_str = ("KerasTensor(type_spec=TensorSpec(shape=(1, 2, 3), "
                    "dtype=tf.float32, name=None))")
    expected_repr = "<KerasTensor: shape=(1, 2, 3) dtype=float32>"
    self.assertEqual(expected_str, str(kt))

            

Reported by Pylint.

              class KerasTensorTest(tf.test.TestCase):

  def test_repr_and_string(self):
    kt = keras_tensor.KerasTensor(
        type_spec=tf.TensorSpec(shape=(1, 2, 3), dtype=tf.float32))
    expected_str = ("KerasTensor(type_spec=TensorSpec(shape=(1, 2, 3), "
                    "dtype=tf.float32, name=None))")
    expected_repr = "<KerasTensor: shape=(1, 2, 3) dtype=float32>"
    self.assertEqual(expected_str, str(kt))

            

Reported by Pylint.

                def test_repr_and_string(self):
    kt = keras_tensor.KerasTensor(
        type_spec=tf.TensorSpec(shape=(1, 2, 3), dtype=tf.float32))
    expected_str = ("KerasTensor(type_spec=TensorSpec(shape=(1, 2, 3), "
                    "dtype=tf.float32, name=None))")
    expected_repr = "<KerasTensor: shape=(1, 2, 3) dtype=float32>"
    self.assertEqual(expected_str, str(kt))
    self.assertEqual(expected_repr, repr(kt))


            

Reported by Pylint.

                      type_spec=tf.TensorSpec(shape=(1, 2, 3), dtype=tf.float32))
    expected_str = ("KerasTensor(type_spec=TensorSpec(shape=(1, 2, 3), "
                    "dtype=tf.float32, name=None))")
    expected_repr = "<KerasTensor: shape=(1, 2, 3) dtype=float32>"
    self.assertEqual(expected_str, str(kt))
    self.assertEqual(expected_repr, repr(kt))

    kt = keras_tensor.KerasTensor(
        type_spec=tf.TensorSpec(shape=(2,), dtype=tf.int32),

            

Reported by Pylint.

                  expected_str = ("KerasTensor(type_spec=TensorSpec(shape=(1, 2, 3), "
                    "dtype=tf.float32, name=None))")
    expected_repr = "<KerasTensor: shape=(1, 2, 3) dtype=float32>"
    self.assertEqual(expected_str, str(kt))
    self.assertEqual(expected_repr, repr(kt))

    kt = keras_tensor.KerasTensor(
        type_spec=tf.TensorSpec(shape=(2,), dtype=tf.int32),
        inferred_value=[2, 3])

            

Reported by Pylint.

              from __future__ import division
from __future__ import print_function

import tensorflow.compat.v2 as tf
from keras import backend
from keras.feature_column import base_feature_layer as kfc
from tensorflow.python.util.tf_export import keras_export

# pylint: disable=protected-access

            

Reported by Pylint.

              import tensorflow.compat.v2 as tf
from keras import backend
from keras.feature_column import base_feature_layer as kfc
from tensorflow.python.util.tf_export import keras_export

# pylint: disable=protected-access


@keras_export('keras.experimental.SequenceFeatures')

            

Reported by Pylint.

              

@keras_export('keras.experimental.SequenceFeatures')
class SequenceFeatures(kfc._BaseFeaturesLayer):
  """A layer for sequence input.

    All `feature_columns` must be sequence dense columns with the same
    `sequence_length`. The output of this method can be fed into sequence
    networks, such as RNN.

            

Reported by Pylint.

                def _target_shape(self, input_shape, total_elements):
    return (input_shape[0], input_shape[1], total_elements)

  def call(self, features, training=None):
    """Returns sequence input corresponding to the `feature_columns`.

    Args:
      features: A dict mapping keys to tensors.
      training: Python boolean or None, indicating whether to the layer is being

            

Reported by Pylint.

              
@keras_export('keras.experimental.SequenceFeatures')
class SequenceFeatures(kfc._BaseFeaturesLayer):
  """A layer for sequence input.

    All `feature_columns` must be sequence dense columns with the same
    `sequence_length`. The output of this method can be fed into sequence
    networks, such as RNN.


            

Reported by Pylint.

                  ```
  """

  def __init__(
      self,
      feature_columns,
      trainable=True,
      name=None,
      **kwargs):

            

Reported by Pylint.

                    trainable=True,
      name=None,
      **kwargs):
    """"Constructs a SequenceFeatures layer.

    Args:
      feature_columns: An iterable of dense sequence columns. Valid columns are
        - `embedding_column` that wraps a `sequence_categorical_column_with_*`
        - `sequence_numeric_column`.

            

Reported by Pylint.

                    ValueError: If any of the `feature_columns` is not a
        `SequenceDenseColumn`.
    """
    super(SequenceFeatures, self).__init__(
        feature_columns=feature_columns,
        trainable=trainable,
        name=name,
        expected_column_type=tf.__internal__.feature_column.SequenceDenseColumn,
        **kwargs)

            

Reported by Pylint.

                    ValueError: If any of the `feature_columns` is not a
        `SequenceDenseColumn`.
    """
    super(SequenceFeatures, self).__init__(
        feature_columns=feature_columns,
        trainable=trainable,
        name=name,
        expected_column_type=tf.__internal__.feature_column.SequenceDenseColumn,
        **kwargs)

            

Reported by Pylint.

                      expected_column_type=tf.__internal__.feature_column.SequenceDenseColumn,
        **kwargs)

  @property
  def _is_feature_layer(self):
    return True

  def _target_shape(self, input_shape, total_elements):
    return (input_shape[0], input_shape[1], total_elements)

            

Reported by Pylint.

              # ==============================================================================
"""Tests Keras integration with enable_mixed_precision_graph_rewrite()."""

import tensorflow.compat.v2 as tf

import os
from keras import combinations
from keras import keras_parameterized
from keras import testing_utils

            

Reported by Pylint.

              
import tensorflow.compat.v2 as tf

import os
from keras import combinations
from keras import keras_parameterized
from keras import testing_utils
from keras.mixed_precision import loss_scale_optimizer as loss_scale_optimizer_v2
from keras.mixed_precision import policy

            

Reported by Pylint.

              from keras.optimizer_v2 import gradient_descent as gradient_descent_v2


class MixedPrecisionTest(keras_parameterized.TestCase):

  IGNORE_PERF_VAR = 'TF_AUTO_MIXED_PRECISION_GRAPH_REWRITE_IGNORE_PERFORMANCE'

  def setUp(self):
    super(MixedPrecisionTest, self).setUp()

            

Reported by Pylint.

              
class MixedPrecisionTest(keras_parameterized.TestCase):

  IGNORE_PERF_VAR = 'TF_AUTO_MIXED_PRECISION_GRAPH_REWRITE_IGNORE_PERFORMANCE'

  def setUp(self):
    super(MixedPrecisionTest, self).setUp()
    # Enable the tests to be run on pre-Volta GPUs by telling the grappler pass
    # to ignore performance and always transform the graph.

            

Reported by Pylint.

              
  IGNORE_PERF_VAR = 'TF_AUTO_MIXED_PRECISION_GRAPH_REWRITE_IGNORE_PERFORMANCE'

  def setUp(self):
    super(MixedPrecisionTest, self).setUp()
    # Enable the tests to be run on pre-Volta GPUs by telling the grappler pass
    # to ignore performance and always transform the graph.
    self._original_ignore_perf_value = os.getenv(self.IGNORE_PERF_VAR)
    os.environ[self.IGNORE_PERF_VAR] = '1'

            

Reported by Pylint.

              
  IGNORE_PERF_VAR = 'TF_AUTO_MIXED_PRECISION_GRAPH_REWRITE_IGNORE_PERFORMANCE'

  def setUp(self):
    super(MixedPrecisionTest, self).setUp()
    # Enable the tests to be run on pre-Volta GPUs by telling the grappler pass
    # to ignore performance and always transform the graph.
    self._original_ignore_perf_value = os.getenv(self.IGNORE_PERF_VAR)
    os.environ[self.IGNORE_PERF_VAR] = '1'

            

Reported by Pylint.

              
  IGNORE_PERF_VAR = 'TF_AUTO_MIXED_PRECISION_GRAPH_REWRITE_IGNORE_PERFORMANCE'

  def setUp(self):
    super(MixedPrecisionTest, self).setUp()
    # Enable the tests to be run on pre-Volta GPUs by telling the grappler pass
    # to ignore performance and always transform the graph.
    self._original_ignore_perf_value = os.getenv(self.IGNORE_PERF_VAR)
    os.environ[self.IGNORE_PERF_VAR] = '1'

            

Reported by Pylint.

                IGNORE_PERF_VAR = 'TF_AUTO_MIXED_PRECISION_GRAPH_REWRITE_IGNORE_PERFORMANCE'

  def setUp(self):
    super(MixedPrecisionTest, self).setUp()
    # Enable the tests to be run on pre-Volta GPUs by telling the grappler pass
    # to ignore performance and always transform the graph.
    self._original_ignore_perf_value = os.getenv(self.IGNORE_PERF_VAR)
    os.environ[self.IGNORE_PERF_VAR] = '1'


            

Reported by Pylint.

                IGNORE_PERF_VAR = 'TF_AUTO_MIXED_PRECISION_GRAPH_REWRITE_IGNORE_PERFORMANCE'

  def setUp(self):
    super(MixedPrecisionTest, self).setUp()
    # Enable the tests to be run on pre-Volta GPUs by telling the grappler pass
    # to ignore performance and always transform the graph.
    self._original_ignore_perf_value = os.getenv(self.IGNORE_PERF_VAR)
    os.environ[self.IGNORE_PERF_VAR] = '1'


            

Reported by Pylint.

                  super(MixedPrecisionTest, self).setUp()
    # Enable the tests to be run on pre-Volta GPUs by telling the grappler pass
    # to ignore performance and always transform the graph.
    self._original_ignore_perf_value = os.getenv(self.IGNORE_PERF_VAR)
    os.environ[self.IGNORE_PERF_VAR] = '1'

  def tearDown(self):
    # Set the IGNORE_PERF_VAR variable back to it's original value.
    if self._original_ignore_perf_value is not None:

            

Reported by Pylint.

              # ==============================================================================
"""Tests for Keras subclassed layers utilizing desired user syntax."""

import tensorflow.compat.v2 as tf

import keras
from keras import keras_parameterized
from keras import testing_utils
from keras.utils import tf_utils

            

Reported by Pylint.

                  class BuildConstantLayer(keras.layers.Layer):

      def build(self, input_shape):
        self.b = tf.convert_to_tensor(2.0)

      def call(self, inputs):
        return self.b * inputs

    layer = BuildConstantLayer()

            

Reported by Pylint.

                    def build(self, input_shape):
        self.b = tf.convert_to_tensor(2.0)

      def call(self, inputs):
        return self.b * inputs

    layer = BuildConstantLayer()
    model = testing_utils.get_model_from_layers(
        [layer, keras.layers.Dense(1)], input_shape=(1,))

            

Reported by Pylint.

                    def build(self, input_shape):
        a = tf.convert_to_tensor(1.0)
        b = 2.0 * a
        self.variable = tf.Variable(b)
        self.constant = tf.convert_to_tensor(self.variable)

      def call(self, inputs):
        return self.variable * self.constant * inputs


            

Reported by Pylint.

                      a = tf.convert_to_tensor(1.0)
        b = 2.0 * a
        self.variable = tf.Variable(b)
        self.constant = tf.convert_to_tensor(self.variable)

      def call(self, inputs):
        return self.variable * self.constant * inputs

    layer = BuildDerivedConstantLayer()

            

Reported by Pylint.

                      self.variable = tf.Variable(b)
        self.constant = tf.convert_to_tensor(self.variable)

      def call(self, inputs):
        return self.variable * self.constant * inputs

    layer = BuildDerivedConstantLayer()
    model = testing_utils.get_model_from_layers(
        [layer, keras.layers.Dense(1)], input_shape=(1,))

            

Reported by Pylint.

              
@keras_parameterized.run_all_keras_modes
@keras_parameterized.run_with_all_model_types
class SubclassedLayersTest(keras_parameterized.TestCase):

  def test_simple_build_with_constant(self):

    class BuildConstantLayer(keras.layers.Layer):


            

Reported by Pylint.

              @keras_parameterized.run_with_all_model_types
class SubclassedLayersTest(keras_parameterized.TestCase):

  def test_simple_build_with_constant(self):

    class BuildConstantLayer(keras.layers.Layer):

      def build(self, input_shape):
        self.b = tf.convert_to_tensor(2.0)

            

Reported by Pylint.

              @keras_parameterized.run_with_all_model_types
class SubclassedLayersTest(keras_parameterized.TestCase):

  def test_simple_build_with_constant(self):

    class BuildConstantLayer(keras.layers.Layer):

      def build(self, input_shape):
        self.b = tf.convert_to_tensor(2.0)

            

Reported by Pylint.

              
  def test_simple_build_with_constant(self):

    class BuildConstantLayer(keras.layers.Layer):

      def build(self, input_shape):
        self.b = tf.convert_to_tensor(2.0)

      def call(self, inputs):

            

Reported by Pylint.