Deep Learning
New TensorFlow Release 1.13.0
January 28, 2019
8 min read
Here is an overview of TensorFlow's latest release 1.13.0.
tf.lite
and source code is now under tensorflow/lite
rather than tensorflow/contrib/lite
.tf.constant
.gain
argument of convolutional orthogonal initializers (convolutional_delta_orthogonal
, convolutional_orthogonal_1D
, convolutional_orthogonal_2D
, convolutional_orthogonal_3D
) have consistent behavior with the tf.initializers.orthogonal
initializer, i.e. scale the output l2-norm by gain
and NOT by sqrt(gain)
. (Note that these functions are currently in tf.contrib
which is not guaranteed backward compatible).tf.acos
, tf.acosh
, tf.add
, tf.as_string
, tf.asin
, tf.asinh
, tf.atan,
tf.atan2,
tf.atanh,
tf.cos,
tf.cosh,
tf.equal,
tf.exp,
tf.floor,
tf.greater,
tf.greater_equal,
tf.less,
tf.less_equal,
tf.log,
tf.logp1,
tf.logical_and,
tf.logical_not,
tf.logical_or,
tf.maximum,
tf.minimum,
tf.not_equal,
tf.sin,
tf.sinh,
tf.tan`tf.data.Dataset.shard
.saved_model.loader.load
which is replaced by saved_model.load
and saved_model.main_op
, which will be replaced by saved_model.main_op
in V2.Variable.count_up_to
and tf.count_up_to
in favor of Dataset.range
.confusion_matrix
op as tf.math.confusion_matrix
instead of tf.train.confusion_matrix
.tf.dtypes.
endpoint for every constant in dtypes.py; moving endpoints in versions.py to corresponding endpoints in tf.sysconfig.
and tf.version.
; moving all constants under tf.saved_model
submodules to tf.saved_model
module. New endpoints are added in V1 and V2 but existing endpoint removals are only applied in V2.tf.register_tensor_conversion_function
.tf.contrib.saved_model.save_keras_model
.LinearOperator.matmul
now returns a new LinearOperator
.ignore_unknown
argument to parse_values
which suppresses ValueError for unknown hyperparameter types. Such * Add tf.linalg.matvec
convenience function.tf.einsum()
raises ValueError
for unsupported equations like "ii->"
.tf.signal.dct
and tf.signal.idct
.round_mode
to QuantizeAndDequantizeV2
op to select rounding algorithm.unicode_encode
, unicode_decode
, unicode_decode_with_offsets
, unicode_split
, unicode_split_with_offset
, and unicode_transcode
ops. Amongst other things, this Op adds the ability to encode, decode, and transcode a variety of input text encoding formats into the main Unicode encodings (UTF-8, UTF-16-BE, UTF-32-BE)SpaceToDepth
supports uint8 data type.tf.nn.safe_embedding_lookup_sparse
, tf.nn.sampled_softmax
and tf.nn.nce_loss
.tf.spectral
into tf.signal
for TensorFlow 2.0.tensorflow/contrib/lite
to tensorflow/lite
.tf.contrib
:rate
argument, keep_prob
is deprecated.tf.contrib.estimator
were changed to tf.estimator
:tf.contrib.estimator.BaselineEstimator
with tf.estimator.BaselineEstimator
tf.contrib.estimator.DNNLinearCombinedEstimator
with tf.estimator.DNNLinearCombinedEstimator
tf.contrib.estimator.DNNEstimator
with tf.estimator.DNNEstimator
tf.contrib.estimator.LinearEstimator
with tf.estimator.LinearEstimator
tf.contrib.estimator.InMemoryEvaluatorHook
and tf.estimator.experimental.InMemoryEvaluatorHook`.tf.contrib.estimator.make_stop_at_checkpoint_step_hook
with tf.estimator.experimental.make_stop_at_checkpoint_step_hook
.tf.contrib.signal
to tf.signal
(preserving aliases in tf.contrib.signal).tf.contrib.estimator.export_all_saved_models
and related should switch to tf.estimator.Estimator.experimental_export_all_saved_models
.tf.data.experimental.StatsOptions()
, to configure options to collect statistics from tf.data.Dataset
pipeline using dStatsAggregatord. Adds option "experimental_stats" to tf.data.Options
which takes tf.data.experimental.StatsOptions
object. Deprecates tf.data.experimental.set_stats_agregator
.tf.data.Dataset.make_one_shot_iterator()
in V1, removed it from V2, and added tf.compat.v1.data.make_one_shot_iterator()`.tf.data.Dataset.make_initializable_iterator()
in V1, removed it from V2, and added tf.compat.v1.data.make_initializable_iterator()
.tf.data
transformations.tf.data.Dataset
implementers: Added tf.data.Dataset._element_structured property
to replace Dataset.output_{types,shapes,classes}
.EVP_MD_CTX_destroy
.:android_tensorflow_lib_selective_registration*
targets, use :android_tensorflow_lib_lite*
targets instead.RoundToEven
function to xla/client/lib/math.h.TF_XLA_DEBUG_OPTIONS_PASSTHROUGH
set to "1" or "true" allows the debug options passed within an XRTCompile op to be passed directly to the XLA compilation backend. If such variable is not set (service side), only a restricted set will be passed through.Here is an overview of TensorFlow's latest release 1.13.0.
tf.lite
and source code is now under tensorflow/lite
rather than tensorflow/contrib/lite
.tf.constant
.gain
argument of convolutional orthogonal initializers (convolutional_delta_orthogonal
, convolutional_orthogonal_1D
, convolutional_orthogonal_2D
, convolutional_orthogonal_3D
) have consistent behavior with the tf.initializers.orthogonal
initializer, i.e. scale the output l2-norm by gain
and NOT by sqrt(gain)
. (Note that these functions are currently in tf.contrib
which is not guaranteed backward compatible).tf.acos
, tf.acosh
, tf.add
, tf.as_string
, tf.asin
, tf.asinh
, tf.atan,
tf.atan2,
tf.atanh,
tf.cos,
tf.cosh,
tf.equal,
tf.exp,
tf.floor,
tf.greater,
tf.greater_equal,
tf.less,
tf.less_equal,
tf.log,
tf.logp1,
tf.logical_and,
tf.logical_not,
tf.logical_or,
tf.maximum,
tf.minimum,
tf.not_equal,
tf.sin,
tf.sinh,
tf.tan`tf.data.Dataset.shard
.saved_model.loader.load
which is replaced by saved_model.load
and saved_model.main_op
, which will be replaced by saved_model.main_op
in V2.Variable.count_up_to
and tf.count_up_to
in favor of Dataset.range
.confusion_matrix
op as tf.math.confusion_matrix
instead of tf.train.confusion_matrix
.tf.dtypes.
endpoint for every constant in dtypes.py; moving endpoints in versions.py to corresponding endpoints in tf.sysconfig.
and tf.version.
; moving all constants under tf.saved_model
submodules to tf.saved_model
module. New endpoints are added in V1 and V2 but existing endpoint removals are only applied in V2.tf.register_tensor_conversion_function
.tf.contrib.saved_model.save_keras_model
.LinearOperator.matmul
now returns a new LinearOperator
.ignore_unknown
argument to parse_values
which suppresses ValueError for unknown hyperparameter types. Such * Add tf.linalg.matvec
convenience function.tf.einsum()
raises ValueError
for unsupported equations like "ii->"
.tf.signal.dct
and tf.signal.idct
.round_mode
to QuantizeAndDequantizeV2
op to select rounding algorithm.unicode_encode
, unicode_decode
, unicode_decode_with_offsets
, unicode_split
, unicode_split_with_offset
, and unicode_transcode
ops. Amongst other things, this Op adds the ability to encode, decode, and transcode a variety of input text encoding formats into the main Unicode encodings (UTF-8, UTF-16-BE, UTF-32-BE)SpaceToDepth
supports uint8 data type.tf.nn.safe_embedding_lookup_sparse
, tf.nn.sampled_softmax
and tf.nn.nce_loss
.tf.spectral
into tf.signal
for TensorFlow 2.0.tensorflow/contrib/lite
to tensorflow/lite
.tf.contrib
:rate
argument, keep_prob
is deprecated.tf.contrib.estimator
were changed to tf.estimator
:tf.contrib.estimator.BaselineEstimator
with tf.estimator.BaselineEstimator
tf.contrib.estimator.DNNLinearCombinedEstimator
with tf.estimator.DNNLinearCombinedEstimator
tf.contrib.estimator.DNNEstimator
with tf.estimator.DNNEstimator
tf.contrib.estimator.LinearEstimator
with tf.estimator.LinearEstimator
tf.contrib.estimator.InMemoryEvaluatorHook
and tf.estimator.experimental.InMemoryEvaluatorHook`.tf.contrib.estimator.make_stop_at_checkpoint_step_hook
with tf.estimator.experimental.make_stop_at_checkpoint_step_hook
.tf.contrib.signal
to tf.signal
(preserving aliases in tf.contrib.signal).tf.contrib.estimator.export_all_saved_models
and related should switch to tf.estimator.Estimator.experimental_export_all_saved_models
.tf.data.experimental.StatsOptions()
, to configure options to collect statistics from tf.data.Dataset
pipeline using dStatsAggregatord. Adds option "experimental_stats" to tf.data.Options
which takes tf.data.experimental.StatsOptions
object. Deprecates tf.data.experimental.set_stats_agregator
.tf.data.Dataset.make_one_shot_iterator()
in V1, removed it from V2, and added tf.compat.v1.data.make_one_shot_iterator()`.tf.data.Dataset.make_initializable_iterator()
in V1, removed it from V2, and added tf.compat.v1.data.make_initializable_iterator()
.tf.data
transformations.tf.data.Dataset
implementers: Added tf.data.Dataset._element_structured property
to replace Dataset.output_{types,shapes,classes}
.EVP_MD_CTX_destroy
.:android_tensorflow_lib_selective_registration*
targets, use :android_tensorflow_lib_lite*
targets instead.RoundToEven
function to xla/client/lib/math.h.TF_XLA_DEBUG_OPTIONS_PASSTHROUGH
set to "1" or "true" allows the debug options passed within an XRTCompile op to be passed directly to the XLA compilation backend. If such variable is not set (service side), only a restricted set will be passed through.