| Status | Accepted |
|---|---|
| Author(s) | Anna Revinskaya (annarev@google.com), Andrew Selle (aselle@google.com) |
| Sponsor | Martin Wicke (wicke@google.com) |
| Updated | 2018-08-27 |
This document proposes organizing TensorFlow API symbols in corresponding logical subnamespaces. As TensorFlow library grows, it is important to structure namespaces in a clear way for easier discoverability and usability.
We define endpoint as full name that can be used to access a TensorFlow
symbol. For e.g.
name_scope can be
accessed using either tf.name_scope or tf.keras.backend.name_scope.
Therefore, name_scope has 2 endpoints: tf.name_scope and
tf.keras.backend.name_scope.
At a high level we have the following goals:
- Add a few additional namespaces.
- Add additional endpoints for TensorFlow symbols in relevant namespaces.
- Remove some of the existing endpoints.
TensorFlow currently has over 2000 endpoints total including over 500 endpoints in the root namespace. As number of symbols grows, it is important to maintain a clear structure to aid discoverability.
Certain API symbol placements could be improved:
- Some namespaces were created recently and might not contain all the
corresponding symbols. For e.g.
tf.mathnamespace was added recently. Symbols such astf.roundare not intf.mathnamespace even though logically they belong in that namespace. - Some symbols are included in the root namespace even though they are rarely
used (for e.g.
tf.zeta). - Some symbols currently start with a prefix that could really be replaced by
introducing a subnamespace (for e.g.
tf.string_stripvstf.strings.strip,tf.sparse_maximumvstf.sparse.maximum). - Certain deep hierarchies seem redundant and could be flattened (for e.g.
tf.saved_model.signature_constants.CLASSIFY_INPUTScould be moved totf.saved_model.CLASSIFY_INPUTS). - To keep clear structure and reduce duplication, we want to collect all layers, losses and metrics under the
tf.kerasnamespace.
In general, we want to balance flatness and browsability. Flat hierarchies allow
for shorter endpoint names that are easy to remember (for e.g. tf.add vs
tf.math.add). At the same time subnamespaces support easier browsability (for
e.g. tf.math namespace would contain all math functions making it easier to discover available symbols).
Furthermore, TensorFlow API has many users. Therefore, we should avoid removing endpoints if they are frequently used.
We plan to add the following additional namespaces:
tf.random - will contain random sampling ops.
tf.keras.layers - will contain all symbols that are currently under tf.layers. Note that signatures of these symbols will likely change to match layers under tf.keras.layers better.
tf.keras.losses - will contain all symbols that are currently under tf.losses. Note that signatures of these symbols will likely change to match losses under tf.keras.losses better.
tf.keras.metrics - will contain all symbols that are currently under tf.metrics. Note that signatures of these symbols will likely change to match metrics under tf.keras.metrics better.
Note that we already introduced some new namespaces earlier in June, specifically
tf.debugging - ops helpful for debugging, such as asserts. We also want to
move TensorFlow Debugger to
tf.debugging namespace.
tf.dtypes - data types.
tf.io - ops for reading and writing.
tf.quantization - ops related to quantization.
We plan to deprecate entire contents of the following namespaces:
tf.logging - Python logging module can be used instead.
tf.manip - We will keep endpoints in root for symbols in tf.manip. tf.manip was added recently but most tensor manipulation ops are frequently used and it makes sense to keep them in root instead.
We will add new endpoints for existing symbols to make sure each namespace contains all relevant endpoints.
See list of endpoints we want to add in Appendix 1: Additional
Endpoints. Note: the list in the appendix does not include new endpoints for symbols under
tf.layers, tf.losses and tf.metrics namespaces since all symbols under
these namespaces will have new endpoints added under tf.keras.layers,
tf.keras.losses and tf.keras.metrics respectively. So, we don't need to
list these endpoint changes individually.
We also want to remove some of the existing endpoints. Specifically we were looking for endpoints to remove based on the following criteria:
- Remove endpoints if they have preferred replacement and if these endpoints are not frequently used.
- Remove all endpoints that have been moved to
tf.quantizationnamespace. - Remove all endpoints that have been moved to
tf.randomnamespace. - Remove all endpoints from
tf.logging.
In total, we propose to remove 214 endpoints, including 171 endpoints in the root namespace.
See the list of endpoints we want to remove in Appendix 2: Deprecated
Endpoints. Note: the list in the appendix does not include endpoints under tf.logging since entire contents of this module will be deprecated. So, we don't need to list these endpoints individually.
Browsing for symbols should become easier. For e.g. page for tf.math namespace should display all math functions that TensorFlow provides. Similarly, tf.sets namespace page should display all available set operations.
Removing symbol endpoints would break references in user code. We plan to apply removals as a part of TensorFlow 2.0 release and provide a conversion script that would replace deprecated references with canonical ones. Initial script is at https://github.com/tensorflow/tensorflow/blob/master/tensorflow/tools/compatibility/tf_upgrade_v2.py. It will be updated to match changes in this doc.
Work will be done in multiple stages:
- We will first add new endpoints. I have some changes ready, so this step should take 1-2 weeks.
- Second step is to remove deprecated endpoints. This will be done later as a part of TensorFlow 2.0 release. Removing endpoints will take about 2 weeks.
In addition to symbols in this table, we plan to add all symbols under
tf.layers, tf.losses, tf.metrics to tf.keras.layers, tf.keras.losses,
tf.keras.metrics respectively.
| Current name | New names |
|---|---|
| tf.Assert | tf.debugging.Assert |
| tf.COMPILER_VERSION | tf.version.COMPILER_VERSION |
| tf.CXX11_ABI_FLAG | tf.sysconfig.CXX11_ABI_FLAG |
| tf.DType | tf.dtypes.DType |
| tf.FixedLenFeature | tf.io.FixedLenFeature |
| tf.FixedLenSequenceFeature | tf.io.FixedLenSequenceFeature |
| tf.GIT_VERSION | tf.version.GIT_VERSION |
| tf.GRAPH_DEF_VERSION | tf.version.GRAPH_DEF_VERSION |
| tf.GRAPH_DEF_VERSION_MIN_CONSUMER | tf.version.GRAPH_DEF_VERSION_MIN_CONSUMER |
| tf.GRAPH_DEF_VERSION_MIN_PRODUCER | tf.version.GRAPH_DEF_VERSION_MIN_PRODUCER |
| tf.HistogramProto | tf.summary.HistogramProto |
| tf.MONOLITHIC_BUILD | tf.sysconfig.MONOLITHIC_BUILD |
| tf.PaddingFIFOQueue | tf.io.PaddingFIFOQueue |
| tf.Print | tf.debugging.Print |
| tf.PriorityQueue | tf.io.PriorityQueue |
| tf.QUANTIZED_DTYPES | tf.dtypes.QUANTIZED_DTYPES |
| tf.QueueBase | tf.io.QueueBase |
| tf.RandomShuffleQueue | tf.io.RandomShuffleQueue |
| tf.SparseConditionalAccumulator | tf.sparse.SparseConditionalAccumulator |
| tf.SparseFeature | tf.io.SparseFeature |
| tf.SparseTensor | tf.sparse.SparseTensor |
| tf.SparseTensorValue | tf.sparse.SparseTensorValue |
| tf.SummaryMetadata | tf.summary.SummaryMetadata |
| tf.VERSION | tf.version.VERSION |
| tf.VarLenFeature | tf.io.VarLenFeature |
| tf.abs | tf.math.abs |
| tf.accumulate_n | tf.math.accumulate_n |
| tf.add_n | tf.math.add_n |
| tf.angle | tf.math.angle |
| tf.argmax | tf.math.argmax |
| tf.argmin | tf.math.argmin |
| tf.as_dtype | tf.dtypes.as_dtype |
| tf.assert_equal | tf.debugging.assert_equal |
| tf.assert_greater | tf.debugging.assert_greater |
| tf.assert_greater_equal | tf.debugging.assert_greater_equal |
| tf.assert_integer | tf.debugging.assert_integer |
| tf.assert_less | tf.debugging.assert_less |
| tf.assert_less_equal | tf.debugging.assert_less_equal |
| tf.assert_near | tf.debugging.assert_near |
| tf.assert_negative | tf.debugging.assert_negative |
| tf.assert_non_negative | tf.debugging.assert_non_negative |
| tf.assert_non_positive | tf.debugging.assert_non_positive |
| tf.assert_none_equal | tf.debugging.assert_none_equal |
| tf.assert_positive | tf.debugging.assert_positive |
| tf.assert_proper_iterable | tf.debugging.assert_proper_iterable |
| tf.assert_rank | tf.debugging.assert_rank |
| tf.assert_rank_at_least | tf.debugging.assert_rank_at_least |
| tf.assert_rank_in | tf.debugging.assert_rank_in |
| tf.assert_same_float_dtype | tf.debugging.assert_same_float_dtype |
| tf.assert_scalar | tf.debugging.assert_scalar |
| tf.assert_type | tf.debugging.assert_type |
| tf.bfloat16 | tf.dtypes.bfloat16 |
| tf.bitcast | tf.dtypes.bitcast |
| tf.bincount | tf.math.bincount |
| tf.bool | tf.dtypes.bool |
| tf.cast | tf.dtypes.cast |
| tf.complex | tf.dtypes.complex |
| tf.complex128 | tf.dtypes.complex128 |
| tf.complex64 | tf.dtypes.complex64 |
| tf.confusion_matrix | tf.train.confusion_matrix |
| tf.conj | tf.math.conj |
| tf.count_nonzero | tf.math.count_nonzero |
| tf.cumprod | tf.math.cumprod |
| tf.cumsum | tf.math.cumsum |
| tf.decode_csv | tf.io.decode_csv |
| tf.depth_to_space | tf.nn.depth_to_space |
| tf.deserialize_many_sparse | tf.io.deserialize_many_sparse |
| tf.divide | tf.math.divide |
| tf.double | tf.dtypes.double |
| tf.erf | tf.math.erf |
| tf.float16 | tf.dtypes.float16 |
| tf.float32 | tf.dtypes.float32 |
| tf.float64 | tf.dtypes.float64 |
| tf.floordiv | tf.math.floordiv |
| tf.floormod | tf.math.floormod |
| tf.get_seed | tf.random.get_seed |
| tf.global_norm | tf.linalg.global_norm |
| tf.half | tf.dtypes.half |
| tf.imag | tf.math.imag |
| tf.import_graph_def | tf.graph_util.import_graph_def |
| tf.int16 | tf.dtypes.int16 |
| tf.int32 | tf.dtypes.int32 |
| tf.int64 | tf.dtypes.int64 |
| tf.int8 | tf.dtypes.int8 |
| tf.is_non_decreasing | tf.debugging.is_non_decreasing |
| tf.is_numeric_tensor | tf.debugging.is_numeric_tensor |
| tf.is_strictly_increasing | tf.debugging.is_strictly_increasing |
| tf.lbeta | tf.math.lbeta |
| tf.log_sigmoid | tf.math.log_sigmoid |
| tf.logical_xor | tf.math.logical_xor |
| tf.manip.roll | tf.roll |
| tf.matmul | tf.linalg.matmul |
| tf.mod | tf.math.mod |
| tf.multinomial | tf.random.multinomial |
| tf.multiply | tf.math.multiply |
| tf.negative | tf.math.negative |
| tf.nn.in_top_k | tf.math.in_top_k |
| tf.nn.l2_normalize | tf.math.l2_normalize, tf.linalg.l2_normalize |
| tf.nn.log_softmax | tf.math.log_softmax |
| tf.nn.log_uniform_candidate_sampler | tf.random.log_uniform_candidate_sampler |
| tf.nn.sigmoid | tf.math.sigmoid |
| tf.nn.softmax | tf.math.softmax |
| tf.nn.top_k | tf.math.top_k |
| tf.nn.uniform_candidate_sampler | tf.random.uniform_candidate_sampler |
| tf.nn.zero_fraction | tf.math.zero_fraction |
| tf.parse_example | tf.io.parse_example |
| tf.parse_single_example | tf.io.parse_single_example |
| tf.parse_single_sequence_example | tf.io.parse_single_sequence_example |
| tf.pow | tf.math.pow |
| tf.python_io.TFRecordCompressionType | tf.io.TFRecordCompressionType |
| tf.python_io.TFRecordOptions | tf.io.TFRecordOptions |
| tf.python_io.TFRecordWriter | tf.io.TFRecordWriter |
| tf.python_io.tf_record_iterator | tf.io.tf_record_iterator |
| tf.qint16 | tf.dtypes.qint16 |
| tf.qint32 | tf.dtypes.qint32 |
| tf.qint8 | tf.dtypes.qint8 |
| tf.quantize | tf.quantization.quantize |
| tf.quantize_v2 | tf.quantization.quantize_v2 |
| tf.quint16 | tf.dtypes.quint16 |
| tf.quint8 | tf.dtypes.quint8 |
| tf.random_crop | tf.image.random_crop |
| tf.random_gamma | tf.random.gamma |
| tf.random_normal | tf.random.normal |
| tf.random_poisson | tf.random.poisson |
| tf.random_shuffle | tf.random.shuffle |
| tf.random_uniform | tf.random.uniform |
| tf.real | tf.math.real |
| tf.realdiv | tf.math.realdiv |
| tf.reduce_all | tf.math.reduce_all |
| tf.reduce_any | tf.math.reduce_any |
| tf.reduce_join | tf.math.reduce_join |
| tf.reduce_logsumexp | tf.math.reduce_logsumexp |
| tf.reduce_max | tf.math.reduce_max |
| tf.reduce_mean | tf.math.reduce_mean |
| tf.reduce_min | tf.math.reduce_min |
| tf.reduce_prod | tf.math.reduce_prod |
| tf.reduce_sum | tf.math.reduce_sum |
| tf.round | tf.math.round |
| tf.saturate_cast | tf.dtypes.saturate_cast |
| tf.saved_model.builder.SavedModelBuilder | tf.saved_model.SavedModelBuilder |
| tf.saved_model.constants.ASSETS_DIRECTORY | tf.saved_model.ASSETS_DIRECTORY |
| tf.saved_model.constants.ASSETS_KEY | tf.saved_model.ASSETS_KEY |
| tf.saved_model.constants.LEGACY_INIT_OP_KEY | tf.saved_model.LEGACY_INIT_OP_KEY |
| tf.saved_model.constants.MAIN_OP_KEY | tf.saved_model.MAIN_OP_KEY |
| tf.saved_model.constants.SAVED_MODEL_FILENAME_PB | tf.saved_model.SAVED_MODEL_FILENAME_PB |
| tf.saved_model.constants.SAVED_MODEL_FILENAME_PBTXT | tf.saved_model.SAVED_MODEL_FILENAME_PBTXT |
| tf.saved_model.constants.SAVED_MODEL_SCHEMA_VERSION | tf.saved_model.SAVED_MODEL_SCHEMA_VERSION |
| tf.saved_model.constants.VARIABLES_DIRECTORY | tf.saved_model.VARIABLES_DIRECTORY |
| tf.saved_model.constants.VARIABLES_FILENAME | tf.saved_model.VARIABLES_FILENAME |
| tf.saved_model.loader.load | tf.saved_model.load |
| tf.saved_model.loader.maybe_saved_model_directory | tf.saved_model.maybe_saved_model_directory |
| tf.saved_model.main_op.main_op_with_restore | tf.saved_model.main_op_with_restore |
| tf.saved_model.signature_constants.CLASSIFY_INPUTS | tf.saved_model.CLASSIFY_INPUTS |
| tf.saved_model.signature_constants.CLASSIFY_METHOD_NAME | tf.saved_model.CLASSIFY_METHOD_NAME |
| tf.saved_model.signature_constants.CLASSIFY_OUTPUT_CLASSES | tf.saved_model.CLASSIFY_OUTPUT_CLASSES |
| tf.saved_model.signature_constants.CLASSIFY_OUTPUT_SCORES | tf.saved_model.CLASSIFY_OUTPUT_SCORES |
| tf.saved_model.signature_constants.DEFAULT_SERVING_SIGNATURE_DEF_KEY | tf.saved_model.DEFAULT_SERVING_SIGNATURE_DEF_KEY |
| tf.saved_model.signature_constants.PREDICT_INPUTS | tf.saved_model.PREDICT_INPUTS |
| tf.saved_model.signature_constants.PREDICT_METHOD_NAME | tf.saved_model.PREDICT_METHOD_NAME |
| tf.saved_model.signature_constants.PREDICT_OUTPUTS | tf.saved_model.PREDICT_OUTPUTS |
| tf.saved_model.signature_constants.REGRESS_INPUTS | tf.saved_model.REGRESS_INPUTS |
| tf.saved_model.signature_constants.REGRESS_METHOD_NAME | tf.saved_model.REGRESS_METHOD_NAME |
| tf.saved_model.signature_constants.REGRESS_OUTPUTS | tf.saved_model.REGRESS_OUTPUTS |
| tf.saved_model.signature_def_utils.build_signature_def | tf.saved_model.build_signature_def |
| tf.saved_model.signature_def_utils.classification_signature_def | tf.saved_model.classification_signature_def |
| tf.saved_model.signature_def_utils.is_valid_signature | tf.saved_model.is_valid_signature |
| tf.saved_model.signature_def_utils.predict_signature_def | tf.saved_model.predict_signature_def |
| tf.saved_model.signature_def_utils.regression_signature_def | tf.saved_model.regression_signature_def |
| tf.saved_model.tag_constants.GPU | tf.saved_model.GPU |
| tf.saved_model.tag_constants.SERVING | tf.saved_model.SERVING |
| tf.saved_model.tag_constants.TPU | tf.saved_model.TPU |
| tf.saved_model.tag_constants.TRAINING | tf.saved_model.TRAINING |
| tf.saved_model.utils.build_tensor_info | tf.saved_model.build_tensor_info |
| tf.saved_model.utils.get_tensor_from_tensor_info | tf.saved_model.get_tensor_from_tensor_info |
| tf.scalar_mul | tf.math.scalar_mul |
| tf.serialize_many_sparse | tf.io.serialize_many_sparse |
| tf.serialize_sparse | tf.io.serialize_sparse |
| tf.set_random_seed | tf.random.set_random_seed |
| tf.sign | tf.math.sign |
| tf.space_to_batch | tf.nn.space_to_batch |
| tf.space_to_depth | tf.nn.space_to_depth |
| tf.sparse_add | tf.sparse.add |
| tf.sparse_concat | tf.sparse.concat |
| tf.sparse_fill_empty_rows | tf.sparse.fill_empty_rows |
| tf.sparse_mask | tf.sparse.mask |
| tf.sparse_maximum | tf.sparse.maximum |
| tf.sparse_merge | tf.sparse.merge |
| tf.sparse_minimum | tf.sparse.minimum |
| tf.sparse_placeholder | tf.sparse.placeholder |
| tf.sparse_reduce_max | tf.sparse.reduce_max |
| tf.sparse_reduce_max_sparse | tf.sparse.reduce_max_sparse |
| tf.sparse_reduce_sum | tf.sparse.reduce_sum |
| tf.sparse_reduce_sum_sparse | tf.sparse.reduce_sum_sparse |
| tf.sparse_reorder | tf.sparse.reorder |
| tf.sparse_reset_shape | tf.sparse.reset_shape |
| tf.sparse_reshape | tf.sparse.reshape |
| tf.sparse_retain | tf.sparse.retain |
| tf.sparse_segment_mean | tf.sparse.segment_mean |
| tf.sparse_segment_sqrt_n | tf.sparse.segment_sqrt_n |
| tf.sparse_segment_sum | tf.sparse.segment_sum |
| tf.sparse_slice | tf.sparse.slice |
| tf.sparse_softmax | tf.sparse.softmax |
| tf.sparse_split | tf.sparse.split |
| tf.sparse_tensor_dense_matmul | tf.sparse.matmul |
| tf.sparse_tensor_to_dense | tf.sparse.to_dense |
| tf.sparse_to_indicator | tf.sparse.to_indicator |
| tf.sparse_transpose | tf.sparse.transpose |
| tf.sqrt | tf.math.sqrt |
| tf.square | tf.math.square |
| tf.string | tf.dtypes.string |
| tf.string_split | tf.strings.split |
| tf.subtract | tf.math.subtract |
| tf.tables_initializer | tf.initializers.tables_initializer |
| tf.tanh | tf.math.tanh, tf.nn.tanh |
| tf.train.match_filenames_once | tf.io.match_filenames_once |
| tf.train.write_graph | tf.io.write_graph |
| tf.truediv | tf.math.truediv |
| tf.truncated_normal | tf.random.truncated_normal |
| tf.truncatediv | tf.math.truncatediv |
| tf.truncatemod | tf.math.truncatemod |
| tf.uint16 | tf.dtypes.uint16 |
| tf.uint32 | tf.dtypes.uint32 |
| tf.uint64 | tf.dtypes.uint64 |
| tf.uint8 | tf.dtypes.uint8 |
| tf.unsorted_segment_mean | tf.math.unsorted_segment_mean |
| tf.unsorted_segment_sqrt_n | tf.math.unsorted_segment_sqrt_n |
| tf.variant | tf.dtypes.variant |
| tf.verify_tensor_all_finite | tf.debugging.verify_tensor_all_finite |
In addition to symbols in this table, we plan to deprecate all symbols under
tf.logging (See Deprecated namespaces section above).
| Symbol that will be removed | Replacement |
|---|---|
| tf.COMPILER_VERSION | replace with tf.version.COMPILER_VERSION |
| tf.CXX11_ABI_FLAG | replace with tf.sysconfig.CXX11_ABI_FLAG |
| tf.Event | replace with tf.summary.Event |
| tf.GIT_VERSION | replace with tf.version.GIT_VERSION |
| tf.GRAPH_DEF_VERSION | replace with tf.version.GRAPH_DEF_VERSION |
| tf.GRAPH_DEF_VERSION_MIN_CONSUMER | replace with tf.version.GRAPH_DEF_VERSION_MIN_CONSUMER |
| tf.GRAPH_DEF_VERSION_MIN_PRODUCER | replace with tf.version.GRAPH_DEF_VERSION_MIN_PRODUCER |
| tf.HistogramProto | replace with tf.summary.HistogramProto |
| tf.MONOLITHIC_BUILD | replace with tf.sysconfig.MONOLITHIC_BUILD |
| tf.OpError | replace with tf.errors.OpError |
| tf.PaddingFIFOQueue | replace with tf.io.PaddingFIFOQueue |
| tf.PriorityQueue | replace with tf.io.PriorityQueue |
| tf.QueueBase | replace with tf.io.QueueBase |
| tf.RandomShuffleQueue | replace with tf.io.RandomShuffleQueue |
| tf.SparseConditionalAccumulator | replace with tf.sparse.SparseConditionalAccumulator |
| tf.SparseFeature | replace with tf.io.SparseFeature |
| tf.SummaryMetadata | replace with tf.summary.SummaryMetadata |
| tf.VERSION | replace with tf.version.VERSION |
| tf.accumulate_n | replace with tf.math.accumulate_n |
| tf.angle | replace with tf.math.angle |
| tf.assert_greater_equal | replace with tf.debugging.assert_greater_equal |
| tf.assert_integer | replace with tf.debugging.assert_integer |
| tf.assert_less_equal | replace with tf.debugging.assert_less_equal |
| tf.assert_near | replace with tf.debugging.assert_near |
| tf.assert_negative | replace with tf.debugging.assert_negative |
| tf.assert_non_negative | replace with tf.debugging.assert_non_negative |
| tf.assert_non_positive | replace with tf.debugging.assert_non_positive |
| tf.assert_none_equal | replace with tf.debugging.assert_none_equal |
| tf.assert_positive | replace with tf.debugging.assert_positive |
| tf.assert_proper_iterable | replace with tf.debugging.assert_proper_iterable |
| tf.assert_rank_at_least | replace with tf.debugging.assert_rank_at_least |
| tf.assert_rank_in | replace with tf.debugging.assert_rank_in |
| tf.assert_same_float_dtype | replace with tf.debugging.assert_same_float_dtype |
| tf.assert_scalar | replace with tf.debugging.assert_scalar |
| tf.assert_type | replace with tf.debugging.assert_type |
| tf.betainc | replace with tf.math.betainc |
| tf.bincount | replace with tf.math.bincount |
| tf.ceil | replace with tf.math.ceil |
| tf.cholesky | replace with tf.linalg.cholesky |
| tf.cholesky_solve | replace with tf.linalg.cholesky_solve |
| tf.confusion_matrix | replace with tf.train.confusion_matrix |
| tf.conj | replace with tf.math.conj |
| tf.cross | replace with tf.linalg.cross |
| tf.cumprod | replace with tf.math.cumprod |
| tf.decode_base64 | replace with tf.io.decode_base64 |
| tf.decode_compressed | replace with tf.io.decode_compressed |
| tf.decode_csv | replace with tf.io.decode_csv |
| tf.decode_json_example | replace with tf.io.decode_json_example |
| tf.depth_to_space | replace with tf.nn.depth_to_space |
| tf.deserialize_many_sparse | replace with tf.io.deserialize_many_sparse |
| tf.diag_part | replace with tf.linalg.tensor_diag_part |
| tf.digamma | replace with tf.math.digamma |
| tf.encode_base64 | replace with tf.io.encode_base64 |
| tf.erf | replace with tf.math.erf |
| tf.erfc | replace with tf.math.erfc |
| tf.expm1 | replace with tf.math.expm1 |
| tf.extract_image_patches | replace with tf.image.extract_image_patches |
| tf.fake_quant_with_min_max_args | replace with tf.quantization.fake_quant_with_min_max_args |
| tf.fake_quant_with_min_max_args_gradient | replace with tf.quantization.fake_quant_with_min_max_args_gradient |
| tf.fake_quant_with_min_max_vars | replace with tf.quantization.fake_quant_with_min_max_vars |
| tf.fake_quant_with_min_max_vars_gradient | replace with tf.quantization.fake_quant_with_min_max_vars_gradient |
| tf.fake_quant_with_min_max_vars_per_channel | replace with tf.quantization.fake_quant_with_min_max_vars_per_channel |
| tf.fake_quant_with_min_max_vars_per_channel_gradient | replace with tf.quantization.fake_quant_with_min_max_vars_per_channel_gradient |
| tf.fft | replace with tf.spectral.fft |
| tf.fft2d | replace with tf.spectral.fft2d |
| tf.fft3d | replace with tf.spectral.fft3d |
| tf.floordiv | replace with tf.math.floordiv |
| tf.floormod | replace with tf.math.floormod |
| tf.get_seed | replace with tf.random.get_seed |
| tf.global_norm | replace with tf.linalg.global_norm |
| tf.glorot_normal_initializer | replace with tf.initializers.glorot_normal |
| tf.ifft | replace with tf.spectral.ifft |
| tf.ifft2d | replace with tf.spectral.ifft2d |
| tf.ifft3d | replace with tf.spectral.ifft3d |
| tf.igamma | replace with tf.math.igamma |
| tf.igammac | replace with tf.math.igammac |
| tf.imag | replace with tf.math.imag |
| tf.invert_permutation | replace with tf.math.invert_permutation |
| tf.is_finite | replace with tf.debugging.is_finite |
| tf.is_inf | replace with tf.debugging.is_inf |
| tf.is_non_decreasing | replace with tf.debugging.is_non_decreasing |
| tf.is_numeric_tensor | replace with tf.debugging.is_numeric_tensor |
| tf.is_strictly_increasing | replace with tf.debugging.is_strictly_increasing |
| tf.lbeta | replace with tf.math.lbeta |
| tf.lgamma | replace with tf.math.lgamma |
| tf.log_sigmoid | replace with tf.math.log_sigmoid |
| tf.logical_xor | replace with tf.math.logical_xor |
| tf.manip.batch_to_space_nd | replace with tf.batch_to_space_nd |
| tf.manip.gather_nd | replace with tf.gather_nd |
| tf.manip.reshape | replace with tf.reshape |
| tf.manip.roll | replace with tf.roll |
| tf.manip.scatter_nd | replace with tf.scatter_nd |
| tf.manip.space_to_batch_nd | replace with tf.space_to_batch_nd |
| tf.manip.tile | replace with tf.tile |
| tf.matching_files | replace with tf.io.matching_files |
| tf.matrix_band_part | replace with tf.linalg.matrix_band_part |
| tf.matrix_determinant | replace with tf.linalg.det |
| tf.matrix_diag | replace with tf.linalg.diag |
| tf.matrix_diag_part | replace with tf.linalg.diag_part |
| tf.matrix_inverse | replace with tf.linalg.matrix_inverse |
| tf.matrix_set_diag | replace with tf.linalg.set_diag |
| tf.matrix_solve | replace with tf.linalg.solve |
| tf.matrix_solve_ls | replace with tf.linalg.matrix_solve_ls |
| tf.matrix_transpose | replace with tf.linalg.transpose |
| tf.matrix_triangular_solve | replace with tf.linalg.triangular_solve |
| tf.nn.log_uniform_candidate_sampler | replace with tf.random.log_uniform_candidate_sampler |
| tf.nn.uniform_candidate_sampler | replace with tf.random.uniform_candidate_sampler |
| tf.orthogonal_initializer | replace with tf.initializers.orthogonal_initializer |
| tf.parse_tensor | replace with tf.io.parse_tensor |
| tf.polygamma | replace with tf.math.polygamma |
| tf.python_io.TFRecordCompressionType | replace with tf.io.TFRecordCompressionType |
| tf.python_io.TFRecordOptions | replace with tf.io.TFRecordOptions |
| tf.qr | replace with tf.linalg.qr |
| tf.quantize_v2 | replace with tf.quantization.quantize_v2 |
| tf.quantized_concat | replace with tf.quantization.quantized_concat |
| tf.random_gamma | replace with tf.random.random_gamma |
| tf.random_poisson | replace with tf.random.random_poisson |
| tf.read_file | replace with tf.io.read_file |
| tf.real | replace with tf.math.real |
| tf.realdiv | replace with tf.math.realdiv |
| tf.reciprocal | replace with tf.math.reciprocal |
| tf.reduce_join | replace with tf.math.reduce_join |
| tf.regex_replace | replace with tf.strings.regex_replace |
| tf.rint | replace with tf.math.rint |
| tf.rsqrt | replace with tf.math.rsqrt |
| tf.saved_model.constants.ASSETS_DIRECTORY | replace with tf.saved_model.ASSETS_DIRECTORY |
| tf.saved_model.constants.ASSETS_KEY | replace with tf.saved_model.ASSETS_KEY |
| tf.saved_model.constants.LEGACY_INIT_OP_KEY | replace with tf.saved_model.LEGACY_INIT_OP_KEY |
| tf.saved_model.constants.MAIN_OP_KEY | replace with tf.saved_model.MAIN_OP_KEY |
| tf.saved_model.constants.SAVED_MODEL_FILENAME_PB | replace with tf.saved_model.SAVED_MODEL_FILENAME_PB |
| tf.saved_model.constants.SAVED_MODEL_FILENAME_PBTXT | replace with tf.saved_model.SAVED_MODEL_FILENAME_PBTXT |
| tf.saved_model.constants.SAVED_MODEL_SCHEMA_VERSION | replace with tf.saved_model.SAVED_MODEL_SCHEMA_VERSION |
| tf.saved_model.constants.VARIABLES_DIRECTORY | replace with tf.saved_model.VARIABLES_DIRECTORY |
| tf.saved_model.constants.VARIABLES_FILENAME | replace with tf.saved_model.VARIABLES_FILENAME |
| tf.saved_model.loader.maybe_saved_model_directory | replace with tf.saved_model.maybe_saved_model_directory |
| tf.saved_model.main_op.main_op | replace with tf.saved_model.main_op |
| tf.saved_model.main_op.main_op_with_restore | replace with tf.saved_model.main_op_with_restore |
| tf.saved_model.signature_constants.CLASSIFY_INPUTS | replace with tf.saved_model.CLASSIFY_INPUTS |
| tf.saved_model.signature_constants.CLASSIFY_METHOD_NAME | replace with tf.saved_model.CLASSIFY_METHOD_NAME |
| tf.saved_model.signature_constants.CLASSIFY_OUTPUT_CLASSES | replace with tf.saved_model.CLASSIFY_OUTPUT_CLASSES |
| tf.saved_model.signature_constants.CLASSIFY_OUTPUT_SCORES | replace with tf.saved_model.CLASSIFY_OUTPUT_SCORES |
| tf.saved_model.signature_constants.PREDICT_INPUTS | replace with tf.saved_model.PREDICT_INPUTS |
| tf.saved_model.signature_constants.PREDICT_METHOD_NAME | replace with tf.saved_model.PREDICT_METHOD_NAME |
| tf.saved_model.signature_constants.PREDICT_OUTPUTS | replace with tf.saved_model.PREDICT_OUTPUTS |
| tf.saved_model.signature_constants.REGRESS_INPUTS | replace with tf.saved_model.REGRESS_INPUTS |
| tf.saved_model.signature_constants.REGRESS_METHOD_NAME | replace with tf.saved_model.REGRESS_METHOD_NAME |
| tf.saved_model.signature_constants.REGRESS_OUTPUTS | replace with tf.saved_model.REGRESS_OUTPUTS |
| tf.saved_model.signature_def_utils.classification_signature_def | replace with tf.saved_model.classification_signature_def |
| tf.saved_model.signature_def_utils.is_valid_signature | replace with tf.saved_model.is_valid_signature |
| tf.saved_model.signature_def_utils.predict_signature_def | replace with tf.saved_model.predict_signature_def |
| tf.saved_model.signature_def_utils.regression_signature_def | replace with tf.saved_model.regression_signature_def |
| tf.saved_model.tag_constants.GPU | replace with tf.saved_model.GPU |
| tf.saved_model.tag_constants.TPU | replace with tf.saved_model.TPU |
| tf.saved_model.tag_constants.TRAINING | replace with tf.saved_model.TRAINING |
| tf.saved_model.utils.get_tensor_from_tensor_info | replace with tf.saved_model.get_tensor_from_tensor_info |
| tf.segment_max | replace with tf.math.segment_max |
| tf.segment_mean | replace with tf.math.segment_mean |
| tf.segment_min | replace with tf.math.segment_min |
| tf.segment_prod | replace with tf.math.segment_prod |
| tf.segment_sum | replace with tf.math.segment_sum |
| tf.self_adjoint_eig | replace with tf.linalg.self_adjoint_eig |
| tf.self_adjoint_eigvals | replace with tf.linalg.self_adjoint_eigvals |
| tf.serialize_many_sparse | replace with tf.io.serialize_many_sparse |
| tf.serialize_sparse | replace with tf.io.serialize_sparse |
| tf.space_to_batch | replace with tf.nn.space_to_batch |
| tf.space_to_depth | replace with tf.nn.space_to_depth |
| tf.sparse_add | replace with tf.sparse.add |
| tf.sparse_concat | replace with tf.sparse.concat |
| tf.sparse_fill_empty_rows | replace with tf.sparse.fill_empty_rows |
| tf.sparse_mask | replace with tf.sparse.mask |
| tf.sparse_maximum | replace with tf.sparse.maximum |
| tf.sparse_merge | replace with tf.sparse.merge |
| tf.sparse_minimum | replace with tf.sparse.minimum |
| tf.sparse_placeholder | replace with tf.sparse.placeholder |
| tf.sparse_reduce_max | replace with tf.sparse.reduce_max |
| tf.sparse_reduce_max_sparse | replace with tf.sparse.reduce_max_sparse |
| tf.sparse_reduce_sum | replace with tf.sparse.reduce_sum |
| tf.sparse_reduce_sum_sparse | replace with tf.sparse.reduce_sum_sparse |
| tf.sparse_reorder | replace with tf.sparse.reorder |
| tf.sparse_reset_shape | replace with tf.sparse.reset_shape |
| tf.sparse_reshape | replace with tf.sparse.reshape |
| tf.sparse_segment_mean | replace with tf.sparse.segment_mean |
| tf.sparse_segment_sqrt_n | replace with tf.sparse.segment_sqrt_n |
| tf.sparse_segment_sum | replace with tf.sparse.segment_sum |
| tf.sparse_slice | replace with tf.sparse.slice |
| tf.sparse_softmax | replace with tf.sparse.softmax |
| tf.sparse_split | replace with tf.sparse.split |
| tf.sparse_tensor_dense_matmul | replace with tf.sparse.matmul |
| tf.sparse_to_dense | replace with tf.sparse.to_dense which takes SparseTensor |
| object | |
| tf.sparse_to_indicator | replace with tf.sparse.to_indicator |
| tf.sparse_tensor_to_dense | replace with tf.sparse.to_dense |
| tf.sparse_transpose | replace with tf.sparse.transpose |
| tf.string_join | replace with tf.strings.join |
| tf.string_strip | replace with tf.strings.strip |
| tf.string_to_hash_bucket | replace with tf.strings.to_hash_bucket |
| tf.string_to_hash_bucket_fast | replace with tf.strings.to_hash_bucket_fast |
| tf.string_to_hash_bucket_strong | replace with tf.strings.to_hash_bucket_strong |
| tf.substr | replace with tf.strings.substr |
| tf.svd | replace with tf.linalg.svd |
| tf.trace | replace with tf.linalg.trace |
| tf.train.VocabInfo | replace with tf.estimator.VocabInfo |
| tf.train.match_filenames_once | replace with tf.io.match_filenames_once |
| tf.truncatediv | replace with tf.math.truncatediv |
| tf.truncatemod | replace with tf.math.truncatemod |
| tf.uniform_unit_scaling_initializer | replace with tf.initializers.uniform_unit_scaling |
| tf.unsorted_segment_max | replace with tf.math.unsorted_segment_max |
| tf.unsorted_segment_mean | replace with tf.math.unsorted_segment_mean |
| tf.unsorted_segment_min | replace with tf.math.unsorted_segment_min |
| tf.unsorted_segment_prod | replace with tf.math.unsorted_segment_prod |
| tf.unsorted_segment_sqrt_n | replace with tf.math.unsorted_segment_sqrt_n |
| tf.unsorted_segment_sum | replace with tf.math.unsorted_segment_sum |
| tf.variance_scaling_initializer | replace with tf.initializers.variance_scaling |
| tf.verify_tensor_all_finite | replace with tf.debugging.verify_tensor_all_finite |
| tf.write_file | replace with tf.io.write_file |
| tf.zeta | replace with tf.math.zeta |