LostTech.TensorFlow : API Documentation

Type Builder

Namespace tensorflow.saved_model

Parent _SavedModelBuilder

Interfaces IBuilder

Builds the `SavedModel` protocol buffer and saves variables and assets.

The `SavedModelBuilder` class provides functionality to build a `SavedModel` protocol buffer. Specifically, this allows multiple meta graphs to be saved as part of a single language-neutral `SavedModel`, while sharing variables and assets.

To build a SavedModel, the first meta graph must be saved with variables. Subsequent meta graphs will simply be saved with their graph definitions. If assets need to be saved and written or copied to disk, they can be provided when the meta graph def is added. If multiple meta graph defs are associated an asset of the same name, only the first version is retained.

Each meta graph added to the SavedModel must be annotated with tags. The tags provide a means to identify the specific meta graph to load and restore, along with the shared set of variables and assets.

Typical usage for the `SavedModelBuilder`: Note: This function will only be available through the v1 compatibility library as tf.compat.v1.saved_model.builder.SavedModelBuilder or tf.compat.v1.saved_model.Builder. Tensorflow 2.0 will introduce a new object-based method of creating SavedModels.
Show Example 
...
            builder = tf.compat.v1.saved_model.Builder(export_dir) 
 with tf.compat.v1.Session(graph=tf.Graph()) as sess:
 ...
  builder.add_meta_graph_and_variables(sess,
                                  ["foo-tag"],
                                  signature_def_map=foo_signatures,
                                  assets_collection=foo_assets)
... 
 with tf.compat.v1.Session(graph=tf.Graph()) as sess:
 ...
  builder.add_meta_graph(["bar-tag", "baz-tag"])
... 
 builder.save()

Methods

Properties

Public instance methods

void add_meta_graph(IEnumerable<string> tags, IDictionary<object, object> signature_def_map, object assets_collection, object legacy_init_op, bool clear_devices, IGraphNodeBase main_op, bool strip_default_attrs, object saver)

Adds the current meta graph to the SavedModel.

Creates a Saver in the current scope and uses the Saver to export the meta graph def. Invoking this API requires the `add_meta_graph_and_variables()` API to have been invoked before.
Parameters
IEnumerable<string> tags
The set of tags to annotate the meta graph def with.
IDictionary<object, object> signature_def_map
The map of signature defs to be added to the meta graph def.
object assets_collection
Assets to be saved with SavedModel. Note that this list should be a subset of the assets saved as part of the first meta graph in the SavedModel.
object legacy_init_op
bool clear_devices
Set to true if the device info on the default graph should be cleared.
IGraphNodeBase main_op
bool strip_default_attrs
object saver
An instance of tf.compat.v1.train.Saver that will be used to export the metagraph. If None, a sharded Saver that restores all variables will be used.

void add_meta_graph(IEnumerable<string> tags, IDictionary<object, object> signature_def_map, object assets_collection, object legacy_init_op, bool clear_devices, IGraphNodeBase main_op, Saver strip_default_attrs, object saver)

Adds the current meta graph to the SavedModel.

Creates a Saver in the current scope and uses the Saver to export the meta graph def. Invoking this API requires the `add_meta_graph_and_variables()` API to have been invoked before.
Parameters
IEnumerable<string> tags
The set of tags to annotate the meta graph def with.
IDictionary<object, object> signature_def_map
The map of signature defs to be added to the meta graph def.
object assets_collection
Assets to be saved with SavedModel. Note that this list should be a subset of the assets saved as part of the first meta graph in the SavedModel.
object legacy_init_op
bool clear_devices
Set to true if the device info on the default graph should be cleared.
IGraphNodeBase main_op
Saver strip_default_attrs
object saver
An instance of tf.compat.v1.train.Saver that will be used to export the metagraph. If None, a sharded Saver that restores all variables will be used.

void add_meta_graph(IEnumerable<string> tags, IDictionary<object, object> signature_def_map, object assets_list, bool clear_devices, object init_op, IGraphNodeBase train_op, bool saver)

void add_meta_graph(IEnumerable<string> tags, IDictionary<object, object> signature_def_map, object assets_list, bool clear_devices, object init_op, IGraphNodeBase train_op, Saver saver)

void add_meta_graph_and_variables(Session sess, IEnumerable<string> tags, IDictionary<object, object> signature_def_map, IEnumerable<object> assets_collection, Nullable<bool> legacy_init_op, Nullable<bool> clear_devices, Operation main_op, bool strip_default_attrs, Saver saver)

Adds the current meta graph to the SavedModel and saves variables.

Creates a Saver to save the variables from the provided session. Exports the corresponding meta graph def. This function assumes that the variables to be saved have been initialized. For a given `SavedModelBuilder`, this API must be called exactly once and for the first meta graph to save. For subsequent meta graph defs to be added, the `add_meta_graph()` API must be used.
Parameters
Session sess
The TensorFlow session from which to save the meta graph and variables.
IEnumerable<string> tags
The set of tags with which to save the meta graph.
IDictionary<object, object> signature_def_map
The map of signature def map to add to the meta graph def.
IEnumerable<object> assets_collection
Assets to be saved with SavedModel.
Nullable<bool> legacy_init_op
Nullable<bool> clear_devices
Set to true if the device info on the default graph should be cleared.
Operation main_op
bool strip_default_attrs
Boolean. If `True`, default-valued attributes will be removed from the NodeDefs. For a detailed guide, see [Stripping Default-Valued Attributes](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/saved_model/README.md#stripping-default-valued-attributes).
Saver saver
An instance of tf.compat.v1.train.Saver that will be used to export the metagraph and save variables. If None, a sharded Saver that restores all variables will be used.

void add_meta_graph_and_variables(Session sess, IEnumerable<string> tags, IDictionary<object, object> signature_def_map, IEnumerable<object> assets_collection, Nullable<bool> legacy_init_op, Nullable<bool> clear_devices, IGraphNodeBase main_op, bool strip_default_attrs, Saver saver)

Adds the current meta graph to the SavedModel and saves variables.

Creates a Saver to save the variables from the provided session. Exports the corresponding meta graph def. This function assumes that the variables to be saved have been initialized. For a given `SavedModelBuilder`, this API must be called exactly once and for the first meta graph to save. For subsequent meta graph defs to be added, the `add_meta_graph()` API must be used.
Parameters
Session sess
The TensorFlow session from which to save the meta graph and variables.
IEnumerable<string> tags
The set of tags with which to save the meta graph.
IDictionary<object, object> signature_def_map
The map of signature def map to add to the meta graph def.
IEnumerable<object> assets_collection
Assets to be saved with SavedModel.
Nullable<bool> legacy_init_op
Nullable<bool> clear_devices
Set to true if the device info on the default graph should be cleared.
IGraphNodeBase main_op
bool strip_default_attrs
Boolean. If `True`, default-valued attributes will be removed from the NodeDefs. For a detailed guide, see [Stripping Default-Valued Attributes](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/saved_model/README.md#stripping-default-valued-attributes).
Saver saver
An instance of tf.compat.v1.train.Saver that will be used to export the metagraph and save variables. If None, a sharded Saver that restores all variables will be used.

void add_meta_graph_and_variables(Session sess, IEnumerable<string> tags, IDictionary<object, object> signature_def_map, IEnumerable<object> assets_collection, Nullable<bool> legacy_init_op, Nullable<bool> clear_devices, object main_op, bool strip_default_attrs, Saver saver)

Adds the current meta graph to the SavedModel and saves variables.

Creates a Saver to save the variables from the provided session. Exports the corresponding meta graph def. This function assumes that the variables to be saved have been initialized. For a given `SavedModelBuilder`, this API must be called exactly once and for the first meta graph to save. For subsequent meta graph defs to be added, the `add_meta_graph()` API must be used.
Parameters
Session sess
The TensorFlow session from which to save the meta graph and variables.
IEnumerable<string> tags
The set of tags with which to save the meta graph.
IDictionary<object, object> signature_def_map
The map of signature def map to add to the meta graph def.
IEnumerable<object> assets_collection
Assets to be saved with SavedModel.
Nullable<bool> legacy_init_op
Nullable<bool> clear_devices
Set to true if the device info on the default graph should be cleared.
object main_op
bool strip_default_attrs
Boolean. If `True`, default-valued attributes will be removed from the NodeDefs. For a detailed guide, see [Stripping Default-Valued Attributes](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/saved_model/README.md#stripping-default-valued-attributes).
Saver saver
An instance of tf.compat.v1.train.Saver that will be used to export the metagraph and save variables. If None, a sharded Saver that restores all variables will be used.

object add_meta_graph_and_variables_dyn(object sess, object tags, object signature_def_map, object assets_collection, object legacy_init_op, ImplicitContainer<T> clear_devices, object main_op, ImplicitContainer<T> strip_default_attrs, object saver)

Adds the current meta graph to the SavedModel and saves variables.

Creates a Saver to save the variables from the provided session. Exports the corresponding meta graph def. This function assumes that the variables to be saved have been initialized. For a given `SavedModelBuilder`, this API must be called exactly once and for the first meta graph to save. For subsequent meta graph defs to be added, the `add_meta_graph()` API must be used.
Parameters
object sess
The TensorFlow session from which to save the meta graph and variables.
object tags
The set of tags with which to save the meta graph.
object signature_def_map
The map of signature def map to add to the meta graph def.
object assets_collection
Assets to be saved with SavedModel.
object legacy_init_op
ImplicitContainer<T> clear_devices
Set to true if the device info on the default graph should be cleared.
object main_op
ImplicitContainer<T> strip_default_attrs
Boolean. If `True`, default-valued attributes will be removed from the NodeDefs. For a detailed guide, see [Stripping Default-Valued Attributes](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/saved_model/README.md#stripping-default-valued-attributes).
object saver
An instance of tf.compat.v1.train.Saver that will be used to export the metagraph and save variables. If None, a sharded Saver that restores all variables will be used.

object add_meta_graph_and_variables_dyn(object sess, object tags, object signature_def_map, object assets_list, ImplicitContainer<T> clear_devices, object init_op, object train_op, ImplicitContainer<T> strip_default_attrs, object saver)

object add_meta_graph_dyn(object tags, object signature_def_map, object assets_collection, object legacy_init_op, ImplicitContainer<T> clear_devices, object main_op, ImplicitContainer<T> strip_default_attrs, object saver)

Adds the current meta graph to the SavedModel.

Creates a Saver in the current scope and uses the Saver to export the meta graph def. Invoking this API requires the `add_meta_graph_and_variables()` API to have been invoked before.
Parameters
object tags
The set of tags to annotate the meta graph def with.
object signature_def_map
The map of signature defs to be added to the meta graph def.
object assets_collection
Assets to be saved with SavedModel. Note that this list should be a subset of the assets saved as part of the first meta graph in the SavedModel.
object legacy_init_op
ImplicitContainer<T> clear_devices
Set to true if the device info on the default graph should be cleared.
object main_op
ImplicitContainer<T> strip_default_attrs
object saver
An instance of tf.compat.v1.train.Saver that will be used to export the metagraph. If None, a sharded Saver that restores all variables will be used.

object add_meta_graph_dyn(object tags, object signature_def_map, object assets_list, ImplicitContainer<T> clear_devices, object init_op, object train_op, object saver)

object save(bool as_text)

object save_dyn(ImplicitContainer<T> as_text)

Public properties

object PythonObject get;