|View source on GitHub|
tf.train.export_meta_graph( filename=None, meta_info_def=None, graph_def=None, saver_def=None, collection_list=None, as_text=False, graph=None, export_scope=None, clear_devices=False, clear_extraneous_savers=False, strip_default_attrs=False, save_debug_info=False, **kwargs )
Optionally writes it to filename.
This function exports the graph, saver, and collection objects into
MetaGraphDef protocol buffer with the intention of it being imported
at a later time or location to restart training, run inference, or be
filename: Optional filename including the path for writing the generated
collection_list: List of string keys to collect.
True, writes the
MetaGraphDefas an ASCII proto.
Graphto export. If
None, use the default graph.
string. Name scope under which to extract the subgraph. The scope name will be striped from the node definitions for easy import later into new name scopes. If
None, the whole graph is exported. graph_def and export_scope cannot both be specified.
clear_devices: Whether or not to clear the device field for an
clear_extraneous_savers: Remove any Saver-related information from the graph (both Save/Restore ops and SaverDefs) that are not associated with the provided SaverDef.
strip_default_attrs: Boolean. If
True, default-valued attributes will be removed from the NodeDefs. For a detailed guide, see Stripping Default-Valued Attributes.
True, save the GraphDebugInfo to a separate file, which in the same directory of filename and with
_debugadded before the file extend.
**kwargs: Optional keyed arguments.
ValueError: When the
GraphDefis larger than 2GB.
RuntimeError: If called with eager execution enabled.
Exporting/importing meta graphs is not supported unless both
graph are provided. No graph exists when eager execution is enabled.