Function import_scoped_meta_graph

tensorflow/python/framework/meta_graph.py:658–708

Recreates a `Graph` saved in a `MetaGraphDef` proto. This function takes a `MetaGraphDef` protocol buffer as input. If the argument is a file containing a `MetaGraphDef` protocol buffer , it constructs a protocol buffer from the file content. The function then adds all the nodes from the `g

(meta_graph_or_file,
                             clear_devices=False,
                             graph=None,
                             import_scope=None,
                             input_map=None,
                             unbound_inputs_col_name="unbound_inputs",
                             restore_collections_predicate=(lambda key: True))

def import_scoped_meta_graph(meta_graph_or_file,
                             clear_devices=False,
                             graph=None,
                             import_scope=None,
                             input_map=None,
                             unbound_inputs_col_name="unbound_inputs",
                             restore_collections_predicate=(lambda key: True)):
  """Recreates a `Graph` saved in a `MetaGraphDef` proto.

  This function takes a `MetaGraphDef` protocol buffer as input. If
  the argument is a file containing a `MetaGraphDef` protocol buffer ,
  it constructs a protocol buffer from the file content. The function
  then adds all the nodes from the `graph_def` field to the
  current graph, recreates the desired collections, and returns a dictionary of
  all the Variables imported into the name scope.

  In combination with `export_scoped_meta_graph()`, this function can be used to

  * Serialize a graph along with other Python objects such as `QueueRunner`,
    `Variable` into a `MetaGraphDef`.

  * Restart training from a saved graph and checkpoints.

  * Run inference from a saved graph and checkpoints.

  Args:
    meta_graph_or_file: `MetaGraphDef` protocol buffer or filename (including
      the path) containing a `MetaGraphDef`.
    clear_devices: Boolean which controls whether to clear device information
      from graph_def. Default false.
    graph: The `Graph` to import into. If `None`, use the default graph.
    import_scope: Optional `string`. Name scope into which to import the
      subgraph. If `None`, the graph is imported to the root name scope.
    input_map: A dictionary mapping input names (as strings) in `graph_def` to
      `Tensor` objects. The values of the named input tensors in the imported
      graph will be re-mapped to the respective `Tensor` values.
    unbound_inputs_col_name: Collection name for looking up unbound inputs.
    restore_collections_predicate: a predicate on collection names. A collection
      named c (i.e whose key is c) will be restored iff
      1) `restore_collections_predicate(c)` is True, and
      2) `c != unbound_inputs_col_name`.

  Returns:
    A dictionary of all the `Variables` imported into the name scope.

  Raises:
    ValueError: If the graph_def contains unbound inputs.
  """
  return import_scoped_meta_graph_with_return_elements(
      meta_graph_or_file, clear_devices, graph, import_scope, input_map,
      unbound_inputs_col_name, restore_collections_predicate)[0]


def import_scoped_meta_graph_with_return_elements(