|View source on GitHub|
Creates multiple dependencies with a synchronized save/restore.
tf.contrib.checkpoint.split_dependency( component_names, component_dtypes, fill_save_buffer_fn, consume_restore_buffer_fn, device )
Useful when a single op produces
Tensors which should each be saved under
different objects, or when
Tensors saved with many different objects need to
be restored together as inputs to a single op (i.e. an object which uses a
single fused op may be swapped out for a subgraph of objects, and these two
programs are checkpoint compatible).
component_names: A sequence of names for the split dependencies.
fill_save_buffer_fnmust add these keys to the dictionary it is passed, and
consume_restore_buffer_fnwill receive a dictionary with these keys.
component_dtypes: Data types for the
Tensors being saved and restored, a sequence corresponding to
fill_save_buffer_fn: A function which takes an empty dictionary as an argument and adds
component_namesas keys. These
Tensors will be saved as if they were individual variables.
consume_restore_buffer_fn: A function which takes a dictionary with
component_namesas keys mapping to restored individual
Tensors and returns a restore op (or if executing eagerly, runs the restoration and may return
device: The device on which to run save and restore operations.
A dictionary mapping from names to Trackable objects. If one is reachable from an object as a dependency, the others should be too; adding dependencies on some but not all of the objects will result in errors.