Apply to speak at TensorFlow World. Deadline April 23rd. Propose talk

tf.contrib.model_pruning.graph_def_from_checkpoint

tf.contrib.model_pruning.graph_def_from_checkpoint(
    checkpoint_dir,
    output_node_names
)

Defined in tensorflow/contrib/model_pruning/python/strip_pruning_vars_lib.py.

Converts checkpoint data to GraphDef.

Reads the latest checkpoint data and produces a GraphDef in which the variables have been converted to constants.

Args:

  • checkpoint_dir: Path to the checkpoints.
  • output_node_names: List of name strings for the result nodes of the graph.

Returns:

A GraphDef from the latest checkpoint

Raises:

  • ValueError: if no checkpoint is found