tf.train.get_checkpoint_state( checkpoint_dir, latest_filename=None )
See the guide: Variables > Saving and Restoring Variables
Returns CheckpointState proto from the "checkpoint" file.
If the "checkpoint" file contains a valid CheckpointState proto, returns it.
checkpoint_dir: The directory of checkpoints.
latest_filename: Optional name of the checkpoint file. Default to 'checkpoint'.
A CheckpointState if the state was available, None otherwise.
ValueError: if the checkpoint read doesn't have model_checkpoint_path set.