def export(checkpoint, destination, use_tf_sampling):
model = None
if use_tf_sampling:
model = lib_tfsampling.CoconetSampleGraph(checkpoint)
model.instantiate_sess_and_restore_checkpoint()
else:
model = lib_graph.load_checkpoint(checkpoint)
tf.logging.info('Loaded graph.')
lib_saved_model.export_saved_model(model, destination,
[tf.saved_model.tag_constants.SERVING],
use_tf_sampling)
def __init__(self, checkpoint_path):
"""Initializes Generator with a wrapped model and strategy name.
Args:
checkpoint_path: A string that gives the full path to the folder that
holds the checkpoint.
"""
self.sampler = lib_tfsampling.CoconetSampleGraph(checkpoint_path)
self.hparams = self.sampler.hparams
self.endecoder = lib_pianoroll.get_pianoroll_encoder_decoder(self.hparams)
self._time_taken = None
self._pianorolls = None
def main(unused_argv):
if FLAGS.checkpoint is None or not FLAGS.checkpoint:
raise ValueError('Need to provide a path to checkpoint directory.')
if FLAGS.destination is None or not FLAGS.destination:
raise ValueError(
'Need to provide a destination directory for the SavedModel.')
model = None
if FLAGS.use_tf_sampling:
model = lib_tfsampling.CoconetSampleGraph(FLAGS.checkpoint)
model.instantiate_sess_and_restore_checkpoint()
else:
model = lib_graph.load_checkpoint(FLAGS.checkpoint)
tf.logging.info('Loaded graph.')
lib_saved_model.export_saved_model(model, FLAGS.destination,
[tf.saved_model.tag_constants.SERVING],
FLAGS.use_tf_sampling)
tf.logging.info('Exported SavedModel to %s.', FLAGS.destination)
