tf.contrib.framework.assign_from_checkpoint(model_path, var_list)

tf.contrib.framework.assign_from_checkpoint(model_path, var_list)

See the guide: Framework (contrib) > Variables

Creates an operation to assign specific variables from a checkpoint.


  • model_path: The full path to the model checkpoint. To get latest checkpoint use model_path = tf.train.latest_checkpoint(checkpoint_dir)
  • var_list: A list of Variable objects or a dictionary mapping names in the checkpoint to the corresponding variables to initialize. If empty or None, it would return no_op(), None.


the restore_op and the feed_dict that need to be run to restore var_list.


  • ValueError: If the checkpoint specified at model_path is missing one of the variables in var_list.

Defined in tensorflow/contrib/framework/python/ops/