Python tensorflow.python.training.saver.get_checkpoint_state() Examples
The following are 5
code examples of tensorflow.python.training.saver.get_checkpoint_state().
You can vote up the ones you like or vote down the ones you don't like,
and go to the original project or source file by following the links above each example.
You may also want to check out all available functions/classes of the module
tensorflow.python.training.saver
, or try the search function
.
Example #1
Source File: session_manager.py From lambda-packs with MIT License | 4 votes |
def _restore_checkpoint(self, master, saver=None, checkpoint_dir=None, checkpoint_filename_with_path=None, wait_for_checkpoint=False, max_wait_secs=7200, config=None): """Creates a `Session`, and tries to restore a checkpoint. Args: master: `String` representation of the TensorFlow master to use. saver: A `Saver` object used to restore a model. checkpoint_dir: Path to the checkpoint files. The latest checkpoint in the dir will be used to restore. checkpoint_filename_with_path: Full file name path to the checkpoint file. wait_for_checkpoint: Whether to wait for checkpoint to become available. max_wait_secs: Maximum time to wait for checkpoints to become available. config: Optional `ConfigProto` proto used to configure the session. Returns: A pair (sess, is_restored) where 'is_restored' is `True` if the session could be restored, `False` otherwise. Raises: ValueError: If both checkpoint_dir and checkpoint_filename_with_path are set. """ self._target = master sess = session.Session(self._target, graph=self._graph, config=config) if checkpoint_dir and checkpoint_filename_with_path: raise ValueError("Can not provide both checkpoint_dir and " "checkpoint_filename_with_path.") # If either saver or checkpoint_* is not specified, cannot restore. Just # return. if not saver or not (checkpoint_dir or checkpoint_filename_with_path): return sess, False if checkpoint_filename_with_path: saver.restore(sess, checkpoint_filename_with_path) return sess, True # Waits up until max_wait_secs for checkpoint to become available. wait_time = 0 ckpt = saver_mod.get_checkpoint_state(checkpoint_dir) while not ckpt or not ckpt.model_checkpoint_path: if wait_for_checkpoint and wait_time < max_wait_secs: logging.info("Waiting for checkpoint to be available.") time.sleep(self._recovery_wait_secs) wait_time += self._recovery_wait_secs ckpt = saver_mod.get_checkpoint_state(checkpoint_dir) else: return sess, False # Loads the checkpoint. saver.restore(sess, ckpt.model_checkpoint_path) saver.recover_last_checkpoints(ckpt.all_model_checkpoint_paths) return sess, True
Example #2
Source File: session_manager.py From auto-alt-text-lambda-api with MIT License | 4 votes |
def _restore_checkpoint(self, master, saver=None, checkpoint_dir=None, checkpoint_filename_with_path=None, wait_for_checkpoint=False, max_wait_secs=7200, config=None): """Creates a `Session`, and tries to restore a checkpoint. Args: master: `String` representation of the TensorFlow master to use. saver: A `Saver` object used to restore a model. checkpoint_dir: Path to the checkpoint files. The latest checkpoint in the dir will be used to restore. checkpoint_filename_with_path: Full file name path to the checkpoint file. wait_for_checkpoint: Whether to wait for checkpoint to become available. max_wait_secs: Maximum time to wait for checkpoints to become available. config: Optional `ConfigProto` proto used to configure the session. Returns: A pair (sess, is_restored) where 'is_restored' is `True` if the session could be restored, `False` otherwise. Raises: ValueError: If both checkpoint_dir and checkpoint_filename_with_path are set. """ self._target = master sess = session.Session(self._target, graph=self._graph, config=config) if checkpoint_dir and checkpoint_filename_with_path: raise ValueError("Can not provide both checkpoint_dir and " "checkpoint_filename_with_path.") # If either saver or checkpoint_* is not specified, cannot restore. Just # return. if not saver or not (checkpoint_dir or checkpoint_filename_with_path): return sess, False if checkpoint_filename_with_path: saver.restore(sess, checkpoint_filename_with_path) return sess, True # Waits up until max_wait_secs for checkpoint to become available. wait_time = 0 ckpt = saver_mod.get_checkpoint_state(checkpoint_dir) while not ckpt or not ckpt.model_checkpoint_path: if wait_for_checkpoint and wait_time < max_wait_secs: logging.info("Waiting for checkpoint to be available.") time.sleep(self._recovery_wait_secs) wait_time += self._recovery_wait_secs ckpt = saver_mod.get_checkpoint_state(checkpoint_dir) else: return sess, False # Loads the checkpoint. saver.restore(sess, ckpt.model_checkpoint_path) saver.recover_last_checkpoints(ckpt.all_model_checkpoint_paths) return sess, True
Example #3
Source File: session_manager.py From deep_image_model with Apache License 2.0 | 4 votes |
def _restore_checkpoint(self, master, saver=None, checkpoint_dir=None, wait_for_checkpoint=False, max_wait_secs=7200, config=None): """Creates a `Session`, and tries to restore a checkpoint. Args: master: `String` representation of the TensorFlow master to use. saver: A `Saver` object used to restore a model. checkpoint_dir: Path to the checkpoint files. wait_for_checkpoint: Whether to wait for checkpoint to become available. max_wait_secs: Maximum time to wait for checkpoints to become available. config: Optional `ConfigProto` proto used to configure the session. Returns: A pair (sess, is_restored) where 'is_restored' is `True` if the session could be restored, `False` otherwise. """ self._target = master sess = session.Session(self._target, graph=self._graph, config=config) # If either saver or checkpoint_dir is not specified, cannot restore. Just # return. if not saver or not checkpoint_dir: return sess, False # Waits up until max_wait_secs for checkpoint to become available. wait_time = 0 ckpt = saver_mod.get_checkpoint_state(checkpoint_dir) while not ckpt or not ckpt.model_checkpoint_path: if wait_for_checkpoint and wait_time < max_wait_secs: logging.info("Waiting for checkpoint to be available.") time.sleep(self._recovery_wait_secs) wait_time += self._recovery_wait_secs ckpt = saver_mod.get_checkpoint_state(checkpoint_dir) else: return sess, False # Loads the checkpoint. saver.restore(sess, ckpt.model_checkpoint_path) saver.recover_last_checkpoints(ckpt.all_model_checkpoint_paths) return sess, True
Example #4
Source File: session_manager.py From Serverless-Deep-Learning-with-TensorFlow-and-AWS-Lambda with MIT License | 4 votes |
def _restore_checkpoint(self, master, saver=None, checkpoint_dir=None, checkpoint_filename_with_path=None, wait_for_checkpoint=False, max_wait_secs=7200, config=None): """Creates a `Session`, and tries to restore a checkpoint. Args: master: `String` representation of the TensorFlow master to use. saver: A `Saver` object used to restore a model. checkpoint_dir: Path to the checkpoint files. The latest checkpoint in the dir will be used to restore. checkpoint_filename_with_path: Full file name path to the checkpoint file. wait_for_checkpoint: Whether to wait for checkpoint to become available. max_wait_secs: Maximum time to wait for checkpoints to become available. config: Optional `ConfigProto` proto used to configure the session. Returns: A pair (sess, is_restored) where 'is_restored' is `True` if the session could be restored, `False` otherwise. Raises: ValueError: If both checkpoint_dir and checkpoint_filename_with_path are set. """ self._target = master sess = session.Session(self._target, graph=self._graph, config=config) if checkpoint_dir and checkpoint_filename_with_path: raise ValueError("Can not provide both checkpoint_dir and " "checkpoint_filename_with_path.") # If either saver or checkpoint_* is not specified, cannot restore. Just # return. if not saver or not (checkpoint_dir or checkpoint_filename_with_path): return sess, False if checkpoint_filename_with_path: saver.restore(sess, checkpoint_filename_with_path) return sess, True # Waits up until max_wait_secs for checkpoint to become available. wait_time = 0 ckpt = saver_mod.get_checkpoint_state(checkpoint_dir) while not ckpt or not ckpt.model_checkpoint_path: if wait_for_checkpoint and wait_time < max_wait_secs: logging.info("Waiting for checkpoint to be available.") time.sleep(self._recovery_wait_secs) wait_time += self._recovery_wait_secs ckpt = saver_mod.get_checkpoint_state(checkpoint_dir) else: return sess, False # Loads the checkpoint. saver.restore(sess, ckpt.model_checkpoint_path) saver.recover_last_checkpoints(ckpt.all_model_checkpoint_paths) return sess, True
Example #5
Source File: session_manager.py From keras-lambda with MIT License | 4 votes |
def _restore_checkpoint(self, master, saver=None, checkpoint_dir=None, checkpoint_filename_with_path=None, wait_for_checkpoint=False, max_wait_secs=7200, config=None): """Creates a `Session`, and tries to restore a checkpoint. Args: master: `String` representation of the TensorFlow master to use. saver: A `Saver` object used to restore a model. checkpoint_dir: Path to the checkpoint files. The latest checkpoint in the dir will be used to restore. checkpoint_filename_with_path: Full file name path to the checkpoint file. wait_for_checkpoint: Whether to wait for checkpoint to become available. max_wait_secs: Maximum time to wait for checkpoints to become available. config: Optional `ConfigProto` proto used to configure the session. Returns: A pair (sess, is_restored) where 'is_restored' is `True` if the session could be restored, `False` otherwise. Raises: ValueError: If both checkpoint_dir and checkpoint_filename_with_path are set. """ self._target = master sess = session.Session(self._target, graph=self._graph, config=config) if checkpoint_dir and checkpoint_filename_with_path: raise ValueError("Can not provide both checkpoint_dir and " "checkpoint_filename_with_path.") # If either saver or checkpoint_* is not specified, cannot restore. Just # return. if not saver or not (checkpoint_dir or checkpoint_filename_with_path): return sess, False if checkpoint_filename_with_path: saver.restore(sess, checkpoint_filename_with_path) return sess, True # Waits up until max_wait_secs for checkpoint to become available. wait_time = 0 ckpt = saver_mod.get_checkpoint_state(checkpoint_dir) while not ckpt or not ckpt.model_checkpoint_path: if wait_for_checkpoint and wait_time < max_wait_secs: logging.info("Waiting for checkpoint to be available.") time.sleep(self._recovery_wait_secs) wait_time += self._recovery_wait_secs ckpt = saver_mod.get_checkpoint_state(checkpoint_dir) else: return sess, False # Loads the checkpoint. saver.restore(sess, ckpt.model_checkpoint_path) saver.recover_last_checkpoints(ckpt.all_model_checkpoint_paths) return sess, True