tf.compat.v1.train.update_checkpoint_state | TensorFlow v2.16.1 (original) (raw)
tf.compat.v1.train.update_checkpoint_state
Stay organized with collections Save and categorize content based on your preferences.
Updates the content of the 'checkpoint' file. (deprecated)
tf.compat.v1.train.update_checkpoint_state(
save_dir,
model_checkpoint_path,
all_model_checkpoint_paths=None,
latest_filename=None,
all_model_checkpoint_timestamps=None,
last_preserved_timestamp=None
)
This updates the checkpoint file containing a CheckpointState proto.
| Args | |
|---|---|
| save_dir | Directory where the model was saved. |
| model_checkpoint_path | The checkpoint file. |
| all_model_checkpoint_paths | List of strings. Paths to all not-yet-deleted checkpoints, sorted from oldest to newest. If this is a non-empty list, the last element must be equal to model_checkpoint_path. These paths are also saved in the CheckpointState proto. |
| latest_filename | Optional name of the checkpoint file. Default to 'checkpoint'. |
| all_model_checkpoint_timestamps | Optional list of timestamps (floats, seconds since the Epoch) indicating when the checkpoints inall_model_checkpoint_paths were created. |
| last_preserved_timestamp | A float, indicating the number of seconds since the Epoch when the last preserved checkpoint was written, e.g. due to akeep_checkpoint_every_n_hours parameter (seetf.train.CheckpointManager for an implementation). |
| Raises | |
|---|---|
| RuntimeError | If any of the model checkpoint paths conflict with the file containing CheckpointSate. |
Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates. Some content is licensed under the numpy license.
Last updated 2024-04-26 UTC.