reference/generated/torchrl.trainers.algorithms.configs.checkpoint.CheckpointConfig

Run in Google Colab

Colab

Download Notebook

Notebook

View on GitHub

GitHub

torchrl.trainers.algorithms.configs.checkpoint.CheckpointConfig

class torchrl.trainers.algorithms.configs.checkpoint.CheckpointConfig(format: str = 'directory', strict: str = 'error', archive_compression: str = 'stored', save_components: list[str] | None = None, _target_: str = 'torchrl.checkpoint.Checkpoint')

Hydra configuration for Checkpoint.

Every kwarg accepted by Checkpoint.__init__ is exposed as a field here. Components are registered by the trainer that receives the checkpoint.

See also

Checkpoint