CheckpointIO — PyTorch Lightning 2.5.1.post0 documentation (original) (raw)

class lightning.pytorch.plugins.io.CheckpointIO[source]¶

Bases: ABC

Interface to save/load checkpoints as they are saved through the Strategy.

Typically most plugins either use the Torch based IO Plugin; TorchCheckpointIO but may require particular handling depending on the plugin.

In addition, you can pass a custom CheckpointIO by extending this class and passing it to the Trainer, i.e Trainer(plugins=[MyCustomCheckpointIO()]).

Note

For some plugins, it is not possible to use a custom checkpoint plugin as checkpointing logic is not modifiable.

abstract load_checkpoint(path, map_location=None)[source]¶

Load checkpoint from a path when resuming or loading ckpt for test/validate/predict stages.

Parameters:

• path¶ (Union[str, Path]) – Path to checkpoint
• map_location¶ (Optional[Any]) – a function, torch.device, string or a dict specifying how to remap storage locations.

Return type:

dict[str, Any]

Returns: The loaded checkpoint.

abstract remove_checkpoint(path)[source]¶

Remove checkpoint file from the filesystem.

Parameters:

path¶ (Union[str, Path]) – Path to checkpoint

Return type:

None

abstract save_checkpoint(checkpoint, path, storage_options=None)[source]¶

Save model/training states as a checkpoint file through state-dump and file-write.

Parameters:

• checkpoint¶ (dict[str, Any]) – dict containing model and trainer state
• path¶ (Union[str, Path]) – write-target path
• storage_options¶ (Optional[Any]) – Optional parameters when saving the model/training states.

Return type:

None

teardown()[source]¶

This method is called to teardown the process.

Return type:

None