Sb3CheckpointArgs
Full path:
schola.scripts.sb3.settings.Sb3CheckpointArgs
schola.scripts.sb3.settings.Sb3CheckpointArgs
Sb3CheckpointArgs
Sb3CheckpointArgs(enable_checkpoints: bool = False, checkpoint_dir: Annotated[pathlib._local.Path, Parameter(validator=Path(exists=False, file_okay=False, dir_okay=True, ext=()))] = WindowsPath('ckpt'), save_freq: Annotated[int, Parameter(validator=Number(lt=None, lte=None, gt=None, gte=0, modulo=None))] = 100000, name_prefix_override: str | None = None, export_onnx: bool = False, save_final_policy: bool = False, save_replay_buffer: 'bool' = False, save_vecnormalize: 'bool' = False)Bases: CheckpointArgs
Methods
| Item | Description |
|---|---|
| init | — |
Attributes
| Item | Description |
|---|---|
checkpoint_dir | Directory to save checkpoints to. |
enable_checkpoints | Enable saving checkpoints |
export_onnx | Whether to export the model to ONNX format instead of just saving a checkpoint. |
name_prefix_override | Override the name prefix for the checkpoint files (e.g. SAC, PPO, etc.). |
save_final_policy | Whether to save the final policy after training is complete. |
save_freq | Frequency with which to save checkpoints. |
| save_replay_buffer | Whether to save the replay buffer when saving a checkpoint. |
| save_vecnormalize | Whether to save the vector normalization statistics when saving a checkpoint. |
Parameters
enable_checkpoints (bool)
checkpoint_dir (Annotated[Path, Parameter(validator=(Path(exists=False, file_okay=False, dir_okay=True, ext=()),))])
save_freq (Annotated[int, Parameter(validator=(Number(lt=None, lte=None, gt=None, gte=0, modulo=None),))])
name_prefix_override (str | None)
export_onnx (bool)
save_final_policy (bool)
save_replay_buffer (bool)
save_vecnormalize (bool)
init
__init__( enable_checkpoints=False, checkpoint_dir=WindowsPath("ckpt"), save_freq=100000, name_prefix_override=None, export_onnx=False, save_final_policy=False, save_replay_buffer=False, save_vecnormalize=False,)Parameters
enable_checkpoints (bool)
checkpoint_dir (Annotated[Path, Parameter(validator=(Path(exists=False, file_okay=False, dir_okay=True, ext=()),))])
save_freq (Annotated[int, Parameter(validator=(Number(lt=None, lte=None, gt=None, gte=0, modulo=None),))])
name_prefix_override (str | None)
export_onnx (bool)
save_final_policy (bool)
save_replay_buffer (bool)
save_vecnormalize (bool)
Returns
None
save_replay_buffer
= False save_replay_buffer: boolWhether to save the replay buffer when saving a checkpoint. This allows for resuming training from the same state of the replay buffer.
save_vecnormalize
= False save_vecnormalize: boolWhether to save the vector normalization statistics when saving a checkpoint. This is useful for environments where observations need to be normalized, and it allows for consistent normalization when resuming training.