← Back to API reference

neuraltrain.utils.WandbLoggerConfig

pydantic model neuraltrain.utils.WandbLoggerConfig

Pydantic configuration for torch-lightning’s wandb logger. See https://lightning.ai/docs/pytorch/stable/extensions/generated/lightning.pytorch.loggers.WandbLogger.html. If you want to resume a run, you can use the id field to specify the run id, either in the config or in the build method.

Fields:

• anonymous (bool | None)

• dir (pathlib.Path | None)

• entity (str | None)

• experiment (Any | None)

• group (str)

• host (str | None)

• id (str | None)

• log_model (str | bool)

• name (str | None)

• offline (bool)

• prefix (str)

• project (str | None)

• resume (Literal['allow', 'never', 'must'])

field name: str | None = None

field group: str [Required]

field entity: str | None = None

field project: str | None = None

field offline: bool = False

field host: str | None = None

field id: str | None = None

field dir: Path | None = None

field anonymous: bool | None = None

field log_model: str | bool = False

field experiment: tp.Any | None = None

field prefix: str = ''

field resume: tp.Literal['allow', 'never', 'must'] = 'allow'

build(save_dir: str | Path, xp_config: dict | BaseModel | None = None, run_id: str | None = None) → Any

← Back to API reference