luml.integrations.lightgbm.packaging

save_lightgbm

def save_lightgbm(
        estimator: "Union[Booster, lgb.LGBMModel]",
        inputs: Any,
        path: str | None = None,
        input_format: Literal["unified", "native"] = "unified",
        dependencies: Literal["default"] | Literal["all"] | list[str] = "default",
        extra_dependencies: list[str] | None = None,
        extra_code_modules: list[str] | Literal["auto"] | None = None,
        manifest_model_name: str | None = None,
        manifest_model_version: str | None = None,
        manifest_model_description: str | None = None,
        manifest_extra_producer_tags: list[str] | None = None
) -> ModelReference

Save a LightGBM model as a Luml model.

Arguments:

• estimator - The LightGBM model to save (Booster or LGBMModel).
• inputs - Example input data for the model.
• path - Path where the model will be saved. Auto-generated if None.
• input_format - Input format for inference:
  • "unified": per-feature NDJSON inputs (default).
  • "native": single JSON payload with dataset structure and optional predict config; supports both dense row data and CSR sparse matrices.
• dependencies - Dependency management strategy ("default", "all", or list).
• extra_dependencies - Additional pip dependencies to include.
• extra_code_modules - Local code modules to package ("auto" or list).
• manifest_model_name - Optional name for the model in manifest.
• manifest_model_version - Optional version for the model in manifest.
• manifest_model_description - Optional description for the model.
• manifest_extra_producer_tags - Additional producer tags for model lineage.

Returns:

• ModelReference - Reference to the saved model.