data_juicer_agents.capabilities.plan

Plan orchestration capability.

class data_juicer_agents.capabilities.plan.ProcessOperatorGenerator(*, model_name: str, api_key: str | None = None, base_url: str | None = None, thinking: bool | None = None)

Bases: object

Generate an operator list for staged plan assembly.

__init__(*, model_name: str, api_key: str | None = None, base_url: str | None = None, thinking: bool | None = None)

generate(*, user_intent: str, retrieval_payload: Dict[str, Any], dataset_spec: Dict[str, Any], dataset_profile: Dict[str, Any] | None = None) → Dict[str, Any]

class data_juicer_agents.capabilities.plan.PlanOrchestrator(*, planner_model_name: str | None = None, llm_api_key: str | None = None, llm_base_url: str | None = None, llm_thinking: bool | None = None)

Bases: object

Fixed orchestration for CLI plan generation.

__init__(*, planner_model_name: str | None = None, llm_api_key: str | None = None, llm_base_url: str | None = None, llm_thinking: bool | None = None)

generate_plan(*, user_intent: str, dataset_path: str, export_path: str, custom_operator_paths: Iterable[Any] | None = None, retrieved_candidates: Dict[str, Any] | None = None, retrieval_top_k: int = 5, retrieval_mode: str = 'auto') → Dict[str, Any]