MCPcopy Create free account
hub / github.com/FareedKhan-dev/ai-long-task / SerializableResult

Class SerializableResult

core/process_parallel.py:50–68  ·  view source on GitHub ↗

Result that can be pickled and sent between processes

Source from the content-addressed store, hash-verified

48
49@dataclass
50class SerializableResult:
51 """Result that can be pickled and sent between processes"""
52
53 # Dictionary representation of the child program (Program objects might not pickle well if they contain complex locks)
54 child_program_dict: Optional[Dict[str, Any]] = None
55 # ID of the parent program
56 parent_id: Optional[str] = None
57 # Time taken for the iteration
58 iteration_time: float = 0.0
59 # Prompt used for generation
60 prompt: Optional[Dict[str, str]] = None
61 # Raw LLM response
62 llm_response: Optional[str] = None
63 # Artifacts generated (images, logs, library docs)
64 artifacts: Optional[Dict[str, Any]] = None
65 # The iteration number
66 iteration: int = 0
67 # Error message if something went wrong
68 error: Optional[str] = None
69
70
71def _worker_init(config_dict: dict, evaluation_file: str, parent_env: dict = None) -> None:

Callers 1

_run_iteration_workerFunction · 0.85

Calls

no outgoing calls

Tested by

no test coverage detected