​
EvolInstructPipeline

class EvolInstructPipeline:
Pipeline for evolving prompts using the Evol-Instruct methodology. Supports custom templates defining evolution strategies and methods. The pipeline leverages language models to iteratively refine prompts through specified evolution strategies. Parameters:
  • templates (Type[EvolInstructTemplates]): Template class containing evolution strategy and method definitions. Must provide EVOL_METHODS and STRATEGY attributes. (default: :obj:EvolInstructTemplates)
  • agent (Optional[ChatAgent]): Chat agent instance for LLM interaction.
  • If: obj:None, initializes with a default ChatAgent. (default: :obj:None)

​
init

def __init__(
    self,
    templates: Type = EvolInstructTemplates,
    agent: Optional[ChatAgent] = None
):
Initialize pipeline with templates and language model agent. Parameters:
  • templates (Type[EvolInstructTemplates]): Template class containing evolution strategy configurations. (default: :obj:EvolInstructTemplates)
  • agent (Optional[ChatAgent]): Preconfigured chat agent instance. Creates a default ChatAgent if not provided. (default: :obj:None)

​
_resolve_evolution_method

def _resolve_evolution_method(self, method_key: str):
Resolve evolution method key to concrete implementation. Parameters:
  • method_key (str): Input method identifier. Can be: - Direct method key from templates.EVOL_METHODS - Strategy name from templates.STRATEGY keys
Returns: str: Resolved method key from EVOL_METHODS

​
_get_evolution_methods

def _get_evolution_methods(self, method: Union[str, List[str]], num_generations: int = 2):
Get list of evolution methods based on input specification. Parameters:
  • method (Union[str, List[str]]): Specification for method selection. Can be: - Strategy name for methods from that strategy - Specific method name - List of method specifications
  • num_generations (int): Number of methods to return.
Returns: List[str]: List of resolved method names

​
_generate_single_evolution

def _generate_single_evolution(
    self,
    prompt: str,
    method: str,
    return_method: bool = False
):
Generate a single evolved prompt from a seed prompt. Parameters:
  • prompt (str): The seed prompt to evolve.
  • method (str): The evolution method key to use.
  • return_method (bool): If True, returns method along with prompt.
Returns: Tuple[str, str]: Evolved prompt and method

​
_generate_multiple_evolutions

def _generate_multiple_evolutions(
    self,
    prompt: str,
    method: Union[str, List[str]],
    num_generations: int = 2,
    keep_original: bool = True,
    num_threads: int = 10
):
Generate multiple evolved versions of a prompt. Parameters:
  • prompt (str): Seed prompt to evolve.
  • method (Union[str, List[str]]): Evolution method specification.
  • num_generations (int): Candidates to generate per iteration.
  • keep_original (bool): Whether to keep the original prompt.
  • num_threads (int): Number of threads for parallel processing.
Returns: List[Tuple[str, str]]: List of (evolved_prompt, method) pairs

​
_generate_iterative_evolutions

def _generate_iterative_evolutions(
    self,
    prompt: str,
    evolution_spec: Union[str, List[Union[str, List[str]]]],
    num_generations: int = 2,
    num_iterations: Optional[int] = None,
    keep_original: bool = True,
    scorer: Optional[BaseScorer] = None,
    num_threads: int = 10
):
Generate iterative evolutions of a prompt with scoring. Parameters:
  • prompt (str): Seed prompt to evolve.
  • evolution_spec (Union[str, List[Union[str, List[str]]]]): Evolution method specification. If a list is provided and num_iterations is None, then num_iterations is set to the length of the list.
  • num_generations (int): Candidates to generate per iteration.
  • num_iterations (Optional[int]): Number of evolution iterations. Defaults to the length of evolution_spec.
  • keep_original (bool): Include original prompt in results.
  • scorer (Optional[BaseScorer]): Scoring model for candidate.
  • num_threads (int): Number of threads for parallel processing.
Returns: Dict[int, List[Dict[str, Any]]]: Evolution results per iteration, where each candidate is represented as a dict with keys: “instruction”, “method”, and “scores”.

​
generate

def generate(
    self,
    prompts: List[str],
    evolution_spec: Union[str, List[Union[str, List[str]]]],
    num_generations: int = 2,
    num_iterations: Optional[int] = None,
    keep_original: bool = True,
    scorer: Optional[BaseScorer] = None,
    num_chunks: int = 1,
    retry_limit: int = 3,
    retry_delay: float = 1.0,
    num_threads: int = 10
):
Evolve a batch of prompts through iterative refinement. Parameters:
  • prompts (List[str]): Seed prompts to evolve.
  • evolution_spec (Union[str, List[Union[str, List[str]]]]): Evolution method specification. If a list is provided and num_iterations is None, then num_iterations is set to the length of the list.
  • num_generations (int): Candidates to generate per iteration.
  • num_iterations (Optional[int]): Number of evolution iterations. Defaults to the length of evolution_spec.
  • keep_original (bool): Include original prompts in results.
  • scorer (Optional[BaseScorer]): Scoring model for candidate.
  • num_chunks (int): Number of parallel processing chunks.
  • retry_limit (int): Max retries for failed generations.
  • retry_delay (float): Delay between retries in seconds.
  • num_threads (int): Number of threads for parallel processing.
Returns: List[Dict[int, List[Dict[str, Any]]]]: Evolution results.