examples/aflow/molqa/optimized_norman/round_5/graph.py

import evoagentx.workflow.operators as operator
import examples.aflow.pertqa.optimized_norman.round_5.prompt as prompt_custom
from evoagentx.models.model_configs import LLMConfig
from evoagentx.benchmark.benchmark import Benchmark
from evoagentx.models.model_utils import create_llm_instance

class Workflow:
    
    def __init__(
        self,
        name: str,
        llm_config: LLMConfig,
        benchmark: Benchmark
    ):
        self.name = name
        self.llm = create_llm_instance(llm_config)
        self.benchmark = benchmark
        self.custom = operator.Custom(self.llm)
        self.answer_generate = operator.AnswerGenerate(self.llm)
        self.sc_ensemble = operator.QAScEnsemble(self.llm)
        self.review = operator.Custom(self.llm)  # Added review operator
        self.self_ask = operator.Custom(self.llm)  # Added self-ask operator
        self.revise = operator.Custom(self.llm)  # Added revise operator for improvement
    
    async def __call__(self, problem: str):
        """
        Implementation of the workflow
        """
        ask_response = await self.self_ask(input=problem, instruction="Generate clarifying questions for the following problem.")  # Self-ask for clarification
        solution = await self.answer_generate(input=ask_response['response'])  # Generate answer based on clarified problem
        review_response = await self.review(input=solution['answer'], instruction="Review the following answer for correctness and clarity.")  # Review the generated answer
        revised_response = await self.revise(input=review_response['response'], instruction="Revise the following answer for improvement.")  # Revise the reviewed answer
        ensemble_response = await self.sc_ensemble(solutions=[revised_response['response']])  # Use ensemble for final selection
        return ensemble_response['response']  # Return the ensemble response