tasks/base_task.py

from __future__ import annotations

from abc import ABC, abstractmethod
from typing import List


class BaseTask(ABC):
    task_id: str
    difficulty: str
    config_path: str
    step_limit: int
    _MIN_SCORE: float = 0.001
    _MAX_SCORE: float = 0.999

    @abstractmethod
    def get_system_prompt(self) -> str:
        raise NotImplementedError

    @abstractmethod
    def grade(
        self,
        declared_service: str | None,
        declared_fault: str | None,
        steps_used: int,
        rewards: List[float],
    ) -> float:
        raise NotImplementedError

    @staticmethod
    def _score_correct(correct: bool, steps_used: int, step_limit: int) -> float:
        if not correct:
            return BaseTask._MIN_SCORE
        efficiency = max(0.0, (step_limit - steps_used) / step_limit)
        score = 0.5 + 0.5 * efficiency
        return float(min(BaseTask._MAX_SCORE, max(BaseTask._MIN_SCORE, score)))