env/config.py

"""Configuration objects for MiniGridEnv."""

from dataclasses import dataclass, field
from typing import Optional


@dataclass
class RewardConfig:
    """Reward-shaping controls for environment step rewards."""

    mode: str = "binary"  # "binary" | "shaped" | "efficiency"
    completion_reward: float = 1.0
    failure_reward: float = 0.0
    step_penalty: float = 0.0
    invalid_action_penalty: float = 0.0
    efficiency_bonus_weight: float = 0.5


@dataclass
class EnvConfig:
    """Episode, observation, and reproducibility settings."""

    # Level selection
    level_name: str = "GoToRedBall"

    # Episode limits
    max_steps_override: Optional[int] = None

    # Reward configuration
    reward: RewardConfig = field(default_factory=RewardConfig)

    # Observation configuration
    include_history: bool = True
    max_history_length: int = 20
    include_raw_grid: bool = False

    # Reproducibility and optional metrics
    seed: int = 42
    compute_optimal: bool = False
    track_carrying: bool = True