Pufferlib 集成

# PufferLib 集成指南 ## 概述 PufferLib 提供了一个仿真层，能够与包括 Gymnasium、OpenAI Gym、PettingZoo 以及许多专业环境库在内的流行 RL 框架无缝集成。该仿真层平铺了观测和动作空间以实现高效向量化，同时保持了兼容性。 ## Gymnasium 集成 ### 基础 Gymnasium 环境 python import gymnasium as gym import pufferlib # 方法 1: 直接包装 gym_env = gym.make('CartPole-v1') puffer_env = pufferlib.emulate(gym_env, num_envs=256) # 方法 2: 使用 make env = pufferlib.make('gym-CartPole-v1', num_envs=256) # 方法 3: 自定义 Gymnasium 环境 class MyGymEnv(gym.Env): def __init__(self): self.observation_space = gym.spaces.Box(low=-1, high=1, shape=(4,)) self.action_space = gym.spaces.Discrete(2) def reset(self, seed=None, options=None): super().reset(seed=seed) return self.observation_space.sample(), {} def step(self, action): obs = self.observation_space.sample() reward = 1.0 terminated = False truncated = False info = {} return obs, reward, terminated, truncated, info # 包装自定义环境 puffer_env = pufferlib.emulate(MyGymEnv, num_envs=128) ### Atari 环境 python import gymnasium as gym from gymnasium.wrappers import AtariPreprocessing, FrameStack import pufferlib # 标准 Atari 设置 def make_atari_env(env_name='ALE/Pong-v5'): env = gym.make(env_name) env = AtariPreprocessing(env, frame_skip=4) env = FrameStack(env, num_stack=4) return env # 使用 PufferLib 向量化 env = pufferlib.emulate(make_atari_env, num_envs=256) # 或者使用内置功能 env = pufferlib.make('atari-pong', num_envs=256, frameskip=4, framestack=4) ### 复杂观测空间 python import gymnasium as gym from gymnasium.spaces import Dict, Box, Discrete import pufferlib class ComplexObsEnv(gym.Env): def __init__(self): # 字典观测空间 self.observation_space = Dict({ 'image': Box(low=0, high=255, shape=(84, 84, 3), dtype=np.uint8), 'vector': Box(low=-np.inf, high=np.inf, shape=(10,), dtype=np.float32), 'discrete': Discrete(5) }) self.action_space = Discrete(4) def reset(self, seed=None, options=None): return { 'image': np.zeros((84, 84, 3), dtype=np.uint8), 'vector': np.zeros(