බහු-සැකසුම්සමග Atari දවටනය

9import multiprocessing
10import multiprocessing.connection
11
12import cv2
13import gym
14import numpy as np

17class Game:

38    def __init__(self, seed: int):

40        self.env = gym.make('BreakoutNoFrameskip-v4')
41        self.env.seed(seed)

44        self.obs_4 = np.zeros((4, 84, 84))

47        self.obs_2_max = np.zeros((2, 84, 84))

50        self.rewards = []

52        self.lives = 0

54    def step(self, action):

66        reward = 0.
67        done = None

70        for i in range(4):

72            obs, r, done, info = self.env.step(action)
73
74            if i >= 2:
75                self.obs_2_max[i % 2] = self._process_obs(obs)
76
77            reward += r

80            lives = self.env.unwrapped.ale.lives()

82            if lives < self.lives:
83                done = True
84                break

87        self.rewards.append(reward)
88
89        if done:

91            episode_info = {"reward": sum(self.rewards), "length": len(self.rewards)}
92            self.reset()
93        else:
94            episode_info = None

97            obs = self.obs_2_max.max(axis=0)

100            self.obs_4 = np.roll(self.obs_4, shift=-1, axis=0)
101            self.obs_4[-1] = obs
102
103        return self.obs_4, reward, done, episode_info

105    def reset(self):

112        obs = self.env.reset()

115        obs = self._process_obs(obs)
116        for i in range(4):
117            self.obs_4[i] = obs
118        self.rewards = []
119
120        self.lives = self.env.unwrapped.ale.lives()
121
122        return self.obs_4

124    @staticmethod
125    def _process_obs(obs):

130        obs = cv2.cvtColor(obs, cv2.COLOR_RGB2GRAY)
131        obs = cv2.resize(obs, (84, 84), interpolation=cv2.INTER_AREA)
132        return obs

135def worker_process(remote: multiprocessing.connection.Connection, seed: int):

143    game = Game(seed)

146    while True:
147        cmd, data = remote.recv()
148        if cmd == "step":
149            remote.send(game.step(data))
150        elif cmd == "reset":
151            remote.send(game.reset())
152        elif cmd == "close":
153            remote.close()
154            break
155        else:
156            raise NotImplementedError

159class Worker:

164    def __init__(self, seed):
165        self.child, parent = multiprocessing.Pipe()
166        self.process = multiprocessing.Process(target=worker_process, args=(parent, seed))
167        self.process.start()