DQN强化学习实践

import torch
import torch.nn as nn
import torch.nn.functional as F
import numpy as np
import gym

# 定义超参数
BATCH_SIZE = 32
LR = 0.01
EPSILON = 0.9  # 最优选择动作百分比
GAMMA = 0.9  # 奖励递减参数
TARGET_REPLACE_ITER = 100  # Q 现实网络的更新频率
MEMORY_CAPACITY = 2000  # 记忆库大小
env = gym.make("CartPole-v0")  # 立杆子游戏
env = env.unwrapped
N_ACTIONS = env.action_space.n  # 杆子能做的动作
N_STATES = env.observation_space.shape[0]  # 杆子能获取的环境信息数
ENV_A_SHAPE = 0 if isinstance(env.action_space.sample(),
                              int) else env.action_space.sample().shape  # to confirm the shape


# 构造DON（Deep Q Network)强化学习神经网络 现实网络 (Target Net) 估计网络 (Eval Net)
class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.fc1 = nn.Linear(N_STATES, 10)
        self.fc1.weight.data.normal_(0, 0.1)  # initialization(初始化）
        self.out = nn.Linear(10, N_ACTIONS)
        self.out.weight.data.normal_(0, 0.1)  # initialization(初始化）

    def forward(self, x):
        x = self.fc1(x)
        x = F.relu(x)
        action_value = self.out(x)
        return action_value


# 构造DQN体系
class DQN(object):
    def __init__(self):
        # 建立 target net 和 eval net 还有 memory
        self.eval_net, self.target_net = Net(), Net()
        self.learn_step_counter = 0  # 用于 target 更新计时
        self.memory_counter = 0  # 记忆库记数
        self.memory = np.zeros((MEMORY_CAPACITY, N_STATES * 2 + 2))  # 初始化记忆库
        self.optimizer = torch.optim.Adam(self.eval_net.parameters(), lr=LR)
        self.lose_func = nn.MSELoss()

    def choose_action(self, x):
        # 根据环境观测值选择动作的机制
        x = torch.unsqueeze(torch.FloatTensor(x), 0)
        # 这里只输入一个 sample
        if np.random.uniform() < EPSILON:  # 选最优动作
            actions_value = self.eval_net.forward(x)
            action = torch.max(actions_value, 1)[1].data.numpy()
            action = action[0] if ENV_A_SHAPE == 0 else action.reshape(ENV_A_SHAPE)  # return the argmax index
        else:
            action = np.random.randint(0, N_ACTIONS)
            action = action if ENV_A_SHAPE == 0 else action.reshape(ENV_A_SHAPE)
        return action

    def store_transition(self, s, a, r, s_):
        # 存储记忆
        transition = np.hstack((s, [a, r], s_))
        # 如果记忆库满了, 就覆盖老数据
        index = self.memory_counter % MEMORY_CAPACITY
        self.memory[index, :] = transition
        self.memory_counter += 1

    def learn(self):
        # target 网络更新 学习记忆库中的记忆
        if self.learn_step_counter % TARGET_REPLACE_ITER == 0:
            self.target_net.load_state_dict(self.eval_net.state_dict())
        self.learn_step_counter += 1
        # 抽取记忆库中的批数据
        sample_index = np.random.choice(MEMORY_CAPACITY, BATCH_SIZE)
        b_memory = self.memory[sample_index, :]
        b_s = torch.FloatTensor(b_memory[:, :N_STATES])
        b_a = torch.LongTensor(b_memory[:, N_STATES:N_STATES + 1].astype(int))
        b_r = torch.FloatTensor(b_memory[:, N_STATES + 1:N_STATES + 2])
        b_s_ = torch.FloatTensor(b_memory[:, -N_STATES:])
        # 针对做过的动作b_a, 来选 q_eval 的值, (q_eval 原本有所有动作的值)
        q_eval = self.eval_net(b_s).gather(1, b_a)  # shape (batch, 1)
        q_next = self.target_net(b_s_).detach()  # q_next 不进行反向传递误差, 所以 detach
        q_target = b_r + GAMMA * q_next.max(1)[0]  # shape (batch, 1)
        loss = self.lose_func(q_eval, q_target)
        # 计算, 更新 eval net
        self.optimizer.zero_grad()
        loss.backward()
        self.optimizer.step()


# 训练 按照 Qlearning 的形式进行 off-policy 的更新. 我们进行回合制更新, 一个回合完了, 进入下一回合. 一直到他们将杆子立起来很久.
dqn = DQN()
for i_episode in range(400):
    s = env.reset()
    while True:
        env.render()  # 显示实验动画
        a = dqn.choose_action(s)
        # 选动作, 得到环境反馈
        s_, r, done, info = env.step(a)
        # 修改 reward, 使 DQN 快速学习
        x, x_dot, theta, theta_dot = s_
        r1 = (env.x_threshold - abs(x)) / env.x_threshold - 0.8
        r2 = (env.theta_threshold_radians - abs(theta)) / env.theta_threshold_radians - 0.5
        r = r1 + r2
        # 存记忆
        dqn.store_transition(s, a, r, s_)
        if dqn.memory_counter > MEMORY_CAPACITY:
            dqn.learn()
        if done:
            break
        s = s_