update projects

projects/codes/HierarchicalDQN/README.md

@@ -0,0 +1,13 @@
+# Hierarchical DQN
+
+## 原理简介
+
+Hierarchical DQN是一种分层强化学习方法，与DQN相比增加了一个meta controller，
+
+![image-20210331153115575](assets/image-20210331153115575.png)
+
+即学习时，meta controller每次会生成一个goal，然后controller或者说下面的actor就会达到这个goal，直到done为止。这就相当于给agent增加了一个队长，队长擅长制定局部目标，指导agent前行，这样应对一些每回合步数较长或者稀疏奖励的问题会有所帮助。
+
+## 伪代码
+
+![image-20210331153542314](assets/image-20210331153542314.png)

projects/codes/HierarchicalDQN/agent.py

@@ -0,0 +1,154 @@
+#!/usr/bin/env python
+# coding=utf-8
+'''
+Author: John
+Email: johnjim0816@gmail.com
+Date: 2021-03-24 22:18:18
+LastEditor: John
+LastEditTime: 2021-05-04 22:39:34
+Discription: 
+Environment: 
+'''
+import torch
+import torch.nn as nn
+import torch.optim as optim
+import torch.nn.functional as F
+import numpy as np
+import random,math
+
+class ReplayBuffer:
+    def __init__(self, capacity):
+        self.capacity = capacity # 经验回放的容量
+        self.buffer = [] # 缓冲区
+        self.position = 0 
+    
+    def push(self, state, action, reward, next_state, done):
+        ''' 缓冲区是一个队列，容量超出时去掉开始存入的转移(transition)
+        '''
+        if len(self.buffer) < self.capacity:
+            self.buffer.append(None)
+        self.buffer[self.position] = (state, action, reward, next_state, done)
+        self.position = (self.position + 1) % self.capacity 
+    
+    def sample(self, batch_size):
+        batch = random.sample(self.buffer, batch_size) # 随机采出小批量转移
+        state, action, reward, next_state, done =  zip(*batch) # 解压成状态，动作等
+        return state, action, reward, next_state, done
+    
+    def __len__(self):
+        ''' 返回当前存储的量
+        '''
+        return len(self.buffer)
+class MLP(nn.Module):
+    def __init__(self, input_dim,output_dim,hidden_dim=128):
+        """ 初始化q网络，为全连接网络
+            input_dim: 输入的特征数即环境的状态维度
+            output_dim: 输出的动作维度
+        """
+        super(MLP, self).__init__()
+        self.fc1 = nn.Linear(input_dim, hidden_dim) # 输入层
+        self.fc2 = nn.Linear(hidden_dim,hidden_dim) # 隐藏层
+        self.fc3 = nn.Linear(hidden_dim, output_dim) # 输出层
+        
+    def forward(self, x):
+        # 各层对应的激活函数
+        x = F.relu(self.fc1(x)) 
+        x = F.relu(self.fc2(x))
+        return self.fc3(x)
+        
+class HierarchicalDQN:
+    def __init__(self,n_states,n_actions,cfg):
+        self.n_states = n_states
+        self.n_actions = n_actions
+        self.gamma = cfg.gamma
+        self.device = cfg.device
+        self.batch_size = cfg.batch_size
+        self.frame_idx = 0  # 用于epsilon的衰减计数
+        self.epsilon = lambda frame_idx: cfg.epsilon_end + (cfg.epsilon_start - cfg.epsilon_end ) * math.exp(-1. * frame_idx / cfg.epsilon_decay)
+        self.policy_net = MLP(2*n_states, n_actions,cfg.hidden_dim).to(self.device)
+        self.meta_policy_net = MLP(n_states, n_states,cfg.hidden_dim).to(self.device)
+        self.optimizer = optim.Adam(self.policy_net.parameters(),lr=cfg.lr)
+        self.meta_optimizer = optim.Adam(self.meta_policy_net.parameters(),lr=cfg.lr)
+        self.memory = ReplayBuffer(cfg.memory_capacity)
+        self.meta_memory = ReplayBuffer(cfg.memory_capacity)
+        self.loss_numpy  = 0
+        self.meta_loss_numpy  = 0
+        self.losses = []
+        self.meta_losses = []
+    def to_onehot(self,x):
+        oh = np.zeros(self.n_states)
+        oh[x - 1] = 1.
+        return oh
+    def set_goal(self,state):
+        if random.random() > self.epsilon(self.frame_idx):
+            with torch.no_grad():
+                state = torch.tensor(state, device=self.device, dtype=torch.float32).unsqueeze(0)
+                goal = self.meta_policy_net(state).max(1)[1].item() 
+        else:
+            goal = random.randrange(self.n_states)
+        return goal
+    def choose_action(self,state):
+        self.frame_idx += 1
+        if random.random() > self.epsilon(self.frame_idx):
+            with torch.no_grad():
+                state = torch.tensor(state, device=self.device, dtype=torch.float32).unsqueeze(0)
+                q_value = self.policy_net(state)
+                action = q_value.max(1)[1].item()  
+        else:
+            action = random.randrange(self.n_actions)
+        return action
+    def update(self):
+        self.update_policy()
+        self.update_meta()
+    def update_policy(self): 
+        if self.batch_size > len(self.memory):
+            return
+        state_batch, action_batch, reward_batch, next_state_batch, done_batch = self.memory.sample(self.batch_size)
+        state_batch = torch.tensor(state_batch,device=self.device,dtype=torch.float)
+        action_batch = torch.tensor(action_batch,device=self.device,dtype=torch.int64).unsqueeze(1)  
+        reward_batch = torch.tensor(reward_batch,device=self.device,dtype=torch.float)  
+        next_state_batch = torch.tensor(next_state_batch,device=self.device, dtype=torch.float)
+        done_batch = torch.tensor(np.float32(done_batch),device=self.device)
+        q_values = self.policy_net(state_batch).gather(dim=1, index=action_batch).squeeze(1)
+        next_state_values = self.policy_net(next_state_batch).max(1)[0].detach()
+        expected_q_values = reward_batch + 0.99 * next_state_values * (1-done_batch)
+        loss = nn.MSELoss()(q_values, expected_q_values) 
+        self.optimizer.zero_grad() 
+        loss.backward()
+        for param in self.policy_net.parameters():  # clip防止梯度爆炸
+            param.grad.data.clamp_(-1, 1)
+        self.optimizer.step()  
+        self.loss_numpy = loss.detach().cpu().numpy()
+        self.losses.append(self.loss_numpy)  
+    def update_meta(self):
+        if self.batch_size > len(self.meta_memory):
+            return
+        state_batch, action_batch, reward_batch, next_state_batch, done_batch = self.meta_memory.sample(self.batch_size)
+        state_batch = torch.tensor(state_batch,device=self.device,dtype=torch.float)
+        action_batch = torch.tensor(action_batch,device=self.device,dtype=torch.int64).unsqueeze(1)  
+        reward_batch = torch.tensor(reward_batch,device=self.device,dtype=torch.float)  
+        next_state_batch = torch.tensor(next_state_batch,device=self.device, dtype=torch.float)
+        done_batch = torch.tensor(np.float32(done_batch),device=self.device)
+        q_values = self.meta_policy_net(state_batch).gather(dim=1, index=action_batch).squeeze(1)
+        next_state_values = self.meta_policy_net(next_state_batch).max(1)[0].detach()
+        expected_q_values = reward_batch + 0.99 * next_state_values * (1-done_batch)
+        meta_loss = nn.MSELoss()(q_values, expected_q_values) 
+        self.meta_optimizer.zero_grad() 
+        meta_loss.backward()
+        for param in self.meta_policy_net.parameters():  # clip防止梯度爆炸
+            param.grad.data.clamp_(-1, 1)
+        self.meta_optimizer.step() 
+        self.meta_loss_numpy = meta_loss.detach().cpu().numpy()
+        self.meta_losses.append(self.meta_loss_numpy)
+
+    def save(self, path):
+        torch.save(self.policy_net.state_dict(), path+'policy_checkpoint.pth')
+        torch.save(self.meta_policy_net.state_dict(), path+'meta_checkpoint.pth')
+
+    def load(self, path):
+        self.policy_net.load_state_dict(torch.load(path+'policy_checkpoint.pth'))
+        self.meta_policy_net.load_state_dict(torch.load(path+'meta_checkpoint.pth'))
+        
+
+        
+        

projects/codes/HierarchicalDQN/task0.py

@@ -0,0 +1,88 @@
+#!/usr/bin/env python
+# coding=utf-8
+'''
+Author: John
+Email: johnjim0816@gmail.com
+Date: 2021-03-29 10:37:32
+LastEditor: John
+LastEditTime: 2021-05-04 22:35:56
+Discription: 
+Environment: 
+'''
+import sys
+import os
+curr_path = os.path.dirname(os.path.abspath(__file__))  # 当前文件所在绝对路径
+parent_path = os.path.dirname(curr_path)  # 父路径
+sys.path.append(parent_path)  # 添加路径到系统路径
+
+import datetime
+import numpy as np
+import torch
+import gym
+
+from common.utils import save_results,make_dir
+from common.utils import plot_rewards
+from HierarchicalDQN.agent import HierarchicalDQN
+from HierarchicalDQN.train import train,test
+
+curr_time = datetime.datetime.now().strftime("%Y%m%d-%H%M%S")  # 获取当前时间
+algo_name = "Hierarchical DQN"  # 算法名称
+env_name = 'CartPole-v0'  # 环境名称
+class HierarchicalDQNConfig:
+    def __init__(self):
+        self.algo_name = algo_name  # 算法名称
+        self.env_name = env_name  # 环境名称
+        self.device = torch.device(
+            "cuda" if torch.cuda.is_available() else "cpu")  # 检测GPU
+        self.train_eps = 300  # 训练的episode数目
+        self.test_eps = 50  # 测试的episode数目
+        self.gamma = 0.99
+        self.epsilon_start = 1  # start epsilon of e-greedy policy
+        self.epsilon_end = 0.01
+        self.epsilon_decay = 200
+        self.lr = 0.0001  # learning rate
+        self.memory_capacity = 10000  # Replay Memory capacity
+        self.batch_size = 32
+        self.target_update = 2  # 目标网络的更新频率
+        self.hidden_dim = 256  # 网络隐藏层
+class PlotConfig:
+    ''' 绘图相关参数设置
+    '''
+
+    def __init__(self) -> None:
+        self.algo_name = algo_name  # 算法名称
+        self.env_name = env_name  # 环境名称
+        self.device = torch.device(
+            "cuda" if torch.cuda.is_available() else "cpu")  # 检测GPU
+        self.result_path = curr_path + "/outputs/" + self.env_name + \
+            '/' + curr_time + '/results/'  # 保存结果的路径
+        self.model_path = curr_path + "/outputs/" + self.env_name + \
+            '/' + curr_time + '/models/'  # 保存模型的路径
+        self.save = True  # 是否保存图片
+
+def env_agent_config(cfg,seed=1):
+    env = gym.make(cfg.env_name)  
+    env.seed(seed)
+    n_states = env.observation_space.shape[0]
+    n_actions = env.action_space.n
+    agent = HierarchicalDQN(n_states,n_actions,cfg)
+    return env,agent
+
+if __name__ == "__main__":
+    cfg = HierarchicalDQNConfig()
+    plot_cfg = PlotConfig()
+    # 训练
+    env, agent = env_agent_config(cfg, seed=1)
+    rewards, ma_rewards = train(cfg, env, agent)
+    make_dir(plot_cfg.result_path, plot_cfg.model_path)  # 创建保存结果和模型路径的文件夹
+    agent.save(path=plot_cfg.model_path)  # 保存模型
+    save_results(rewards, ma_rewards, tag='train',
+                path=plot_cfg.result_path)  # 保存结果
+    plot_rewards(rewards, ma_rewards, plot_cfg, tag="train")  # 画出结果
+    # 测试
+    env, agent = env_agent_config(cfg, seed=10)
+    agent.load(path=plot_cfg.model_path)  # 导入模型
+    rewards, ma_rewards = test(cfg, env, agent)
+    save_results(rewards, ma_rewards, tag='test', path=plot_cfg.result_path)  # 保存结果
+    plot_rewards(rewards, ma_rewards, plot_cfg, tag="test")  # 画出结果
+

projects/codes/HierarchicalDQN/train.py

@@ -0,0 +1,77 @@
+import sys
+import os
+curr_path = os.path.dirname(os.path.abspath(__file__))  # 当前文件所在绝对路径
+parent_path = os.path.dirname(curr_path)  # 父路径
+sys.path.append(parent_path)  # 添加路径到系统路径
+
+import numpy as np
+
+def train(cfg, env, agent):
+    print('开始训练!')
+    print(f'环境：{cfg.env_name}, 算法：{cfg.algo_name}, 设备：{cfg.device}')
+    rewards = [] # 记录所有回合的奖励
+    ma_rewards = []  # 记录所有回合的滑动平均奖励
+    for i_ep in range(cfg.train_eps):
+        state = env.reset()
+        done = False
+        ep_reward = 0
+        while not done:
+            goal = agent.set_goal(state)
+            onehot_goal = agent.to_onehot(goal)
+            meta_state = state
+            extrinsic_reward = 0
+            while not done and goal != np.argmax(state):
+                goal_state = np.concatenate([state, onehot_goal])
+                action = agent.choose_action(goal_state)
+                next_state, reward, done, _ = env.step(action)
+                ep_reward += reward
+                extrinsic_reward += reward
+                intrinsic_reward = 1.0 if goal == np.argmax(
+                    next_state) else 0.0
+                agent.memory.push(goal_state, action, intrinsic_reward, np.concatenate(
+                    [next_state, onehot_goal]), done)
+                state = next_state
+                agent.update()
+        if (i_ep+1)%10 == 0: 
+            print(f'回合：{i_ep+1}/{cfg.train_eps}，奖励：{ep_reward}，Loss:{agent.loss_numpy:.2f}， Meta_Loss:{agent.meta_loss_numpy:.2f}')
+        agent.meta_memory.push(meta_state, goal, extrinsic_reward, state, done)
+        rewards.append(ep_reward)
+        if ma_rewards:
+            ma_rewards.append(
+                0.9*ma_rewards[-1]+0.1*ep_reward)
+        else:
+            ma_rewards.append(ep_reward)
+    print('完成训练！')
+    return rewards, ma_rewards
+
+def test(cfg, env, agent):
+    print('开始测试!')
+    print(f'环境：{cfg.env_name}, 算法：{cfg.algo_name}, 设备：{cfg.device}')
+    rewards = [] # 记录所有回合的奖励
+    ma_rewards = []  # 记录所有回合的滑动平均奖励
+    for i_ep in range(cfg.train_eps):
+        state = env.reset()
+        done = False
+        ep_reward = 0
+        while not done:
+            goal = agent.set_goal(state)
+            onehot_goal = agent.to_onehot(goal)
+            extrinsic_reward = 0
+            while not done and goal != np.argmax(state):
+                goal_state = np.concatenate([state, onehot_goal])
+                action = agent.choose_action(goal_state)
+                next_state, reward, done, _ = env.step(action)
+                ep_reward += reward
+                extrinsic_reward += reward
+                state = next_state
+                agent.update()
+        if (i_ep+1)%10 == 0: 
+            print(f'回合：{i_ep+1}/{cfg.train_eps}，奖励：{ep_reward}，Loss:{agent.loss_numpy:.2f}， Meta_Loss:{agent.meta_loss_numpy:.2f}')
+        rewards.append(ep_reward)
+        if ma_rewards:
+            ma_rewards.append(
+                0.9*ma_rewards[-1]+0.1*ep_reward)
+        else:
+            ma_rewards.append(ep_reward)
+    print('完成训练！')
+    return rewards, ma_rewards