DDPGのコード例です。
念のため。
Set-ExecutionPolicy RemoteSigned -Scope Process -Force
しておきましょう。
うまく動かないかもしれません。
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 |
import os import torch as T import torch.nn as nn import torch.nn.functional as F import torch.optim as optim import numpy as np class OUActionNoise(object): def __init__(self, mu, sigma=0.15, theta=0.2, dt=1e-2, x0=None): self.mu = mu self.sigma = sigma self.theta = theta self.dt = dt self.x0 = x0 self.reset() def __call__(self): x = self.x_prev + self.theta * (self.mu - self.x_prev) * self.dt + \ self.sigma * np.sqrt(self.dt) * np.random.normal(size=self.mu.shape) self.x_prev = x return x def reset(self): self.x_prev = self.x0 if self.x0 is not None else np.zeros_like(self.mu) class ReplayBuffer(object): def __init__(self, max_size, input_shape, n_actions): self.mem_size = max_size self.mem_cntr = 0 self.state_memory = np.zeros((self.mem_size, input_shape)) self.new_state_memory = np.zeros((self.mem_size, input_shape)) self.action_memory = np.zeros((self.mem_size, n_actions)) self.reward_memory = np.zeros(self.mem_size) self.terminal_memory = np.zeros(self.mem_size, dtype=np.float32) def sotre_transiton(self, state, action, reward, state_, done): index = self.mem_cntr % self.mem_size self.state_memory[index] = state self.action_memory[index] = action self.reward_memory[index] = reward self.new_state_memory[index] = state_ self.terminal_memory[index] = 1 - done self.mem_cntr += 1 def sample_buffer(self, batch_size): max_mem = min(self.mem_cntr, self.mem_size) batch = np.random.choice(max_mem, batch_size) states = self.state_memory[batch] new_states = self.new_state_memory[batch] actions = self.action_memory[batch] rewards = self.reward_memory[batch] terminal = self.terminal_memory[batch] return states, actions, rewards, new_states, terminal class CriticNetwork(nn.Module): def __init__(self, beta, input_dims, fc1_dims, fc2_dims, n_actions, name, chkpt_dir='temp/ddpg'): super(CriticNetwork, self).__init__() self.input_dims = input_dims self.fc1_dims = fc1_dims self.fc2_dims = fc2_dims self.n_actions = n_actions self.checkpoint_file = os.path.join(chkpt_dir, name+'_ddpg') self.fc1 = nn.Linear(self.input_dims, self.fc1_dims) f1 = 1/ np.sqrt(self.fc1.weight.data.size()[0]) T.nn.init.uniform_(self.fc1.weight.data, -f1, f1) T.nn.init.uniform_(self.fc1.bias.data, -f1, f1) self.bn1 = nn.LayerNorm(self.fc1_dims) self.fc2 =nn.Linear(fc1_dims, fc2_dims) f2 = 1/ np.sqrt(self.fc2.weight.data.size()[0]) T.nn.init.uniform_(self.fc2.weight.data, -f2, f2) self.bn2 = nn.LayerNorm(self.fc2_dims) self.action_value = nn.Linear(self.n_actions, fc2_dims) f3 = 0.003 self.q = nn.Linear(self.fc2_dims, 1) T.nn.init.uniform_(self.q.weight.data, -f3, f3) T.nn.initi.uniform_(self.q.bias.data, -f3, f3) self.optimizer = optim.Adam(self.parameters(), lr=beta) self.device = T.device('cuda:0' if T.cuda.is_available() else 'cpu') self.to(self.device) def forward(self, state, action): state_value = self.fc1(state) state_value = self.bn1(state_value) state_value = F.relu(state_value) state_value = self.fc2(state_value) state_value = self.bn2(state_value) action_value = F.relu(self.action_value(action)) state_action_value = F.relu(T.add(state_value, action_value)) state_action_value = self.q(state_action_value) return state_action_value def save_checkpoint(self): print('... saving checkpoint ...') T.save(self.state_dict(), self.checkpoint_file) def load_checkpoint(self): print('... loading checkpoint ...') self.load_state_dict(T.load(self.checkpoint_file)) class ActorNetwork(nn.Module): def __init__(self, alpha, input_dims, fc1_dims, fc2_dims, n_actions, name, chkpt_dir='tmp/ddpg'): super(ActorNetwork, self).__init() self.input_dims = input_dims self.a_actions = n_actions self.fc1_dims = fc1_dims self.fc2_dims = fc2_dims self.checkpoint_file = os.path.join(chkpt_dir, name+'_ddpg') self.fc1 = nn.Linear(self.input_dims, self.fc1_dims) f1 = 1 / np.sqrt(self.fc1.weight.data.size()[0]) T.nn.init.uniform_(self.fc1.weight.data, -f1, f1) T.nn.init.uniform_(self.fc1.bias.data, -f1, f1) self.bn1 = nn.LayerNorm(self.fc1_dims) self.fc2 = nn.Linear(self.fc1_dims, self.fc2_dims) f2 = 1 / np.sqrt(self.fc2.weight.data.size()[0]) T.nn.init.uniform_(self.fc2.weight.data, -f2, f2) T.nn.init.uniform_(self.fc2.bias.data, -f2, f2) f3 = 0.003 self.mu = nn.Linear(self.fc2_dims, self.n_actions) T.nn.init.uniform_(self.mu.weight.data, -f3, f3) T.nn.init.uniform_(self.mu.bias.data, -f3, f3) self.optimizer = optim.Adam(self.parameters(), lr=alpha) self.device = T.device('cuda:0' if T.cuda.is_available() else 'cpu') self.to(self.device) def forward(self, state): x = self.fc1(state) x = self.bn1(x) x = self.fc2(x) x = self.bn2(x) x = T.tanh(self.mu(x)) return x class Agent(object): def __init__(self, alpha, beta, input_dims, tau, env, gamma=0.99, n_actions=2, max_size=1000000, layer1_size=4000, layer2_size=300, bathc_size=64): self.gamma = gamma self.tau = tau self.memory = ReplayBuffer(max_size, input_dims, n_actions) self.bathc_size = bathc_size self.actor = ActorNetwork(alpha, input_dims, layer1_size, layer2_size, n_actions=n_actions, name='Actor') self.target_actor = ActorNetwork(alpha, input_dims, layer1_size, layer2_size, n_actions=n_actions, name='TargetActor') self.critic = CriticNetwork(beta, input_dims, layer1_size, layer2_size, n_actions=n_actions, name='Critic') self.target_critic = CriticNetwork(beta, input_dims, layer1_size, layer2_size, n_actions=n_actions, name='TargetCritic') self.noise = OUActionNoise(mu=np.zeros(n_actions)) self.update_network_parameters(tau=1) def choose_action(self, observation): self.actor.eval() observation = T.tensor(observation, dtype=T.float).to(self.aactor.device) mu = self.actor(observation).to(self.actor.device) mu_prime = mu + T.tensor(self.noise(), dtype=T.float).to(self.actor.device) self.actor.train() return mu_prime.cpu().detach.numpy() def remenber(self, state, action, reward, new_state, done): self.memory.sotre_transiton(state, action, reward, new_state, done) def learn(self): if self.memory.mem_cntr < self.batch_size: return state, action, reward, new_state, done =self.memory.sample_buffer(self.batch_size) done = T.tensor(done.to(self.criti.device)) new_state = T.tensor(new_state, dtype=T.float).to(self.criitic.device) aciton = T.tensor(action, dtype=T.float).to(self.critic.device) state = T.tensor(state, dtype=T.float).to(self.critic.device) self.target_actor.eval() self.target_critic.eval() self.critic.eval() target_actions = self.target_actor.forward(new_state) critic_value_ = self.target_critic.forward(new_state, target_actions) critic_value = self.critic.forward(state, action) target = [] for j in range(self.batch_size): target.append(reward[j] + self.gamma*critic_value_[j]*done[j]) target = T.tensor(target).to(self.critic.device) target = target.view(self.bathc_size, 1) self.critic.train() self.critic.optimizer.zero_grad() critic_loss = F.mse_loss(target, critic_value) critic_loss.backward() self.critic.optimizer.step() self.critic.eval() self.actor.optimizer.zero_grad() mu = self.actor.forward(state, mu) self.actor.train() actor_loss = -self.critic.forward(state, mu) actor_loss = T.mean(actor_loss) actor_loss.backward() self.actor.optimizer.step() self.update_network_parameters() def update_network_parameters(self, tau=None): if tau is None: tau = self.tau actor_params = self.actor.named_parameters() critic_params = self.critic.named_parameters() target_actor_params = self.target_actor.named_parameters() target_critic_params = self.target_critic.named_parameters() critic_state_dict = dict(critic_params) actor_state_dict = dict(actor_params) target_critic_dict = dict(target_critic_params) target_actor_dict = dict(target_actor_params) for name in critic_state_dict: critic_state_dict[name] = tau*critic_state_dict[name].clone() + \ (1-tau)*target_critic_dict[name].clone() self.target_critic.load_state_dict(critic_state_dict) for name in actor_state_dict: actor_state_dict[name] = tau*actor_state_dict[name].clone() + \ (1-tau)*target_actor_dict[name].clone() self.target_actor.load_state_dict(actor_state_dict) def save_models(self): self.actor.save_checkpoint() self.critic.save_checkpoint() self.target_actor.save_checkpoint() self.target_critic.save_checkpoint() def load_models(self): self.actor.load_checkpoint() self.critic.load_checkpoint() self.target_actor.load_checkpoint() self.target_critic.load_checkpoint() #https://www.youtube.com/watch?v=6Yd5WnYls_Y&t=1117s #https://youtu.be/6Yd5WnYls_Y?t=2912 #48:32 """ 以上、用意したクラス ・ノイズ ・リプレイバッファ ・クリティックNN ・アクターNN ・エージェント """ |
そしてメインスクリプトです。
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 |
from ddpg_torch import Agent import gym import numpy as np #from utils import plotLearning env = gym.make('LunarLanderContinuous-v2') agent = Agent(alpha=0.000025, neta=0.00025, input_dims=[8], tau=0.001, env=env,bathc_size=64, layer1_size=400, layer2_size=300, n_actions=2) np.random.seed(0) score_history = [] for i in range(1000): done = Falsescore = 0 score = 0 obs =env.reset() while not done: act =agent.choose_action(obs) new_state, reward, done, info = env.step(act) agent.remenber(obs, act, reward, new_state, int(done)) agent.learn() score += reward obs = new_state score_history.appen(score) print('epsisode', i, 'score %.2f' % score, '100 game sverage %.2f' % np.mean(score_history[-100:])) if i % 25 == 0: agent.save_models() filename = ' lunar-lander.png' #plotLearning(score_history, filename, window=100) |
