-
Notifications
You must be signed in to change notification settings - Fork 15
/
td3.py
430 lines (361 loc) · 18.1 KB
/
td3.py
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
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
import os
import torch as T
#import torch.cuda as T
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
import numpy as np
from data_manager import DataManager
class OUActionNoise(object):
def __init__(self, mu, sigma=0.15, theta=.2, dt=1e-2, x0=None):
self.theta = theta
self.mu = mu
self.sigma = sigma
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)
def __repr__(self):
return 'OrnsteinUhlenbeckActionNoise(mu={}, sigma={})'.format(
self.mu, self.sigma)
class AWGNActionNoise(object):
def __init__(self, mu = 0, sigma=1):
self.mu = mu
self.sigma = sigma
def __call__(self):
#self.mu = mu
#self.sigma = sigma
x = np.random.normal(size=self.mu.shape) * self.sigma
return x
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 store_transition(self, state, action, reward, state_, done):
index = self.mem_cntr % self.mem_size
self.state_memory[index] = state
self.new_state_memory[index] = state_
self.action_memory[index] = action
self.reward_memory[index] = reward
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]
actions = self.action_memory[batch]
rewards = self.reward_memory[batch]
states_ = self.new_state_memory[batch]
terminal = self.terminal_memory[batch]
return states, actions, rewards, states_, terminal
class CriticNetwork(nn.Module):
def __init__(self, beta, input_dims, fc1_dims, fc2_dims, fc3_dims, fc4_dims, n_actions, name,
chkpt_dir='C:\\demo\\IRS_TD3_minimal\\main_foder\\tmp\\TD3', load_file = ''):
super(CriticNetwork, self).__init__()
self.input_dims = input_dims
self.fc1_dims = fc1_dims
self.fc2_dims = fc2_dims
self.fc3_dims = fc3_dims
self.fc4_dims = fc4_dims
self.n_actions = n_actions
self.checkpoint_file = os.path.join(chkpt_dir,name+'_TD3')
self.load_file = 'C:\\demo\\other_branch\\Learning-based_Secure_Transmission_for_RIS_Aided_mmWave-UAV_Communications_with_Imperfect_CSI\\data\\mannal_store\\models\\Critic_UAV_TD3'
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.fc1.weight.data.uniform_(-f1, f1)
#self.fc1.bias.data.uniform_(-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])
#f2 = 0.002
T.nn.init.uniform_(self.fc2.weight.data, -f2, f2)
T.nn.init.uniform_(self.fc2.bias.data, -f2, f2)
#self.fc2.weight.data.uniform_(-f2, f2)
#self.fc2.bias.data.uniform_(-f2, f2)
self.bn2 = nn.LayerNorm(self.fc2_dims)
self.fc3 = nn.Linear(self.fc2_dims, self.fc3_dims)
f3 = 1./np.sqrt(self.fc3.weight.data.size()[0])
T.nn.init.uniform_(self.fc3.weight.data, -f3, f3)
T.nn.init.uniform_(self.fc3.bias.data, -f3, f3)
#self.fc2.weight.data.uniform_(-f2, f2)
#self.fc2.bias.data.uniform_(-f2, f2)
self.bn3 = nn.LayerNorm(self.fc3_dims)
self.fc4 = nn.Linear(self.fc3_dims, self.fc4_dims)
f4 = 1./np.sqrt(self.fc4.weight.data.size()[0])
T.nn.init.uniform_(self.fc4.weight.data, -f4, f4)
T.nn.init.uniform_(self.fc4.bias.data, -f4, f4)
#self.fc2.weight.data.uniform_(-f2, f2)
#self.fc2.bias.data.uniform_(-f2, f2)
self.bn4 = nn.LayerNorm(self.fc4_dims)
self.action_value = nn.Linear(self.n_actions, self.fc4_dims)
f5 = 0.003
self.q = nn.Linear(self.fc4_dims, 1)
T.nn.init.uniform_(self.q.weight.data, -f5, f5)
T.nn.init.uniform_(self.q.bias.data, -f5, f5)
#self.q.weight.data.uniform_(-f3, f3)
#self.q.bias.data.uniform_(-f3, f3)
self.optimizer = optim.Adam(self.parameters(), lr=beta)
# if torch.cuda.available():
# import torch.cuda as T
# else:
# import torch as T
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)
state_value = F.relu(state_value)
state_value = self.fc3(state_value)
state_value = self.bn3(state_value)
state_value = F.relu(state_value)
state_value = self.fc4(state_value)
state_value = self.bn4(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,load_file = ''):
print('... loading checkpoint ...')
self.load_state_dict(T.load(load_file))
class ActorNetwork(nn.Module):
def __init__(self, alpha, input_dims, fc1_dims, fc2_dims, fc3_dims, fc4_dims, n_actions, name,
chkpt_dir='C:\\demo\\IRS_TD3_minimal\\main_foder\\tmp\\TD3', load_file = ''):
super(ActorNetwork, self).__init__()
self.input_dims = input_dims
self.fc1_dims = fc1_dims
self.fc2_dims = fc2_dims
self.fc3_dims = fc3_dims
self.fc4_dims = fc4_dims
self.n_actions = n_actions
self.checkpoint_file = os.path.join(chkpt_dir,name+'_TD3')
self.load_file = 'C:\\demo\\other_branch\\Learning-based_Secure_Transmission_for_RIS_Aided_mmWave-UAV_Communications_with_Imperfect_CSI\\data\\mannal_store\\models\\Actor_UAV_TD3'
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.fc1.weight.data.uniform_(-f1, f1)
self.fc1.bias.data.uniform_(-f1, f1)
self.bn1 = nn.LayerNorm(self.fc1_dims)
self.fc2 = nn.Linear(self.fc1_dims, self.fc2_dims)
#f2 = 0.002
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)
self.fc2.weight.data.uniform_(-f2, f2)
self.fc2.bias.data.uniform_(-f2, f2)
self.bn2 = nn.LayerNorm(self.fc2_dims)
self.fc3 = nn.Linear(self.fc2_dims, self.fc3_dims)
f3 = 1./np.sqrt(self.fc3.weight.data.size()[0])
# T.nn.init.uniform_(self.fc3.weight.data, -f3, f3)
# T.nn.init.uniform_(self.fc3.bias.data, -f3, f3)
self.fc2.weight.data.uniform_(-f2, f2)
self.fc2.bias.data.uniform_(-f2, f2)
self.bn3 = nn.LayerNorm(self.fc3_dims)
self.fc4 = nn.Linear(self.fc3_dims, self.fc4_dims)
f4 = 1./np.sqrt(self.fc4.weight.data.size()[0])
# T.nn.init.uniform_(self.fc4.weight.data, -f4, f4)
# T.nn.init.uniform_(self.fc4.bias.data, -f4, f4)
self.fc2.weight.data.uniform_(-f2, f2)
self.fc2.bias.data.uniform_(-f2, f2)
self.bn4 = nn.LayerNorm(self.fc4_dims)
#f3 = 0.004
f5 = 0.003
self.mu = nn.Linear(self.fc4_dims, self.n_actions)
# T.nn.init.uniform_(self.mu.weight.data, -f5, f5)
# T.nn.init.uniform_(self.mu.bias.data, -f5, f5)
self.mu.weight.data.uniform_(-f3, f3)
self.mu.bias.data.uniform_(-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 = F.relu(x)
x = self.fc2(x)
x = self.bn2(x)
x = F.relu(x)
x = self.fc3(x)
x = self.bn3(x)
x = F.relu(x)
x = self.fc4(x)
x = self.bn4(x)
x = F.relu(x)
x = T.tanh(self.mu(x))
return x
def save_checkpoint(self):
print('... saving checkpoint ...')
T.save(self.state_dict(), self.checkpoint_file)
def load_checkpoint(self, load_file=''):
print('... loading checkpoint ...')
self.load_state_dict(T.load(load_file))
class Agent(object):
def __init__(self, alpha, beta, input_dims, tau, env, gamma=0.99,
n_actions=2, max_size=1000000, layer1_size=400,
layer2_size=300, layer3_size=256, layer4_size=128, batch_size=64,
update_actor_interval=2, noise = 'AWGN', agent_name = 'default', load_file = ''):
self.load_file = load_file
self.layer1_size = layer1_size
self.layer2_size = layer2_size
self.layer3_size = layer3_size
self.layer4_size = layer4_size
self.gamma = gamma
self.tau = tau
self.memory = ReplayBuffer(max_size, input_dims, n_actions)
self.batch_size = batch_size
self.learn_step_cntr = 0
self.update_actor_iter = update_actor_interval
self.actor = ActorNetwork(alpha, input_dims, layer1_size,
layer2_size, layer3_size, layer4_size, n_actions=n_actions,
name='Actor_' + agent_name,chkpt_dir=env.data_manager.store_path )
self.critic_1 = CriticNetwork(beta, input_dims, layer1_size,
layer2_size, layer3_size, layer4_size, n_actions=n_actions,
name='Critic_1_' + agent_name,chkpt_dir=env.data_manager.store_path)
self.critic_2 = CriticNetwork(beta, input_dims, layer1_size,
layer2_size, layer3_size, layer4_size, n_actions=n_actions,
name='Critic_2_' + agent_name,chkpt_dir=env.data_manager.store_path)
self.target_actor = ActorNetwork(alpha, input_dims, layer1_size,
layer2_size, layer3_size, layer4_size, n_actions=n_actions,
name='TargetActor_' + agent_name,chkpt_dir=env.data_manager.store_path)
self.target_critic_1 = CriticNetwork(beta, input_dims, layer1_size,
layer2_size, layer3_size, layer4_size, n_actions=n_actions,
name='TargetCritic_1_' + agent_name,chkpt_dir=env.data_manager.store_path)
self.target_critic_2 = CriticNetwork(beta, input_dims, layer1_size,
layer2_size, layer3_size, layer4_size, n_actions=n_actions,
name='TargetCritic_2_' + agent_name,chkpt_dir=env.data_manager.store_path)
if noise == 'OU':
self.noise = OUActionNoise(mu=np.zeros(n_actions))
elif noise == 'AWGN':
self.noise = AWGNActionNoise(mu = np.zeros(n_actions))
# tau = 1 means copy parameters to target
self.update_network_parameters(tau=1)
def choose_action(self, observation, greedy=0.5, epsilon = 1):
self.actor.eval()
observation = T.tensor(observation, dtype=T.float).to(self.actor.device)
mu = self.actor.forward(observation).to(self.actor.device)
mu_prime = mu + T.tensor(greedy * self.noise(),
dtype=T.float).to(self.actor.device)
self.actor.train()
return mu_prime.cpu().detach().numpy()
def remember(self, state, action, reward, new_state, done):
self.memory.store_transition(state, action, reward, new_state, done)
def learn(self):
if self.memory.mem_cntr < self.batch_size:
return
# the done here is opposite of the done in the env
state, action, reward, new_state, done = \
self.memory.sample_buffer(self.batch_size)
# trun s, a, r, new_s into tensor
reward = T.tensor(reward, dtype=T.float).to(self.critic_1.device)
done = T.tensor(done).to(self.critic_1.device)
new_state = T.tensor(new_state, dtype=T.float).to(self.critic_1.device)
action = T.tensor(action, dtype=T.float).to(self.critic_1.device)
state = T.tensor(state, dtype=T.float).to(self.critic_1.device)
# trun on evaliation mode of target actor, target critic, critic net
# fix these three nets
self.target_actor.eval()
self.target_critic_1.eval()
self.target_critic_2.eval()
self.critic_1.eval()
self.critic_2.eval()
target_actions = self.target_actor.forward(new_state)
# target_actions = target_actions + \
# T.clamp(T.tensor(np.random.normal(scale=0.2)), -0.5, 0.5)
# # might break if elements of min and max are not all equal
# target_actions = T.clamp(target_actions, self.min_action[0], self.max_action[0])
critic_value_1_ = self.target_critic_1.forward(new_state, target_actions)
critic_value_2_ = self.target_critic_2.forward(new_state, target_actions)
critic_value_1 = self.critic_1.forward(state, action)
critic_value_2 = self.critic_2.forward(state, action)
critic_value_ = T.min(critic_value_1_, critic_value_2_)
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_1.device)
target = target.view(self.batch_size, 1)
# here update the critic net using mse of (r + gamma * Q_argmax_a*(newstate, a*)) - Q(state, action)
self.critic_1.train()
self.critic_2.train()
self.critic_1.optimizer.zero_grad()
self.critic_2.optimizer.zero_grad()
critic_1_loss = F.mse_loss(target, critic_value_1)
critic_2_loss = F.mse_loss(target, critic_value_2)
critic_loss = critic_1_loss + critic_2_loss
critic_loss.backward()
self.critic_1.optimizer.step()
self.critic_2.optimizer.step()
self.learn_step_cntr += 1
if self.learn_step_cntr % self.update_actor_iter != 0:
return
# here update the actor net by policy gradient
# first fix the critic net
self.critic_1.eval()
self.critic_2.eval()
self.actor.optimizer.zero_grad()
mu = self.actor.forward(state)
self.actor.train()
actor_q1_loss = self.critic_1.forward(state, mu)
actor_loss = -T.mean(actor_q1_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_1_params = self.critic_1.named_parameters()
critic_2_params = self.critic_2.named_parameters()
target_actor_params = self.target_actor.named_parameters()
target_critic_1_params = self.target_critic_1.named_parameters()
target_critic_2_params = self.target_critic_2.named_parameters()
critic_1_state_dict = dict(critic_1_params)
critic_2_state_dict = dict(critic_2_params)
actor_state_dict = dict(actor_params)
target_actor_state_dict = dict(target_actor_params)
target_critic_1_state_dict = dict(target_critic_1_params)
target_critic_2_state_dict = dict(target_critic_2_params)
for name in critic_1_state_dict:
critic_1_state_dict[name] = tau*critic_1_state_dict[name].clone() + \
(1-tau)*target_critic_1_state_dict[name].clone()
for name in critic_2_state_dict:
critic_2_state_dict[name] = tau*critic_2_state_dict[name].clone() + \
(1-tau)*target_critic_2_state_dict[name].clone()
for name in actor_state_dict:
actor_state_dict[name] = tau*actor_state_dict[name].clone() + \
(1-tau)*target_actor_state_dict[name].clone()
self.target_critic_1.load_state_dict(critic_1_state_dict)
self.target_critic_2.load_state_dict(critic_2_state_dict)
self.target_actor.load_state_dict(actor_state_dict)
def save_models(self):
self.actor.save_checkpoint()
self.target_actor.save_checkpoint()
self.critic_1.save_checkpoint()
self.critic_2.save_checkpoint()
self.target_critic_1.save_checkpoint()
self.target_critic_2.save_checkpoint()
def load_models(self, load_file_actor = '',load_file_critic_1 ='',load_file_critic_2 =''):
self.actor.load_checkpoint(load_file = load_file_actor)
self.target_actor.load_checkpoint(load_file = load_file_actor)
self.critic_1.load_checkpoint(load_file = load_file_critic_1)
self.critic_2.load_checkpoint(load_file = load_file_critic_2)
self.target_critic_1.load_checkpoint(load_file = load_file_critic_1)
self.target_critic_2.load_checkpoint(load_file = load_file_critic_2)