-
Notifications
You must be signed in to change notification settings - Fork 0
/
main_2.py
244 lines (194 loc) · 9.61 KB
/
main_2.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
import argparse
import gym
import numpy as np
import os
import torch
import pickle
import BCQ
import DDPG
import utils
# Handles interactions with the environment, i.e. train behavioral or generate buffer
def interact_with_environment(env, state_dim, action_dim, max_action, device, args):
# For saving files
setting = f"{args.env}_{args.seed}"
buffer_name = f"{args.buffer_name}_{setting}"
# Initialize and load policy
policy = DDPG.DDPG(state_dim, action_dim, max_action, device) # , args.discount, args.tau)
if args.generate_buffer: policy.load(f"./models/behavioral_{setting}")
# Initialize buffer
replay_buffer = utils.ReplayBuffer(state_dim, action_dim, device)
evaluations = []
state, done = env.reset(), False
episode_reward = 0
episode_timesteps = 0
episode_num = 0
# Interact with the environment for max_timesteps
for t in range(int(args.max_timesteps)):
episode_timesteps += 1
# Select action with noise
if (
(args.generate_buffer and np.random.uniform(0, 1) < args.rand_action_p) or
(args.train_behavioral and t < args.start_timesteps)
):
action = env.action_space.sample()
else:
action = (
policy.select_action(np.array(state))
+ np.random.normal(0, max_action * args.gaussian_std, size=action_dim)
).clip(-max_action, max_action)
# Perform action
next_state, reward, done, _ = env.step(action)
done_bool = float(done) if episode_timesteps < env._max_episode_steps else 0
# Store data in replay buffer
replay_buffer.add(state, action, next_state, reward, done_bool)
state = next_state
episode_reward += reward
# Train agent after collecting sufficient data
if args.train_behavioral and t >= args.start_timesteps:
policy.train(replay_buffer, args.batch_size)
if done:
# +1 to account for 0 indexing. +0 on ep_timesteps since it will increment +1 even if done=True
print(
f"Total T: {t + 1} Episode Num: {episode_num + 1} Episode T: {episode_timesteps} Reward: {episode_reward:.3f}")
# Reset environment
state, done = env.reset(), False
episode_reward = 0
episode_timesteps = 0
episode_num += 1
# Evaluate episode
if args.train_behavioral and (t + 1) % args.eval_freq == 0:
evaluations.append(eval_policy(policy, args.env, args.seed))
np.save(f"./results/behavioral_{setting}", evaluations)
policy.save(f"./models/behavioral_{setting}")
# Save final policy
if args.train_behavioral:
policy.save(f"./models/behavioral_{setting}")
# Save final buffer and performance
else:
evaluations.append(eval_policy(policy, args.env, args.seed))
np.save(f"./results/buffer_performance_{setting}", evaluations)
replay_buffer.save(f"./buffers/{buffer_name}")
# Trains BCQ offline
def train_BCQ(state_dim, action_dim, max_action, device, args):
# For saving files
setting = f"{args.env}_{args.seed}"
buffer_name = f"{args.buffer_name}_{setting}"
# Initialize policy
policy = BCQ.BCQ(state_dim, action_dim, max_action, device, args.discount, args.tau, args.lmbda, args.phi)
# Load buffer
replay_buffer = utils.ReplayBuffer(state_dim, action_dim, device)
replay_buffer.load(f"./buffers/{buffer_name}")
evaluations = []
episode_num = 0
done = True
training_iters = 0
while training_iters < args.max_timesteps:
pol_vals = policy.train(replay_buffer, iterations=int(args.eval_freq), batch_size=args.batch_size)
# evaluations.append(eval_policy(policy, args.env, args.seed))
# np.save(f"./results/BCQ_{setting}", evaluations)
training_iters += args.eval_freq
print(f"Training iterations: {training_iters}")
if training_iters >= args.max_timesteps:
torch.save(policy.actor, 'trained_actor_na')
torch.save(policy.critic, 'trained_critic_na')
torch.save(policy.vae, 'trained_vae_na')
# Runs policy for X episodes and returns average reward
# A fixed seed is used for the eval environment
def eval_policy(policy, env_name, seed, eval_episodes=10):
eval_env = gym.make(env_name)
eval_env.seed(seed + 100)
avg_reward = 0.
for _ in range(eval_episodes):
state, done = eval_env.reset(), False
while not done:
action = policy.select_action(np.array(state))
state, reward, done, _ = eval_env.step(action)
avg_reward += reward
avg_reward /= eval_episodes
print("---------------------------------------")
print(f"Evaluation over {eval_episodes} episodes: {avg_reward:.3f}")
print("---------------------------------------")
return avg_reward
class InputObject:
def __init__(self, buffer_name):
infile = open(buffer_name, 'rb')
self.dict = pickle.load(infile)
infile.close()
self.buffer_name = buffer_name.replace('.pickle', '', 1)
self.buffer = self.dict[2]
self.env = "" # useless
self.seed = 0
self.eval_freq = 5e3
self.max_timesteps = 50000
self.buffer_size = len(self.buffer)
self.start_timesteps = 200 # useless
self.rand_action_p = 0.01 # useless
self.gaussian_std = 0.01 # useless
self.batch_size = 30 # mini batch size for the NNs
self.discount = 0.99
self.tau = 0.005 # target network update rate
self.lmbda = 0.75 # Weighting for clipped double Q-learning in BCQ
self.phi = 0.05
self.train_behavioral = False
self.generate_buffer = False
if __name__ == "__main__":
"""Load the buffer"""
# infile = open('Helsinki_TD3_GroupMask_site04[0][A]_2021-05-08.pickle', 'rb')
# new_dict = pickle.load(infile)
# infile.close()
args = InputObject('Helsinki_TD3_GroupMask_site04[0][A]_2021-05-08.pickle')
state_batch = [args.dict[2][x][0] for x in range(args.buffer_size)]
action_batch = [args.dict[2][x][1] for x in range(args.buffer_size)]
reward_batch = [args.dict[2][x][2] for x in range(args.buffer_size)]
next_state_batch = [args.dict[2][x][3] for x in range(args.buffer_size)]
# parser = argparse.ArgumentParser()
# parser.add_argument("--env", default="Hopper-v3") # OpenAI gym environment name
# parser.add_argument("--seed", default=0, type=int) # Sets Gym, PyTorch and Numpy seeds
# parser.add_argument("--buffer_name", default="Robust") # Prepends name to filename
# parser.add_argument("--eval_freq", default=5e3, type=float) # How often (time steps) we evaluate
# parser.add_argument("--max_timesteps", default=1e6,
# type=int) # Max time steps to run environment or train for (this defines buffer size)
# parser.add_argument("--start_timesteps", default=25e3,
# type=int) # Time steps initial random policy is used before training behavioral
# parser.add_argument("--rand_action_p", default=0.3,
# type=float) # Probability of selecting random action during batch generation
# parser.add_argument("--gaussian_std", default=0.3,
# type=float) # Std of Gaussian exploration noise (Set to 0.1 if DDPG trains poorly)
# parser.add_argument("--batch_size", default=100, type=int) # Mini batch size for networks
# parser.add_argument("--discount", default=0.99) # Discount factor
# parser.add_argument("--tau", default=0.005) # Target network update rate
# parser.add_argument("--lmbda", default=0.75) # Weighting for clipped double Q-learning in BCQ
# parser.add_argument("--phi", default=0.05) # Max perturbation hyper-parameter for BCQ
# parser.add_argument("--train_behavioral", action="store_true") # If true, train behavioral (DDPG)
# parser.add_argument("--generate_buffer", action="store_true") # If true, generate buffer
# args = parser.parse_args()
print("---------------------------------------")
if args.train_behavioral:
print(f"Setting: Training behavioral, Env: {args.env}, Seed: {args.seed}")
elif args.generate_buffer:
print(f"Setting: Generating buffer, Env: {args.env}, Seed: {args.seed}")
else:
print(f"Setting: Training BCQ, Env: {args.env}, Seed: {args.seed}")
print("---------------------------------------")
if args.train_behavioral and args.generate_buffer:
print("Train_behavioral and generate_buffer cannot both be true.")
exit()
if not os.path.exists("./results"):
os.makedirs("./results")
if not os.path.exists("./models"):
os.makedirs("./models")
if not os.path.exists("./buffers"):
os.makedirs("./buffers")
# env = gym.make(args.env)
# env.seed(args.seed)
# env.action_space.seed(args.seed)
torch.manual_seed(args.seed)
np.random.seed(args.seed)
state_dim = len(state_batch[0])
action_dim = len(action_batch[0])
max_action = 6
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
# if args.train_behavioral or args.generate_buffer:
# interact_with_environment(env, state_dim, action_dim, max_action, device, args)
# else: # the next line should be indented if the if-else structure is activated
train_BCQ(state_dim, action_dim, max_action, device, args)