forked from thu-ml/tianshou
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathmujoco_td3.py
executable file
·149 lines (135 loc) · 6.43 KB
/
mujoco_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
#!/usr/bin/env python3
import os
import gym
import torch
import pprint
import datetime
import argparse
import numpy as np
from torch.utils.tensorboard import SummaryWriter
from tianshou.policy import TD3Policy
from tianshou.utils import BasicLogger
from tianshou.env import SubprocVectorEnv
from tianshou.utils.net.common import Net
from tianshou.exploration import GaussianNoise
from tianshou.trainer import offpolicy_trainer
from tianshou.utils.net.continuous import Actor, Critic
from tianshou.data import Collector, ReplayBuffer, VectorReplayBuffer
def get_args():
parser = argparse.ArgumentParser()
parser.add_argument('--task', type=str, default='Ant-v3')
parser.add_argument('--seed', type=int, default=0)
parser.add_argument('--buffer-size', type=int, default=1000000)
parser.add_argument('--hidden-sizes', type=int, nargs='*', default=[256, 256])
parser.add_argument('--actor-lr', type=float, default=3e-4)
parser.add_argument('--critic-lr', type=float, default=3e-4)
parser.add_argument('--gamma', type=float, default=0.99)
parser.add_argument('--tau', type=float, default=0.005)
parser.add_argument('--exploration-noise', type=float, default=0.1)
parser.add_argument('--policy-noise', type=float, default=0.2)
parser.add_argument('--noise-clip', type=float, default=0.5)
parser.add_argument('--update-actor-freq', type=int, default=2)
parser.add_argument("--start-timesteps", type=int, default=25000)
parser.add_argument('--epoch', type=int, default=200)
parser.add_argument('--step-per-epoch', type=int, default=5000)
parser.add_argument('--step-per-collect', type=int, default=1)
parser.add_argument('--update-per-step', type=int, default=1)
parser.add_argument('--n-step', type=int, default=1)
parser.add_argument('--batch-size', type=int, default=256)
parser.add_argument('--training-num', type=int, default=1)
parser.add_argument('--test-num', type=int, default=10)
parser.add_argument('--logdir', type=str, default='log')
parser.add_argument('--render', type=float, default=0.)
parser.add_argument(
'--device', type=str,
default='cuda' if torch.cuda.is_available() else 'cpu')
parser.add_argument('--resume-path', type=str, default=None)
parser.add_argument('--watch', default=False, action='store_true',
help='watch the play of pre-trained policy only')
return parser.parse_args()
def test_td3(args=get_args()):
env = gym.make(args.task)
args.state_shape = env.observation_space.shape or env.observation_space.n
args.action_shape = env.action_space.shape or env.action_space.n
args.max_action = env.action_space.high[0]
args.exploration_noise = args.exploration_noise * args.max_action
args.policy_noise = args.policy_noise * args.max_action
args.noise_clip = args.noise_clip * args.max_action
print("Observations shape:", args.state_shape)
print("Actions shape:", args.action_shape)
print("Action range:", np.min(env.action_space.low),
np.max(env.action_space.high))
# train_envs = gym.make(args.task)
if args.training_num > 1:
train_envs = SubprocVectorEnv(
[lambda: gym.make(args.task) for _ in range(args.training_num)])
else:
train_envs = gym.make(args.task)
# test_envs = gym.make(args.task)
test_envs = SubprocVectorEnv(
[lambda: gym.make(args.task) for _ in range(args.test_num)])
# seed
np.random.seed(args.seed)
torch.manual_seed(args.seed)
train_envs.seed(args.seed)
test_envs.seed(args.seed)
# model
net_a = Net(args.state_shape, hidden_sizes=args.hidden_sizes, device=args.device)
actor = Actor(
net_a, args.action_shape, max_action=args.max_action,
device=args.device).to(args.device)
actor_optim = torch.optim.Adam(actor.parameters(), lr=args.actor_lr)
net_c1 = Net(args.state_shape, args.action_shape,
hidden_sizes=args.hidden_sizes,
concat=True, device=args.device)
net_c2 = Net(args.state_shape, args.action_shape,
hidden_sizes=args.hidden_sizes,
concat=True, device=args.device)
critic1 = Critic(net_c1, device=args.device).to(args.device)
critic1_optim = torch.optim.Adam(critic1.parameters(), lr=args.critic_lr)
critic2 = Critic(net_c2, device=args.device).to(args.device)
critic2_optim = torch.optim.Adam(critic2.parameters(), lr=args.critic_lr)
policy = TD3Policy(
actor, actor_optim, critic1, critic1_optim, critic2, critic2_optim,
tau=args.tau, gamma=args.gamma,
exploration_noise=GaussianNoise(sigma=args.exploration_noise),
policy_noise=args.policy_noise, update_actor_freq=args.update_actor_freq,
noise_clip=args.noise_clip, estimation_step=args.n_step,
action_space=env.action_space)
# load a previous policy
if args.resume_path:
policy.load_state_dict(torch.load(args.resume_path, map_location=args.device))
print("Loaded agent from: ", args.resume_path)
# collector
if args.training_num > 1:
buffer = VectorReplayBuffer(args.buffer_size, len(train_envs))
else:
buffer = ReplayBuffer(args.buffer_size)
train_collector = Collector(policy, train_envs, buffer, exploration_noise=True)
test_collector = Collector(policy, test_envs)
train_collector.collect(n_step=args.start_timesteps, random=True)
# log
t0 = datetime.datetime.now().strftime("%m%d_%H%M%S")
log_file = f'seed_{args.seed}_{t0}-{args.task.replace("-", "_")}_td3'
log_path = os.path.join(args.logdir, args.task, 'td3', log_file)
writer = SummaryWriter(log_path)
writer.add_text("args", str(args))
logger = BasicLogger(writer)
def save_fn(policy):
torch.save(policy.state_dict(), os.path.join(log_path, 'policy.pth'))
if not args.watch:
# trainer
result = offpolicy_trainer(
policy, train_collector, test_collector, args.epoch,
args.step_per_epoch, args.step_per_collect, args.test_num,
args.batch_size, save_fn=save_fn, logger=logger,
update_per_step=args.update_per_step, test_in_train=False)
pprint.pprint(result)
# Let's watch its performance!
policy.eval()
test_envs.seed(args.seed)
test_collector.reset()
result = test_collector.collect(n_episode=args.test_num, render=args.render)
print(f'Final reward: {result["rews"].mean()}, length: {result["lens"].mean()}')
if __name__ == '__main__':
test_td3()