-
Notifications
You must be signed in to change notification settings - Fork 0
/
arguments.py
86 lines (54 loc) · 3.18 KB
/
arguments.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
import argparse
import os
parser = argparse.ArgumentParser(description='Train, test or visualize an agent to solve the platform environment')
parser.add_argument("--task", type=str,default="all",choices=["all","train","test","visu"])
parser.add_argument("--agent_name",type=str,choices=["MH_DQN","MH_TD3"], default="MH_DQN")
parser.add_argument("--buffer_size", type=int, default=10000,
help="maximum size of buffer")
parser.add_argument("--batch_size", type=int, default=128,
help="size of each training batch")
parser.add_argument("--random_seed", type=int, default=128)
parser.add_argument("--hidden_sizes_actor", type=int, default=[128, 64], nargs='*',
help="list of hidden layers sizes of the actor network")
parser.add_argument("--lr_actor", type=float, default=1e-4,
help="the actor leraning rate")
parser.add_argument("--tau_actor", type=float, default=0.001,
help="the target actor update rate")
parser.add_argument("--hidden_sizes_critic", type=int, default=[128, 64], nargs='*',
help="list of hidden layers sizes of the critic network")
parser.add_argument("--lr_critic", type=float, default=1e-3,
help="the critic leraning rate")
parser.add_argument("--tau_critic", type=float, default=0.1,
help="the target critic update rate")
parser.add_argument("--grad_max", type=float, default=10.,
help="max norm of the gradients")
parser.add_argument("--device",type=str,choices=["cpu","cuda"],default="cuda")
parser.add_argument("--update_every", type=int, default=1,
help="update frequency")
parser.add_argument("--gamma", type=float, default=0.99,
help="discount factor")
parser.add_argument("--path",type=str,default=None,
help="path to save/load models")
parser.add_argument("--OUN",type=bool,default=True,
help="if True we use Ornstein-uhlenbeck noise else uniform sampling")
parser.add_argument("--d", type=int, default=16,
help="update frequency of the actor and the targets")
parser.add_argument("--n_episodes", type=int, default=20000,
help="number of episode to train on")
parser.add_argument("--max_t", type=int, default=1000,
help="number of instant per episode")
parser.add_argument("--maxlen", type=int, default=1000,
help="max number of episode to computer the average")
parser.add_argument("--eps_start", type=float, default=1.,
help="initial epsilon")
parser.add_argument("--eps_end", type=float, default=0.01,
help="minmum epsilon")
parser.add_argument("--eps_decay", type=float, default=0.995,
help="decrease rate")
parser.add_argument("--use_gamma",type=bool,default=False,
help="if True evaluate episode score unsing agent.gamma else gamma=1")
parser.add_argument("--n_episodes_test", type=int, default=5000,
help="number of episode to test on")
args=parser.parse_args()
if args.path is None:
args.path=os.path.join("saved",args.agent_name)