-
Notifications
You must be signed in to change notification settings - Fork 40
/
Copy pathtest.py
70 lines (64 loc) · 1.58 KB
/
test.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
import click
from Algorithms.tf2.TRPO.trpo import TRPO
@click.command()
@click.option(
"--env_id", type=str, default="MountainCar-v0", help="Environment Id"
)
@click.option(
"--render", type=bool, default=True, help="Render environment or not"
)
@click.option(
"--num_process",
type=int,
default=1,
help="Number of process to run environment",
)
@click.option(
"--lr_v", type=float, default=3e-4, help="Learning rate for Value Net"
)
@click.option("--gamma", type=float, default=0.99, help="Discount factor")
@click.option("--tau", type=float, default=0.95, help="GAE factor")
@click.option(
"--max_kl", type=float, default=1e-2, help="kl constraint for TRPO"
)
@click.option("--damping", type=float, default=1e-2, help="damping for TRPO")
@click.option("--batch_size", type=int, default=1000, help="Batch size")
@click.option(
"--model_path",
type=str,
default="trained_models",
help="Directory to store model",
)
@click.option("--seed", type=int, default=1, help="Seed for reproducing")
@click.option("--test_epochs", type=int, default=1, help="Epochs for testing")
def main(
env_id,
render,
num_process,
lr_v,
gamma,
tau,
max_kl,
damping,
batch_size,
model_path,
seed,
test_epochs,
):
trpo = TRPO(
env_id,
render,
num_process,
batch_size,
lr_v,
gamma,
tau,
max_kl,
damping,
seed=seed,
model_path=model_path,
)
for i_iter in range(1, test_epochs):
trpo.eval(i_iter)
if __name__ == "__main__":
main()