We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
1 parent 6074482 commit de05e0bCopy full SHA for de05e0b
1 file changed
config/four_rooms.ini
@@ -8,16 +8,16 @@ num_threads = 8
8
9
[env]
10
size = 19
11
-# 0 derives the timeout from the map size: 4 * size. Positive values override it.
+# if 0, max_steps = 4 * size. Positive values override it.
12
max_steps = 0
13
14
[policy]
15
-hidden_size = 256
16
-num_layers = 6
+hidden_size = 128
+num_layers = 2
17
expansion_factor = 1
18
19
[train]
20
-total_timesteps = 300_000_000
+total_timesteps = 100_000_000
21
gamma = 0.99
22
gae_lambda = 0.95
23
learning_rate = 0.005
0 commit comments