Skip to content

Commit 28eb484

Browse files
authored
Merge branch 'PufferAI:3.0' into vision_test
2 parents c0996dd + 9f56612 commit 28eb484

File tree

9 files changed

+879
-114
lines changed

9 files changed

+879
-114
lines changed

pufferlib/config/nethack.ini

Lines changed: 22 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -1,16 +1,31 @@
11
[base]
22
package = nethack
33
env_name = nethack
4+
policy_name = Policy
5+
rnn_name = Recurrent
46

57
[vec]
6-
num_envs = 128
8+
num_envs = 8192
79
num_workers = 16
8-
batch_size = 64
10+
batch_size = 4096
911

1012
[train]
11-
batch_size = 8192
12-
minibatch_size = 2048
13-
update_epochs = 1
13+
total_timesteps = 90_000_000
14+
adam_beta1 = 0.8946507418260217
15+
adam_beta2 = 0.9
16+
adam_eps = 0.0001
17+
batch_size = auto
1418
bptt_horizon = 64
15-
total_timesteps = 10_000_000
16-
anneal_lr = False
19+
clip_coef = 0.19696765958267629
20+
ent_coef = 0.0005690816545012474
21+
gae_lambda = 0.747650023961198
22+
gamma = 0.9997053654668936
23+
learning_rate = 0.044482546441415506
24+
max_grad_norm = 2.2356112188495723
25+
minibatch_size = 32768
26+
prio_alpha = 0.98967001208896
27+
prio_beta0 = 0.09999999999999998
28+
vf_clip_coef = 2.178492167689251
29+
vf_coef = 1.6832989594296321
30+
vtrace_c_clip = 2.878171091654008
31+
vtrace_rho_clip = 0.7876748061547312

pufferlib/config/ocean/whisker_racer.ini

Lines changed: 11 additions & 10 deletions
Original file line numberDiff line numberDiff line change
@@ -10,31 +10,32 @@ num_envs = 8
1010
[env]
1111
num_envs = 1024
1212
frameskip = 4
13-
width = 1920
14-
height = 1080
15-
track_width = 50
13+
width = 1080
14+
height = 720
15+
track_width = 75
1616
num_radial_sectors = 180
1717
num_points = 16
1818
bezier_resolution = 4
1919
turn_pi_frac = 40
20-
w_ang = 0.586 # 0.523
20+
w_ang = 0.777 # 0.586 # 0.523
2121
reward_yellow = 0.2
2222
reward_green = -0.001
2323
corner_thresh = 0.5 # dot product for hairpins
2424
ftmp1 = 0.5 #0.9
2525
ftmp2 = 3.0 #1.05
2626
ftmp3 = 0.3 # 0.2
2727
ftmp4 = 0.0
28-
render_many = 1
28+
mode7 = 0
29+
render_many = 0
2930
rng = 6
30-
method = -1
31+
method = 2
3132

3233
[policy]
33-
hidden_size = 64
34+
hidden_size = 128
3435

3536
[rnn]
36-
input_size = 64
37-
hidden_size = 64
37+
input_size = 128
38+
hidden_size = 128
3839

3940
[train]
4041
adam_beta1 = 0.9446160612709289
@@ -53,7 +54,7 @@ max_grad_norm = 1.1402446026380597
5354
minibatch_size = 32768 #16384 32768
5455
prio_alpha = 0.8186786991771018
5556
prio_beta0 = 0.49639773186725333
56-
total_timesteps = 20_000_000
57+
total_timesteps = 200_000_000
5758
vf_clip_coef = 1.1492700894337171
5859
vf_coef = 1.2551354745222134
5960
vtrace_c_clip = 0

0 commit comments

Comments
 (0)