Do we really need actor and learner devices
Created on February 25|Last edited on February 26
Comment
meta
18m 1s
38m 9s
38m 15s
38m 1s
config
15
20
20
20
1
3
3
3
["gpu:0","gpu:1","gpu:2","gpu:3","gpu:4","gpu:5","gpu:6","gpu:7"]
["gpu:1","gpu:2","gpu:3","gpu:5","gpu:6","gpu:7"]
["gpu:1","gpu:2","gpu:3","gpu:5","gpu:6","gpu:7"]
["gpu:1","gpu:2","gpu:3","gpu:5","gpu:6","gpu:7"]
cleanrl
cleanrl
cleanrl
[0]
[1,2,3]
[1,2,3]
[1,2,3]
["gpu:0"]
["gpu:1","gpu:2","gpu:3"]
["gpu:1","gpu:2","gpu:3"]
["gpu:1","gpu:2","gpu:3"]
1920
7680
7680
7680
480
1920
1920
1920
15
60
60
60
false
true
true
true
1
2
3
1
false
true
true
true
8
2
2
2
summary
_wandb
-
2278
2284
-
charts
4900.1333
11641.2168
10197.4668
8842.91699
644.53333
729.65002
804.73334
771.11664
0.00012573
7.6815e-8
7.6815e-8
1.5362e-7
23308
25854
25856
25739
24465
25308
26763
25877
eval
-
864
864
-
losses
0.0015253
1.0240e-7
9.3194e-8
1.1383e-7
1.06023
1.08573
1.11612
1.10855
0.041031
0.027367
0.01293
0.0028086
Run set
1
Run set 2
3
Add a comment