Skip to main content
costa-huang
Projects
cleanRL
Reports
sebulba

sebulba

Costa
Created on February 4|Last edited on February 4

50100150200250300Time (minutes)2000400060008000
50100150200250300Time (minutes)200040006000800010000
10M20M30M40M50Mglobal_step0100200300400500
10M20M30M40M50Mglobal_step11.21.41.61.82
10M20M30M40M50Mglobal_step-2-1012
10M20M30M40M50Mglobal_step246810
10M20M30M40M50Mglobal_step0.20.40.60.81
50100150200250300Time (minutes)0100200300400500
sebulba_ppo_envpool
1
baseline
1
sebulba_ppo_envpool (1 GPU)
1
sebulba_ppo_envpool (pmap 2GPU)
1
sebulba_ppo_envpool (lambdalabs pmap 4GPU)
1
throughput (no a-l sync)
1
sebulba_ppo_envpool_1gpu
2
sebulba_ppo_envpool_a0_l1_rollout_is_faster_timeout
1
pmap (2GPU)
1
timeout=0.02
1
device_put_replicated
1
pmap (a0-l1,2,3,4)
1
pmap (a0-l1,2,3,4) num-threads 64
1
pmap (a0-l1,2,3,4) timeout
1
pmap (a0-l1,2,3,4) timeout, warmup
1
pmap (a0-l1,2,3,4) 128 steps
1







sebulba_ppo_envpool
1
baseline
1
sebulba_ppo_envpool (1 GPU)
1
sebulba_ppo_envpool (pmap 2GPU)
1
sebulba_ppo_envpool (lambdalabs pmap 4GPU)
1
sebulba_ppo_envpool (lambdalabs pmap 2GPU)
1
sebulba_ppo_envpool (lambdalabs pmap 8GPU)
1
sebulba_ppo_envpool (lambdalabs pmap 1GPU a, 4GPU l)
1


Created with ❤️ on Weights & Biases.

https://wandb.ai/costa-huang/cleanRL/reports/sebulba--VmlldzozNDgwODU5
Made with Weights & Biases. Sign up or log in to create reports like this one.