Skip to main content

Procgen: CleanRL's PPG vs PPO vs openai/phasic-policy-gradient

Created on May 27|Last edited on May 27

videos
This run didn't log media for key "videos", step 9879, index 0. Docs →
This run didn't log media for key "videos", step 5887, index 0. Docs →
This run didn't log media for key "videos", step 6069, index 0. Docs →
This run didn't log media for key "videos", step 6050, index 0. Docs →
This run didn't log media for key "videos", step 5750, index 0. Docs →
This run didn't log media for key "videos", step 5928, index 0. Docs →
5M10M15M20M25MStep020406080Episodic Return
CleanRL's ppg_procgen.py
CleanRL's ppo_procgen.py
openai/phasic-policy-gradient
CleanRL's ppg_procgen.py
3
CleanRL's ppo_procgen.py
6
openai/phasic-policy-gradient
3
Name
3 visualized
3
State
Notes
User
Tags
Created
Runtime
Sweep
arch
clip_param
distribution_mode
env_name
interacts_total
kl_penalty
n_aux_epochs
n_epoch_pi
n_epoch_vf
n_pi
num_envs
track
wandb_entity
wandb_project_name
charts/episodic_length
charts/episodic_return
global_step
Finished
dipamc77
4h 19m 52s
-
shared
0.2
easy
starpilot
25000000
0
6
1
1
32
64
true
openrlbenchmark
phasic-policy-gradient
514
46
25001024
1-1
of 1



CleanRL's ppg_procgen.py
3
CleanRL's ppo_procgen.py
6
openai/phasic-policy-gradient
3



CleanRL's ppg_procgen.py
3
CleanRL's ppo_procgen.py
6
openai/phasic-policy-gradient
3