dodox/metastable-baselines
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Split PPO_SDE into PPO_BASE_SDE and PPO_LATENT_SDE

Browse Source
This commit is contained in:
Dominik Moritz Roth 2022-06-19 22:47:04 +02:00
parent c0a5b6650d
commit b9303416ac
1 changed files with 15 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
test.py
View File

@ -28,11 +28,22 @@ def main():
ent_coef=0.0001, ent_coef=0.0001,
learning_rate=0.0004 learning_rate=0.0004
) )
ppo_sde = PPO( ppo_base_sde = PPO(
"MlpPolicy", "MlpPolicy",
env, env,
verbose=1, verbose=1,
tensorboard_log="./logs_tb/test/ppo_sde/", tensorboard_log="./logs_tb/test/ppo_base_sde/",
use_sde=True,
sde_sample_freq=30*20,
sde_net_arch=[],
ent_coef=0.000001,
learning_rate=0.0003
)
ppo_latent_sde = PPO(
"MlpPolicy",
env,
verbose=1,
tensorboard_log="./logs_tb/test/ppo_latent_sde/",
use_sde=True, use_sde=True,
sde_sample_freq=30*20, sde_sample_freq=30*20,
ent_coef=0.000001, ent_coef=0.000001,
@ -53,8 +64,8 @@ def main():
#print('PPO:') #print('PPO:')
#testModel(ppo, 500000, showRes = True, saveModel=True, n_eval_episodes=4) #testModel(ppo, 500000, showRes = True, saveModel=True, n_eval_episodes=4)
print('PPO_SDE:') print('PPO_BASE_SDE:')
testModel(ppo_sde, 100000, showRes = True, saveModel=True, n_eval_episodes=0) testModel(ppo_base_sde, 200000, showRes = True, saveModel=True, n_eval_episodes=0)
#print('A2C:') #print('A2C:')
#testModel(a2c, showRes = True) #testModel(a2c, showRes = True)
#print('TRL_PG:') #print('TRL_PG:')