wassname/DeepRL
1
0
Fork 0
mirror of https://github.com/wassname/DeepRL.git synced 2026-09-10 11:40:58 +08:00
Code Issues Packages Projects Releases Wiki Activity

Minor update

Browse Source
This commit is contained in:
Shangtong Zhang
2018-04-10 23:40:45 -06:00
parent bb5ef4fbfc
commit e1df46aeec
2 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
agent/PPO_agent.py
+1 -1
View File
@@ -87,7 +87,7 @@ class PPOAgent(BaseAgent):
obj = ratio * sampled_advantages
obj_clipped = ratio.clamp(1.0 - self.config.ppo_ratio_clip,
1.0 + self.config.ppo_ratio_clip) * sampled_advantages
policy_loss = -torch.min(obj, obj_clipped).mean(0) + config.entropy_weight * entropy_loss.mean(0)
policy_loss = -torch.min(obj, obj_clipped).mean(0) + config.entropy_weight * entropy_loss
value_loss = 0.5 * (sampled_returns - values).pow(2).mean()