hongshaorou/dreamerv3-torch
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge pull request #81 from Ruicky8/patch-1

Browse Source 
Fix missing advantage computation when reward_EMA is disabled
This commit is contained in:
Naoki Morihira
2026-03-08 13:31:20 +09:00
committed by GitHub
parent 6253f988fe a8456e95bc
commit 6ef8646d80
1 changed files with 2 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
models.py
View File

@@ -409,6 +409,8 @@ class ImagBehavior(nn.Module):
metrics.update(tools.tensorstats(normed_target, "normed_target")) metrics.update(tools.tensorstats(normed_target, "normed_target"))
metrics["EMA_005"] = to_np(self.ema_vals[0]) metrics["EMA_005"] = to_np(self.ema_vals[0])
metrics["EMA_095"] = to_np(self.ema_vals[1]) metrics["EMA_095"] = to_np(self.ema_vals[1])
else:
adv = target - base
if self._config.imag_gradient == "dynamics": if self._config.imag_gradient == "dynamics":
actor_target = adv actor_target = adv