Finalized factoring out projections

This commit is contained in:
2022-09-03 11:59:16 +02:00
parent 0aeea4e2e5
commit 4532135812
3 changed files with 2 additions and 34 deletions
+1 -2
View File
@@ -336,8 +336,7 @@ class PPO(GaussianRolloutCollectorAuxclass, OnPolicyAlgorithm):
policy_loss = surrogate_loss + self.ent_coef * entropy_loss + \
trust_region_loss + self.action_coef * action_loss
import pdb
pdb.set_trace()
pg_losses.append(policy_loss.item())
loss = policy_loss + self.vf_coef * value_loss