Closed Jaroan closed 1 year ago
Hi! So I can't speak for other repos but generally you have freedom with the indexing as long as you correctly account for it when computing the loss. However, in this case a warning that we are confident about the implementation of SharedBuffer since we took it from our repo https://github.com/marlbenchmark/on-policy which has SOTA results in a bunch of envs but the SeparateBuffer is not extensively tested.
Thank you for the clarification!
Question
[Edit: I noticed an error in my post, but I modified my question appropriately] I notice a difference in the way
active_masks
are being copied in the ppo buffers forinsert()
andchooseinsert()
:In the
chooseinsert()
it isself.active_masks[self.step] = active_masks.copy()
line 184While in the
insert()
it isself.active_masks[self.step + 1] = active_masks.copy()
line138Thanks in advance!