Fix three issues blocking HELM_MiCE from running on repo defaults - #9
Open
kaisaraka wants to merge 1 commit into
Open
Fix three issues blocking HELM_MiCE from running on repo defaults#9kaisaraka wants to merge 1 commit into
kaisaraka wants to merge 1 commit into
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Building the model from repo defaults (torch 2.11) hits three errors in helm/modules/mice.py:
moe_inter_dim undefined (line 190-191): the config defines mice_inter_dim, but the experts read args.moe_inter_dim → AttributeError: 'Namespace' object has no attribute 'moe_inter_dim'. Renamed to mice_inter_dim.
bool flag passed as int (line 189): args.train_curv is 1 (int), and geoopt's Lorentz(learnable=...) rejects it on torch 2.11 (require_grad must be bool, not int). Wrapped in bool(...).
.eval() unpack error (line 215): LorentzMoE.forward unpacks three values from Gate.forward, which returns two in eval mode → ValueError: not enough values to unpack. Made the unpack tolerant of both.
Verified the model builds and runs a forward in both train and eval mode after the fixes. These block anyone running the repo on defaults, not specific to my setup.