Skip to content

Conversation

@sophie-xhonneux
Copy link
Contributor

Description

Make the bilinear layer more memory efficient.

Issue Number

Closes #1683

Draft stage, needs option for bias, and reset parameters

Checklist before asking for review

  • I have performed a self-review of my code
  • My changes comply with basic sanity checks:
    • I have fixed formatting issues with ./scripts/actions.sh lint
    • I have run unit tests with ./scripts/actions.sh unit-test
    • I have documented my code and I have updated the docstrings.
    • I have added unit tests, if relevant
  • I have tried my changes with data and code:
    • I have run the integration tests with ./scripts/actions.sh integration-test
    • (bigger changes) I have run a full training and I have written in the comment the run_id(s): launch-slurm.py --time 60
    • (bigger changes and experiments) I have shared a hegdedoc in the github issue with all the configurations and runs for this experiments
  • I have informed and aligned with people impacted by my change:
    • for config changes: the MatterMost channels and/or a design doc
    • for changes of dependencies: the MatterMost software development channel

kctezcan and others added 30 commits January 13, 2026 08:31
…ae_aggregation_engine. More checnking needed.
…iex/dev/include-reg-tokens-in-query-agg-engine
…herGenerator into sophiex/dev/include-reg-tokens-in-query-agg-engine
TODO

bias flag
reset params
clean up
return torch.cat(outputs, dim=1)


class EfficientBilinear(torch.nn.Module):
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is there a reason to retain the old implementation?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Not really!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

Status: No status

Development

Successfully merging this pull request may close these issues.

Make the bilinear layer more memory efficient

5 participants