Skip to content

Conversation

@elseml
Copy link
Member

@elseml elseml commented Jun 11, 2025

Minimal fix providing missing serialization of the final logits_projector layer handled by the ModelComparisonApproximator class itself. Both the weights and the predictions of an original and a reloaded approximator of the example in #506 are now exactly equal. Please let me know if any further adaptations for other cases are needed.

As a more general discussion point: As there were some issues related to saving/loading recently, should we add some tests (e.g., using tempfiles) for the approximators that ensure that the weights and predictions of an original and a reloaded approximator match?

@codecov
Copy link

codecov bot commented Jun 11, 2025

Codecov Report

All modified and coverable lines are covered by tests ✅

Files with missing lines Coverage Δ
...low/approximators/model_comparison_approximator.py 81.81% <100.00%> (ø)

@stefanradev93
Copy link
Contributor

Thx, Lasse! This was mainly due to the way the build() function was handled before the stanradize_an_approx branch. I will make sure in a bit that the error is also resolved there.

@elseml elseml closed this Jun 12, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants