Fix sigma_noise gradient propagation in BaseLaplace.log_likelihood
#212
Conversation
This file contains 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
Commit 1eea27c introduced type hints in baselaplace.py but broke the gradient flow through `BaseLaplace.log_likelihood` since `sigma_noise` loses its gradient via `torch.tensor(...)`. This is fixed by using `torch.as_tensor(...)` as it converts floats but preserves the gradient.
|
Thanks a lot for bisecting the issue and fixing it as well. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Commit 1eea27c introduces type hints in
baselaplace.pybut breaks the gradient flow throughBaseLaplace.log_likelihood(...)sincesigma_noiseloses its gradient viatorch.tensor(...).This is fixed by using
torch.as_tensor(...)as it converts floats but preserves the gradient. (See highlighted note here.)This change fixes the
UserWarningbelow and passes type checking with Pylance.