Make linear operators pickleable#124
Merged
gpleiss merged 3 commits intocornellius-gp:mainfrom Feb 17, 2026
Merged
Conversation
Replace lambda in defaultdict with a module-level function so that num_nonbatch_dimensions can be pickled. Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Context
This PR aims to fix cornellius-gp/gpytorch#2717
The recent changes in GPyTorch v1.15 make variational models not pickle-able anymore. After digging around, it seems that the root cause lies in linear operators, as linear operators are not pickle-able in the first place. See here for a code snippet.
This PR
Claude suggested that the issue is caused by the
@cachedecorator. By default, the cache uses the original unwrapped function in the dictionary key; see here. But pickle looks up the function by the qualified name, which points to the wrapped function. This eventually causes a check failure in pickle dumps. This PR fixes this by using the qualified name (a string) in the dictionary key.P.S. There is still a test failure in the kernel linear operator, which is likely caused by lambda functions.