Stabilize special function in fast_compile#2289
Open
ricardoV94 wants to merge 1 commit into
Open
Conversation
ricardoV94
force-pushed
the
tree1
branch
2 times, most recently
from
902da44 to
968296b
Compare
kratsg
approved these changes
Jul 11, 2026
kratsg
left a comment
kratsg
left a comment
There was a problem hiding this comment.
This seems fine to me on some level. I imagine if a user wants the graph to fast-compile without the stabilized form, they'd use the naive version? I would also probably expect something in the documentation about stabilization to mention api equivalency with scipy (re: stablization) just so that it isn't a hidden "feature"?
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes #2288
The compromise is: if the user writes code using helpers that resemble the stable scipy/numpy look-alike (e.g., pt.special.logsumexp), we make an effort to stabilize them in fast_compile (if the returned graph/op is not already the stable from, which it is for e.g., log1p).
We don't promise to stabilize the equivalent graphs written in naive form. If it happens to be stabilized today, because the naive form is the same those helpers return that's luck, not a promise.