Map multiple names of the same object to the same wrapped object #716
Conversation
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
Avoid creating two different wrappers for the same object in wrap_namespace. Generally cleaner, but more importantly towards better ufuncs support, which will require mapping from ufuncs to their wrapped version. Another minor pro is that this theoretically avoids needing to define grads for two names of the same object, e.g., we could define the VJP for anp.conj but not for anp.conjugate since they are actually the same wrapped object. However, it's minor and theoretical because whether or not two "equivalent" numpy functions/ufuncs are the same object is apparently an implementation detail, e.g., np.amax is documented as an alias of np.max, but they actually aren't the same object, and the equivalent ufuncs np.degrees and np.rad2deg are two different ones (tested on numpy 2.3.1). Because this might change between different numpy version, it's more robust to keep the grad definitions for both names.
agriyakhetarpal
approved these changes
Oct 19, 2025
Collaborator
agriyakhetarpal
left a comment
agriyakhetarpal
left a comment
There was a problem hiding this comment.
Thanks, @danra! This makes sense to me.
agriyakhetarpal
requested review from
Copilot and
fjosw
and removed request for
Copilot
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Outdated suggestions cannot be applied.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.
Avoid creating two different wrappers for the same object in
wrap_namespace.Generally cleaner, but more importantly towards better ufuncs support, which will require mapping from ufuncs to their wrapped version.
Another minor pro is that this theoretically avoids needing to define grads for two names of the same object, e.g., we could define the VJP for anp.conj but not for anp.conjugate since they are actually the same wrapped object. However, it's minor and theoretical because whether or not two "equivalent" numpy functions/ufuncs are the same object is apparently an implementation detail, e.g.,
np.amaxis documented as an alias ofnp.max, but they actually aren't the same object, and the equivalent ufuncsnp.degreesandnp.rad2degare two different ones (tested on numpy 2.3.1). Because this might change between different numpy version, it's more robust to keep the grad definitions for both names.