Rename sparse functions to match numpy array API

[jessegrabowski]

Description

Rename tensor.sparse.mul -> tensor.sparse.multiply and tensor.sparse.sub -> tensor.sparse.subtract

Related Issue

• Closes Rename to multiply to match numpy #1370, Rename to subtract to make numpy API compat #1371
• Related to #

Checklist

• Checked that the pre-commit linting/style checks pass
• Included tests that prove the fix is effective or that the new feature works
• Added necessary documentation (docstrings and/or example notebooks)
• If you are a pro: each commit corresponds to a relevant logical change

Type of change

• New feature / enhancement
• Bug fix
• Documentation
• Maintenance
• Other (please specify):

---

📚 Documentation preview 📚: https://pytensor--1663.org.readthedocs.build/en/1663/

[ricardoV94]

hmm I'm sure that issue was meant for tensor operations not sparse. Those don't exist on numpy so we can't "match their API". How does scipy call them?

not sure

[Armavica]

I would vote for yes, scipy also calls them multiply. And as far as I understand it's not numpy's API that we want to match but the Array API which is supposed to apply to whatever behaves like an array?

But why not also change the tensor operations in tensor/math.py while we are at it?

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 81.57%. Comparing base (95acdb3) to head (b331f9a).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1663   +/-   ##
=======================================
  Coverage   81.56%   81.57%           
=======================================
  Files         242      242           
  Lines       53819    53827    +8     
  Branches     9485     9485           
=======================================
+ Hits        43899    43907    +8     
  Misses       7431     7431           
  Partials     2489     2489           

Files with missing lines	Coverage Δ
pytensor/sparse/basic.py	82.71% <100.00%> (+0.07%)	⬆️

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[ricardoV94]

We want to match numpy/scipy API, in case it doesn't have an equivalent (like subtract it seems), we can use the array API as the next best thing. That's where numpy/scipy want to converge anyway

[ricardoV94]

So both changes are fine

[jessegrabowski]

Do you want me to do the pytensor.tensor stuff in this PR too? Happy to. I did sparse in this PR because I searched for open issues related to sparse and this came up.

[ricardoV94]

As you prefer, it's your work

[jessegrabowski]

But why not also change the tensor operations in tensor/math.py while we are at it?

I started looking into this, but it seems like a pretty big job. I made the trivial replacements, then \started hitting all kinds of random bugs with rewrites and prettyprint and was quickly demotivated.