Paddle qr

[jindalai]

Close #20785
qr frontend function and test for the paddlepaddle linear algebra module
Closes the todolist subtask #20785
(Previous PR was closed due to merge conflict)

[ivy-leaves]

If you are working on an open task, please edit the PR description to link to the issue you've created.

For more information, please check ToDo List Issues Guide.

Thank you 🤗

[jindalai]

Resolved merge conflicts

[rishabgit]

Hi! Thanks for working on this 😄 Could you please take care of the lint issues? Looks great otherwise!

Hope this helps - pre-commit

[jindalai]

Hi @rishabgit! Thanks a lot for the review! 😄
I have fixed the formatting with pre-commit and now there are no errors.
Please let me know if something needs further work.

[jindalai]

Hi @rishabgit! PR was accidentally closed. I have reopened it.

[rishabgit]

Hey, I'm getting an odd error on local testing - ERROR ivy_tests/test_ivy/test_frontends/test_paddle/test_tensor/test_linalg.py - ivy.utils.exceptions.IvyAttributeError: paddle: function_supported_devices_and_dtypes: paddle: function_supported_devices_and_dtypes: 'tuple' object has no attribute 'values'

Looks like it was due to a bug on the testing pipeline which was recently fixed, could you please sync your branch with the latest updates?

[jindalai]

Thanks for letting me know!
I have synced the branch with the ivy/master branch.

Some tests are failing here. Please let me know of what might be causing them and the required changes in the code/test/deployment
😄

[rishabgit]

Hi, I resolved the issue with incorrect fixtures in testing, but it throws error on qr implementation now. Could you please look into it? Thanks!

[jindalai]

Thanks a lot for the changes!
Reg the implementation I cant find anything wrong as of now

I have referred the Paddle documentation and the corresponding ivy code for it.

Can you please let me know if you find something?

The qr function implementation seems to be quite simple. I have accounted for the:

• Supported datatypes
• Positional Args

Could this be due to some other testing/infra reason?

[rishabgit]

Hi, your testing fn was off, fixed it now but getting AssertionError: the results from backend torch and ground truth framework paddle do not match which I'm not sure how to fix exactly

@hirwa-nshuti saw your changes here which I believe deals with similar "issue" with numpy testing?

Do you have any idea on what could be the ideal workaround here? Already set rtol=1e-01

[jindalai]

Hello @rishabgit
How should we proceed with it?

Are there some changes I should be trying?

[rishabgit]

Hi! Fixed the issue with testing pipeline, could you please take care of merge conflicts? LGTM to merge otherwise. Thanks! 😄

[rishabgit]

Needed slight updates, but LGTM otherwise! Thanks for adding this @yashj8 😄

[jindalai]

Thanks for everything @rishabgit! This was my first PR at ivy!
Sorry for the delay and mess up in the merge conflict.
Can you please shed some light on what might be going wrong initially (which you fixed with the matrix_is_stable?) Would be helpful later on.

[rishabgit]

Thanks for everything @rishabgit! This was my first PR at ivy! Sorry for the delay and mess up in the merge conflict. Can you please shed some light on what might be going wrong initially (which you fixed with the matrix_is_stable?) Would be helpful later on.

Hey @yashj8, I'll suggest diving into the code of matrix_is_stable (especially its docstring) and googling from there if you're unclear about any key word - would explain things much better + exhaustively than I could imo 😄