fix: Ivy Failing Test: paddle - activations.relu for all backend

[marvlyngkhoi]

PR Description

500 test cases passed locally for all backend

Related Issue

Closes #28232

Checklist

• Did you add a function?
• Did you add the tests?
• Did you run your tests and are your tests passing?
• Did pre-commit not fail on any check?
• Did you follow the steps we provided?

[github-actions]

PR Compliance Checks

Thank you for your Pull Request! We have run several checks on this pull request in order to make sure it's suitable for merging into this project. The results are listed in the following section.

Issue Reference

In order to be considered for merging, the pull request description must refer to a specific issue number. This is described in our contributing guide and our PR template.
This check is looking for a phrase similar to: "Fixes #XYZ" or "Resolves #XYZ" where XYZ is the issue number that this PR is meant to address.

[samthakur587]

handle the complex dtype at paddle backend.

[marvlyngkhoi]

Did as you've asked and tested again found out that the error for complex numbers happen due to cases where complex_mode=st.sampled_from(["jax","split","magnitude"]), was set to jax at the main ivy test_activation.py file so should I add a check at the paddle backend or should i just removed jax from the list of stategy to sample from @samthakur587

[samthakur587]

Did as you've asked and tested again found out that the error for complex numbers happen due to cases where complex_mode=st.sampled_from(["jax","split","magnitude"]), was set to jax at the main ivy test_activation.py file so should I add a check at the paddle backend or should i just removed jax from the list of stategy to sample from @samthakur587

No we should not remove the jax from complex_mode=st.sampled_from(["jax","split","magnitude"]) according to the _handle_complex_input handling the complex data. if it's 'jax' then it's calling the function based on the backend like here the paddle and call the relu impl. for complex dtype. as i have seen the we don't have the correct implementation for complex dtype at paddle backend. so we should try to fix this. @fleventy-5

[marvlyngkhoi]

@samthakur587 did the fix and found out the root cause of the problem complex128 dtype which is not supported in paddle I think
Pass 500 cases tested locally for paddle backend

Also tested 200 cases for all backend just to be sure that the fix didn't break other backend

[samthakur587]

Hii @fleventy-5 can you please solve the merge conflict to run the CI.

[marvlyngkhoi]

Hii @fleventy-5 can you please solve the merge conflict to run the CI.

@samthakur587 done

[samthakur587]

import ivy
import paddle
import jax
x = paddle.to_tensor([1-1j, 1+1j])
y = jax.numpy.array([1-1j, 1+1j])
#paddle output: ivy.array([1.+0.j, 1.+1.j])
print(ivy.relu(x))
#jax output: ivy.array([1.-1.j, 1.+1.j])
print(ivy.relu(y))

as the logic for the complex dtype at paddle is return paddle.complex(F.relu(x.real()), F.relu(x.imag())) it's dosen't pass this case bcz the x.imag() part is -ve and if we apply the relu activation function then it returns the zero that is the wrong logic here for paddle backend for complex dtype.

[samthakur587]

Hii @fleventy-5 sorry for the late reply. i have added the comment what changes are required.

[marvlyngkhoi]

@samthakur587 done the implementation at the paddle back-end in the below code

def relu(
    x: paddle.Tensor, /, *, complex_mode="jax", out: Optional[paddle.Tensor] = None
) -> paddle.Tensor:
    x_dtype = str(x.dtype).split('.')[1]
    if paddle.is_complex(x):
        if complex_mode=="jax":
            x_real = paddle.real(x)
            x_imag = paddle.imag(x)
    
            condition = paddle.logical_or(
               paddle.logical_and(x_real <= 0, x_imag < 0)
               , x_real < 0)
            
            x_imag = paddle.where(condition, paddle.zeros_like(x_imag), x_imag)  
            return paddle.complex(F.relu(x_real),x_imag)
        
        return paddle.complex(F.relu(x.real()), F.relu(x.imag()))
    return F.relu(x)

but there is still issue with complex128

and when I tried the function all by itself on isolation using the above function definition it was working fine as shown below

but it was working fine for complex64

[Ishticode]

Hi @fleventy-5
Thanks for the PR. Merging this now.
@samthakur587 Could you fix the value mismatches for some of the complex vlue cases that you outlined in paddle :)
Thank you both