[Numpy-discussion] Test suite for duck arrays

2023-01-16 Thread Mark Harfouche
I am trying to create a few different duck arrays that are backed by
different files.

Is there a standard test suite that we can inspire ourselves from that
helps us assert:

"Arrays should implement all these different functions and the results
should be correct"
___
NumPy-Discussion mailing list -- numpy-discussion@python.org
To unsubscribe send an email to numpy-discussion-le...@python.org
https://mail.python.org/mailman3/lists/numpy-discussion.python.org/
Member address: arch...@mail-archive.com


[Numpy-discussion] Re: Test suite for duck arrays

2023-01-16 Thread Ralf Gommers
On Mon, Jan 16, 2023 at 3:28 PM Mark Harfouche 
wrote:

> I am trying to create a few different duck arrays that are backed by
> different files.
>
> Is there a standard test suite that we can inspire ourselves from that
> helps us assert:
>
> "Arrays should implement all these different functions and the results
> should be correct"
>

Have a look at https://github.com/data-apis/array-api-tests/, it does
pretty much that and is meant to easily be configurable and can run against
an arbitrary array library.

For everything in NumPy that's not in the array API standard, I'm not aware
of such a test suite (and it's then a fairly ill-defined problem anyway).

Cheers,
Ralf
___
NumPy-Discussion mailing list -- numpy-discussion@python.org
To unsubscribe send an email to numpy-discussion-le...@python.org
https://mail.python.org/mailman3/lists/numpy-discussion.python.org/
Member address: arch...@mail-archive.com


[Numpy-discussion] Re: Test suite for duck arrays

2023-01-16 Thread Mark Harfouche
Wow this looks great!

On Mon, Jan 16, 2023 at 10:52 AM Ralf Gommers 
wrote:

>
>
> On Mon, Jan 16, 2023 at 3:28 PM Mark Harfouche 
> wrote:
>
>> I am trying to create a few different duck arrays that are backed by
>> different files.
>>
>> Is there a standard test suite that we can inspire ourselves from that
>> helps us assert:
>>
>> "Arrays should implement all these different functions and the results
>> should be correct"
>>
>
> Have a look at https://github.com/data-apis/array-api-tests/, it does
> pretty much that and is meant to easily be configurable and can run against
> an arbitrary array library.
>
> For everything in NumPy that's not in the array API standard, I'm not
> aware of such a test suite (and it's then a fairly ill-defined problem
> anyway).
>
> Cheers,
> Ralf
>
> ___
> NumPy-Discussion mailing list -- numpy-discussion@python.org
> To unsubscribe send an email to numpy-discussion-le...@python.org
> https://mail.python.org/mailman3/lists/numpy-discussion.python.org/
> Member address: mark.harfou...@gmail.com
>
___
NumPy-Discussion mailing list -- numpy-discussion@python.org
To unsubscribe send an email to numpy-discussion-le...@python.org
https://mail.python.org/mailman3/lists/numpy-discussion.python.org/
Member address: arch...@mail-archive.com