mir.algorithm - Sorting, searching and counting

[wilzbach]

Ideally this would seamlessly integrate with the existing algorithms in Phobos from std.algorithm.

I don't know what we do with new algorithms, but mir.algorithm sounds like a good place? At least sorting, partitioning, counting and searching should be checked.

See also: https://github.com/DlangScience/mir/wiki/NumPy:-Sorting,-searching,-and-counting

[9il]

Probably this name would never be used. What you can place here?

[9il]

Sorting, searching and counting can be done with std.algorithm. Any use case, which should be put here?

[wilzbach]

AFAIK phobos doesn't have any windowing function.

[9il]

but ndslice already has windows

[wilzbach]

This issue is also about creating some documentation and tests that ensure that std.algorithm works fine with Ndslice - especially the documentation is important to users.

[wilzbach]

but ndslice already has windows

Okay then this is only to write a guideguide or tutorial.

[9il]

This issue is also about creating some documentation and tests that ensure that std.algorithm works fine with Ndslice - especially the documentation is important to users.

The best option is mir.ndslice.example with a lot of unittests, which contains a 1-3 lines functions to cover numpy.

[9il]

BTW, half of check boxes in our wiki about numpy should be turn on. They are 1-3 lines functions.

[9il]

BTW, half of check boxes in our wiki about numpy should be turn on. They are 1-3 lines functions.

However this is useless without examples

[wilzbach]

Closing in favor of #77 and #78

[wilzbach]

I found a good example in ndslice - median (from the examples in package.d) could be part of such a package.

[9il]

I found a good example in ndslice - median (from the examples in package.d) could be part of such a package.

This would in filters or something specialized. algorithm assumes something general purposed