PR 114: [WIP] count_allele_calls

[quansight-bot]

Issue metadata

• url: https://github.com/pystatgen/sgkit/issues/185
• state: open
• milestone: No milestone assigned
• created by: Quansight Bot @quansight-bot
• labels:
• assigned:

Issue description

Pull request metadata

• url: https://github.com/pystatgen/sgkit/pull/114
• state: closed
• milestone: No milestone assigned
• created by: Tim Millar [@timothymillar](https://github.com/timothymillar)
• labels:
  • auto-merge
• assigned:

Pull request description

See issue #85

This implements an additional jitted function count_call_alleles_ndarray_jit for ndarrays only rather than doing it all in dask.
This approach seems to be inline with the goals outlined here but I'm happy to replicate the approach of the original count_alleles function if that is preferred.

Likewise I can re-write count_alleles using this approach which should improve performance (mainly on chucked arrays due to njit(..., nogil=True))

I haven't added numba to requirements.txt or setup.py Because that is done in #76

I guess add numba now for CI and fix conflict later