Ben Jeffery

Python 3.12 support

1

Python 3.12 has been released. The good news is that tskit builds and almost all tests pass. The failing ones are those that depend on `numba` via `lshmm` as `numba`...

Remove legacy h5py based formats

5

Opening this up for discussion. As mentioned by @jeromekelleher at https://github.com/tskit-dev/tskit/pull/2811#issuecomment-1663778875 it has been a long time since these legacy formats were used. When dropping them we should add a...

Error if boolean passed as row index or slice?

3

I was recently helping a new user who has R experience who tried `tables.nodes['individual'==0]` which gave the first node as it is equal to `tables.nodes[False]` which gets interpreted as `tables.nodes[0]`....

Modify getsize to return total size, not just the top level

2

Fixes #253 TODO: * [X] Changes documented in docs/release.rst * [ ] GitHub Actions have all passed * [ ] Test coverage is 100% (Codecov passes)

Add sample mask

9

Needs a couple of extra tests for weird masks, but mostly there.

Low density regions and splitting

2

Currently for large scale inference were splitting at regions of low, or no, site density. This results in some chrom arms with 3-4 tree sequences, which is annoying for downstream...

Currently, sample and variant masks form an sgkit dataset are "True=keep". This is the opposite meaning to sgkit dataset masks which are "True=mask". These should be made consistent.

Sample masking from sgkit dataset

1

To restrict inference to a subset of samples a subset-copy of the sgkit dataset needs to be performed. For large datasets this is less than ideal. If tsinfer read a...