Jerome Kelleher

Well, we could add an option to *either* store the intermediate results or recompute them. I don't think this would add too much complexity. That is, assuming there *is* a...

Is this still an open issue? I think we should probably close it unless someone is intending to follow it up.

OK, let's keep it open.

I'm going to close this because we're addressing the general problem with pairwise statistics using a different framework now (starting from the divergence matrix in #2736) The short version I...

Out of interest @agladstein, if you're working with mamba already, why not install all the requirements from there in the first place?

msprime has binary wheels including GSL built-in for some time now (although there is https://github.com/tskit-dev/msprime/issues/2343 )

Whichever you prefer.

Thanks @tomwhite that's very interesting. I think the limitations of having to use a numpy-like API rather than simple loops, and this significant performance drop on key functionality is a...

It's pretty obscure code too. Numba FTW!

This isn't very easy right now all right - @hyanwong I think we have some hacks for this in sc2ts? Something like, identify the recombinants as the nodes that have...