Kevin R. Thornton

Any idea how much this drops RAM use? For me, the case that motivated opening the issue on RAM use was computing the pairwise distance matrix. It blew up to...

Is `memory-profiler` measuring actual process memory, or just memory occupied by Python object? The caching is happening on the C side, and so may not be visible? If you are...

Keeping things idiomatic by returning an iterator seems like a good idea to me. We could have a second function that returns an array if anyone wants to deal with...

I have implemented discrete chromosomes in fwdpy11. Length refers to the total genome length, and the recombination functions work to ensure correct recording of transmitted edges. In practice, keeping the...

One question: if you add chromosome labels, and edges on chromosome i are in [0,i-1), then do you plan to simplify separately by chromosome or change how edges are sorted...

I'd probably populate the tables when "dumping" to tskit's format. But I'm practice, we keep everything in the fwdpp format as the metadata are much smaller. I may be thinking...

> > I like the original idea proposed by @jeromekelleher, and don't see it as problematic to create a new table, after all, if @molpopgen recommends... > > > keeping...

From a forward simulation perspective, an alternative that seems more natural is to keep an array of table collections, and simplify each independently. Further, I then know immediately how to...

> > From a forward simulation perspective, an alternative that seems more natural is to keep an array of table collections, and simplify each independently. Further, I then know immediately...

> I haven't thought this through in detail, but I imagine that we have another `int32_t` column on the edge table, which is `-1` by default (== no chromosome), similar...