Kyle McDonald

seems like this wasn't fully fixed?

Random cost matrices are definitely not representative for many applications. But I wanted to toss in another toy problem that generates a more difficult cost matrix: https://gist.github.com/kylemcdonald/ac9e2e84552e60543071c33a250111c9 ```python import numpy...

@cheind that's exactly it. I use it for [this sort of thing](https://twitter.com/kcimc/status/1072218450519441413). A friend has [a heuristic](https://github.com/Quasimondo/RasterFairy/) for this that provides poor solutions for non-uniform data. For comparison, on my...

I just changed to uint32 for pylapsolver and I saw a huge speed increase for the random cost matrix. For the cloud-to-grid problem I saw it drop from 40s to...

i think @arturoc was the one working on this, but i don't know what system he was planning on using to solve this.

there's definitely a lot going on in the header, and some of them could be grouped like you say... what do you think the main groups are exactly? community, download,...

just wanted to say this is basically ok with everyone else, we just have to hack something together to get it to work.

i think we're in agreement that your suggestion works, we just need someone to take the time to implement it. if you're up for it, go for it! otherwise we...

maybe @ofZach could chime in and explain better what he meant -- i was adding this issue for him :)

thanks so much! i'm happy to merge this if you can also add a (very short, couple sentence) explanation to the readme about how to use cmake to build the...