Leland McInnes
Leland McInnes
This is ... very peculiar. It certainly isn't obvious to me how this could happen. It is possible to end up with infinite distances in a result if an insufficient...
Hmm, definitely still some issues -- things got brought over from umap, but apparently there are some issues. I don't have time to dig into this right now, but hopefully...
I finally got some time and dug into this. The short answer was that not everything got fully ported over from umap, and there were some issues with sparse support...
Thanks for #68 @tomwhite, that shoudl take care of a lot of them. This is popping up in umap as well, so I will have to take care of it...
That's not a bad idea if people have a KNN graph already. So right now we "seed" a knn-graph based on results from an rp-forest. It would not be hard,...
To ensure I am understanding correctly, you want to find, for each _y_ in _Y_, the points _{x_1, ..., x_k}_ in _X_ that are closest to _y_? Presuming you have...
I agree that a pruning approach may yield some benefits, but as you note, that is going to require some careful analysis as to how best to apply such a...
Do you want the peak RAM usage (harder to calculate) or the size of an index? The latter is going to be a small multiple of the dataset RAM usage...
I'm not sure quite what you mean in suggestion one -- can you explain a little further? For option two there used to be an option that tried to detect...
Ah, I see what you want to do now. It isn't so obvious how to integrate that into the code in a way that wouldn't be a very special case...