Using k-d trees to efficiently calculate nearest neighbors in 3D vector space

Re:Using k-d trees to efficiently calculate nearest neighbor... by joshvm on Sunday February 11, 2018 @12:36AM (#16351362)

There's also FLANN which is an approximate nearest neighbour algorithm which is routinely used for things like feature/keypoint matching (e.g. SIFT keypoints are 128 dimensional).

https://www.cs.ubc.ca/research/flann/

Barnes Hut is a fun one to implement for N-body simulations.

[ Parent | on HN ]

Re:Using k-d trees to efficiently calculate nearest neighbor... by _wmd on Sunday February 11, 2018 @02:24AM (#16351626)

I was toying with compqring similarity of texts just last night, is there anything that could handle 10k+ sized sparse vectors? Obviously there are tools like lucene, but I wanted to mess around in the interactive interpreter

[ Parent | on HN ]

Re:Using k-d trees to efficiently calculate nearest neighbor... by agibsonccc on Sunday February 11, 2018 @03:08AM (#16351729)

Check out LargeVis then. The original github repo by the authors has a sparse mode: https://github.com/lferry007/LargeVis/

TSNE didn't scale well no matter what you did to it (hence the newer technique) - in general though you can't beat cosine similarity of TFIDF vectors as a baseline.

[ Parent | on HN ]