RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing high performance applications.
As the name suggests, nn-descent is immensely useful on its own, especially for computing all-neighbors graphs. It has its own c++ API but we need a Python API for it.
Intron7
commented
1 year ago
As the name suggests, nn-descent is immensely useful on its own, especially for computing all-neighbors graphs. It has its own c++ API but we need a Python API for it.