Product quantization

This repository is an introduction to Product Quantization (PQ). ProductQuantization class allows us to do mathematical operations directly on the compressed matrix representations. The class has the sklearn standard methods: fit, encode, transform and its written in pure python.

References

• https://github.com/dblalock/bolt
• https://arxiv.org/pdf/2106.10860.pdf
• https://github.com/matsui528/nanopq
• https://www.jstage.jst.go.jp/article/mta/6/1/6_2/_pdf/
• https://hal.inria.fr/inria-00514462v2/document