

Automatic index selection and tuning library for FAISS that selects optimal KNN index configurations to maximize recall given memory and query speed constraints, eliminating manual hyperparameter tuning.
Loading more......
Autofaiss is a Python library developed by Criteo that automatically creates FAISS KNN indices with optimal similarity search parameters. It wraps the FAISS library and automatically tunes the best KNN index to maximize recall given RAM and query speed constraints.
FAISS offers hundreds of algorithm combinations with up to 6 hyperparameters per index type. Autofaiss eliminates the need to manually select and tune these parameters by automatically finding the optimal configuration for your specific requirements.
Available via pip:
pip install autofaiss