#DRAFT

Revisiting the Inverted Indices for Billion-Scale Approximate Nearest Neighbors

Build

Today we provide the C++ implementation supporting only the CPU version, which requires a BLAS library.

The code requires a C++ compiler that understands:

• the Intel intrinsics for SSE instructions
• the GCC intrinsic for the popcount instruction
• basic OpenMP

1) Configure FAISS

There are a few models for makefile.inc in the faiss/example_makefiles/ subdirectory. Copy the relevant one for your system to faiss/ and adjust to your needs. In particular, you need to set a proper BLAS library paths. There are also indications for specific configurations in the troubleshooting section of the FAISS wiki.

https://github.com/facebookresearch/faiss/wiki/Troubleshooting

2) Replace FAISS CMakeList.txt Replace faiss/CMakeList.txt with CMakeList.txt.faiss in order to deactivate building of unnecessary tests and the GPU version.

3) Build project ``` cmake . make ```

Data

• The proposed methods are tested on two 1 billion datasets: SIFT1B and DEEP1B. For proposed examples, all data files have to be in data/SIFT1B and data/DEEP1B. ``` cd data/SIFT1B bash load_sift1b.sh cd data/DEEP1B python load_deep1b.py ```
• We also provide learned centroids and labels of base points assigned to particular centroid for both datasets. ``` TODO ```