SIFT1B_read.py

# -*- coding: UTF-8 -*-
"""
Python script to read SIFT1B dataset files, including .bvecs, .ivecs, .fvecs
Dataset source: http://corpus-texmex.irisa.fr/
"""
import numpy as np

def bvecs_read(fname):
    a = np.fromfile(fname, dtype=np.int32, count=1)
    b = np.fromfile(fname, dtype=np.uint8)
    d = a[0]
    return b.reshape(-1, d + 4)[:, 4:].copy()


def ivecs_read(fname):
    a = np.fromfile(fname, dtype='int32')
    d = a[0]
    return a.reshape(-1, d + 1)[:, 1:].copy()


def fvecs_read(fname):
    return ivecs_read(fname).view('float32')