DelightRun

#include <iostream>
#include <random>
#include <chrono>
#include <x86intrin.h>
#include <cassert>

// Runtime evaluation for squared Eucliden distance functions
// - fvec_L2_sqr_ref: naive reference impl from Faiss
// - fvec_L2_sqr_sse: SSE impl from Faiss
// - fvec_L2_sqr_avx: AVX impl from Faiss

DelightRun

{"lastUpload":"2020-04-29T08:23:10.533Z","extensionVersion":"v3.4.3"}

DelightRun

"""
Implementation of 'Maximum Likelihood Estimation of Intrinsic Dimension' by Elizaveta Levina and Peter J. Bickel

how to use
----------

The goal is to estimate intrinsic dimensionality of data, the estimation of dimensionality is scale dependent
(depending on how much you zoom into the data distribution you can find different dimesionality), so they
propose to average it over different scales, the interval of the scales [k1, k2] are the only parameters of the algorithm.

DelightRun

import time
import argparse
import faiss
import numpy as np
from utils import load_vecs, save_vecs


parser = argparse.ArgumentParser(description="Build KNN Graph using GPU")
parser.add_argument('dataset', type=str, help='Dataset')
parser.add_argument('k', type=int, nargs='+', help='K')

DelightRun

sudo find /etc/apt/sources.list.d/ -type f -name "*.list" -exec  sed  -i.bak -r  's#deb(-src)?\s*http(s)?://ppa.launchpad.net#deb\1 http\2://launchpad.proxy.ustclug.org#ig' {} \;