metrics sparse comparison.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from scipy.misc import comb\n",
    "from scipy.sparse import coo_matrix, find\n",
    "from scipy.sparse.data import _data_matrix\n",
    "\n",
    "from random import randint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "labels_true = [randint(0,5000) for _ in range(1000000)]\n",
    "labels_pred = [randint(0,5000) for _ in range(1000000)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "classes, class_idx = np.unique(labels_true, return_inverse=True)\n",
    "clusters, cluster_idx = np.unique(labels_pred, return_inverse=True)\n",
    "n_classes = classes.shape[0]\n",
    "n_clusters = clusters.shape[0]\n",
    "\n",
    "c = coo_matrix((np.ones(class_idx.shape[0]),(class_idx, cluster_idx)),\n",
    "                             shape=(n_classes, n_clusters),\n",
    "                             dtype=np.int)\n",
    "c_array = c.toarray()\n",
    "csr = c.tocsr()\n",
    "csc = c.tocsc()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100 loops, best of 3: 11.6 ms per loop\n",
      "100 loops, best of 3: 4.06 ms per loop\n",
      "10 loops, best of 3: 107 ms per loop\n",
      "10 loops, best of 3: 93.9 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%timeit c.sum(axis=0)\n",
    "%timeit c.sum(axis=1)\n",
    "%timeit find(c)\n",
    "%timeit c.tocsc().data.flatten()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10 loops, best of 3: 34.8 ms per loop\n",
      "10 loops, best of 3: 29.2 ms per loop\n",
      "10 loops, best of 3: 51.3 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%timeit c_array.sum(axis=0)\n",
    "%timeit c_array.sum(axis=1)\n",
    "%timeit c_array.flatten()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100 loops, best of 3: 2.52 ms per loop\n",
      "1000 loops, best of 3: 1.42 ms per loop\n",
      "10 loops, best of 3: 38.4 ms per loop\n",
      "1000 loops, best of 3: 1.03 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%timeit csr.sum(axis=0)\n",
    "%timeit csr.sum(axis=1)\n",
    "%timeit find(csr)\n",
    "%timeit csr.data.flatten()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1000 loops, best of 3: 1.42 ms per loop\n",
      "100 loops, best of 3: 2.41 ms per loop\n",
      "10 loops, best of 3: 56.4 ms per loop\n",
      "1000 loops, best of 3: 1.02 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%timeit csc.sum(axis=0)\n",
    "%timeit csc.sum(axis=1)\n",
    "%timeit find(csc)\n",
    "%timeit csc.data.flatten()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 loop, best of 3: 1.41 s per loop\n",
      "1 loop, best of 3: 3.38 s per loop\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import cluster\n",
    "%timeit cluster.homogeneity_completeness_v_measure(labels_true, labels_pred)\n",
    "%timeit cluster.homogeneity_completeness_v_measure(labels_true, labels_pred, sparse=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 loop, best of 3: 415 ms per loop\n",
      "1 loop, best of 3: 2.46 s per loop\n",
      "1 loop, best of 3: 3.28 s per loop\n",
      "1 loop, best of 3: 3.36 s per loop\n"
     ]
    }
   ],
   "source": [
    "%timeit cluster.mutual_info_score(labels_true, labels_pred, contingency=c_array)\n",
    "%timeit cluster.mutual_info_score(labels_true, labels_pred, contingency=c)\n",
    "%timeit cluster.mutual_info_score(labels_true, labels_pred, contingency=csr)\n",
    "%timeit cluster.mutual_info_score(labels_true, labels_pred, contingency=csc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}