Shuffle once benchmarks

shuffle_once.ipynb

{
 "metadata": {
  "name": "sgdclassifier shuffling"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "from sklearn.linear_model import SGDClassifier\nfrom sklearn.datasets import fetch_mldata",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "mnist = fetch_mldata(\"MNIST original\")\nX, y = mnist.data / 255., mnist.target",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "%timeit SGDClassifier(shuffle_once=False).fit(X, y)\n%timeit SGDClassifier(shuffle=True, shuffle_once=False).fit(X, y)\n%timeit SGDClassifier(shuffle_once=True).fit(X, y)\n%timeit SGDClassifier(shuffle_once=False, shuffle=2).fit(X, y)",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "1 loops, best of 3: 2.8 s per loop\n1 loops, best of 3: 3.32 s per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n1 loops, best of 3: 3.48 s per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n1 loops, best of 3: 3.64 s per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n"
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "%timeit SGDClassifier(shuffle_once=False, n_iter=100).fit(X, y)\n%timeit SGDClassifier(shuffle=True, n_iter=100, shuffle_once=False).fit(X, y)\n%timeit SGDClassifier(shuffle_once=True, n_iter=100).fit(X, y)\n%timeit SGDClassifier(shuffle_once=False, shuffle=2, n_iter=100).fit(X, y)",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "1 loops, best of 3: 10.2 s per loop\n1 loops, best of 3: 12.8 s per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n1 loops, best of 3: 10.6 s per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n1 loops, best of 3: 13 s per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n"
      }
     ],
     "prompt_number": 15
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "from sklearn.datasets import load_digits\ndigits = load_digits()\nX, y = digits.data / 16., digits.target",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "%timeit SGDClassifier(shuffle_once=False).fit(X, y)\n%timeit SGDClassifier(shuffle=True, shuffle_once=False).fit(X, y)\n%timeit SGDClassifier(shuffle_once=True).fit(X, y)\n%timeit SGDClassifier(shuffle_once=False, shuffle=2).fit(X, y)",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "100 loops, best of 3: 7.87 ms per loop\n100 loops, best of 3: 9.23 ms per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n100 loops, best of 3: 9.29 ms per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n100 loops, best of 3: 8.84 ms per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n"
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "%timeit SGDClassifier(shuffle_once=False, n_iter=100).fit(X, y)\n%timeit SGDClassifier(shuffle=True, n_iter=100, shuffle_once=False).fit(X, y)\n%timeit SGDClassifier(shuffle_once=True, n_iter=100).fit(X, y)\n%timeit SGDClassifier(shuffle_once=False, shuffle=2, n_iter=100).fit(X, y)",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "10 loops, best of 3: 137 ms per loop\n10 loops, best of 3: 161 ms per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n10 loops, best of 3: 134 ms per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n10 loops, best of 3: 152 ms per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n"
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "from sklearn.datasets import fetch_20newsgroups_vectorized",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 8
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "news = fetch_20newsgroups_vectorized()",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 9
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "news.data.shape",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 10,
       "text": "(11314, 130107)"
      }
     ],
     "prompt_number": 10
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "X, y = news.data, news.target",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 11
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "%timeit SGDClassifier(shuffle_once=False).fit(X, y)\n%timeit SGDClassifier(shuffle=True, shuffle_once=False).fit(X, y)\n%timeit SGDClassifier(shuffle_once=True).fit(X, y)\n%timeit SGDClassifier(shuffle_once=False, shuffle=2).fit(X, y)",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "1 loops, best of 3: 550 ms per loop\n1 loops, best of 3: 654 ms per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n1 loops, best of 3: 582 ms per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n1 loops, best of 3: 630 ms per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n"
      }
     ],
     "prompt_number": 13
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "%timeit SGDClassifier(shuffle_once=False, n_iter=100).fit(X, y)\n%timeit SGDClassifier(shuffle=True, n_iter=100, shuffle_once=False).fit(X, y)\n%timeit SGDClassifier(shuffle_once=False, n_iter=100, shuffle=2).fit(X, y)\n%timeit SGDClassifier(shuffle_once=False, shuffle=2, n_iter=100).fit(X, y)",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "1 loops, best of 3: 10.5 s per loop\n1 loops, best of 3: 12.3 s per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n1 loops, best of 3: 12.4 s per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n1 loops, best of 3: 12.9 s per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n"
      }
     ],
     "prompt_number": 14
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "",
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
}