Perf-tests on Cython string comparisons

testing_out_string_comparisons_for_inf_like_values.ipynb

{
 "metadata": {
  "name": "Testing out string comparison methods for inf-like values"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": "Testing for multiple string representations of Infinity"
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Testing out various options for adding additional infinity tests (for pandas)"
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "%load_ext cythonmagic\ninpt1 = [\"inf\", \"Inf\", \"-inf\", \"-Inf\", \"infinity\", \"-infinity\", \"RASCAL\", \"rabbit\", \"sicko\", \"grabbit\"] * 500\ninpt2 = [\"a\", \"b\", \"c\", \"d\"] * 500\ninpt3 = [\"a\" * 10000, \"b\" * 10000, \"inf\" * 10000, \"-INF\" * 100, \"-INF\", \"Inf\"] * 500",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "%%cython\ncimport numpy as np\nimport numpy as np\ncdef char* cinf = b'inf'\ncdef char* negcinf = b'-inf'\ncdef char* cinf2 = b'Inf'\ncdef char* negcinf2 = b'-Inf'\ncdef np.ndarray infinities\n\ninfinities = np.array([b'inf', b'-inf', b'Inf', b'-Inf'])\n\nfrom libc.string cimport strcmp, strcasecmp\n\ncpdef compare_with_strcmp(lst):\n    cdef int count = 0\n    for elem in lst:\n        if strcmp(elem, cinf) == 0 or strcmp(elem, cinf2) == 0:\n            count += 1\n        elif strcmp(elem, negcinf) == 0 or strcmp(elem, negcinf2) == 0:\n            count += 1\n        else:\n            pass\n    return count\n        \ncpdef compare_with_strcmp_reversed(lst):\n    cdef int count = 0\n    for elem in lst:\n        if strcmp(cinf, elem) == 0 or strcmp(cinf2, elem) == 0:\n            count += 1\n        elif strcmp(negcinf, elem) == 0 or strcmp(negcinf2, elem) == 0:\n            count += 1\n        else:\n            pass\n    return count\n\ncpdef compare_with_in_operator(lst):\n    cdef int count = 0\n    for elem in lst:\n        if elem in infinities:\n            count += 1\n    return count\n\ncpdef compare_with_strcasecmp(lst):\n    cdef int count = 0\n    for elem in lst:\n        if strcasecmp(elem, cinf) == 0:\n            count += 1\n        elif strcasecmp(elem, negcinf) == 0:\n            count +=1\n    return count",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 6
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "INPUT = inpt1\nprint \"Strlen\"\n%timeit compare_with_strcmp(INPUT)\nprint \"Strlen reversed\"\n%timeit compare_with_strcmp_reversed(INPUT)\nprint \"In operator\"\n%timeit compare_with_in_operator(INPUT)\nprint \"Case insensitive\"\n%timeit compare_with_strcasecmp(INPUT)",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "Strlen\n1000 loops, best of 3: 273 us per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\nStrlen reversed\n1000 loops, best of 3: 266 us per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\nIn operator\n10 loops, best of 3: 48.3 ms per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\nCase insensitive\n1000 loops, best of 3: 285 us per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n"
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "INPUT = inpt2\nprint \"Strlen\"\n%timeit compare_with_strcmp(INPUT)\nprint \"Strlen reversed\"\n%timeit compare_with_strcmp_reversed(INPUT)\nprint \"In operator\"\n%timeit compare_with_in_operator(INPUT)\nprint \"Case insensitive\"\n%timeit compare_with_strcasecmp(INPUT)",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "Strlen\n10000 loops, best of 3: 125 us per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\nStrlen reversed\n10000 loops, best of 3: 123 us per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\nIn operator\n100 loops, best of 3: 19.3 ms per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\nCase insensitive\n10000 loops, best of 3: 104 us per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n"
      }
     ],
     "prompt_number": 8
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "INPUT = inpt3\nprint \"Strlen\"\n%timeit compare_with_strcmp(INPUT)\nprint \"Strlen reversed\"\n%timeit compare_with_strcmp_reversed(INPUT)\nprint \"In operator\"\n%timeit compare_with_in_operator(INPUT)\nprint \"Case insensitive\"\n%timeit compare_with_strcasecmp(INPUT)",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "Strlen\n10000 loops, best of 3: 105 us per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\nStrlen reversed\n10000 loops, best of 3: 92.5 us per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\nIn operator\n100 loops, best of 3: 19.2 ms per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\nCase insensitive\n10000 loops, best of 3: 89.1 us per loop"
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "\n"
      }
     ],
     "prompt_number": 9
    }
   ],
   "metadata": {}
  }
 ]
}