timing of str concat in python

time_str_contact

{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "based on http://www.skymind.com/~ocrow/python_string/"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import sys\n",
      "print sys.version"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "2.7.5 |Anaconda 1.6.1 (x86_64)| (default, Jun 28 2013, 22:20:13) \n",
        "[GCC 4.0.1 (Apple Inc. build 5493)]\n"
       ]
      }
     ],
     "prompt_number": 1
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "MAX_NUM = 20000"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def method_0_str(max_num=MAX_NUM):\n",
      "    out_str = ''\n",
      "    for num in xrange(max_num):\n",
      "        out_str += str(num)\n",
      "    return out_str"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def method_0_repr(max_num=MAX_NUM):\n",
      "    out_str = ''\n",
      "    for num in xrange(max_num):\n",
      "        out_str += repr(num)\n",
      "    return out_str"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def method_1_bt(max_num=MAX_NUM):\n",
      "    out_str = ''\n",
      "    for num in xrange(max_num):\n",
      "        out_str += `num`\n",
      "    return out_str"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from array import array\n",
      "\n",
      "def method_3_array(max_num=MAX_NUM):\n",
      "    char_array = array('c') \n",
      "    for num in xrange(max_num):\n",
      "        char_array.fromstring(repr(num))\n",
      "    return char_array.tostring()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 6
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def method_4_list(max_num=MAX_NUM):\n",
      "    str_list = []\n",
      "    for num in xrange(max_num):\n",
      "        str_list.append(repr(num))\n",
      "    return ''.join(str_list)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 7
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from cStringIO import StringIO\n",
      "\n",
      "def method_5_cStr(max_num=MAX_NUM):\n",
      "    file_str = StringIO()\n",
      "    for num in xrange(max_num):\n",
      "        file_str.write(repr(num))\n",
      "    return file_str.getvalue()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 8
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def method_6_list_compr(max_num=MAX_NUM):\n",
      "    return ''.join([repr(num) for num in xrange(max_num)])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 9
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def method_7_buffer(max_num=MAX_NUM):\n",
      "    templist = [0]*max_num\n",
      "    num = 0\n",
      "    while num < max_num:\n",
      "        templist[num] = repr(num)\n",
      "        num += 1\n",
      "    return ''.join(templist)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 10
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# a bit of cheating\n",
      "import numpy as np\n",
      "def method_8_numpy(max_num=MAX_NUM):\n",
      "    arr = np.arange(max_num)\n",
      "    return np.array2string(arr, separator='')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 11
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "method_5_cStr(10)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 12,
       "text": [
        "'0123456789'"
       ]
      }
     ],
     "prompt_number": 12
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## timing"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import time\n",
      "def _timeit(stmt, repeat=1):\n",
      "    code = compile(stmt, '<_timeit>', 'exec')\n",
      "    t0 = time.time()  # use time.clock() if on windows\n",
      "    exec code\n",
      "    t1 = time.time()\n",
      "    ms = (t1-t0)*1000\n",
      "    \n",
      "    cnt = 1\n",
      "    while cnt < repeat:\n",
      "        t0 = time.time()  # use time.clock() if on windows\n",
      "        exec code\n",
      "        t1 = time.time()\n",
      "        ms = 0.5 * (ms + (t1-t0)*1000)\n",
      "        cnt += 1\n",
      "    return ms"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 13
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "meth = [m for m in dir() if m.startswith('method_')]\n",
      "timing = {}\n",
      "for m in meth:\n",
      "    timing[m] = _timeit(m+'()', repeat=10)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 14
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "sorted(timing.items(), key=lambda x: x[1])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 15,
       "text": [
        "[('method_8_numpy', 0.3586830571293831),\n",
        " ('method_6_list_compr', 3.315247595310211),\n",
        " ('method_1_bt', 3.7367017939686775),\n",
        " ('method_4_list', 4.171660635620356),\n",
        " ('method_7_buffer', 5.43083855882287),\n",
        " ('method_0_repr', 6.246051285415888),\n",
        " ('method_3_array', 7.965344935655594),\n",
        " ('method_0_str', 7.995434571057558),\n",
        " ('method_5_cStr', 12.76054373010993)]"
       ]
      }
     ],
     "prompt_number": 15
    }
   ],
   "metadata": {}
  }
 ]
}