Numpy serialization benchmark.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "import numpy as np\n",
    "from levy_data import cdf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Numpy Dump / Load to file (calls pickle)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 loop, best of 3: 611 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%timeit cdf.dump('test')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The slowest run took 18.17 times longer than the fastest. This could mean that an intermediate result is being cached.\n",
      "1 loop, best of 3: 32.9 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%timeit test = np.load('test')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Note: we should memoize the data everywhere it's loaded to improve performance*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Numpy Dumps / Loads and write to file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 loop, best of 3: 751 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%%timeit\n",
    "a = cdf.dumps()\n",
    "with open('test', 'bw') as f:\n",
    "    f.write(a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The slowest run took 16.75 times longer than the fastest. This could mean that an intermediate result is being cached.\n",
      "1 loop, best of 3: 34.6 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%%timeit\n",
    "with open('test', 'rb') as f:\n",
    "    test = np.loads(f.read())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Direct to pickle file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 loop, best of 3: 394 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%%timeit\n",
    "with open('test', 'bw') as f:\n",
    "    pickle.dump(cdf, f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The slowest run took 195.77 times longer than the fastest. This could mean that an intermediate result is being cached.\n",
      "1 loop, best of 3: 5.38 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%%timeit\n",
    "with open('test', 'br') as f:\n",
    "    pickle.load(f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Numpy savez / load"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 loop, best of 3: 872 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%timeit np.savez('test', cdf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The slowest run took 78.33 times longer than the fastest. This could mean that an intermediate result is being cached.\n",
      "1 loop, best of 3: 4.77 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%timeit test = np.load('test')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Numpy savez_compressed / load"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 loop, best of 3: 908 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%timeit np.savez_compressed('test', cdf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100 loops, best of 3: 6.17 ms per loop\n"
     ]
    }
   ],
   "source": [
    "%timeit test = np.load('test')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [conda env:vol_tools3.5]",
   "language": "python",
   "name": "conda-env-vol_tools3.5-py"
  },
  "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": 1
}