{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "# CR\u00dcCIAL PYTHON Week 8: Easy parallelization with joblib"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from IPython.core.display import Image\n",
      "Image(url='http://labrosa.ee.columbia.edu/crucialpython/logo.png', width=600)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<img src=\"http://labrosa.ee.columbia.edu/crucialpython/logo.png\" width=\"600\"/>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 1,
       "text": [
        "<IPython.core.display.Image at 0x10c711f90>"
       ]
      }
     ],
     "prompt_number": 1
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## What is joblib?\n",
      "joblib is a python module which includes functionality for doing various useful tasks, like caching the output of functions given certain input, logging/tracing, and easy parallelization.  Today, we'll focus on the latter, which essentially is a wrapper around Python's built-in `multiprocessing` module.  It's essentially just an extremely easy way to write a parallelized `for` loop."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## joblib.Parallel and joblib.delayed\n",
      "Two write a simple parallelized `for` loop using joblib, you need to use the `Parallel` class and the `delayed` function.  Together, they form a construct which looks a lot like list comprehension.  `delayed` is a [decorator](http://nbviewer.ipython.org/github/craffel/crucialpython/blob/master/week7/decorators.ipynbdecorator) which simply returns the function, its arguments, and its keyword arguments as a tuple.  `Parallel` constructs a class which can be called with a list of tuples, where each tuple includes a function and its arguments.  It subsequently calls each function from each tuple with the corresponding arguments.  But, what you really need to know in practice is simple:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from joblib import Parallel, delayed\n",
      "import numpy as np\n",
      "print [np.power(i, 2) for i in xrange(10)]\n",
      "print Parallel()(delayed(np.power)(i, 2) for i in xrange(10))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "[0, 1, 4, 9, 16, 25, 36, 49, 64, 81]\n",
        "[0, 1, 4, 9, 16, 25, 36, 49, 64, 81]\n"
       ]
      }
     ],
     "prompt_number": 2
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# To parallelize, simply set the n_jobs argument!\n",
      "Parallel(n_jobs=8)(delayed(np.power)(i, 2) for i in xrange(10))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 3,
       "text": [
        "[0, 1, 4, 9, 16, 25, 36, 49, 64, 81]"
       ]
      }
     ],
     "prompt_number": 3
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Practical example\n",
      "`np.power` is not a function you'd need to parallelize calls to in general, because it's so fast and the overhead for parallelization makes it not worth while (we'll get to that later).  Here's a function which actually takes a while to compute:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def convolve_random(size):\n",
      "    ''' Convolve two random arrays of length \"size\" '''\n",
      "    return np.convolve(np.random.random_sample(size), np.random.random_sample(size))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Time to run once with length-40000 arrays\n",
      "%timeit convolve_random(40000)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "1 loops, best of 3: 904 ms per loop\n"
       ]
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Time to run sequentially for length 40000, 41000, 42000, ... 47000 arrays\n",
      "%timeit [convolve_random(40000 + i*1000) for i in xrange(8)]\n",
      "# In parallel, with 8 jobs\n",
      "%timeit Parallel(n_jobs=8)(delayed(convolve_random)(40000 + i*1000) for i in xrange(8))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "1 loops, best of 3: 8.69 s per loop\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "1 loops, best of 3: 2.88 s per loop\n"
       ]
      }
     ],
     "prompt_number": 6
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Other useful features of Parallel\n",
      "The `Parallel` class is meant to be more convenient to use than `multiprocessing`.  As such, it can dump out progress messages, cleanly report errors, and be interrupted with ctrl-C, all of which takes some effort to do in `multiprocessing`."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Use the verbose argument to display progress messages.\n",
      "# The frequency of the messages increases with the verbosity level. \n",
      "result = Parallel(n_jobs=8, verbose=50)(delayed(convolve_random)(40000 + i*1000) for i in xrange(16))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "[Parallel(n_jobs=8)]: Done   1 out of  16 | elapsed:    2.1s remaining:   32.0s\n",
        "[Parallel(n_jobs=8)]: Done   2 out of  16 | elapsed:    2.2s remaining:   15.7s\n",
        "[Parallel(n_jobs=8)]: Done   3 out of  16 | elapsed:    2.4s remaining:   10.3s\n",
        "[Parallel(n_jobs=8)]: Done   4 out of  16 | elapsed:    2.5s remaining:    7.4s\n",
        "[Parallel(n_jobs=8)]: Done   5 out of  16 | elapsed:    2.6s remaining:    5.6s\n",
        "[Parallel(n_jobs=8)]: Done   6 out of  16 | elapsed:    2.7s remaining:    4.5s\n",
        "[Parallel(n_jobs=8)]: Done   7 out of  16 | elapsed:    2.8s remaining:    3.6s\n",
        "[Parallel(n_jobs=8)]: Done   8 out of  16 | elapsed:    3.0s remaining:    3.0s\n",
        "[Parallel(n_jobs=8)]: Done   9 out of  16 | elapsed:    5.3s remaining:    4.1s\n",
        "[Parallel(n_jobs=8)]: Done  10 out of  16 | elapsed:    5.6s remaining:    3.3s\n",
        "[Parallel(n_jobs=8)]: Done  11 out of  16 | elapsed:    5.9s remaining:    2.7s\n",
        "[Parallel(n_jobs=8)]: Done  12 out of  16 | elapsed:    6.0s remaining:    2.0s\n",
        "[Parallel(n_jobs=8)]: Done  13 out of  16 | elapsed:    6.3s remaining:    1.4s\n",
        "[Parallel(n_jobs=8)]: Done  14 out of  16 | elapsed:    6.3s remaining:    0.9s\n",
        "[Parallel(n_jobs=8)]: Done  15 out of  16 | elapsed:    6.4s remaining:    0.4s\n",
        "[Parallel(n_jobs=8)]: Done  16 out of  16 | elapsed:    6.6s finished\n"
       ]
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## When should you use Parallel?\n",
      "The easiest case (and the only case we'll cover here) is when you want to run the same process many times with different arguments, where each run of the process does not depend on any other runs.  This is the intended use of `joblib.Parallel`.\n",
      "\n",
      "Furthermore, as mentioned above, there's some overhead to using parallelization in general.  As a simple rule, for very fast functions, it's not worth it, and for very slow functions, you'll get a speedup which levels off as you approach the number of cores you have.  The following code seeks to test this in a principled way using our `convolve_random` function."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Try convolution sizes [5000, 10000, 15000 ... 50000]\n",
      "sizes = 5000*(1 + np.arange(10))\n",
      "# Try n_jobs from [1, ..., max_jobs]\n",
      "max_jobs = 8\n",
      "n_jobs_values = 1 + np.arange(max_jobs)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 8
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import time\n",
      "# Store the timing for each setting\n",
      "times = np.zeros((n_jobs_values.shape[0], sizes.shape[0]))\n",
      "for n, n_jobs in enumerate(n_jobs_values):\n",
      "    for m, size in enumerate(sizes):\n",
      "        start = time.time()\n",
      "        result = Parallel(n_jobs=n_jobs)(delayed(convolve_random)(size) for i in xrange(max_jobs))\n",
      "        # Compute and store elapsed time\n",
      "        times[n, m] = time.time() - start\n",
      "# Save it out so we don't have to run it twice\n",
      "np.savetxt('times.txt', times)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import matplotlib.pyplot as plt\n",
      "# Load in our pre-computed times\n",
      "times = np.genfromtxt('times.txt')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 9
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "plt.figure(figsize=(10, 6))\n",
      "for size_index in xrange(4):\n",
      "    plt.subplot(2, 2, size_index + 1)\n",
      "    # Plot the times for this size by n_jobs\n",
      "    plt.plot(n_jobs_values, times[:, size_index], '.')\n",
      "    # Set up axes and labels\n",
      "    plt.xticks(n_jobs_values)\n",
      "    plt.xlim([.8, max_jobs + .2])\n",
      "    plt.title('Size = {}'.format(sizes[size_index]))\n",
      "    plt.xlabel('n_jobs')\n",
      "    plt.ylabel('Time (s)')\n",
      "plt.tight_layout()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAskAAAGrCAYAAAAy4QohAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x10cb8a610>"
       ]
      }
     ],
     "prompt_number": 10
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "plt.figure(figsize=(10, 6))\n",
      "# Need as many colors as max_jobs!\n",
      "colors = ['r', 'k', 'Purple', 'IndianRed', 'Chartreuse', 'DarkRed',  'Teal', 'b']\n",
      "ax = plt.gca().set_color_cycle(colors)\n",
      "# Plot each n_jobs setting as a different line\n",
      "for job_results in times:\n",
      "    plt.plot(sizes, job_results)\n",
      "# Set up labels and legend\n",
      "plt.xlabel('Convolution size')\n",
      "plt.ylabel('Time (s)')\n",
      "plt.legend(n_jobs_values, 'upper left', title='n_jobs')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 11,
       "text": [
        "<matplotlib.legend.Legend at 0x10cb54290>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAmkAAAF/CAYAAAASFl7JAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x10cb84f90>"
       ]
      }
     ],
     "prompt_number": 11
    }
   ],
   "metadata": {}
  }
 ]
}