PANDAS: Python has R envy

Python has R envy

{
 "metadata": {
  "name": "Python has R envy"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "# http://goo.gl/c84WQ\n",
      "\n",
      "Eugene Van den Bulke  \n",
      "http://twitter.com/3kwa  \n",
      "eugene.vandenbulke@gmail.com"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "# Python is BIG in Science:\n",
      "+ CERN [The Holy Grail of programming](http://cds.cern.ch/record/974627)\n",
      "+ JPL [Lesson from middle age](http://www.youtube.com/watch?v=e3lTby5RI54)\n",
      "+ ..."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import datetime\n",
      "from IPython.display import YouTubeVideo\n",
      "\n",
      "YouTubeVideo(\"e3lTby5RI54\", start=int(datetime.timedelta(minutes=14, seconds=53).total_seconds()))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "\n",
        "        <iframe\n",
        "            width=\"400\"\n",
        "            height=300\"\n",
        "            src=\"http://www.youtube.com/embed/e3lTby5RI54?start=893\"\n",
        "            frameborder=\"0\"\n",
        "            allowfullscreen\n",
        "        ></iframe>\n",
        "        "
       ],
       "output_type": "pyout",
       "prompt_number": 1,
       "text": [
        "<IPython.lib.display.YouTubeVideo at 0x1046a9ad0>"
       ]
      }
     ],
     "prompt_number": 1
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "[John Hunter Memorial Fund](http://numfocus.org/johnhunter/)"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "# Numpy a powerful N-dimensional array object\n",
      "\n",
      "## Introduction"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "N = 100000\n",
      "\n",
      "def c_like_python():\n",
      "    a = range(N)\n",
      "    b = range(N)\n",
      "    c = []\n",
      "    for i in range(len(a)):\n",
      "        c.append(a[i] + b[i])\n",
      "    return c\n",
      "\n",
      "%timeit -n 10 c_like_python()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "10 loops, best of 3: 23.6 ms per loop\n"
       ]
      }
     ],
     "prompt_number": 2
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import itertools\n",
      "\n",
      "def hettingeresque():\n",
      "    return [a + b for a, b in itertools.izip(*itertools.tee(xrange(N), 2))]\n",
      "\n",
      "%timeit -n 10 hettingeresque()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "10 loops, best of 3: 10.7 ms per loop\n"
       ]
      }
     ],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import numpy\n",
      "\n",
      "def math_head():\n",
      "    return numpy.arange(N) + numpy.arange(N)\n",
      "\n",
      "%timeit -n 10 math_head()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "10 loops, best of 3: 193 \u00b5s per loop\n"
       ]
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Fibonacci (digression)\n",
      "\n",
      "$$F_0 = 0$$\n",
      "$$F_1 = 1$$\n",
      "$$F_{n+2} = F_{n+1} + F_{n}$$\n",
      "\n",
      "### Naive implementation"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def recursive(n):\n",
      "    if n == 0:\n",
      "        return 0\n",
      "    elif n == 1:\n",
      "        return 1\n",
      "    else:\n",
      "        return recursive(n-1) + recursive(n-2)\n",
      "    \n",
      "%timeit -n 10 recursive(20)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "10 loops, best of 3: 4.45 ms per loop\n"
       ]
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Linear algebra ... too the rescue?\n",
      "\n",
      "[33 miniatures: mathematical and algorithmic applications of linear algebra](http://kam.mff.cuni.cz/~matousek/la-ams.html)\n",
      "\n",
      "$$\n",
      "\\left(\\begin{array}{c}F_{n+1}\\\\\\\\F_n\\end{array}\\right) = \n",
      "\\begin{bmatrix}\n",
      "1 & 1 \\\\\\\\\n",
      "1 & 0 \\end{bmatrix}^n\n",
      "\\left(\\begin{array}{c}1\\\\\\\\0\\end{array}\\right)$$\n"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def vectorized(n):\n",
      "    M = numpy.array([1, 1, 1, 0]).reshape(2, 2)\n",
      "    result = numpy.linalg.matrix_power(M, n).dot([1, 0])\n",
      "    return result[1]\n",
      "\n",
      "%timeit -n 10 vectorized(20)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "10 loops, best of 3: 71.9 \u00b5s per loop\n"
       ]
      }
     ],
     "prompt_number": 6
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Neither\n",
      "\n",
      "There is a reason [functools.lru_cache](http://docs.python.org/3.4/library/functools.html#functools.lru_cache) made it into the standard library with Python 3.2."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from functools import wraps\n",
      "\n",
      "def cache(f):\n",
      "    cache = {}\n",
      "    @wraps(f)\n",
      "    def wrapper(n):\n",
      "        if n not in cache:\n",
      "            cache[n] = f(n)\n",
      "        return cache[n]\n",
      "    return wrapper\n",
      "\n",
      "@cache\n",
      "def recursive(n):\n",
      "    if n == 0:\n",
      "        return 0\n",
      "    elif n == 1:\n",
      "        return 1\n",
      "    else:\n",
      "        return recursive(n-1) + recursive(n-2)\n",
      "\n",
      "%timeit -n 10 recursive(20)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "10 loops, best of 3: 286 ns per loop\n"
       ]
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Types %$^&@ :("
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "recursive(100)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 8,
       "text": [
        "354224848179261915075L"
       ]
      }
     ],
     "prompt_number": 8
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "vectorized(100)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 9,
       "text": [
        "3736710778780434371"
       ]
      }
     ],
     "prompt_number": 9
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "M = numpy.array([1, 1, 1, 0]).reshape(2, 2)\n",
      "numpy.linalg.matrix_power(M, 100)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 10,
       "text": [
        "array([[ 1298777728820984005,  3736710778780434371],\n",
        "       [ 3736710778780434371, -2437933049959450366]])"
       ]
      }
     ],
     "prompt_number": 10
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "M = numpy.array([1, 1, 1, 0], dtype='float64').reshape(2, 2)\n",
      "numpy.linalg.matrix_power(M, 100)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 11,
       "text": [
        "array([[  5.73147844e+20,   3.54224848e+20],\n",
        "       [  3.54224848e+20,   2.18922996e+20]])"
       ]
      }
     ],
     "prompt_number": 11
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "# numpy array rocks"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import random\n",
      "\n",
      "array = [random.randint(-5, 5) for _ in xrange(10)]\n",
      "odd = [i for i in array if i % 2]\n",
      "array, odd"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 12,
       "text": [
        "([-3, -3, 0, 4, -5, 5, 3, -2, 1, 2], [-3, -3, -5, 5, 3, 1])"
       ]
      }
     ],
     "prompt_number": 12
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "array = numpy.random.randint(-5, 5, 10)\n",
      "array"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 13,
       "text": [
        "array([-4,  0,  2,  2,  4, -2,  1, -5, -3, -1])"
       ]
      }
     ],
     "prompt_number": 13
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## universal functions"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "numpy.mod(4, 2)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 14,
       "text": [
        "0"
       ]
      }
     ],
     "prompt_number": 14
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "mod = numpy.mod(array, 2)\n",
      "mod"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 15,
       "text": [
        "array([0, 0, 0, 0, 0, 0, 1, 1, 1, 1])"
       ]
      }
     ],
     "prompt_number": 15
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## broadcasting"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "mask = mod != 0\n",
      "mask"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 16,
       "text": [
        "array([False, False, False, False, False, False,  True,  True,  True,  True], dtype=bool)"
       ]
      }
     ],
     "prompt_number": 16
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## (boolean) indexing"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "odd = array[mask]\n",
      "array, odd"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 17,
       "text": [
        "(array([-4,  0,  2,  2,  4, -2,  1, -5, -3, -1]), array([ 1, -5, -3, -1]))"
       ]
      }
     ],
     "prompt_number": 17
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Why? \n",
      "\n",
      "+ ~40 times faster\n",
      "+ expressiveness (debatable)"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%timeit -n 10 [i for i in (random.randint(-5, 5) for _ in xrange(N)) if i % 2]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "10 loops, best of 3: 197 ms per loop\n"
       ]
      }
     ],
     "prompt_number": 18
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%%timeit -n 10\n",
      "array = numpy.random.randint(-5, 5, N)\n",
      "array[numpy.mod(array, 2) != 0]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "10 loops, best of 3: 4.5 ms per loop\n"
       ]
      }
     ],
     "prompt_number": 19
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "# PANDAS (Python is not so big in statistics ... YET)\n",
      "\n",
      "Python used for data munging and preparation not so much for analysis and modelling where tools like R dominate.\n",
      "\n",
      "Time series are big business, [scikit.timeseries](http://pytseries.sourceforge.net/) use to be the reference implementation but ..."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from IPython.display import VimeoVideo, ScribdDocument\n",
      "\n",
      "ScribdDocument(71048089)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "\n",
        "        <iframe\n",
        "            width=\"400\"\n",
        "            height=300\"\n",
        "            src=\"http://www.scribd.com/embeds/71048089/content\"\n",
        "            frameborder=\"0\"\n",
        "            allowfullscreen\n",
        "        ></iframe>\n",
        "        "
       ],
       "output_type": "pyout",
       "prompt_number": 20,
       "text": [
        "<IPython.lib.display.ScribdDocument at 0x1046b4750>"
       ]
      }
     ],
     "prompt_number": 20
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "VimeoVideo(\"59324550\")"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "\n",
        "        <iframe\n",
        "            width=\"400\"\n",
        "            height=300\"\n",
        "            src=\"http://player.vimeo.com/video/59324550\"\n",
        "            frameborder=\"0\"\n",
        "            allowfullscreen\n",
        "        ></iframe>\n",
        "        "
       ],
       "output_type": "pyout",
       "prompt_number": 21,
       "text": [
        "<IPython.lib.display.VimeoVideo at 0x1046b4d90>"
       ]
      }
     ],
     "prompt_number": 21
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "VimeoVideo, ScribdDocument are my tiny [contribution](https://github.com/ipython/ipython/pull/3200) to IPython\n",
      "\n",
      "## Series"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import pandas\n",
      "\n",
      "series = pandas.Series([1, 2, 3, 4, 5], index=list('abcde'), name='demo')\n",
      "series"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 22,
       "text": [
        "a    1\n",
        "b    2\n",
        "c    3\n",
        "d    4\n",
        "e    5\n",
        "Name: demo"
       ]
      }
     ],
     "prompt_number": 22
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Think of it as a dictionary"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "series['e']"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 23,
       "text": [
        "5"
       ]
      }
     ],
     "prompt_number": 23
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### An old friend"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "series.index, type(series.index.values)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 24,
       "text": [
        "(Index([a, b, c, d, e], dtype=object), numpy.ndarray)"
       ]
      }
     ],
     "prompt_number": 24
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "series.values, type(series.values)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 25,
       "text": [
        "(array([1, 2, 3, 4, 5]), numpy.ndarray)"
       ]
      }
     ],
     "prompt_number": 25
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Alignment"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "another = pandas.Series([6, 7, 8, 9, 10], index=list('acegi'))\n",
      "another"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 26,
       "text": [
        "a     6\n",
        "c     7\n",
        "e     8\n",
        "g     9\n",
        "i    10"
       ]
      }
     ],
     "prompt_number": 26
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "series / another"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 27,
       "text": [
        "a    0.166667\n",
        "b         NaN\n",
        "c    0.428571\n",
        "d         NaN\n",
        "e    0.625000\n",
        "g         NaN\n",
        "i         NaN"
       ]
      }
     ],
     "prompt_number": 27
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "_.fillna(method='bfill')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 28,
       "text": [
        "a    0.166667\n",
        "b    0.428571\n",
        "c    0.428571\n",
        "d    0.625000\n",
        "e    0.625000\n",
        "g         NaN\n",
        "i         NaN"
       ]
      }
     ],
     "prompt_number": 28
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "_.dropna()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 29,
       "text": [
        "a    0.166667\n",
        "b    0.428571\n",
        "c    0.428571\n",
        "d    0.625000\n",
        "e    0.625000"
       ]
      }
     ],
     "prompt_number": 29
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### describe"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "_.describe()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 30,
       "text": [
        "count    5.000000\n",
        "mean     0.454762\n",
        "std      0.188635\n",
        "min      0.166667\n",
        "25%      0.428571\n",
        "50%      0.428571\n",
        "75%      0.625000\n",
        "max      0.625000"
       ]
      }
     ],
     "prompt_number": 30
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### plot"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "normal = pandas.Series( numpy.random.normal(size=N), \n",
      "                        index=xrange(N),\n",
      "                        name='normal')\n",
      "normal"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 31,
       "text": [
        "0    -1.177525\n",
        "1    -0.139446\n",
        "2     0.040140\n",
        "3     1.397148\n",
        "4     0.094213\n",
        "5     0.908324\n",
        "6     0.180137\n",
        "7     0.834102\n",
        "8     0.248740\n",
        "9    -0.976949\n",
        "10   -0.309291\n",
        "11   -0.221863\n",
        "12   -0.320174\n",
        "13   -0.034669\n",
        "14   -1.942757\n",
        "...\n",
        "99985   -1.346713\n",
        "99986   -0.563065\n",
        "99987   -0.170557\n",
        "99988   -1.006635\n",
        "99989    0.640849\n",
        "99990   -0.006355\n",
        "99991    1.462928\n",
        "99992   -0.543633\n",
        "99993    1.639055\n",
        "99994    1.113526\n",
        "99995    0.379972\n",
        "99996    0.215632\n",
        "99997    1.381902\n",
        "99998   -0.918391\n",
        "99999   -1.341989\n",
        "Name: normal, Length: 100000"
       ]
      }
     ],
     "prompt_number": 31
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "normal.hist(bins=100)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 32,
       "text": [
        "<matplotlib.axes.AxesSubplot at 0x1053e9a10>"
       ]
      },
      {
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXwAAAD9CAYAAAC/fMwDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x1046b0fd0>"
       ]
      }
     ],
     "prompt_number": 32
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "normal.describe()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 33,
       "text": [
        "count    100000.000000\n",
        "mean         -0.000221\n",
        "std           1.001505\n",
        "min          -5.529533\n",
        "25%          -0.679099\n",
        "50%          -0.000121\n",
        "75%           0.676216\n",
        "max           4.395072"
       ]
      }
     ],
     "prompt_number": 33
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### skewness & kurtosis (digression)"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "normal.skew(), normal.kurt()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 34,
       "text": [
        "(0.0033964687260883385, 0.0013031617003046747)"
       ]
      }
     ],
     "prompt_number": 34
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Dataframe\n",
      "\n",
      "Think of it as a dictionary of dictionaries with more constructors than we have fingers and toes"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df = pandas.DataFrame({'A': series, 'B': another})\n",
      "df"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>A</th>\n",
        "      <th>B</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>a</th>\n",
        "      <td>  1</td>\n",
        "      <td>  6</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>b</th>\n",
        "      <td>  2</td>\n",
        "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>c</th>\n",
        "      <td>  3</td>\n",
        "      <td>  7</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>d</th>\n",
        "      <td>  4</td>\n",
        "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>e</th>\n",
        "      <td>  5</td>\n",
        "      <td>  8</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>g</th>\n",
        "      <td>NaN</td>\n",
        "      <td>  9</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>i</th>\n",
        "      <td>NaN</td>\n",
        "      <td> 10</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "output_type": "pyout",
       "prompt_number": 35,
       "text": [
        "    A   B\n",
        "a   1   6\n",
        "b   2 NaN\n",
        "c   3   7\n",
        "d   4 NaN\n",
        "e   5   8\n",
        "g NaN   9\n",
        "i NaN  10"
       ]
      }
     ],
     "prompt_number": 35
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "pandas.set_option('display.notebook_repr_html', False)\n",
      "df"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 36,
       "text": [
        "    A   B\n",
        "a   1   6\n",
        "b   2 NaN\n",
        "c   3   7\n",
        "d   4 NaN\n",
        "e   5   8\n",
        "g NaN   9\n",
        "i NaN  10"
       ]
      }
     ],
     "prompt_number": 36
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "A dictionary of dictionaries ... with a twist"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df['A']"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 37,
       "text": [
        "a     1\n",
        "b     2\n",
        "c     3\n",
        "d     4\n",
        "e     5\n",
        "g   NaN\n",
        "i   NaN\n",
        "Name: A"
       ]
      }
     ],
     "prompt_number": 37
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df.ix['e']"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 38,
       "text": [
        "A    5\n",
        "B    8\n",
        "Name: e"
       ]
      }
     ],
     "prompt_number": 38
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df.plot(subplots=True, title='Split view')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 39,
       "text": [
        "array([<matplotlib.axes.AxesSubplot object at 0x1054066d0>,\n",
        "       <matplotlib.axes.AxesSubplot object at 0x10560d390>], dtype=object)"
       ]
      },
      {
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXUAAAD7CAYAAACVMATUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkaQuhBB25P8BOwsI8PWyTlwAAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x10556d0d0>"
       ]
      }
     ],
     "prompt_number": 39
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df.fillna(method='ffill')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 40,
       "text": [
        "   A   B\n",
        "a  1   6\n",
        "b  2   6\n",
        "c  3   7\n",
        "d  4   7\n",
        "e  5   8\n",
        "g  5   9\n",
        "i  5  10"
       ]
      }
     ],
     "prompt_number": 40
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df.describe()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 41,
       "text": [
        "              A          B\n",
        "count  5.000000   5.000000\n",
        "mean   3.000000   8.000000\n",
        "std    1.581139   1.581139\n",
        "min    1.000000   6.000000\n",
        "25%    2.000000   7.000000\n",
        "50%    3.000000   8.000000\n",
        "75%    4.000000   9.000000\n",
        "max    5.000000  10.000000"
       ]
      }
     ],
     "prompt_number": 41
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df.cov()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 42,
       "text": [
        "     A    B\n",
        "A  2.5  2.0\n",
        "B  2.0  2.5"
       ]
      }
     ],
     "prompt_number": 42
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Panel\n",
      "\n",
      "Think of it as ..."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "panel = pandas.Panel({'source': df, 'squared': df * df, 'cube': df ** 3})\n",
      "panel"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 43,
       "text": [
        "<class 'pandas.core.panel.Panel'>\n",
        "Dimensions: 3 (items) x 7 (major_axis) x 2 (minor_axis)\n",
        "Items axis: cube to squared\n",
        "Major_axis axis: a to i\n",
        "Minor_axis axis: A to B"
       ]
      }
     ],
     "prompt_number": 43
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "panel.to_frame()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 44,
       "text": [
        "             cube  source  squared\n",
        "major minor                       \n",
        "a     A         1       1        1\n",
        "      B       216       6       36\n",
        "b     A         8       2        4\n",
        "c     A        27       3        9\n",
        "      B       343       7       49\n",
        "d     A        64       4       16\n",
        "e     A       125       5       25\n",
        "      B       512       8       64\n",
        "g     B       729       9       81\n",
        "i     B      1000      10      100"
       ]
      }
     ],
     "prompt_number": 44
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "##IO\n",
      "\n",
      "Text, CSV, Excel, HDF5 (PyTables) ..."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "help(pandas.io)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Help on package pandas.io in pandas:\n",
        "\n",
        "NAME\n",
        "    pandas.io\n",
        "\n",
        "FILE\n",
        "    /usr/local/lib/python2.7/site-packages/pandas/io/__init__.py\n",
        "\n",
        "PACKAGE CONTENTS\n",
        "    auth\n",
        "    data\n",
        "    date_converters\n",
        "    ga\n",
        "    parsers\n",
        "    pytables\n",
        "    sql\n",
        "    tests (package)\n",
        "    wb\n",
        "\n",
        "\n"
       ]
      }
     ],
     "prompt_number": 45
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "**`pandas.io.ga`** stands for Google Analytics - [yhat](http://blog.yhathq.com/posts/pandas-google-analytics.html)\n",
      "\n",
      "Every one wants a piece of Pandas [Adobe SiteCatalyst/Omniture](https://github.com/lexual/sitecat-py)!\n",
      "\n",
      "## grouping, joining etc.\n"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "grouped = df.groupby(lambda x: ord(x) % 2)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 46
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "grouped.sum()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 47,
       "text": [
        "minor  A   B\n",
        "0      6 NaN\n",
        "1      9  40"
       ]
      }
     ],
     "prompt_number": 47
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Like SQL better?"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from pandasql import sqldf\n",
      "\n",
      "query = lambda query: sqldf(query, globals())\n",
      "\n",
      "query('select A+B from df')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 48,
       "text": [
        "   A+B\n",
        "0    7\n",
        "1  NaN\n",
        "2   10\n",
        "3  NaN\n",
        "4   13\n",
        "5  NaN\n",
        "6  NaN"
       ]
      }
     ],
     "prompt_number": 48
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### How about SQLALchemy?\n",
      "\n",
      "Check [CALCHIPAN](https://bitbucket.org/zzzeek/calchipan/) (Crouching Alchemist Hidden Panda)"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "# Statsmodels\n",
      "\n",
      "Pandas does simple linear regression.\n",
      "\n",
      "If you need more advanced statistical analysis and modelling [statsmodels](http://statsmodels.sourceforge.net/) is your next port of call!\n",
      "\n",
      "# Machine Learning\n",
      "\n",
      "[scikti-learn](http://scikit-learn.org/stable/) and [sklearn-pandas](https://github.com/paulgb/sklearn-pandas)\n",
      "    \n",
      "R is awesome and has a lot of traction ... if you'd rather use Python to **explore & exploit** your data ... yes you can!"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "# Boundary testing and isolation"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import datetime\n",
      "\n",
      "one_hour = datetime.timedelta(hours=1)\n",
      "start = datetime.datetime(2013, 1, 1)\n",
      "\n",
      "index = [start + i * one_hour for i in xrange(N)]\n",
      "temperature = numpy.random.randint(15, 30, N)\n",
      "\n",
      "series = pandas.Series(temperature, index=index)\n",
      "series"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 49,
       "text": [
        "2013-01-01 00:00:00    27\n",
        "2013-01-01 01:00:00    25\n",
        "2013-01-01 02:00:00    25\n",
        "2013-01-01 03:00:00    26\n",
        "2013-01-01 04:00:00    27\n",
        "2013-01-01 05:00:00    29\n",
        "2013-01-01 06:00:00    21\n",
        "2013-01-01 07:00:00    20\n",
        "2013-01-01 08:00:00    26\n",
        "2013-01-01 09:00:00    22\n",
        "2013-01-01 10:00:00    21\n",
        "2013-01-01 11:00:00    21\n",
        "2013-01-01 12:00:00    20\n",
        "2013-01-01 13:00:00    15\n",
        "2013-01-01 14:00:00    21\n",
        "...\n",
        "2024-05-29 01:00:00    25\n",
        "2024-05-29 02:00:00    24\n",
        "2024-05-29 03:00:00    28\n",
        "2024-05-29 04:00:00    24\n",
        "2024-05-29 05:00:00    23\n",
        "2024-05-29 06:00:00    18\n",
        "2024-05-29 07:00:00    26\n",
        "2024-05-29 08:00:00    15\n",
        "2024-05-29 09:00:00    16\n",
        "2024-05-29 10:00:00    20\n",
        "2024-05-29 11:00:00    29\n",
        "2024-05-29 12:00:00    23\n",
        "2024-05-29 13:00:00    22\n",
        "2024-05-29 14:00:00    25\n",
        "2024-05-29 15:00:00    16\n",
        "Length: 100000"
       ]
      }
     ],
     "prompt_number": 49
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "series.resample('D', how='mean').head()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 50,
       "text": [
        "2013-01-01    22.458333\n",
        "2013-01-02    21.666667\n",
        "2013-01-03    21.875000\n",
        "2013-01-04    21.791667\n",
        "2013-01-05    21.375000\n",
        "Freq: D"
       ]
      }
     ],
     "prompt_number": 50
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "**Altered resample default behavior:**\n",
      "\n",
      "The default time series resample binning behavior of daily D and higher frequencies has been changed to closed='left', label='left'. Lower nfrequencies are unaffected. The prior defaults were causing a great deal of confusion for users, especially resampling data to daily frequency (which labeled the aggregated group with the end of the interval: the next day)."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "series.resample('D', how='mean', closed='right', label='right').head()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 51,
       "text": [
        "2013-01-01    27.000000\n",
        "2013-01-02    22.291667\n",
        "2013-01-03    21.500000\n",
        "2013-01-04    22.125000\n",
        "2013-01-05    21.500000\n",
        "Freq: D"
       ]
      }
     ],
     "prompt_number": 51
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Isolating third party library is ignored too often despite being both USEFUL and EASY\n",
      "\n",
      "It helps when switching from scikit.timeseries to pandas and when upgrading pandas (under intense development)"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def resample_daily(resamplable):\n",
      "    return resamplable.resample('D', how='mean')\n",
      "\n",
      "def test_resample_daily():\n",
      "    # load resamplable fixture and expected\n",
      "    # result = resample_daily(fixture)\n",
      "    # assert result = expected\n",
      "    pass"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 52
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Invitation to (re)read chapter 8 of [Clean Code](http://www.amazon.com/Clean-Code-Handbook-Software-Craftsmanship/dp/0132350882)"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from IPython.core.display import Image \n",
      "Image(filename='boundaries.png') "
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAdoAAAF2CAIAAABGWM5tAAAKsGlDQ1BJQ0MgUHJvZmlsZQAASA2t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\ngwAAAABJRU5ErkJggg==\n",
       "prompt_number": 53,
       "text": [
        "<IPython.core.display.Image at 0x105675790>"
       ]
      }
     ],
     "prompt_number": 53
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "# We can all be Data Scientist\n",
      "\n",
      "It is becoming trendier than being a **software engineer** and we can do it *all* in Python ;)\n",
      "\n",
      "# Questions?"
     ]
    }
   ],
   "metadata": {}
  }
 ]
}