{
 "metadata": {
  "name": "",
  "signature": "sha256:a318835b4224090dfb7278ec7e82129b5966061077556fe99ce04391d435ef3f"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import numpy as np\n",
      "import pandas as pd\n",
      "from pandas import Series, DataFrame"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 23
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Let's see how we would find outliers in a dataset\n",
      "\n",
      "# First we'll seed the numpy generator\n",
      "np.random.seed(12345)\n",
      "\n",
      "#Next we'll create the dataframe\n",
      "dframe = DataFrame(np.random.randn(1000,4))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 24
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "#Show preview\n",
      "dframe.head()"
     ],
     "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>0</th>\n",
        "      <th>1</th>\n",
        "      <th>2</th>\n",
        "      <th>3</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
        "      <td>-0.204708</td>\n",
        "      <td> 0.478943</td>\n",
        "      <td>-0.519439</td>\n",
        "      <td>-0.555730</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td> 1.965781</td>\n",
        "      <td> 1.393406</td>\n",
        "      <td> 0.092908</td>\n",
        "      <td> 0.281746</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td> 0.769023</td>\n",
        "      <td> 1.246435</td>\n",
        "      <td> 1.007189</td>\n",
        "      <td>-1.296221</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td> 0.274992</td>\n",
        "      <td> 0.228913</td>\n",
        "      <td> 1.352917</td>\n",
        "      <td> 0.886429</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
        "      <td>-2.001637</td>\n",
        "      <td>-0.371843</td>\n",
        "      <td> 1.669025</td>\n",
        "      <td>-0.438570</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 25,
       "text": [
        "          0         1         2         3\n",
        "0 -0.204708  0.478943 -0.519439 -0.555730\n",
        "1  1.965781  1.393406  0.092908  0.281746\n",
        "2  0.769023  1.246435  1.007189 -1.296221\n",
        "3  0.274992  0.228913  1.352917  0.886429\n",
        "4 -2.001637 -0.371843  1.669025 -0.438570"
       ]
      }
     ],
     "prompt_number": 25
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Lets describe the data\n",
      "dframe.describe()"
     ],
     "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>0</th>\n",
        "      <th>1</th>\n",
        "      <th>2</th>\n",
        "      <th>3</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>count</th>\n",
        "      <td> 1000.000000</td>\n",
        "      <td> 1000.000000</td>\n",
        "      <td> 1000.000000</td>\n",
        "      <td> 1000.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>mean</th>\n",
        "      <td>   -0.067684</td>\n",
        "      <td>    0.067924</td>\n",
        "      <td>    0.025598</td>\n",
        "      <td>   -0.002298</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>std</th>\n",
        "      <td>    0.998035</td>\n",
        "      <td>    0.992106</td>\n",
        "      <td>    1.006835</td>\n",
        "      <td>    0.996794</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>min</th>\n",
        "      <td>   -3.428254</td>\n",
        "      <td>   -3.548824</td>\n",
        "      <td>   -3.184377</td>\n",
        "      <td>   -3.745356</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>25%</th>\n",
        "      <td>   -0.774890</td>\n",
        "      <td>   -0.591841</td>\n",
        "      <td>   -0.641675</td>\n",
        "      <td>   -0.644144</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>50%</th>\n",
        "      <td>   -0.116401</td>\n",
        "      <td>    0.101143</td>\n",
        "      <td>    0.002073</td>\n",
        "      <td>   -0.013611</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>75%</th>\n",
        "      <td>    0.616366</td>\n",
        "      <td>    0.780282</td>\n",
        "      <td>    0.680391</td>\n",
        "      <td>    0.654328</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>max</th>\n",
        "      <td>    3.366626</td>\n",
        "      <td>    2.653656</td>\n",
        "      <td>    3.260383</td>\n",
        "      <td>    3.927528</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 26,
       "text": [
        "                 0            1            2            3\n",
        "count  1000.000000  1000.000000  1000.000000  1000.000000\n",
        "mean     -0.067684     0.067924     0.025598    -0.002298\n",
        "std       0.998035     0.992106     1.006835     0.996794\n",
        "min      -3.428254    -3.548824    -3.184377    -3.745356\n",
        "25%      -0.774890    -0.591841    -0.641675    -0.644144\n",
        "50%      -0.116401     0.101143     0.002073    -0.013611\n",
        "75%       0.616366     0.780282     0.680391     0.654328\n",
        "max       3.366626     2.653656     3.260383     3.927528"
       ]
      }
     ],
     "prompt_number": 26
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Lets select the first column\n",
      "col = dframe[0]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 27
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# NOw we can check which values in the column are greater than 3, for instance.\n",
      "col[np.abs(col)>3]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 28,
       "text": [
        "523   -3.428254\n",
        "900    3.366626\n",
        "Name: 0, dtype: float64"
       ]
      }
     ],
     "prompt_number": 28
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# So we now know in column[0], rows 523 and 900 have values with abs > 3\n",
      "\n",
      "#How about all the columns?\n",
      "\n",
      "# We can use the \"any\" method\n",
      "dframe[(np.abs(dframe)>3).any(1)]"
     ],
     "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>0</th>\n",
        "      <th>1</th>\n",
        "      <th>2</th>\n",
        "      <th>3</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>5  </th>\n",
        "      <td>-0.539741</td>\n",
        "      <td> 0.476985</td>\n",
        "      <td> 3.248944</td>\n",
        "      <td>-1.021228</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>97 </th>\n",
        "      <td>-0.774363</td>\n",
        "      <td> 0.552936</td>\n",
        "      <td> 0.106061</td>\n",
        "      <td> 3.927528</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>102</th>\n",
        "      <td>-0.655054</td>\n",
        "      <td>-0.565230</td>\n",
        "      <td> 3.176873</td>\n",
        "      <td> 0.959533</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>305</th>\n",
        "      <td>-2.315555</td>\n",
        "      <td> 0.457246</td>\n",
        "      <td>-0.025907</td>\n",
        "      <td>-3.399312</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>324</th>\n",
        "      <td> 0.050188</td>\n",
        "      <td> 1.951312</td>\n",
        "      <td> 3.260383</td>\n",
        "      <td> 0.963301</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>400</th>\n",
        "      <td> 0.146326</td>\n",
        "      <td> 0.508391</td>\n",
        "      <td>-0.196713</td>\n",
        "      <td>-3.745356</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>499</th>\n",
        "      <td>-0.293333</td>\n",
        "      <td>-0.242459</td>\n",
        "      <td>-3.056990</td>\n",
        "      <td> 1.918403</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>523</th>\n",
        "      <td>-3.428254</td>\n",
        "      <td>-0.296336</td>\n",
        "      <td>-0.439938</td>\n",
        "      <td>-0.867165</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>586</th>\n",
        "      <td> 0.275144</td>\n",
        "      <td> 1.179227</td>\n",
        "      <td>-3.184377</td>\n",
        "      <td> 1.369891</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>808</th>\n",
        "      <td>-0.362528</td>\n",
        "      <td>-3.548824</td>\n",
        "      <td> 1.553205</td>\n",
        "      <td>-2.186301</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>900</th>\n",
        "      <td> 3.366626</td>\n",
        "      <td>-2.372214</td>\n",
        "      <td> 0.851010</td>\n",
        "      <td> 1.332846</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 29,
       "text": [
        "            0         1         2         3\n",
        "5   -0.539741  0.476985  3.248944 -1.021228\n",
        "97  -0.774363  0.552936  0.106061  3.927528\n",
        "102 -0.655054 -0.565230  3.176873  0.959533\n",
        "305 -2.315555  0.457246 -0.025907 -3.399312\n",
        "324  0.050188  1.951312  3.260383  0.963301\n",
        "400  0.146326  0.508391 -0.196713 -3.745356\n",
        "499 -0.293333 -0.242459 -3.056990  1.918403\n",
        "523 -3.428254 -0.296336 -0.439938 -0.867165\n",
        "586  0.275144  1.179227 -3.184377  1.369891\n",
        "808 -0.362528 -3.548824  1.553205 -2.186301\n",
        "900  3.366626 -2.372214  0.851010  1.332846"
       ]
      }
     ],
     "prompt_number": 29
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# WE could also possibly cap the data at 3\n",
      "\n",
      "dframe[np.abs(dframe)>3] = np.sign(dframe) *3"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 33
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "dframe.describe()"
     ],
     "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>0</th>\n",
        "      <th>1</th>\n",
        "      <th>2</th>\n",
        "      <th>3</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>count</th>\n",
        "      <td> 1000.000000</td>\n",
        "      <td> 1000.000000</td>\n",
        "      <td> 1000.000000</td>\n",
        "      <td> 1000.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>mean</th>\n",
        "      <td>   -0.061623</td>\n",
        "      <td>    0.074473</td>\n",
        "      <td>    0.037153</td>\n",
        "      <td>    0.009919</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>std</th>\n",
        "      <td>    0.995875</td>\n",
        "      <td>    0.989820</td>\n",
        "      <td>    1.003604</td>\n",
        "      <td>    0.989688</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>min</th>\n",
        "      <td>   -2.969411</td>\n",
        "      <td>   -2.989741</td>\n",
        "      <td>   -2.925113</td>\n",
        "      <td>   -2.881858</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>25%</th>\n",
        "      <td>   -0.774132</td>\n",
        "      <td>   -0.588138</td>\n",
        "      <td>   -0.622310</td>\n",
        "      <td>   -0.636641</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>50%</th>\n",
        "      <td>   -0.115171</td>\n",
        "      <td>    0.102787</td>\n",
        "      <td>    0.012889</td>\n",
        "      <td>   -0.010997</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>75%</th>\n",
        "      <td>    0.619779</td>\n",
        "      <td>    0.787953</td>\n",
        "      <td>    0.682401</td>\n",
        "      <td>    0.659019</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>max</th>\n",
        "      <td>    3.000000</td>\n",
        "      <td>    3.000000</td>\n",
        "      <td>    3.000000</td>\n",
        "      <td>    3.000000</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 34,
       "text": [
        "                 0            1            2            3\n",
        "count  1000.000000  1000.000000  1000.000000  1000.000000\n",
        "mean     -0.061623     0.074473     0.037153     0.009919\n",
        "std       0.995875     0.989820     1.003604     0.989688\n",
        "min      -2.969411    -2.989741    -2.925113    -2.881858\n",
        "25%      -0.774132    -0.588138    -0.622310    -0.636641\n",
        "50%      -0.115171     0.102787     0.012889    -0.010997\n",
        "75%       0.619779     0.787953     0.682401     0.659019\n",
        "max       3.000000     3.000000     3.000000     3.000000"
       ]
      }
     ],
     "prompt_number": 34
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Next we'll learn about Permutation!"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
}