{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "#Agenda\n",
      "\n",
      "- Define the problem and the approach\n",
      "- Data basics: loading data, looking at your data, basic commands\n",
      "- Handling missing values\n",
      "- Intro to scikit-learn\n",
      "- Grouping and aggregating data\n",
      "- <p style=\"color: red\">Feature selection</p>\n",
      "- Fitting and evaluating a model\n",
      "- Deploying your work"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "#In this workbook you will\n",
      "- Use RandomForest to calculate the best features for your model\n",
      "- Handling outlying data using `apply` and `groupby`\n",
      "- Discretize (or bucket) data into groups\n",
      "- Make your own features for your model"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import pandas as pd\n",
      "import numpy as np\n",
      "import pylab as pl"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 1
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df = pd.read_csv(\"./data/credit-data-trainingset.csv\")"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df.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>serious_dlqin2yrs</th>\n",
        "      <th>revolving_utilization_of_unsecured_lines</th>\n",
        "      <th>age</th>\n",
        "      <th>number_of_time30-59_days_past_due_not_worse</th>\n",
        "      <th>debt_ratio</th>\n",
        "      <th>monthly_income</th>\n",
        "      <th>number_of_open_credit_lines_and_loans</th>\n",
        "      <th>number_of_times90_days_late</th>\n",
        "      <th>number_real_estate_loans_or_lines</th>\n",
        "      <th>number_of_time60-89_days_past_due_not_worse</th>\n",
        "      <th>number_of_dependents</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
        "      <td> 1</td>\n",
        "      <td> 0.766127</td>\n",
        "      <td> 45</td>\n",
        "      <td> 2</td>\n",
        "      <td> 0.802982</td>\n",
        "      <td>  9120</td>\n",
        "      <td> 13</td>\n",
        "      <td> 0</td>\n",
        "      <td> 6</td>\n",
        "      <td> 0</td>\n",
        "      <td> 2</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td> 0</td>\n",
        "      <td> 0.957151</td>\n",
        "      <td> 40</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0.121876</td>\n",
        "      <td>  2600</td>\n",
        "      <td>  4</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0</td>\n",
        "      <td> 1</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td> 0</td>\n",
        "      <td> 0.658180</td>\n",
        "      <td> 38</td>\n",
        "      <td> 1</td>\n",
        "      <td> 0.085113</td>\n",
        "      <td>  3042</td>\n",
        "      <td>  2</td>\n",
        "      <td> 1</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td> 0</td>\n",
        "      <td> 0.907239</td>\n",
        "      <td> 49</td>\n",
        "      <td> 1</td>\n",
        "      <td> 0.024926</td>\n",
        "      <td> 63588</td>\n",
        "      <td>  7</td>\n",
        "      <td> 0</td>\n",
        "      <td> 1</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
        "      <td> 0</td>\n",
        "      <td> 0.213179</td>\n",
        "      <td> 74</td>\n",
        "      <td> 0</td>\n",
        "      <td> 0.375607</td>\n",
        "      <td>  3500</td>\n",
        "      <td>  3</td>\n",
        "      <td> 0</td>\n",
        "      <td> 1</td>\n",
        "      <td> 0</td>\n",
        "      <td> 1</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 3,
       "text": [
        "   serious_dlqin2yrs  revolving_utilization_of_unsecured_lines  age  \\\n",
        "0                  1                                  0.766127   45   \n",
        "1                  0                                  0.957151   40   \n",
        "2                  0                                  0.658180   38   \n",
        "3                  0                                  0.907239   49   \n",
        "4                  0                                  0.213179   74   \n",
        "\n",
        "   number_of_time30-59_days_past_due_not_worse  debt_ratio  monthly_income  \\\n",
        "0                                            2    0.802982            9120   \n",
        "1                                            0    0.121876            2600   \n",
        "2                                            1    0.085113            3042   \n",
        "3                                            1    0.024926           63588   \n",
        "4                                            0    0.375607            3500   \n",
        "\n",
        "   number_of_open_credit_lines_and_loans  number_of_times90_days_late  \\\n",
        "0                                     13                            0   \n",
        "1                                      4                            0   \n",
        "2                                      2                            1   \n",
        "3                                      7                            0   \n",
        "4                                      3                            0   \n",
        "\n",
        "   number_real_estate_loans_or_lines  \\\n",
        "0                                  6   \n",
        "1                                  0   \n",
        "2                                  0   \n",
        "3                                  1   \n",
        "4                                  1   \n",
        "\n",
        "   number_of_time60-89_days_past_due_not_worse  number_of_dependents  \n",
        "0                                            0                     2  \n",
        "1                                            0                     1  \n",
        "2                                            0                     0  \n",
        "3                                            0                     0  \n",
        "4                                            0                     1  "
       ]
      }
     ],
     "prompt_number": 3
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "##Finding Important Features\n",
      "We're going to let scikit-learn help us determine which variables are the best at predicting risk. To do this, we're going to use an algorithm called `RandomForest`. `RandomForest` randomly generates a \"forest\" of decision trees. As the trees are randomly generated, the algorithm takes turns leaving out each variable in fitting the model. This allows the `RandomForest` to calculate just how much worse a model does when each variable is left out."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.ensemble import RandomForestClassifier\n",
      "\n",
      "features = np.array(['revolving_utilization_of_unsecured_lines',\n",
      "                     'age', 'number_of_time30-59_days_past_due_not_worse',\n",
      "                     'debt_ratio', 'monthly_income','number_of_open_credit_lines_and_loans', \n",
      "                     'number_of_times90_days_late', 'number_real_estate_loans_or_lines',\n",
      "                     'number_of_time60-89_days_past_due_not_worse', 'number_of_dependents'])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "clf = RandomForestClassifier(compute_importances=True)\n",
      "clf.fit(df[features], df['serious_dlqin2yrs'])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": [
        "/usr/local/lib/python2.7/site-packages/sklearn/ensemble/forest.py:783: DeprecationWarning: Setting compute_importances is no longer required as version 0.14. Variable importances are now computed on the fly when accessing the feature_importances_ attribute. This parameter will be removed in 0.16.\n",
        "  DeprecationWarning)\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 5,
       "text": [
        "RandomForestClassifier(bootstrap=True, compute_importances=None,\n",
        "            criterion='gini', max_depth=None, max_features='auto',\n",
        "            min_density=None, min_samples_leaf=1, min_samples_split=2,\n",
        "            n_estimators=10, n_jobs=1, oob_score=False, random_state=None,\n",
        "            verbose=0)"
       ]
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# from the calculated importances, order them from most to least important\n",
      "# and make a barplot so we can visualize what is/isn't important\n",
      "importances = clf.feature_importances_\n",
      "sorted_idx = np.argsort(importances)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 6
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "padding = np.arange(len(features)) + 0.5\n",
      "pl.barh(padding, importances[sorted_idx], align='center')\n",
      "pl.yticks(padding, features[sorted_idx])\n",
      "pl.xlabel(\"Relative Importance\")\n",
      "pl.title(\"Variable Importance\")\n",
      "pl.show()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAl8AAAEZCAYAAACpVkDxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nnS/GGGOMsXbEnS/GGGOMsXb0v+l6eDcDVsB5AAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x109721290>"
       ]
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "So you can see that the best variable is `revolving_utilization_of_unsecured_lines` while the worst is `number_real_estate_loans_or_lines`. There's also a dramatic drop off after `number_of_open_credit_lines_and_loans`. This is where you need to use your own discretion. How many variables should you include in your model?"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "###Engineering Additional Features\n",
      "Feature selection/engineering will likely have the biggest impact in determining the success/failure of your model. Even if you're using the latest and greatest algorithm, if you put in non-important features you're going to get poor results. Remember, it's math, not magic.\n",
      "\n",
      "Feature engineering is a skill that will take time to get the hang of. Sometimes the best way is to just talk to people. Ask questions, brainstorm with others, etc. Oftentimes 2 features might not be helpful when used individually, but when combined they can be extremely powerful.\n",
      "\n"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df['income_bins'] = pd.cut(df.monthly_income, bins=15)\n",
      "pd.value_counts(df['income_bins'])\n",
      "# not very helpful"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 8,
       "text": [
        "(-3008.75, 200583.333]        112392\n",
        "(200583.333, 401166.667]          11\n",
        "(601750, 802333.333]               5\n",
        "(401166.667, 601750]               4\n",
        "(1404083.333, 1604666.667]         1\n",
        "(802333.333, 1002916.667]          1\n",
        "(2808166.667, 3008750]             1\n",
        "dtype: int64"
       ]
      }
     ],
     "prompt_number": 8
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "###Bucketing Continuous Values"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "####Use the cap_value function you wrote previously to cap monthly_income at $15,000"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def cap_values(x, cap):\n",
      "    if x > cap:\n",
      "        return cap\n",
      "    else:\n",
      "        return x\n",
      "    \n",
      "df.monthly_income = df.monthly_income.apply(lambda x: cap_values(x, 15000))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 9
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df.monthly_income.describe()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 10,
       "text": [
        "count    112415.000000\n",
        "mean       5916.167344\n",
        "std        3644.715884\n",
        "min           0.000000\n",
        "25%        3235.000000\n",
        "50%        5200.000000\n",
        "75%        8000.000000\n",
        "max       15000.000000\n",
        "dtype: float64"
       ]
      }
     ],
     "prompt_number": 10
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df['income_bins'] = pd.cut(df.monthly_income, bins=15, labels=False)\n",
      "pd.value_counts(df.income_bins)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 11,
       "text": [
        "3     14614\n",
        "4     14412\n",
        "2     13192\n",
        "5     12034\n",
        "6     10713\n",
        "1      8042\n",
        "7      7839\n",
        "8      5743\n",
        "9      5597\n",
        "14     5439\n",
        "0      4465\n",
        "10     4307\n",
        "11     2536\n",
        "12     1959\n",
        "13     1523\n",
        "dtype: int64"
       ]
      }
     ],
     "prompt_number": 11
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df[[\"income_bins\", \"serious_dlqin2yrs\"]].groupby(\"income_bins\").mean()"
     ],
     "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>serious_dlqin2yrs</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>income_bins</th>\n",
        "      <th></th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0 </th>\n",
        "      <td> 0.047256</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1 </th>\n",
        "      <td> 0.098980</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2 </th>\n",
        "      <td> 0.092480</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3 </th>\n",
        "      <td> 0.079239</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4 </th>\n",
        "      <td> 0.068485</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>5 </th>\n",
        "      <td> 0.067309</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>6 </th>\n",
        "      <td> 0.057220</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>7 </th>\n",
        "      <td> 0.056767</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>8 </th>\n",
        "      <td> 0.053108</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>9 </th>\n",
        "      <td> 0.048061</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>10</th>\n",
        "      <td> 0.042721</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>11</th>\n",
        "      <td> 0.041009</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>12</th>\n",
        "      <td> 0.040327</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>13</th>\n",
        "      <td> 0.042679</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>14</th>\n",
        "      <td> 0.048538</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 12,
       "text": [
        "             serious_dlqin2yrs\n",
        "income_bins                   \n",
        "0                     0.047256\n",
        "1                     0.098980\n",
        "2                     0.092480\n",
        "3                     0.079239\n",
        "4                     0.068485\n",
        "5                     0.067309\n",
        "6                     0.057220\n",
        "7                     0.056767\n",
        "8                     0.053108\n",
        "9                     0.048061\n",
        "10                    0.042721\n",
        "11                    0.041009\n",
        "12                    0.040327\n",
        "13                    0.042679\n",
        "14                    0.048538"
       ]
      }
     ],
     "prompt_number": 12
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "cols = [\"income_bins\", \"serious_dlqin2yrs\"]\n",
      "df[cols].groupby(\"income_bins\").mean().plot()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 13,
       "text": [
        "<matplotlib.axes.AxesSubplot at 0x10bf4cb50>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXsAAAEQCAYAAABRDdSSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x10970e990>"
       ]
      }
     ],
     "prompt_number": 13
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "cols = ['age', 'serious_dlqin2yrs']\n",
      "age_means = df[cols].groupby(\"age\").mean()\n",
      "age_means.plot()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 14,
       "text": [
        "<matplotlib.axes.AxesSubplot at 0x10bf70350>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEPCAYAAABFpK+YAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x109729790>"
       ]
      }
     ],
     "prompt_number": 14
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "####Bin `age` into 14 different groups. Then make a frequency table that shows the number of customers that were/were not delinquent for each bin.\n",
      "*HINT: You might want to have larger bins near the min/max values to account for outliers.*"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "mybins = [0] + range(20, 80, 5) + [120]\n",
      "df['age_bucket'] = pd.cut(df.age, bins=mybins)\n",
      "pd.value_counts(df['age_bin'])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 24,
       "text": [
        "(45, 50]     14112\n",
        "(50, 55]     13390\n",
        "(55, 60]     12629\n",
        "(60, 65]     12317\n",
        "(40, 45]     12053\n",
        "(35, 40]     10241\n",
        "(65, 70]      8315\n",
        "(30, 35]      8123\n",
        "(75, 120]     7581\n",
        "(25, 30]      5803\n",
        "(70, 75]      5600\n",
        "(20, 25]      2250\n",
        "dtype: int64"
       ]
      }
     ],
     "prompt_number": 24
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "####Using the age bins, calculate the percent of customers that were delinquent for each bucket"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df[[\"age_bucket\", \"serious_dlqin2yrs\"]].groupby(\"age_bucket\").mean()"
     ],
     "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>serious_dlqin2yrs</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>age_bucket</th>\n",
        "      <th></th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>(20, 25]</th>\n",
        "      <td> 0.109778</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>(25, 30]</th>\n",
        "      <td> 0.116319</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>(30, 35]</th>\n",
        "      <td> 0.108211</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>(35, 40]</th>\n",
        "      <td> 0.088956</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>(40, 45]</th>\n",
        "      <td> 0.085124</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>(45, 50]</th>\n",
        "      <td> 0.080995</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>(50, 55]</th>\n",
        "      <td> 0.072890</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>(55, 60]</th>\n",
        "      <td> 0.050598</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>(60, 65]</th>\n",
        "      <td> 0.039864</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>(65, 70]</th>\n",
        "      <td> 0.026458</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>(70, 75]</th>\n",
        "      <td> 0.026607</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>(75, 120]</th>\n",
        "      <td> 0.020314</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 25,
       "text": [
        "            serious_dlqin2yrs\n",
        "age_bucket                   \n",
        "(20, 25]             0.109778\n",
        "(25, 30]             0.116319\n",
        "(30, 35]             0.108211\n",
        "(35, 40]             0.088956\n",
        "(40, 45]             0.085124\n",
        "(45, 50]             0.080995\n",
        "(50, 55]             0.072890\n",
        "(55, 60]             0.050598\n",
        "(60, 65]             0.039864\n",
        "(65, 70]             0.026458\n",
        "(70, 75]             0.026607\n",
        "(75, 120]            0.020314"
       ]
      }
     ],
     "prompt_number": 25
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df[[\"age_bucket\", \"serious_dlqin2yrs\"]].groupby(\"age_bucket\").mean().plot()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 26,
       "text": [
        "<matplotlib.axes.AxesSubplot at 0x1078df650>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXYAAAEQCAYAAACk818iAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nf6YpAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x1078dfc10>"
       ]
      }
     ],
     "prompt_number": 26
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "labels, levels = pd.factorize(df.age_bucket)\n",
      "df.age_bucket = labels"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 27
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "####Write something that buckets debt_ratio into 4 (nearly) equally sized groups.\n",
      "*Hint: use the `quantile` method for Series*"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "bins = []\n",
      "\n",
      "for q in [0.2, 0.4, 0.6, 0.8, 1.0]:\n",
      "    bins.append(df.debt_ratio.quantile(q))\n",
      "\n",
      "debt_ratio_binned = pd.cut(df.debt_ratio, bins=bins)\n",
      "debt_ratio_binned\n",
      "print pd.value_counts(debt_ratio_binned)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "(0.467, 3.838]     22483\n",
        "(3.838, 307001]    22483\n",
        "(0.134, 0.287]     22483\n",
        "(0.287, 0.467]     22483\n",
        "dtype: int64\n"
       ]
      }
     ],
     "prompt_number": 34
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "###Scaling Features\n",
      "Some algorithms will work better if your data is centered around 0. The `StandardScaler` module in `scikit-learn` makes it very easy to quickly scale columns in your data frame."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.preprocessing import StandardScaler\n",
      "\n",
      "df['monthly_income_scaled'] = StandardScaler().fit_transform(df.monthly_income)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 35
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print df.monthly_income_scaled.describe()\n",
      "print\n",
      "print \"Mean at 0?\", round(df.monthly_income_scaled.mean(), 10)==0\n",
      "\n",
      "pl.hist(df.monthly_income_scaled)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "count    1.124150e+05\n",
        "mean     1.023365e-15\n",
        "std      1.000004e+00\n",
        "min     -1.623225e+00\n",
        "25%     -7.356346e-01\n",
        "50%     -1.964956e-01\n",
        "75%      5.717433e-01\n",
        "max      2.492341e+00\n",
        "dtype: float64\n",
        "\n",
        "Mean at 0? True\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 36,
       "text": [
        "(array([  8669.,  15715.,  22969.,  17668.,  15316.,  10149.,   7454.,\n",
        "         5079.,   2939.,   6457.]),\n",
        " array([-1.62322492, -1.21166838, -0.80011184, -0.38855529,  0.02300125,\n",
        "        0.4345578 ,  0.84611434,  1.25767088,  1.66922743,  2.08078397,\n",
        "        2.49234051]),\n",
        " <a list of 10 Patch objects>)"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAEACAYAAAC3adEgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAGcZJREFUeJzt3X9Mlef9//HXWWHrZ1GsmHrAc1xo4VBE8Mjmji6ZCw09\nVkikdnRStiq29I/iftiuaUz6z2iyCs2yLbbVtGloxrpE7JpNTFYJi9nZj25iZjFZpKlnLSockG8n\nxR5bC6Vc3z+sJ73kiIcf59xYno/kJIfrXPd9v6/rHM7r3Dfn5nYZY4wAAPjMl5wuAAAwtxAMAAAL\nwQAAsBAMAAALwQAAsBAMAADLpMHQ29urO++8UytXrlRRUZGeffZZSVJDQ4O8Xq9KSkpUUlKiw4cP\nx5ZpbGyUz+dTQUGBOjo6Yu3Hjx9XcXGxfD6fdu7cGWsfGRlRdXW1fD6f1q1bpzNnzsz2GAEAU2Em\nMTAwYLq6uowxxkSjUZOfn2+6u7tNQ0OD+eUvfzmh/8mTJ43f7zejo6Omp6fH5ObmmvHxcWOMMd/8\n5jdNZ2enMcaY8vJyc/jwYWOMMXv37jX19fXGGGNaW1tNdXX1ZCUBAJJs0j2GrKwsrV69WpK0YMEC\nrVixQpFI5EqgTOjf1tammpoapaenKycnR3l5eers7NTAwICi0agCgYAkadu2bTp48KAk6dChQ6qt\nrZUkVVVV6ciRI7OXegCAKUv4bwynT59WV1eX1q1bJ0l67rnn5Pf7VVdXp+HhYUlSf3+/vF5vbBmv\n16tIJDKh3ePxxAImEolo+fLlkqS0tDQtWrRIQ0NDMx8ZAGBaEgqGixcv6r777tOePXu0YMEC1dfX\nq6enRydOnFB2drYef/zxZNcJAEiRtOt1+OSTT1RVVaUHHnhAmzdvliQtXbo09vjDDz+sTZs2Sbq8\nJ9Db2xt7rK+vT16vVx6PR319fRParyxz9uxZLVu2TGNjY7pw4YIyMzMn1JGXl6d33nlnmsMEgPkn\nNzdX//3vf6e83KR7DMYY1dXVqbCwUI8++misfWBgIHb/j3/8o4qLiyVJlZWVam1t1ejoqHp6ehQO\nhxUIBJSVlaWMjAx1dnbKGKNXXnlF99xzT2yZlpYWSdJrr72msrKyuLW88847MsbMqdvPfvYzx2ug\npi9OTXO1Lmq6cWua7ofpSfcY3njjDf3ud7/TqlWrVFJSIknavXu39u/frxMnTsjlcum2227Tiy++\nKEkqLCzUli1bVFhYqLS0NO3bt08ul0uStG/fPm3fvl2XLl1SRUWFNm7cKEmqq6vT1q1b5fP5tGTJ\nErW2tk5rIACA2TFpMHz729/W+Pj4hPby8vJrLvPkk0/qySefnND+jW98Q//5z38mtH/lK1/Rq6++\nmkitAIAU4MznGSgtLXW6hAmoKTFzsSZpbtZFTYmZizVNl8sYc0NcqMflcukGKRUA5oTpvm+yxwAA\nsBAMAAALwQAAsBAMAAALwQAAsBAMAAALwQAAsBAMAAALwQAAsBAMAAALwQAAsBAMAADLda/ghvkj\nIyNT0ej7Kd/uwoWL9cEHXOcbmCv476qIuXxRJSfmmOcWSAb+uyoAYFYQDAAAC8EAALAQDAAAC8EA\nALAQDAAAC8EAALAQDAAAC8EAALAQDAAAC8EAALAQDAAAC8EAALAQDAAAC8EAALAQDAAAC8EAALAQ\nDAAAC8EAALAQDAAAy6TB0NvbqzvvvFMrV65UUVGRnn32WUnS0NCQgsGg8vPztWHDBg0PD8eWaWxs\nlM/nU0FBgTo6OmLtx48fV3FxsXw+n3bu3BlrHxkZUXV1tXw+n9atW6czZ87M9hgBAFMwaTCkp6fr\n17/+tU6ePKmjR49q7969euutt9TU1KRgMKhTp06prKxMTU1NkqTu7m4dOHBA3d3dam9v144dO2SM\nkSTV19erublZ4XBY4XBY7e3tkqTm5mYtWbJE4XBYjz32mHbt2pXkIQMAJjNpMGRlZWn16tWSpAUL\nFmjFihWKRCI6dOiQamtrJUm1tbU6ePCgJKmtrU01NTVKT09XTk6O8vLy1NnZqYGBAUWjUQUCAUnS\ntm3bYst8fl1VVVU6cuRIckYKAEhIwn9jOH36tLq6urR27VoNDg7K7XZLktxutwYHByVJ/f398nq9\nsWW8Xq8ikciEdo/Ho0gkIkmKRCJavny5JCktLU2LFi3S0NDQzEcGAJiWtEQ6Xbx4UVVVVdqzZ48W\nLlxoPeZyueRyuZJS3NUaGhpi90tLS1VaWpqS7QLAjSAUCikUCs14PdcNhk8++URVVVXaunWrNm/e\nLOnyXsK5c+eUlZWlgYEBLV26VNLlPYHe3t7Ysn19ffJ6vfJ4POrr65vQfmWZs2fPatmyZRobG9OF\nCxeUmZkZt5bPBwMAwHb1B+annnpqWuuZ9FCSMUZ1dXUqLCzUo48+GmuvrKxUS0uLJKmlpSUWGJWV\nlWptbdXo6Kh6enoUDocVCASUlZWljIwMdXZ2yhijV155Rffcc8+Edb322msqKyub1kAAALPDZa58\nbSiOf/zjH/rOd76jVatWxQ4XNTY2KhAIaMuWLTp79qxycnL06quv6pZbbpEk7d69Wy+//LLS0tK0\nZ88e3X333ZIuf111+/btunTpkioqKmJffR0ZGdHWrVvV1dWlJUuWqLW1VTk5ORMLdbk0SamYBZef\nYyfmmOcWSIbpvm9OGgxzCcGQfAQD8MUy3fdNznwGAFgIBgCAhWAAAFgIBgCAhWAAAFgIBgCAhWAA\nAFgIBgCAhWAAAFgIBgCAhWAAAFgIBgCAhWAAAFgIBgCAhWAAAFgIBgCA5brXfAaSLy12hcBUW7hw\nsT74YMiRbQNzFVdwQ4yTV3BzZruXt83rCl9UXMENADArCAYAgIVgAABYCAYAgIVgAABYCAYAgIVg\nAABYCAYAgIVgAABYCAYAgIVgAABYCAYAgIVgAABYCAYAgIVgAABYuFDPHJORkalo9H2nywAwj3Gh\nnjnGuYvlSM5dMIcL9QDJkLQL9Tz00ENyu90qLi6OtTU0NMjr9aqkpEQlJSU6fPhw7LHGxkb5fD4V\nFBSoo6Mj1n78+HEVFxfL5/Np586dsfaRkRFVV1fL5/Np3bp1OnPmzJQHAQCYPdcNhgcffFDt7e1W\nm8vl0k9/+lN1dXWpq6tL5eXlkqTu7m4dOHBA3d3dam9v144dO2JpVV9fr+bmZoXDYYXD4dg6m5ub\ntWTJEoXDYT322GPatWvXbI8RADAF1w2G9evXa/HixRPa4+2etLW1qaamRunp6crJyVFeXp46Ozs1\nMDCgaDSqQCAgSdq2bZsOHjwoSTp06JBqa2slSVVVVTpy5MiMBgRMTZpcLpcjt4yMTKcHD8Q17W8l\nPffcc/L7/aqrq9Pw8LAkqb+/X16vN9bH6/UqEolMaPd4PIpEIpKkSCSi5cuXS5LS0tK0aNEiDQ0N\nTbcsYIrGdPnvG6m/8SUDzFXTCob6+nr19PToxIkTys7O1uOPPz7bdQEAHDKtr6suXbo0dv/hhx/W\npk2bJF3eE+jt7Y091tfXJ6/XK4/Ho76+vgntV5Y5e/asli1bprGxMV24cEGZmfF3sRsaGmL3S0tL\nVVpaOp3yAeALKRQKKRQKzXxFJgE9PT2mqKgo9nN/f3/s/q9+9StTU1NjjDHm5MmTxu/3m5GREfPu\nu++a22+/3YyPjxtjjAkEAubo0aNmfHzclJeXm8OHDxtjjNm7d6955JFHjDHG7N+/31RXV8etIcFS\nb3iSjGQcujm17fk45vnzmoZzpvsau+4eQ01Njf7617/qf//7n5YvX66nnnpKoVBIJ06ckMvl0m23\n3aYXX3xRklRYWKgtW7aosLBQaWlp2rdv32ffy5f27dun7du369KlS6qoqNDGjRslSXV1ddq6dat8\nPp+WLFmi1tbWmacdAGDaOMFtjuEEt/m17fnwmoZzknaCGwBgfiEYAAAWggEAYCEYAAAWggEAYCEY\nAAAWggEAYCEYAAAWggEAYCEYAAAWggEAYCEYAAAWggEAYCEYAAAWggEAYCEYAAAWggEAYCEYAAAW\nggEAYCEYAAAWggEAYCEYAAAWggEAYCEYAAAWggEAYCEYAAAWggEAYCEYAAAWggEAYCEYAAAWggEA\nYCEYAAAWggEAYCEYAAAWggEAYCEYAACW6wbDQw89JLfbreLi4ljb0NCQgsGg8vPztWHDBg0PD8ce\na2xslM/nU0FBgTo6OmLtx48fV3FxsXw+n3bu3BlrHxkZUXV1tXw+n9atW6czZ87M1tgAANNw3WB4\n8MEH1d7ebrU1NTUpGAzq1KlTKisrU1NTkySpu7tbBw4cUHd3t9rb27Vjxw4ZYyRJ9fX1am5uVjgc\nVjgcjq2zublZS5YsUTgc1mOPPaZdu3bN9hgBAFNw3WBYv369Fi9ebLUdOnRItbW1kqTa2lodPHhQ\nktTW1qaamhqlp6crJydHeXl56uzs1MDAgKLRqAKBgCRp27ZtsWU+v66qqiodOXJk9kYHAJiyaf2N\nYXBwUG63W5Lkdrs1ODgoServ75fX643183q9ikQiE9o9Ho8ikYgkKRKJaPny5ZKktLQ0LVq0SEND\nQ9MbDXBDSZPL5Ur5LSMj0+mBY45Lm+kKrrzYAEzVmCST8q1Go/y+YnLTCga3261z584pKytLAwMD\nWrp0qaTLewK9vb2xfn19ffJ6vfJ4POrr65vQfmWZs2fPatmyZRobG9OFCxeUmRn/E01DQ0Psfmlp\nqUpLS6dTPgB8IYVCIYVCoZmvyCSgp6fHFBUVxX5+4oknTFNTkzHGmMbGRrNr1y5jjDEnT540fr/f\njIyMmHfffdfcfvvtZnx83BhjTCAQMEePHjXj4+OmvLzcHD582BhjzN69e80jjzxijDFm//79prq6\nOm4NCZZ6w5NkJOPQzaltz8cxOzvfmB+m+1xfd6n777/fZGdnm/T0dOP1es3LL79szp8/b8rKyozP\n5zPBYNC8//77sf5PP/20yc3NNXfccYdpb2+Ptf/73/82RUVFJjc31/z4xz+OtX/88cfme9/7nsnL\nyzNr1641PT09szrAG818faOaf2N2dr4xP0z3uXZ9tvCc53K5dIOUOiOX/17j1Did2vZ8HLOT254f\nv0uY/vsmZz4DACwEAwDAQjAAACwEAwDAQjAAACwEAwDAMuN/ifFFFI1GdfHiRafLAABHEAxxrFhR\novPnP5DLdVNKtzs+PpLS7QFAPARDHB9+eEkff9wlyZPiLXdIujvF2wQAG39jAABYCAYAgIVgAABY\nCAYAgIVgAABYCAYAgIVgAABYCAYAgIVgAABYOPMZmHfSPruEbOotXLhYH3ww5Mi2kTiCAZh3xuTU\nda6jUWcCCVPDoSQAgIVgAABYCAYAgIVgAABYCAYAgIVgAABYCAYAgIXzGAAgiTIyMhWNvu90GVNC\nMABAEl0OBWdOKJSmd0Ihh5IAABaCAQBgIRgAABaCAQBgIRgAABaCAQBgIRgAAJYZBUNOTo5WrVql\nkpISBQIBSdLQ0JCCwaDy8/O1YcMGDQ8Px/o3NjbK5/OpoKBAHR0dsfbjx4+ruLhYPp9PO3funElJ\nAIAZmlEwuFwuhUIhdXV16dixY5KkpqYmBYNBnTp1SmVlZWpqapIkdXd368CBA+ru7lZ7e7t27Ngh\nYy6f9FFfX6/m5maFw2GFw2G1t7fPcFgAgOma8aGkK2/uVxw6dEi1tbWSpNraWh08eFCS1NbWppqa\nGqWnpysnJ0d5eXnq7OzUwMCAotFobI9j27ZtsWUAAKk34z2Gu+66S2vWrNFLL70kSRocHJTb7ZYk\nud1uDQ4OSpL6+/vl9Xpjy3q9XkUikQntHo9HkUhkJmUBmLPS5HK5HLllZGQ6Pfgbxoz+V9Ibb7yh\n7OxsvffeewoGgyooKLAev/KEzJaGhobY/dLSUpWWls7augGkwpic+r9B0ejsvRfNXaHPbjMzo2DI\nzs6WJN1666269957dezYMbndbp07d05ZWVkaGBjQ0qVLJV3eE+jt7Y0t29fXJ6/XK4/Ho76+Pqvd\n4/HE3d7ngwEAcLXSz25XPDWttUz7UNJHH32kaDQqSfrwww/V0dGh4uJiVVZWqqWlRZLU0tKizZs3\nS5IqKyvV2tqq0dFR9fT0KBwOKxAIKCsrSxkZGers7JQxRq+88kpsGQBA6k17j2FwcFD33nuvJGls\nbEw/+MEPtGHDBq1Zs0ZbtmxRc3OzcnJy9Oqrr0qSCgsLtWXLFhUWFiotLU379u2LHWbat2+ftm/f\nrkuXLqmiokIbN26chaEBAKbDZa7+WtEc5XK5JnwDKlkWL/ZoePiYpPiHtJKnQ9LdcvZ/tzuxbae2\nO1+3PR/HfHnbTrzdXf4AfGONmTOfAQAWggEAYCEYAAAWggEAYCEYAAAWggEAYCEYAAAWggEAYCEY\nAAAWggEAYCEYAAAWggEAYCEYAAAWggEAYJnRFdwA4MaRNquXGv4iIxgAzBNOXW/6xgsjDiUBACwE\nAwDAQjAAACwEAwDAQjAAACwEAwDAQjAAACwEAwDAQjAAACwEAwDAQjAAACwEAwDAQjAAACwEAwDA\nQjAAACwEAwDAQjAAACwEAwDAQjAAACxzJhja29tVUFAgn8+nZ555xulyAGDemhPB8Omnn+pHP/qR\n2tvb1d3drf379+utt95yuqwEhJwuII6Q0wXEEXK6gDhCThdwDSGnC4gj5HQBcYScLiCOkNMFzJo5\nEQzHjh1TXl6ecnJylJ6ervvvv19tbW1Ol5WAkNMFxBFyuoA4Qk4XEEfI6QKuIeR0AXGEnC4gjpDT\nBcQRcrqAWTMngiESiWj58uWxn71eryKRiIMVAcD8leZ0AZLkcrmcLsFy001f0sKFtXK5/m/Sfh9/\n/LZuvvn4rG13bOz/6aOPZm11ADAtcyIYPB6Pent7Yz/39vbK6/VafXJzc1McIH0J9RodDSdh2zMd\n51MObvtarleTEx8OrtTk5AeTeNueyfM3k+1OZjZrmq35nk5NyX6ur1WTM6+x3NzcaS3nMsaYWa5l\nysbGxnTHHXfoyJEjWrZsmQKBgPbv368VK1Y4XRoAzDtzYo8hLS1Nzz//vO6++259+umnqqurIxQA\nwCFzYo8BADB3zIlvJV3tiSee0IoVK+T3+/Xd735XFy5ciNsv1SfF/f73v9fKlSt100036c0337xm\nv5ycHK1atUolJSUKBAJzoqZUztXQ0JCCwaDy8/O1YcMGDQ8Px+2XinlKZNw/+clP5PP55Pf71dXV\nlZQ6plJTKBTSokWLVFJSopKSEv385z9Pek0PPfSQ3G63iouLr9kn1fN0vZqcmKfe3l7deeedWrly\npYqKivTss8/G7ZfKuUqkpinPlZmDOjo6zKeffmqMMWbXrl1m165dE/qMjY2Z3Nxc09PTY0ZHR43f\n7zfd3d1Jreutt94yb7/9tiktLTXHjx+/Zr+cnBxz/vz5pNYylZpSPVdPPPGEeeaZZ4wxxjQ1NcV9\n/oxJ/jwlMu4//elPpry83BhjzNGjR83atWuTVk+iNf3lL38xmzZtSmodV/vb3/5m3nzzTVNUVBT3\n8VTPUyI1OTFPAwMDpquryxhjTDQaNfn5+Y6/phKpaapzNSf3GILBoL70pculrV27Vn19E78h5MRJ\ncQUFBcrPz0+or0nREbpEakr1XB06dEi1tbWSpNraWh08ePCafZM5T4mM+/O1rl27VsPDwxocHHS0\nJil1r58r1q9fr8WLF1/z8VTPUyI1Samfp6ysLK1evVqStGDBAq1YsUL9/f1Wn1TPVSI1SVObqzkZ\nDJ/38ssvq6KiYkL7XD4pzuVy6a677tKaNWv00ksvOV1OyudqcHBQbrdbkuR2u6/5S5HseUpk3PH6\nxPsgksqaXC6X/vnPf8rv96uiokLd3d1JqydRqZ6nRDg9T6dPn1ZXV5fWrl1rtTs5V9eqaapz5di3\nkoLBoM6dOzehfffu3dq0aZMk6emnn9aXv/xlff/735/QL1nnNCRS1/W88cYbys7O1nvvvadgMKiC\nggKtX7/esZqSMVfXqunpp5+esO1rbX+25+lqiY776k9SyTxfJpF1f/3rX1dvb6+++tWv6vDhw9q8\nebNOnTqVtJoSlcp5SoST83Tx4kXdd9992rNnjxYsWDDhcSfmarKapjpXjgXDn//850kf/81vfqPX\nX39dR44cift4IifFJaOuRGRnZ0uSbr31Vt177706duzYjN7wZlpTMuZqsprcbrfOnTunrKwsDQwM\naOnSpXH7zfY8XS2RcV/dp6+vTx6PZ9ZqmE5NCxcujN0vLy/Xjh07NDQ0pMzMzKTVdT2pnqdEODVP\nn3zyiaqqqvTAAw9o8+bNEx53Yq6uV9NU52pOHkpqb2/XL37xC7W1tenmm2+O22fNmjUKh8M6ffq0\nRkdHdeDAAVVWVqasxmsdr/voo48UjUYlSR9++KE6Ojom/aZHKmpK9VxVVlaqpaVFktTS0hL3hZqK\neUpk3JWVlfrtb38rSTp69KhuueWW2GGwZEikpsHBwdhzeezYMRljHA0FKfXzlAgn5skYo7q6OhUW\nFurRRx+N2yfVc5VITVOeqxn8MTxp8vLyzNe+9jWzevVqs3r1alNfX2+MMSYSiZiKiopYv9dff93k\n5+eb3Nxcs3v37qTX9Yc//MF4vV5z8803G7fbbTZu3Dihrnfeecf4/X7j9/vNypUrk15XIjUZk9q5\nOn/+vCkrKzM+n88Eg0Hz/vvvT6gpVfMUb9wvvPCCeeGFF2J9fvjDH5rc3FyzatWqSb9tlqqann/+\nebNy5Urj9/vNt771LfOvf/0r6TXdf//9Jjs726Snpxuv12uam5sdn6fr1eTEPP397383LpfL+P3+\n2PvT66+/7uhcJVLTVOeKE9wAAJY5eSgJAOAcggEAYCEYAAAWggEAYCEYAAAWggEAYCEYAAAWggEA\nYPn/GO7gc7SyCyIAAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x1078cc050>"
       ]
      }
     ],
     "prompt_number": 36
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "##Redo our feature importance"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "features = np.array(['revolving_utilization_of_unsecured_lines',\n",
      "                     'age', 'number_of_time30-59_days_past_due_not_worse',\n",
      "                     'debt_ratio', 'monthly_income','number_of_open_credit_lines_and_loans', \n",
      "                     'number_of_times90_days_late', 'number_real_estate_loans_or_lines',\n",
      "                     'number_of_time60-89_days_past_due_not_worse', 'number_of_dependents',\n",
      "                     'income_bins', 'age_bucket', 'monthly_income_scaled'])\n",
      "\n",
      "clf = RandomForestClassifier(compute_importances=True)\n",
      "clf.fit(df[features], df['serious_dlqin2yrs'])\n",
      "\n",
      "importances = clf.feature_importances_\n",
      "sorted_idx = np.argsort(importances)\n",
      "\n",
      "padding = np.arange(len(features)) + 0.5\n",
      "pl.barh(padding, importances[sorted_idx], align='center')\n",
      "pl.yticks(padding, features[sorted_idx])\n",
      "pl.xlabel(\"Relative Importance\")\n",
      "pl.title(\"Variable Importance\")\n",
      "pl.show()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": [
        "/usr/local/lib/python2.7/site-packages/sklearn/ensemble/forest.py:783: DeprecationWarning: Setting compute_importances is no longer required as version 0.14. Variable importances are now computed on the fly when accessing the feature_importances_ attribute. This parameter will be removed in 0.16.\n",
        "  DeprecationWarning)\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAl8AAAEZCAYAAACpVkDxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\njDHGmtD/AwmVkB3EkoWpAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x1081a49d0>"
       ]
      }
     ],
     "prompt_number": 37
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "best_features = features[sorted_idx][::-1]\n",
      "best_features"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 24,
       "text": [
        "array(['revolving_utilization_of_unsecured_lines', 'debt_ratio',\n",
        "       'monthly_income', 'number_of_times90_days_late', 'age',\n",
        "       'monthly_income_scaled', 'number_of_open_credit_lines_and_loans',\n",
        "       'number_of_time30-59_days_past_due_not_worse',\n",
        "       'number_of_time60-89_days_past_due_not_worse', 'age_bucket',\n",
        "       'number_of_dependents', 'number_real_estate_loans_or_lines',\n",
        "       'income_bins'], \n",
        "      dtype='|S43')"
       ]
      }
     ],
     "prompt_number": 24
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "##Test out your own features\n",
      "Partner with the person sitting next to you and see if you can come up with some new features that outperform the basic variables.\n",
      "\n",
      "- brainstorm new features\n",
      "- create them using pandas\n",
      "- test their effectiveness w/ scikit-learn\n",
      "\n",
      "Things you might try:\n",
      "\n",
      "- Try using some of the scikit-learn [preprecessing functions](http://scikit-learn.org/stable/modules/preprocessing.html#preprocessing)\n",
      "- Look at `pd.get_dummies`\n",
      "- Combine features together (total times late, log(monthly_income), age^2)"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "##We just did the following\n",
      "\n",
      "- Ranked feature importance using RandomForest\n",
      "- Created some of our own features using `pandas` and `scikit-learn`\n",
      "- Scaled data using `scikit-learn`"
     ]
    }
   ],
   "metadata": {}
  }
 ]
}