{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Cross validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn import metrics\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.linear_model import LogisticRegressionCV\n",
    "from sklearn.cross_validation import KFold, cross_val_score\n",
    "from sklearn.cross_validation import train_test_split\n",
    "%matplotlib inline \n",
    "#Don't use this command in canopy or in python\n",
    "\n",
    "path = 'data/adult.csv'\n",
    "data = pd.read_csv(path)\n",
    "# remove rows where occupation is unknown\n",
    "data = data[data.occupation != '?']\n",
    "raw_data = data[data.occupation != '?']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>workclass</th>\n",
       "      <th>fnlwgt</th>\n",
       "      <th>education</th>\n",
       "      <th>education.num</th>\n",
       "      <th>marital.status</th>\n",
       "      <th>occupation</th>\n",
       "      <th>relationship</th>\n",
       "      <th>race</th>\n",
       "      <th>sex</th>\n",
       "      <th>capital.gain</th>\n",
       "      <th>capital.loss</th>\n",
       "      <th>hours.per.week</th>\n",
       "      <th>native.country</th>\n",
       "      <th>income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>82</td>\n",
       "      <td>Private</td>\n",
       "      <td>132870</td>\n",
       "      <td>HS-grad</td>\n",
       "      <td>9</td>\n",
       "      <td>Widowed</td>\n",
       "      <td>Exec-managerial</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>White</td>\n",
       "      <td>Female</td>\n",
       "      <td>0</td>\n",
       "      <td>4356</td>\n",
       "      <td>18</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>54</td>\n",
       "      <td>Private</td>\n",
       "      <td>140359</td>\n",
       "      <td>7th-8th</td>\n",
       "      <td>4</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>Machine-op-inspct</td>\n",
       "      <td>Unmarried</td>\n",
       "      <td>White</td>\n",
       "      <td>Female</td>\n",
       "      <td>0</td>\n",
       "      <td>3900</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>41</td>\n",
       "      <td>Private</td>\n",
       "      <td>264663</td>\n",
       "      <td>Some-college</td>\n",
       "      <td>10</td>\n",
       "      <td>Separated</td>\n",
       "      <td>Prof-specialty</td>\n",
       "      <td>Own-child</td>\n",
       "      <td>White</td>\n",
       "      <td>Female</td>\n",
       "      <td>0</td>\n",
       "      <td>3900</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>34</td>\n",
       "      <td>Private</td>\n",
       "      <td>216864</td>\n",
       "      <td>HS-grad</td>\n",
       "      <td>9</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>Other-service</td>\n",
       "      <td>Unmarried</td>\n",
       "      <td>White</td>\n",
       "      <td>Female</td>\n",
       "      <td>0</td>\n",
       "      <td>3770</td>\n",
       "      <td>45</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>38</td>\n",
       "      <td>Private</td>\n",
       "      <td>150601</td>\n",
       "      <td>10th</td>\n",
       "      <td>6</td>\n",
       "      <td>Separated</td>\n",
       "      <td>Adm-clerical</td>\n",
       "      <td>Unmarried</td>\n",
       "      <td>White</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>3770</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   age workclass  fnlwgt     education  education.num marital.status  \\\n",
       "1   82   Private  132870       HS-grad              9        Widowed   \n",
       "3   54   Private  140359       7th-8th              4       Divorced   \n",
       "4   41   Private  264663  Some-college             10      Separated   \n",
       "5   34   Private  216864       HS-grad              9       Divorced   \n",
       "6   38   Private  150601          10th              6      Separated   \n",
       "\n",
       "          occupation   relationship   race     sex  capital.gain  \\\n",
       "1    Exec-managerial  Not-in-family  White  Female             0   \n",
       "3  Machine-op-inspct      Unmarried  White  Female             0   \n",
       "4     Prof-specialty      Own-child  White  Female             0   \n",
       "5      Other-service      Unmarried  White  Female             0   \n",
       "6       Adm-clerical      Unmarried  White    Male             0   \n",
       "\n",
       "   capital.loss  hours.per.week native.country income  \n",
       "1          4356              18  United-States  <=50K  \n",
       "3          3900              40  United-States  <=50K  \n",
       "4          3900              40  United-States  <=50K  \n",
       "5          3770              45  United-States  <=50K  \n",
       "6          3770              40  United-States  <=50K  "
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>workclass</th>\n",
       "      <th>fnlwgt</th>\n",
       "      <th>education</th>\n",
       "      <th>education.num</th>\n",
       "      <th>marital.status</th>\n",
       "      <th>occupation</th>\n",
       "      <th>relationship</th>\n",
       "      <th>race</th>\n",
       "      <th>sex</th>\n",
       "      <th>...</th>\n",
       "      <th>capital.loss</th>\n",
       "      <th>hours.per.week</th>\n",
       "      <th>native.country</th>\n",
       "      <th>income</th>\n",
       "      <th>workclass_num</th>\n",
       "      <th>over50K</th>\n",
       "      <th>marital_num</th>\n",
       "      <th>race_num</th>\n",
       "      <th>sex_num</th>\n",
       "      <th>rel_num</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>82</td>\n",
       "      <td>Private</td>\n",
       "      <td>132870</td>\n",
       "      <td>HS-grad</td>\n",
       "      <td>9</td>\n",
       "      <td>Widowed</td>\n",
       "      <td>Exec-managerial</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>White</td>\n",
       "      <td>Female</td>\n",
       "      <td>...</td>\n",
       "      <td>4356</td>\n",
       "      <td>18</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>54</td>\n",
       "      <td>Private</td>\n",
       "      <td>140359</td>\n",
       "      <td>7th-8th</td>\n",
       "      <td>4</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>Machine-op-inspct</td>\n",
       "      <td>Unmarried</td>\n",
       "      <td>White</td>\n",
       "      <td>Female</td>\n",
       "      <td>...</td>\n",
       "      <td>3900</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>41</td>\n",
       "      <td>Private</td>\n",
       "      <td>264663</td>\n",
       "      <td>Some-college</td>\n",
       "      <td>10</td>\n",
       "      <td>Separated</td>\n",
       "      <td>Prof-specialty</td>\n",
       "      <td>Own-child</td>\n",
       "      <td>White</td>\n",
       "      <td>Female</td>\n",
       "      <td>...</td>\n",
       "      <td>3900</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>34</td>\n",
       "      <td>Private</td>\n",
       "      <td>216864</td>\n",
       "      <td>HS-grad</td>\n",
       "      <td>9</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>Other-service</td>\n",
       "      <td>Unmarried</td>\n",
       "      <td>White</td>\n",
       "      <td>Female</td>\n",
       "      <td>...</td>\n",
       "      <td>3770</td>\n",
       "      <td>45</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>38</td>\n",
       "      <td>Private</td>\n",
       "      <td>150601</td>\n",
       "      <td>10th</td>\n",
       "      <td>6</td>\n",
       "      <td>Separated</td>\n",
       "      <td>Adm-clerical</td>\n",
       "      <td>Unmarried</td>\n",
       "      <td>White</td>\n",
       "      <td>Male</td>\n",
       "      <td>...</td>\n",
       "      <td>3770</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   age workclass  fnlwgt     education  education.num marital.status  \\\n",
       "1   82   Private  132870       HS-grad              9        Widowed   \n",
       "3   54   Private  140359       7th-8th              4       Divorced   \n",
       "4   41   Private  264663  Some-college             10      Separated   \n",
       "5   34   Private  216864       HS-grad              9       Divorced   \n",
       "6   38   Private  150601          10th              6      Separated   \n",
       "\n",
       "          occupation   relationship   race     sex   ...     capital.loss  \\\n",
       "1    Exec-managerial  Not-in-family  White  Female   ...             4356   \n",
       "3  Machine-op-inspct      Unmarried  White  Female   ...             3900   \n",
       "4     Prof-specialty      Own-child  White  Female   ...             3900   \n",
       "5      Other-service      Unmarried  White  Female   ...             3770   \n",
       "6       Adm-clerical      Unmarried  White    Male   ...             3770   \n",
       "\n",
       "   hours.per.week  native.country income workclass_num  over50K  marital_num  \\\n",
       "1              18   United-States  <=50K             0        0            0   \n",
       "3              40   United-States  <=50K             0        0            1   \n",
       "4              40   United-States  <=50K             0        0            2   \n",
       "5              45   United-States  <=50K             0        0            1   \n",
       "6              40   United-States  <=50K             0        0            2   \n",
       "\n",
       "   race_num  sex_num  rel_num  \n",
       "1         0        0        0  \n",
       "3         0        0        0  \n",
       "4         0        0        0  \n",
       "5         0        0        0  \n",
       "6         0        1        0  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#First we need to create dummy variables\n",
    "data['workclass_num'] = data.workclass.map({'Private':0, 'State-gov':1, 'Federal-gov':2, 'Self-emp-not-inc':3, 'Self-emp-inc':4, 'Local-gov':5, 'Without-pay':6})\n",
    "data['over50K'] = np.where(data.income == '<=50K', 0, 1)\n",
    "data['marital_num'] = data['marital.status'].map({'Widowed':0, 'Divorced':1, 'Separated':2, 'Never-married':3, 'Married-civ-spouse':4, 'Married-AF-spouse':4, 'Married-spouse-absent':5})\n",
    "data['race_num'] = data.race.map({'White':0, 'Black':1, 'Asian-Pac-Islander':2, 'Amer-Indian-Eskimo':3, 'Other':4})\n",
    "data['sex_num'] = np.where(data.sex == 'Female', 0, 1)\n",
    "data['rel_num'] = data.relationship.map({'Not-in-family':0, 'Unmarried':0, 'Own-child':0, 'Other-relative':0, 'Husband':1, 'Wife':1})\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>workclass</th>\n",
       "      <th>fnlwgt</th>\n",
       "      <th>education</th>\n",
       "      <th>education.num</th>\n",
       "      <th>marital.status</th>\n",
       "      <th>occupation</th>\n",
       "      <th>relationship</th>\n",
       "      <th>race</th>\n",
       "      <th>sex</th>\n",
       "      <th>...</th>\n",
       "      <th>capital.loss</th>\n",
       "      <th>hours.per.week</th>\n",
       "      <th>native.country</th>\n",
       "      <th>income</th>\n",
       "      <th>workclass_num</th>\n",
       "      <th>over50K</th>\n",
       "      <th>marital_num</th>\n",
       "      <th>race_num</th>\n",
       "      <th>sex_num</th>\n",
       "      <th>rel_num</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>82</td>\n",
       "      <td>Private</td>\n",
       "      <td>132870</td>\n",
       "      <td>HS-grad</td>\n",
       "      <td>9</td>\n",
       "      <td>Widowed</td>\n",
       "      <td>Exec-managerial</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>White</td>\n",
       "      <td>Female</td>\n",
       "      <td>...</td>\n",
       "      <td>4356</td>\n",
       "      <td>18</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>54</td>\n",
       "      <td>Private</td>\n",
       "      <td>140359</td>\n",
       "      <td>7th-8th</td>\n",
       "      <td>4</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>Machine-op-inspct</td>\n",
       "      <td>Unmarried</td>\n",
       "      <td>White</td>\n",
       "      <td>Female</td>\n",
       "      <td>...</td>\n",
       "      <td>3900</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>41</td>\n",
       "      <td>Private</td>\n",
       "      <td>264663</td>\n",
       "      <td>Some-college</td>\n",
       "      <td>10</td>\n",
       "      <td>Separated</td>\n",
       "      <td>Prof-specialty</td>\n",
       "      <td>Own-child</td>\n",
       "      <td>White</td>\n",
       "      <td>Female</td>\n",
       "      <td>...</td>\n",
       "      <td>3900</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>34</td>\n",
       "      <td>Private</td>\n",
       "      <td>216864</td>\n",
       "      <td>HS-grad</td>\n",
       "      <td>9</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>Other-service</td>\n",
       "      <td>Unmarried</td>\n",
       "      <td>White</td>\n",
       "      <td>Female</td>\n",
       "      <td>...</td>\n",
       "      <td>3770</td>\n",
       "      <td>45</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>38</td>\n",
       "      <td>Private</td>\n",
       "      <td>150601</td>\n",
       "      <td>10th</td>\n",
       "      <td>6</td>\n",
       "      <td>Separated</td>\n",
       "      <td>Adm-clerical</td>\n",
       "      <td>Unmarried</td>\n",
       "      <td>White</td>\n",
       "      <td>Male</td>\n",
       "      <td>...</td>\n",
       "      <td>3770</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   age workclass  fnlwgt     education  education.num marital.status  \\\n",
       "1   82   Private  132870       HS-grad              9        Widowed   \n",
       "3   54   Private  140359       7th-8th              4       Divorced   \n",
       "4   41   Private  264663  Some-college             10      Separated   \n",
       "5   34   Private  216864       HS-grad              9       Divorced   \n",
       "6   38   Private  150601          10th              6      Separated   \n",
       "\n",
       "          occupation   relationship   race     sex   ...     capital.loss  \\\n",
       "1    Exec-managerial  Not-in-family  White  Female   ...             4356   \n",
       "3  Machine-op-inspct      Unmarried  White  Female   ...             3900   \n",
       "4     Prof-specialty      Own-child  White  Female   ...             3900   \n",
       "5      Other-service      Unmarried  White  Female   ...             3770   \n",
       "6       Adm-clerical      Unmarried  White    Male   ...             3770   \n",
       "\n",
       "   hours.per.week  native.country income workclass_num  over50K  marital_num  \\\n",
       "1              18   United-States  <=50K             0        0            0   \n",
       "3              40   United-States  <=50K             0        0            1   \n",
       "4              40   United-States  <=50K             0        0            2   \n",
       "5              45   United-States  <=50K             0        0            1   \n",
       "6              40   United-States  <=50K             0        0            2   \n",
       "\n",
       "   race_num  sex_num  rel_num  \n",
       "1         0        0        0  \n",
       "3         0        0        0  \n",
       "4         0        0        0  \n",
       "5         0        0        0  \n",
       "6         0        1        0  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x11447780>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAEACAYAAABWLgY0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDdj/AXuwY7rmwqaqAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x143da470>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data.race_num.hist(normed = 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The dataset is mostly composed of white people (1994)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x116c5898>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEACAYAAACj0I2EAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x12aa2978>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data.workclass_num.hist(normed = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[  1.48590902e-02   3.52480852e-01  -3.63766335e-01  -1.71074311e-01\n",
      "   3.29860790e-01   3.06367805e+00   3.24647793e-04   6.86984115e-04]\n"
     ]
    }
   ],
   "source": [
    "X = data[['workclass_num', 'education.num', 'marital_num', 'race_num', 'sex_num', 'rel_num', 'capital.gain', 'capital.loss']]\n",
    "y = data.over50K\n",
    "# fit an Extra Tree model to the data\n",
    "logr = LogisticRegression()\n",
    "logr.fit(X, y)\n",
    "# display the relative importance of each attribute\n",
    "relval = logr.coef_[0]\n",
    "print relval"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdgAAAEZCAYAAAA0bTSIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nwfzHLOZaaAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x164d59b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# horizontal bar plot of feature importance\n",
    "pos = np.arange(8) + 0.5\n",
    "plt.barh(pos, relval, align='center')\n",
    "plt.title(\"Feature Importance\")\n",
    "plt.xlabel(\"Model Accuracy\")\n",
    "plt.ylabel(\"Features\")\n",
    "plt.yticks(pos, ('Working Class', 'Education', 'Marital Status', 'Race', 'Sex', 'Relationship Status', 'Capital Gain', 'Capital Loss'))\n",
    "plt.grid(True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Vanilla Cross Validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=13) #deaults at 25% for the test size\n",
    "logr = LogisticRegression(penalty = 'l2')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "c_list = np.logspace(-3, 10, num=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1602b1d0>]"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAENCAYAAADgwHn9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nMRQkSY2hIElqDAVJUmMoSJKa/w+n3UN7Cm5EYQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x15048518>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ac_list = []\n",
    "for c_val in c_list:\n",
    "    logr.C = c_val\n",
    "    logr.fit(X_train, y_train)\n",
    "    y_pred = logr.predict(X_test)\n",
    "    ac_list.append(metrics.accuracy_score(y_test, y_pred))\n",
    "plt.plot(c_list, ac_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x14a792b0>]"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYkAAAEECAYAAADd88i7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3XuUFPWd9/H3B1GMgkpgwQQVEwnqIbLENazxOig5EnOy\nGJNdIZoE1zwSEyKai8CzZp24sl4SPcdd4w2JwWwUjc+qGEXBy2hEEiHKiDIISiAiDihgvBCVy/f5\no4rQNNPT3TPd0zPdn9c5c+iurqrv78f01KerflXVigjMzMxa0q3SDTAzs87LIWFmZjk5JMzMLCeH\nhJmZ5eSQMDOznBwSZmaWU0EhIWmUpKWSlkma1MLrfSTNlrRI0mJJ49LpPST9QdJzkl6U9J8Zy1wi\nabWkZ9OfUSXrlZmZlYTyXSchqRuwDDgZWAMsAMZExNKMeS4B9oyIKZL6Ai8B/SNii6S9ImKTpN2A\necAPImJeusw7EXFNebpmZmbtVciexHBgeUSsiojNwExgdNY8zUCv9HEvYH1EbAGIiE3p9B5pvY0Z\ny6mtDTczs/IrJCQGAK9mPF+dTss0DRgiaQ3QCEzc/oKkbpKeIwmShohYkrHchPQQ1S2S9m1TD8zM\nrGy6l2g9U4DGiBgh6RBgrqShEfFuRGwDPiNpH2COpBMj4gngeuDSiAhJlwHXAOdkr1iS7xtiZtYG\nEdHuozWF7Em8BhyU8fyAdFqmY4HfpI16BfgTcFjmDBHxNvAAcFT6/I3YMSAyDfhsrgZERNl/Tjzx\nxKqo4Tqdt4brdN4a1VinVAoJiQXAIEkDJe0BjAFmZc3TBIwEkNQfGAyskNR3+2EkSR8BPg8sSp/v\nn7H86cAL7elIex188MFVUcN1Om8N1+m8NaqxTqnkPdwUEVslTQDmkITK9IhokjQ+eTluBi4HbpXU\nSDIYfVFEbJB0BDBDktJlfxURj6arvkrSMGAbsBIYX+rOFcNv9tquU019qbY61dSXjqxTKgWNSUTE\nQ8ChWdNuynj8JvClFpZbDByZY53fKKqlZVZXV1cVNVyn89Zwnc5boxrrlEre6yQqTVJ09jaamXU2\nkogOGrg2M7Ma5ZAwM7OcSnWdhFWpCHj5Zdi2DQYPBlXoGvmtW+HFF6F3bzjwwMq0oaNt3gzPP5/0\nt1+/8tbasgUaG+Evf9kxrU8f+PSnYbfdylu73F59NXnf9OxZ6ZZ0TQ6JKtfYCOvWwciRO2/gI+Ch\nh5IAaMmHH8KCBfDkk8lGYrfd4K9/hRNOgH/8R+jRI3/tffaBr38duhWxv7pwIcyfv+P5u+/C00/D\nU08lG8oNG6BXr6QdJ56Y/HziEzv6tm4d/O53sGZN4TXLRYKxY5ONbbbf/Aaam1tebuPGpA+//z0c\ndBC89hp8/ONJXw8/vLCg/vjHYfRo6J71Fz5vHjz77I7nf/lL8n87f/6uYbRmDaxdC8cdB8cc0/k3\nsn37Ju+LAen9IJqaYOpU+O1vk9/BzJnw2Yyrsd5/H+6+O/n/ttw8cF2lFiyA//gP+OMf4aMfhd13\nhx//GP7pn+C++5LXIuD441tevls3GDZs543wqlXwxBPJRmbr1vxteOQRmDQJxo3LP+/8+UmbFi+G\nL31px6fXHj2SUDrhBOjfP2lzU1PSju0/3bvD8OGwdGmyQT322J2Do1LmzUv6fv75O0/fsCHZ+J99\ndsvL9eyZbJSPOy75BLx1axL2TzwBK1YUVvu555IN/L/9G5x5ZhI6l16afKr+whd2/N/stRd87nPJ\n+6ClMGtuTj4oPPMMfPBBwV2viNWrk7b27p38/p9/Hi64AL77XZg7F847L3k/fvvbcMstcNVVMHQo\nDBpU6ZaXx3XXlWbg2iHRCWzenGx4n3wy+WPesKGw5caOTf4AMr3/PnzjG8lGd/JkOOcc2GMPuP/+\nZCPxyivJH8W//3uyMS7nhvSZZ+C005KN9z777Jje2Jj88W7enDx/91146y2YMiXZqBayl7JdBCxf\nnuyBHHpoEmyd5fDI7bcnewz33LPz9HvugRtvhIcfLl/tiCRULr00eW/16wcXXwxf+9quexfVZNu2\n5LBkUxOceurOez8rV8KYMckHkVNOSf4/jmzxBP3qUKqzmxwSFbB5c7IB3f5JeP58OPjg5FP7CSfA\nxz6Wfx0ffJB8IvrWt5JPRwCbNiUb5b594dZbd93Ybh9fGDSo4z5ln302/N3fJZ/aIDkcNHw4/PCH\nyQYdko36P/xDEmbV5PXXYcgQePPNnQ+5nX9+ckhk0i7fzFIeL7+cfLLuLOFZSZs3J3tHtTCu5ZDo\ngj74AH75S7jiCthvPxgxIgmG449PDgkV67XX4OST4ayz4MILkz2DAw5IAqKzbBCam5PBz6efToJw\n5Mikv1OnVrplHeOww+COO+Azn9kx7YgjYPr0JCzNysUh0cX8+tfJ4Z9PfzoZGzjmmNKst7kZTjop\n2Ys46SSYNq3zBMR2P/1pssd04IHJceP77ituMLsrO++8ZM/tBz9Inr/xBnzqU8neRTUf9rHKc0h0\nIW++mWwoHnoIjj669Otfty459n3eeZ1z4/vhh0k4du+enLGTOT5R7e66C267LTnDBpLf04wZO56b\nlYtDogv52c+SwbIZMyrdkspZtgz23nvH6Ym1Yt265PqS7XsO3/kOHHLIjj0Ls3LxbTkqZNu25HqB\nYua/6aZkkLmWDR5cewEByVlFBx6449qExx9PxqLMugqHRJGmTUvO17/4Yli/Pv/8jz6afIIux2Em\n6xpGjEjCYfvFaX//95VukVnhHBJFuvtu+MlPksMHgwcnp3I++WRyfUJLbrwx2Yuo9IVdVjnbQ6Kh\nITmbrbOdWGDWGo9JFGHjRhg4MDn/fe+9k6tXr78+2VtYsgSOOgomToQvfzmZf82aZMB21arkVhJW\nmzZsSE7/Pf305OKt7CuwzcrBA9cV8Otfw513wqzsL28F3nkn+bR44YXJ1ZxXX52c+vn663DDDR3f\nVutcjjwyuRJ44cLkOgmzcvPAdQXce++OvYRsvXol90V69tnkk+PRR3vA2nYYMSI59XfIkEq3xKw4\nvpynQO+/D3PmJIeXWrPvvskVtrfcktyHyYOUBvCVryQ3WeyM17GYtcaHmwr0wAPJ/YeeeKLSLTEz\ny8+HmzrYvfcmN88zM6sl3pMowNatyZe4/P73yd00zcw6O+9JdKD585PbdzsgzKzWFBQSkkZJWipp\nmaRd7oIvqY+k2ZIWSVosaVw6vYekP0h6TtKLkv4zY5nekuZIeknSw5L2LVmvSsyHmsysVuUNCUnd\ngOuAU4AhwFhJh2XNNgFYFBHDgBHA1ZK6R8QHwIiI+AwwFDhJ0rHpMpOBRyLiUOAxYEpJelQGzzyT\nXClrZlZrCtmTGA4sj4hVEbEZmAmMzpqnGdh+TXEvYH1EbAGIiE3p9B5pve1fOz4a2H5f1BlAp/2s\nvn598u1qZma1ppCQGAC8mvF8dTot0zRgiKQ1QCMwcfsLkrpJeo4kSBoiYkn6Ur+IWAsQEc1Av7Z1\nofw2bGjbN8eZmXV1pbqYbgrQGBEjJB0CzJU0NCLejYhtwGck7QPMkXRiRLR0tUHOU5jq6+v/9riu\nro66uroSNTu/CIeEmXV+DQ0NNDQ0lHy9eU+BlXQ0UB8Ro9Lnk4GIiCsz5nkQmBoR89LnjwKTImJh\n1rp+DGyKiKslNQF1EbFW0v7A4xFxeAv1K3oK7HvvJd8J8N57FWuCmVnROvIU2AXAIEkDJe0BjAGy\nb3HXBIxMG9YfGAyskNR3+1lLkj4CfB5YlC4zCxiXPv4mcF87+lE269d7L8LMalfew00RsVXSBGAO\nSahMj4gmSeOTl+Nm4HLgVkmNgICLImKDpCOAGZKULvuriHg0XfWVwF2S/hVYBfxLyXtXAj7UZGa1\nzFdc5/HYY3DZZcm/ZmZdha+47iA+3GRmtcwhkYcPN5lZLXNI5LFhA/TpU+lWmJlVhkMiD+9JmFkt\nc0jk4TEJM6tlDok8vCdhZrXMIZGHxyTMrJY5JPLw4SYzq2UOiTx8uMnMaplDohW+A6yZ1TqHRCs2\nbYLu3WHPPSvdEjOzynBItMLjEWZW6xwSrfChJjOrdQ6JVvj0VzOrdQ6JVvhwk5nVOodEK3y4ycxq\nnUOiFT7cZGa1ziHRCu9JmFmtc0i0wmMSZlbrHBKt8OEmM6t1DolW+HCTmdU6h0QrHBJmVuscEq3w\nmISZ1bqCQkLSKElLJS2TNKmF1/tImi1pkaTFksal0w+Q9JikF9Pp52csc4mk1ZKeTX9GlaxXJeA7\nwJqZgSKi9RmkbsAy4GRgDbAAGBMRSzPmuQTYMyKmSOoLvAT0B/oC+0fEIkk9gT8CoyNiabrMOxFx\nTZ76ka+N5fDee9CvX/KvmVlXI4mIUHvXU8iexHBgeUSsiojNwExgdNY8zUCv9HEvYH1EbImI5ohY\nBBAR7wJNwICM5drdgXLxoSYzs8JCYgDwasbz1ey8oQeYBgyRtAZoBCZmr0TSwcAw4A8Zkyekh6hu\nkbRvEe0uO5/+amYG3Uu0nilAY0SMkHQIMFfS0HTvgfRQ093AxO3TgOuBSyMiJF0GXAOc09LK6+vr\n//a4rq6Ourq6EjU7N49HmFlX0tDQQENDQ8nXW8iYxNFAfUSMSp9PBiIirsyY50FgakTMS58/CkyK\niIWSugO/BWZHxLU5agwE7o+IoS28VpExibvvhpkzk3/NzLqajhyTWAAMkjRQ0h7AGGBW1jxNwMi0\nYf2BwcCK9LVfAEuyA0LS/hlPTwdeKL755eMxCTOzAg43RcRWSROAOSShMj0imiSNT16Om4HLgVsl\nNZIMRl8UERskHQucCSyW9BwQwP+NiIeAqyQNA7YBK4HxZehfm3lMwsysgMNNlVapw00/+lFyCuyP\nftThpc3M2q0jDzfVJB9uMjNzSOTkw01mZg6JnHwKrJmZQyInh4SZmUMiJ49JmJk5JFrkO8CamSUc\nEi3YtAm6d4c996x0S8zMKssh0YI33oC+fSvdCjOzynNItODll+GTn6x0K8zMKs8h0YLly2Hw4Eq3\nwsys8hwSLVi2DD71qUq3wsys8hwSLVi+3CFhZgYOiRY5JMzMEr4LbJYtW6BnT3jrLZ8Ca2Zdl+8C\nWyYrV8LHPuaAMDMDh8QufKjJzGwHh0QWh4SZ2Q4OiSwOCTOzHRwSWZYt84V0ZmbbOSSyeE/CzGwH\nnwKb4cMPYZ994J13YPfdO6SkmVlZ+BTYMlixAg480AFhZradQyKDxyPMzHZWUEhIGiVpqaRlkia1\n8HofSbMlLZK0WNK4dPoBkh6T9GI6/fyMZXpLmiPpJUkPS9q3ZL1qI49HmJntLG9ISOoGXAecAgwB\nxko6LGu2CcCiiBgGjACultQd2AJ8PyKGAJ8Dvpux7GTgkYg4FHgMmFKKDrWHQ8LMbGeF7EkMB5ZH\nxKqI2AzMBEZnzdMM9Eof9wLWR8SWiGiOiEUAEfEu0AQMSOcbDcxIH88ATmt7N0rDIWFmtrPuBcwz\nAHg14/lqkuDINA14VNIaoCdwRvZKJB0MDAN+n07qFxFrASKiWVK/olpeBh6TMDPbWSEhUYgpQGNE\njJB0CDBX0tB07wFJPYG7gYkR8V6OdeQ8z7W+vv5vj+vq6qirqytRs3fYtAnefDM5u8nMrKtpaGig\noaGh5OvNe52EpKOB+ogYlT6fDEREXJkxz4PA1IiYlz5/FJgUEQvTsYnfArMj4tqMZZqAuohYK2l/\n4PGIOLyF+h1yncTixXDGGbBkSdlLmZmVXUdeJ7EAGCRpoKQ9gDHArKx5moCRacP6A4OBFelrvwCW\nZAZEahYwLn38TeC+oltfQh6PMDPbVd6QiIitJGcvzQFeBGZGRJOk8ZLOTWe7HDhKUiMwF7goIjZI\nOhY4EzhJ0nOSnpU0Kl3mSuDzkl4CTgauKG3XiuOQMDPblW/Lkfre92DQIJg4seylzMzKzrflKLHX\nX0++kc7MzHZwSKQcEmZmu3JIpJqbHRJmZtkcEkBEsiex//6VbomZWefikADefht22w169qx0S8zM\nOheHBB6PMDPLxSGBQ8LMLBeHBA4JM7NcHBIkZzZ50NrMbFcOCbwnYWaWi0MCh4SZWS4OCRwSZma5\nOCRwSJiZ5eKQwFdbm5nlUvMh8f77yVeX9ulT6ZaYmXU+NR8Szc3Qvz+o3XddNzOrPjUfEh6PMDPL\nzSHhkDAzy8kh4ZAwM8vJIeEzm8zMcqr5kPA30pmZ5VbzIeHDTWZmuTkkHBJmZjkVFBKSRklaKmmZ\npEktvN5H0mxJiyQtljQu47XpktZKej5rmUskrZb0bPozqt29aQOHhJlZboqI1meQugHLgJOBNcAC\nYExELM2Y5xJgz4iYIqkv8BLQPyK2SDoOeBe4LSKGZi3zTkRck6d+5GtjW23dCnvumVxxvfvuZSlh\nZlYRkoiIdl8mXMiexHBgeUSsiojNwExgdNY8zUCv9HEvYH1EbAGIiKeAjTnWXdHrnNetg969HRBm\nZrkUEhIDgFcznq9Op2WaBgyRtAZoBCYWWH9CeojqFkn7FrhMyfjMJjOz1nUv0XqmAI0RMULSIcBc\nSUMj4t1WlrkeuDQiQtJlwDXAOS3NWF9f/7fHdXV11NXVlaTRHo8ws2rR0NBAQ0NDyddbyJjE0UB9\nRIxKn08GIiKuzJjnQWBqRMxLnz8KTIqIhenzgcD9mWMSWTVyvl7OMYnp0+Gpp+DWW8uyejOziunI\nMYkFwCBJAyXtAYwBZmXN0wSMTBvWHxgMrMhsL1njD5Iyr3M+HXihuKa3n6+2NjNrXd6QiIitwARg\nDvAiMDMimiSNl3RuOtvlwFGSGoG5wEURsQFA0u3A08BgSX+WdHa6zFWSnpe0CDgRuLCkPSuADzeZ\nmbUu7+GmSivn4abTT4exY+Gf/7ksqzczq5iOPNxUtXx2k5lZ62o6JNascUiYmbWmZg83bdkCe+8N\nb78NPXqUfPVmZhXlw03ttGYN9O3rgDAza03NhsSqVTBwYKVbYWbWudVsSKxcCQcfXOlWmJl1bjUb\nEqtWOSTMzPKp2ZBYudKHm8zM8qnpkPCehJlZ62o2JDxwbWaWX01eJ7FtG+y1F2zcCB/5SElXbWbW\nKfg6iXZobob99nNAmJnlU5Mh4UFrM7PC1GxIeNDazCy/mgwJXyNhZlaYmgwJH24yMytMTYaE9yTM\nzApTkyHhPQkzs8LU3HUSEck1Em+8AT17lmy1Zmadiq+TaKN165JwcECYmeVXcyHhQ01mZoWruZDw\noLWZWeFqLiS8J2FmVriCQkLSKElLJS2TNKmF1/tImi1pkaTFksZlvDZd0lpJz2ct01vSHEkvSXpY\n0r7t7k0BfLW1mVnh8oaEpG7AdcApwBBgrKTDsmabACyKiGHACOBqSd3T125Nl802GXgkIg4FHgOm\ntK0LxfHhJjOzwhWyJzEcWB4RqyJiMzATGJ01TzPQK33cC1gfEVsAIuIpYGML6x0NzEgfzwBOK7Lt\nbeLDTWZmheuefxYGAK9mPF9NEhyZpgGPSloD9ATOKGC9/SJiLUBENEvqV8Ay7RLhLxsyMytGISFR\niClAY0SMkHQIMFfS0Ih4t4h15Lxirr6+/m+P6+rqqKura1Mj16+H3XeHfTtk9MPMrOM0NDTQ0NBQ\n8vUWEhKvAQdlPD8gnZbpWGAqQES8IulPwGHAwlbWu1ZS/4hYK2l/YF2uGTNDoj08aG1m1Sr7A/RP\nfvKTkqy3kDGJBcAgSQMl7QGMAWZlzdMEjASQ1B8YDKzIeF3pT6ZZwLj08TeB+4pqeRssWQKHHlru\nKmZm1SNvSETEVpKzl+YALwIzI6JJ0nhJ56azXQ4cJakRmAtcFBEbACTdDjwNDJb0Z0lnp8tcCXxe\n0kvAycAVpexYSxYuhM9+ttxVzMyqR03d4O+YY+Dyy+HEE0uyOjOzTqtUN/irmZDYvBn22w9efx32\n2acEDTMz68R8F9giLVkCBx3kgDAzK0bNhITHI8zMildTIXHUUZVuhZlZ1+KQMDOznGpi4PqDD6B3\nb3jzzeSrS83Mqp0HrovwwgswaJADwsysWDUREj7UZGbWNjUREgsWOCTMzNqiJkLCp7+ambVN1Q9c\n//Wv0KcPbNwIPXqUsGFmZp2YB64L1NgIhx/ugDAza4uqDwkPWpuZtV3VhkQEPP443HgjfO5zlW6N\nmVnXVKqvL+1UHnkE6uth3Tq4+GI488xKt8jMrGuqupCYMwfGjYOf/QzOOAN2263SLTIz67qq6uym\nt96CoUPhF7+AkSPL3DAzs07MXzrUgnHjYO+94ec/L2+bzMw6u1KFRNUcbpo1C556ChYtqnRLzMyq\nR5cNiddegz//OXn84Yfw7W/DnXdCz56VbZeZWTXpkiHx/vvJtQ8DB4LSnakLL4Tjj69su8zMqk2X\nDIn/+R848kh44IFKt8TMrLp1uZDYtg2uvhquv77SLTEzq34FXXEtaZSkpZKWSZrUwut9JM2WtEjS\nYknj8i0r6RJJqyU9m/6MKqQtDz6YfHlQXV0hc5uZWXvkPQVWUjdgGXAysAZYAIyJiKUZ81wC7BkR\nUyT1BV4C+gPbci2bLvNORFyTp/5Op8COGAHnngtjxxbdVzOzmtGRd4EdDiyPiFURsRmYCYzOmqcZ\n6JU+7gWsj4gtBSxbVAcWLoRXXoGvfrWYpczMrK0KCYkBwKsZz1en0zJNA4ZIWgM0AhMLXHZCeojq\nFkn75mvI1VfDBRfA7rsX0GozM2u3Ug1cTwEaI2KEpEOAuZKG5lnmeuDSiAhJlwHXAOe0NGN9fT1v\nvw333gtnnVUH1JWo2WZm1aGhoYGGhoaSr7eQMYmjgfqIGJU+nwxERFyZMc+DwNSImJc+fxSYRBJC\nrS6bTh8I3B8RuwTL9jGJG26A+fPhttva0VszsxrRkWMSC4BBkgZK2gMYA8zKmqcJGJk2rD8wGFjR\n2rKS9s9Y/nTghdYa8eCD8MUvFtBaMzMrmYJu8JeennotSahMj4grJI0n2Su4OT2j6VbgIJLB6Msj\n4o5cy6bTbwOGkZwBtRIYHxFrW6gdmzYF/fvDqlXQu3e7+2xmVvVq6i6ws2cHU6fC735X6daYmXUN\nHXm4qeIeeMCHmszMKqFLhMSDD8Kpp1a6FWZmtadLhMSHH8IRR1S6FWZmtadLhMSpp+64JbiZmXWc\nLhESHo8wM6uMLnF20zvvhL9xzsysCDV1Cmxnb6OZWWdTU6fAmplZZTgkzMwsJ4eEmZnl5JAwM7Oc\nHBJmZpaTQ8LMzHJySJiZWU4OCTMzy8khYWZmOTkkzMwsJ4eEmZnl5JAwM7OcHBJmZpaTQ8LMzHJy\nSJiZWU4FhYSkUZKWSlomaVILr/eRNFvSIkmLJY3Lt6yk3pLmSHpJ0sOS9i1Jj9qooaGhKmq4Tuet\n4Tqdt0Y11imVvCEhqRtwHXAKMAQYK+mwrNkmAIsiYhgwArhaUvc8y04GHomIQ4HHgCml6FBb+c1e\n23WqqS/VVqea+tKRdUqlkD2J4cDyiFgVEZuBmcDorHmagV7p417A+ojYkmfZ0cCM9PEM4LS2d6P9\nVq5cWRU1XKfz1nCdzlujGuuUSiEhMQB4NeP56nRapmnAEElrgEZgYgHL9o+ItQAR0Qz0K67ppeU3\ne23Xqaa+VFudaupLR9Yple4lWs8UoDEiRkg6BJgraWiR68j5RdZSu7+mtSAdUaea+lJtdaqpL9VW\np5r60pF1SqGQkHgNOCjj+QHptEzHAlMBIuIVSX8CDsuzbLOk/hGxVtL+wLqWipfii7zNzKxtCjnc\ntAAYJGmgpD2AMcCsrHmagJEAkvoDg4EVeZadBYxLH38TuK8d/TAzszJQRM6jPDtmkkYB15KEyvSI\nuELSeCAi4mZJfYFbSfYaBFweEXfkWjad/lHgLuBAYBXwLxHxVqk7aGZmbVdQSJiZWW3yFddmZpaT\nQ8LMzHLqkiEh6TBJN0i6U9I5ZawzWtLNku6Q9Pky1vmEpFsk3VXGGntJ+qWkmyR9rUw1yt6PtE5H\n/V465H2W1tpL0gJJp5Zp/SdKejLtzwnlqJHWkaTLJP2XpK+Xsc5xaV+mSXqqjHUGSPrf9H29yy2J\nSlTj8PQ99nNJXynD+nf6uyx6WxARXfaHZJD8zg6osx8wrQPq3FXGdZ8FfDF9PLOr9qNCv5eyv8+A\nnwA/BE4t0/pPAB4AfgF8soz9OA34JfAzYEQH/G5GA/+njOv/AvC19PEdZarxfeDY9PF9ZezLXem/\nRW0LKronIWm6pLWSns+a3uoNBdN5vkTypp9Zzjqpi4Gfd0CdgrWh1gHsuPp9a5lqtEk76hT0e2lP\nnWLeZ22tI2kksAR4gySQSl4jIp6MiC+S3DPt0nL1BTgUmBcRPwS+U8Y6230NuL2MdeYB4yU9AjxU\nphq/AsZIugr4aBnWn624bUG5kz5Psh0HDAOez5jWDXgZGAjsDiwCDktf+zpwDfCxjPnzJm876nwc\nuAI4qSP6A/ymjP93Z5J+SgVuL0eNjHkK7kdb6xTze2lvfwp9n7Xjd3NZ+j54GLinzL+bPShiT6+N\n77Ovpo8L3mNt43vgQOCmcr4HgAuA44p5X7fjd9OtkN9/O9b/m/TfsyhiW1DRPYmIeArYmDU5500B\nI+JXEfF9YLCkayXdBDxexjpfAU4Gvirp3DLW+UDSDcCwQj+ZF1sLuCftx8+B+8tRQ9JHi+1HG+t8\njyJ+L+2oc2Ix77O21omIi9P3wa9J7oNWjr58WdKNJDfTvK5cfQH+Fxgl6VrgiTLWATiH5PqsgrWh\nzmPABen7+k/lqKHkYuObSH43Py3D+rP/Lv8fRWwLSnXvplJq6aaAwzNniIgnKOIN2I46/w38dwfU\n2QCc1846rdaKiE3Av5a5Rqn6ka9OKX4vhdQpxfssb53tIuK2ctWIiHtIPiiUQmt1/gp8q9x10lr1\n5a4TEc8DXy1zjVXA+DKuv6W/y4K3BV3y7CYzM+sYnTEkCrmhoOtUrla1/b9VU51q6ku11Sl3jbKt\nvzOEhNj5jI5CbijoOh1Xq9r+36qpTjX1pdrqlLtGx21nijkzoNQ/JKeurQE+AP4MnJ1O/wLwErAc\nmOw6lakCOQWZAAAAT0lEQVRVbf9v1VSnmvpSbXXKXaMjtzMR4Rv8mZlZbp3hcJOZmXVSDgkzM8vJ\nIWFmZjk5JMzMLCeHhJmZ5eSQMDOznBwSZmaWk0PCzMxy+v9j6+faGAw/owAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x15048438>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.semilogx(c_list, ac_list)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# No Free Lunch Theorem\n",
    "\n",
    "There isn't a single set of parameters that will work for every single dataset. The parameter C that we just obtained might not work for a different dataset.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYkAAAEOCAYAAAB8aOvdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nLJOThJmZZXKSMDOzTP8fkQ5XqFLMnzkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xe991c18>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYMAAAEOCAYAAABrSnsUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x164aae80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYMAAAEOCAYAAABrSnsUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/1MbK57/6PExAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x16ddf0f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "logr = LogisticRegression(penalty = 'l1')\n",
    "for test_split in [0.3, 0.5, 0.7]:\n",
    "        X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = test_split, random_state=13)\n",
    "        ac_list = []\n",
    "        for c_val in c_list:\n",
    "            logr.C = c_val\n",
    "            logr.fit(X_train, y_train)\n",
    "            y_pred = logr.predict(X_test)\n",
    "            ac_list.append(metrics.accuracy_score(y_test, y_pred))\n",
    "        plt.figure()\n",
    "        plt.semilogx(c_list, ac_list)\n",
    "        title = 'Split = '+str(test_split)\n",
    "        plt.title(title)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# K-Fold Cross Validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0.76552734,  0.84285575,  0.83709347])"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "kf = KFold(len(data), n_folds=3, shuffle=False)\n",
    "logr = LogisticRegression(penalty = 'l1')\n",
    "c_list = np.logspace(-3, 10, num=5)\n",
    "cross_val_score(logr, X, y, cv=kf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.815712291977\n"
     ]
    }
   ],
   "source": [
    "print np.mean(cross_val_score(logr, X, y, cv=kf, n_jobs = -1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.001\n",
      "1.77827941004\n",
      "3162.27766017\n",
      "5623413.2519\n",
      "10000000000.0\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x12b3f470>]"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYMAAAEECAYAAADK0VhyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAHKpJREFUeJzt3X2UHHWd7/H3J4QogQiiHJQgCRADHFZELoa4IJmQqAMC\nQUVNYLkHXE1UIgQMJuzZc5x7jx6XmwnIgzwEWAQExvC0ZJdFgkiLQReDhAAhT5AQA0FWRFAEIQ/f\n+0dV2HbSmemZqe7qrv68zsmZrqpf1fdbmU5/86tf1a8VEZiZWWsblHcCZmaWPxcDMzNzMTAzMxcD\nMzPDxcDMzHAxMDMzqiwGktolrZC0StKsCtvfI+keSY9JekLS6en6vSX9TNKydP1ZGedvZmYZUG/P\nGUgaBKwCJgAbgMXA5IhYUdbm28A7I+J8Se8FVgJ7Au8F3hcRj0naBfgNMKl8XzMzy181PYMxwOqI\nWBcRG4EuYFK3Nr8DhqWvhwF/iIhNEfG7iHgMICJeA5YDw7NJ3czMsjK4ijbDgfVly8+RFIhyVwP3\nS9oA7AJ8sftBJI0EDgUe7k+iZmZWO9UUg2qcDyyNiPGS9gfuk3RI2hsgvUR0G3D21nXdSfK8GGZm\nfRQRyuI41Vwmeh7Yp2x573RduSOBW9PEngHWAgcCSBpMUghujIi7egoUETX9M27cuJrHcJzGjeE4\njRvDcfr3J0vVFIPFwChJIyQNASYDC7q1WQ5MBJC0JzAaWJNu+1fgqYi4OJuU+2/kyJGO06BxinQu\nRYtTpHMpYpys9FoMImIzMB1YCCwDuiJiuaRpkqamzb4HHC5pKXAf8K2IeFnSkcCpwDGSlkh6VFJ7\nbU6ld0V7ExQpTpHOpWhxinQuRYyTlarGDCLiJ8AB3dZdVfb6JeCECvs9BOwwwBwz09bW5jgNGqdI\n51K0OEU6lyLGyUqvzxnUi6RolFzMzJqBJKKOA8hmZlZwLgZmZuZiYGZmLgZmZoaLgZmZkd10FGYN\nKwK2bIGNG2HTpu3/3N627je5Sdtf7mlbI7RtxPysMbgYtLgtW6r7QGz2NoMGweDBsOOOPf+stG5Q\nWf+5e2EoX+5pWyO0bcT8rHG4GBTMsmVw+unw+uvVfWhu2ZJ86PX0YVjNtr7sv+OOMHRo//fva5vu\nH+hmRZFlL8sPnRXMqafCyJFwyinVfVDu0DDPh5tZX2X50JmLQYGsWweHHQZr18K73pV3NmZWa34C\n2Sq6+GL40pdcCMys79wzKIhXXoH994elS2HvvfPOxszqwT0D28aVV8Lxx7sQmFn/uGdQAG++Cfvt\nB/fcA4ccknc2ZlYv7hnY37j5ZvjQh1wIzKz//JxBk9uyBTo7k8FjM7P+cs+gyf3kJzBkCEyYkHcm\nZtbMXAya3Jw5cN55nu/FzAamqmIgqV3SCkmrJM2qsP09ku6R9JikJySdXrbtWkkvSno8w7wNeOQR\nWLMGPv/5vDMxs2bXazGQNAi4DPgUcDAwRdKB3ZpNBx6LiEOB8cBcSVvHI65L97WMdXbCjBnJ1BJm\nZgNRTc9gDLA6ItZFxEagC5jUrc3vgGHp62HAHyJiE0BELAL+mFG+llq7Fn76U/jyl/POxMyKoJq7\niYYD68uWnyMpEOWuBu6XtAHYBfhiNunZ9nz/+0khGDas97ZmZr3J6tbS84GlETFe0v7AfZIOiYjX\n+nKQjo6Ot1+3tbXR1taWUXrF8vLLcOON8OSTeWdiZvVUKpUolUo1OXavTyBLGgt0RER7ujwbiIi4\noKzNfwLfjYiH0uX7gVkR8Ui6PAL494jY7mNRfgK5et/9Ljz9NFx3Xd6ZmFmesnwCuZqewWJgVPqB\n/gIwGZjSrc1yYCLwkKQ9gdHAmrLtSv/YAP31r3DZZcl4gZlZVnodQI6IzSR3Cy0ElgFdEbFc0jRJ\nU9Nm3wMOl7QUuA/4VkS8DCDpZuCXwGhJv5V0Ri1OpFX86EfJdxYcfHDemZhZkXiiuiayZUtSBC6/\nHMaPzzsbM8ubJ6prUXffDTvvDB5XN7OsuRg0kTlzYOZMTz1hZtlzMWgSDz8M69fDySfnnYmZFZGL\nQZPo7IRzzoHBnnTczGrAA8hN4JlnYOzYZAqKXXbJOxszaxQeQG4xF10EU6e6EJhZ7bhn0OBeeglG\nj4annoL3vS/vbMyskbhn0EKuuAI++1kXAjOrLfcMGtgbb8C++8IDD8BBB+WdjZk1GvcMWsQNN8BH\nP+pCYGa1555Bg9qyBQ48EK65Bo4+Ou9szKwRuWfQAhYsgHe/Gz7+8bwzMbNW4GLQoDo7PfWEmdWP\ni0ED+tWvYMOG5C4iM7N6cDFoQHPmwLnnwg475J2JmbUKDyA3mNWr4cgjk6kndt4572zMrJF5ALnA\nLrwQvvpVFwIzqy/3DBrIf/93cjvp8uWw5555Z2Nmjc49g4K6/HL4/OddCMys/qoqBpLaJa2QtErS\nrArb3yPpHkmPSXpC0unV7muJ119P5iE699y8MzGzVtRrMZA0CLgM+BRwMDBF0oHdmk0HHouIQ4Hx\nwFxJg6vc14Drr4ePfQwOOCDvTMysFVXTMxgDrI6IdRGxEegCJnVr8ztgWPp6GPCHiNhU5b4tb/Nm\nmDsXzjsv70zMrFVVUwyGA+vLlp9L15W7GjhY0gZgKXB2H/Ztef/2b7DHHvD3f593JmbWqrL6Rt3z\ngaURMV7S/sB9kg7p60E6Ojreft3W1kZbW1tG6TWuiOQhs299y1NPmFnPSqUSpVKpJsfu9dZSSWOB\njohoT5dnAxERF5S1+U/guxHxULp8PzCLpNj0uG/ZMVry1tJFi+CMM2DFCj9xbGZ9U+9bSxcDoySN\nkDQEmAws6NZmOTAxTW5PYDSwpsp9W5qnnjCzRtDrZaKI2CxpOrCQpHhcGxHLJU1LNsc84HvAdZKW\nAgK+FREvA1Tat0bn0nRWroT/+i+45Za8MzGzVucnkHM0dSrstReUDZWYmVUty8tELgY5efHF5Oss\nV65M7iQyM+srT0dRAJddBpMnuxCYWWNwzyAHf/kLjBwJv/wlfPCDeWdjZs3KPYMmd911yZfcuxCY\nWaNwz6DONm2C0aPh5pth7Ni8szGzZuaeQRO7887kDiIXAjNrJC4GdbR16glPSGdmjcbFoI4efBBe\nfRVOOCHvTMzM/paLQR11dsI3vwmD/LduZg3GA8h18tRTcMwx8Oyz8M535p2NmRWBB5Cb0Ny5cOaZ\nLgRm1pjcM6iDF16Agw+G1avhPe/JOxszKwr3DJrMpZfCKae4EJhZ43LPoMZeew323Rcefhj22y/v\nbMysSNwzaCLXXgvjx7sQmFljc8+ghjZtglGjYP58GDMm72zMrGjcM2gSt90GI0a4EJhZ43MxqBFP\nPWFmzcTFoEYeeABefx2OOy7vTMzMeldVMZDULmmFpFWSZlXYPlPSEkmPSnpC0iZJu6Xbzk7XPSHp\nrKxPoFF1dsLMmZ56wsyaQ68DyJIGAauACcAGYDEwOSJWbKf98cCMiJgo6WDgFuCjwCbgHuCrEbGm\nwn6FGUB+8kn4xCdg7Vo/cWxmtVPvAeQxwOqIWBcRG4EuYFIP7aeQFACAg4CHI+LNiNgMPAh8diAJ\nN4O5c+Eb33AhMLPmMbiKNsOB9WXLz5EUiG1I2gloB85MVz0JfEfSu4E3geNIehaF9fzzcNdd8PTT\neWdiZla9aopBX5wALIqIVwAiYoWkC4D7gNeAJcDm7e3c0dHx9uu2tjba2toyTq/2LrkETjsNdt89\n70zMrGhKpRKlUqkmx65mzGAs0BER7enybCAi4oIKbe8A5kdE13aO9V1gfURcWWFb048Z/OlPydQT\nv/kNjByZdzZmVnT1HjNYDIySNELSEGAysKBCUrsC44C7uq3fI/25D/AZ4OaBJt2orrkmGTh2ITCz\nZtPrZaKI2CxpOrCQpHhcGxHLJU1LNse8tOlJwL0R8Ua3Q9wuaXdgI/D1iPhThvk3jI0b4fvfT77w\n3sys2XhuoozcdFMyKd3PfpZ3JmbWKjw3UYPZOvXEzJl5Z2Jm1j8uBhm4//5khtJjj807EzOz/nEx\nyMDWXoEy6ayZmdWfxwwGaOnSZDK6NWvgHe/IOxszayUeM2ggc+fCWWe5EJhZc3PPYADWr4cPfzjp\nFey2W97ZmFmrcc+gQVxyCZx+uguBmTU/9wz66dVXky+5X7IE9tkn72zMrBW5Z9AA5s2D9nYXAjMr\nBvcM+uGtt5JewX/8Bxx6aN7ZmFmrcs8gZ11dcNBBLgRmVhwuBn0U8T/fb2xmVhQuBn20cGHy85Of\nzDcPM7MsuRj0kaeeMLMicjHogyVLYOVKmDw570zMzLLlYtAHnZ1w9tkwZEjemZiZZcu3llZp3To4\n7LBk6oldd807GzMz31qai4svhi99yYXAzIrJPYMqvPJK8pDZ0qXwgQ/knY2ZWaLuPQNJ7ZJWSFol\naVaF7TMlLZH0qKQnJG2StFu67XxJyyQ9LukmSU13xf2qq+D4410IzKy4eu0ZSBoErAImABuAxcDk\niFixnfbHAzMiYqKkEcADwIER8ZakHwN3R8QNFfZryJ7Bm28mvYJ77oFDDsk7GzOz/1HvnsEYYHVE\nrIuIjUAXMKmH9lOAW9LXfwLeAnaWNBgYSlJQmsbNN8Pf/Z0LgZkVWzXFYDiwvmz5uXTdNiTtBLQD\ntwNExB+BucBvgeeBVyLipwNJuJ62Tj1x3nl5Z2JmVluDMz7eCcCiiHgFQNJ+wDnACOBV4DZJp0TE\nzZV27ujoePt1W1sbbW1tGafXN/fckzxTMGFCrmmYmQFQKpUolUo1OXY1YwZjgY6IaE+XZwMRERdU\naHsHMD8iutLlLwCfiIivpMunAUdExPQK+zbcmMH48fDlL8Opp+adiZnZtuo9ZrAYGCVpRHon0GRg\nQYWkdgXGAXeVrV4JjJX0TkkiGYRePvC0a++RR+CZZ+ALX8g7EzOz2uv1MlFEbJY0HVhIUjyujYjl\nkqYlm2Ne2vQk4N6IeKNs36WSbgB+A2wGlgDzaAKdnTBjBuy4Y96ZmJnVnh86q+DZZ+Hww2HtWhg2\nLO9szMwq83QUNXbRRclYgQuBmbUK9wy6efllGDUKnnwS9tor72zMzLbPPYMauvJKmDTJhcDMWot7\nBmX++lfYd1+4777kqWMzs0bmnkGN/OhH8JGPuBCYWevJ+gnkprVlC8ydC5dfnncmZmb1555B6u67\nYehQyHkGDDOzXLgYpLZOSKdMrr6ZmTUXFwPg179OvuP45JPzzsTMLB8uBsCcOXDOOTDYIyhm1qJa\n/tbSZ56BI45IpqDYZZe6hzcz6zffWpqhiy6CqVNdCMystbV0z+Cll2D0aFi2DN7//rqGNjMbMPcM\nMnLFFfDZz7oQmJm1bM/gjTeSqSceeAAOOqhuYc3MMuOeQQZuvBE++lEXAjMzaNGewZYtSRGYNw/G\njatLSDOzzLlnMEALFsCuu8LRR+ediZlZY2jJYuCpJ8zM/lZVxUBSu6QVklZJmlVh+0xJSyQ9KukJ\nSZsk7SZpdNn6JZJelXRW9qdRvV/9CjZsgM98Js8szMwaS69jBpIGAauACcAGYDEwOSJWbKf98cCM\niJhY4TjPAUdExPoK+9VlzOBzn4Px42H69JqHMjOrqXqPGYwBVkfEuojYCHQBk3poPwW4pcL6icAz\nlQpBvaxeDb/4BZxxRl4ZmJk1pmqKwXCg/AP8uXTdNiTtBLQDt1fY/EUqF4m6ufBCmDYNdt45zyzM\nzBpP1vN0ngAsiohXyldK2hE4EZjd084dHR1vv25ra6Mtw2+a+f3voasLVlS8uGVm1vhKpRKlUqkm\nx65mzGAs0BER7enybCAi4oIKbe8A5kdEV7f1JwJf33qM7cSp6ZhBR0cycDxvXs1CmJnVVZZjBtUU\ngx2AlSQDyC8AvwamRMTybu12BdYAe0fEG9223QL8JCKu7yFOzYrB668nU088+CAccEBNQpiZ1V2W\nxaDXy0QRsVnSdGAhyRjDtRGxXNK0ZHNs/b/2ScC9FQrBUJLB46lZJNwf118PH/uYC4GZ2fYUfjqK\nzZuTIvDDH8JRR2V+eDOz3Hg6ij646y7YYw848si8MzEza1yFLgYRyfcbe+oJM7OeFboYPPRQ8m1m\nk3p6RM7MzIpdDDo74dxzYYcd8s7EzKyxFXYAeeXKZIrqtWth6NDMDmtm1jA8gFyFuXPha19zITAz\nq0YhewYvvggHHgirViV3EpmZFZF7Br247DKYPNmFwMysWoXrGfzlLzByJPzyl/DBDw48LzOzRuWe\nQQ+uuw4+/nEXAjOzvihUz2DTJhg9Gm66KZmLyMysyNwz2I4774S99nIhMDPrq8IUg61TT8ycmXcm\nZmbNpzDF4Be/gFdfhRNPzDsTM7PmU5hiMGcOfPObMKgwZ2RmVj+FGEB+6ik45phk6omddso4MTOz\nBuUB5G4uvBDOPNOFwMysv5q+Z/DCC3DwwcnUE+99bw0SMzNrUO4ZlLn0UjjlFBcCM7OBqKoYSGqX\ntELSKkmzKmyfKWmJpEclPSFpk6Td0m27SrpV0nJJyyQdkVXyr70G8+bBOedkdUQzs9bU62UiSYOA\nVcAEYAOwGJgcESu20/54YEZETEyXfwj8PCKukzQYGBoRf6qwX58vE118MSxaBLfe2qfdzMwKIcvL\nRIOraDMGWB0R69LgXcAkoGIxAKYAt6Rt3wV8PCJOB4iITcA2haA/Nm2Ciy6CH/84i6OZmbW2ai4T\nDQfWly0/l67bhqSdgHbg9nTVvsBLkq5LLyHNS9sM2G23wT77wBGZXXQyM2td1fQM+uIEYFFEvFJ2\n/MOAMyPiEUnfB2YD3660c0dHx9uv29raaGtrqxhk69QTZc3NzAqvVCpRKpVqcuxqxgzGAh0R0Z4u\nzwYiIi6o0PYOYH5EdKXLewK/ioj90uWjgFkRcUKFfaseM3jgAfj612HZMj9xbGatq963li4GRkka\nIWkIMBlYUCGpXYFxwF1b10XEi8B6SaPTVROApwaatKeeMDPLVq+XiSJis6TpwEKS4nFtRCyXNC3Z\nHPPSpicB90bEG90OcRZwk6QdgTXAGQNJ+MknYckSuOOOgRzFzMzKNd0TyGeckXyL2T/9Ux2SMjNr\nYFleJmqqYvD88/ChD8HTT8Puu9cpMTOzBtWy01FceimcdpoLgZlZ1pqmZ/DnP8O++8LixclPM7NW\n15I9g6uvhokTXQjMzGqhKXoGGzfC/vsndxAdfnidEzMza1At1zOYPz8pBi4EZma10fDFIAI6O+G8\n8/LOxMysuBq+GNx/P7z1FrS3552JmVlxNXwxmDMHZs701BNmZrXU0APIjz8Oxx4La9bAO96RU2Jm\nZg2qZQaQOzvhG99wITAzq7WG7RmsXw8f/nDSK9httxwTMzNrUC3RM7jkEjj9dBcCM7N6aMiewauv\nwn77waOPwogROSdmZtagCt8zuPrq5FZSFwIzs/pouJ7BW28lTxsvWAAf+UjeWZmZNa5C9wy6uuCA\nA1wIzMzqqaGKgaeeMDPLR0MVg4ULk5+f/GS+eZiZtZqqioGkdkkrJK2SNKvC9pmSlkh6VNITkjZJ\n2i3d9qykpen2X/cUZ+vUE8rkCpiZmVWr1wFkSYOAVcAEYAOwGJgcESu20/54YEZETEyX1wD/KyL+\n2EucGD48WLMGhgzp+4mYmbWaeg8gjwFWR8S6iNgIdAGTemg/BbilbFlVxuHss10IzMzyUM2H9HBg\nfdnyc+m6bUjaCWgHbi9bHcB9khZL+kpPgaZOrSIbMzPL3OCMj3cCsCgiXilbd2REvCBpD5KisDwi\nFlXa+aKLOt5+3dbWRltbW8bpmZk1r1KpRKlUqsmxqxkzGAt0RER7ujwbiIi4oELbO4D5EdG1nWN9\nG/hzRFxYYdt2vwPZzMy2Ve8xg8XAKEkjJA0BJgMLKiS1KzAOuKts3VBJu6SvdwY+CTyZReJmZpad\nXi8TRcRmSdOBhSTF49qIWC5pWrI55qVNTwLujYg3ynbfE7hTUqSxboqIhdmegpmZDVTDzU1kZmbV\nKfTcRGZmVn8uBmZm5mJgZmYuBmZmhouBmZnhYmBmZrgYmJkZLgZmZoaLgZmZ4WJgZma4GJiZGS4G\nZmaGi4GZmeFiYGZmuBiYmRkuBmZmhouBmZnhYmBmZlRZDCS1S1ohaZWkWRW2z5S0RNKjkp6QtEnS\nbmXbB6XbFmSZfF+VSiXHadA4RTqXosUp0rkUMU5Wei0GkgYBlwGfAg4Gpkg6sLxNRHRGxEci4jDg\nfKAUEa+UNTkbeCq7tPunaG+CIsUp0rkULU6RzqWIcbJSTc9gDLA6ItZFxEagC5jUQ/spwC1bFyTt\nDRwHXDOQRLPw7LPPOk6DxinSuRQtTpHOpYhxslJNMRgOrC9bfi5dtw1JOwHtwO1lqy8CzgOinzlm\npmhvgiLFKdK5FC1Okc6liHGyMjjj450ALNp6iUjSp4EXI+IxSW2AetpZ6nFzJuoRw3EaN4bjNG4M\nx8lXNcXgeWCfsuW903WVTKbsEhFwJHCipOOAnYBhkm6IiP/dfceIaJ6/NTOzglFEz1dvJO0ArAQm\nAC8AvwamRMTybu12BdYAe0fEGxWOMw74ZkScmFHuZmaWkV57BhGxWdJ0YCHJGMO1EbFc0rRkc8xL\nm54E3FupEJiZWWPrtWdgZmbF5yeQzczMxcDMzBq8GEg6UNIVkn4s6R9rGGeSpHmSbpH0iRrF2FfS\nNZLm1+L4aYyhkn4o6SpJp9QwTs3PJY1Tj99LXd5jaayhkhand9fVKsY4SQ+m53R0DeNI0nckXSLp\ntBrGOSo9l6slLapRjOGS7kjf09tMt5NhnIPS99kPJH2uBsf/m3+Xff48iIiG/0PyfMKP6xBnN+Dq\nGseYX8Nj/wPw6fR1Vx3+vmp2Ljn8Xmr+HgP+DzATOK6GMY4G7gb+FdivhnFOAn4IdALj6/AemAR8\npUbHPhY4JX19Sw3P4VzgyPT1XTWMMz/92afPg7r0DCRdK+lFSY93W9/jBHhpmxNI3txdtYyT+mfg\nBzWOUbV+xNqb/3lafHMN4/TLAOL0+nsZSIy+vMf6G0fSRJL5uX5PLw9fDiRORDwYEZ8GZgP/t1Zx\ngAOAhyJiJvD1GsbZ6hTg5hrFeAiYJumnwE9qeC43ApMl/T9g9xocv7u+fR7UuqKnVeko4FDg8bJ1\ng4CngRHAjsBjwIHpttOAC4H3l7XvtZIOIM5ewL8Ax9T6XIBba/j3dirp/zqBm2sVp6xN1efS3zjV\n/l4Gei7VvscG8Lv5Tvo+uBe4sw6/myH0oefWz/fayenrqnuh/XwPfAC4qobnMgM4qq/v6QH8bgZV\n8x4YwPFvTX/+A334PKhLzyAiFgF/7LZ6uxPgRcSNEXEuMFrSxZKuAh6oYZzPkTxUd7KkqTWK8aak\nK4BDq/1fdl9jAXem5/AD4N+ridGfOJJ27+u59DPON6jy9zKAGOP68h7rb5yI+Of0fXATcHWt4kj6\njKQrgetJZhuuSRzgDqBd0sXAz2sYB+AfgetqGONnwIz0Pb22VnEkjUjfZ9cDc2pw/O7/Lm+nD58H\nWc9N1BeVJsAbU94gIn5OH95oA4hzKXBpjWO8DHxtADF6jRURrwNfyiBGb3GyOpfe4gz091JNjCze\nY73G2SoibqhlnIi4k+Q/BVnoKc4bwJdrHSeN1VHLGBHxOHByBjF6i7MOmFbD41f6d1n150FD301k\nZmb1kWcx6MsEeI0ep17nUs9YRYpTpHNxnMaNUY84NTt+PYuB+Nu7KBYDo9LraENIZjzN4msx6xGn\nXudSz1hFilOkc3Gcxo1Rjzj1+6ypduR8IH9IbgnbALwJ/BY4I11/LMmMqKuB2c0Qp17nUrS/t6L9\nfhynceMU5Vzq+VkTEZ6ozszMPIBsZma4GJiZGS4GZmaGi4GZmeFiYGZmuBiYmRkuBmZmhouBmZkB\n/x/3tXx54UkFMwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1670df28>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ac_list = []\n",
    "for c_val in c_list:\n",
    "    print c_val\n",
    "    logr.C = c_val\n",
    "    logr.fit(X_train, y_train)\n",
    "    y_pred = logr.predict(X_test)\n",
    "    ac_list.append(cross_val_score(logr, X, y, cv=kf, n_jobs = -1).mean())\n",
    "plt.semilogx(c_list, ac_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegressionCV(Cs=array([  1.00000e-03,   1.77828e+00,   3.16228e+03,   5.62341e+06,\n",
       "         1.00000e+10]),\n",
       "           class_weight=None, cv=3, dual=False, fit_intercept=True,\n",
       "           intercept_scaling=1.0, max_iter=100, multi_class='ovr',\n",
       "           n_jobs=-1, penalty='l2', random_state=None, refit=True,\n",
       "           scoring=None, solver='lbfgs', tol=0.0001, verbose=0)"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "logr = LogisticRegressionCV(c_list, cv = 3, n_jobs = -1)\n",
    "logr.fit(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 3162.27766017])"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "logr.C_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}