project notebook.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>type_employer</th>\n",
       "      <th>fnlwgt</th>\n",
       "      <th>education</th>\n",
       "      <th>education_num</th>\n",
       "      <th>marital</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>hr_per_week</th>\n",
       "      <th>country</th>\n",
       "      <th>income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>39</td>\n",
       "      <td>State-gov</td>\n",
       "      <td>77516</td>\n",
       "      <td>Bachelors</td>\n",
       "      <td>13</td>\n",
       "      <td>Never-married</td>\n",
       "      <td>Adm-clerical</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>White</td>\n",
       "      <td>Male</td>\n",
       "      <td>2174</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>50</td>\n",
       "      <td>Self-emp-not-inc</td>\n",
       "      <td>83311</td>\n",
       "      <td>Bachelors</td>\n",
       "      <td>13</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Exec-managerial</td>\n",
       "      <td>Husband</td>\n",
       "      <td>White</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>13</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>38</td>\n",
       "      <td>Private</td>\n",
       "      <td>215646</td>\n",
       "      <td>HS-grad</td>\n",
       "      <td>9</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>Handlers-cleaners</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>White</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>53</td>\n",
       "      <td>Private</td>\n",
       "      <td>234721</td>\n",
       "      <td>11th</td>\n",
       "      <td>7</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Handlers-cleaners</td>\n",
       "      <td>Husband</td>\n",
       "      <td>Black</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>28</td>\n",
       "      <td>Private</td>\n",
       "      <td>338409</td>\n",
       "      <td>Bachelors</td>\n",
       "      <td>13</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Prof-specialty</td>\n",
       "      <td>Wife</td>\n",
       "      <td>Black</td>\n",
       "      <td>Female</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>Cuba</td>\n",
       "      <td>&lt;=50K</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   age     type_employer  fnlwgt  education  education_num  \\\n",
       "0   39         State-gov   77516  Bachelors             13   \n",
       "1   50  Self-emp-not-inc   83311  Bachelors             13   \n",
       "2   38           Private  215646    HS-grad              9   \n",
       "3   53           Private  234721       11th              7   \n",
       "4   28           Private  338409  Bachelors             13   \n",
       "\n",
       "              marital         occupation   relationship   race     sex  \\\n",
       "0       Never-married       Adm-clerical  Not-in-family  White    Male   \n",
       "1  Married-civ-spouse    Exec-managerial        Husband  White    Male   \n",
       "2            Divorced  Handlers-cleaners  Not-in-family  White    Male   \n",
       "3  Married-civ-spouse  Handlers-cleaners        Husband  Black    Male   \n",
       "4  Married-civ-spouse     Prof-specialty           Wife  Black  Female   \n",
       "\n",
       "   capital_gain  capital_loss  hr_per_week        country income  \n",
       "0          2174             0           40  United-States  <=50K  \n",
       "1             0             0           13  United-States  <=50K  \n",
       "2             0             0           40  United-States  <=50K  \n",
       "3             0             0           40  United-States  <=50K  \n",
       "4             0             0           40           Cuba  <=50K  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#reading csv file into dataframe\n",
    "adult_salary = pd.read_csv(\"adult_salary.csv\").iloc[:, 1:]\n",
    "adult_salary.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 32561 entries, 0 to 32560\n",
      "Data columns (total 15 columns):\n",
      " #   Column         Non-Null Count  Dtype \n",
      "---  ------         --------------  ----- \n",
      " 0   age            32561 non-null  int64 \n",
      " 1   type_employer  32561 non-null  object\n",
      " 2   fnlwgt         32561 non-null  int64 \n",
      " 3   education      32561 non-null  object\n",
      " 4   education_num  32561 non-null  int64 \n",
      " 5   marital        32561 non-null  object\n",
      " 6   occupation     32561 non-null  object\n",
      " 7   relationship   32561 non-null  object\n",
      " 8   race           32561 non-null  object\n",
      " 9   sex            32561 non-null  object\n",
      " 10  capital_gain   32561 non-null  int64 \n",
      " 11  capital_loss   32561 non-null  int64 \n",
      " 12  hr_per_week    32561 non-null  int64 \n",
      " 13  country        32561 non-null  object\n",
      " 14  income         32561 non-null  object\n",
      "dtypes: int64(6), object(9)\n",
      "memory usage: 3.7+ MB\n"
     ]
    }
   ],
   "source": [
    "#checking if null values exist\n",
    "adult_salary.info()\n",
    "#looks like no null values, probably other missing data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'?',\n",
       " 'Federal-gov',\n",
       " 'Local-gov',\n",
       " 'Never-worked',\n",
       " 'Private',\n",
       " 'Self-emp-inc',\n",
       " 'Self-emp-not-inc',\n",
       " 'State-gov',\n",
       " 'Without-pay'}"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#checking unique values for all text based columns to see if any missing data\n",
    "set(adult_salary[\"type_employer\"])\n",
    "#? indicating unknown employers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "#lets remove any data where employer type is unknown\n",
    "adult_salary = adult_salary[adult_salary[\"type_employer\"] != \"?\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'10th',\n",
       " '11th',\n",
       " '12th',\n",
       " '1st-4th',\n",
       " '5th-6th',\n",
       " '7th-8th',\n",
       " '9th',\n",
       " 'Assoc-acdm',\n",
       " 'Assoc-voc',\n",
       " 'Bachelors',\n",
       " 'Doctorate',\n",
       " 'HS-grad',\n",
       " 'Masters',\n",
       " 'Preschool',\n",
       " 'Prof-school',\n",
       " 'Some-college'}"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set(adult_salary[\"education\"])\n",
    "#looks like nothing missing in education"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Divorced',\n",
       " 'Married-AF-spouse',\n",
       " 'Married-civ-spouse',\n",
       " 'Married-spouse-absent',\n",
       " 'Never-married',\n",
       " 'Separated',\n",
       " 'Widowed'}"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set(adult_salary[\"marital\"])\n",
    "#nothing missing in marital status"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'?',\n",
       " 'Adm-clerical',\n",
       " 'Armed-Forces',\n",
       " 'Craft-repair',\n",
       " 'Exec-managerial',\n",
       " 'Farming-fishing',\n",
       " 'Handlers-cleaners',\n",
       " 'Machine-op-inspct',\n",
       " 'Other-service',\n",
       " 'Priv-house-serv',\n",
       " 'Prof-specialty',\n",
       " 'Protective-serv',\n",
       " 'Sales',\n",
       " 'Tech-support',\n",
       " 'Transport-moving'}"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set(adult_salary[\"occupation\"])\n",
    "#looks like some missing occupation details, should definitely drop any data where its a ? as occupation is \n",
    "#probably a very important factor in determining income"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "adult_salary = adult_salary[adult_salary[\"occupation\"] != \"?\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Adm-clerical',\n",
       " 'Armed-Forces',\n",
       " 'Craft-repair',\n",
       " 'Exec-managerial',\n",
       " 'Farming-fishing',\n",
       " 'Handlers-cleaners',\n",
       " 'Machine-op-inspct',\n",
       " 'Other-service',\n",
       " 'Priv-house-serv',\n",
       " 'Prof-specialty',\n",
       " 'Protective-serv',\n",
       " 'Sales',\n",
       " 'Tech-support',\n",
       " 'Transport-moving'}"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set(adult_salary[\"occupation\"])\n",
    "#relationship column is all set, probably not much of a factor so doesn't matter either way"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Amer-Indian-Eskimo', 'Asian-Pac-Islander', 'Black', 'Other', 'White'}"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set(adult_salary[\"race\"])\n",
    "#nothing missing here, 'other' might impact model, potentially making this column irrelevant if not a lot of people\n",
    "#report race, we will check correlations later, and drop if needed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Female', 'Male'}"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set(adult_salary[\"sex\"])\n",
    "#all set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'?',\n",
       " 'Cambodia',\n",
       " 'Canada',\n",
       " 'China',\n",
       " 'Columbia',\n",
       " 'Cuba',\n",
       " 'Dominican-Republic',\n",
       " 'Ecuador',\n",
       " 'El-Salvador',\n",
       " 'England',\n",
       " 'France',\n",
       " 'Germany',\n",
       " 'Greece',\n",
       " 'Guatemala',\n",
       " 'Haiti',\n",
       " 'Holand-Netherlands',\n",
       " 'Honduras',\n",
       " 'Hong',\n",
       " 'Hungary',\n",
       " 'India',\n",
       " 'Iran',\n",
       " 'Ireland',\n",
       " 'Italy',\n",
       " 'Jamaica',\n",
       " 'Japan',\n",
       " 'Laos',\n",
       " 'Mexico',\n",
       " 'Nicaragua',\n",
       " 'Outlying-US(Guam-USVI-etc)',\n",
       " 'Peru',\n",
       " 'Philippines',\n",
       " 'Poland',\n",
       " 'Portugal',\n",
       " 'Puerto-Rico',\n",
       " 'Scotland',\n",
       " 'South',\n",
       " 'Taiwan',\n",
       " 'Thailand',\n",
       " 'Trinadad&Tobago',\n",
       " 'United-States',\n",
       " 'Vietnam',\n",
       " 'Yugoslavia'}"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set(adult_salary[\"country\"])\n",
    "#? inidcating misisng country, will drop these datapoints"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "adult_salary = adult_salary[adult_salary[\"country\"] != \"?\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'<=50K', '>50K'}"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#lets check if anything missing in target variable\n",
    "set(adult_salary[\"income\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "False\n",
      "False\n",
      "False\n",
      "False\n"
     ]
    }
   ],
   "source": [
    "#categorical data seems normal, lets hceck for negative values in capital gains/losses, education, hrs per week and\n",
    "#education num as these numbers should not be negative\n",
    "print(min(adult_salary[\"education_num\"]) < 0)\n",
    "print(min(adult_salary[\"capital_gain\"]) < 0)\n",
    "print(min(adult_salary[\"capital_loss\"]) < 0)\n",
    "print(min(adult_salary[\"hr_per_week\"]) < 0)\n",
    "#all false so no negative values, assuming now that data is clean"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fdaf3da52e8>"
      ]
     },
     "execution_count": 152,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "corr = adult_salary.corr()\n",
    "sns.heatmap(corr, \n",
    "        xticklabels=corr.columns,\n",
    "        yticklabels=corr.columns)\n",
    "#heatmap of non categorigcal features, nothing highly correlated to each other, so we should initially use all features \n",
    "#for training, and then eliminate some"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 232,
   "metadata": {},
   "outputs": [],
   "source": [
    "#lets split into features and target\n",
    "features, target = adult_salary.iloc[:, :14], adult_salary[\"income\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 233,
   "metadata": {},
   "outputs": [],
   "source": [
    "#lets set target variable to 0 if below or equal 50K and 1 if above 50K\n",
    "target = target.astype('category').cat.codes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['age', 'type_employer', 'fnlwgt', 'education', 'education_num',\n",
       "       'marital', 'occupation', 'relationship', 'race', 'sex', 'capital_gain',\n",
       "       'capital_loss', 'hr_per_week', 'country'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 236,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 237,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fdb2b9a95c0>"
      ]
     },
     "execution_count": 237,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#encoded categorical features for the purpose of this visualization\n",
    "#feature correlation\n",
    "from yellowbrick.target import FeatureCorrelation\n",
    "encoded_features = features.copy()\n",
    "encoded_features[\"type_employer\"] = encoded_features[\"type_employer\"].astype('category').cat.codes\n",
    "encoded_features[\"education\"] = encoded_features[\"education\"].astype('category').cat.codes\n",
    "encoded_features[\"marital\"] = encoded_features[\"marital\"].astype('category').cat.codes\n",
    "encoded_features[\"occupation\"] = encoded_features[\"occupation\"].astype('category').cat.codes\n",
    "encoded_features[\"relationship\"] = encoded_features[\"relationship\"].astype('category').cat.codes\n",
    "encoded_features[\"race\"] = encoded_features[\"race\"].astype('category').cat.codes\n",
    "encoded_features[\"sex\"] = encoded_features[\"sex\"].astype('category').cat.codes\n",
    "encoded_features[\"country\"] = encoded_features[\"country\"].astype('category').cat.codes\n",
    "\n",
    "\n",
    "visualizer = FeatureCorrelation(labels=encoded_features.columns)\n",
    "visualizer.fit(encoded_features, target)       \n",
    "visualizer.show()           "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 238,
   "metadata": {},
   "outputs": [],
   "source": [
    "#cant judge categorical variables from above visualization, they will be split up below and explained later.\n",
    "#but age has a very high correlation to income as expected, could be correlated to how much further somoene is in\n",
    "#their career and could be them making more money at later stages.\n",
    "#education_num is the most highly correlated indicating people with more secialized education tend to find higher paying\n",
    "#jobs\n",
    "#hrs per week is also highly correlated, and is in line with the thinking that a lot of high paying jobs have long hours,\n",
    "#ex: being an Investment Banker at Goldman Sachs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>fnlwgt</th>\n",
       "      <th>education_num</th>\n",
       "      <th>capital_gain</th>\n",
       "      <th>capital_loss</th>\n",
       "      <th>hr_per_week</th>\n",
       "      <th>employer_Federal-gov</th>\n",
       "      <th>employer_Local-gov</th>\n",
       "      <th>employer_Private</th>\n",
       "      <th>employer_Self-emp-inc</th>\n",
       "      <th>...</th>\n",
       "      <th>country_Portugal</th>\n",
       "      <th>country_Puerto-Rico</th>\n",
       "      <th>country_Scotland</th>\n",
       "      <th>country_South</th>\n",
       "      <th>country_Taiwan</th>\n",
       "      <th>country_Thailand</th>\n",
       "      <th>country_Trinadad&amp;Tobago</th>\n",
       "      <th>country_United-States</th>\n",
       "      <th>country_Vietnam</th>\n",
       "      <th>country_Yugoslavia</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>39</td>\n",
       "      <td>77516</td>\n",
       "      <td>13</td>\n",
       "      <td>2174</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>50</td>\n",
       "      <td>83311</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>38</td>\n",
       "      <td>215646</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>53</td>\n",
       "      <td>234721</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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",
       "      <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>28</td>\n",
       "      <td>338409</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 104 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   age  fnlwgt  education_num  capital_gain  capital_loss  hr_per_week  \\\n",
       "0   39   77516             13          2174             0           40   \n",
       "1   50   83311             13             0             0           13   \n",
       "2   38  215646              9             0             0           40   \n",
       "3   53  234721              7             0             0           40   \n",
       "4   28  338409             13             0             0           40   \n",
       "\n",
       "   employer_Federal-gov  employer_Local-gov  employer_Private  \\\n",
       "0                     0                   0                 0   \n",
       "1                     0                   0                 0   \n",
       "2                     0                   0                 1   \n",
       "3                     0                   0                 1   \n",
       "4                     0                   0                 1   \n",
       "\n",
       "   employer_Self-emp-inc  ...  country_Portugal  country_Puerto-Rico  \\\n",
       "0                      0  ...                 0                    0   \n",
       "1                      0  ...                 0                    0   \n",
       "2                      0  ...                 0                    0   \n",
       "3                      0  ...                 0                    0   \n",
       "4                      0  ...                 0                    0   \n",
       "\n",
       "   country_Scotland  country_South  country_Taiwan  country_Thailand  \\\n",
       "0                 0              0               0                 0   \n",
       "1                 0              0               0                 0   \n",
       "2                 0              0               0                 0   \n",
       "3                 0              0               0                 0   \n",
       "4                 0              0               0                 0   \n",
       "\n",
       "   country_Trinadad&Tobago  country_United-States  country_Vietnam  \\\n",
       "0                        0                      1                0   \n",
       "1                        0                      1                0   \n",
       "2                        0                      1                0   \n",
       "3                        0                      1                0   \n",
       "4                        0                      0                0   \n",
       "\n",
       "   country_Yugoslavia  \n",
       "0                   0  \n",
       "1                   0  \n",
       "2                   0  \n",
       "3                   0  \n",
       "4                   0  \n",
       "\n",
       "[5 rows x 104 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#lets also encode categorical features, may be a good idea to one hot encode, so every instance in a feature\n",
    "#has its own column with a true or false, so models don't inappropriately assign weights\n",
    "features = pd.get_dummies(features, columns=[\"type_employer\", \"education\", \"marital\", \"occupation\", \"relationship\"\n",
    "                                 , \"race\", \"sex\", \"country\"], prefix=[\"employer\", \"education\", \"marital\", \"occupation\", \"relationship\"\n",
    "                                 , \"race\", \"sex\", \"country\"])\n",
    "features.head()\n",
    "#to do"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 231,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fdb2bbd9f60>"
      ]
     },
     "execution_count": 231,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#tough to visualize a 100 features, let's do some principal omponen analysis to decompose the features into fewer \n",
    "#dimensions\n",
    "from yellowbrick.features import PCA\n",
    "\n",
    "classes = [\"<=50K\", \">50K\"]\n",
    "\n",
    "visualizer = PCA(scale=True, classes=classes)\n",
    "visualizer.fit_transform(features, target)\n",
    "visualizer.show()\n",
    "#based on this diagram, the decision boundary looks curved with an upwards hump,\n",
    "#Logistic Regression or polynomail regression with degree 2 could be good here"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "#lets split data into training and testing\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(features, target, test_size=0.25, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "#preprocessing featurs so loss functions can converge faster\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "trainscaler = MinMaxScaler().fit(X_train)\n",
    "X_train = trainscaler.transform(X_train)\n",
    "testscaler = MinMaxScaler().fit(X_test)\n",
    "X_test = trainscaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/vaibhavguglani/anaconda3/lib/python3.6/site-packages/sklearn/linear_model/_logistic.py:765: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
     ]
    }
   ],
   "source": [
    "#only two discrete target variables, this is a classification problem\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "#lets try logistic regression\n",
    "logreg = LogisticRegression().fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8525394510011934"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#prediction and accuracy\n",
    "ypred = logreg.predict(X_test)\n",
    "\n",
    "from sklearn.metrics import accuracy_score\n",
    "accuracy_score(y_test, ypred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8522742341864474"
      ]
     },
     "execution_count": 186,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#failed to converge lets try using a different solver\n",
    "logreg = LogisticRegression(solver=\"liblinear\").fit(X_train, y_train)\n",
    "#prediction and accuracy\n",
    "ypred = logreg.predict(X_test)\n",
    "\n",
    "\n",
    "accuracy_score(y_test, ypred)\n",
    "#0.85% accuracy still\n",
    "#maybe using different penalty may improve performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8553242275560271"
      ]
     },
     "execution_count": 184,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "logreg = LogisticRegression(penalty=\"l1\", solver=\"liblinear\").fit(X_train, y_train)\n",
    "#prediction and accuracy\n",
    "ypred = logreg.predict(X_test)\n",
    "\n",
    "\n",
    "accuracy_score(y_test, ypred)\n",
    "#slightly better with l1 penalty 0.855 vs 0.852, lasso is eliminating some variables it deems unecessary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "22654\n",
      "7508\n"
     ]
    }
   ],
   "source": [
    "print(len(target[target==0]))\n",
    "print(len(target[target==1]))\n",
    "#high sccuracydoesnt necessarily mean its an amazing model, dataset is unbalanced with there being much more people\n",
    "#making less than or equal to 50K, lets check out ROC curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fdb38aaa278>"
      ]
     },
     "execution_count": 180,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from yellowbrick.classifier import ClassificationScoreVisualizer\n",
    "\n",
    "\n",
    "  \n",
    "from yellowbrick.classifier import ROCAUC\n",
    "\n",
    "\n",
    "\n",
    "visualizer = ROCAUC(logreg, classes=[\"<=50K\", \">50K\"])\n",
    "\n",
    "visualizer.fit(X_train, y_train)        \n",
    "visualizer.score(X_test, y_test)        \n",
    "visualizer.show()    \n",
    "\n",
    "#the high AUC scores suggest that the model is good at distinguishing between the two classes and the dataset imbalance \n",
    "#is not having a negative impact on it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fdb3edcb978>"
      ]
     },
     "execution_count": 182,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from yellowbrick.classifier import ClassPredictionError\n",
    "\n",
    "visualizer = ClassPredictionError(\n",
    "    logreg, classes=[\"<=50K\", \">50K\"]\n",
    ")\n",
    "\n",
    "\n",
    "visualizer.fit(X_train, y_train)\n",
    "\n",
    "\n",
    "visualizer.score(X_test, y_test)\n",
    "\n",
    "\n",
    "visualizer.show()\n",
    "#based on this visualization, it seems that the model is better at classifying people making less than 50K\n",
    "#this is because a significantly larger number of people are making less than 50K and this is evident in the \n",
    "#class imbalance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.3746103080112907"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#we can also try a linear regression model that can output 1 or 0 depending on what the predicted value is closer to\n",
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "linreg = LinearRegression().fit(X_train, y_train)\n",
    "ypred = linreg.predict(X_test)\n",
    "\n",
    "from sklearn.metrics import r2_score\n",
    "r2_score(y_test, ypred)\n",
    "\n",
    "#very low r2 score but maybe rounding to closest value could change results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8408699111523671"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#rounds each result\n",
    "ypred_class = [i.round() for i in ypred]\n",
    "#now lets check accuracy\n",
    "accuracy_score(y_test, ypred_class)\n",
    "#pretty accurate but slightly worse than logistic regression, we can potentially try polynomial regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 187,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8558546611855191"
      ]
     },
     "execution_count": 187,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import PolynomialFeatures\n",
    "#polynomial could overfit so lets use Ridge regularization\n",
    "from sklearn.linear_model import Ridge\n",
    "#lets try degree 2\n",
    "poly = PolynomialFeatures(2)\n",
    "degree2 = poly.fit_transform(X_train)\n",
    "\n",
    "degree2_reg = Ridge().fit(degree2, y_train)\n",
    "ypred = degree2_reg.predict(PolynomialFeatures(2).fit_transform(X_test))\n",
    "ypred_class =  [i.round() for i in ypred]\n",
    "accuracy_score(y_test, ypred_class)\n",
    "#looks like this is slightly better than Logistic by a very small amount"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fdb38aea470>"
      ]
     },
     "execution_count": 192,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from yellowbrick.regressor import PredictionError\n",
    "visualizer = PredictionError(degree2_reg, hist=False)\n",
    "visualizer.fit(degree2, y_train)\n",
    "visualizer.score(PolynomialFeatures(2).fit_transform(X_test), y_test)\n",
    "visualizer.show()\n",
    "#the prediction error here does not really mean much, and accuracy is the better metric to use, as this is\n",
    "#a classification problem, not a regression problem"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8228351677496353"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "\n",
    "knn = KNeighborsClassifier().fit(X_train, y_train)\n",
    "ypred = knn.predict(X_test)\n",
    "accuracy_score(y_test, ypred)\n",
    "#by default uses only 5 neighbors, b a good amount of neighbors to use is equal to square root of datapoints"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8311894974141361"
      ]
     },
     "execution_count": 193,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "neighbors = int(np.sqrt(len(X_train)))\n",
    "\n",
    "knn = KNeighborsClassifier(n_neighbors=neighbors).fit(X_train, y_train)\n",
    "ypred = knn.predict(X_test)\n",
    "accuracy_score(y_test, ypred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 194,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8158069221588649"
      ]
     },
     "execution_count": 194,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#wont ty naive bayes as there is a mix of categorical and non categorical data, may not work here\n",
    "#can try decision tree, as better suited for such data\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "dectree = DecisionTreeClassifier().fit(X_train, y_train)\n",
    "ypred = dectree.predict(X_test)\n",
    "accuracy_score(y_test, ypred)\n",
    "#base decision tree not as good as Logistic Regression, may be worth tuning different parameters, could be overfitted"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 195,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 195,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ypredtrain = dectree.predict(X_train)\n",
    "accuracy_score(y_train, ypredtrain)\n",
    "#as suspected, heavily overfit with training accuracy of 100% lets mess around with depth"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 196,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAecAAAFXCAYAAACYx4YhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOzdd3hUZf7+8fe0TCaZ9EAgpACB0AKEgIJgQJqIoiI2cMX6BRvsriLWtfwQlWXVVXTRtWFXrCsCgtKLqBAJvUNIqAnpdZLJzO+PQBRFEjRhksn9uq5cmZlzzsznPITcOeV5HoPb7XYjIiIiDYbR0wWIiIjIyRTOIiIiDYzCWUREpIFROIuIiDQwCmcREZEGRuEsIiLSwJg9XYCIJ02dOpW1a9cCsGfPHlq1aoWvry8As2fPrn5ck3HjxnH//ffTrl27313nhRdeIDY2lpEjR/75wv+EutpngI0bN/Lpp58yZcoUNm3axGuvvcaMGTPqpM6xY8dy8OBBAgICTnr9jjvu4KKLLqqTzxBpqAzq5yxSZdCgQbzwwgt07drV06WcNX92nz///HMWLlzIf//73zqurCqc//KXvyiIpUnSkbPI73jxxRdJTU0lMzOTDh068MADD/Doo4+SnZ1NVlYWrVq14vnnnycsLKw65EpKSvj3v/9NdHQ0u3btory8nEcffZQ+ffrwwAMP0L59e2699Va6du3K+PHjWb16NZmZmdxwww3cdNNNVFZWMn36dJYsWUJAQADdunVjz549vPvuuyfVNnr0aG666abq4HrmmWdwu93cdNNN3H///eTm5gIwYMAA/v73v9d6nz/55BM+/PBDXC4XwcHBPPLII8TFxbFu3TqmTZuGy+UC4LbbbqNbt27MmDGDwsJCHnzwQUaOHMkTTzzB3LlzeeCBB7Db7ezYsYMjR47Qtm1bnnvuOfz9/Vm+fDnPPPMMRqORTp068d133/HBBx8QFRV1Rv8+CQkJDB48mO3bt/PMM88wZsyYk547HA6mT59OaWkpFouFv//97/Tv35/PP/+cTz/9lNLSUux2+2/aVqQhUDiLnMbBgweZO3cuZrOZt99+m8TERMaPH4/b7Wb8+PF8+eWX3HLLLSdts3HjRh577DE6derEm2++yUsvvUSfPn1OWqe8vJyQkBA++ugjNm/ezJgxYxgzZgxffPEFW7ZsYe7cuRgMBu64445T1nX11VfzxRdfcNFFF1FZWcmcOXN45513+Pjjj4mKiuLNN9+kpKSEhx9+mMLCwt+cGj6VH3/8kf/973+8//772Gw2Vq1axcSJE5k/fz4vvvgiN998M5dccgnbt29n9uzZDBs2jL/+9a8sXLiQp59+mh9++OGk99u8eTPvvPMOBoOBa665hgULFjBo0CDuu+8+3n77bTp27MgXX3zBF1988bs1TZ8+nZdffvmk19566y1CQkKoqKhg4MCBvPDCCwAnPc/NzeWSSy7h5Zdfpnv37uzatYvrr7+eTz/9FIDdu3ezZMkS7HZ7je0i4gkKZ5HTSExMxGyu+m9y4403sm7dOmbNmkVaWhq7du2ie/fuv9kmMjKSTp06AdC5c+ffDZ/BgwcD0KVLF8rLyykpKWH58uVcfvnlWK1WAK699tpTHtkNHz6c6dOnk5WVxdatW4mNjaV169YkJyczfvx4Dh8+TN++fZk0aVKtghlg2bJl7N+/n9GjR1e/lp+fT15eHsOHD2fKlCksWbKEvn37cs8999T4fsnJyfj4+AAQHx9Pfn4+69atIy4ujo4dOwJwxRVXMHXq1N99j/vuu++0p7V79ep1yucbN24kJiam+t+nffv2JCUl8eOPP2IwGOjQoYOCWRo0hbPIafj5+VU//te//sXGjRu58sor6d27N06nk1PdsvHLG6oMBsMp1wGqA9hgMADgdrur/xA4wWg8dYcKPz8/hg0bxty5c1m/fj1XX301AN26dWPx4sWsWbOG77//nquvvpr//Oc/JCUl1bivLpeLyy+/nMmTJ1c/z8zMJCgoiNGjRzNw4EBWr17NypUreemll5gzZ85p3+9U7WAymX7THr+3j7Xxy3+fXz4/cfr9l9xuN06nE4vF8pvtRBoadaUSqaVVq1Zx4403MnLkSMLCwvjuu++orKys088YMGAAc+bMoby8HKfTedpTvtdccw2ff/4569evZ9iwYUDVteeZM2cyZMgQHn74Ydq1a0daWlqtPrtfv37MmzePzMxMAD788ENuvPFGoOoa97Zt2xg1ahRPPPEEBQUF5OfnYzKZcDqdtd6/pKQk0tLS2L59OwALFy6koKCg+g+UutK9e3f27dvHxo0bAdi1axdr167l3HPPrdPPEakvOnIWqaW77rqL6dOnM3PmTEwmE0lJSaSnp9fpZ4waNYp9+/YxcuRI/Pz8iIqKwmaznXLdhIQEzGYzw4YNqz4Kv/HGG3nggQcYMWIEPj4+dOjQgREjRtTqs5OTkxk3bhy33HILBoMBu93OSy+9hMFg4N577+Wpp57i+eefx2g0MmHCBKKionC5XDz//PPcdddd3HDDDTV+RnBwMM899xz3338/RqOxeh9+bx9Pdc156NChTJgw4bSfExoaygsvvMATTzxBWVkZBoOBp59+mjZt2rB+/fpatYeIJ6krlUgDsmrVKrKzs7n88suBqj7JVqu1+lRzY1dUVMTMmTOZOHEiNpuNLVu2cNttt7Fy5co6P3oWacx05CzSgLRv35433niDN954g8rKSjp27Mjjjz/u6bLqjN1ux2KxcNVVV2E2mzGbzTz//PMKZpFf0ZGziIhIA6MbwkRERBoYhbOIiEgD0yCuObtcLoqLi7FYLLr2JCIiXs/tdlNRUYG/v/8p+/o3iHAuLi5m586dni5DRETkrIqPjz/lKH4NIpwtFgtQVeSJ4f5qsnnzZhISEuqzrCZB7Vh31JZ1R21Zd9SWdacu27K8vJydO3dW59+vNYhwPnEq28fHp3owhdo4k3Xl96kd647asu6oLeuO2rLu1HVb/t6lXN0QJiIi0sAonEVERBoYhbOIiEgDo3AWERFpYBTOIiIiDYzCWUREpIGpsSuVy+Xi8ccfZ8eOHfj4+DB16lRiY2Orl7/11lvMmzcPqJoofsKECbjdbvr370/r1q0BSExMZNKkSfWzByIiIl6mxnBetGgR5eXlzJ49m9TUVKZNm1Y9+XlGRgZz5szhk08+wWg0MmbMGIYMGYLNZqNLly688sor9b4DIiIi3qbG09opKSkkJycDVUfAmzdvrl7WokULXn/9dUwmEwaDAafTidVqZcuWLRw9epSxY8cybtw49u7dW397ICIi4mVqDOeioiLsdnv1c5PJhNPpBKqG3QwNDcXtdvPPf/6Tzp0706ZNG5o1a8b48eN59913ue2225g8eXL97cEppB3bTm7x0bP6mSIiInWlxtPadrud4uLi6uculwuz+efNHA4HDz30EP7+/jz22GMAJCQkYDKZAOjVqxeZmZm43e4aZ5z65VF5baSkpPzmNZfLxfqSL8gvCWBQ8wvP6P2aqlO1o/wxasu6o7asO2rLunO22rLGcE5KSmLp0qVcfPHFpKamEh8fX73M7XZz55130rt3b8aPH1/9+ksvvURwcDDjxo1j+/bttGzZslZTQSYkJNR63NKUlBR69ux5ymXfL/4Ss9nxu8vlZ6drRzkzasu6o7asO2rLulOXbelwOE57QFpjOA8dOpTVq1czevRo3G43Tz31FLNmzSImJgaXy8WPP/5IeXk5K1euBOCee+5h/PjxTJ48meXLl2MymXj66afrZGdqq9TpQ6C17Kx+poiISF2pMZyNRiNTpkw56bW4uLjqx5s2bTrldq+++uqfLO2Pq3TZ8DWXUFBWTKCvv8fqEBER+SO8chASg9EPgP05WR6uRERE5Mx5ZThbzVV3lx/Kz/ZwJSIiImfOK8PZ7hsEQGZhrocrEREROXNeGc4hflXhnF+S5+FKREREzpxXhnNEQCgAxeWFHq5ERETkzHllOLcKDgeg3Fnk4UpERETOnFeGc1RQVTgbKPFwJSIiImfOK8PZx2KlzGnCbNRAJCIi0vh4ZTgDlDl98LOUU+lyeboUERGRM+K14VzpthForeRIYXHNK4uIiDQgXhvOJmPVsJ3pORqIREREGhevDWdfS9UoYQfzNISniIg0Ll4bznZrIADZxRqIREREGhevDedQ/2AA8krzPVyJiIjImfHacG4RGAZAiaPAw5WIiIicGa8N56jjo4Q5K3W3toiINC5eG86BtqDjjzRKmIiINC5eG85Wsx+VLgMWowO32+3pckRERGrNa8PZYDBQVmklwFpBQVmFp8sRERGpNa8NZwCX20aQr5MDebruLCIijYdXh7PJ6I/Z6CYjL8fTpYiIiNSaV4ezr0/VKGFH8hXOIiLSeHh1OAf6Vt2xfaxY4SwiIo2HV4dz2PFRwvJLNRCJiIg0Hl4dzi0CqkYJK60o9HAlIiIitefV4RxuDwHAWVnk4UpERERqz6vD2e/4zFQGSj1ciYiISO15dTjbLHbcbvA1Oyh3Vnq6HBERkVrx6nA2Gk1UuHwIsjo5XKCjZxERaRy8OpwBXFSNEnYwXxNgiIhI4+D14Ww2+mOzuMjIy/d0KSIiIrXi9eFs86m6KexoQbaHKxEREakdrw/nQN+qcM4u1pGziIg0Dl4fzmHH+zoXlCqcRUSkcfD6cI4ICAU0SpiIiDQeXh/OdmvV5BeVLo0SJiIijYPXh7Pt+LSRRspwu90erkZERKRmTSCcq24Is1sryC52eLgaERGRmnl9OFtMPlS6zAT7VnBAA5GIiEgj4PXhDODWKGEiItKINIlwNpvs2H0qOZinm8JERKThaxLh7OcTgNEARwtzPV2KiIhIjZpEOAfZqrpTZRcrnEVEpOFrEuEcfnyUsMKyAg9XIiIiUrMmEc7BfsEAlGmUMBERaQSaRDjbLAEAuFzFHq5ERESkZk0jnH2qwtnH5KCk3OnhakRERE7PXNMKLpeLxx9/nB07duDj48PUqVOJjY2tXv7WW28xb948AAYMGMCECRMoKytj8uTJZGdn4+/vzz//+U9CQ0Prby9qcCKcg4/3dW7fLNBjtYiIiNSkxiPnRYsWUV5ezuzZs5k0aRLTpk2rXpaRkcGcOXP46KOP+Pjjj1m1ahXbt2/nww8/JD4+ng8++ICRI0cyc+bMet2JmljNfrjdRg1EIiIijUKN4ZySkkJycjIAiYmJbN68uXpZixYteP311zGZTBgMBpxOJ1ar9aRt+vfvz5o1a+qp/NoxGAxg0ChhIiLSONR4WruoqAi73V793GQy4XQ6MZvNWCwWQkNDcbvdTJ8+nc6dO9OmTRuKiooICKg6lezv709hYe3ukv5l8NdGSkpKrdetrDATaC3hh6076Yj6O//SmbSjnJ7asu6oLeuO2rLunK22rDGc7XY7xcU/3+Xscrkwm3/ezOFw8NBDD+Hv789jjz32m22Ki4sJDKzdNd6EhASsVmut1k1JSaFnz561Whdgf8p6CkrzMQf4n9F23u5M21F+n9qy7qgt647asu7UZVs6HI7THpDWeFo7KSmJFStWAJCamkp8fHz1MrfbzZ133kmHDh2YMmUKJpOpepvly5cDsGLFigbxgxHsVzVKWG5JvocrEREROb0aj5yHDh3K6tWrGT16NG63m6eeeopZs2YRExODy+Xixx9/pLy8nJUrVwJwzz33MGbMGO6//37GjBmDxWLh2WefrfcdqUmIXzDp2VCkUcJERKSBqzGcjUYjU6ZMOem1uLi46sebNm065XYzZsz4k6XVLT9r1an1MqdGCRMRkYatSQxCAuB3fJQw3CVUulyeLUZEROQ0mkw423yqjpwDrBUcLSzzcDUiIiK/rwmFc9WRc5BVfZ1FRKRhazrhbLHjdkOQr5MDCmcREWnAmkw4G40mDAZfgn2dHFI4i4hIA9ZkwhnAx2zXEJ4iItLgNalw9rcGYrO4OFyg7lQiItJwNalwDradGCVMY2uLiEjDVeMgJN7E7lsVzkUOHTmLSMNVVlHM/uzN5Jdk0SKoLS2D22Ex+Xi6LDmLmlQ4n+hOVV5RhNvtrppKUkSkAaiodJCRvY29WakczNuJ2101WNLWQ6swGsy0DI4jOrQjUSEdsfuG1Olnu91uKiodOJwllDtLcThLcFRUfa90VeBvDSbAN4xA3zAs5tpNTiR/TtMK5+OjhFnNDgrKKgiy6S9REfGcSpeTg7k72Ze1gYycrThdFQCE+kfStll3wuytOJy3h4ycbRzM3cHB3B3Al4T4tSAqtCOtQuIxGIyUV5TgOBGqzlLKfxGueY5cjmz4AZfbhdtdicvtOumxs7KCcmcpbmo3cqKvxU6gbxgBtjACfEMJ8A3DarZhMlowmyxV341V301GM2ajBaerghJHPiXlBZSUF1D8i8cljnxc7krC7FE0D4yleWAswX4RGA0N46qr2+0mt+QI2YUHcLrLz9rnNqlw9jt+5Bx8vK+zwllEzoTb7aawLJuj+fvIKzmK3TeUEP8WhPi1wGrxq3H7sopicooPkVN0iGNFBzmUt4tyZykAgb7htGnWnTbNuhPs17x6m5bB7UhqPYyisjwO5G7nQM42DuXtIffAMjYdWFarukuLjBgNJoyGqu8GgxGjwYjBYMTX4keQLRwfsw2rxQ+r2YbV7Ff13OyHyWimyJFLYVk2BaXZFJblkFWYQWbh/j/Uhr9mMVkBA/lZ69mbtb76tWYBMTQLiCEisDXhAdH4mH3r5PNqo7Ash8N5uzmct4fD+XsoqygCoIW5K9DvrNTQpMK5epSw492purQI9nBFItKQud1u8kuzOJK/l6MF+ziav4+S8lPPbOfnE0iwX0R1WAf7NaekvJCc4kNkFx0ip/ggxY7832zTPrInbZonEubf6rSX2uy+wXRs2YeOLftQUVnOkbzdHCnYh8lgPh6kNnx+E642UtdvpFevXnXaLi5XJUWOXArKsikqy6Hc6aDSVUGlqwLn8e+VLmf1Y5PBjJ81CD+fwKqv44/9fYKwmK243S7yS4+RWbCfzII0MgvTOZS3i0N5u6o/02Aw/uaI/JdH6YYajrSNBiMWkxWL2Vr1vfrLF4vJistdyZH8vRzO20ORI6d6O5tPAG2b9aBlcBx5GWdvXoYmHc4i0jS43C5yiw9zJH8fR/P3crRgP06X41e/pH/+RW0xWXE4SzhasI+yiuLq9/G12IkN60qLoDaE+rekyJFHbvERckuOkFdy5DeB8ks2SwCtQjoQ5h9JqD2SUP9IAnxDagyVU7GYfIgO60x0WOca162Pe2uMRhOBtnACbeF18n4Gg5Fgv+YE+zUnvsU5QNVZhqzCdDIL9pNddJCKSsfxsHdS6arA4Sypel7prPUp+dqwmHyJCe1My+B2tAxuR5CtWXUbphxIqbPPqUmTCmeLyYrBYCHIqlHCRDyloPQYZqNP9TSu9cHlqiS7+ODxMN7H0YI0Kip/nvCm6ganECoqHVRUOiirKKaishxwn/Q+fj6BtG2WSERQGyIC25z0ixog4lef63CWkld8lNySI+SXZOLrYyfMvxWh9sjqy2pSO74Wf6JDOxEd2qnGdV2uSlw1BLTLVVn9733Sl7OMispy3LhoHhBDqL1Vg7je3aTCGU6MElbA9jyFs8jZVFiWw7p9X7M/u2oO+CBbc1oGx9EyuB0tgtpiNdv+8Hu73S5yio9UXSfM38PR/H04XT/fvBPgG0br8ARaBLUlIrDNKe92dh+/OerEL22T0Yy/NfiMjjytZhsRQa2JCGr9h/dFzpzRaMKIqYaVLGf1uvWf1eTC2W4NpLQ8l0P5RZ4uRaRRcVSUsOvoOoodebRu1o3mAbG1Cq6KSgebMpax+eBKXG4n4QHRWM02jubvY/vhNWw/vAYDBsLsrY6fSozD6XZQUVmO2Wg+5WnfEzdmVd2ws5vDeXtxOH8+/Rxka0aLoDgigtrQIrBNrY7SDQZj1fVIdRWSBqDJhXOAbxDZRZBXcuqbOkTkZNlFB9l26Dv2HdtApcsJwLbD3xFoC6d9RC/imifh5/Pb8HO7XezJXE/K/gWUlhfi5xNIr9YX06ZZdwwGA5UuJ8cKMzh0/Gg3qzCdY0UHqu9A3rZmDgBGg+mkG3/MJjPlTgcl5T/fXOXnE0Rc8yQig9vRIigOf2tQ/TeMSD1qcuF84qawIofCWeT3VLqcpB3bxPbDa8gqTAeqTg13bNmHYL8I9mT+RNqxzaSkLeCntG+ICu1Au4heRId0xGg0kVmQzo97v+JYUQYmo5nu0YNJiBpw0ihXJqO56lpuUBt6MJSKSgdH8/dxOH8PGYf3EhBk//nu38qf7/4tLS/DYDASG5ZQfaQd6BuuQYXEqzS5cD5xU4bLXcLhghJaBtbcN1HEG5Q7y9iT+RNOVwVmo/l4l5STB4wwGc0cyNnOzqM/Hr9L2UBUSEc6tjyPViHtq08xtwqJp3dcCfuyNrLr6FoycraRkbMNX4udMHskB3N3AtA6vBu9Wg+v1YhWFpOVqNCORIV2xJiTQs8uPeuzOUQatCYXzidGCQv2reCFFduZNiLJwxWJ1C+3283erPWs3Te/ejCFmljNfnRp1Z+OLXsT4Bv2u+uc6HebU3SIXUfXVQ09mbuTUP9Iere9lIigNnW5KyJNRpML5xPXxloFwn/X7OTBwQkaKUy8VnbRIX7Y+yWZBfsxGS0kxgwhzN6q6nRx5ckDRVSdQnYS7NecNuHdMZsstf6cUHskve2X0avNxRSUHiPIr3mD6I4i0lg1uXA+cc25b+sAPtxYwSvf7eT+wQkerkqkbjkqSlif/g07Dv+AGzexYQmc0+aSOp8w4ddMRjMh/i3q9TNEmoImG85xYSYCfS28sHIbf+vfCV9LDX3kRBoBl9vF7qPrSElbgMNZQpCtGb3bXkZkSHtPlyYiZ6DJhbPV7IfRYKKisojbzuvKv5Zu4Z11exh/XrynSxP5wyoqyzmYu51NB1aQXXQAs8mHXq0vplNkX0zGJvffXKTRa3L/aw0GAzYfO6Xlhfytf0deWLGNZ5dt5dbe7TAZdY1MGo+KSgcHcraTdmwTB3J3UHl8usG2zXrQq/Xweh0eU0TqV5MLZwCbJZCc4kNEBPhywzltef373Xy+KYOru8d6ujSR06pwOsjI3c7+YxuPB3LVoCCBtma0Du9Km/BuuuYr4gWaZDiHB7TiWFEGacc2cu8FXXjjh91MX7KZq7rFaCADaXAKSrM5mLuTQ3k7OZS3qzqQg44HcuvwbgT7RehnV8SLNMlw7tKqPzsO/8iG9CWMTOrGqK4xfLYxncW7jjAkvqWnyxMv43a7OZy/h31ZqVgt/gT6hhHgG0agLQw/n8DfjB1dUengSN4eDubt4mDuTgrLsquXBdmaHw/krgpkES/WJMM5wDeUuOY92J2ZQtqxzdw3KIHPNqYzfclmhbPUqSP5e1m//1uOFuw75XKjwUygLZQA3zDs1hDySo5ytCANl7sSqBo1Kya0M61COhAZ0p4A39CzWb6IeEiTDGeAbtED2ZO5ng0Zi7m8x98Y3L4Fi3cdYV1GNr2iTz0ikkhtHc1PY336NxzJ3wtAVEhHukZdAEBB2TEKy3IoLM2moCybgtJj5JVkVm8b5t+KViHxRIa0p3lALEajuvmJNDVNNpwDbeG0bZ7InsyfSM/eyn2DEli86wjTl2zm4xsHeLo8aaQyC/aTmr6IQ3m7gKoxqBNjhtIsILp6nV/P9et2u3E4Sygqy8XfGlTdF19Emq4mG84A3aIGsjdzPakZi7m0+wSSokL5fFM6u7IKaN9M3VCkdpyVFVXXlB0r2bTxCACRwe1JjBlC88CaewAYDAZ8Lf74Wvzru1QRaSSadDgH+TWjTbPu7M1K5UDudu4blMDod1bwzLIt/Pfq8zxdnjRgxY58DuRsJyNnG4fz91T3MW4ZFEdizNDfHB2LiJyJJh3OAN2iB7E3awOp6Yu5ottdtAsP4J21e3l8WHdNJynV3G4Xx4oOkJGznQM528gpPly9LNgvgqjQjpRmmUjueqEHqxQRb9HkwznYr6prStqxjRzO38mkCzpzx6c/aDrJJsLtdnEgdye7j6ZQVlF0fGamqtmaKl2/nLXJCbgBMBpMRAa3J/r43MMnplRMyU7x4J6IiDdp8uEM0D16EGnHNrIhfTFje97O4ws38Mp3O7FZTPiaTVjNRqxm0/EvI74WEzaLmZ5RoUQE2DxdvvwBFZUOdh9NYduh7ygoO1b9usloxmS0YDKaMRst+JhtmIwWzEYzgbZwokI6EhncHovZ6sHqRcTbKZyBEP8WxIZ1ZX/2JrKL9nDvBV2Y/FUKU77ZeNrtDAY4L7YZIxOiubxrNO3CdRNZQ1dYlsP2w2vYeWQtFZVlGA1m2kf0olPLvoT4t/jNgCAiIp6gcD6ue/RA9mdvIjVjEX9LvoP+cREUOipwOCspq6jE4XThqKx6XO50UeCoYNHOw6zcm8l3aVncN/cnEloEc3lCNJcnRJMUFarRmxoAt9uN01VOdtFBth1aTXr2Vty4sVkCSGiVTHyL3th87J4uU0TkJArn40LtkcSEdiY9ZytHC/bQK7rm+W8fGtKVY0VlfLX1AF9uzuDbHYd5ctEmnly0iehgP+7s14HJA7sopOvZ0fx97M/egsNZQrmzFIezBEdFafXzE6NtQdUAH51bnU/r8K6aSlFEGiz9dvqF7jGDSc/ZSmrGIloGt6tVqIbbfbn53HbcfG47ihwVfLPjMP/bnM7cLQd4cN569uUU8dKoczUdZT3JyNnG0m3vnRTABgz4mP2wWmwE+IZiNduw+QTSrnlPmgfG6o8lEWnwFM6/EGZvRXRoJzJytnEkfw8tg9ud0fZ2q4VR3WIY1S2GzMJShr+6mFfX7KKgrIK3xvTDYlJA16WDuTtZuu09DAYjAzuMIdQeidXsh8Xko2vHItKoKZx/pXv0YDJytpGavpgQ/5a43S5cbhcud2X14xPfA23hWEw+p3yf5gE2Ft95IZe9voSP1qdRUFbBxzf2x2ZRk9eFw3l7WLLtHcDA4M43EBlc82UIEZHGQknxK+EBUUSFdOBA7g4++uGJ065rNJiICGxDq5B4WouebqUAACAASURBVIXE/2YKv2CbDwtuG8KVby1n/raDXPLaEv53ywUE+p460KV2jhaksXjr27jdbgZ1UjCLiPdROJ9CrzaXYDJacOPGaDBhNBgxGIwYDUaMBtPxU6ZuMgvSOZy/m8P5u1mXNh8/n0Aig9tXTe8X3A6rxQ8/HzNf3nIB17+/is82pjP0lUXM+79BhNt9Pb2bjVJWYQaLtsyi0u1kYMe/EBXawdMliYjUOYXzKQT7NWdgp+trtW5JeSGHcndyKG8XB3N3sjszhd2ZKRgwEBueQL/2V+Nj9uHDscnc9sn3zPpxDwNnfsOC24bQKkjDg56J7KKDfLv5DZyV5QzoOIaYsC6eLklEpF4onP8kP58A2kX0pF1ET1xuFzlFBzmYu5P92ZtJO7aJIkceQzrfhK/Fn9euOY8gXx+eX7GNAS8t5Jvbh9A2rOlOD1jpcrLj8PfszvyJAN8Qmge2pllALGH2yN90c8otPsI3m9+gvNJBcvw1tA7v5qGqRUTqX43h7HK5ePzxx9mxYwc+Pj5MnTqV2NiTp8HLyclhzJgxzJkzB6vVitvtpn///rRu3RqAxMREJk2aVC870JAYDUbCA6IJD4ima9QFrNr1KXuz1vP1xlcY2uVW7L7BPHNZT4JtPjy+cAPnv7iA6Zf25LoebTAam073Hrfbxb6sjfy0/xuKHDkYMJBTfIj92VsAMBrMhAe0onlALM0DY/G12Fmy7V0czhL6tbuSuOY9PLwHIiL1q8ZwXrRoEeXl5cyePZvU1FSmTZvGyy+/XL185cqVPPvss2RlZVW/lp6eTpcuXXjllVfqp+pGwGg0kRx/NTYfO1sOrmT+xpkM7XIrIf4RPHJhN0L9fJj8VQo3frCal1Zu55nLenF+2+aeLrveHcrdxbq0r8kpPoTRYKJzZD+6RQ/CWVlOZuF+Mgv2k1WQTlZBOpkF++Hgz9v2iRtJ+xbneK54EZGzpMZwTklJITk5Gag6At68efNJy41GI7NmzeLKK6+sfm3Lli0cPXqUsWPH4uvry4MPPkjbtm3ruPSGz2Awck6bS7BZAliXNp+vN77M4C43ERHYmrvO78glnaN4aN56ZqemMeA/C7mqeyzTLulBGy881Z1ddJCUtAUcytsFQNtmifSIvZAA39CqFSz+2H1DaNssEaiamOJY4QEyC/dzrDCDqNCOdGjR21Pli4icVQa32+0+3QoPP/wwF154IQMGDADgggsuYNGiRZjNJ+f6oEGD+Prrr7Faraxdu5Zjx44xfPhw1q1bx9NPP81nn332u5/hcDh+E/reJteZxoGKdRgwEONzHoGmyOplG7NKeP6no2zOLsViNDC6Qyg3dwnH7mPyYMV/ntNdTpkrl5zKNPIr0wGwGyNoYemKzRji4epERDwvISEBq/W3s9zVeORst9spLi6ufu5yuX4TzKf6MJOpKlh69epFZmYmbre7xmETf6/IU0lJSaFnz561Wrdh6MmBnC4s3f4+6eVr6Nt+FO0jeh1fAjcNc/PR+jQemr+ed7dlsyCjiMeHJTKuT7t6HfqzLtrR7XZTUl5ATtFBsosPkVN0iOziQxQ78qrXCfNvRc82F3l1n+TG9zPZcKkt647asu7UZVvWdFBaYzgnJSWxdOlSLr74YlJTU4mPj6/xQ1966SWCg4MZN24c27dvp2XLlhrPGIgK7ciwhP9j0da3WL3rU0rLC0mIGnC8H7WBMUltGNk1mhdWbOPpxZu567MfmL/tAB9cn4zdavF0+dWqTjlnkFmw//hp54M4nMUnreNrsdMqJJ5Q/0iaB7YmKiReQ2qKiNRSjeE8dOhQVq9ezejRo3G73Tz11FPMmjWLmJgYBg8efMptxo8fz+TJk1m+fDkmk4mnn366zgtvrJoHxnJxt9v5ZvOb/LR/IXsy15MYM5jW4V0xGIzYLGYeGNyVm85px00frmbe1oMMeGkhX946kKhg/zqrw+EsJb8ki/zKDDJy/LCYrMe/fKu+m62YDFU/HkWO3KobtI7fsJVbfBg3P18NsVtDaBHUmlD/SELtrQizR+Lno7mtRUT+qBqvOZ8NJw7vvfu09slKHAWsT/+G3Ud/wo2LYL8IEmOGEBvWpfoI01npYuIXP/Lqml1EBtqYc+sgekSF1voz3G43xY588kszyS/JJL80q+qrJIvSisIatzcYjJiMZpyV5dWv/dzNqTXNA2NoFhCr+ZCPa+w/kw2J2rLuqC3rTn2c1v7D15ylfvhZA+nX/iq6Rg1kQ8Zi9mauZ9n29wnxa0FizBBiwrpgNhmZeWVv2ocHct/cFPr/ZwEfXJ/MpV2iT/veRWV57MlMYdfRFIocOb9aasBuDaFVSAeCbM3IzSoislVLKiodVDjLqr7/4stZWU6gLZzmgbG/O0CIiIjULf2W9bBAWxjJ8dfQLXogG9KXsC8rlaXb3yPUP5LEmMFEBrfnngs60zbMzvXvr+KKWct47rJeTEzueNJ1/EqXk/Tsrew6uu54dyU3ZqOF2LAEQvxbEGRrRpCtGYG2ZphNP1+/TslNoWuU/qoWEWlIFM4NRJCtGf07XFsV0hmL2Ze1kSXb3qXqSDeYIL9mfDgmgDd+zOflVcvZl53F9Ev7UVh2lF1H17E3KxWHswSAZgExtI84hzbh3bCYa3eZQEREGg6FcwMT7NecAR3G0C1qEDuP/EBuyRHyS7M4mLsTgJGdgE4Aaby1eiEWUyVQdXd0Qqv+tIvoRbCf9480JiLizRTODVSIfwS94y6rfu5wllJw/GauzMIjLNm1HYO7gJKKQIZ1Hsj5cT0xGhv3oCUiIlJFHU8bCavZRrOAGNpF9KRvu0t44MK7yam4lGkrIhn++m5e+2EPDeDGexERqQMK50bKbDLy7OW9+OLmC7BZTNz56Q+MeXcl+aXlNW8sIiINmsK5kbssIZr1k0bQr3UzPtmwn17/nsfa9GOeLktERP4EhbMXiA7xZ8mdF/LQkAT25RSR/NJCnl++Vae5RUQaKYWzlzCbjDwxvAcLxg8h1M+HSXNSuPzNpWQXOzxdmoiInCGFs5cZEt+Sn+4ZwZD4lszbepAez85l2e4jni5LRETOgMLZC7UItPH1uME8eXEiRwpLGfLKtzzy9XoqKl2eLk1ERGpB4eyljEYDDwzuyooJw2gdYuepRZu54D8L2Ztd84QXIiLiWQpnL9cnthkp91zCmB6t+X7/MZKenccHP+3zdFkiInIaCucmIMjmw3vXJ/PWmH64cTP2/VXc9OFqCssqPF2aiIicgsK5CRnbqy0p91zCOdFhvLtuLz2fm8fW7FJPlyUiIr+icG5i2oUHsmLCMO4f1IW9OYXc+s0+3lm3x9NliYjILyicmyAfs4mnLkli4fgh+FuM3Pzhd7ywYpunyxIRkeMUzk3Y4PiW/HdIa1oG2rjny3U88vV6jSomItIAKJybuLhgX1ZOGEa78ACeWrSZOz/7gUqX+kOLiHiSwlloExbAignDSIwM4dU1uxjz7koczkpPlyUi0mQpnAWAiAAbS+68kAFxEXy2MZ1LX1+irlYiIh6icJZqQTYf5o8bzGVdoli86whDX/mWY0Vlni5LRKTJUTjLSXwtJj65cQA3nhPH2oxsBvxnIftzijxdlohIk6Jwlt8wm4y8ce15TLqgM9szC+j173nM3XrA02WJiDQZCmc5JYPBwPRLe/LyVb0pLndy+RtLuf+rFM1sJSJyFiic5bTGnxfPd38dTvvwAJ5ZtpWB//mGjNxiT5clIuLVFM5So8RWofx498VckxjLmv1ZJD03l3k6zS0iUm8UzlIrgb4+fHB9MjOPn+a+7I2lPDD3J53mFhGpBwpnqTWDwcBtx09ztwsP4F9LtzBopk5zi4jUNYWznLHEVqGsPX6a+7u0LHo+N48F2w96uiwREa+hcJY/5MRp7peuPJdCRwUjXl/CYwtSNS63iEgdUDjLH2YwGLijbwdWTryI2BB/pn67ieGvLiazsNTTpYmINGoKZ/nTekWHse7uSxjRuWrYz6Tn5rFy71FPlyUi0mgpnKVOhPhZ+eLmC5h2SRKZRWUMfvlbnlm6RfNDi4j8AQpnqTNGo4HJg7qw6PahNLf7cv/cn7hi1jJySxyeLk1EpFFROEud6x8XQco9lzCoXQu+2nKAc5+fT7q6W4mI1JrCWepFRICNBbcN5r6BXdibXcSw/y7iqG4UExGpFYWz1BuT0cjTI5K4f1AXdmYVMPzVxeSVlnu6LBGRBk/hLPXuyYt7cNt58Ww4lMulry+h2FHh6ZJERBo0hbPUO4PBwEujzmVMj9Z8l5bFqLeW43BWerosEZEGS+EsZ4XRaGDWmH6M6BzFop2H+ct7q3Bq0gwRkVNSOMtZYzEZmX1Dfwa2i+CLTemM+3gNLpf6QYuI/JrCWc4qX4uJL24eyDnRYbyzbi93f7lWA5WIiPyKwlnOugBfC/PGDSahRTAvrdrB/1u40dMliYg0KApn8YgwfysLbhtM2zA7T3y7kccWpOoIWkTkuBrD2eVy8eijj3LttdcyduxY9u/f/5t1cnJyGDZsGA5H1TCNZWVlTJw4keuuu45x48aRk5NT95VLo9cy0I9vbhtC2zA7U7/dxK2z11Chm8RERGoO50WLFlFeXs7s2bOZNGkS06ZNO2n5ypUrueWWW8jKyqp+7cMPPyQ+Pp4PPviAkSNHMnPmzLqvXLxCm7AAVk+8iHOiw3h77R4ue2MphWXqBy0iTVuN4ZySkkJycjIAiYmJbN68+eQ3MBqZNWsWwcHBp9ymf//+rFmzpi5rFi/TPMDG4juGcnGnVnyz4xADZ37D4YIST5clIuIx5ppWKCoqwm63Vz83mUw4nU7M5qpN+/Xrd8ptAgICAPD396ewsLBWxfw6+GuSkpJyRuvLqTWUdvxH90B8yov5354cznlmDi9cEEPrIKunyzojDaUtvYHasu6oLevO2WrLGsPZbrdTXPzzjEIul6s6mGuzTXFxMYGBgbUqJiEhAau1dr+MU1JS6NmzZ63Wld/X0Nrx015unly0iccWbOD2pRn875aB9GvT3NNl1UpDa8vGTG1Zd9SWdacu29LhcJz2gLTG09pJSUmsWLECgNTUVOLj42v80KSkJJYvXw7AihUr9IMhtWYwGPjH0G68fu155JdVMPSVb/l8Y7qnyxIROatqDOehQ4fi4+PD6NGjefrpp3nwwQeZNWsWixcv/t1txowZw65duxgzZgyzZ89mwoQJdVq0eL+bz23HV7cOwmIycs07y5m5aoenSxIROWtqPK1tNBqZMmXKSa/FxcX9Zr0lS5ZUP7bZbMyYMaMOypOmbFjHSJbeeSEjXl/CxC9+pKi8gvsGJXi6LBGReqdBSKRBS4oKY/ldw4gO9uPBeet59GsNViIi3k/hLA1e+2aBLL9rGHFhATy5aBP3zklRQIuIV1M4S6MQG2pn2V0X0ikiiOdXbOOOT3/QjFYi4rUUztJoRAb5sfTOC0mMDOG173dx00erNSe0iHglhbM0Ks3sviy+80L6xIbzfso+Rr+7knJnpafLEhGpUwpnaXSCbT4sGD+EC+Ii+GJTOlfMWkZphdPTZYmI1BmFszRKAb4W5o4bxEUdI1mw/RAjXltCsUMTZoiId1A4S6Nls5j5/OYLuKJrDMv2HOXSN5YqoEXEKyicpVGzmk18ODaZK7vFsHzPUUa8voQiBbSINHIKZ2n0LCYj71+fzFXdY1mxN1MBLSKNnsJZvILFZOS9v5zP1d1jWbk3k0teW0JhmQJaRBonhbN4jRMBfU1iLKv2ZXLJa4sV0CLSKCmcxauYTUbeve58rk1szeq0LAW0iDRKCmfxOmaTkXeu68foHlUBffFriykoK/d0WSIitaZwFq9kNhl5e0w/xvRozXdpWVz86hJySxyeLktEpFYUzuK1zCYjb43px3VJbVizP4vklxayP6fI02WJiNRI4SxerSqg+/K3/h3ZdjSfvjMWkJKR7emyREROS+EsXs9kNPLc5efw78t7cbSolIEzv2H+toOeLktE5HcpnKXJ+Gv/Tnxy4wAqXW4uf2Mp/12z09MliYicksJZmpQrusaw+M6hhPr5cOenP/DQvJ9wudyeLktE5CQKZ2ly+sQ2Y/VfL6J9eAD/XLKF699fhUNzQotIA6JwliapXXggqyZeRN/WzZidmsZF/11EdrG6WolIw6BwliYr3O7LN7cP4cpuMazYm0niM1/xzY5Dni5LREThLE2bzWLmo7H9mTo8kcyiMoa/upiJn/9ISbnT06WJSBOmcJYmz2g08OCQrqz523A6RwQxc/UOej43j7Xpxzxdmog0UQpnkeOSosL48e6L+Xv/TuzMKqDfiwv4fws3UFHp8nRpItLEKJxFfsFmMfPs5b349vYhRAbamPLNRpJfXMCOzHxPlyYiTYjCWeQUBrVvSeq9l/KXnm1Ym5FNz+fm8Z9V29UnWkTOCoWzyO8ItvnwznXnM/uG/tgsJv76xVqG/XcRaZo8Q0TqmcJZpAZXdY9l4+RLGdE5iiW7j9D9ma947ftduN06ihaR+qFwFqmFloF+/O+WC3hzdF9MBgO3f/I9l7y+hAN5xZ4uTUS8kMJZpJYMBgM3nhPHxsmXcmGHSBZuP0S3f33FO+v26ChaROqUwlnkDEUF+zN/3CBeuboPlW43N3/4HVfMWsaxUg1cIiJ1Q+Es8gcYDAbG9WnPhnsvZWC7CL7acoAx8/fwxaZ0T5cmIl5A4SzyJ7QOtfPNbUOZccU5lDldXPXWcm77ZA3FjgpPlyYijZjCWeRPMhoN3HV+R96+qC3dI0N4/fvd9Pr3fNZlZHu6NBFppBTOInWkbZCVNX8bzqQLOlcN/znja6Yt3kSlS8N/isiZUTiL1CGr2cT0S3uy8LYhNLf78vD8VIa8/C3puepyJSK1p3AWqQdD4quG/xzZNbp6rujZ69M8XZaINBIKZ5F6EuZv5dMbB/DqNX2ocLm47r2V/OW9lWQXOzxdmog0cApnkXpkMBi4tXd7frpnBH1iw/lofRpd/zWH/6nLlYichsJZ5Cxo3yyQFROG8c8RSeSVlnPlW8sZ+/4qckp0FC0iv6VwFjlLTEYj9w7sQso9Izg3JowPftpH1+lfMWdzhqdLE5EGRuEscpZ1ighi5YSLeOriHuSUOLhi1jJu/GA1uTqKFpHjFM4iHmA2Gbl/cALr7rmEXtFhvJeyl67/+oovdRQtIiicRTyqS4tgVk+8iKnDEzlW7GDUrGWMmrWMDPWLFmnSFM4iHmY2GXlwSFfWTxpBctvmfLk5g4R/zeGFFdtwVmp0MZGmqMZwdrlcPProo1x77bWMHTuW/fv3n7T8448/ZtSoUVxzzTUsXboUgLy8PHr37s3YsWMZO3Ysb7/9dv1UL+JFOkUEseSOC3ntmvPwMRm558t1nDfja43RLdIEmWtaYdGiRZSXlzN79mxSU1OZNm0aL7/8MgBZWVm8++67fPbZZzgcDq677jr69evH1q1bGTFiBI888ki974CINzEaDdzSux2Xdonivq9SeGfdXs574WvuOr8DUy7qTqCvj6dLFJGzoMYj55SUFJKTkwFITExk8+bN1cs2btxIjx498PHxISAggJiYGLZv387mzZvZsmUL119/PX/961/JzMysvz0Q8ULN7L7MGtOPRXcMpV14AC+u3E6Xf87RDWMiTUSNR85FRUXY7fbq5yaTCafTidlspqioiICAgOpl/v7+FBUV0bZtWxISEujbty9z5sxh6tSpzJgxo8Zifhn8tZGSknJG68upqR3rTl23ZSDw5qBI3t5q5a0txxg1axm3JoQzrmszjAZDnX5WQ6Ofy7qjtqw7Z6stawxnu91OcfHPd466XC7MZvMplxUXFxMQEEC3bt2w2WwADB06tFbBDJCQkIDVaq3VuikpKfTs2bNW68rvUzvWnfpsy/POhQmHcxn55jLe2HyMXKMfb43ui7/VUi+f52n6uaw7asu6U5dt6XA4TntAWuNp7aSkJFasWAFAamoq8fHx1cu6detGSkoKDoeDwsJC9uzZQ3x8PP/4xz9YuHAhAGvWrKFLly5/dj9EmryEliF8/7fh9G/bnM83pnPBzG84kKcuVyLeqMYj56FDh7J69WpGjx6N2+3mqaeeYtasWcTExDB48GDGjh3Lddddh9vt5u6778ZqtTJp0iQeeughPvzwQ2w2G1OnTj0b+yLi9cLtviy8bQgTPv+RN37YTe/nv+aLWy7g3JhwT5cmInWoxnA2Go1MmTLlpNfi4uKqH19zzTVcc801Jy2Pjo7m3XffraMSReSXfMwm/nt1H7q0CObeOSkM/M83vH7teYxJauPp0kSkjmgQEpFGyGAw8Lf+nZhz60B8zEauf38Vj36disvl9nRpIlIHFM4ijdjwTq1YPfEi2obZeXLRJq55Z4Um0BDxAgpnkUauc4tg1vx1OAPiIvhiUzpdps/h49Q03G4dRYs0VgpnES9w4kaxpy7uQX5pBWPeXcllbywlXRNoiDRKCmcRL2E5Pg1l6r0jGNSuBfO3HSRh+hxmrNhGpUsTaIg0JgpnES/Tvlkg39w+hDeu7YvVbOTuL9fRb8YCNhzK8XRpIlJLCmcRL2QwGLjp3Di23HcZY3q0Zm1GNuf8ez4Pzv2J0gqnp8sTkRoonEW8WPMAG+9dn8y8cYOIDvZj+tItnPPv+fx0QNNQijRkCmeRJuCijq3YeO+lTDi/A9uO5tN3xgKmLd6ka9EiDZTCWaSJ8LdaeOGKc5k/bjDh/lYenp/KoJnfsi+70NOlicivKJxFmphhHSPZcO+lXNkthlX7Munx7Dze+nGP+kWLNCAKZ5EmKMzfyuwb+jNrTF8Abp39HVe/vYJjRWUerkxEQOEs0mQZDAZu6BVH6r0jSG7bnC82pdP9mbks2H7Q06WJNHkKZ5EmrnWoncV3DGXaJUlklzi45LUl3PHp9xQ5KjxdmkiTpXAWEUxGI5MHdeGHvw+na8tgXl2zix7PzmX1vkxPlybSJCmcRaRa98hQfvj7xdw3sAv7coq44D/f8ODcn3A4Kz1dmkiTonAWkZNYzSaeHpHEsjuH0TrUn+lLt9D7+fka/lPkLFI4i8gpnd+2OesnjWD8ee3ZdDiP3s9/zbTFm3BWauASkfqmcBaR32W3Wnj5qj7M/b9B1QOXnP/iAr7cnIHLpX7RIvVF4SwiNRreqRUbJ1/K6OOTaIyatYzO//ySl7/bQUm5JtIQqWsKZxGplVA/K+9fn8yGe0dw87lx7M8tZsJnPxL7xGc88vV6jhSUerpEEa+hcBaRM5LQMoTXr+3Lvn+M4uEhXTFg4KlFm2kz9XNu/eg7Nh/O9XSJIo2ewllE/pAWgTamDE8k7ZFR/OfK3rQOtfPW2j10f2YuQ1/+ls83puvmMZE/yOzpAkSkcfPzMXN733jG92nPvG0HeGHFNpbsPsKS3UeICvJj/Hnt+b8+7YkIsHm6VJFGQ0fOIlInjEYDl3aJZtEdF7Jp8qXc2a8DeWXlPLpgA7FPfM71763ku32Zmv1KpBYUziJS5zq3CObFUeeS8eiVvHjFubQLD+DD9Wkkv7SQXs/N45XvdpJb4vB0mSINlsJZROpNoK8Pd57fgU2TL+Xb24dwRdcYNh3J467PfqDV//uU0e+sYP62g7o2LfIruuYsIvXOYDAwqH1LBrVvycH8Et5P2cs76/byyYb9fLJhPy0CbPylZxtu6NWWhJYhni5XxOMUziJyVrUK8uO+QQlMHtiFtRnZvLN2Dx+tT+PZZVt5dtlWekaFckGED9HxpTTXTWTSRCmcRcQjDAYD58aEc25MOM9e3ouvthzgnXV7WLD9ECkH3Dy//jOGdYjkLz3bcHlCNDaLfl1J06GfdhHxOKvZxFXdY7mqeyxHCkp55quVrMisYP62g8zfdpAAq4VR3WK4vmcbBsRFYDLqdhnxbgpnEWlQWgTaGNMxjGf+0pOtR/L44Kd9vP/TPt5eu4e31+6hVZAf1yW1YWyvtnRpEezpcqUWpk2bxpYtW8jKyqKsrIzo6GhCQkKYMWNGjdu++uqr9OnTh27dup1y+ZNPPsnNN99MZGTkH6rt888/Z8aMGURHR1e/Fh8fzyOPPPKH3q+uKJxFpMHq3CKYqRf3YMpFiazal8l7KXv5dMN+/rV0C/9auoWkqFDG9mzL6B6tdX26AXvggQeAqiDcu3cv9957b623HT9+/GmXP/zww3+qNoARI0acUU1ng8JZRBo8o9FA/7gI+sdFMOOKc/lq6wHeW7eXBdsPcveBddz7VQrDOkQytldbLusSja/F5OmSG6z7vkrh0w376/Q9r+oey/RLe57xdg888AB5eXnk5eXx8ssv88wzz3DkyBEyMzMZNGgQd999Nw888AAXX3wxx44dY/ny5ZSVlZGens64ceMYNWoUY8eO5fHHH2f+/PkcOHCA7OxsDh06xIMPPkhycjJLly5lxowZ2O12goKC6NChAxMnTqxVfSNGjKB169ZYLBbatm3L8uXLMZvNPPnkkyxfvpx58+ZhNpvp1asXkydP5sUXX2T9+vWUlJTw5JNPEhcXd8ZtcoLCWUQaFV+Liau7x3J191gyC0uZnZrGeyn7qq9PB/lauDoxliu7xXJBXAQ+ZgV1Q9anTx9uuukmDhw4QGJiIldffTUOh4P+/ftz9913n7RuUVERb7zxBmlpadx+++2MGjXqpOU+Pj68/vrrrF69mjfffJO+ffsydepUZs+eTXh4OJMmTTplDXPnzmXDhg3Vz6+88kpGjhxJSUkJd955J507d+bFF18kMjKSGTNmsGPHDr7++ms++ugjzGYzEydOZOnSpQC0bduWf/zjH3+6XRTOItJoNQ+wMTG5ExOTO7H1SB7vpezlvZR9vP79bl7/fjeBvhYu6hjJZV2iubhTK4JsPp4u2eOmX9rzDx3l1pc2bdoAEBwczKZNm/j++++x2+2Ul5f///buYyN4lgAAECFJREFUPTiqOkvg+LffnXR3XoSYEExIgkEgRh7BFyGKomDkXYoYAZEZCuK6KgwzEQcUJIKoqFXUyMKUrjPIbgVBHUedAnRHMw6rshkDBk2iEJA8hDxJupN0p3N/+0dIQwAF8jAdOZ+qW9330fceTnU4fX/33t/vnG2vvvpqAKKios67fujQoQBERkbi8XioqanBbrcTHh4OQEpKClVVVed87qeatdvjA3zXtQ8fPsy1116LyWTy7ffbb789Z/uukFsehRC/CMMiQ1h71yhKVszgo8zbeWTc1YQFmtmef5Q52z4l4snt3PEfe/jDp4Ucq3X1drjiFJ1OB7Rdj3Y4HGzYsIEFCxbQ3Nx8Tj/s7dteaF/t+vXrh8vloqamBqDD2fHF0p/xZED7/uPj4zlw4ABerxelFPv27fMVZX03PUkgZ85CiF8Ug17PLYMjuWVwJC9OS+GrijrePXiMdwuO8dG3P/DRtz/wyNv7GDEglLuGDeSuYdGMuTIcvf6n/+MXPevGG2/kN7/5Dfn5+ZjNZmJjYzlx4kSX9qnX61m5ciULFy7E4XCgaRqxsbHnbHd2s7bdbmfTpk0/ut8hQ4Zw5513ct9996FpGqNHj2bChAkUFhZ2Kd4z6ZQfDBHjdrspKCggKSkJi8VyUZ/Jy8tj9Gj/aZrpqySP3Udy2X16KpeldS7++nUp7xaU8vF3P+A51ad3hN3KnUOjuWtYNLcnRhFk/eU0f1/u38vNmzfz4IMPYjabWbZsGampqUyfPr1T++rOXF6o7smZsxDisjEwxEbmTUPIvGkITncLHxZX8P7XZbz/TanvOWqTQU9afAR3DBnAtQNCuXZAqDym1YfZbDZmzZqF1WolOjqa9PT03g7pokhxFkJcluwWE9OviWH6NTFomuJfZTW8/3Up739d6mv+bneFw8o1UaEkR4WSPCCU5AEhXB0RjEXuBPd7c+bMYc6cOb0dxiWT4iyEuOzp9TpSruxHypX9eGritZSdbOSfJScoqKhjf3ktX1XU8mFxBR8WV/g+Y9TrGB4ZwojoMEZGhzIiOowRA8JwWE29+C8RvxRSnIUQ4izRwYHMGjGIWSNOL6tr8lBQUceB8loOVNT6XveX1/Knfae3GxzuYGR0GCOjw0iMCCIuzM6gMDsh8hiXuARSnIUQ4iKEBJhJjY8gNT7Ct8zbqlFUWc+XZTXkn5q+LKv1jVN99ufbC3XcqenK0ECig9um/jar3DEufKQ4CyFEJxkNeoZHhjA8MoQ5o+MBUEpxtNZFflkNh6oaOFLroqTGyZEaJ4UnTvJlWc1592Uy6IkKCiA6KJCo4ACigwMZEBRAhD2AKxxWIuynJodVrnVfBqQ4CyFEN9LpdAw6dYZ8NqUUxxuaOVLrpKTaybE6F+X1TZSdbKT8ZCNlJxvZd6wK79GffsI12Goiwm6lv91KcIAZh8WIw2IiyGryvdotJoIsJirLnRBRTT+bhXCbBZvZeMHOPLpbV0alAigqKqK+vp4xY8awZMkS1q9fj9ncucsEGzdu5L333iMi4nQLyE033URmZman9tdTLlicNU1j1apVFBUVYTabyc7O7vAQ9/bt2339i2ZmZjJ+/HhqampYtmwZzc3NREREsG7dOgIC5FEEIcTlTafTERkUQGRQADfE9j/vNpqmOOFsbivY9Y1UOt2ccDZxwtnMiYbmttdT06FqJ9rFdFXx8fe+t2aDnnCbhX42C/0CLYTZLIRYzYQGmgkJMBNiNRPie28iJMBMcICZYKup04W9K6NSAezevZvw8HDGjBnDSy+9dMnHP9v8+fO57777uryfnnTB4vzhhx/i8XjIyckhPz+fZ5991tdzSmVlJVu3bmXnzp243W4yMjIYO3Ysr7zyCpMnT2bmzJls2bKFnJwc5s+f39P/FiGE6PP0+tMFfDT9fnJbTVM0tnhpcLdQ39xCg9tLfbOHBnfbsobmFgq+K8ES0o9ql4cqVzOJod8QG3ICTVNoZ9f1Vmh0tU3l5zugDvQ6HXrfqw6dDk64ojjakITVaMBi1GMxGjq8Nxn06HVQWFxB7Q81bPpnEVprKx/9+RVOHi9HKcVt9z7AkOSR/E/O6xw+uB+ltXL9uPGMnziJ7Tt2YjSZ6HdlPNkrsvivnX9hw7psrBYLFeXlVFdVsm7dOpKSknjzzTfZtm0bwcHBmEwm0tPTzxkg48eMHz+e+Ph4EhISqK+v942YtXnzZjZt2kRubi42m43JkyfzwAMPdBhVa/PmzQQHB1/UcS7GBYtzXl4e48aNA2DEiBEUFBT41h04cICRI0diNpsxm83ExMRQWFhIXl4eixYtAiAtLY0XX3xRirMQQnQzvV6H3dLWhB0VdP5t8qzODr1a7Stxc6TK6ZtvL9KaUmdMZ8yfs77tvVfTfMW9qLKeNwsOXzBex7ESzM5qct76gqAjeZianFQPnYHe08jhl5/n2C2Lif3ofcpunIvXYueLL/bzzPF9hDkG47XaufPtYmLrGhm05i36f3WEFlsYtVeNJ6jhX0xY8gz1w24l6h+vUXN7JkajiaBP/pOPWv6PZV+3XdM36nWQX0TroRyyt2xFR9uPi5DrJ2KPG055eQUBM5dQGWjnh7w/YY0YSP9Js7nnuT9z8qt8HlvyBHffcj0ZGRnccMMNwOlRtbrbBYuz0+nEbj997cRgMOD1ejEajTidThwOh2+dzWbD6XR2WG6z2WhoaLioYM4s/BcjLy/vkrYX5yd57D6Sy+4juew+Z+ZSzxXE62/njAWd1qopmrwa19gUcxI0PK3q9KSdnveeanov3Pc9dZUeUsZGs7cyl+OVR7n6YA5KQaNFz7Shdur6PcjhLz7B7awnKG4o/RPDOFZpxRAYgGNQMCVGPbcMtFN1yIQlLg5DpI2mlgg8TT8QoW/AE3YFkY4AvJpC9b8StFbcbg8upWjVFNZGNy2DxtAwaDReTdGqAA9QVM4gcwC7jpwEThJR14gzwEjjV8cIOfQNyhDO1zXNHDhwgKioKPbs2UN1dTVer7dHvqsXLM52ux2X6/QILpqmYTQaz7vO5XLhcDh8y61WKy6Xi6CgH/lJdxbpW/vnJ3nsPpLL7iO57D7+lMu3tFoOH7awbOat/NlZSmPjaBYvXkxzczObNm3i3+bdxfPPP8/rb/83AOnp6bywailvt5QQGhrK/fdP5da3X+b9f5/KU1V5pKffRFpaGrm5uXzwwUl+97v5ZGTs5p1lUzCbzcyf/x7Tp1/foVl748ZqwsPDfdeclVJ4tbaWgNv+dxP7185GU4pVKw4wYWIaN4xN5R8fR/DBX//C9MGhJCcnU1ZWxqJFiygtLSUxMbFT+W3vW/vHXLA4jxo1ir///e+kp6eTn59PYmKib11ycjIvv/wybrcbj8fDoUOHSExMZNSoUXzyySfMnDmT3Nxcv/liCCGE8A+zZ89mxYoVzJkzB6fTSUZGBmazmeDgYF9f2GPHjmXAgAEkJSXx3HPPkZCQ8JP7DAsLY+HChWRkZBASEoLb7fadTJ7p9ddf54MPPvDNx8XF8fTTT6PTtV0mADAbDdgtJsICLUxLn0jhV/k8+eSTWCwWJk2axPDhw7s3IWe54KhU7XdrFxcXo5Ri7dq15ObmEhMTw2233cb27dvJyclBKcWiRYuYOHEiVVVVZGVl4XK5CA0NZcOGDQQGBv7oMWRUqt4jeew+ksvuI7nsPpdTLr1eL3/84x/JzMxEKcX999/PkiVLGDNmTLfs369GpdLr9Tz99NMdlp3562XWrFnMmjWrw/rw8HBeffXVzsYshBBCXDKj0UhTUxMzZszAZDKRnJxMSkpKb4fVKdIJiRBCiF+MpUuXsnTp0t4Oo8u6cJ+eEEIIIXqCFGchhBDCz0hxFkIIIfyMFGchhBDCz0hxFkIIIfyMFGchhBDCz0hxFkIIIfyMXzzn3N5JmcfjuaTPud3ungjnsiN57D6Sy+4juew+ksvu0125bK93P9ZJ5wW77/w5NDQ0UFxc3NthCCGEED+rxMTEDqM7tvOL4qxpGi6XC5PJhE6n6+1whBBCiB6llKKlpQWbzYZef+4VZr8ozkIIIYQ4TW4IE0IIIfyMFGchhBDCz0hxFkIIIfyMFGchhBDCz/jFc84XS9M0Vq1aRVFREWazmezsbGJjY3s7rD5n//79vPDCC2zdupWjR4/y+OOPo9PpuOqqq3jqqafOe+eg6KilpYUnnniCsrIyPB4PmZmZDB48WHLZCa2traxYsYKSkhJ0Oh2rV6/GYrFILjupurqamTNn8tprr2E0GiWPXTBjxgzsdjsAAwcO5N577+WZZ57BYDCQmprKww8/3HMHV33Irl27VFZWllJKqS+//FItXry4lyPqe7Zs2aImT56s7rnnHqWUUosWLVKfffaZUkqplStXqt27d/dmeH3Gjh07VHZ2tlJKqdraWnXzzTdLLjtpz5496vHHH1dKKfXZZ5+pxYsXSy47yePxqIceekjdcccd6rvvvpM8dkFzc7OaNm1ah2VTp05VR48eVZqmqV//+tfq4MGDPXb8PvUTKi8vj3HjxgEwYsQICgoKejmivicmJoaNGzf65g8ePMh1110HQFpaGnv37u2t0PqUSZMm8eijjwJtzysaDAbJZSdNmDCBNWvWAFBeXk5QUJDkspPWr1/P7NmziYiIAOTvuysKCwtpampiwYIFzJs3j3379uHxeIiJiUGn05Gamtqj+exTxdnpdPqaGAAMBgNer7cXI+p7Jk6ciNF4+mqGUsrX8YvNZqOhoaG3QutTbDYbdrsdp9PJI488wmOPPSa57AKj0UhWVhZr1qxhypQpkstOeOuttwgLC/OdwID8fXeF1WrlV7/6Fa+++iqrV69m+fLlBAQE+Nb3dD77VHG22+24XC7fvKZpHQqNuHRnXn9yuVwEBQX1YjR9S0VFBfPmzWPatGlMmTJFctlF69evZ9euXaxcubJD/8WSy4uzc+dO9u7dy9y5c/nmm2/IysqipqbGt17yeGni4uKYOnUqOp2OuLg4HA4HdXV1vvU9nc8+VZxHjRpFbm4uAPn5+SQmJvZyRH3fsGHD+PzzzwHIzc0lJSWllyPqG6qqqliwYAG//e1vufvuuwHJZWe98847bN68GYCAgAB0Oh1JSUmSy0u0bds23njjDbZu3crQoUNZv349aWlpksdO2rFjB88++ywAx48fp6mpicDAQL7//nuUUnz66ac9ms8+1X1n+93axcXFKKVYu3YtCQkJvR1Wn1NaWsrSpUvZvn07JSUlrFy5kpaWFuLj48nOzsZgMPR2iH4vOzubv/3tb8THx/uW/f73vyc7O1tyeYkaGxtZvnw5VVVVeL1eFi5cSEJCgnwvu2Du3LmsWrUKvV4veewkj8fD8uXLKS8vR6fTsWzZMvR6PWvXrqW1tZXU1FSWLFnSY8fvU8VZCCGEuBz0qWZtIYQQ4nIgxVkIIYTwM1KchRBCCD8jxVkIIYTwM1KchRBCCD8jxVkIIYTwM1KchRBCCD8jxVkIIYTwM/8PWD08/UNyLY0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#current tree depth\n",
    "max_depth = dectree.get_depth()\n",
    "#lets try all possible depth until 50 to see what gives us the best accuracy\n",
    "training_errors = []\n",
    "testing_errors = []\n",
    "depth = [i for i in range(1, 51)]\n",
    "for i in range(1, 51):\n",
    "    tree = DecisionTreeClassifier(max_depth=i).fit(X_train, y_train)\n",
    "    train_pred = tree.predict(X_train)\n",
    "    test_pred = tree.predict(X_test)\n",
    "    training_error = 1 - accuracy_score(y_train, train_pred)\n",
    "    testing_error = 1 - accuracy_score(y_test, test_pred)\n",
    "    training_errors.append(training_error)\n",
    "    testing_errors.append(testing_error)\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.title('Training vs Testing Error')\n",
    "plt.plot(depth, training_errors, 'b', label = 'Training Error')\n",
    "plt.plot(depth, testing_errors, 'g', label = 'Testing Error')\n",
    "plt.legend(loc = 'lower right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "8\n"
     ]
    }
   ],
   "source": [
    "#looks like testing error is at its lowest with a much shorter depth lets find it\n",
    "#depth that minimizes error\n",
    "for i in range(1, 51):\n",
    "    if testing_errors[i - 1] == min(testing_errors):\n",
    "        print(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 203,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8585068293329797"
      ]
     },
     "execution_count": 203,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#lets check accuracy with tree of depth 8\n",
    "dectree = DecisionTreeClassifier(max_depth=8).fit(X_train, y_train)\n",
    "ypred = dectree.predict(X_test)\n",
    "accuracy_score(y_test, ypred)\n",
    "#now the tree is slightly better than logistic regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8378199177827874"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#let us try support vector machines\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "svc = SVC().fit(X_train, y_train)\n",
    "ypred = svc.predict(X_test)\n",
    "accuracy_score(y_test, ypred)\n",
    "\n",
    "#svc is worse right now but lets try some other hyperparameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8517438005569553"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#it is a radial kernel by default, lets try linear\n",
    "svc = SVC(kernel=\"linear\").fit(X_train, y_train)\n",
    "ypred = svc.predict(X_test)\n",
    "accuracy_score(y_test, ypred)\n",
    "#performs better with linear, lets also try polynomial"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8415329531892322"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc = SVC(kernel=\"poly\", degree=2).fit(X_train, y_train)\n",
    "ypred = svc.predict(X_test)\n",
    "accuracy_score(y_test, ypred)\n",
    "#degree 2 slightly worse lets try 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8402068691155019"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc = SVC(kernel=\"poly\", degree=3).fit(X_train, y_train)\n",
    "ypred = svc.predict(X_test)\n",
    "accuracy_score(y_test, ypred)\n",
    "#this is even worse, lets work more with the linear kernel and give it a larger error budget on training data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8542633602970429"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc = SVC(kernel=\"linear\", C=10).fit(X_train, y_train)\n",
    "ypred = svc.predict(X_test)\n",
    "accuracy_score(y_test, ypred)\n",
    "#slightly better lets increase error budget further"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8541307518896698"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc = SVC(kernel=\"linear\", C=100).fit(X_train, y_train)\n",
    "ypred = svc.predict(X_test)\n",
    "accuracy_score(y_test, ypred)\n",
    "#seems that best svm has a linear kernel with error budget 10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8538655350749238"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#lets try some ensemble models, starting with random forest\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "rf = RandomForestClassifier().fit(X_train, y_train)\n",
    "ypred = rf.predict(X_test)\n",
    "accuracy_score(y_test, ypred)\n",
    "#since the decision tree on its own with no tuning was overfit, lets try using a shorter max depth on random forests to\n",
    "#see if it makes the performance better"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "21\n"
     ]
    }
   ],
   "source": [
    "#lets try all possible depth until 50 to see what gives us the best accuracy\n",
    "training_errors = []\n",
    "testing_errors = []\n",
    "depth = [i for i in range(1, 51)]\n",
    "for i in range(1, 51):\n",
    "    trees = RandomForestClassifier(max_depth=i).fit(X_train, y_train)\n",
    "    train_pred = trees.predict(X_train)\n",
    "    test_pred = trees.predict(X_test)\n",
    "    training_error = 1 - accuracy_score(y_train, train_pred)\n",
    "    testing_error = 1 - accuracy_score(y_test, test_pred)\n",
    "    training_errors.append(training_error)\n",
    "    testing_errors.append(testing_error)\n",
    "for i in range(1, 51):\n",
    "    if testing_errors[i - 1] == min(testing_errors):\n",
    "        print(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8635459488131547"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#looks like best depth is 18 lets try this new random forest now\n",
    "rf = RandomForestClassifier(max_depth=21).fit(X_train, y_train)\n",
    "ypred = rf.predict(X_test)\n",
    "accuracy_score(y_test, ypred)\n",
    "#best model so far with 86% accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8618220395173054"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#lets try a boosting classifier as well\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "\n",
    "adaclf = AdaBoostClassifier().fit(X_train, y_train)\n",
    "ypred = adaclf.predict(X_test)\n",
    "accuracy_score(y_test, ypred)\n",
    "#not as good as random forest so far but pretty close, lets try increasing number of estimatorsd in ensemble"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8722981036997746"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#default estimators are 50, now there are 500\n",
    "adaclf = AdaBoostClassifier(n_estimators=500).fit(X_train, y_train)\n",
    "ypred = adaclf.predict(X_test)\n",
    "accuracy_score(y_test, ypred)\n",
    "#increasing the number of estimators has improved the model, and its better than random forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8732263625513857"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#lets try xgboost, most popular algorithm for structured data on Kaggle\n",
    "from xgboost import XGBClassifier\n",
    "\n",
    "xgb = XGBClassifier().fit(X_train, y_train)\n",
    "ypred = xgb.predict(X_test)\n",
    "accuracy_score(y_test, ypred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9045577118606605"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#lets try the algorithm with a linear booster instead, as the model had good results with linear models\n",
    "xgb = XGBClassifier().fit(X_train, y_train)\n",
    "ypred = xgb.predict(X_train)\n",
    "accuracy_score(y_train, ypred)\n",
    "#it is slightly overfitted, lets try changing some parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 200,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8762763559209654"
      ]
     },
     "execution_count": 200,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgb = XGBClassifier(max_depth=4, tree_method='hist', eta=0.25).fit(X_train, y_train)\n",
    "ypred = xgb.predict(X_test)\n",
    "accuracy_score(y_test, ypred)\n",
    "#best model so far"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 202,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fdb38084240>"
      ]
     },
     "execution_count": 202,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "visualizer = ROCAUC(xgb, classes=[\"<=50K\", \">50K\"])\n",
    "\n",
    "visualizer.fit(X_train, y_train)        \n",
    "visualizer.score(X_test, y_test)        \n",
    "visualizer.show() \n",
    "#aisde from an increase in accuracy, compared to Logistic regression, XGBoost is better at class separation, \n",
    "#as tree based methods are better for imbalanced datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [],
   "source": [
    "#now that we have an idea of what models work well, lets do some feature elimination to see if the model improves\n",
    "#will try this for XGboost and Logistic Regression\n",
    "#since logistic regression is just a linear relationship with a sigmoid, we can analyze pearson correlation, to\n",
    "#eliminate irrelevant features\n",
    "categorical_salary = features.copy()\n",
    "categorical_salary[\"income\"] = target\n",
    "corr_df = pd.DataFrame(categorical_salary.corr()['income'][:])\n",
    "corr_df = corr_df.sort_values(by='income', ascending=False)\n",
    "#lets use top 10 positive and negative correlated features\n",
    "top10pos = corr_df.iloc[:31, :]\n",
    "top10neg = corr_df.iloc[95:, :]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [],
   "source": [
    "#lets use all features with an absolute value correlation of 0.01 or higher\n",
    "poscor = corr_df[corr_df[\"income\"] >= 0.01]\n",
    "negcor = corr_df[corr_df[\"income\"] <= -0.01]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>income</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>marital_Married-civ-spouse</th>\n",
       "      <td>0.445418</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>relationship_Husband</th>\n",
       "      <td>0.401236</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_num</th>\n",
       "      <td>0.335286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>age</th>\n",
       "      <td>0.241998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hr_per_week</th>\n",
       "      <td>0.229480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>capital_gain</th>\n",
       "      <td>0.221196</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sex_Male</th>\n",
       "      <td>0.216699</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>occupation_Exec-managerial</th>\n",
       "      <td>0.213442</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>occupation_Prof-specialty</th>\n",
       "      <td>0.181458</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_Bachelors</th>\n",
       "      <td>0.178847</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_Masters</th>\n",
       "      <td>0.174126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_Prof-school</th>\n",
       "      <td>0.156472</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>capital_loss</th>\n",
       "      <td>0.150053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>employer_Self-emp-inc</th>\n",
       "      <td>0.137646</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_Doctorate</th>\n",
       "      <td>0.129162</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>relationship_Wife</th>\n",
       "      <td>0.125126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race_White</th>\n",
       "      <td>0.084735</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>employer_Federal-gov</th>\n",
       "      <td>0.057394</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_United-States</th>\n",
       "      <td>0.040204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>employer_Local-gov</th>\n",
       "      <td>0.028673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>occupation_Protective-serv</th>\n",
       "      <td>0.026360</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>employer_Self-emp-not-inc</th>\n",
       "      <td>0.025575</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>occupation_Tech-support</th>\n",
       "      <td>0.022829</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_India</th>\n",
       "      <td>0.020152</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>occupation_Sales</th>\n",
       "      <td>0.018450</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Taiwan</th>\n",
       "      <td>0.017571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Iran</th>\n",
       "      <td>0.015515</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Japan</th>\n",
       "      <td>0.014427</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Germany</th>\n",
       "      <td>0.014317</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>marital_Married-AF-spouse</th>\n",
       "      <td>0.013874</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_France</th>\n",
       "      <td>0.013535</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Philippines</th>\n",
       "      <td>0.012863</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_England</th>\n",
       "      <td>0.012356</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Canada</th>\n",
       "      <td>0.012078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Italy</th>\n",
       "      <td>0.011435</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race_Asian-Pac-Islander</th>\n",
       "      <td>0.011394</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                              income\n",
       "income                      1.000000\n",
       "marital_Married-civ-spouse  0.445418\n",
       "relationship_Husband        0.401236\n",
       "education_num               0.335286\n",
       "age                         0.241998\n",
       "hr_per_week                 0.229480\n",
       "capital_gain                0.221196\n",
       "sex_Male                    0.216699\n",
       "occupation_Exec-managerial  0.213442\n",
       "occupation_Prof-specialty   0.181458\n",
       "education_Bachelors         0.178847\n",
       "education_Masters           0.174126\n",
       "education_Prof-school       0.156472\n",
       "capital_loss                0.150053\n",
       "employer_Self-emp-inc       0.137646\n",
       "education_Doctorate         0.129162\n",
       "relationship_Wife           0.125126\n",
       "race_White                  0.084735\n",
       "employer_Federal-gov        0.057394\n",
       "country_United-States       0.040204\n",
       "employer_Local-gov          0.028673\n",
       "occupation_Protective-serv  0.026360\n",
       "employer_Self-emp-not-inc   0.025575\n",
       "occupation_Tech-support     0.022829\n",
       "country_India               0.020152\n",
       "occupation_Sales            0.018450\n",
       "country_Taiwan              0.017571\n",
       "country_Iran                0.015515\n",
       "country_Japan               0.014427\n",
       "country_Germany             0.014317\n",
       "marital_Married-AF-spouse   0.013874\n",
       "country_France              0.013535\n",
       "country_Philippines         0.012863\n",
       "country_England             0.012356\n",
       "country_Canada              0.012078\n",
       "country_Italy               0.011435\n",
       "race_Asian-Pac-Islander     0.011394"
      ]
     },
     "execution_count": 138,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "poscor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>country_Portugal</th>\n",
       "      <td>-0.010199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>employer_Without-pay</th>\n",
       "      <td>-0.012406</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Outlying-US(Guam-USVI-etc)</th>\n",
       "      <td>-0.012406</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Haiti</th>\n",
       "      <td>-0.013272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Peru</th>\n",
       "      <td>-0.013300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Nicaragua</th>\n",
       "      <td>-0.014414</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Jamaica</th>\n",
       "      <td>-0.014780</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Vietnam</th>\n",
       "      <td>-0.018215</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Puerto-Rico</th>\n",
       "      <td>-0.019337</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_El-Salvador</th>\n",
       "      <td>-0.021198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Columbia</th>\n",
       "      <td>-0.021267</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Guatemala</th>\n",
       "      <td>-0.021300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>occupation_Craft-repair</th>\n",
       "      <td>-0.021445</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_Preschool</th>\n",
       "      <td>-0.022253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Dominican-Republic</th>\n",
       "      <td>-0.023906</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>occupation_Transport-moving</th>\n",
       "      <td>-0.024944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race_Amer-Indian-Eskimo</th>\n",
       "      <td>-0.029426</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race_Other</th>\n",
       "      <td>-0.032104</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_1st-4th</th>\n",
       "      <td>-0.034317</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>occupation_Priv-house-serv</th>\n",
       "      <td>-0.038618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>marital_Married-spouse-absent</th>\n",
       "      <td>-0.042562</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_12th</th>\n",
       "      <td>-0.044753</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_5th-6th</th>\n",
       "      <td>-0.047063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_9th</th>\n",
       "      <td>-0.055520</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>occupation_Farming-fishing</th>\n",
       "      <td>-0.056483</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_7th-8th</th>\n",
       "      <td>-0.059032</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>marital_Widowed</th>\n",
       "      <td>-0.059097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_Some-college</th>\n",
       "      <td>-0.060261</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_Mexico</th>\n",
       "      <td>-0.064735</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_10th</th>\n",
       "      <td>-0.068421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>marital_Separated</th>\n",
       "      <td>-0.074056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>occupation_Machine-op-inspct</th>\n",
       "      <td>-0.075912</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>relationship_Other-relative</th>\n",
       "      <td>-0.084457</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_11th</th>\n",
       "      <td>-0.084522</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race_Black</th>\n",
       "      <td>-0.088332</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>occupation_Handlers-cleaners</th>\n",
       "      <td>-0.093836</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>occupation_Adm-clerical</th>\n",
       "      <td>-0.099849</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>employer_Private</th>\n",
       "      <td>-0.117218</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>marital_Divorced</th>\n",
       "      <td>-0.132030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education_HS-grad</th>\n",
       "      <td>-0.136137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>relationship_Unmarried</th>\n",
       "      <td>-0.145799</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>occupation_Other-service</th>\n",
       "      <td>-0.165934</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>relationship_Not-in-family</th>\n",
       "      <td>-0.193258</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sex_Female</th>\n",
       "      <td>-0.216699</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>relationship_Own-child</th>\n",
       "      <td>-0.226186</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>marital_Never-married</th>\n",
       "      <td>-0.320053</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      income\n",
       "country_Portugal                   -0.010199\n",
       "employer_Without-pay               -0.012406\n",
       "country_Outlying-US(Guam-USVI-etc) -0.012406\n",
       "country_Haiti                      -0.013272\n",
       "country_Peru                       -0.013300\n",
       "country_Nicaragua                  -0.014414\n",
       "country_Jamaica                    -0.014780\n",
       "country_Vietnam                    -0.018215\n",
       "country_Puerto-Rico                -0.019337\n",
       "country_El-Salvador                -0.021198\n",
       "country_Columbia                   -0.021267\n",
       "country_Guatemala                  -0.021300\n",
       "occupation_Craft-repair            -0.021445\n",
       "education_Preschool                -0.022253\n",
       "country_Dominican-Republic         -0.023906\n",
       "occupation_Transport-moving        -0.024944\n",
       "race_Amer-Indian-Eskimo            -0.029426\n",
       "race_Other                         -0.032104\n",
       "education_1st-4th                  -0.034317\n",
       "occupation_Priv-house-serv         -0.038618\n",
       "marital_Married-spouse-absent      -0.042562\n",
       "education_12th                     -0.044753\n",
       "education_5th-6th                  -0.047063\n",
       "education_9th                      -0.055520\n",
       "occupation_Farming-fishing         -0.056483\n",
       "education_7th-8th                  -0.059032\n",
       "marital_Widowed                    -0.059097\n",
       "education_Some-college             -0.060261\n",
       "country_Mexico                     -0.064735\n",
       "education_10th                     -0.068421\n",
       "marital_Separated                  -0.074056\n",
       "occupation_Machine-op-inspct       -0.075912\n",
       "relationship_Other-relative        -0.084457\n",
       "education_11th                     -0.084522\n",
       "race_Black                         -0.088332\n",
       "occupation_Handlers-cleaners       -0.093836\n",
       "occupation_Adm-clerical            -0.099849\n",
       "employer_Private                   -0.117218\n",
       "marital_Divorced                   -0.132030\n",
       "education_HS-grad                  -0.136137\n",
       "relationship_Unmarried             -0.145799\n",
       "occupation_Other-service           -0.165934\n",
       "relationship_Not-in-family         -0.193258\n",
       "sex_Female                         -0.216699\n",
       "relationship_Own-child             -0.226186\n",
       "marital_Never-married              -0.320053"
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "negcor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [],
   "source": [
    "#we notice that country of employment is a very prevalent feature which makes sense as countries like the US have\n",
    "#a higher rate of pay compared to others, it also seems that occupation at a manegerial level, or having a \n",
    "#speciality is linked to higher pay.\n",
    "#its a given and visible here that amount of education and degree level is positively correlated too, and there\n",
    "#are apparent race issues clearly visible with race white having the highest postive correlation to income\n",
    "#also hrs_per_week is postively correalted, reaffirming that a lot of higher paying jobs have longer hours\n",
    "#it also seems thhat marriage is linked to high pay, probably because a lot of people marry afterv they have \n",
    "#settled down well into their careers, age is similar too"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [],
   "source": [
    "new_feature_list = []\n",
    "for i in list(poscor.index)[1:]:\n",
    "    new_feature_list.append(i)\n",
    "for i in list(negcor.index):\n",
    "    new_feature_list.append(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>marital_Married-civ-spouse</th>\n",
       "      <th>relationship_Husband</th>\n",
       "      <th>education_num</th>\n",
       "      <th>age</th>\n",
       "      <th>hr_per_week</th>\n",
       "      <th>capital_gain</th>\n",
       "      <th>sex_Male</th>\n",
       "      <th>occupation_Exec-managerial</th>\n",
       "      <th>occupation_Prof-specialty</th>\n",
       "      <th>education_Bachelors</th>\n",
       "      <th>...</th>\n",
       "      <th>occupation_Adm-clerical</th>\n",
       "      <th>employer_Private</th>\n",
       "      <th>marital_Divorced</th>\n",
       "      <th>education_HS-grad</th>\n",
       "      <th>relationship_Unmarried</th>\n",
       "      <th>occupation_Other-service</th>\n",
       "      <th>relationship_Not-in-family</th>\n",
       "      <th>sex_Female</th>\n",
       "      <th>relationship_Own-child</th>\n",
       "      <th>marital_Never-married</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>13</td>\n",
       "      <td>39</td>\n",
       "      <td>40</td>\n",
       "      <td>2174</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>50</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</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",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>38</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</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>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>53</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</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",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>13</td>\n",
       "      <td>28</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32556</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>27</td>\n",
       "      <td>38</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32557</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>40</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</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>32558</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>58</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32559</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>22</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32560</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>52</td>\n",
       "      <td>40</td>\n",
       "      <td>15024</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</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",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>30162 rows × 82 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       marital_Married-civ-spouse  relationship_Husband  education_num  age  \\\n",
       "0                               0                     0             13   39   \n",
       "1                               1                     1             13   50   \n",
       "2                               0                     0              9   38   \n",
       "3                               1                     1              7   53   \n",
       "4                               1                     0             13   28   \n",
       "...                           ...                   ...            ...  ...   \n",
       "32556                           1                     0             12   27   \n",
       "32557                           1                     1              9   40   \n",
       "32558                           0                     0              9   58   \n",
       "32559                           0                     0              9   22   \n",
       "32560                           1                     0              9   52   \n",
       "\n",
       "       hr_per_week  capital_gain  sex_Male  occupation_Exec-managerial  \\\n",
       "0               40          2174         1                           0   \n",
       "1               13             0         1                           1   \n",
       "2               40             0         1                           0   \n",
       "3               40             0         1                           0   \n",
       "4               40             0         0                           0   \n",
       "...            ...           ...       ...                         ...   \n",
       "32556           38             0         0                           0   \n",
       "32557           40             0         1                           0   \n",
       "32558           40             0         0                           0   \n",
       "32559           20             0         1                           0   \n",
       "32560           40         15024         0                           1   \n",
       "\n",
       "       occupation_Prof-specialty  education_Bachelors  ...  \\\n",
       "0                              0                    1  ...   \n",
       "1                              0                    1  ...   \n",
       "2                              0                    0  ...   \n",
       "3                              0                    0  ...   \n",
       "4                              1                    1  ...   \n",
       "...                          ...                  ...  ...   \n",
       "32556                          0                    0  ...   \n",
       "32557                          0                    0  ...   \n",
       "32558                          0                    0  ...   \n",
       "32559                          0                    0  ...   \n",
       "32560                          0                    0  ...   \n",
       "\n",
       "       occupation_Adm-clerical  employer_Private  marital_Divorced  \\\n",
       "0                            1                 0                 0   \n",
       "1                            0                 0                 0   \n",
       "2                            0                 1                 1   \n",
       "3                            0                 1                 0   \n",
       "4                            0                 1                 0   \n",
       "...                        ...               ...               ...   \n",
       "32556                        0                 1                 0   \n",
       "32557                        0                 1                 0   \n",
       "32558                        1                 1                 0   \n",
       "32559                        1                 1                 0   \n",
       "32560                        0                 0                 0   \n",
       "\n",
       "       education_HS-grad  relationship_Unmarried  occupation_Other-service  \\\n",
       "0                      0                       0                         0   \n",
       "1                      0                       0                         0   \n",
       "2                      1                       0                         0   \n",
       "3                      0                       0                         0   \n",
       "4                      0                       0                         0   \n",
       "...                  ...                     ...                       ...   \n",
       "32556                  0                       0                         0   \n",
       "32557                  1                       0                         0   \n",
       "32558                  1                       1                         0   \n",
       "32559                  1                       0                         0   \n",
       "32560                  1                       0                         0   \n",
       "\n",
       "       relationship_Not-in-family  sex_Female  relationship_Own-child  \\\n",
       "0                               1           0                       0   \n",
       "1                               0           0                       0   \n",
       "2                               1           0                       0   \n",
       "3                               0           0                       0   \n",
       "4                               0           1                       0   \n",
       "...                           ...         ...                     ...   \n",
       "32556                           0           1                       0   \n",
       "32557                           0           0                       0   \n",
       "32558                           0           1                       0   \n",
       "32559                           0           0                       1   \n",
       "32560                           0           1                       0   \n",
       "\n",
       "       marital_Never-married  \n",
       "0                          1  \n",
       "1                          0  \n",
       "2                          0  \n",
       "3                          0  \n",
       "4                          0  \n",
       "...                      ...  \n",
       "32556                      0  \n",
       "32557                      0  \n",
       "32558                      0  \n",
       "32559                      1  \n",
       "32560                      0  \n",
       "\n",
       "[30162 rows x 82 columns]"
      ]
     },
     "execution_count": 142,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_df = pd.DataFrame()\n",
    "for c in new_feature_list:\n",
    "    new_df[c] = features[c]\n",
    "new_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [],
   "source": [
    "#lets resplit this data, and preprocess it\n",
    "X_trainc, X_testc, y_trainc, y_testc = train_test_split(new_df, target, test_size=0.25, random_state=42)\n",
    "trainscaler = MinMaxScaler().fit(X_trainc)\n",
    "X_trainc = trainscaler.transform(X_trainc)\n",
    "testscaler = MinMaxScaler().fit(X_testc)\n",
    "X_testc = trainscaler.transform(X_testc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 207,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.855191619148654"
      ]
     },
     "execution_count": 207,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#logistic regression\n",
    "logreg = LogisticRegression(penalty=\"l1\", solver=\"liblinear\").fit(X_trainc, y_trainc)\n",
    "ypred = logreg.predict(X_testc)\n",
    "accuracy_score(y_testc, ypred)\n",
    "#looks like logistic regression isnt better with these selected features, lets try the polynomial regression from before\n",
    "#instead"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 208,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fdb38876a20>"
      ]
     },
     "execution_count": 208,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "visualizer = ROCAUC(logreg, classes=[\"<=50K\", \">50K\"])\n",
    "\n",
    "visualizer.fit(X_trainc, y_trainc)        \n",
    "visualizer.score(X_testc, y_testc)        \n",
    "visualizer.show() \n",
    "#AUC hasnt changed from initial logistic regression AUC, might as well use all the features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8541307518896698"
      ]
     },
     "execution_count": 206,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "poly = PolynomialFeatures(2)\n",
    "degree2 = poly.fit_transform(X_trainc)\n",
    "\n",
    "degree2_reg = Ridge().fit(degree2, y_trainc)\n",
    "ypred = degree2_reg.predict(PolynomialFeatures(2).fit_transform(X_testc))\n",
    "ypred_class =  [i.round() for i in ypred]\n",
    "accuracy_score(y_testc, ypred_class)\n",
    "#in this case logistic regression is better, maybe an algorithmic approach to eature selection could be better"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 209,
   "metadata": {},
   "outputs": [],
   "source": [
    "#tires all possible combinations of features on our best XGboost model\n",
    "from sklearn.feature_selection import RFE\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "#using 50 top features\n",
    "def RFE_feature_selection(n):\n",
    "    select = RFE(DecisionTreeRegressor(random_state = 3000), n_features_to_select = n)\n",
    "    \n",
    "    #fit the RFE selector to the training data\n",
    "    select.fit(X_train, y_train)\n",
    "\n",
    "    #transform training and testing sets so only the selected features are retained\n",
    "    X_train_selected = select.transform(X_train)\n",
    "    X_test_selected = select.transform(X_test)\n",
    "\n",
    "    model = xgclf = XGBClassifier(max_depth=4, tree_method='hist', eta=0.25).fit(X=X_train_selected, y=y_train)\n",
    "    \n",
    "    accuracy = model.score(X_test_selected, y_test)\n",
    "    \n",
    "    #print(\"Selected features after RFE:\")\n",
    "    #for i in range(0, len(select.get_support())):\n",
    "     #   if select.get_support()[i] == True:\n",
    "      #      print(\"\\t\" + str(features.columns[i]))\n",
    "    \n",
    "    #print(\"Logistic Regression performance with selected features:\")\n",
    "    #print(accuracy)\n",
    "    return accuracy, xgclf\n",
    "\n",
    "#lets try up to 105 features and see what selection gives us the best performance\n",
    "accuracies = []\n",
    "models = []\n",
    "for i in range(1, 105):\n",
    "    accuracy, model = RFE_feature_selection(i)\n",
    "    accuracies.append(accuracy)\n",
    "    models.append(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 210,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8765415727357114"
      ]
     },
     "execution_count": 210,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#lets find the features in the amx accuracy xgboost\n",
    "np.array(accuracies).max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 211,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "93\n",
      "94\n",
      "95\n"
     ]
    }
   ],
   "source": [
    "for i in range(1, 105):\n",
    "    if accuracies[i - 1] == max(accuracies):\n",
    "        print(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Selected features after RFE:\n",
      "\tage\n",
      "\tfnlwgt\n",
      "\teducation_num\n",
      "\tcapital_gain\n",
      "\tcapital_loss\n",
      "\thr_per_week\n",
      "\temployer_Federal-gov\n",
      "\temployer_Local-gov\n",
      "\temployer_Private\n",
      "\temployer_Self-emp-inc\n",
      "\temployer_Self-emp-not-inc\n",
      "\temployer_State-gov\n",
      "\teducation_10th\n",
      "\teducation_11th\n",
      "\teducation_12th\n",
      "\teducation_5th-6th\n",
      "\teducation_7th-8th\n",
      "\teducation_9th\n",
      "\teducation_Assoc-acdm\n",
      "\teducation_Assoc-voc\n",
      "\teducation_Bachelors\n",
      "\teducation_Doctorate\n",
      "\teducation_HS-grad\n",
      "\teducation_Masters\n",
      "\teducation_Preschool\n",
      "\teducation_Prof-school\n",
      "\teducation_Some-college\n",
      "\tmarital_Divorced\n",
      "\tmarital_Married-AF-spouse\n",
      "\tmarital_Married-civ-spouse\n",
      "\tmarital_Married-spouse-absent\n",
      "\tmarital_Never-married\n",
      "\tmarital_Separated\n",
      "\tmarital_Widowed\n",
      "\toccupation_Adm-clerical\n",
      "\toccupation_Craft-repair\n",
      "\toccupation_Exec-managerial\n",
      "\toccupation_Farming-fishing\n",
      "\toccupation_Handlers-cleaners\n",
      "\toccupation_Machine-op-inspct\n",
      "\toccupation_Other-service\n",
      "\toccupation_Priv-house-serv\n",
      "\toccupation_Prof-specialty\n",
      "\toccupation_Protective-serv\n",
      "\toccupation_Sales\n",
      "\toccupation_Tech-support\n",
      "\toccupation_Transport-moving\n",
      "\trelationship_Husband\n",
      "\trelationship_Not-in-family\n",
      "\trelationship_Other-relative\n",
      "\trelationship_Own-child\n",
      "\trelationship_Unmarried\n",
      "\trelationship_Wife\n",
      "\trace_Amer-Indian-Eskimo\n",
      "\trace_Asian-Pac-Islander\n",
      "\trace_Black\n",
      "\trace_Other\n",
      "\trace_White\n",
      "\tsex_Female\n",
      "\tsex_Male\n",
      "\tcountry_Cambodia\n",
      "\tcountry_Canada\n",
      "\tcountry_China\n",
      "\tcountry_Cuba\n",
      "\tcountry_Dominican-Republic\n",
      "\tcountry_Ecuador\n",
      "\tcountry_England\n",
      "\tcountry_France\n",
      "\tcountry_Germany\n",
      "\tcountry_Greece\n",
      "\tcountry_Guatemala\n",
      "\tcountry_Haiti\n",
      "\tcountry_Holand-Netherlands\n",
      "\tcountry_Hong\n",
      "\tcountry_Hungary\n",
      "\tcountry_India\n",
      "\tcountry_Iran\n",
      "\tcountry_Ireland\n",
      "\tcountry_Italy\n",
      "\tcountry_Jamaica\n",
      "\tcountry_Japan\n",
      "\tcountry_Laos\n",
      "\tcountry_Mexico\n",
      "\tcountry_Philippines\n",
      "\tcountry_Poland\n",
      "\tcountry_Portugal\n",
      "\tcountry_Puerto-Rico\n",
      "\tcountry_South\n",
      "\tcountry_Taiwan\n",
      "\tcountry_Thailand\n",
      "\tcountry_United-States\n",
      "\tcountry_Vietnam\n",
      "\tcountry_Yugoslavia\n"
     ]
    }
   ],
   "source": [
    "#looks like highest accuracy is with 93 features, lets find these features\n",
    "def RFE_feature_selection():\n",
    "    select = RFE(DecisionTreeRegressor(random_state = 3000), n_features_to_select = 93)\n",
    "    \n",
    "    #fit the RFE selector to the training data\n",
    "    select.fit(X_train, y_train)\n",
    "\n",
    "    #transform training and testing sets so only the selected features are retained\n",
    "    X_train_selected = select.transform(X_train)\n",
    "    X_test_selected = select.transform(X_test)\n",
    "\n",
    "    model = adaclf = XGBClassifier(max_depth=4, tree_method='hist', eta=0.25).fit(X=X_train_selected, y=y_train)\n",
    "    \n",
    "    accuracy = model.score(X_test_selected, y_test)\n",
    "    \n",
    "    print(\"Selected features after RFE:\")\n",
    "    for i in range(0, len(select.get_support())):\n",
    "        if select.get_support()[i] == True:\n",
    "            print(\"\\t\" + str(features.columns[i]))\n",
    "    \n",
    "    return accuracy, X_train_selected, X_test_selected\n",
    "a, xtr, xtst = RFE_feature_selection()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#based on this algorithmic eliminaton, the most important feature seems to be age, fnlwgt (unsure what this is), \n",
    "#capitak gain/loss (maybe indicating profitbaility of the eomployer in relationship with higher pay, makes sense as \n",
    "#a lot of companies withh high stock market values like Apple tend to pay more), hours per week, type of employer\n",
    "#(which makes sense as the private sector will usually pay more than the government), followed by education level and \n",
    "#years (indicating income relationship with skillset, and specialization), followed by marital status (maybe data \n",
    "#consists of household income, even if it doesn't, it could indicate the tendency of marrying after making some progress\n",
    "#with a person's career), followed by occupation (maybe if data was more specific with type of occupation, there\n",
    "#would be a stronger relationship), followed by race and sex (indicating racial and gender based discrimination in jobs)\n",
    "#and lastly has countries, which makes sense as first world countries have significantly higher salaries and GDP\n",
    "#per capita\n",
    "#Also, should consider that teh algorithm performed best with all these features, so all these features are very important"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}