First Portfolio Project.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Salary Predictions Based on Job Descriptions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Part 1 - DEFINE"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  Objective \n",
    "\n",
    "\n",
    "Accurate salary predictions help any company to recruit and retain the best employers or talent. In order to do this, there has to be a guideline to providing competitive compensation, job satisfaction, and conducive work environment for future employers. By doing so, the company will effectively control its financial system. \n",
    "\n",
    "\n",
    "Machine learning techniques are very useful in this regard to create a model that will predict a salary using features, such as years of experience, college degree, job type, industry, and miles from a metropolis. To accomplish the objective, a training dataset, which contains the abovementioned features and the respective salaries. Part of this training dataset, which is twenty percent, was split into a test dataset with corresponding salaries to determine the accuracy and error of the model. \n",
    "\n",
    "\n",
    "A testing dataset with no salary information was used as a substitute for real-world data. \n",
    "\n",
    "1. Years Experience: How many years of experience\n",
    "2. Job Type: The position held (CEO, CFO, CTO, Vice President, Manager, Janitor, and senior or junior position)\n",
    "3. College Degree: Doctoral, Masters, Bachelors, High School, or None\n",
    "4. College Major: Biology, Business, Chemistry, Computer Science, Engineering, Literature, Math, Physics, or None\n",
    "5. Industry: Auto, Education, Finance, Health, Oil, Service, or Web\n",
    "6. Miles From Metropolis: How many miles away from a major city\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\user\\Anaconda3\\lib\\site-packages\\sklearn\\ensemble\\weight_boosting.py:29: DeprecationWarning: numpy.core.umath_tests is an internal NumPy module and should not be imported. It will be removed in a future NumPy release.\n",
      "  from numpy.core.umath_tests import inner1d\n"
     ]
    }
   ],
   "source": [
    "#import your libraries\n",
    "import pandas as pd\n",
    "import sklearn as sk\n",
    "import seaborn as sns \n",
    "sns.set_color_codes('deep')\n",
    "import numpy as np \n",
    "import matplotlib.pyplot as plt \n",
    "%matplotlib inline\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "from scipy.stats import norm\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.model_selection import cross_val_predict\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.linear_model import Ridge\n",
    "from sklearn.preprocessing import PolynomialFeatures\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.externals import joblib\n",
    "__author__ = \"Ermias Kassaye\"\n",
    "__email__ = \"jeremyabebe@gmail.com\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 2 - DISCOVER"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ---- 2 Load the data ----"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#load the data into a Pandas dataframe\n",
    "Train_data = pd.read_csv(\"data/train_features.csv\", index_col='jobId')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The following output shows the first ten rows of the train data\n"
     ]
    },
    {
     "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>companyId</th>\n",
       "      <th>jobType</th>\n",
       "      <th>degree</th>\n",
       "      <th>major</th>\n",
       "      <th>industry</th>\n",
       "      <th>yearsExperience</th>\n",
       "      <th>milesFromMetropolis</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>jobId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>JOB1362684407687</th>\n",
       "      <td>COMP37</td>\n",
       "      <td>CFO</td>\n",
       "      <td>MASTERS</td>\n",
       "      <td>MATH</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>10</td>\n",
       "      <td>83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407688</th>\n",
       "      <td>COMP19</td>\n",
       "      <td>CEO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>WEB</td>\n",
       "      <td>3</td>\n",
       "      <td>73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407689</th>\n",
       "      <td>COMP52</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>DOCTORAL</td>\n",
       "      <td>PHYSICS</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>10</td>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407690</th>\n",
       "      <td>COMP38</td>\n",
       "      <td>MANAGER</td>\n",
       "      <td>DOCTORAL</td>\n",
       "      <td>CHEMISTRY</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>8</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407691</th>\n",
       "      <td>COMP7</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>BACHELORS</td>\n",
       "      <td>PHYSICS</td>\n",
       "      <td>FINANCE</td>\n",
       "      <td>8</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407692</th>\n",
       "      <td>COMP15</td>\n",
       "      <td>MANAGER</td>\n",
       "      <td>DOCTORAL</td>\n",
       "      <td>COMPSCI</td>\n",
       "      <td>FINANCE</td>\n",
       "      <td>2</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407693</th>\n",
       "      <td>COMP15</td>\n",
       "      <td>CFO</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>23</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407694</th>\n",
       "      <td>COMP24</td>\n",
       "      <td>JUNIOR</td>\n",
       "      <td>BACHELORS</td>\n",
       "      <td>CHEMISTRY</td>\n",
       "      <td>EDUCATION</td>\n",
       "      <td>9</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407695</th>\n",
       "      <td>COMP20</td>\n",
       "      <td>JANITOR</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>EDUCATION</td>\n",
       "      <td>1</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407696</th>\n",
       "      <td>COMP41</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>BACHELORS</td>\n",
       "      <td>CHEMISTRY</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>17</td>\n",
       "      <td>68</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 companyId         jobType       degree      major   industry  \\\n",
       "jobId                                                                           \n",
       "JOB1362684407687    COMP37             CFO      MASTERS       MATH     HEALTH   \n",
       "JOB1362684407688    COMP19             CEO  HIGH_SCHOOL       NONE        WEB   \n",
       "JOB1362684407689    COMP52  VICE_PRESIDENT     DOCTORAL    PHYSICS     HEALTH   \n",
       "JOB1362684407690    COMP38         MANAGER     DOCTORAL  CHEMISTRY       AUTO   \n",
       "JOB1362684407691     COMP7  VICE_PRESIDENT    BACHELORS    PHYSICS    FINANCE   \n",
       "JOB1362684407692    COMP15         MANAGER     DOCTORAL    COMPSCI    FINANCE   \n",
       "JOB1362684407693    COMP15             CFO         NONE       NONE     HEALTH   \n",
       "JOB1362684407694    COMP24          JUNIOR    BACHELORS  CHEMISTRY  EDUCATION   \n",
       "JOB1362684407695    COMP20         JANITOR  HIGH_SCHOOL       NONE  EDUCATION   \n",
       "JOB1362684407696    COMP41  VICE_PRESIDENT    BACHELORS  CHEMISTRY       AUTO   \n",
       "\n",
       "                  yearsExperience  milesFromMetropolis  \n",
       "jobId                                                   \n",
       "JOB1362684407687               10                   83  \n",
       "JOB1362684407688                3                   73  \n",
       "JOB1362684407689               10                   38  \n",
       "JOB1362684407690                8                   17  \n",
       "JOB1362684407691                8                   16  \n",
       "JOB1362684407692                2                   31  \n",
       "JOB1362684407693               23                   24  \n",
       "JOB1362684407694                9                   70  \n",
       "JOB1362684407695                1                   54  \n",
       "JOB1362684407696               17                   68  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Open data here. This command will print 10 rows of training data\n",
    "print('The following output shows the first ten rows of the train data')\n",
    "Train_data.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The shape of the data can be seen here. \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(1000000, 7)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# This part will print the shape of the data o\n",
    "print('The shape of the data can be seen here. ')\n",
    "Train_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The descriptive statistics is shown below\n"
     ]
    },
    {
     "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>yearsExperience</th>\n",
       "      <th>milesFromMetropolis</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1000000.000000</td>\n",
       "      <td>1000000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>11.992386</td>\n",
       "      <td>49.529260</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>7.212391</td>\n",
       "      <td>28.877733</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>6.000000</td>\n",
       "      <td>25.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>12.000000</td>\n",
       "      <td>50.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>18.000000</td>\n",
       "      <td>75.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>24.000000</td>\n",
       "      <td>99.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       yearsExperience  milesFromMetropolis\n",
       "count   1000000.000000       1000000.000000\n",
       "mean         11.992386            49.529260\n",
       "std           7.212391            28.877733\n",
       "min           0.000000             0.000000\n",
       "25%           6.000000            25.000000\n",
       "50%          12.000000            50.000000\n",
       "75%          18.000000            75.000000\n",
       "max          24.000000            99.000000"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# This part of the code will prnt descriptive statistics of the numerical data\n",
    "\n",
    "print('The descriptive statistics is shown below')\n",
    "\n",
    "Train_data.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ---- 3 Clean the data ----"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "We can see the info about the date here.\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Index: 1000000 entries, JOB1362684407687 to JOB1362685407686\n",
      "Data columns (total 7 columns):\n",
      "companyId              1000000 non-null object\n",
      "jobType                1000000 non-null object\n",
      "degree                 1000000 non-null object\n",
      "major                  1000000 non-null object\n",
      "industry               1000000 non-null object\n",
      "yearsExperience        1000000 non-null int64\n",
      "milesFromMetropolis    1000000 non-null int64\n",
      "dtypes: int64(2), object(5)\n",
      "memory usage: 61.0+ MB\n"
     ]
    }
   ],
   "source": [
    "#Let us print the descriptive statistics of the categorical data \n",
    "print('We can see the info about the date here.')\n",
    "\n",
    "Train_data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The null values of the given data are dispalyed here.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "companyId              False\n",
       "jobType                False\n",
       "degree                 False\n",
       "major                  False\n",
       "industry               False\n",
       "yearsExperience        False\n",
       "milesFromMetropolis    False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Let check whether there are columns with null values \n",
    "print('The null values of the given data are dispalyed here.')\n",
    "\n",
    "Train_data.isnull().any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Duplicated dataset in the training set:  8001\n"
     ]
    }
   ],
   "source": [
    "# Let us check that whether there are duplicated records in the dataset\n",
    "\n",
    "print(\"Duplicated dataset in the training set: \", Train_data.duplicated().sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The unique values of the training data are shown below.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "milesFromMetropolis    100\n",
       "companyId               63\n",
       "yearsExperience         25\n",
       "major                    9\n",
       "jobType                  8\n",
       "industry                 7\n",
       "degree                   5\n",
       "dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let us check for unique values in training data \n",
    "\n",
    "print('The unique values of the training data are shown below.')\n",
    "\n",
    "Train_data.nunique().sort_values(ascending = False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Observation:\n",
    "\n",
    "1.The training data has 100000 and 8 columns. \n",
    "2. The data has 2 int and 6 objects\n",
    "3. milesFromMetropolis is the target valiable.\n",
    "4. No null values recorded in the training data. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The copy of the data\n"
     ]
    }
   ],
   "source": [
    "#A copy of data before starting the training\n",
    "\n",
    "print('The copy of the data')\n",
    "Train_data_First = Train_data.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "99    0.010180\n",
       "62    0.010175\n",
       "63    0.010159\n",
       "97    0.010155\n",
       "41    0.010147\n",
       "92    0.010141\n",
       "39    0.010135\n",
       "85    0.010128\n",
       "95    0.010125\n",
       "64    0.010123\n",
       "0     0.010123\n",
       "81    0.010119\n",
       "58    0.010119\n",
       "17    0.010109\n",
       "3     0.010106\n",
       "48    0.010104\n",
       "28    0.010095\n",
       "71    0.010095\n",
       "56    0.010092\n",
       "1     0.010091\n",
       "96    0.010090\n",
       "93    0.010089\n",
       "53    0.010088\n",
       "51    0.010080\n",
       "30    0.010078\n",
       "72    0.010063\n",
       "19    0.010058\n",
       "88    0.010057\n",
       "37    0.010055\n",
       "68    0.010053\n",
       "        ...   \n",
       "43    0.009946\n",
       "31    0.009937\n",
       "76    0.009935\n",
       "59    0.009935\n",
       "79    0.009933\n",
       "70    0.009932\n",
       "86    0.009931\n",
       "91    0.009929\n",
       "47    0.009929\n",
       "35    0.009928\n",
       "82    0.009928\n",
       "15    0.009927\n",
       "74    0.009926\n",
       "42    0.009925\n",
       "77    0.009924\n",
       "80    0.009913\n",
       "55    0.009909\n",
       "46    0.009902\n",
       "89    0.009893\n",
       "32    0.009888\n",
       "8     0.009888\n",
       "65    0.009882\n",
       "52    0.009875\n",
       "73    0.009855\n",
       "24    0.009854\n",
       "26    0.009845\n",
       "6     0.009840\n",
       "18    0.009824\n",
       "87    0.009815\n",
       "20    0.009807\n",
       "Name: milesFromMetropolis, Length: 100, dtype: float64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Data imbalance is calculated here \n",
    "\n",
    "Train_data.milesFromMetropolis.value_counts(normalize= True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x22c28274e10>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Let us plot the ratio of data imbalance \n",
    "\n",
    "Train_data['milesFromMetropolis'].hist(figsize=(5,4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Show me the rows and columns in train salaries: (1000000, 1)\n"
     ]
    },
    {
     "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>salary</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>jobId</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>JOB1362684407687</th>\n",
       "      <td>130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407688</th>\n",
       "      <td>101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407689</th>\n",
       "      <td>137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407690</th>\n",
       "      <td>142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407691</th>\n",
       "      <td>163</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  salary\n",
       "jobId                   \n",
       "JOB1362684407687     130\n",
       "JOB1362684407688     101\n",
       "JOB1362684407689     137\n",
       "JOB1362684407690     142\n",
       "JOB1362684407691     163"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Train_Salaries = pd.read_csv('data/train_salaries.csv',index_col= 'jobId')\n",
    "\n",
    "print(\"Show me the rows and columns in train salaries:\", Train_Salaries.shape)\n",
    "# print 5 rows of the train salaries.\n",
    "Train_Salaries.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Show me the rows and columns in the test features: (1000000, 7)\n"
     ]
    },
    {
     "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>companyId</th>\n",
       "      <th>jobType</th>\n",
       "      <th>degree</th>\n",
       "      <th>major</th>\n",
       "      <th>industry</th>\n",
       "      <th>yearsExperience</th>\n",
       "      <th>milesFromMetropolis</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>jobId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>JOB1362685407687</th>\n",
       "      <td>COMP33</td>\n",
       "      <td>MANAGER</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>22</td>\n",
       "      <td>73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407688</th>\n",
       "      <td>COMP13</td>\n",
       "      <td>JUNIOR</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>20</td>\n",
       "      <td>47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407689</th>\n",
       "      <td>COMP10</td>\n",
       "      <td>CTO</td>\n",
       "      <td>MASTERS</td>\n",
       "      <td>BIOLOGY</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>17</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 companyId  jobType       degree    major industry  \\\n",
       "jobId                                                                \n",
       "JOB1362685407687    COMP33  MANAGER  HIGH_SCHOOL     NONE   HEALTH   \n",
       "JOB1362685407688    COMP13   JUNIOR         NONE     NONE     AUTO   \n",
       "JOB1362685407689    COMP10      CTO      MASTERS  BIOLOGY   HEALTH   \n",
       "\n",
       "                  yearsExperience  milesFromMetropolis  \n",
       "jobId                                                   \n",
       "JOB1362685407687               22                   73  \n",
       "JOB1362685407688               20                   47  \n",
       "JOB1362685407689               17                    9  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Test_Fetaures = pd.read_csv('data/test_features.csv', index_col= 'jobId')\n",
    "\n",
    "print (\"Show me the rows and columns in the test features:\", Test_Fetaures.shape)\n",
    "\n",
    "Test_Fetaures.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Give me number of rows with duplicate data in the train dataset: 15917\n"
     ]
    }
   ],
   "source": [
    "#The features in the data are given below\n",
    "\n",
    "Features = ['companyId','jobType', 'degree', 'major','industry','yearsExperience','milesFromMetropolis']\n",
    "\n",
    "# Check the duplicated values here\n",
    "\n",
    "Train_data_Dupls = Train_data[Train_data.duplicated(Features, keep = False)]\n",
    "\n",
    "print('Give me number of rows with duplicate data in the train dataset:', len(Train_data_Dupls))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Show me rows and columns in the Train_add_salary: (1000000, 7)\n"
     ]
    },
    {
     "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>companyId</th>\n",
       "      <th>jobType</th>\n",
       "      <th>degree</th>\n",
       "      <th>major</th>\n",
       "      <th>industry</th>\n",
       "      <th>yearsExperience</th>\n",
       "      <th>milesFromMetropolis</th>\n",
       "      <th>salary</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>jobId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>JOB1362684407687</th>\n",
       "      <td>COMP37</td>\n",
       "      <td>CFO</td>\n",
       "      <td>MASTERS</td>\n",
       "      <td>MATH</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>10</td>\n",
       "      <td>83</td>\n",
       "      <td>130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407688</th>\n",
       "      <td>COMP19</td>\n",
       "      <td>CEO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>WEB</td>\n",
       "      <td>3</td>\n",
       "      <td>73</td>\n",
       "      <td>101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407689</th>\n",
       "      <td>COMP52</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>DOCTORAL</td>\n",
       "      <td>PHYSICS</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>10</td>\n",
       "      <td>38</td>\n",
       "      <td>137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407690</th>\n",
       "      <td>COMP38</td>\n",
       "      <td>MANAGER</td>\n",
       "      <td>DOCTORAL</td>\n",
       "      <td>CHEMISTRY</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>8</td>\n",
       "      <td>17</td>\n",
       "      <td>142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407691</th>\n",
       "      <td>COMP7</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>BACHELORS</td>\n",
       "      <td>PHYSICS</td>\n",
       "      <td>FINANCE</td>\n",
       "      <td>8</td>\n",
       "      <td>16</td>\n",
       "      <td>163</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407692</th>\n",
       "      <td>COMP15</td>\n",
       "      <td>MANAGER</td>\n",
       "      <td>DOCTORAL</td>\n",
       "      <td>COMPSCI</td>\n",
       "      <td>FINANCE</td>\n",
       "      <td>2</td>\n",
       "      <td>31</td>\n",
       "      <td>113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407693</th>\n",
       "      <td>COMP15</td>\n",
       "      <td>CFO</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>23</td>\n",
       "      <td>24</td>\n",
       "      <td>178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407694</th>\n",
       "      <td>COMP24</td>\n",
       "      <td>JUNIOR</td>\n",
       "      <td>BACHELORS</td>\n",
       "      <td>CHEMISTRY</td>\n",
       "      <td>EDUCATION</td>\n",
       "      <td>9</td>\n",
       "      <td>70</td>\n",
       "      <td>73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407695</th>\n",
       "      <td>COMP20</td>\n",
       "      <td>JANITOR</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>EDUCATION</td>\n",
       "      <td>1</td>\n",
       "      <td>54</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407696</th>\n",
       "      <td>COMP41</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>BACHELORS</td>\n",
       "      <td>CHEMISTRY</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>17</td>\n",
       "      <td>68</td>\n",
       "      <td>104</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 companyId         jobType       degree      major   industry  \\\n",
       "jobId                                                                           \n",
       "JOB1362684407687    COMP37             CFO      MASTERS       MATH     HEALTH   \n",
       "JOB1362684407688    COMP19             CEO  HIGH_SCHOOL       NONE        WEB   \n",
       "JOB1362684407689    COMP52  VICE_PRESIDENT     DOCTORAL    PHYSICS     HEALTH   \n",
       "JOB1362684407690    COMP38         MANAGER     DOCTORAL  CHEMISTRY       AUTO   \n",
       "JOB1362684407691     COMP7  VICE_PRESIDENT    BACHELORS    PHYSICS    FINANCE   \n",
       "JOB1362684407692    COMP15         MANAGER     DOCTORAL    COMPSCI    FINANCE   \n",
       "JOB1362684407693    COMP15             CFO         NONE       NONE     HEALTH   \n",
       "JOB1362684407694    COMP24          JUNIOR    BACHELORS  CHEMISTRY  EDUCATION   \n",
       "JOB1362684407695    COMP20         JANITOR  HIGH_SCHOOL       NONE  EDUCATION   \n",
       "JOB1362684407696    COMP41  VICE_PRESIDENT    BACHELORS  CHEMISTRY       AUTO   \n",
       "\n",
       "                  yearsExperience  milesFromMetropolis  salary  \n",
       "jobId                                                           \n",
       "JOB1362684407687               10                   83     130  \n",
       "JOB1362684407688                3                   73     101  \n",
       "JOB1362684407689               10                   38     137  \n",
       "JOB1362684407690                8                   17     142  \n",
       "JOB1362684407691                8                   16     163  \n",
       "JOB1362684407692                2                   31     113  \n",
       "JOB1362684407693               23                   24     178  \n",
       "JOB1362684407694                9                   70      73  \n",
       "JOB1362684407695                1                   54      31  \n",
       "JOB1362684407696               17                   68     104  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# let us salary column to 'Train_data'\n",
    "\n",
    "Train_add_salary = Train_data.join(Train_Salaries)\n",
    "\n",
    "print (\"Show me rows and columns in the Train_add_salary:\",Train_data.shape)\n",
    "\n",
    "Train_add_salary.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "I want to see the number of rows with duplicated data in Train_add_salary: 372\n"
     ]
    }
   ],
   "source": [
    "Features2 =  ['companyId','jobType', 'degree', 'major','industry','yearsExperience','milesFromMetropolis','salary']\n",
    "\n",
    "#Figure out the duplicated records for the Train_add_salary\n",
    "\n",
    "Train_add_salary_dupls = Train_add_salary[Train_add_salary.duplicated(Features2, keep = False)]\n",
    "\n",
    "print (\"I want to see the number of rows with duplicated data in Train_add_salary:\", len(Train_add_salary_dupls))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>companyId</th>\n",
       "      <th>jobType</th>\n",
       "      <th>degree</th>\n",
       "      <th>major</th>\n",
       "      <th>industry</th>\n",
       "      <th>yearsExperience</th>\n",
       "      <th>milesFromMetropolis</th>\n",
       "      <th>salary</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>jobId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>JOB1362685003735</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>BACHELORS</td>\n",
       "      <td>BIOLOGY</td>\n",
       "      <td>SERVICE</td>\n",
       "      <td>23</td>\n",
       "      <td>34</td>\n",
       "      <td>214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685283347</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>BACHELORS</td>\n",
       "      <td>BIOLOGY</td>\n",
       "      <td>SERVICE</td>\n",
       "      <td>23</td>\n",
       "      <td>34</td>\n",
       "      <td>122</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685004580</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>0</td>\n",
       "      <td>82</td>\n",
       "      <td>129</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685288664</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>0</td>\n",
       "      <td>82</td>\n",
       "      <td>97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685165440</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>15</td>\n",
       "      <td>13</td>\n",
       "      <td>125</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685283913</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>15</td>\n",
       "      <td>13</td>\n",
       "      <td>156</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684435928</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>23</td>\n",
       "      <td>94</td>\n",
       "      <td>136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684748853</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>23</td>\n",
       "      <td>94</td>\n",
       "      <td>105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684556793</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>EDUCATION</td>\n",
       "      <td>11</td>\n",
       "      <td>63</td>\n",
       "      <td>106</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684734837</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>EDUCATION</td>\n",
       "      <td>11</td>\n",
       "      <td>63</td>\n",
       "      <td>83</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 companyId jobType       degree    major   industry  \\\n",
       "jobId                                                                 \n",
       "JOB1362685003735     COMP0     CEO    BACHELORS  BIOLOGY    SERVICE   \n",
       "JOB1362685283347     COMP0     CEO    BACHELORS  BIOLOGY    SERVICE   \n",
       "JOB1362685004580     COMP0     CEO  HIGH_SCHOOL     NONE       AUTO   \n",
       "JOB1362685288664     COMP0     CEO  HIGH_SCHOOL     NONE       AUTO   \n",
       "JOB1362685165440     COMP0     CEO  HIGH_SCHOOL     NONE       AUTO   \n",
       "JOB1362685283913     COMP0     CEO  HIGH_SCHOOL     NONE       AUTO   \n",
       "JOB1362684435928     COMP0     CEO  HIGH_SCHOOL     NONE       AUTO   \n",
       "JOB1362684748853     COMP0     CEO  HIGH_SCHOOL     NONE       AUTO   \n",
       "JOB1362684556793     COMP0     CEO  HIGH_SCHOOL     NONE  EDUCATION   \n",
       "JOB1362684734837     COMP0     CEO  HIGH_SCHOOL     NONE  EDUCATION   \n",
       "\n",
       "                  yearsExperience  milesFromMetropolis  salary  \n",
       "jobId                                                           \n",
       "JOB1362685003735               23                   34     214  \n",
       "JOB1362685283347               23                   34     122  \n",
       "JOB1362685004580                0                   82     129  \n",
       "JOB1362685288664                0                   82      97  \n",
       "JOB1362685165440               15                   13     125  \n",
       "JOB1362685283913               15                   13     156  \n",
       "JOB1362684435928               23                   94     136  \n",
       "JOB1362684748853               23                   94     105  \n",
       "JOB1362684556793               11                   63     106  \n",
       "JOB1362684734837               11                   63      83  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Let us see the duplicated records \n",
    "\n",
    "Dupls_add_salary = Train_add_salary[Train_data.duplicated(Features, keep = False)]\n",
    "\n",
    "Dupls_add_salary.sort_values(Features).head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "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>yearsExperience</th>\n",
       "      <th>milesFromMetropolis</th>\n",
       "      <th>salary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>15917.000000</td>\n",
       "      <td>15917.000000</td>\n",
       "      <td>15917.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>11.869825</td>\n",
       "      <td>49.637683</td>\n",
       "      <td>95.311679</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>7.196198</td>\n",
       "      <td>29.097380</td>\n",
       "      <td>37.297904</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>17.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>6.000000</td>\n",
       "      <td>24.000000</td>\n",
       "      <td>67.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>12.000000</td>\n",
       "      <td>49.000000</td>\n",
       "      <td>91.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>18.000000</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>119.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>24.000000</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>260.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       yearsExperience  milesFromMetropolis        salary\n",
       "count     15917.000000         15917.000000  15917.000000\n",
       "mean         11.869825            49.637683     95.311679\n",
       "std           7.196198            29.097380     37.297904\n",
       "min           0.000000             0.000000     17.000000\n",
       "25%           6.000000            24.000000     67.000000\n",
       "50%          12.000000            49.000000     91.000000\n",
       "75%          18.000000            75.000000    119.000000\n",
       "max          24.000000            99.000000    260.000000"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print('')\n",
    "Dupls_add_salary.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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>companyId</th>\n",
       "      <th>jobType</th>\n",
       "      <th>degree</th>\n",
       "      <th>major</th>\n",
       "      <th>industry</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>15917</td>\n",
       "      <td>15917</td>\n",
       "      <td>15917</td>\n",
       "      <td>15917</td>\n",
       "      <td>15917</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>63</td>\n",
       "      <td>8</td>\n",
       "      <td>5</td>\n",
       "      <td>9</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>COMP32</td>\n",
       "      <td>JANITOR</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>FINANCE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>308</td>\n",
       "      <td>6811</td>\n",
       "      <td>7409</td>\n",
       "      <td>14733</td>\n",
       "      <td>2380</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       companyId  jobType degree  major industry\n",
       "count      15917    15917  15917  15917    15917\n",
       "unique        63        8      5      9        7\n",
       "top       COMP32  JANITOR   NONE   NONE  FINANCE\n",
       "freq         308     6811   7409  14733     2380"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# It is good to see the non numerical values \n",
    "\n",
    "Text_Features = ['companyId', 'jobType', 'degree', 'major', 'industry']\n",
    "\n",
    "Dupls_add_salary[Text_Features].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(984083, 7)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Drop duplicates from train_feat dataset \n",
    "\n",
    "Train_data.drop(Dupls_add_salary.index, inplace= True)\n",
    "\n",
    "Train_data.shape\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(984083, 1)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Drop duplicates from train_feat dataset \n",
    "\n",
    "Train_Salaries.drop(Dupls_add_salary.index, inplace= True)\n",
    "Train_Salaries.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": true
   },
   "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>companyId</th>\n",
       "      <th>jobType</th>\n",
       "      <th>degree</th>\n",
       "      <th>major</th>\n",
       "      <th>industry</th>\n",
       "      <th>yearsExperience</th>\n",
       "      <th>milesFromMetropolis</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>jobId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>JOB1362685464148</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>4</td>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686275559</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>4</td>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685526996</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>22</td>\n",
       "      <td>83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686243166</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>22</td>\n",
       "      <td>83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685975334</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>EDUCATION</td>\n",
       "      <td>16</td>\n",
       "      <td>61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686139961</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>EDUCATION</td>\n",
       "      <td>16</td>\n",
       "      <td>61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685579600</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>SERVICE</td>\n",
       "      <td>17</td>\n",
       "      <td>97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686036582</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>SERVICE</td>\n",
       "      <td>17</td>\n",
       "      <td>97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685597246</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>WEB</td>\n",
       "      <td>20</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685740129</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CEO</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>WEB</td>\n",
       "      <td>20</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685718750</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CFO</td>\n",
       "      <td>DOCTORAL</td>\n",
       "      <td>CHEMISTRY</td>\n",
       "      <td>FINANCE</td>\n",
       "      <td>4</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685735025</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CFO</td>\n",
       "      <td>DOCTORAL</td>\n",
       "      <td>CHEMISTRY</td>\n",
       "      <td>FINANCE</td>\n",
       "      <td>4</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685529715</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CFO</td>\n",
       "      <td>DOCTORAL</td>\n",
       "      <td>MATH</td>\n",
       "      <td>WEB</td>\n",
       "      <td>5</td>\n",
       "      <td>35</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685865848</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CFO</td>\n",
       "      <td>DOCTORAL</td>\n",
       "      <td>MATH</td>\n",
       "      <td>WEB</td>\n",
       "      <td>5</td>\n",
       "      <td>35</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685678043</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CFO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>3</td>\n",
       "      <td>29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685989023</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CFO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>3</td>\n",
       "      <td>29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685510279</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CFO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686064987</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CFO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685866111</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CFO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>22</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686032975</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CFO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>22</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685562700</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CFO</td>\n",
       "      <td>MASTERS</td>\n",
       "      <td>MATH</td>\n",
       "      <td>FINANCE</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685861237</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CFO</td>\n",
       "      <td>MASTERS</td>\n",
       "      <td>MATH</td>\n",
       "      <td>FINANCE</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685561641</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CFO</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>WEB</td>\n",
       "      <td>0</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686042880</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CFO</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>WEB</td>\n",
       "      <td>0</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685985678</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CTO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>22</td>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686319224</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CTO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>22</td>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685950614</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CTO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>FINANCE</td>\n",
       "      <td>0</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686147468</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CTO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>FINANCE</td>\n",
       "      <td>0</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685511806</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CTO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>OIL</td>\n",
       "      <td>23</td>\n",
       "      <td>73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685946974</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CTO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>OIL</td>\n",
       "      <td>23</td>\n",
       "      <td>73</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685410456</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>SENIOR</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>3</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686059381</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>SENIOR</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>3</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685434185</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>9</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686162428</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>9</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685964669</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>22</td>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686125268</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>22</td>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685862739</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>OIL</td>\n",
       "      <td>17</td>\n",
       "      <td>87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686139591</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>OIL</td>\n",
       "      <td>17</td>\n",
       "      <td>87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685941525</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>MASTERS</td>\n",
       "      <td>BIOLOGY</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>23</td>\n",
       "      <td>92</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686268467</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>MASTERS</td>\n",
       "      <td>BIOLOGY</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>23</td>\n",
       "      <td>92</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685905201</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>MASTERS</td>\n",
       "      <td>ENGINEERING</td>\n",
       "      <td>OIL</td>\n",
       "      <td>5</td>\n",
       "      <td>96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686100346</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>MASTERS</td>\n",
       "      <td>ENGINEERING</td>\n",
       "      <td>OIL</td>\n",
       "      <td>5</td>\n",
       "      <td>96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685819993</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>23</td>\n",
       "      <td>87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686281378</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>23</td>\n",
       "      <td>87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685692212</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>EDUCATION</td>\n",
       "      <td>7</td>\n",
       "      <td>65</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686355462</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>EDUCATION</td>\n",
       "      <td>7</td>\n",
       "      <td>65</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686150298</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>EDUCATION</td>\n",
       "      <td>15</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686336619</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>EDUCATION</td>\n",
       "      <td>15</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686167124</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>EDUCATION</td>\n",
       "      <td>19</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686229505</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>EDUCATION</td>\n",
       "      <td>19</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685622438</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>FINANCE</td>\n",
       "      <td>21</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685726139</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>FINANCE</td>\n",
       "      <td>21</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686366911</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>5</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686406949</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>5</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685818380</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>24</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686221098</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>24</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685955412</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>SERVICE</td>\n",
       "      <td>5</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686128323</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>SERVICE</td>\n",
       "      <td>5</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685832750</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>SERVICE</td>\n",
       "      <td>23</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362686377041</th>\n",
       "      <td>COMP9</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>SERVICE</td>\n",
       "      <td>23</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>15995 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 companyId         jobType       degree        major  \\\n",
       "jobId                                                                  \n",
       "JOB1362685464148     COMP0             CEO  HIGH_SCHOOL         NONE   \n",
       "JOB1362686275559     COMP0             CEO  HIGH_SCHOOL         NONE   \n",
       "JOB1362685526996     COMP0             CEO  HIGH_SCHOOL         NONE   \n",
       "JOB1362686243166     COMP0             CEO  HIGH_SCHOOL         NONE   \n",
       "JOB1362685975334     COMP0             CEO         NONE         NONE   \n",
       "JOB1362686139961     COMP0             CEO         NONE         NONE   \n",
       "JOB1362685579600     COMP0             CEO         NONE         NONE   \n",
       "JOB1362686036582     COMP0             CEO         NONE         NONE   \n",
       "JOB1362685597246     COMP0             CEO         NONE         NONE   \n",
       "JOB1362685740129     COMP0             CEO         NONE         NONE   \n",
       "JOB1362685718750     COMP0             CFO     DOCTORAL    CHEMISTRY   \n",
       "JOB1362685735025     COMP0             CFO     DOCTORAL    CHEMISTRY   \n",
       "JOB1362685529715     COMP0             CFO     DOCTORAL         MATH   \n",
       "JOB1362685865848     COMP0             CFO     DOCTORAL         MATH   \n",
       "JOB1362685678043     COMP0             CFO  HIGH_SCHOOL         NONE   \n",
       "JOB1362685989023     COMP0             CFO  HIGH_SCHOOL         NONE   \n",
       "JOB1362685510279     COMP0             CFO  HIGH_SCHOOL         NONE   \n",
       "JOB1362686064987     COMP0             CFO  HIGH_SCHOOL         NONE   \n",
       "JOB1362685866111     COMP0             CFO  HIGH_SCHOOL         NONE   \n",
       "JOB1362686032975     COMP0             CFO  HIGH_SCHOOL         NONE   \n",
       "JOB1362685562700     COMP0             CFO      MASTERS         MATH   \n",
       "JOB1362685861237     COMP0             CFO      MASTERS         MATH   \n",
       "JOB1362685561641     COMP0             CFO         NONE         NONE   \n",
       "JOB1362686042880     COMP0             CFO         NONE         NONE   \n",
       "JOB1362685985678     COMP0             CTO  HIGH_SCHOOL         NONE   \n",
       "JOB1362686319224     COMP0             CTO  HIGH_SCHOOL         NONE   \n",
       "JOB1362685950614     COMP0             CTO  HIGH_SCHOOL         NONE   \n",
       "JOB1362686147468     COMP0             CTO  HIGH_SCHOOL         NONE   \n",
       "JOB1362685511806     COMP0             CTO  HIGH_SCHOOL         NONE   \n",
       "JOB1362685946974     COMP0             CTO  HIGH_SCHOOL         NONE   \n",
       "...                    ...             ...          ...          ...   \n",
       "JOB1362685410456     COMP9          SENIOR         NONE         NONE   \n",
       "JOB1362686059381     COMP9          SENIOR         NONE         NONE   \n",
       "JOB1362685434185     COMP9  VICE_PRESIDENT  HIGH_SCHOOL         NONE   \n",
       "JOB1362686162428     COMP9  VICE_PRESIDENT  HIGH_SCHOOL         NONE   \n",
       "JOB1362685964669     COMP9  VICE_PRESIDENT  HIGH_SCHOOL         NONE   \n",
       "JOB1362686125268     COMP9  VICE_PRESIDENT  HIGH_SCHOOL         NONE   \n",
       "JOB1362685862739     COMP9  VICE_PRESIDENT  HIGH_SCHOOL         NONE   \n",
       "JOB1362686139591     COMP9  VICE_PRESIDENT  HIGH_SCHOOL         NONE   \n",
       "JOB1362685941525     COMP9  VICE_PRESIDENT      MASTERS      BIOLOGY   \n",
       "JOB1362686268467     COMP9  VICE_PRESIDENT      MASTERS      BIOLOGY   \n",
       "JOB1362685905201     COMP9  VICE_PRESIDENT      MASTERS  ENGINEERING   \n",
       "JOB1362686100346     COMP9  VICE_PRESIDENT      MASTERS  ENGINEERING   \n",
       "JOB1362685819993     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362686281378     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362685692212     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362686355462     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362686150298     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362686336619     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362686167124     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362686229505     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362685622438     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362685726139     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362686366911     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362686406949     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362685818380     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362686221098     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362685955412     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362686128323     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362685832750     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "JOB1362686377041     COMP9  VICE_PRESIDENT         NONE         NONE   \n",
       "\n",
       "                   industry  yearsExperience  milesFromMetropolis  \n",
       "jobId                                                              \n",
       "JOB1362685464148       AUTO                4                   80  \n",
       "JOB1362686275559       AUTO                4                   80  \n",
       "JOB1362685526996     HEALTH               22                   83  \n",
       "JOB1362686243166     HEALTH               22                   83  \n",
       "JOB1362685975334  EDUCATION               16                   61  \n",
       "JOB1362686139961  EDUCATION               16                   61  \n",
       "JOB1362685579600    SERVICE               17                   97  \n",
       "JOB1362686036582    SERVICE               17                   97  \n",
       "JOB1362685597246        WEB               20                   43  \n",
       "JOB1362685740129        WEB               20                   43  \n",
       "JOB1362685718750    FINANCE                4                   60  \n",
       "JOB1362685735025    FINANCE                4                   60  \n",
       "JOB1362685529715        WEB                5                   35  \n",
       "JOB1362685865848        WEB                5                   35  \n",
       "JOB1362685678043       AUTO                3                   29  \n",
       "JOB1362685989023       AUTO                3                   29  \n",
       "JOB1362685510279     HEALTH               12                   10  \n",
       "JOB1362686064987     HEALTH               12                   10  \n",
       "JOB1362685866111     HEALTH               22                    9  \n",
       "JOB1362686032975     HEALTH               22                    9  \n",
       "JOB1362685562700    FINANCE                4                    0  \n",
       "JOB1362685861237    FINANCE                4                    0  \n",
       "JOB1362685561641        WEB                0                   43  \n",
       "JOB1362686042880        WEB                0                   43  \n",
       "JOB1362685985678       AUTO               22                   36  \n",
       "JOB1362686319224       AUTO               22                   36  \n",
       "JOB1362685950614    FINANCE                0                   24  \n",
       "JOB1362686147468    FINANCE                0                   24  \n",
       "JOB1362685511806        OIL               23                   73  \n",
       "JOB1362685946974        OIL               23                   73  \n",
       "...                     ...              ...                  ...  \n",
       "JOB1362685410456     HEALTH                3                   26  \n",
       "JOB1362686059381     HEALTH                3                   26  \n",
       "JOB1362685434185       AUTO                9                    2  \n",
       "JOB1362686162428       AUTO                9                    2  \n",
       "JOB1362685964669       AUTO               22                   81  \n",
       "JOB1362686125268       AUTO               22                   81  \n",
       "JOB1362685862739        OIL               17                   87  \n",
       "JOB1362686139591        OIL               17                   87  \n",
       "JOB1362685941525     HEALTH               23                   92  \n",
       "JOB1362686268467     HEALTH               23                   92  \n",
       "JOB1362685905201        OIL                5                   96  \n",
       "JOB1362686100346        OIL                5                   96  \n",
       "JOB1362685819993       AUTO               23                   87  \n",
       "JOB1362686281378       AUTO               23                   87  \n",
       "JOB1362685692212  EDUCATION                7                   65  \n",
       "JOB1362686355462  EDUCATION                7                   65  \n",
       "JOB1362686150298  EDUCATION               15                    4  \n",
       "JOB1362686336619  EDUCATION               15                    4  \n",
       "JOB1362686167124  EDUCATION               19                   13  \n",
       "JOB1362686229505  EDUCATION               19                   13  \n",
       "JOB1362685622438    FINANCE               21                   16  \n",
       "JOB1362685726139    FINANCE               21                   16  \n",
       "JOB1362686366911     HEALTH                5                   32  \n",
       "JOB1362686406949     HEALTH                5                   32  \n",
       "JOB1362685818380     HEALTH               24                   23  \n",
       "JOB1362686221098     HEALTH               24                   23  \n",
       "JOB1362685955412    SERVICE                5                   15  \n",
       "JOB1362686128323    SERVICE                5                   15  \n",
       "JOB1362685832750    SERVICE               23                    8  \n",
       "JOB1362686377041    SERVICE               23                    8  \n",
       "\n",
       "[15995 rows x 7 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Review duplicates in test_feat dataset \n",
    "\n",
    "Filt = Test_Fetaures.duplicated(Features, keep = False)\n",
    "Test_Fetaures[Filt].sort_values(Features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(991954, 7)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# To predict the saleries, let us drop all non-unique rows \n",
    "\n",
    "Test_Fetaures.drop_duplicates(Features, inplace= True)\n",
    "Test_Fetaures.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(984083, 8)"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Train_add_salary.drop(Dupls_add_salary.index, inplace = True)\n",
    "Train_add_salary.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>yearsExperience</th>\n",
       "      <th>milesFromMetropolis</th>\n",
       "      <th>salary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>984083.000000</td>\n",
       "      <td>984083.000000</td>\n",
       "      <td>984083.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>11.994368</td>\n",
       "      <td>49.527506</td>\n",
       "      <td>116.397440</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>7.212639</td>\n",
       "      <td>28.874178</td>\n",
       "      <td>38.649051</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>6.000000</td>\n",
       "      <td>25.000000</td>\n",
       "      <td>89.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>12.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>114.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>18.000000</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>141.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>24.000000</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>301.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       yearsExperience  milesFromMetropolis         salary\n",
       "count    984083.000000        984083.000000  984083.000000\n",
       "mean         11.994368            49.527506     116.397440\n",
       "std           7.212639            28.874178      38.649051\n",
       "min           0.000000             0.000000       0.000000\n",
       "25%           6.000000            25.000000      89.000000\n",
       "50%          12.000000            50.000000     114.000000\n",
       "75%          18.000000            75.000000     141.000000\n",
       "max          24.000000            99.000000     301.000000"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Train_add_salary.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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>companyId</th>\n",
       "      <th>jobType</th>\n",
       "      <th>degree</th>\n",
       "      <th>major</th>\n",
       "      <th>industry</th>\n",
       "      <th>yearsExperience</th>\n",
       "      <th>milesFromMetropolis</th>\n",
       "      <th>salary</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>jobId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>JOB1362684438246</th>\n",
       "      <td>COMP44</td>\n",
       "      <td>JUNIOR</td>\n",
       "      <td>DOCTORAL</td>\n",
       "      <td>MATH</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>11</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684903671</th>\n",
       "      <td>COMP34</td>\n",
       "      <td>JUNIOR</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>OIL</td>\n",
       "      <td>1</td>\n",
       "      <td>25</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685059763</th>\n",
       "      <td>COMP25</td>\n",
       "      <td>CTO</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>6</td>\n",
       "      <td>60</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685223816</th>\n",
       "      <td>COMP42</td>\n",
       "      <td>MANAGER</td>\n",
       "      <td>DOCTORAL</td>\n",
       "      <td>ENGINEERING</td>\n",
       "      <td>FINANCE</td>\n",
       "      <td>18</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685235843</th>\n",
       "      <td>COMP40</td>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>MASTERS</td>\n",
       "      <td>ENGINEERING</td>\n",
       "      <td>WEB</td>\n",
       "      <td>3</td>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 companyId         jobType       degree        major industry  \\\n",
       "jobId                                                                           \n",
       "JOB1362684438246    COMP44          JUNIOR     DOCTORAL         MATH     AUTO   \n",
       "JOB1362684903671    COMP34          JUNIOR         NONE         NONE      OIL   \n",
       "JOB1362685059763    COMP25             CTO  HIGH_SCHOOL         NONE     AUTO   \n",
       "JOB1362685223816    COMP42         MANAGER     DOCTORAL  ENGINEERING  FINANCE   \n",
       "JOB1362685235843    COMP40  VICE_PRESIDENT      MASTERS  ENGINEERING      WEB   \n",
       "\n",
       "                  yearsExperience  milesFromMetropolis  salary  \n",
       "jobId                                                           \n",
       "JOB1362684438246               11                    7       0  \n",
       "JOB1362684903671                1                   25       0  \n",
       "JOB1362685059763                6                   60       0  \n",
       "JOB1362685223816               18                    6       0  \n",
       "JOB1362685235843                3                   29       0  "
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Let us see rows with zero salary \n",
    "\n",
    "Zero_salary = Train_add_salary['salary'] <= 0\n",
    "Train_add_salary[Zero_salary]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>salary</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>jobId</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>JOB1362684407687</th>\n",
       "      <td>130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407688</th>\n",
       "      <td>101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407689</th>\n",
       "      <td>137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407690</th>\n",
       "      <td>142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407691</th>\n",
       "      <td>163</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407692</th>\n",
       "      <td>113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407693</th>\n",
       "      <td>178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407694</th>\n",
       "      <td>73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407695</th>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407696</th>\n",
       "      <td>104</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407697</th>\n",
       "      <td>102</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407698</th>\n",
       "      <td>144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407699</th>\n",
       "      <td>79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407700</th>\n",
       "      <td>193</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407701</th>\n",
       "      <td>47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407702</th>\n",
       "      <td>172</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407703</th>\n",
       "      <td>47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407704</th>\n",
       "      <td>126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407705</th>\n",
       "      <td>122</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407706</th>\n",
       "      <td>95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407707</th>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407708</th>\n",
       "      <td>68</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407709</th>\n",
       "      <td>105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407710</th>\n",
       "      <td>76</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407711</th>\n",
       "      <td>202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407712</th>\n",
       "      <td>131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407713</th>\n",
       "      <td>158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407714</th>\n",
       "      <td>130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407715</th>\n",
       "      <td>101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362684407716</th>\n",
       "      <td>82</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407657</th>\n",
       "      <td>216</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407658</th>\n",
       "      <td>115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407659</th>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407660</th>\n",
       "      <td>117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407661</th>\n",
       "      <td>131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407662</th>\n",
       "      <td>97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407663</th>\n",
       "      <td>112</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407664</th>\n",
       "      <td>142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407665</th>\n",
       "      <td>93</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407666</th>\n",
       "      <td>181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407667</th>\n",
       "      <td>84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407668</th>\n",
       "      <td>132</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407669</th>\n",
       "      <td>113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407670</th>\n",
       "      <td>122</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407671</th>\n",
       "      <td>110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407672</th>\n",
       "      <td>147</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407673</th>\n",
       "      <td>117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407674</th>\n",
       "      <td>124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407675</th>\n",
       "      <td>79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407676</th>\n",
       "      <td>72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407677</th>\n",
       "      <td>89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407678</th>\n",
       "      <td>139</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407679</th>\n",
       "      <td>143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407680</th>\n",
       "      <td>136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407681</th>\n",
       "      <td>115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407682</th>\n",
       "      <td>88</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407683</th>\n",
       "      <td>160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407684</th>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407685</th>\n",
       "      <td>149</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOB1362685407686</th>\n",
       "      <td>88</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>984078 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  salary\n",
       "jobId                   \n",
       "JOB1362684407687     130\n",
       "JOB1362684407688     101\n",
       "JOB1362684407689     137\n",
       "JOB1362684407690     142\n",
       "JOB1362684407691     163\n",
       "JOB1362684407692     113\n",
       "JOB1362684407693     178\n",
       "JOB1362684407694      73\n",
       "JOB1362684407695      31\n",
       "JOB1362684407696     104\n",
       "JOB1362684407697     102\n",
       "JOB1362684407698     144\n",
       "JOB1362684407699      79\n",
       "JOB1362684407700     193\n",
       "JOB1362684407701      47\n",
       "JOB1362684407702     172\n",
       "JOB1362684407703      47\n",
       "JOB1362684407704     126\n",
       "JOB1362684407705     122\n",
       "JOB1362684407706      95\n",
       "JOB1362684407707      32\n",
       "JOB1362684407708      68\n",
       "JOB1362684407709     105\n",
       "JOB1362684407710      76\n",
       "JOB1362684407711     202\n",
       "JOB1362684407712     131\n",
       "JOB1362684407713     158\n",
       "JOB1362684407714     130\n",
       "JOB1362684407715     101\n",
       "JOB1362684407716      82\n",
       "...                  ...\n",
       "JOB1362685407657     216\n",
       "JOB1362685407658     115\n",
       "JOB1362685407659      56\n",
       "JOB1362685407660     117\n",
       "JOB1362685407661     131\n",
       "JOB1362685407662      97\n",
       "JOB1362685407663     112\n",
       "JOB1362685407664     142\n",
       "JOB1362685407665      93\n",
       "JOB1362685407666     181\n",
       "JOB1362685407667      84\n",
       "JOB1362685407668     132\n",
       "JOB1362685407669     113\n",
       "JOB1362685407670     122\n",
       "JOB1362685407671     110\n",
       "JOB1362685407672     147\n",
       "JOB1362685407673     117\n",
       "JOB1362685407674     124\n",
       "JOB1362685407675      79\n",
       "JOB1362685407676      72\n",
       "JOB1362685407677      89\n",
       "JOB1362685407678     139\n",
       "JOB1362685407679     143\n",
       "JOB1362685407680     136\n",
       "JOB1362685407681     115\n",
       "JOB1362685407682      88\n",
       "JOB1362685407683     160\n",
       "JOB1362685407684      64\n",
       "JOB1362685407685     149\n",
       "JOB1362685407686      88\n",
       "\n",
       "[984078 rows x 1 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Let us drop rows with zero salary from the dataset\n",
    "\n",
    "Train_data.drop(Train_add_salary[Zero_salary].index, inplace = True)\n",
    "Train_Salaries.drop(Train_add_salary[Zero_salary].index, inplace=True)\n",
    "Train_add_salary.drop(Train_add_salary[Zero_salary].index, inplace = True)\n",
    "Train_Salaries\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>yearsExperience</th>\n",
       "      <th>milesFromMetropolis</th>\n",
       "      <th>salary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>984078.000000</td>\n",
       "      <td>984078.000000</td>\n",
       "      <td>984078.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>11.994390</td>\n",
       "      <td>49.527629</td>\n",
       "      <td>116.398031</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>7.212638</td>\n",
       "      <td>28.874166</td>\n",
       "      <td>38.648258</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>18.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>6.000000</td>\n",
       "      <td>25.000000</td>\n",
       "      <td>89.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>12.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>114.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>18.000000</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>141.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>24.000000</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>301.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       yearsExperience  milesFromMetropolis         salary\n",
       "count    984078.000000        984078.000000  984078.000000\n",
       "mean         11.994390            49.527629     116.398031\n",
       "std           7.212638            28.874166      38.648258\n",
       "min           0.000000             0.000000      18.000000\n",
       "25%           6.000000            25.000000      89.000000\n",
       "50%          12.000000            50.000000     114.000000\n",
       "75%          18.000000            75.000000     141.000000\n",
       "max          24.000000            99.000000     301.000000"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Let us summerize continuous variables \n",
    "Train_add_salary.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\user\\Anaconda3\\lib\\site-packages\\scipy\\stats\\stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<function matplotlib.pyplot.show(*args, **kw)>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYkAAAEKCAYAAADn+anLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzt3Xl81NW9//FXZstkluyT7SSQlRBEcUGoG+IuaqX210Xbe9va9lK9eq22t61b1dZq7fW2Vm97bWm1autaN6gCda1WERXZV0EgkAPIvmXPzPf3xwy9KSYYMJNt3s/Hg8dMvvOdmc+XgXnnnPP9npPmOA4iIiJdcfV3ASIiMnApJEREpFsKCRER6ZZCQkREuqWQEBGRbikkRESkWwoJERHplkJCRES6pZAQEZFuefq7gE9i9OjRTllZWX+XISIyqCxatGibtTbSk30HdUiUlZUxc+bM/i5DRGRQMcbU93RfdTeJiEi3FBIiItIthYSIiHRLISEiIt1SSIiISLcUEiIi0i2FhIiIdEshISIi3UraxXTGmPuBC4At1trRBzz2n8CdQMRau80YkwbcDZwHNAFfs9bOS1ZtIiKDleM4bN3bQmtHjLLcYNLfL5ktiQeAcw/caIwpA84C1nfaPAmoSfyZAtybxLpERAYdx3HYtreF3772AZ/7zVv88a31H/+kXpC0kLDWvg7s6OKhu4DvA06nbZOBh6y1jrV2DpBtjClOVm0iIoOF4zjs2NfKfX9fyxd+O4c7Zq2ktSNGbtDXJ+/fp3M3GWMuBKy1dqEx5p8eAjZ0+rkhsW1TH5YnIjJgOI7D7uZ2nplveeTt9azaso+CcDr/cXo1Y4fnkB0YYiFhjAkANwBnd/FwWhfbnC62YYyZQrxLigOCRkRkSNjd1M5zizbyxzn1rNi8l7ygj3+fWMXJVfm0RqME/R4qIskfj4C+bUlUARXA/lZEKTDPGDOOeMuh85zfpcDGrl7EWjsVmAowadKkLoNERGQw2t3czqwlm3hwdj3LNu0hJ+BlyimVnDoiQls0hs/jYmRJJjkBL2lpXf1u3fv6LCSstYuBgv0/G2PWAWMTZzdNB640xjwGjAd2W2vV1SQiKWFvSzsvLN3MA7PXsdjuISvDy9dPKuf02gLaHQefJ42RxVnkBn19Fg77JfMU2EeBiUC+MaYBuNlae183u88gfvrrauKnwF6arLpERAaKaMzhtfe3cM/Lq1mwYRfhdA9fOWE4Z40qpCPq4PW4qI0EyQ+l93k47Je0kLDWXvIxj5d3uu8AVySrFhGRgWZnYyt3/nUlT8xtIN3r4l/GD+OcI4rocGJ43S5qC+Ph4HL1TzjsN6hXphMRGWxiMYc3Vm/llunLWLOtkZOq8vjGyZW4XOBxp1EbyRoQ4bCfQkJEpI/saW7n5y+s5OF31pPhcfO9s0dwZGk2LhfURELkh/24B0g47KeQEBFJsljM4Z21O/jhtCWs2rKP8RW5fPOUClxpaRRkplOZH8LnGZhT6SkkRESSqLGlg7teep+H3qrH60njmjNrOKYsG5c7jbqiTPJC6f1d4kEpJEREksBxHObV7+SGZ5ewYvNexg7P4VsTqnC7IZLppyoycFsPnSkkRER6WVNrB796dTX3vbEWtyuNb59ew3Hl2bjS0qgtChMJ+/u7xB5TSIiI9KLFDTu59uklLN24h2PKsrl8YhUeVxr5oXSqCkKke9z9XeIhUUiIiPSCaMzhvjfW8osXVwJwxcRqTqjKxQFqC8NEwv13QdwnoZAQEfmEdja28f2nFvLisi2MKs7kqtOr8Xld5AR8VBWE8HsHV+uhM4WEiMgn8N66HVzzxALW72jmc8eWcuGYEtLSSIw9DM7WQ2cKCRGRwxCNOdz/5lr++68r8Xlc3HTBKCojQbICXmqLwoO69dCZQkJE5BDtamzj+08t4oVlHzKyKMzVZ9bgdbsYnhdgeG5wwEyp0RsUEiIih2B+/Q6ufmIh9dub+OwxhslHl+BypTG6JIucPlpStC8pJEREeiAac3hg9jru/OsKPC4XN55fR1UkRGaGl7rizCHTvXQghYSIyMfY3dTOD55eyKwlH1JbFOaaM0bg86YxPC9Aed7Q6l46kEJCROQgFqzfwdWPL2Td9iY+c7Ths8eWkJaWxqjigT/vUm9QSIiIdKG1I8of31rHz19YhSsNrj9vJDUFYULpHkaVZA3Z7qUDKSRERDpxHIelG3fzo78s4911OxlREOKaM0fg87kYluheGmhrPiSTQkJEJGFPczv3vLyKP71dTywGX/nUcM6qK4Q0qCsOkz+IJubrLQoJEUl5HdEYr67Ywm0zlrNuexPHDsvm306pxO91E/C7GVWcRYYvNbqXDqSQEJGUZnc28ZPnlzNr6WayM7x8/5xajirNIhpzqC4IU5zlH9JnL30chYSIpKTmtg4efns9v351Nbua2jl3dBEXH19G1HGGxMR8vUUhISIpxXEcFq7fxY+eW8b8Dbsozwtw3aQ6CrPS8bpdHFEQSsmxh+4oJEQkZexqauOXL73Po+9sAODSE8s5o66AtmiMspwAZbkBvO6Bv6RoX1JIiMiQF405vLh0M7fPXMH6HU0cX57DN0+uwOd1keFzM6Ywm7Df299lDkgKCREZ0jbtbuYnzy1jxpLN5AR8XHvuSI4oCeMANQVhCjNTe2D64ygkRGRIao/GePq9Bu584X227Wtl0ugivjg2PjCdF0qnMqKB6Z5QSIjIkLN2WyM3T1/C6+9voyTbz+0XjcZkZ+DzuBhZlDkkp/ROFoWEiAwZLe0dPPRWPb96ZTX7Wjv4/HHx5UTbozFMToDyvAAeDUwfEoWEiAx6juOwbOMebpq+lPfqd1IVCXLLp48gJ+jD73VxVFk2WRkamD4cCgkRGdSa2zr49aurue+NdURjDl87sZyz6gppj8WozA9icgIpNSFfb0taSBhj7gcuALZYa0cntt0KTAZiwBbga9bajcaYNOBu4DygKbF9XrJqE5HBLxZzmFu/g5umLWXF5r0cabK4/NQqMtJd+H0uxhRlE0rX78GfVDL/Bh8AfgU81GnbndbaHwIYY64CbgIuAyYBNYk/44F7E7ciIv/EcRy27m3lf15ZxWPvbsDrdnHladV8qjKXjphDjeZb6lVJCwlr7evGmPIDtu3p9GMQcBL3JwMPWWsdYI4xJtsYU2yt3ZSs+kRkcHEch+2Nrdz/5jr+/O4Gtu5rY3xFLt88uQKPx0XY72VEYThlZ2tNlj5vixljbgO+AuwGTtu/GdjQabeGxDaFhIiwbV8LD86u5/F3N7BlbytVkSCXT6xmRGEIx4ERhWEKMtNJS1Probf1eUhYa28AbjDGXAdcCdwMdPXJOl1swxgzBZiSuJ+sMkVkANixr40H31rLY+9u4MM9rVTmB5kyoZIjijNp7oiSG/Tporgk689RnUeA54mHRANQ1umxUmBjV0+y1k4FpgJMmjSpyyARkcFtR2Mbf5pTz8Nv1/PhnlbK8wJcP2kkR5osmjuiBBLrTGcFdFprsvVpSBhjaqy1qxI/XgisSNyfDlxpjHmM+ID1bo1HiKSeXU3xcHjk7fVs3N3C8NwA1547kiNLs2jdHw4mS9c89KFkngL7KDARyDfGNBBvMZxnjKklfgpsPfEzmwBmED/9dTXxU2AvTVZdIjLwNLZ08NCcdfxxTj0bd7VQlhvgB+eO5KjSLNo6ooT8HkbnKRz6QzLPbrqki833dbOvA1yRrFpEZOCavXob1z69iPU7minNyeB7Z9dydFk2rdF4OJQrHPqVrjQRkX7R3hHl7pdX8dvX1xD0efjuWSM4tiyH1liUcIZaDgOFQkJE+pzd1cw1jy/gnbU7OGZYNpdPqMLjSSMc8HBkfhaZWgBowFBIiEifcRyHvy7ZzA3PLmFXUzuXnljOqbUR/F43dcWZajkMQAoJEekTzW1R7pi5nD/OqScSjq/xUJCZTmGmn+qCkNaWHqAUEiKSdKs+3MvVjy9g6cY9TKjJ52snluNypTGyMExhll9XSg9gCgkRSZpYzOHJ9zbw4+eW0xaN8R+nVXPc8GxC6V7qSjIJapbWAU+fkIgkxZ7mdn747BKmLdxIeV6A75w1gpDfS0mOn8r8kFaIGyQUEiLS6xas38k1Tyxk7bZGzjuymC8eV0aay6GuOJNI2N/f5ckhUEiISK9pbY/yu7+v4Vevrsadlsa1k0ZSWxgmlOFmVHGWpvEehBQSIvKJOY7DooZd3DhtKYsbdjOqOJP/OL2adK+L4XkBhucFtYToIKWQEJFPpLG1g1++9D4PvlWPKw2+NaGSk6vzSUuDI0qyyAn6+rtE+QQUEiJyWBzH4Y1V27h5+lLWbGtk7PAc/u2USnweF1mB+CpxWudh8FNIiMgh29nYyu0zVvD0PEsw3c1/nj2Co0wWLlcaIwrDRMJaJW6oUEiISI/FYg7PLdrI7TNWsHlPC6ePLOBfxg/DSYOCLD8V+UHSPWo9DCUKCRHpEbuziVumL+XF5VsozEznRxcewfC8AH6vm5FFYbIDGnsYihQSInJQHdEYD7+9nrtefJ/dLe1cdIxh8pgSHGB4XoCynIAujBvCFBIi0q01W/dx7VOLeWfdDiryg9x4fh15YR9hv4cRhWHCmtJ7yFNIiMhHxGIOj7xdzx2zVtLaEeWrJ5Rz2sgIaUBNQZiiLD8uXfeQEhQSIvJPtuxp4dqnF/PKii1URYJcdXoNmRke8kM+qgt0WmuqUUiICPB/CwLdOG0J2xvb+MLYMs4bXYjH7aK2KEx+SKe1piKFhIiwt6WdW/+yjD/Pa6AgnM5tk0dTnJ1BJJxOVUFIrYcUppAQSXFz1+3gu39eSP32Js4aVcglY4fhcqMFgQRQSIikrLb2KHe/vIqpf19DhtfNteeOpLYoTHbAS21RmIBPXw+ikBBJSau37OU7jy9kkd2dmHOpAp/HTVUkSGlOQGcuyT8oJERSSCzm8Mc59fzXrBV0xBwuP7WK8RU5ZPg8jCrJ1HUP8hEKCZEU4DgOKzfv5SfPL+eN1duoKQhx5WnVhDM8lOYGKc/TVdPSNYWEyBC3YUcjd720iucWbiLqOFx8fBnnHFGIz+NmVHGm1nuQg1JIiAxRW/e2cO/fPuDxuRtobI0yoSbCF8eWkeFzUZjpp7ogjM+j1oMcnEJCZIjZ09zOg7PX8sDserY3tnFMWTZfHj+M/HA6XreLmoKQ1nuQHlNIiAwRzW0dPDXP8pvXPqBhZzPViSk1yvMDuNLSqIgEKcr0a+xBDolCQmSQa+uI8eKyzdzz8mpWfriXokw/3z+7ltGlmcQcGJYXwGQH1LUkh0UhITJIxWIOsz/Yxi9fWsXc+p1kZ3i57NQqTqrKoyMWoyDTT3leUFNqyCeStJAwxtwPXABssdaOTmy7E/g00AZ8AFxqrd2VeOw64BtAFLjKWvvXZNUmMtjtbmrj+meWMHPJJnweF18eP4yzRxUSdRyyAl4qIyFC6fodUD65ZLY/HwDOPWDbi8Boa+1RwPvAdQDGmFHAxcARief8rzFGv/6IdGHhhl1c+Os3eX7xJs4dXcyvv3QsZ9YVEvR7OG5YLkeVZisgpNck7V+StfZ1Y0z5Adte6PTjHOBzifuTgcesta3AWmPMamAc8Fay6hMZbGIxh4feWscds1bgcaVxw3l11BTGZ2itigQ1lbckRX/+uvF14PHEfUM8NPZrSGz7CGPMFGBK4n4y6xMZMPY0t3Pt04uZsXgTtYVhvn1GDRk+N1WRECXZGbg115IkSb+EhDHmBqADeDixqat/4U5Xz7XWTgWmAkyaNKnLfUSGksUNu7jqsQWs3dbIZ44uYfKYEvw+N0eYLDI115IkWY9CwhjjttZGe+MNjTFfJT6gfYa1dv+XfANQ1mm3UmBjb7yfyGAVizk8/HY9t89YgSuNf0zlXZiZrqulpc/0tCWx2hjzJPAHa+2yw30zY8y5wA+AU621TZ0emg48Yoz5BVAC1ADvHO77iAx2e5rbueHZxfxl4SZqCkJcdXo1Ib+XkUVhirQQkPShnobEUcTPPvq9McYF3E98oHlPd08wxjwKTATyjTENwM3Ez2ZKB15MjCfMsdZeZq1daox5AlhGvBvqit5quYgMNks37uaqR+fzwdZGPn1UMZOPLiHs92oqb+kXaY5zaN36xpgJwKNANvAkcKu1dnUSavtYkyZNcmbOnNkfby3S62Ixh8feWc9PZiwH4IqJ1dQVhynJyaAqEsKr6TSklxhj3rPWju3Jvj0ekwDOBy4FyoGfEx90PgWYAYw4rEpFBIBte1u59fllTFuwkcrEnEs5Qd8/1pkW6S897W5aBbwK3Gmtnd1p+5OJloWIHIZte1v5w5treXzuBrbta+O80UVcdIwhJ+CjriSToC6Kk372sf8CE62IB6y1P+7qcWvtVb1elcgQt21fKw+8uY7H3l3Ptn1t1BSEuPzUKioiIUpz/FTmhzRbqwwIHxsS1tqoMeY0oMuQEJGe297Yyh/eXMdj7/xfOFw2oYqqgiBet4vaojCRsLqXZODoaVt2tjHmV8SvkG7cv9FaOy8pVYkMMTsaW3lg9joefXsDW/e1Ul0Q4rJTq6iOBHG7XZTnBSnK8mtwWgacnobEiYnbzq0JBzi9d8sRGVp2NrXx4JtrebhTOHxrQiXVBUHcLhfD8wIUZWXowjgZsHoUEtba05JdiMhQsqe5nfvfWMuf3q5n2762f4RDTWEIgOF5QUqyFQ4y8PX41AljzPnEp/L+R4dpd4PZIqnspeWbuf7pJWzZG285XH5qFTUFIWJAWW4Ak5NBukcz4cvg0NPrJH4DBIDTgN8Tn+Jb02aIdNLeEeO/Zq3gvjfXEgmnc+P5dYwoDBFzoCQnQGlOhlaJk0Gnx2MS1tqjjDGLrLU/Msb8HHg6mYWJDCZ2ZxNXPjKf+Rt2cWJVHt84qQKXK43irAzKcgMKBxm0ehoSzYnbJmNMCbAdqEhOSSKDywtLN/ODpxaxt6WDb02oZHxFLlkBL3XFmQR8uhhOBree/gt+zhiTDdwJzCN+ZtPvk1aVyCDQ1h7ljlkreGD2OgrCfq6/qI7csI/yvCDD84JaCEiGhJ6e3XRr4u5TxpjnAL+1dnfyyhIZ2NbvaOKqR+exYMNuTqnO5ysnDMfvczOqOJO8UHp/lyfSaw4aEsaYzx7kMay1GpeQlOI4DjMXb+L6Z5fQ2NrBZadWMq48l5ygj7riTI09yJDzcS2JTx/kMQcNXksKaW2PctuM5fxxTj1FmX5uOK+O3KCPikiQ4blBXOpekiHooCFhrb20rwoRGcjWbdvHVY8tYFHDbibU5POvnyonw+fiiJIscoK+/i5PJGl0MZ3IQbRHY0xfaLn1L8tpbOvg3ydWMXZ4DnmhdGqLwupekiFPF9OJdMFxHJbY3dw+YwVvrdlOSbafGy+oIzeYTlV+kLLcgLqXJCXoYjqRA2zf18rPX1jJk/MsOHDx8WWcfUQhPo+L0SVZZAfUvSSpo6ch0ZK43X8x3Q50MZ0MMa3tUR59ZwO/enUV2/a1cWJVHl8eP4yAz01+OJ2aAnUvSerpaUj8pYuL6X6XtKpE+pDjOMxZs52fPL+cpRv3MCw3wI8+XcOwvAAZPjcjCsPkBLykpal7SVJPT0NiBRC11j5ljBkFHAs8m7yyRPpGw84mbp+xnFlLNpPhdTPllEpOrM7DlQaVkRDFWRm6clpSWk9D4ofW2j8bY04GzgJ+DtwLjE9aZSJJ1NTWwdTX1vD7N9bS2NrBuaOLuOgYg8eVRnGWn+F5QXUtidDzkIgmbs8HfmOtnWaMuSU5JYkkTyzm8MLSzfx01grqtzcxqjiTr59UQV7YR1aGl+qCEJl+b3+XKTJg9DQkrDHmt8CZwM+MMemAltSSQaVhZxM/fHYJr67cSl7Qx3fPGsGRJhOvx01NQYhIOF3jDiIH6GlIfAE4F/hva+0uY0wx8L3klSXSezqiMR6cvY5fvryKptYonzu2lHNHF+Fxp1GRF8TkZOBx63ceka70dBbYJjpdF2Gt3QRsSlZRIr1l+abdXP/0EuZv2EVNYp3pnKCP4iw/FfkhMnwadxA5GK2IIkNSc1uUe155n/vfWAfAN06u4OSqPPw+NyOLMjXfkkgPKSRkSHEchzkfbOfGaUv4YGsjxw7L5tKTygmlexieF2RYbkBdSyKHQCEhQ8auxjZ+Oms5T861BNPdXHPmCI4qzSSc7mVEcVhnLYkcBoWEDHqxmMPMJZu49fnlbN7dwmm1Eb54fBl+r5uqSAiTnaHJ+EQOU9JCwhhzP3ABsMVaOzqx7fPALUAdMM5aO7fT/tcB3yB+TcZV1tq/Jqs2GTo27mrilunLeGHZhxSE0/nh+XVURkLkBn3UFIYI+PR7kMgnkcz/QQ8AvwIe6rRtCfBZ4Ledd0xM9XEx8fUqSoCXjDEjrLVRRLrgOA7Pzrf8+Lll7GpuZ/KYEi4YU4zf66a2MKxrHkR6SdJCwlr7ujGm/IBtyyG+PvYBJgOPWWtbgbXGmNXAOOCtZNUng9ee5nZumraEZxdspDQng++fM5KCzHRKsjOojARJ9+i0VpHeMlDa4gaY0+nnhsQ2kX8yd+0OvvPnBazf0cz5RxbzmaNLCPo91Om0VpGkGCgh0VW/gNPVjsaYKcCUxP1k1iQDSEc0xj0vr+Le1z4gw+vm+vNGUl0QoiQ7g6pICK9OaxVJioESEg1AWaefS4GNXe1orZ0KTAWYNGlSl0EiQ8uGHY18+/EFzKvfxXHDcvjmyRUE/W7qijOJhP0f/wIictgGSkhMBx4xxvyC+MB1DVpDO+U5jsMz8yy3/GUpze1RvnlyBSdW5ZEbSqe2MKwpNUT6QDJPgX0UmAjkG2MagJuJL3v6P0AEeN4Ys8Bae461dqkx5glgGdABXKEzm1LbnuZ2bnx2CdMXbmR4boCbz6ghN+ilKhKiNCeg6x5E+kgyz266pJuHnulm/9uA25JVjwwec9ft4JrHF7BhZzMXjinmwjGGULqbupIssjJ01bRIXxoo3U0itHfEuOeVVdz7tw8Ipnu48fw6KvODFGdnUF2gwWmR/qCQkAFh4Yad3PjsUhbb3RxfnsM3T6rE73NRV5xJQaYGp0X6i0JC+tWOxlbumLmCp+ZZfG4Xl02oZHxlLlkZXuqKszQ4LdLPFBLSLzqiMR55Zz2/fGkVOxrbOK02wsXHl+F1u6jMD1GWq8FpkYFAISF97r36ndwyPd61VJ4X5LtnjaAkx0843UtNYViD0yIDiEJC+szWvS3cMXMFzy7YSLrHxZRTKjmhKg+PO42aSIiCTL9aDyIDjEJCkq69I8qf5qznnldWsbOpnTPrCvj8cWV43WmU5mQwLC+gSflEBiiFhCSN4zi8vXY7P/7LcpZt2kNlJMj3zxlJYZafrAwPIwrDhLVanMiAppCQpNi2t5WfPL+MvyzcRIbPxWWnVjK+PBev18WIAq33IDJYKCSkVzmOw/QFlh8/t5wdjW2cPaqQzx5Xis/toiw3wLDcAD6PLooTGSwUEtJrtu5t4YZnlvDCsg8x2Rnc9pkjKc72kxv0UVUQIpSuf24ig43+18on5jgO0xKth51Nbfy/Yw3njS4m3eeitjBMfkhdSyKDlUJCPpEP97RwwzOLeWn5FspyMvje2aMpyPRjcjKozA+pa0lkkFNIyGF76r0GfvL8MnY3t/P540qZNLqIdK+bkUVh8kLp/V2eiPQChYQcsg/3NHPd04t5ZcVWhuUGuHbSSPJD6ZSo9SAy5Cgk5JA8OXcDtz6/nH0tHXxxbCnnHBFvPdQVZ5Ib9PV3eSLSyxQS0iObd7dw7VOL+Nv7WynPC3DD+XXkBX1qPYgMcQoJ+VjTF1hunLaExpYolxxfxlmjCtV6EEkRCgnpVmNrB9c/s5hpC+LrTN90fjW5IR8l2Wo9iKQKhYR0aV79Tq56bD4NO5u5cEwJk8eU4Pep9SCSahQS8k+i0Rj3vLKaX7+6mlC6hx+eX0dFYp3pqohaDyKpRiEh/9Cws4mrHp3PvPW7GDs8h2+eXEEg3c3IokxNyCeSohQSAsC0+fHB6Zb2KN+aUMmnKnPJDvgYWZSpdaZFUphCIsU1tnZw3dOLmb5wI+V5Aa46vYbsoJeqSIjSHK0zLZLqFBIpbG79Dq5+bME/BqcvHFNMyO+lrjhT60yLCKCQSEkdHTHueWUV//u3Dwile7jpglGU5wcoSQxOe90anBaROIVECnEch7nrdvLj55ax2O5mXHkul55YTsjvYWRxmEjY398lisgAo5BIEau37OWOmSt4ecUWMrxuLptQybjKHPKCfmqLwvi9GpwWkY9SSAxxW/e2cNeL7/PkPEs05vDpo0q44MhifF4XVZEQJjtDg9Mi0i2FxBC1r6Wd3/99Lfe/uZY9LR2cXJ3HF8YOI+T3kB/yURnRcqIi8vH0LTHEtHVEeXpeA/e8vJqNu1uoKw5z3afKKcpOJ+TzUF0QJkfTaohIDykkhohozOH1lVu584UVLNu0l5IsPz84ZyQji0P4PG5qCkLkh9LVtSQihyRpIWGMuR+4ANhirR2d2JYLPA6UA+uAL1hrdxpj0oC7gfOAJuBr1tp5yaptKHEch8V2Fz+btZI3V28n0+9hyimVnFCdi4s0KvKDlGRn4NFprSJyGJLZkngA+BXwUKdt1wIvW2vvMMZcm/j5B8AkoCbxZzxwb+JWDmLr3hZ+Nmslz863uNLgc8eWcu7oQtwuF6XZGZTlBnTWkoh8IkkLCWvt68aY8gM2TwYmJu4/CPyNeEhMBh6y1jrAHGNMtjGm2Fq7KVn1DWbtHVEef3cDd720iu2NbUwcEeHzY0vJ8LkpCKdTkR8iqEFpEekFff1NUrj/i99au8kYU5DYboANnfZrSGz7SEgYY6YAUxL3k1vtALRowy5unr6U+Rt2MSw3wNVn1lCaEyDs91BTECYroOk0RKT3DJRfN7saTXW62tFaOxWYCjBp0qQu9xmKdjW18V+zVvDE3AY87jS+flI5J9fk43O7GFEY1lTeIpIUfR0SH+7vRjLGFANbEtsbgLJO+5UCG/u4tgEpFnN4al4Dd/51JVv2tjKhJp9Lxg0j3eOiLCfA8LygFgISkaTp65CYDnx4BoQWAAAMSElEQVQVuCNxO63T9iuNMY8RH7DerfEIWLZpDzdPW8K763ZisjP40aePYFh+BmG/lxGFYTL96loSkeRK5imwjxIfpM43xjQANxMPhyeMMd8A1gOfT+w+g/jpr6uJnwJ7abLqGgz2tbTz3y+8zyNvryctDb56wnAmjojgcbuoKQhRkOnX9Q4i0ieSeXbTJd08dEYX+zrAFcmqZbBwHIfnFm3kthkr2Ly7hROr8vjy+GFk+DyUZPspzw+S7tEprSLSdwbKwHXKq9/eyA3PLOGN1dsozvJz0/l1VERChNM91BSFtQiQiPQLhUQ/a++Ice/fVnPva2voiMX40rgyzhpViNuVRnUkTFGWupZEpP8oJPrRO2u3c8MzS1i1ZR9Hmiy+eXIF4QwPRVkZVOQHdbW0iPQ7hUQ/2NnUyu3Pr+CpeQ0E0z18+4waji7LIsPnYWRRmOyAZmkVkYFBIdGHHMfhmXkN3D5zBdv2tXFWXQGfO64Un8dNVSSIyQngVteSiAwgCok+smbrPm58dgmzP9hOaU4Gt04eTWmOn7xQOtUFIQI+fRQiMvDomynJWtuj/Ppvq/nd62vpiMX48vhhnF4bId3nprYwTH5I02mIyMClkEiiOWu2c8Mzi/lgayNjSrP4+kkVhP0ehuUFGZYb0HQaIjLgKSSSoLmtg5/OXMHDc9YT8nu45owajizNIjPDy4giTachIoOHQqKXzV+/k+8+sZA12xo5dUSEL40bht/rorogTLGueRCRQUYh0UvaO2Lc9dL7TH19DRk+N/959ghGFWdSlOWnMhLSNQ8iMigpJHrBik17uOaJBSzftJdxFbl87YRywn4PtUVa50FEBjeFxCcQi8X4zWtruPvlVbjS0viP06s5piybSGY6NQVhtR5EZNBTSBym+u2NXP34Auav38WY0iy+cVIFWQEvtYVhCrP8aj2IyJCgkDhEsViMP86p52ezVhKNOUw5pYJxlblEQn5GFIbJ8Kn1ICJDh0LiEGza3cx3n1jI7A+2U1sU5rIJVeQEvdQUhCnJVutBRIYehUQPPTl3Az9+bhlNbVG+csJwTq2JkB30UlecqSk1RGTI0rfbx9i0u5lrn1rMa+9vpSI/yA8nVhEJp1MVCWGyM3Tdg4gMaQqJbjiOw+PvbuD2GctpbIvypXHDOG1khJwMHyNLMgml669ORIY+fdN1YeOuZn7w5CL+vnoblYnWQ34oncr8IKW5ms5bRFKHQqITx3F49O31/HTWCpraonx5fLz1kO33UVscJqw5l0QkxSgkEuzOJr735CJmf7CdqkiQy06tIi/koyo/pNaDiKSslA+JWCzGw2+v52ezVtLSHuVfPzWcibURMv0eaoszNWOriKS0lA6J9Tsa+cGTi3lrzXZqCkJcdmolOUEfFflBhuUG1XoQkZSXkiHhOA6PvrOe22Ysp60jxldPGM6EEWo9iIgcKCVD4s9zG7j+mSVURYL8+8RqcoNeytV6EBH5iJQMiU+PKWHz7maqIiEimemMVOtBRKRLKRkSGT43F48bRmNbh1oPIiIHkZIhAVCQ6e/vEkREBjxXfxcgIiIDl0JCRES61S/dTcaYbwP/BqQBv7PW/tIYkws8DpQD64AvWGt39kd9IiIS1+ctCWPMaOIBMQ4YA1xgjKkBrgVettbWAC8nfhYRkX7UH91NdcAca22TtbYDeA24CJgMPJjY50HgM/1Qm4iIdNIfIbEEmGCMyTPGBIDzgDKg0Fq7CSBxW9APtYmISCd9PiZhrV1ujPkZ8CKwD1gIdPT0+caYKcCUxP2k1CgiInH9cnaTtfY+a+2x1toJwA5gFfChMaYYIHG7pZvnTrXWjrXWjs3Ly+u7okVEUlB/nd1UYK3dYowZBnwWOAGoAL4K3JG4nfZxr7No0aJtxpj6wywjH9h2mM8dClL5+FP52CG1j1/HHje8p09KcxwnOeUchDHm70Ae0A58x1r7sjEmD3gCGAasBz5vrd2RxBrmWmvHJuv1B7pUPv5UPnZI7ePXsR/6sfdLS8Jae0oX27YDZ/RDOSIi0g1dcS0iIt1K5ZCY2t8F9LNUPv5UPnZI7ePXsR+ifhmTEBGRwSGVWxIiIvIxUnI9CWPMucDdgBv4vbX2jn4uqc8YY9YBe4Eo0DHUz/QwxtwPXABssdaOTmxLickkuzn2W4jPnbY1sdv11toZ/VNh8hhjyoCHgCIgBky11t6dQp99d8d/C4f4+adcS8IY4wZ+DUwCRgGXGGNG9W9Vfe40a+3RQz0gEh4Azj1gW6pMJvkAHz12gLsSn//RQzEgEjqA71pr64BPAVck/p+nymff3fHDIX7+KRcSxGefXW2tXWOtbQMeIz65oAxB1trXiV/V31lKTCbZzbGnBGvtJmvtvMT9vcBywJA6n313x3/IUjEkDLCh088NHOZf3iDlAC8YY95LzIOVilJ9MskrjTGLjDH3G2Ny+ruYZDPGlAPHAG+Tgp/9AccPh/j5p2JIpHWxLZVO8TrJWnss8e62K4wxE/q7IOlT9wJVwNHAJuDn/VtOchljQsBTwNXW2j39XU9f6+L4D/nzT8WQaCA+Nfl+pcDGfqqlz1lrNyZutwDPEO9+SzU9mkxyKLLWfmitjVprY8DvGMKfvzHGS/wL8mFr7dOJzSnz2Xd1/Ifz+adiSLwL1BhjKowxPuBiYHo/19QnjDFBY0x4/33gbOLre6Sa6cQnkYQeTiY5VOz/gky4iCH6+Rtj0oD7gOXW2l90eiglPvvujv9wPv+UvJjOGHMe8Evip8Deb629rZ9L6hPGmErirQeIn/78yFA/dmPMo8BE4jNgfgjcDDxLH04m2V+6OfaJxLsaHOKngH5rfx/9UGKMORn4O7CY+CmgANcT75dPhc++u+O/hEP8/FMyJEREpGdSsbtJRER6SCEhIiLdUkiIiEi3FBIiItIthYSIiHQrJWeBFempLmbNBJhord2VxPecAXwpme8h0lMKCRHiswNba6PdPHyXtfa/+6CGNCDNWntest9LpKcUEjLoGGNuBbZZa+9O/Hwb8YvF0oEvJG6fsdbenHj8WeJTsfiBu621UxPb9wG/AM4BvmuMuQC4kPg0yy9Ya//zIDV8Bxhtrf26MeZI4FHiUxx8n/jcOCbxnv9lrf1d4jnfO7C+xORrM4FXgROAzxhjXgPGWmu3GWP+BbgK8BG/EOzfrbXRRO13E18vohmYbK390BhTCPwGqEyUerm1dnZ3r3PIf/mScjQmIYPRfSSmVjDGuIhPrfIhUEP8i/po4LhOkxd+3Vp7HDAWuMoYk5fYHgSWWGvHA8uIT1NwhLX2KOAnnd7vGmPMgsSfVxPbfglUG2MuAv5A/MrVpsRjRwHnE//Sv8kYU2KMOfsg9dUCD1lrj7HW1u9/U2NMHfBF4pMyHk18oagvd6p9jrV2DPA68S4xgHuA1xLbjwWWfszriByUWhIy6Fhr1xljthtjjgEKgfnA8cTnopqf2C1E/Ev5deLBcFFie1li+3biX5ZPJbbvAVqA3xtjngee6/SWH+lustbGjDFfAxYBv7XWvtnp4WnW2magOREq44CTu6lvPVBvrZ3TxaGeARwHvGuMAcjg/yaka+tU43vAWYn7pwNfSdQYBXYbY/71IK8jclAKCRmsfg98jfjyjPcT/0L9qbX2t513MsZMBM4ETrDWNhlj/ka82wmgZX+Xi7W2wxgzLvE6FwNXEv/CPZgaYB9QcsD2A+e6cYhPUd9VfeVAYzevnwY8aK29rovH2q21+98nysH/Lx/sdUQOSt1NMlg9Q3xpzuOBvyb+fD0xfz4mrgDIAnYmAmIk8aUcPyLxvKzEco5XE+8S6pYxJov4mMAEIM8Y87lOD082xvgT3VoTic883F19B/My8Ln9+xljco0xw3vwnMsT+7uNMZmH+ToigFoSMkhZa9sSXTm7Eq2BFxJ9728lulT2Af8CzAIuM8YsAlYCXXXrAISBacYYP/HfvK/p9Ng1iYHf/T4D3AT8r7X2fWPMN4BXjTGvJx5/B3ie+EyjtybW8NjYTX3dDh5ba5cZY25MHJsLaAeuAOq7ew7wbWBqoqYo8YHrtw7jdUQAzQIrg1Tiy24e8ameV/V3PfslrqvY1xenzIr0BXU3yaBjjBkFrAZeHkgBITIUqSUhIiLdUktCRES6pZAQEZFuKSRERKRbCgkREemWQkJERLqlkBARkW79f986B04mMjqEAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure()\n",
    "sns.lineplot(x ='yearsExperience', y = 'salary', data = Train_add_salary)\n",
    "plt.figure()\n",
    "sns.lineplot(x = 'milesFromMetropolis', y ='salary', data = Train_add_salary)\n",
    "#plt.grid()\n",
    "plt.show\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "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>companyId</th>\n",
       "      <th>jobType</th>\n",
       "      <th>degree</th>\n",
       "      <th>major</th>\n",
       "      <th>industry</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>984078</td>\n",
       "      <td>984078</td>\n",
       "      <td>984078</td>\n",
       "      <td>984078</td>\n",
       "      <td>984078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>63</td>\n",
       "      <td>8</td>\n",
       "      <td>5</td>\n",
       "      <td>9</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>COMP39</td>\n",
       "      <td>SENIOR</td>\n",
       "      <td>HIGH_SCHOOL</td>\n",
       "      <td>NONE</td>\n",
       "      <td>WEB</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>15933</td>\n",
       "      <td>124528</td>\n",
       "      <td>229819</td>\n",
       "      <td>517620</td>\n",
       "      <td>140904</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       companyId jobType       degree   major industry\n",
       "count     984078  984078       984078  984078   984078\n",
       "unique        63       8            5       9        7\n",
       "top       COMP39  SENIOR  HIGH_SCHOOL    NONE      WEB\n",
       "freq       15933  124528       229819  517620   140904"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let us summarize categorical variables \n",
    "\n",
    "Train_add_salary[Text_Features].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>companyId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>COMP53</th>\n",
       "      <td>15540.0</td>\n",
       "      <td>115.669305</td>\n",
       "      <td>38.624364</td>\n",
       "      <td>22.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>278.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP37</th>\n",
       "      <td>15658.0</td>\n",
       "      <td>115.716567</td>\n",
       "      <td>38.578881</td>\n",
       "      <td>20.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>113.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>285.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP57</th>\n",
       "      <td>15405.0</td>\n",
       "      <td>115.737163</td>\n",
       "      <td>38.407085</td>\n",
       "      <td>19.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>263.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP42</th>\n",
       "      <td>15659.0</td>\n",
       "      <td>115.833003</td>\n",
       "      <td>38.720676</td>\n",
       "      <td>19.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>113.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>281.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP31</th>\n",
       "      <td>15499.0</td>\n",
       "      <td>115.897090</td>\n",
       "      <td>38.432014</td>\n",
       "      <td>24.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>140.00</td>\n",
       "      <td>283.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP54</th>\n",
       "      <td>15712.0</td>\n",
       "      <td>115.953921</td>\n",
       "      <td>38.298060</td>\n",
       "      <td>22.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>140.00</td>\n",
       "      <td>274.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP18</th>\n",
       "      <td>15695.0</td>\n",
       "      <td>115.966805</td>\n",
       "      <td>38.371254</td>\n",
       "      <td>21.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>292.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP45</th>\n",
       "      <td>15608.0</td>\n",
       "      <td>115.978152</td>\n",
       "      <td>38.671857</td>\n",
       "      <td>20.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>301.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP30</th>\n",
       "      <td>15784.0</td>\n",
       "      <td>115.991067</td>\n",
       "      <td>38.927359</td>\n",
       "      <td>20.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>282.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP55</th>\n",
       "      <td>15488.0</td>\n",
       "      <td>115.996191</td>\n",
       "      <td>38.799304</td>\n",
       "      <td>20.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>283.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP19</th>\n",
       "      <td>15702.0</td>\n",
       "      <td>116.026939</td>\n",
       "      <td>38.592960</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>273.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP43</th>\n",
       "      <td>15694.0</td>\n",
       "      <td>116.060087</td>\n",
       "      <td>38.742795</td>\n",
       "      <td>20.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>289.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP35</th>\n",
       "      <td>15849.0</td>\n",
       "      <td>116.107830</td>\n",
       "      <td>38.744032</td>\n",
       "      <td>21.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>293.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP15</th>\n",
       "      <td>15649.0</td>\n",
       "      <td>116.141031</td>\n",
       "      <td>38.462651</td>\n",
       "      <td>21.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>273.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP48</th>\n",
       "      <td>15672.0</td>\n",
       "      <td>116.143441</td>\n",
       "      <td>38.505668</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>283.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP0</th>\n",
       "      <td>15462.0</td>\n",
       "      <td>116.151468</td>\n",
       "      <td>38.518017</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>285.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP50</th>\n",
       "      <td>15556.0</td>\n",
       "      <td>116.170738</td>\n",
       "      <td>38.641241</td>\n",
       "      <td>21.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>266.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP7</th>\n",
       "      <td>15538.0</td>\n",
       "      <td>116.185996</td>\n",
       "      <td>38.509411</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>275.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP27</th>\n",
       "      <td>15516.0</td>\n",
       "      <td>116.211459</td>\n",
       "      <td>38.821440</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>273.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP59</th>\n",
       "      <td>15807.0</td>\n",
       "      <td>116.214778</td>\n",
       "      <td>38.731666</td>\n",
       "      <td>21.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>288.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP26</th>\n",
       "      <td>15512.0</td>\n",
       "      <td>116.237687</td>\n",
       "      <td>38.441005</td>\n",
       "      <td>18.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>270.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP20</th>\n",
       "      <td>15657.0</td>\n",
       "      <td>116.242511</td>\n",
       "      <td>38.636869</td>\n",
       "      <td>19.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>275.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP47</th>\n",
       "      <td>15552.0</td>\n",
       "      <td>116.260995</td>\n",
       "      <td>38.597732</td>\n",
       "      <td>21.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>301.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP28</th>\n",
       "      <td>15541.0</td>\n",
       "      <td>116.262017</td>\n",
       "      <td>38.796515</td>\n",
       "      <td>19.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>273.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP8</th>\n",
       "      <td>15399.0</td>\n",
       "      <td>116.271381</td>\n",
       "      <td>38.694449</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>277.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP5</th>\n",
       "      <td>15536.0</td>\n",
       "      <td>116.287654</td>\n",
       "      <td>38.465242</td>\n",
       "      <td>23.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>290.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP10</th>\n",
       "      <td>15662.0</td>\n",
       "      <td>116.293002</td>\n",
       "      <td>38.741160</td>\n",
       "      <td>24.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>278.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP32</th>\n",
       "      <td>15650.0</td>\n",
       "      <td>116.357891</td>\n",
       "      <td>38.606844</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>282.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP46</th>\n",
       "      <td>15715.0</td>\n",
       "      <td>116.378683</td>\n",
       "      <td>38.438527</td>\n",
       "      <td>18.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>284.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP33</th>\n",
       "      <td>15530.0</td>\n",
       "      <td>116.391500</td>\n",
       "      <td>38.757705</td>\n",
       "      <td>22.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>268.0</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP24</th>\n",
       "      <td>15540.0</td>\n",
       "      <td>116.453797</td>\n",
       "      <td>38.664120</td>\n",
       "      <td>21.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>285.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP6</th>\n",
       "      <td>15615.0</td>\n",
       "      <td>116.459558</td>\n",
       "      <td>38.979509</td>\n",
       "      <td>20.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>279.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP12</th>\n",
       "      <td>15536.0</td>\n",
       "      <td>116.531411</td>\n",
       "      <td>38.611113</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>281.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP36</th>\n",
       "      <td>15616.0</td>\n",
       "      <td>116.536117</td>\n",
       "      <td>38.681094</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>284.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP3</th>\n",
       "      <td>15769.0</td>\n",
       "      <td>116.537383</td>\n",
       "      <td>38.418483</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>294.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP22</th>\n",
       "      <td>15655.0</td>\n",
       "      <td>116.545129</td>\n",
       "      <td>38.851992</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>287.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP51</th>\n",
       "      <td>15741.0</td>\n",
       "      <td>116.550346</td>\n",
       "      <td>38.674928</td>\n",
       "      <td>21.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>278.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP52</th>\n",
       "      <td>15499.0</td>\n",
       "      <td>116.558617</td>\n",
       "      <td>38.557005</td>\n",
       "      <td>19.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>271.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP4</th>\n",
       "      <td>15698.0</td>\n",
       "      <td>116.565741</td>\n",
       "      <td>38.873825</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>278.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP56</th>\n",
       "      <td>15716.0</td>\n",
       "      <td>116.567065</td>\n",
       "      <td>39.112377</td>\n",
       "      <td>21.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>289.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP38</th>\n",
       "      <td>15601.0</td>\n",
       "      <td>116.569835</td>\n",
       "      <td>38.767127</td>\n",
       "      <td>19.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>283.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP23</th>\n",
       "      <td>15677.0</td>\n",
       "      <td>116.602539</td>\n",
       "      <td>38.539240</td>\n",
       "      <td>19.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>279.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP16</th>\n",
       "      <td>15578.0</td>\n",
       "      <td>116.603800</td>\n",
       "      <td>38.745942</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>274.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP61</th>\n",
       "      <td>15558.0</td>\n",
       "      <td>116.609911</td>\n",
       "      <td>38.502785</td>\n",
       "      <td>19.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>275.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP17</th>\n",
       "      <td>15485.0</td>\n",
       "      <td>116.610591</td>\n",
       "      <td>39.166633</td>\n",
       "      <td>22.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>289.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP25</th>\n",
       "      <td>15649.0</td>\n",
       "      <td>116.641063</td>\n",
       "      <td>38.246480</td>\n",
       "      <td>20.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>267.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP14</th>\n",
       "      <td>15644.0</td>\n",
       "      <td>116.663833</td>\n",
       "      <td>38.607292</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>283.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP62</th>\n",
       "      <td>15680.0</td>\n",
       "      <td>116.672513</td>\n",
       "      <td>39.118087</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>286.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP49</th>\n",
       "      <td>15562.0</td>\n",
       "      <td>116.680118</td>\n",
       "      <td>38.812663</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>141.75</td>\n",
       "      <td>271.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP60</th>\n",
       "      <td>15547.0</td>\n",
       "      <td>116.691838</td>\n",
       "      <td>38.131621</td>\n",
       "      <td>20.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>141.00</td>\n",
       "      <td>298.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP29</th>\n",
       "      <td>15532.0</td>\n",
       "      <td>116.731264</td>\n",
       "      <td>38.528423</td>\n",
       "      <td>22.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>284.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP34</th>\n",
       "      <td>15561.0</td>\n",
       "      <td>116.759527</td>\n",
       "      <td>38.865911</td>\n",
       "      <td>23.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>272.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP58</th>\n",
       "      <td>15696.0</td>\n",
       "      <td>116.816896</td>\n",
       "      <td>38.771087</td>\n",
       "      <td>23.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>274.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP21</th>\n",
       "      <td>15729.0</td>\n",
       "      <td>116.818107</td>\n",
       "      <td>38.733401</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>287.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP9</th>\n",
       "      <td>15711.0</td>\n",
       "      <td>116.848959</td>\n",
       "      <td>38.779373</td>\n",
       "      <td>20.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>264.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP41</th>\n",
       "      <td>15574.0</td>\n",
       "      <td>116.910107</td>\n",
       "      <td>38.978871</td>\n",
       "      <td>21.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>114.0</td>\n",
       "      <td>143.00</td>\n",
       "      <td>268.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP44</th>\n",
       "      <td>15724.0</td>\n",
       "      <td>116.972144</td>\n",
       "      <td>38.742389</td>\n",
       "      <td>20.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>294.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP13</th>\n",
       "      <td>15689.0</td>\n",
       "      <td>117.001976</td>\n",
       "      <td>38.795415</td>\n",
       "      <td>21.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>273.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP2</th>\n",
       "      <td>15401.0</td>\n",
       "      <td>117.011623</td>\n",
       "      <td>38.563394</td>\n",
       "      <td>20.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>268.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COMP40</th>\n",
       "      <td>15747.0</td>\n",
       "      <td>117.077983</td>\n",
       "      <td>38.613580</td>\n",
       "      <td>21.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>142.00</td>\n",
       "      <td>272.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>63 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             count        mean        std   min   25%    50%     75%    max\n",
       "companyId                                                                  \n",
       "COMP53     15540.0  115.669305  38.624364  22.0  88.0  114.0  141.00  278.0\n",
       "COMP37     15658.0  115.716567  38.578881  20.0  88.0  113.0  141.00  285.0\n",
       "COMP57     15405.0  115.737163  38.407085  19.0  88.0  114.0  141.00  263.0\n",
       "COMP42     15659.0  115.833003  38.720676  19.0  88.0  113.0  141.00  281.0\n",
       "COMP31     15499.0  115.897090  38.432014  24.0  89.0  114.0  140.00  283.0\n",
       "COMP54     15712.0  115.953921  38.298060  22.0  89.0  114.0  140.00  274.0\n",
       "COMP18     15695.0  115.966805  38.371254  21.0  89.0  114.0  141.00  292.0\n",
       "COMP45     15608.0  115.978152  38.671857  20.0  88.0  114.0  141.00  301.0\n",
       "COMP30     15784.0  115.991067  38.927359  20.0  88.0  114.0  141.00  282.0\n",
       "COMP55     15488.0  115.996191  38.799304  20.0  88.0  114.0  141.00  283.0\n",
       "COMP19     15702.0  116.026939  38.592960  20.0  89.0  114.0  141.00  273.0\n",
       "COMP43     15694.0  116.060087  38.742795  20.0  88.0  114.0  141.00  289.0\n",
       "COMP35     15849.0  116.107830  38.744032  21.0  88.0  114.0  141.00  293.0\n",
       "COMP15     15649.0  116.141031  38.462651  21.0  89.0  114.0  141.00  273.0\n",
       "COMP48     15672.0  116.143441  38.505668  20.0  89.0  114.0  141.00  283.0\n",
       "COMP0      15462.0  116.151468  38.518017  20.0  89.0  114.0  141.00  285.0\n",
       "COMP50     15556.0  116.170738  38.641241  21.0  89.0  114.0  141.00  266.0\n",
       "COMP7      15538.0  116.185996  38.509411  20.0  89.0  114.0  141.00  275.0\n",
       "COMP27     15516.0  116.211459  38.821440  20.0  89.0  114.0  142.00  273.0\n",
       "COMP59     15807.0  116.214778  38.731666  21.0  89.0  114.0  141.00  288.0\n",
       "COMP26     15512.0  116.237687  38.441005  18.0  89.0  114.0  141.00  270.0\n",
       "COMP20     15657.0  116.242511  38.636869  19.0  89.0  114.0  141.00  275.0\n",
       "COMP47     15552.0  116.260995  38.597732  21.0  89.0  114.0  141.00  301.0\n",
       "COMP28     15541.0  116.262017  38.796515  19.0  89.0  114.0  142.00  273.0\n",
       "COMP8      15399.0  116.271381  38.694449  20.0  89.0  114.0  141.00  277.0\n",
       "COMP5      15536.0  116.287654  38.465242  23.0  89.0  114.0  142.00  290.0\n",
       "COMP10     15662.0  116.293002  38.741160  24.0  89.0  114.0  141.00  278.0\n",
       "COMP32     15650.0  116.357891  38.606844  20.0  89.0  114.0  142.00  282.0\n",
       "COMP46     15715.0  116.378683  38.438527  18.0  89.0  114.0  142.00  284.0\n",
       "COMP33     15530.0  116.391500  38.757705  22.0  89.0  114.0  141.00  268.0\n",
       "...            ...         ...        ...   ...   ...    ...     ...    ...\n",
       "COMP24     15540.0  116.453797  38.664120  21.0  89.0  114.0  141.00  285.0\n",
       "COMP6      15615.0  116.459558  38.979509  20.0  88.0  114.0  142.00  279.0\n",
       "COMP12     15536.0  116.531411  38.611113  20.0  89.0  114.0  141.00  281.0\n",
       "COMP36     15616.0  116.536117  38.681094  20.0  89.0  114.0  142.00  284.0\n",
       "COMP3      15769.0  116.537383  38.418483  20.0  89.0  115.0  141.00  294.0\n",
       "COMP22     15655.0  116.545129  38.851992  20.0  89.0  114.0  142.00  287.0\n",
       "COMP51     15741.0  116.550346  38.674928  21.0  89.0  114.0  142.00  278.0\n",
       "COMP52     15499.0  116.558617  38.557005  19.0  90.0  114.0  141.00  271.0\n",
       "COMP4      15698.0  116.565741  38.873825  20.0  89.0  115.0  142.00  278.0\n",
       "COMP56     15716.0  116.567065  39.112377  21.0  89.0  114.0  142.00  289.0\n",
       "COMP38     15601.0  116.569835  38.767127  19.0  89.0  115.0  141.00  283.0\n",
       "COMP23     15677.0  116.602539  38.539240  19.0  89.0  114.0  141.00  279.0\n",
       "COMP16     15578.0  116.603800  38.745942  20.0  89.0  114.0  141.00  274.0\n",
       "COMP61     15558.0  116.609911  38.502785  19.0  89.0  114.0  142.00  275.0\n",
       "COMP17     15485.0  116.610591  39.166633  22.0  89.0  114.0  142.00  289.0\n",
       "COMP25     15649.0  116.641063  38.246480  20.0  90.0  114.0  141.00  267.0\n",
       "COMP14     15644.0  116.663833  38.607292  20.0  89.0  114.0  142.00  283.0\n",
       "COMP62     15680.0  116.672513  39.118087  20.0  89.0  114.0  141.00  286.0\n",
       "COMP49     15562.0  116.680118  38.812663  20.0  89.0  115.0  141.75  271.0\n",
       "COMP60     15547.0  116.691838  38.131621  20.0  90.0  115.0  141.00  298.0\n",
       "COMP29     15532.0  116.731264  38.528423  22.0  89.0  114.0  142.00  284.0\n",
       "COMP34     15561.0  116.759527  38.865911  23.0  89.0  114.0  142.00  272.0\n",
       "COMP58     15696.0  116.816896  38.771087  23.0  89.0  114.0  142.00  274.0\n",
       "COMP21     15729.0  116.818107  38.733401  20.0  89.0  114.0  142.00  287.0\n",
       "COMP9      15711.0  116.848959  38.779373  20.0  89.0  114.0  142.00  264.0\n",
       "COMP41     15574.0  116.910107  38.978871  21.0  89.0  114.0  143.00  268.0\n",
       "COMP44     15724.0  116.972144  38.742389  20.0  90.0  115.0  142.00  294.0\n",
       "COMP13     15689.0  117.001976  38.795415  21.0  89.0  115.0  142.00  273.0\n",
       "COMP2      15401.0  117.011623  38.563394  20.0  90.0  115.0  142.00  268.0\n",
       "COMP40     15747.0  117.077983  38.613580  21.0  90.0  115.0  142.00  272.0\n",
       "\n",
       "[63 rows x 8 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Train_add_salary.groupby('companyId')['salary'].describe().sort_values('mean')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x22c3a1c0d30>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Correlation between categorical variables and salary\n",
    "\n",
    "#to_list() may not work. Hence, use tolist()\n",
    "\n",
    "Jobtype_ranking = Train_add_salary.groupby('jobType')['salary'].median().sort_values().index.tolist()\n",
    "fig_dims = (10, 5)\n",
    "\n",
    "fig, ax = plt.subplots(figsize = fig_dims)\n",
    "Jobtype_distro = sns.boxenplot(x = 'jobType', y = 'salary', color = 'b', order =Jobtype_ranking, data = Train_add_salary, ax = ax)\n",
    "\n",
    "fig_dims = (10, 5) \n",
    "\n",
    "fig, ax = plt.subplots(figsize = fig_dims)\n",
    "\n",
    "Degree_ranking = Train_add_salary.groupby('degree')['salary'].median().sort_values().index.tolist()\n",
    "sns.boxenplot(data = Train_add_salary, x = 'degree', y = 'salary', color = 'b', order= Degree_ranking, ax = ax)\n",
    "\n",
    "\n",
    "fig_dims = (10, 5) \n",
    "\n",
    "\n",
    "\n",
    "\n",
    "fig, ax = plt.subplots(figsize = fig_dims)\n",
    "\n",
    "Major_ranking = Train_add_salary.groupby('major')['salary'].median().sort_values().index.tolist()\n",
    "\n",
    "sns.boxenplot(data = Train_add_salary, x ='major', y = 'salary', color = 'b', order= Major_ranking, ax=ax)\n",
    "\n",
    "\n",
    "fig_dims = (10, 5)\n",
    "\n",
    "fig, ax = plt.subplots(figsize = fig_dims)\n",
    "\n",
    "Industry_ranking = Train_add_salary.groupby('industry')['salary'].median().sort_values().index.tolist()\n",
    "\n",
    "\n",
    "sns.boxenplot(data = Train_add_salary, x = 'industry', y = 'salary', color = 'b',order= Industry_ranking, ax = ax )\n",
    "\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>jobType</th>\n",
       "      <th>avg_salary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>JANITOR</td>\n",
       "      <td>70.83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>JUNIOR</td>\n",
       "      <td>95.42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>SENIOR</td>\n",
       "      <td>105.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>MANAGER</td>\n",
       "      <td>115.46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>VICE_PRESIDENT</td>\n",
       "      <td>125.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CFO</td>\n",
       "      <td>135.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CTO</td>\n",
       "      <td>135.58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CEO</td>\n",
       "      <td>145.41</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          jobType  avg_salary\n",
       "3         JANITOR       70.83\n",
       "4          JUNIOR       95.42\n",
       "6          SENIOR      105.59\n",
       "5         MANAGER      115.46\n",
       "7  VICE_PRESIDENT      125.47\n",
       "1             CFO      135.55\n",
       "2             CTO      135.58\n",
       "0             CEO      145.41"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Baseline_df = Train_add_salary.groupby('jobType', as_index = False).mean().sort_values('salary').round(2)\n",
    "\n",
    "Baseline_df.rename(columns = {'salary':'avg_salary'}, inplace = True)\n",
    "\n",
    "Baseline_df = Baseline_df[['jobType','avg_salary']]\n",
    "\n",
    "Baseline_df\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "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>companyId</th>\n",
       "      <th>jobType</th>\n",
       "      <th>degree</th>\n",
       "      <th>major</th>\n",
       "      <th>industry</th>\n",
       "      <th>yearsExperience</th>\n",
       "      <th>milesFromMetropolis</th>\n",
       "      <th>salary</th>\n",
       "      <th>avg_salary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>COMP37</td>\n",
       "      <td>CFO</td>\n",
       "      <td>MASTERS</td>\n",
       "      <td>MATH</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>10</td>\n",
       "      <td>83</td>\n",
       "      <td>130</td>\n",
       "      <td>135.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>COMP15</td>\n",
       "      <td>CFO</td>\n",
       "      <td>NONE</td>\n",
       "      <td>NONE</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>23</td>\n",
       "      <td>24</td>\n",
       "      <td>178</td>\n",
       "      <td>135.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>COMP27</td>\n",
       "      <td>CFO</td>\n",
       "      <td>BACHELORS</td>\n",
       "      <td>CHEMISTRY</td>\n",
       "      <td>AUTO</td>\n",
       "      <td>9</td>\n",
       "      <td>70</td>\n",
       "      <td>131</td>\n",
       "      <td>135.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>COMP0</td>\n",
       "      <td>CFO</td>\n",
       "      <td>BACHELORS</td>\n",
       "      <td>PHYSICS</td>\n",
       "      <td>HEALTH</td>\n",
       "      <td>18</td>\n",
       "      <td>32</td>\n",
       "      <td>132</td>\n",
       "      <td>135.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>COMP7</td>\n",
       "      <td>CFO</td>\n",
       "      <td>BACHELORS</td>\n",
       "      <td>COMPSCI</td>\n",
       "      <td>EDUCATION</td>\n",
       "      <td>15</td>\n",
       "      <td>25</td>\n",
       "      <td>140</td>\n",
       "      <td>135.55</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  companyId jobType     degree      major   industry  yearsExperience  \\\n",
       "0    COMP37     CFO    MASTERS       MATH     HEALTH               10   \n",
       "1    COMP15     CFO       NONE       NONE     HEALTH               23   \n",
       "2    COMP27     CFO  BACHELORS  CHEMISTRY       AUTO                9   \n",
       "3     COMP0     CFO  BACHELORS    PHYSICS     HEALTH               18   \n",
       "4     COMP7     CFO  BACHELORS    COMPSCI  EDUCATION               15   \n",
       "\n",
       "   milesFromMetropolis  salary  avg_salary  \n",
       "0                   83     130      135.55  \n",
       "1                   24     178      135.55  \n",
       "2                   70     131      135.55  \n",
       "3                   32     132      135.55  \n",
       "4                   25     140      135.55  "
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Baseline_df = pd.merge(Train_add_salary, Baseline_df, on = 'jobType')\n",
    "\n",
    "Baseline_df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "966.0"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "mean_squared_error(Baseline_df['salary'], Baseline_df ['avg_salary']).round(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "Train_add_salary['jobType_num'] = Train_add_salary['jobType'].apply(lambda x :  Jobtype_ranking.index(x))                                                                                                \n",
    "\n",
    "Train_add_salary['degree_num'] = Train_add_salary['degree'].apply(lambda x :  Degree_ranking.index(x))  \n",
    "\n",
    "Train_add_salary['major_num'] = Train_add_salary['major'].apply(lambda x :  Major_ranking.index(x))  \n",
    "\n",
    "Train_add_salary['industry_num'] = Train_add_salary['industry'].apply(lambda x : Industry_ranking.index(x))  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Let us separate target from predicators \n",
    "\n",
    "y = Train_add_salary.salary\n",
    "\n",
    "\n",
    "Train_add_salary.drop(['salary'], axis = 1, inplace = True)\n",
    "\n",
    "#Let us select numerical columns only\n",
    "\n",
    "Numerical_cols = [cname for cname in Train_add_salary.columns if Train_add_salary[cname].dtype in ['int64', 'float64']]\n",
    "\n",
    "X = Train_add_salary[Numerical_cols].copy()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\user\\Anaconda3\\lib\\site-packages\\sklearn\\ensemble\\weight_boosting.py:29: DeprecationWarning: numpy.core.umath_tests is an internal NumPy module and should not be imported. It will be removed in a future NumPy release.\n",
      "  from numpy.core.umath_tests import inner1d\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "RFG_model = RandomForestRegressor(n_estimators = 50, n_jobs = -1, random_state = 0)\n",
    "\n",
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "\n",
    "KNR_model = KNeighborsRegressor(n_jobs=-1)\n",
    "\n",
    "#from xgboost import XGBRegressor\n",
    "\n",
    "#XG_model = XGBRegressor()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE scores for Random Forest Regression:\n",
      " [447.20043632 446.41239079 449.02204147 443.55111076 443.97782131]\n",
      "Average MSE for random Forest Regressor: 446.03276013138213\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "scores = -1 * cross_val_score(RFG_model, X, y, cv = 5, n_jobs = -1, scoring = 'neg_mean_squared_error')\n",
    "\n",
    "print('MSE scores for Random Forest Regression:\\n', scores)\n",
    "\n",
    "print('Average MSE for random Forest Regressor:', sum(scores)/5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE scores for K nearest Neighbors Regressor:\n",
      " [442.88415982 440.94390111 443.14422303 440.80718705 439.47936753]\n",
      "Average MSE for K nearest Neighbors Regressors: 441.4517677084735\n"
     ]
    }
   ],
   "source": [
    "scores = -1 * cross_val_score(KNR_model, X, y, cv = 5, n_jobs = -1, scoring = 'neg_mean_squared_error')\n",
    "\n",
    "print('MSE scores for K nearest Neighbors Regressor:\\n', scores)\n",
    "\n",
    "print('Average MSE for K nearest Neighbors Regressors:', sum(scores)/5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "#look for duplicate data, invalid data (e.g. salaries <=0), or corrupt data and remove it\n",
    "\n",
    "# Lke\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ---- 4 Explore the data (EDA) ----"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "#summarize each feature variable\n",
    "#summarize the target variable\n",
    "#look for correlation between each feature and the target\n",
    "#look for correlation between features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ---- 5 Establish a baseline ----"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "#select a reasonable metric (MSE in this case)\n",
    "#create an extremely simple model and measure its efficacy\n",
    "#e.g. use \"average salary\" for each industry as your model and then measure MSE\n",
    "#during 5-fold cross-validation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ---- 6 Hypothesize solution ----"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "#brainstorm 3 models that you think may improve results over the baseline model based\n",
    "#on your "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Brainstorm 3 models that you think may improve results over the baseline model based on your EDA and explain why they're reasonable solutions here.\n",
    "\n",
    "Also write down any new features that you think you should try adding to the model based on your EDA, e.g. interaction variables, summary statistics for each group, etc"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 3 - DEVELOP"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You will cycle through creating features, tuning models, and training/validing models (steps 7-9) until you've reached your efficacy goal\n",
    "\n",
    "#### Your metric will be MSE and your goal is:\n",
    " - <360 for entry-level data science roles\n",
    " - <320 for senior data science roles"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ---- 7 Engineer features  ----"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "#make sure that data is ready for modeling\n",
    "#create any new features needed to potentially enhance model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ---- 8 Create models ----"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "#create and tune the models that you brainstormed during part 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ---- 9 Test models ----"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "#do 5-fold cross validation on models and measure MSE"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ---- 10 Select best model  ----"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "#select the model with the lowest error as your \"prodcuction\" model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 4 - DEPLOY"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ---- 11 Automate pipeline ----"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "#write script that trains model on entire training set, saves model to disk,\n",
    "#and scores the \"test\" dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ---- 12 Deploy solution ----"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "#save your prediction to a csv file or optionally save them as a table in a SQL database\n",
    "#additionally, you want to save a visualization and summary of your prediction and feature importances\n",
    "#these visualizations and summaries will be extremely useful to business stakeholders"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ---- 13 Measure efficacy ----"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We'll skip this step since we don't have the outcomes for the test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}