{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.model_selection import train_test_split,GridSearchCV\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "import pickle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "df=pd.read_csv(\"survey.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "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>Respondent</th>\n",
       "      <th>MainBranch</th>\n",
       "      <th>Hobbyist</th>\n",
       "      <th>Age</th>\n",
       "      <th>Age1stCode</th>\n",
       "      <th>CompFreq</th>\n",
       "      <th>CompTotal</th>\n",
       "      <th>ConvertedComp</th>\n",
       "      <th>Country</th>\n",
       "      <th>CurrencyDesc</th>\n",
       "      <th>...</th>\n",
       "      <th>SurveyEase</th>\n",
       "      <th>SurveyLength</th>\n",
       "      <th>Trans</th>\n",
       "      <th>UndergradMajor</th>\n",
       "      <th>WebframeDesireNextYear</th>\n",
       "      <th>WebframeWorkedWith</th>\n",
       "      <th>WelcomeChange</th>\n",
       "      <th>WorkWeekHrs</th>\n",
       "      <th>YearsCode</th>\n",
       "      <th>YearsCodePro</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>I am a developer by profession</td>\n",
       "      <td>Yes</td>\n",
       "      <td>NaN</td>\n",
       "      <td>13</td>\n",
       "      <td>Monthly</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Germany</td>\n",
       "      <td>European Euro</td>\n",
       "      <td>...</td>\n",
       "      <td>Neither easy nor difficult</td>\n",
       "      <td>Appropriate in length</td>\n",
       "      <td>No</td>\n",
       "      <td>Computer science, computer engineering, or sof...</td>\n",
       "      <td>ASP.NET Core</td>\n",
       "      <td>ASP.NET;ASP.NET Core</td>\n",
       "      <td>Just as welcome now as I felt last year</td>\n",
       "      <td>50.0</td>\n",
       "      <td>36</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>I am a developer by profession</td>\n",
       "      <td>No</td>\n",
       "      <td>NaN</td>\n",
       "      <td>19</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>Pound sterling</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Computer science, computer engineering, or sof...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Somewhat more welcome now than last year</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>I code primarily as a hobby</td>\n",
       "      <td>Yes</td>\n",
       "      <td>NaN</td>\n",
       "      <td>15</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Russian Federation</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>Neither easy nor difficult</td>\n",
       "      <td>Appropriate in length</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Somewhat more welcome now than last year</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>I am a developer by profession</td>\n",
       "      <td>Yes</td>\n",
       "      <td>25.0</td>\n",
       "      <td>18</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Albania</td>\n",
       "      <td>Albanian lek</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>No</td>\n",
       "      <td>Computer science, computer engineering, or sof...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Somewhat less welcome now than last year</td>\n",
       "      <td>40.0</td>\n",
       "      <td>7</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>I used to be a developer by profession, but no...</td>\n",
       "      <td>Yes</td>\n",
       "      <td>31.0</td>\n",
       "      <td>16</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>United States</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>Easy</td>\n",
       "      <td>Too short</td>\n",
       "      <td>No</td>\n",
       "      <td>Computer science, computer engineering, or sof...</td>\n",
       "      <td>Django;Ruby on Rails</td>\n",
       "      <td>Ruby on Rails</td>\n",
       "      <td>Just as welcome now as I felt last year</td>\n",
       "      <td>NaN</td>\n",
       "      <td>15</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 61 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Respondent                                         MainBranch Hobbyist  \\\n",
       "0           1                     I am a developer by profession      Yes   \n",
       "1           2                     I am a developer by profession       No   \n",
       "2           3                        I code primarily as a hobby      Yes   \n",
       "3           4                     I am a developer by profession      Yes   \n",
       "4           5  I used to be a developer by profession, but no...      Yes   \n",
       "\n",
       "    Age Age1stCode CompFreq  CompTotal  ConvertedComp             Country  \\\n",
       "0   NaN         13  Monthly        NaN            NaN             Germany   \n",
       "1   NaN         19      NaN        NaN            NaN      United Kingdom   \n",
       "2   NaN         15      NaN        NaN            NaN  Russian Federation   \n",
       "3  25.0         18      NaN        NaN            NaN             Albania   \n",
       "4  31.0         16      NaN        NaN            NaN       United States   \n",
       "\n",
       "     CurrencyDesc  ...                  SurveyEase           SurveyLength  \\\n",
       "0   European Euro  ...  Neither easy nor difficult  Appropriate in length   \n",
       "1  Pound sterling  ...                         NaN                    NaN   \n",
       "2             NaN  ...  Neither easy nor difficult  Appropriate in length   \n",
       "3    Albanian lek  ...                         NaN                    NaN   \n",
       "4             NaN  ...                        Easy              Too short   \n",
       "\n",
       "  Trans                                     UndergradMajor  \\\n",
       "0    No  Computer science, computer engineering, or sof...   \n",
       "1   NaN  Computer science, computer engineering, or sof...   \n",
       "2   NaN                                                NaN   \n",
       "3    No  Computer science, computer engineering, or sof...   \n",
       "4    No  Computer science, computer engineering, or sof...   \n",
       "\n",
       "  WebframeDesireNextYear    WebframeWorkedWith  \\\n",
       "0           ASP.NET Core  ASP.NET;ASP.NET Core   \n",
       "1                    NaN                   NaN   \n",
       "2                    NaN                   NaN   \n",
       "3                    NaN                   NaN   \n",
       "4   Django;Ruby on Rails         Ruby on Rails   \n",
       "\n",
       "                              WelcomeChange WorkWeekHrs YearsCode YearsCodePro  \n",
       "0   Just as welcome now as I felt last year        50.0        36           27  \n",
       "1  Somewhat more welcome now than last year         NaN         7            4  \n",
       "2  Somewhat more welcome now than last year         NaN         4          NaN  \n",
       "3  Somewhat less welcome now than last year        40.0         7            4  \n",
       "4   Just as welcome now as I felt last year         NaN        15            8  \n",
       "\n",
       "[5 rows x 61 columns]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "column_needed=df.loc[:,[\"Country\",\"EdLevel\",\"YearsCodePro\",\"Employment\",\"ConvertedComp\"]]\n",
    "column_needed=column_needed.rename({\"ConvertedComp\":\"Salary\"},axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "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>Country</th>\n",
       "      <th>EdLevel</th>\n",
       "      <th>YearsCodePro</th>\n",
       "      <th>Employment</th>\n",
       "      <th>Salary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Germany</td>\n",
       "      <td>Master’s degree (M.A., M.S., M.Eng., MBA, etc.)</td>\n",
       "      <td>27</td>\n",
       "      <td>Independent contractor, freelancer, or self-em...</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>Bachelor’s degree (B.A., B.S., B.Eng., etc.)</td>\n",
       "      <td>4</td>\n",
       "      <td>Employed full-time</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Russian Federation</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Master’s degree (M.A., M.S., M.Eng., MBA, etc.)</td>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>United States</td>\n",
       "      <td>Bachelor’s degree (B.A., B.S., B.Eng., etc.)</td>\n",
       "      <td>8</td>\n",
       "      <td>Employed full-time</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Country                                          EdLevel  \\\n",
       "0             Germany  Master’s degree (M.A., M.S., M.Eng., MBA, etc.)   \n",
       "1      United Kingdom     Bachelor’s degree (B.A., B.S., B.Eng., etc.)   \n",
       "2  Russian Federation                                              NaN   \n",
       "3             Albania  Master’s degree (M.A., M.S., M.Eng., MBA, etc.)   \n",
       "4       United States     Bachelor’s degree (B.A., B.S., B.Eng., etc.)   \n",
       "\n",
       "  YearsCodePro                                         Employment  Salary  \n",
       "0           27  Independent contractor, freelancer, or self-em...     NaN  \n",
       "1            4                                 Employed full-time     NaN  \n",
       "2          NaN                                                NaN     NaN  \n",
       "3            4                                                NaN     NaN  \n",
       "4            8                                 Employed full-time     NaN  "
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "column_needed.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "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>Country</th>\n",
       "      <th>EdLevel</th>\n",
       "      <th>YearsCodePro</th>\n",
       "      <th>Employment</th>\n",
       "      <th>Salary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Germany</td>\n",
       "      <td>Master’s degree (M.A., M.S., M.Eng., MBA, etc.)</td>\n",
       "      <td>27</td>\n",
       "      <td>Independent contractor, freelancer, or self-em...</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>Bachelor’s degree (B.A., B.S., B.Eng., etc.)</td>\n",
       "      <td>4</td>\n",
       "      <td>Employed full-time</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Russian Federation</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Albania</td>\n",
       "      <td>Master’s degree (M.A., M.S., M.Eng., MBA, etc.)</td>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>United States</td>\n",
       "      <td>Bachelor’s degree (B.A., B.S., B.Eng., etc.)</td>\n",
       "      <td>8</td>\n",
       "      <td>Employed full-time</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Country                                          EdLevel  \\\n",
       "0             Germany  Master’s degree (M.A., M.S., M.Eng., MBA, etc.)   \n",
       "1      United Kingdom     Bachelor’s degree (B.A., B.S., B.Eng., etc.)   \n",
       "2  Russian Federation                                              NaN   \n",
       "3             Albania  Master’s degree (M.A., M.S., M.Eng., MBA, etc.)   \n",
       "4       United States     Bachelor’s degree (B.A., B.S., B.Eng., etc.)   \n",
       "\n",
       "  YearsCodePro                                         Employment  Salary  \n",
       "0           27  Independent contractor, freelancer, or self-em...     NaN  \n",
       "1            4                                 Employed full-time     NaN  \n",
       "2          NaN                                                NaN     NaN  \n",
       "3            4                                                NaN     NaN  \n",
       "4            8                                 Employed full-time     NaN  "
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "column_needed.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "column_needed=column_needed[~column_needed[\"Salary\"].isnull()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Country           0\n",
       "EdLevel         568\n",
       "YearsCodePro    135\n",
       "Employment       39\n",
       "Salary            0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "column_needed.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "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>Country</th>\n",
       "      <th>EdLevel</th>\n",
       "      <th>YearsCodePro</th>\n",
       "      <th>Employment</th>\n",
       "      <th>Salary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>United States</td>\n",
       "      <td>Bachelor’s degree (B.A., B.S., B.Eng., etc.)</td>\n",
       "      <td>13</td>\n",
       "      <td>Employed full-time</td>\n",
       "      <td>116000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>Master’s degree (M.A., M.S., M.Eng., MBA, etc.)</td>\n",
       "      <td>4</td>\n",
       "      <td>Employed full-time</td>\n",
       "      <td>32315.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>Bachelor’s degree (B.A., B.S., B.Eng., etc.)</td>\n",
       "      <td>2</td>\n",
       "      <td>Employed full-time</td>\n",
       "      <td>40070.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Spain</td>\n",
       "      <td>Some college/university study without earning ...</td>\n",
       "      <td>7</td>\n",
       "      <td>Employed full-time</td>\n",
       "      <td>14268.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Netherlands</td>\n",
       "      <td>Secondary school (e.g. American high school, G...</td>\n",
       "      <td>20</td>\n",
       "      <td>Employed full-time</td>\n",
       "      <td>38916.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Country                                            EdLevel  \\\n",
       "7    United States       Bachelor’s degree (B.A., B.S., B.Eng., etc.)   \n",
       "9   United Kingdom    Master’s degree (M.A., M.S., M.Eng., MBA, etc.)   \n",
       "10  United Kingdom       Bachelor’s degree (B.A., B.S., B.Eng., etc.)   \n",
       "11           Spain  Some college/university study without earning ...   \n",
       "12     Netherlands  Secondary school (e.g. American high school, G...   \n",
       "\n",
       "   YearsCodePro          Employment    Salary  \n",
       "7            13  Employed full-time  116000.0  \n",
       "9             4  Employed full-time   32315.0  \n",
       "10            2  Employed full-time   40070.0  \n",
       "11            7  Employed full-time   14268.0  \n",
       "12           20  Employed full-time   38916.0  "
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "column_needed.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 34756 entries, 7 to 64154\n",
      "Data columns (total 5 columns):\n",
      " #   Column        Non-Null Count  Dtype  \n",
      "---  ------        --------------  -----  \n",
      " 0   Country       34756 non-null  object \n",
      " 1   EdLevel       34188 non-null  object \n",
      " 2   YearsCodePro  34621 non-null  object \n",
      " 3   Employment    34717 non-null  object \n",
      " 4   Salary        34756 non-null  float64\n",
      "dtypes: float64(1), object(4)\n",
      "memory usage: 1.6+ MB\n"
     ]
    }
   ],
   "source": [
    "column_needed.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "column_needed=column_needed.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Country         0\n",
       "EdLevel         0\n",
       "YearsCodePro    0\n",
       "Employment      0\n",
       "Salary          0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "column_needed.isnull().sum()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(34025, 5)"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "column_needed.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 30019 entries, 7 to 64154\n",
      "Data columns (total 4 columns):\n",
      " #   Column        Non-Null Count  Dtype  \n",
      "---  ------        --------------  -----  \n",
      " 0   Country       30019 non-null  object \n",
      " 1   EdLevel       30019 non-null  object \n",
      " 2   YearsCodePro  30019 non-null  object \n",
      " 3   Salary        30019 non-null  float64\n",
      "dtypes: float64(1), object(3)\n",
      "memory usage: 1.1+ MB\n"
     ]
    }
   ],
   "source": [
    "column_needed= column_needed[column_needed[\"Employment\"]==\"Employed full-time\"]\n",
    "\n",
    "column_needed=column_needed.drop(\"Employment\",axis=1)\n",
    "column_needed.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "United States     7569\n",
       "India             2425\n",
       "United Kingdom    2287\n",
       "Germany           1903\n",
       "Canada            1178\n",
       "                  ... \n",
       "Bahamas              1\n",
       "Niger                1\n",
       "San Marino           1\n",
       "Fiji                 1\n",
       "Guinea               1\n",
       "Name: Country, Length: 154, dtype: int64"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "column_needed[\"Country\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "def join_country(cat,cutoff):\n",
    "    cat_join={}\n",
    "    for i in cat.index:\n",
    "        if cat[i]>=cutoff:\n",
    "            cat_join[i]=i\n",
    "\n",
    "        else:\n",
    "            cat_join[i]=\"others\"\n",
    "    return cat_join\n",
    "            "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "country_map=join_country(column_needed[\"Country\"].value_counts(),400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "others                8549\n",
       "United States         7569\n",
       "India                 2425\n",
       "United Kingdom        2287\n",
       "Germany               1903\n",
       "Canada                1178\n",
       "Brazil                 991\n",
       "France                 972\n",
       "Spain                  670\n",
       "Australia              659\n",
       "Netherlands            654\n",
       "Poland                 566\n",
       "Italy                  560\n",
       "Russian Federation     522\n",
       "Sweden                 514\n",
       "Name: Country, dtype: int64"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "column_needed[\"Country\"]=column_needed[\"Country\"].map(country_map)\n",
    "column_needed[\"Country\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig,ax=plt.subplots(1,1,figsize=(12,7))\n",
    "column_needed.boxplot(\"Salary\",\"Country\",ax=ax)\n",
    "plt.suptitle(\"Salary vs Country\")\n",
    "plt.ylabel(\"Salary\")\n",
    "plt.title(\"\")\n",
    "plt.xticks(rotation=90)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "column_needed=column_needed[column_needed[\"Salary\"]<=250000]\n",
    "column_needed=column_needed[column_needed[\"Salary\"]>10000]\n",
    "column_needed=column_needed[column_needed[\"Country\"]!=\"others\"]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig,ax=plt.subplots(1,1,figsize=(12,7))\n",
    "column_needed.boxplot(\"Salary\",\"Country\",ax=ax)\n",
    "plt.suptitle(\"Salary vs Country\")\n",
    "plt.ylabel(\"Salary\")\n",
    "plt.title(\"\")\n",
    "plt.xticks(rotation=90)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['13', '4', '2', '7', '20', '1', '3', '10', '12', '29', '6', '28',\n",
       "       '8', '23', '15', '25', '9', '11', 'Less than 1 year', '5', '21',\n",
       "       '16', '18', '14', '32', '19', '22', '38', '30', '26', '27', '17',\n",
       "       '24', '34', '35', '33', '36', '40', '39', 'More than 50 years',\n",
       "       '31', '37', '41', '45', '42', '44', '43', '50', '49'], dtype=object)"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "column_needed[\"YearsCodePro\"].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "def clean_experience(x):\n",
    "    if x==\"More than 50 years\":\n",
    "        return 50\n",
    "    if x==\"Less than 1 year\":\n",
    "        return 0.5\n",
    "    return float(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "column_needed[\"YearsCodePro\"]=column_needed[\"YearsCodePro\"].apply(clean_experience)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([13. ,  4. ,  2. ,  7. , 20. ,  1. ,  3. , 10. , 12. , 29. ,  6. ,\n",
       "       28. ,  8. , 23. , 15. , 25. ,  9. , 11. ,  0.5,  5. , 21. , 16. ,\n",
       "       18. , 14. , 32. , 19. , 22. , 38. , 30. , 26. , 27. , 17. , 24. ,\n",
       "       34. , 35. , 33. , 36. , 40. , 39. , 50. , 31. , 37. , 41. , 45. ,\n",
       "       42. , 44. , 43. , 49. ])"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "column_needed[\"YearsCodePro\"].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Bachelor’s degree (B.A., B.S., B.Eng., etc.)',\n",
       "       'Master’s degree (M.A., M.S., M.Eng., MBA, etc.)',\n",
       "       'Some college/university study without earning a degree',\n",
       "       'Secondary school (e.g. American high school, German Realschule or Gymnasium, etc.)',\n",
       "       'Associate degree (A.A., A.S., etc.)',\n",
       "       'Professional degree (JD, MD, etc.)',\n",
       "       'Other doctoral degree (Ph.D., Ed.D., etc.)',\n",
       "       'I never completed any formal education',\n",
       "       'Primary/elementary school'], dtype=object)"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "column_needed[\"EdLevel\"].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "def clean_education(x):\n",
    "    if \"Bachelor’s degree\" in x:\n",
    "        return \"Bachelor’s degree\" \n",
    "    if \"Master’s degree\" in x:\n",
    "        return \"Master’s degree\"\n",
    "    if \"Professional degree\" in x or \"Other doctoral\" in x:\n",
    "            return \"Post grad\"\n",
    "    return \"Less than a Bachelors\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "column_needed[\"EdLevel\"]=column_needed[\"EdLevel\"].apply(clean_education)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Bachelor’s degree', 'Master’s degree', 'Less than a Bachelors',\n",
       "       'Post grad'], dtype=object)"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "column_needed[\"EdLevel\"].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "trans_edu=LabelEncoder()\n",
    "trans_edu.fit(column_needed[\"EdLevel\"])\n",
    "column_needed[\"EdLevel\"]=trans_edu.transform(column_needed[\"EdLevel\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "trans_count=LabelEncoder()\n",
    "trans_count.fit(column_needed[\"Country\"])\n",
    "column_needed[\"Country\"]=trans_count.transform(column_needed[\"Country\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([13, 12, 10,  7,  4,  2,  6,  1,  3,  5, 11,  8,  0,  9])"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "column_needed[\"Country\"].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "features=column_needed.drop(\"Salary\",axis=1)\n",
    "target=column_needed[\"Salary\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(18489, 3)"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train,X_test,Y_train,Y_test=train_test_split(features,target,test_size=0.15,random_state=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(estimator=RandomForestRegressor(random_state=0),\n",
       "             param_grid={'max_depth': [None, 2, 4, 6, 8, 10, 12],\n",
       "                         'max_features': [3]},\n",
       "             scoring='neg_mean_squared_error')"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "max_depth=[None,2,4,6,8,10,12]\n",
    "parameters={\"max_depth\":max_depth,\"max_features\":[3]}\n",
    "regressor=RandomForestRegressor(random_state=0)\n",
    "search=GridSearchCV(regressor,parameters,scoring=\"neg_mean_squared_error\")\n",
    "search.fit(X_train,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'max_depth': 8, 'max_features': 3}"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(max_depth=8, max_features=3, random_state=0)"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "regressor=RandomForestRegressor(random_state=0,max_depth=8,max_features=3)\n",
    "regressor.fit(X_train,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "30214.412107087024"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_pred=regressor.predict(X_train)\n",
    "np.sqrt(mean_squared_error(Y_train,train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "30840.508458729575"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_pred=regressor.predict(X_test)\n",
    "np.sqrt(mean_squared_error(Y_test,test_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([['United States', 'Master’s degree', '15']], dtype='<U15')"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test=np.array([[\"United States\",\"Master’s degree\",15]])\n",
    "test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "test[:,0]=trans_count.transform(test[:,0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "test[:,1]=trans_edu.transform(test[:,1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\HP\\anaconda3\\lib\\site-packages\\sklearn\\base.py:450: UserWarning: X does not have valid feature names, but RandomForestRegressor was fitted with feature names\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([144096.55476784])"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "regressor.predict(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression()"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model= LinearRegression()\n",
    "model.fit(X_train,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle.dump(model,open(\"model.pkl\",\"wb\"))"
   ]
  },
  {
   "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}