{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conforming to Statistical Assumptions\n",
    "> In this chapter, you will focus on analyzing the underlying distribution of your data and whether it will impact your machine learning pipeline. You will learn how to deal with skewed data and situations where outliers may be negatively impacting your analysis. This is the Summary of lecture \"Feature Engineering for Machine Learning in Python\", via datacamp.\n",
    "\n",
    "- toc: true \n",
    "- badges: true\n",
    "- comments: true\n",
    "- author: Chanseok Kang\n",
    "- categories: [Python, Datacamp, Machine Learning]\n",
    "- image: images/log_trans.png"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "plt.rcParams['figure.figsize'] = (8, 8)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data distributions\n",
    "- All model requires normal distribution\n",
    "![normal](image/distributions.png)\n",
    "- InterQuatile Range (IQR)\n",
    "![iqr](image/iqr.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### What does your data look like? (I)\n",
    "Up until now you have focused on creating new features and dealing with issues in your data. Feature engineering can also be used to make the most out of the data that you already have and use it more effectively when creating machine learning models.\n",
    "Many algorithms may assume that your data is normally distributed, or at least that all your columns are on the same scale. This will often not be the case, e.g. one feature may be measured in thousands of dollars while another would be number of years. In this exercise, you will create plots to examine the distributions of some numeric columns in the `so_survey_df` DataFrame, stored in `so_numeric_df`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SurveyDate</th>\n",
       "      <th>FormalEducation</th>\n",
       "      <th>ConvertedSalary</th>\n",
       "      <th>Hobby</th>\n",
       "      <th>Country</th>\n",
       "      <th>StackOverflowJobsRecommend</th>\n",
       "      <th>VersionControl</th>\n",
       "      <th>Age</th>\n",
       "      <th>Years Experience</th>\n",
       "      <th>Gender</th>\n",
       "      <th>RawSalary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2/28/18 20:20</td>\n",
       "      <td>Bachelor's degree (BA. BS. B.Eng.. etc.)</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Yes</td>\n",
       "      <td>South Africa</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Git</td>\n",
       "      <td>21</td>\n",
       "      <td>13</td>\n",
       "      <td>Male</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6/28/18 13:26</td>\n",
       "      <td>Bachelor's degree (BA. BS. B.Eng.. etc.)</td>\n",
       "      <td>70841.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Sweeden</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Git;Subversion</td>\n",
       "      <td>38</td>\n",
       "      <td>9</td>\n",
       "      <td>Male</td>\n",
       "      <td>70,841.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6/6/18 3:37</td>\n",
       "      <td>Bachelor's degree (BA. BS. B.Eng.. etc.)</td>\n",
       "      <td>NaN</td>\n",
       "      <td>No</td>\n",
       "      <td>Sweeden</td>\n",
       "      <td>8.0</td>\n",
       "      <td>Git</td>\n",
       "      <td>45</td>\n",
       "      <td>11</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5/9/18 1:06</td>\n",
       "      <td>Some college/university study without earning ...</td>\n",
       "      <td>21426.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Sweeden</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Zip file back-ups</td>\n",
       "      <td>46</td>\n",
       "      <td>12</td>\n",
       "      <td>Male</td>\n",
       "      <td>21,426.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4/12/18 22:41</td>\n",
       "      <td>Bachelor's degree (BA. BS. B.Eng.. etc.)</td>\n",
       "      <td>41671.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>UK</td>\n",
       "      <td>8.0</td>\n",
       "      <td>Git</td>\n",
       "      <td>39</td>\n",
       "      <td>7</td>\n",
       "      <td>Male</td>\n",
       "      <td>£41,671.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      SurveyDate                                    FormalEducation  \\\n",
       "0  2/28/18 20:20           Bachelor's degree (BA. BS. B.Eng.. etc.)   \n",
       "1  6/28/18 13:26           Bachelor's degree (BA. BS. B.Eng.. etc.)   \n",
       "2    6/6/18 3:37           Bachelor's degree (BA. BS. B.Eng.. etc.)   \n",
       "3    5/9/18 1:06  Some college/university study without earning ...   \n",
       "4  4/12/18 22:41           Bachelor's degree (BA. BS. B.Eng.. etc.)   \n",
       "\n",
       "   ConvertedSalary Hobby       Country  StackOverflowJobsRecommend  \\\n",
       "0              NaN   Yes  South Africa                         NaN   \n",
       "1          70841.0   Yes       Sweeden                         7.0   \n",
       "2              NaN    No       Sweeden                         8.0   \n",
       "3          21426.0   Yes       Sweeden                         NaN   \n",
       "4          41671.0   Yes            UK                         8.0   \n",
       "\n",
       "      VersionControl  Age  Years Experience Gender   RawSalary  \n",
       "0                Git   21                13   Male         NaN  \n",
       "1     Git;Subversion   38                 9   Male   70,841.00  \n",
       "2                Git   45                11    NaN         NaN  \n",
       "3  Zip file back-ups   46                12   Male   21,426.00  \n",
       "4                Git   39                 7   Male  £41,671.00  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "so_survey_df = pd.read_csv('./dataset/Combined_DS_v10.csv')\n",
    "so_survey_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "so_numeric_df = so_survey_df[['ConvertedSalary', 'Age', 'Years Experience']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create a histogram\n",
    "so_numeric_df.hist();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeAAAAHSCAYAAAA0ZhgzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAWLElEQVR4nO3df6zdd33f8dcb3wRCoPwKeCxxMJWi9qauWDsXQeKNa4LaMqaSainD66poulK0bs1YmTo8vC5Dk7VEWged1Ha1ZqZUdCYkZYURlkLh3mqOu3QJ41e4Y1BKnYxQQIOAMwQ4fPbHPQET/OPG91y/7ePHQzryOd/z/X7P27n+5unv9x6fW2OMAABn1pO6BwCA85EAA0ADAQaABgIMAA0EGAAaCDAANJg7ky92ySWXjK1bt57Jl2RKHnnkkVx88cXdY8B5yfF37rrvvvu+NMZ47vGeO6MB3rp1a+69994z+ZJMyfLychYWFrrHgPOS4+/cVVV/fqLnXIIGgAYCDAANBBgAGggwADQQYABoIMAA0ECAAaCBAANAAwEGgAYCDAANBBgAGggwADQQYABoIMAA0ECAAaCBAHNSBw4cyLZt23LNNddk27ZtOXDgQPdIADNhrnsAzl4HDhzInj17sn///jz66KPZtGlTFhcXkyS7du1qng7g3OYMmBPau3dv9u/fn507d2Zubi47d+7M/v37s3fv3u7RAM55AswJraysZMeOHd+zbMeOHVlZWWmaCGB2CDAnND8/n4MHD37PsoMHD2Z+fr5pIoDZIcCc0J49e7K4uJilpaUcPXo0S0tLWVxczJ49e7pHAzjneRMWJ/TYG61uvPHGrKysZH5+Pnv37vUGLIApEGBOateuXdm1a1eWl5ezsLDQPQ7AzHAJGgAaCDAANBBgAGggwADQQIABoMGaAlxVv1xV91fVx6vqQFU9papeWFX3VNWnquq2qrpwo4cFgFlxygBX1aVJ/lGS7WOMbUk2JXltkluSvHmMcUWSLydZ3MhBAWCWrPUS9FySi6pqLslTkzyU5OVJ7pg8f2uSa6c/HgDMplMGeIzxf5L8mySHsxreh5Pcl+QrY4yjk9UeTHLpRg0JALPmlJ+EVVXPSvLqJC9M8pUktyd55XFWHSfY/oYkNyTJ5s2bs7y8fLqz0ujIkSO+dtDE8Teb1vJRlK9I8mdjjC8mSVW9M8lVSZ5ZVXOTs+DLknzueBuPMfYl2Zck27dvHz7O8Nzkoyihj+NvNq3le8CHk7ykqp5aVZXkmiSfSLKU5LrJOtcnedfGjAgAs2ct3wO+J6tvtvpQko9NttmX5A1JXl9Vn07ynCT7N3BOAJgpa/ppSGOMm5Lc9LjFn0ny4qlPBADnAZ+EBQANBBgAGggwADQQYABoIMAA0ECAAaCBAANAAwEGgAYCDAANBBgAGggwADQQYABoIMAA0ECAAaCBAANAAwEGgAYCDAANBBgAGggwADSY6x6As0NVTWU/Y4yp7Adg1jkDJslqOE92e8Eb3nPKdcQXYO0EGAAaCDAANBBgAGggwADQQIABoIEAA0ADAQaABgIMAA0EGAAaCDAANBBgAGggwADQQIABoIEAA0ADAQaABgIMAA0EGAAaCDAANBBgAGggwADQQIABoIEAA0ADAQaABgIMAA0EGAAaCDAANBBgAGggwADQQIABoIEAA0ADAQaABgIMAA0EGAAaCDAANBBgAGggwADQQIABoIEAA0ADAQaABgIMAA0EGAAaCDAANBBgAGggwADQQIABoIEAA0ADAQaABgIMAA0EGAAaCDAANBBgAGggwADQQIABoIEAA0ADAQaABgIMAA0EGAAaCDAANBBgAGggwADQQIABoIEAA0ADAQaABgIMAA0EGAAaCDAANFhTgKvqmVV1R1X9r6paqaqXVtWzq+r9VfWpya/P2uhhAWBWrPUM+NeT3DXG+OEkL0qykmR3kg+MMa5I8oHJYwBgDU4Z4Kr6gSR/Pcn+JBljfHOM8ZUkr05y62S1W5Ncu1FDAsCsmVvDOj+Y5ItJ/mNVvSjJfUlel2TzGOOhJBljPFRVzzvexlV1Q5IbkmTz5s1ZXl6extw08LWDHkeOHHH8zaC1BHguyY8nuXGMcU9V/XqewOXmMca+JPuSZPv27WNhYeF05qTbXXfG1w56LC8vO/5m0Fq+B/xgkgfHGPdMHt+R1SD/RVU9P0kmv35hY0YEgNlzygCPMT6f5IGq+qHJomuSfCLJu5NcP1l2fZJ3bciEADCD1nIJOkluTPK7VXVhks8k+XtZjfc7qmoxyeEkP7cxIwLA7FlTgMcYH06y/ThPXTPdcQDg/OCTsACggQADQAMBBoAGAgwADQQYABoIMAA0EGAAaCDAANBAgAGggQADQAMBBoAGAgwADQQYABoIMAA0EGAAaCDAANBAgAGggQADQAMBBoAGAgwADQQYABoIMAA0EGAAaCDAANBAgAGggQADQAMBBoAGAgwADQQYABoIMAA0EGAAaCDAANBAgAGggQADQAMBBoAGAgwADQQYABoIMAA0EGAAaCDAANBAgAGggQADQAMBBoAGAgwADQQYABoIMAA0EGAAaCDAANBAgAGggQADQAMBBoAGAgwADQQYABoIMAA0EGAAaCDAANBAgAGggQADQAMBBoAGAgwADQQYABoIMAA0EGAAaCDAANBAgAGggQADQIO57gE4M170pvfl4a9/a1372Lr7znVt/4yLLshHbvrJde0DYFYI8Hni4a9/K5+9+VWnvf3y8nIWFhbWNcN6Aw4wS1yCBoAGAgwADQQYABoIMAA0EGAAaCDAANBAgAGggQADQAMBBoAGAgwADQQYABoIMAA0EGAAaCDAANBAgAGggQADQAMBBoAGAgwADdYc4KraVFX/s6reM3n8wqq6p6o+VVW3VdWFGzcmAMyWJ3IG/LokK8c8viXJm8cYVyT5cpLFaQ4GALNsTQGuqsuSvCrJf5g8riQvT3LHZJVbk1y7EQMCwCxa6xnwW5L80yTfnjx+TpKvjDGOTh4/mOTSKc8GADNr7lQrVNXfTPKFMcZ9VbXw2OLjrDpOsP0NSW5Iks2bN2d5efn0JmXd1vPf/siRI1P52vn6wxM3reOPs8spA5zk6iQ/U1V/I8lTkvxAVs+In1lVc5Oz4MuSfO54G48x9iXZlyTbt28fCwsL05ibJ+quO7Oe//bLy8vr2n4aM8D5airHH2edU16CHmP8szHGZWOMrUlem+SDY4yfT7KU5LrJatcnedeGTQkAM2Y9/w74DUleX1Wfzur3hPdPZyQAmH1ruQT9HWOM5STLk/ufSfLi6Y8EALPPJ2EBQAMBBoAGAgwADQQYABoIMAA0EGAAaCDAANBAgAGggQADQAMBBoAGAgwADQQYABoIMAA0EGAAaCDAANBAgAGggQADQAMBBoAGc90DcGY8fX53fvTW3evbya3rnSFJXrW+nQDMCAE+T3xt5eZ89ubTj9/y8nIWFhbWNcPW3Xeua3uAWeISNAA0EGAAaCDAANBAgAGggQADQAMBBoAGAgwADQQYABoIMAA0EGAAaCDAANBAgAGggQADQAMBBoAGAgwADQQYABoIMAA0EGAAaCDAANBAgAGggQADQAMBBoAGAgwADQQYABoIMAA0EGAAaCDAANBAgAGggQADQAMBBoAGAgwADQQYABoIMAA0EGCAs9SBAweybdu2XHPNNdm2bVsOHDjQPRJTNNc9AADf78CBA9mzZ0/279+fRx99NJs2bcri4mKSZNeuXc3TMQ3OgAHOQnv37s3+/fuzc+fOzM3NZefOndm/f3/27t3bPRpTIsAAZ6GVlZXs2LHje5bt2LEjKysrTRMxbQIMcBaan5/PwYMHv2fZwYMHMz8/3zQR0ybAAGehPXv2ZHFxMUtLSzl69GiWlpayuLiYPXv2dI/GlHgTFsBZ6LE3Wt14441ZWVnJ/Px89u7d6w1YM0SAAc5Su3btyq5du7K8vJyFhYXucZgyl6ABoIEAA0ADAQaABgIMAA0EGAAaeBf0eWTr7jvXt4O71rf9My66YH2vDzBDBPg88dmbX7Wu7bfuvnPd+wDgu1yCBoAGAgwADQQYABoIMAA08CYsgLPU5ZdfngceeOA7j7ds2ZLDhw83TsQ0OQMGOAs9Ft+rrroqt99+e6666qo88MADufzyy7tHY0oEGOAs9Fh877777lxyySW5++67vxNhZoMAA5yl7rjjjpM+5twmwABnqeuuu+6kjzm3CTDAWWjLli05dOhQrr766nzpS1/K1VdfnUOHDmXLli3dozEl3gUNcBY6fPhwLr/88hw6dCiHDh1K4l3Qs8YZMMBZ6vDhwxljZGlpKWMM8Z0xzoABGlXVVPYzxpjKfjhznAEDNBpjnPL2gje855TrcO4RYABoIMAA0ECAAaCBAANAg1MGuKq2VNVSVa1U1f1V9brJ8mdX1fur6lOTX5+18eMCwGxYyxnw0ST/ZIwxn+QlSf5hVV2ZZHeSD4wxrkjygcljAGANThngMcZDY4wPTe5/LclKkkuTvDrJrZPVbk1y7UYNCQCz5gl9D7iqtib5sST3JNk8xngoWY10kudNezgAmFVr/iSsqnpakt9L8o/HGF9d66e3VNUNSW5Iks2bN2d5efk0xuRs4GsHfRx/s2dNAa6qC7Ia398dY7xzsvgvqur5Y4yHqur5Sb5wvG3HGPuS7EuS7du3j4WFhfVPzZl3153xtYMmjr+ZtJZ3QVeS/UlWxhj/9pin3p3k+sn965O8a/rjAcBsWssZ8NVJfiHJx6rqw5Nlb0xyc5J3VNViksNJfm5jRgSA2XPKAI8xDiY50Td8r5nuOABwfvBJWADQQIABoIEAA0ADAQaABmv+IA5m21o+WKVuOfV+xhhTmAZg9jkDJslqOE92W1paOuU64guwdgIMAA0EGAAaCDAANBBgAGggwADQQIABoIEAA0ADAQaABgIMAA0EGAAaCDAANBBgAGggwADQQIABoIEAA0ADAQaABgIMAA0EGAAaCDAANBBgAGggwADQQIABoMFc9wAAs+xFb3pfHv76t9a9n6277zztbZ9x0QX5yE0/ue4ZmC4BBthAD3/9W/nsza9a1z6Wl5ezsLBw2tuvJ95sHJegAaCBAANAAwEGgAYCDAANBBgAGggwADQQYABoIMAA0ECAAaCBAANAAwEGgAYCDAAN/DAGgA309Pnd+dFbd69/R7euZ4YkWd8PhGD6BBhgA31t5WY/DYnjcgkaABoIMAA0EGAAaCDAANBAgAGggQADQAMBBoAGAgwADQQYABoIMAA0EGAAaCDAANBAgAGggZ+GBLDBpvLTiO46/X0846IL1v/6TJ0AA2yg9f4owmQ14NPYD2cXl6ABoIEAA0ADAQaABgIMAA0EGAAaCDAANBBgAGggwADQQIABoIEAA0ADAQaABgIMAA0EGAAaCDAANBBgAGggwADQQIABoIEAA0ADAQaABgIMAA0EGAAaCDAANBBgAGggwADQQIABoIEAA0ADAQaABusKcFX9dFV9sqo+XVW7pzUUAMy60w5wVW1K8htJXpnkyiS7qurKaQ0GALNsPWfAL07y6THGZ8YY30zy9iSvns5YADDb1hPgS5M8cMzjByfLAIBTmFvHtnWcZeP7Vqq6IckNSbJ58+YsLy+v4yXpcuTIEV872AA7d+5c03p1y8mfX1pamsI0nEnrCfCDSbYc8/iyJJ97/EpjjH1J9iXJ9u3bx8LCwjpeki7Ly8vxtYPpG+P7zlu+j+NvNq3nEvT/SHJFVb2wqi5M8tok757OWAAw2077DHiMcbSqfinJHyTZlOStY4z7pzYZAMyw9VyCzhjjvUneO6VZAOC84ZOwAKCBAANAAwEGgAYCDAANBBgAGggwADQQYABoIMAA0ECAAaCBAANAAwEGgAYCDAANBBgAGggwADQQYABoUGOMM/diVV9M8udn7AWZpkuSfKl7CDhPOf7OXS8YYzz3eE+c0QBz7qqqe8cY27vngPOR4282uQQNAA0EGAAaCDBrta97ADiPOf5mkO8BA0ADZ8AA0ECA+Y6q+tmqGlX1w92zwEaqVQer6pXHLHtNVd11Bl77FVX1cFV9+Jjbzg1+zb0b/Ro8cS5B8x1V9Y4kz0/ygTHGv2weBzZUVW1LcnuSH0uyKcmHk/z0GONP17HPuTHG0VOs84okvzTGuPZ0X2faM9HDGTBJkqp6WpKrkywmee1k2ZOq6jer6v6qek9Vvbeqrps891er6o+q6r6q+oOqen7j+PCEjTE+nuS/JHlDkpuS/M4Y40+r6vqq+pPJmelvVtWTkqSq9lXVvZPj4V88tp+qerCqfrWq7k7ys1X1y1X1iar6SFW9ba3zVNVLJ695YVU9bbKP+ckZ81JV/f5k2W9UVU22eWVV/XFVfaiqbquqi08w09uq6trJcz9xzLH7X6tq82T5waq6efJ7/2RVXTVZPldVb66qj1fVR6vqH5xsPzwBYww3tyT5u0n2T+4fSvLjSa5L8t6s/kXtLyX58mTZBZN1njtZ/28neWv378HN7Yneklyc5JNJPpbkyUm2Jfn9JHOT5/cl+TuT+8+e/DqX5L8luXLy+MEkrz9mnw8luXBy/5nHec1XJHk4q2fcj922Tp67eXL77SS/csz6/y/J1qyeqX8wybVJnpfkj5I8dbLeniRvPMFMb5ts8+TJsXvJZPnPJ9k3uX8wyS2T+z+T5K7J/RuT3JZk02P/HU62H7e13+ZO0WfOH7uSvGVy/+2TxxckuX2M8e0kn6+qpcnzP5TV/1G9f/IX8U1Z/Z8OnFPGGI9U1W1JjowxvjG5PPwTSe6d/Nm+KMkDk9V3VdViVgP8l5NcmeQTk+duO2a39yd5W1W9K6sxP56lcfxL0DcluS/JV5P84jHL//sY47NJUlVvT7JjsvzKJIcms16Y1Yg+5tiZHjOf5EeS/OExx+6Dxzz/zsmv92U1+MnqXwDeMsZ4NEnGGP+3qv7KKfbDGggwqarnJHl5km1VNbJ6MI0k//lEmyS5f4zx0jM0Imykb09uyeqf7beOMX712BWq6ookr0vy4jHGVyaXlp9yzCqPHHP/p5K8LMmrk/zzqtr2WLzW4JIkT53cf3KSr0/uP/7NOmMy611jjF84wb4eOc6ySvLRMcZfO8E235j8+mi+24c6zuufaj+sge8Bk6xeVv6dMcYLxhhbxxhbkvxZVj/8/W9Nvhe8OcnCZP1PJnluVb00Sarqgqr6kY7BYcr+MMlrquqSZPUvp1V1eZIfSPK1JF+dvN/hp463cVVtSnLZGOODSX4lyXPz3aCuxb4ku7P65rB/fczyl1TV5ZP9vyarZ7qHkrysqn5w8toXT/6icDKfSHJpVb14ss2Fazh235fkFyevnap69mnuh8dxBkyyern55sct+72sXq56MMnHk/zvJPckeXiM8c3Jm7H+XVU9I6t/jt6S1UtvcM4aY3ysqt6U1UurT0ryrSR/P8m9WY3Ox5N8JsndJ9jFXJL/VFVPz+oJzi1jjK8dZ72dVfXhYx6/KcmzkjwyxnhHVc0l+eOqetnk+UNJfi2rl32Xk7x7jDEml8Rvq6oLJ+u9McmnTvL7+8Yxx+7TJ/P+Wk5+7P52kiuSfLSqjib5rTHGvz+N/fA4/hkSJ1VVTxtjHJlcpv6TJFePMT7fPRecL+oM/7MlzhxnwJzKe6rqmVl9g8e/El+A6XAGDAANvAkLABoIMAA0EGAAaCDAANBAgAGggQADQIP/Dxr3mqiefNCzAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create a boxplot of two columns\n",
    "so_numeric_df[['Age', 'Years Experience']].boxplot();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAf8AAAHSCAYAAADiw8JWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nO3df5TddX3v++fbCXKpqKBoSkgo9hJ7B0cOlTlCJe2aMTWAnmv0VC8J3kuEWaZ0aWy9Zy3AM65Lr3a6hK4ez7XH0pWeiUCvDv4uUaBpFp1tb0CUpFAITj1ES2FMCocTREKEMuF9/9ifwZ3pJBOSbPZOPs/HWnvNd7+/n8/3+5msNXnt72d/9ndHZiJJkurxsk4PQJIkvbQMf0mSKmP4S5JUGcNfkqTKGP6SJFXG8JckqTLzOj2Al8pJJ52Up512WqeHIVXp6aef5hWveEWnhyFVZ8uWLY9n5utm1qsJ/9NOO43Nmzd3ehhSlRqNBgMDA50ehlSdiPin2epO+0uSVBnDX5Kkyhj+kiRVxvCXJKkyhr8kSZUx/CVJqozhL0lSZQx/SZIqY/hLklQZw1+SpMoY/pIkVcbwlySpMoa/JEmVMfwlSarMnOEfEYsiYjwiJiLigYj43VJ/TURsjIgHy88TSz0i4rMRsS0i7ouIt7Qca1Vp/2BErGqpnx0R95c+n42IONhzSJKk/TuQK/8p4D9kZi9wLvDhiDgDuAq4PTMXA7eX5wAXAovLYzVwHTSDHLgaOAd4K3D1dJiXNqtb+l1Q6i/qHJK6y6mnnkpEMDg4SERw6qmndnpIkjiA8M/MHZn5d2X7KWACOAVYDtxQmt0AvKdsLwduzKa7gBMi4mTgfGBjZu7MzCeAjcAFZd+rMvM7mZnAjTOO9WLOIalLnHrqqTzyyCO87W1v4ytf+Qpve9vbeOSRR3wBIHWBF/Wef0ScBvwq8F1gfmbugOYLBOD1pdkpwCMt3SZLbX/1yVnqHMQ5JHWJ6eC/4447OOmkk7jjjjteeAEgqbPmHWjDiDge+Brwe5n50/K2/KxNZ6nlQdT3O5wD6RMRq2m+LcD8+fNpNBpzHFbS4fSxj32MRqPBrl27aDQafOxjH+POO+/0b1HqsAMK/4g4hmbwfyEzv17Kj0bEyZm5o0y5P1bqk8Cilu4Lge2lPjCj3ij1hbO0P5hz7CUz1wJrAfr7+3NgYGBmE0lt9JnPfIY77riDRqPBwMAA5513HgD+LUqddSCr/QMYBSYy8z+17FoPTK/YXwXc3FK/pKzIPxd4skzZbwCWRcSJZaHfMmBD2fdURJxbznXJjGO9mHNI6hKLFi3izjvv5LzzzuPxxx/nvPPO484772TRokVzd5bUVtFcY7efBhFLgP8PuB94vpT/I833/b8MnAo8DLw/M3eWAP8vNFfs7wYuzczN5ViXlb4AI5n5+VLvB64HjgNuA9ZkZkbEa1/sOfalv78/N2/ebxNJh9n0or9pixYt4uGHH+7giKS6RMSWzOz/V/W5wv9oYfhLnTM97S/ppbWv8PcOf5IkVcbwlySpMoa/JEmVMfwlSaqM4S9JUmUMf0mSKmP4S5JUGcNfkqTKGP6SJFXG8JckqTKGvyRJlTH8JUmqjOEvSVJlDH9Jkipj+EuSVBnDX5Kkyhj+kiRVxvCXJKkyhr8kSZUx/CVJqozhL0lSZQx/SZIqY/hLklQZw1+SpMoY/pIkVcbwlySpMoa/JEmVMfwlSaqM4S9JUmUMf0mSKmP4S5JUGcNfkqTKGP6SJFXG8JckqTKGvyRJlTH8JUmqjOEvSVJlDH9Jkipj+EuSVBnDX5KkyswZ/hGxLiIei4itLbUvRcS95fFQRNxb6qdFxM9a9v1ZS5+zI+L+iNgWEZ+NiCj110TExoh4sPw8sdSjtNsWEfdFxFtajrWqtH8wIlYdzn8QSZKOdgdy5X89cEFrITMvysyzMvMs4GvA11t2/3B6X2Ze3lK/DlgNLC6P6WNeBdyemYuB28tzgAtb2q4u/YmI1wBXA+cAbwWunn7BIEmS5jZn+Gfm3wI7Z9tXrt7/N2Bsf8eIiJOBV2XmdzIzgRuB95Tdy4EbyvYNM+o3ZtNdwAnlOOcDGzNzZ2Y+AWxkxosTSZK0b4f6nv+vA49m5oMttTdExD0R8e2I+PVSOwWYbGkzWWoA8zNzB0D5+fqWPo/M0mdfdUmSdADmHWL/lex91b8DODUz/0dEnA38ZUS8CYhZ+uYcx95XnwM+VkSspvmWAfPnz6fRaMxxSkntsGvXLv/+pC5y0OEfEfOAfw+cPV3LzGeBZ8v2loj4IfBGmlfnC1u6LwS2l+1HI+LkzNxRpvUfK/VJYNEsfSaBgRn1xmxjzMy1wFqA/v7+HBgYmK2ZpDZrNBr49yd1j0OZ9v9N4B8y84Xp/Ih4XUT0lO1fprlY70dlOv+piDi3rBO4BLi5dFsPTK/YXzWjfklZ9X8u8GQ5zgZgWUScWBb6LSs1SZJ0AOa88o+IMZpX2idFxCRwdWaOAiv41wv9fgP4ZERMAXuAyzNzerHg79D85MBxwG3lAfBp4MsRMQQ8DLy/1G8F3glsA3YDlwJk5s6I+BRwd2n3yZZzSJKkOURz8f3Rr7+/Pzdv3tzpYUhVctpf6oyI2JKZ/TPr3uFPkqTKGP6SJFXG8JckqTKGvyRJlTH8JUmqjOEvSVJlDH9Jkipj+EuSVBnDX5Kkyhj+kiRVxvCXJKkyhr8kSZUx/CVJqozhL0lSZQx/SZIqY/hLklQZw1+SpMoY/pIkVcbwlySpMoa/JEmVMfwlSaqM4S9JUmUMf0mSKmP4S5JUGcNfkqTKGP6SJFXG8JckqTKGvyRJlTH8JUmqjOEvSVJlDH9Jkipj+EuSVBnDX5Kkyhj+kiRVxvCXJKkyhr8kSZUx/CVJqozhL0lSZQx/SZIqY/hLklQZw1+SpMrMGf4RsS4iHouIrS2134+IH0fEveXxzpZ9H4+IbRHxg4g4v6V+Qalti4irWupviIjvRsSDEfGliHh5qR9bnm8r+0+b6xySJGluB3Llfz1wwSz1z2TmWeVxK0BEnAGsAN5U+vxpRPRERA/wOeBC4AxgZWkLcE051mLgCWCo1IeAJzLzdOAzpd0+z/Hifm1Jkuo1Z/hn5t8COw/weMuBmzLz2cz8R2Ab8Nby2JaZP8rMfwFuApZHRABvB75a+t8AvKflWDeU7a8CS0v7fZ1DkiQdgEN5z/8jEXFfeVvgxFI7BXikpc1kqe2r/lrgJ5k5NaO+17HK/idL+30dS5IkHYB5B9nvOuBTQJaffwxcBsQsbZPZX2Tkftqzn33767OXiFgNrAaYP38+jUZjtmaS2mzXrl3+/Uld5KDCPzMfnd6OiD8HvlWeTgKLWpouBLaX7dnqjwMnRMS8cnXf2n76WJMRMQ94Nc23H/Z3jpnjXAusBejv78+BgYEX9XtKOjwajQb+/Und46Cm/SPi5Jan7wWmPwmwHlhRVuq/AVgMfA+4G1hcVva/nOaCvfWZmcA48L7SfxVwc8uxVpXt9wF/U9rv6xySJOkAzHnlHxFjwABwUkRMAlcDAxFxFs3p9oeA3wbIzAci4svA94Ep4MOZuacc5yPABqAHWJeZD5RTXAncFBF/ANwDjJb6KPAXEbGN5hX/irnOIUmS5hbNi+mjX39/f27evLnTw5Cq5LS/1BkRsSUz+2fWvcOfJEmVMfwlSaqM4S9JUmUMf0mSKmP4S5JUGcNfkqTKGP6SJFXG8JckqTKGvyRJlTH8JUmqjOEvSVJlDH9Jkipj+EuSVBnDX5Kkyhj+kiRVxvCXJKkyhr8kSZUx/CVJqozhL0lSZQx/SZIqY/hLklQZw1+SpMoY/pIkVcbwlySpMoa/JEmVMfwlSaqM4S9JUmUMf0mSKmP4S5JUGcNfkqTKGP6SJFXG8JckqTKGvyRJlTH8JUmqjOEvSVJlDH9Jkipj+EuSVBnDX5Kkyhj+kiRVxvCXJKkyc4Z/RKyLiMciYmtL7Y8i4h8i4r6I+EZEnFDqp0XEzyLi3vL4s5Y+Z0fE/RGxLSI+GxFR6q+JiI0R8WD5eWKpR2m3rZznLS3HWlXaPxgRqw7nP4gkSUe7A7nyvx64YEZtI9CXmWcC/w34eMu+H2bmWeVxeUv9OmA1sLg8po95FXB7Zi4Gbi/PAS5sabu69CciXgNcDZwDvBW4evoFgyRJmtuc4Z+ZfwvsnFH768ycKk/vAhbu7xgRcTLwqsz8TmYmcCPwnrJ7OXBD2b5hRv3GbLoLOKEc53xgY2buzMwnaL4QmfniRJIk7cPheM//MuC2ludviIh7IuLbEfHrpXYKMNnSZrLUAOZn5g6A8vP1LX0emaXPvuqSJOkAzDuUzhExDEwBXyilHcCpmfk/IuJs4C8j4k1AzNI95zr8Pvoc8LEiYjXNtwyYP38+jUZjjlNKaoddu3b59yd1kYMO/7LQ7t8BS8tUPpn5LPBs2d4SET8E3kjz6rz1rYGFwPay/WhEnJyZO8q0/mOlPgksmqXPJDAwo96YbYyZuRZYC9Df358DAwOzNZPUZo1GA//+pO5xUNP+EXEBcCXw7szc3VJ/XUT0lO1fprlY70dlOv+piDi3rPK/BLi5dFsPTK/YXzWjfklZ9X8u8GQ5zgZgWUScWBb6LSs1SZJ0AOa88o+IMZpX2idFxCTNlfYfB44FNpZP7N1VVvb/BvDJiJgC9gCXZ+b0YsHfofnJgeNorhGYXifwaeDLETEEPAy8v9RvBd4JbAN2A5cCZObOiPgUcHdp98mWc0iSpDlEmbE/6vX39+fmzZs7PQypSk77S50REVsys39m3Tv8SZJUGcNfkqTKGP6SJFXG8JckqTKGvyRJlTH8JUmqjOEvSVJlDH9Jkipj+EuSVBnDX5Kkyhj+kiRVxvCXJKkyhr8kSZUx/CVJqozhL6ltxsbG6OvrY+nSpfT19TE2NtbpIUkC5nV6AJKOTmNjYwwPDzM6OsqePXvo6elhaGgIgJUrV3Z4dFLdvPKX1BYjIyOMjo4yODjIvHnzGBwcZHR0lJGRkU4PTaqe4S+pLSYmJliyZMletSVLljAxMdGhEUmaZvhLaove3l42bdq0V23Tpk309vZ2aESSphn+ktpieHiYoaEhxsfHmZqaYnx8nKGhIYaHhzs9NKl6LviT1BbTi/rWrFnDxMQEvb29jIyMuNhP6gKRmZ0ew0uiv78/N2/e3OlhSFVqNBoMDAx0ehhSdSJiS2b2z6w77S9JUmUMf0mSKmP4S5JUGcNfkqTKGP6S2sZ7+0vdyY/6SWoL7+0vdS+v/CW1hff2l7qX4S+pLby3v9S9DH9JbeG9/aXuZfhLagvv7S91Lxf8SWoL7+0vdS/v7S+p7by3v9QZ3ttfkiQBhr8kSdUx/CVJqozhL0lSZQx/SZIqY/hLklQZw1+SpMocUPhHxLqIeCwitrbUXhMRGyPiwfLzxFKPiPhsRGyLiPsi4i0tfVaV9g9GxKqW+tkRcX/p89mIiIM9h6TuceaZZxIRDA4OEhGceeaZnR6SJA78yv964IIZtauA2zNzMXB7eQ5wIbC4PFYD10EzyIGrgXOAtwJXT4d5abO6pd8FB3MOSd3jzDPP5P777+fd73433/jGN3j3u9/N/fff7wsAqQscUPhn5t8CO2eUlwM3lO0bgPe01G/MpruAEyLiZOB8YGNm7szMJ4CNwAVl36sy8zvZvN3gjTOO9WLOIalLTAf/zTffzAknnMDNN9/8wgsASZ11KPf2n5+ZOwAyc0dEvL7UTwEeaWk3WWr7q0/OUj+Yc+xoHWBErKY5M8D8+fNpNBov/reUdNAuvfRSGo0Gu3btotFocOmll7J+/Xr/FqUOa8cX+8QstTyI+sGcY+9C5lpgLTTv7e+9xaWX1uc//3luvvnmF+7tv3z5cgDv8y912KGs9n90eqq9/Hys1CeBRS3tFgLb56gvnKV+MOeQ1CXe/OY3s379epYvX85PfvITli9fzvr163nzm9/c6aFJ1TuU8F8PTK/YXwXc3FK/pKzIPxd4skzdbwCWRcSJZaHfMmBD2fdURJxbVvlfMuNYL+YckrrEfffd98ILgPe+970vBP99993X6aFJ1Tugaf+IGAMGgJMiYpLmqv1PA1+OiCHgYeD9pfmtwDuBbcBu4FKAzNwZEZ8C7i7tPpmZ04sIf4fmJwqOA24rD17sOSR1l+mg9yt9pe4SzQX2R7/+/v7cvHlzp4chVWVsbIyRkREmJibo7e1leHiYlStXdnpYUjUiYktm9s+st2PBnyQxNjbG8PAwo6Oj7Nmzh56eHoaGhgB8ASB1mLf3ldQWIyMjjI6OMjg4yLx58xgcHGR0dJSRkZFOD02qnuEvqS0mJiZYsmTJXrUlS5YwMTHRoRFJmmb4S2qL3t5eNm3atFdt06ZN9Pb2dmhEkqYZ/pLaYnh4mKGhIcbHx5mammJ8fJyhoSGGh4c7PTSpei74k9QW04v61qxZ88Jq/5GRERf7SV3Aj/pJajs/5y91xr4+6ue0vyRJlTH8JbXN2NgYfX19LF26lL6+PsbGxjo9JEn4nr+kNvEmP1L38spfUlt4kx+pexn+ktpiYmKCycnJvab9JycnvcmP1AWc9pfUFgsWLOCKK67gi1/84gvT/hdffDELFizo9NCk6nnlL6ltImK/zyV1hlf+ktpi+/btXH/99Xvd5Oeaa67hgx/8YKeHJlXPK39JbdHb28vChQvZunUrt99+O1u3bmXhwoXe21/qAoa/pLbw3v5S93LaX1JbeG9/qXt5b39Jbee9/aXO8N7+kiQJMPwlSaqO4S9JUmUMf0mSKmP4S5JUGcNfkqTKGP6SJFXG8JckqTKGvyRJlTH8JUmqjOEvSVJlDH9Jkipj+EuSVBnDX5Kkyhj+kiRVxvCXJKkyhr8kSZUx/CVJqozhL0lSZQx/SZIqY/hLklSZgw7/iPiViLi35fHTiPi9iPj9iPhxS/2dLX0+HhHbIuIHEXF+S/2CUtsWEVe11N8QEd+NiAcj4ksR8fJSP7Y831b2n3awv4ckSbU56PDPzB9k5lmZeRZwNrAb+EbZ/ZnpfZl5K0BEnAGsAN4EXAD8aUT0REQP8DngQuAMYGVpC3BNOdZi4AlgqNSHgCcy83TgM6WdJEk6AIdr2n8p8MPM/Kf9tFkO3JSZz2bmPwLbgLeWx7bM/FFm/gtwE7A8IgJ4O/DV0v8G4D0tx7qhbH8VWFraS5KkORyu8F8BjLU8/0hE3BcR6yLixFI7BXikpc1kqe2r/lrgJ5k5NaO+17HK/idLe0mSNId5h3qA8j78u4GPl9J1wKeALD//GLgMmO3KPJn9BUjupz1z7Gsd22pgNcD8+fNpNBr7+jUktdGuXbv8+5O6yCGHP8336v8uMx8FmP4JEBF/DnyrPJ0EFrX0WwhsL9uz1R8HToiIeeXqvrX99LEmI2Ie8Gpg58yBZeZaYC1Af39/DgwMHPxvKemgNRoN/PuTusfhmPZfScuUf0Sc3LLvvcDWsr0eWFFW6r8BWAx8D7gbWFxW9r+c5lsI6zMzgXHgfaX/KuDmlmOtKtvvA/6mtJfURcbGxujr62Pp0qX09fUxNjY2dydJbXdIV/4R8QvAO4DfbilfGxFn0ZyGf2h6X2Y+EBFfBr4PTAEfzsw95TgfATYAPcC6zHygHOtK4KaI+APgHmC01EeBv4iIbTSv+Fccyu8h6fAbGxtjeHiY0dFR9uzZQ09PD0NDzQ/srFy5ssOjk+oWtVww9/f35+bNmzs9DKkafX19/Mmf/AmDg4MvTPuPj4+zZs0atm7dOvcBJB2yiNiSmf0z697hT1JbTExMsGTJkr1qS5YsYWJiokMjkjTN8JfUFr29vWzatGmv2qZNm+jt7e3QiCRNM/wltcXw8DBDQ0OMj48zNTXF+Pg4Q0NDDA8Pd3poUvUOx0f9JOlfmV7Ut2bNGiYmJujt7WVkZMTFflIXcMGfpLbzc/5SZ7jgT5IkAYa/JEnVMfwlSaqM4S9JUmUMf0mSKmP4S5JUGcNfkqTKGP6SJFXG8JckqTKGvyRJlTH8JUmqjOEvSVJlDH9Jkipj+Etqm7GxMfr6+li6dCl9fX2MjY11ekiSgHmdHoCko9PY2BjDw8OMjo6yZ88eenp6GBoaAmDlypUdHp1UN6/8JbXFyMgIo6OjDA4OMm/ePAYHBxkdHWVkZKTTQ5OqZ/hLaouJiQmWLFmyV23JkiVMTEx0aESSphn+ktqit7eXTZs27VXbtGkTvb29HRqRpGmGv6S2GB4eZmhoiPHxcaamphgfH2doaIjh4eFOD02qngv+JLXFypUrufPOO7nwwgt59tlnOfbYY/nQhz7kYj+pCxj+ktpibGyMW265hdtuu22v1f5ve9vbfAEgdZjT/pLawtX+Uvcy/CW1hav9pe5l+EtqC1f7S93L8JfUFq72l7qXC/4ktcX0or41a9YwMTFBb28vIyMjLvaTukBkZqfH8JLo7+/PzZs3d3oYUpUajQYDAwOdHoZUnYjYkpn9M+tO+0uSVBnDX5Kkyhj+kiRVxvCXJKkyhr8kSZUx/CVJqozhL6ltxsbG6OvrY+nSpfT19TE2NtbpIUnCm/xIapOxsTGGh4cZHR3d61v9AG/0I3XYIV/5R8RDEXF/RNwbEZtL7TURsTEiHiw/Tyz1iIjPRsS2iLgvIt7ScpxVpf2DEbGqpX52Of620jf2dw5J3cFv9ZO61+Ga9h/MzLNa7iJ0FXB7Zi4Gbi/PAS4EFpfHauA6aAY5cDVwDvBW4OqWML+utJ3ud8Ec55DUBSYmJpicnNxr2n9yctJv9ZO6QLum/ZcDA2X7BqABXFnqN2bznsJ3RcQJEXFyabsxM3cCRMRG4IKIaACvyszvlPqNwHuA2/ZzDkldYMGCBVx55ZV84QtfeGHa/wMf+AALFizo9NCk6h2OK/8E/joitkTE6lKbn5k7AMrP15f6KcAjLX0nS21/9clZ6vs7h6QuMfO7Q2r5LhGp2x2OK//zMnN7RLwe2BgR/7CftjFLLQ+ifkDKi5HVAPPnz6fRaBxoV0mHaPv27Vx55ZVcdtllPPzww5x66qlceumlXHPNNf4tSh12yOGfmdvLz8ci4hs037N/NCJOzswdZVr/sdJ8EljU0n0hsL3UB2bUG6W+cJb27OccrWNbC6yF5rf6+a1i0kunt7eXd7zjHfzhH/7hC9/qNz4+zvr16/2GP6nDDmnaPyJeERGvnN4GlgFbgfXA9Ir9VcDNZXs9cElZ9X8u8GSZst8ALIuIE8tCv2XAhrLvqYg4t6zyv2TGsWY7h6QuMDw8zNDQEOPj40xNTTE+Ps7Q0BDDw8OdHppUvUO98p8PfKN8+m4e8MXM/KuIuBv4ckQMAQ8D7y/tbwXeCWwDdgOXAmTmzoj4FHB3affJ6cV/wO8A1wPH0Vzod1upf3of55DUBaY/y79mzRomJibo7e1lZGTEz/hLXSBqWYDT39+fmzdv7vQwpCpNT/tLemlFxJaWj+G/wNv7SpJUGcNfkqTKGP6SJFXG8JckqTKGv6S28St9pe7kV/pKagu/0lfqXl75S2oLv9JX6l6Gv6S2mJiYYMmSJXvVlixZ4lf6Sl3A8JfUFr29vWzatGmv2qZNm+jt7e3QiCRNM/wltYX39pe6lwv+JLWF9/aXupf39pfUdt7bX+oM7+0vSZIAw1+SpOoY/pIkVcbwlySpMoa/JEmVMfwlSaqM4S9JUmUMf0mSKmP4S5JUGcNfkqTKGP6SJFXG8JckqTKGvyRJlTH8JUmqjOEvSVJlDH9Jkipj+EuSVBnDX5Kkyhj+kiRVxvCXJKkyhr+kthkbG6Ovr4+lS5fS19fH2NhYp4ckCZjX6QFIOjqNjY0xPDzM6Ogoe/bsoaenh6GhIQBWrlzZ4dFJdfPKX1JbjIyMMDo6yuDgIPPmzWNwcJDR0VFGRkY6PTSpeoa/pLaYmJhgcnJyr2n/yclJJiYmOj00qXqGv6S2WLBgAR/96Ed5+umnyUyefvppPvrRj7JgwYJOD02qnuEvqS12797NT3/6U9asWcOtt97KmjVr+OlPf8ru3bs7PTSpeoa/pLbYuXMnV1xxBevWreNd73oX69at44orrmDnzp2dHppUPcNfUtu8/e1vZ+vWrdx+++1s3bqVt7/97Z0ekiQMf0ltsnDhQlatWsX4+DhTU1OMj4+zatUqFi5c2OmhSdU76PCPiEURMR4RExHxQET8bqn/fkT8OCLuLY93tvT5eERsi4gfRMT5LfULSm1bRFzVUn9DRHw3Ih6MiC9FxMtL/djyfFvZf9rB/h6S2uPaa69lamqKyy67jPPPP5/LLruMqakprr322k4PTareoVz5TwH/ITN7gXOBD0fEGWXfZzLzrPK4FaDsWwG8CbgA+NOI6ImIHuBzwIXAGcDKluNcU461GHgCGCr1IeCJzDwd+ExpJ6mLrFy5kosuuogdO3bw/PPPs2PHDi666CJv8CN1gYMO/8zckZl/V7afAiaAU/bTZTlwU2Y+m5n/CGwD3loe2zLzR5n5L8BNwPKICODtwFdL/xuA97Qc64ay/VVgaWkvqUuMjY1xyy23cNttt7Fx40Zuu+02brnlFm/xK3WBw/Kef5l2/1Xgu6X0kYi4LyLWRcSJpXYK8EhLt8lS21f9tcBPMnNqRn2vY5X9T5b2krrEyMgIF198MWvWrOH8889nzZo1XHzxxd7hT+oCh3xv/4g4Hvga8HuZ+dOIuA74FJDl5x8DlwGzXZkns78Ayf20Z459rWNbDawGmD9/Po1GY7+/i6TD5/vf/z4//vGPOfbYY8lMHn/8cT73uc/x5JNP+rcoddghhX9EHEMz+L+QmV8HyMxHW/b/OfCt8nQSWNTSfSGwvWzPVn8cOCEi5pWr+9b208eajIh5wKuBf/Xh4cxcC6wF6O/vz4GBgYP+XSW9OD09PTz33HOccMIJABx33HHs2rWLnp4e/FuUOutQVvsHMApMZOZ/aqmf3NLsvcDWsr0eWFFW6r8BWAx8D7gbWFxW9r+c5qLA9ZmZwDjwvtJ/FXBzy7FWle33AX9T2kvqElNTU+zevXuvO/zt3r2bqampuTtLaqtDufI/D/g/gPsj4t5S+480V+ufRQtWozAAAAiDSURBVHMa/iHgtwEy84GI+DLwfZqfFPhwZu4BiIiPABuAHmBdZj5QjnclcFNE/AFwD80XG5SffxER22he8a84hN9DUpusWLGCdevWMTExQW9vLytWrHDBn9QFopYL5v7+/ty8eXOnhyFVIyL4xV/8Rb74xS+yZ88eenp6uPjii/nnf/5navl/R+q0iNiSmf0z64e84E+SZrNw4UJ27drFZZddxsMPP8ypp57KM8884x3+pC7g7X0ltcW1117LM888w0MPPcTzzz/PQw89xDPPPOMd/qQuYPhLaos777yTZ599lp6eHqC5+v/ZZ5/lzjvv7PDIJPmev6S2OOaYY3jlK1/J1772tRfe8/+t3/otnnrqKZ577rlOD0+qgu/5S3pJTU1Ncdppp7F06VIyk4jgrLPO4p577un00KTqOe0vqW3uueceLr/8cr75zW9y+eWXG/xSlzD8JbXV6aefzrx58zj99NM7PRRJhdP+ktrm+OOP56qrruK5557jmGOO4fjjj2fXrl2dHpZUPa/8JbVFRHDOOefwxje+kZe97GW88Y1v5JxzzsFv35Y6z9X+kvarW8K6lv+rpMNpX6v9vfKXtF+ZedCPZcuWvfDiISJYtmzZQR9L0uFj+Etqmw0bNvD888/zS1d+i+eff54NGzZ0ekiSMPwlSaqO4S9JUmUMf0mSKmP4S5JUGcNfkqTKGP6SJFXG8JckqTKGvyRJlTH8JUmqjOEvSVJlDH9Jkipj+EuSVBnDX5Kkyhj+kiRVxvCXJKky8zo9AEnt82/+77/myZ891+lhAHDaVbd09PyvPu4Y/v7qZR0dg9QtDH/pKPbkz57joU+/q9PDoNFoMDAw0NExdPrFh9RNnPaXJKkyhr8kSZUx/CVJqozhL0lSZVzwJx3FXtl7FW++4apOD6Pphs6e/pW9AJ1f/Ch1A8NfOoo9NfFpV/sXrvaXfs5pf0mSKmP4S5JUGcNfkqTKGP6SJFXGBX/SUa5rFrr9Vefv7S+p6YgO/4i4APh/gB7gv2bmpzs8JKmrdMNKf2i+AOmWsUg6gqf9I6IH+BxwIXAGsDIizujsqCRJ6n5HbPgDbwW2ZeaPMvNfgJuA5R0ekyRJXe9IDv9TgEdank+WmiRJ2o8j+T3/mKWWezWIWA2sBpg/fz6NRuMlGJZ0dBkcHDwsx4lrDq3/+Pj4YRmHpCM7/CeBRS3PFwLbWxtk5lpgLUB/f392+vai0pEoM+duNIduuL2vpJ87kqf97wYWR8QbIuLlwApgfYfHJElS1ztir/wzcyoiPgJsoPlRv3WZ+UCHhyVJUtc7YsMfIDNvBW7t9DgkSTqSHMnT/pIk6SAY/pIkVcbwlySpMoa/JEmVMfwlSaqM4S9JUmUMf0mSKmP4S5JUGcNfkqTKGP6SJFXG8JckqTKGvyRJlTH8JUmqjOEvSVJlDH9JkioTmdnpMbwkIuK/A//U6XFIlToJeLzTg5Aq9EuZ+bqZxWrCX1LnRMTmzOzv9DgkNTntL0lSZQx/SZIqY/hLeims7fQAJP2c7/lLklQZr/wlSaqM4S8d4SLiFyPipoj4YUR8PyJujYg3dmAcH4yIBS+yz2kRsbVs/0JEfCEi7o+IrRGxKSKOn6P/QxFx0qGMW6rRvE4PQNLBi4gAvgHckJkrSu0sYD7w317CcfQAHwS2AtsP8jC/CzyamW8ux/wV4LnDMsAiInoyc8/hPKZ0JPLKXzqyDQLPZeafTRcy815gU0T8UbmCvj8iLgKIiIGIaETEVyPiH8qVdkTEhRHx5eljlHbfLNvLIuI7EfF3EfGV6avxctX9f0XEJmAl0A98ISLujYjjIuLsiPh2RGyJiA0RcXLpd3ZE/H1EfAf4cMvvcjLw45bf4weZ+Wzp85flOA9ExOrZ/iH21SYidkXEJyPiu8AnIuIbLfveERFfP8h/e+mIZfhLR7Y+YMss9X8PnAX8G+A3gT+aDl/gV4HfA84Afhk4D9gInBsRryhtLgK+VKbUPwH8Zma+BdgM/J8t53kmM5dk5v9b9n0gM88CpoA/Ad6XmWcD64CR0ufzwEcz89dmjHkdcGV5ofEHEbG4Zd9l5Tj9wEcj4rWz/M77avMKYGtmngN8EuiNiOk7nl1axiNVxfCXjk5LgLHM3JOZjwLfBv5t2fe9zJzMzOeBe4HTMnMK+Cvgf42IecC7gJuBc2m+SLgjIu4FVgG/1HKeL+3j/L9C84XJxtLvE8DCiHg1cEJmfru0+4vpDmXG4peBPwJeA9wdEb1l90cj4u+Bu4BFQOsLA+Zoswf4WjlHlnP+7xFxAvBrwG37+B2ko5bv+UtHtgeA981Sj/30ebZlew8//3/gSzSn4XcCd2fmU2VNwcbMXLmPYz29j3oAD8y8ui+Bu8/PF2fmLuDrwNcj4nngnRExn+bsxa9l5u6IaAD/04zjDuynzTMz3uf/PPBN4BngK+WFj1QVr/ylI9vfAMdGxIemCxHxb4EngIsioqdMcf8G8L05jtUA3gJ8iJ9f0d8FnBcRp5dj/8J+PknwFPDKsv0D4HUR8Wul3zER8abM/AnwZEQsKe0+0DLu8yLixLL9cpozDv8EvBp4ooT6/0JzNmKmA2kDQGZup7ko8RPA9ftqJx3NvPKXjmCZmRHxXuA/R8RVNK9mH6L5nv7xwN/TvNK+IjP/uQTjvo61JyK+RXPV/qpS++8R8UFgLCKOLU0/weyfJLge+LOI+BnN6fT3AZ8tU/3zgP9Mc6biUmBdROwGNrT0/5+B68psw8uAW2hO178cuDwi7qP5ouKuWc79VwfQptUXgNdl5vfnaCcdlbzDn6TqRMR/Ae7JzNFOj0XqBMNfUlUiYgvNtQrvmP4ooVQbw1+SpMq44E+SpMoY/pIkVcbwlySpMoa/JEmVMfwlSaqM4S9JUmX+fzLYZgGtbNkrAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "so_numeric_df[['ConvertedSalary']].boxplot();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### What does your data look like? (II)\n",
    "In the previous exercise you looked at the distribution of individual columns. While this is a good start, a more detailed view of how different features interact with each other may be useful as this can impact your decision on what to transform and how."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 540x540 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot pairwise relationships\n",
    "sns.pairplot(so_numeric_df);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       ConvertedSalary         Age  Years Experience\n",
      "count     6.650000e+02  999.000000        999.000000\n",
      "mean      9.256517e+04   36.003003          9.961962\n",
      "std       2.091344e+05   13.255127          4.878129\n",
      "min       0.000000e+00   18.000000          0.000000\n",
      "25%       2.755000e+04   25.000000          7.000000\n",
      "50%       5.556200e+04   35.000000         10.000000\n",
      "75%       8.823800e+04   45.000000         13.000000\n",
      "max       2.000000e+06   83.000000         27.000000\n"
     ]
    }
   ],
   "source": [
    "# Print summary statistics\n",
    "print(so_numeric_df.describe())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Scaling and transformations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Normalization\n",
    "As discussed in the video, in normalization you linearly scale the entire column between 0 and 1, with 0 corresponding with the lowest value in the column, and 1 with the largest.\n",
    "When using scikit-learn (the most commonly used machine learning library in Python) you can use a `MinMaxScaler` to apply normalization. (It is called this as it scales your values between a minimum and maximum value.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\kcsgo\\anaconda3\\lib\\site-packages\\ipykernel_launcher.py:10: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  # Remove the CWD from sys.path while we load stuff.\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>Age_MM</th>\n",
       "      <th>Age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.046154</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.307692</td>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.415385</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.430769</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.323077</td>\n",
       "      <td>39</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Age_MM  Age\n",
       "0  0.046154   21\n",
       "1  0.307692   38\n",
       "2  0.415385   45\n",
       "3  0.430769   46\n",
       "4  0.323077   39"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "# Instantiate MinMaxScaler\n",
    "MM_scaler = MinMaxScaler()\n",
    "\n",
    "# Fit MM_scaler to the data\n",
    "MM_scaler.fit(so_numeric_df[['Age']])\n",
    "\n",
    "# Transform the data using the fitted scaler\n",
    "so_numeric_df['Age_MM'] = MM_scaler.transform(so_numeric_df[['Age']])\n",
    "\n",
    "# Compare the original and transformed column\n",
    "so_numeric_df[['Age_MM', 'Age']].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Standardization\n",
    "While normalization can be useful for scaling a column between two data points, it is hard to compare two scaled columns if even one of them is overly affected by outliers. One commonly used solution to this is called standardization, where instead of having a strict upper and lower bound, you center the data around its mean, and calculate the number of standard deviations away from mean each data point is.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\kcsgo\\anaconda3\\lib\\site-packages\\ipykernel_launcher.py:10: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  # Remove the CWD from sys.path while we load stuff.\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>Age_SS</th>\n",
       "      <th>Age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.132431</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.150734</td>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.679096</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.754576</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.226214</td>\n",
       "      <td>39</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Age_SS  Age\n",
       "0 -1.132431   21\n",
       "1  0.150734   38\n",
       "2  0.679096   45\n",
       "3  0.754576   46\n",
       "4  0.226214   39"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "# Instantiate StandardScaler\n",
    "SS_scaler = StandardScaler()\n",
    "\n",
    "# Fit SS_scaler to the data\n",
    "SS_scaler.fit(so_numeric_df[['Age']])\n",
    "\n",
    "# Transform the data using the fitted scaler\n",
    "so_numeric_df['Age_SS'] = SS_scaler.transform(so_numeric_df[['Age']])\n",
    "\n",
    "# Compare the original and transformed column\n",
    "so_numeric_df[['Age_SS', 'Age']].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Log transformation\n",
    "In the previous exercises you scaled the data linearly, which will not affect the data's shape. This works great if your data is normally distributed (or closely normally distributed), an assumption that a lot of machine learning models make. Sometimes you will work with data that closely conforms to normality, e.g the height or weight of a population. On the other hand, many variables in the real world do not follow this pattern e.g, wages or age of a population. In this exercise you will use a log transform on the `ConvertedSalary` column in the `so_numeric_df` DataFrame as it has a large amount of its data centered around the lower values, but contains very high values also. These distributions are said to have a long right tail.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\kcsgo\\anaconda3\\lib\\site-packages\\ipykernel_launcher.py:10: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  # Remove the CWD from sys.path while we load stuff.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.preprocessing import PowerTransformer\n",
    "\n",
    "# Instantiate PowerTransformer\n",
    "pow_trans = PowerTransformer()\n",
    "\n",
    "# Train the transform on the data\n",
    "pow_trans.fit(so_numeric_df[['ConvertedSalary']])\n",
    "\n",
    "# Apply the power transform to the data\n",
    "so_numeric_df['ConvertedSalary_LG'] = pow_trans.transform(so_numeric_df[['ConvertedSalary']])\n",
    "\n",
    "# Plot the data before and after the transformation\n",
    "so_numeric_df[['ConvertedSalary', 'ConvertedSalary_LG']].hist();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Removing outliers\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Percentage based outlier removal\n",
    "One way to ensure a small portion of data is not having an overly adverse effect is by removing a certain percentage of the largest and/or smallest values in the column. This can be achieved by finding the relevant quantile and trimming the data using it with a mask. This approach is particularly useful if you are concerned that the highest values in your dataset should be avoided. When using this approach, you must remember that even if there are no outliers, this will still remove the same top N percentage from the dataset.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Find the 95th quantile\n",
    "quantile = so_numeric_df['ConvertedSalary'].quantile(0.95)\n",
    "\n",
    "# Trim the outlier\n",
    "trimmed_df = so_numeric_df[so_numeric_df['ConvertedSalary'] < quantile]\n",
    "\n",
    "# The original histogram\n",
    "so_numeric_df[['ConvertedSalary']].hist();\n",
    "\n",
    "# The trimmed histogram\n",
    "trimmed_df[['ConvertedSalary']].hist();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Statistical outlier removal\n",
    "While removing the top N% of your data is useful for ensuring that very spurious points are removed, it does have the disadvantage of always removing the same proportion of points, even if the data is correct. A commonly used alternative approach is to remove data that sits further than three standard deviations from the mean. You can implement this by first calculating the mean and standard deviation of the relevant column to find upper and lower bounds, and applying these bounds as a mask to the DataFrame. This method ensures that only data that is genuinely different from the rest is removed, and will remove fewer points if the data is close together.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Find the mean and standard dev\n",
    "std = so_numeric_df['ConvertedSalary'].std()\n",
    "mean = so_numeric_df['ConvertedSalary'].mean()\n",
    "\n",
    "# Calculate the cutoff\n",
    "cut_off = std * 3\n",
    "lower, upper = mean - cut_off, mean + cut_off\n",
    "\n",
    "# Trim the outlier\n",
    "trimmed_df = so_numeric_df[\n",
    "    (so_numeric_df['ConvertedSalary'] < upper) &\n",
    "    (so_numeric_df['ConvertedSalary'] > lower)\n",
    "]\n",
    "\n",
    "# Trimmed box plot\n",
    "trimmed_df[['ConvertedSalary']].boxplot();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Scaling and transforming new data\n",
    "- Why only use training data during scaling?\n",
    "    - Data leakage: Using data that you won't have access to when assessing the performance of your model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train and testing transformations (I)\n",
    "So far you have created scalers based on a column, and then applied the scaler to the same data that it was trained on. When creating machine learning models you will generally build your models on historic data (train set) and apply your model to new unseen data (test set). In these cases you will need to ensure that the same scaling is being applied to both the training and test data.\n",
    "To do this in practice you train the scaler on the train set, and keep the trained scaler to apply it to the test set. You should never retrain a scaler on the test set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "so_numeric_df = pd.read_csv('./dataset/Combined_DS_v10.csv')[['ConvertedSalary', 'Age', 'Years Experience']]\n",
    "\n",
    "so_train_numeric, so_test_numeric = train_test_split(so_numeric_df, test_size=0.3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\kcsgo\\anaconda3\\lib\\site-packages\\ipykernel_launcher.py:8: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \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>Age</th>\n",
       "      <th>Age_ss</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>380</th>\n",
       "      <td>42</td>\n",
       "      <td>0.445336</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>287</th>\n",
       "      <td>35</td>\n",
       "      <td>-0.082274</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>851</th>\n",
       "      <td>33</td>\n",
       "      <td>-0.233020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>219</th>\n",
       "      <td>18</td>\n",
       "      <td>-1.363613</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>252</th>\n",
       "      <td>30</td>\n",
       "      <td>-0.459138</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Age    Age_ss\n",
       "380   42  0.445336\n",
       "287   35 -0.082274\n",
       "851   33 -0.233020\n",
       "219   18 -1.363613\n",
       "252   30 -0.459138"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Apply a standard scaler to the data\n",
    "SS_scaler = StandardScaler()\n",
    "\n",
    "# Fit the standard scaler to the data\n",
    "SS_scaler.fit(so_train_numeric[['Age']])\n",
    "\n",
    "# Transform the test data using the fitted scaler\n",
    "so_test_numeric['Age_ss'] = SS_scaler.transform(so_test_numeric[['Age']])\n",
    "so_test_numeric[['Age', 'Age_ss']].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train and testing transformations (II)\n",
    "Similar to applying the same scaler to both your training and test sets, if you have removed outliers from the train set, you probably want to do the same on the test set as well. Once again you should ensure that you use the thresholds calculated only from the train set to remove outliers from the test set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_std = so_train_numeric['ConvertedSalary'].std()\n",
    "train_mean = so_train_numeric['ConvertedSalary'].mean()\n",
    "\n",
    "cut_off = train_std * 3\n",
    "train_lower, train_upper = train_mean - cut_off, train_mean + cut_off\n",
    "\n",
    "# Trim the test DataFrame\n",
    "trimmed_df = so_test_numeric[(so_test_numeric['ConvertedSalary'] < train_upper) &\n",
    "                             (so_test_numeric['ConvertedSalary'] > train_lower)]"
   ]
  }
 ],
 "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}