Data Preprocessing.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Created by Berkay Alan**\n",
    "\n",
    "**Data Preprocessing**\n",
    "\n",
    "**24 March 2021**\n",
    "\n",
    "**For more Tutorial:** https://github.com/berkayalan"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Content\n",
    "\n",
    "- **Data Cleaning / Cleasing**\n",
    "    - Noisy Data\n",
    "    - Missing Data Analysis\n",
    "    - Outlier Detection\n",
    "- **Data Standardization / Feature Scaling**\n",
    "    - Normalization(0-1 Scaling)\n",
    "    - Standardization(Z Score Scaling)\n",
    "    - Min-Max Scaling\n",
    "    - Binary Transformation\n",
    "- **Variable Transformation**\n",
    "    - Label Encoding\n",
    "    - One Hot Encoding"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Resources"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- [Kaggle Data Cleaning Course](https://www.kaggle.com/learn/data-cleaning)\n",
    "- [How to Identify Outliers in your Data](https://machinelearningmastery.com/how-to-identify-outliers-in-your-data/)\n",
    "- [A Brief Overview of Outlier Detection Techniques](https://towardsdatascience.com/a-brief-overview-of-outlier-detection-techniques-1e0b2c19e561)\n",
    "- [A Comprehensive Guide to Data Exploration](https://www.analyticsvidhya.com/blog/2016/01/guide-data-exploration/)\n",
    "- [Outlier Detection Methods](https://docs.oracle.com/cd/E17236_01/epm.1112/cb_statistical/frameset.htm?ch07s02s10s01.html#:~:text=If%20a%20value%20is%20a,deviations%20is%20called%20the%20threshold.&text=This%20method%20can%20fail%20to,outliers%20increase%20the%20standard%20deviation.)\n",
    "- [Detecting And Treating Outliers In Python — Part 2](https://towardsdatascience.com/local-outlier-factor-lof-algorithm-for-outlier-identification-8efb887d9843)\n",
    "- [Outlier detection with Local Outlier Factor(LOF)](https://scikit-learn.org/stable/auto_examples/neighbors/plot_lof_outlier_detection.html#:~:text=The%20Local%20Outlier%20Factor%20(LOF,lower%20density%20than%20their%20neighbors.))\n",
    "- [About Feature Scaling and Normalization](https://sebastianraschka.com/Articles/2014_about_feature_scaling.html)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## Importing Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 389,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 390,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 391,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import math"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 392,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 393,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 394,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "from sklearn.neighbors import LocalOutlierFactor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 395,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import missingno as msno"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 396,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "from ycimpute.imputer import knnimput"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 397,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "from ycimpute.imputer import iterforest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 398,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "from ycimpute.imputer import EM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 410,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "from sklearn import preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 459,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import LabelEncoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 477,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import OneHotEncoder"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## Data Cleaning / Cleasing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "### Outlier Detection"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "In statistics, an outlier is a data point that differs significantly from other observations. An outlier may be due to variability in the measurement or it may indicate experimental error; the latter are sometimes excluded from the data set. An outlier can cause serious problems in statistical analyses."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**How to Identify Outliers in our Data?**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**Mean and Standard Deviation Method:** For this outlier detection method, the mean and standard deviation of the residuals are calculated and compared. If a value is a certain number of standard deviations away from the mean, that data point is identified as an outlier. The specified number of standard deviations is called the threshold. The default value is 3. This method can fail to detect outliers because the outliers increase the standard deviation. The more extreme the outlier, the more the standard deviation is affected.\n",
    "\n",
    "**Z Score:** Z score is an important concept in statistics. Z score is also called standard score. This score helps to understand if a data value is greater or smaller than mean and how far away it is from the mean. More specifically, Z score tells how many standard deviations away a data point is from the mean.\n",
    "\n",
    "**IQR-Boxplot:** For this outlier detection method, the median of the residuals is calculated, along with the 25th percentile and the 75th percentile. The difference between the 25th and 75th percentile is the interquartile deviation (IQD). Then, the difference is calculated between each historical value and the residual median. If the historical value is a certain number of Median Absolute Deviation(MAD) away from the median of the residuals, that value is classified as an outlier. The default threshold is 2.22, which is equivalent to 3 standard deviations or MADs. This method is somewhat susceptible to influence from extreme outliers, but less so than the mean and standard deviation method. Box plots are based on this approach. The median and interquartile deviation method can be used for both symmetric and asymmetric data.\n",
    "\n",
    "- lower inner fence: Q1 - 1.5*IQ\n",
    "- upper inner fence: Q3 + 1.5*IQ\n",
    "- lower outer fence: Q1 - 3*IQ\n",
    "- upper outer fence: Q3 + 3*IQ"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**Outlier Detection**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "For a real world example, we will use *diamonds* dataset from seaborn."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "hidden": 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>carat</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   carat  depth  table  price     x     y     z\n",
       "0   0.23   61.5   55.0    326  3.95  3.98  2.43\n",
       "1   0.21   59.8   61.0    326  3.89  3.84  2.31\n",
       "2   0.23   56.9   65.0    327  4.05  4.07  2.31\n",
       "3   0.29   62.4   58.0    334  4.20  4.23  2.63\n",
       "4   0.31   63.3   58.0    335  4.34  4.35  2.75"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = sns.load_dataset(\"diamonds\")\n",
    "df = df.select_dtypes(include=[\"float64\",\"int64\"])\n",
    "df.dropna(inplace=True)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(53940, 7)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    55.0\n",
       "1    61.0\n",
       "2    65.0\n",
       "3    58.0\n",
       "4    58.0\n",
       "Name: table, dtype: float64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_table = df[\"table\"]\n",
    "df_table.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Let's plot table variable in order to see outliers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "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": [
    "sns.boxplot(x=df_table)\n",
    "plt.title(\"Boxplot of Table Variable of Diamonds Dataset\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    53940.000000\n",
       "mean        57.457184\n",
       "std          2.234491\n",
       "min         43.000000\n",
       "25%         56.000000\n",
       "50%         57.000000\n",
       "75%         59.000000\n",
       "max         95.000000\n",
       "Name: table, dtype: float64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_table.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Let's calculate the IQR."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Q1:  56.0\n",
      "Q3:  59.0\n",
      "IQR:  3.0\n"
     ]
    }
   ],
   "source": [
    "quantile1 = df_table.quantile(0.25)\n",
    "quantile3 = df_table.quantile(0.75)\n",
    "IQR = quantile3-quantile1\n",
    "\n",
    "print(\"Q1: \", quantile1)\n",
    "print(\"Q3: \", quantile3)\n",
    "print(\"IQR: \", IQR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "lower_inner_fence = quantile1 - 1.5*IQR\n",
    "upper_inner_fence = quantile3 + 1.5*IQR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lower_inner_fence 51.5\n",
      "upper_inner_fence 63.5\n"
     ]
    }
   ],
   "source": [
    "print(\"lower_inner_fence\",lower_inner_fence)\n",
    "print(\"upper_inner_fence\",upper_inner_fence)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of Outliers:  605\n"
     ]
    }
   ],
   "source": [
    "print(\"Number of Outliers: \",len(df_table[(df_table<lower_inner_fence) | (df_table>upper_inner_fence)]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**Solutions for Bivariate Outliers**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**1. Deleting the outlier observations**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "hidden": 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>table</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>55.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>61.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>65.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>58.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>58.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   table\n",
       "0   55.0\n",
       "1   61.0\n",
       "2   65.0\n",
       "3   58.0\n",
       "4   58.0"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_table = pd.DataFrame(df_table)\n",
    "df_table.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(53940, 1)"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_table.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "new_df = df_table[~((df_table<lower_inner_fence) | (df_table>upper_inner_fence)).any(axis=1)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(53335, 1)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**2. Filling with Average**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "hidden": 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>carat</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   carat  depth  table  price     x     y     z\n",
       "0   0.23   61.5   55.0    326  3.95  3.98  2.43\n",
       "1   0.21   59.8   61.0    326  3.89  3.84  2.31\n",
       "2   0.23   56.9   65.0    327  4.05  4.07  2.31\n",
       "3   0.29   62.4   58.0    334  4.20  4.23  2.63\n",
       "4   0.31   63.3   58.0    335  4.34  4.35  2.75"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = sns.load_dataset(\"diamonds\")\n",
    "df = df.select_dtypes(include=[\"float64\",\"int64\"])\n",
    "df.dropna(inplace=True)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    55.0\n",
       "1    61.0\n",
       "2    65.0\n",
       "3    58.0\n",
       "4    58.0\n",
       "Name: table, dtype: float64"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_table = df[\"table\"]\n",
    "df_table.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-50-be31fa38afc0>:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\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",
      "  df_table[(df_table<lower_inner_fence) | (df_table>upper_inner_fence)] = df_table.mean()\n"
     ]
    }
   ],
   "source": [
    "df_table[(df_table<lower_inner_fence) | (df_table>upper_inner_fence)] = df_table.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Series([], Name: table, dtype: float64)"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_table[(df_table<lower_inner_fence) | (df_table>upper_inner_fence)] # There is no outlier after filling."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**3. Suppression of the Outliers**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "If a outlier is close to upper bound, we make it upper bound. If it's close to lower bound, we make it lower bound."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "hidden": 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>carat</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   carat  depth  table  price     x     y     z\n",
       "0   0.23   61.5   55.0    326  3.95  3.98  2.43\n",
       "1   0.21   59.8   61.0    326  3.89  3.84  2.31\n",
       "2   0.23   56.9   65.0    327  4.05  4.07  2.31\n",
       "3   0.29   62.4   58.0    334  4.20  4.23  2.63\n",
       "4   0.31   63.3   58.0    335  4.34  4.35  2.75"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = sns.load_dataset(\"diamonds\")\n",
    "df = df.select_dtypes(include=[\"float64\",\"int64\"])\n",
    "df.dropna(inplace=True)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    55.0\n",
       "1    61.0\n",
       "2    65.0\n",
       "3    58.0\n",
       "4    58.0\n",
       "Name: table, dtype: float64"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_table = df[\"table\"]\n",
    "df_table.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2        65.0\n",
       "91       69.0\n",
       "145      64.0\n",
       "219      64.0\n",
       "227      67.0\n",
       "         ... \n",
       "53695    65.0\n",
       "53697    65.0\n",
       "53756    64.0\n",
       "53757    64.0\n",
       "53785    65.0\n",
       "Name: table, Length: 605, dtype: float64"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_table[(df_table<lower_inner_fence) | (df_table>upper_inner_fence)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-62-b792e6eea236>:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\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",
      "  df_table[df_table<lower_inner_fence] =lower_inner_fence\n",
      "<ipython-input-62-b792e6eea236>:2: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\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",
      "  df_table[df_table>upper_inner_fence] =upper_inner_fence\n"
     ]
    }
   ],
   "source": [
    "df_table[df_table<lower_inner_fence] =lower_inner_fence\n",
    "df_table[df_table>upper_inner_fence] =upper_inner_fence"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**Solutions for Multivariate Outliers**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    " A multivariate outlier is a combination of unusual scores on at least two variables. Both types of outliers can influence the outcome of statistical analyses. For example, it could be a human with a height measurement of 2 meters (in the 95th percentile) and weight measurement of 50kg (in the 5th percentile)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**1. Local outlier factor (LOF):** The Local Outlier Factor (LOF) algorithm is an unsupervised anomaly detection method which computes the local density deviation of a given data point with respect to its neighbors. It considers as outliers the samples that have a substantially lower density than their neighbors. "
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "![image.png](attachment:image.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "hidden": 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>total_bill</th>\n",
       "      <th>tip</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>16.99</td>\n",
       "      <td>1.01</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10.34</td>\n",
       "      <td>1.66</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>21.01</td>\n",
       "      <td>3.50</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23.68</td>\n",
       "      <td>3.31</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>24.59</td>\n",
       "      <td>3.61</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   total_bill   tip  size\n",
       "0       16.99  1.01     2\n",
       "1       10.34  1.66     3\n",
       "2       21.01  3.50     3\n",
       "3       23.68  3.31     2\n",
       "4       24.59  3.61     4"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = sns.load_dataset(\"tips\")\n",
    "df = df.select_dtypes(include=[\"float64\",\"int64\"])\n",
    "df.dropna(inplace=True)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "lof = LocalOutlierFactor(n_neighbors=20, contamination=0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,\n",
       "        1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,\n",
       "        1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,\n",
       "        1,  1,  1,  1,  1,  1,  1,  1, -1,  1,  1,  1,  1,  1,  1,  1, -1,\n",
       "        1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,\n",
       "        1,  1,  1,  1,  1,  1,  1, -1,  1,  1,  1,  1,  1,  1,  1,  1,  1,\n",
       "       -1,  1,  1,  1,  1,  1,  1, -1,  1, -1,  1,  1,  1,  1,  1,  1,  1,\n",
       "        1,  1,  1,  1,  1,  1,  1, -1,  1,  1,  1,  1,  1,  1,  1,  1, -1,\n",
       "        1,  1,  1,  1,  1,  1,  1,  1,  1, -1,  1,  1,  1, -1,  1,  1,  1,\n",
       "        1,  1,  1, -1,  1,  1, -1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,\n",
       "       -1,  1, -1,  1,  1,  1,  1,  1, -1,  1,  1,  1, -1, -1, -1, -1,  1,\n",
       "        1,  1,  1,  1,  1,  1,  1,  1, -1,  1, -1,  1,  1,  1,  1,  1,  1,\n",
       "        1,  1,  1,  1,  1,  1,  1,  1, -1,  1,  1,  1,  1,  1, -1,  1,  1,\n",
       "        1, -1,  1,  1,  1,  1,  1,  1,  1,  1,  1, -1,  1,  1,  1,  1,  1,\n",
       "        1,  1,  1,  1,  1,  1])"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lof.fit_predict(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "df_scores = lof.negative_outlier_factor_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-1.2259471 , -1.00746956, -1.00220147, -1.02155569, -0.99932425,\n",
       "       -1.02361463, -1.22935985, -1.03244471, -1.0201844 , -1.06164523])"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_scores[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-3.30666549, -2.33154258, -2.18736263, -1.98330455, -1.81981065,\n",
       "       -1.79815601, -1.73853455, -1.67893803, -1.67288622, -1.64176048,\n",
       "       -1.62372071, -1.56936213, -1.49508486, -1.48216547, -1.47085665,\n",
       "       -1.41998804, -1.4016868 , -1.368511  , -1.35970075, -1.35623266])"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sort(df_scores)[:20]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Let's specify a treshold value randomly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "treshold = np.sort(df_scores)[7]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "outliers = df_scores>treshold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "new_df = df[outliers]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "hidden": 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>total_bill</th>\n",
       "      <th>tip</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>16.99</td>\n",
       "      <td>1.01</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10.34</td>\n",
       "      <td>1.66</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>21.01</td>\n",
       "      <td>3.50</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23.68</td>\n",
       "      <td>3.31</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>24.59</td>\n",
       "      <td>3.61</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>239</th>\n",
       "      <td>29.03</td>\n",
       "      <td>5.92</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>240</th>\n",
       "      <td>27.18</td>\n",
       "      <td>2.00</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>241</th>\n",
       "      <td>22.67</td>\n",
       "      <td>2.00</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>242</th>\n",
       "      <td>17.82</td>\n",
       "      <td>1.75</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>243</th>\n",
       "      <td>18.78</td>\n",
       "      <td>3.00</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>236 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     total_bill   tip  size\n",
       "0         16.99  1.01     2\n",
       "1         10.34  1.66     3\n",
       "2         21.01  3.50     3\n",
       "3         23.68  3.31     2\n",
       "4         24.59  3.61     4\n",
       "..          ...   ...   ...\n",
       "239       29.03  5.92     3\n",
       "240       27.18  2.00     2\n",
       "241       22.67  2.00     2\n",
       "242       17.82  1.75     2\n",
       "243       18.78  3.00     2\n",
       "\n",
       "[236 rows x 3 columns]"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "hidden": 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>total_bill</th>\n",
       "      <th>tip</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>3.07</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>92</th>\n",
       "      <td>5.75</td>\n",
       "      <td>1.00</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>111</th>\n",
       "      <td>7.25</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>170</th>\n",
       "      <td>50.81</td>\n",
       "      <td>10.00</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>172</th>\n",
       "      <td>7.25</td>\n",
       "      <td>5.15</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>178</th>\n",
       "      <td>9.60</td>\n",
       "      <td>4.00</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>212</th>\n",
       "      <td>48.33</td>\n",
       "      <td>9.00</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     total_bill    tip  size\n",
       "67         3.07   1.00     1\n",
       "92         5.75   1.00     2\n",
       "111        7.25   1.00     1\n",
       "170       50.81  10.00     3\n",
       "172        7.25   5.15     2\n",
       "178        9.60   4.00     2\n",
       "212       48.33   9.00     4"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df_scores<treshold] # Let's see outliers"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**2. Suppression of the Outliers**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "hidden": 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>total_bill</th>\n",
       "      <th>tip</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>59</th>\n",
       "      <td>48.27</td>\n",
       "      <td>6.73</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    total_bill   tip  size\n",
       "59       48.27  6.73     4"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "suppression_value = df[df_scores==treshold]\n",
    "suppression_value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "hidden": 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>total_bill</th>\n",
       "      <th>tip</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>3.07</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>92</th>\n",
       "      <td>5.75</td>\n",
       "      <td>1.00</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>111</th>\n",
       "      <td>7.25</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>170</th>\n",
       "      <td>50.81</td>\n",
       "      <td>10.00</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>172</th>\n",
       "      <td>7.25</td>\n",
       "      <td>5.15</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>178</th>\n",
       "      <td>9.60</td>\n",
       "      <td>4.00</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>212</th>\n",
       "      <td>48.33</td>\n",
       "      <td>9.00</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     total_bill    tip  size\n",
       "67         3.07   1.00     1\n",
       "92         5.75   1.00     2\n",
       "111        7.25   1.00     1\n",
       "170       50.81  10.00     3\n",
       "172        7.25   5.15     2\n",
       "178        9.60   4.00     2\n",
       "212       48.33   9.00     4"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "outliers = df[df_scores<treshold] \n",
    "outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "outliers_new = outliers.to_records(index=False) # Lets convert it to a numpy array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "rec.array([( 3.07,  1.  , 1), ( 5.75,  1.  , 2), ( 7.25,  1.  , 1),\n",
       "           (50.81, 10.  , 3), ( 7.25,  5.15, 2), ( 9.6 ,  4.  , 2),\n",
       "           (48.33,  9.  , 4)],\n",
       "          dtype=[('total_bill', '<f8'), ('tip', '<f8'), ('size', '<i8')])"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "outliers_new"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "outliers_new[:] = suppression_value.to_records(index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "rec.array([(48.27, 6.73, 4), (48.27, 6.73, 4), (48.27, 6.73, 4),\n",
       "           (48.27, 6.73, 4), (48.27, 6.73, 4), (48.27, 6.73, 4),\n",
       "           (48.27, 6.73, 4)],\n",
       "          dtype=[('total_bill', '<f8'), ('tip', '<f8'), ('size', '<i8')])"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "outliers_new"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "df[df_scores<treshold] = pd.DataFrame(outliers_new,index=df[df_scores<treshold].index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {
    "hidden": 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>total_bill</th>\n",
       "      <th>tip</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>48.27</td>\n",
       "      <td>6.73</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>92</th>\n",
       "      <td>48.27</td>\n",
       "      <td>6.73</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>111</th>\n",
       "      <td>48.27</td>\n",
       "      <td>6.73</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>170</th>\n",
       "      <td>48.27</td>\n",
       "      <td>6.73</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>172</th>\n",
       "      <td>48.27</td>\n",
       "      <td>6.73</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>178</th>\n",
       "      <td>48.27</td>\n",
       "      <td>6.73</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>212</th>\n",
       "      <td>48.27</td>\n",
       "      <td>6.73</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     total_bill   tip  size\n",
       "67        48.27  6.73     4\n",
       "92        48.27  6.73     4\n",
       "111       48.27  6.73     4\n",
       "170       48.27  6.73     4\n",
       "172       48.27  6.73     4\n",
       "178       48.27  6.73     4\n",
       "212       48.27  6.73     4"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df_scores<treshold] "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "### Missing Data Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "In statistics, missing data, or missing values, occur when no data value is stored for the variable in an observation.\n",
    "\n",
    "Missing data in the training data set can reduce the power / fit of a model or can lead to a biased model because we have not analysed the behavior and relationship with other variables correctly. It can lead to wrong prediction or classification."
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "![image.png](attachment:image.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**Solutions for Missing Data**\n",
    "\n",
    "*1)* Deletion : List Wise Deletion and Pair Wise Deletion\n",
    "\n",
    "*2)* Mean/ Mode/ Median Imputation\n",
    "\n",
    "*3)* Prediction Model\n",
    "\n",
    "*4)* KNN Imputation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**Note:** If a value is missing becuase it doesn't exist (like the height of the oldest child of someone who doesn't have any children) then it doesn't make sense to try and guess what it might be. These values you probably do want to keep as NaN. On the other hand, if a value is missing because it wasn't recorded, then you can try to guess what it might have been based on the other values in that column and row."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**1) Deletion : List Wise Deletion and Pair Wise Deletion**\n",
    "\n",
    "Deletion methods are used when the nature of missing data is “Missing completely at random” else non random missing values can bias the model output."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 247,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "C1 = np.array([1,5,np.NaN,42,np.NaN,54,22,85,19,35])\n",
    "C2 = np.array([42,24,53,3,np.NaN,np.NaN,42,88,9,12])\n",
    "C3 = np.array([np.NaN,13,53,57,np.NaN,43,76,18,np.NaN,35])\n",
    "df = pd.DataFrame({\"C1\":C1,\"C2\":C2,\"C3\":C3})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 248,
   "metadata": {
    "hidden": 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>C1</th>\n",
       "      <th>C2</th>\n",
       "      <th>C3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>53.0</td>\n",
       "      <td>53.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>42.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>57.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>54.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>43.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>22.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>76.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>85.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>19.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>35.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     C1    C2    C3\n",
       "0   1.0  42.0   NaN\n",
       "1   5.0  24.0  13.0\n",
       "2   NaN  53.0  53.0\n",
       "3  42.0   3.0  57.0\n",
       "4   NaN   NaN   NaN\n",
       "5  54.0   NaN  43.0\n",
       "6  22.0  42.0  76.0\n",
       "7  85.0  88.0  18.0\n",
       "8  19.0   9.0   NaN\n",
       "9  35.0  12.0  35.0"
      ]
     },
     "execution_count": 248,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 249,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "C1    2\n",
       "C2    2\n",
       "C3    3\n",
       "dtype: int64"
      ]
     },
     "execution_count": 249,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 250,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "C1    8\n",
       "C2    8\n",
       "C3    7\n",
       "dtype: int64"
      ]
     },
     "execution_count": 250,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.notnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 251,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "7"
      ]
     },
     "execution_count": 251,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 252,
   "metadata": {
    "hidden": 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>C1</th>\n",
       "      <th>C2</th>\n",
       "      <th>C3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>53.0</td>\n",
       "      <td>53.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>54.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>43.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>19.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     C1    C2    C3\n",
       "0   1.0  42.0   NaN\n",
       "2   NaN  53.0  53.0\n",
       "4   NaN   NaN   NaN\n",
       "5  54.0   NaN  43.0\n",
       "8  19.0   9.0   NaN"
      ]
     },
     "execution_count": 252,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df.isnull().any(axis=1)] # All missing values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 253,
   "metadata": {
    "hidden": 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>C1</th>\n",
       "      <th>C2</th>\n",
       "      <th>C3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>42.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>57.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>22.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>76.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>85.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>35.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     C1    C2    C3\n",
       "1   5.0  24.0  13.0\n",
       "3  42.0   3.0  57.0\n",
       "6  22.0  42.0  76.0\n",
       "7  85.0  88.0  18.0\n",
       "9  35.0  12.0  35.0"
      ]
     },
     "execution_count": 253,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df.notnull().all(axis=1)] # All non-missing values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 254,
   "metadata": {
    "hidden": 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>C1</th>\n",
       "      <th>C2</th>\n",
       "      <th>C3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>53.0</td>\n",
       "      <td>53.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   C1    C2    C3\n",
       "2 NaN  53.0  53.0"
      ]
     },
     "execution_count": 254,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df[\"C1\"].isnull() & df[\"C3\"].notnull()]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Let's delete missing values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 255,
   "metadata": {
    "hidden": 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>C1</th>\n",
       "      <th>C2</th>\n",
       "      <th>C3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>42.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>57.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>22.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>76.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>85.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>35.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     C1    C2    C3\n",
       "1   5.0  24.0  13.0\n",
       "3  42.0   3.0  57.0\n",
       "6  22.0  42.0  76.0\n",
       "7  85.0  88.0  18.0\n",
       "9  35.0  12.0  35.0"
      ]
     },
     "execution_count": 255,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dropna() # If we want to drop na values permanently, we can say inplace argument True"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "If we want to delete rows that all values are NaN, we can use *how* argument."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 256,
   "metadata": {
    "hidden": 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>C1</th>\n",
       "      <th>C2</th>\n",
       "      <th>C3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>53.0</td>\n",
       "      <td>53.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>42.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>57.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>54.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>43.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>22.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>76.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>85.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>19.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>35.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     C1    C2    C3\n",
       "0   1.0  42.0   NaN\n",
       "1   5.0  24.0  13.0\n",
       "2   NaN  53.0  53.0\n",
       "3  42.0   3.0  57.0\n",
       "5  54.0   NaN  43.0\n",
       "6  22.0  42.0  76.0\n",
       "7  85.0  88.0  18.0\n",
       "8  19.0   9.0   NaN\n",
       "9  35.0  12.0  35.0"
      ]
     },
     "execution_count": 256,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dropna(how=\"all\") "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**2) Mean/ Mode/ Median Imputation**\n",
    "\n",
    "Imputation is a method to fill in the missing values with estimated ones."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 271,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "C1 = np.array([1,5,np.NaN,42,np.NaN,54,22,85,19,35])\n",
    "C2 = np.array([42,24,53,3,np.NaN,np.NaN,42,88,9,12])\n",
    "C3 = np.array([np.NaN,13,53,57,np.NaN,43,76,18,np.NaN,35])\n",
    "df = pd.DataFrame({\"C1\":C1,\"C2\":C2,\"C3\":C3})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 272,
   "metadata": {
    "hidden": 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>C1</th>\n",
       "      <th>C2</th>\n",
       "      <th>C3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>53.0</td>\n",
       "      <td>53.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>42.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>57.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>54.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>43.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>22.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>76.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>85.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>19.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>35.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     C1    C2    C3\n",
       "0   1.0  42.0   NaN\n",
       "1   5.0  24.0  13.0\n",
       "2   NaN  53.0  53.0\n",
       "3  42.0   3.0  57.0\n",
       "4   NaN   NaN   NaN\n",
       "5  54.0   NaN  43.0\n",
       "6  22.0  42.0  76.0\n",
       "7  85.0  88.0  18.0\n",
       "8  19.0   9.0   NaN\n",
       "9  35.0  12.0  35.0"
      ]
     },
     "execution_count": 272,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Let's give mean of column to NA values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 273,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "df[\"C1\"][df[\"C1\"].isnull()] = df[\"C1\"].mean()\n",
    "df[\"C2\"][df[\"C2\"].isnull()] = df[\"C2\"].mean()\n",
    "df[\"C3\"][df[\"C3\"].isnull()] = df[\"C3\"].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 274,
   "metadata": {
    "hidden": 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>C1</th>\n",
       "      <th>C2</th>\n",
       "      <th>C3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.000</td>\n",
       "      <td>42.000</td>\n",
       "      <td>42.142857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.000</td>\n",
       "      <td>24.000</td>\n",
       "      <td>13.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>32.875</td>\n",
       "      <td>53.000</td>\n",
       "      <td>53.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>42.000</td>\n",
       "      <td>3.000</td>\n",
       "      <td>57.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>32.875</td>\n",
       "      <td>34.125</td>\n",
       "      <td>42.142857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>54.000</td>\n",
       "      <td>34.125</td>\n",
       "      <td>43.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>22.000</td>\n",
       "      <td>42.000</td>\n",
       "      <td>76.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>85.000</td>\n",
       "      <td>88.000</td>\n",
       "      <td>18.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>19.000</td>\n",
       "      <td>9.000</td>\n",
       "      <td>42.142857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>35.000</td>\n",
       "      <td>12.000</td>\n",
       "      <td>35.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       C1      C2         C3\n",
       "0   1.000  42.000  42.142857\n",
       "1   5.000  24.000  13.000000\n",
       "2  32.875  53.000  53.000000\n",
       "3  42.000   3.000  57.000000\n",
       "4  32.875  34.125  42.142857\n",
       "5  54.000  34.125  43.000000\n",
       "6  22.000  42.000  76.000000\n",
       "7  85.000  88.000  18.000000\n",
       "8  19.000   9.000  42.142857\n",
       "9  35.000  12.000  35.000000"
      ]
     },
     "execution_count": 274,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "We can also use **fillna()** function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 275,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "C1 = np.array([1,5,np.NaN,42,np.NaN,54,22,85,19,35])\n",
    "C2 = np.array([42,24,53,3,np.NaN,np.NaN,42,88,9,12])\n",
    "C3 = np.array([np.NaN,13,53,57,np.NaN,43,76,18,np.NaN,35])\n",
    "df = pd.DataFrame({\"C1\":C1,\"C2\":C2,\"C3\":C3})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 276,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    34.125\n",
       "1    13.000\n",
       "2    53.000\n",
       "3    57.000\n",
       "4    34.125\n",
       "5    43.000\n",
       "6    76.000\n",
       "7    18.000\n",
       "8    34.125\n",
       "9    35.000\n",
       "Name: C3, dtype: float64"
      ]
     },
     "execution_count": 276,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"C1\"].fillna(df[\"C1\"].mean())\n",
    "df[\"C2\"].fillna(df[\"C3\"].mean())\n",
    "df[\"C3\"].fillna(df[\"C2\"].mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "We can also use **apply()** function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 277,
   "metadata": {
    "hidden": 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>C1</th>\n",
       "      <th>C2</th>\n",
       "      <th>C3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.000</td>\n",
       "      <td>42.000</td>\n",
       "      <td>42.142857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.000</td>\n",
       "      <td>24.000</td>\n",
       "      <td>13.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>32.875</td>\n",
       "      <td>53.000</td>\n",
       "      <td>53.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>42.000</td>\n",
       "      <td>3.000</td>\n",
       "      <td>57.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>32.875</td>\n",
       "      <td>34.125</td>\n",
       "      <td>42.142857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>54.000</td>\n",
       "      <td>34.125</td>\n",
       "      <td>43.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>22.000</td>\n",
       "      <td>42.000</td>\n",
       "      <td>76.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>85.000</td>\n",
       "      <td>88.000</td>\n",
       "      <td>18.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>19.000</td>\n",
       "      <td>9.000</td>\n",
       "      <td>42.142857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>35.000</td>\n",
       "      <td>12.000</td>\n",
       "      <td>35.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       C1      C2         C3\n",
       "0   1.000  42.000  42.142857\n",
       "1   5.000  24.000  13.000000\n",
       "2  32.875  53.000  53.000000\n",
       "3  42.000   3.000  57.000000\n",
       "4  32.875  34.125  42.142857\n",
       "5  54.000  34.125  43.000000\n",
       "6  22.000  42.000  76.000000\n",
       "7  85.000  88.000  18.000000\n",
       "8  19.000   9.000  42.142857\n",
       "9  35.000  12.000  35.000000"
      ]
     },
     "execution_count": 277,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.apply(lambda col: col.fillna(col.mean()),axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "We can also use **where()** function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 279,
   "metadata": {
    "hidden": 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>C1</th>\n",
       "      <th>C2</th>\n",
       "      <th>C3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>43.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>28.5</td>\n",
       "      <td>53.0</td>\n",
       "      <td>53.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>42.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>57.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>28.5</td>\n",
       "      <td>33.0</td>\n",
       "      <td>43.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>54.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>43.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>22.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>76.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>85.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>19.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>43.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>35.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     C1    C2    C3\n",
       "0   1.0  42.0  43.0\n",
       "1   5.0  24.0  13.0\n",
       "2  28.5  53.0  53.0\n",
       "3  42.0   3.0  57.0\n",
       "4  28.5  33.0  43.0\n",
       "5  54.0  33.0  43.0\n",
       "6  22.0  42.0  76.0\n",
       "7  85.0  88.0  18.0\n",
       "8  19.0   9.0  43.0\n",
       "9  35.0  12.0  35.0"
      ]
     },
     "execution_count": 279,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.where(pd.notna(df),df.median(),axis=\"columns\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "We can also impute missing data by categorical variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 284,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "Salary = np.array([12900,5850,np.NaN,7600,np.NaN,6300,7500,8250,11900,9350])\n",
    "C2 = np.array([42,24,53,3,np.NaN,np.NaN,42,88,9,12])\n",
    "C3 = np.array([np.NaN,13,53,57,np.NaN,43,76,18,np.NaN,35])\n",
    "gender = np.array([\"F\",\"M\",\"M\",\"M\",\"F\",\"F\",\"F\",\"F\",\"M\",\"M\"])\n",
    "df = pd.DataFrame({\"Salary\":Salary,\"C2\":C2,\"C3\":C3, \"Gender\":gender})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 285,
   "metadata": {
    "hidden": 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>Salary</th>\n",
       "      <th>C2</th>\n",
       "      <th>C3</th>\n",
       "      <th>Gender</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>12900.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5850.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>53.0</td>\n",
       "      <td>53.0</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7600.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6300.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>43.0</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7500.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>76.0</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8250.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>11900.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>9350.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Salary    C2    C3 Gender\n",
       "0  12900.0  42.0   NaN      F\n",
       "1   5850.0  24.0  13.0      M\n",
       "2      NaN  53.0  53.0      M\n",
       "3   7600.0   3.0  57.0      M\n",
       "4      NaN   NaN   NaN      F\n",
       "5   6300.0   NaN  43.0      F\n",
       "6   7500.0  42.0  76.0      F\n",
       "7   8250.0  88.0  18.0      F\n",
       "8  11900.0   9.0   NaN      M\n",
       "9   9350.0  12.0  35.0      M"
      ]
     },
     "execution_count": 285,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 287,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Gender\n",
       "F    7875.0\n",
       "M    8475.0\n",
       "Name: Salary, dtype: float64"
      ]
     },
     "execution_count": 287,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(\"Gender\").median()[\"Salary\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Let's assign values depending on Gender."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 289,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    12900.0\n",
       "1     5850.0\n",
       "2     8475.0\n",
       "3     7600.0\n",
       "4     7875.0\n",
       "5     6300.0\n",
       "6     7500.0\n",
       "7     8250.0\n",
       "8    11900.0\n",
       "9     9350.0\n",
       "Name: Salary, dtype: float64"
      ]
     },
     "execution_count": 289,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"Salary\"].fillna(df.groupby(\"Gender\")[\"Salary\"].transform(\"median\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "We can also impute categorical variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 292,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "Salary = np.array([12900,5850,np.NaN,7600,np.NaN,6300,7500,8250,11900,9350])\n",
    "C2 = np.array([42,24,53,3,np.NaN,np.NaN,42,88,9,12])\n",
    "C3 = np.array([np.NaN,13,53,57,np.NaN,43,76,18,np.NaN,35])\n",
    "gender = np.array([\"F\",np.NaN,\"M\",\"M\",np.NaN,\"F\",\"F\",\"F\",np.NaN,\"M\"],dtype=object)\n",
    "df = pd.DataFrame({\"Salary\":Salary,\"C2\":C2,\"C3\":C3, \"Gender\":gender})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 293,
   "metadata": {
    "hidden": 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>Salary</th>\n",
       "      <th>C2</th>\n",
       "      <th>C3</th>\n",
       "      <th>Gender</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>12900.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5850.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>53.0</td>\n",
       "      <td>53.0</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7600.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6300.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>43.0</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7500.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>76.0</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8250.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>11900.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>9350.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Salary    C2    C3 Gender\n",
       "0  12900.0  42.0   NaN      F\n",
       "1   5850.0  24.0  13.0    NaN\n",
       "2      NaN  53.0  53.0      M\n",
       "3   7600.0   3.0  57.0      M\n",
       "4      NaN   NaN   NaN    NaN\n",
       "5   6300.0   NaN  43.0      F\n",
       "6   7500.0  42.0  76.0      F\n",
       "7   8250.0  88.0  18.0      F\n",
       "8  11900.0   9.0   NaN    NaN\n",
       "9   9350.0  12.0  35.0      M"
      ]
     },
     "execution_count": 293,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Let's assign mode value of Gender."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 294,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    F\n",
       "1    F\n",
       "2    M\n",
       "3    M\n",
       "4    F\n",
       "5    F\n",
       "6    F\n",
       "7    F\n",
       "8    F\n",
       "9    M\n",
       "Name: Gender, dtype: object"
      ]
     },
     "execution_count": 294,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"Gender\"].fillna(df.Gender.mode().item())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "We could also be a bit more savvy and replace missing values with whatever value comes directly after it in the same column. (This makes a lot of sense for datasets where the observations have some sort of logical order to them.) \n",
    "\n",
    "We will use *method* argument for that."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 295,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    F\n",
       "1    M\n",
       "2    M\n",
       "3    M\n",
       "4    F\n",
       "5    F\n",
       "6    F\n",
       "7    F\n",
       "8    M\n",
       "9    M\n",
       "Name: Gender, dtype: object"
      ]
     },
     "execution_count": 295,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.Gender.fillna(method=\"bfill\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 296,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    F\n",
       "1    F\n",
       "2    M\n",
       "3    M\n",
       "4    M\n",
       "5    F\n",
       "6    F\n",
       "7    F\n",
       "8    F\n",
       "9    M\n",
       "Name: Gender, dtype: object"
      ]
     },
     "execution_count": 296,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.Gender.fillna(method=\"ffill\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**3) Prediction Model**\n",
    "\n",
    "Prediction model is one of the sophisticated method for handling missing data. Here, we create a predictive model to estimate values that will substitute the missing data. In this case, we divide our data set into two sets: One set with no missing values for the variable and another one with missing values. First data set become training data set of the model while second data set with missing values is test data set and variable with missing values is treated as target variable. Next, we create a model to predict target variable based on other attributes of the training data set and populate missing values of test data set."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "For a real world example, we will work with *titanic* dataset of Seaborn."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 344,
   "metadata": {
    "hidden": 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>survived</th>\n",
       "      <th>pclass</th>\n",
       "      <th>age</th>\n",
       "      <th>sibsp</th>\n",
       "      <th>parch</th>\n",
       "      <th>fare</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>7.2500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>71.2833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.9250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>53.1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.0500</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   survived  pclass   age  sibsp  parch     fare\n",
       "0         0       3  22.0      1      0   7.2500\n",
       "1         1       1  38.0      1      0  71.2833\n",
       "2         1       3  26.0      0      0   7.9250\n",
       "3         1       1  35.0      1      0  53.1000\n",
       "4         0       3  35.0      0      0   8.0500"
      ]
     },
     "execution_count": 344,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = sns.load_dataset(\"titanic\")\n",
    "df = df.select_dtypes(include= [\"float64\",\"int64\"])\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 345,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "survived      0\n",
       "pclass        0\n",
       "age         177\n",
       "sibsp         0\n",
       "parch         0\n",
       "fare          0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 345,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "We will use *EM* algorithm."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 346,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "variables = list(df.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 347,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "new_df = np.array(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 348,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(891, 6)"
      ]
     },
     "execution_count": 348,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 349,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "EM_df = EM().complete(new_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 350,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "EM_df = pd.DataFrame(EM_df,columns=variables)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 351,
   "metadata": {
    "hidden": 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>survived</th>\n",
       "      <th>pclass</th>\n",
       "      <th>age</th>\n",
       "      <th>sibsp</th>\n",
       "      <th>parch</th>\n",
       "      <th>fare</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.2500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>71.2833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.9250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>53.1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>8.0500</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   survived  pclass   age  sibsp  parch     fare\n",
       "0       0.0     3.0  22.0    1.0    0.0   7.2500\n",
       "1       1.0     1.0  38.0    1.0    0.0  71.2833\n",
       "2       1.0     3.0  26.0    0.0    0.0   7.9250\n",
       "3       1.0     1.0  35.0    1.0    0.0  53.1000\n",
       "4       0.0     3.0  35.0    0.0    0.0   8.0500"
      ]
     },
     "execution_count": 351,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "EM_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 352,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "survived    0\n",
       "pclass      0\n",
       "age         0\n",
       "sibsp       0\n",
       "parch       0\n",
       "fare        0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 352,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "EM_df.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**4) KNN Imputation**\n",
    "\n",
    "Here the missing values of an attribute are imputed using the given number of attributes that are most similar to the attribute whose values are missing. The similarity of two attributes is determined using a distance function. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "For a real world example, we will work with *titanic* dataset of Seaborn."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 353,
   "metadata": {
    "hidden": 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>survived</th>\n",
       "      <th>pclass</th>\n",
       "      <th>age</th>\n",
       "      <th>sibsp</th>\n",
       "      <th>parch</th>\n",
       "      <th>fare</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>7.2500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>71.2833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.9250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>53.1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.0500</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   survived  pclass   age  sibsp  parch     fare\n",
       "0         0       3  22.0      1      0   7.2500\n",
       "1         1       1  38.0      1      0  71.2833\n",
       "2         1       3  26.0      0      0   7.9250\n",
       "3         1       1  35.0      1      0  53.1000\n",
       "4         0       3  35.0      0      0   8.0500"
      ]
     },
     "execution_count": 353,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = sns.load_dataset(\"titanic\")\n",
    "df = df.select_dtypes(include= [\"float64\",\"int64\"])\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 354,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "survived      0\n",
       "pclass        0\n",
       "age         177\n",
       "sibsp         0\n",
       "parch         0\n",
       "fare          0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 354,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 355,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "variables = list(df.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 356,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "new_df = np.array(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 357,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(891, 6)"
      ]
     },
     "execution_count": 357,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 358,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.    ,  3.    , 22.    ,  1.    ,  0.    ,  7.25  ],\n",
       "       [ 1.    ,  1.    , 38.    ,  1.    ,  0.    , 71.2833],\n",
       "       [ 1.    ,  3.    , 26.    ,  0.    ,  0.    ,  7.925 ],\n",
       "       [ 1.    ,  1.    , 35.    ,  1.    ,  0.    , 53.1   ],\n",
       "       [ 0.    ,  3.    , 35.    ,  0.    ,  0.    ,  8.05  ],\n",
       "       [ 0.    ,  3.    ,     nan,  0.    ,  0.    ,  8.4583],\n",
       "       [ 0.    ,  1.    , 54.    ,  0.    ,  0.    , 51.8625],\n",
       "       [ 0.    ,  3.    ,  2.    ,  3.    ,  1.    , 21.075 ],\n",
       "       [ 1.    ,  3.    , 27.    ,  0.    ,  2.    , 11.1333],\n",
       "       [ 1.    ,  2.    , 14.    ,  1.    ,  0.    , 30.0708]])"
      ]
     },
     "execution_count": 358,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_df[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 359,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Imputing row 1/891 with 0 missing, elapsed time: 0.193\n",
      "Imputing row 101/891 with 0 missing, elapsed time: 0.195\n",
      "Imputing row 201/891 with 0 missing, elapsed time: 0.197\n",
      "Imputing row 301/891 with 1 missing, elapsed time: 0.198\n",
      "Imputing row 401/891 with 0 missing, elapsed time: 0.200\n",
      "Imputing row 501/891 with 0 missing, elapsed time: 0.202\n",
      "Imputing row 601/891 with 0 missing, elapsed time: 0.203\n",
      "Imputing row 701/891 with 0 missing, elapsed time: 0.204\n",
      "Imputing row 801/891 with 0 missing, elapsed time: 0.206\n"
     ]
    }
   ],
   "source": [
    "knn_df = knnimput.KNN(k=4).complete(new_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 360,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "knn_df = pd.DataFrame(knn_df,columns=variables)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 361,
   "metadata": {
    "hidden": 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>survived</th>\n",
       "      <th>pclass</th>\n",
       "      <th>age</th>\n",
       "      <th>sibsp</th>\n",
       "      <th>parch</th>\n",
       "      <th>fare</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.2500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>71.2833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.9250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>53.1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>8.0500</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   survived  pclass   age  sibsp  parch     fare\n",
       "0       0.0     3.0  22.0    1.0    0.0   7.2500\n",
       "1       1.0     1.0  38.0    1.0    0.0  71.2833\n",
       "2       1.0     3.0  26.0    0.0    0.0   7.9250\n",
       "3       1.0     1.0  35.0    1.0    0.0  53.1000\n",
       "4       0.0     3.0  35.0    0.0    0.0   8.0500"
      ]
     },
     "execution_count": 361,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 362,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "survived    0\n",
       "pclass      0\n",
       "age         0\n",
       "sibsp       0\n",
       "parch       0\n",
       "fare        0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 362,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn_df.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "**Vizualization of Missing Data**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "We will use *missingno* library."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 363,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "C1 = np.array([1,5,np.NaN,42,np.NaN,54,22,85,19,35])\n",
    "C2 = np.array([42,24,53,3,np.NaN,np.NaN,42,88,9,12])\n",
    "C3 = np.array([np.NaN,13,53,57,np.NaN,43,76,18,np.NaN,35])\n",
    "df = pd.DataFrame({\"C1\":C1,\"C2\":C2,\"C3\":C3})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 364,
   "metadata": {
    "hidden": 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>C1</th>\n",
       "      <th>C2</th>\n",
       "      <th>C3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>53.0</td>\n",
       "      <td>53.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>42.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>57.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>54.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>43.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>22.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>76.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>85.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>19.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>35.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     C1    C2    C3\n",
       "0   1.0  42.0   NaN\n",
       "1   5.0  24.0  13.0\n",
       "2   NaN  53.0  53.0\n",
       "3  42.0   3.0  57.0\n",
       "4   NaN   NaN   NaN\n",
       "5  54.0   NaN  43.0\n",
       "6  22.0  42.0  76.0\n",
       "7  85.0  88.0  18.0\n",
       "8  19.0   9.0   NaN\n",
       "9  35.0  12.0  35.0"
      ]
     },
     "execution_count": 364,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 365,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1728x720 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "msno.bar(df)\n",
    "\n",
    "plt.title(\"Vizualization of Missing Data\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 366,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABaoAAAJlCAYAAAA7AMWIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzde7jddX0n+vd371wJVwW5BRKSrBVvjDi2OJWq8YJWEtB6qq2tt6povcK0fZwzM306jtPLGU/biZRyOFRHylFLOzrVDjepUqTWC5YRqxWzNpeEQAwCEgghCcnO9/yx1rZpJOS692//Vl6v58mTvX9rrd/vvS8Pv+w33/39lFprAAAAAACgKSNNBwAAAAAA4NCmqAYAAAAAoFGKagAAAAAAGqWoBgAAAACgUYpqAAAAAAAapagGAAAAAKBRimoAgH1QSvmnUsqySTz/wlJKLaXMGLx/bSnlLZNwnUn9OHZzzVJK+UQp5aFSys0HcJ5TSymPllJGD+Acj5ZSFu3v6wEAgINLUQ0AMFBK+UIp5cNPcPzVpZT1pZQZtdZn1VpvnKpMtdZX1Vr/7EDOUUq5vJTyO7ucd0o/joGfTXJ2kvm11jN3fbCU8tZBSf9Huxx/zeD45UlSa7271np4rXV8f4MMXn/n/r5+d0opHyqlbCulbBz86ZVSLi6lnLgP57ixlPKOg50NAACmM0U1AMA/uzzJm0opZZfjb0ryqVrr9qmPNFQWJFlda930JM+5I8kvTqwoH3hzkt6kJju4/qLWekSSpyT5+SQnJLllX8pqAAA41CiqAQD+2efSLxdfOHGglHJMkhVJrhi8v7qU8vLB2xsGW0g8WkrZNFj1u3CwMvgrO5948NiSwdvLSynfKqU8UkpZW0r50O4C7by6tpTy7Z2u9+jgnMsGj/2Pwarvh0spN5VSnjU4/s4kv5Lkg4PX/K8n+Dhml1JWllLWDf6sLKXMHjy2rJRyTynlN0opPyyl/KCU8qtPkvekUspfl1J+VEq5vZRy/uD425N8LMnPDHL8592cYn2S7yR55eB1T0nygiR/vdM1dt0e5a2llDsHK5jvKqX8yuD4klLKlwefkwdKKX+xm6/H5aWUPymlXD04xzdKKYt3eu4rSimrBue5ZHDOPa54rrVuq7X+U5JfTHJ/kt8YnO+YUspVpZT7B9ugXFVKmT947HfT//67ePB5unhw/KOD75VHSim3lFJeuJvLAgBAKymqAQAGaq2bk/xl+it4J7w+yfdrrd9+gucfPdhC4vAkH03yd0nu3YtLbRpc4+gky5O8u5Tymr3I95ydrvfrSVYl+d+Dh69N0knytMGxTw1ec9ng7Y8MXnvuE5z6Pyb5N0nOSPKcJGcm+a2dHj8hyVFJTk7y9iR/Mijwn8ifJ7knyUlJfiHJ75VSXlZr/XiSX0vytUGO//QkH+oV+eevwS8l+XySrU/0xFLKvCQXJXnVYBXzC5LcOnj4vyS5PskxSeYn+eMnueYbkvznwXNvT/K7g/Mfm+QzSf59kqem/zl/wZOc5ycMtij5fP75f4CMJPlE+ivMT02yOcnFg+f+x/S/j943+Dy9b/Cab6b/9XlKkk8n+R+llDn7kgMAAKYzRTUAwL/0Z0leV0qZO3j/zYNju1VK+cUkv5zk/6i1btvTBWqtN9Zav1Nr3VFr/cf0y90X723AUsrPJvmdJOfVWh8ZnPO/11o31lq3JvlQkueUUo7ay1P+SpIP11p/WGu9P/3C9k07Pb5t8Pi2Wus1SR5NsvQJcp2S/j7U/67WuqXWemv6q6jftOtz9+Cvkiwb5H9zBqvZn8SOJM8upcyttf5gsIp5IveCJCcN8nxl96fI/6y13jzY3uVT6ZfCSXJOkn+qtf7PwWMXpb/qe1+tS79kTq31wVrrZ2utj9VaN6Zfij/p17/W+snB67bXWv8wyew8wdcAAADaSlENALCTQZl5f5JXl1IWJfnp9FewPqFSynPTXw3784OSd49KKc8vpfztYOuHh9NfaXzsXr72lPRXfb+l1tobHBstpfxfpZQ7SimPJFk9ePpenTP91c9rdnp/zeDYhAd32Z/7sSSH7+Y8PxqUrzuf6+S9zJHkxyvbr05/Vfextda/f5Lnbkp/a41fS/KDwfYdTx88/MEkJcnNpZR/KqW87Ukuu3P5vPPHd1KStTtdr6a/YnxfnZzkR0lSSjmslPL/llLWDL5eNyU5upQyursXD7ZeuW2w/ciG9Fe47+3XFwAApj1FNQDAT5rYeuJNSa6vtd73RE8qpRyX/urf99Vav7XTQ5uSHLbT807Y5aWfTn/P5VNqrUcluTT9QvVJDVZ5fy7JylrrtTs99MtJXp3k5ekXmAsnXjL4u+7h1OvSX3k84dTBsX21LslTSilH7HKuvdkOZVdXpL+n8/+3pyfWWr9Qaz07yYlJvp/kTwfH19daz6+1npTkXUkumdiXeh/8IP1tQ5IkpZSy8/t7o5QykuTc9Lf0SPof19Ikz6+1HpnkRRNPnfiQdnn9C5P8u/S3oTmm1np0koezF98zAADQFopqAICfdEX6pe/52c22H4NBfp9N8qla61/s8vC3kzyrlHLGYB/hD+3y+BHprzzeUko5M/2ieW/89/T3y/7IE5xva5IH0y/If2+Xx+9LsuhJzvvnSX6rlHLcYE/m307yyb3M9GO11rVJvprk90spc0op/yr9Pa0/ta/nSvLlJGfnyfeVTinl+FLKeYO9qremvy3J+OCx100MKUzyUPoF8Pg+5rg6yemllNcMvubvTX/P7j0qpcwspTwj/c/vCUn+aPDQEenvS71hMCxy1/26d/16HZFke/or/WeUUn47yZH7+HEAAMC0pqgGANhFrXV1+oXrvPRXPj+R+ekPx7uwlPLoTn9OHWzJ8eEkX0wylmTXvZHfk+TDpZSN6ZfCf7mX0X4pyc/vcr0Xpl+sr0l/5fL3knx9l9d9PMkzSykbSimfe4Lz/k6Sf0jyj0m+k/4wxt/Zy0y7ekP6K7rXpb/a/D/VWv9mX09S+75Ua/3RHp46kv4K5XXpb63x4vQ/v0l/25ZvlFIeTf/reEGt9a59zPFAktcl+Uj6/yPgmel/rp5wuOPALw6uuWFw3QeTPK/WOrFKfWWSuUkeSP9rdd0ur/9okl8opTxUSrkoyRfSH5bZS//rvCU7bUcCAADDoPS32QMAAPZksI3HPUl+pdb6t03nAQCAYWFFNQAAPIlSyitLKUeXUmYn+Q/p7w2966p1AADgACiqAQDgyf1MkjvS36rj3CSvqbVubjYSAAAMF1t/AAAAAADQKCuqAQAAAABolKIaAAAAAIBGKaoBAAAAAGiUohoAAAAAgEYpqgEAmJbKQNM5AACAyTej6QAAALAbo7XW7U2HAODQU0oZTTIvyeFJHqq1bm44EsDQs6Ia+LFSyrxSyutLKUc2nQWAQ9PgXvT7pZTrktxaSrmglLKw4VgAHEJKKYcn+USSm5L0klxaSvnXzaYCGH5WVANJklLKYUluSdJNcmEp5U+tGgBgKpVSjkjy1SQbkjyQ5O4k/y3JglLKf0ryaK21NhgRgCE3uBd9I8n9Sa5JMjfJ+5P8IMn/bjAawNAr/q0PlFJmJlmZZEWS1UlekOT/THKJshqAqVBKmZ3kr9P/jb93Jllba91eSvlIkvcmeV6t9ftNZgRguJVS5ib5QpKtSd5Ra10zOP6xJGcl+Ve11m0NRgQYarb+AJLkuUnemOSGJL+S5E+S/Nck7xn8Yw0AJturkxyf5A+SrN5pb+qPDf5+eSOpADiU/FKSI5P8l1rrmsGCniRZlf5q6teXUt5QSnl+YwkBhpitP4AkuTXJ/53+CuoflVJ+P/3/PvzXJCmlWFkNwGQ7MclYkpt23t6j1torpTyc5KTGkgFwqPhWkj9KcnOS1Fq3lVLmJDk/yXFJXpbksCQbSykX1Fo/01hSgCGkqIZD1OAfXK9O8tNJ/leSP6i1bimljNRa7yulfGjw1Imy+tJa66bB2/Mm3gaA/TW4F70mydPTn5NwTa118+BetKOUMnPwK9YPJBnd5bXFftUAHKid7kXPSfKlJJ+ptW4ZPDaa5MYkDyd5e5KvJDk7ye8l+fellJtqrT9sIjfAMFJUwyFoMCDk6iSzkhyR/p6gM5Kk1rpj8PcDu5TVtZTy/yQ5IcmHSymfrbV+bqqzAzAcdrkXHZXk+vQHVf34XpRkfPD35iTH7PTaeUneUkr5Rq31likLDcBQeYKfi65JUnZ6yrGDxz9ea103OHZ9KWVJkouSHJ1EUQ1wkCiq4RBTSjk8yT8kuSfJf0jyDxMrBnY1KKs/nKQm+UiSpyQ5PckrB+8DwD7b23vRToX1jvQLhIlS4b8leW2Sn5qSwAAMnT3diwa/uXNfKeV3J+5HO/02z+wka5Lc30B0gKGlqIZDSCllRvpDqX6Q5O211tWD4yM7lQH/Qq31h4OV1Yen/w+4R5I8v9b6nSkJDcBQ2Zd7USlltNY6nuTR9H+zZ3b6e4f+UpIX1VrvnNLwAAyFvbkXTWwvNdiKqtSBUspxSf5N+sMVtzbyAQAMKUU1HFpOSPLMJJckWTtxcHcl9U6OSvK09Pdme0Gt9bZJSwjAsNuXe9HEHtQTW39cnOSN6d+LvjXJOQEYXvv0c9FEaV1K6Sb5YJIXJ1lWa31s8qMCHDoU1XBo+ddJnp3kxsEKtd0qpcwavDkj/T2ql0VJDcCB2+t7Ufr3oMeT3Jf+EKtHo6QG4MDt689F40nemf4w+mckeWWt9fuTnhLgEDPSdABgSo0k2TbxTinlJ/4bsNOxRemXAY+lP936BbXWb09JSgCG2b7ci5aUUl6W5EdJ1qW/9ZSSGoADta8/F70yyUlJViV5uZ+LACZHGfwGC3AIGPyq2q1JLqm1/ubg2MRAkF2fe02SB2utb5rimAAMsX28F12X5PYkv5Hk2FrrvVMaFoChtI/3oi8kGau1vq+UMrvWal9qgEliRTUcWu5LckuSN5RSXp3091vbdQVBKeUZSWYl+frURwRgyO3LvWhGkl6tdauSGoCDaF/uRaPpr6SOkhpgcimq4RBSa304yfvSH47426WUcwfHfzw0pJRyTJLfTHJKkquayAnA8NqPe9Hnm8gJwPDaj3vRXzeRE+BQY+sPOASVUn4uyWeSPJjkz5L8Sfp7tL0wyS8kOS/Ji+y9BsBkcS8CoGnuRQDTi6IaDlGllOcluSzJ6UkmVg7ck+TeJO+ttX63qWwAHBrciwBomnsRwPShqIZDWCnlqUkWJjkj/b3Xvp7knlrrj5rMBcChw70IgKa5FwFMD4pqAAAAAAAaZZgiAAAAAACNUlQDAAAAANAoRTUAAAAAAI1SVAMAAAAA0ChFNQAAAAAAjVJUAwAAAADQKEU1AAAAAACNUlQDAAAAANAoRTUAAAAAAI2a0XSAPVm2bFltOgMAh66VK1cmSS688MKGkwBwqFq5cmXOOOOMpmMAQJKUpgNMglZ2j7XWnHfeeXnJS16SX//1X286zs72+3vEimoAAAAAgBYppaTT6aTX6zUd5aBRVAMAAAAAtEy3282dd96Zbdu2NR3loFBUAwAAAAC0TLfbzbZt27J69eqmoxwUimoAAAAAgJZZunRpkgzN9h+KagAAAACAljnxxBMzb948RTUAAAAAAM0YGRkZqoGKimoAAAAAgBbqdru54447sn379qajHDBFNQAAAABAC00MVFyzZk3TUQ6YohoAAAAAoIU6nU6S4RioqKgGAAAAAGih+fPn57DDDlNUAwAAAADQjJGRkSxZskRRDQAAAABAcyYGKo6Pjzcd5YAoqgEAAAAAWqrb7Wbr1q25++67m45yQBTVAAAAAAAt1e12kySrVq1qOMmBUVQDAAAAALTU/PnzM2fOnNbvU62oBgAAAABoqdHR0XQ6HUU1AAAAAADN6XQ6rR+oqKgGAAAAAGixbrebLVu2ZO3atU1H2W+KagAAAACAFpsYqNjm7T8U1QAAAAAALXbqqadm9uzZimoAAAAAAJoxOjqaJUuWZGxsrOko+23SiupSyi+UUv64lPJ3pZRHSim1lPLJyboeAAAAAMChqtvtZmxsLDt27Gg6yn6ZzBXVv5XkfUnOSHLvJF4HAAAAAOCQ1u12s3nz5txzzz1NR9kvk1lU/9sk3SRHJnn3JF4HAAAAAOCQNjFQcdWqVQ0n2T+TVlTXWv+21jpWa62TdQ0AAAAAAJIFCxZk1qxZrd2n2jBFAAAAAICWGx0dzeLFi9Pr9ZqOsl8U1QAAAAAAQ6DNAxUV1QAAAAAAQ6Db7eaxxx7Lvffe23SUfaaoBgAAAAAYAhMDFdu4T7WiGgAAAABgCCxcuDAzZ85s5T7VimoAAAAAgCEwY8aM1g5UVFQDAAAAAAyJbrebXq+XWmvTUfaJohoAAAAAYEh0u91s2rQp69atazrKPpkxWScupbwmyWsG754w+PtnSimXD95+oNb6m5N1fQAAAACAQ02n00mS9Hq9nHzyyQ2n2XuTVlQnOSPJW3Y5tmjwJ0nWJFFUAwAAAAAcJKeddtqPByq+5CUvaTrOXpu0rT9qrR+qtZYn+bNwsq4NAAAAAHAomjlzZk477bTWDVS0RzUAAAAAwBDpdrsZGxtr1UBFRTUAAAAAwBDpdDrZuHFj1q9f33SUvaaoBgAAAAAYIkuXLk2SVm3/oagGAAAAABgip512WmbMmKGoBgAAAACgGbNmzWrdQEVFNQAAAADAkOl0Oun1eq0ZqKioBgAAAAAYMt1uN4888kjuu+++pqPsFUU1AAAAAMCQ6Xa7SdozUFFRDQAAAAAwZBYvXpyRkRFFNQAAAAAAzZg1a1YWLlyoqAYAAAAAoDndbjdjY2OtGKioqAYAAAAAGELdbjcbNmzI/fff33SUPVJUAwAAAAAMoaVLlyZpx0BFRTUAAAAAwBBatGhRawYqKqoBAAAAAIbQnDlzsmDBAkU1AAAAAADN6Xa76fV6036goqIaAAAAAGBIdbvdPPTQQ3nggQeajvKkZjQdAAAAgCd366235sILL2w6BgCHuBtvvLHpCOyHTqeTpD9Q8bjjjms4ze5ZUQ0AAAAAMKSWLFmSkZGRjI2NNR3lSSmqAQAAAACG1Ny5c3PKKadM+4GKimoAAAAAgCE2MVBxOlNUAwAAAAAMsW63mwcffDAPPvhg01F2S1ENAAAAADDEut1ukkzrVdWKagAAAACAIbZw4cIkyZo1a5oN8iQU1QAAAAAAQ2z9+vVJkuOPP77hJLunqAYAAAAAGGITW35MbAEyHSmqAQAAAACGWK/Xy+GHH56TTjqp6Si7pagGAAAAABhivV4vnU4npZSmo+yWohoAAAAAYEht27Ytd95557Te9iNRVAMAAAAADK01a9Zk27ZtimoAAAAAAJqxatWqJNN7kGKiqAYAAAAAGFq9Xi/z5s2b1oMUE0U1AAAAAMDQmhikODIyvavg6Z0OAAAAAID9sn379txxxx3pdDpNR9kjRTUAAAAAwBBqyyDFRFENAAAAADCUer1ekuk/SDFRVAMAAAAADKVer5e5c+dm/vz5TUfZI0U1AAAAAMAQGhsba8UgxURRDQAAAAAwdMbHx3P77be3YtuPRFENAAAAADB07r777mzdulVRDQAAAABAM1atWpWkHYMUE0U1AAAAAMDQ6fV6mTNnTisGKSaKagAAAACAoTM2NpYlS5ZkdHS06Sh7RVENAAAAADBE2jZIMVFUAwAAAAAMlbVr12bLli2KagAAAAAAmtHr9ZK0Z5BioqgGAAAAABgqY2NjmT17dk499dSmo+w1RTUAAAAAwBDp9XqtGqSYKKoBAAAAAIbGjh07MjY21qptPxJFNQAAAADA0LjnnnuyefNmRTUAAAAAAM2YGKTY6XQaTrJvFNUAAAAAAEOi1+tl1qxZWbhwYdNR9omiGgAAAABgSPR6vSxevLhVgxQTRTUAAAAAwFBo6yDFRFENAAAAADAU7r333jz22GOt2586UVQDAAAAAAyFsbGxJMnSpUsbTrLvFNUAAAAAAEOg1+tl5syZrRukmCiqAQAAAACGwsQgxRkzZjQdZZ8pqgEAAAAAWq7Wml6v18pBiomiGgAAAACg9datW5dNmza1cpBioqgGAAAAAGi9Xq+XJFZUAwAAAADQjIlBiqeddlrTUfaLohoAAAAAoOV6vV5OO+20zJw5s+ko+0VRDQAAAADQYrXWjI2NtXZ/6kRRDQAAAADQauvXr8/GjRtbuz91oqgGAAAAAGi1iUGKS5cubTjJ/lNUAwAAAAC0WK/Xy+joaGsHKSaKagAAAACAVpsYpDhr1qymo+w3RTUAAAAAQEvVWtPr9Vq9P3WiqAYAAAAAaK377rsvjzzyiKIaAAAAAIBmTAxSVFQDAAAAANCIXq+XkZGRLF68uOkoB2RG0wH25MYbb2w6AgCHuFtvvbXpCAAAAPCExsbGsnDhwlYPUkysqAYAAAAAaKVhGaSYKKoBAAAAAFrp/vvvz4YNGxTVAAAAAAA0Y2KQ4tKlSxtOcuAU1QAAAAAALTQxSHHRokVNRzlgimoAAAAAgBbq9XpZsGBB5syZ03SUA6aoBgAAAABomWEapJgoqgEAAAAAWueBBx7IQw89pKgGAAAAAKAZE4MUO51Ow0kODkU1AAAAAEDLjI2NZWRkJEuWLGk6ykGhqAYAAAAAaJler5dTTjklc+fObTrKQaGoBgAAAABombGxsaHZnzpRVAMAAAAA0DBFNQAAAABAy3Q6nR8PVBwGimoAAAAAgJbpdrtZu3ZtNm/e3HSUg0JRDQAAAADQMp1OJzt27Mjtt9/edJSDQlENAAAAANAyE4MUx8bGGk5ycCiqAQAAAABa5thjj80xxxwzNPtUK6oBAAAAAFqmlJJut6uoBgAAAACgOd1uN2vWrMmWLVuajnLAFNUAAAAAAC3U7XazY8eO3HnnnU1HOWCKagAAAACAFpoYqLhq1aqGkxw4RTUAAAAAQAsdd9xxOfroo4din2pFNQAAAABACw3TQEVFNQAAAABAS3U6naxevTqPP/5401EOiKIaAAAAAKClJgYq3nHHHU1HOSCKagAAAACAlpoYqNj27T8U1QAAAAAALXX88cfnyCOPVFQDAAAAANCMYRmoqKgGAAAAAGixbrebu+66q9UDFRXVAAAAAAAt1u12Mz4+nrvuuqvpKPtNUQ0AAAAA0GITAxVXrVrVcJL9p6gGAAAAAGixE044IUcccUSr96lWVAMAAAAAtFgpJZ1OJ2NjY01H2W+KagAAAACAlpsYqLht27amo+wXRTUAAAAAQMt1u91s27attQMVFdUAAAAAAC03MVCxrftUK6oBAAAAAFrupJNOyrx581q7T7WiGgAAAACg5Uop6Xa7VlQDAAAAANCcbrebO+64I9u3b286yj5TVAMAAAAADIGJgYqrV69uOso+U1QDAAAAAAyBTqeTJFm1alXDSfadohoAAAAAYAicfPLJOeyww1o5UFFRDQAAAAAwBEZGRtLpdFo5UFFRDQAAAAAwJCYGKo6PjzcdZZ8oqgEAAAAAhkSn08njjz/euoGKimoAAAAAgCGxdOnSJGndPtWKagAAAACAITF//vzMnTu3dftUK6oBAAAAAIZEWwcqKqoBAAAAAIZIp9PJ7bff3qqBiopqAAAAAIAh0u12s3Xr1tx9991NR9lrimoAAAAAgCHS7XaTpFXbfyiqAQAAAACGyCmnnJI5c+YoqgEAAAAAaMbo6GgWL17cqqJ6RtMB9mTZsmVNRwDgELZy5cqmIwAAAMA+63a7ufbaazM+Pp7R0dGm4+yRFdUAAAAAAEOm2+1my5Ytueeee5qOslcU1QAAAAAAQ2bp0qVJ2jNQUVENAAAAADBkTj311MyePVtRDQAAAABAM9o2UFFRDQAAAAAwhLrdbsbGxrJjx46mo+yRohoAAAAAYAh1u91s3ry5FQMVFdUAAAAAAEOo2+0macdARUU1AAAAAMAQWrBgQWbOnKmoBgAAAACgGTNmzMjixYszNjbWdJQ9UlQDAAAAAAyptgxUVFQDAAAAAAypbrebTZs2Zd26dU1HeVKKagAAAACAIdWWgYqKagAAAACAIbVw4cJWDFRUVAMAAAAADKmZM2dm0aJFimoAAAAAAJozMVCx1tp0lN1SVAMAAAAADLFOp5NHH310Wg9UVFQDAAAAAAyxNgxUVFQDAAAAAAyxHTt2JEk2bNjQcJLdU1QDAAAAAAypHTt25OKLL84xxxyTs88+u+k4u6WoBgAAAAAYUtdff32+973v5Z3vfGcOP/zwpuPslqIaAAAAAGAIPfroo7nsssvyrGc9K694xSuajvOkFNUAAAAAAEPo8ssvz4YNG3LBBRdkZGR6V8HTOx0AAAAAAPvszjvvzF/91V/l3HPPTafTaTrOHimqAQAAAACGSK01H/3oR3P44Yfn7W9/e9Nx9oqiGgAAAABgiNxwww35x3/8x5x//vk58sgjm46zVxTVAAAAAABD4rHHHsull16apUuX5lWvelXTcfaaohoAAAAAYEhcccUVeeCBB/KBD3wgo6OjTcfZa4pqAAAAAIAhcPfdd+czn/lMzjnnnDzzmc9sOs4+UVQDAAAAALRcrTUXXXRR5s6dm/PPP7/pOPtMUQ0AAAAA0HI33XRTbrnllrztbW/L0Ucf3XScfaaoBgAAAABosS1btuSSSy7J4sWLc9555zUdZ78oqgEAAAAAWuzTn/50fvjDH+aCCy5o1QDFnSmqAQAAAABa6t57782VV16Zs88+O6effnrTcfabohoAAAAAoKUuvvjizJw5M+9617uajnJAFNUAAAAAAC30ta99LV//+tfzlre8JU996lObjnNAFNUAAAAAAC3z+OOP5+KLL86CBQvy2te+tuk4B0xRDQAAAADQMldeeWXWrVuX97///ZkxY0bTcQ6YohoAAAAAoEXWr1+fT33qU3nxi1+c5z3veU3HOSgU1QAAAAAALXLJJZdkZGQk73nPe5qOctAoqgEAAAAAWuKb3/xm/u7v/i5vfOMb87SnPa3pOAeNohoAAAAAoAW2bduWP/7jP878+fPzute9ruk4B5WiGgAAAACgBT7zmc9k7dq1ef/7359Zs2Y1HVYEuG4AAB0iSURBVOegUlQDAAAAAExz999/f6644oqcddZZOfPMM5uOc9ApqgEAAAAAprlLL7004+Pjee9739t0lEmhqAYAAAAAmMZuvfXW3HDDDfnlX/7lnHjiiU3HmRSKagAAAACAaWr79u256KKLcsIJJ+QNb3hD03EmjaIaAAAAAGCa+tznPpe77ror733vezN79uym40waRTUAAAAAwDT0ox/9KJdffnnOPPPMnHXWWU3HmVSKagAAAACAaeiyyy7L1q1b8773vS+llKbjTCpFNQAAAADANPPd7343X/jCF/L6178+p5xyStNxJp2iGgAAAABgGhkfH89FF12U4447Lm984xubjjMlFNUAAAAAANPIVVddlbGxsbz73e/O3Llzm44zJRTVAAAAAADTxMMPP5yPf/zjee5zn5tly5Y1HWfKlFpr0xn2ZNoHBAAAAIBDwDBO85t23eMf/uEf5tprr83HPvaxLFy4sOk4+2q/v0esqAYAAAAAmAZWrVqVq6++Oq997WvbWFIfEEU1AAAAAEDDduzYkZUrV+aYY47JW97ylqbjTDlFNQAAAABAw6677rp8//vfz6/92q9l3rx5TceZcopqAAAAAIAGbdy4MZdddllOP/30vPzlL286TiMU1QAAAAAADfrEJz6RjRs35gMf+EBKGcaZlXumqAYAAAAAaMjtt9+ez3/+8znvvPOyZMmSpuM0RlENAAAAANCAWmsuuuiiHHHEEXnb297WdJxGKaoBAAAAABrwxS9+Md/5zndy/vnn54gjjmg6TqMU1QAAAAAAU2zTpk259NJL8/SnPz2vetWrmo7TOEU1AAAAAMAUu+KKK/LQQw/lggsuyMiImtZnAAAAAABgCq1evTqf/exnc8455+TpT39603GmBUU1AAAAAMAUmRigeNhhh+X8889vOs60oagGAAAAAJgiX/7yl/Otb30rb3vb23LUUUc1HWfaUFQDAAAAAEyBzZs355JLLsmSJUty7rnnNh1nWlFUAwAAAABMgU9+8pO5//7784EPfCCjo6NNx5lWFNUAAAAAAJNs7dq1+cu//Mu84hWvyOmnn950nGlHUQ0AAAAAMIlqrbn44osze/bsvOtd72o6zrSkqAYAAAAAmERf/epXc/PNN+etb31rnvKUpzQdZ1pSVAMAAAAAQ62U8p5Syl2llC2llFtKKS+cqmtv3bo1F198cRYuXJjXvOY1U3XZ1lFUAwAAAABDq5Tyi0k+muT3kjw3yVeTXFtKOXUqrn/llVdm/fr1ueCCCzJjxoypuGQrKaoBAAAAgGH260kur7X+aa31tlrr+5P8IMm7J/vCP/jBD/LpT386L33pS3PGGWdM9uVaTYUPAAAAAAylUsqsJM9L8ge7PHR9khdM1nXHx8dz880359JLL02tNe985zsn61JDw4pqAAAAAGBYHZtkNMl9uxy/L8kJk3HB8fHxfPCDH8yHPvSh3H333am15iMf+UjGx8cn43JDQ1ENAAAAAAy7usv75QmOHRQ333xzbrvttjz++ONJku3bt+d73/tebr755sm43NBQVAMAAAAAw+qBJOP5ydXTT8tPrrI+KMbGxrJly5Z/cWzr1q25/fbbJ+NyQ0NRDQAAAAAMpVrr40luSXL2Lg+dneSrk3HNTqeTOXPm/Itjs2fPzpIlSybjckNDUQ0AAAAADLM/SvLWUso7SinPKKV8NMlJSS6djIudeeaZecYznvHjsnpkZCTPfOYzc+aZZ07G5YZGqXVStmI5mKZ9QAAAAAA4BJSmA+yvUsp7knwwyYlJvpvk39Zab8okdY/j4+O5+eabc+WVV+Y73/lO/vzP/zzHH3/8ZFxqutnv7xFFNQAAAACwN1pbVD+JSe0e77333rzxjW/Mr/7qr+bNb37zZF5qutjv7xFbfwAAAAAATIKTTz45z3ve83LNNddkfHy86TjTmqIaAAAAAGCSLF++PPfdd19uueWWpqNMa4pqAAAAAIBJ8rM/+7M5+uijc9VVVzUdZVpTVAMAAAAATJKZM2fmla98Zb761a/mwQcfbDrOtKWoBgAAAACYRMuXL8/4+Hiuu+66pqNMW4pqAAAAAIBJdMopp+Q5z3lOrrnmmuzYsaPpONOSohoAAAAAYJKtWLEi69aty7e+9a2mo0xLimoAAAAAgEn2ohe9KEceeaShiruhqAYAAAAAmGSzZs3KK17xinzlK1/Jhg0bmo4z7SiqAQAAAACmwPLly7N9+/Z84QtfaDrKtKOoBgAAAACYAgsXLsyzn/3sXH311am1Nh1nWlFUAwAAAABMkeXLl2ft2rX59re/3XSUaUVRDQAAAAAwRZYtW5Z58+YZqrgLRTUAAAAAwBSZM2dOzj777Nx00015+OGHm44zbSiqAQAAAACm0PLly7Nt27b8zd/8TdNRpg1FNQAAAADAFFqyZEme/vSn56qrrjJUcUBRDQAAAAAwxVasWJE1a9bku9/9btNRpgVFNQAAAADAFHvpS1+auXPnGqo4oKgGAAAAAJhic+fOzctf/vLceOON2bhxY9NxGqeoBgAAAABowPLly/P444/ni1/8YtNRGqeoBgAAAABowNKlS9PpdAxVjKIaAAAAAKAxK1asyJ133pnbbrut6SiNUlQDAAAAADTkZS97WebMmZOrr7666SiNUlQDAAAAADRk3rx5eelLX5obbrghmzZtajpOYxTVAAAAAAANWr58ebZs2ZIvfelLTUdpTJnum3QvW7ZsegcEYKitXLkyZ5xxRtMxADjE3XrrrbnwwgubjgHAIe7GG28sTWeYBNOie6y15h3veEdGR0dz2WWXNR3nQOz394gV1QAAAAAADSqlZMWKFRkbG8uqVauajtMIRTUAAAAAQMPOPvvszJo165AdqqioBgAAAABo2OGHH55ly5blS1/6UjZv3tx0nCmnqAYAAAAAmAZWrFiRxx57LDfccEPTUaacohoAAAAAYBp49rOfnQULFhyS238oqgEAAAAApoGJoYq33XZb7rjjjqbjTClFNQAAAADANHH22Wdn5syZh9yqakU1AAAAAMA0cdRRR+VFL3pRrr/++mzZsqXpOFNGUQ0AAAAAMI2sWLEimzZtype//OWmo0wZRTUAAAAAwDTynOc8J/Pnz89VV13VdJQpo6gGAAAAAJhGSilZvnx5vvvd72b16tVNx5kSimoAAAAAgGnmla98ZWbMmHHIDFVUVAMAAAAATDPHHHNMzjrrrFx//fV5/PHHm44z6RTVAAAAAADT0LnnnptHHnkkN910U9NRJp2iGgAAAABgGnruc5+bE0888ZAYqqioBgAAAACYhkZGRrJ8+fJ8+9vfztq1a5uOM6kU1QAAAAAA09TP/dzPZWRkZOiHKiqqAQAAAACmqac+9ak566yzct111w31UEVFNQAAAADANLZ8+fI8/PDD+fu///umo0waRTUAAAAAwDT2Uz/1Uzn++OOHevsPRTUAAAAAwDQ2OjqaV73qVbnlllty7733Nh1nUiiqAQAAAACmuXPOOScjIyO55pprmo4yKRTVAAAAAADT3HHHHZfnP//5ufbaa7N9+/am4xx0imoAAAAAgBZYsWJFHnrooXzta19rOspBp6gGAAAAAGiB5z//+Tn22GNz1VVXNR3loFNUAwAAAAC0wMRQxW9+85tZv35903EOKkU1AAAAAEBLnHPOOUkydEMVFdUAAAAAAC1xwgkn5Kd/+qdz7bXXZnx8vOk4B42iGgAAAACgRZYvX54HHngg3/jGN5qOctAoqgEAAAAAWuQFL3hB5s2bl6997WtNRzloFNUAAAAAAC0yY8aMdLvdjI2NNR3loFFUAwAAAAC0TLfbzZ133plt27Y1HeWgUFQDAAAAALRMt9vNtm3bsnr16qajHBSKagAAAACAlul2u0mSXq/XcJKDQ1ENAAAAANAyJ510Ug477DBFNQAAAAAAzRgZGUmn01FUAwAAAADQnG63mzvuuCPbt29vOsoBU1QDAAAAALTQMA1UVFQDAAAAALTQxEDFsbGxhpMcOEU1AAAAAEALzZ8/P3Pnzh2KfaoV1QAAAAAALTRMAxUV1QAAAAAALTUxUHF8fLzpKAdEUQ0AAAAA0FKdTidbt27N3Xff3XSUA6KoBgAAAABoqYmBiqtWrWo4yYFRVAMAAAAAtNQpp5ySOXPmtH6fakU1AAAAAEBLjY6OZsmSJYpqAAAAAACa0+l0Wj9QUVENAAAAANBi3W43W7Zsydq1a5uOst8U1QAAAAAALbZ06dIkafX2H4pqAAAAAIAWO/XUUzN79mxFNQAAAAAAzRiGgYqKagAAAACAlut0OhkbG8uOHTuajrJfFNUAAAAAAC03MVDxnnvuaTrKfpnRdAAAmO5uvfXWXHjhhU3HAOAQtXLlyqYjAAAt0O12kySrVq3Kqaee2nCafWdFNQAAAABAyy1cuDCzZs1q7T7VimoAAAAAgJYbHR3NokWLMjY21nSU/aKoBgAAAAAYAt1ut7UDFRXVAAAAAABDoNvt5rHHHsu9997bdJR9pqgGAAAAABgCS5cuTZJW7lOtqAYAAAAAGAILFizIzJkzW7lPtaIaAAAAAGAIzJw5M4sWLbKiGgAAAACA5nS73fR6vdRam46yTxTVAAAAAABDotvtZtOmTVm3bl3TUfaJohoAAAAAYEh0Op0k7RuoqKgGAAAAABgSp512WmbMmKGoBgAAAACgGbNmzcppp52mqAYAAAAAoDndbjdjY2OtGqioqAYAAAAAGCLdbjcbN27M+vXrm46y1xTVAAAAAABDpNvtJmnXQEVFNQAAAADAEFm0aFFGR0cV1QAAAAAANKONAxUV1QAAAAAAQ6bb7abX67VmoKKiGgAAAABgyHQ6nTzyyCO57777mo6yVxTVAAAAAABDpm0DFRXVAAAAAABDZvHixRkZGVFUAwAAAADQjNmzZ2fhwoWKagAAAAAAmtPtdjM2NtaKgYqKagAAAACAIdTtdrNhw4bcf//9TUfZI0U1AAAAwP/f3v2E+H3ndRx/ffJ3pk0TE1LSpmn+bOb3HQoqQfbQ3cO69aQu6Ip7KCJ4WDwseqiy4mEVe9iD9CIosgd7UNhCBcH1oChIGypuIHgotRTm+x3abm3SJpnpZJI2UbrJ10NmtNbOZPJn8un3m8cDhl9mfr9fvu/zkw+fN8AIDWmholANAAAAADBCQ1qoKFQDAAAAAIzQ1NRUjhw5IlQDAAAAAFDPZDJJ27af+4WKQjUAAAAAwEg1TZOlpaUsLCzUHmVdQjUAAAAAwEgNZaGiUA0AAAAAMFIzMzMppaTrutqjrEuoBgAAAAAYqenp6Rw+fNiJagAAAAAA6mmaRqgGAAAAAKCepmmyuLiYxcXF2qOsSagGAAAAABixISxUFKoBAAAAAEZsZmYmSfLWW29VnmRtQjUAAAAAwIhNT08nST7++OPKk6xNqAYAAAAAoCqhGgAAAACAqoRqAAAAAACqEqoBAAAAAKhKqAYAAAAAoCqhGgAAAACAqoRqAAAAAACqEqoBAAAAAKhKqAYAAAAAoCqhGgAAAACAqoRqAAAAAACqEqoBAAAAAKhKqAYAAAAAoCqhGgAAAACAqoRqAAAAAACqEqoBAAAAAKhKqAYAAAAAoCqhGgAAAACAqoRqAAAAAACqEqoBAAAAAKhKqAYAAAAAoCqhGgAAAACAqoRqAAAAAACqEqoBAAAAAKhKqAYAAAAAoCqhGgAAAACAqoRqAAAAAACqEqoBAAAAAKhKqAYAAAAAoCqhGgAAAACAqoRqAAAAAACqEqoBAAAAAKhKqAYAAAAAoCqhGgAAAACAqoRqAAAAAACqEqoBAAAAAKhKqAYAAAAAoCqhGgAAAACAqoRqAAAAAACqEqoBAAAAAKhKqAYAAAAAoCqhGgAAAACAqoRqAAAAAACqEqoBAAAAAKhKqAYAAAAAoCqhGgAAAACAqoRqAAAAAACqEqoBAAAAAKhKqAYAAAAAoCqhGgAAAACAqoRqAAAAAACqEqoBAAAAAKhKqAYAAAAAoCqhGgAAAACAqoRqAAAAAACqEqoBAAAAAKhKqAYAAAAAoCqhGgAAAACAqoRqAAAAAIAR6/u+9gg3JVQDAAAAAIzY+fPnkyR79uypPMnahGoAAAAAgBFr2zZJ0jRN5UnWJlQDAAAAAIxY27bZsmVLjh8/XnuUNQnVAAAAAAAj1nVdjh49mp07d9YeZU1CNQAAAADASPV9n7ZtP9fXfiRCNQAAAADAaF24cCEXL14UqgEAAAAAqGMIixQToRoAAAAAYLSGsEgxEaoBAAAAAEarbdscPnw4U1NTtUdZl1ANAAAAADBCQ1mkmAjVAAAAAACjtLCwkKWlJaEaAAAAAIA6uq5L8vlfpJgk22oPAAAAwPpOnDiRkydP1h4DABiYtm1TSsnMzEztUW7KiWoAAAAAgBFaXaQ4PT1de5SbEqoBAAAAAEZoKIsUE6EaAAAAAGB0FhcXs7i4KFQDAAAAAFBH27ZJkslkUnmSjRGqAQAAAABGZnWRolANAAAAAEAVbdvm0KFDeeCBB2qPsiFCNQAAAADAyHRdN5j7qROhGgAAAABgVJaWlnLhwgWhGgAAAACAOlYXKQrVAAAAAABUsRqqZ2ZmKk+ycUI1AAAAAMCIdF2XQ4cOZdeuXbVH2TChGgAAAABgRNq2HdS1H4lQDQAAAAAwGsvLyzl37lwmk0ntUW6JUA0AAAAAMBJzc3NJhrVIMRGqAQAAAABGo+u6JEI1AAAAAACVtG2bgwcPDmqRYiJUAwAAAACMRtu2g7ufOhGqAQAAAABGYXl5Oe+//35mZ2drj3LLhGoAAAAAgBGYn59PMrz7qROhGgAAAABgFNq2TRJXfwAAAAAAUMfc3FweffTR7N69u/Yot0yoBgAAAAAYga7rBnmaOhGqAQAAAAAG7/Llyzl79uwg76dOhGoAAAAAgMHrui7JMBcpJkI1AAAAAMDgrS5SFKoBAAAAAKiibdscOHAge/bsqT3KbRGqAQAAAAAGruu6wZ6mToRqAAAAAIBB+/DDD/Puu+8K1QAAAAAA1DE/P59kuPdTJ0I1AAAAAMCgrS5SnEwmlSe5fUI1AAAAAMCAzc3N5eGHH87evXtrj3LbhGoAAAAAgAEb+iLFRKgGAAAAABisK1euDH6RYiJUAwAAAAAMVtd16fteqAYAAAAAoI7VRYpCNQAAAAAAVXRdl/3792ffvn21R7kjQjUAAAAAwEC1bTv409SJUA0AAAAAMEhXr17NO++8I1QDAAAAAFDH/Px8+r7PZDKpPcodE6oBAAAAAAZoLIsUE6EaAAAAAGCQ2rbNvn37sn///tqj3DGhGgAAAABggMaySDERqgEAAAAABmd1keIY7qdOhGoAAAAAgMF58803c/36dSeqAQAAAACoY3WR4uzsbOVJ7g6hGgAAAABgYNq2zd69e0exSDERqgEAAAAABmd1kWIppfYod4VQDQAAAAAwMOfOncuBAwdqj3HXCNUAAAAAAANz5MiRvP3227XHuGuEagAAAACAgZlMJpmfn8/169drj3JXCNUAAAAAAAPTNE2uXLmSM2fO1B7lrhCqAQAAAAAGpmmaJDeWKo6BUA0AAAAAMDBHjx7N9u3bRxOqt9UeAAAAgPW9+uqreeaZZ2qPAcB97uTJk7VH4BO2bduW48ePjyZUO1ENAAAAADBATdOk67pRLFQUqgEAAAAABqhpmnz00Ud57733ao9yx4RqAAAAAIABWl2oODc3V3mSOydUAwAAAAAM0JgWKgrVAAAAAAADtH379hw7dixd19Ue5Y4J1QAAAAAAA7W6ULHv+9qj3BGhGgAAAABgoJqmyeXLlwe/UFGoBgAAAAAYqNWFikO/p1qoBgAAAAAYqGPHjmXbtm2Dv6daqAYAAAAAGKgdO3bk2LFjTlQDAAAAAFBP0zRp23bQCxWFagAAAACAAWuaJpcuXcq5c+dqj3LbhGoAAAAAgAGbTCZJhr1QUagGAAAAABiw48ePZ+vWrUI1AAAAAAB17NixI0ePHhWqAQAAAACoZ+gLFYVqAAAAAICBm0wmWV5ezoULF2qPcluEagAAAACAgZudnU0y3IWKQjUAAAAAwMAdP348W7ZsEaoBAAAAAKhj586dg16oKFQDAAAAAIzAZDIZ7ELFDYXqUso3Sil/Vkr5l1LKpVJKX0r5/k2+8+VSyj+UUj4opVwppbxWSnmmlLL17owOAAAAAMCqpmmytLSUhYWF2qPcso2eqP6DJL+d5ESSMzf7cCnll5O8kuQrSf42yZ8n2ZHkT5K8eFuTAgAAAACwpqZpkgxzoeJGQ/XvJGmS7E7yrfU+WErZneQvklxL8tW+77/Z9/3v5UbkPpXkG6WUp29/ZAAAAAAAPm1mZmawCxU3FKr7vn+57/uu39jlJt9I8nCSF/u+/7dP/B//mRsns5ObxG4AAAAAAG7N1NRUDh8+PN5QfYt+buX1Hz/jvVeSXEny5VLKzk14NgAAAADAfWsymaTrutpj3LLNCNWzK6//L9v3ff/jJG8l2ZbkC5vwbAAAAACA+1bTNFlcXMzi4mLtUW7JZoTqPSuvy2u8v/r3n9iEZwMAAAAA3LdmZ2+cIx7a9R+bEapvpqy8buS+awAAAAAANmhmZialFKE6/3ties8a7+/+1OcAAAAAALgLpqen8/jjjwvVSeZWXptPv1FK2ZbkWJIfJ3lzE54NAAAAAHBfa5pGqE7y0srrz3/Ge19J8kCSH/Z9/1+b8GwAAAAAgPta0zRZWFjIBx98UHuUDduMUP03SRaSPF1K+eLqH0spU0m+u/Lr9zbhuQAAAAAA972muXHZxZBOVW/byIdKKV9P8vWVXx9Zef1SKeUvV/690Pf9t5Ok7/tLpZTfzI1gfbKU8mKSD5L8UpLZlb//9d0ZHwAAAACAT5qZmUmSdF2XJ598svI0G7OhUJ3kRJLf+NTfvrDykyQ/SvLt1Tf6vv9BKeVnk3wnya8mmUoyn+R3k/xp3/f9nQwNAAAAAMBne/DBBwe3UHFDobrv+2eTPHsr/3Hf9/+a5BdvfSQAAAAAAO7EZDLJ66+/XnuMDduMO6oBAAAAAKhodnY258+fz8WLF2uPsiFCNQAAAADAyEwmkyTDWagoVAMAAAAAjIxQDQAAAABAVbt27cpjjz0mVAMAAAAAUE/TNEI1AAAAAAD1NE2Tc+fOZXl5ufYoNyVUAwAAAACM0Oo91V3XVZ7k5oRqAAAAAIARapomyTAWKgrVAAAAAAAj9NBDD+XgwYNCNQAAAAAA9UwmE1d/AAAAAABQT9M0OXv2bC5fvlx7lHUJ1QAAAAAAIzWUe6qFagAAAACAkZpMJknuXagupfxWKeW1UsqllZ9TpZSv3ex7QjUAAAAAwEjt2bMnjzzyyL28p/rdJL+f5GeSfDHJS0l+UEr56fW+tO0eDAYAAAAAQCWTyeSenaju+/7vPvWn75RSvpXkS0leW+t7TlQDAAAAAIzYzMxMzpw5k+effz6nTp3KtWvX7slzSylbSylPJ9mV5IfrfdaJagAAAACAkbp27VpeeeWVJMkLL7yQ6enpPPHEE3nuueeydevWTXlmKeWnkpxKMpXkwyS/0vf9v6/3HSeqAQAAAABG6vTp0zlz5sz//H716tW88cYbOX369GY+di7JiSRPJvlekr8qpfzkel8ofd9v5kAAAAAAAFTy1FNP/WGSZ/N/Dy1fT/JHL7/88nfvxQyllH9O8qO+77+55meEagAAAAAANksp5aUkZ/u+//W1PuOOagAAAAAA7opSyh8n+fsk/5HkoSS/luSrSb623veEagAAAAAA7pZHknx/5XU5yWtJfqHv+39a70uu/gAAAAAAoKotN/8IAAAAAABsHqEaAAAAAICqhGoAAAAAAKoSqgEAAAAAqEqoBgAAAACgKqEaAAAAAICqhGoAAAAAAKoSqgEAAAAAqEqoBgAAAACgqv8G9MgHttOhRCsAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1800x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "msno.matrix(df)\n",
    "\n",
    "plt.title(\"Vizualization of Missing Data\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "For a real world example we will use planets dataset of seaborn."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 367,
   "metadata": {
    "hidden": 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>method</th>\n",
       "      <th>number</th>\n",
       "      <th>orbital_period</th>\n",
       "      <th>mass</th>\n",
       "      <th>distance</th>\n",
       "      <th>year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Radial Velocity</td>\n",
       "      <td>1</td>\n",
       "      <td>269.300</td>\n",
       "      <td>7.10</td>\n",
       "      <td>77.40</td>\n",
       "      <td>2006</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Radial Velocity</td>\n",
       "      <td>1</td>\n",
       "      <td>874.774</td>\n",
       "      <td>2.21</td>\n",
       "      <td>56.95</td>\n",
       "      <td>2008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Radial Velocity</td>\n",
       "      <td>1</td>\n",
       "      <td>763.000</td>\n",
       "      <td>2.60</td>\n",
       "      <td>19.84</td>\n",
       "      <td>2011</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Radial Velocity</td>\n",
       "      <td>1</td>\n",
       "      <td>326.030</td>\n",
       "      <td>19.40</td>\n",
       "      <td>110.62</td>\n",
       "      <td>2007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Radial Velocity</td>\n",
       "      <td>1</td>\n",
       "      <td>516.220</td>\n",
       "      <td>10.50</td>\n",
       "      <td>119.47</td>\n",
       "      <td>2009</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            method  number  orbital_period   mass  distance  year\n",
       "0  Radial Velocity       1         269.300   7.10     77.40  2006\n",
       "1  Radial Velocity       1         874.774   2.21     56.95  2008\n",
       "2  Radial Velocity       1         763.000   2.60     19.84  2011\n",
       "3  Radial Velocity       1         326.030  19.40    110.62  2007\n",
       "4  Radial Velocity       1         516.220  10.50    119.47  2009"
      ]
     },
     "execution_count": 367,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = sns.load_dataset(\"planets\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 368,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "method              0\n",
       "number              0\n",
       "orbital_period     43\n",
       "mass              522\n",
       "distance          227\n",
       "year                0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 368,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 369,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "msno.matrix(df)\n",
    "\n",
    "plt.title(\"Vizualization of Planets' Missing Data\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 370,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x864 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "msno.heatmap(df)\n",
    "\n",
    "plt.title(\"Vizualization of Planets' Missing Data\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## Data Standardization / Feature Scaling"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "### Normalization(0-1 Scaling)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Normalization is the process of scaling individual samples to have unit norm. Normalization typically means rescales the values into a range of [0,1].\n",
    "\n",
    "We will use **normalize()** function for normalization."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 447,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "C1 = np.array([1,5,43,54,22,85,19,35])\n",
    "C2 = np.array([42,24,53,342,88,9,54,12])\n",
    "C3 = np.array([13,53,57,43,76,18,45,35])\n",
    "df = pd.DataFrame({\"C1\":C1,\"C2\":C2,\"C3\":C3})\n",
    "\n",
    "df = df.astype(float)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 448,
   "metadata": {
    "hidden": 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>C1</th>\n",
       "      <th>C2</th>\n",
       "      <th>C3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>53.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>43.0</td>\n",
       "      <td>53.0</td>\n",
       "      <td>57.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>54.0</td>\n",
       "      <td>342.0</td>\n",
       "      <td>43.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>22.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>76.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>85.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>19.0</td>\n",
       "      <td>54.0</td>\n",
       "      <td>45.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>35.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     C1     C2    C3\n",
       "0   1.0   42.0  13.0\n",
       "1   5.0   24.0  53.0\n",
       "2  43.0   53.0  57.0\n",
       "3  54.0  342.0  43.0\n",
       "4  22.0   88.0  76.0\n",
       "5  85.0    9.0  18.0\n",
       "6  19.0   54.0  45.0\n",
       "7  35.0   12.0  35.0"
      ]
     },
     "execution_count": 448,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 449,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.02273902, 0.95503889, 0.29560727],\n",
       "       [0.08562347, 0.41099265, 0.90760876],\n",
       "       [0.4835736 , 0.59603258, 0.64101617],\n",
       "       [0.15477355, 0.98023246, 0.1232456 ],\n",
       "       [0.18590738, 0.74362952, 0.64222549],\n",
       "       [0.97309817, 0.10303392, 0.20606785],\n",
       "       [0.26093584, 0.74160713, 0.61800594],\n",
       "       [0.68719985, 0.23561138, 0.68719985]])"
      ]
     },
     "execution_count": 449,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "preprocessing.normalize(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "### Standardization(Z Score Scaling)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "The result of standardization (or Z-score normalization) is that the features will be rescaled so that they’ll have the properties of a standard normal distribution with\n",
    "\n",
    "                                        μ=0 and σ=1\n",
    "\n",
    "where μ is the mean (average) and σ is the standard deviation from the mean; standard scores (also called z scores) of the samples are calculated as follows:\n",
    "\n",
    "                                          z=x−μσ\n",
    "                                          \n",
    "We will use **preprocessing.StandardScaler()** or **scale()** function to make standardization.                   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 437,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "df = np.array([[ 1., -1.,  2.],\n",
    "              [ 2.,  0.,  0.],\n",
    "              [ 0.,  1., -1.]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 438,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1., -1.,  2.],\n",
       "       [ 2.,  0.,  0.],\n",
       "       [ 0.,  1., -1.]])"
      ]
     },
     "execution_count": 438,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 442,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.4444444444444444"
      ]
     },
     "execution_count": 442,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 443,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0657403385139377"
      ]
     },
     "execution_count": 443,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.std()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 444,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "df = preprocessing.StandardScaler().fit_transform(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 445,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 0., 0.])"
      ]
     },
     "execution_count": 445,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.mean(axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 446,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1., 1., 1.])"
      ]
     },
     "execution_count": 446,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.std(axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "### Min-Max Scaling"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "An alternative standardization is scaling features to lie between a given minimum and maximum value, often between zero and one, or so that the maximum absolute value of each feature is scaled to unit size. This can be achieved using MinMaxScaler or MaxAbsScaler, respectively.\n",
    "\n",
    "We will use **MinMaxScaler()** function for that."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 480,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "C1 = np.array([1,5,43,54,22,85,19,35])\n",
    "C2 = np.array([42,24,53,342,88,9,54,12])\n",
    "C3 = np.array([13,53,57,43,76,18,45,35])\n",
    "df = pd.DataFrame({\"C1\":C1,\"C2\":C2,\"C3\":C3})\n",
    "\n",
    "df = df.astype(float)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 481,
   "metadata": {
    "hidden": 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>C1</th>\n",
       "      <th>C2</th>\n",
       "      <th>C3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>53.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>43.0</td>\n",
       "      <td>53.0</td>\n",
       "      <td>57.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>54.0</td>\n",
       "      <td>342.0</td>\n",
       "      <td>43.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>22.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>76.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>85.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>19.0</td>\n",
       "      <td>54.0</td>\n",
       "      <td>45.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>35.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     C1     C2    C3\n",
       "0   1.0   42.0  13.0\n",
       "1   5.0   24.0  53.0\n",
       "2  43.0   53.0  57.0\n",
       "3  54.0  342.0  43.0\n",
       "4  22.0   88.0  76.0\n",
       "5  85.0    9.0  18.0\n",
       "6  19.0   54.0  45.0\n",
       "7  35.0   12.0  35.0"
      ]
     },
     "execution_count": 481,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 482,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "mmscaler = preprocessing.MinMaxScaler(feature_range=(15,30))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 483,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[15.        , 16.48648649, 15.        ],\n",
       "       [15.71428571, 15.67567568, 24.52380952],\n",
       "       [22.5       , 16.98198198, 25.47619048],\n",
       "       [24.46428571, 30.        , 22.14285714],\n",
       "       [18.75      , 18.55855856, 30.        ],\n",
       "       [30.        , 15.        , 16.19047619],\n",
       "       [18.21428571, 17.02702703, 22.61904762],\n",
       "       [21.07142857, 15.13513514, 20.23809524]])"
      ]
     },
     "execution_count": 483,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mmscaler.fit_transform(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "### Binary Transformation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "An easy method of converting decimal to binary number equivalents is to write down the decimal number and to continually divide-by-2 (two) to give a result and a remainder of either a “1” or a “0” until the final result equals zero."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 484,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "C1 = np.array([1,5,43,54,22,85,19,35])\n",
    "C2 = np.array([42,24,53,342,88,9,54,12])\n",
    "C3 = np.array([13,53,57,43,76,18,45,35])\n",
    "df = pd.DataFrame({\"C1\":C1,\"C2\":C2,\"C3\":C3})\n",
    "\n",
    "df = df.astype(float)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 485,
   "metadata": {
    "hidden": 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>C1</th>\n",
       "      <th>C2</th>\n",
       "      <th>C3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>53.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>43.0</td>\n",
       "      <td>53.0</td>\n",
       "      <td>57.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>54.0</td>\n",
       "      <td>342.0</td>\n",
       "      <td>43.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>22.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>76.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>85.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>19.0</td>\n",
       "      <td>54.0</td>\n",
       "      <td>45.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>35.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     C1     C2    C3\n",
       "0   1.0   42.0  13.0\n",
       "1   5.0   24.0  53.0\n",
       "2  43.0   53.0  57.0\n",
       "3  54.0  342.0  43.0\n",
       "4  22.0   88.0  76.0\n",
       "5  85.0    9.0  18.0\n",
       "6  19.0   54.0  45.0\n",
       "7  35.0   12.0  35.0"
      ]
     },
     "execution_count": 485,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 486,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "binarizer = preprocessing.Binarizer(threshold=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 487,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 1., 1.],\n",
       "       [0., 1., 1.],\n",
       "       [1., 1., 1.],\n",
       "       [1., 1., 1.],\n",
       "       [1., 1., 1.],\n",
       "       [1., 0., 1.],\n",
       "       [1., 1., 1.],\n",
       "       [1., 1., 1.]])"
      ]
     },
     "execution_count": 487,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "binarizer.fit_transform(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## Variable Transformation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "### Label Encoding"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Encode categorical target labels with value between 0 and n_classes-1."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "For a real world example, we will use tips dataset of Seaborn."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 497,
   "metadata": {
    "hidden": 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>total_bill</th>\n",
       "      <th>tip</th>\n",
       "      <th>sex</th>\n",
       "      <th>smoker</th>\n",
       "      <th>day</th>\n",
       "      <th>time</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>16.99</td>\n",
       "      <td>1.01</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10.34</td>\n",
       "      <td>1.66</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>21.01</td>\n",
       "      <td>3.50</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23.68</td>\n",
       "      <td>3.31</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>24.59</td>\n",
       "      <td>3.61</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   total_bill   tip     sex smoker  day    time  size\n",
       "0       16.99  1.01  Female     No  Sun  Dinner     2\n",
       "1       10.34  1.66    Male     No  Sun  Dinner     3\n",
       "2       21.01  3.50    Male     No  Sun  Dinner     3\n",
       "3       23.68  3.31    Male     No  Sun  Dinner     2\n",
       "4       24.59  3.61  Female     No  Sun  Dinner     4"
      ]
     },
     "execution_count": 497,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = sns.load_dataset(\"tips\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 498,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "label_encoder = LabelEncoder()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 499,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 0, 1, 1, 0, 1, 0, 1, 0, 1, 1, 0,\n",
       "       0, 1, 1, 1, 1, 1, 1, 0, 1, 1, 0, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       0, 0, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 0, 1, 1,\n",
       "       1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0,\n",
       "       1, 0, 1, 1, 0, 0, 1, 0, 0, 0, 1, 0, 1, 1, 0, 0, 1, 0, 0, 1, 1, 0,\n",
       "       0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1,\n",
       "       1, 0, 1, 0, 0, 1, 1, 1, 0, 1, 0, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 0, 1, 1, 1, 1, 1, 0,\n",
       "       0, 1, 1, 0, 0, 0, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0,\n",
       "       1, 0, 1, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1,\n",
       "       1, 0])"
      ]
     },
     "execution_count": 499,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "label_encoder.fit_transform(df[\"sex\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 500,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "df[\"encoded_sex\"] = label_encoder.fit_transform(df[\"sex\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 501,
   "metadata": {
    "hidden": 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>total_bill</th>\n",
       "      <th>tip</th>\n",
       "      <th>sex</th>\n",
       "      <th>smoker</th>\n",
       "      <th>day</th>\n",
       "      <th>time</th>\n",
       "      <th>size</th>\n",
       "      <th>encoded_sex</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>16.99</td>\n",
       "      <td>1.01</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10.34</td>\n",
       "      <td>1.66</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>21.01</td>\n",
       "      <td>3.50</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23.68</td>\n",
       "      <td>3.31</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>24.59</td>\n",
       "      <td>3.61</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   total_bill   tip     sex smoker  day    time  size  encoded_sex\n",
       "0       16.99  1.01  Female     No  Sun  Dinner     2            0\n",
       "1       10.34  1.66    Male     No  Sun  Dinner     3            1\n",
       "2       21.01  3.50    Male     No  Sun  Dinner     3            1\n",
       "3       23.68  3.31    Male     No  Sun  Dinner     2            1\n",
       "4       24.59  3.61  Female     No  Sun  Dinner     4            0"
      ]
     },
     "execution_count": 501,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "If we want a binary transformation, we can also use **cat.codes** function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 502,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "df[\"smoker_encoded\"] = df[\"smoker\"].cat.codes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 503,
   "metadata": {
    "hidden": 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>total_bill</th>\n",
       "      <th>tip</th>\n",
       "      <th>sex</th>\n",
       "      <th>smoker</th>\n",
       "      <th>day</th>\n",
       "      <th>time</th>\n",
       "      <th>size</th>\n",
       "      <th>encoded_sex</th>\n",
       "      <th>smoker_encoded</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>16.99</td>\n",
       "      <td>1.01</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10.34</td>\n",
       "      <td>1.66</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>21.01</td>\n",
       "      <td>3.50</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23.68</td>\n",
       "      <td>3.31</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>24.59</td>\n",
       "      <td>3.61</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>239</th>\n",
       "      <td>29.03</td>\n",
       "      <td>5.92</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sat</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>240</th>\n",
       "      <td>27.18</td>\n",
       "      <td>2.00</td>\n",
       "      <td>Female</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Sat</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>241</th>\n",
       "      <td>22.67</td>\n",
       "      <td>2.00</td>\n",
       "      <td>Male</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Sat</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>242</th>\n",
       "      <td>17.82</td>\n",
       "      <td>1.75</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sat</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>243</th>\n",
       "      <td>18.78</td>\n",
       "      <td>3.00</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Thur</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>244 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     total_bill   tip     sex smoker   day    time  size  encoded_sex  \\\n",
       "0         16.99  1.01  Female     No   Sun  Dinner     2            0   \n",
       "1         10.34  1.66    Male     No   Sun  Dinner     3            1   \n",
       "2         21.01  3.50    Male     No   Sun  Dinner     3            1   \n",
       "3         23.68  3.31    Male     No   Sun  Dinner     2            1   \n",
       "4         24.59  3.61  Female     No   Sun  Dinner     4            0   \n",
       "..          ...   ...     ...    ...   ...     ...   ...          ...   \n",
       "239       29.03  5.92    Male     No   Sat  Dinner     3            1   \n",
       "240       27.18  2.00  Female    Yes   Sat  Dinner     2            0   \n",
       "241       22.67  2.00    Male    Yes   Sat  Dinner     2            1   \n",
       "242       17.82  1.75    Male     No   Sat  Dinner     2            1   \n",
       "243       18.78  3.00  Female     No  Thur  Dinner     2            0   \n",
       "\n",
       "     smoker_encoded  \n",
       "0                 1  \n",
       "1                 1  \n",
       "2                 1  \n",
       "3                 1  \n",
       "4                 1  \n",
       "..              ...  \n",
       "239               1  \n",
       "240               0  \n",
       "241               0  \n",
       "242               1  \n",
       "243               1  \n",
       "\n",
       "[244 rows x 9 columns]"
      ]
     },
     "execution_count": 503,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "We can also do that for multiple classes but its not recommended. *One Hot Encoding* can be used for that."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 504,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "df[\"day_encoded\"] =label_encoder.fit_transform(df[\"day\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 505,
   "metadata": {
    "hidden": 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>total_bill</th>\n",
       "      <th>tip</th>\n",
       "      <th>sex</th>\n",
       "      <th>smoker</th>\n",
       "      <th>day</th>\n",
       "      <th>time</th>\n",
       "      <th>size</th>\n",
       "      <th>encoded_sex</th>\n",
       "      <th>smoker_encoded</th>\n",
       "      <th>day_encoded</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>16.99</td>\n",
       "      <td>1.01</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10.34</td>\n",
       "      <td>1.66</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>21.01</td>\n",
       "      <td>3.50</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23.68</td>\n",
       "      <td>3.31</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>24.59</td>\n",
       "      <td>3.61</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>239</th>\n",
       "      <td>29.03</td>\n",
       "      <td>5.92</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sat</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>240</th>\n",
       "      <td>27.18</td>\n",
       "      <td>2.00</td>\n",
       "      <td>Female</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Sat</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>241</th>\n",
       "      <td>22.67</td>\n",
       "      <td>2.00</td>\n",
       "      <td>Male</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Sat</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>242</th>\n",
       "      <td>17.82</td>\n",
       "      <td>1.75</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sat</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>243</th>\n",
       "      <td>18.78</td>\n",
       "      <td>3.00</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Thur</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>244 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     total_bill   tip     sex smoker   day    time  size  encoded_sex  \\\n",
       "0         16.99  1.01  Female     No   Sun  Dinner     2            0   \n",
       "1         10.34  1.66    Male     No   Sun  Dinner     3            1   \n",
       "2         21.01  3.50    Male     No   Sun  Dinner     3            1   \n",
       "3         23.68  3.31    Male     No   Sun  Dinner     2            1   \n",
       "4         24.59  3.61  Female     No   Sun  Dinner     4            0   \n",
       "..          ...   ...     ...    ...   ...     ...   ...          ...   \n",
       "239       29.03  5.92    Male     No   Sat  Dinner     3            1   \n",
       "240       27.18  2.00  Female    Yes   Sat  Dinner     2            0   \n",
       "241       22.67  2.00    Male    Yes   Sat  Dinner     2            1   \n",
       "242       17.82  1.75    Male     No   Sat  Dinner     2            1   \n",
       "243       18.78  3.00  Female     No  Thur  Dinner     2            0   \n",
       "\n",
       "     smoker_encoded  day_encoded  \n",
       "0                 1            2  \n",
       "1                 1            2  \n",
       "2                 1            2  \n",
       "3                 1            2  \n",
       "4                 1            2  \n",
       "..              ...          ...  \n",
       "239               1            1  \n",
       "240               0            1  \n",
       "241               0            1  \n",
       "242               1            1  \n",
       "243               1            3  \n",
       "\n",
       "[244 rows x 10 columns]"
      ]
     },
     "execution_count": 505,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "### One Hot Encoding"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "A one hot encoding is a representation of categorical variables as binary vectors.\n",
    "\n",
    "This first requires that the categorical values be mapped to integer values."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "For a real world example, we will use tips dataset of Seaborn."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 476,
   "metadata": {
    "hidden": 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>total_bill</th>\n",
       "      <th>tip</th>\n",
       "      <th>sex</th>\n",
       "      <th>smoker</th>\n",
       "      <th>day</th>\n",
       "      <th>time</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>16.99</td>\n",
       "      <td>1.01</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10.34</td>\n",
       "      <td>1.66</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>21.01</td>\n",
       "      <td>3.50</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23.68</td>\n",
       "      <td>3.31</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>24.59</td>\n",
       "      <td>3.61</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>Dinner</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   total_bill   tip     sex smoker  day    time  size\n",
       "0       16.99  1.01  Female     No  Sun  Dinner     2\n",
       "1       10.34  1.66    Male     No  Sun  Dinner     3\n",
       "2       21.01  3.50    Male     No  Sun  Dinner     3\n",
       "3       23.68  3.31    Male     No  Sun  Dinner     2\n",
       "4       24.59  3.61  Female     No  Sun  Dinner     4"
      ]
     },
     "execution_count": 476,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = sns.load_dataset(\"tips\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 479,
   "metadata": {
    "hidden": 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>total_bill</th>\n",
       "      <th>tip</th>\n",
       "      <th>sex</th>\n",
       "      <th>smoker</th>\n",
       "      <th>day</th>\n",
       "      <th>size</th>\n",
       "      <th>time_Lunch</th>\n",
       "      <th>time_Dinner</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>16.99</td>\n",
       "      <td>1.01</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10.34</td>\n",
       "      <td>1.66</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>21.01</td>\n",
       "      <td>3.50</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23.68</td>\n",
       "      <td>3.31</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>24.59</td>\n",
       "      <td>3.61</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Sun</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</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>239</th>\n",
       "      <td>29.03</td>\n",
       "      <td>5.92</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sat</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>240</th>\n",
       "      <td>27.18</td>\n",
       "      <td>2.00</td>\n",
       "      <td>Female</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Sat</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>241</th>\n",
       "      <td>22.67</td>\n",
       "      <td>2.00</td>\n",
       "      <td>Male</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Sat</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>242</th>\n",
       "      <td>17.82</td>\n",
       "      <td>1.75</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Sat</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>243</th>\n",
       "      <td>18.78</td>\n",
       "      <td>3.00</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Thur</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>244 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     total_bill   tip     sex smoker   day  size  time_Lunch  time_Dinner\n",
       "0         16.99  1.01  Female     No   Sun     2           0            1\n",
       "1         10.34  1.66    Male     No   Sun     3           0            1\n",
       "2         21.01  3.50    Male     No   Sun     3           0            1\n",
       "3         23.68  3.31    Male     No   Sun     2           0            1\n",
       "4         24.59  3.61  Female     No   Sun     4           0            1\n",
       "..          ...   ...     ...    ...   ...   ...         ...          ...\n",
       "239       29.03  5.92    Male     No   Sat     3           0            1\n",
       "240       27.18  2.00  Female    Yes   Sat     2           0            1\n",
       "241       22.67  2.00    Male    Yes   Sat     2           0            1\n",
       "242       17.82  1.75    Male     No   Sat     2           0            1\n",
       "243       18.78  3.00  Female     No  Thur     2           0            1\n",
       "\n",
       "[244 rows x 8 columns]"
      ]
     },
     "execution_count": 479,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.get_dummies(df,columns=[\"time\"],prefix=[\"time\"])"
   ]
  }
 ],
 "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
}