InternalLab_Residency2_Hypothesis_Testing_and_KNN.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "IEJOH3pMy_EG"
   },
   "source": [
    "\n",
    "\n",
    "## Lab 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "yRG2Rb6By_EK"
   },
   "source": [
    "Welcome to the second lab of the AIML!\n",
    "In addition to implementing code, there will be questions that you must answer which relate to the lab and your implementation. Each question you will answer is preceded by a **'Question X'** header. Carefully read each question and provide you answer or code in the following textboxes with **'Answer:'** header. Your lab submission will be evaluated based on your answers to each of the questions and the implementation you provide."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Wv9-kA4vy_EN"
   },
   "source": [
    "# Every question is of 1 mark."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "c7F5I7f9y_EQ"
   },
   "source": [
    "## Overview"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "DbanWtFwzc-A"
   },
   "outputs": [],
   "source": [
    "# To enable plotting graphs in Jupyter notebook\n",
    "%matplotlib inline "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "s3TbncQ0y_ES"
   },
   "source": [
    "This module covers,\n",
    "\n",
    "1) One sample and Two sample t-tests\n",
    "\n",
    "2) ANOVA\n",
    "\n",
    "3) Type I and Type II errors\n",
    "\n",
    "4) Probabilty Distributions\n",
    "\n",
    "5) Linear Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "i8UE0UE6y_EV"
   },
   "source": [
    "## Hypothesis Testing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "kImA4Gk-y_EW"
   },
   "source": [
    "The purpose of the test is to tell if there is any significant difference between two data sets.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "yItvHo3By_EZ"
   },
   "source": [
    "## Question 1 \n",
    "\n",
    "*A student is trying to decide between two GPUs. He want to use the GPU for his research to run Deep learning algorithms, so the only thing he is concerned with is speed.*\n",
    "\n",
    "*He picks a Deep Learning algorithm on a large data set and runs it on both GPUs 15 times, timing each run in hours. Results are given in the below lists GPU1 and GPU2.*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "nnQ2Ftdgy_Ea"
   },
   "source": [
    "Hint: You can import ttest function from scipy to perform t tests "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "JbQ05nufy_Ec"
   },
   "source": [
    "Given,\n",
    "\n",
    "Null Hypothesis : There is no significant difference between data sets\n",
    "\n",
    "Alternate Hypothesis : There is a significant difference\n",
    "\n",
    "*Do two-sample testing and check whether to reject Null Hypothesis or not.*\n",
    "\n",
    "#### Answer:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Zg9r4020y_Ee"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from scipy.stats import ttest_ind\n",
    "GPU1 = pd.DataFrame([11,9,10,11,10,12,9,11,12,9,11,12,9,10,9])\n",
    "GPU2 = pd.DataFrame([11,13,10,13,12,9,11,12,12,11,12,12,10,11,13])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Check for equal variance test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SD of GPU1:  0    1.175139\n",
      "dtype: float64\n",
      "SD of GPU2:  0    1.187234\n",
      "dtype: float64\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LeveneResult(statistic=array([0.05511811]), pvalue=array([0.81609376]))"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"SD of GPU1: \",GPU1.std())\n",
    "print(\"SD of GPU2: \",GPU2.std())\n",
    "\n",
    "from scipy.stats import levene\n",
    "\n",
    "levene(GPU1, GPU2)   # pvalue > alpha (0.05) so accept H0 that equal variance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ttest_indResult(statistic=array([-2.62762951]), pvalue=array([0.01379428]))"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ttest = ttest_ind(GPU1, GPU2, equal_var=True)\n",
    "ttest # pvalue < alpha (0.05) so reject H0 --> So accept the H1 --> So there is a significant difference"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "VqYtQt37y_El"
   },
   "source": [
    "## Question 2\n",
    "\n",
    "He is trying a third GPU which is GPU3."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "JM3dI6Kpy_Eo"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ttest_indResult(statistic=array([-1.49889438]), pvalue=array([0.14509211]))"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "GPU3 = pd.DataFrame([9,10,9,11,10,13,12,9,12,12,13,12,13,10,11])\n",
    "\n",
    "ttest1 = ttest_ind(GPU1, GPU3, equal_var=True)\n",
    "ttest1 \n",
    "\n",
    "#Assumption: Both the datasets (GPU1 & GPU 3) are random, independent, parametric & normally distributed"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "xccR5TB4y_Ev"
   },
   "source": [
    "*Do two-sample testing and check whether there is significant differene between speeds of two GPUs GPU1 and GPU3.*\n",
    "\n",
    "#### Answer:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "uQUx17MFy_Ex"
   },
   "source": [
    "pvalue(0.145) > alpha (0.05) so accept H0. So there is no significant difference"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "1ntmaD07y_E2"
   },
   "source": [
    "### ANOVA"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "G0pj7K4jy_E4"
   },
   "source": [
    "## Question 3 \n",
    "\n",
    "If you need to compare more than two data sets at a time, an ANOVA is your best bet. \n",
    "\n",
    "*The results from three experiments with overlapping 95% confidence intervals are given below, and we want to confirm that the results for all three experiments are not significantly different.*\n",
    "\n",
    "#Assumption: All the 3 datasets (e1,e2 & e3) are random, independent, parametric & normally distributed\n",
    "But before conducting ANOVA, test equality of variances (using Levene's test) is satisfied or not. If not, then mention that we cannot depend on the result of ANOVA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "AKzdGmBWy_E7"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LeveneResult(statistic=array([2.67417257]), pvalue=array([0.12259793]))"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "e1 = pd.DataFrame([1.595440,1.419730,0.000000,0.000000])\n",
    "e2 = pd.DataFrame([1.433800,2.079700,0.892139,2.384740])\n",
    "e3 = pd.DataFrame([0.036930,0.938018,0.995956,1.006970])\n",
    "levene(e1, e2, e3)  # pvalue> alpha (0.05) So accept H0: Variances of above datasets are equal"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "iZ5OpNTZy_FH"
   },
   "source": [
    "Hint - You can use stats.levene function and stats.f_oneway function."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "LS7fFigZy_FM"
   },
   "source": [
    "#### Answer:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "JeTpweFsy_FP"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2.51357623]\n",
      "[0.13574645]\n"
     ]
    }
   ],
   "source": [
    "from scipy.stats import f_oneway\n",
    "F, p = f_oneway(e1,e2,e3)\n",
    "print(F)\n",
    "print(p)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "pvalue(0.13)>alpha(0.05)  So accept the H0: We confirm that the results for all three experiments are not \n",
    "significantly different"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "-QkO7Zdky_FY"
   },
   "source": [
    "## Question 4 \n",
    "\n",
    "*In one or two sentences explain about **TypeI** and **TypeII** errors.*\n",
    "\n",
    "#### Answer:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "7GMrA5hiy_Fc"
   },
   "source": [
    "Type-I error(False positive) is a level of significance of the test which is also called as alpha, \n",
    "if pvalue > alpha then we can accept the null hypothesis test (H0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "6x2cZFpyy_Fi"
   },
   "source": [
    "Type-II error (False Negative) is power of test, most of the time in the real time situations it won't be considered as real experiment itself got failed. But sometimes it may consider for experimenting to know the cause of the failure."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "HjY6P__6y_Fm"
   },
   "source": [
    "# Question 5\n",
    "You are a manager of a chinese restaurant. You want to determine whether the waiting time to place an order has changed in the past month from its previous population mean value of 4.5 minutes. \n",
    "State the null and alternative hypothesis."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "lV5GQ6T8y_Fp"
   },
   "source": [
    "Answer: \n",
    "\n",
    "Null Hypothesis (H0): There is no significat change in the waiting time to place an order in the past month from its pervious population mean value of 4.5 minutes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "rKmKCB_Fy_Fu"
   },
   "source": [
    "Alternate Hypothesis (Ha): There is a significat change in the waiting time to place an order in the past month from its pervious population mean value of 4.5 minutes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Hh8GpzDny_Fy"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "gflwdh5qy_F4"
   },
   "source": [
    "# Question 6 \n",
    "Get the binomial distribution with n = 10, p = .7 and k = [0, 1, 2, 3, 4, 5, 6, 7]\n",
    "\n",
    "Hint - Use stats.binom.pmf() function for this.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "emY4gr4Jy_F_"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5.90490000e-06, 1.37781000e-04, 1.44670050e-03, 9.00169200e-03,\n",
       "       3.67569090e-02, 1.02919345e-01, 2.00120949e-01, 2.66827932e-01])"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import scipy\n",
    "scipy.stats.binom.pmf(k = [0, 1, 2, 3, 4, 5, 6, 7],n = 10, p = .7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "u43Z8ZM-y_F5"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "import warnings \n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## K-Nearest-Neighbors"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "KNN falls in the supervised learning family of algorithms. Informally, this means that we are given a labelled dataset consiting of training observations (x,y) and would like to capture the relationship between x and y. More formally, our goal is to learn a function h:X→Y so that given an unseen observation x, h(x) can confidently predict the corresponding output y.\n",
    "\n",
    "In this module we will explore the inner workings of KNN, choosing the optimal K values and using KNN from scikit-learn."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Overview\n",
    "\n",
    "1.Read the problem statement.\n",
    "\n",
    "2.Get the dataset.\n",
    "\n",
    "3.Explore the dataset.\n",
    "\n",
    "4.Pre-processing of dataset.\n",
    "\n",
    "5.Visualization\n",
    "\n",
    "6.Transform the dataset for building machine learning model.\n",
    "\n",
    "7.Split data into train, test set.\n",
    "\n",
    "7.Build Model.\n",
    "\n",
    "8.Apply the model.\n",
    "\n",
    "9.Evaluate the model.\n",
    "\n",
    "10.Finding Optimal K value\n",
    "\n",
    "11.Repeat 7,8,9 steps."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Problem statement\n",
    "\n",
    "### Dataset\n",
    "\n",
    "The data set we’ll be using is the Iris Flower Dataset which was first introduced in 1936 by the famous statistician Ronald Fisher and consists of 50 observations from each of three species of Iris (Iris setosa, Iris virginica and Iris versicolor). Four features were measured from each sample: the length and the width of the sepals and petals.\n",
    "\n",
    "**Attributes of the dataset:** https://archive.ics.uci.edu/ml/datasets/Iris\n",
    "\n",
    "**Train the KNN algorithm to be able to distinguish the species from one another given the measurements of the 4 features.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Question 1\n",
    "\n",
    "Import the data set and print 10 random rows from the data set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Sepal Length (in cm)</th>\n",
       "      <th>Sepal Width in (cm)</th>\n",
       "      <th>Petal length (in cm)</th>\n",
       "      <th>Petal width (in cm)</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Sepal Length (in cm)  Sepal Width in (cm)  Petal length (in cm)  \\\n",
       "0                   5.1                  3.5                   1.4   \n",
       "1                   4.9                  3.0                   1.4   \n",
       "2                   4.7                  3.2                   1.3   \n",
       "3                   4.6                  3.1                   1.5   \n",
       "4                   5.0                  3.6                   1.4   \n",
       "\n",
       "   Petal width (in cm)        Class  \n",
       "0                  0.2  Iris-setosa  \n",
       "1                  0.2  Iris-setosa  \n",
       "2                  0.2  Iris-setosa  \n",
       "3                  0.2  Iris-setosa  \n",
       "4                  0.2  Iris-setosa  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris_df = pd.read_csv('Iris.csv')\n",
    "iris_df.sample(n=10)\n",
    "iris_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Pre-processing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Question 2 - Estimating missing values\n",
    "\n",
    "*Its not good to remove the records having missing values all the time. We may end up loosing some data points. So, we will have to see how to replace those missing values with some estimated values (median) *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sepal Length (in cm)    0\n",
       "Sepal Width in (cm)     0\n",
       "Petal length (in cm)    0\n",
       "Petal width (in cm)     0\n",
       "Class                   0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris_df.describe()\n",
    "\n",
    "iris_df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Sepal Length (in cm)</th>\n",
       "      <th>Sepal Width in (cm)</th>\n",
       "      <th>Petal length (in cm)</th>\n",
       "      <th>Petal width (in cm)</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Sepal Length (in cm)  Sepal Width in (cm)  Petal length (in cm)  \\\n",
       "0                   5.1                  3.5                   1.4   \n",
       "1                   4.9                  3.0                   1.4   \n",
       "2                   4.7                  3.2                   1.3   \n",
       "3                   4.6                  3.1                   1.5   \n",
       "4                   5.0                  3.6                   1.4   \n",
       "\n",
       "   Petal width (in cm)        Class  \n",
       "0                  0.2  Iris-setosa  \n",
       "1                  0.2  Iris-setosa  \n",
       "2                  0.2  Iris-setosa  \n",
       "3                  0.2  Iris-setosa  \n",
       "4                  0.2  Iris-setosa  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris_df[\"Petal length (in cm)\"].median()\n",
    "\n",
    "iris_df[\"Petal length (in cm)\"]=iris_df[\"Petal length (in cm)\"].fillna(iris_df[\"Petal length (in cm)\"].median())\n",
    "iris_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Question 3 - Dealing with categorical data\n",
    "\n",
    "Change all the classes to numericals (0to2)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sepal Length (in cm)    float64\n",
       "Sepal Width in (cm)     float64\n",
       "Petal length (in cm)    float64\n",
       "Petal width (in cm)     float64\n",
       "Class                    object\n",
       "dtype: object"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris_df.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Question 4\n",
    "\n",
    "*Observe the independent variables variance and drop such variables having no variance or almost zero variance(variance < 0.1). They will be having almost no influence on the classification.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sepal Length (in cm)    0.685694\n",
       "Sepal Width in (cm)     0.188004\n",
       "Petal length (in cm)    3.113179\n",
       "Petal width (in cm)     0.582414\n",
       "dtype: float64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris_df.var()\n",
    "\n",
    "# there is no variance <0.1 so nothing can be dropped off"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Question 5\n",
    "\n",
    "*Plot the scatter matrix for all the variables.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Sepal Length (in cm)</th>\n",
       "      <th>Sepal Width in (cm)</th>\n",
       "      <th>Petal length (in cm)</th>\n",
       "      <th>Petal width (in cm)</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>Iris-setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Sepal Length (in cm)  Sepal Width in (cm)  Petal length (in cm)  \\\n",
       "0                   5.1                  3.5                   1.4   \n",
       "1                   4.9                  3.0                   1.4   \n",
       "2                   4.7                  3.2                   1.3   \n",
       "3                   4.6                  3.1                   1.5   \n",
       "4                   5.0                  3.6                   1.4   \n",
       "\n",
       "   Petal width (in cm)        Class  \n",
       "0                  0.2  Iris-setosa  \n",
       "1                  0.2  Iris-setosa  \n",
       "2                  0.2  Iris-setosa  \n",
       "3                  0.2  Iris-setosa  \n",
       "4                  0.2  Iris-setosa  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<matplotlib.axes._subplots.AxesSubplot object at 0x0000000009612BA8>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x00000000096C1668>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x00000000096ED978>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x0000000009717C88>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x0000000009960F98>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x0000000009960FD0>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x00000000099BC780>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x00000000099E3E10>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x0000000009A144E0>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x0000000009A3CB38>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x0000000009A6C208>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x0000000009A92898>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x0000000009ABCF28>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x0000000009EFC5F8>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x0000000009F22C88>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x0000000009F55358>]],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 16 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.plotting.scatter_matrix(iris_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'sns' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-17-f644dbc6e26b>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m()\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0msns\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpairplot\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0miris_df\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m: name 'sns' is not defined"
     ]
    }
   ],
   "source": [
    "sns.pairplot(iris_df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Question 6 - Model\n",
    "\n",
    "*Split the data in test and train sets and build the model and train and test on training and test sets respectively using **scikit-learn**. Print the Accuracy of the model with different values of **k=3,5,9**.*\n",
    "\n",
    "**Hint:** For accuracy you can check **accuracy_score()** in scikit-learn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sepal Length (in cm)     float64\n",
       "Sepal Width in (cm)      float64\n",
       "Petal length (in cm)     float64\n",
       "Petal width (in cm)      float64\n",
       "Class                   category\n",
       "dtype: object"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from scipy.stats import zscore\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "\n",
    "NNH = KNeighborsClassifier(n_neighbors= 3 , weights = 'distance' )\n",
    "iris_df['Class'] = iris_df.Class.astype('category')\n",
    "iris_df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Sepal Length (in cm)</th>\n",
       "      <th>Sepal Width in (cm)</th>\n",
       "      <th>Petal length (in cm)</th>\n",
       "      <th>Petal width (in cm)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Class</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Iris-setosa</th>\n",
       "      <td>50</td>\n",
       "      <td>50</td>\n",
       "      <td>50</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Iris-versicolor</th>\n",
       "      <td>50</td>\n",
       "      <td>50</td>\n",
       "      <td>50</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Iris-virginica</th>\n",
       "      <td>50</td>\n",
       "      <td>50</td>\n",
       "      <td>50</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 Sepal Length (in cm)  Sepal Width in (cm)  \\\n",
       "Class                                                        \n",
       "Iris-setosa                        50                   50   \n",
       "Iris-versicolor                    50                   50   \n",
       "Iris-virginica                     50                   50   \n",
       "\n",
       "                 Petal length (in cm)  Petal width (in cm)  \n",
       "Class                                                       \n",
       "Iris-setosa                        50                   50  \n",
       "Iris-versicolor                    50                   50  \n",
       "Iris-virginica                     50                   50  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris_df.groupby(\"Class\").count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "iris_df_features = iris_df.drop(labels=\"Class\", axis = 1)\n",
    "iris_df_features_z = iris_df_features.apply(zscore)\n",
    "#iris_df_features_z"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(150, 4)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris_df_labels = iris_df[\"Class\"]\n",
    "X = np.array(iris_df_features_z)\n",
    "X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(150,)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y = np.array(iris_df_labels)\n",
    "y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.30, random_state=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
       "           metric_params=None, n_jobs=1, n_neighbors=3, p=2,\n",
       "           weights='distance')"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "NNH.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model accuracy: 0.9777777777777777 \n"
     ]
    }
   ],
   "source": [
    "predicted_labels = NNH.predict(X_test)\n",
    "NNH.score(X_test, y_test)\n",
    "\n",
    "print(\"Model accuracy: {} \".format(NNH.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[14  0  0]\n",
      " [ 0 18  0]\n",
      " [ 0  1 12]]\n",
      "This Model has pridected wrongly for one time as per the confusion matrix above against test data\n"
     ]
    }
   ],
   "source": [
    "from sklearn import metrics\n",
    "\n",
    "print(metrics.confusion_matrix(y_test, predicted_labels))\n",
    "print(\"This Model has pridected wrongly for one time as per the confusion matrix above against test data\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Question 7 - Cross Validation\n",
    "\n",
    "Run the KNN with no of neighbours to be 1,3,5..19 and *Find the **optimal number of neighbours** from the above list using the Mis classification error"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Hint:\n",
    "\n",
    "Misclassification error (MSE) = 1 - Test accuracy score. Calculated MSE for each model with neighbours = 1,3,5...19 and find the model with lowest MSE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The optimal number of neighbors is 5\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import KFold\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "# creating odd list of K for KNN\n",
    "myList = list(range(1,20,2))\n",
    "\n",
    "\n",
    "# empty list that will hold cv scores\n",
    "cv_scores = []\n",
    "k_neighbors = []\n",
    "\n",
    "# perform 10-fold cross validation\n",
    "for k in myList:\n",
    "    knn = KNeighborsClassifier(n_neighbors=k)\n",
    "    scores = cross_val_score(knn, X_train, y_train, cv=10, scoring='accuracy')\n",
    "    cv_scores.append(scores.mean())\n",
    "    k_neighbors.append(k)\n",
    "\n",
    "\n",
    "MSE = [1 - x for x in cv_scores]\n",
    "min(MSE)\n",
    "MSE.index(min(MSE))\n",
    "best_k = myList[MSE.index(min(MSE))]\n",
    "print (\"The optimal number of neighbors is %d\" % best_k)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Question 8\n",
    "\n",
    "*Plot misclassification error vs k (with k value on X-axis) using matplotlib.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline \n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "fig_size = plt.rcParams[\"figure.figsize\"]\n",
    "fig_size[0] = 18\n",
    "fig_size[1] = 9\n",
    "plt.rcParams[\"figure.figsize\"] = fig_size\n",
    "\n",
    "plt.xlim(0,25)\n",
    "\n",
    "\n",
    "# plot misclassification error vs k\n",
    "plt.plot(k_neighbors, MSE)\n",
    "\n",
    "\n",
    "\n",
    "plt.xlabel('Number of Neighbors K')\n",
    "plt.ylabel('Misclassification Error')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "YRGRNbqay_Hi"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "name": "InternalLab_Residency2_Hypothesis_Testing_and_Linear_Regression-updated.ipynb",
   "provenance": [],
   "version": "0.3.2"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}