LDAonIRIS.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Working on a multi-class problem using the famous Iris dataset that has been deposited on the UCI machine learning repository\n",
    "(https://archive.ics.uci.edu/ml/datasets/Iris).\n",
    "\n",
    "The iris dataset contains measurements for 150 iris flowers from three different species.\n",
    "\n",
    "The three classes in the Iris dataset:\n",
    "1. Iris-setosa (n=50)\n",
    "2. Iris-versicolor (n=50)\n",
    "3. Iris-virginica (n=50)\n",
    "\n",
    "The four features of the Iris dataset:\n",
    "1. sepal length in cm\n",
    "2. sepal width in cm\n",
    "3. petal length in cm\n",
    "4. petal width in cm\n",
    "\n",
    "<img src=\"iris_petal_sepal.png\">\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import make_classification\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import seaborn as sns; sns.set();\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from numpy import pi\n",
    "from sklearn.naive_bayes import GaussianNB"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Importing the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>sepal-length</th>\n",
       "      <th>sepal-width</th>\n",
       "      <th>petal-length</th>\n",
       "      <th>petal-width</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>145</th>\n",
       "      <td>6.7</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.2</td>\n",
       "      <td>2.3</td>\n",
       "      <td>Iris-virginica</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146</th>\n",
       "      <td>6.3</td>\n",
       "      <td>2.5</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1.9</td>\n",
       "      <td>Iris-virginica</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>147</th>\n",
       "      <td>6.5</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.2</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Iris-virginica</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148</th>\n",
       "      <td>6.2</td>\n",
       "      <td>3.4</td>\n",
       "      <td>5.4</td>\n",
       "      <td>2.3</td>\n",
       "      <td>Iris-virginica</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149</th>\n",
       "      <td>5.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.1</td>\n",
       "      <td>1.8</td>\n",
       "      <td>Iris-virginica</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     sepal-length  sepal-width  petal-length  petal-width           Class\n",
       "145           6.7          3.0           5.2          2.3  Iris-virginica\n",
       "146           6.3          2.5           5.0          1.9  Iris-virginica\n",
       "147           6.5          3.0           5.2          2.0  Iris-virginica\n",
       "148           6.2          3.4           5.4          2.3  Iris-virginica\n",
       "149           5.9          3.0           5.1          1.8  Iris-virginica"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url = \"https://archive.ics.uci.edu/ml/machine-learning-databases/iris/iris.data\"\n",
    "names = ['sepal-length', 'sepal-width', 'petal-length', 'petal-width', 'Class']\n",
    "dataset = pd.read_csv(url, names=names)\n",
    "\n",
    "dataset.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data preprocessing\n",
    "\n",
    "Once dataset is loaded into a pandas data frame object, the first step is to divide dataset into features and corresponding labels and then divide the resultant dataset into training and test sets. The following code divides data into labels and feature set:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = dataset.iloc[:, 0:4].values\n",
    "y = dataset.iloc[:, 4].values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The above script assigns the first four columns of the dataset i.e. the feature set to X variable while the values in the fifth column (labels) are assigned to the y variable.\n",
    "\n",
    "The following code divides data into training and test sets:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Feature Scaling\n",
    "\n",
    "We will now perform feature scaling as part of data preprocessing too. For this task, we will be using scikit learn `StandardScalar`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "sc = StandardScaler()\n",
    "X_train = sc.fit_transform(X_train)\n",
    "X_test = sc.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " LDA on the IRIS dataset and computing the overall accuracy of the classifier."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "### WRITE YOUR CODE HERE ####\n",
    "\n",
    "def LDA(X_train, y_train):\n",
    "    # Collecting all class 0 and class 1 into separate variables\n",
    "    class_1 = X_train[np.argwhere(y_train == 'Iris-setosa')[:, 0:4]].reshape(-1,4)\n",
    "    class_2 = X_train[np.argwhere(y_train == 'Iris-versicolor')[:, 0:4]].reshape(-1,4)\n",
    "    class_3 = X_train[np.argwhere(y_train == 'Iris-virginica')[:, 0:4]].reshape(-1,4)\n",
    "\n",
    "    # Getting number of examples in each class\n",
    "    num_class_1_samples = np.shape(class_1)[0]\n",
    "    num_class_2_samples = np.shape(class_2)[0]\n",
    "    num_class_3_samples = np.shape(class_3)[0]\n",
    "\n",
    "    # Computing class mean for each label and calculating the difference between them.\n",
    "    class_1_mean = np.mean(class_1,axis = 0).reshape(4,1);\n",
    "    class_2_mean = np.mean(class_2,axis = 0).reshape(4,1);\n",
    "    class_3_mean = np.mean(class_3,axis = 0).reshape(4,1);\n",
    "    overall_mean = np.mean(X_train,axis=0).reshape(4,1);\n",
    "\n",
    "    SB_1 = num_class_1_samples*(class_1_mean-overall_mean)@(class_1_mean-overall_mean).T;\n",
    "    SB_2 = num_class_2_samples*(class_2_mean-overall_mean)@(class_2_mean-overall_mean).T;\n",
    "    SB_3 = num_class_3_samples*(class_3_mean-overall_mean)@(class_3_mean-overall_mean).T;\n",
    "\n",
    "    SB = SB_1 + SB_2 + SB_3;\n",
    "\n",
    "    #Calculating covariance matrix\n",
    "    cov_1 = (class_1.T - class_1_mean)@(class_1.T - class_1_mean).T\n",
    "    cov_2 = (class_2.T - class_2_mean)@(class_2.T - class_2_mean).T\n",
    "    cov_3 = (class_3.T - class_3_mean)@(class_3.T - class_3_mean).T\n",
    "\n",
    "    SW = cov_1 + cov_2 + cov_3;\n",
    "\n",
    "    eigvals, eigvecs = np.linalg.eig(np.linalg.pinv(SW)@(SB))\n",
    "    \n",
    "    # Getting the eigenvectors with the maximum eigenvalue.\n",
    "    idx = np.real(eigvals).argsort()[::-1]\n",
    "    weights = np.atleast_1d(eigvecs[:, idx])[:, :2]\n",
    "    return np.real(weights)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "lda_weights = LDA(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\t\tLDA on train data\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\t\tLDA on test data\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "X_train_lda = X_train@lda_weights\n",
    "X_test_lda = X_test@lda_weights\n",
    "\n",
    "def plot_LDA(X_input,label):\n",
    "    data = pd.DataFrame(X_input)\n",
    "    data['class'] = label\n",
    "    data.columns=[\"LD1\",\"LD2\",\"class\"]\n",
    "    data.head()\n",
    "    markers = ['o', 'x','s']\n",
    "    colors = ['b', 'g','r']\n",
    "    customPalette = sns.set_palette(sns.color_palette(colors))\n",
    "    sns.lmplot(x=\"LD1\", y=\"LD2\", data=data, hue='class', markers=markers,fit_reg=False,legend=False)\n",
    "    plt.legend(loc='upper center')\n",
    "    plt.show()\n",
    "    \n",
    "print(\"\\n\\t\\tLDA on train data\")\n",
    "plot_LDA(X_train_lda,y_train)\n",
    "print(\"\\t\\tLDA on test data\")\n",
    "plot_LDA(X_test_lda,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set accuracy:  97.5%\n",
      "Test set accuracy:  100.0%\n"
     ]
    }
   ],
   "source": [
    "model = GaussianNB(priors = None)\n",
    "model.fit(X_train_lda,y_train)\n",
    "\n",
    "acc= str(model.score(X_train_lda,y_train)*100)+'%'\n",
    "\n",
    "print(\"Training set accuracy: \",acc)\n",
    "\n",
    "    \n",
    "model = GaussianNB(priors = None)\n",
    "model.fit(X_test_lda,y_test)\n",
    "\n",
    "acc= str(model.score(X_test_lda,y_test)*100)+'%'\n",
    "\n",
    "print(\"Test set accuracy: \", acc)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_plot = sc.fit_transform(X)@lda_weights\n",
    "Y_plot = y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "N = 500\n",
    "X1 = np.linspace(-2.5, 2.5, N)\n",
    "X2 = np.linspace(-1, 1, N)\n",
    "X1, X2 = np.meshgrid(X1,X2)\n",
    "\n",
    "temp = np.copy(Y_plot)\n",
    "temp[temp == 'Iris-setosa'] = int(0)\n",
    "temp[temp == 'Iris-versicolor'] = int(1)\n",
    "temp[temp == 'Iris-virginica'] = int(2)\n",
    "temp = temp.tolist()\n",
    "\n",
    "model = GaussianNB(priors = None)\n",
    "model.fit(X_plot,temp)\n",
    "\n",
    "temp = np.array([model.predict([[x1,x2]])[0] for x1, x2 in zip(X1.flatten(), X2.flatten())])\n",
    "Z = temp.reshape(X1.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_decision(model,X1,X2,Z):\n",
    "    fig = plt.figure(figsize = (7,7))\n",
    "    ax = fig.add_subplot()\n",
    "\n",
    "    ax.contour( X1, X2, Z, 2, alpha = 0.6, colors = ('r','g','b'))\n",
    "    ax.contourf( X1, X2, Z, 2, alpha = 0.6, colors = ('r','g','b'))\n",
    "    ax.scatter(X_plot[np.argwhere(Y_plot == 'Iris-setosa')[:, 0]][:,0], X_plot[np.argwhere(Y_plot == 'Iris-setosa')[:, 0]][:,1], alpha = 1, color='r',label ='Iris-setosa')\n",
    "    ax.scatter(X_plot[np.argwhere(Y_plot == 'Iris-versicolor')[:, 0]][:,0], X_plot[np.argwhere(Y_plot == 'Iris-versicolor')[:, 0]][:,1], alpha = 1, color='g',label ='Iris-versicolor')\n",
    "    ax.scatter(X_plot[np.argwhere(Y_plot == 'Iris-virginica')[:, 0]][:,0], X_plot[np.argwhere(Y_plot == 'Iris-virginica')[:, 0]][:,1], alpha = 1, color='b',label ='Iris-virginica')\n",
    "\n",
    "    ax.set_xlabel('LD1')\n",
    "    ax.set_ylabel('LD2')\n",
    "\n",
    "    ax.set_title('LDA Multi-Class Decision Boundary')\n",
    "    plt.legend()\n",
    "    plt.show()\n",
    "\n",
    "plot_decision(model,X1,X2,Z)"
   ]
  }
 ],
 "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}