5. Multi-class SVM Classifier using sklearn.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Library imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn import svm\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading dataset into pandas dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>x1</th>\n",
       "      <th>x2</th>\n",
       "      <th>y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    x1   x2  y\n",
       "0  1.0  6.0  1\n",
       "1  1.0  7.0  1\n",
       "2  2.0  5.0  1\n",
       "3  2.0  8.0  1\n",
       "4  4.0  2.0  2"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv('multi_class_svm.csv', names=['x1', 'x2', 'y'])\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## defining x and y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = data[['x1', 'x2']]\n",
    "\n",
    "y_1 = data.y.where(data.y == 1, other=-1)\n",
    "y_2 = data.y.where(data.y == 2, other=-1)\n",
    "y_3 = data.y.where(data.y == 3, other=-1)\n",
    "y_4 = data.y.where(data.y == 4, other=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEWCAYAAABrDZDcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x20356b3d780>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_list = [y_1, y_2, y_3, y_4]\n",
    "\n",
    "clf = svm.SVC(kernel='linear', C=1000)\n",
    "\n",
    "data_1 = data[data.y == 1]\n",
    "data_2 = data[data.y == 2]\n",
    "data_3 = data[data.y == 3]\n",
    "data_4 = data[data.y == 4]\n",
    "\n",
    "plt.plot(data_1.x1, data_1.x2, 'b+', label='Class 1')\n",
    "plt.plot(data_2.x1, data_2.x2, 'rx', label='Class 2')\n",
    "plt.plot(data_3.x1, data_3.x2, 'g*', label='Class 3')\n",
    "plt.plot(data_4.x1, data_4.x2, 'o', label='Class 4')\n",
    "\n",
    "for i, y_i in enumerate(y_list, start=1):\n",
    "    clf.fit(x, y_i)\n",
    "    \n",
    "    W = clf.coef_[0]\n",
    "    b = clf.intercept_[0]\n",
    "\n",
    "    xx = np.linspace(0, 12)\n",
    "    yy = -b/W[1] - xx*W[0]/W[1]\n",
    "    \n",
    "    for c in range(len(yy)):\n",
    "        if yy[c] < 0:\n",
    "            yy[c] = 0\n",
    "        elif yy[c] > 14:\n",
    "            yy[c] = 14\n",
    "    \n",
    "    plt.plot(xx, yy, label='Decision Boundary ' + str(i))\n",
    "\n",
    "plt.title('Multiclass Support Vector Machine')\n",
    "plt.xlabel('x1')\n",
    "plt.ylabel('x2')\n",
    "#plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}