SVM-from-scratch.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "np.random.seed(6)\n",
    "import math"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "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": [
    "from sklearn.datasets.samples_generator import make_blobs\n",
    "(X,y)=make_blobs(n_samples=50,n_features=2,centers=2,cluster_std=1.05,random_state=40)\n",
    "X1=np.c_[np.ones((X.shape[0])),X]\n",
    "plt.scatter(X1[:,1],X1[:,2],marker='o',c=y)\n",
    "plt.axis([-5,10,-12,-1])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "positiveX=[]\n",
    "negativeX=[]\n",
    "for i,v in enumerate(y):\n",
    "    if v==0:\n",
    "        negativeX.append(X[i])\n",
    "    else:\n",
    "        positiveX.append(X[i])\n",
    "\n",
    "data_dict={-1:np.array(negativeX),1:np.array(positiveX)}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "w=[]\n",
    "b=[]\n",
    "\n",
    "max_feature_value=float('-inf')\n",
    "min_feature_value=float('+inf')\n",
    "\n",
    "for yi in data_dict:\n",
    "    if np.amax(data_dict[yi])>max_feature_value:\n",
    "        max_feature_value=np.amax(data_dict[yi])\n",
    "        \n",
    "    if np.amin(data_dict[yi])<min_feature_value:\n",
    "        min_feature_value=np.amin(data_dict[yi])\n",
    "        \n",
    "learning_rate=[max_feature_value*0.1, max_feature_value-0.01, max_feature_value*0.001,]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def SVM_Training(data_dict):\n",
    "    i=1\n",
    "    global w\n",
    "    global b\n",
    "    length_Wvector={}\n",
    "    transforms=[[1,1],[-1,1],[-1,-1],[1,-1]]\n",
    "    \n",
    "    b_step_size=2\n",
    "    b_multiple=5\n",
    "    w_optimum=max_feature_value*0.5\n",
    "    \n",
    "    for lrate in learning_rate:\n",
    "        w=np.array([w_optimum,w_optimum])\n",
    "        optimized=False\n",
    "        while not optimized:\n",
    "            for b in np.arange(-1*(max_feature_value*b_step_size),max_feature_value*b_step_size,lrate*b_multiple):\n",
    "                for transformation in transforms:\n",
    "                    w_t=w*transformation\n",
    "                    correctly_classified=True\n",
    "                \n",
    "                    for yi in data_dict:\n",
    "                        for xi in data_dict[yi]:\n",
    "                            if yi*(np.dot(w_t,xi)+b)<1:\n",
    "                                correctly_classified=False\n",
    "                    if correctly_classified:\n",
    "                        length_Wvector[np.linalg.norm(w_t)]=[w_t,b]\n",
    "            if w[0]<0:\n",
    "                optimized=True\n",
    "            else:\n",
    "                w=w-lrate\n",
    "        norms=sorted([n for n in length_Wvector])\n",
    "        minimum_Wlength=length_Wvector[norms[0]]\n",
    "        w=minimum_Wlength[0]\n",
    "        b=minimum_Wlength[1]\n",
    "        w_optimum=w[0]+lrate*2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "SVM_Training(data_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "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": [
    "colors={1:'r',-1:'b'}\n",
    "fig=plt.figure()\n",
    "ax=fig.add_subplot(1,1,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "def visualise(data_dict):\n",
    "    plt.scatter(X1[:,1],X1[:,2],marker='o',c=y)\n",
    "    \n",
    "    def hyperplane_value(x,w,b,v):\n",
    "        return (-w[0]*x-b+v)/w[1]\n",
    "    \n",
    "    datarange=(min_feature_value*0.9,max_feature_value*1.)\n",
    "    hyp_x_min=datarange[0]\n",
    "    hyp_x_max=datarange[1]\n",
    "    \n",
    "    psv1=hyperplane_value(hyp_x_min,w,b,1)\n",
    "    psv2=hyperplane_value(hyp_x_max,w,b,1)\n",
    "    ax.plot([hyp_x_min,hyp_x_max],[psv1,psv2],'k')\n",
    "    \n",
    "    nsv1=hyperplane_value(hyp_x_min,w,b,-1)\n",
    "    nsv2=hyperplane_value(hyp_x_max,w,b,-1)\n",
    "    ax.plot([hyp_x_min,hyp_x_max],[nsv1,nsv2],'k')\n",
    "    \n",
    "    db1=hyperplane_value(hyp_x_min,w,b,0)\n",
    "    db2=hyperplane_value(hyp_x_max,w,b,0)\n",
    "    ax.plot([hyp_x_min,hyp_x_max],[db1,db2],'y--')\n",
    "    \n",
    "    plt.axis([-5,10,-12,-1])\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "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": [
    "visualise(data_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1, 1, -1, 1, -1, "
     ]
    }
   ],
   "source": [
    "def predict(features):\n",
    "    dot_result=np.sign(np.dot(np.array(features),w)+b)\n",
    "    return dot_result.astype(int)\n",
    "\n",
    "for i in X[:5]:\n",
    "    print(predict(i), end=', ')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "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])"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "l=[]\n",
    "for xi in X:\n",
    "    l.append(predict(xi[:6]))\n",
    "l=np.array(l).astype(int)\n",
    "l"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-1.8171622 , -9.22909875])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X[4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "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])"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "for i,v in enumerate(y):\n",
    "    if v==0:\n",
    "        y[i]=-1\n",
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100.0\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "print(accuracy_score(y,l)*100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}