ALGORITHM9(I).ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "DM_ASSIGN_9(A).ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gpEI4QtuTSji"
      },
      "source": [
        "**QUESTION 1**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GbiIW9d2TVkt"
      },
      "source": [
        "%matplotlib inline\n",
        "import matplotlib.pyplot as plt\n",
        "from matplotlib import style\n",
        "style.use('ggplot')\n",
        "import numpy as np\n",
        "import pandas as pd"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QRWi2qofTVuN"
      },
      "source": [
        "class SVM(object):\n",
        "    def __init__(self,visualization=True):\n",
        "        self.visualization = visualization\n",
        "        self.colors = {1:'r',-1:'b'}\n",
        "        if self.visualization:\n",
        "            self.fig = plt.figure()\n",
        "            self.ax = self.fig.add_subplot(1,1,1)\n",
        "    \n",
        "    def fit(self,data):\n",
        "        #train with data\n",
        "        self.data = data\n",
        "        # { |\\w\\|:{w,b}}\n",
        "        opt_dict = {}\n",
        "        \n",
        "        transforms = [[1,1],[-1,1],[-1,-1],[1,-1]]\n",
        "        \n",
        "        all_data = np.array([])\n",
        "        for yi in self.data:\n",
        "            all_data = np.append(all_data,self.data[yi])\n",
        "                    \n",
        "        self.max_feature_value = max(all_data)         \n",
        "        self.min_feature_value = min(all_data)\n",
        "        all_data = None\n",
        "        \n",
        "        step_sizes = [self.max_feature_value * 0.1,\n",
        "                      self.max_feature_value * 0.01,\n",
        "                      self.max_feature_value * 0.001,]\n",
        "        \n",
        "        b_range_multiple = 5\n",
        "        b_multiple = 5\n",
        "        \n",
        "        latest_optimum = self.max_feature_value*10\n",
        "        \n",
        "        \n",
        "        for step in step_sizes:\n",
        "            w = np.array([latest_optimum,latest_optimum])\n",
        "            \n",
        "            optimized = False\n",
        "            while not optimized:\n",
        "                for b in np.arange(-1*self.max_feature_value*b_range_multiple,\n",
        "                                   self.max_feature_value*b_range_multiple,\n",
        "                                   step*b_multiple):\n",
        "                    for transformation in transforms:\n",
        "                        w_t = w*transformation\n",
        "                        found_option = True\n",
        "                        \n",
        "                        for i in self.data:\n",
        "                            for xi in self.data[i]:\n",
        "                                yi=i\n",
        "                                if not yi*(np.dot(w_t,xi)+b)>=1:\n",
        "                                    found_option=False\n",
        "                        if found_option:\n",
        "                            \n",
        "                            opt_dict[np.linalg.norm(w_t)]=[w_t,b]\n",
        "                \n",
        "                if w[0]<0:\n",
        "                    optimized=True\n",
        "                    print(\"optimized a step\")\n",
        "                else:\n",
        "                    w = w-step\n",
        "                    \n",
        "            norms = sorted([n for n in opt_dict])\n",
        "            opt_choice = opt_dict[norms[0]]\n",
        "\n",
        "            self.w=opt_choice[0]\n",
        "            self.b=opt_choice[1]\n",
        "            \n",
        "            latest_optimum = opt_choice[0][0]+step*2\n",
        "    \n",
        "    def predict(self,features):\n",
        "        classification = np.sign(np.dot(np.array(features),self.w)+self.b)\n",
        "        if classification!=0 and self.visualization:\n",
        "            self.ax.scatter(features[0],features[1],s=200,marker='*',c=self.colors[classification])\n",
        "        return (classification,np.dot(np.array(features),self.w)+self.b)\n",
        "    \n",
        "    def visualize(self):\n",
        "        [[self.ax.scatter(x[0],x[1],s=100,c=self.colors[i]) for x in data_dict[i]] for i in data_dict]\n",
        "        \n",
        "        def hyperplane(x,w,b,v):\n",
        "            #returns a x2 value on line when given x1\n",
        "            return (-w[0]*x-b+v)/w[1]\n",
        "       \n",
        "        hyp_x_min= self.min_feature_value*0.9\n",
        "        hyp_x_max = self.max_feature_value*1.1\n",
        "        \n",
        "        pav1 = hyperplane(hyp_x_min,self.w,self.b,1)\n",
        "        pav2 = hyperplane(hyp_x_max,self.w,self.b,1)\n",
        "        self.ax.plot([hyp_x_min,hyp_x_max],[pav1,pav2],'k')\n",
        "        \n",
        "        nav1 = hyperplane(hyp_x_min,self.w,self.b,-1)\n",
        "        nav2 = hyperplane(hyp_x_max,self.w,self.b,-1)\n",
        "        self.ax.plot([hyp_x_min,hyp_x_max],[nav1,nav2],'k')\n",
        "        \n",
        "        db1 = hyperplane(hyp_x_min,self.w,self.b,0)\n",
        "        db2 = hyperplane(hyp_x_max,self.w,self.b,0)\n",
        "        self.ax.plot([hyp_x_min,hyp_x_max],[db1,db2],'y--')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "d2VlGIb8TVw8",
        "outputId": "0e3983de-9cc1-4fc0-f5a3-d86ef76446db",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "data_dict = {-1:np.array([[1,7],[2,8],[3,8]]),1:np.array([[5,1],[6,-1],[7,3]])}\n",
        "data_dict[-1][0]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([1, 7])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TQsglVzUTV3M",
        "outputId": "a35ecfcc-3b46-4935-c75d-9485359a0879",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 315
        }
      },
      "source": [
        "svm = SVM() \n",
        "svm.fit(data=data_dict)\n",
        "svm.visualize()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "optimized a step\n",
            "optimized a step\n",
            "optimized a step\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LaDyNMNPTVtI",
        "outputId": "0185ae12-931b-4e9e-988c-de66cc060911",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "svm.predict([3,8])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(-1.0, -1.000000000000098)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qCu88RCNTpyC"
      },
      "source": [
        "df = pd.read_excel('data.xlsx')\n",
        "d = {-1.000000:[],1.000000:[]}"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4g3b1pEOTp3f"
      },
      "source": [
        "for i in range(350):\n",
        "  l = [df.iloc[i][0],df.iloc[1][1]]\n",
        "  d[df.iloc[i][2]].append(l)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VnsEu-pMTp7E"
      },
      "source": [
        "data2 = {}"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GNbz9GzKTp2H"
      },
      "source": [
        "data2[-1.0]=np.array(d[-1.0])\n",
        "data2[1.0]=np.array(d[1.0])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KbkHRzloTpwV",
        "outputId": "b1a5d34a-3e4c-411d-e4ae-6713f160a71f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "data2"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{-1.0: array([[8.40116942, 7.91302285],\n",
              "        [6.25113397, 7.91302285],\n",
              "        [8.26864113, 7.91302285],\n",
              "        [7.7914091 , 7.91302285],\n",
              "        [5.59151257, 7.91302285],\n",
              "        [8.250217  , 7.91302285],\n",
              "        [8.72897066, 7.91302285],\n",
              "        [7.22309596, 7.91302285],\n",
              "        [8.41904079, 7.91302285],\n",
              "        [5.9597379 , 7.91302285],\n",
              "        [8.15924215, 7.91302285],\n",
              "        [8.74007663, 7.91302285],\n",
              "        [8.41447651, 7.91302285],\n",
              "        [6.40790406, 7.91302285],\n",
              "        [6.89074985, 7.91302285],\n",
              "        [7.17688424, 7.91302285],\n",
              "        [8.97866274, 7.91302285],\n",
              "        [7.6156746 , 7.91302285],\n",
              "        [8.80718026, 7.91302285],\n",
              "        [7.21654775, 7.91302285],\n",
              "        [7.03281296, 7.91302285],\n",
              "        [8.2042945 , 7.91302285],\n",
              "        [8.10423299, 7.91302285],\n",
              "        [7.05016424, 7.91302285],\n",
              "        [5.68550018, 7.91302285],\n",
              "        [5.80737645, 7.91302285],\n",
              "        [8.29295085, 7.91302285],\n",
              "        [7.79453831, 7.91302285],\n",
              "        [5.61317444, 7.91302285],\n",
              "        [7.02863175, 7.91302285],\n",
              "        [7.50289161, 7.91302285],\n",
              "        [7.47773506, 7.91302285],\n",
              "        [8.38335799, 7.91302285],\n",
              "        [5.9413459 , 7.91302285],\n",
              "        [6.55040872, 7.91302285],\n",
              "        [5.50742709, 7.91302285],\n",
              "        [8.82887575, 7.91302285],\n",
              "        [8.18204777, 7.91302285],\n",
              "        [8.12956457, 7.91302285],\n",
              "        [5.98602693, 7.91302285],\n",
              "        [6.7226184 , 7.91302285],\n",
              "        [6.02969196, 7.91302285],\n",
              "        [7.23852237, 7.91302285],\n",
              "        [8.33028289, 7.91302285],\n",
              "        [7.71504078, 7.91302285],\n",
              "        [7.90940524, 7.91302285],\n",
              "        [7.73849576, 7.91302285],\n",
              "        [8.05262604, 7.91302285],\n",
              "        [8.50945998, 7.91302285],\n",
              "        [7.69434383, 7.91302285],\n",
              "        [6.13206981, 7.91302285],\n",
              "        [7.50657291, 7.91302285],\n",
              "        [6.07248155, 7.91302285],\n",
              "        [8.67118107, 7.91302285],\n",
              "        [5.77069975, 7.91302285],\n",
              "        [6.68487128, 7.91302285],\n",
              "        [7.53216142, 7.91302285],\n",
              "        [7.16332305, 7.91302285],\n",
              "        [8.3186217 , 7.91302285],\n",
              "        [7.35772349, 7.91302285],\n",
              "        [6.29558616, 7.91302285],\n",
              "        [7.98319741, 7.91302285],\n",
              "        [6.02021394, 7.91302285],\n",
              "        [7.80340663, 7.91302285],\n",
              "        [7.71893892, 7.91302285],\n",
              "        [6.30257188, 7.91302285],\n",
              "        [6.13779805, 7.91302285],\n",
              "        [6.08223378, 7.91302285],\n",
              "        [6.02362482, 7.91302285],\n",
              "        [6.20961589, 7.91302285],\n",
              "        [8.84235616, 7.91302285],\n",
              "        [5.55567919, 7.91302285],\n",
              "        [8.85130023, 7.91302285],\n",
              "        [5.589923  , 7.91302285],\n",
              "        [8.89888732, 7.91302285],\n",
              "        [6.54158487, 7.91302285],\n",
              "        [7.33775578, 7.91302285],\n",
              "        [8.51818724, 7.91302285],\n",
              "        [8.63741716, 7.91302285],\n",
              "        [6.16153694, 7.91302285],\n",
              "        [7.81251845, 7.91302285],\n",
              "        [8.79430828, 7.91302285],\n",
              "        [8.91497494, 7.91302285],\n",
              "        [5.87777339, 7.91302285],\n",
              "        [6.1261473 , 7.91302285],\n",
              "        [8.11292999, 7.91302285],\n",
              "        [5.67313983, 7.91302285],\n",
              "        [5.74949709, 7.91302285],\n",
              "        [7.21193932, 7.91302285],\n",
              "        [8.47462837, 7.91302285],\n",
              "        [8.98964505, 7.91302285],\n",
              "        [5.51537414, 7.91302285],\n",
              "        [7.39912612, 7.91302285],\n",
              "        [8.51471862, 7.91302285],\n",
              "        [8.68198659, 7.91302285],\n",
              "        [8.45872697, 7.91302285],\n",
              "        [8.5760551 , 7.91302285],\n",
              "        [8.11163783, 7.91302285],\n",
              "        [5.91121283, 7.91302285],\n",
              "        [7.2815779 , 7.91302285],\n",
              "        [6.09091117, 7.91302285],\n",
              "        [8.40889026, 7.91302285],\n",
              "        [8.748038  , 7.91302285],\n",
              "        [6.0931955 , 7.91302285],\n",
              "        [8.59308062, 7.91302285],\n",
              "        [6.85752013, 7.91302285],\n",
              "        [6.83899311, 7.91302285],\n",
              "        [6.45008523, 7.91302285],\n",
              "        [8.537591  , 7.91302285],\n",
              "        [8.0952586 , 7.91302285],\n",
              "        [7.06755456, 7.91302285],\n",
              "        [7.983736  , 7.91302285],\n",
              "        [8.80515949, 7.91302285],\n",
              "        [6.10941256, 7.91302285],\n",
              "        [6.35608699, 7.91302285],\n",
              "        [7.74325194, 7.91302285],\n",
              "        [8.33014378, 7.91302285],\n",
              "        [8.48679852, 7.91302285],\n",
              "        [6.89341176, 7.91302285],\n",
              "        [5.90422833, 7.91302285],\n",
              "        [5.78098333, 7.91302285],\n",
              "        [6.76163667, 7.91302285],\n",
              "        [6.24557619, 7.91302285],\n",
              "        [7.32810174, 7.91302285],\n",
              "        [7.4497339 , 7.91302285],\n",
              "        [8.01930544, 7.91302285],\n",
              "        [5.88642817, 7.91302285],\n",
              "        [6.25826582, 7.91302285],\n",
              "        [8.33857075, 7.91302285],\n",
              "        [5.98531659, 7.91302285],\n",
              "        [8.5866472 , 7.91302285],\n",
              "        [8.73244622, 7.91302285],\n",
              "        [5.54464447, 7.91302285],\n",
              "        [6.82005739, 7.91302285],\n",
              "        [6.08734091, 7.91302285],\n",
              "        [7.39077979, 7.91302285],\n",
              "        [5.85581852, 7.91302285],\n",
              "        [5.63743711, 7.91302285],\n",
              "        [8.76630189, 7.91302285],\n",
              "        [8.90057159, 7.91302285],\n",
              "        [6.76324807, 7.91302285],\n",
              "        [7.75471891, 7.91302285],\n",
              "        [5.73781556, 7.91302285],\n",
              "        [6.22769199, 7.91302285],\n",
              "        [5.63861346, 7.91302285],\n",
              "        [7.1427576 , 7.91302285],\n",
              "        [6.02533853, 7.91302285],\n",
              "        [8.96957422, 7.91302285],\n",
              "        [6.99471822, 7.91302285],\n",
              "        [8.84380236, 7.91302285]]),\n",
              " 1.0: array([[3.73834988e+00, 7.91302285e+00],\n",
              "        [1.60121779e+00, 7.91302285e+00],\n",
              "        [2.46467159e+00, 7.91302285e+00],\n",
              "        [1.10826377e+00, 7.91302285e+00],\n",
              "        [4.69636871e+00, 7.91302285e+00],\n",
              "        [2.41152645e+00, 7.91302285e+00],\n",
              "        [2.69998043e+00, 7.91302285e+00],\n",
              "        [1.10528550e+00, 7.91302285e+00],\n",
              "        [2.44991836e+00, 7.91302285e+00],\n",
              "        [2.91752235e+00, 7.91302285e+00],\n",
              "        [4.16348738e-01, 7.91302285e+00],\n",
              "        [3.30077315e+00, 7.91302285e+00],\n",
              "        [2.61526104e-01, 7.91302285e+00],\n",
              "        [2.78415267e+00, 7.91302285e+00],\n",
              "        [3.56012625e+00, 7.91302285e+00],\n",
              "        [2.43955313e+00, 7.91302285e+00],\n",
              "        [3.08800323e+00, 7.91302285e+00],\n",
              "        [1.06888787e+00, 7.91302285e+00],\n",
              "        [3.22829331e+00, 7.91302285e+00],\n",
              "        [1.90321146e+00, 7.91302285e+00],\n",
              "        [5.18564201e-01, 7.91302285e+00],\n",
              "        [1.88755759e+00, 7.91302285e+00],\n",
              "        [1.31430404e+00, 7.91302285e+00],\n",
              "        [1.20642806e+00, 7.91302285e+00],\n",
              "        [3.11462025e+00, 7.91302285e+00],\n",
              "        [2.61463881e+00, 7.91302285e+00],\n",
              "        [2.06618873e+00, 7.91302285e+00],\n",
              "        [1.08896018e+00, 7.91302285e+00],\n",
              "        [4.29276809e+00, 7.91302285e+00],\n",
              "        [4.30504127e+00, 7.91302285e+00],\n",
              "        [1.41969506e+00, 7.91302285e+00],\n",
              "        [3.07696424e+00, 7.91302285e+00],\n",
              "        [3.89744751e+00, 7.91302285e+00],\n",
              "        [4.77423656e+00, 7.91302285e+00],\n",
              "        [4.59801623e+00, 7.91302285e+00],\n",
              "        [1.92407852e+00, 7.91302285e+00],\n",
              "        [8.13215082e-01, 7.91302285e+00],\n",
              "        [3.98375710e+00, 7.91302285e+00],\n",
              "        [5.69085898e-01, 7.91302285e+00],\n",
              "        [7.94121493e-01, 7.91302285e+00],\n",
              "        [1.77914058e+00, 7.91302285e+00],\n",
              "        [4.23879282e+00, 7.91302285e+00],\n",
              "        [2.91390598e+00, 7.91302285e+00],\n",
              "        [2.93088833e+00, 7.91302285e+00],\n",
              "        [4.62918994e+00, 7.91302285e+00],\n",
              "        [2.87539134e+00, 7.91302285e+00],\n",
              "        [4.98869748e-02, 7.91302285e+00],\n",
              "        [4.04688643e+00, 7.91302285e+00],\n",
              "        [3.04404180e+00, 7.91302285e+00],\n",
              "        [2.39956840e+00, 7.91302285e+00],\n",
              "        [1.34220079e+00, 7.91302285e+00],\n",
              "        [1.29048192e+00, 7.91302285e+00],\n",
              "        [2.40504768e+00, 7.91302285e+00],\n",
              "        [1.13670742e+00, 7.91302285e+00],\n",
              "        [2.43011491e-01, 7.91302285e+00],\n",
              "        [8.46209073e-01, 7.91302285e+00],\n",
              "        [1.29223154e+00, 7.91302285e+00],\n",
              "        [9.89550288e-01, 7.91302285e+00],\n",
              "        [3.02846532e+00, 7.91302285e+00],\n",
              "        [4.11851823e+00, 7.91302285e+00],\n",
              "        [4.05307481e+00, 7.91302285e+00],\n",
              "        [4.01118508e+00, 7.91302285e+00],\n",
              "        [3.54056688e+00, 7.91302285e+00],\n",
              "        [4.29688927e+00, 7.91302285e+00],\n",
              "        [3.90542259e+00, 7.91302285e+00],\n",
              "        [1.01898426e+00, 7.91302285e+00],\n",
              "        [4.96671625e+00, 7.91302285e+00],\n",
              "        [4.68122641e-01, 7.91302285e+00],\n",
              "        [3.25309272e+00, 7.91302285e+00],\n",
              "        [1.07588263e+00, 7.91302285e+00],\n",
              "        [1.21936551e+00, 7.91302285e+00],\n",
              "        [1.69843903e+00, 7.91302285e+00],\n",
              "        [9.89311703e-01, 7.91302285e+00],\n",
              "        [2.53416688e+00, 7.91302285e+00],\n",
              "        [4.75378991e+00, 7.91302285e+00],\n",
              "        [1.97303609e+00, 7.91302285e+00],\n",
              "        [2.92232420e+00, 7.91302285e+00],\n",
              "        [3.03268623e+00, 7.91302285e+00],\n",
              "        [3.57321434e+00, 7.91302285e+00],\n",
              "        [2.00768177e+00, 7.91302285e+00],\n",
              "        [4.29345511e+00, 7.91302285e+00],\n",
              "        [4.60242491e+00, 7.91302285e+00],\n",
              "        [3.75419445e+00, 7.91302285e+00],\n",
              "        [1.42795853e+00, 7.91302285e+00],\n",
              "        [3.98407894e+00, 7.91302285e+00],\n",
              "        [7.13829151e-01, 7.91302285e+00],\n",
              "        [2.52275095e+00, 7.91302285e+00],\n",
              "        [3.05343976e+00, 7.91302285e+00],\n",
              "        [3.51898368e+00, 7.91302285e+00],\n",
              "        [1.91669966e+00, 7.91302285e+00],\n",
              "        [3.64341718e+00, 7.91302285e+00],\n",
              "        [4.43642340e+00, 7.91302285e+00],\n",
              "        [2.79241606e-01, 7.91302285e+00],\n",
              "        [6.91081713e-01, 7.91302285e+00],\n",
              "        [4.31532489e+00, 7.91302285e+00],\n",
              "        [2.10873241e+00, 7.91302285e+00],\n",
              "        [2.05657087e+00, 7.91302285e+00],\n",
              "        [4.79570670e+00, 7.91302285e+00],\n",
              "        [3.75122646e+00, 7.91302285e+00],\n",
              "        [4.90498557e+00, 7.91302285e+00],\n",
              "        [1.16757903e+00, 7.91302285e+00],\n",
              "        [4.81133806e-01, 7.91302285e+00],\n",
              "        [1.92291100e+00, 7.91302285e+00],\n",
              "        [2.50136275e+00, 7.91302285e+00],\n",
              "        [2.85127934e+00, 7.91302285e+00],\n",
              "        [4.88315798e+00, 7.91302285e+00],\n",
              "        [2.46427591e+00, 7.91302285e+00],\n",
              "        [2.00441433e+00, 7.91302285e+00],\n",
              "        [4.97495131e+00, 7.91302285e+00],\n",
              "        [1.30499004e+00, 7.91302285e+00],\n",
              "        [3.32662355e+00, 7.91302285e+00],\n",
              "        [4.82128308e+00, 7.91302285e+00],\n",
              "        [3.35575518e+00, 7.91302285e+00],\n",
              "        [1.49587428e+00, 7.91302285e+00],\n",
              "        [2.65563263e+00, 7.91302285e+00],\n",
              "        [7.31587586e-03, 7.91302285e+00],\n",
              "        [4.41894405e+00, 7.91302285e+00],\n",
              "        [2.02191929e+00, 7.91302285e+00],\n",
              "        [1.50602793e+00, 7.91302285e+00],\n",
              "        [4.75290796e+00, 7.91302285e+00],\n",
              "        [2.30324097e+00, 7.91302285e+00],\n",
              "        [1.43823756e+00, 7.91302285e+00],\n",
              "        [4.23133193e-01, 7.91302285e+00],\n",
              "        [2.91090668e+00, 7.91302285e+00],\n",
              "        [7.65343487e-01, 7.91302285e+00],\n",
              "        [3.65471531e-01, 7.91302285e+00],\n",
              "        [2.90282388e+00, 7.91302285e+00],\n",
              "        [1.43507376e+00, 7.91302285e+00],\n",
              "        [1.80960042e+00, 7.91302285e+00],\n",
              "        [3.62412347e+00, 7.91302285e+00],\n",
              "        [4.29155839e+00, 7.91302285e+00],\n",
              "        [1.73957843e+00, 7.91302285e+00],\n",
              "        [4.80872803e+00, 7.91302285e+00],\n",
              "        [4.76784526e+00, 7.91302285e+00],\n",
              "        [1.03018688e+00, 7.91302285e+00],\n",
              "        [3.84122430e+00, 7.91302285e+00],\n",
              "        [3.07765940e+00, 7.91302285e+00],\n",
              "        [4.59461244e+00, 7.91302285e+00],\n",
              "        [3.01268631e+00, 7.91302285e+00],\n",
              "        [3.51069114e+00, 7.91302285e+00],\n",
              "        [3.71837478e+00, 7.91302285e+00],\n",
              "        [1.92551138e+00, 7.91302285e+00],\n",
              "        [1.25751895e+00, 7.91302285e+00],\n",
              "        [1.83798687e-01, 7.91302285e+00],\n",
              "        [2.36061437e+00, 7.91302285e+00],\n",
              "        [3.22536164e+00, 7.91302285e+00],\n",
              "        [1.39484970e+00, 7.91302285e+00],\n",
              "        [2.58930455e+00, 7.91302285e+00],\n",
              "        [1.22833319e+00, 7.91302285e+00],\n",
              "        [1.48754087e+00, 7.91302285e+00],\n",
              "        [3.25237047e+00, 7.91302285e+00],\n",
              "        [4.45696653e+00, 7.91302285e+00],\n",
              "        [4.30553233e+00, 7.91302285e+00],\n",
              "        [1.04957538e+00, 7.91302285e+00],\n",
              "        [1.99546956e+00, 7.91302285e+00],\n",
              "        [4.43941512e+00, 7.91302285e+00],\n",
              "        [1.28264006e+00, 7.91302285e+00],\n",
              "        [4.83400802e+00, 7.91302285e+00],\n",
              "        [3.09578676e+00, 7.91302285e+00],\n",
              "        [8.26728833e-01, 7.91302285e+00],\n",
              "        [4.13099514e+00, 7.91302285e+00],\n",
              "        [3.27846551e+00, 7.91302285e+00],\n",
              "        [2.73226292e+00, 7.91302285e+00],\n",
              "        [1.25664642e+00, 7.91302285e+00],\n",
              "        [2.00781923e-01, 7.91302285e+00],\n",
              "        [1.16687706e+00, 7.91302285e+00],\n",
              "        [1.80556468e+00, 7.91302285e+00],\n",
              "        [3.16727518e+00, 7.91302285e+00],\n",
              "        [4.93049048e+00, 7.91302285e+00],\n",
              "        [1.03578323e+00, 7.91302285e+00],\n",
              "        [3.78541971e+00, 7.91302285e+00],\n",
              "        [4.43164055e+00, 7.91302285e+00],\n",
              "        [2.36114930e+00, 7.91302285e+00],\n",
              "        [7.94567630e-01, 7.91302285e+00],\n",
              "        [4.05460393e+00, 7.91302285e+00],\n",
              "        [2.38253876e+00, 7.91302285e+00],\n",
              "        [5.81437661e-01, 7.91302285e+00],\n",
              "        [4.37860306e+00, 7.91302285e+00],\n",
              "        [3.17593190e+00, 7.91302285e+00],\n",
              "        [4.86406745e-01, 7.91302285e+00],\n",
              "        [4.54219728e+00, 7.91302285e+00],\n",
              "        [1.75080108e-01, 7.91302285e+00],\n",
              "        [1.98748471e-01, 7.91302285e+00],\n",
              "        [4.94283946e+00, 7.91302285e+00],\n",
              "        [3.43090799e+00, 7.91302285e+00],\n",
              "        [1.88344596e+00, 7.91302285e+00],\n",
              "        [2.52162126e+00, 7.91302285e+00],\n",
              "        [3.81747840e+00, 7.91302285e+00],\n",
              "        [2.44373984e-01, 7.91302285e+00],\n",
              "        [4.79725009e-01, 7.91302285e+00],\n",
              "        [3.00827082e-01, 7.91302285e+00],\n",
              "        [4.09754437e+00, 7.91302285e+00],\n",
              "        [3.85739152e+00, 7.91302285e+00],\n",
              "        [9.78478655e-01, 7.91302285e+00],\n",
              "        [4.47559077e+00, 7.91302285e+00],\n",
              "        [3.42150073e+00, 7.91302285e+00],\n",
              "        [3.28423241e+00, 7.91302285e+00],\n",
              "        [4.95190601e+00, 7.91302285e+00],\n",
              "        [1.68458966e-01, 7.91302285e+00],\n",
              "        [2.12126635e+00, 7.91302285e+00]])}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5BE_70JVTpuh",
        "outputId": "f3dae1ae-cc7e-4fc8-92d8-e71c3254e953",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 315
        }
      },
      "source": [
        "svm2 = SVM()\n",
        "svm2.fit(data2)\n",
        "svm2.visualize()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "optimized a step\n",
            "optimized a step\n",
            "optimized a step\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e4CrodhVT6ut",
        "outputId": "836fb463-16f9-41d1-bbe7-1fcadfc59936",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "data2[-1][5]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([8.250217  , 7.91302285])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QdOUO5rPT8-h",
        "outputId": "41e5f20a-9af7-45e3-903d-0f258f369e15",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "svm2.predict(data2[1][56])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1.0, 24.74074460210661)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 19
        }
      ]
    }
  ]
}