EXample_SVM_Primal_Dual_Classification_.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/OptimizationExpert/Pyomo/blob/main/EXample_SVM_Primal_Dual_Classification_.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VvJL0bOSIm84"
      },
      "source": [
        "Developed by Alireza Soroudi\n",
        "alireza.soroudi@gmail.com\n",
        "\n",
        "Linkedin News letter : https://www.linkedin.com/newsletters/optimization-in-open-source-6874020019009859585/\n",
        "\n",
        "Git hub: https://github.com/OptimizationExpert/Pyomo\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 61,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "w_3yuwV8XtMF",
        "outputId": "789dcd6b-0d87-494c-fdff-0b7078caaedc"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: pyomo in /usr/local/lib/python3.10/dist-packages (6.7.3)\n",
            "Requirement already satisfied: ply in /usr/local/lib/python3.10/dist-packages (from pyomo) (3.11)\n"
          ]
        }
      ],
      "source": [
        "!pip install pyomo\n",
        "import numpy as np\n",
        "from random import randint\n",
        "import random\n",
        "import time\n",
        "from math import ceil\n",
        "import matplotlib.pyplot as plt # Data visualization\n",
        "from pyomo.environ import *\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "!wget -N -q \"https://github.com/thomasfork/ipopt_linux/raw/main/ipopt-linux64.zip\"\n",
        "!unzip -o -q ipopt-linux64\n",
        "opt = SolverFactory('ipopt', executable='/content/ipopt')\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df=pd.DataFrame()\n",
        "N = 20\n",
        "special = [0]*10+[1,-0.9]\n",
        "\n",
        "df['x1'] = [random.random() for i in range(N)]\n",
        "df['x2'] = [(2-0.4*random.random())*x+1-random.random() + random.sample([-2,2],1)[0] +random.sample(special,1)[0]  for x in df['x1']]\n",
        "Y = []\n",
        "for i in range(N):\n",
        "  a,b = 2-0.1*random.random() , 0.5*random.random()\n",
        "  y= df.loc[i,'x2'] -  (a*df.loc[i,'x1']-b)\n",
        "  label = 1 if y > 0 else 0\n",
        "  Y.append(label)\n",
        "df['output'] = Y\n",
        "df.describe()"
      ],
      "metadata": {
        "id": "4aq0puvcQmM0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "outputId": "d6ce38bd-ea43-4708-efb5-0318305efb7a"
      },
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "              x1         x2     output\n",
              "count  20.000000  20.000000  20.000000\n",
              "mean    0.490289   1.494480   0.550000\n",
              "std     0.287524   2.150709   0.510418\n",
              "min     0.069764  -1.786427   0.000000\n",
              "25%     0.250316  -0.457237   0.000000\n",
              "50%     0.485279   2.498731   1.000000\n",
              "75%     0.798228   3.310282   1.000000\n",
              "max     0.954320   4.267448   1.000000"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-9c073cbe-530b-4780-8081-da82c76e2aeb\" class=\"colab-df-container\">\n",
              "    <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>x1</th>\n",
              "      <th>x2</th>\n",
              "      <th>output</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>20.000000</td>\n",
              "      <td>20.000000</td>\n",
              "      <td>20.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>0.490289</td>\n",
              "      <td>1.494480</td>\n",
              "      <td>0.550000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>0.287524</td>\n",
              "      <td>2.150709</td>\n",
              "      <td>0.510418</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.069764</td>\n",
              "      <td>-1.786427</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>0.250316</td>\n",
              "      <td>-0.457237</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>0.485279</td>\n",
              "      <td>2.498731</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>0.798228</td>\n",
              "      <td>3.310282</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>0.954320</td>\n",
              "      <td>4.267448</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-9c073cbe-530b-4780-8081-da82c76e2aeb')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-9c073cbe-530b-4780-8081-da82c76e2aeb button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-9c073cbe-530b-4780-8081-da82c76e2aeb');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-439e6262-ce51-4a38-999c-4020bd9286a4\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-439e6262-ce51-4a38-999c-4020bd9286a4')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-439e6262-ce51-4a38-999c-4020bd9286a4 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "summary": "{\n  \"name\": \"df\",\n  \"rows\": 8,\n  \"fields\": [\n    {\n      \"column\": \"x1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 6.9086331205225875,\n        \"min\": 0.0697644579208746,\n        \"max\": 20.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          0.4902893681506244,\n          0.48527908798547953,\n          20.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 6.780372848770985,\n        \"min\": -1.7864272079551724,\n        \"max\": 20.0,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          1.4944802522945748,\n          2.4987311971974933,\n          20.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"output\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 6.8785387081225755,\n        \"min\": 0.0,\n        \"max\": 20.0,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          0.55,\n          1.0,\n          0.5104177855340405\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 62
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Clean the data"
      ],
      "metadata": {
        "id": "ez4Vb1L3qzo3"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df"
      ],
      "metadata": {
        "id": "0vmcQWl0O257",
        "outputId": "2e4a8cea-4d56-4858-f516-151913b80f85",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 676
        }
      },
      "execution_count": 63,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "          x1        x2  output\n",
              "0   0.822319  4.267448       1\n",
              "1   0.508125  3.682452       1\n",
              "2   0.602915  3.113640       1\n",
              "3   0.370307 -0.941068       0\n",
              "4   0.069764 -1.786427       0\n",
              "5   0.954320  4.194997       1\n",
              "6   0.797793  2.668337       1\n",
              "7   0.179543 -1.414971       0\n",
              "8   0.462433 -0.566184       0\n",
              "9   0.265365  3.214378       1\n",
              "10  0.353555  3.597996       1\n",
              "11  0.286864 -1.249000       0\n",
              "12  0.160554  2.388695       1\n",
              "13  0.205169  2.714797       1\n",
              "14  0.542187  3.864249       1\n",
              "15  0.102677  2.608767       1\n",
              "16  0.852984  0.086284       0\n",
              "17  0.799536 -0.317349       0\n",
              "18  0.578547 -0.420921       0\n",
              "19  0.890831  0.183486       0"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-80c03b46-51b8-48f8-bbc2-de1ada789615\" class=\"colab-df-container\">\n",
              "    <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>x1</th>\n",
              "      <th>x2</th>\n",
              "      <th>output</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.822319</td>\n",
              "      <td>4.267448</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.508125</td>\n",
              "      <td>3.682452</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.602915</td>\n",
              "      <td>3.113640</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.370307</td>\n",
              "      <td>-0.941068</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.069764</td>\n",
              "      <td>-1.786427</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>0.954320</td>\n",
              "      <td>4.194997</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>0.797793</td>\n",
              "      <td>2.668337</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>0.179543</td>\n",
              "      <td>-1.414971</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>0.462433</td>\n",
              "      <td>-0.566184</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>0.265365</td>\n",
              "      <td>3.214378</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>0.353555</td>\n",
              "      <td>3.597996</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>0.286864</td>\n",
              "      <td>-1.249000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>0.160554</td>\n",
              "      <td>2.388695</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>0.205169</td>\n",
              "      <td>2.714797</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>0.542187</td>\n",
              "      <td>3.864249</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>0.102677</td>\n",
              "      <td>2.608767</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>0.852984</td>\n",
              "      <td>0.086284</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>0.799536</td>\n",
              "      <td>-0.317349</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>0.578547</td>\n",
              "      <td>-0.420921</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>0.890831</td>\n",
              "      <td>0.183486</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-80c03b46-51b8-48f8-bbc2-de1ada789615')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-80c03b46-51b8-48f8-bbc2-de1ada789615 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-80c03b46-51b8-48f8-bbc2-de1ada789615');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-a3c86a71-e3e2-4ad4-a23c-3ebd8f4734aa\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-a3c86a71-e3e2-4ad4-a23c-3ebd8f4734aa')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-a3c86a71-e3e2-4ad4-a23c-3ebd8f4734aa button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "  <div id=\"id_8c681427-4c1d-4542-9a39-ab6eb1c5ac38\">\n",
              "    <style>\n",
              "      .colab-df-generate {\n",
              "        background-color: #E8F0FE;\n",
              "        border: none;\n",
              "        border-radius: 50%;\n",
              "        cursor: pointer;\n",
              "        display: none;\n",
              "        fill: #1967D2;\n",
              "        height: 32px;\n",
              "        padding: 0 0 0 0;\n",
              "        width: 32px;\n",
              "      }\n",
              "\n",
              "      .colab-df-generate:hover {\n",
              "        background-color: #E2EBFA;\n",
              "        box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "        fill: #174EA6;\n",
              "      }\n",
              "\n",
              "      [theme=dark] .colab-df-generate {\n",
              "        background-color: #3B4455;\n",
              "        fill: #D2E3FC;\n",
              "      }\n",
              "\n",
              "      [theme=dark] .colab-df-generate:hover {\n",
              "        background-color: #434B5C;\n",
              "        box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "        filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "        fill: #FFFFFF;\n",
              "      }\n",
              "    </style>\n",
              "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('df')\"\n",
              "            title=\"Generate code using this dataframe.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "    <script>\n",
              "      (() => {\n",
              "      const buttonEl =\n",
              "        document.querySelector('#id_8c681427-4c1d-4542-9a39-ab6eb1c5ac38 button.colab-df-generate');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      buttonEl.onclick = () => {\n",
              "        google.colab.notebook.generateWithVariable('df');\n",
              "      }\n",
              "      })();\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "df",
              "summary": "{\n  \"name\": \"df\",\n  \"rows\": 20,\n  \"fields\": [\n    {\n      \"column\": \"x1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.2875242187318021,\n        \"min\": 0.0697644579208746,\n        \"max\": 0.9543198880835225,\n        \"num_unique_values\": 20,\n        \"samples\": [\n          0.8223190965106344,\n          0.7995356800490375,\n          0.10267668732750002\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 2.150708903512592,\n        \"min\": -1.7864272079551724,\n        \"max\": 4.267448151739432,\n        \"num_unique_values\": 20,\n        \"samples\": [\n          4.267448151739432,\n          -0.3173489701811305,\n          2.60876704906537\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"output\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0,\n        \"min\": 0,\n        \"max\": 1,\n        \"num_unique_values\": 2,\n        \"samples\": [\n          0,\n          1\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 63
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.scatter(df['x1'],df['x2'],c=df['output'])\n",
        "for i in range(len(df)):\n",
        "  plt.text(df.loc[i,'x1'],df.loc[i,'x2'],s=str(i))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 430
        },
        "id": "2V56FDCSwVnO",
        "outputId": "07722a2f-7c95-4212-917c-bddfb20f9f60"
      },
      "execution_count": 64,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "J_set = [c for c in df.columns if c not in [\"output\"] ]\n",
        "\n",
        "J_set"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4ohtoAz29E2z",
        "outputId": "1c42b041-3918-4198-a1a3-6d1f55e5d7f9"
      },
      "execution_count": 65,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['x1', 'x2']"
            ]
          },
          "metadata": {},
          "execution_count": 65
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#Data Processing\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "wwHh7pzAO1TM"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df"
      ],
      "metadata": {
        "id": "dVsmHHns-6kp",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 676
        },
        "outputId": "d72f36fb-ebea-49bf-a635-d104554af887"
      },
      "execution_count": 66,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "          x1        x2  output\n",
              "0   0.822319  4.267448       1\n",
              "1   0.508125  3.682452       1\n",
              "2   0.602915  3.113640       1\n",
              "3   0.370307 -0.941068       0\n",
              "4   0.069764 -1.786427       0\n",
              "5   0.954320  4.194997       1\n",
              "6   0.797793  2.668337       1\n",
              "7   0.179543 -1.414971       0\n",
              "8   0.462433 -0.566184       0\n",
              "9   0.265365  3.214378       1\n",
              "10  0.353555  3.597996       1\n",
              "11  0.286864 -1.249000       0\n",
              "12  0.160554  2.388695       1\n",
              "13  0.205169  2.714797       1\n",
              "14  0.542187  3.864249       1\n",
              "15  0.102677  2.608767       1\n",
              "16  0.852984  0.086284       0\n",
              "17  0.799536 -0.317349       0\n",
              "18  0.578547 -0.420921       0\n",
              "19  0.890831  0.183486       0"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-48a94c18-f4e7-49ea-8ebc-c402784882c9\" class=\"colab-df-container\">\n",
              "    <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>x1</th>\n",
              "      <th>x2</th>\n",
              "      <th>output</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.822319</td>\n",
              "      <td>4.267448</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.508125</td>\n",
              "      <td>3.682452</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.602915</td>\n",
              "      <td>3.113640</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.370307</td>\n",
              "      <td>-0.941068</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.069764</td>\n",
              "      <td>-1.786427</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>0.954320</td>\n",
              "      <td>4.194997</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>0.797793</td>\n",
              "      <td>2.668337</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>0.179543</td>\n",
              "      <td>-1.414971</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>0.462433</td>\n",
              "      <td>-0.566184</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>0.265365</td>\n",
              "      <td>3.214378</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>0.353555</td>\n",
              "      <td>3.597996</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>0.286864</td>\n",
              "      <td>-1.249000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>0.160554</td>\n",
              "      <td>2.388695</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>0.205169</td>\n",
              "      <td>2.714797</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>0.542187</td>\n",
              "      <td>3.864249</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>0.102677</td>\n",
              "      <td>2.608767</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>0.852984</td>\n",
              "      <td>0.086284</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>0.799536</td>\n",
              "      <td>-0.317349</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>0.578547</td>\n",
              "      <td>-0.420921</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>0.890831</td>\n",
              "      <td>0.183486</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-48a94c18-f4e7-49ea-8ebc-c402784882c9')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-48a94c18-f4e7-49ea-8ebc-c402784882c9 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-48a94c18-f4e7-49ea-8ebc-c402784882c9');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-358e7730-7856-4a0e-b52c-dc2eafa670b1\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-358e7730-7856-4a0e-b52c-dc2eafa670b1')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-358e7730-7856-4a0e-b52c-dc2eafa670b1 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "  <div id=\"id_993c9a8f-ad1f-40f2-b8d8-d57fa881ed04\">\n",
              "    <style>\n",
              "      .colab-df-generate {\n",
              "        background-color: #E8F0FE;\n",
              "        border: none;\n",
              "        border-radius: 50%;\n",
              "        cursor: pointer;\n",
              "        display: none;\n",
              "        fill: #1967D2;\n",
              "        height: 32px;\n",
              "        padding: 0 0 0 0;\n",
              "        width: 32px;\n",
              "      }\n",
              "\n",
              "      .colab-df-generate:hover {\n",
              "        background-color: #E2EBFA;\n",
              "        box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "        fill: #174EA6;\n",
              "      }\n",
              "\n",
              "      [theme=dark] .colab-df-generate {\n",
              "        background-color: #3B4455;\n",
              "        fill: #D2E3FC;\n",
              "      }\n",
              "\n",
              "      [theme=dark] .colab-df-generate:hover {\n",
              "        background-color: #434B5C;\n",
              "        box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "        filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "        fill: #FFFFFF;\n",
              "      }\n",
              "    </style>\n",
              "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('df')\"\n",
              "            title=\"Generate code using this dataframe.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "    <script>\n",
              "      (() => {\n",
              "      const buttonEl =\n",
              "        document.querySelector('#id_993c9a8f-ad1f-40f2-b8d8-d57fa881ed04 button.colab-df-generate');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      buttonEl.onclick = () => {\n",
              "        google.colab.notebook.generateWithVariable('df');\n",
              "      }\n",
              "      })();\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "df",
              "summary": "{\n  \"name\": \"df\",\n  \"rows\": 20,\n  \"fields\": [\n    {\n      \"column\": \"x1\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.2875242187318021,\n        \"min\": 0.0697644579208746,\n        \"max\": 0.9543198880835225,\n        \"num_unique_values\": 20,\n        \"samples\": [\n          0.8223190965106344,\n          0.7995356800490375,\n          0.10267668732750002\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"x2\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 2.150708903512592,\n        \"min\": -1.7864272079551724,\n        \"max\": 4.267448151739432,\n        \"num_unique_values\": 20,\n        \"samples\": [\n          4.267448151739432,\n          -0.3173489701811305,\n          2.60876704906537\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"output\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0,\n        \"min\": 0,\n        \"max\": 1,\n        \"num_unique_values\": 2,\n        \"samples\": [\n          0,\n          1\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 66
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "eoezWqgXje1e"
      },
      "execution_count": 66,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "n = len(df)\n",
        "from sklearn.model_selection import train_test_split\n",
        "train_df, test_df = train_test_split(df, test_size=0.2, random_state=42)\n",
        "test_df.head(), len(test_df)"
      ],
      "metadata": {
        "id": "Cv5EhKYtkSpR",
        "outputId": "51c42b25-dfdf-46af-d687-99b9081b9cbc",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(          x1        x2  output\n",
              " 0   0.822319  4.267448       1\n",
              " 17  0.799536 -0.317349       0\n",
              " 15  0.102677  2.608767       1\n",
              " 1   0.508125  3.682452       1,\n",
              " 4)"
            ]
          },
          "metadata": {},
          "execution_count": 67
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "qsifuCh2FhHW"
      },
      "execution_count": 67,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "![Screenshot 2024-08-08 at 22.32.18.png](data:image/png;base64,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)"
      ],
      "metadata": {
        "id": "czXw9avc6EeX"
      }
    },
    {
      "cell_type": "markdown",
      "source": [],
      "metadata": {
        "id": "5WER2_FwCyjK"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "I_set = [i for i in train_df.index]\n",
        "m = len(J_set)\n",
        "Y={i:2*train_df.loc[i,\"output\"]-1 for i in I_set}\n",
        "Y"
      ],
      "metadata": {
        "id": "359022mJLLZn",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f553434a-0596-4dc6-97d8-5e6edcddaf17"
      },
      "execution_count": 68,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{8: -1,\n",
              " 5: 1,\n",
              " 11: -1,\n",
              " 3: -1,\n",
              " 18: -1,\n",
              " 16: -1,\n",
              " 13: 1,\n",
              " 2: 1,\n",
              " 9: 1,\n",
              " 19: -1,\n",
              " 4: -1,\n",
              " 12: 1,\n",
              " 7: -1,\n",
              " 10: 1,\n",
              " 14: 1,\n",
              " 6: 1}"
            ]
          },
          "metadata": {},
          "execution_count": 68
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model = AbstractModel()\n",
        "model.i = Set(initialize= I_set)\n",
        "model.j= Set(initialize= J_set)\n",
        "C = 0.5\n",
        "def rule_x(model,i,j):\n",
        "  return train_df.loc[i,j]\n",
        "model.X = Param(model.i,model.j,initialize= rule_x, mutable = True)\n",
        "\n",
        "model.w = Var(model.j, bounds=(-200,200), within=Reals)\n",
        "model.beta = Var(bounds=(-200,200), within=Reals)\n",
        "model.eps = Var(model.i, bounds=(0,200), within=Reals)\n",
        "\n",
        "def rule_con1(model,i):\n",
        "  return  Y[i]*( sum(model.w[j]*model.X[i,j] for j in model.j)+ model.beta) >= 1- model.eps[i]\n",
        "model.con1= Constraint(model.i, rule=rule_con1)\n",
        "\n",
        "def rule_of(model):\n",
        "  return 0.5*sum(model.w[j]**2 for j in model.j) + C * sum(model.eps[i] for i in model.i)\n",
        "model.obj = Objective(rule=rule_of, sense=minimize)\n",
        "instance = model.create_instance()\n",
        "st= time.time()\n",
        "result = opt.solve(instance)\n",
        "ft= time.time()\n",
        "print(\"time\", ft-st)\n",
        "print(\"obj\", value(instance.obj))"
      ],
      "metadata": {
        "id": "LxcJDJCfWxsS",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "48a77216-b992-4fcc-b1c0-ffe7869f366d"
      },
      "execution_count": 69,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time 0.0240933895111084\n",
            "obj 0.37062761847982184\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "error = 0\n",
        "for i in test_df.index:\n",
        "  actual = 2*test_df.loc[i,\"output\"]-1\n",
        "  predict = sum(test_df.loc[i,j]*value(instance.w[j]) for j in instance.j) + value(instance.beta)\n",
        "  if predict >=1:\n",
        "    predict = 1\n",
        "  else:\n",
        "    predict = -1\n",
        "  error += 1 if abs(actual-predict)>0 else 0\n",
        "print(\"error\", error/len(test_df))"
      ],
      "metadata": {
        "id": "F8_rHbIqWxxM",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ebdd1354-84bc-43f0-bc3c-01c86cfe27f2"
      },
      "execution_count": 70,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "error 0.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "value(instance.w['x1']),value(instance.w['x2']), value(instance.beta)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "07Lb_FpAzRqb",
        "outputId": "414bd881-fd9c-44bd-aa28-892e2db6e72a"
      },
      "execution_count": 71,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(-0.2706605209687525, 0.8173114991319608, -0.9088525223505491)"
            ]
          },
          "metadata": {},
          "execution_count": 71
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.scatter(df['x1'],df['x2'],c=df['output'])\n",
        "#wx+b =1.  w[x1]*x1 + w[x2]*x2 + b = 0\n",
        "#wx+b =-1\n",
        "#wx+b =0\n",
        "X = np.linspace(-1,1,100)\n",
        "\n",
        "y1 = [(0-value(instance.beta)-value(instance.w['x1'])*x)/value(instance.w['x2'])  for x in X]\n",
        "plt.plot(X,y1 , c= 'g', lw = 3)\n",
        "\n",
        "y2 = [(1-value(instance.beta)-value(instance.w['x1'])*x)/value(instance.w['x2'])  for x in X]\n",
        "plt.plot(X,y2 , c= 'r', lw = 3)\n",
        "\n",
        "\n",
        "y3= [(-1-value(instance.beta)-value(instance.w['x1'])*x)/value(instance.w['x2'])  for x in X]\n",
        "plt.plot(X,y3 , c= 'r', lw = 3)\n",
        "\n",
        "\n",
        "for i in range(len(df)):\n",
        "  plt.text(df.loc[i,'x1'],df.loc[i,'x2'],s=str(i))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 430
        },
        "id": "36AasrYoxyb4",
        "outputId": "96625c7f-0a1a-4eb5-91c9-392c25ecf6f8"
      },
      "execution_count": 72,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiIAAAGdCAYAAAAvwBgXAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABNSklEQVR4nO3deXwTZf4H8M8kbdP7bimVtlwKylEQpYIoIBXosggrPxWWxXKsqIsi1kXBVRDRBYRFdlkWEDl03RWP5fACFBQqyg1VQECKIFdbeqdN27RJnt8fpaFpkzRpk0ySft6vV18lM5PMdwhtPsx853kkIYQAERERkQwUchdARERErReDCBEREcmGQYSIiIhkwyBCREREsmEQISIiItkwiBAREZFsGESIiIhINgwiREREJBsfuQuwxmAw4OrVqwgJCYEkSXKXQ0RERDYQQqCsrAzx8fFQKKyf83DrIHL16lUkJCTIXQYRERE1w6VLl9CuXTur27h1EAkJCQFQeyChoaEyV0NERES2UKvVSEhIMH6OW+PWQaTuckxoaCiDCBERkYexpa2CzapEREQkGwYRIiIikg2DCBEREcmGQYSIiIhkwyBCRETkAitWrED79u3h7++PlJQUHDx4UO6S3AKDCBERkZN98MEHyMjIwNy5c3H06FEkJydj2LBhuHbtmtylyY5BhIiIyMmWLl2Kxx57DJMmTcJtt92GVatWITAwEOvWrZO7NNkxiBARETlRdXU1jhw5gtTUVOMyhUKB1NRU7Nu3T8bK3AODCBERkYMJYYCoPgRRuQn5l7dAr9ejTZs2Jtu0adMGubm5TqvhlVdegSRJJl9du3Z12v6ay61HViUiIvI0Qvs9hPplQH+p9rFaV/u9chuEuMulk7h269YNO3fuND728XG/j333q4iIiMhDiepDEMVTABiMy6IjlVAqgdxf3wYq2gNBkwEAeXl5iIuLc2o9Pj4+Tt9HS/HSDBERkYOIsjcAiOtftfz8JPTpqcLXeysgypZBGMphMBiwa9cu9OvXz6n1nD17FvHx8ejYsSPGjx+PixcvOnV/zcEzIkRERA4gdL8CNT+YXTfj8QhMeiYPfZKvIeWedfj7qpPQaDSYNGmSY2sQ1UD1AcBQir63R2P9+vXo2rUrcnJyMG/ePNxzzz04ceKETbPiugqDCBERkSMYCiyuemRUCAoK9XjljULkznwOvXrdju3btzdqYG0JUfE+RNlSQJQCAIb3BaDsACnsNfTsOQwpKSlISkrChx9+iClTpjhsvy3FIEJEROQIilirq6dNDse0yeGQwhZDChjl0F0LzQaIsr82XqH/FaIoHYh8D+Hht+OWW25Bdna2Q/fdUuwRISIicgDJJwHw7QOrH61SAKBKtby+GYShvPZMiFkGAAaIsjdQXl6Oc+fOoW3btg7df0sxiBARETmIFDILgBKWPl6lkOchKYIcu1PtlwCqGi2eOS8fe76vwIVLWnz//Xf43eg0KJVKjBs3zrH7byFemiEiInIQyS8ZiPw3hHoOoPv5xgpFNKTg5yAFjnH8TvX5qA0/epPFl3N0GP+nXBQWGxATpcSAAbdh//79iImJcXwNLcAgQkRE5ECS3+1A1KeA7idAfxmQwgC/OyBJTvrIVcagYQgBgPdXmV6CkaLfrr185GZ4aYaIiMjBJEmC5NsNkv8wSKq7nBdCAEA1FIC/lQ0UgG8ftwwhAIMIERGRR5MUwZBCnrOwVgFAASnkeVeWZBcGESIiIg8nBaVDCp0HSOGmK5TtIUW+A8mvtyx12YI9IkRERF5AChwHBIwxjqwKZQLg29Olk+w1B4MIERGRl5AkP0B1j9xl2MVll2YWLlwISZIwY8YMV+2SiIiI3JxLgsihQ4ewevVq9OzZ0xW7IyIiIg/h9CBSXl6O8ePHY82aNYiIiHD27oiIyE1lZmZi5MiRiI+PhyRJ2LJli8Vtn3jiCUiShGXLlrmsPpKH04PItGnTMGLECKSmNj22vlarhVqtNvkiIiLvoNFokJycjBUrVljdbvPmzdi/fz/i4+NdVBnJyanNqhs3bsTRo0dx6NAhm7ZfsGAB5s2b58ySiIhIJmlpaUhLS7O6zZUrV/D0009jx44dGDFihIsqIzk57YzIpUuX8Mwzz+A///kP/P2tjfh2w+zZs1FaWmr8unTpkrPKIyIiN2MwGDBhwgTMnDkT3bp1k7scchGnnRE5cuQIrl27httvv924TK/XIzMzE//85z+h1WqhVCpNnqNSqaBSqZxVEhERyUAIYdNYFosWLYKPjw+mT5/ugqrIXTgtiAwZMgTHjx83WTZp0iR07doVL7zwQqMQQkRE3kPor0FUbAAq/geIEghFFBDwSO0IoIrwRtsfOXIEf//733H06FG3H4CLHMtpQSQkJATdu3c3WRYUFISoqKhGy4mIyHsI3UWIorGAoRjGWWENBYBmJUTVViByY6PnfPvtt7h27RoSExONy/R6PZ577jksW7YMFy5ccE3x5HIcWZWIiBxKlP7ZNIQYGQB9DoR6bqPnTJgwodHdlcOGDcOECRMwadIk5xVLsnNpENm9e7crd0dERC4mak4DNVlm15VrDMg+rwXwBQDg/PnzyMrKQmRkJBITExEVFWWyva+vL+Li4tClSxcnV01y4hkRIiJynJqTFlcd/qEKQ8ZcMT7OyMgAAKSnp2PDhg3OrozcFIMIERE5RGZmJhYvXIQjR39BTp4e/1vXFqPTgo3rB/UPhD7nZgCAFLEWUhOTs7EvpHVw2aR3RETk3TQaDXr2GoDlf42zvqEUAPjebn0bajV4RoSIiByibuRUQ6k/MGWBha0kIPBRSIogl9ZG7otnRIiIyKGk0Dn1HilNv/uPhBT8jKtLIjfGMyJERORQklQ7rYcU/CwQUAzo8wFlW0gBYwDfnhywjEzwjAgRkQfLzMzEyJEjER8fD0mSsGXLFpP1QgjMmTMHbdu2RUBAAFJTU3H27NkW71cIAVH1FQxF6TDkpcBwbRAM6gUQusvGbSTfm6EI+ysUkWugCHsVkl8yQwg1wiBCROTBNBoNkpOTsWLFCrPr33jjDfzjH//AqlWrcODAAQQFBWHYsGGoqqpq9j6FMECUzoYomQZUHwREMWC4ClS8C1EwAqLathnXiQBemiEi8mh1DaLmCCGwbNkyvPTSSxg1ahQA4N1330WbNm2wZcsWjB07tnk7rfwQqNp0/UH90VP1ALQQxU8273WpVeIZESIiL3X+/Hnk5uaaDJ0eFhaGlJQU7Nu3r1mvKYSA0KwH0PgSS7nGgKwTlcg6nm/cf1ZWFi5evNisfVHrwDMiREQeRhgqAP05AErApzMkyc/sdrm5uQCANm3amCxv06aNcZ39O9cA+vNmV3HkVGoOBhEiIg8hRCVE2TKgciMgKmsXShFA0EQgaKqLqrB8Iv3GyKlKIOBBKMJed1FN5Ml4aYaIyAMIUQ1R9Eeg4p0bIQQARDFE+ZsQpbMaPScurnaE07y8PJPleXl5xnX2khSBgE93WP/40EPy69es16fWh0GEiMgTVG4Bag4BMJhfX7W10aIOHTogLi4Ou3btMi5Tq9U4cOAA+vVrflCQgv5ouQ4oAUUbwH9os1+fWhdemiEi8gCi4n3UNogKk+XlGgOyz9eg7v+VdQ2ikZGRSExMxIwZM/Daa6/h5ptvRocOHfDyyy8jPj4eo0ePbnYtUsBvIHRnAc0K1I6YWnfnjAQowmontLPQt0LUEIMIEZEn0F9CwxACNN0g+vzzz0Oj0WDq1KkoKSnBgAEDsH37dvj7+7eoHEXIMxD+90Fo/gvoTgFSACT/YUDAg5AUoS16bWpdJCFE43/ZbkKtViMsLAylpaUIDeU/bCJqvQz59wH6y1a2UAB+d0MRudZlNRFZYs/nN3tEiIg8gf9oWP+VbYAUMNJFxRA5DoMIEZEHkALHA4oI3JjNtr7a8UTgb36EVSJ3xiBCROQBJGUUpMj/AMoO15coYfwV7tsbUsS7kCSVXOURNRubVYmIPITk0xGI/hyoOQxUHwUkJeDXH5LvbXKXRtRsDCJERB5EkiTA787aLyIvwEszREREJBsGESIiIpINgwgRERHJhkGEiIiIZMMgQkRERLJhECEicrGysjLMmDEDSUlJCAgIQP/+/XHo0CG5yyKSBYMIEZGL/fGPf8RXX32Ff//73zh+/DiGDh2K1NRUXLlypeknE3kZTnpHRORClZWVCAkJwdatWzFixAjj8j59+iAtLQ2vvfaajNUROQYnvSMiclM6nQ56vR7+/v4mywMCArB3716ZqiKSD4MIEZELhYSEoF+/fpg/fz6uXr0KvV6P9957D/v27UNOTo7c5RG5HIMIEZETiZofYSh5Doa8FBjy+sJQ/Ce8+/ZzEELgpptugkqlwj/+8Q+MGzcOCgV/JVPrw3/1REROIio+hih8CKj6AhDFgCgBtN+gY+Rs7P58HMrLy3Hp0iUcPHgQNTU16Nixo9wlu50FCxbgzjvvREhICGJjYzF69GicOXNG7rLIgRhEiIicQOjOQaj/AkAA0NdbU/tnUb4Ugb4/oW3btiguLsaOHTswatQoOUp1a3v27MG0adOwf/9+fPXVV6ipqcHQoUOh0WjkLo0chHfNEBE5gUH9GlDxH5iGkFo7vtFACAW63HYffsn/A2bOnAl/f398++238PX1dX2xHiQ/Px+xsbHYs2cP7r33XrnLIQt41wwRkdyqD8NcCAGA0jIDnn4xD7el/BuPPvooBgwYgB07djCE2KC0tBQAEBkZKXMl5Cg8I0JE5ASGgjGA7rj1jRRtoIj91jUFeRBhKAOqDwGoAXxuheSTCAAwGAx44IEHUFJSwlud3Zw9n98+LqqJiKh1Ud0L6E4CMFjYQFm7DRkJUQNRthSoeA+A9sZyvwGQwl7HtKdexYkTJxhCvAyDCBGRE0iBj0Bo3gZQjdqG1YYEpMBHXVyV+xJCQJQ+X3uHUcO/r+p9eGpqH3z2pYTMzL1o166dLDWSc7BHhIjICSRlHKSIVQBUMP1VqwSghBT2N0i+XeQpzh3V/ABUfY6GIUQIgadfzMGWL/Kx85NH0aFDB3nqI6fhGREiIieRVHcDMbuAyg8htHsBGAC/OyAFjIPkw//V1ycqN6E2pJk2+D41Ox/vby7D5vVtEeKzA7m5zwAAwsLCEBAQ4PpCyeHYrEpERLIzFD0OVH/TaLmy7Vmz269fvx4TJ050clXUXGxWJSIiz6KMgbkzIvqcm288kMKhaHPQpWWR87FHhIiIZCcFjIalcVdqKYHAh1xUDbmSU4PIypUr0bNnT4SGhiI0NBT9+vXDtm3bnLlLIiLyRL59ANX9ACQzK5WAIhJS4EQXF0Wu4NQg0q5dOyxcuBBHjhzB4cOHcd9992HUqFE4efKkM3dLREQeRpIkSOFvAgHjATQYYdb3dkiRGyEpY2SpjZzL5c2qkZGRWLx4MaZMmdLktmxWJSJqfYShGKg+AIhqwPc2SD6d5S6J7OSWzap6vR4fffQRNBoN+vXrZ3YbrVYLrfbGaHpqtdpV5RERkZuQFBGA/3C5yyAXcXqz6vHjxxEcHAyVSoUnnngCmzdvxm233WZ22wULFiAsLMz4lZCQ4OzyiIiISEZOvzRTXV2NixcvorS0FB9//DHefvtt7Nmzx2wYMXdGJCEhgZdmiIiIPIg9l2Zc3iOSmpqKTp06YfXq1U1uyx4RIiIiz2PP57fLxxExGAwmZz2IiIio9XJqs+rs2bORlpaGxMRElJWV4b///S92796NHTt2OHO3RERE5CGcGkSuXbuGRx99FDk5OQgLC0PPnj2xY8cO3H///c7cLREREXkIp16aWbt2LS5cuACtVotr165h586dDCFE5DUyMzMxcuRIxMfHQ5IkbNmyxWT9K6+8gq5duyIoKAgRERFITU3FgQMH5CmWyE1xrhkiombSaDRITk7GihUrzK6/5ZZb8M9//hPHjx/H3r170b59ewwdOhT5+fkurpTIfbn8rhl78K4ZIvIUkiRh8+bNGD16tMVt6n6n7dy5E0OGDHFdcUQu5tZ3zRARtUbV1dV46623EBYWhuTkZLnLISe6cuUK/vCHPyAqKgoBAQHo0aMHDh8+LHdZbstlQ7wTEXkDoc8FDIWAIgaSMrbJ7T/77DOMHTsWFRUVaNu2Lb766itER0e7oFKSQ3FxMe6++24MHjwY27ZtQ0xMDM6ePYuIiAi5S3NbDCJERDYQ1T9AlC0Bam40mwq/uyGFzITka37aCgAYPHgwsrKyUFBQgDVr1uDhhx/GgQMHEBvbdIghz7No0SIkJCRg/fr1xmUdOnSQsSL3x0szROTVmrqzZeLEibVT0Nf7Gj7cdMI1UX0Iouj3QM0h0xev3gdR+AhE9Q8W9x8UFITOnTvjrrvuwtq1a+Hj44O1a9c66vDIzXzyySe444478NBDDyE2Nha9e/fGmjVr5C7LrfGMCBF5tbo7WyZPnowHH3zQ7DbDhw83+R+sSqUy/lkIAVH6IgA9AEODZxoA1ECo59hcD0eX9i5C6ADt1xDavQD0+OWXbKxc+QsyMjLw4osv4tChQ5g+fTr8/PyQnp4ud7luiUGEiLxaWloa0tLSrG6jUqkQFxdnfmXNUUD/q9lV5RoDss9rAWQBAM6fP4+srCxERkYiKioKr7/+Oh544AG0bdsWBQUFWLFiBa5cuYKHHnqoBUdE7kLozkMUTwH0l1H7cSpgMNTgjuQgvD73AUiq3ujduzdOnDiBVatWMYhYwEszRNTq7d69G7GxsejSpQuefPJJFBYW3lhpIYQAwOEfqtDn/ovoc/9FAEBGRgZ69+6NOXPmQKlU4vTp0xgzZgxuueUWjBw5EoWFhfj222/RrVs3Zx8SOZkwaCCKHgX0OdeX6ADo0TbWB7fe4gNR/BiE7gIA4NZbb8XFixflKtXt8YwIEXkVIXRAzTHAUAIoEyH5drG6/fDhw/Hggw+iQ4cOOHfuHF588UWkpaVh3759UCqVgBRm8bmD+gdCn3MzAECKeAeSqp/J+k2bNrX4eMhNVX0KGPIaLe7f1x8/Z2sB6CAq3oUUOgc///wzkpKSXF9jdTWQn3/jq6DA9HH95cOGAW++6foawSBCRF5EVGyCKP8bYLgxcqnw6QYp9BVIfubH7hg7dqzxzz169EDPnj3RqVMn7N69u3bQMdXdgBQMiHLLO1ZEAn53OOw4yP2Jqi8BSABMxwSdMTUCA0ZewoK/5+Oh0Ztx+FxnvPXWW3jrrbdauEMBaDSWw4S5oKFW2/76nTq1rL4WYBAhIq8gKv4DoZ7XeIXuFETRH4Co9216nY4dOyI6OhrZ2dkYMmQIJMkfCJ4GUbbI4nOk4BmQJN/mlk6eSFSgYQgBgDt7+eN/69riL38txPw3f0CHDvOxbNkyjB8/3nRDgwEoLbU9VOTnA1VVzjseGacdYBAhIo8nDOUQ6jcsrL1+Z0uZpfWmLl++jMLCQrRt2/bGwsDJkEQ1RPly1N49o7z+3RdSyLOQAseafzHyXr5dgZofUPvvAIBOACV6oFCP3wYp8NuMaKAkDoqKB4AffgAeeaRx0NDrZT0EEwwiREQtoP0SQKXZVTfubNkDwPTOlsjISMybNw9jxoxBXFwczp07h+effx6dO3fGsGHDjK8hSRIQ/CQQOBao2g6hz4ekbAP4p0FScB4sr6TVWj1DIeX9AuRcAAr1QJEeKDFAanSC5CqAo66vvSGFAoiKAmJigOjo2u/1v6Kjgfh42cpjECEiz6fPw42zFKYO/1CFIWOuGB9nZGQAANLT07Fy5Ur8+OOPeOedd1BSUoL4+HgMHToU8+fPNxlLpI6kiAACx0Fy1nGQcwgBlJdbb9hsuK6szOpLyvpvwNfXcqgw9xURASiVclZsFYMIEXk+RTQaDzZWy+TOluhdkHwSTNbv2LHD2dWRoxkMQEmJff0V7jyIXGCgbYGiLnyEhgKS98RhBhEi8nz+QwH1PADVFjZQAL49G4UQchM6nfnwYOnsRWGhe/VXNBQebjlEmFsWGCh3xbJiECEijycpwoDgpyDKl5pbC0CCFDLT5tfLzMzE4sWLceTIEeTk5GDz5s0YPXo0AKCmpgYvvfQSvvjiC/zyyy8ICwtDamoqFi5ciHgZr7O7lcpK2wJF3briYrkrtkyhqA0L1i6D1F8XHV176YRsxiBCRN4h6HFIkl/tnS1Cc2O5Mh5S6GuQ/O60+aWszU9TUVGBo0eP4uWXX0ZycjKKi4vxzDPP4IEHHsDhw4cddTTuQ4ja8Sjs6a/QaJp+Xbn4+TV96aNhf4WCg5A7kySEaHwjtJtQq9UICwtDaWkpQkPZmU5ETROiEtB+CxhKAWUi4HcnJKn5HySSJJmcETHn0KFD6Nu3L3799VckJiY2e18uYTDUnoGwp7+i2tIlLzcQHGzbJZC6r+Bgr+qvcFf2fH7zjAgReRVJCqjtGXGh0tJSSJKE8PBwl+4XAFBTY39/hcF8Y69biIy0LVDUrfP3l7tiaiEGESIiXD+TYigDFOGQJD+bn1dVVYUXXngB48aNc8yZ24oK289UFBTU3j3irnx8boxfYculkKio2udQq8J3nIhaNVFzBqJ8xfVB0QwAVBABv4MU/CdIyjirz62pqcHDDz8MIQRWrlxp5sXFjWG8be2vqKhwynE6hEple29FbGzt3SO8DEJNYBAholZLVB+rncodOtwYh0QLVH4Eod0JRH5o+gS9HigqAvLzUZOTg4dnz8avV6/i6z/8AaF/+UvjsFFQUHvpxF2FhNjXXxEUxGBBDsdmVSJqlYQwQFwdAuRfAgpraofqvv4lFeqBQgNQHAHFZ+ex+aabMLqqqjaECIEaAA8DOAvgGwAx8h5KLUmyrb+i/nIzo8cSOQKbVYmoddJoLF/+aHQZJBcKtfnbTMsBZAMASgAA569cQRaASABtAfwfamcQ+Qy1g8rnXn9eJADbu0ua4OPT9BDe9ddHRrK/gjwS/9USkXsSorYR09beivz82oG0bGTtAsNhAIPrPc64/j0dwCsAPrn+uFeD530DYJClFw0IsL2/IiYGCAvjZRBqFRhEiMg19PraW0ftuSNEp5Ol1EEArF2zFkBtULCnv6KVD+NNZAmDCBE1T8Np0pu6M6S4uPYshxsSCgARSiDq+lfk9e/RwUDSnyHFxpmGjejo2hE6iajFGESIyHSadFtvNW1imnRZ2ThNusH/AyBoBxAuAcrGl0Gk0LmQAsfLcABErQeDCJE3qhvG257+Ck+cJt1S0LBxmnRJ3AVR+hJQtQWAErWdIwYAAgiaBgT83rnHRUQMIkQeQaez3F/hDdOkNzWUt5P6KyTJD1L4GxC6qRCVnwKGIkjKeCDgd00OZkZEjsEgQiQHW6ZJr7/ME6ZJtzVUuOE06ZJPZ0ghz8pdBlGrxCBC1FJC1PZL2DR2Rb77T5NeN4y3LROOxcRwmnQiahEGEaKGDAbjMN42na3gNOlERM3GIELez9I06ZZChadMk94wTFgKF5wmnYjcGIMIeZ6G06RbCxn5+bWzn7orpdJ8gLAUKjhNOhF5Gf5GI3nVnybd1ssg7jxNur+/7U2bMTG1d4+wv4KIWjEGEXKsetOk2zyMt6dNk24taHCadCIiuzCIkHXV1ZYvfZhbXlTkvv0V9adJt2VW0+ho9lcQETkZg0hrY2madEthQ62Wu2LLfHzMn52w1l+hVMpdNRER1cMg4snqpkm3dW6Q/Hy7pkl3OUvTpJsLGbGxnCadiMgLMIi4E53Ocn+FubAh4zTpNgkNtX7pIzaW06QTEbVyDCLOZG6adGuXQtx4mnRIUu2lDVsaNuv6K1QquasmIiI359QgsmDBAmzatAmnT59GQEAA+vfvj0WLFqFLly7O3K1zNJwm3ZbLIZ4+TXr9dZGR7K8gIiKHc2oQ2bNnD6ZNm4Y777wTOp0OL774IoYOHYqffvoJQUFBztx10wyGG/0VtjZuuvM06UFBtoeKmBibp0knIiJyJkkI110LyM/PR2xsLPbs2YN77723ye3VajXCwsJQWlqK0NBQxxSxYQMwa1Zt2PCkadKbuhTC/goiInIT9nx+u7RHpPT6UNuRkZFm12u1WmjrnXVQO+PWUUkC8vIc/7rWNJwmvalLIlFRbjdNOhERkTO4LIgYDAbMmDEDd999N7p37252mwULFmDevHnOLSQ6uuWv4ednfTCshmGD06QTERGZ5bJLM08++SS2bduGvXv3ol27dma3MXdGJCEhwbGXZg4eBFJSTJcFBzc90mb9YBESwv4KIiIiC9zu0sxTTz2Fzz77DJmZmRZDCACoVCqonH3L5623Atu3m4aLgADn7pOIiIjMcmoQEULg6aefxubNm7F792506NDBmbuzTUgIMGyY3FUQERERnBxEpk2bhv/+97/YunUrQkJCkJubCwAICwtDAM9CEBERtXpO7RGRLPRRrF+/HhMnTmzy+U65fZeIiIicym16RFw4RAkRERF5IN5TSkRERLJhECEiIiLZMIgQERGRbBhEiIiISDYMIkRERCQbBhEiIiKSDYMIERERyYZBhIiIiGTDIEJERESyYRAhIiIi2TCIEBERkWwYRIiIiEg2DCJEREQkGwYRIiIikg2DCBEREcmGQYSIiIhkwyBCREREsmEQISIiItkwiBAREZFsGESIiIhINgwiREREJBsGESIiolZOCCHbvn1k2zMRERE5RWVNJfIr8pGvyTd+L6goMF1Wb/norqOxbtQ6WWplECEiInJjQgiUVZcZA0RBRYFJwKgfKuq+a2o0du0jT5PnpOqbxiBCRETkQgZhQFFlkflAUT9s1Htcra92ak35mnynvr41DCJEREQtUKOvaXzZo/4lkUrTwFFYWQiDMMhdton8CgYRIiIit1BRU2H+DIWFyyCl2lK5S7ZIKSkRHRiNmKAYxATG1P45MMb4uO57XHCcbDUyiBARkdcSQqBUW2pTf0XdWY2Kmgq5y7bI38f/RqCoCxMNgkX9sBHuHw6F5N43yDKIEBGRx9Ab9CisLLS5v6KgogA1hhq5y7YoxC+k0dkJc4Gi7nuQbxAkSZK7bIdiECEiItlU66tRUFGAa5prTTZs5mvyUVRZBAH5xrywRoKEyIDIJi+D1F+v8lHJXbbsGESIiMghhBDQ1GjMXgax1GOh1qrlLtsiH4VP47MTFoJFdGA0ogKioFQo5S7b4zCIEBGRWUIIlFSVNDkgVv3vVboqucu2KNA3sPFZCiv9FWGqMK+7DOKOGESIiFoJnUGHwopCq02bDc9i6Aw6ucu2KFQVitig2EbBwqSRs973QN9AuUsmMxhEiIg8lFantTh2hblgUVRZJHfJFkmQEBUY1aiHwtqlED+ln9xlkwMwiBARuQEhBMqryy3PD2ImaJRXl8tdtkW+Cl+zzZmWLoVEBkSyv6KVYhAhInICgzCguLLYpv6KurMWWr1W7rItCvQNNH92wsKtpqGqUPZXkE0YRIiIbKAz6FBQUWB1/Ir6fy6sKIRe6OUu26Jw/3Cbx66ICYxBgG+A3CWTl2IQIaJWqUpXZfOEY/mafBRXFctdskUKSdEoQFi7DBIdGA1fpa/cZRMBYBAhIi/QcJp0W+YHsXeadFfyU/oZ7waxZSjviIAItx/Gm8gSBhEicjt106Sb7a2wcEeIs6dJb4lgv+Amx6+o/z3YL5j9FdRqMIgQkdNV66uN41d46jTp9UX4R5ht2jQ3fkV0YDT7K4isYBAhIruZmybd2q2mnjRNekxQDKIDzA+IVXebKfsriByHQYSolWs4Tbott5p6wjTpTc0PUhc+PGGadCJvxiBC5GX0Bn1tf4WNZyvcfZr0UFWoxf6K1jJNOpE3YxAhcnN106TbMnaFp02T3tRQ3pwmncj7OTWIZGZmYvHixThy5AhycnKwefNmjB492pm7JHJ7mmqN5VtMPXiadFv7KziMNxHV59QgotFokJycjMmTJ+PBBx905q6IZFF/mvSmAkXdZZBKXaXcZVsU4BPQqIfC2q2mnCadiFrKqUEkLS0NaWlpztwFkUPpDXoUVhY2eRmkrt/C3adJD1OFWW3abHhJJMgvSO6SiaiVcaseEa1WC632xqRParX7npImz1Clq2p0psLcHSF127h7fwWnSScib+NWQWTBggWYN2+e3GWQm2oN06RbGhQrJigGEf4R7K8gIq/jVkFk9uzZyMjIMD5Wq9VISEiQsSJyJoMw1PZX2Dh+hbtPkx7kG2T+soeFu0E4TToRkZsFEZVKBZWKt+p5qrpp0q1NPFY/aBRUFHjMNOnWQgWnSSciaj63CiLkXqxNk25ufhBPmSa9qbtBogOjOU06EZGLODWIlJeXIzs72/j4/PnzyMrKQmRkJBITE525a2rA3DTpTV0Gcfdp0u0Zv4LTpBMRuSenBpHDhw9j8ODBxsd1/R/p6enYsGGDM3ft9epPk27LGBYFFQVeMU163dmMEL8Q9lcQEXkBpwaRQYMGQQj3vBXS3dToaxqfoTDXa+Hh06SbGyyL06QTEbVe7BFxEkvTpDcMFHWPS6pK5C7ZInPTpFsbwyIqMAo+Cv7TIiKipvHTwgZCCKi1apvPVuRX5HvMNOnW+is4TToRETlbqwwilqZJNxcoPGGa9BC/EJsvg3CadCIicietLoj869C/8NQXT7n1MN5106SbG7uiYaiIDoyGv4+/3GUTERE1S6sLIiF+IS4NIT4KH7N3g1gaxjsqIIrDeBMRUavR6oJIdGB0i55ff5r0hpdBGgaO6MBohPuH8zIIERGRBa0uiMQExZg8DlOFmR1p09IdIZwmnYiIyHFaXRDpFtMNWY9ncZp0IiIiN9DqgkiAbwCS45LlLoOIiIgAcHAIIiIikg2DCBEREcmGQYSIiIhkwyBCREREsmEQISIiItkwiBAREZFsGESIiIhINgwiREREJBsGESIiIpINgwgRERHJhkGEiIiIZMMgQkRERLJhECEiIiLZMIgQERGRbBhEiIiISDYMIkRERCQbBhEiIiKSDYMIERERyYZBhIiIiGTDIEJEROTmMjMzMXLkSMTHx0OSJGzZssVkfV5eHiZOnIj4+HgEBgZi+PDhOHv2rDzF2olBhIiIyM1pNBokJydjxYoVjdYJITB69Gj88ssv2Lp1K44dO4akpCSkpqZCo9HIUK19fOQugIiIiKxLS0tDWlqa2XVnz57F/v37ceLECXTr1g0AsHLlSsTFxeH999/HH//4R1eWajeeESEiIvJgWq0WAODv729cplAooFKpsHfvXrnKshmDCBERkQfr2rUrEhMTMXv2bBQXF6O6uhqLFi3C5cuXkZOTI3d5TeKlGSIiIjeiUVdg13vf4uiuH2HQGXBbv1swfMp9CI8JM7u9r68vNm3ahClTpiAyMhJKpRKpqalIS0uDEKLpHWq1QHU1EBLi4COxjSRsqlIearUaYWFhKC0tRWhoqNzlEBEROVxmZiYWL16MI0eOICcnB32DByO0IgoSJAghICkkaH00MCSX4cfTP0CtVqNTp074+uuvkZiYaPJapaWlqC4uRgyAlN/+Fne0b48VDz0E5OcDBQW13xt+qdXAhAnAu+867Jjs+fzmGREiIiIZ1d0R88j/jcWEiX9AdUU1IAABAQgB6ItxQJ+JHofaYtOfn0Hq4vl4uF07+P/1r4BGYxIqwgoKgMpKnAVwGMD8kyeBzz9vuoj8fGcfpkUMIkRERK6i0wGFhSZnKNLy85Hm64uzq/8LAJhgOI7ROIEwaBEGLf4AgdEAZomLwOL5AICYPXtwdc8eVAFIBPARgJjrfz4O4BkAowEMtbUuBhEiIiIPVFVl+ZKHucshxcW1ZznMuPn6945Qo+P1PxsAfA7gIQC9622bcf17OoANAHKuL8sD0BbAowBetuc4GESIiIhkJgRQXm5boKj7Ki93aknXAJQD+ADAUgCDAWwH8CKAbwAMvL7d9OtfJnx9gZgY06/oaPPLYmOdehzWMIgQEZF3MhiAkhLLZyvMhY3rY3K4C8P176MAPAEfVAaEYFb3zvj+11+xKjAQA8eMsRw2QkMBSZKzfJswiBARkWfQ6W6cmbDlckhBAaDXy121RWr4Qg0VSqFCCVSIv6s7sP9LbEMSTiMWpVChCL6QsBtH0BUPSN3w+v9eRN+03rj1hRdqBytbskTuw2gxBhEiIpJHZaX9/RXuSqEwf9nj+hkKER2NxTM/QPbVCpQY/KCGH/RSvTFFJSDyYgQAYD/iccEnAQZ97fmQUBEJNTQY8+xvcefwXgCAn3/+GUlJSa4+SqdgECEiopYTAigrs6+/wp0nZPPzM99HYS5oxMQA4eG1YcSCMwfP4qsr/wMQCjS4WqITOlSKcpRdLQEApD4xACUnKnDuwK9QiUAM6DwIO85uhaKLFufOncP27dvx6aefYvfu3c46epdiECEiosYMBqCoyLZAUbeuulruqi0LDrbcrGmuvyIkxKH9FQVXiiyuU6MIR5FpfLx01WIAQHp6OtavXw9JkrBu3TosWLAAzzzzDLp06YL//e9/GDBggMPqk5NLRlZdsWIFFi9ejNzcXCQnJ2P58uXo27dvk8/jyKpERA5SXW1ff0VhYW0YcVcREU2Hifpf9SaEk8PJ789gxoCXmtzutc9mI+U3t7ugIudyq5FVP/jgA2RkZGDVqlVISUnBsmXLMGzYMJw5cwaxMt4uRETk0Soqmr4DpP7j0lK5K7ZMqbR8W6m5oBEVBfh41gn9W++6GbFJMbh2MR+w8N//0OgQ3J7aw7WFuQGnnxFJSUnBnXfeiX/+858AAIPBgISEBDz99NOYNWuW1efyjAgRtQpC1M73YU9/RUWF3FVb5u/fdJiov76J/gpvsXfzAcwbY/kulz+v+xOGTRzswoqcx23OiFRXV+PIkSOYPXu2cZlCoUBqair27dvXaHutVgttvXu41Wq1M8sjInIOvd7+/oqaGrmrtiw01PbeipgYICjII8avcLUBv0vBSx9kYMX0tSjOu3GGKjQqBFMXT/CaEGIvpwaRgoIC6PV6tGnTxmR5mzZtcPr06UbbL1iwAPPmzXNmSURE9quutj1Q1PVXuOvE5pIEREba3l8RHQ2oVHJX7TUGPtQPA37XF0d3/oiCK0WIaBOOPkN7wtfPV+7SZONWF9lmz56NjIwM42O1Wo2EhAQZKyIiryNE7W2j9oxf4c5nZ3187O+vUCrlrrpVU/oocefw3k1v2Eo4NYhER0dDqVQiLy/PZHleXh7i4uIaba9SqaBi8iYiewhheRhvc2Hj+jTpbisgwLZAUb+/gpdByIM5NYj4+fmhT58+2LVrF0aPHg2gtll1165deOqpp5y5ayLyVPWnSbflUkhBQe1z3FVYmG2Bon5/BVEr4vRLMxkZGUhPT8cdd9yBvn37YtmyZdBoNJg0aZKzd01E7kCrbTpQNBzG2537K6Ki7Ouv8POTu2oit+b0IPLII48gPz8fc+bMQW5uLnr16oXt27c3amAlIg9Q119ha29Ffn7tsN/uytw06dbOXkREsL+CyMFcMrJqc3EcESInszRNurWzF242TbqJoCD7+is8ZJp0Ik/jNuOIEJGL2TpNev3+CjeeJh3h4baPXRETU9voSUQehUGEyJ1VVto+dkVdf4W7Uigs91eYO3MRFcX+CvJImZmZWLx4MY4cOYKcnBxs3rzZeMMGAEgWzsK98cYbmDlzpouqdB8MIkSu4u3TpNvSX9EKhvEm0mg0SE5OxuTJk/Hggw82Wp+Tk2PyeNu2bZgyZQrGjBnjqhLdCoMIUXOZmybdWqjwlGnSbe2vcPA06UTeIi0tDWlpaRbXNxxHa+vWrRg8eDA6duzo7NLcEoMIUZ2aGtMQUVAAXLtmfRhvd54mvW4Yb1vHr5B5mnSi1igvLw+ff/453nnnHblLkQ2DCHmvumnSbe2vKCmRu2LLLE2TbunshQdOk07UGr3zzjsICQkxewmnteBvKvIM3jxNui2XQjiMN5Fb0qgrsO3tXdi+7msU5ZYgOj4SaX8cguGTByMguOm7uNatW4fx48fDvxWfkWQQIXmYmya9qf4Kd54mPSTENFDExlo/e8Fp0okaaepuk/LycsyaNQtbtmxBYWEhOnTogOnTp+OJJ56Qpd7ivBI8e+8cXM3OhYAABFBWXI6Vz27AF2t24m+7rc8m/+233+LMmTP44IMPXFSxe2IQIceoP016U4EiP782hLhrf0XdNOnWzlDUDxXR0eyvIHKApu42ycjIwNdff4333nsP7du3x5dffok//elPiI+PxwMPPODyepdM/hdyfsmDybigAhAQuHj6Cv7+5FtWn7927Vr06dMHycnJTq7UvTGIUGOtZZp0S5dDIiPZX0Ekg6buNvn++++Rnp6OQYMGAQCmTp2K1atX4+DBgy4PIlfP5eLgtmNm1+mEDpW6cnzxvx0AgPPnzyMrKwuRkZFITEwEUDvy6EcffYS//e1vLqvZXfG3bWtg7zTp+flAVZXcVVvWcJr0pvorwsJ4GYTIC/Tv3x+ffPIJJk+ejPj4eOzevRs///wz3nzzTZfXcmr/WYvr1CjCUWQC10/6ZmRkAADS09OxYcMGAMDGjRshhMC4ceOcXarbYxDxRHq96TTpTZ218IZp0uuv5zTpRF6joqwSBVeKEBweiMi4CKvbLl++HFOnTkW7du3g4+MDhUKBNWvW4N5773VRtTcolJYH54uUYpGK/wMAzNvyPPo/cGejbaZOnYqpU6c6rT5PwiDiDuqmSbf1VtOiIvedJl2huDF+hS23mnKadKJWqeBKIda/tBFfv78Xuura/yh1u7sL0uc9gt739TD7nOXLl2P//v345JNPkJSUhMzMTEybNg3x8fFITU11Zfnoce+tUCgVMOgt97r5+CrR/e6uLqzKMzGIOJoQQHm5ff0VnjRNelNnLThNOhE1If9yIZ5KmY3S/FLodTc+yE/t+xkvDJ2Plz/IaPScyspKvPjii9i8eTNGjBgBAOjZsyeysrKwZMkSlweR6PhIDB57N77Z+J3ZMKJQSBg2aTBCo0JcWpcnYhBpiqVp0q2dwXDnadIDA+3rr+A06UTkYG/Peg8l10obfYAbDAKQgL/9cWWj59TU1KCmpgaKBvMVKZVKGGS6A2/6vx5D7oVrOPndGePZkbrvyYO64ck3J8pSl6dpfUFEp7Otv6Luq7DQs6ZJb2r0zcBAuSsmIg+i1+vxyiuv4L333kNubi7i4+MxceJEvPTSSxZnkbWmrLgcez7cZ/Ysgk7oUCnKUXZ9lOOGd5sMHDgQM2fOREBAAJKSkrBnzx68++67WLp0aUsPs1kCQwLwt2/m4futh7BjwzcouFKE2MRoDJ90H1J+ezuUPDtsE0kId202qL29KSwsDKWlpQgNDXXMiy5bBjz7rGNey9GsTZNuqb/C11fuqonIi/31r3/F0qVL8c4776Bbt244fPgwJk2ahNdffx3Tp0+3+/Wyj53Hk32eN7uuSFyrvdukgbq7TXJzczF79mx8+eWXKCoqQlJSEqZOnYpnn322WaGInMeez+/Wd0YkOtp1+7I0TbqlMxecJp2I3Mz333+PUaNGGfsy2rdvj/fffx8HDx5s1usFhVk+K1t3t4mkkDD1jQn4v4yRJuvj4uKwfv36Zu2X3FfrCyIxMc1/bsNp0pu6FBIczP4KIvJo/fv3x1tvvYWff/4Zt9xyC3744Qfs3bu32ZdD4jrEolNyEn45fhHCYP6EvBAC94y5qyVlkwdp3UEkIsK2Ibzr/sxhvImolaiuqkZ5iQYZMzKgVqvRtWtXKJVK6PV6vP766xg/fnyzXleSJDw67xHMHf2G+fUKCfc/OhBtklrwn0byKK0viHTvDuTk1PZisL+CiMjE5Z+v4r3XPsaeD76HrkaPfOUVXPA7hRXLVmLA4P7IysrCjBkzEB8fj/T09Gbto/8Dd+LP6/6E5dPehraqGj4+ShgMAga9Aal/uBfPrORAX61J62tWJSIis3758Vc8e+/L0FZojeN7fCs+R0dFV3QN74ll372OxK434bXXXsN7772H06dPt2h/FWWV2L3xO1zJzkVweBAGPtwP8Z3iHHEoJDM2qxIRkd3emPhPVGm0JrfWGqCHMAAadSXenLoKb2bOd9jYHYEhAfjNY64diIzcD4MIERHh7NFfcC7rQqPl0WiL8zgNf10gNN+WYfXyNVi6dCkmT57s+iLJKzGIEBERLpy8ZHZ5F/TCOZzEaRxDNapw/vUf8fjjj2POnDkurpC8FYMIERHBP8j8XYE+ki+6oBe6oBcA4K8bXsSdw3u7sDLydhw9i4iIcHtqD/j5W7+TMDA0ED0H3uaiiqi1YBAhIiIEhQbWjmRqZQzGsS+MhipA5bqiqFVgECEiIgDAo/MexsgnhgIAFEoFlL5KKJQKQAL+L2Mkxs4aLW+B5JU4jggREZm4fDYHu97LRFFOMaJvikLqo/eibYc2cpdFHsSez28GESIiInIoez6/eWmGiIiIZMMgQkRERLJhECEiIiLZMIgQERGRbBhEiIiISDYMIkREXmTlypXo2bMnQkNDERoain79+mHbtm1yl0VkEYMIEZEXadeuHRYuXIgjR47g8OHDuO+++zBq1CicPHlS7tKIzOI4IkREXi4yMhKLFy/GlClT5C6FWgl7Pr85+y4RkZfS6/X46KOPoNFo0K9fP7nLITKLQYSIyMNdPZeLyz/nIDDEH7fedQt+OvUT+vXrh6qqKgQHB2Pz5s247TbOmkvuiUGEiMhDXTx9BcunvY2sb04Yl0W0CcPYv/wOx44dg1qtxscff4z09HTs2bOHYYTcEntEiIg80NVzuZjWdxYq1JUw6A2N1k9+/fcYN/t3AIDU1FR06tQJq1evdnWZ1EpxrhkiIi+34eWNFkMIALwzdyOK80oAAAaDAVqt1oXVEdmOl2aIiDxMRVklMj/eZzaEZIvjiEIcAvRBWLfk3yjwuYrdu3djx44dMlRK1DSnBZHXX38dn3/+ObKysuDn54eSkhJn7YqIqFUpuVYKvc78mZBqaHESh6A1VCFrxbdI6d8XO3bswP333+/iKols47QgUl1djYceegj9+vXD2rVrnbUbIqJWJyQyGJIkwVyL323SHQAAhVKBiS+PNfaJELkrpwWRefPmAQA2bNjgrF0QEbVKIRHBSBlxOw5uO2axR0QYBO77/QAXV0ZkP7dqVtVqtVCr1SZfRETUWPq8R6D0VUKhkBqvlIBRTw1Hm6QY1xdGZCe3CiILFixAWFiY8SshIUHukoiI3FLn3h3wxldz0KZ9rMlyX5UPxj4/Gk8sTZepMiL72DWOyKxZs7Bo0SKr25w6dQpdu3Y1Pt6wYQNmzJhhU7OqVqs1ucVMrVYjISGB44gQEVkghMCPe37CpTNXERjij76/uR3B4UFyl0WtnNPmmnnuuecwceJEq9t07NjRnpc0oVKpoFKpmv18IqLWRpIkJA/qhuRB3eQuhahZ7AoiMTExiInhNUciIiJyDKf1iFy8eBFZWVm4ePEi9Ho9srKykJWVhfLycmftkojII2VmZmLkyJGIj4+HJEnYsmWLyfpNmzZh6NChiIqKgiRJyMrKkqVOImdwWhCZM2cOevfujblz56K8vBy9e/dG7969cfjwYWftkojII2k0GiQnJ2PFihUW1w8YMKDJHj0iT8RJ74iI3IgkSdi8eTNGjx7daN2FCxfQoUMHHDt2DL169XJ5bUS24qR3RERE5BE46R0RkQyqKrS4/PNVKH2USOx6E5Q+SrlLIpIFgwgRkQtVaqrwzssb8fnbu1BVXgUAiIqPwMN/HoXR09Nkro7I9RhEiIhcpLqqGrOGzsfpg9kmc8QUXi3GyowNuHz2qozVEcmDQYSIyEW2rf0ap/b/DEu3CHy68kvXFkTkBhhEiIhc5NOVO2Aug+iEDpUoh8JHAnTA+fPnkZWVhcjISCQmJqKoqAgXL17E1au1Z0zOnDkDAIiLi0NcXJwLj4DI8XjXDBFRE9q3bw9Jkhp9TZs2za7XyT1/DeaSiBpFOICd2Kf7CgCQkZGB3r17Y86cOQCATz75BL1798aIESMAAGPHjkXv3r2xatWqlh0YkRvgOCJERE3Iz8+HXq83Pj5x4gTuv/9+fPPNNxg0aJDNr/NI/GMoyi2xuF6hUKDfA33wyqbnW1Atkfw4jggRkQPFxMQYL4PExcXhs88+Q6dOnTBw4EC7XmfI+HugUFr+tWswGDB43D0tLZfIozCIEBHZobq6Gu+99x4mT54MSZLseu7o6b9BYEiA2TCi8FGgQ49E3D36TkeVSuQRGESIiOywZcsWlJSUYOLEiXY/NzYhGn/bPQ9tkmpnMVf6KIyhpHv/rnhj5xz4+PIeAmpd2CNCRFRPdVU1Plv9FT5ZuQM52bnwD/bH4EfuxpiM3yKhy00YNmwY/Pz88OmnnzZ7HwaDAUd3HseZg9lQ+ihwx7Be6Ny7gwOPgkhe9nx+M4gQEV2nrdRi1rDXcPK7MxAQxjtclD4K+Pj64OkNk/CbscOwadMmjBo1St5iidyYPZ/fPAdIRHTdf177H376/gwa/v9MrzNAGGrw7MSZiI2NNd5GS0Qtxx4RIiIANdU1+HTVlzAYzJ8k1usNOF/xMwanpMLHh/+HI3IUBhEiIgDXLhagvFhjcX0R8lCFCtwW3cOFVRF5P8Z6IiIAvn7Wfx1GSXEY5vMIbopLcFFFRK0Dz4gQEQGISYhGQtd4WBsaRK/TI2XE7a4riqgVYBAhIgIgSRLGzXrQ4sy4Ch8FuqbcjFvvusW1hRF5OQYRIqLrUifciwlzHgJQGzwAGAccS+x6E+Ztnmn3aKpEZB3HESEiauDXU5exbc1OXDpzFYFhgRj4UD/0G3kHlD5KuUsj8ggc0IyIiIhkw9l3iYiIyCMwiBAREZFsGESIiIhINgwiREREJBsGESIiIpINgwgReZWFCxdCkiTMmDFD7lKIyAYMIkTkNQ4dOoTVq1ejZ8+ecpdCRDZiECEir1BeXo7x48djzZo1iIiIkLscIrIRgwgReYVp06ZhxIgRSE1NlbsUIrKD9XmviYjc0NVzuTi1/ywkhYSe996Knd9+haNHj+LQoUNyl0ZEdmIQISKPUZxXgiWT/4WD244Zl2mlShz13YNdX++Ev7+/jNURUXNwrhki8ggVZZWYducLuHouDwa9wbj8mriCH7EPEiTjTLl6vR6SJEGhUECr1UKp5GR1RK7EuWaIyOtsX/c1rpzNNQkhABCJWNyF+5GCVKxZtAFZWVm44447MH78eGRlZTGEELk5BhEi8gjb134NgcYncH0kXwRLYQj1icDZry+ie/fuCAoKQlRUFLp37y5DpURkDwYRIvIIRbnFMJNDjAx6AwqvFLmuICJyCDarEpFHiG4XhdLCMothRKFUIDYxGgCwe/du1xVGRC3CMyJE5BHSpgyxut6gN2DYpMEuqoaIHIVBhIg8wtCJg9Che6Lxzpj6FAoJyYO6od/IO2SojIhagkGEiDxCQJA//rZ7Hu4Zc5dJGPHxVWL4lPvw2mezofThHTJEnobjiBCRxym4WoSfD52DpJDQrX8XhEaFyF0SEdVjz+c3m1WJyONEx0cielSk3GUQkQPw0gwRERHJxmlB5MKFC5gyZQo6dOiAgIAAdOrUCXPnzkV1dbWzdklEREQexmmXZk6fPg2DwYDVq1ejc+fOOHHiBB577DFoNBosWbLEWbslIiIiD+LSZtXFixdj5cqV+OWXX2zans2qREREnsdtm1VLS0sRGWm5wUyr1UKr1Rofq9VqV5RFREREMnFZs2p2djaWL1+Oxx9/3OI2CxYsQFhYmPErISHBVeURERGRDOwOIrNmzYIkSVa/Tp8+bfKcK1euYPjw4XjooYfw2GOPWXzt2bNno7S01Ph16dIl+4+IiIiIPIbdPSL5+fkoLCy0uk3Hjh3h5+cHALh69SoGDRqEu+66Cxs2bIBCYXv2YY8IERGR53Fqj0hMTAxiYmJs2vbKlSsYPHgw+vTpg/Xr19sVQoiIiMj7Oa1Z9cqVKxg0aBCSkpKwZMkS5OfnG9fFxcXZ9Bp1J2vYtEpEROQ56j63bbno4rQg8tVXXyE7OxvZ2dlo166dyTpbrwaVlZUBAJtWiYiIPFBZWRnCwsKsbuPWk94ZDAZcvXoVISEhkCTJoa+tVquRkJCAS5cueWX/CY/P83n7MXr78QHef4w8Ps/nrGMUQqCsrAzx8fFNtmW49aR3CoWi0dkURwsNDfXaf2AAj88bePsxevvxAd5/jDw+z+eMY2zqTEgddo8SERGRbBhEiIiISDatNoioVCrMnTsXKpVK7lKcgsfn+bz9GL39+ADvP0Yen+dzh2N062ZVIiIi8m6t9owIERERyY9BhIiIiGTDIEJERESyYRAhIiIi2XhtEHn99dfRv39/BAYGIjw83KbnCCEwZ84ctG3bFgEBAUhNTcXZs2dNtikqKsL48eMRGhqK8PBwTJkyBeXl5U44AuvsrePChQuQJMns10cffWTcztz6jRs3uuKQGmnO3/WgQYMa1f/EE0+YbHPx4kWMGDECgYGBiI2NxcyZM6HT6Zx5KGbZe3xFRUV4+umn0aVLFwQEBCAxMRHTp09HaWmpyXZyvocrVqxA+/bt4e/vj5SUFBw8eNDq9h999BG6du0Kf39/9OjRA1988YXJelt+Jl3JnuNbs2YN7rnnHkRERCAiIgKpqamNtp84cWKj92r48OHOPgyr7DnGDRs2NKrf39/fZBtPfg/N/T6RJAkjRowwbuNO72FmZiZGjhyJ+Ph4SJKELVu2NPmc3bt34/bbb4dKpULnzp2xYcOGRtvY+3NtN+Gl5syZI5YuXSoyMjJEWFiYTc9ZuHChCAsLE1u2bBE//PCDeOCBB0SHDh1EZWWlcZvhw4eL5ORksX//fvHtt9+Kzp07i3HjxjnpKCyztw6dTidycnJMvubNmyeCg4NFWVmZcTsAYv369Sbb1T9+V2rO3/XAgQPFY489ZlJ/aWmpcb1OpxPdu3cXqamp4tixY+KLL74Q0dHRYvbs2c4+nEbsPb7jx4+LBx98UHzyySciOztb7Nq1S9x8881izJgxJtvJ9R5u3LhR+Pn5iXXr1omTJ0+Kxx57TISHh4u8vDyz23/33XdCqVSKN954Q/z000/ipZdeEr6+vuL48ePGbWz5mXQVe4/v97//vVixYoU4duyYOHXqlJg4caIICwsTly9fNm6Tnp4uhg8fbvJeFRUVueqQGrH3GNevXy9CQ0NN6s/NzTXZxpPfw8LCQpNjO3HihFAqlWL9+vXGbdzpPfziiy/EX/7yF7Fp0yYBQGzevNnq9r/88osIDAwUGRkZ4qeffhLLly8XSqVSbN++3biNvX9nzeG1QaTO+vXrbQoiBoNBxMXFicWLFxuXlZSUCJVKJd5//30hhBA//fSTACAOHTpk3Gbbtm1CkiRx5coVh9duiaPq6NWrl5g8ebLJMlv+8bpCc49x4MCB4plnnrG4/osvvhAKhcLkl+XKlStFaGio0Gq1DqndFo56Dz/88EPh5+cnampqjMvkeg/79u0rpk2bZnys1+tFfHy8WLBggdntH374YTFixAiTZSkpKeLxxx8XQtj2M+lK9h5fQzqdToSEhIh33nnHuCw9PV2MGjXK0aU2m73H2NTvV297D998800REhIiysvLjcvc7T2sY8vvgeeff15069bNZNkjjzwihg0bZnzc0r8zW3jtpRl7nT9/Hrm5uUhNTTUuCwsLQ0pKCvbt2wcA2LdvH8LDw3HHHXcYt0lNTYVCocCBAwdcVqsj6jhy5AiysrIwZcqURuumTZuG6Oho9O3bF+vWrbN5tmRHaskx/uc//0F0dDS6d++O2bNno6KiwuR1e/TogTZt2hiXDRs2DGq1GidPnnT8gVjgqH9LpaWlCA0NhY+P6bRRrn4Pq6urceTIEZOfH4VCgdTUVOPPT0P79u0z2R6ofS/qtrflZ9JVmnN8DVVUVKCmpgaRkZEmy3fv3o3Y2Fh06dIFTz75JAoLCx1au62ae4zl5eVISkpCQkICRo0aZfJz5G3v4dq1azF27FgEBQWZLHeX99BeTf0MOuLvzBZuPemdK+Xm5gKAyQdU3eO6dbm5uYiNjTVZ7+Pjg8jISOM2ruCIOtauXYtbb70V/fv3N1n+6quv4r777kNgYCC+/PJL/OlPf0J5eTmmT5/usPpt0dxj/P3vf4+kpCTEx8fjxx9/xAsvvIAzZ85g06ZNxtc19x7XrXMVR7yHBQUFmD9/PqZOnWqyXI73sKCgAHq93uzf7enTp80+x9J7Uf/nrW6ZpW1cpTnH19ALL7yA+Ph4k1/qw4cPx4MPPogOHTrg3LlzePHFF5GWloZ9+/ZBqVQ69Bia0pxj7NKlC9atW4eePXuitLQUS5YsQf/+/XHy5Em0a9fOq97DgwcP4sSJE1i7dq3Jcnd6D+1l6WdQrVajsrISxcXFLf53bwuPCiKzZs3CokWLrG5z6tQpdO3a1UUVOZatx9dSlZWV+O9//4uXX3650br6y3r37g2NRoPFixc77EPM2cdY/0O5R48eaNu2LYYMGYJz586hU6dOzX5dW7nqPVSr1RgxYgRuu+02vPLKKybrnP0ekv0WLlyIjRs3Yvfu3SbNnGPHjjX+uUePHujZsyc6deqE3bt3Y8iQIXKUapd+/fqhX79+xsf9+/fHrbfeitWrV2P+/PkyVuZ4a9euRY8ePdC3b1+T5Z7+HroDjwoizz33HCZOnGh1m44dOzbrtePi4gAAeXl5aNu2rXF5Xl4eevXqZdzm2rVrJs/T6XQoKioyPr8lbD2+ltbx8ccfo6KiAo8++miT26akpGD+/PnQarUOmYvAVcdYJyUlBQCQnZ2NTp06IS4urlHHd15eHgB4zHtYVlaG4cOHIyQkBJs3b4avr6/V7R39HpoTHR0NpVJp/Lusk5eXZ/F44uLirG5vy8+kqzTn+OosWbIECxcuxM6dO9GzZ0+r23bs2BHR0dHIzs52+YdYS46xjq+vL3r37o3s7GwA3vMeajQabNy4Ea+++mqT+5HzPbSXpZ/B0NBQBAQEQKlUtvjfhE0c1m3ipuxtVl2yZIlxWWlpqdlm1cOHDxu32bFjh2zNqs2tY+DAgY3utLDktddeExEREc2utbkc9Xe9d+9eAUD88MMPQogbzar1O75Xr14tQkNDRVVVleMOoAnNPb7S0lJx1113iYEDBwqNRmPTvlz1Hvbt21c89dRTxsd6vV7cdNNNVptVf/vb35os69evX6NmVWs/k65k7/EJIcSiRYtEaGio2Ldvn037uHTpkpAkSWzdurXF9TZHc46xPp1OJ7p06SKeffZZIYR3vIdC1H6OqFQqUVBQ0OQ+5H4P68DGZtXu3bubLBs3blyjZtWW/JuwqVaHvZKb+fXXX8WxY8eMt6geO3ZMHDt2zORW1S5duohNmzYZHy9cuFCEh4eLrVu3ih9//FGMGjXK7O27vXv3FgcOHBB79+4VN998s2y371qr4/Lly6JLly7iwIEDJs87e/askCRJbNu2rdFrfvLJJ2LNmjXi+PHj4uzZs+Jf//qXCAwMFHPmzHH68Zhj7zFmZ2eLV199VRw+fFicP39ebN26VXTs2FHce++9xufU3b47dOhQkZWVJbZv3y5iYmJku33XnuMrLS0VKSkpokePHiI7O9vkdkGdTieEkPc93Lhxo1CpVGLDhg3ip59+ElOnThXh4eHGO5QmTJggZs2aZdz+u+++Ez4+PmLJkiXi1KlTYu7cuWZv323qZ9JV7D2+hQsXCj8/P/Hxxx+bvFd1v4PKysrEn//8Z7Fv3z5x/vx5sXPnTnH77beLm2++2aWhuCXHOG/ePLFjxw5x7tw5ceTIETF27Fjh7+8vTp48adzGk9/DOgMGDBCPPPJIo+Xu9h6WlZUZP+sAiKVLl4pjx46JX3/9VQghxKxZs8SECROM29fdvjtz5kxx6tQpsWLFCrO371r7O3MErw0i6enpAkCjr2+++ca4Da6Pt1DHYDCIl19+WbRp00aoVCoxZMgQcebMGZPXLSwsFOPGjRPBwcEiNDRUTJo0ySTcuEpTdZw/f77R8QohxOzZs0VCQoLQ6/WNXnPbtm2iV69eIjg4WAQFBYnk5GSxatUqs9u6gr3HePHiRXHvvfeKyMhIoVKpROfOncXMmTNNxhERQogLFy6ItLQ0ERAQIKKjo8Vzzz1ncvurq9h7fN98843Zf9MAxPnz54UQ8r+Hy5cvF4mJicLPz0/07dtX7N+/37hu4MCBIj093WT7Dz/8UNxyyy3Cz89PdOvWTXz++ecm6235mXQle44vKSnJ7Hs1d+5cIYQQFRUVYujQoSImJkb4+vqKpKQk8dhjjzn0F3xz2HOMM2bMMG7bpk0b8Zvf/EYcPXrU5PU8+T0UQojTp08LAOLLL79s9Fru9h5a+h1Rd0zp6eli4MCBjZ7Tq1cv4efnJzp27GjymVjH2t+ZI0hCyHBvJhERERG8eIh3IiIicn8MIkRERCQbBhEiIiKSDYMIERERyYZBhIiIiGTDIEJERESyYRAhIiIi2TCIEBERkWwYRIiIiEg2DCJEREQkGwYRIiIikg2DCBEREcnm/wG1OxIAZuONpgAAAABJRU5ErkJggg==\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "xleBuBrq4YlQ"
      },
      "execution_count": 72,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "![Screenshot 2024-08-09 at 12.25.35.png](data:image/png;base64,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)"
      ],
      "metadata": {
        "id": "dh5hZ-_a4aSY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model = AbstractModel()\n",
        "model.i = Set(initialize= I_set)\n",
        "model.j= Set(initialize= J_set)\n",
        "C = 0.5\n",
        "def rule_x(model,i,j):\n",
        "  return train_df.loc[i,j]\n",
        "model.X = Param(model.i,model.j,initialize= rule_x, mutable = True)\n",
        "model.alpha = Var(model.i, bounds=(0,C), within=Reals)\n",
        "\n",
        "def rule_con1(model):\n",
        "  return  sum(Y[i]*model.alpha[i] for i in model.i) == 0\n",
        "model.con1= Constraint(rule=rule_con1)\n",
        "\n",
        "def rule_of(model):\n",
        "  return sum(model.alpha[i] for i in model.i) -0.5*sum(Y[i]*Y[ii]*model.alpha[i]*model.alpha[ii]*sum(model.X[i,j]*model.X[ii,j] for j in model.j) for i in model.i for ii in model.i )\n",
        "\n",
        "model.obj = Objective(rule=rule_of, sense=maximize)\n",
        "instance = model.create_instance()\n",
        "st= time.time()\n",
        "result = opt.solve(instance)\n",
        "ft= time.time()\n",
        "print(\"time\", ft-st)\n",
        "print(\"obj\", value(instance.obj))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0xljDFb94a-H",
        "outputId": "9c8b8605-a85e-4c40-89d4-7278bfb07e93"
      },
      "execution_count": 73,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time 0.026998043060302734\n",
            "obj 0.37062761709734804\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"Total samples\", len(I_set))\n",
        "nonzero = 0\n",
        "for i in instance.i:\n",
        "  if value(instance.alpha[i])>0.01:\n",
        "    print(i, value(instance.alpha[i]))\n",
        "    nonzero+=1\n",
        "print(f\"Number of nonzero alpha = {nonzero}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DWAKZ3ll6TnA",
        "outputId": "f672579f-d6e5-4d49-ce98-4b8d955a610a"
      },
      "execution_count": 74,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Total samples 16\n",
            "19 0.37062765506746154\n",
            "12 0.37062762576774877\n",
            "Number of nonzero alpha = 2\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "all = [value(instance.alpha[i])*Y[i]*np.array(df.loc[i,J_set]) for i in instance.i if value(instance.alpha[i])>0.01]\n",
        "W = sum(all)\n",
        "all,W"
      ],
      "metadata": {
        "id": "Htx5gyz97OsX",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8641e7a4-7c0e-4b55-e239-b1bae78db517"
      },
      "execution_count": 78,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "([array([-0.33016644, -0.06800498]), array([0.05950581, 0.88531648])],\n",
              " array([-0.27066063,  0.81731151]))"
            ]
          },
          "metadata": {},
          "execution_count": 78
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "$$Y_i(WX_i+b) =1$$  "
      ],
      "metadata": {
        "id": "Ug2JcpuSa2yY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "i = 19\n",
        "X = np.array(df.loc[i,J_set])\n",
        "X,W\n",
        "Y[i]-np.dot(X,W)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MztlqX1yZghO",
        "outputId": "74f60e3a-cb84-4297-e08d-dfbafbef89e9"
      },
      "execution_count": 79,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "-0.908852447879709"
            ]
          },
          "metadata": {},
          "execution_count": 79
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "eX_YZM4O7-8y"
      },
      "execution_count": 76,
      "outputs": []
    }
  ],
  "metadata": {
    "colab": {
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}