Logistic_Regression.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyPQiRRcDoY051iRXfHFiM/3",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Armin-Abdollahi/Iris-Flower-Classification/blob/main/Logistic_Regression.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "kBEkwJhaWfK5"
      },
      "outputs": [],
      "source": [
        "# Importing Libraries\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "from mpl_toolkits import mplot3d\n",
        "from sklearn.multiclass import OneVsOneClassifier\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.datasets import load_iris\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "\n",
        "%matplotlib inline"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Load Data\n",
        "data = load_iris()"
      ],
      "metadata": {
        "id": "yeOktaNVjtWg"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Exploring Data\n",
        "labels = data['target_names']\n",
        "description = data['DESCR']\n",
        "feature_names = data['feature_names']\n",
        "\n",
        "print(f\"This is a dataset with {len(data.data)} samples.\\n{len(labels)} classes.\\n{len(feature_names)} features for each sample.\")\n",
        "print(f\"Feature names: {feature_names}\")\n",
        "print(f\"Label Names:{labels}\")\n",
        "print(f\"---------\\nSample input feature vector: {data['data'][0]}\\nLabel: {data['target'][0]}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Vm_nXSpij3dr",
        "outputId": "28483a95-10b7-4f8a-cf0d-a8d1e34695ba"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "This is a dataset with 150 samples.\n",
            "3 classes.\n",
            "4 features for each sample.\n",
            "Feature names: ['sepal length (cm)', 'sepal width (cm)', 'petal length (cm)', 'petal width (cm)']\n",
            "Label Names:['setosa' 'versicolor' 'virginica']\n",
            "---------\n",
            "Sample input feature vector: [5.1 3.5 1.4 0.2]\n",
            "Label: 0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Seeing if the data is linearly seperable or not\n",
        "x, y = data['data'][:,2],data['data'][:,3] # petal length and petal width \n",
        "c = data['target']\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(8,6))\n",
        "scatter = ax.scatter(x, y, c=c,label='x')\n",
        "\n",
        "# produce a legend with the unique colors from the scatter\n",
        "legend1 = ax.legend(*scatter.legend_elements(),\n",
        "                    loc=\"best\", title=\"Classes\",fontsize=16)\n",
        "ax.add_artist(legend1)\n",
        "plt.title(\"0-Setosa,1-Versioclor,2-Virginica\",fontsize=16)\n",
        "plt.xlabel('Petal Length(cm)',fontsize=18)\n",
        "plt.ylabel('Petal Width(cm)',fontsize=18)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 414
        },
        "id": "1PdP5tarj_lp",
        "outputId": "ffa30267-7915-4812-aa12-d1311bf2ef08"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# 3-D Plot\n",
        "c = []\n",
        "for i in data['target']:\n",
        "    if i == 0:\n",
        "        c.append('red')\n",
        "    elif i==1:\n",
        "        c.append('green')\n",
        "    else:\n",
        "        c.append('blue')"
      ],
      "metadata": {
        "id": "ANLiTUs4kGM3"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "data['feature_names']"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "S_e9k_70kONk",
        "outputId": "a1895ba1-d786-4811-e5e7-e44bf5e5d1eb"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['sepal length (cm)',\n",
              " 'sepal width (cm)',\n",
              " 'petal length (cm)',\n",
              " 'petal width (cm)']"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "fig = plt.figure(figsize=(5,5))\n",
        "ax = mplot3d.Axes3D(fig)\n",
        "\n",
        "# Data for a three-dimensional line\n",
        "sepal_width,petal_length,petal_width  = data['data'][:,0],data['data'][:,1],data['data'][:,2]\n",
        "plt.title('3D Plot')\n",
        "ax.set_xlabel('Sepal Width(cm)')\n",
        "ax.set_ylabel('Petal Length(cm)')\n",
        "ax.set_zlabel('Petal Width(cm)')\n",
        "ax.scatter(sepal_width,petal_length,petal_width,c=c)\n",
        "\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 391
        },
        "id": "bOvYFfK6kOwy",
        "outputId": "5468223f-b4d8-4f58-f51d-639084649832"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 360x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Binary classifier\n",
        "# for this simplistic example I am just using one feature: sepal length\n",
        "x = data['data'][:,:2]\n",
        "y = (data['target']==0).astype('int')"
      ],
      "metadata": {
        "id": "mdc0EtPSkR5m"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# rough plot to see what we're dealing with \n",
        "plt.figure(figsize=(7,7))\n",
        "for i in range(len(x)):\n",
        "    if y[i] == 1:\n",
        "        c = 'red'\n",
        "        marker = '^'\n",
        "    else:\n",
        "        c = 'green'\n",
        "        marker = '.'\n",
        "    plt.plot(x[i,0],x[i,1],c=c,marker=marker)\n",
        "plt.xlabel('Sepal Length(cm)',fontsize=14)\n",
        "plt.ylabel('Sepal Width(cm)',fontsize=14)\n",
        "plt.title(\"Red Square - Iris Setosa | Green Dot - Not iris Setosa\",fontsize=12)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 462
        },
        "id": "Z3gwvWQPkcna",
        "outputId": "a02aa32f-10a2-4a96-f887-e3b723b2ec81"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 504x504 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAb4AAAG9CAYAAABqGRNHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3de5gcZZn38e/PwSERVASiIkTiAXUTRZCIxBMDURFF3Ne4ghgRVxYET7i6eGYRFMTFw7KwyggvZ09LXF9EUTAYATOiiQQwwQNiIMgg4aiBwMDkfv94aqDTmempmanu6u76fa6rr5k6zFP3U9XTdz/V1XUrIjAzM6uKx5UdgJmZWSs58ZmZWaU48ZmZWaU48ZmZWaU48ZmZWaU48ZmZWaU48XUAScdKOr/sONqNpFdJ+n3ZcbSCpLMlHVJ2HNaYpEskvavB8ndIurSVMdmmnPiaRNJqSeslrZN0e/bCtWWTtvVJSX/OtnWrpO80YzutJmmJpEPHWh4RV0bE8yfR7g6SFkm6U9J9kn6bN6mMF1NZJPVKOkbS7yXdL+kv2Yvw68qODTb6f/i7pHslLZX0Xkm5XoMkHSLpqinGcKykkPS2mnmbZfNmFRFDROwbEec0WH5BREz4mEiaI+lSSXdn+2+5pDfk/NvVkl4z0W12Mye+5npTRGwJ7ALsCnyi6A1k7y7fCbwm29ZcYHHR28kRx2YdtL3zgDXAjsA2pP331yLiKtGFwJuBg4GnAM8C/hN442grt/p4Zd4UEU8k7fcvAB8DzmxxDHcDn5XU0+LtTnWf/wC4DHg68FTgg8DfioirkiLCjyY8gNWkZDQy/UXghzXTewBLgXuBa4G+mmXPAn4O/J30ZD8VOH+M7ZwKfLVBHGO2BfQBt44VN7A7MJDFOJj9bW/NugG8D/gj8Ods3n7AiuxvlgI7T2EfLgEOrY2V9GJ5Oyl5bRR/tuwvWV9/D8wfo911wC4NtjvqsQE+DwwDD2ZtnJrNfznwa+C+7OfLa9o6BLgpi+nPwDuy+c8BLgfuAu4ELgC2ahDT2cAhYyx7DbAe2CHHc/JjwHXAQ8Bm4zwPn0xKTIPZfv0c0FPTr6uAk4F7sr7tm/f/oeb5tQF4Yc32zgXWAjcDnya9Of+HbJ8PZ/v93kk+n47N9vO1wLuyeZtlz+NZRcTAxs/ZQ4BfAF/JjvPnRvZbtlzZsjtISez6kX1R1+a2WYyNnh+j/t+R/k82ZM+PdcDR2fz9gZXZ+kuAfxjv/4hxXg866VF6AN36YOMEskP2pP7PbHr77B/hDdk/1Wuz6RnZ8gHgy8DmwKuzJ+BYiW8h6V3sv5FGez11y8dsi/ET326kF8bNgFnADcBRNesGKZluDUwnjWrvAF4G9ADvytrbfJL7sPZFpA94BDgp68v02viB55NGcc/IpmcBzxmj3Z+SXpAOBJ5Zt2y8Y/NoTNn01qQX/ndm++nt2fQ2wBakF7TnZ+tuB8zJfn9u1vbmwAzgChq/gTmbsRPfF4AlOZ+TK4CZ2f4br6//C5ye9eOpwK+Aw7NlhwAPA/+SHesjgNsAjff/UDf/FuCI7Pdzgf8HPDE7fn8A3lOzvaum+D95LHA+6UX/JuDxbJr4phQDmya+R4APZNuZzsaJbx9gObAVKQn+A7DdKG2K9ObyYuAfgafVLW/4f1e/74HnAfdnx/vxwNHAjUAvDf6PGOf1oJMePtXZXN+X9HfSE+kO4N+z+QuBH0XEjyJiQ0RcBiwD3iDpmcBLgc9ExEMRcQXpNMeoIuJ80j/WPqSR3R2SPgYw0bZGaXt5RPwyIh6JiNWkF8E961Y7MSLujoj1wGHA6RFxdUQMR/qs4yHSP0sRNgD/nvVlfd2yYVISmS3p8RGxOiL+NEY7/wRcCXwG+LOkFZJemi0b89iM0dYbgT9GxHnZfvoW8DvgTTUxv1DS9IgYjIiVABFxY0RclvVlLenNSf2+zWtb0igYAElbZ58D3Sfpwbp1T4mINdn+a/Q8fFrW56Mi4v6IuIM0Ojmwpq2bI+IbETEMnENK7E+bYOy3AVtnpx4PBD4REX/Pnm9fIr2hKFREXEQa0W30WW2TYrgtIv4re27UP2cfJiXYF5DeMNwQEYOjxBvAXqQE9iVgUNIVknbKVpno/90BpLNPl0XEw6RR+3TSmYsx/49yvh50BCe+5vrHSJ9p9JGe3Ntm83cE/il7cbpX0r3AK0kvHM8A7omI+2vaubnRRiJ9YP4a0jvH9wLHS9pnMm3VkvQ8SRdnF+f8DTihpg8j1tT8viPwkbp+zcziqG/7k9nFOOskfT1nSGsjov6FHEiJBDiK9K7+DknflrTJdrN174mIj0fEHNIL9QrSmxTR+NiM5hlsuk9vBrbP9vsBpGMyKOmHkl6Q9f9pWYx/yfbt+Wy6b/O6qza+7I3IVqR36JvXrVt/vMbq646k0cBgzbLTSSO/EY8m24h4IPt1ohdwbU86Y7Fttr3afXlztnxc2RW+I8+nlTn+5NPAp4BpNfOmFMMY1oy1ICIuJ50uPI30nO2X9KQx1r01It4fEc8hHZv7SaNTmMD/XWaj52xEbMji3L7R/1HO14OO4MTXAhHxc9KpqpOzWWuA8yJiq5rHFhHxBdK586dI2qKmiWfm3M7DEfE/pM9wXpijrfuBJ4xMZO94Z9Qs/xpp9LJTRDwJ+CTptMtGm635fQ3w+bp+PSEbBdXHekJEbJk93punf3Xb2nRhxDcj4pWkF4IgnRZt3GDEnaTj8gzSactGx2a0GG7LtlfrmaTPSIiIn0TEa0nJ5HfAN7J1TsjaelG2bxey6b7NazHwUkk75Fi3/niN1dc1pFHDtjXLnpS9WShENsrenvRZ4Z2kEVDtvnx0PzL+sb+y5vk0bozZ6PZG4Mia2VOKYaxNjRPHKRGxGzCbdAry38ZtMGINKVm+MJs13v9dw+ds9oZvJo89Z8f6P8rzetARnPha56vAayW9mPTu/k2S9pHUI2mapD5JO0TEzaTTTZ/NLlF/JY+dNttEdon1GyU9UdLjJO0LzAGuztHWH4Bp2d8/nvQuuHaE8ETSZ1TrspHKEeP08RvAeyW9TMkWI7FNYD9NiqTnS9pb0uakixDWk04zjrbuSZJemF3K/kRSv26MiLtocGyyP/8r8Oya5n4EPE/SQVl7B5BexC7ORnVvzt54PES6uGAkpidm0/dJ2p4cL3hjiYhLgZ+RRq0vy4714xn/FHOj5+EgcCnwJUlPyp5bz5E05VNbWXv7Ad8mfd58fXa69LvA57Pn8o7Av2YxQtrvO0jqner2a3yK9PkWAK2OQdJLs+P1eNKb0AcZ5Tkr6SmSPivpudlx2Bb4Z+CX2Srj/d/VP2e/C7xR0vxs2x8hPT+XjvN/NNHXg/YVbfBBYzc+GP0qtq8Bi7LfX0b6TO5u0ucNPyS70IL0JL2S9MI43lWdbyFdqHEPj10ZdkjN8oZtkT5sHyR9BvlRNr645dWkd3jrsjaOo+bDfdK7wefWxfN60pWNI1d+/Q/wxEnuwyXUXdVZt/zRecDOpIsv/p7t04vJPqAfpd3/Il0ssC7b9xez8VVtjY7NPNIbhntIn5dBOj24nHRV53Lgldn87bJ27uOxq+dmZ8vmZOuuI51q/Uh9/+piPpsxLm7JlveSTk/9EXiAdAXsJcDrxnlONurrk0nP2VuzPlwDHFjzvLmqrq1Nng91216fHZ/7SBddvY+ai7FIX8M4P4tjDXAM8Lia/v0wi/POST6fjqXu/4j0xqX24pYpxcCmF7fU76NH5wHzSWdn1vHYlb1bjtLmFqTPUFdn694OfIt0anLc/zvS11xuyZZ9NJv3f4BV2bH4OY9ddDXm/xHjvB500kNZh6wiJB1LenFaWHYslp+ks0lXbp5dcihmHc+nOs3MrFLKuHuDmU3c90mnusxsinyq08zMKsWnOs3MrFJafqoz+67YMuAvEbFf3bJDgP/gse/NnBoRZzRqb9ttt41Zs2Y1IVIzM+tUy5cvvzMiZoy2rIzP+D5EusfbqHcoAL4TEe/P29isWbNYtmxZIYGZmVl3kDTmXapaeqoz+xLwG4GGozgzM7NmafVnfF8l3Slh1DtqZBZIuk7ShZJmjraCpMMkLZO0bO3atU0J1MzMulPLEl92i6I7ImJ5g9V+QLqDws6ku4yMWsk4IvojYm5EzJ0xY9RTuGZmZqNq5YjvFcD+klaT7tG3t6Tza1eIiLsi4qFs8gzS3eXNzMwK07LEFxGfiIgdImIWqebV5fW3zZJUW/plf9JFMGZmZoUp/c4tko4DlkUqDvlBSfuTqhbfTbqhq5mZWWE6/s4tc+fODX+dwczMaklaHhFzR1vmO7eYmVmlOPGZmVmlOPGZmVmlOPGZmVmlOPGZmVmlOPGZmVmlOPGZ5TU4CHvuCbffXnYkZjYFTnxmeR1/PFx1VfppZh3Lic8sj8FBOOss2LAh/fSoz6xjOfGZ5XH88SnpAQwPe9Rn1sGc+MzGMzLaGxpK00NDHvWZdTAnPrPx1I72RnjUZ9axnPjMxjMw8Nhob8TQECxdWk48ZjYlpZclMmt711xTdgRmViCP+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+MzMrFKc+Kx7DQ7CnnvC7beXHYmZtREnPutexx8PV12VfpqZZZz4rDsNDsJZZ8GGDemnR31mlnHis+50/PEp6QEMD3vUZ2aPcuKz7jMy2hsaStNDQx71mdmjnPis+9SO9kZ41GdmGSc+6z4DA4+N9kYMDcHSpeXEY2ZtZbOyAzAr3DXXlB2BmbUxj/jMzKxSnPjMzKxSnPjMzKxSnPjMzKxSnPjMzKxSnPjMzKxSnPjMzKxSnPisNVwiyMzahBOftYZLBJlZm3Dis+ZziSAzayNOfNZ8LhFkZm3Eic+ayyWCzKzNOPFZc7lEkJm1GSc+ay6XCDKzNuOyRNZcLhFkZm3GIz4zM6sUJz4zM6sUJz4zM6sUJz4zM6sUJz4zM6sUJz4zM6uUlic+ST2SrpF08SjLNpf0HUk3Srpa0qxWx2c2Za5EYdbWyhjxfQi4YYxl7wHuiYjnAl8BTmpZVGZFcSUKs7bW0sQnaQfgjcAZY6zyZuCc7PcLgfmS1IrYzArhShRmba/VI76vAkcDG8ZYvj2wBiAiHgHuA7apX0nSYZKWSVq2du3aZsVqNnGuRGHW9lqW+CTtB9wREcun2lZE9EfE3IiYO2PGjAKiMyuAK1GYdYRWjvheAewvaTXwbWBvSefXrfMXYCaApM2AJwN3tTBGs8lzJQqzjtCyxBcRn4iIHSJiFnAgcHlELKxb7SLgXdnvb83WiVbFaDYlrkRh1hFKr84g6ThgWURcBJwJnCfpRuBuUoI06wyuRGHWEUpJfBGxBFiS/X5MzfwHgX8qIyYzM6sG37nFzMwqxYnPzMwqxYnPzMwqxYnPzMwqxYnPzMwqxYnPzMwqxYnPOs+KFbDVVnDddWVHYmYdyInPOs/ChXDffXDQQWVHYmYdyInPOsuKFbByZfp95UqP+sxswpz4rLMsrLu9q0d9ZjZBTnzWOWpHeyM86jOzCXLis85RP9ob4VGfmU2AE591jj/9aWLzzcxGUXpZIrPc1q8vOwIz6wIe8ZmZWaU48ZmZWaU48ZmZWaU48ZmZWaU48ZmZWaU48ZmZWaU48dnUXHYZbLYZXH552ZE03+Ag7Lkn3H572ZGY2RQ48dnUHHAADA/DW99adiTNd/zxcNVV6aeZMbBmgBOvPJGBNQNt2d5Y/AV2m7zLLoN77km/33NPGvXtvXe5MTXL4CCcdRZs2JB+fuYz8PSnlx2VWWkG1gww/9z5DA0P0dvTy+KDFzNv5ry2aa8Rj/hs8g44YOPpbh71HX98SnqQRrge9VnFLVm9hKHhIYZjmKHhIZasXtJW7TXixGeTUzvaGzEy6us2I6O9oaE0PTSUpv1Zn1VY36w+ent66VEPvT299M3qa6v2GlFENK3xVpg7d24sW7as7DCqZ+utN018AE95Ctx9d+vjaaYjj4Qzz3ws8QH09sKhh8Jpp5UXl1nJBtYMsGT1Evpm9RVyWrLI9iQtj4i5oy5z4rNJedzjYLTnjvTYKcFuseuuqRZgvV12gWuuaX08ZjauRonPF7fY5HRbcmvEyc2sq/gzPjMzqxQnPjMzqxQnPjMzqxQnPjMzqxQnPjMzqxQnPjMzqxQnPpuaoisW5GnPVRLMbAqc+Gxqiq5YkKc9V0kwsynwnVts8gYH4dnPhgcfhOnT4aabplaxIE97RW/TzLpSozu3eMRnk1d0xYI87blKgplNkUd8Njm1I68RUxmB5Wmv6G2aWdfyiM+KVzvyGjGVEVie9orepplVkhOfTc7AwMZleiBNL13avPaK3qaZVZKrM9jkFF2xIE97rpJgZgXwiM/MzCrFic/MzCrFic/MzCrFic/MzCrFic/MzCrFic/MzCrFic/MzCrFia9MZZTXybtNl/4xsy7lxFemMsrr5N2mS/+YNdXAmgFOvPJEBtYMlB1K5fgm1WUpo7xO3m269I9ZUw2sGWD+ufMZGh6it6eXxQcvZt7MeWWH1VV8k+p2VEZ5nbzbdOkfs6ZasnoJQ8NDDMcwQ8NDLFm9pOyQKsWJrwyDg3DWWY/dcHloKE038/O0vNssIzaziumb1UdvTy896qG3p5e+WX1lh1QpuROfpGdJ2kvSGyS9VNK0ZgbW1coor5N3my79Y9Z082bOY/HBizl+r+N9mrMEDaszSJoFHAG8HdgeUM3iIUlXAv3AoojYsEkDNroyyuvk3aZL/5i1xLyZ85zwSjLmxS2STgHeBVwKXAT8CrgNWA9sDbwQeBUpKT4CvDsift2CmDfSsRe3mJlZ0zS6uKXRiO9B4DkRcecoy+4ALs8en5X0BmBHoOWJz8zMbCLGTHwRcXTeRiLiR8WEY2Zm1ly+qtPMzCql4cUtIyQ9BTgW2At4KnUJMyKeWnhkZmZmTZAr8QHnAnOAc4C/Ap19uxczM6usvImvD9gzIn7TxFjMzMyaLu9nfH+awLqjkjRN0q8kXStppaTPjrLOIZLWSlqRPQ6dyjZtClasgK22guuuK2a9dq5EYWaVkjeZfQg4UdKLJfVMclsPAXtHxIuBXYDXS9pjlPW+ExG7ZI8zJrktm6qFC+G+++Cgg4pZr50rUZhZpeRNfDcC04HfkO7YMlz7yNNAJOuyycdnD39W2I5WrICVK9PvK1eOPZrLu97I/T83bGjdfT/L2KaZdYS8ie9bwJOBDwIHAG+re+QiqUfSCtIX4C+LiKtHWW2BpOskXShp5hjtHCZpmaRla9euzbt5y2vhwo2nxxrN5V2vnStRmFnl5KrHJ+kBYPeI+G0hG5W2Av4X+EBtm5K2AdZFxEOSDgcOiIi9G7XlW5YVbMUK2HXXTedfey3svPPE16ut7Tei2TX+ytimmbWVIurxrQKeVFRAEXEv8DPg9XXz74qIh7LJM4Dditqm5VQ/ihtRP5rLu147V6Iws0rKm/g+DXxZ0mskPU3S1rWPPA1ImpGN9JA0HXgt8Lu6dbarmdwfuCFnfFaUP/0p3/y867VzJQozq6S83+MbuRfnpWx8QYqy6TxXem4HnJNdFfo44LsRcbGk44BlEXER8EFJ+5OqPdwNHJIzPivK+vXFrnfNNZOPZbLK2KaZdYy8iW+vqW4oIq4DNvlQKCKOqfn9E8AnprotMzOzseRKfBHx82YHYmZm1gq5PuOT9H5Jm1zNIGmhpCOLD8vMzKw58l7cchSwZpT5q4EPFxaNmZlZk+VNfDsAN48y/9ZsmZmZWUfIm/huJ91fs95LgDuLC8fMzKy58l7V+U3gFEn3A0uyeXsBXwUuaEJcZmZmTZF3xPfvwC+AnwAPZI9LgKXAZ5oTWgUUWTYnb3mgotvL24d27quZVUquxBcRD0fE24HnAQdljxdExIER8XAzA+xqRZbNyVseqOj28vahnfva5gbWDHDilScysGagZW0VuU2zdpPrJtXtrGNvUl17I+Wp3kC5/obR9TeKblZ7efvQzn1tcwNrBph/7nyGhofo7ell8cGLmTdzXlPbKnKbZmWZ1E2qJX1a0hY5N/AKSW+abICVVGTZnLzlgYpuL28f2rmvbW7J6iUMDQ8xHMMMDQ+xZPWSprdV5DbN2lGjU53PAW6R1C/pTbU3kJY0TdJLJH1Q0q+A84B7mh1s1xgpkjpyI+WhockXS60tBjuiUVHYotrL24d27msH6JvVR29PLz3qobenl75ZfU1vq8htmrWliBjzAbwIOJ2U1IZJN49en/2+AVgGHAZs3qidZj5222236DhHHBHR2xsBjz16eyOOPHLibc2Zs3E7I485cyYXW9728vahnfvaIZbesjROuOKEWHrL0pa1VeQ2zcpAKn4wat7IW4j2ccDOwI7AdNJ391ZEROnf4evIz/h23TWNXurtssvEKwtMn75xwdUR06blr6Awmfby9qGd+2pmXavRZ3y+uMXMzLpOo8SX9wvsIw09A3gqdZ8NRsRvJh+emZlZ6+RKfJJ2Bc4HXkAqPlsrbyFaMzOz0uUd8fWTqjP8C3AbG1dhNzMz6xh5E99sYNeI+EMzgzEzM2u2vPfqvB6Y5K02zMzM2seYIz5JW9dMfhL4oqRPk5LgRvfnjIi7mxOemZlZsRqN+O4E1maPy4DdgUtJn/GNzB9Zx7pNGVUXzMxaoFHi2wvYu+ax1xjz9m5yjFaGMqouWNvoX97PPuftQ//y/rJDGVW7x2ftLe+dW54JrIm6lSUJmBkRtzQpvnH5C+xNUEbVBWsb/cv7Ofziwx+dPn2/0zlst8NKjGhj7R6ftYdJVWeo82dgxijzt86WWTcpo+qCtY1FqxY1nC5bu8dn7S9v4hOjf3dvS2CUmydaxyqj6oK1lQWzFzScLlu7x2ftr+H3+CSdkv0awImSHqhZ3EO64GWUOxBbx6odxY0YGc2ddtrE17OOM3LacNGqRSyYvaDtTiO2e3zW/hp+xifpZ9mvewIDwFDN4iFgNXByRPyxWQGOx5/xFayMqgtmZgWb9E2qI2KvrIGzgA9FxN+aEJ+1k7xJy8nNzDpUrluWRcS7mx2ImZlZKzS6c8vPyHkz6ojwd/nMzKwjNBrx/bbm9x7gHcDtwNXZvN2B7UjliszMzDrCmIkvIj4w8rukrwDnkD7ni5r5X2XT+nxmZmZtK+/3+A4GTq2/cwvw38A7iw3JzMyseSbyBfYXjTJ/tHlmZmZtK28h2v8LnCFpJ+CX2bw9gKOBs5oRmJmZWTPkHfEdDZwIfAC4PHt8APhCtqwaii7Bk6e9FStgq63guuuK2WZeLjdkZl0qV+KLiA0R8cWI2B7YCtgqIrbP5g03N8Q2UnQJnjztLVwI990HBx1UzDbzcrmhphtYM8CJV57IwJqBskOZtLzlgcrqa57t5o2t6D50w/HvVLnKErWzlt2yrOgSPHnaW7Ei3RpsxLXXws47T36bRcZmUzKwZoD5585naHiI3p5eFh+8mHkz55Ud1oTkLQ9UVl/zbDdvbEX3oRuOf7ubVFkiSddJekr2+/XZ9KiPZgXeVoouwZOnvYULN55u1ajP5YaabsnqJQwNDzEcwwwND7Fk9ZKyQ5qwvOWByuprnu3mja3oPnTD8e9kjU51LgIeyn6/MJse69Hdii7Bk6e9FStg5cqN/27lyuZ/1udyQy3RN6uP3p5eetRDb08vfbP6yg5pwvKWByqrr3m2mze2ovvQDce/o0VERz922223aLojjojo7Y2Axx69vRFHHtm89ubM2Xj5yGPOnGL6NJXYrBBLb1kaJ1xxQiy9ZWnZoUza6ctOj9ed+7o4fdnpDdcrq695tps3tqL70A3Hv50By2KMvDFeWaKfA4tJV3FeHREPtygf59aSz/iKLsGTp73p09NnbPWmTYP16ye+zSJjMzNrc5MuSwSsAQ4DjgUekLQU+BkpEf46IjY0+NvuUfQLfp72mpncGnFyM7Mu1/DrDBGxMCJ2AJ4PfAS4C/ggqSjtPZIulvTh5odpZmZWjLzf4/tjRJweEW+PiO2A2cApwKuAk5sZoJmZWZHy3rIMSTOAPmCv7OdzgeXAkibEZWZm1hQNE5+kt5CS3N5snOiOAn4REfc3OT4zM7NCjTfiuxC4hXSfznMjoqQrLszMzIox3md8xwA3Al8BfifpXEn/LOnZzQ/NzMyseONd1fm5iHgN6cbUBwN/yn6ulHRzlgjf3YI4u1ORFRDytlVWtQczszaR96rOoYj4eUR8NiL6gO2Ac4E3A2c0Mb7uVmQFhLxtlVXtwbpSWRUGPvbTj7HTKTvxsZ9+rK3ass6QqzqDpM1IhWf3yh57ANOAPwM/i4hDmxlkIy2rzlC0Iisg5G2rrGoP1pXKqjDwsZ9+jC/+4ouPTh/9iqM56TUnld6WtZdJVWfI/vCTki4F7gWuAA4lXexyBDArIp5TZtLraEVWQMjbVlnVHqwrlVVh4Hurvtdwuqy2rHOMd6rz/cBa0tcXdoqIHSPikIg4JyJuaX54XarICgh52yqr2oN1rbIqDLxl9lsaTpfVlnWOhl9niIhntCqQSqkdoY0YGamddlpz2qof7Y046CD47W8ntk0zYN7MeSw+eDFLVi+hb1ZfywqpjpyK/N6q7/GW2W+Z0qnJItuyzuEK7GUosgJC3rbKqvZgZlaCqVRnsGYosgJC3rac3MzMgJxfZzAzM+sWTnxmZlYpTnxmZlYpY37GJ+mivI1ExP7FhGNmZtZcjS5uuatlUZiZmbXImIkvInzzaTMz6zot+4xP0jRJv5J0raSVkj47yjqbS/qOpBslXS1pVkuCy1vZoMhqCkXLW3WhG/pqZjYFuROfpL0k9Uv6saTLax85m3gI2DsiXgzsArxe0h5167wHuCcinkuqAdia2yjkrWxQZDWFouWtutAFfS26IkD/8n72OW8f+pf3d/R287ZXhrwVEIruaxnVI/Jus6zKFnl0Qx8aiohxH8AhwIPAt7Kf3wGuId28+tQ8ba8gw/UAABjBSURBVNS19wTgN8DL6ub/BJiX/b4ZcCfZ3WXGeuy2224xJbfdFjFtWgRETJ8eMTg4tfXKcM01Ka6Rx7XXjr5eF/R16S1LY/rnpkfPZ3ti+uemx9Jblk6pvdOXnR4cy6OP05ed3pHbzdteGY6+7OiNYjv6sqNHXa/ovhZ9zPLIu80yYsurG/oQEQEsizHyRt4R30eB90fE24GHgU9ExK7A+cC6vElWUo+kFcAdwGURcXXdKtsDa7KE/AhwH7DNKO0cJmmZpGVr167Nu/nR5a1sUGQ1haLlrbrQBX0tuiLAolWLGk53ynbztleGvBUQiu5rGdUj8m6zrMoWeXRDH8aTN/E9G/hp9vtDwJbZ76eSRoO5RMRwROwC7ADsLumFef+2rp3+iJgbEXNnzJgxmSaSvJUNiqymULS8VRe6oa8UXxFgwewFDac7Zbt52ytD3goIRfe1jOoRebdZVmWLPLqhD+MaaygYG5+CXAO8KPv9WuCg7PdXAPflaWOUNo8BPlo3r7WnOo84IqK3d+PThL29EUceObn1yjBnzsZxjTzmzNl4vW7oa2bpLUvjhCtOKOzUyunLTo/Xnfu6cU8Ptvt287ZXhqMvOzqe+5/PHfM054ii+1r0Mcsj7zbLiC2vbugDDU515q3A/k1geUR8SdKngA8DPwDmA7+KiLfmaGMG8HBE3CtpOnApcFJEXFyzzvuyBPteSQcCb4mItzVqd0rVGfJWNiiymkLR8lZd6Ia+mpnl1Kg6Q97EtzUwLSJuk/Q44N9Io70/AJ+LiHtztLEzcA7QQzrF+t2IOE7ScaTMfJGkacB5wK7A3cCBEXFTo3Y7siyRmZk11ZQTXztz4jMzs3qF1OPLRmMHAbOzWauAb0WEC72ZmVnHyHVVp6SXAH8CvgTsnj1OBm7KlpmZmXWEvF9n6Ad+AewQEa+OiFcDM4ErsmVmZmYdIe+pzjnAwRFx/8iMiLh/5MKUpkRmZmbWBHlHfL8DnjHK/O1IV3aamZl1hLwjvk8Dp2QjvF9m8/bI5n88+7oDABFxd7EhmpmZFSdv4vtB9vObwMj3H5T9/H8100H6nl53GhyEAw+E73wHnv70sqMxM7NJyJv49mpqFJ2itlTPaaeVHY0VZGDNAEtWL6FvVh/zZs6b8nrtHF/RfS16n5S1j9uZ90nx/AX2vAYH4dnPTrcHmz4dbrrJo74uMLBmgPnnzmdoeIjenl4WH7x41BeXvOu1c3xF97XofVLWPm5n3ieT1+gL7BMpRPsiSadKukTSdtm8f5S0a1GBtrU2LtVjk9fuJViKjK/ovha9Tzq5zE2zeJ80R94vsL8O+DWpXt7ewPRs0XOAf29OaG2kzUv12OS1ewmWIuMruq9F75OOLnPTJN4nzZH3JtVXA+dExH9L+jvw4oi4SdJuwA8iYrSvOrRES051HnkknHnmY4kPoLcXDj3Un/V1AX/G1/z18vLnWZvyPpmcIqoz3A/MiYjVdYnvWcANETGt2JDza0nic6keM7OOUsRNqu8mneZcXTf/JcCtkw+tQzi5mZl1jbwXt3wT+A9JO5C+q7eZpD1JN6o+t1nBmZmZFS1v4vs08GfgZmBLUkmiy4GrgM83JzQzM7Pi5TrVGREPA++Q9BnS6c3HAddExB+bGZyZmVnRcheiBYiIm0g1+DYDSrugxczMbLIanuqUNF/S2+rmfRxYB9wr6ceStmpmgGZmZkUa7zO+jwM7jExI2h04ATgPOBp4MfCppkVnZmZWsPES34uAn9dM/xOwNCL+JSK+DHwQ2L9ZwZmZmRVtvMS3FXBHzfQrgB/XTI/cxsysYw2sGeDEK09kYM1Aw/X6l/ezz3n70L+8v5D2ipYnvryxdcs+KWu7ZSijr526f8e7uGWQdD/ONZI2B3YFPlOz/InAQ02Kzazp8t79vn95P4dffDgAl950KQCH7XbYpNsrWp74iq660O77pEqVDcroayfv3/FGfJcAX5S0N3AScD9wZc3ynYEbmxSbWdPlvfv9olWLGk5PtL2i5Ymv6KoL7b5PqlTZoIy+dvL+HS/xHQM8CPwU+GfgXyKi5k7N/DNwWZNiM2u6vHe/XzB7QcPpibZXtDzxFV11od33SZUqG5TR107ev3lvUv1kYF1EDNfN3zqbPzT6XzZfywrRWtfKe/f7/uX9LFq1iAWzF4x6Sm+i7RUtT3xFV11o931SpcoGZfS1nffvlKsztDMnPjMzq1dIBXYzM7Nu4MRnZmaV4sRnZmaV4sRnZmaV4sRnZmaV4sRnZmaV4sRnZmaV4sRnZmaV4sRnZmaV4sRnLdHO5UuKjm3h9xayzUnbsPB7Cwtpr53LzbTzcTUby3hlicymrJ3LlxQd28LvLeSC6y8AePTn+W85v23iK3Kb7XxczRrxiM+arp3LlxQd2yV/vKTh9ES1c7mZdj6uZo048VnTtXP5kqJj23enfRtOT1Q7l5tp5+Nq1oirM1hLtHP5kqJjW/i9hVzyx0vYd6d9p3Sas1nxFbnNdj6uVm0uS2RmZpXiskRmZmYZJz4zM6sUJz4zM6sUJz4zM6sUJz4zM6sUJz4zM6sUJz4zM6sUJz4zM6sUJz7rWkVXDmj3igWulGCt1qnPOVdnsK5UdOWAdq9Y4EoJ1mqd/JzziM+6UtGVA9q9YoErJVirdfJzzonPulLRlQPavWKBKyVYq3Xyc843qbauVXTlgHavWOBKCdZq7fycc3UGMzOrFFdnMDMzyzjxmZlZpTjxmZlZpTjxmZlZpTjxmZlZpTjxmZlZpTjxmZlZpbQs8UmaKelnklZJWinpQ6Os0yfpPkkrsscxrYrPzMyqoZUjvkeAj0TEbGAP4H2SZo+y3pURsUv2OK6F8VVCWRULylB0NYV27mvR+pf3s895+9C/vL/sUEZVpWNhxWtZdYaIGAQGs9//LukGYHtgVatiqLqyKhaUoehqCu3c16L1L+/n8IsPB+DSmy4F4LDdDiszpI1U6VhYc5TyGZ+kWcCuwNWjLJ4n6VpJl0iaM8bfHyZpmaRla9eubWKk3aWsigVlKLqaQjv3tWiLVi1qOF22Kh0La46WJz5JWwKLgKMi4m91i38D7BgRLwb+C/j+aG1ERH9EzI2IuTNmzGhuwF2krIoFZSi6mkI797VoC2YvaDhdtiodC2uOlt6kWtLjgYuBn0TEl3OsvxqYGxF3jrWOb1I9MWVVLChD0dUU2rmvRetf3s+iVYtYMHtBW53mHFGlY2GT0xbVGSQJOAe4OyKOGmOdpwN/jYiQtDtwIWkEOGaQTnxmZlavUeJr2cUtwCuAdwLXS1qRzfsk8EyAiPg68FbgCEmPAOuBAxslPTMzs4lq5VWdVwEaZ51TgVNbE5GZmVWR79xiZmaV4sRnZmaV4sRnZmaV4sRnZmaV4sRnZmaV4sRnZmaV4sRnZmaV4sRXMVUq59LupXXMrBytvHOLlaxK5VzavbSOmZXHI74KqVI5l3YvrWNm5XHiq5AqlXNp99I6ZlYen+qskHkz57H44MWVKOcyclqznUvrmFk5WlqPrxlclsjMzOo1KkvkU51mZlYpTnxmZlYpTnxmZlYpTnxmZlYpTnxmZlYpTnxmZlYpTnxmZlYpTnzNMDgIe+4Jt99ediRmZlbHia8Zjj8erroq/exQ3VDFoRv6UBbvO+tmvnNL0QYH4dnPhgcfhOnT4aab4OlPLzuqCemGKg7d0IeyeN9ZN/CdW1rp+ONhw4b0+/BwR476uqGKQzf0oSzed9btnPiKNDgIZ50FQ0NpemgoTXfYZ33dUMWhG/pQFu8763Y+1VmkI4+EM898LPEB9PbCoYfCaaeVF9ckDKwZ6PgqDt3Qh7J431mna3Sq04mvSLvuCitWbDp/l13gmmtaH4+ZWUU1Snyux1ckJzczs7bnz/jMzKxSnPjMzKxSnPjMzKxSnPjMzKxSnPjMzKxSnPjMzKxSnPjMzKxSnPjMzKxSnPjM2lz/8n72OW8f+pf3T7ktlxsy851bzNpa//J+Dr/4cAAuvelSAA7b7bBJteVyQ2aJR3xmbWzRqkUNpyfC5YbMEic+sza2YPaChtMT4XJDZolPdZq1sZHTmotWLWLB7AWTPs0JMG/mPBYfvNjlhqzyXJbIzMy6TqOyRD7VaWZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmleLEZ2ZmldKyxCdppqSfSVolaaWkD42yjiSdIulGSddJekmr4jMzs2rYrIXbegT4SET8RtITgeWSLouIVTXr7AvslD1eBnwt+2lmZlaIlo34ImIwIn6T/f534AZg+7rV3gycG8kvga0kbdeqGO0xA2sGOPHKExlYM1B2KGZmhWrliO9RkmYBuwJX1y3aHlhTM31rNm+wJYEZkJLe/HPnMzQ8RG9PL4sPXsy8mfPKDsvMrBAtv7hF0pbAIuCoiPjbJNs4TNIyScvWrl1bbIDGktVLGBoeYjiGGRoeYsnqJWWHZGZWmJYmPkmPJyW9CyLie6Os8hdgZs30Dtm8jUREf0TMjYi5M2bMaE6wFdY3q4/enl561ENvTy99s/rKDsnMrDAtO9UpScCZwA0R8eUxVrsIeL+kb5MuarkvInyas8XmzZzH4oMXs2T1Evpm9fk0p5l1lVZ+xvcK4J3A9ZJWZPM+CTwTICK+DvwIeANwI/AA8O4Wxmc15s2c54RnZl2pZYkvIq4CNM46AbyvNRGZmVkV+c4tZmZWKU58ZmZWKU58ZmZWKU58ZmZWKU58ZmZWKU58ZmZWKU58ZmZWKU58ZmZWKU58ZmZWKU58ZmZWKU58ZmZWKU58ZmZWKUr3he5cktYCN5cdxyi2Be4sO4gpch/aRzf0w31oH93Qj/H6sGNEjFqwteMTX7uStCwi5pYdx1S4D+2jG/rhPrSPbujHVPrgU51mZlYpTnxmZlYpTnzN0192AAVwH9pHN/TDfWgf3dCPSffBn/GZmVmleMRnZmaV4sRnZmaV4sQ3RZJ6JF0j6eJRlh0iaa2kFdnj0DJiHI+k1ZKuz2JcNspySTpF0o2SrpP0kjLibCRHH/ok3VdzLI4pI85GJG0l6UJJv5N0g6R5dcvb/jhArn609bGQ9Pya2FZI+puko+rWaetjkbMPbX0cRkj6sKSVkn4r6VuSptUt31zSd7JjcbWkWeO1uVmzgq2QDwE3AE8aY/l3IuL9LYxnsvaKiLG+DLovsFP2eBnwtexnu2nUB4ArI2K/lkUzcf8J/Dgi3iqpF3hC3fJOOQ7j9QPa+FhExO+BXSC9sQX+Avxv3WptfSxy9gHa+DgASNoe+CAwOyLWS/oucCBwds1q7wHuiYjnSjoQOAk4oFG7HvFNgaQdgDcCZ5QdS5O9GTg3kl8CW0naruyguomkJwOvBs4EiIihiLi3brW2Pw45+9FJ5gN/ioj6u0O1/bGoMVYfOsVmwHRJm5HeRN1Wt/zNwDnZ7xcC8yWpUYNOfFPzVeBoYEODdRZkp0IulDSzRXFNVACXSlou6bBRlm8PrKmZvjWb107G6wPAPEnXSrpE0pxWBpfDs4C1wFnZqfMzJG1Rt04nHIc8/YD2Pha1DgS+Ncr8TjgWI8bqA7T5cYiIvwAnA7cAg8B9EXFp3WqPHouIeAS4D9imUbtOfJMkaT/gjohY3mC1HwCzImJn4DIee1fSbl4ZES8hnb55n6RXlx3QJIzXh9+Q7t33YuC/gO+3OsBxbAa8BPhaROwK3A98vNyQJiVPP9r9WACQnabdH/ifsmOZrHH60PbHQdJTSCO6ZwHPALaQtHCq7TrxTd4rgP0lrQa+Dewt6fzaFSLiroh4KJs8A9ittSHmk72rIiLuIH0OsHvdKn8BakerO2Tz2sZ4fYiIv0XEuuz3HwGPl7RtywMd263ArRFxdTZ9ISmB1Gr740COfnTAsRixL/CbiPjrKMs64VhAgz50yHF4DfDniFgbEQ8D3wNeXrfOo8ciOx36ZOCuRo068U1SRHwiInaIiFmkUwmXR8RG70TqzvnvT7oIpq1I2kLSE0d+B14H/LZutYuAg7Mr2fYgnW4YbHGoY8rTB0lPHznvL2l30nO/4T9HK0XE7cAaSc/PZs0HVtWt1tbHAfL1o92PRY23M/YpwrY/Fpkx+9Ahx+EWYA9JT8hinc+mr6MXAe/Kfn8r6bW44Z1ZfFVnwSQdByyLiIuAD0raH3gEuBs4pMzYxvA04H+z5/9mwDcj4seS3gsQEV8HfgS8AbgReAB4d0mxjiVPH94KHCHpEWA9cOB4/xwl+ABwQXZ66ibg3R12HEaM14+2PxbZG6jXAofXzOuoY5GjD21/HCLiakkXkk7LPgJcA/TXvc6eCZwn6UbS6+yB47XrW5aZmVml+FSnmZlVihOfmZlVihOfmZlVihOfmZlVihOfmZlVihOfWQtIOlujVPDoFErVLz6aY709Jf0huzFys2J5kaS/jHErNLNxOfFZ15I0Q9J/Zy/aD0n6q6TFkl5bdmz1JM2SFJLmlhzHIZLWTaGJ/wA+HxHDRcVULyKuB34J/GuztmHdzYnPutki0q3L3gM8D9gPuIRxbmBrkyPp5cALgO+2YHNnkb587Ztw2IQ58VlXkrQV8Crg4xGxOCJujohfR8TJEfHtmvV6JZ0k6VZJD0j6taR9apb3ZSOx/ZSKdT6YVYDYrWadbZQKZN4qab1S0cxC7+SR3RrraEl/yrZxfe3NemtGjAskXZb1ZVX96FbSGyX9PuvHFZIOzP5ulqQ+UkLZIpsXko6t+fNpkk5XKmp6q6R/qwvzIGBxRKyv2+YblAqErpd0l6QfKCsmmo3Gj8lOBf9d0hpJBygVs/22pHWS/ijpdXXbuhTYGuibwm61inLis261Lnvsr7qKzXXOAvYkvWi/kFRB4weSXly33snAx4C5pNtwXSxppMDqNNItlfYD5pAKsZ4uaX5BfQH4HGnk+j5gNnBito031q33eeAU4MXAr4FvS9oSQNIzSTf5/WG2/BTgizV/uxQ4inQLru2yx8k1yz8MXE+66fRJwBe1cXX1VwHLaoOR9HrSvRQvI92kfS/g52z82nMU8Kus3e+SjsE3SbcF2wW4Aji/9jhGxBCwgnTszCYmIvzwoysfwALSvfseBAZIL+Ivq1n+HFItxWfW/d33gf/Ofu8j1fp7R83yLYF7gUMbbPvbwBk102cDFzdYf1a2nbmjLNuCdC/FV9XN/yrwo7q/P7xm+fbZvFdm0ycCN9S18clsnVnZ9CHAulFiWA18q27eH4FP10zfC7y7bp1fAN9u0O+N2s32bQCnjLdvSEn8vLKfZ3503sMjPutaEbGIVMPrTaTP9l4O/FLSJ7NVXgIIWJWdUluXXdjxRlJSrDVQ0+460shnNoCkHkmfUio4fFfWxluAZxbUldmkUeWP6+I8YpQ4r6v5faRS9VOzny8gjQJrXU1+19VN31bTNsB00puMWrsCi/O2m+3bB0j7d8RISZ3abUF6MzB9nLbNNuEPhq2rRcSDpNNslwHHSToDOFbSyaTTbQG8FHi47k/Xk99HgY8AHyK9YK8DTmDTF+rJGnmD+iZSmZZa9XE/Oh0RoVSxoqg3uPXbirq27wSeUlC7D9dNw6b92Jo0YjSbECc+q5pVpOf9NFKJEwFPj4ifjfN3e5A+2xsp9/JC4Nxs2SuBH0TEedlyka4ivbfAmB8iVcu+fArt/I5UzbpWfdHhIWCy38G7hmwUXDdvPvCNSbbZyAtJpzvNJsSJz7qSpG2A/wH+L+lU2t9JF6YcTbry8G/A3yRdAJwt6SOkC1RGrhS8KSJqX1Q/LWkt6fTeMaQE8c1s2R+AAyS9kjTq+QDwLNKL/kQ9T6k+Wq3fkT6fPDlLqleQPgvbA9gQEf052/468K/ZaPcbpAtxRmq1jYyqVpOu3nxtFv8DEfFAzvZ/QroAp9bnSRcL3UjaXyIVCj59Au1uQtIs0meYl062Dasuf8Zn3Wod6UvOHyJdRbiSdPrxm8ABNeu9m3Rl5xdJCeZi4NXAzXXtfRz4Eik57gTsFxH3Z8s+R7oq8RJSUrofuGCScV9ASji1j+cCnwGOJZ1WXUk6dbsA+HPehiPi5uxv9geuJV2l+dls8YPZOktJCfJbwFrSG4W8zicl7jk12/wR8H+AfbO+/Jx0ZeeGCbQ7mrcDl2Z9MpsQF6I1ayD7btvPgBkRcWfJ4RRO0oeA44CtooAXA0lfIO2r+pFfYSRtTrqi9O0R8Ytmbce6l0d8ZhUi6X2Sdpf0LElvJ40kzy4i6WVOAG5SE+/VCexIui2ak55Nikd8Zg1024hP0leAt5Fu23Yr6fuGx0X6QrhZJTjxmZlZpfhUp5mZVYoTn5mZVYoTn5mZVYoTn5mZVYoTn5mZVcr/B7N+h+bnKM0HAAAAAElFTkSuQmCC\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "lr_binary = LogisticRegression() \n",
        "lr_binary.fit(x,y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OMaIey9kkl79",
        "outputId": "3d655f2b-27ea-4501-bbc7-fafcfb561da8"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LogisticRegression()"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Exploring attributes of the LogisticRegression() object\n",
        "lr_binary.__dict__"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "X2Cucnm5kpd2",
        "outputId": "bbb3fd18-c08f-4c39-892c-9cd666083b14"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'penalty': 'l2',\n",
              " 'dual': False,\n",
              " 'tol': 0.0001,\n",
              " 'C': 1.0,\n",
              " 'fit_intercept': True,\n",
              " 'intercept_scaling': 1,\n",
              " 'class_weight': None,\n",
              " 'random_state': None,\n",
              " 'solver': 'lbfgs',\n",
              " 'max_iter': 100,\n",
              " 'multi_class': 'auto',\n",
              " 'verbose': 0,\n",
              " 'warm_start': False,\n",
              " 'n_jobs': None,\n",
              " 'l1_ratio': None,\n",
              " 'n_features_in_': 2,\n",
              " 'classes_': array([0, 1]),\n",
              " 'n_iter_': array([23], dtype=int32),\n",
              " 'coef_': array([[-3.38829757,  3.1645277 ]]),\n",
              " 'intercept_': array([8.32330389])}"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# let us check the score now\n",
        "lr_binary.score(x,y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Eg9VQdAfkwkE",
        "outputId": "e371ce97-f16b-460b-9da9-dc4131a73671"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1.0"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Using a single feature + labels are iris setosa and not iris setosa\n",
        "sepal_l = data['data'][:,0] #sepal length"
      ],
      "metadata": {
        "id": "8e48EQNjkzD3"
      },
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "single_feature_lr_binary = LogisticRegression()"
      ],
      "metadata": {
        "id": "mOSIvAJCk508"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "single_feature_lr_binary.fit(sepal_l.reshape(-1,1),y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WXzZkLerk_N-",
        "outputId": "da4c15f6-fbbf-49b6-b915-bbada293d5c3"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LogisticRegression()"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y_probab = single_feature_lr_binary.predict_proba(np.linspace(4.5,8.0,1000).reshape(-1,1))"
      ],
      "metadata": {
        "id": "0W8O-030k_08"
      },
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# h(x) = theta_0 + theta_1*X\n",
        "theta_0,theta_1 = single_feature_lr_binary.intercept_,single_feature_lr_binary.coef_\n",
        "print(theta_0.shape)\n",
        "print(theta_1.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KyJJjSm5lC0E",
        "outputId": "b2ec50d7-d0e2-41f0-e60d-80e5abe22aaf"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(1,)\n",
            "(1, 1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "single_feature_lr_binary.decision_function(sepal_l[0].reshape(-1,1))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9G7ulMKxlFZT",
        "outputId": "2666f237-d5dd-4a02-d3c4-081483b3efe8"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0.91116385])"
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def h(theta_0,theta_1,x):\n",
        "    return theta_0 + theta_1*x"
      ],
      "metadata": {
        "id": "Dq2jv2B7lH7h"
      },
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "single_feature_lr_binary.decision_function([[4.5],[8]])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XSoer_aPlKQD",
        "outputId": "afe24972-ba86-43c1-9cf4-909e8c581bc4"
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ 2.98579885, -9.11623862])"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "single_feature_lr_binary.score(sepal_l.reshape(-1,1),y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pcNXiwRYlMVF",
        "outputId": "5ad7f6e4-baf5-4d31-939c-3af782ef9add"
      },
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.8933333333333333"
            ]
          },
          "metadata": {},
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# 0.893 is not really a good score, so let us try to tweak our model to make this score better.."
      ],
      "metadata": {
        "id": "JEAh47fylOj_"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# new regularization constant \n",
        "lr_single_feature_high_regularization = LogisticRegression(C=0.5) # C is inverse of Regulariation Strength\n",
        "lr_single_feature_low_regularization = LogisticRegression(C=10) \n",
        "crazy_regularized = LogisticRegression(C=0.01)"
      ],
      "metadata": {
        "id": "cU_haIsDlTdm"
      },
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "lr_single_feature_high_regularization.fit(sepal_l.reshape(-1,1),y)\n",
        "lr_single_feature_low_regularization.fit(sepal_l.reshape(-1,1),y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Xuj86QgOlVrQ",
        "outputId": "28a8445e-22e0-4bf1-9098-658b927b74c1"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LogisticRegression(C=10)"
            ]
          },
          "metadata": {},
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "crazy_regularized.fit(sepal_l.reshape(-1,1),y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "k4dpDNDslX_f",
        "outputId": "e0784a45-829a-48d7-e626-26fee04bb58e"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LogisticRegression(C=0.01)"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "no_regularization = LogisticRegression(C=10**5)"
      ],
      "metadata": {
        "id": "8XWbyHnhlaa4"
      },
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "y_prob_high = lr_single_feature_high_regularization.predict_proba(np.linspace(4.5,8.0,1000).reshape(-1,1))\n",
        "y_prob_low = lr_single_feature_low_regularization.predict_proba(np.linspace(4.5,8.0,1000).reshape(-1,1))"
      ],
      "metadata": {
        "id": "ZPpQJDjolcym"
      },
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Comparing three models with different regularization constants"
      ],
      "metadata": {
        "id": "Xu4Qd6T4le12"
      },
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(10,10))\n",
        "\n",
        "plt.subplot(221)\n",
        "plt.title(\"C=1.0\")\n",
        "plt.plot(sepal_l,y,'ro')\n",
        "plt.plot(np.linspace(4.5,8.0,1000),y_probab[:,0],label=\"P(Not Iris Setosa) | X \")\n",
        "plt.plot(np.linspace(4.5,8.0,1000),y_probab[:,1],label=\"P(Iris Setosa) | X \")\n",
        "plt.legend(loc='best')\n",
        "plt.xlabel('Sepal Length(in cm)')\n",
        "\n",
        "# C = 10 , LOW REGULARIZATION \n",
        "plt.subplot(222)\n",
        "plt.title(\"C = 10\")\n",
        "plt.plot(sepal_l,y,'ro')\n",
        "plt.plot(np.linspace(4.5,8.0,1000),y_prob_low[:,0],label=\"P(Not Iris Setosa) | X \")\n",
        "plt.plot(np.linspace(4.5,8.0,1000),y_prob_low[:,1],label=\"P(Iris Setosa) | X \",color='black')\n",
        "plt.legend(loc='best')\n",
        "plt.xlabel('Sepal Length(in cm)')\n",
        "\n",
        "\n",
        "# C = 0.5 , HIGH REULARIZATION \n",
        "plt.subplot(223)\n",
        "plt.title(\"C = 0.5\")\n",
        "plt.plot(sepal_l,y,'ro')\n",
        "plt.plot(np.linspace(4.5,8.0,1000),y_prob_high[:,0],label=\"P(Not Iris Setosa) | X \")\n",
        "plt.plot(np.linspace(4.5,8.0,1000),y_prob_high[:,1],label=\"P(Iris Setosa) | X \",color='black')\n",
        "plt.legend(loc='best')\n",
        "plt.xlabel('Sepal Length(in cm)')\n",
        "plt.show()\n",
        "\n",
        "plt.tight_layout()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 638
        },
        "id": "gHs5SeRbliGY",
        "outputId": "d55bd850-0570-4adb-9dad-0cb6c7cba6b6"
      },
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x720 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# scores \n",
        "print(lr_single_feature_high_regularization.score(sepal_l.reshape(-1,1),y))\n",
        "print(lr_single_feature_low_regularization.score(sepal_l.reshape(-1,1),y))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "az2LFhHeloPc",
        "outputId": "a27d1f3c-8431-45fb-9884-1fc541456bcc"
      },
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.8933333333333333\n",
            "0.8933333333333333\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "crazy_regularized.score(sepal_l.reshape(-1,1),y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pN0dXO9olr1S",
        "outputId": "dd744f02-28e5-4b58-a661-b9de601a16f6"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6666666666666666"
            ]
          },
          "metadata": {},
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "no_regularization.fit(sepal_l.reshape(-1,1),y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZZMw0j9sluMS",
        "outputId": "e4b59f6f-99f7-4692-c299-972b752576ed"
      },
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LogisticRegression(C=100000)"
            ]
          },
          "metadata": {},
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "no_regularization.score(sepal_l.reshape(-1,1),y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GHzb2dyclwPM",
        "outputId": "cc881f3b-3e50-494a-bcab-c3f0550e0f3b"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.8933333333333333"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Multi-Class Classification"
      ],
      "metadata": {
        "id": "xYQBsXfAlzs0"
      },
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X = data['data'] # using all 4 features \n",
        "y = data['target'] # labels"
      ],
      "metadata": {
        "id": "o0o3aHIXl-lJ"
      },
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "lr_multi_class = LogisticRegression(multi_class='ovr') # we set the `multi_class` attribute to 'ovr'\n",
        "lr_multi_class.fit(X,y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bmE39rPjmCfB",
        "outputId": "4fad1ea0-6469-4fdd-e087-7e488c09b548"
      },
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LogisticRegression(multi_class='ovr')"
            ]
          },
          "metadata": {},
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y_hat = lr_multi_class.predict(X) # make predictions"
      ],
      "metadata": {
        "id": "RiHWo4GHmGK0"
      },
      "execution_count": 37,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "lr_multi_class.score(X,y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zcU-S4v7mIlc",
        "outputId": "fd47e1f8-f3ec-42a5-e7af-1c40aa53233a"
      },
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9533333333333334"
            ]
          },
          "metadata": {},
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Visually Inspecting where the model mis-classified instances"
      ],
      "metadata": {
        "id": "D5AhzA8smLIT"
      },
      "execution_count": 39,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(6,6))\n",
        "plt.title(\"OvR | Green+ = Correct Prediction | Redx = Incorrect | Mean Accuracy: 95.3%\",fontsize=14)\n",
        "for i in range(len(X)):\n",
        "    # correctly predicted\n",
        "    if y[i] == y_hat[i]:\n",
        "        plt.plot(i,X[i,0],'g+')\n",
        "    else:\n",
        "        plt.plot(i,X[i,0],'rx')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 391
        },
        "id": "EUyC4PHbmOTe",
        "outputId": "239c9270-3195-4dff-eaa2-671ef4785265"
      },
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x432 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model = LogisticRegression()\n",
        "ovo = OneVsOneClassifier(model)\n",
        "# fit\n",
        "ovo.fit(X,y)\n",
        "# score\n",
        "print(\"Score:\",ovo.score(X,y))\n",
        "# predictions\n",
        "y_hat = ovo.predict(X)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CnxnG7oPmRSu",
        "outputId": "4eb5521a-7b34-4e57-e2d4-2565d375d05a"
      },
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Score: 0.9733333333333334\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(6,6))\n",
        "plt.title(\"OvO | Green+ = Correct Prediction | Redx = Incorrect | Mean Accuracy: 95.3%\",fontsize=14)\n",
        "for i in range(len(X)):\n",
        "    # correctly predicted\n",
        "    if y[i] == y_hat[i]:\n",
        "        plt.plot(i,X[i,0],'g+')\n",
        "    else:\n",
        "        plt.plot(i,X[i,0],'rx')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 391
        },
        "id": "su3eHRXJmUOE",
        "outputId": "e68d65a5-5c66-432e-8114-726abb812b63"
      },
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x432 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# 3.Multinomial"
      ],
      "metadata": {
        "id": "1jf0HaCQmWZA"
      },
      "execution_count": 43,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "lr_muticlasss_multinomial = LogisticRegression(multi_class='multinomial',max_iter=125)\n",
        "lr_muticlasss_multinomial.fit(X,y)\n",
        "print(lr_muticlasss_multinomial.score(X,y))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UvhnW_j0mbtF",
        "outputId": "4c1456d4-c8f7-44ed-e995-358510c95add"
      },
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.9733333333333334\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y_hat = lr_muticlasss_multinomial.predict(X)"
      ],
      "metadata": {
        "id": "Nv9gqZ9SmcOQ"
      },
      "execution_count": 45,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(6,6))\n",
        "plt.title(\"OvO | Green+ = Correct Prediction | Redx = Incorrect | Mean Accuracy: 95.3%\",fontsize=14)\n",
        "for i in range(len(X)):\n",
        "    # correctly predicted\n",
        "    if y[i] == y_hat[i]:\n",
        "        plt.plot(i,X[i,0],'g+')\n",
        "    else:\n",
        "        plt.plot(i,X[i,0],'rx')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 391
        },
        "id": "3rejtdjomeOi",
        "outputId": "af36330f-5e13-4560-8710-6d850e13f57c"
      },
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x432 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Standard Scaler to the rescue!"
      ],
      "metadata": {
        "id": "dYCDYQBVmhAV"
      },
      "execution_count": 47,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_scaled = StandardScaler().fit_transform(X)"
      ],
      "metadata": {
        "id": "PptsVMtImlvz"
      },
      "execution_count": 48,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_scaled.std(axis=0)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HpW9J_CrmmWI",
        "outputId": "6ac34ea2-e69b-4c60-fd4f-01c563c4410c"
      },
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([1., 1., 1., 1.])"
            ]
          },
          "metadata": {},
          "execution_count": 49
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# 4. One-vs-Rest with Scaled Features"
      ],
      "metadata": {
        "id": "MKXQqJjMmoC2"
      },
      "execution_count": 50,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "lr_scaled_features = LogisticRegression(multi_class=\"ovr\") # one vs rest\n",
        "lr_scaled_features.fit(X_scaled,y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Cm9Pi-79mrLv",
        "outputId": "7693a884-c819-4d9b-cfc1-5c53cdd7a775"
      },
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LogisticRegression(multi_class='ovr')"
            ]
          },
          "metadata": {},
          "execution_count": 51
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# score on unscale\n",
        "lr_scaled_features.score(X_scaled,y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "E5dtSeq-mtd6",
        "outputId": "bb554600-1fe1-4b9f-9405-9ddd7f585e5a"
      },
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9466666666666667"
            ]
          },
          "metadata": {},
          "execution_count": 52
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# 5 One-vs-One with scaled features"
      ],
      "metadata": {
        "id": "BC8Z-aAFmvUK"
      },
      "execution_count": 53,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "lr_ovo_scaled = LogisticRegression()\n",
        "ovo_scaled = OneVsOneClassifier(lr_ovo_scaled)\n",
        "ovo_scaled.fit(X_scaled,y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9LjegSLrm0Fm",
        "outputId": "fd927ebc-5a89-410a-abb9-f21284f935af"
      },
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "OneVsOneClassifier(estimator=LogisticRegression())"
            ]
          },
          "metadata": {},
          "execution_count": 54
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "ovo_scaled.score(X_scaled,y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rK-pyY4Hm2De",
        "outputId": "ef1d2e0f-839b-47a4-a14a-04b3d4e59715"
      },
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9733333333333334"
            ]
          },
          "metadata": {},
          "execution_count": 55
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# 6. Mutlinomial with scaled features"
      ],
      "metadata": {
        "id": "Q0XagtMdm3xt"
      },
      "execution_count": 56,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "lr_scaled_multinomial = LogisticRegression(multi_class=\"multinomial\")\n",
        "lr_scaled_multinomial.fit(X_scaled,y)\n",
        "lr_scaled_multinomial.score(X_scaled,y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "27h6fhg0m8nu",
        "outputId": "4f2e564e-074e-48e1-d8e0-81467e25833b"
      },
      "execution_count": 57,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9733333333333334"
            ]
          },
          "metadata": {},
          "execution_count": 57
        }
      ]
    }
  ]
}