机器学习实战_支持向量机_SVM.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyOkRmO0XG+9oToAgC97ORGP",
      "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/leiyi-ihb/ColabPractics/blob/main/%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E5%AE%9E%E6%88%98_%E6%94%AF%E6%8C%81%E5%90%91%E9%87%8F%E6%9C%BA_SVM.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": "ypAmQ3lxiVz_"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "! if [ ! -f \"Social_Network_Ads.csv\" ]; then wget https://raw.githubusercontent.com/MLEveryday/100-Days-Of-ML-Code/master/datasets/Social_Network_Ads.csv ; else echo \"Found previous downloaded data.\"; fi"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jpDwkADHjIQw",
        "outputId": "1bbc7a85-3e81-47ed-d267-6d3504fa8ae4"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2022-10-10 15:01:46--  https://raw.githubusercontent.com/MLEveryday/100-Days-Of-ML-Code/master/datasets/Social_Network_Ads.csv\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.110.133, 185.199.109.133, 185.199.108.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.110.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 10926 (11K) [text/plain]\n",
            "Saving to: ‘Social_Network_Ads.csv’\n",
            "\n",
            "\rSocial_Network_Ads.   0%[                    ]       0  --.-KB/s               \rSocial_Network_Ads. 100%[===================>]  10.67K  --.-KB/s    in 0s      \n",
            "\n",
            "2022-10-10 15:01:46 (57.8 MB/s) - ‘Social_Network_Ads.csv’ saved [10926/10926]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "dataset = pd.read_csv('Social_Network_Ads.csv')\n",
        "print(dataset)\n",
        "print(type(dataset))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "23Vt43tpkgfC",
        "outputId": "4ccfb1fb-d915-40aa-b7dd-8f41eb68ba55"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "      User ID  Gender  Age  EstimatedSalary  Purchased\n",
            "0    15624510    Male   19            19000          0\n",
            "1    15810944    Male   35            20000          0\n",
            "2    15668575  Female   26            43000          0\n",
            "3    15603246  Female   27            57000          0\n",
            "4    15804002    Male   19            76000          0\n",
            "..        ...     ...  ...              ...        ...\n",
            "395  15691863  Female   46            41000          1\n",
            "396  15706071    Male   51            23000          1\n",
            "397  15654296  Female   50            20000          1\n",
            "398  15755018    Male   36            33000          0\n",
            "399  15594041  Female   49            36000          1\n",
            "\n",
            "[400 rows x 5 columns]\n",
            "<class 'pandas.core.frame.DataFrame'>\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X = dataset.iloc[:, [2,3]].values #https://blog.csdn.net/Fwuyi/article/details/123127754#:~:text=iloc%20%5B%5D%E5%87%BD%E6%95%B0%EF%BC%8C%E5%B1%9E%E4%BA%8E%20pandas%20%E5%BA%93%EF%BC%8C%E5%85%A8%E7%A7%B0%E4%B8%BAindex%20location%EF%BC%8C%E5%8D%B3%E5%AF%B9%E6%95%B0%E6%8D%AE%E8%BF%9B%E8%A1%8C%E4%BD%8D%E7%BD%AE%E7%B4%A2%E5%BC%95%EF%BC%8C%E4%BB%8E%E8%80%8C%E5%9C%A8%E6%95%B0%E6%8D%AE%E8%A1%A8%E4%B8%AD,%E6%8F%90%E5%8F%96%E5%87%BA%E7%9B%B8%E5%BA%94%E7%9A%84%E6%95%B0%E6%8D%AE%E3%80%82%202%20iloc%E5%87%BD%E6%95%B0%E4%BD%BF%E7%94%A8%20df.iloc%20%5Ba%2Cb%5D%EF%BC%8C%E5%85%B6%E4%B8%ADdf%E6%98%AFDataFrame%E6%95%B0%E6%8D%AE%E7%BB%93%E6%9E%84%E7%9A%84%E6%95%B0%E6%8D%AE%EF%BC%88%E8%A1%A81%E5%B0%B1%E6%98%AFdf%EF%BC%89%EF%BC%8Ca%E6%98%AF%E8%A1%8C%E7%B4%A2%E5%BC%95%EF%BC%88%E8%A7%81%E8%A1%A81%EF%BC%89%EF%BC%8Cb%E6%98%AF%E5%88%97%E7%B4%A2%E5%BC%95%EF%BC%88%E8%A7%81%E8%A1%A81%EF%BC%89%E3%80%82\n",
        "# print(type(X))\n",
        "y = dataset.iloc[:, 4].values\n",
        "# y\n"
      ],
      "metadata": {
        "id": "VhBx00AlouMj"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=0)\n",
        "# print(X_train, X_test)\n",
        "# print(y_train, y_test)"
      ],
      "metadata": {
        "id": "26cTSOUnrjzT"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.preprocessing import StandardScaler\n",
        "sc = StandardScaler()\n",
        "X_train = sc.fit_transform(X_train)\n",
        "X_test = sc.fit_transform(X_test)\n",
        "# print(X_train, X_test)\n",
        "# print(X_test.mean())\n"
      ],
      "metadata": {
        "id": "mTdVg6spoykP"
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.svm import SVC\n",
        "classifer = SVC(kernel='linear', random_state=0)\n",
        "classifer.fit(X_train, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uddG5BAhemmQ",
        "outputId": "657956de-4986-4d02-f6b7-4b15398cf9be"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "SVC(kernel='linear', random_state=0)"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y_pred = classifer.predict(X_test)\n",
        "print(y_pred)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QnVA3qjFfVsu",
        "outputId": "1c043331-af86-4c1c-8e5d-edf679d70000"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 1 0 0 1 0 1 0 1 0 0 0 0 0 0 1 0 0 0 0\n",
            " 0 0 1 0 0 0 0 1 0 0 1 0 1 1 0 0 1 1 0 0 0 0 0 0 1 0 1 0 1 0 0 0 1 1 0 0 0\n",
            " 0 0 1 0 1 1 1 1 0 0 1 1 0 1 0 0 0 1 0 0 0 0 0 0 1 1]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import confusion_matrix\n",
        "cm = confusion_matrix(y_test, y_pred)\n",
        "cm"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Xk0IF1RpgiPt",
        "outputId": "5d4fd138-a8b4-4215-87be-3699ecba2498"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[63,  5],\n",
              "       [ 7, 25]])"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from matplotlib.colors import ListedColormap\n",
        "X_set, y_set = X_train, y_train\n",
        "X1, X2 = np.meshgrid(np.arange(start=X_set[:, 0].min() - 1, stop=X_set[:, 0].max() + 1, step=0.01),\n",
        "                     np.arange(start=X_set[:, 1].min() - 1, stop=X_set[:, 1].max() + 1, step=0.01))\n",
        "plt.contourf(X1, X2, classifer.predict(np.array([X1.ravel(), X2.ravel()]).T).reshape(X1.shape),\n",
        "             alpha=0.75, cmap=ListedColormap(('red', 'green')))\n",
        "plt.xlim(X1.min(), X1.max())\n",
        "plt.ylim(X2.min(), X2.max())\n",
        "for i, j in enumerate(np.unique(y_set)):\n",
        "    plt.scatter(X_set[y_set == j, 0], X_set[y_set == j, 1],\n",
        "                c=ListedColormap(('red', 'green'))(i), label=j)\n",
        "plt.title('SVM (Training set)')\n",
        "plt.xlabel('Age')\n",
        "plt.ylabel('Estimated Salary')\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 352
        },
        "id": "td4jzAnimZb_",
        "outputId": "9b3db56b-62ed-4919-cbd7-5ce19368dc08"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:matplotlib.axes._axes:*c* argument looks like a single numeric RGB or RGBA sequence, which should be avoided as value-mapping will have precedence in case its length matches with *x* & *y*.  Please use the *color* keyword-argument or provide a 2-D array with a single row if you intend to specify the same RGB or RGBA value for all points.\n",
            "WARNING:matplotlib.axes._axes:*c* argument looks like a single numeric RGB or RGBA sequence, which should be avoided as value-mapping will have precedence in case its length matches with *x* & *y*.  Please use the *color* keyword-argument or provide a 2-D array with a single row if you intend to specify the same RGB or RGBA value for all points.\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from matplotlib.colors import ListedColormap\n",
        "X_set, y_set = X_test, y_test\n",
        "X1, X2 = np.meshgrid(np.arange(start=X_set[:, 0].min() - 1, stop=X_set[:, 0].max() + 1, step=0.01),\n",
        "                     np.arange(start=X_set[:, 1].min() - 1, stop=X_set[:, 1].max() + 1, step=0.01))\n",
        "plt.contourf(X1, X2, classifer.predict(np.array([X1.ravel(), X2.ravel()]).T).reshape(X1.shape),\n",
        "             alpha=0.75, cmap=ListedColormap(('red', 'green')))\n",
        "plt.xlim(X1.min(), X1.max())\n",
        "plt.ylim(X2.min(), X2.max())\n",
        "for i, j in enumerate(np.unique(y_set)):\n",
        "    plt.scatter(X_set[y_set == j, 0], X_set[y_set == j, 1],\n",
        "                c=ListedColormap(('red', 'green'))(i), label=j)\n",
        "plt.title('SVM (Test set)')\n",
        "plt.xlabel('Age')\n",
        "plt.ylabel('Estimated Salary')\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 352
        },
        "id": "PWptfF7im5NF",
        "outputId": "1d85e743-3b22-44f9-d416-be0c5139408b"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:matplotlib.axes._axes:*c* argument looks like a single numeric RGB or RGBA sequence, which should be avoided as value-mapping will have precedence in case its length matches with *x* & *y*.  Please use the *color* keyword-argument or provide a 2-D array with a single row if you intend to specify the same RGB or RGBA value for all points.\n",
            "WARNING:matplotlib.axes._axes:*c* argument looks like a single numeric RGB or RGBA sequence, which should be avoided as value-mapping will have precedence in case its length matches with *x* & *y*.  Please use the *color* keyword-argument or provide a 2-D array with a single row if you intend to specify the same RGB or RGBA value for all points.\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAEWCAYAAABv+EDhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3df5xcdX3v8ddnN8kGTISYkk2AhLgPJaKIoUYlGB67BfUCir+wfUBTLFaa6q23ilJvbS5t9d7c1ttevbR6HxppBTWFS0EUBGuRZlcoSzRgRDFZxCUJMckG1iQQSDa7O5/7x5lJdmfPzJzZmTPnzJz38/HYx+6cmXPmO7vJ93O+38/3h7k7IiKSPW1JF0BERJKhACAiklEKACIiGaUAICKSUQoAIiIZpQAgIpJRCgAiZZjZLWb27qTLUQsz6zSzrWbWkXRZJF0UACT1zGyVmT1kZgfN7Ndm9h9m9gYzO8/MXjCzOSHn/NjMPmJmS83MzezHRc//hpkdNbPtZd73HOB1wLfN7M/N7FD+64iZjU94/Pg0PlOPme2q9ryI195uZm8pPHb3IWAjsCaO95PmpQAgqWZmLwW+A/wD8DLgNODTwIi7PwzsAt5XdM7ZwKuBWyYcPjF/vOB3gacqvP0fARs88D/dfY67zwE+BPQXHrv7a2r4iI2ygeDziByjACBpdyaAu9/i7uPuftjd/83dH8s/fzPw/qJz3g/c6+7DE459Hfj9otd8rcJ7XwL0VSqgmb3KzO7Lt04GzOx3Jjx3qZn93MyeN7Nfmdl1ZvYS4LvAqRNaEaeGXHfKuROee4eZbTGzA/nW0Tn5418HlgB356/7yfwpm4AuMzuj0ueRDHF3fekrtV/AS4Fhgor+EmBe0fOLgTFgcf5xG0Gr4N35x0sBz39/GmgnaB1sA94CbC/xvi/Jn3dKyHNXAw9OeN3TwAeAGcC5wLPAq/PP7wEuyP88D/jN/M89wK4Kn73UuecC+4A35T/P7wPbgY7889uBt4Rc7zHgnUn/TfWVni+1ACTV3P05YBVBZfwV4Bkzu8vMOvPPPw30AlflT7kI6ADuKbrULmCAoNJ/P0GLoJyT89+fr/C6dxAEka+6+5i7/xi4A/jt/POjwKvN7KXuvt/dH61wvYlKnbsG+LK7b/KgVXQzMAKcV+F6z0/4XCIKAJJ+7r7V3a9299OBs4FTgf8z4SU3czwAXAXc6u6jIZf6GsHd+5VUDgAH8t/nVnjdGcCb8l0xB8zsALAaWJh//nLgUmCHmfWZ2coK15uo1LlnAJ8oes/FBL+XcuZO+FwiCgDSXNx9G3ATQSAo+CZwupn9FvBegoAQ5g7g7cCgu++s8D4vAL8kn4Mo42mgz91PnvA1x90/nL/Oj9z9XcAC4FvAbYW3qHDdcuc+Dawres8T3b2Q9J5ybTObAbwC+Eml95XsUACQVMsnWD9hZqfnHy8muIN/uPCafGV9O/BVYIe7bw67Vv51FwLXRHz7e4HuCq/5DnCmmV1lZjPzX28ws7PMbJaZrTazk/ItkueAXP68IWC+mZ1U4nOXO/crwIfM7E0WeImZvd3M5k64dlfRJd9I0FW1I+JnlwxQAJC0e54g2bnJzF4gqPh/Bnyi6HU3E3SNlB3Z4+6b3f2XEd97PbDazKzM9Z4H3gZcAewG9gKfJchDQNAltd3MniMYPro6f942gmGqg/lunLDum1Lnbgb+EPgCsB94kqBrq+Cvgf+Wv25h5NBq4EsRP7dkhLlrQxiRUszsn4Hb3P1bSZdlusxsAcFw1nPd/UjS5ZH0UAAQEckodQGJiGSUAoCISEYpAIiIZNSMpAtQjd+YOdOXzp6ddDHid/gwjywYZ07HlEUuRUSqdmj7oWfd/ZTi400VAJbOns3mFSuSLkZDzLyglxfaDtH98p6kiyIiTa736t7Q+R/qAkqp0Qd6aM9B31O9SRdFRFqUAkCKjT7QAygIiEg8FABSLtfXAygIiEj9NVUOIKtyfT20dffS91SvcgIiCZjTPocrllzBohMW0ZbS++YcOfYc3sOtO2/l0PihSOcoADQJBQGR5Fyx5ArOPv1sOuZ2UGZpqES5O/Ofn88VXMGNT90Y6Zx0hjIJpe4gkWQsOmFRqit/ADOjY24Hi05YFPkcBYAmoyAg0nhttKW68i8ws6q6qBQAmpCCgIjUgwJAk1IQEMmeB+5/gIvPu5i3veFtrL9hfc3XUwBoYrk+TRYTyYrx8XE+82ef4Su3foXv/Md3uOfOe3hy4MmarqkA0ORGH+ihe4cpCIikyNzb76br3As5c8FZdJ17IXNvv7vmaz726GMsWbqExUsXM2vWLC5996Xc/937a7pmYgHAzGab2Q/N7Cdm9riZfTqpsjS7jduDbWsVBESSN/f2u1n48euZuWs35s7MXbtZ+PHraw4CQ3uGWHTa8RE+C09dyNCeoZqumWQLYAS40N1fBywHLjaz8xIsT1NTTkAkHU5Z93naDk/eebPt8BFOWff5hEpUWmIBwAOF6Woz81/an7IGCgIiyZvxqz1VHY+qc1EneyZcY+/uvXQu6qzpmonmAMys3cy2APuA+9x9U8hr1pjZZjPb/MzoaOML2WQUBESSNXZa+ESsUsejeu25r2XHUzvYtWMXR48e5d5v3cuFF19Y0zUTDQDuPu7uy4HTgTea2dkhr1nv7ivcfcUpM2c2vpBNSEFAJDnPrL2W3AmTN67KnTCbZ9ZeW9N1Z8yYwfV/fT0f/J0P8vY3v51L3nkJr3zVK2u7Zk1n14m7HzCzjcDFwM+SLk8ryPX1MPMCrR0k0mjPv+8yIMgFzPjVHsZOW8Qza689drwW3W/tpvut3TVfpyCxAGBmpwCj+cr/BOCtwGeTKk8rGn1AQUAkCc+/77K6VPhxS7ILaBGw0cweA35EkAP4ToLlaUnaVEZESklyFNBj7n6uu5/j7me7+2eSKkur04xhEQmjmcAZoZaAiBRTAMgQjQ4SkYkUADJG3UEiUqAAkEHqDhJpTn/+J3/O+Wedz2UX1GeEkQJARqk7SKT5vOeK9/CVW79St+spAGSYuoNE4nH3E3dz4c0XctYXz+LCmy/k7idqXw4a4A3nv4GT5p1Ul2uBAkDmjT7Qw0lHFARE6uXuJ+7m+o3Xs/vQbhxn96HdXL/x+roFgXpSABD2b9KmMiL18vn+z3NkbPJy0EfGjvD5fi0HLSk1cVOZLXu3JFwakea151D4ss+ljidJAUCOyfUF3UEHDx9IuigiTWvRnPBln0sdT5ICgEyyf1MPoJyAyHRdu/JaZs+YvBz07BmzuXZlbctBA3x8zce58pIreerJp+g+p5vbv3F7TddLxXLQki65vh7autO5iujQoSEG9w8yMj5CR3sHXfO66JxT265IIvV02ZnBGP3P93+ePYf2sGjOIq5dee2x47X43PrP1XyNiRQAJFQag8DQoSEGhgfIeQ6AkfERBoYHABQEJFUuO/OyulT4cVMXkJSUtnkCg/sHj1X+BTnPMbh/MKESiTQ3BQApK03zBEbGR6o6LlIvOXK4e9LFqMjdyZGr/MI8BQCpaP+mdASBjvaOqo7XYujQEP1P99O7vZf+p/sZOjRU9/eQ5rHn8B5Gnh9JdRBwd0aeH2HP4ejDTZUDkEj2b+ph3puCnMBJJ5zM8oXLG16Grnldk3IAAG3WRte8rrq+j3INUuzWnbdyBVew6IRFtKX0vjlHjj2H93Drzlsjn6MAIJHt39QDDz5I29pk5gkUKt96jwIqHlk07uMlcw3NGgCyMHoqzs94aPwQNz51Y12ulSYKAFKdVauA5FoCnXM661pxhd3tl9KsuYYstGiy8BnjkM62jKRaK80YDhtZVEocuYZGyMLoqSx8xjgoAMi0tMqM4ah39XHkGhqlmtFTzZr81gix6VEAkGlrhU1lSt3Vt1v7sec62jtYNn9Z03YlRB09VehGKVSahW6UZggCjRwh1koUAKQmaZssVq2ueV202eT/Bm3Wxpnzz2Tl4pX0LO1h5eKVTVv5Q+nPWNyiaeZulKifUSZTAJCapWmyWLU653SybP6ylrnbDxP1MzZzN0oW/o5x0CggqYv9m3r4raV99JHcPIHpqvfIojSK8hk72jtCK/tm6UbJwt+x3tQCkLrZuL2b3LoZHDx8gAd3Pph0caRK6kbJnsQCgJktNrONZvZzM3vczD6aVFmkjlatonuHMT4+piDQZNSNkj1JdgGNAZ9w90fNbC7wiJnd5+4/T7BMUgcbt3fDNx6kbe1Y0kWRKqkbJVsSawG4+x53fzT/8/PAVuC0pMojdbZqFRAkhtUSEEmnVOQAzGwpcC6wKeS5NWa22cw2PzM62uiiSQ0KM4bHx9USEEmjxAOAmc0B7gA+5u7PFT/v7uvdfYW7rzhl5szGF1Bqsn9Tc88TEGlliQYAM5tJUPlvcPdvJlkWiU8zzxMQaWWJJYHNzIB/BLa6e313OpbUScN+ApJucSznnIVlsGuRZAvgzcBVwIVmtiX/dWmC5ZGY7d/UQ/cO4+DhA2zZuyXp4kiKxLEOUTOvbdQoSY4CetDdzd3Pcffl+a97kyqPNMbEyWIiBXGsQ9TMaxs1SuJJYMmgVauOJYY1RFQgnnWImnlto0ZRAJBEjD7Qc2zGsJLDEsdyzloiujIFAEnMxu3d5G44OeliSArEsQ6R1jaqTAFAkrV8ubqDJJZ1iLS2UWVaDloSN/pAMET04GzNGM6yONYh0tpG5akFIKmwf5Mmi4k0mloAkhrNvKmMNJYmeNWHWgCSKtpURirRBK/6UQCQ9Fm1ity6GdpURkJpglf9KABIOq1aRe6GkxkfH9OyETKJJnjVjwKApFd+iOjBwwfo296XdGkkJTTBq34qBgAz+y9mNq8RhREpVpgxjDsAFz00xC2f6Of+q3u55RP9XPSQ+n2zRhO86idKC6AT+JGZ3WZmF+eXcRZpmI3bu2nPwal39XLdTQMsHB6hDVg4PMJ1Nw0oCGSMJnjVj3n+zqrsi4JK/23AB4AVwG3AP7r7L+Mt3mQr5s71zStWNPItJUV2PtrLkil7xsHe+R1c+b9XNr5AIk2i9+reR9x9SuUZaR6Au7uZ7QX2AmPAPOB2M7vP3T9Z36KKhAur/AEWDIcn/y56aIhr7hhkwfAI++Z3cOPlXdx/vu4SRQoqBgAz+yjwfuBZ4EbgT9191MzagF8ACgDSGB0dMDK1st83f2ry76KHhrjupgFmHw2GCxa6iwAFAZG8KDmAecB73f0/ufu/uPsogLvngHfEWjqRibq6oG3yP9kXZsKNl09N/l1zx+Cxyr9g9tEc19yhseIiBWVbAGbWDlzh7n8V9ry7b42jUCKhOvN37oODQUvAjDWXObecuo1uJt/Vl+oWKnU8TbTMgTRK2RaAu48DA2a2pEHlESmvsxNWroSeHujuZsNdM8B9yiJyYd1C5Y6nhZY5kEaK2gX0uJndb2Z3Fb7iLphIJKtWkevrAZg0Y/jGy7s4MmvyP+8js9pCu4vSRMscSCNFGQV0feylEKnRxBnD3Uu7jyV6m20UkJY5kEaqGADcXXPwJfVGH8gvJX3G8Xkt95/fmfoKv1hHe0doZa9lDiQOUZaCOM/MfmRmh8zsqJmNm1mJEdkiydm4vbvpN5WJa5mDoUND9D/dT+/2Xvqf7ldOQYBoOYAvAFcSjPk/AbgG+GKchRKZrv2bgrWD+p7qbcpVRONY5kCJZSkl6kzgJ82sPT8q6Ktm9mPgU/EWTWR6Nm7vhm88SNvaYFOZVUtWJV2kqtR7H9tyiWUNL822KC2AF81sFrDFzP6XmV0b8TyR5EzYVCbrS0krsSylRKnIrwLagY8ALwCLgcvjLJRIXeQ3lcE9tUFAffOSpIoBwN13uPthd3/O3T/t7h939yfr8eZm9k9mts/MflaP64lMsXx5ME8ghUFAffOStJIBwMx+amaPlfqq0/vfBFxcp2uJlJRbFz5jOEmNmvSlHbSklHJJ4NgXenP3H5jZ0rjfRySYMQxt3cHooOULlyddoob1zXfN62JgeGBSsNEOWgJlAoC772hkQUoxszXAGoAlHbpjkdqcdAQOcnzGcJIaNemrMNJHC8xJsSj7AZwH/ANwFjCLICH8gru/NOayAeDu64H1EOwI1oj3lNa1f1MPPPggbWvHjrUEklp9s5F35vUeWiqtQRPBJHtWraJ7h3Hw8AE27dqUWCJWe9tK0jQRTDKpMFnM1h4Gm/xcIydJ6c5ckpToRDAzuwXoB5aZ2S4z+2A9riuNt2HBEEvP66etu5el5/WzYUETDGVctaq47j9Gk6QkC6K0AK4iqPA/AlxLHSeCufuV9biOJGvDgiHWLBvgxfagL3vH7BHWLAv23129L913t0tGOtgxW6tvSjZFnQh2BDgM3AX8Tb0mgklrWNs1eKzyL3ixPcfarvRvYrJusIsTx+u/+qZIMyg3EexLZvaa/M8nAT8Bvgb82Mx05y7H7OwI7y4pdTxNVu/rZP3AMs440oE54JDL5dQvL5lQrgVwgbs/nv/5A8AT7v5a4PXAJ2MvmTSNJSPh3SWljqfN6n2dbH94Jbm+HnzdDIzm3lNAJKpyAeDohJ/fCnwLwN33xloiaTph3SgnjrexbrAJu1FK7DEs0orKBYADZvYOMzsXeDPwrwBmNoNgPoAIMLUb5YwjHawfWJb6BHA5hXkCaVtATqSeyo0C+iPg74GFwMcm3PlfBNwTd8Gkuaze19nUFX6x45vKjDXlpjJRxDEDOqlZ1TI95t48qyusmDvXN69YkXQxWtfQEAwOwsgIdHRAVxd0RvzPW8u5aZZfNgKzxNcOqqfCUtTFy1DUMhM5jmtKffRe3fuIu0+pPCPNBJYmUWsFPjAAufx/3pGR4DFUvsbQEGzbBoWbiZGR4HGUc9Nu1SpyN2yh7aPpWECuXuLYJjKNW0+qRVKeAkCrqKUChyBw5Cb/5yWXC45XOv8Xvzhe+Re4B8cL164UlNLcgli+/NhS0n1P9dL98p6kS1SzOJaiTtvWk8UtksI6T4CCQJ4CQLMqrjDHx6dfgUNwnWqOTzQ2Vvp4lKBUTfBKMFDkbjg5aAm0QBAotRT1jLYZ9D/dP6075kYtbx1VGlskaVMyAJjZx8ud6O6fq39xJFRxpTd/PuzdO7nCLCVKBQ7BdcNeW+seDFGCUtTWR62tnFpNaAk0e2I4bClqwxjLjTFGENCrvWNO28YzaWuRpFG5YaBz818rgA8Dp+W/PgT8ZvxFE+B4pVeonEdGYPfuqRVmKVEr8K4S/0lLHZ+ovT3aexQUB5qorY9ygaKBuncY4+NjTT1ZLGwp6jabWh1Us0Vl2pa31laYlZXbEezTAGb2A+A33f35/OO/QsNAGyes0ouqrS1aBV5gNrkv30qtlVmkszMISpWuV9DeDv39x1s07e1BF1ax4uBVSzdVHW3c3g3f2tL03UHFS1H3bu8NfV01d8xpWt66a14XT+7bxmjb8X+DM3PG/Dnzp93N1WqiLOvcyeRZwUfzxyQOQ0NB5djbe7ySjKq9/Xil2dEBy5ZF7xoZHAxP5Ea5ux4eLl2etpB/Yrnc5BZNWIALC16lWjNJbBW6fPmxGcMP7nyw8e8fg1a7Y/7dx2D9Xc4ZBwgmKB6ADzziPPPcnkQ2AEqjKEngrwE/NLM784/fDdwcX5EyLKyPO6q2NjjzzOn3hddyd13qNWNjcNZZU5PVxUljd5gxIwgY5ZK7XV2Tfz9QfSunznLrZtC2dqypWwIFaevDr9U1dwyycBiunrCix9KPMalFANlODFcMAO6+zsy+C1yQP/QBd/9xvMXKqKjdPW1tsHBhcOcdZTRMlJEztSSBy53b2Tn5vXp7w68xNgarKiRVC9dJ03DRFpon0Gqbxy8YnvpvcudJ4a/NamI46jDQE4Hn3P2rZnaKmb3c3Z+Ks2CZVO5uu1DJxjXBq5q760qjksqdW+too+KAEpdqhpvWME8gbROV0tSHX6t98ztYWBQElhyEHSdPfW2zdnPVqmIOwMz+EvivHN8DeCbwjTgLlVnl+rhXroSenuB7NRVg1JEznZ1BzqBSDiFsVNLevUGLJEr+Yf788HKecMLk3MdQgn2yQ0NsmLGVpR8eoe0vYemHR9gwY2vFMhVyAlFHBw0dGuLJfdsm9Uc/uW9bZvuj6+3Gy7s4MmtyFfdXvUEieKJm7uaqVZQWwHuAc4FHAdx9t5nNjbVUWRVHH3c1fftR7q5LBZTh4SA4VVIqYXzgwOSyNXJ8f5ENs59gzSXw4qzg8Y6TYc1lwHefYHWF8Q+5vp7ILYHde3/B6IzJ/dGjbc7uvb+g8xWtcReepPvPD36H19wxyILhEfbN72DXG7t4xYLW6eaqVZQAcNTd3cwcwMxeEnOZsiuOPu56T/CqdShm1NdVM4u5GhG6dtZ2jx+r/AtenBUcX128RUDI9aLOGH6uPXwGdanjUr37z+88FggKOtFSEAVRAsBtZvZl4GQz+0PgD4Ab4y1WhoXdhdey/EG9WxW1BpRS54ep9/j+iPmQUonCKcdLXW/ZMnJ9yyvOGF58EHaG9EcvPljFZxKpQZRN4f8OuB24A1gG/IW7/33cBZO8sD73gYHofeRR+/aj6uqaOra/moASdn4p9R7fHzEfsqREBTzleLnrDQ2x++9g5C/G+OpHe7nooal/r//6cDsnHp187MSjwXGRRoiSBP6su9/n7n/q7te5+31m9tlGFE6oz/IHnZ3TTyKHXauWgBJ2/qmn1hZUoorYfbXufkIr5nX3V3G9gQEWvhD8B1t6EK796tYpQeDX55zJF+9l0kSlL94bHBdphChdQG8lGAU00SUhxyQOKVn+YJJah2KGnX/SSakZ37/6iQ64e4S1FwXdPksOBpX/6ic6YGKeu1x3VlHQfsko/N7/2zqpP/r+8zu5CHj468eTlDde3jWlz1okLuVWA/0w8J+BLjN7bMJTc4H/iLtgkhfXKp1p06jx/VF0dbH68QFW/7Qob7Ksa8rrQvMrJSbzLTnIlMRwWJJSpFHKdQH9M3AZcFf+e+Hr9e7+ew0om0Dtfe5yXNS1hKJ2c5V6XYn3acsfb+ZVRKW1lFsN9CBwELgSwMwWALOBOWY2x9131vrmZnYxcAPQDtzo7n9T6zVbThqXP2hW1YyIitoiKfW6Eu+T6+tsqZ3FpLlVzAGY2WXA54BTgX3AGcBW4DW1vLGZtQNfJMgx7AJ+ZGZ3ufvPa7luS0pT90gza1QwrfA+EyeLfWbPWZMmKikHII0UJQn8P4DzgO+7+7lm9ltAPbqA3gg86e6DAGZ2K/AuQAFA4tOoYFrhfXJ9Pax+WS/X3r2Vl4wGxxYOj3DdTcG8BAUBaYQoA7JH3X0YaDOzNnffSLBLWK1OA56e8HhX/tgkZrbGzDab2eZnRkfr8LYpV7wfQJJr4kisNtzTcazyL5h9NMc1dzR2hzPJrigtgANmNgf4AbDBzPYBL8RbrOPcfT2wHmDF3Lkh20u1kKT3vJXGKjGENGwZY5E4RGkBvAs4DFwL/CvwS4LRQLX6FbB4wuPT88eyKyV73kqDlBgttG9+iw3xldSKshTEC+4+TrAnwN0ES0HX4078R8ArzezlZjYLuIJgyGl2pXHSl8QnZIjvCzODZYxFGiHKUhB/ZGZ7gceAzcAj+e81cfcx4CPA9whGFd3m7o/Xet2mlqY9byV+nZ3BPgoT3PQ6+ItFWxMqkGRNlBzAdcDZ7v5svd/c3e8F7q33dZtWXHvehq0mCppbkLShoWAznQn++NE2HlqS4xY0T0DiFyUA/BJ4Me6CCPGMUw9LLG/dCmbBZuyFY0o2T1bLEtxRlcj5bLing1teO6LJYhK7KAHgU8BDZrYJONYZ7e5/Elupsqze49RLbTTvRWmcuDZgaUaNGo1VJudTzc5iItMVJQB8Gfh34KdA+CpXkl7VJJCVbA6UG40VNQBE6XabMQPGQnb/yud8FARKGzo0pG0d6yBKAJjp7h+PvSQSj2p24FKyOVDraKyo3W4w+RhMyfkoCEw1dGiIgeEBch78fkfGRxgYDlpoCgLViTIP4Lv52biLzOxlha/YSyb1UWoHLrPJj7XC6HG1jsaK2u0Gwe+9wqqjub4eQKuIFgzuHzxW+RfkPMfgfs2XqVaUFsCV+e+fmnDMAdUWzaBUYrnUsf7+6SU+n3gCdu8+/vjUU+HMJt3ZqtbRWNV0pY2PwwUXVHzZsZbA9j66l3ZHv34LGhkP//2WOi6lVQwA7v7yRhREYlQqsTzxWC2Jz+LKH44/TlsQiDK6p9bRWDF1u+VuOJm2jx7IfHdQR3tHaGXf0a4uzGqV2xHsQnf/dzN7b9jz7v7N+IolDVdL4rO48p94PE0BoJogV8torLAWBFTs769o+XJyfWQ+J9A1r2tSDgCgzdromqdOiWqVawF0E4z+CVv3xwEFgFYS1zIUjRhPH1U1Qa6WclfT7TaN30XWE8OFRK9GAdWu3I5gf5n/8TPu/tTE58xM3UKtpr096I8OO16LNK1uGjXI1WMeQJRutxooCHSqwq+DKKOA7gg5dnu9CyIJKx4VVOn4RKeeWvrcNK1uGnV0T5OsyqrRQVKrcjmAVxFs+3hSUR7gpQR7A0sckuoyCZuQVO74RIV+/uJRQKVyA0lNOIs6uqeJusOy3hKQ2pTLASwD3gGczOQ8wPPAH8ZZqMxKckOYUiNXoo5SOfPMqQnf4eHarllvjdoTOMzQEGzbNnki2LZtk8s1TQoCMl3lcgDfBr5tZivdvb+BZcqueixBMF1xrEQa1+qmSZruPIlf/GLqRDD34Hgd/rYKAjIdUXIA7zGzl5rZTDO738yeMbN6bAovxZLcEKazM5iFWmFWamzXbMReyIUWVuH3WWhhFb9XuRZKpXNLqaWLLSLlBKRaUWYCv83dP2lm7wG2A+8l2B/4G3EWLJNq7YapRqn+6Hq3NKJcs1FdX1FbWKXG8RdL4QqqaglINaK0AGbmv78d+Bd3PxhjebItbN2eOLpMot4JN0qjRt1EbWGFtVyqvWaxUsNpax1mG0ItAYkqSgvgbjPbRrAx/IfN7BTgSLzFyqhGJSmTzDWEaVTXVzUtrOKWS6HvP8q5MLWF1dkZPioqppnSaglIFFE2hf8z4HxghY1X4koAAA0ASURBVLuPEuwO9q64C5ZZnZ2wciX09ATf46iQ07b5fKP2Qp4/v7rjE1XTOgtrYe3dGwyNndiqOOusWAOuWgJSSckAYGafnPDwIncfB3D3FwDtBtbM0rb5fKO6voaHqzs+UTUJ7VItrOHh+IN7EQUBKadcC+CKCT9/qui5i2MoizRKoyrcqOIYgRSm1pZP1NZZylpYCgJSSrkcgJX4OeyxNJMkJ0SVK1O937+4H77Uekf1bvk0cjRXRLnbzmL77q0sOdjLvvkd3Hh5F/efn57RS5KMcgHAS/wc9liaTRwVbpqEDS0NW9cojpZP2ibA5X8XS/PFWTg8wnU3BcNssxgEtJ/wceW6gF5nZs+Z2fPAOfmfC49f26DyiUxPWD+8e7ARe9xdTY3q0iphw4Ihlp7XT1t3L0vP62fD7Cem/C5mH81xzR3pWtyuEQr7CRc2lCnsJzx0KKEh0AkrtxRE/QcoizRKqf72sTFYtSr+90+ohbVhwRBrlg3wYntQ4e+YPcKaS4AxWP3Tya9dMJy9LRTL7SecxVZAlIlgIs0nbSOdGmRt1+Cxyr/gxVmw9qKpr915UvYSw9pPeDIFAGlNaRvp1CA7O8Irsp0nFR1oa2PpqWcBrRMEhg4N0f90P73be+l/uj+0W6fUvsFZ3U84kQBgZr9tZo+bWc7MViRRBmlxCffDJ2XJSHhFtuTF8NxHqwwRjdq33zWvizabXO1leT/hKEtBxOFnBIvKfTmh95csaPWRTiHWDXZNygEAnDjexrodr4SV4b+LictGvOqUs5pyhEzUvn3tJzxZIgHA3bcCWJTtBkUkstX7gopsbdcgOztGWDLSwbrBrmPHS8n19WDdvWx9ZuuxWT6Fu2gg9RVkNX372k/4uKRaAJGZ2RpgDcCSFk/gNY2ktq2USFbv66xY4Yc5Y6SDHbMnV5jNMkKmo70jtLLPat9+VLEFADP7PrAw5Km1+d3GInH39cB6gBVz52oCWtIauW2lAk1DlUog1zpCphETr7rmdTEwPDCpGyjLfftRxRYA3P0tcV1bGqi4Eh4ba8xS0knuj5wRGxYMTeoqetloO8Ozpi6VUctddCE5W6iY4+pWUt/+9KS+C0gSFFYJl1Lvhc7StmdBShVX4lH6+wvnFU8Ym5UzZuZgdOIgGaemu+hGTrxS3371khoG+h4z2wWsBO4xs+8lUQ6pIKwSLqXe+ZmUraiZRoVKfMfsEdzys36XDbBhQeVlDcImjB1tc146NoMzjnRgDmccCf6m257ZOu0yauJVuiU1CuhO4M4k3luqELWyNav/BKsUrqiZNqGzfttzrO0arNgKKNXf/+uZYzzbN3mpjFp2FlNyNt00E1hKi1rZegy5+YzO5K1GyVm/JY5PVHLCWMjxWiaLaeJVuikAZNXQULDPbW9v8D1sQ/iwSriUem/gntGZvNWophIvtm6wixPHJ/9tTxxv49Jn509eSTTfnTTdINA5p5Nl85cdu+PvaO9g2fxl6qtPCSWBsyjqCJuwjWMa2TefwZm81Sg563ew8t112ISxS5+dz82L9k5eSXTZwLHXT3ejeSVn08s8juZ7TFbMneubV2Rw6aB6j4fv7y/dv75yZXznSt1NdxRQmKXn9U+ZCAZBMnj7w8f/tm3dvQDTyglIMnqv7n3E3adUnmoBpF0c4+FruYtP225X5WRgItl0Z/2GiZpTyPX1MPOC6SeGJT2UA0i7cuPhp6uWtfKbpW++EDgLQa0QOMNyHQJUl1MYfaAHaP5VRLNOASDt4uhzr3WETWdn0N3T0xN8T1vlD/EEzhZXKjFcKqfQKktJZ5kCQNrFsbNVs9zF10ITyaq2el8n6weWTZoItn5gWdkuJgWB5qYkcNoV5wAguFtvtQq73kolq2fMgPb2lskL1DMJXAslhtOtVBJYLYC0y8LdehzCurnMgsXsWiQvUMtSEPWmlkBz0iigZhDHePh6j5BJ24ibsDkMY2MwXrTaZRMvMFfLUhBxmO48AUmOWgBZVO8RMmkdcVOcrC6u/AuaNC9Qy1IQcVFLoLkoAGRRvUfINHLETZQlLDKilqUg4qQg0DwUALKo3iNkGjXiJq0tjYRUO2yzkRQEmoMCQBbVe2hpHENVw9Ta0mhUORtkOsM2GynX10N7TkEgzZQEzqJ6L+cQ1/IQxYnlWlsazbSMRUT1XAoiDqMPKDGcZmoBZFG9h5bGMVQ1rLunlKh38BpSmwh1B6WXWgBZVe+hpfW+XtTtKKu9g9cS04nQENF0UgtA0inKHb/u4JuKWgLpowAg6VQuYZv2heikJAWBdFEAkHTSnsAtS0EgPRQAJJ2UsG1pGiKaDkoCS3opYdvSNEQ0eWoBiEhi1B2ULAUAEUmUuoOSowAgIonTHsPJSCQAmNnfmtk2M3vMzO40s5OTKIeIpIe6gxovqRbAfcDZ7n4O8ATwqYTKISIpoiDQWIkEAHf/N3cfyz98GDg9iXKISPooCDROGnIAfwB8t9STZrbGzDab2eZnRkcbWCwRSYqCQGPEFgDM7Ptm9rOQr3dNeM1aYAzYUOo67r7e3Ve4+4pTZs6Mq7gikjIKAvGLbSKYu7+l3PNmdjXwDuAid/e4yiEizSvX18O8N2myWFySGgV0MfBJ4J3u/mISZRCR5rB/Uw+glkAcksoBfAGYC9xnZlvM7EsJlUNEmoC6g+KR1CigV7j7Yndfnv/6UBLlEJHmoSBQf2kYBSQiEomCQH0pAIhIU1EQqB8FABFpOgoC9aEAICJNSauI1k4BQESa1ugDPZx0REFguhQARKSpTZwnsGXvlmQL02QUAESk6eX6gpbAwcMHki5KU1EAEJGWoBnD1VMAEJGWodFB1VEAEJGWoiAQnQKAiLQcBYFoFABEpCUpCFSmACAiLUuTxcpTABCRlqbJYqUpAIhIy9u/qYfuHaYgUEQBQEQyYeP2biBoCTy488GES5MOCgAikhmFGcPj42NJFyUVFABEJFM0Y/g4BQARyRwNEQ2YuyddhsjM7BlgR9LlyPsN4NmkC9Eg+qytSZ+1NYV91jPc/ZTiFzZVAEgTM9vs7iuSLkcj6LO2Jn3W1lTNZ1UXkIhIRikAiIhklALA9K1PugANpM/amvRZW1Pkz6ocgIhIRqkFICKSUQoAIiIZpQBQAzP7WzPbZmaPmdmdZnZy0mWKi5n9tpk9bmY5M2vJ4XRmdrGZDZjZk2b2Z0mXJy5m9k9mts/MfpZ0WeJmZovNbKOZ/Tz/7/ejSZcpLmY228x+aGY/yX/WT1c6RwGgNvcBZ7v7OcATwKcSLk+cfga8F/hB0gWJg5m1A18ELgFeDVxpZq9OtlSxuQm4OOlCNMgY8Al3fzVwHvDHLfx3HQEudPfXAcuBi83svHInKADUwN3/zd0Lq0o9DJyeZHni5O5b3X0g6XLE6I3Ak+4+6O5HgVuBdyVcpli4+w+AXyddjkZw9z3u/mj+5+eBrcBpyZYqHh44lH84M/9VdpSPAkD9/AHw3aQLIdN2GvD0hMe7aNGKIqvMbClwLrAp2ZLEx8zazWwLsA+4z93LftYZjSlW8zKz7wMLQ55a6+7fzr9mLUFTc0Mjy1ZvUT6rSDMysznAHcDH3P25pMsTF3cfB5bn85F3mtnZ7l4y16MAUIG7v6Xc82Z2NfAO4CJv8kkVlT5ri/sVsHjC49Pzx6TJmdlMgsp/g7t/M+nyNIK7HzCzjQS5npIBQF1ANTCzi4FPAu909xeTLo/U5EfAK83s5WY2C7gCuCvhMkmNzMyAfwS2uvvnki5PnMzslMJIRDM7AXgrsK3cOQoAtfkCMBe4z8y2mNmXki5QXMzsPWa2C1gJ3GNm30u6TPWUT+Z/BPgeQaLwNnd/PNlSxcPMbgH6gWVmtsvMPph0mWL0ZuAq4ML8/9EtZnZp0oWKySJgo5k9RnBDc5+7f6fcCVoKQkQko9QCEBHJKAUAEZGMUgAQEckoBQARkYxSABARySgFAJGIzOzdZuZm9qqkyyJSDwoAItFdCTyY/y7S9BQARCLIryWzCvggwSxhzKzNzP5vfk+I+8zsXjN7X/6515tZn5k9YmbfM7NFCRZfJJQCgEg07wL+1d2fAIbN7PUE+yMsJdg/4CqCWdKFtWf+AXifu78e+CdgXRKFFilHi8GJRHMlcEP+51vzj2cA/+LuOWBvfvEtgGXA2QRLhAC0A3saW1yRyhQARCows5cBFwKvNTMnqNAduLPUKcDj7r6yQUUUmRZ1AYlU9j7g6+5+hrsvdffFwFMEu2pdns8FdAI9+dcPAKeY2bEuITN7TRIFFylHAUCksiuZerd/B8HmObuAnwPfAB4FDua3lHwf8Fkz+wmwBTi/ccUViUargYrUwMzmuPshM5sP/BB4s7vvTbpcIlEoByBSm+/kN+GYBfx3Vf7STNQCEBHJKOUAREQySgFARCSjFABERDJKAUBEJKMUAEREMur/A6rLLWtTc3RGAAAAAElFTkSuQmCC\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.pipeline import Pipeline\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "from sklearn.svm import LinearSVC\n",
        "svm_clf = Pipeline((\n",
        "    (\"scaler\", StandardScaler()), \n",
        "    (\"linear_svc\", LinearSVC(C=1, loss=\"hinge\", dual=True)),\n",
        "    )) \n",
        "svm_clf.fit(X, y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BBvBLcDroHt_",
        "outputId": "92e02456-d055-402c-aa23-785b21bd8d5b"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Pipeline(steps=[('scaler', StandardScaler()),\n",
              "                ('linear_svc', LinearSVC(C=1, loss='hinge'))])"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.datasets import make_moons\n",
        "from sklearn.pipeline import Pipeline\n",
        "from sklearn.preprocessing import PolynomialFeatures\n",
        "\n",
        "polynomial_svm_clf = Pipeline((\n",
        "        (\"poly_features\", PolynomialFeatures(degree=3)),\n",
        "        (\"scaler\", StandardScaler()),\n",
        "        (\"svm_clf\", LinearSVC(C=10, loss=\"hinge\"))\n",
        "    ))\n",
        "\n",
        "polynomial_svm_clf.fit(X, y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "A7zL2gZcqYEm",
        "outputId": "cec3a061-5c3c-4368-8213-152446a71590"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/svm/_base.py:1208: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.\n",
            "  ConvergenceWarning,\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Pipeline(steps=[('poly_features', PolynomialFeatures(degree=3)),\n",
              "                ('scaler', StandardScaler()),\n",
              "                ('svm_clf', LinearSVC(C=10, loss='hinge'))])"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.svm import SVC\n",
        "poly_kernel_svm_clf = Pipeline((\n",
        "        (\"scaler\", StandardScaler()),\n",
        "        (\"svm_clf\", SVC(kernel=\"poly\", degree=3, coef0=1, C=5))\n",
        "    ))\n",
        "poly_kernel_svm_clf.fit(X, y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KbQwn_jorBiD",
        "outputId": "9f845e81-4f7e-4fee-fd29-4728f62f328e"
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Pipeline(steps=[('scaler', StandardScaler()),\n",
              "                ('svm_clf', SVC(C=5, coef0=1, kernel='poly'))])"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#调参实例  \n",
        "以高斯函数NuSVC进行调参示例演练，这里我选取《kaggle 泰坦尼克：从灾难中学习算法》中经过数据处理的数据集。假设 train_X 为训练特征， train_y 为训练标签，高斯核函数支持向量分类器主要需要调整的参数有两个 nu 和 gamma. 基于分类任务的评价指标指标有很多，我们这里选取 scoring='accuracy' ，有兴趣的同学也可以尝试其他的评价指标，比如 'f1' 'roc_auc' 等。"
      ],
      "metadata": {
        "id": "UiKmTaKAbyjL"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.svm import NuSVC\n",
        "from sklearn.model_selection import cross_val_score\n",
        "from sklearn.model_selection import GridSearchCV\n"
      ],
      "metadata": {
        "id": "aj755GISu2hj"
      },
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "nu_list = np.linspace(0.1, 0.7, 7)\n",
        "accuracy_scroes = []\n",
        "for nu in nu_list:\n",
        "  musc = NuSVC(nu=nu)\n",
        "  scores_accuracy = cross_val_score(musc, X_train, y_train, scoring='accuracy', cv=5)\n",
        "  accuracy_scroes.append(scores_accuracy.mean())\n",
        "print(accuracy_scroes)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vx2gT2dwcm6F",
        "outputId": "19c6d7de-1280-49ed-d692-fd1968086d03"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[0.4833333333333334, 0.57, 0.6266666666666667, 0.6033333333333333, 0.7733333333333334, 0.76, 0.76]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(nu_list, accuracy_scroes)\n",
        "plt.title(\"nu for accuracy_scroes\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 299
        },
        "id": "FAOs_wBRf9tr",
        "outputId": "8d72292a-72ce-4491-eeb9-68e7c9840196"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'nu for accuracy_scroes')"
            ]
          },
          "metadata": {},
          "execution_count": 25
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "由上结果可知，最佳nu值出现在`0.5`左右，所以网格搜索我们设定nu的取值范围在`[0.4, 0.6]`之间  \n",
        "# **搜索gamma的最优取值范围**"
      ],
      "metadata": {
        "id": "UYrBlPG6gfFE"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "gamma_list = np.linspace(0.01, 1, 91)\n",
        "accuracy_scroes = []\n",
        "for gamma in gamma_list:\n",
        "  nsvc = NuSVC(gamma=gamma, nu=0.5)\n",
        "  scores_accuracy = cross_val_score(nsvc, X_train, y_train, scoring='accuracy', cv=5)\n",
        "  accuracy_scroes.append(scores_accuracy)\n",
        "print(accuracy_scroes)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_NL_eHKUgO7o",
        "outputId": "40b8570e-5fbf-4e93-d60d-ffdc89e095e7"
      },
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[array([0.61666667, 0.71666667, 0.65      , 0.7       , 0.75      ]), array([0.63333333, 0.7       , 0.66666667, 0.7       , 0.73333333]), array([0.61666667, 0.65      , 0.63333333, 0.68333333, 0.66666667]), array([0.63333333, 0.65      , 0.61666667, 0.68333333, 0.66666667]), array([0.63333333, 0.65      , 0.61666667, 0.68333333, 0.66666667]), array([0.63333333, 0.63333333, 0.61666667, 0.68333333, 0.66666667]), array([0.65      , 0.63333333, 0.61666667, 0.68333333, 0.66666667]), array([0.65      , 0.63333333, 0.61666667, 0.66666667, 0.68333333]), array([0.65      , 0.63333333, 0.61666667, 0.66666667, 0.68333333]), array([0.65      , 0.63333333, 0.61666667, 0.66666667, 0.68333333]), array([0.66666667, 0.63333333, 0.61666667, 0.66666667, 0.68333333]), array([0.66666667, 0.63333333, 0.61666667, 0.66666667, 0.68333333]), array([0.66666667, 0.63333333, 0.61666667, 0.66666667, 0.68333333]), array([0.66666667, 0.63333333, 0.61666667, 0.66666667, 0.68333333]), array([0.66666667, 0.63333333, 0.61666667, 0.66666667, 0.68333333]), array([0.66666667, 0.63333333, 0.6       , 0.66666667, 0.68333333]), array([0.66666667, 0.63333333, 0.6       , 0.66666667, 0.68333333]), array([0.66666667, 0.63333333, 0.6       , 0.66666667, 0.7       ]), array([0.65      , 0.63333333, 0.6       , 0.66666667, 0.7       ]), array([0.65      , 0.63333333, 0.58333333, 0.65      , 0.65      ]), array([0.65      , 0.63333333, 0.58333333, 0.65      , 0.65      ]), array([0.65      , 0.63333333, 0.58333333, 0.65      , 0.65      ]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.65      ]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.65      ]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.65      ]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.65      ]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.65      ]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.65      ]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.65      ]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.65      ]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.65      ]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.65      ]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.63333333]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.63333333]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.63333333]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.63333333]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.63333333]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.63333333]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.63333333]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.63333333]), array([0.65      , 0.63333333, 0.6       , 0.65      , 0.63333333]), array([0.65      , 0.63333333, 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.63333333, 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.63333333, 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.63333333, 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.63333333, 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.63333333, 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333]), array([0.65      , 0.65      , 0.6       , 0.63333333, 0.63333333])]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(gamma_list, accuracy_scroes)\n",
        "plt.title('gamma for accuracy_scores')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 299
        },
        "id": "caIc7BhYj1Xi",
        "outputId": "72b4e697-47a4-4714-ce6e-0c01e12a8fee"
      },
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'gamma for accuracy_scores')"
            ]
          },
          "metadata": {},
          "execution_count": 27
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "由上结果就可以知道`gamma`值的最优结果在`0.02`左右，所以网格搜索设定ganmma的取值范围在`(0, 0.2]`之间  \n",
        "# 网格搜索"
      ],
      "metadata": {
        "id": "kLk86-kdkX04"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "from joblib.logger import PrintTime\n",
        "param_grid = {'nu': np.linspace(0.3, 0.6, 31), 'gamma': np.linspace(0.01, 0.2, 191)}\n",
        "gsearch = GridSearchCV(NuSVC(), param_grid=param_grid, scoring='accuracy', cv=5)\n",
        "gsearch.fit(X_train, y_train)\n",
        "print(gsearch.best_params_)\n",
        "print(gsearch.best_score_)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Pj0CBXNsk-bG",
        "outputId": "150e6913-3786-47c9-fa35-95a24023f008"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{'gamma': 0.015, 'nu': 0.5700000000000001}\n",
            "0.6966666666666667\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "输出结果为  \n",
        "`{'gamma': 0.015, 'nu': 0.5700000000000001}\n",
        "\n",
        "0.6966666666666667`  \n",
        "所以`gamma=0.015 nu=0.57`即为最优参数"
      ],
      "metadata": {
        "id": "qjvCFyjYnh61"
      }
    },
    {
      "cell_type": "markdown",
      "source": [],
      "metadata": {
        "id": "Oq0YH5_-iUlT"
      }
    }
  ]
}