MachingLearning/SVM/SVM.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 实例化一个朴素的 SVM 分类器，并看下其准确率"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import datasets\n",
    "from sklearn.svm import SVC\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "digits = datasets.load_digits()\n",
    "\n",
    "X = digits.data\n",
    "y = digits.target"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 显示数据集的几个数字样例"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_digits(digits):\n",
    "    fig, axes = plt.subplots(2, 5, figsize=(10,10),\n",
    "                             subplot_kw={'xticks':[], 'yticks':[]},\n",
    "                             gridspec_kw=dict(hspace=0.1, wspace=0.1))\n",
    "    for i, ax in enumerate(axes.flat):\n",
    "        ax.imshow(digits[i].reshape(8, 8), cmap = 'bone')\n",
    "    plt.show()\n",
    "    \n",
    "    \n",
    "# 可视化几个数字样例\n",
    "example_digits = X[:10]\n",
    "plot_digits(example_digits)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### SVM重要超参数\n",
    "\n",
    "参数  |  说明 \n",
    ":-: | :-\n",
    "**penalty**|字符串，可选'l1'或'l2'，默认'l2'，指定模型的正则方式\n",
    "**loss**| 字符串，可选'hinge'或'squared_hinge'，默认'squared_hinge'，用于指定模型的损失函数\n",
    "**kenel**| 字符串，可选'linear','poly','rbf','sigmoid','precomputed\n",
    "**degree**| 整型数字，当使用多项式核时，用来确定多项式的阶次\n",
    "**dual**| 布尔值，默认值为'True'，选择算法以解决双优化或原始优化问题\n",
    "**tol**| 浮点数，默认为 1e-4，用于判断是否停止迭代的容差\n",
    "**C**| 浮点数， 默认为1.0，容错空间系数，用于调整容错空间在优化迭代中所占的重要性\n",
    "**multi_class**| 字符串，可选'ovr'和'crammer_singer'，但面临多分类问题时，用于确定多分类策略，'ovr'指定了使用 One VS Rest 策略进行多分类任务,而'crammer_singer'则是在所有的类上建立一个联合的目标损失函数进行优化\n",
    "**verbose**| 整型数字，默认值为0，若为大于0的整数，则会在训练过程中不断输入与训练相关的条件与参数\n",
    "**max_iter**| 整型数字，默认值为1000，用于指定迭代的最大次数"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 调参前的SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state = 233)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  0.39111111111111113\n",
      "Wall time: 321 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "svc = SVC()\n",
    "svc.fit(X_train, y_train)\n",
    "print('Accuracy: ', svc.score(X_test, y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 调参后的SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  0.9911111111111112\n",
      "Wall time: 99 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "svc2 = SVC(C=10000, kernel='rbf', gamma=0.001)\n",
    "svc2.fit(X_train, y_train)\n",
    "print('Accuracy: ', svc2.score(X_test, y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 与其他分类器性能比较\n",
    "* KNN\n",
    "* LogisticRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  0.9866666666666667\n",
      "Wall time: 20.2 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "# KNN\n",
    "\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "param_grid = [\n",
    "    {\n",
    "        'weights': ['uniform'],\n",
    "        'n_neighbors': [i for i in range(3, 11)],\n",
    "        'p': [i for i in range(1, 4)]\n",
    "    },\n",
    "    {\n",
    "        'weights': ['distance'],\n",
    "        'n_neighbors': [i for i in range(3, 11)],\n",
    "        'p': [i for i in range(1, 4)]\n",
    "    }\n",
    "]\n",
    "\n",
    "knn_gs_clf = GridSearchCV(KNeighborsClassifier(), param_grid=param_grid, n_jobs=-1)\n",
    "knn_gs_clf.fit(X_train, y_train)\n",
    "print('Accuracy: ', knn_gs_clf.score(X_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  0.9377777777777778\n",
      "Wall time: 132 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "# LogisticRegression\n",
    "\n",
    "log_clf = LogisticRegression()\n",
    "log_clf.fit(X_train, y_train)\n",
    "print('Accuracy: ', log_clf.score(X_test, y_test))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": false,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}