Regression.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Lasso回归预测"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.linear_model import Lasso\n",
    "\n",
    "data = pd.read_csv('data.csv')\n",
    "\n",
    "import warnings\n",
    "\n",
    "def ignore_warn(*arfs, **kwargs):\n",
    "    pass\n",
    "\n",
    "warnings.warn = ignore_warn  # 忽略无意义的警告\n",
    "\n",
    "l_train = len(data[data['SalePrice'].notnull()])  # 训练集长度\n",
    "train = data[:l_train]  # 训练集\n",
    "y = train['SalePrice']  # 预测目标\n",
    "X = train.drop('SalePrice', axis=1).values  # 特征向量"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "使用交叉验证法计算精度。不同于分类问题，回归问题，我们通过计算预测房价与实际房价的误差平方的平均值（偏差）来表征误差。参考 Mean squared error\n",
    "\n",
    "Lasso 算法重要的参数是正则项系数，这里选择 0.0005 时误差较低。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "偏差: 0.0121 (0.0013)\n"
     ]
    }
   ],
   "source": [
    "def scoring(model):  # 定义函数 scoring，利用5折交叉验证法计算测试误差\n",
    "    r = cross_val_score(model, X, y, scoring=\"neg_mean_squared_error\", cv=5)\n",
    "    score = -r\n",
    "    return(score)\n",
    "\n",
    "clf = Lasso(alpha=0.0005)  # 参数设置\n",
    "score = scoring(clf)  # 调用 scoring 函数计算回归偏差\n",
    "print(\"偏差: {:.4f} ({:.4f})\".format(score.mean(), score.std()))\n",
    "# 交叉验证法偏差平均值及标准差"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "sklearn 提供 coef_ 函数返回特征的权重系数。我们使用这个函数观察嵌入式选择后，还有多少特征保留下来。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "特征总数：367\n",
      "嵌入式选择后，保留特征数：120\n"
     ]
    }
   ],
   "source": [
    "clf = Lasso(alpha=0.0005)  # 参数设置\n",
    "clf.fit(X, y)  # 训练模型\n",
    "print('特征总数：%d' % len(data.columns))\n",
    "print('嵌入式选择后，保留特征数：%d' % np.sum(clf.coef_ != 0))  # 计算并显示嵌入式选择后，保留的特征数。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "用学习曲线来观察学习器性能"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEWCAYAAACXGLsWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x102a5abe0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.model_selection import learning_curve\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "def plot_learning_curve(estimator, title, X, y, cv=10,\n",
    "                        train_sizes=np.linspace(.1, 1.0, 5)):  # 定义 plot_learning_curve 函数绘制学习曲线\n",
    "    plt.figure()\n",
    "    plt.title(title)  # 图片标题\n",
    "    plt.xlabel('Training examples')  # 横坐标\n",
    "    plt.ylabel('Score')  # 纵坐标\n",
    "    train_sizes, train_scores, test_scores = learning_curve(estimator, X, y, cv=cv, scoring=\"neg_mean_squared_error\",\n",
    "                                                            train_sizes=train_sizes)  # 交叉验证法计算训练误差，测试误差\n",
    "    train_scores_mean = np.mean(-train_scores, axis=1)  # 计算训练误差平均值\n",
    "    train_scores_std = np.std(-train_scores, axis=1)  # 训练误差方差\n",
    "    test_scores_mean = np.mean(-test_scores, axis=1)  # 测试误差平均值\n",
    "    test_scores_std = np.std(-test_scores, axis=1)  # 测试误差方差\n",
    "    plt.grid()  # 增加网格\n",
    "\n",
    "    plt.fill_between(train_sizes, train_scores_mean - train_scores_std,\n",
    "                     train_scores_mean + train_scores_std,\n",
    "                     alpha=0.1, color='g')  # 颜色填充\n",
    "    plt.fill_between(train_sizes, test_scores_mean - test_scores_std,\n",
    "                     test_scores_mean + test_scores_std,\n",
    "                     alpha=0.1, color='r')  # 颜色填充\n",
    "    plt.plot(train_sizes, train_scores_mean, 'o-', color='g',\n",
    "             label='traning score')  # 绘制训练误差曲线\n",
    "    plt.plot(train_sizes, test_scores_mean, 'o-', color='r',\n",
    "             label='testing score')  # 绘制测试误差曲线\n",
    "    plt.legend(loc='best')\n",
    "    return plt\n",
    "\n",
    "clf = Lasso(alpha=0.0005)\n",
    "g = plot_learning_curve(clf, 'Lasso', X, y)  # 调用 plot_learning_curve 绘制学习曲线"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "进行预测"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "clf = Lasso(alpha=0.0005)  # 参数设置\n",
    "clf.fit(X, y)\n",
    "test = data[l_train:].drop('SalePrice', axis=1).values  # 测试集数据\n",
    "predict = np.exp(clf.predict(test))  # 预测\n",
    "resul = pd.DataFrame()\n",
    "resul['SalePrice'] = predict\n",
    "resul.to_csv('submission.csv', index=False)  # 将结果写入 submission.csv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SalePrice</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>121481.128294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>155375.149059</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>185628.353482</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>199841.767030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>196858.642644</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>170424.175271</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>178996.402624</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>161678.140894</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>189810.889093</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>117325.356936</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>205292.113572</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>96925.791690</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>92606.160768</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>144357.851972</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>114028.578514</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>371088.443782</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>250878.095662</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>274572.145009</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>274429.908476</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>475775.616533</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>320294.812479</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>208277.225936</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>181830.441581</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>161200.039783</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>185943.397495</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>195701.282774</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>332231.922810</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>235727.817049</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>198701.106361</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>242593.034312</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1429</th>\n",
       "      <td>83720.850636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1430</th>\n",
       "      <td>138064.545818</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1431</th>\n",
       "      <td>44611.087821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1432</th>\n",
       "      <td>68528.145421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1433</th>\n",
       "      <td>55593.655873</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1434</th>\n",
       "      <td>318907.095473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1435</th>\n",
       "      <td>276272.123419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1436</th>\n",
       "      <td>208564.946544</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1437</th>\n",
       "      <td>148419.142667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1438</th>\n",
       "      <td>204561.896490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1439</th>\n",
       "      <td>154210.808114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1440</th>\n",
       "      <td>208789.324463</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1441</th>\n",
       "      <td>184357.355709</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1442</th>\n",
       "      <td>324132.802128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1443</th>\n",
       "      <td>334882.298097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1444</th>\n",
       "      <td>83451.158063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1445</th>\n",
       "      <td>191048.337634</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1446</th>\n",
       "      <td>108335.143373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1447</th>\n",
       "      <td>128638.159244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1448</th>\n",
       "      <td>152194.809762</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1449</th>\n",
       "      <td>80471.896793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1450</th>\n",
       "      <td>81091.184511</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1451</th>\n",
       "      <td>144276.166019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1452</th>\n",
       "      <td>78767.218736</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1453</th>\n",
       "      <td>78786.860532</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1454</th>\n",
       "      <td>87752.688321</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1455</th>\n",
       "      <td>79773.574929</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1456</th>\n",
       "      <td>163530.859859</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1457</th>\n",
       "      <td>118462.460432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1458</th>\n",
       "      <td>225774.125584</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1459 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          SalePrice\n",
       "0     121481.128294\n",
       "1     155375.149059\n",
       "2     185628.353482\n",
       "3     199841.767030\n",
       "4     196858.642644\n",
       "5     170424.175271\n",
       "6     178996.402624\n",
       "7     161678.140894\n",
       "8     189810.889093\n",
       "9     117325.356936\n",
       "10    205292.113572\n",
       "11     96925.791690\n",
       "12     92606.160768\n",
       "13    144357.851972\n",
       "14    114028.578514\n",
       "15    371088.443782\n",
       "16    250878.095662\n",
       "17    274572.145009\n",
       "18    274429.908476\n",
       "19    475775.616533\n",
       "20    320294.812479\n",
       "21    208277.225936\n",
       "22    181830.441581\n",
       "23    161200.039783\n",
       "24    185943.397495\n",
       "25    195701.282774\n",
       "26    332231.922810\n",
       "27    235727.817049\n",
       "28    198701.106361\n",
       "29    242593.034312\n",
       "...             ...\n",
       "1429   83720.850636\n",
       "1430  138064.545818\n",
       "1431   44611.087821\n",
       "1432   68528.145421\n",
       "1433   55593.655873\n",
       "1434  318907.095473\n",
       "1435  276272.123419\n",
       "1436  208564.946544\n",
       "1437  148419.142667\n",
       "1438  204561.896490\n",
       "1439  154210.808114\n",
       "1440  208789.324463\n",
       "1441  184357.355709\n",
       "1442  324132.802128\n",
       "1443  334882.298097\n",
       "1444   83451.158063\n",
       "1445  191048.337634\n",
       "1446  108335.143373\n",
       "1447  128638.159244\n",
       "1448  152194.809762\n",
       "1449   80471.896793\n",
       "1450   81091.184511\n",
       "1451  144276.166019\n",
       "1452   78767.218736\n",
       "1453   78786.860532\n",
       "1454   87752.688321\n",
       "1455   79773.574929\n",
       "1456  163530.859859\n",
       "1457  118462.460432\n",
       "1458  225774.125584\n",
       "\n",
       "[1459 rows x 1 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 查看 submission.csv\n",
    "pd.read_csv('submission.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}