diff --git a/机器学习竞赛实战_优胜解决方案/建筑能源利用率预测/.ipynb_checkpoints/3_模型解释与分析_建筑能源利用率预测-checkpoint.ipynb b/机器学习竞赛实战_优胜解决方案/建筑能源利用率预测/.ipynb_checkpoints/3_模型解释与分析_建筑能源利用率预测-checkpoint.ipynb
new file mode 100644
index 0000000..b032e27
--- /dev/null
+++ b/机器学习竞赛实战_优胜解决方案/建筑能源利用率预测/.ipynb_checkpoints/3_模型解释与分析_建筑能源利用率预测-checkpoint.ipynb
@@ -0,0 +1,439 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 理解模型"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "\n",
+    "pd.options.mode.chained_assignment = None\n",
+    "pd.set_option('display.max_columns', 60)\n",
+    "\n",
+    "import matplotlib.pyplot as plt\n",
+    "%matplotlib inline\n",
+    "\n",
+    "plt.rcParams['font.size'] = 24\n",
+    "\n",
+    "from IPython.core.pylabtools import figsize\n",
+    "\n",
+    "import seaborn as sns\n",
+    "sns.set(font_scale = 2)\n",
+    "\n",
+    "from sklearn.preprocessing import Imputer, MinMaxScaler\n",
+    "\n",
+    "from sklearn.linear_model import LinearRegression\n",
+    "from sklearn.ensemble import GradientBoostingRegressor\n",
+    "\n",
+    "from sklearn import tree\n",
+    "\n",
+    "# LIME用于解释预测\n",
+    "import lime  # pip install lime\n",
+    "import lime.lime_tabular"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Read in data into dataframes \n",
+    "train_features = pd.read_csv('data/training_features.csv')\n",
+    "test_features = pd.read_csv('data/testing_features.csv')\n",
+    "train_labels = pd.read_csv('data/training_labels.csv')\n",
+    "test_labels = pd.read_csv('data/testing_labels.csv')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "D:\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\deprecation.py:66: DeprecationWarning: Class Imputer is deprecated; Imputer was deprecated in version 0.20 and will be removed in 0.22. Import impute.SimpleImputer from sklearn instead.\n",
+      "  warnings.warn(msg, category=DeprecationWarning)\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Create an imputer object with a median filling strategy\n",
+    "imputer = Imputer(strategy = 'median')\n",
+    "\n",
+    "# Train on the training features\n",
+    "imputer.fit(train_features)\n",
+    "\n",
+    "# Transform both training data and testing data\n",
+    "X = imputer.transform(train_features)\n",
+    "X_test = imputer.transform(test_features)\n",
+    "\n",
+    "# Convert y to one-dimensional array (vector)\n",
+    "y = np.array(train_labels).reshape((-1, ))\n",
+    "y_test = np.array(test_labels).reshape((-1, ))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Function to calculation mean absolute error\n",
+    "def mae (y_true,y_pred):\n",
+    "    return np.mean(abs(y_true - y_pred))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "GradientBoostingRegressor(alpha=0.9, criterion='friedman_mse', init=None,\n",
+       "                          learning_rate=0.1, loss='lad', max_depth=5,\n",
+       "                          max_features=None, max_leaf_nodes=None,\n",
+       "                          min_impurity_decrease=0.0, min_impurity_split=None,\n",
+       "                          min_samples_leaf=6, min_samples_split=6,\n",
+       "                          min_weight_fraction_leaf=0.0, n_estimators=800,\n",
+       "                          n_iter_no_change=None, presort='auto',\n",
+       "                          random_state=42, subsample=1.0, tol=0.0001,\n",
+       "                          validation_fraction=0.1, verbose=0, warm_start=False)"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "model = GradientBoostingRegressor(loss='lad', max_depth=5, max_features=None,\n",
+    "                                  min_samples_leaf=6, min_samples_split=6, \n",
+    "                                  n_estimators=800, random_state=42)\n",
+    "\n",
+    "model.fit(X, y)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 解释模型\n",
+    "* 特征重要性\n",
+    "* Locally Interpretable Model-agnostic Explainer(LIME)\n",
+    "* 建立一棵树模"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 特征重要性"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>feature</th>\n",
+       "      <th>importance</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Site EUI (kBtu/ft²)</td>\n",
+       "      <td>0.452163</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Weather Normalized Site Electricity Intensity ...</td>\n",
+       "      <td>0.249107</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Water Intensity (All Water Sources) (gal/ft²)</td>\n",
+       "      <td>0.056662</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>Property Id</td>\n",
+       "      <td>0.031396</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Largest Property Use Type_Non-Refrigerated War...</td>\n",
+       "      <td>0.025153</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>DOF Gross Floor Area</td>\n",
+       "      <td>0.025003</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td>log_Water Intensity (All Water Sources) (gal/ft²)</td>\n",
+       "      <td>0.022335</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
+       "      <td>Largest Property Use Type_Multifamily Housing</td>\n",
+       "      <td>0.021462</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8</th>\n",
+       "      <td>Order</td>\n",
+       "      <td>0.020169</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9</th>\n",
+       "      <td>log_Direct GHG Emissions (Metric Tons CO2e)</td>\n",
+       "      <td>0.019410</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                             feature  importance\n",
+       "0                                Site EUI (kBtu/ft²)    0.452163\n",
+       "1  Weather Normalized Site Electricity Intensity ...    0.249107\n",
+       "2      Water Intensity (All Water Sources) (gal/ft²)    0.056662\n",
+       "3                                        Property Id    0.031396\n",
+       "4  Largest Property Use Type_Non-Refrigerated War...    0.025153\n",
+       "5                               DOF Gross Floor Area    0.025003\n",
+       "6  log_Water Intensity (All Water Sources) (gal/ft²)    0.022335\n",
+       "7      Largest Property Use Type_Multifamily Housing    0.021462\n",
+       "8                                              Order    0.020169\n",
+       "9        log_Direct GHG Emissions (Metric Tons CO2e)    0.019410"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Extract the feature importances into a dataframe\n",
+    "feature_results = pd.DataFrame({'feature': list(train_features.columns), \n",
+    "                                'importance': model.feature_importances_})\n",
+    "\n",
+    "# Show the top 10 most important\n",
+    "feature_results = feature_results.sort_values('importance', ascending = False).reset_index(drop=True)\n",
+    "\n",
+    "feature_results.head(10)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x720 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# 画图\n",
+    "figsize(12, 10)\n",
+    "plt.style.use('fivethirtyeight')\n",
+    "\n",
+    "# Plot the 10 most important features in a horizontal bar chart\n",
+    "feature_results.loc[:9, :].plot(x = 'feature', y = 'importance', \n",
+    "                                 edgecolor = 'k',\n",
+    "                                 kind='barh', color = 'blue');\n",
+    "plt.xlabel('Relative Importance', size = 20); plt.ylabel('')\n",
+    "plt.title('Feature Importances from Random Forest', size = 30);"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 基于特征重要性进行选择\n",
+    "拿最重要的部分特征来建模"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Most important training features shape:  (6622, 10)\n",
+      "Most important testing  features shape:  (2839, 10)\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Extract the names of the most important features\n",
+    "most_important_features = feature_results['feature'][:10]\n",
+    "\n",
+    "# Find the index that corresponds to each feature name\n",
+    "indices = [list(train_features.columns).index(x) for x in most_important_features]\n",
+    "\n",
+    "# Keep only the most important features\n",
+    "X_reduced = X[:, indices]\n",
+    "X_test_reduced = X_test[:, indices]\n",
+    "\n",
+    "print('Most important training features shape: ', X_reduced.shape)\n",
+    "print('Most important testing  features shape: ', X_test_reduced.shape)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Linear Regression Full Results: MAE =    13.4651.\n",
+      "Linear Regression Reduced Results: MAE = 14.5095.\n"
+     ]
+    }
+   ],
+   "source": [
+    "lr = LinearRegression()\n",
+    "\n",
+    "# Fit on full set of features\n",
+    "lr.fit(X, y)\n",
+    "lr_full_pred = lr.predict(X_test)\n",
+    "\n",
+    "# Fit on reduced set of features\n",
+    "lr.fit(X_reduced, y)\n",
+    "lr_reduced_pred = lr.predict(X_test_reduced)\n",
+    "\n",
+    "# Display results\n",
+    "print('Linear Regression Full Results: MAE = %0.4f.' % mae(y_test, lr_full_pred))\n",
+    "print('Linear Regression Reduced Results: MAE = %0.4f.' % mae(y_test, lr_reduced_pred))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "结果表明，试验结果不如原来的"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Linear Regression Full Results: MAE = 9.0839.\n",
+      "Gradient Boosted Reduced Results: MAE = 9.3735\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Create the model with the same hyperparamters\n",
+    "model_gbdt = GradientBoostingRegressor(loss='lad', max_depth=5, max_features=None,\n",
+    "                                  min_samples_leaf=6, min_samples_split=6, \n",
+    "                                  n_estimators=800, random_state=42)\n",
+    "\n",
+    "model_gbdt.fit(X, y)\n",
+    "model_full_pred = model_gbdt.predict(X_test)\n",
+    "\n",
+    "# Fit and test on the reduced set of features\n",
+    "model_gbdt.fit(X_reduced, y)\n",
+    "model_reduced_pred = model_gbdt.predict(X_test_reduced)\n",
+    "\n",
+    "print('Gradient Boosted Full Results: MAE = %0.4f.' % mae(y_test, model_full_pred))\n",
+    "print('Gradient Boosted Reduced Results: MAE = %0.4f' % mae(y_test, model_reduced_pred))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "GBDT出现同样的问题，最终选择使用全部特征。"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Locally Interpretable Model-agnostic Explanations"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "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": "3.7.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}