Day_047_hyper_parameter_tunning.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import datasets, metrics\n",
    "from sklearn.model_selection import train_test_split, KFold, GridSearchCV\n",
    "from sklearn.ensemble import GradientBoostingRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 讀取波士頓房價資料集\n",
    "boston = datasets.load_boston()\n",
    "# 切分訓練集/測試集\n",
    "x_train, x_test, y_train, y_test = train_test_split(boston.data, boston.target, test_size=0.25, random_state=42)\n",
    "\n",
    "# 建立模型\n",
    "clf = GradientBoostingRegressor(random_state=7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "506"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "boston.data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "8.913775994322062\n"
     ]
    }
   ],
   "source": [
    "# 先看看使用預設參數得到的結果，約為 8.379 的 MSE\n",
    "clf.fit(x_train, y_train)\n",
    "y_pred = clf.predict(x_test)\n",
    "print(metrics.mean_squared_error(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 9 candidates, totalling 45 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 4 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  45 out of  45 | elapsed:    5.1s finished\n"
     ]
    }
   ],
   "source": [
    "# 設定要訓練的超參數組合\n",
    "n_estimators = [100, 200, 300]\n",
    "max_depth = [1, 3, 5]\n",
    "param_grid = dict(n_estimators=n_estimators, max_depth=max_depth)\n",
    "\n",
    "## 建立搜尋物件，放入模型及參數組合字典 (n_jobs=-1 會使用全部 cpu 平行運算)\n",
    "grid_search = GridSearchCV(clf, param_grid, scoring=\"neg_mean_squared_error\", n_jobs=-1, verbose=1)\n",
    "\n",
    "# 開始搜尋最佳參數\n",
    "grid_result = grid_search.fit(x_train, y_train)\n",
    "\n",
    "# 預設會跑 3-fold cross-validadtion，總共 9 種參數組合，總共要 train 27 次模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Accuracy: -10.843075 using {'max_depth': 3, 'n_estimators': 100}\n"
     ]
    }
   ],
   "source": [
    "# 印出最佳結果與最佳參數\n",
    "print(\"Best Accuracy: %f using %s\" % (grid_result.best_score_, grid_result.best_params_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'max_depth': 3, 'n_estimators': 100}"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid_result.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 使用最佳參數重新建立模型\n",
    "clf_bestparam = GradientBoostingRegressor(max_depth=grid_result.best_params_['max_depth'],\n",
    "                                           n_estimators=grid_result.best_params_['n_estimators'])\n",
    "\n",
    "# 訓練模型\n",
    "clf_bestparam.fit(x_train, y_train)\n",
    "\n",
    "# 預測測試集\n",
    "y_pred = clf_bestparam.predict(x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "8.640384158903457\n"
     ]
    }
   ],
   "source": [
    "# 調整參數後約可降至 8.30 的 MSE\n",
    "print(metrics.mean_squared_error(y_test, y_pred))"
   ]
  },
  {
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}