ARIMA.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing the required libraries\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from IPython.display import clear_output,display\n",
    "\n",
    "from sklearn.model_selection import cross_val_score,GridSearchCV,RandomizedSearchCV\n",
    "from sklearn.metrics import mean_squared_log_error,mean_squared_error\n",
    "from sklearn.linear_model import LinearRegression,Lasso,Ridge\n",
    "\n",
    "from sklearn.preprocessing import StandardScaler,MinMaxScaler\n",
    "\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.ensemble import RandomForestRegressor,AdaBoostRegressor,BaggingRegressor\n",
    "\n",
    "from xgboost import XGBRegressor\n",
    "from catboost import CatBoostRegressor\n",
    "from lightgbm import LGBMRegressor\n",
    "\n",
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# importing the train and test datasets\n",
    "\n",
    "tr=pd.read_csv(\"train.csv\")\n",
    "test=pd.read_csv(\"test.csv\")\n",
    "sample_sub=pd.read_csv(\"sample_submission.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>ID</th>\n",
       "      <th>Day_No</th>\n",
       "      <th>Sales</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512082</th>\n",
       "      <td>548023</td>\n",
       "      <td>878</td>\n",
       "      <td>114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512083</th>\n",
       "      <td>548024</td>\n",
       "      <td>879</td>\n",
       "      <td>145</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512084</th>\n",
       "      <td>548025</td>\n",
       "      <td>880</td>\n",
       "      <td>167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512085</th>\n",
       "      <td>548026</td>\n",
       "      <td>881</td>\n",
       "      <td>107</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512086</th>\n",
       "      <td>548027</td>\n",
       "      <td>882</td>\n",
       "      <td>152</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>512087 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            ID  Day_No  Sales\n",
       "0            1       1     81\n",
       "1            2       2     79\n",
       "2            3       3     75\n",
       "3            4       4     80\n",
       "4            5       5     41\n",
       "...        ...     ...    ...\n",
       "512082  548023     878    114\n",
       "512083  548024     879    145\n",
       "512084  548025     880    167\n",
       "512085  548026     881    107\n",
       "512086  548027     882    152\n",
       "\n",
       "[512087 rows x 3 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tr=tr[[\"ID\",\"Day_No\",\"Sales\"]]\n",
    "tr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Day_No\n",
       "1       78603\n",
       "2       67389\n",
       "3       61725\n",
       "4       64464\n",
       "5       59094\n",
       "        ...  \n",
       "878     67278\n",
       "879     80955\n",
       "880     74450\n",
       "881     58999\n",
       "882    114140\n",
       "Name: Sales, Length: 882, dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train=tr.groupby(\"Day_No\").sum()[\"Sales\"]\n",
    "train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "train,val=train[train.index<=822],train[train.index>822]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Day_No\n",
       "1       78603\n",
       "2       67389\n",
       "3       61725\n",
       "4       64464\n",
       "5       59094\n",
       "        ...  \n",
       "818     55825\n",
       "819    121131\n",
       "820    109930\n",
       "821    108359\n",
       "822    111420\n",
       "Name: Sales, Length: 822, dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Day_No\n",
       "823     72473\n",
       "824     92689\n",
       "825     86765\n",
       "826     19841\n",
       "827     76109\n",
       "828     65411\n",
       "829     63288\n",
       "830     64181\n",
       "831     63316\n",
       "832     53589\n",
       "833    107470\n",
       "834     88302\n",
       "835     77462\n",
       "836     81115\n",
       "837     84558\n",
       "838     64261\n",
       "839     73982\n",
       "840     63139\n",
       "841     57717\n",
       "842     58373\n",
       "843     59251\n",
       "844     61208\n",
       "845     62124\n",
       "846     53817\n",
       "847    103197\n",
       "848     90322\n",
       "849     92394\n",
       "850    110168\n",
       "851     65563\n",
       "852     86302\n",
       "853     77250\n",
       "854    106799\n",
       "855     84955\n",
       "856     84031\n",
       "857     82790\n",
       "858     83976\n",
       "859     70517\n",
       "860     77046\n",
       "861     64724\n",
       "862     64257\n",
       "863     79071\n",
       "864     56357\n",
       "865     83557\n",
       "866     63641\n",
       "867     54667\n",
       "868    103094\n",
       "869     86248\n",
       "870     80920\n",
       "871     79551\n",
       "872     80478\n",
       "873     66427\n",
       "874     74019\n",
       "875     63190\n",
       "876     75119\n",
       "877     66405\n",
       "878     67278\n",
       "879     80955\n",
       "880     74450\n",
       "881     58999\n",
       "882    114140\n",
       "Name: Sales, dtype: int64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "val"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Stationarity test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# dickey fuller, KPSS\n",
    "from statsmodels.tsa.stattools import adfuller, kpss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "def adf_test(timeseries):\n",
    "    \n",
    "    #Perform Dickey-Fuller test:\n",
    "    print ('Results of Dickey-Fuller Test:')\n",
    "    dftest = adfuller(timeseries, autolag='AIC')\n",
    "    dfoutput=pd.Series(dftest[0:4], index=['Test Statistic','p-value','#Lags Used','Number of Observations Used'])\n",
    "\n",
    "    for key,value in dftest[4].items():\n",
    "        dfoutput['Critical Value (%s)'%key] = value\n",
    "    print (dfoutput)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Results of Dickey-Fuller Test:\n",
      "Test Statistic                  -4.465763\n",
      "p-value                          0.000227\n",
      "#Lags Used                      20.000000\n",
      "Number of Observations Used    801.000000\n",
      "Critical Value (1%)             -3.438540\n",
      "Critical Value (5%)             -2.865155\n",
      "Critical Value (10%)            -2.568695\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "adf_test(train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Our series is Stationary**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "def kpss_test(timeseries):\n",
    "    print ('Results of KPSS Test:')\n",
    "    kpsstest = kpss(timeseries, regression='c')\n",
    "    kpss_output = pd.Series(kpsstest[0:3], index=['Test Statistic','p-value','Lags Used'])\n",
    "    for key,value in kpsstest[3].items():\n",
    "        kpss_output['Critical Value (%s)'%key] = value\n",
    "    print (kpss_output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Results of KPSS Test:\n",
      "Test Statistic            0.199919\n",
      "p-value                   0.100000\n",
      "Lags Used                21.000000\n",
      "Critical Value (10%)      0.347000\n",
      "Critical Value (5%)       0.463000\n",
      "Critical Value (2.5%)     0.574000\n",
      "Critical Value (1%)       0.739000\n",
      "dtype: float64\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\statsmodels\\tsa\\stattools.py:1843: FutureWarning: The behavior of using nlags=None will change in release 0.13.Currently nlags=None is the same as nlags=\"legacy\", and so a sample-size lag length is used. After the next release, the default will change to be the same as nlags=\"auto\" which uses an automatic lag length selection method. To silence this warning, either use \"auto\" or \"legacy\"\n",
      "  warnings.warn(msg, FutureWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\statsmodels\\tsa\\stattools.py:1876: InterpolationWarning: The test statistic is outside of the range of p-values available in the\n",
      "look-up table. The actual p-value is greater than the p-value returned.\n",
      "\n",
      "  warnings.warn(warn_msg.format(direction=\"greater\"), InterpolationWarning)\n"
     ]
    }
   ],
   "source": [
    "kpss_test(train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ACF and PACF plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "from statsmodels.tsa.stattools import acf, pacf "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\statsmodels\\tsa\\stattools.py:666: FutureWarning: fft=True will become the default after the release of the 0.12 release of statsmodels. To suppress this warning, explicitly set fft=False.\n",
      "  FutureWarning,\n"
     ]
    }
   ],
   "source": [
    "lag_acf = acf(train, nlags=25) \n",
    "lag_pacf = pacf(train, nlags=25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(lag_acf,  marker='o') \n",
    "plt.axhline(y=0,linestyle='--',color='gray') \n",
    "plt.axhline(y=-1.96/np.sqrt(len(train)),linestyle='--',color='gray') \n",
    "plt.axhline(y=1.96/np.sqrt(len(train)),linestyle='--',color='gray') \n",
    "plt.title('Autocorrelation Function') \n",
    "plt.show() \n",
    "\n",
    "plt.plot(lag_pacf, marker='o') \n",
    "plt.axhline(y=0,linestyle='--',color='gray') \n",
    "plt.axhline(y=-1.96/np.sqrt(len(train)),linestyle='--',color='gray') \n",
    "plt.axhline(y=1.96/np.sqrt(len(train)),linestyle='--',color='gray') \n",
    "plt.title('Partial Autocorrelation Function') \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## AR Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [],
   "source": [
    "from statsmodels.tsa.arima_model import ARIMA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [],
   "source": [
    "# fit AR model\n",
    "model = ARIMA(train, order=(6,0,10))\n",
    "model_fit= model.fit(disp=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [],
   "source": [
    "output = model_fit.forecast(len(val))\n",
    "pred_valid = (pd.DataFrame(output[0])).values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(14,6))\n",
    "\n",
    "plt.plot(train[600:].index, train[600:], label='train_data')\n",
    "plt.plot(val.index, val, label='valid')\n",
    "plt.plot(val.index, pred_valid, label='predicted')\n",
    "\n",
    "plt.legend(loc='best')\n",
    "plt.title(\"AR mode\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The RMSE value for AR is 16179.368380266495\n"
     ]
    }
   ],
   "source": [
    "# calculating RMSE \n",
    "from math import sqrt\n",
    "rmse = sqrt(mean_squared_error(val, pred_valid))\n",
    "print('The RMSE value for AR is', rmse)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>ARMA Model Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>       <td>Sales</td>      <th>  No. Observations:  </th>    <td>822</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>            <td>ARMA(6, 10)</td>   <th>  Log Likelihood     </th> <td>-8962.557</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>css-mle</td>     <th>  S.D. of innovations</th> <td>13015.404</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>          <td>Mon, 19 Oct 2020</td> <th>  AIC                </th> <td>17961.114</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>              <td>15:24:29</td>     <th>  BIC                </th> <td>18045.926</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>                <td>0</td>        <th>  HQIC               </th> <td>17993.653</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                       <td> </td>        <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "        <td></td>          <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th>        <td> 6.979e+04</td> <td> 1377.382</td> <td>   50.671</td> <td> 0.000</td> <td> 6.71e+04</td> <td> 7.25e+04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L1.Sales</th>  <td>    1.1927</td> <td>  2.5e-05</td> <td> 4.78e+04</td> <td> 0.000</td> <td>    1.193</td> <td>    1.193</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L2.Sales</th>  <td>   -0.1863</td> <td>      nan</td> <td>      nan</td> <td>   nan</td> <td>      nan</td> <td>      nan</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L3.Sales</th>  <td>    0.1677</td> <td>      nan</td> <td>      nan</td> <td>   nan</td> <td>      nan</td> <td>      nan</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L4.Sales</th>  <td>    0.1870</td> <td>    0.001</td> <td>  298.953</td> <td> 0.000</td> <td>    0.186</td> <td>    0.188</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L5.Sales</th>  <td>   -1.1642</td> <td> 6.86e-06</td> <td> -1.7e+05</td> <td> 0.000</td> <td>   -1.164</td> <td>   -1.164</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L6.Sales</th>  <td>    0.7723</td> <td> 3.79e-05</td> <td> 2.04e+04</td> <td> 0.000</td> <td>    0.772</td> <td>    0.772</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L1.Sales</th>  <td>   -0.9501</td> <td>    0.033</td> <td>  -28.552</td> <td> 0.000</td> <td>   -1.015</td> <td>   -0.885</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L2.Sales</th>  <td>    0.2298</td> <td>    0.045</td> <td>    5.121</td> <td> 0.000</td> <td>    0.142</td> <td>    0.318</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L3.Sales</th>  <td>   -0.2496</td> <td>    0.044</td> <td>   -5.690</td> <td> 0.000</td> <td>   -0.336</td> <td>   -0.164</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L4.Sales</th>  <td>   -0.4104</td> <td>    0.047</td> <td>   -8.642</td> <td> 0.000</td> <td>   -0.503</td> <td>   -0.317</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L5.Sales</th>  <td>    1.2197</td> <td>    0.041</td> <td>   29.442</td> <td> 0.000</td> <td>    1.138</td> <td>    1.301</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L6.Sales</th>  <td>   -0.6644</td> <td>    0.043</td> <td>  -15.481</td> <td> 0.000</td> <td>   -0.749</td> <td>   -0.580</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L7.Sales</th>  <td>   -0.0839</td> <td>    0.046</td> <td>   -1.816</td> <td> 0.069</td> <td>   -0.175</td> <td>    0.007</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L8.Sales</th>  <td>    0.0306</td> <td>    0.044</td> <td>    0.693</td> <td> 0.488</td> <td>   -0.056</td> <td>    0.117</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L9.Sales</th>  <td>   -0.3282</td> <td>    0.045</td> <td>   -7.276</td> <td> 0.000</td> <td>   -0.417</td> <td>   -0.240</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L10.Sales</th> <td>    0.3017</td> <td>    0.034</td> <td>    8.977</td> <td> 0.000</td> <td>    0.236</td> <td>    0.368</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Roots</caption>\n",
       "<tr>\n",
       "    <td></td>    <th>            Real</th>  <th>         Imaginary</th> <th>         Modulus</th>  <th>        Frequency</th>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>AR.1</th>  <td>          -1.0416</td> <td>          -0.0000j</td> <td>           1.0416</td> <td>          -0.5000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>AR.2</th>  <td>          -0.2224</td> <td>          -0.9750j</td> <td>           1.0000</td> <td>          -0.2857</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>AR.3</th>  <td>          -0.2224</td> <td>          +0.9750j</td> <td>           1.0000</td> <td>           0.2857</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>AR.4</th>  <td>           0.9805</td> <td>          -0.4921j</td> <td>           1.0971</td> <td>          -0.0740</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>AR.5</th>  <td>           0.9805</td> <td>          +0.4921j</td> <td>           1.0971</td> <td>           0.0740</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>AR.6</th>  <td>           1.0328</td> <td>          -0.0000j</td> <td>           1.0328</td> <td>          -0.0000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>MA.1</th>  <td>          -1.1466</td> <td>          -0.1476j</td> <td>           1.1561</td> <td>          -0.4796</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>MA.2</th>  <td>          -1.1466</td> <td>          +0.1476j</td> <td>           1.1561</td> <td>           0.4796</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>MA.3</th>  <td>          -0.2272</td> <td>          -0.9851j</td> <td>           1.0110</td> <td>          -0.2861</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>MA.4</th>  <td>          -0.2272</td> <td>          +0.9851j</td> <td>           1.0110</td> <td>           0.2861</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>MA.5</th>  <td>          -0.0990</td> <td>          -1.2589j</td> <td>           1.2627</td> <td>          -0.2625</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>MA.6</th>  <td>          -0.0990</td> <td>          +1.2589j</td> <td>           1.2627</td> <td>           0.2625</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>MA.7</th>  <td>           0.8482</td> <td>          -0.6316j</td> <td>           1.0576</td> <td>          -0.1019</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>MA.8</th>  <td>           0.8482</td> <td>          +0.6316j</td> <td>           1.0576</td> <td>           0.1019</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>MA.9</th>  <td>           1.0971</td> <td>          -0.0000j</td> <td>           1.0971</td> <td>          -0.0000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>MA.10</th> <td>           1.2402</td> <td>          -0.0000j</td> <td>           1.2402</td> <td>          -0.0000</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                              ARMA Model Results                              \n",
       "==============================================================================\n",
       "Dep. Variable:                  Sales   No. Observations:                  822\n",
       "Model:                    ARMA(6, 10)   Log Likelihood               -8962.557\n",
       "Method:                       css-mle   S.D. of innovations          13015.404\n",
       "Date:                Mon, 19 Oct 2020   AIC                          17961.114\n",
       "Time:                        15:24:29   BIC                          18045.926\n",
       "Sample:                             0   HQIC                         17993.653\n",
       "                                                                              \n",
       "================================================================================\n",
       "                   coef    std err          z      P>|z|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------\n",
       "const         6.979e+04   1377.382     50.671      0.000    6.71e+04    7.25e+04\n",
       "ar.L1.Sales      1.1927    2.5e-05   4.78e+04      0.000       1.193       1.193\n",
       "ar.L2.Sales     -0.1863        nan        nan        nan         nan         nan\n",
       "ar.L3.Sales      0.1677        nan        nan        nan         nan         nan\n",
       "ar.L4.Sales      0.1870      0.001    298.953      0.000       0.186       0.188\n",
       "ar.L5.Sales     -1.1642   6.86e-06   -1.7e+05      0.000      -1.164      -1.164\n",
       "ar.L6.Sales      0.7723   3.79e-05   2.04e+04      0.000       0.772       0.772\n",
       "ma.L1.Sales     -0.9501      0.033    -28.552      0.000      -1.015      -0.885\n",
       "ma.L2.Sales      0.2298      0.045      5.121      0.000       0.142       0.318\n",
       "ma.L3.Sales     -0.2496      0.044     -5.690      0.000      -0.336      -0.164\n",
       "ma.L4.Sales     -0.4104      0.047     -8.642      0.000      -0.503      -0.317\n",
       "ma.L5.Sales      1.2197      0.041     29.442      0.000       1.138       1.301\n",
       "ma.L6.Sales     -0.6644      0.043    -15.481      0.000      -0.749      -0.580\n",
       "ma.L7.Sales     -0.0839      0.046     -1.816      0.069      -0.175       0.007\n",
       "ma.L8.Sales      0.0306      0.044      0.693      0.488      -0.056       0.117\n",
       "ma.L9.Sales     -0.3282      0.045     -7.276      0.000      -0.417      -0.240\n",
       "ma.L10.Sales     0.3017      0.034      8.977      0.000       0.236       0.368\n",
       "                                    Roots                                     \n",
       "==============================================================================\n",
       "                   Real          Imaginary           Modulus         Frequency\n",
       "------------------------------------------------------------------------------\n",
       "AR.1            -1.0416           -0.0000j            1.0416           -0.5000\n",
       "AR.2            -0.2224           -0.9750j            1.0000           -0.2857\n",
       "AR.3            -0.2224           +0.9750j            1.0000            0.2857\n",
       "AR.4             0.9805           -0.4921j            1.0971           -0.0740\n",
       "AR.5             0.9805           +0.4921j            1.0971            0.0740\n",
       "AR.6             1.0328           -0.0000j            1.0328           -0.0000\n",
       "MA.1            -1.1466           -0.1476j            1.1561           -0.4796\n",
       "MA.2            -1.1466           +0.1476j            1.1561            0.4796\n",
       "MA.3            -0.2272           -0.9851j            1.0110           -0.2861\n",
       "MA.4            -0.2272           +0.9851j            1.0110            0.2861\n",
       "MA.5            -0.0990           -1.2589j            1.2627           -0.2625\n",
       "MA.6            -0.0990           +1.2589j            1.2627            0.2625\n",
       "MA.7             0.8482           -0.6316j            1.0576           -0.1019\n",
       "MA.8             0.8482           +0.6316j            1.0576            0.1019\n",
       "MA.9             1.0971           -0.0000j            1.0971           -0.0000\n",
       "MA.10            1.2402           -0.0000j            1.2402           -0.0000\n",
       "------------------------------------------------------------------------------\n",
       "\"\"\""
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_fit.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Seasonal ACF and PACF plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "from statsmodels.tsa.stattools import acf, pacf "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Day_No\n",
       "8       6213.0\n",
       "9       8208.0\n",
       "10     13815.0\n",
       "11     13839.0\n",
       "12       428.0\n",
       "        ...   \n",
       "818   -16315.0\n",
       "819    55632.0\n",
       "820    50211.0\n",
       "821    49903.0\n",
       "822    50101.0\n",
       "Name: Sales, Length: 815, dtype: float64"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diff_7=train-train.shift(7)\n",
    "diff_7.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [],
   "source": [
    "lag_acf = acf(diff_7.dropna(), nlags=25) \n",
    "lag_pacf = pacf(diff_7.dropna(), nlags=25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(lag_acf,  marker='o') \n",
    "plt.axhline(y=0,linestyle='--',color='gray') \n",
    "plt.axhline(y=-1.96/np.sqrt(len(diff_7.dropna())),linestyle='--',color='gray') \n",
    "plt.axhline(y=1.96/np.sqrt(len(diff_7.dropna())),linestyle='--',color='gray') \n",
    "plt.title('Autocorrelation Function') \n",
    "plt.show() \n",
    "\n",
    "plt.plot(lag_pacf, marker='o') \n",
    "plt.axhline(y=0,linestyle='--',color='gray') \n",
    "plt.axhline(y=-1.96/np.sqrt(len(diff_7.dropna())),linestyle='--',color='gray') \n",
    "plt.axhline(y=1.96/np.sqrt(len(diff_7.dropna())),linestyle='--',color='gray') \n",
    "plt.title('Partial Autocorrelation Function') \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## SARIMA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "from statsmodels.tsa.statespace import sarimax"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "Invalid model: moving average lag(s) {7} are in both the seasonal and non-seasonal moving average components.",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-116-62cf66f90fce>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mmodel\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0msarimax\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mSARIMAX\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtrain\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mseasonal_order\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m2\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m4\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m7\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0morder\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m6\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m10\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mfit1\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      4\u001b[0m \u001b[0mvalid_pred\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mfit1\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mstart\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m823\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mend\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m882\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdynamic\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mTrue\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\statsmodels\\tsa\\statespace\\sarimax.py\u001b[0m in \u001b[0;36m__init__\u001b[1;34m(self, endog, exog, order, seasonal_order, trend, measurement_error, time_varying_regression, mle_regression, simple_differencing, enforce_stationarity, enforce_invertibility, hamilton_representation, concentrate_scale, trend_offset, use_exact_diffuse, dates, freq, missing, **kwargs)\u001b[0m\n\u001b[0;32m    329\u001b[0m             \u001b[0mtrend\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mtrend\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0menforce_stationarity\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mNone\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0menforce_invertibility\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mNone\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    330\u001b[0m             \u001b[0mconcentrate_scale\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mconcentrate_scale\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdates\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mdates\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfreq\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mfreq\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 331\u001b[1;33m             missing=missing)\n\u001b[0m\u001b[0;32m    332\u001b[0m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_params\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mSARIMAXParams\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_spec\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    333\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\statsmodels\\tsa\\arima\\specification.py\u001b[0m in \u001b[0;36m__init__\u001b[1;34m(self, endog, exog, order, seasonal_order, ar_order, diff, ma_order, seasonal_ar_order, seasonal_diff, seasonal_ma_order, seasonal_periods, trend, enforce_stationarity, enforce_invertibility, concentrate_scale, trend_offset, dates, freq, missing)\u001b[0m\n\u001b[0;32m    368\u001b[0m                              \u001b[1;34m' in both the seasonal and non-seasonal'\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    369\u001b[0m                              \u001b[1;34m' moving average components.'\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 370\u001b[1;33m                              % duplicate_ma_lags)\n\u001b[0m\u001b[0;32m    371\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    372\u001b[0m         \u001b[1;31m# Handle trend\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mValueError\u001b[0m: Invalid model: moving average lag(s) {7} are in both the seasonal and non-seasonal moving average components."
     ]
    }
   ],
   "source": [
    "model = sarimax.SARIMAX(train, seasonal_order=(3,0,4,7), order=(6,0,10))\n",
    "fit1 = model.fit()\n",
    "\n",
    "valid_pred = fit1.predict(start=823, end=882, dynamic=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [],
   "source": [
    "# val_User_Traffic_MA=[]\n",
    "\n",
    "# for i in range(1,601):\n",
    "    \n",
    "# #     tr,valid=train[(train.Day_No<=822)&(train.Course_ID==i)],train[(train.Day_No>822)&(train.Course_ID==i)]\n",
    "#     tr=train[train.Course_ID==i]\n",
    "    \n",
    "#     for j in range(823,883):\n",
    "#         df=tr.loc[(tr.Day_No>=j-14)&(tr.Day_No<j)]\n",
    "#         val_User_Traffic_MA.append(df[\"User_Traffic\"].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\pandas\\core\\indexing.py:965: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  self.obj[item] = s\n"
     ]
    }
   ],
   "source": [
    "val.loc[:,'User_Traffic']=val_User_Traffic_MA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [],
   "source": [
    "# mean_UT=train.groupby([\"Course_ID\",\"Week_No\",\"Short_Promotion\",\"Day_of_2_Weeks\"])[\"User_Traffic\"].mean().to_frame()\n",
    "# mean_UT.reset_index(inplace=True)\n",
    "# mean_UT[\"Group_Traffic\"]=mean_UT[\"User_Traffic\"]\n",
    "# mean_UT.drop(columns=\"User_Traffic\",inplace=True)\n",
    "# mean_UT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [],
   "source": [
    "# train=train.merge(mean_UT,how=\"left\")\n",
    "# test=test.merge(mean_UT,how=\"left\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [],
   "source": [
    "# train.drop(columns=[\"ID\",\"User_Traffic\",\"Day_No\"],inplace=True)\n",
    "# test.drop(columns=[\"ID\",\"Day_No\"],inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:3: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:4: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  after removing the cwd from sys.path.\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:7: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  import sys\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:8: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \n"
     ]
    }
   ],
   "source": [
    "# train[\"Sales\"]=np.log1p(train[\"Sales\"])\n",
    "\n",
    "traaiin[\"Sales\"]=np.log1p(traaiin[\"Sales\"])\n",
    "traaiin[\"User_Traffic\"]=np.log1p(traaiin[\"User_Traffic\"])\n",
    "\n",
    "\n",
    "val[\"Sales\"]=np.log1p(val[\"Sales\"])\n",
    "val[\"User_Traffic\"]=np.log1p(val[\"User_Traffic\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [],
   "source": [
    "def export_submission(predictions,filename=\"Sample.csv\"):\n",
    "    \n",
    "    ss=sample_sub.copy()\n",
    "    ss[\"Sales\"]=np.expm1(predictions)\n",
    "    ss.to_csv(f\"{filename}.csv\",index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ID                      int64\n",
       "Day_No                  int64\n",
       "Course_ID               int64\n",
       "Course_Domain          object\n",
       "Course_Type            object\n",
       "Short_Promotion         int64\n",
       "Public_Holiday          int64\n",
       "Long_Promotion          int64\n",
       "User_Traffic            int64\n",
       "Competition_Metric    float64\n",
       "Sales                   int64\n",
       "Day_of_2_Weeks          int64\n",
       "Week_No                 int64\n",
       "Year_No                 int32\n",
       "dtype: object"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Course_Domain\n",
       "Business                186.592971\n",
       "Development             120.016391\n",
       "Finance & Accounting    124.520231\n",
       "Software Marketing      118.654665\n",
       "Name: Sales, dtype: float64"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.groupby(\"Course_Domain\").mean()[\"Sales\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Course_Type\n",
       "Course     113.411236\n",
       "Degree     184.061224\n",
       "Program    128.246482\n",
       "Name: Sales, dtype: float64"
      ]
     },
     "execution_count": 135,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.groupby(\"Course_Type\").mean()[\"Sales\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:2: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \n"
     ]
    }
   ],
   "source": [
    "traaiin[\"Course_Domain\"]=traaiin[\"Course_Domain\"].map({'Business':0,\"Development\":1,\"Finance & Accounting\":2,\"Software Marketing\":3})\n",
    "val[\"Course_Domain\"]=val[\"Course_Domain\"].map({'Business':0,\"Development\":1,\"Finance & Accounting\":2,\"Software Marketing\":3})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:2: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \n"
     ]
    }
   ],
   "source": [
    "traaiin[\"Course_Type\"]=traaiin[\"Course_Type\"].map({'Course':0,\"Degree\":1,\"Program\":2})\n",
    "val[\"Course_Type\"]=val[\"Course_Type\"].map({'Course':0,\"Degree\":1,\"Program\":2})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "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>ID</th>\n",
       "      <th>Day_No</th>\n",
       "      <th>Course_ID</th>\n",
       "      <th>Course_Domain</th>\n",
       "      <th>Course_Type</th>\n",
       "      <th>Short_Promotion</th>\n",
       "      <th>Public_Holiday</th>\n",
       "      <th>Long_Promotion</th>\n",
       "      <th>User_Traffic</th>\n",
       "      <th>Competition_Metric</th>\n",
       "      <th>Sales</th>\n",
       "      <th>Day_of_2_Weeks</th>\n",
       "      <th>Week_No</th>\n",
       "      <th>Year_No</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9.306105</td>\n",
       "      <td>0.007</td>\n",
       "      <td>4.406719</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.521568</td>\n",
       "      <td>0.007</td>\n",
       "      <td>4.382027</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.363576</td>\n",
       "      <td>0.007</td>\n",
       "      <td>4.330733</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.397235</td>\n",
       "      <td>0.007</td>\n",
       "      <td>4.394449</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>8.825413</td>\n",
       "      <td>0.007</td>\n",
       "      <td>3.737670</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512022</th>\n",
       "      <td>547963</td>\n",
       "      <td>818</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>8.941807</td>\n",
       "      <td>0.070</td>\n",
       "      <td>4.418841</td>\n",
       "      <td>6</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512023</th>\n",
       "      <td>547964</td>\n",
       "      <td>819</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.538348</td>\n",
       "      <td>0.070</td>\n",
       "      <td>5.351858</td>\n",
       "      <td>7</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512024</th>\n",
       "      <td>547965</td>\n",
       "      <td>820</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.367173</td>\n",
       "      <td>0.070</td>\n",
       "      <td>5.099866</td>\n",
       "      <td>8</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512025</th>\n",
       "      <td>547966</td>\n",
       "      <td>821</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.438193</td>\n",
       "      <td>0.070</td>\n",
       "      <td>5.252273</td>\n",
       "      <td>9</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512026</th>\n",
       "      <td>547967</td>\n",
       "      <td>822</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.602788</td>\n",
       "      <td>0.070</td>\n",
       "      <td>5.303305</td>\n",
       "      <td>10</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>476087 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            ID  Day_No  Course_ID  Course_Domain  Course_Type  \\\n",
       "0            1       1          1              1            0   \n",
       "1            2       2          1              1            0   \n",
       "2            3       3          1              1            0   \n",
       "3            4       4          1              1            0   \n",
       "4            5       5          1              1            0   \n",
       "...        ...     ...        ...            ...          ...   \n",
       "512022  547963     818        600              3            2   \n",
       "512023  547964     819        600              3            2   \n",
       "512024  547965     820        600              3            2   \n",
       "512025  547966     821        600              3            2   \n",
       "512026  547967     822        600              3            2   \n",
       "\n",
       "        Short_Promotion  Public_Holiday  Long_Promotion  User_Traffic  \\\n",
       "0                     0               1               1      9.306105   \n",
       "1                     0               0               1      9.521568   \n",
       "2                     0               0               1      9.363576   \n",
       "3                     0               0               1      9.397235   \n",
       "4                     0               0               1      8.825413   \n",
       "...                 ...             ...             ...           ...   \n",
       "512022                0               0               1      8.941807   \n",
       "512023                1               0               1      9.538348   \n",
       "512024                1               0               1      9.367173   \n",
       "512025                1               0               1      9.438193   \n",
       "512026                1               0               1      9.602788   \n",
       "\n",
       "        Competition_Metric     Sales  Day_of_2_Weeks  Week_No  Year_No  \n",
       "0                    0.007  4.406719               1        1        1  \n",
       "1                    0.007  4.382027               2        1        1  \n",
       "2                    0.007  4.330733               3        1        1  \n",
       "3                    0.007  4.394449               4        1        1  \n",
       "4                    0.007  3.737670               5        1        1  \n",
       "...                    ...       ...             ...      ...      ...  \n",
       "512022               0.070  4.418841               6       13        3  \n",
       "512023               0.070  5.351858               7       13        3  \n",
       "512024               0.070  5.099866               8       14        3  \n",
       "512025               0.070  5.252273               9       14        3  \n",
       "512026               0.070  5.303305              10       14        3  \n",
       "\n",
       "[476087 rows x 14 columns]"
      ]
     },
     "execution_count": 138,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "traaiin"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "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>ID</th>\n",
       "      <th>Day_No</th>\n",
       "      <th>Course_ID</th>\n",
       "      <th>Course_Domain</th>\n",
       "      <th>Course_Type</th>\n",
       "      <th>Short_Promotion</th>\n",
       "      <th>Public_Holiday</th>\n",
       "      <th>Long_Promotion</th>\n",
       "      <th>User_Traffic</th>\n",
       "      <th>Competition_Metric</th>\n",
       "      <th>Sales</th>\n",
       "      <th>Day_of_2_Weeks</th>\n",
       "      <th>Week_No</th>\n",
       "      <th>Year_No</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>822</th>\n",
       "      <td>823</td>\n",
       "      <td>823</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9.384336</td>\n",
       "      <td>0.007</td>\n",
       "      <td>4.143135</td>\n",
       "      <td>11</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>823</th>\n",
       "      <td>824</td>\n",
       "      <td>824</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.370118</td>\n",
       "      <td>0.007</td>\n",
       "      <td>4.394449</td>\n",
       "      <td>12</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>824</th>\n",
       "      <td>825</td>\n",
       "      <td>825</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.387733</td>\n",
       "      <td>0.007</td>\n",
       "      <td>4.418841</td>\n",
       "      <td>13</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>825</th>\n",
       "      <td>826</td>\n",
       "      <td>826</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9.385595</td>\n",
       "      <td>0.007</td>\n",
       "      <td>2.772589</td>\n",
       "      <td>14</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>826</th>\n",
       "      <td>827</td>\n",
       "      <td>827</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.334724</td>\n",
       "      <td>0.007</td>\n",
       "      <td>4.418841</td>\n",
       "      <td>1</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512082</th>\n",
       "      <td>548023</td>\n",
       "      <td>878</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.229211</td>\n",
       "      <td>0.070</td>\n",
       "      <td>4.744932</td>\n",
       "      <td>10</td>\n",
       "      <td>22</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512083</th>\n",
       "      <td>548024</td>\n",
       "      <td>879</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.234789</td>\n",
       "      <td>0.070</td>\n",
       "      <td>4.983607</td>\n",
       "      <td>11</td>\n",
       "      <td>22</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512084</th>\n",
       "      <td>548025</td>\n",
       "      <td>880</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.228033</td>\n",
       "      <td>0.070</td>\n",
       "      <td>5.123964</td>\n",
       "      <td>12</td>\n",
       "      <td>22</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512085</th>\n",
       "      <td>548026</td>\n",
       "      <td>881</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.243098</td>\n",
       "      <td>0.070</td>\n",
       "      <td>4.682131</td>\n",
       "      <td>13</td>\n",
       "      <td>22</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512086</th>\n",
       "      <td>548027</td>\n",
       "      <td>882</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.248166</td>\n",
       "      <td>0.070</td>\n",
       "      <td>5.030438</td>\n",
       "      <td>14</td>\n",
       "      <td>22</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>36000 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            ID  Day_No  Course_ID  Course_Domain  Course_Type  \\\n",
       "822        823     823          1              1            0   \n",
       "823        824     824          1              1            0   \n",
       "824        825     825          1              1            0   \n",
       "825        826     826          1              1            0   \n",
       "826        827     827          1              1            0   \n",
       "...        ...     ...        ...            ...          ...   \n",
       "512082  548023     878        600              3            2   \n",
       "512083  548024     879        600              3            2   \n",
       "512084  548025     880        600              3            2   \n",
       "512085  548026     881        600              3            2   \n",
       "512086  548027     882        600              3            2   \n",
       "\n",
       "        Short_Promotion  Public_Holiday  Long_Promotion  User_Traffic  \\\n",
       "822                   1               1               1      9.384336   \n",
       "823                   0               0               1      9.370118   \n",
       "824                   0               0               1      9.387733   \n",
       "825                   0               1               1      9.385595   \n",
       "826                   0               0               1      9.334724   \n",
       "...                 ...             ...             ...           ...   \n",
       "512082                0               0               1      9.229211   \n",
       "512083                0               0               1      9.234789   \n",
       "512084                0               0               1      9.228033   \n",
       "512085                0               0               1      9.243098   \n",
       "512086                1               0               1      9.248166   \n",
       "\n",
       "        Competition_Metric     Sales  Day_of_2_Weeks  Week_No  Year_No  \n",
       "822                  0.007  4.143135              11       14        3  \n",
       "823                  0.007  4.394449              12       14        3  \n",
       "824                  0.007  4.418841              13       14        3  \n",
       "825                  0.007  2.772589              14       14        3  \n",
       "826                  0.007  4.418841               1       15        3  \n",
       "...                    ...       ...             ...      ...      ...  \n",
       "512082               0.070  4.744932              10       22        3  \n",
       "512083               0.070  4.983607              11       22        3  \n",
       "512084               0.070  5.123964              12       22        3  \n",
       "512085               0.070  4.682131              13       22        3  \n",
       "512086               0.070  5.030438              14       22        3  \n",
       "\n",
       "[36000 rows x 14 columns]"
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train,y_train,X_val,y_val=traaiin.drop(columns=\"Sales\"),traaiin[\"Sales\"],val.drop(columns=\"Sales\"),val[\"Sales\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "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>ID</th>\n",
       "      <th>Day_No</th>\n",
       "      <th>Course_ID</th>\n",
       "      <th>Course_Domain</th>\n",
       "      <th>Course_Type</th>\n",
       "      <th>Short_Promotion</th>\n",
       "      <th>Public_Holiday</th>\n",
       "      <th>Long_Promotion</th>\n",
       "      <th>User_Traffic</th>\n",
       "      <th>Competition_Metric</th>\n",
       "      <th>Day_of_2_Weeks</th>\n",
       "      <th>Week_No</th>\n",
       "      <th>Year_No</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>11004</td>\n",
       "      <td>0.007</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>13650</td>\n",
       "      <td>0.007</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>11655</td>\n",
       "      <td>0.007</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>12054</td>\n",
       "      <td>0.007</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>6804</td>\n",
       "      <td>0.007</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512022</th>\n",
       "      <td>547963</td>\n",
       "      <td>818</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>7644</td>\n",
       "      <td>0.070</td>\n",
       "      <td>6</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512023</th>\n",
       "      <td>547964</td>\n",
       "      <td>819</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>13881</td>\n",
       "      <td>0.070</td>\n",
       "      <td>7</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512024</th>\n",
       "      <td>547965</td>\n",
       "      <td>820</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>11697</td>\n",
       "      <td>0.070</td>\n",
       "      <td>8</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512025</th>\n",
       "      <td>547966</td>\n",
       "      <td>821</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>12558</td>\n",
       "      <td>0.070</td>\n",
       "      <td>9</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512026</th>\n",
       "      <td>547967</td>\n",
       "      <td>822</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>14805</td>\n",
       "      <td>0.070</td>\n",
       "      <td>10</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>476087 rows × 13 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            ID  Day_No  Course_ID  Course_Domain  Course_Type  \\\n",
       "0            1       1          1              1            0   \n",
       "1            2       2          1              1            0   \n",
       "2            3       3          1              1            0   \n",
       "3            4       4          1              1            0   \n",
       "4            5       5          1              1            0   \n",
       "...        ...     ...        ...            ...          ...   \n",
       "512022  547963     818        600              3            2   \n",
       "512023  547964     819        600              3            2   \n",
       "512024  547965     820        600              3            2   \n",
       "512025  547966     821        600              3            2   \n",
       "512026  547967     822        600              3            2   \n",
       "\n",
       "        Short_Promotion  Public_Holiday  Long_Promotion  User_Traffic  \\\n",
       "0                     0               1               1         11004   \n",
       "1                     0               0               1         13650   \n",
       "2                     0               0               1         11655   \n",
       "3                     0               0               1         12054   \n",
       "4                     0               0               1          6804   \n",
       "...                 ...             ...             ...           ...   \n",
       "512022                0               0               1          7644   \n",
       "512023                1               0               1         13881   \n",
       "512024                1               0               1         11697   \n",
       "512025                1               0               1         12558   \n",
       "512026                1               0               1         14805   \n",
       "\n",
       "        Competition_Metric  Day_of_2_Weeks  Week_No  Year_No  \n",
       "0                    0.007               1        1        1  \n",
       "1                    0.007               2        1        1  \n",
       "2                    0.007               3        1        1  \n",
       "3                    0.007               4        1        1  \n",
       "4                    0.007               5        1        1  \n",
       "...                    ...             ...      ...      ...  \n",
       "512022               0.070               6       13        3  \n",
       "512023               0.070               7       13        3  \n",
       "512024               0.070               8       14        3  \n",
       "512025               0.070               9       14        3  \n",
       "512026               0.070              10       14        3  \n",
       "\n",
       "[476087 rows x 13 columns]"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "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>ID</th>\n",
       "      <th>Day_No</th>\n",
       "      <th>Course_ID</th>\n",
       "      <th>Course_Domain</th>\n",
       "      <th>Course_Type</th>\n",
       "      <th>Short_Promotion</th>\n",
       "      <th>Public_Holiday</th>\n",
       "      <th>Long_Promotion</th>\n",
       "      <th>User_Traffic</th>\n",
       "      <th>Competition_Metric</th>\n",
       "      <th>Day_of_2_Weeks</th>\n",
       "      <th>Week_No</th>\n",
       "      <th>Year_No</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>822</th>\n",
       "      <td>823</td>\n",
       "      <td>823</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9.384336</td>\n",
       "      <td>0.007</td>\n",
       "      <td>11</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>823</th>\n",
       "      <td>824</td>\n",
       "      <td>824</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.370118</td>\n",
       "      <td>0.007</td>\n",
       "      <td>12</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>824</th>\n",
       "      <td>825</td>\n",
       "      <td>825</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.387733</td>\n",
       "      <td>0.007</td>\n",
       "      <td>13</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>825</th>\n",
       "      <td>826</td>\n",
       "      <td>826</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9.385595</td>\n",
       "      <td>0.007</td>\n",
       "      <td>14</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>826</th>\n",
       "      <td>827</td>\n",
       "      <td>827</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.334724</td>\n",
       "      <td>0.007</td>\n",
       "      <td>1</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512082</th>\n",
       "      <td>548023</td>\n",
       "      <td>878</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.229211</td>\n",
       "      <td>0.070</td>\n",
       "      <td>10</td>\n",
       "      <td>22</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512083</th>\n",
       "      <td>548024</td>\n",
       "      <td>879</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.234789</td>\n",
       "      <td>0.070</td>\n",
       "      <td>11</td>\n",
       "      <td>22</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512084</th>\n",
       "      <td>548025</td>\n",
       "      <td>880</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.228033</td>\n",
       "      <td>0.070</td>\n",
       "      <td>12</td>\n",
       "      <td>22</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512085</th>\n",
       "      <td>548026</td>\n",
       "      <td>881</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.243098</td>\n",
       "      <td>0.070</td>\n",
       "      <td>13</td>\n",
       "      <td>22</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512086</th>\n",
       "      <td>548027</td>\n",
       "      <td>882</td>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>9.248166</td>\n",
       "      <td>0.070</td>\n",
       "      <td>14</td>\n",
       "      <td>22</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>36000 rows × 13 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            ID  Day_No  Course_ID  Course_Domain  Course_Type  \\\n",
       "822        823     823          1              1            0   \n",
       "823        824     824          1              1            0   \n",
       "824        825     825          1              1            0   \n",
       "825        826     826          1              1            0   \n",
       "826        827     827          1              1            0   \n",
       "...        ...     ...        ...            ...          ...   \n",
       "512082  548023     878        600              3            2   \n",
       "512083  548024     879        600              3            2   \n",
       "512084  548025     880        600              3            2   \n",
       "512085  548026     881        600              3            2   \n",
       "512086  548027     882        600              3            2   \n",
       "\n",
       "        Short_Promotion  Public_Holiday  Long_Promotion  User_Traffic  \\\n",
       "822                   1               1               1      9.384336   \n",
       "823                   0               0               1      9.370118   \n",
       "824                   0               0               1      9.387733   \n",
       "825                   0               1               1      9.385595   \n",
       "826                   0               0               1      9.334724   \n",
       "...                 ...             ...             ...           ...   \n",
       "512082                0               0               1      9.229211   \n",
       "512083                0               0               1      9.234789   \n",
       "512084                0               0               1      9.228033   \n",
       "512085                0               0               1      9.243098   \n",
       "512086                1               0               1      9.248166   \n",
       "\n",
       "        Competition_Metric  Day_of_2_Weeks  Week_No  Year_No  \n",
       "822                  0.007              11       14        3  \n",
       "823                  0.007              12       14        3  \n",
       "824                  0.007              13       14        3  \n",
       "825                  0.007              14       14        3  \n",
       "826                  0.007               1       15        3  \n",
       "...                    ...             ...      ...      ...  \n",
       "512082               0.070              10       22        3  \n",
       "512083               0.070              11       22        3  \n",
       "512084               0.070              12       22        3  \n",
       "512085               0.070              13       22        3  \n",
       "512086               0.070              14       22        3  \n",
       "\n",
       "[36000 rows x 13 columns]"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 203,
   "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>Day_No</th>\n",
       "      <th>Course_ID</th>\n",
       "      <th>Course_Domain</th>\n",
       "      <th>Course_Type</th>\n",
       "      <th>Short_Promotion</th>\n",
       "      <th>Public_Holiday</th>\n",
       "      <th>Long_Promotion</th>\n",
       "      <th>Competition_Metric</th>\n",
       "      <th>Sales</th>\n",
       "      <th>Day_of_2_Weeks</th>\n",
       "      <th>Week_No</th>\n",
       "      <th>Year_No</th>\n",
       "      <th>Target_Course_ID</th>\n",
       "      <th>Group_Traffic</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Day_No</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.000506</td>\n",
       "      <td>-0.007710</td>\n",
       "      <td>0.005606</td>\n",
       "      <td>0.014645</td>\n",
       "      <td>0.019112</td>\n",
       "      <td>-0.020204</td>\n",
       "      <td>0.003223</td>\n",
       "      <td>0.051783</td>\n",
       "      <td>0.016370</td>\n",
       "      <td>0.077342</td>\n",
       "      <td>0.921999</td>\n",
       "      <td>0.038702</td>\n",
       "      <td>0.023682</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Course_ID</th>\n",
       "      <td>-0.000506</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.028393</td>\n",
       "      <td>0.041155</td>\n",
       "      <td>-0.000027</td>\n",
       "      <td>0.001242</td>\n",
       "      <td>-0.039605</td>\n",
       "      <td>-0.007379</td>\n",
       "      <td>0.019872</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>-0.000672</td>\n",
       "      <td>-0.000221</td>\n",
       "      <td>0.017519</td>\n",
       "      <td>0.034256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Course_Domain</th>\n",
       "      <td>-0.007710</td>\n",
       "      <td>0.028393</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.207093</td>\n",
       "      <td>-0.000439</td>\n",
       "      <td>0.002895</td>\n",
       "      <td>0.147668</td>\n",
       "      <td>0.116191</td>\n",
       "      <td>0.007674</td>\n",
       "      <td>0.000159</td>\n",
       "      <td>-0.010210</td>\n",
       "      <td>-0.003377</td>\n",
       "      <td>-0.029696</td>\n",
       "      <td>-0.276193</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Course_Type</th>\n",
       "      <td>0.005606</td>\n",
       "      <td>0.041155</td>\n",
       "      <td>0.207093</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000317</td>\n",
       "      <td>-0.002437</td>\n",
       "      <td>0.020147</td>\n",
       "      <td>0.138144</td>\n",
       "      <td>0.139948</td>\n",
       "      <td>-0.000121</td>\n",
       "      <td>0.007425</td>\n",
       "      <td>0.002455</td>\n",
       "      <td>0.141331</td>\n",
       "      <td>0.036488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Short_Promotion</th>\n",
       "      <td>0.014645</td>\n",
       "      <td>-0.000027</td>\n",
       "      <td>-0.000439</td>\n",
       "      <td>0.000317</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.018770</td>\n",
       "      <td>-0.001149</td>\n",
       "      <td>0.000182</td>\n",
       "      <td>0.372352</td>\n",
       "      <td>-0.117440</td>\n",
       "      <td>-0.009894</td>\n",
       "      <td>0.019498</td>\n",
       "      <td>0.381366</td>\n",
       "      <td>0.203155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Public_Holiday</th>\n",
       "      <td>0.019112</td>\n",
       "      <td>0.001242</td>\n",
       "      <td>0.002895</td>\n",
       "      <td>-0.002437</td>\n",
       "      <td>-0.018770</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.008319</td>\n",
       "      <td>-0.000491</td>\n",
       "      <td>-0.233028</td>\n",
       "      <td>0.047473</td>\n",
       "      <td>-0.002278</td>\n",
       "      <td>0.019078</td>\n",
       "      <td>-0.081046</td>\n",
       "      <td>-0.063492</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Long_Promotion</th>\n",
       "      <td>-0.020204</td>\n",
       "      <td>-0.039605</td>\n",
       "      <td>0.147668</td>\n",
       "      <td>0.020147</td>\n",
       "      <td>-0.001149</td>\n",
       "      <td>0.008319</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.119171</td>\n",
       "      <td>-0.093620</td>\n",
       "      <td>0.000422</td>\n",
       "      <td>-0.026756</td>\n",
       "      <td>-0.008849</td>\n",
       "      <td>-0.113504</td>\n",
       "      <td>-0.208779</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Competition_Metric</th>\n",
       "      <td>0.003223</td>\n",
       "      <td>-0.007379</td>\n",
       "      <td>0.116191</td>\n",
       "      <td>0.138144</td>\n",
       "      <td>0.000182</td>\n",
       "      <td>-0.000491</td>\n",
       "      <td>-0.119171</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.017942</td>\n",
       "      <td>-0.000070</td>\n",
       "      <td>0.004269</td>\n",
       "      <td>0.001411</td>\n",
       "      <td>-0.022864</td>\n",
       "      <td>-0.151883</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sales</th>\n",
       "      <td>0.051783</td>\n",
       "      <td>0.019872</td>\n",
       "      <td>0.007674</td>\n",
       "      <td>0.139948</td>\n",
       "      <td>0.372352</td>\n",
       "      <td>-0.233028</td>\n",
       "      <td>-0.093620</td>\n",
       "      <td>-0.017942</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.017544</td>\n",
       "      <td>0.061220</td>\n",
       "      <td>0.025625</td>\n",
       "      <td>0.875030</td>\n",
       "      <td>0.711351</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Day_of_2_Weeks</th>\n",
       "      <td>0.016370</td>\n",
       "      <td>0.000015</td>\n",
       "      <td>0.000159</td>\n",
       "      <td>-0.000121</td>\n",
       "      <td>-0.117440</td>\n",
       "      <td>0.047473</td>\n",
       "      <td>0.000422</td>\n",
       "      <td>-0.000070</td>\n",
       "      <td>-0.017544</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.028303</td>\n",
       "      <td>0.000909</td>\n",
       "      <td>0.003095</td>\n",
       "      <td>-0.000686</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Week_No</th>\n",
       "      <td>0.077342</td>\n",
       "      <td>-0.000672</td>\n",
       "      <td>-0.010210</td>\n",
       "      <td>0.007425</td>\n",
       "      <td>-0.009894</td>\n",
       "      <td>-0.002278</td>\n",
       "      <td>-0.026756</td>\n",
       "      <td>0.004269</td>\n",
       "      <td>0.061220</td>\n",
       "      <td>0.028303</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.314653</td>\n",
       "      <td>0.079132</td>\n",
       "      <td>0.060720</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Year_No</th>\n",
       "      <td>0.921999</td>\n",
       "      <td>-0.000221</td>\n",
       "      <td>-0.003377</td>\n",
       "      <td>0.002455</td>\n",
       "      <td>0.019498</td>\n",
       "      <td>0.019078</td>\n",
       "      <td>-0.008849</td>\n",
       "      <td>0.001411</td>\n",
       "      <td>0.025625</td>\n",
       "      <td>0.000909</td>\n",
       "      <td>-0.314653</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.005960</td>\n",
       "      <td>-0.001114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Target_Course_ID</th>\n",
       "      <td>0.038702</td>\n",
       "      <td>0.017519</td>\n",
       "      <td>-0.029696</td>\n",
       "      <td>0.141331</td>\n",
       "      <td>0.381366</td>\n",
       "      <td>-0.081046</td>\n",
       "      <td>-0.113504</td>\n",
       "      <td>-0.022864</td>\n",
       "      <td>0.875030</td>\n",
       "      <td>0.003095</td>\n",
       "      <td>0.079132</td>\n",
       "      <td>0.005960</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.826938</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Group_Traffic</th>\n",
       "      <td>0.023682</td>\n",
       "      <td>0.034256</td>\n",
       "      <td>-0.276193</td>\n",
       "      <td>0.036488</td>\n",
       "      <td>0.203155</td>\n",
       "      <td>-0.063492</td>\n",
       "      <td>-0.208779</td>\n",
       "      <td>-0.151883</td>\n",
       "      <td>0.711351</td>\n",
       "      <td>-0.000686</td>\n",
       "      <td>0.060720</td>\n",
       "      <td>-0.001114</td>\n",
       "      <td>0.826938</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Day_No  Course_ID  Course_Domain  Course_Type  \\\n",
       "Day_No              1.000000  -0.000506      -0.007710     0.005606   \n",
       "Course_ID          -0.000506   1.000000       0.028393     0.041155   \n",
       "Course_Domain      -0.007710   0.028393       1.000000     0.207093   \n",
       "Course_Type         0.005606   0.041155       0.207093     1.000000   \n",
       "Short_Promotion     0.014645  -0.000027      -0.000439     0.000317   \n",
       "Public_Holiday      0.019112   0.001242       0.002895    -0.002437   \n",
       "Long_Promotion     -0.020204  -0.039605       0.147668     0.020147   \n",
       "Competition_Metric  0.003223  -0.007379       0.116191     0.138144   \n",
       "Sales               0.051783   0.019872       0.007674     0.139948   \n",
       "Day_of_2_Weeks      0.016370   0.000015       0.000159    -0.000121   \n",
       "Week_No             0.077342  -0.000672      -0.010210     0.007425   \n",
       "Year_No             0.921999  -0.000221      -0.003377     0.002455   \n",
       "Target_Course_ID    0.038702   0.017519      -0.029696     0.141331   \n",
       "Group_Traffic       0.023682   0.034256      -0.276193     0.036488   \n",
       "\n",
       "                    Short_Promotion  Public_Holiday  Long_Promotion  \\\n",
       "Day_No                     0.014645        0.019112       -0.020204   \n",
       "Course_ID                 -0.000027        0.001242       -0.039605   \n",
       "Course_Domain             -0.000439        0.002895        0.147668   \n",
       "Course_Type                0.000317       -0.002437        0.020147   \n",
       "Short_Promotion            1.000000       -0.018770       -0.001149   \n",
       "Public_Holiday            -0.018770        1.000000        0.008319   \n",
       "Long_Promotion            -0.001149        0.008319        1.000000   \n",
       "Competition_Metric         0.000182       -0.000491       -0.119171   \n",
       "Sales                      0.372352       -0.233028       -0.093620   \n",
       "Day_of_2_Weeks            -0.117440        0.047473        0.000422   \n",
       "Week_No                   -0.009894       -0.002278       -0.026756   \n",
       "Year_No                    0.019498        0.019078       -0.008849   \n",
       "Target_Course_ID           0.381366       -0.081046       -0.113504   \n",
       "Group_Traffic              0.203155       -0.063492       -0.208779   \n",
       "\n",
       "                    Competition_Metric     Sales  Day_of_2_Weeks   Week_No  \\\n",
       "Day_No                        0.003223  0.051783        0.016370  0.077342   \n",
       "Course_ID                    -0.007379  0.019872        0.000015 -0.000672   \n",
       "Course_Domain                 0.116191  0.007674        0.000159 -0.010210   \n",
       "Course_Type                   0.138144  0.139948       -0.000121  0.007425   \n",
       "Short_Promotion               0.000182  0.372352       -0.117440 -0.009894   \n",
       "Public_Holiday               -0.000491 -0.233028        0.047473 -0.002278   \n",
       "Long_Promotion               -0.119171 -0.093620        0.000422 -0.026756   \n",
       "Competition_Metric            1.000000 -0.017942       -0.000070  0.004269   \n",
       "Sales                        -0.017942  1.000000       -0.017544  0.061220   \n",
       "Day_of_2_Weeks               -0.000070 -0.017544        1.000000  0.028303   \n",
       "Week_No                       0.004269  0.061220        0.028303  1.000000   \n",
       "Year_No                       0.001411  0.025625        0.000909 -0.314653   \n",
       "Target_Course_ID             -0.022864  0.875030        0.003095  0.079132   \n",
       "Group_Traffic                -0.151883  0.711351       -0.000686  0.060720   \n",
       "\n",
       "                     Year_No  Target_Course_ID  Group_Traffic  \n",
       "Day_No              0.921999          0.038702       0.023682  \n",
       "Course_ID          -0.000221          0.017519       0.034256  \n",
       "Course_Domain      -0.003377         -0.029696      -0.276193  \n",
       "Course_Type         0.002455          0.141331       0.036488  \n",
       "Short_Promotion     0.019498          0.381366       0.203155  \n",
       "Public_Holiday      0.019078         -0.081046      -0.063492  \n",
       "Long_Promotion     -0.008849         -0.113504      -0.208779  \n",
       "Competition_Metric  0.001411         -0.022864      -0.151883  \n",
       "Sales               0.025625          0.875030       0.711351  \n",
       "Day_of_2_Weeks      0.000909          0.003095      -0.000686  \n",
       "Week_No            -0.314653          0.079132       0.060720  \n",
       "Year_No             1.000000          0.005960      -0.001114  \n",
       "Target_Course_ID    0.005960          1.000000       0.826938  \n",
       "Group_Traffic      -0.001114          0.826938       1.000000  "
      ]
     },
     "execution_count": 203,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 204,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Day_No                  int64\n",
       "Course_ID               int64\n",
       "Course_Domain           int64\n",
       "Course_Type             int64\n",
       "Short_Promotion         int64\n",
       "Public_Holiday          int64\n",
       "Long_Promotion          int64\n",
       "Competition_Metric    float64\n",
       "Sales                 float64\n",
       "Day_of_2_Weeks          int64\n",
       "Week_No                 int64\n",
       "Year_No                 int32\n",
       "Target_Course_ID      float64\n",
       "Group_Traffic         float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 204,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ACF and PACF plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from statsmodels.tsa.stattools import acf,pacf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>ID</th>\n",
       "      <th>Day_No</th>\n",
       "      <th>Course_ID</th>\n",
       "      <th>Course_Domain</th>\n",
       "      <th>Course_Type</th>\n",
       "      <th>Short_Promotion</th>\n",
       "      <th>Public_Holiday</th>\n",
       "      <th>Long_Promotion</th>\n",
       "      <th>User_Traffic</th>\n",
       "      <th>Competition_Metric</th>\n",
       "      <th>Sales</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Development</td>\n",
       "      <td>Course</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>11004</td>\n",
       "      <td>0.007</td>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>Development</td>\n",
       "      <td>Course</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>13650</td>\n",
       "      <td>0.007</td>\n",
       "      <td>79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>Development</td>\n",
       "      <td>Course</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>11655</td>\n",
       "      <td>0.007</td>\n",
       "      <td>75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>Development</td>\n",
       "      <td>Course</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>12054</td>\n",
       "      <td>0.007</td>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>Development</td>\n",
       "      <td>Course</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>6804</td>\n",
       "      <td>0.007</td>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>877</th>\n",
       "      <td>878</td>\n",
       "      <td>878</td>\n",
       "      <td>1</td>\n",
       "      <td>Development</td>\n",
       "      <td>Course</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>10878</td>\n",
       "      <td>0.007</td>\n",
       "      <td>73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>878</th>\n",
       "      <td>879</td>\n",
       "      <td>879</td>\n",
       "      <td>1</td>\n",
       "      <td>Development</td>\n",
       "      <td>Course</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>12096</td>\n",
       "      <td>0.007</td>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>879</th>\n",
       "      <td>880</td>\n",
       "      <td>880</td>\n",
       "      <td>1</td>\n",
       "      <td>Development</td>\n",
       "      <td>Course</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>10773</td>\n",
       "      <td>0.007</td>\n",
       "      <td>61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>880</th>\n",
       "      <td>881</td>\n",
       "      <td>881</td>\n",
       "      <td>1</td>\n",
       "      <td>Development</td>\n",
       "      <td>Course</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>10773</td>\n",
       "      <td>0.007</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>881</th>\n",
       "      <td>882</td>\n",
       "      <td>882</td>\n",
       "      <td>1</td>\n",
       "      <td>Development</td>\n",
       "      <td>Course</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>14133</td>\n",
       "      <td>0.007</td>\n",
       "      <td>120</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>882 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      ID  Day_No  Course_ID Course_Domain Course_Type  Short_Promotion  \\\n",
       "0      1       1          1   Development      Course                0   \n",
       "1      2       2          1   Development      Course                0   \n",
       "2      3       3          1   Development      Course                0   \n",
       "3      4       4          1   Development      Course                0   \n",
       "4      5       5          1   Development      Course                0   \n",
       "..   ...     ...        ...           ...         ...              ...   \n",
       "877  878     878          1   Development      Course                0   \n",
       "878  879     879          1   Development      Course                0   \n",
       "879  880     880          1   Development      Course                0   \n",
       "880  881     881          1   Development      Course                0   \n",
       "881  882     882          1   Development      Course                1   \n",
       "\n",
       "     Public_Holiday  Long_Promotion  User_Traffic  Competition_Metric  Sales  \n",
       "0                 1               1         11004               0.007     81  \n",
       "1                 0               1         13650               0.007     79  \n",
       "2                 0               1         11655               0.007     75  \n",
       "3                 0               1         12054               0.007     80  \n",
       "4                 0               1          6804               0.007     41  \n",
       "..              ...             ...           ...                 ...    ...  \n",
       "877               0               1         10878               0.007     73  \n",
       "878               0               1         12096               0.007     81  \n",
       "879               0               1         10773               0.007     61  \n",
       "880               0               1         10773               0.007     60  \n",
       "881               0               1         14133               0.007    120  \n",
       "\n",
       "[882 rows x 11 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train[train.Course_ID==1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Day_No\n",
       "1      15628.136895\n",
       "2      15572.165000\n",
       "3      14241.430000\n",
       "4      14644.700000\n",
       "5      13198.465000\n",
       "           ...     \n",
       "878    14442.330000\n",
       "879    16397.360000\n",
       "880    14763.735000\n",
       "881    12974.010000\n",
       "882    19959.100000\n",
       "Name: User_Traffic, Length: 882, dtype: float64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bydates=train[train.Course_ID==1].groupby(\"Day_No\")[\"User_Traffic\"].mean()\n",
    "bydates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\statsmodels\\tsa\\stattools.py:666: FutureWarning: fft=True will become the default after the release of the 0.12 release of statsmodels. To suppress this warning, explicitly set fft=False.\n",
      "  FutureWarning,\n"
     ]
    }
   ],
   "source": [
    "autocorrfunc=acf(train[train.Course_ID==1][\"User_Traffic\"],nlags=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "51"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(autocorrfunc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(14,7))\n",
    "sns.barplot(x=list(range(0,51)),y=autocorrfunc,color='lightblue')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(14,7))\n",
    "plt.plot(autocorrfunc,  marker='o') \n",
    "plt.axhline(y=0,linestyle='--',color='gray') \n",
    "plt.axhline(y=-1.96/np.sqrt(882),linestyle='--',color='gray') \n",
    "plt.axhline(y=1.96/np.sqrt(882),linestyle='--',color='gray') \n",
    "plt.title('Autocorrelation Function') \n",
    "plt.show() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# dickey fuller, KPSS\n",
    "from statsmodels.tsa.stattools import adfuller, kpss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "def adf_test(timeseries):\n",
    "    \n",
    "    #Perform Dickey-Fuller test:\n",
    "    print ('Results of Dickey-Fuller Test:')\n",
    "    dftest = adfuller(timeseries, autolag='AIC')\n",
    "    dfoutput=pd.Series(dftest[0:4], index=['Test Statistic','p-value','#Lags Used','Number of Observations Used'])\n",
    "\n",
    "    for key,value in dftest[4].items():\n",
    "        dfoutput['Critical Value (%s)'%key] = value\n",
    "    print (dfoutput)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Results of Dickey-Fuller Test:\n",
      "Test Statistic                -6.814058e+00\n",
      "p-value                        2.080814e-09\n",
      "#Lags Used                     2.100000e+01\n",
      "Number of Observations Used    8.600000e+02\n",
      "Critical Value (1%)           -3.437977e+00\n",
      "Critical Value (5%)           -2.864907e+00\n",
      "Critical Value (10%)          -2.568563e+00\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "adf_test(train[train.Course_ID==1][\"User_Traffic\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "def kpss_test(timeseries):\n",
    "    print ('Results of KPSS Test:')\n",
    "    kpsstest = kpss(timeseries, regression='c')\n",
    "    kpss_output = pd.Series(kpsstest[0:3], index=['Test Statistic','p-value','Lags Used'])\n",
    "    for key,value in kpsstest[3].items():\n",
    "        kpss_output['Critical Value (%s)'%key] = value\n",
    "    print (kpss_output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Results of KPSS Test:\n",
      "Test Statistic            0.072358\n",
      "p-value                   0.100000\n",
      "Lags Used                21.000000\n",
      "Critical Value (10%)      0.347000\n",
      "Critical Value (5%)       0.463000\n",
      "Critical Value (2.5%)     0.574000\n",
      "Critical Value (1%)       0.739000\n",
      "dtype: float64\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\statsmodels\\tsa\\stattools.py:1843: FutureWarning: The behavior of using nlags=None will change in release 0.13.Currently nlags=None is the same as nlags=\"legacy\", and so a sample-size lag length is used. After the next release, the default will change to be the same as nlags=\"auto\" which uses an automatic lag length selection method. To silence this warning, either use \"auto\" or \"legacy\"\n",
      "  warnings.warn(msg, FutureWarning)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\statsmodels\\tsa\\stattools.py:1876: InterpolationWarning: The test statistic is outside of the range of p-values available in the\n",
      "look-up table. The actual p-value is greater than the p-value returned.\n",
      "\n",
      "  warnings.warn(warn_msg.format(direction=\"greater\"), InterpolationWarning)\n"
     ]
    }
   ],
   "source": [
    "kpss_test(train[train.Course_ID==1][\"User_Traffic\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fitting Linear Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None, normalize=True)"
      ]
     },
     "execution_count": 141,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr=LinearRegression(normalize=True)\n",
    "lr.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_pred=lr.predict(X_train)\n",
    "y_val_pred=lr.predict(X_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "174.6139972801859"
      ]
     },
     "execution_count": 144,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1000*np.sqrt(mean_squared_error(y_train,y_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "309.2379920358871"
      ]
     },
     "execution_count": 145,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1000*np.sqrt(mean_squared_error(y_val,y_val_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 7.54603922e-06,  7.17176282e-03, -6.91828625e-03,  9.63880514e-02,\n",
       "        2.31193095e-02,  1.53933669e-01, -1.42143108e-01,  7.48556237e-02,\n",
       "        9.05218287e-01,  3.84867986e-01, -4.26562134e-04, -4.94774843e-02,\n",
       "       -2.59676153e+00])"
      ]
     },
     "execution_count": 146,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr.coef_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "metadata": {},
   "outputs": [],
   "source": [
    "export_submission(y_test_pred,\"Linear Regression Featuress\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Grid Search For Linear Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [],
   "source": [
    "parameters={\"normalize\":[\"True\",\"False\"]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {},
   "outputs": [],
   "source": [
    "search=GridSearchCV(LinearRegression(),param_grid=parameters,scoring='neg_mean_squared_error',cv=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score=nan,\n",
       "             estimator=LinearRegression(copy_X=True, fit_intercept=True,\n",
       "                                        n_jobs=None, normalize=False),\n",
       "             iid='deprecated', n_jobs=None,\n",
       "             param_grid={'normalize': ['True', 'False']},\n",
       "             pre_dispatch='2*n_jobs', refit=True, return_train_score=False,\n",
       "             scoring='neg_mean_squared_error', verbose=0)"
      ]
     },
     "execution_count": 172,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'normalize': 'True'}"
      ]
     },
     "execution_count": 173,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.06520298933937438"
      ]
     },
     "execution_count": 174,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_score_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Grid Search For Lasso Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [],
   "source": [
    "parameters={\"normalize\":[\"True\",\"False\"],\n",
    "           \"alpha\":[0.0000001,0.000001,0.00001,0.0001,0.001,0.01,0.1]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "metadata": {},
   "outputs": [],
   "source": [
    "search=GridSearchCV(Lasso(random_state=10),parameters,scoring='neg_mean_squared_error',cv=5,n_jobs=-1,verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 14 candidates, totalling 70 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  17 tasks      | elapsed:   50.9s\n",
      "[Parallel(n_jobs=-1)]: Done  70 out of  70 | elapsed:  4.3min finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score=nan,\n",
       "             estimator=Lasso(alpha=1.0, copy_X=True, fit_intercept=True,\n",
       "                             max_iter=1000, normalize=False, positive=False,\n",
       "                             precompute=False, random_state=10,\n",
       "                             selection='cyclic', tol=0.0001, warm_start=False),\n",
       "             iid='deprecated', n_jobs=-1,\n",
       "             param_grid={'alpha': [1e-07, 1e-06, 1e-05, 0.0001, 0.001, 0.01,\n",
       "                                   0.1],\n",
       "                         'normalize': ['True', 'False']},\n",
       "             pre_dispatch='2*n_jobs', refit=True, return_train_score=False,\n",
       "             scoring='neg_mean_squared_error', verbose=2)"
      ]
     },
     "execution_count": 150,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'alpha': 1e-07, 'normalize': 'True'}"
      ]
     },
     "execution_count": 151,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.17638559246615532"
      ]
     },
     "execution_count": 152,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sqrt(-1*search.best_score_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Lasso(alpha=1e-07, copy_X=True, fit_intercept=True, max_iter=1000,\n",
       "      normalize=True, positive=False, precompute=False, random_state=None,\n",
       "      selection='cyclic', tol=0.0001, warm_start=False)"
      ]
     },
     "execution_count": 154,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lasso=Lasso(alpha=1e-07,normalize=True)\n",
    "lasso.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_pred=lasso.predict(X_train)\n",
    "y_val_pred=lasso.predict(X_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "175.05504922579527"
      ]
     },
     "execution_count": 156,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1000*np.sqrt(mean_squared_error(y_train,y_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "308.6264962712279"
      ]
     },
     "execution_count": 157,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1000*np.sqrt(mean_squared_error(y_val,y_val_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-1.48947301e-08,  4.51244553e-05, -1.72221193e-05,  9.65075669e-02,\n",
       "        2.30145994e-02,  1.48223720e-01, -1.43841918e-01,  7.51544379e-02,\n",
       "        9.06740729e-01,  3.84702035e-01,  1.22983786e-03,  4.40325648e-04,\n",
       "       -0.00000000e+00])"
      ]
     },
     "execution_count": 158,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lasso.coef_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 582,
   "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>Course_ID</th>\n",
       "      <th>Course_Domain</th>\n",
       "      <th>Course_Type</th>\n",
       "      <th>Short_Promotion</th>\n",
       "      <th>Public_Holiday</th>\n",
       "      <th>Long_Promotion</th>\n",
       "      <th>Competition_Metric</th>\n",
       "      <th>Day_of_2_Weeks</th>\n",
       "      <th>Week_No</th>\n",
       "      <th>Year_No</th>\n",
       "      <th>Target_Course_ID</th>\n",
       "      <th>Group_Traffic</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.007</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>72.666667</td>\n",
       "      <td>9968.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.007</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>72.666667</td>\n",
       "      <td>9968.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.007</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>72.666667</td>\n",
       "      <td>9968.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.007</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>41.000000</td>\n",
       "      <td>6755.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.007</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>41.000000</td>\n",
       "      <td>6755.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512082</th>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.070</td>\n",
       "      <td>14</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>144.000000</td>\n",
       "      <td>10164.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512083</th>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.070</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>134.000000</td>\n",
       "      <td>9618.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512084</th>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.070</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>130.000000</td>\n",
       "      <td>9807.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512085</th>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.070</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>121.000000</td>\n",
       "      <td>9219.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>512086</th>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.070</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>171.000000</td>\n",
       "      <td>12180.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>512087 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        Course_ID  Course_Domain  Course_Type  Short_Promotion  \\\n",
       "0               1              1            0                0   \n",
       "1               1              1            0                0   \n",
       "2               1              1            0                0   \n",
       "3               1              1            0                0   \n",
       "4               1              1            0                0   \n",
       "...           ...            ...          ...              ...   \n",
       "512082        600              3            2                1   \n",
       "512083        600              3            2                1   \n",
       "512084        600              3            2                1   \n",
       "512085        600              3            2                1   \n",
       "512086        600              3            2                1   \n",
       "\n",
       "        Public_Holiday  Long_Promotion  Competition_Metric  Day_of_2_Weeks  \\\n",
       "0                    1               1               0.007               1   \n",
       "1                    0               1               0.007               1   \n",
       "2                    0               1               0.007               1   \n",
       "3                    0               1               0.007               2   \n",
       "4                    1               1               0.007               2   \n",
       "...                ...             ...                 ...             ...   \n",
       "512082               0               1               0.070              14   \n",
       "512083               0               1               0.070               1   \n",
       "512084               0               1               0.070               2   \n",
       "512085               0               1               0.070               3   \n",
       "512086               0               1               0.070               4   \n",
       "\n",
       "        Week_No  Year_No  Target_Course_ID  Group_Traffic  \n",
       "0             1        1         72.666667         9968.0  \n",
       "1             1        2         72.666667         9968.0  \n",
       "2             1        3         72.666667         9968.0  \n",
       "3             1        1         41.000000         6755.0  \n",
       "4             1        2         41.000000         6755.0  \n",
       "...         ...      ...               ...            ...  \n",
       "512082        4        3        144.000000        10164.0  \n",
       "512083        5        3        134.000000         9618.0  \n",
       "512084        5        3        130.000000         9807.0  \n",
       "512085        5        3        121.000000         9219.0  \n",
       "512086        5        3        171.000000        12180.0  \n",
       "\n",
       "[512087 rows x 12 columns]"
      ]
     },
     "execution_count": 582,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "export_submission(y_test_pred,\"Lasso Regression Engineered Features\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Grid Search For Tree Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {},
   "outputs": [],
   "source": [
    "tree=DecisionTreeRegressor()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeRegressor(ccp_alpha=0.0, criterion='mse', max_depth=None,\n",
       "                      max_features=None, max_leaf_nodes=None,\n",
       "                      min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                      min_samples_leaf=1, min_samples_split=2,\n",
       "                      min_weight_fraction_leaf=0.0, presort='deprecated',\n",
       "                      random_state=None, splitter='best')"
      ]
     },
     "execution_count": 160,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tree.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_pred=tree.predict(X_train)\n",
    "y_val_pred=tree.predict(X_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.830350265476785e-14"
      ]
     },
     "execution_count": 163,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1000*np.sqrt(mean_squared_error(y_train,y_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "304.4192083946222"
      ]
     },
     "execution_count": 164,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1000*np.sqrt(mean_squared_error(y_val,y_val_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 587,
   "metadata": {},
   "outputs": [],
   "source": [
    "export_submission(y_test_pred,\"Tree Engineered Features\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "44"
      ]
     },
     "execution_count": 165,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tree.get_depth()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "path=tree.cost_complexity_pruning_path(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [],
   "source": [
    "parameters={\"max_depth\":range(7,25)}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [],
   "source": [
    "search=GridSearchCV(DecisionTreeRegressor(random_state=10),parameters,scoring='neg_mean_squared_error',cv=5,verbose=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 18 candidates, totalling 90 fits\n",
      "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[CV] max_depth=7 .....................................................\n",
      "[CV] ........................ max_depth=7, score=-0.024, total=   2.6s\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    2.5s remaining:    0.0s\n",
      "[CV] max_depth=7 .....................................................\n",
      "[CV] ........................ max_depth=7, score=-0.028, total=   2.4s\n",
      "[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    4.8s remaining:    0.0s\n",
      "[CV] max_depth=7 .....................................................\n",
      "[CV] ........................ max_depth=7, score=-0.035, total=   2.4s\n",
      "[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    7.3s remaining:    0.0s\n",
      "[CV] max_depth=7 .....................................................\n",
      "[CV] ........................ max_depth=7, score=-0.037, total=   2.3s\n",
      "[Parallel(n_jobs=1)]: Done   4 out of   4 | elapsed:    9.6s remaining:    0.0s\n",
      "[CV] max_depth=7 .....................................................\n",
      "[CV] ........................ max_depth=7, score=-0.031, total=   2.4s\n",
      "[Parallel(n_jobs=1)]: Done   5 out of   5 | elapsed:   12.0s remaining:    0.0s\n",
      "[CV] max_depth=8 .....................................................\n",
      "[CV] ........................ max_depth=8, score=-0.026, total=   2.7s\n",
      "[Parallel(n_jobs=1)]: Done   6 out of   6 | elapsed:   14.6s remaining:    0.0s\n",
      "[CV] max_depth=8 .....................................................\n",
      "[CV] ........................ max_depth=8, score=-0.028, total=   2.7s\n",
      "[Parallel(n_jobs=1)]: Done   7 out of   7 | elapsed:   17.3s remaining:    0.0s\n",
      "[CV] max_depth=8 .....................................................\n",
      "[CV] ........................ max_depth=8, score=-0.035, total=   2.6s\n",
      "[Parallel(n_jobs=1)]: Done   8 out of   8 | elapsed:   19.9s remaining:    0.0s\n",
      "[CV] max_depth=8 .....................................................\n",
      "[CV] ........................ max_depth=8, score=-0.039, total=   2.7s\n",
      "[Parallel(n_jobs=1)]: Done   9 out of   9 | elapsed:   22.6s remaining:    0.0s\n",
      "[CV] max_depth=8 .....................................................\n",
      "[CV] ........................ max_depth=8, score=-0.036, total=   2.7s\n",
      "[Parallel(n_jobs=1)]: Done  10 out of  10 | elapsed:   25.3s remaining:    0.0s\n",
      "[CV] max_depth=9 .....................................................\n",
      "[CV] ........................ max_depth=9, score=-0.028, total=   2.9s\n",
      "[Parallel(n_jobs=1)]: Done  11 out of  11 | elapsed:   28.2s remaining:    0.0s\n",
      "[CV] max_depth=9 .....................................................\n",
      "[CV] ........................ max_depth=9, score=-0.028, total=   3.0s\n",
      "[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:   31.2s remaining:    0.0s\n",
      "[CV] max_depth=9 .....................................................\n",
      "[CV] ........................ max_depth=9, score=-0.036, total=   2.9s\n",
      "[Parallel(n_jobs=1)]: Done  13 out of  13 | elapsed:   34.2s remaining:    0.0s\n",
      "[CV] max_depth=9 .....................................................\n",
      "[CV] ........................ max_depth=9, score=-0.039, total=   2.8s\n",
      "[Parallel(n_jobs=1)]: Done  14 out of  14 | elapsed:   37.0s remaining:    0.0s\n",
      "[CV] max_depth=9 .....................................................\n",
      "[CV] ........................ max_depth=9, score=-0.038, total=   3.1s\n",
      "[Parallel(n_jobs=1)]: Done  15 out of  15 | elapsed:   40.1s remaining:    0.0s\n",
      "[CV] max_depth=10 ....................................................\n",
      "[CV] ....................... max_depth=10, score=-0.030, total=   3.2s\n",
      "[Parallel(n_jobs=1)]: Done  16 out of  16 | elapsed:   43.3s remaining:    0.0s\n",
      "[CV] max_depth=10 ....................................................\n",
      "[CV] ....................... max_depth=10, score=-0.029, total=   3.1s\n",
      "[Parallel(n_jobs=1)]: Done  17 out of  17 | elapsed:   46.5s remaining:    0.0s\n",
      "[CV] max_depth=10 ....................................................\n",
      "[CV] ....................... max_depth=10, score=-0.037, total=   3.1s\n",
      "[Parallel(n_jobs=1)]: Done  18 out of  18 | elapsed:   49.6s remaining:    0.0s\n",
      "[CV] max_depth=10 ....................................................\n",
      "[CV] ....................... max_depth=10, score=-0.041, total=   3.1s\n",
      "[Parallel(n_jobs=1)]: Done  19 out of  19 | elapsed:   52.7s remaining:    0.0s\n",
      "[CV] max_depth=10 ....................................................\n",
      "[CV] ....................... max_depth=10, score=-0.043, total=   3.1s\n",
      "[Parallel(n_jobs=1)]: Done  20 out of  20 | elapsed:   55.9s remaining:    0.0s\n",
      "[CV] max_depth=11 ....................................................\n",
      "[CV] ....................... max_depth=11, score=-0.032, total=   3.2s\n",
      "[Parallel(n_jobs=1)]: Done  21 out of  21 | elapsed:   59.1s remaining:    0.0s\n",
      "[CV] max_depth=11 ....................................................\n",
      "[CV] ....................... max_depth=11, score=-0.032, total=   3.4s\n",
      "[Parallel(n_jobs=1)]: Done  22 out of  22 | elapsed:  1.0min remaining:    0.0s\n",
      "[CV] max_depth=11 ....................................................\n",
      "[CV] ....................... max_depth=11, score=-0.038, total=   3.3s\n",
      "[Parallel(n_jobs=1)]: Done  23 out of  23 | elapsed:  1.1min remaining:    0.0s\n",
      "[CV] max_depth=11 ....................................................\n",
      "[CV] ....................... max_depth=11, score=-0.041, total=   3.2s\n",
      "[Parallel(n_jobs=1)]: Done  24 out of  24 | elapsed:  1.2min remaining:    0.0s\n",
      "[CV] max_depth=11 ....................................................\n",
      "[CV] ....................... max_depth=11, score=-0.050, total=   3.4s\n",
      "[Parallel(n_jobs=1)]: Done  25 out of  25 | elapsed:  1.2min remaining:    0.0s\n",
      "[CV] max_depth=12 ....................................................\n",
      "[CV] ....................... max_depth=12, score=-0.036, total=   3.4s\n",
      "[Parallel(n_jobs=1)]: Done  26 out of  26 | elapsed:  1.3min remaining:    0.0s\n",
      "[CV] max_depth=12 ....................................................\n",
      "[CV] ....................... max_depth=12, score=-0.035, total=   3.6s\n",
      "[Parallel(n_jobs=1)]: Done  27 out of  27 | elapsed:  1.3min remaining:    0.0s\n",
      "[CV] max_depth=12 ....................................................\n",
      "[CV] ....................... max_depth=12, score=-0.040, total=   3.6s\n",
      "[Parallel(n_jobs=1)]: Done  28 out of  28 | elapsed:  1.4min remaining:    0.0s\n",
      "[CV] max_depth=12 ....................................................\n",
      "[CV] ....................... max_depth=12, score=-0.042, total=   3.5s\n",
      "[Parallel(n_jobs=1)]: Done  29 out of  29 | elapsed:  1.4min remaining:    0.0s\n",
      "[CV] max_depth=12 ....................................................\n",
      "[CV] ....................... max_depth=12, score=-0.053, total=   3.5s\n",
      "[Parallel(n_jobs=1)]: Done  30 out of  30 | elapsed:  1.5min remaining:    0.0s\n",
      "[CV] max_depth=13 ....................................................\n",
      "[CV] ....................... max_depth=13, score=-0.039, total=   3.7s\n",
      "[Parallel(n_jobs=1)]: Done  31 out of  31 | elapsed:  1.6min remaining:    0.0s\n",
      "[CV] max_depth=13 ....................................................\n",
      "[CV] ....................... max_depth=13, score=-0.037, total=   3.8s\n",
      "[Parallel(n_jobs=1)]: Done  32 out of  32 | elapsed:  1.6min remaining:    0.0s\n",
      "[CV] max_depth=13 ....................................................\n",
      "[CV] ....................... max_depth=13, score=-0.041, total=   3.7s\n",
      "[Parallel(n_jobs=1)]: Done  33 out of  33 | elapsed:  1.7min remaining:    0.0s\n",
      "[CV] max_depth=13 ....................................................\n",
      "[CV] ....................... max_depth=13, score=-0.043, total=   3.8s\n",
      "[Parallel(n_jobs=1)]: Done  34 out of  34 | elapsed:  1.8min remaining:    0.0s\n",
      "[CV] max_depth=13 ....................................................\n",
      "[CV] ....................... max_depth=13, score=-0.057, total=   3.7s\n",
      "[Parallel(n_jobs=1)]: Done  35 out of  35 | elapsed:  1.8min remaining:    0.0s\n",
      "[CV] max_depth=14 ....................................................\n",
      "[CV] ....................... max_depth=14, score=-0.042, total=   3.8s\n",
      "[Parallel(n_jobs=1)]: Done  36 out of  36 | elapsed:  1.9min remaining:    0.0s\n",
      "[CV] max_depth=14 ....................................................\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] ....................... max_depth=14, score=-0.038, total=   4.3s\n",
      "[Parallel(n_jobs=1)]: Done  37 out of  37 | elapsed:  1.9min remaining:    0.0s\n",
      "[CV] max_depth=14 ....................................................\n",
      "[CV] ....................... max_depth=14, score=-0.043, total=   4.0s\n",
      "[Parallel(n_jobs=1)]: Done  38 out of  38 | elapsed:  2.0min remaining:    0.0s\n",
      "[CV] max_depth=14 ....................................................\n",
      "[CV] ....................... max_depth=14, score=-0.044, total=   4.0s\n",
      "[Parallel(n_jobs=1)]: Done  39 out of  39 | elapsed:  2.1min remaining:    0.0s\n",
      "[CV] max_depth=14 ....................................................\n",
      "[CV] ....................... max_depth=14, score=-0.059, total=   3.9s\n",
      "[Parallel(n_jobs=1)]: Done  40 out of  40 | elapsed:  2.1min remaining:    0.0s\n",
      "[CV] max_depth=15 ....................................................\n",
      "[CV] ....................... max_depth=15, score=-0.044, total=   4.2s\n",
      "[Parallel(n_jobs=1)]: Done  41 out of  41 | elapsed:  2.2min remaining:    0.0s\n",
      "[CV] max_depth=15 ....................................................\n",
      "[CV] ....................... max_depth=15, score=-0.040, total=   4.1s\n",
      "[Parallel(n_jobs=1)]: Done  42 out of  42 | elapsed:  2.3min remaining:    0.0s\n",
      "[CV] max_depth=15 ....................................................\n",
      "[CV] ....................... max_depth=15, score=-0.045, total=   4.1s\n",
      "[Parallel(n_jobs=1)]: Done  43 out of  43 | elapsed:  2.4min remaining:    0.0s\n",
      "[CV] max_depth=15 ....................................................\n",
      "[CV] ....................... max_depth=15, score=-0.046, total=   4.2s\n",
      "[Parallel(n_jobs=1)]: Done  44 out of  44 | elapsed:  2.4min remaining:    0.0s\n",
      "[CV] max_depth=15 ....................................................\n",
      "[CV] ....................... max_depth=15, score=-0.062, total=   4.1s\n",
      "[Parallel(n_jobs=1)]: Done  45 out of  45 | elapsed:  2.5min remaining:    0.0s\n",
      "[CV] max_depth=16 ....................................................\n",
      "[CV] ....................... max_depth=16, score=-0.046, total=   4.1s\n",
      "[Parallel(n_jobs=1)]: Done  46 out of  46 | elapsed:  2.6min remaining:    0.0s\n",
      "[CV] max_depth=16 ....................................................\n",
      "[CV] ....................... max_depth=16, score=-0.042, total=   4.2s\n",
      "[Parallel(n_jobs=1)]: Done  47 out of  47 | elapsed:  2.6min remaining:    0.0s\n",
      "[CV] max_depth=16 ....................................................\n",
      "[CV] ....................... max_depth=16, score=-0.046, total=   4.3s\n",
      "[Parallel(n_jobs=1)]: Done  48 out of  48 | elapsed:  2.7min remaining:    0.0s\n",
      "[CV] max_depth=16 ....................................................\n",
      "[CV] ....................... max_depth=16, score=-0.048, total=   4.4s\n",
      "[Parallel(n_jobs=1)]: Done  49 out of  49 | elapsed:  2.8min remaining:    0.0s\n",
      "[CV] max_depth=16 ....................................................\n",
      "[CV] ....................... max_depth=16, score=-0.066, total=   4.2s\n",
      "[Parallel(n_jobs=1)]: Done  50 out of  50 | elapsed:  2.8min remaining:    0.0s\n",
      "[CV] max_depth=17 ....................................................\n",
      "[CV] ....................... max_depth=17, score=-0.047, total=   4.4s\n",
      "[Parallel(n_jobs=1)]: Done  51 out of  51 | elapsed:  2.9min remaining:    0.0s\n",
      "[CV] max_depth=17 ....................................................\n",
      "[CV] ....................... max_depth=17, score=-0.044, total=   4.5s\n",
      "[Parallel(n_jobs=1)]: Done  52 out of  52 | elapsed:  3.0min remaining:    0.0s\n",
      "[CV] max_depth=17 ....................................................\n",
      "[CV] ....................... max_depth=17, score=-0.048, total=   4.4s\n",
      "[Parallel(n_jobs=1)]: Done  53 out of  53 | elapsed:  3.1min remaining:    0.0s\n",
      "[CV] max_depth=17 ....................................................\n",
      "[CV] ....................... max_depth=17, score=-0.049, total=   4.4s\n",
      "[Parallel(n_jobs=1)]: Done  54 out of  54 | elapsed:  3.1min remaining:    0.0s\n",
      "[CV] max_depth=17 ....................................................\n",
      "[CV] ....................... max_depth=17, score=-0.065, total=   4.3s\n",
      "[Parallel(n_jobs=1)]: Done  55 out of  55 | elapsed:  3.2min remaining:    0.0s\n",
      "[CV] max_depth=18 ....................................................\n",
      "[CV] ....................... max_depth=18, score=-0.047, total=   4.5s\n",
      "[Parallel(n_jobs=1)]: Done  56 out of  56 | elapsed:  3.3min remaining:    0.0s\n",
      "[CV] max_depth=18 ....................................................\n",
      "[CV] ....................... max_depth=18, score=-0.044, total=   4.8s\n",
      "[Parallel(n_jobs=1)]: Done  57 out of  57 | elapsed:  3.4min remaining:    0.0s\n",
      "[CV] max_depth=18 ....................................................\n",
      "[CV] ....................... max_depth=18, score=-0.052, total=   4.7s\n",
      "[Parallel(n_jobs=1)]: Done  58 out of  58 | elapsed:  3.4min remaining:    0.0s\n",
      "[CV] max_depth=18 ....................................................\n",
      "[CV] ....................... max_depth=18, score=-0.051, total=   4.6s\n",
      "[Parallel(n_jobs=1)]: Done  59 out of  59 | elapsed:  3.5min remaining:    0.0s\n",
      "[CV] max_depth=18 ....................................................\n",
      "[CV] ....................... max_depth=18, score=-0.066, total=   4.5s\n",
      "[Parallel(n_jobs=1)]: Done  60 out of  60 | elapsed:  3.6min remaining:    0.0s\n",
      "[CV] max_depth=19 ....................................................\n",
      "[CV] ....................... max_depth=19, score=-0.050, total=   4.7s\n",
      "[Parallel(n_jobs=1)]: Done  61 out of  61 | elapsed:  3.7min remaining:    0.0s\n",
      "[CV] max_depth=19 ....................................................\n",
      "[CV] ....................... max_depth=19, score=-0.046, total=   4.8s\n",
      "[Parallel(n_jobs=1)]: Done  62 out of  62 | elapsed:  3.8min remaining:    0.0s\n",
      "[CV] max_depth=19 ....................................................\n",
      "[CV] ....................... max_depth=19, score=-0.050, total=   4.6s\n",
      "[Parallel(n_jobs=1)]: Done  63 out of  63 | elapsed:  3.8min remaining:    0.0s\n",
      "[CV] max_depth=19 ....................................................\n",
      "[CV] ....................... max_depth=19, score=-0.051, total=   4.8s\n",
      "[Parallel(n_jobs=1)]: Done  64 out of  64 | elapsed:  3.9min remaining:    0.0s\n",
      "[CV] max_depth=19 ....................................................\n",
      "[CV] ....................... max_depth=19, score=-0.066, total=   4.7s\n",
      "[Parallel(n_jobs=1)]: Done  65 out of  65 | elapsed:  4.0min remaining:    0.0s\n",
      "[CV] max_depth=20 ....................................................\n",
      "[CV] ....................... max_depth=20, score=-0.049, total=   4.9s\n",
      "[Parallel(n_jobs=1)]: Done  66 out of  66 | elapsed:  4.1min remaining:    0.0s\n",
      "[CV] max_depth=20 ....................................................\n",
      "[CV] ....................... max_depth=20, score=-0.047, total=   5.0s\n",
      "[Parallel(n_jobs=1)]: Done  67 out of  67 | elapsed:  4.2min remaining:    0.0s\n",
      "[CV] max_depth=20 ....................................................\n",
      "[CV] ....................... max_depth=20, score=-0.053, total=   5.0s\n",
      "[Parallel(n_jobs=1)]: Done  68 out of  68 | elapsed:  4.2min remaining:    0.0s\n",
      "[CV] max_depth=20 ....................................................\n",
      "[CV] ....................... max_depth=20, score=-0.052, total=   5.0s\n",
      "[Parallel(n_jobs=1)]: Done  69 out of  69 | elapsed:  4.3min remaining:    0.0s\n",
      "[CV] max_depth=20 ....................................................\n",
      "[CV] ....................... max_depth=20, score=-0.069, total=   4.9s\n",
      "[Parallel(n_jobs=1)]: Done  70 out of  70 | elapsed:  4.4min remaining:    0.0s\n",
      "[CV] max_depth=21 ....................................................\n",
      "[CV] ....................... max_depth=21, score=-0.048, total=   4.9s\n",
      "[Parallel(n_jobs=1)]: Done  71 out of  71 | elapsed:  4.5min remaining:    0.0s\n",
      "[CV] max_depth=21 ....................................................\n",
      "[CV] ....................... max_depth=21, score=-0.048, total=   5.1s\n",
      "[Parallel(n_jobs=1)]: Done  72 out of  72 | elapsed:  4.6min remaining:    0.0s\n",
      "[CV] max_depth=21 ....................................................\n",
      "[CV] ....................... max_depth=21, score=-0.053, total=   5.1s\n",
      "[Parallel(n_jobs=1)]: Done  73 out of  73 | elapsed:  4.7min remaining:    0.0s\n",
      "[CV] max_depth=21 ....................................................\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] ....................... max_depth=21, score=-0.053, total=   5.1s\n",
      "[Parallel(n_jobs=1)]: Done  74 out of  74 | elapsed:  4.7min remaining:    0.0s\n",
      "[CV] max_depth=21 ....................................................\n",
      "[CV] ....................... max_depth=21, score=-0.069, total=   5.0s\n",
      "[Parallel(n_jobs=1)]: Done  75 out of  75 | elapsed:  4.8min remaining:    0.0s\n",
      "[CV] max_depth=22 ....................................................\n",
      "[CV] ....................... max_depth=22, score=-0.050, total=   5.1s\n",
      "[Parallel(n_jobs=1)]: Done  76 out of  76 | elapsed:  4.9min remaining:    0.0s\n",
      "[CV] max_depth=22 ....................................................\n",
      "[CV] ....................... max_depth=22, score=-0.048, total=   5.1s\n",
      "[Parallel(n_jobs=1)]: Done  77 out of  77 | elapsed:  5.0min remaining:    0.0s\n",
      "[CV] max_depth=22 ....................................................\n",
      "[CV] ....................... max_depth=22, score=-0.053, total=   5.1s\n",
      "[Parallel(n_jobs=1)]: Done  78 out of  78 | elapsed:  5.1min remaining:    0.0s\n",
      "[CV] max_depth=22 ....................................................\n",
      "[CV] ....................... max_depth=22, score=-0.052, total=   5.3s\n",
      "[Parallel(n_jobs=1)]: Done  79 out of  79 | elapsed:  5.2min remaining:    0.0s\n",
      "[CV] max_depth=22 ....................................................\n",
      "[CV] ....................... max_depth=22, score=-0.068, total=   5.0s\n",
      "[Parallel(n_jobs=1)]: Done  80 out of  80 | elapsed:  5.3min remaining:    0.0s\n",
      "[CV] max_depth=23 ....................................................\n",
      "[CV] ....................... max_depth=23, score=-0.051, total=   5.2s\n",
      "[Parallel(n_jobs=1)]: Done  81 out of  81 | elapsed:  5.3min remaining:    0.0s\n",
      "[CV] max_depth=23 ....................................................\n",
      "[CV] ....................... max_depth=23, score=-0.049, total=   5.2s\n",
      "[Parallel(n_jobs=1)]: Done  82 out of  82 | elapsed:  5.4min remaining:    0.0s\n",
      "[CV] max_depth=23 ....................................................\n",
      "[CV] ....................... max_depth=23, score=-0.053, total=   5.3s\n",
      "[Parallel(n_jobs=1)]: Done  83 out of  83 | elapsed:  5.5min remaining:    0.0s\n",
      "[CV] max_depth=23 ....................................................\n",
      "[CV] ....................... max_depth=23, score=-0.054, total=   5.2s\n",
      "[Parallel(n_jobs=1)]: Done  84 out of  84 | elapsed:  5.6min remaining:    0.0s\n",
      "[CV] max_depth=23 ....................................................\n",
      "[CV] ....................... max_depth=23, score=-0.068, total=   5.2s\n",
      "[Parallel(n_jobs=1)]: Done  85 out of  85 | elapsed:  5.7min remaining:    0.0s\n",
      "[CV] max_depth=24 ....................................................\n",
      "[CV] ....................... max_depth=24, score=-0.051, total=   5.4s\n",
      "[Parallel(n_jobs=1)]: Done  86 out of  86 | elapsed:  5.8min remaining:    0.0s\n",
      "[CV] max_depth=24 ....................................................\n",
      "[CV] ....................... max_depth=24, score=-0.052, total=   5.4s\n",
      "[Parallel(n_jobs=1)]: Done  87 out of  87 | elapsed:  5.9min remaining:    0.0s\n",
      "[CV] max_depth=24 ....................................................\n"
     ]
    }
   ],
   "source": [
    "search.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 224,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.013210302124428008"
      ]
     },
     "execution_count": 224,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "metadata": {},
   "outputs": [],
   "source": [
    "tree=DecisionTreeRegressor(max_depth= 12)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 226,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeRegressor(ccp_alpha=0.0, criterion='mse', max_depth=12,\n",
       "                      max_features=None, max_leaf_nodes=None,\n",
       "                      min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                      min_samples_leaf=1, min_samples_split=2,\n",
       "                      min_weight_fraction_leaf=0.0, presort='deprecated',\n",
       "                      random_state=None, splitter='best')"
      ]
     },
     "execution_count": 226,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tree.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 227,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2.4849066497880004, 6.1779441140506)"
      ]
     },
     "execution_count": 227,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test_pred.min(),y_test_pred.max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 228,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.3277473750014732e-12"
      ]
     },
     "execution_count": 228,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1000*np.sqrt(mean_squared_error(y_train,y_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 229,
   "metadata": {},
   "outputs": [],
   "source": [
    "export_submission(y_test_pred,\"Tree Max Depth Engineered Features\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Random Forests"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 405,
   "metadata": {},
   "outputs": [],
   "source": [
    "forest=RandomForestRegressor(max_features=None,n_estimators=400,n_jobs=8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 406,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(bootstrap=True, ccp_alpha=0.0, criterion='mse',\n",
       "                      max_depth=None, max_features=None, max_leaf_nodes=None,\n",
       "                      max_samples=None, min_impurity_decrease=0.0,\n",
       "                      min_impurity_split=None, min_samples_leaf=1,\n",
       "                      min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "                      n_estimators=400, n_jobs=8, oob_score=False,\n",
       "                      random_state=None, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 406,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forest.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 407,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_pred=forest.predict(X_train)\n",
    "y_test_pred=forest.predict(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 408,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "107.22308899248416"
      ]
     },
     "execution_count": 408,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1000*np.sqrt(mean_squared_error(y_train,y_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 409,
   "metadata": {},
   "outputs": [],
   "source": [
    "export_submission(y_test_pred,\"Forest Engineered Features\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [],
   "source": [
    "parameters={\"max_depth\":range(2,20),'max_samples':[0.5,0.6,0.7,0.8]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [],
   "source": [
    "search=RandomizedSearchCV(RandomForestRegressor(random_state=10,max_features=None,verbose=2,n_estimators=150,n_jobs=-1),parameters,scoring='neg_mean_squared_error',cv=5,verbose=2,n_iter=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 20 candidates, totalling 100 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=6)]: Using backend LokyBackend with 6 concurrent workers.\n",
      "[Parallel(n_jobs=6)]: Done  29 tasks      | elapsed: 17.2min\n",
      "[Parallel(n_jobs=6)]: Done 100 out of 100 | elapsed: 77.8min finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=5, error_score=nan,\n",
       "                   estimator=RandomForestRegressor(bootstrap=True,\n",
       "                                                   ccp_alpha=0.0,\n",
       "                                                   criterion='mse',\n",
       "                                                   max_depth=None,\n",
       "                                                   max_features=None,\n",
       "                                                   max_leaf_nodes=None,\n",
       "                                                   max_samples=None,\n",
       "                                                   min_impurity_decrease=0.0,\n",
       "                                                   min_impurity_split=None,\n",
       "                                                   min_samples_leaf=1,\n",
       "                                                   min_samples_split=2,\n",
       "                                                   min_weight_fraction_leaf=0.0,\n",
       "                                                   n_estimators=200, n_jobs=6,\n",
       "                                                   oob_score=False,\n",
       "                                                   random_state=10, verbose=0,\n",
       "                                                   warm_start=False),\n",
       "                   iid='deprecated', n_iter=20, n_jobs=6,\n",
       "                   param_distributions={'max_depth': range(2, 20),\n",
       "                                        'max_samples': [0.5, 0.6, 0.7, 0.8]},\n",
       "                   pre_dispatch='2*n_jobs', random_state=None, refit=True,\n",
       "                   return_train_score=False, scoring='neg_mean_squared_error',\n",
       "                   verbose=2)"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'max_samples': 0.8, 'max_depth': 7}"
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.03500106369665605"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 412,
   "metadata": {},
   "outputs": [],
   "source": [
    "tree=RandomForestRegressor(max_depth= 7, max_samples=0.8, verbose=2, n_jobs=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 413,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend ThreadingBackend with 12 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "building tree 1 of 100\n",
      "building tree 2 of 100\n",
      "building tree 3 of 100\n",
      "building tree 4 of 100\n",
      "building tree 5 of 100\n",
      "building tree 6 of 100\n",
      "building tree 7 of 100\n",
      "building tree 8 of 100\n",
      "building tree 9 of 100\n",
      "building tree 10 of 100\n",
      "building tree 11 of 100\n",
      "building tree 12 of 100\n",
      "building tree 13 of 100\n",
      "building tree 14 of 100\n",
      "building tree 15 of 100\n",
      "building tree 16 of 100\n",
      "building tree 17 of 100\n",
      "building tree 18 of 100\n",
      "building tree 19 of 100\n",
      "building tree 20 of 100\n",
      "building tree 21 of 100\n",
      "building tree 22 of 100\n",
      "building tree 23 of 100\n",
      "building tree 24 of 100\n",
      "building tree 25 of 100building tree 26 of 100\n",
      "\n",
      "building tree 27 of 100\n",
      "building tree 28 of 100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done  17 tasks      | elapsed:    6.4s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "building tree 29 of 100\n",
      "building tree 30 of 100\n",
      "building tree 31 of 100\n",
      "building tree 32 of 100\n",
      "building tree 33 of 100\n",
      "building tree 34 of 100\n",
      "building tree 35 of 100\n",
      "building tree 36 of 100\n",
      "building tree 37 of 100\n",
      "building tree 38 of 100\n",
      "building tree 39 of 100\n",
      "building tree 40 of 100\n",
      "building tree 41 of 100\n",
      "building tree 42 of 100\n",
      "building tree 43 of 100\n",
      "building tree 44 of 100\n",
      "building tree 45 of 100\n",
      "building tree 46 of 100building tree 47 of 100\n",
      "\n",
      "building tree 48 of 100\n",
      "building tree 49 of 100\n",
      "building tree 50 of 100\n",
      "building tree 51 of 100\n",
      "building tree 52 of 100\n",
      "building tree 53 of 100\n",
      "building tree 54 of 100\n",
      "building tree 55 of 100\n",
      "building tree 56 of 100\n",
      "building tree 57 of 100\n",
      "building tree 58 of 100\n",
      "building tree 59 of 100\n",
      "building tree 60 of 100\n",
      "building tree 61 of 100\n",
      "building tree 62 of 100\n",
      "building tree 63 of 100\n",
      "building tree 64 of 100\n",
      "building tree 65 of 100\n",
      "building tree 66 of 100\n",
      "building tree 67 of 100\n",
      "building tree 68 of 100\n",
      "building tree 69 of 100\n",
      "building tree 70 of 100\n",
      "building tree 71 of 100\n",
      "building tree 72 of 100\n",
      "building tree 73 of 100\n",
      "building tree 74 of 100\n",
      "building tree 75 of 100\n",
      "building tree 76 of 100\n",
      "building tree 77 of 100\n",
      "building tree 78 of 100\n",
      "building tree 79 of 100\n",
      "building tree 80 of 100\n",
      "building tree 81 of 100\n",
      "building tree 82 of 100\n",
      "building tree 83 of 100\n",
      "building tree 84 of 100\n",
      "building tree 85 of 100\n",
      "building tree 86 of 100\n",
      "building tree 87 of 100\n",
      "building tree 88 of 100\n",
      "building tree 89 of 100\n",
      "building tree 90 of 100\n",
      "building tree 91 of 100\n",
      "building tree 92 of 100\n",
      "building tree 93 of 100\n",
      "building tree 94 of 100\n",
      "building tree 95 of 100\n",
      "building tree 96 of 100\n",
      "building tree 97 of 100\n",
      "building tree 98 of 100\n",
      "building tree 99 of 100\n",
      "building tree 100 of 100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done 100 out of 100 | elapsed:   26.6s finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(bootstrap=True, ccp_alpha=0.0, criterion='mse',\n",
       "                      max_depth=7, max_features='auto', max_leaf_nodes=None,\n",
       "                      max_samples=0.8, min_impurity_decrease=0.0,\n",
       "                      min_impurity_split=None, min_samples_leaf=1,\n",
       "                      min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "                      n_estimators=100, n_jobs=-1, oob_score=False,\n",
       "                      random_state=None, verbose=2, warm_start=False)"
      ]
     },
     "execution_count": 413,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tree.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 414,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=12)]: Using backend ThreadingBackend with 12 concurrent workers.\n",
      "[Parallel(n_jobs=12)]: Done  17 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=12)]: Done 100 out of 100 | elapsed:    0.4s finished\n",
      "[Parallel(n_jobs=12)]: Using backend ThreadingBackend with 12 concurrent workers.\n",
      "[Parallel(n_jobs=12)]: Done  17 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=12)]: Done 100 out of 100 | elapsed:    0.0s finished\n"
     ]
    }
   ],
   "source": [
    "y_train_pred=tree.predict(X_train)\n",
    "y_test_pred=tree.predict(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 415,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "131.3696498980016"
      ]
     },
     "execution_count": 415,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1000*np.sqrt(mean_squared_error(y_train,y_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 416,
   "metadata": {},
   "outputs": [],
   "source": [
    "export_submission(y_test_pred,\"Random Forest Tuned Engineered Features\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## LightGBM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 430,
   "metadata": {},
   "outputs": [],
   "source": [
    "regressor=LGBMRegressor()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 431,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LGBMRegressor(boosting_type='gbdt', class_weight=None, colsample_bytree=1.0,\n",
       "              importance_type='split', learning_rate=0.1, max_depth=-1,\n",
       "              min_child_samples=20, min_child_weight=0.001, min_split_gain=0.0,\n",
       "              n_estimators=100, n_jobs=-1, num_leaves=31, objective=None,\n",
       "              random_state=None, reg_alpha=0.0, reg_lambda=0.0, silent=True,\n",
       "              subsample=1.0, subsample_for_bin=200000, subsample_freq=0)"
      ]
     },
     "execution_count": 431,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "regressor.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 434,
   "metadata": {},
   "outputs": [
    {
     "ename": "AttributeError",
     "evalue": "'LGBMRegressor' object has no attribute 'max_bin'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-434-46da86873b75>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mregressor\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmax_bin\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;31mAttributeError\u001b[0m: 'LGBMRegressor' object has no attribute 'max_bin'"
     ]
    }
   ],
   "source": [
    "regressor.max_bin"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 419,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_pred=regressor.predict(X_train)\n",
    "y_test_pred=regressor.predict(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 420,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "117.33328243727972"
      ]
     },
     "execution_count": 420,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1000*np.sqrt(mean_squared_error(y_train,y_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 421,
   "metadata": {},
   "outputs": [],
   "source": [
    "export_submission(y_test_pred,\"LGBM Default Engineered Features\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 427,
   "metadata": {},
   "outputs": [],
   "source": [
    "parameters={'max_depth':range(2,10),\n",
    "           }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 428,
   "metadata": {},
   "outputs": [],
   "source": [
    "search=RandomizedSearchCV(LGBMRegressor(n_estimators=200,random_state=10),parameters,scoring='neg_mean_squared_error',n_jobs=-1,cv=5,verbose=2,n_iter=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 429,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 10 candidates, totalling 50 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  17 tasks      | elapsed:   39.3s\n",
      "[Parallel(n_jobs=-1)]: Done  18 tasks      | elapsed:   39.3s\n",
      "[Parallel(n_jobs=-1)]: Done  19 tasks      | elapsed:   39.3s\n",
      "[Parallel(n_jobs=-1)]: Done  20 tasks      | elapsed:   39.3s\n",
      "[Parallel(n_jobs=-1)]: Done  21 tasks      | elapsed:   39.3s\n",
      "[Parallel(n_jobs=-1)]: Done  22 tasks      | elapsed:   39.3s\n",
      "[Parallel(n_jobs=-1)]: Done  23 tasks      | elapsed:   39.3s\n",
      "[Parallel(n_jobs=-1)]: Done  24 tasks      | elapsed:   39.3s\n",
      "[Parallel(n_jobs=-1)]: Done  25 tasks      | elapsed:   39.3s\n",
      "[Parallel(n_jobs=-1)]: Done  26 tasks      | elapsed:   39.3s\n",
      "[Parallel(n_jobs=-1)]: Done  27 tasks      | elapsed:   39.3s\n",
      "[Parallel(n_jobs=-1)]: Done  38 out of  40 | elapsed:   39.3s remaining:    2.0s\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-429-c04fc3a23fe8>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0msearch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0my_train\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py\u001b[0m in \u001b[0;36mfit\u001b[1;34m(self, X, y, groups, **fit_params)\u001b[0m\n\u001b[0;32m    708\u001b[0m                 \u001b[1;32mreturn\u001b[0m \u001b[0mresults\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    709\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 710\u001b[1;33m             \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_run_search\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mevaluate_candidates\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    711\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    712\u001b[0m         \u001b[1;31m# For multi-metric evaluation, store the best_index_, best_params_ and\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py\u001b[0m in \u001b[0;36m_run_search\u001b[1;34m(self, evaluate_candidates)\u001b[0m\n\u001b[0;32m   1482\u001b[0m         evaluate_candidates(ParameterSampler(\n\u001b[0;32m   1483\u001b[0m             \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mparam_distributions\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mn_iter\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1484\u001b[1;33m             random_state=self.random_state))\n\u001b[0m",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py\u001b[0m in \u001b[0;36mevaluate_candidates\u001b[1;34m(candidate_params)\u001b[0m\n\u001b[0;32m    687\u001b[0m                                \u001b[1;32mfor\u001b[0m \u001b[0mparameters\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m(\u001b[0m\u001b[0mtrain\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtest\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    688\u001b[0m                                in product(candidate_params,\n\u001b[1;32m--> 689\u001b[1;33m                                           cv.split(X, y, groups)))\n\u001b[0m\u001b[0;32m    690\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    691\u001b[0m                 \u001b[1;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mout\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m<\u001b[0m \u001b[1;36m1\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\joblib\\parallel.py\u001b[0m in \u001b[0;36m__call__\u001b[1;34m(self, iterable)\u001b[0m\n\u001b[0;32m   1015\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1016\u001b[0m             \u001b[1;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mretrieval_context\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1017\u001b[1;33m                 \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mretrieve\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1018\u001b[0m             \u001b[1;31m# Make sure that we get a last message telling us we are done\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1019\u001b[0m             \u001b[0melapsed_time\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_start_time\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\joblib\\parallel.py\u001b[0m in \u001b[0;36mretrieve\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m    907\u001b[0m             \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    908\u001b[0m                 \u001b[1;32mif\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'supports_timeout'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;32mFalse\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 909\u001b[1;33m                     \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_output\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mextend\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mjob\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    910\u001b[0m                 \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    911\u001b[0m                     \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_output\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mextend\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mjob\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\joblib\\_parallel_backends.py\u001b[0m in \u001b[0;36mwrap_future_result\u001b[1;34m(future, timeout)\u001b[0m\n\u001b[0;32m    560\u001b[0m         AsyncResults.get from multiprocessing.\"\"\"\n\u001b[0;32m    561\u001b[0m         \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 562\u001b[1;33m             \u001b[1;32mreturn\u001b[0m \u001b[0mfuture\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mresult\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    563\u001b[0m         \u001b[1;32mexcept\u001b[0m \u001b[0mLokyTimeoutError\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    564\u001b[0m             \u001b[1;32mraise\u001b[0m \u001b[0mTimeoutError\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\concurrent\\futures\\_base.py\u001b[0m in \u001b[0;36mresult\u001b[1;34m(self, timeout)\u001b[0m\n\u001b[0;32m    428\u001b[0m                 \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m__get_result\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    429\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 430\u001b[1;33m             \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_condition\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mwait\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    431\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    432\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_state\u001b[0m \u001b[1;32min\u001b[0m \u001b[1;33m[\u001b[0m\u001b[0mCANCELLED\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mCANCELLED_AND_NOTIFIED\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\threading.py\u001b[0m in \u001b[0;36mwait\u001b[1;34m(self, timeout)\u001b[0m\n\u001b[0;32m    294\u001b[0m         \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m    \u001b[1;31m# restore state no matter what (e.g., KeyboardInterrupt)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    295\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mtimeout\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 296\u001b[1;33m                 \u001b[0mwaiter\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0macquire\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    297\u001b[0m                 \u001b[0mgotit\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;32mTrue\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    298\u001b[0m             \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "search.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'learning_rate': 0.017}"
      ]
     },
     "execution_count": 174,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.05427853762690869"
      ]
     },
     "execution_count": 175,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "metadata": {},
   "outputs": [
    {
     "ename": "AttributeError",
     "evalue": "'RandomizedSearchCV' object has no attribute 'cv_results_'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-170-caa5ef9be8fb>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0msearch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcv_results_\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;31mAttributeError\u001b[0m: 'RandomizedSearchCV' object has no attribute 'cv_results_'"
     ]
    }
   ],
   "source": [
    "search.cv_results_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 437,
   "metadata": {},
   "outputs": [],
   "source": [
    "lightgbm = LGBMRegressor( task= 'train',\n",
    "          boosting_type= 'gbdt',\n",
    "          objective= 'regression',\n",
    "          metric= 'l2',\n",
    "          num_leaves= 300,\n",
    "          learning_rate= 0.32,\n",
    "          feature_fraction= 0.9,\n",
    "          bagging_fraction= .9,\n",
    "          bagging_freq= 70,\n",
    "          verbose= 100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 453,
   "metadata": {},
   "outputs": [],
   "source": [
    "booster=LGBMRegressor( task= 'train',\n",
    "          boosting_type= 'gbdt',\n",
    "          objective= 'regression',\n",
    "          metric= 'l1',\n",
    "          num_leaves= 300,\n",
    "          learning_rate= 0.32,  \n",
    "          bagging_fraction= .9,\n",
    "          bagging_freq= 70,\n",
    "          verbose= 100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 454,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[LightGBM] [Warning] bagging_fraction is set=0.9, subsample=1.0 will be ignored. Current value: bagging_fraction=0.9\n",
      "[LightGBM] [Warning] bagging_freq is set=70, subsample_freq=0 will be ignored. Current value: bagging_freq=70\n",
      "[LightGBM] [Warning] bagging_fraction is set=0.9, subsample=1.0 will be ignored. Current value: bagging_fraction=0.9\n",
      "[LightGBM] [Warning] bagging_freq is set=70, subsample_freq=0 will be ignored. Current value: bagging_freq=70\n",
      "[LightGBM] [Debug] Dataset::GetMultiBinFromSparseFeatures: sparse rate 0.793997\n",
      "[LightGBM] [Debug] Dataset::GetMultiBinFromAllFeatures: sparse rate 0.239815\n",
      "[LightGBM] [Debug] init for col-wise cost 0.022439 seconds, init for row-wise cost 0.042293 seconds\n",
      "[LightGBM] [Debug] col-wise cost 0.009825 seconds, row-wise cost 0.003463 seconds\n",
      "[LightGBM] [Warning] Auto-choosing row-wise multi-threading, the overhead of testing was 0.035727 seconds.\n",
      "You can set `force_row_wise=true` to remove the overhead.\n",
      "And if memory is not enough, you can set `force_col_wise=true`.\n",
      "[LightGBM] [Debug] Using Dense Multi-Val Bin\n",
      "[LightGBM] [Info] Total Bins 1042\n",
      "[LightGBM] [Info] Number of data points in the train set: 512087, number of used features: 11\n",
      "[LightGBM] [Debug] Use subset for bagging\n",
      "[LightGBM] [Info] Start training from score 4.705091\n",
      "[LightGBM] [Debug] Re-bagging, using 460970 data to train\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 12\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 13\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 16\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 15\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 16\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 18\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 21\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 16\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 17\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 19\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 18\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 22\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 18\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 21\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 21\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 20\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 21\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 27\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 32\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 22\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 22\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 30\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 20\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 27\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 34\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 21\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 27\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 24\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 30\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 33\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 25\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 23\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 24\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 22\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 26\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 28\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 24\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 29\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 25\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 17\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 23\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 28\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 26\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 27\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 37\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 31\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 31\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 26\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 30\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 28\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 35\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 26\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 21\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 20\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 26\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 26\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 24\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 39\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 26\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 22\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 29\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 19\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 27\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 22\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 29\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 25\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 25\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 24\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 22\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 22\n",
      "[LightGBM] [Debug] Re-bagging, using 460653 data to train\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 19\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 20\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 24\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 23\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 25\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 25\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 27\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 25\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 23\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 28\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 24\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 23\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 22\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 26\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 25\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 22\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 23\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 26\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 24\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 38\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 41\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 23\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 22\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 29\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 20\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 21\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 21\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 19\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 24\n",
      "[LightGBM] [Debug] Trained a tree with leaves = 300 and max_depth = 24\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LGBMRegressor(bagging_fraction=0.9, bagging_freq=70, boosting_type='gbdt',\n",
       "              class_weight=None, colsample_bytree=1.0, importance_type='split',\n",
       "              learning_rate=0.32, max_depth=-1, metric='l1',\n",
       "              min_child_samples=20, min_child_weight=0.001, min_split_gain=0.0,\n",
       "              n_estimators=100, n_jobs=-1, num_leaves=300,\n",
       "              objective='regression', random_state=None, reg_alpha=0.0,\n",
       "              reg_lambda=0.0, silent=True, subsample=1.0,\n",
       "              subsample_for_bin=200000, subsample_freq=0, task='train',\n",
       "              verbose=100)"
      ]
     },
     "execution_count": 454,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "booster.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 455,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_pred=booster.predict(X_train)\n",
    "y_test_pred=booster.predict(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 456,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "109.82900129001581"
      ]
     },
     "execution_count": 456,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1000*np.sqrt(mean_squared_error(y_train,y_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 458,
   "metadata": {},
   "outputs": [],
   "source": [
    "export_submission(y_test_pred,\"LGBM Optimized Engineered Features\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler=MinMaxScaler()\n",
    "X_train=scaler.fit_transform(X_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "knn=KNeighborsRegressor(n_neighbors= 17,n_jobs=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "knn.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_pred=forest.predict(X_train)\n",
    "y_test_pred=forest.predict(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "216.13655175210727"
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1000*np.sqrt(mean_squared_error(y_train,y_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [],
   "source": [
    "export_submission(y_test_pred,\"KNN 17 Engineered Features\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [],
   "source": [
    "parameters={'n_neighbors':range(1,50,1)}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [],
   "source": [
    "search=GridSearchCV(KNeighborsRegressor(n_jobs=6),parameters,scoring='neg_mean_squared_error',cv=5,verbose=2,n_jobs=6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 49 candidates, totalling 245 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=10)]: Using backend LokyBackend with 10 concurrent workers.\n",
      "[Parallel(n_jobs=10)]: Done  21 tasks      | elapsed:   30.4s\n",
      "[Parallel(n_jobs=10)]: Done 142 tasks      | elapsed:  3.2min\n",
      "[Parallel(n_jobs=10)]: Done 245 out of 245 | elapsed:  5.7min finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score=nan,\n",
       "             estimator=KNeighborsRegressor(algorithm='auto', leaf_size=30,\n",
       "                                           metric='minkowski',\n",
       "                                           metric_params=None, n_jobs=None,\n",
       "                                           n_neighbors=5, p=2,\n",
       "                                           weights='uniform'),\n",
       "             iid='deprecated', n_jobs=10,\n",
       "             param_grid={'n_neighbors': range(1, 50)}, pre_dispatch='2*n_jobs',\n",
       "             refit=True, return_train_score=False,\n",
       "             scoring='neg_mean_squared_error', verbose=2)"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'n_neighbors': 17}"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.06775012845127092"
      ]
     },
     "execution_count": 135,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Catboost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {},
   "outputs": [],
   "source": [
    "boost=CatBoostRegressor(learning_rate=0.01)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0:\tlearn: 0.4576319\ttotal: 47.7ms\tremaining: 47.6s\n",
      "1:\tlearn: 0.4539557\ttotal: 92.2ms\tremaining: 46s\n",
      "2:\tlearn: 0.4503334\ttotal: 140ms\tremaining: 46.6s\n",
      "3:\tlearn: 0.4467243\ttotal: 192ms\tremaining: 47.7s\n",
      "4:\tlearn: 0.4431456\ttotal: 247ms\tremaining: 49.2s\n",
      "5:\tlearn: 0.4395896\ttotal: 308ms\tremaining: 51s\n",
      "6:\tlearn: 0.4360925\ttotal: 358ms\tremaining: 50.8s\n",
      "7:\tlearn: 0.4326444\ttotal: 409ms\tremaining: 50.7s\n",
      "8:\tlearn: 0.4292267\ttotal: 462ms\tremaining: 50.8s\n",
      "9:\tlearn: 0.4258651\ttotal: 512ms\tremaining: 50.7s\n",
      "10:\tlearn: 0.4225387\ttotal: 563ms\tremaining: 50.6s\n",
      "11:\tlearn: 0.4192125\ttotal: 614ms\tremaining: 50.5s\n",
      "12:\tlearn: 0.4159608\ttotal: 662ms\tremaining: 50.3s\n",
      "13:\tlearn: 0.4127267\ttotal: 710ms\tremaining: 50s\n",
      "14:\tlearn: 0.4095260\ttotal: 761ms\tremaining: 50s\n",
      "15:\tlearn: 0.4063475\ttotal: 813ms\tremaining: 50s\n",
      "16:\tlearn: 0.4032326\ttotal: 858ms\tremaining: 49.6s\n",
      "17:\tlearn: 0.4001225\ttotal: 906ms\tremaining: 49.4s\n",
      "18:\tlearn: 0.3970596\ttotal: 952ms\tremaining: 49.1s\n",
      "19:\tlearn: 0.3940225\ttotal: 1s\tremaining: 49.2s\n",
      "20:\tlearn: 0.3910309\ttotal: 1.05s\tremaining: 48.9s\n",
      "21:\tlearn: 0.3880544\ttotal: 1.09s\tremaining: 48.7s\n",
      "22:\tlearn: 0.3851025\ttotal: 1.14s\tremaining: 48.5s\n",
      "23:\tlearn: 0.3821643\ttotal: 1.19s\tremaining: 48.5s\n",
      "24:\tlearn: 0.3792916\ttotal: 1.24s\tremaining: 48.2s\n",
      "25:\tlearn: 0.3764522\ttotal: 1.29s\tremaining: 48.3s\n",
      "26:\tlearn: 0.3736548\ttotal: 1.34s\tremaining: 48.2s\n",
      "27:\tlearn: 0.3708520\ttotal: 1.39s\tremaining: 48.1s\n",
      "28:\tlearn: 0.3681167\ttotal: 1.44s\tremaining: 48.1s\n",
      "29:\tlearn: 0.3653685\ttotal: 1.49s\tremaining: 48.1s\n",
      "30:\tlearn: 0.3626623\ttotal: 1.53s\tremaining: 48s\n",
      "31:\tlearn: 0.3600155\ttotal: 1.58s\tremaining: 48s\n",
      "32:\tlearn: 0.3574092\ttotal: 1.63s\tremaining: 47.9s\n",
      "33:\tlearn: 0.3548097\ttotal: 1.68s\tremaining: 47.8s\n",
      "34:\tlearn: 0.3522468\ttotal: 1.73s\tremaining: 47.6s\n",
      "35:\tlearn: 0.3496808\ttotal: 1.78s\tremaining: 47.6s\n",
      "36:\tlearn: 0.3471447\ttotal: 1.82s\tremaining: 47.4s\n",
      "37:\tlearn: 0.3446378\ttotal: 1.87s\tremaining: 47.4s\n",
      "38:\tlearn: 0.3421839\ttotal: 1.92s\tremaining: 47.3s\n",
      "39:\tlearn: 0.3397438\ttotal: 1.97s\tremaining: 47.2s\n",
      "40:\tlearn: 0.3373404\ttotal: 2.01s\tremaining: 47s\n",
      "41:\tlearn: 0.3349551\ttotal: 2.06s\tremaining: 47s\n",
      "42:\tlearn: 0.3325942\ttotal: 2.11s\tremaining: 46.9s\n",
      "43:\tlearn: 0.3302550\ttotal: 2.16s\tremaining: 46.9s\n",
      "44:\tlearn: 0.3279541\ttotal: 2.2s\tremaining: 46.8s\n",
      "45:\tlearn: 0.3256462\ttotal: 2.25s\tremaining: 46.6s\n",
      "46:\tlearn: 0.3233862\ttotal: 2.29s\tremaining: 46.5s\n",
      "47:\tlearn: 0.3211370\ttotal: 2.35s\tremaining: 46.7s\n",
      "48:\tlearn: 0.3189201\ttotal: 2.41s\tremaining: 46.7s\n",
      "49:\tlearn: 0.3167450\ttotal: 2.46s\tremaining: 46.7s\n",
      "50:\tlearn: 0.3145761\ttotal: 2.51s\tremaining: 46.8s\n",
      "51:\tlearn: 0.3124506\ttotal: 2.57s\tremaining: 46.9s\n",
      "52:\tlearn: 0.3103684\ttotal: 2.62s\tremaining: 46.9s\n",
      "53:\tlearn: 0.3083047\ttotal: 2.67s\tremaining: 46.8s\n",
      "54:\tlearn: 0.3062283\ttotal: 2.72s\tremaining: 46.8s\n",
      "55:\tlearn: 0.3042073\ttotal: 2.78s\tremaining: 46.9s\n",
      "56:\tlearn: 0.3021799\ttotal: 2.83s\tremaining: 46.8s\n",
      "57:\tlearn: 0.3001864\ttotal: 2.88s\tremaining: 46.8s\n",
      "58:\tlearn: 0.2982232\ttotal: 2.93s\tremaining: 46.8s\n",
      "59:\tlearn: 0.2962416\ttotal: 2.98s\tremaining: 46.7s\n",
      "60:\tlearn: 0.2942884\ttotal: 3.03s\tremaining: 46.6s\n",
      "61:\tlearn: 0.2924070\ttotal: 3.07s\tremaining: 46.5s\n",
      "62:\tlearn: 0.2905272\ttotal: 3.12s\tremaining: 46.4s\n",
      "63:\tlearn: 0.2886250\ttotal: 3.17s\tremaining: 46.3s\n",
      "64:\tlearn: 0.2867986\ttotal: 3.21s\tremaining: 46.2s\n",
      "65:\tlearn: 0.2849957\ttotal: 3.26s\tremaining: 46.1s\n",
      "66:\tlearn: 0.2831605\ttotal: 3.31s\tremaining: 46s\n",
      "67:\tlearn: 0.2813629\ttotal: 3.36s\tremaining: 46.1s\n",
      "68:\tlearn: 0.2796206\ttotal: 3.41s\tremaining: 46s\n",
      "69:\tlearn: 0.2779112\ttotal: 3.45s\tremaining: 45.9s\n",
      "70:\tlearn: 0.2761789\ttotal: 3.5s\tremaining: 45.8s\n",
      "71:\tlearn: 0.2744484\ttotal: 3.55s\tremaining: 45.7s\n",
      "72:\tlearn: 0.2727479\ttotal: 3.6s\tremaining: 45.7s\n",
      "73:\tlearn: 0.2710926\ttotal: 3.65s\tremaining: 45.6s\n",
      "74:\tlearn: 0.2694506\ttotal: 3.69s\tremaining: 45.5s\n",
      "75:\tlearn: 0.2678444\ttotal: 3.74s\tremaining: 45.4s\n",
      "76:\tlearn: 0.2662649\ttotal: 3.78s\tremaining: 45.3s\n",
      "77:\tlearn: 0.2647017\ttotal: 3.83s\tremaining: 45.2s\n",
      "78:\tlearn: 0.2631180\ttotal: 3.87s\tremaining: 45.2s\n",
      "79:\tlearn: 0.2615446\ttotal: 3.92s\tremaining: 45.1s\n",
      "80:\tlearn: 0.2600051\ttotal: 3.97s\tremaining: 45s\n",
      "81:\tlearn: 0.2585160\ttotal: 4.01s\tremaining: 45s\n",
      "82:\tlearn: 0.2570013\ttotal: 4.06s\tremaining: 44.9s\n",
      "83:\tlearn: 0.2555603\ttotal: 4.11s\tremaining: 44.8s\n",
      "84:\tlearn: 0.2540943\ttotal: 4.16s\tremaining: 44.8s\n",
      "85:\tlearn: 0.2526591\ttotal: 4.21s\tremaining: 44.7s\n",
      "86:\tlearn: 0.2512597\ttotal: 4.26s\tremaining: 44.7s\n",
      "87:\tlearn: 0.2498527\ttotal: 4.3s\tremaining: 44.6s\n",
      "88:\tlearn: 0.2484848\ttotal: 4.35s\tremaining: 44.5s\n",
      "89:\tlearn: 0.2470866\ttotal: 4.4s\tremaining: 44.5s\n",
      "90:\tlearn: 0.2457105\ttotal: 4.45s\tremaining: 44.4s\n",
      "91:\tlearn: 0.2443567\ttotal: 4.5s\tremaining: 44.4s\n",
      "92:\tlearn: 0.2430531\ttotal: 4.54s\tremaining: 44.3s\n",
      "93:\tlearn: 0.2417476\ttotal: 4.59s\tremaining: 44.2s\n",
      "94:\tlearn: 0.2404844\ttotal: 4.63s\tremaining: 44.2s\n",
      "95:\tlearn: 0.2392350\ttotal: 4.68s\tremaining: 44.1s\n",
      "96:\tlearn: 0.2380047\ttotal: 4.73s\tremaining: 44.1s\n",
      "97:\tlearn: 0.2367583\ttotal: 4.78s\tremaining: 44s\n",
      "98:\tlearn: 0.2355524\ttotal: 4.83s\tremaining: 43.9s\n",
      "99:\tlearn: 0.2343289\ttotal: 4.87s\tremaining: 43.9s\n",
      "100:\tlearn: 0.2331280\ttotal: 4.92s\tremaining: 43.8s\n",
      "101:\tlearn: 0.2319811\ttotal: 4.97s\tremaining: 43.7s\n",
      "102:\tlearn: 0.2308034\ttotal: 5.01s\tremaining: 43.7s\n",
      "103:\tlearn: 0.2296655\ttotal: 5.06s\tremaining: 43.6s\n",
      "104:\tlearn: 0.2285548\ttotal: 5.11s\tremaining: 43.5s\n",
      "105:\tlearn: 0.2274603\ttotal: 5.15s\tremaining: 43.5s\n",
      "106:\tlearn: 0.2263828\ttotal: 5.2s\tremaining: 43.4s\n",
      "107:\tlearn: 0.2252936\ttotal: 5.24s\tremaining: 43.3s\n",
      "108:\tlearn: 0.2242273\ttotal: 5.29s\tremaining: 43.2s\n",
      "109:\tlearn: 0.2231462\ttotal: 5.34s\tremaining: 43.2s\n",
      "110:\tlearn: 0.2220969\ttotal: 5.38s\tremaining: 43.1s\n",
      "111:\tlearn: 0.2210813\ttotal: 5.43s\tremaining: 43.1s\n",
      "112:\tlearn: 0.2200735\ttotal: 5.48s\tremaining: 43s\n",
      "113:\tlearn: 0.2190638\ttotal: 5.53s\tremaining: 43s\n",
      "114:\tlearn: 0.2180873\ttotal: 5.58s\tremaining: 42.9s\n",
      "115:\tlearn: 0.2171155\ttotal: 5.63s\tremaining: 42.9s\n",
      "116:\tlearn: 0.2161626\ttotal: 5.68s\tremaining: 42.8s\n",
      "117:\tlearn: 0.2152315\ttotal: 5.73s\tremaining: 42.8s\n",
      "118:\tlearn: 0.2142845\ttotal: 5.78s\tremaining: 42.8s\n",
      "119:\tlearn: 0.2133158\ttotal: 5.84s\tremaining: 42.8s\n",
      "120:\tlearn: 0.2123661\ttotal: 5.89s\tremaining: 42.8s\n",
      "121:\tlearn: 0.2114318\ttotal: 5.93s\tremaining: 42.7s\n",
      "122:\tlearn: 0.2105185\ttotal: 5.98s\tremaining: 42.6s\n",
      "123:\tlearn: 0.2096429\ttotal: 6.03s\tremaining: 42.6s\n",
      "124:\tlearn: 0.2087945\ttotal: 6.1s\tremaining: 42.7s\n",
      "125:\tlearn: 0.2079610\ttotal: 6.15s\tremaining: 42.7s\n",
      "126:\tlearn: 0.2071160\ttotal: 6.2s\tremaining: 42.6s\n",
      "127:\tlearn: 0.2062719\ttotal: 6.25s\tremaining: 42.6s\n",
      "128:\tlearn: 0.2054319\ttotal: 6.3s\tremaining: 42.6s\n",
      "129:\tlearn: 0.2046430\ttotal: 6.35s\tremaining: 42.5s\n",
      "130:\tlearn: 0.2038154\ttotal: 6.4s\tremaining: 42.5s\n",
      "131:\tlearn: 0.2030321\ttotal: 6.46s\tremaining: 42.5s\n",
      "132:\tlearn: 0.2022743\ttotal: 6.52s\tremaining: 42.5s\n",
      "133:\tlearn: 0.2015416\ttotal: 6.57s\tremaining: 42.4s\n",
      "134:\tlearn: 0.2007974\ttotal: 6.61s\tremaining: 42.4s\n",
      "135:\tlearn: 0.2000155\ttotal: 6.66s\tremaining: 42.3s\n",
      "136:\tlearn: 0.1992823\ttotal: 6.71s\tremaining: 42.3s\n",
      "137:\tlearn: 0.1985693\ttotal: 6.76s\tremaining: 42.2s\n",
      "138:\tlearn: 0.1978545\ttotal: 6.8s\tremaining: 42.1s\n",
      "139:\tlearn: 0.1971134\ttotal: 6.86s\tremaining: 42.1s\n",
      "140:\tlearn: 0.1963923\ttotal: 6.91s\tremaining: 42.1s\n",
      "141:\tlearn: 0.1957213\ttotal: 6.96s\tremaining: 42s\n",
      "142:\tlearn: 0.1950594\ttotal: 7s\tremaining: 42s\n",
      "143:\tlearn: 0.1944192\ttotal: 7.05s\tremaining: 41.9s\n",
      "144:\tlearn: 0.1937797\ttotal: 7.09s\tremaining: 41.8s\n",
      "145:\tlearn: 0.1931343\ttotal: 7.14s\tremaining: 41.8s\n",
      "146:\tlearn: 0.1925058\ttotal: 7.19s\tremaining: 41.7s\n",
      "147:\tlearn: 0.1918836\ttotal: 7.23s\tremaining: 41.6s\n",
      "148:\tlearn: 0.1912843\ttotal: 7.28s\tremaining: 41.6s\n",
      "149:\tlearn: 0.1906852\ttotal: 7.33s\tremaining: 41.5s\n",
      "150:\tlearn: 0.1900568\ttotal: 7.37s\tremaining: 41.4s\n",
      "151:\tlearn: 0.1894840\ttotal: 7.42s\tremaining: 41.4s\n",
      "152:\tlearn: 0.1888976\ttotal: 7.46s\tremaining: 41.3s\n",
      "153:\tlearn: 0.1883280\ttotal: 7.51s\tremaining: 41.3s\n",
      "154:\tlearn: 0.1877690\ttotal: 7.56s\tremaining: 41.2s\n",
      "155:\tlearn: 0.1872177\ttotal: 7.6s\tremaining: 41.1s\n",
      "156:\tlearn: 0.1866428\ttotal: 7.65s\tremaining: 41.1s\n",
      "157:\tlearn: 0.1860773\ttotal: 7.69s\tremaining: 41s\n",
      "158:\tlearn: 0.1855111\ttotal: 7.74s\tremaining: 40.9s\n",
      "159:\tlearn: 0.1849928\ttotal: 7.78s\tremaining: 40.9s\n",
      "160:\tlearn: 0.1844831\ttotal: 7.83s\tremaining: 40.8s\n",
      "161:\tlearn: 0.1839904\ttotal: 7.87s\tremaining: 40.7s\n",
      "162:\tlearn: 0.1834552\ttotal: 7.92s\tremaining: 40.7s\n",
      "163:\tlearn: 0.1827794\ttotal: 7.98s\tremaining: 40.7s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "164:\tlearn: 0.1822514\ttotal: 8.04s\tremaining: 40.7s\n",
      "165:\tlearn: 0.1817667\ttotal: 8.09s\tremaining: 40.6s\n",
      "166:\tlearn: 0.1812595\ttotal: 8.14s\tremaining: 40.6s\n",
      "167:\tlearn: 0.1807526\ttotal: 8.19s\tremaining: 40.5s\n",
      "168:\tlearn: 0.1803009\ttotal: 8.24s\tremaining: 40.5s\n",
      "169:\tlearn: 0.1798649\ttotal: 8.29s\tremaining: 40.5s\n",
      "170:\tlearn: 0.1794144\ttotal: 8.34s\tremaining: 40.4s\n",
      "171:\tlearn: 0.1789758\ttotal: 8.38s\tremaining: 40.4s\n",
      "172:\tlearn: 0.1785404\ttotal: 8.43s\tremaining: 40.3s\n",
      "173:\tlearn: 0.1781191\ttotal: 8.48s\tremaining: 40.3s\n",
      "174:\tlearn: 0.1777024\ttotal: 8.54s\tremaining: 40.2s\n",
      "175:\tlearn: 0.1772936\ttotal: 8.58s\tremaining: 40.2s\n",
      "176:\tlearn: 0.1768856\ttotal: 8.63s\tremaining: 40.1s\n",
      "177:\tlearn: 0.1764968\ttotal: 8.68s\tremaining: 40.1s\n",
      "178:\tlearn: 0.1760992\ttotal: 8.72s\tremaining: 40s\n",
      "179:\tlearn: 0.1757170\ttotal: 8.77s\tremaining: 39.9s\n",
      "180:\tlearn: 0.1753414\ttotal: 8.81s\tremaining: 39.9s\n",
      "181:\tlearn: 0.1749706\ttotal: 8.86s\tremaining: 39.8s\n",
      "182:\tlearn: 0.1746019\ttotal: 8.91s\tremaining: 39.8s\n",
      "183:\tlearn: 0.1742488\ttotal: 8.96s\tremaining: 39.7s\n",
      "184:\tlearn: 0.1738486\ttotal: 9.01s\tremaining: 39.7s\n",
      "185:\tlearn: 0.1735020\ttotal: 9.06s\tremaining: 39.7s\n",
      "186:\tlearn: 0.1731624\ttotal: 9.11s\tremaining: 39.6s\n",
      "187:\tlearn: 0.1727755\ttotal: 9.16s\tremaining: 39.5s\n",
      "188:\tlearn: 0.1724454\ttotal: 9.2s\tremaining: 39.5s\n",
      "189:\tlearn: 0.1721103\ttotal: 9.25s\tremaining: 39.4s\n",
      "190:\tlearn: 0.1717930\ttotal: 9.3s\tremaining: 39.4s\n",
      "191:\tlearn: 0.1714771\ttotal: 9.36s\tremaining: 39.4s\n",
      "192:\tlearn: 0.1711604\ttotal: 9.41s\tremaining: 39.3s\n",
      "193:\tlearn: 0.1708434\ttotal: 9.46s\tremaining: 39.3s\n",
      "194:\tlearn: 0.1703283\ttotal: 9.51s\tremaining: 39.2s\n",
      "195:\tlearn: 0.1700319\ttotal: 9.56s\tremaining: 39.2s\n",
      "196:\tlearn: 0.1697359\ttotal: 9.61s\tremaining: 39.2s\n",
      "197:\tlearn: 0.1694473\ttotal: 9.66s\tremaining: 39.1s\n",
      "198:\tlearn: 0.1691508\ttotal: 9.71s\tremaining: 39.1s\n",
      "199:\tlearn: 0.1688596\ttotal: 9.75s\tremaining: 39s\n",
      "200:\tlearn: 0.1685803\ttotal: 9.8s\tremaining: 39s\n",
      "201:\tlearn: 0.1683144\ttotal: 9.85s\tremaining: 38.9s\n",
      "202:\tlearn: 0.1680469\ttotal: 9.89s\tremaining: 38.8s\n",
      "203:\tlearn: 0.1677791\ttotal: 9.93s\tremaining: 38.8s\n",
      "204:\tlearn: 0.1674194\ttotal: 9.98s\tremaining: 38.7s\n",
      "205:\tlearn: 0.1671157\ttotal: 10s\tremaining: 38.7s\n",
      "206:\tlearn: 0.1668702\ttotal: 10.1s\tremaining: 38.6s\n",
      "207:\tlearn: 0.1666204\ttotal: 10.1s\tremaining: 38.5s\n",
      "208:\tlearn: 0.1663787\ttotal: 10.2s\tremaining: 38.5s\n",
      "209:\tlearn: 0.1661411\ttotal: 10.2s\tremaining: 38.4s\n",
      "210:\tlearn: 0.1659006\ttotal: 10.3s\tremaining: 38.4s\n",
      "211:\tlearn: 0.1656135\ttotal: 10.3s\tremaining: 38.3s\n",
      "212:\tlearn: 0.1653840\ttotal: 10.3s\tremaining: 38.2s\n",
      "213:\tlearn: 0.1651594\ttotal: 10.4s\tremaining: 38.2s\n",
      "214:\tlearn: 0.1649352\ttotal: 10.4s\tremaining: 38.1s\n",
      "215:\tlearn: 0.1647108\ttotal: 10.5s\tremaining: 38s\n",
      "216:\tlearn: 0.1644969\ttotal: 10.5s\tremaining: 38s\n",
      "217:\tlearn: 0.1642827\ttotal: 10.6s\tremaining: 37.9s\n",
      "218:\tlearn: 0.1640715\ttotal: 10.6s\tremaining: 37.9s\n",
      "219:\tlearn: 0.1638649\ttotal: 10.7s\tremaining: 37.8s\n",
      "220:\tlearn: 0.1636665\ttotal: 10.7s\tremaining: 37.7s\n",
      "221:\tlearn: 0.1634661\ttotal: 10.8s\tremaining: 37.7s\n",
      "222:\tlearn: 0.1632734\ttotal: 10.8s\tremaining: 37.6s\n",
      "223:\tlearn: 0.1630773\ttotal: 10.8s\tremaining: 37.6s\n",
      "224:\tlearn: 0.1628881\ttotal: 10.9s\tremaining: 37.5s\n",
      "225:\tlearn: 0.1626041\ttotal: 10.9s\tremaining: 37.4s\n",
      "226:\tlearn: 0.1624298\ttotal: 11s\tremaining: 37.4s\n",
      "227:\tlearn: 0.1622411\ttotal: 11s\tremaining: 37.3s\n",
      "228:\tlearn: 0.1620580\ttotal: 11.1s\tremaining: 37.3s\n",
      "229:\tlearn: 0.1618843\ttotal: 11.1s\tremaining: 37.2s\n",
      "230:\tlearn: 0.1617072\ttotal: 11.2s\tremaining: 37.1s\n",
      "231:\tlearn: 0.1615317\ttotal: 11.2s\tremaining: 37.1s\n",
      "232:\tlearn: 0.1613672\ttotal: 11.2s\tremaining: 37s\n",
      "233:\tlearn: 0.1612053\ttotal: 11.3s\tremaining: 37s\n",
      "234:\tlearn: 0.1610452\ttotal: 11.3s\tremaining: 36.9s\n",
      "235:\tlearn: 0.1608277\ttotal: 11.4s\tremaining: 36.9s\n",
      "236:\tlearn: 0.1604328\ttotal: 11.4s\tremaining: 36.8s\n",
      "237:\tlearn: 0.1602799\ttotal: 11.5s\tremaining: 36.7s\n",
      "238:\tlearn: 0.1601272\ttotal: 11.5s\tremaining: 36.7s\n",
      "239:\tlearn: 0.1599812\ttotal: 11.6s\tremaining: 36.6s\n",
      "240:\tlearn: 0.1598341\ttotal: 11.6s\tremaining: 36.6s\n",
      "241:\tlearn: 0.1596932\ttotal: 11.7s\tremaining: 36.5s\n",
      "242:\tlearn: 0.1594455\ttotal: 11.7s\tremaining: 36.5s\n",
      "243:\tlearn: 0.1592550\ttotal: 11.8s\tremaining: 36.4s\n",
      "244:\tlearn: 0.1590574\ttotal: 11.8s\tremaining: 36.4s\n",
      "245:\tlearn: 0.1589184\ttotal: 11.9s\tremaining: 36.3s\n",
      "246:\tlearn: 0.1587785\ttotal: 11.9s\tremaining: 36.3s\n",
      "247:\tlearn: 0.1586493\ttotal: 11.9s\tremaining: 36.2s\n",
      "248:\tlearn: 0.1582860\ttotal: 12s\tremaining: 36.2s\n",
      "249:\tlearn: 0.1581567\ttotal: 12s\tremaining: 36.1s\n",
      "250:\tlearn: 0.1580309\ttotal: 12.1s\tremaining: 36.1s\n",
      "251:\tlearn: 0.1579055\ttotal: 12.1s\tremaining: 36s\n",
      "252:\tlearn: 0.1577825\ttotal: 12.2s\tremaining: 36s\n",
      "253:\tlearn: 0.1576664\ttotal: 12.2s\tremaining: 35.9s\n",
      "254:\tlearn: 0.1575523\ttotal: 12.3s\tremaining: 35.9s\n",
      "255:\tlearn: 0.1574334\ttotal: 12.3s\tremaining: 35.8s\n",
      "256:\tlearn: 0.1572646\ttotal: 12.4s\tremaining: 35.8s\n",
      "257:\tlearn: 0.1571511\ttotal: 12.4s\tremaining: 35.7s\n",
      "258:\tlearn: 0.1570244\ttotal: 12.5s\tremaining: 35.7s\n",
      "259:\tlearn: 0.1569115\ttotal: 12.5s\tremaining: 35.6s\n",
      "260:\tlearn: 0.1568022\ttotal: 12.6s\tremaining: 35.6s\n",
      "261:\tlearn: 0.1566938\ttotal: 12.6s\tremaining: 35.5s\n",
      "262:\tlearn: 0.1565308\ttotal: 12.7s\tremaining: 35.5s\n",
      "263:\tlearn: 0.1564300\ttotal: 12.7s\tremaining: 35.4s\n",
      "264:\tlearn: 0.1563276\ttotal: 12.7s\tremaining: 35.3s\n",
      "265:\tlearn: 0.1562256\ttotal: 12.8s\tremaining: 35.3s\n",
      "266:\tlearn: 0.1561267\ttotal: 12.8s\tremaining: 35.2s\n",
      "267:\tlearn: 0.1560320\ttotal: 12.9s\tremaining: 35.2s\n",
      "268:\tlearn: 0.1559362\ttotal: 12.9s\tremaining: 35.1s\n",
      "269:\tlearn: 0.1558450\ttotal: 13s\tremaining: 35s\n",
      "270:\tlearn: 0.1557453\ttotal: 13s\tremaining: 35s\n",
      "271:\tlearn: 0.1556568\ttotal: 13s\tremaining: 34.9s\n",
      "272:\tlearn: 0.1555702\ttotal: 13.1s\tremaining: 34.9s\n",
      "273:\tlearn: 0.1554713\ttotal: 13.1s\tremaining: 34.8s\n",
      "274:\tlearn: 0.1553820\ttotal: 13.2s\tremaining: 34.8s\n",
      "275:\tlearn: 0.1552986\ttotal: 13.2s\tremaining: 34.7s\n",
      "276:\tlearn: 0.1552181\ttotal: 13.3s\tremaining: 34.6s\n",
      "277:\tlearn: 0.1551395\ttotal: 13.3s\tremaining: 34.6s\n",
      "278:\tlearn: 0.1550610\ttotal: 13.4s\tremaining: 34.5s\n",
      "279:\tlearn: 0.1549644\ttotal: 13.4s\tremaining: 34.5s\n",
      "280:\tlearn: 0.1548823\ttotal: 13.5s\tremaining: 34.5s\n",
      "281:\tlearn: 0.1547874\ttotal: 13.5s\tremaining: 34.4s\n",
      "282:\tlearn: 0.1546548\ttotal: 13.6s\tremaining: 34.4s\n",
      "283:\tlearn: 0.1543469\ttotal: 13.6s\tremaining: 34.3s\n",
      "284:\tlearn: 0.1542554\ttotal: 13.7s\tremaining: 34.3s\n",
      "285:\tlearn: 0.1541802\ttotal: 13.7s\tremaining: 34.2s\n",
      "286:\tlearn: 0.1540934\ttotal: 13.8s\tremaining: 34.2s\n",
      "287:\tlearn: 0.1540197\ttotal: 13.8s\tremaining: 34.2s\n",
      "288:\tlearn: 0.1538733\ttotal: 13.9s\tremaining: 34.1s\n",
      "289:\tlearn: 0.1537101\ttotal: 13.9s\tremaining: 34.1s\n",
      "290:\tlearn: 0.1536435\ttotal: 14s\tremaining: 34s\n",
      "291:\tlearn: 0.1535604\ttotal: 14s\tremaining: 34s\n",
      "292:\tlearn: 0.1534799\ttotal: 14.1s\tremaining: 33.9s\n",
      "293:\tlearn: 0.1534143\ttotal: 14.1s\tremaining: 33.9s\n",
      "294:\tlearn: 0.1533355\ttotal: 14.2s\tremaining: 33.8s\n",
      "295:\tlearn: 0.1532715\ttotal: 14.2s\tremaining: 33.8s\n",
      "296:\tlearn: 0.1531586\ttotal: 14.2s\tremaining: 33.7s\n",
      "297:\tlearn: 0.1530948\ttotal: 14.3s\tremaining: 33.7s\n",
      "298:\tlearn: 0.1530198\ttotal: 14.3s\tremaining: 33.6s\n",
      "299:\tlearn: 0.1529612\ttotal: 14.4s\tremaining: 33.6s\n",
      "300:\tlearn: 0.1528961\ttotal: 14.4s\tremaining: 33.5s\n",
      "301:\tlearn: 0.1527709\ttotal: 14.5s\tremaining: 33.5s\n",
      "302:\tlearn: 0.1527041\ttotal: 14.5s\tremaining: 33.4s\n",
      "303:\tlearn: 0.1526448\ttotal: 14.6s\tremaining: 33.4s\n",
      "304:\tlearn: 0.1525430\ttotal: 14.6s\tremaining: 33.3s\n",
      "305:\tlearn: 0.1524833\ttotal: 14.7s\tremaining: 33.3s\n",
      "306:\tlearn: 0.1523388\ttotal: 14.7s\tremaining: 33.2s\n",
      "307:\tlearn: 0.1522839\ttotal: 14.8s\tremaining: 33.1s\n",
      "308:\tlearn: 0.1522341\ttotal: 14.8s\tremaining: 33.1s\n",
      "309:\tlearn: 0.1521682\ttotal: 14.8s\tremaining: 33s\n",
      "310:\tlearn: 0.1520093\ttotal: 14.9s\tremaining: 33s\n",
      "311:\tlearn: 0.1519076\ttotal: 14.9s\tremaining: 32.9s\n",
      "312:\tlearn: 0.1516302\ttotal: 15s\tremaining: 32.9s\n",
      "313:\tlearn: 0.1515673\ttotal: 15s\tremaining: 32.8s\n",
      "314:\tlearn: 0.1514313\ttotal: 15.1s\tremaining: 32.8s\n",
      "315:\tlearn: 0.1513825\ttotal: 15.1s\tremaining: 32.7s\n",
      "316:\tlearn: 0.1513386\ttotal: 15.2s\tremaining: 32.6s\n",
      "317:\tlearn: 0.1512489\ttotal: 15.2s\tremaining: 32.6s\n",
      "318:\tlearn: 0.1511544\ttotal: 15.2s\tremaining: 32.5s\n",
      "319:\tlearn: 0.1510442\ttotal: 15.3s\tremaining: 32.5s\n",
      "320:\tlearn: 0.1509881\ttotal: 15.3s\tremaining: 32.4s\n",
      "321:\tlearn: 0.1508969\ttotal: 15.4s\tremaining: 32.4s\n",
      "322:\tlearn: 0.1508430\ttotal: 15.4s\tremaining: 32.3s\n",
      "323:\tlearn: 0.1507167\ttotal: 15.5s\tremaining: 32.3s\n",
      "324:\tlearn: 0.1506703\ttotal: 15.5s\tremaining: 32.2s\n",
      "325:\tlearn: 0.1505433\ttotal: 15.6s\tremaining: 32.2s\n",
      "326:\tlearn: 0.1504023\ttotal: 15.6s\tremaining: 32.1s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "327:\tlearn: 0.1503545\ttotal: 15.7s\tremaining: 32.1s\n",
      "328:\tlearn: 0.1503109\ttotal: 15.7s\tremaining: 32.1s\n",
      "329:\tlearn: 0.1502683\ttotal: 15.8s\tremaining: 32s\n",
      "330:\tlearn: 0.1501448\ttotal: 15.8s\tremaining: 31.9s\n",
      "331:\tlearn: 0.1498822\ttotal: 15.9s\tremaining: 31.9s\n",
      "332:\tlearn: 0.1497633\ttotal: 15.9s\tremaining: 31.9s\n",
      "333:\tlearn: 0.1497168\ttotal: 15.9s\tremaining: 31.8s\n",
      "334:\tlearn: 0.1496026\ttotal: 16s\tremaining: 31.7s\n",
      "335:\tlearn: 0.1495647\ttotal: 16s\tremaining: 31.7s\n",
      "336:\tlearn: 0.1494886\ttotal: 16.1s\tremaining: 31.6s\n",
      "337:\tlearn: 0.1494545\ttotal: 16.1s\tremaining: 31.6s\n",
      "338:\tlearn: 0.1493424\ttotal: 16.2s\tremaining: 31.5s\n",
      "339:\tlearn: 0.1492453\ttotal: 16.2s\tremaining: 31.5s\n",
      "340:\tlearn: 0.1491364\ttotal: 16.3s\tremaining: 31.4s\n",
      "341:\tlearn: 0.1491015\ttotal: 16.3s\tremaining: 31.4s\n",
      "342:\tlearn: 0.1488331\ttotal: 16.4s\tremaining: 31.3s\n",
      "343:\tlearn: 0.1486446\ttotal: 16.4s\tremaining: 31.3s\n",
      "344:\tlearn: 0.1486044\ttotal: 16.5s\tremaining: 31.2s\n",
      "345:\tlearn: 0.1484741\ttotal: 16.5s\tremaining: 31.2s\n",
      "346:\tlearn: 0.1482333\ttotal: 16.5s\tremaining: 31.1s\n",
      "347:\tlearn: 0.1481330\ttotal: 16.6s\tremaining: 31.1s\n",
      "348:\tlearn: 0.1480309\ttotal: 16.6s\tremaining: 31s\n",
      "349:\tlearn: 0.1479638\ttotal: 16.7s\tremaining: 31s\n",
      "350:\tlearn: 0.1479302\ttotal: 16.7s\tremaining: 30.9s\n",
      "351:\tlearn: 0.1478914\ttotal: 16.8s\tremaining: 30.9s\n",
      "352:\tlearn: 0.1478517\ttotal: 16.8s\tremaining: 30.9s\n",
      "353:\tlearn: 0.1478126\ttotal: 16.9s\tremaining: 30.8s\n",
      "354:\tlearn: 0.1477165\ttotal: 16.9s\tremaining: 30.7s\n",
      "355:\tlearn: 0.1476230\ttotal: 17s\tremaining: 30.7s\n",
      "356:\tlearn: 0.1475976\ttotal: 17s\tremaining: 30.6s\n",
      "357:\tlearn: 0.1475692\ttotal: 17.1s\tremaining: 30.6s\n",
      "358:\tlearn: 0.1474497\ttotal: 17.1s\tremaining: 30.5s\n",
      "359:\tlearn: 0.1474203\ttotal: 17.1s\tremaining: 30.5s\n",
      "360:\tlearn: 0.1472074\ttotal: 17.2s\tremaining: 30.4s\n",
      "361:\tlearn: 0.1470958\ttotal: 17.2s\tremaining: 30.4s\n",
      "362:\tlearn: 0.1470067\ttotal: 17.3s\tremaining: 30.3s\n",
      "363:\tlearn: 0.1468448\ttotal: 17.3s\tremaining: 30.3s\n",
      "364:\tlearn: 0.1467933\ttotal: 17.4s\tremaining: 30.3s\n",
      "365:\tlearn: 0.1467072\ttotal: 17.4s\tremaining: 30.2s\n",
      "366:\tlearn: 0.1466529\ttotal: 17.5s\tremaining: 30.2s\n",
      "367:\tlearn: 0.1465722\ttotal: 17.5s\tremaining: 30.1s\n",
      "368:\tlearn: 0.1465419\ttotal: 17.6s\tremaining: 30.1s\n",
      "369:\tlearn: 0.1463160\ttotal: 17.6s\tremaining: 30s\n",
      "370:\tlearn: 0.1461138\ttotal: 17.7s\tremaining: 30s\n",
      "371:\tlearn: 0.1460311\ttotal: 17.7s\tremaining: 29.9s\n",
      "372:\tlearn: 0.1459325\ttotal: 17.8s\tremaining: 29.9s\n",
      "373:\tlearn: 0.1458828\ttotal: 17.8s\tremaining: 29.9s\n",
      "374:\tlearn: 0.1458633\ttotal: 17.9s\tremaining: 29.8s\n",
      "375:\tlearn: 0.1457547\ttotal: 17.9s\tremaining: 29.7s\n",
      "376:\tlearn: 0.1456793\ttotal: 18s\tremaining: 29.7s\n",
      "377:\tlearn: 0.1454073\ttotal: 18s\tremaining: 29.6s\n",
      "378:\tlearn: 0.1452198\ttotal: 18.1s\tremaining: 29.6s\n",
      "379:\tlearn: 0.1449965\ttotal: 18.1s\tremaining: 29.5s\n",
      "380:\tlearn: 0.1447620\ttotal: 18.2s\tremaining: 29.5s\n",
      "381:\tlearn: 0.1446679\ttotal: 18.2s\tremaining: 29.5s\n",
      "382:\tlearn: 0.1446457\ttotal: 18.3s\tremaining: 29.4s\n",
      "383:\tlearn: 0.1444603\ttotal: 18.3s\tremaining: 29.4s\n",
      "384:\tlearn: 0.1444384\ttotal: 18.3s\tremaining: 29.3s\n",
      "385:\tlearn: 0.1442564\ttotal: 18.4s\tremaining: 29.3s\n",
      "386:\tlearn: 0.1442112\ttotal: 18.4s\tremaining: 29.2s\n",
      "387:\tlearn: 0.1441842\ttotal: 18.5s\tremaining: 29.1s\n",
      "388:\tlearn: 0.1440916\ttotal: 18.5s\tremaining: 29.1s\n",
      "389:\tlearn: 0.1439962\ttotal: 18.6s\tremaining: 29s\n",
      "390:\tlearn: 0.1439224\ttotal: 18.6s\tremaining: 29s\n",
      "391:\tlearn: 0.1438949\ttotal: 18.7s\tremaining: 28.9s\n",
      "392:\tlearn: 0.1437134\ttotal: 18.7s\tremaining: 28.9s\n",
      "393:\tlearn: 0.1436464\ttotal: 18.7s\tremaining: 28.8s\n",
      "394:\tlearn: 0.1435184\ttotal: 18.8s\tremaining: 28.8s\n",
      "395:\tlearn: 0.1432799\ttotal: 18.8s\tremaining: 28.7s\n",
      "396:\tlearn: 0.1431645\ttotal: 18.9s\tremaining: 28.7s\n",
      "397:\tlearn: 0.1431018\ttotal: 18.9s\tremaining: 28.6s\n",
      "398:\tlearn: 0.1430844\ttotal: 19s\tremaining: 28.6s\n",
      "399:\tlearn: 0.1430226\ttotal: 19s\tremaining: 28.6s\n",
      "400:\tlearn: 0.1428091\ttotal: 19.1s\tremaining: 28.5s\n",
      "401:\tlearn: 0.1427450\ttotal: 19.1s\tremaining: 28.5s\n",
      "402:\tlearn: 0.1425847\ttotal: 19.2s\tremaining: 28.4s\n",
      "403:\tlearn: 0.1425604\ttotal: 19.2s\tremaining: 28.4s\n",
      "404:\tlearn: 0.1422351\ttotal: 19.3s\tremaining: 28.3s\n",
      "405:\tlearn: 0.1420806\ttotal: 19.3s\tremaining: 28.3s\n",
      "406:\tlearn: 0.1420243\ttotal: 19.4s\tremaining: 28.3s\n",
      "407:\tlearn: 0.1418095\ttotal: 19.4s\tremaining: 28.2s\n",
      "408:\tlearn: 0.1416448\ttotal: 19.5s\tremaining: 28.2s\n",
      "409:\tlearn: 0.1415521\ttotal: 19.5s\tremaining: 28.1s\n",
      "410:\tlearn: 0.1413911\ttotal: 19.6s\tremaining: 28.1s\n",
      "411:\tlearn: 0.1412421\ttotal: 19.6s\tremaining: 28s\n",
      "412:\tlearn: 0.1412232\ttotal: 19.7s\tremaining: 28s\n",
      "413:\tlearn: 0.1410728\ttotal: 19.7s\tremaining: 27.9s\n",
      "414:\tlearn: 0.1408720\ttotal: 19.8s\tremaining: 27.9s\n",
      "415:\tlearn: 0.1407529\ttotal: 19.8s\tremaining: 27.8s\n",
      "416:\tlearn: 0.1406629\ttotal: 19.9s\tremaining: 27.8s\n",
      "417:\tlearn: 0.1406087\ttotal: 19.9s\tremaining: 27.7s\n",
      "418:\tlearn: 0.1404711\ttotal: 20s\tremaining: 27.7s\n",
      "419:\tlearn: 0.1403840\ttotal: 20s\tremaining: 27.6s\n",
      "420:\tlearn: 0.1401798\ttotal: 20.1s\tremaining: 27.6s\n",
      "421:\tlearn: 0.1399825\ttotal: 20.1s\tremaining: 27.5s\n",
      "422:\tlearn: 0.1398381\ttotal: 20.1s\tremaining: 27.5s\n",
      "423:\tlearn: 0.1397930\ttotal: 20.2s\tremaining: 27.4s\n",
      "424:\tlearn: 0.1394942\ttotal: 20.2s\tremaining: 27.4s\n",
      "425:\tlearn: 0.1393549\ttotal: 20.3s\tremaining: 27.3s\n",
      "426:\tlearn: 0.1391229\ttotal: 20.3s\tremaining: 27.3s\n",
      "427:\tlearn: 0.1389429\ttotal: 20.4s\tremaining: 27.2s\n",
      "428:\tlearn: 0.1388238\ttotal: 20.4s\tremaining: 27.2s\n",
      "429:\tlearn: 0.1385433\ttotal: 20.5s\tremaining: 27.1s\n",
      "430:\tlearn: 0.1383625\ttotal: 20.5s\tremaining: 27.1s\n",
      "431:\tlearn: 0.1381367\ttotal: 20.6s\tremaining: 27.1s\n",
      "432:\tlearn: 0.1380855\ttotal: 20.6s\tremaining: 27s\n",
      "433:\tlearn: 0.1379742\ttotal: 20.7s\tremaining: 27s\n",
      "434:\tlearn: 0.1377930\ttotal: 20.7s\tremaining: 26.9s\n",
      "435:\tlearn: 0.1375400\ttotal: 20.8s\tremaining: 26.9s\n",
      "436:\tlearn: 0.1374382\ttotal: 20.8s\tremaining: 26.8s\n",
      "437:\tlearn: 0.1373897\ttotal: 20.9s\tremaining: 26.8s\n",
      "438:\tlearn: 0.1373385\ttotal: 20.9s\tremaining: 26.8s\n",
      "439:\tlearn: 0.1373088\ttotal: 21s\tremaining: 26.7s\n",
      "440:\tlearn: 0.1372944\ttotal: 21s\tremaining: 26.7s\n",
      "441:\tlearn: 0.1371835\ttotal: 21.1s\tremaining: 26.6s\n",
      "442:\tlearn: 0.1371098\ttotal: 21.1s\tremaining: 26.6s\n",
      "443:\tlearn: 0.1369390\ttotal: 21.2s\tremaining: 26.5s\n",
      "444:\tlearn: 0.1369030\ttotal: 21.2s\tremaining: 26.5s\n",
      "445:\tlearn: 0.1367439\ttotal: 21.3s\tremaining: 26.4s\n",
      "446:\tlearn: 0.1365917\ttotal: 21.3s\tremaining: 26.4s\n",
      "447:\tlearn: 0.1364481\ttotal: 21.4s\tremaining: 26.3s\n",
      "448:\tlearn: 0.1362868\ttotal: 21.4s\tremaining: 26.3s\n",
      "449:\tlearn: 0.1361944\ttotal: 21.5s\tremaining: 26.3s\n",
      "450:\tlearn: 0.1360937\ttotal: 21.5s\tremaining: 26.2s\n",
      "451:\tlearn: 0.1359937\ttotal: 21.6s\tremaining: 26.2s\n",
      "452:\tlearn: 0.1358447\ttotal: 21.6s\tremaining: 26.1s\n",
      "453:\tlearn: 0.1356870\ttotal: 21.7s\tremaining: 26.1s\n",
      "454:\tlearn: 0.1355891\ttotal: 21.7s\tremaining: 26s\n",
      "455:\tlearn: 0.1355070\ttotal: 21.8s\tremaining: 26s\n",
      "456:\tlearn: 0.1353719\ttotal: 21.8s\tremaining: 25.9s\n",
      "457:\tlearn: 0.1353394\ttotal: 21.9s\tremaining: 25.9s\n",
      "458:\tlearn: 0.1352465\ttotal: 21.9s\tremaining: 25.9s\n",
      "459:\tlearn: 0.1351040\ttotal: 22s\tremaining: 25.8s\n",
      "460:\tlearn: 0.1350369\ttotal: 22.1s\tremaining: 25.8s\n",
      "461:\tlearn: 0.1348879\ttotal: 22.1s\tremaining: 25.7s\n",
      "462:\tlearn: 0.1347359\ttotal: 22.2s\tremaining: 25.7s\n",
      "463:\tlearn: 0.1346557\ttotal: 22.2s\tremaining: 25.7s\n",
      "464:\tlearn: 0.1346404\ttotal: 22.3s\tremaining: 25.6s\n",
      "465:\tlearn: 0.1346133\ttotal: 22.3s\tremaining: 25.6s\n",
      "466:\tlearn: 0.1345061\ttotal: 22.4s\tremaining: 25.5s\n",
      "467:\tlearn: 0.1343764\ttotal: 22.4s\tremaining: 25.5s\n",
      "468:\tlearn: 0.1342694\ttotal: 22.5s\tremaining: 25.4s\n",
      "469:\tlearn: 0.1341664\ttotal: 22.5s\tremaining: 25.4s\n",
      "470:\tlearn: 0.1340798\ttotal: 22.6s\tremaining: 25.3s\n",
      "471:\tlearn: 0.1339384\ttotal: 22.6s\tremaining: 25.3s\n",
      "472:\tlearn: 0.1338653\ttotal: 22.7s\tremaining: 25.2s\n",
      "473:\tlearn: 0.1336658\ttotal: 22.7s\tremaining: 25.2s\n",
      "474:\tlearn: 0.1334609\ttotal: 22.8s\tremaining: 25.2s\n",
      "475:\tlearn: 0.1333264\ttotal: 22.8s\tremaining: 25.1s\n",
      "476:\tlearn: 0.1332910\ttotal: 22.9s\tremaining: 25.1s\n",
      "477:\tlearn: 0.1332755\ttotal: 22.9s\tremaining: 25s\n",
      "478:\tlearn: 0.1330865\ttotal: 23s\tremaining: 25s\n",
      "479:\tlearn: 0.1330716\ttotal: 23s\tremaining: 24.9s\n",
      "480:\tlearn: 0.1330370\ttotal: 23.1s\tremaining: 24.9s\n",
      "481:\tlearn: 0.1329130\ttotal: 23.1s\tremaining: 24.9s\n",
      "482:\tlearn: 0.1328379\ttotal: 23.2s\tremaining: 24.8s\n",
      "483:\tlearn: 0.1327254\ttotal: 23.2s\tremaining: 24.8s\n",
      "484:\tlearn: 0.1325356\ttotal: 23.3s\tremaining: 24.7s\n",
      "485:\tlearn: 0.1324008\ttotal: 23.3s\tremaining: 24.7s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "486:\tlearn: 0.1323613\ttotal: 23.4s\tremaining: 24.6s\n",
      "487:\tlearn: 0.1322527\ttotal: 23.4s\tremaining: 24.6s\n",
      "488:\tlearn: 0.1321548\ttotal: 23.5s\tremaining: 24.5s\n",
      "489:\tlearn: 0.1320376\ttotal: 23.5s\tremaining: 24.5s\n",
      "490:\tlearn: 0.1318594\ttotal: 23.6s\tremaining: 24.4s\n",
      "491:\tlearn: 0.1317876\ttotal: 23.6s\tremaining: 24.4s\n",
      "492:\tlearn: 0.1316942\ttotal: 23.7s\tremaining: 24.3s\n",
      "493:\tlearn: 0.1316843\ttotal: 23.7s\tremaining: 24.3s\n",
      "494:\tlearn: 0.1315194\ttotal: 23.7s\tremaining: 24.2s\n",
      "495:\tlearn: 0.1314596\ttotal: 23.8s\tremaining: 24.2s\n",
      "496:\tlearn: 0.1312742\ttotal: 23.8s\tremaining: 24.1s\n",
      "497:\tlearn: 0.1312206\ttotal: 23.9s\tremaining: 24.1s\n",
      "498:\tlearn: 0.1311571\ttotal: 23.9s\tremaining: 24s\n",
      "499:\tlearn: 0.1310546\ttotal: 24s\tremaining: 24s\n",
      "500:\tlearn: 0.1308477\ttotal: 24s\tremaining: 23.9s\n",
      "501:\tlearn: 0.1307316\ttotal: 24.1s\tremaining: 23.9s\n",
      "502:\tlearn: 0.1306793\ttotal: 24.1s\tremaining: 23.8s\n",
      "503:\tlearn: 0.1305269\ttotal: 24.2s\tremaining: 23.8s\n",
      "504:\tlearn: 0.1304144\ttotal: 24.2s\tremaining: 23.8s\n",
      "505:\tlearn: 0.1302686\ttotal: 24.3s\tremaining: 23.7s\n",
      "506:\tlearn: 0.1302089\ttotal: 24.3s\tremaining: 23.7s\n",
      "507:\tlearn: 0.1301856\ttotal: 24.4s\tremaining: 23.6s\n",
      "508:\tlearn: 0.1301405\ttotal: 24.4s\tremaining: 23.6s\n",
      "509:\tlearn: 0.1300410\ttotal: 24.5s\tremaining: 23.5s\n",
      "510:\tlearn: 0.1299573\ttotal: 24.5s\tremaining: 23.5s\n",
      "511:\tlearn: 0.1298758\ttotal: 24.6s\tremaining: 23.4s\n",
      "512:\tlearn: 0.1298273\ttotal: 24.6s\tremaining: 23.4s\n",
      "513:\tlearn: 0.1297773\ttotal: 24.7s\tremaining: 23.3s\n",
      "514:\tlearn: 0.1297632\ttotal: 24.7s\tremaining: 23.3s\n",
      "515:\tlearn: 0.1297156\ttotal: 24.8s\tremaining: 23.3s\n",
      "516:\tlearn: 0.1296082\ttotal: 24.8s\tremaining: 23.2s\n",
      "517:\tlearn: 0.1295894\ttotal: 24.9s\tremaining: 23.2s\n",
      "518:\tlearn: 0.1295433\ttotal: 24.9s\tremaining: 23.1s\n",
      "519:\tlearn: 0.1294505\ttotal: 25s\tremaining: 23.1s\n",
      "520:\tlearn: 0.1293339\ttotal: 25.1s\tremaining: 23s\n",
      "521:\tlearn: 0.1292617\ttotal: 25.1s\tremaining: 23s\n",
      "522:\tlearn: 0.1290791\ttotal: 25.2s\tremaining: 22.9s\n",
      "523:\tlearn: 0.1290051\ttotal: 25.2s\tremaining: 22.9s\n",
      "524:\tlearn: 0.1289360\ttotal: 25.3s\tremaining: 22.9s\n",
      "525:\tlearn: 0.1288640\ttotal: 25.3s\tremaining: 22.8s\n",
      "526:\tlearn: 0.1288061\ttotal: 25.4s\tremaining: 22.8s\n",
      "527:\tlearn: 0.1287235\ttotal: 25.4s\tremaining: 22.7s\n",
      "528:\tlearn: 0.1286228\ttotal: 25.4s\tremaining: 22.7s\n",
      "529:\tlearn: 0.1285968\ttotal: 25.5s\tremaining: 22.6s\n",
      "530:\tlearn: 0.1285595\ttotal: 25.6s\tremaining: 22.6s\n",
      "531:\tlearn: 0.1285012\ttotal: 25.6s\tremaining: 22.5s\n",
      "532:\tlearn: 0.1284046\ttotal: 25.6s\tremaining: 22.5s\n",
      "533:\tlearn: 0.1282822\ttotal: 25.7s\tremaining: 22.4s\n",
      "534:\tlearn: 0.1281050\ttotal: 25.7s\tremaining: 22.4s\n",
      "535:\tlearn: 0.1280916\ttotal: 25.8s\tremaining: 22.3s\n",
      "536:\tlearn: 0.1280509\ttotal: 25.8s\tremaining: 22.3s\n",
      "537:\tlearn: 0.1280097\ttotal: 25.9s\tremaining: 22.2s\n",
      "538:\tlearn: 0.1279432\ttotal: 25.9s\tremaining: 22.2s\n",
      "539:\tlearn: 0.1278868\ttotal: 26s\tremaining: 22.1s\n",
      "540:\tlearn: 0.1278633\ttotal: 26s\tremaining: 22.1s\n",
      "541:\tlearn: 0.1276979\ttotal: 26.1s\tremaining: 22s\n",
      "542:\tlearn: 0.1275616\ttotal: 26.1s\tremaining: 22s\n",
      "543:\tlearn: 0.1274699\ttotal: 26.2s\tremaining: 21.9s\n",
      "544:\tlearn: 0.1273971\ttotal: 26.2s\tremaining: 21.9s\n",
      "545:\tlearn: 0.1273525\ttotal: 26.3s\tremaining: 21.8s\n",
      "546:\tlearn: 0.1273155\ttotal: 26.3s\tremaining: 21.8s\n",
      "547:\tlearn: 0.1272806\ttotal: 26.4s\tremaining: 21.7s\n",
      "548:\tlearn: 0.1271546\ttotal: 26.4s\tremaining: 21.7s\n",
      "549:\tlearn: 0.1271409\ttotal: 26.4s\tremaining: 21.6s\n",
      "550:\tlearn: 0.1270683\ttotal: 26.5s\tremaining: 21.6s\n",
      "551:\tlearn: 0.1269776\ttotal: 26.5s\tremaining: 21.5s\n",
      "552:\tlearn: 0.1268678\ttotal: 26.6s\tremaining: 21.5s\n",
      "553:\tlearn: 0.1267832\ttotal: 26.6s\tremaining: 21.4s\n",
      "554:\tlearn: 0.1267251\ttotal: 26.7s\tremaining: 21.4s\n",
      "555:\tlearn: 0.1266411\ttotal: 26.7s\tremaining: 21.3s\n",
      "556:\tlearn: 0.1266245\ttotal: 26.8s\tremaining: 21.3s\n",
      "557:\tlearn: 0.1265697\ttotal: 26.8s\tremaining: 21.2s\n",
      "558:\tlearn: 0.1265533\ttotal: 26.9s\tremaining: 21.2s\n",
      "559:\tlearn: 0.1265139\ttotal: 26.9s\tremaining: 21.1s\n",
      "560:\tlearn: 0.1264843\ttotal: 27s\tremaining: 21.1s\n",
      "561:\tlearn: 0.1264037\ttotal: 27s\tremaining: 21s\n",
      "562:\tlearn: 0.1262981\ttotal: 27.1s\tremaining: 21s\n",
      "563:\tlearn: 0.1261903\ttotal: 27.1s\tremaining: 21s\n",
      "564:\tlearn: 0.1261501\ttotal: 27.2s\tremaining: 20.9s\n",
      "565:\tlearn: 0.1261205\ttotal: 27.2s\tremaining: 20.9s\n",
      "566:\tlearn: 0.1261032\ttotal: 27.2s\tremaining: 20.8s\n",
      "567:\tlearn: 0.1260347\ttotal: 27.3s\tremaining: 20.8s\n",
      "568:\tlearn: 0.1260164\ttotal: 27.3s\tremaining: 20.7s\n",
      "569:\tlearn: 0.1258931\ttotal: 27.4s\tremaining: 20.7s\n",
      "570:\tlearn: 0.1258755\ttotal: 27.4s\tremaining: 20.6s\n",
      "571:\tlearn: 0.1258085\ttotal: 27.5s\tremaining: 20.6s\n",
      "572:\tlearn: 0.1257507\ttotal: 27.5s\tremaining: 20.5s\n",
      "573:\tlearn: 0.1257203\ttotal: 27.6s\tremaining: 20.5s\n",
      "574:\tlearn: 0.1256901\ttotal: 27.6s\tremaining: 20.4s\n",
      "575:\tlearn: 0.1256179\ttotal: 27.7s\tremaining: 20.4s\n",
      "576:\tlearn: 0.1255472\ttotal: 27.7s\tremaining: 20.3s\n",
      "577:\tlearn: 0.1254928\ttotal: 27.8s\tremaining: 20.3s\n",
      "578:\tlearn: 0.1254650\ttotal: 27.8s\tremaining: 20.2s\n",
      "579:\tlearn: 0.1254367\ttotal: 27.9s\tremaining: 20.2s\n",
      "580:\tlearn: 0.1253601\ttotal: 27.9s\tremaining: 20.1s\n",
      "581:\tlearn: 0.1252846\ttotal: 27.9s\tremaining: 20.1s\n",
      "582:\tlearn: 0.1252416\ttotal: 28s\tremaining: 20s\n",
      "583:\tlearn: 0.1251665\ttotal: 28s\tremaining: 20s\n",
      "584:\tlearn: 0.1250451\ttotal: 28.1s\tremaining: 19.9s\n",
      "585:\tlearn: 0.1249507\ttotal: 28.1s\tremaining: 19.9s\n",
      "586:\tlearn: 0.1249168\ttotal: 28.2s\tremaining: 19.8s\n",
      "587:\tlearn: 0.1248378\ttotal: 28.2s\tremaining: 19.8s\n",
      "588:\tlearn: 0.1247984\ttotal: 28.3s\tremaining: 19.7s\n",
      "589:\tlearn: 0.1247882\ttotal: 28.3s\tremaining: 19.7s\n",
      "590:\tlearn: 0.1247120\ttotal: 28.4s\tremaining: 19.6s\n",
      "591:\tlearn: 0.1246261\ttotal: 28.4s\tremaining: 19.6s\n",
      "592:\tlearn: 0.1246103\ttotal: 28.5s\tremaining: 19.5s\n",
      "593:\tlearn: 0.1245850\ttotal: 28.5s\tremaining: 19.5s\n",
      "594:\tlearn: 0.1245590\ttotal: 28.6s\tremaining: 19.4s\n",
      "595:\tlearn: 0.1244818\ttotal: 28.6s\tremaining: 19.4s\n",
      "596:\tlearn: 0.1244313\ttotal: 28.7s\tremaining: 19.4s\n",
      "597:\tlearn: 0.1243621\ttotal: 28.7s\tremaining: 19.3s\n",
      "598:\tlearn: 0.1242736\ttotal: 28.8s\tremaining: 19.3s\n",
      "599:\tlearn: 0.1242119\ttotal: 28.8s\tremaining: 19.2s\n",
      "600:\tlearn: 0.1241865\ttotal: 28.9s\tremaining: 19.2s\n",
      "601:\tlearn: 0.1241713\ttotal: 28.9s\tremaining: 19.1s\n",
      "602:\tlearn: 0.1241238\ttotal: 29s\tremaining: 19.1s\n",
      "603:\tlearn: 0.1240926\ttotal: 29s\tremaining: 19s\n",
      "604:\tlearn: 0.1240294\ttotal: 29.1s\tremaining: 19s\n",
      "605:\tlearn: 0.1240062\ttotal: 29.1s\tremaining: 18.9s\n",
      "606:\tlearn: 0.1239754\ttotal: 29.2s\tremaining: 18.9s\n",
      "607:\tlearn: 0.1238412\ttotal: 29.2s\tremaining: 18.8s\n",
      "608:\tlearn: 0.1237562\ttotal: 29.3s\tremaining: 18.8s\n",
      "609:\tlearn: 0.1237322\ttotal: 29.3s\tremaining: 18.7s\n",
      "610:\tlearn: 0.1236766\ttotal: 29.4s\tremaining: 18.7s\n",
      "611:\tlearn: 0.1235585\ttotal: 29.4s\tremaining: 18.6s\n",
      "612:\tlearn: 0.1235125\ttotal: 29.4s\tremaining: 18.6s\n",
      "613:\tlearn: 0.1234543\ttotal: 29.5s\tremaining: 18.5s\n",
      "614:\tlearn: 0.1234315\ttotal: 29.5s\tremaining: 18.5s\n",
      "615:\tlearn: 0.1234184\ttotal: 29.6s\tremaining: 18.4s\n",
      "616:\tlearn: 0.1233380\ttotal: 29.6s\tremaining: 18.4s\n",
      "617:\tlearn: 0.1233209\ttotal: 29.7s\tremaining: 18.3s\n",
      "618:\tlearn: 0.1232881\ttotal: 29.7s\tremaining: 18.3s\n",
      "619:\tlearn: 0.1232744\ttotal: 29.8s\tremaining: 18.2s\n",
      "620:\tlearn: 0.1232301\ttotal: 29.8s\tremaining: 18.2s\n",
      "621:\tlearn: 0.1232182\ttotal: 29.9s\tremaining: 18.1s\n",
      "622:\tlearn: 0.1231723\ttotal: 29.9s\tremaining: 18.1s\n",
      "623:\tlearn: 0.1230438\ttotal: 30s\tremaining: 18.1s\n",
      "624:\tlearn: 0.1229654\ttotal: 30s\tremaining: 18s\n",
      "625:\tlearn: 0.1229062\ttotal: 30.1s\tremaining: 18s\n",
      "626:\tlearn: 0.1227986\ttotal: 30.1s\tremaining: 17.9s\n",
      "627:\tlearn: 0.1227379\ttotal: 30.2s\tremaining: 17.9s\n",
      "628:\tlearn: 0.1227234\ttotal: 30.2s\tremaining: 17.8s\n",
      "629:\tlearn: 0.1226769\ttotal: 30.2s\tremaining: 17.8s\n",
      "630:\tlearn: 0.1226036\ttotal: 30.3s\tremaining: 17.7s\n",
      "631:\tlearn: 0.1225956\ttotal: 30.4s\tremaining: 17.7s\n",
      "632:\tlearn: 0.1225028\ttotal: 30.4s\tremaining: 17.6s\n",
      "633:\tlearn: 0.1224405\ttotal: 30.4s\tremaining: 17.6s\n",
      "634:\tlearn: 0.1223687\ttotal: 30.5s\tremaining: 17.5s\n",
      "635:\tlearn: 0.1223034\ttotal: 30.6s\tremaining: 17.5s\n",
      "636:\tlearn: 0.1222720\ttotal: 30.6s\tremaining: 17.4s\n",
      "637:\tlearn: 0.1222085\ttotal: 30.7s\tremaining: 17.4s\n",
      "638:\tlearn: 0.1221655\ttotal: 30.7s\tremaining: 17.3s\n",
      "639:\tlearn: 0.1221458\ttotal: 30.8s\tremaining: 17.3s\n",
      "640:\tlearn: 0.1220947\ttotal: 30.8s\tremaining: 17.3s\n",
      "641:\tlearn: 0.1220392\ttotal: 30.9s\tremaining: 17.2s\n",
      "642:\tlearn: 0.1220254\ttotal: 30.9s\tremaining: 17.2s\n",
      "643:\tlearn: 0.1219957\ttotal: 30.9s\tremaining: 17.1s\n",
      "644:\tlearn: 0.1219616\ttotal: 31s\tremaining: 17.1s\n",
      "645:\tlearn: 0.1219406\ttotal: 31s\tremaining: 17s\n",
      "646:\tlearn: 0.1218943\ttotal: 31.1s\tremaining: 17s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "647:\tlearn: 0.1218521\ttotal: 31.1s\tremaining: 16.9s\n",
      "648:\tlearn: 0.1217810\ttotal: 31.2s\tremaining: 16.9s\n",
      "649:\tlearn: 0.1217686\ttotal: 31.2s\tremaining: 16.8s\n",
      "650:\tlearn: 0.1217404\ttotal: 31.3s\tremaining: 16.8s\n",
      "651:\tlearn: 0.1216440\ttotal: 31.3s\tremaining: 16.7s\n",
      "652:\tlearn: 0.1216048\ttotal: 31.4s\tremaining: 16.7s\n",
      "653:\tlearn: 0.1215621\ttotal: 31.4s\tremaining: 16.6s\n",
      "654:\tlearn: 0.1215537\ttotal: 31.5s\tremaining: 16.6s\n",
      "655:\tlearn: 0.1215241\ttotal: 31.5s\tremaining: 16.5s\n",
      "656:\tlearn: 0.1214678\ttotal: 31.6s\tremaining: 16.5s\n",
      "657:\tlearn: 0.1214454\ttotal: 31.6s\tremaining: 16.4s\n",
      "658:\tlearn: 0.1214067\ttotal: 31.6s\tremaining: 16.4s\n",
      "659:\tlearn: 0.1213583\ttotal: 31.7s\tremaining: 16.3s\n",
      "660:\tlearn: 0.1212655\ttotal: 31.7s\tremaining: 16.3s\n",
      "661:\tlearn: 0.1211959\ttotal: 31.8s\tremaining: 16.2s\n",
      "662:\tlearn: 0.1211763\ttotal: 31.8s\tremaining: 16.2s\n",
      "663:\tlearn: 0.1211336\ttotal: 31.9s\tremaining: 16.1s\n",
      "664:\tlearn: 0.1211217\ttotal: 31.9s\tremaining: 16.1s\n",
      "665:\tlearn: 0.1210817\ttotal: 32s\tremaining: 16s\n",
      "666:\tlearn: 0.1210423\ttotal: 32s\tremaining: 16s\n",
      "667:\tlearn: 0.1210336\ttotal: 32.1s\tremaining: 15.9s\n",
      "668:\tlearn: 0.1210254\ttotal: 32.1s\tremaining: 15.9s\n",
      "669:\tlearn: 0.1209810\ttotal: 32.2s\tremaining: 15.8s\n",
      "670:\tlearn: 0.1209320\ttotal: 32.2s\tremaining: 15.8s\n",
      "671:\tlearn: 0.1209138\ttotal: 32.3s\tremaining: 15.7s\n",
      "672:\tlearn: 0.1208612\ttotal: 32.3s\tremaining: 15.7s\n",
      "673:\tlearn: 0.1207514\ttotal: 32.4s\tremaining: 15.6s\n",
      "674:\tlearn: 0.1207153\ttotal: 32.4s\tremaining: 15.6s\n",
      "675:\tlearn: 0.1206802\ttotal: 32.4s\tremaining: 15.6s\n",
      "676:\tlearn: 0.1206395\ttotal: 32.5s\tremaining: 15.5s\n",
      "677:\tlearn: 0.1206002\ttotal: 32.5s\tremaining: 15.5s\n",
      "678:\tlearn: 0.1205657\ttotal: 32.6s\tremaining: 15.4s\n",
      "679:\tlearn: 0.1205291\ttotal: 32.6s\tremaining: 15.4s\n",
      "680:\tlearn: 0.1204821\ttotal: 32.7s\tremaining: 15.3s\n",
      "681:\tlearn: 0.1204371\ttotal: 32.7s\tremaining: 15.3s\n",
      "682:\tlearn: 0.1203992\ttotal: 32.8s\tremaining: 15.2s\n",
      "683:\tlearn: 0.1202928\ttotal: 32.8s\tremaining: 15.2s\n",
      "684:\tlearn: 0.1202742\ttotal: 32.9s\tremaining: 15.1s\n",
      "685:\tlearn: 0.1202134\ttotal: 32.9s\tremaining: 15.1s\n",
      "686:\tlearn: 0.1201790\ttotal: 33s\tremaining: 15s\n",
      "687:\tlearn: 0.1201432\ttotal: 33s\tremaining: 15s\n",
      "688:\tlearn: 0.1200391\ttotal: 33.1s\tremaining: 14.9s\n",
      "689:\tlearn: 0.1200289\ttotal: 33.1s\tremaining: 14.9s\n",
      "690:\tlearn: 0.1199937\ttotal: 33.2s\tremaining: 14.8s\n",
      "691:\tlearn: 0.1199783\ttotal: 33.2s\tremaining: 14.8s\n",
      "692:\tlearn: 0.1199703\ttotal: 33.3s\tremaining: 14.7s\n",
      "693:\tlearn: 0.1199429\ttotal: 33.3s\tremaining: 14.7s\n",
      "694:\tlearn: 0.1199165\ttotal: 33.4s\tremaining: 14.6s\n",
      "695:\tlearn: 0.1199052\ttotal: 33.4s\tremaining: 14.6s\n",
      "696:\tlearn: 0.1198899\ttotal: 33.4s\tremaining: 14.5s\n",
      "697:\tlearn: 0.1198814\ttotal: 33.5s\tremaining: 14.5s\n",
      "698:\tlearn: 0.1198183\ttotal: 33.5s\tremaining: 14.4s\n",
      "699:\tlearn: 0.1197176\ttotal: 33.6s\tremaining: 14.4s\n",
      "700:\tlearn: 0.1196191\ttotal: 33.6s\tremaining: 14.3s\n",
      "701:\tlearn: 0.1195219\ttotal: 33.7s\tremaining: 14.3s\n",
      "702:\tlearn: 0.1194834\ttotal: 33.7s\tremaining: 14.3s\n",
      "703:\tlearn: 0.1194722\ttotal: 33.8s\tremaining: 14.2s\n",
      "704:\tlearn: 0.1194345\ttotal: 33.8s\tremaining: 14.2s\n",
      "705:\tlearn: 0.1193981\ttotal: 33.9s\tremaining: 14.1s\n",
      "706:\tlearn: 0.1193027\ttotal: 33.9s\tremaining: 14.1s\n",
      "707:\tlearn: 0.1192699\ttotal: 34s\tremaining: 14s\n",
      "708:\tlearn: 0.1192603\ttotal: 34s\tremaining: 14s\n",
      "709:\tlearn: 0.1192248\ttotal: 34.1s\tremaining: 13.9s\n",
      "710:\tlearn: 0.1191926\ttotal: 34.1s\tremaining: 13.9s\n",
      "711:\tlearn: 0.1190994\ttotal: 34.2s\tremaining: 13.8s\n",
      "712:\tlearn: 0.1190870\ttotal: 34.3s\tremaining: 13.8s\n",
      "713:\tlearn: 0.1190329\ttotal: 34.3s\tremaining: 13.7s\n",
      "714:\tlearn: 0.1189864\ttotal: 34.3s\tremaining: 13.7s\n",
      "715:\tlearn: 0.1189598\ttotal: 34.4s\tremaining: 13.6s\n",
      "716:\tlearn: 0.1189316\ttotal: 34.4s\tremaining: 13.6s\n",
      "717:\tlearn: 0.1189254\ttotal: 34.5s\tremaining: 13.5s\n",
      "718:\tlearn: 0.1188952\ttotal: 34.5s\tremaining: 13.5s\n",
      "719:\tlearn: 0.1188547\ttotal: 34.6s\tremaining: 13.4s\n",
      "720:\tlearn: 0.1187980\ttotal: 34.6s\tremaining: 13.4s\n",
      "721:\tlearn: 0.1187867\ttotal: 34.7s\tremaining: 13.4s\n",
      "722:\tlearn: 0.1187707\ttotal: 34.7s\tremaining: 13.3s\n",
      "723:\tlearn: 0.1187415\ttotal: 34.8s\tremaining: 13.3s\n",
      "724:\tlearn: 0.1187319\ttotal: 34.8s\tremaining: 13.2s\n",
      "725:\tlearn: 0.1187259\ttotal: 34.9s\tremaining: 13.2s\n",
      "726:\tlearn: 0.1186738\ttotal: 34.9s\tremaining: 13.1s\n",
      "727:\tlearn: 0.1186625\ttotal: 35s\tremaining: 13.1s\n",
      "728:\tlearn: 0.1186266\ttotal: 35s\tremaining: 13s\n",
      "729:\tlearn: 0.1185975\ttotal: 35.1s\tremaining: 13s\n",
      "730:\tlearn: 0.1185302\ttotal: 35.1s\tremaining: 12.9s\n",
      "731:\tlearn: 0.1184960\ttotal: 35.2s\tremaining: 12.9s\n",
      "732:\tlearn: 0.1184692\ttotal: 35.2s\tremaining: 12.8s\n",
      "733:\tlearn: 0.1184148\ttotal: 35.3s\tremaining: 12.8s\n",
      "734:\tlearn: 0.1183866\ttotal: 35.3s\tremaining: 12.7s\n",
      "735:\tlearn: 0.1183526\ttotal: 35.4s\tremaining: 12.7s\n",
      "736:\tlearn: 0.1183345\ttotal: 35.4s\tremaining: 12.6s\n",
      "737:\tlearn: 0.1182457\ttotal: 35.5s\tremaining: 12.6s\n",
      "738:\tlearn: 0.1181930\ttotal: 35.5s\tremaining: 12.5s\n",
      "739:\tlearn: 0.1181414\ttotal: 35.6s\tremaining: 12.5s\n",
      "740:\tlearn: 0.1181117\ttotal: 35.6s\tremaining: 12.4s\n",
      "741:\tlearn: 0.1181030\ttotal: 35.6s\tremaining: 12.4s\n",
      "742:\tlearn: 0.1180757\ttotal: 35.7s\tremaining: 12.3s\n",
      "743:\tlearn: 0.1180481\ttotal: 35.7s\tremaining: 12.3s\n",
      "744:\tlearn: 0.1180211\ttotal: 35.8s\tremaining: 12.3s\n",
      "745:\tlearn: 0.1179906\ttotal: 35.9s\tremaining: 12.2s\n",
      "746:\tlearn: 0.1179046\ttotal: 35.9s\tremaining: 12.2s\n",
      "747:\tlearn: 0.1178787\ttotal: 36s\tremaining: 12.1s\n",
      "748:\tlearn: 0.1178531\ttotal: 36s\tremaining: 12.1s\n",
      "749:\tlearn: 0.1178071\ttotal: 36.1s\tremaining: 12s\n",
      "750:\tlearn: 0.1177916\ttotal: 36.1s\tremaining: 12s\n",
      "751:\tlearn: 0.1177471\ttotal: 36.2s\tremaining: 11.9s\n",
      "752:\tlearn: 0.1177404\ttotal: 36.2s\tremaining: 11.9s\n",
      "753:\tlearn: 0.1177281\ttotal: 36.3s\tremaining: 11.8s\n",
      "754:\tlearn: 0.1177064\ttotal: 36.3s\tremaining: 11.8s\n",
      "755:\tlearn: 0.1176790\ttotal: 36.4s\tremaining: 11.7s\n",
      "756:\tlearn: 0.1176526\ttotal: 36.4s\tremaining: 11.7s\n",
      "757:\tlearn: 0.1176331\ttotal: 36.5s\tremaining: 11.6s\n",
      "758:\tlearn: 0.1176217\ttotal: 36.5s\tremaining: 11.6s\n",
      "759:\tlearn: 0.1176030\ttotal: 36.6s\tremaining: 11.5s\n",
      "760:\tlearn: 0.1175686\ttotal: 36.6s\tremaining: 11.5s\n",
      "761:\tlearn: 0.1174854\ttotal: 36.6s\tremaining: 11.4s\n",
      "762:\tlearn: 0.1174739\ttotal: 36.7s\tremaining: 11.4s\n",
      "763:\tlearn: 0.1174208\ttotal: 36.7s\tremaining: 11.3s\n",
      "764:\tlearn: 0.1173946\ttotal: 36.8s\tremaining: 11.3s\n",
      "765:\tlearn: 0.1173695\ttotal: 36.8s\tremaining: 11.3s\n",
      "766:\tlearn: 0.1173320\ttotal: 36.9s\tremaining: 11.2s\n",
      "767:\tlearn: 0.1173001\ttotal: 36.9s\tremaining: 11.2s\n",
      "768:\tlearn: 0.1172737\ttotal: 37s\tremaining: 11.1s\n",
      "769:\tlearn: 0.1172385\ttotal: 37s\tremaining: 11.1s\n",
      "770:\tlearn: 0.1172119\ttotal: 37.1s\tremaining: 11s\n",
      "771:\tlearn: 0.1171892\ttotal: 37.1s\tremaining: 11s\n",
      "772:\tlearn: 0.1171800\ttotal: 37.2s\tremaining: 10.9s\n",
      "773:\tlearn: 0.1171451\ttotal: 37.2s\tremaining: 10.9s\n",
      "774:\tlearn: 0.1170946\ttotal: 37.3s\tremaining: 10.8s\n",
      "775:\tlearn: 0.1170630\ttotal: 37.3s\tremaining: 10.8s\n",
      "776:\tlearn: 0.1170407\ttotal: 37.4s\tremaining: 10.7s\n",
      "777:\tlearn: 0.1170107\ttotal: 37.4s\tremaining: 10.7s\n",
      "778:\tlearn: 0.1169952\ttotal: 37.5s\tremaining: 10.6s\n",
      "779:\tlearn: 0.1169540\ttotal: 37.5s\tremaining: 10.6s\n",
      "780:\tlearn: 0.1169487\ttotal: 37.5s\tremaining: 10.5s\n",
      "781:\tlearn: 0.1169183\ttotal: 37.6s\tremaining: 10.5s\n",
      "782:\tlearn: 0.1168995\ttotal: 37.6s\tremaining: 10.4s\n",
      "783:\tlearn: 0.1168271\ttotal: 37.7s\tremaining: 10.4s\n",
      "784:\tlearn: 0.1168063\ttotal: 37.7s\tremaining: 10.3s\n",
      "785:\tlearn: 0.1167860\ttotal: 37.8s\tremaining: 10.3s\n",
      "786:\tlearn: 0.1167583\ttotal: 37.8s\tremaining: 10.2s\n",
      "787:\tlearn: 0.1167098\ttotal: 37.9s\tremaining: 10.2s\n",
      "788:\tlearn: 0.1166443\ttotal: 37.9s\tremaining: 10.1s\n",
      "789:\tlearn: 0.1165670\ttotal: 38s\tremaining: 10.1s\n",
      "790:\tlearn: 0.1165532\ttotal: 38s\tremaining: 10s\n",
      "791:\tlearn: 0.1165056\ttotal: 38.1s\tremaining: 9.99s\n",
      "792:\tlearn: 0.1164671\ttotal: 38.1s\tremaining: 9.95s\n",
      "793:\tlearn: 0.1164214\ttotal: 38.2s\tremaining: 9.9s\n",
      "794:\tlearn: 0.1163459\ttotal: 38.2s\tremaining: 9.85s\n",
      "795:\tlearn: 0.1163260\ttotal: 38.3s\tremaining: 9.8s\n",
      "796:\tlearn: 0.1162632\ttotal: 38.3s\tremaining: 9.76s\n",
      "797:\tlearn: 0.1162261\ttotal: 38.4s\tremaining: 9.71s\n",
      "798:\tlearn: 0.1162141\ttotal: 38.4s\tremaining: 9.66s\n",
      "799:\tlearn: 0.1161772\ttotal: 38.5s\tremaining: 9.62s\n",
      "800:\tlearn: 0.1161586\ttotal: 38.5s\tremaining: 9.57s\n",
      "801:\tlearn: 0.1161481\ttotal: 38.6s\tremaining: 9.52s\n",
      "802:\tlearn: 0.1161283\ttotal: 38.6s\tremaining: 9.47s\n",
      "803:\tlearn: 0.1160952\ttotal: 38.7s\tremaining: 9.42s\n",
      "804:\tlearn: 0.1160762\ttotal: 38.7s\tremaining: 9.38s\n",
      "805:\tlearn: 0.1160601\ttotal: 38.8s\tremaining: 9.33s\n",
      "806:\tlearn: 0.1160455\ttotal: 38.8s\tremaining: 9.28s\n",
      "807:\tlearn: 0.1160376\ttotal: 38.9s\tremaining: 9.23s\n",
      "808:\tlearn: 0.1159994\ttotal: 38.9s\tremaining: 9.19s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "809:\tlearn: 0.1159663\ttotal: 39s\tremaining: 9.14s\n",
      "810:\tlearn: 0.1159227\ttotal: 39s\tremaining: 9.09s\n",
      "811:\tlearn: 0.1158907\ttotal: 39.1s\tremaining: 9.04s\n",
      "812:\tlearn: 0.1158696\ttotal: 39.1s\tremaining: 8.99s\n",
      "813:\tlearn: 0.1158394\ttotal: 39.2s\tremaining: 8.95s\n",
      "814:\tlearn: 0.1158055\ttotal: 39.2s\tremaining: 8.9s\n",
      "815:\tlearn: 0.1157850\ttotal: 39.3s\tremaining: 8.85s\n",
      "816:\tlearn: 0.1157539\ttotal: 39.3s\tremaining: 8.81s\n",
      "817:\tlearn: 0.1157124\ttotal: 39.4s\tremaining: 8.76s\n",
      "818:\tlearn: 0.1156405\ttotal: 39.4s\tremaining: 8.71s\n",
      "819:\tlearn: 0.1156260\ttotal: 39.5s\tremaining: 8.66s\n",
      "820:\tlearn: 0.1155959\ttotal: 39.5s\tremaining: 8.62s\n",
      "821:\tlearn: 0.1155875\ttotal: 39.6s\tremaining: 8.57s\n",
      "822:\tlearn: 0.1155578\ttotal: 39.6s\tremaining: 8.52s\n",
      "823:\tlearn: 0.1155322\ttotal: 39.7s\tremaining: 8.48s\n",
      "824:\tlearn: 0.1154955\ttotal: 39.7s\tremaining: 8.43s\n",
      "825:\tlearn: 0.1154749\ttotal: 39.8s\tremaining: 8.38s\n",
      "826:\tlearn: 0.1154105\ttotal: 39.8s\tremaining: 8.33s\n",
      "827:\tlearn: 0.1153872\ttotal: 39.9s\tremaining: 8.29s\n",
      "828:\tlearn: 0.1153421\ttotal: 39.9s\tremaining: 8.24s\n",
      "829:\tlearn: 0.1152816\ttotal: 40s\tremaining: 8.19s\n",
      "830:\tlearn: 0.1152644\ttotal: 40s\tremaining: 8.14s\n",
      "831:\tlearn: 0.1152280\ttotal: 40.1s\tremaining: 8.1s\n",
      "832:\tlearn: 0.1151885\ttotal: 40.1s\tremaining: 8.05s\n",
      "833:\tlearn: 0.1151640\ttotal: 40.2s\tremaining: 8s\n",
      "834:\tlearn: 0.1151535\ttotal: 40.3s\tremaining: 7.96s\n",
      "835:\tlearn: 0.1151450\ttotal: 40.3s\tremaining: 7.91s\n",
      "836:\tlearn: 0.1151100\ttotal: 40.4s\tremaining: 7.86s\n",
      "837:\tlearn: 0.1150915\ttotal: 40.4s\tremaining: 7.81s\n",
      "838:\tlearn: 0.1150636\ttotal: 40.4s\tremaining: 7.76s\n",
      "839:\tlearn: 0.1150365\ttotal: 40.5s\tremaining: 7.71s\n",
      "840:\tlearn: 0.1150015\ttotal: 40.5s\tremaining: 7.67s\n",
      "841:\tlearn: 0.1149931\ttotal: 40.6s\tremaining: 7.62s\n",
      "842:\tlearn: 0.1149773\ttotal: 40.6s\tremaining: 7.57s\n",
      "843:\tlearn: 0.1149075\ttotal: 40.7s\tremaining: 7.52s\n",
      "844:\tlearn: 0.1148842\ttotal: 40.8s\tremaining: 7.48s\n",
      "845:\tlearn: 0.1148633\ttotal: 40.8s\tremaining: 7.43s\n",
      "846:\tlearn: 0.1148530\ttotal: 40.9s\tremaining: 7.38s\n",
      "847:\tlearn: 0.1148228\ttotal: 40.9s\tremaining: 7.33s\n",
      "848:\tlearn: 0.1147851\ttotal: 41s\tremaining: 7.29s\n",
      "849:\tlearn: 0.1147613\ttotal: 41s\tremaining: 7.24s\n",
      "850:\tlearn: 0.1147509\ttotal: 41.1s\tremaining: 7.19s\n",
      "851:\tlearn: 0.1147177\ttotal: 41.1s\tremaining: 7.14s\n",
      "852:\tlearn: 0.1147069\ttotal: 41.2s\tremaining: 7.09s\n",
      "853:\tlearn: 0.1146998\ttotal: 41.2s\tremaining: 7.05s\n",
      "854:\tlearn: 0.1146657\ttotal: 41.3s\tremaining: 7s\n",
      "855:\tlearn: 0.1146549\ttotal: 41.3s\tremaining: 6.95s\n",
      "856:\tlearn: 0.1146458\ttotal: 41.4s\tremaining: 6.9s\n",
      "857:\tlearn: 0.1146149\ttotal: 41.4s\tremaining: 6.86s\n",
      "858:\tlearn: 0.1145783\ttotal: 41.5s\tremaining: 6.81s\n",
      "859:\tlearn: 0.1145627\ttotal: 41.5s\tremaining: 6.76s\n",
      "860:\tlearn: 0.1144978\ttotal: 41.6s\tremaining: 6.72s\n",
      "861:\tlearn: 0.1144634\ttotal: 41.7s\tremaining: 6.67s\n",
      "862:\tlearn: 0.1144378\ttotal: 41.7s\tremaining: 6.62s\n",
      "863:\tlearn: 0.1144130\ttotal: 41.8s\tremaining: 6.58s\n",
      "864:\tlearn: 0.1143967\ttotal: 41.8s\tremaining: 6.53s\n",
      "865:\tlearn: 0.1143497\ttotal: 41.9s\tremaining: 6.48s\n",
      "866:\tlearn: 0.1143256\ttotal: 41.9s\tremaining: 6.43s\n",
      "867:\tlearn: 0.1143087\ttotal: 42s\tremaining: 6.38s\n",
      "868:\tlearn: 0.1142745\ttotal: 42s\tremaining: 6.33s\n",
      "869:\tlearn: 0.1142589\ttotal: 42.1s\tremaining: 6.29s\n",
      "870:\tlearn: 0.1142499\ttotal: 42.1s\tremaining: 6.24s\n",
      "871:\tlearn: 0.1142300\ttotal: 42.2s\tremaining: 6.19s\n",
      "872:\tlearn: 0.1142129\ttotal: 42.2s\tremaining: 6.14s\n",
      "873:\tlearn: 0.1141569\ttotal: 42.3s\tremaining: 6.09s\n",
      "874:\tlearn: 0.1141410\ttotal: 42.3s\tremaining: 6.05s\n",
      "875:\tlearn: 0.1141216\ttotal: 42.4s\tremaining: 6s\n",
      "876:\tlearn: 0.1140801\ttotal: 42.4s\tremaining: 5.95s\n",
      "877:\tlearn: 0.1140648\ttotal: 42.5s\tremaining: 5.9s\n",
      "878:\tlearn: 0.1140567\ttotal: 42.5s\tremaining: 5.86s\n",
      "879:\tlearn: 0.1140487\ttotal: 42.6s\tremaining: 5.81s\n",
      "880:\tlearn: 0.1140288\ttotal: 42.6s\tremaining: 5.76s\n",
      "881:\tlearn: 0.1139758\ttotal: 42.7s\tremaining: 5.71s\n",
      "882:\tlearn: 0.1139548\ttotal: 42.7s\tremaining: 5.66s\n",
      "883:\tlearn: 0.1139311\ttotal: 42.8s\tremaining: 5.61s\n",
      "884:\tlearn: 0.1139034\ttotal: 42.8s\tremaining: 5.57s\n",
      "885:\tlearn: 0.1138686\ttotal: 42.9s\tremaining: 5.52s\n",
      "886:\tlearn: 0.1138379\ttotal: 42.9s\tremaining: 5.47s\n",
      "887:\tlearn: 0.1138140\ttotal: 43s\tremaining: 5.42s\n",
      "888:\tlearn: 0.1137803\ttotal: 43s\tremaining: 5.37s\n",
      "889:\tlearn: 0.1137472\ttotal: 43.1s\tremaining: 5.32s\n",
      "890:\tlearn: 0.1137343\ttotal: 43.1s\tremaining: 5.28s\n",
      "891:\tlearn: 0.1136872\ttotal: 43.2s\tremaining: 5.23s\n",
      "892:\tlearn: 0.1136587\ttotal: 43.2s\tremaining: 5.18s\n",
      "893:\tlearn: 0.1136280\ttotal: 43.3s\tremaining: 5.13s\n",
      "894:\tlearn: 0.1136201\ttotal: 43.3s\tremaining: 5.08s\n",
      "895:\tlearn: 0.1136088\ttotal: 43.4s\tremaining: 5.03s\n",
      "896:\tlearn: 0.1135882\ttotal: 43.4s\tremaining: 4.99s\n",
      "897:\tlearn: 0.1135703\ttotal: 43.5s\tremaining: 4.94s\n",
      "898:\tlearn: 0.1135403\ttotal: 43.5s\tremaining: 4.89s\n",
      "899:\tlearn: 0.1135103\ttotal: 43.6s\tremaining: 4.84s\n",
      "900:\tlearn: 0.1134782\ttotal: 43.6s\tremaining: 4.79s\n",
      "901:\tlearn: 0.1134626\ttotal: 43.7s\tremaining: 4.75s\n",
      "902:\tlearn: 0.1134476\ttotal: 43.7s\tremaining: 4.7s\n",
      "903:\tlearn: 0.1134258\ttotal: 43.8s\tremaining: 4.65s\n",
      "904:\tlearn: 0.1134141\ttotal: 43.8s\tremaining: 4.6s\n",
      "905:\tlearn: 0.1133846\ttotal: 43.9s\tremaining: 4.55s\n",
      "906:\tlearn: 0.1133233\ttotal: 44s\tremaining: 4.51s\n",
      "907:\tlearn: 0.1132922\ttotal: 44s\tremaining: 4.46s\n",
      "908:\tlearn: 0.1132707\ttotal: 44.1s\tremaining: 4.41s\n",
      "909:\tlearn: 0.1132619\ttotal: 44.1s\tremaining: 4.36s\n",
      "910:\tlearn: 0.1132425\ttotal: 44.2s\tremaining: 4.32s\n",
      "911:\tlearn: 0.1132127\ttotal: 44.2s\tremaining: 4.27s\n",
      "912:\tlearn: 0.1131951\ttotal: 44.3s\tremaining: 4.22s\n",
      "913:\tlearn: 0.1131848\ttotal: 44.3s\tremaining: 4.17s\n",
      "914:\tlearn: 0.1131672\ttotal: 44.4s\tremaining: 4.12s\n",
      "915:\tlearn: 0.1131379\ttotal: 44.4s\tremaining: 4.07s\n",
      "916:\tlearn: 0.1131159\ttotal: 44.5s\tremaining: 4.03s\n",
      "917:\tlearn: 0.1130977\ttotal: 44.5s\tremaining: 3.98s\n",
      "918:\tlearn: 0.1130485\ttotal: 44.6s\tremaining: 3.93s\n",
      "919:\tlearn: 0.1130380\ttotal: 44.7s\tremaining: 3.88s\n",
      "920:\tlearn: 0.1130155\ttotal: 44.7s\tremaining: 3.83s\n",
      "921:\tlearn: 0.1129897\ttotal: 44.8s\tremaining: 3.79s\n",
      "922:\tlearn: 0.1129683\ttotal: 44.8s\tremaining: 3.74s\n",
      "923:\tlearn: 0.1129610\ttotal: 44.9s\tremaining: 3.69s\n",
      "924:\tlearn: 0.1129548\ttotal: 44.9s\tremaining: 3.64s\n",
      "925:\tlearn: 0.1129234\ttotal: 45s\tremaining: 3.6s\n",
      "926:\tlearn: 0.1128948\ttotal: 45s\tremaining: 3.55s\n",
      "927:\tlearn: 0.1128366\ttotal: 45.1s\tremaining: 3.5s\n",
      "928:\tlearn: 0.1128175\ttotal: 45.2s\tremaining: 3.45s\n",
      "929:\tlearn: 0.1127887\ttotal: 45.2s\tremaining: 3.4s\n",
      "930:\tlearn: 0.1127782\ttotal: 45.3s\tremaining: 3.35s\n",
      "931:\tlearn: 0.1127570\ttotal: 45.3s\tremaining: 3.31s\n",
      "932:\tlearn: 0.1127462\ttotal: 45.4s\tremaining: 3.26s\n",
      "933:\tlearn: 0.1127189\ttotal: 45.4s\tremaining: 3.21s\n",
      "934:\tlearn: 0.1127052\ttotal: 45.5s\tremaining: 3.16s\n",
      "935:\tlearn: 0.1126899\ttotal: 45.5s\tremaining: 3.11s\n",
      "936:\tlearn: 0.1126709\ttotal: 45.6s\tremaining: 3.06s\n",
      "937:\tlearn: 0.1126427\ttotal: 45.6s\tremaining: 3.02s\n",
      "938:\tlearn: 0.1125929\ttotal: 45.7s\tremaining: 2.97s\n",
      "939:\tlearn: 0.1125766\ttotal: 45.7s\tremaining: 2.92s\n",
      "940:\tlearn: 0.1125543\ttotal: 45.8s\tremaining: 2.87s\n",
      "941:\tlearn: 0.1125121\ttotal: 45.8s\tremaining: 2.82s\n",
      "942:\tlearn: 0.1124751\ttotal: 45.9s\tremaining: 2.77s\n",
      "943:\tlearn: 0.1124399\ttotal: 45.9s\tremaining: 2.73s\n",
      "944:\tlearn: 0.1123888\ttotal: 46s\tremaining: 2.68s\n",
      "945:\tlearn: 0.1123689\ttotal: 46s\tremaining: 2.63s\n",
      "946:\tlearn: 0.1123316\ttotal: 46.1s\tremaining: 2.58s\n",
      "947:\tlearn: 0.1123129\ttotal: 46.1s\tremaining: 2.53s\n",
      "948:\tlearn: 0.1122932\ttotal: 46.2s\tremaining: 2.48s\n",
      "949:\tlearn: 0.1122665\ttotal: 46.2s\tremaining: 2.43s\n",
      "950:\tlearn: 0.1122548\ttotal: 46.3s\tremaining: 2.38s\n",
      "951:\tlearn: 0.1122372\ttotal: 46.4s\tremaining: 2.34s\n",
      "952:\tlearn: 0.1121935\ttotal: 46.4s\tremaining: 2.29s\n",
      "953:\tlearn: 0.1121419\ttotal: 46.4s\tremaining: 2.24s\n",
      "954:\tlearn: 0.1121331\ttotal: 46.5s\tremaining: 2.19s\n",
      "955:\tlearn: 0.1121056\ttotal: 46.5s\tremaining: 2.14s\n",
      "956:\tlearn: 0.1120550\ttotal: 46.6s\tremaining: 2.09s\n",
      "957:\tlearn: 0.1120467\ttotal: 46.6s\tremaining: 2.04s\n",
      "958:\tlearn: 0.1120351\ttotal: 46.7s\tremaining: 2s\n",
      "959:\tlearn: 0.1120165\ttotal: 46.8s\tremaining: 1.95s\n",
      "960:\tlearn: 0.1120058\ttotal: 46.8s\tremaining: 1.9s\n",
      "961:\tlearn: 0.1119832\ttotal: 46.9s\tremaining: 1.85s\n",
      "962:\tlearn: 0.1119661\ttotal: 46.9s\tremaining: 1.8s\n",
      "963:\tlearn: 0.1119467\ttotal: 47s\tremaining: 1.75s\n",
      "964:\tlearn: 0.1119298\ttotal: 47s\tremaining: 1.71s\n",
      "965:\tlearn: 0.1119196\ttotal: 47.1s\tremaining: 1.66s\n",
      "966:\tlearn: 0.1118966\ttotal: 47.1s\tremaining: 1.61s\n",
      "967:\tlearn: 0.1118844\ttotal: 47.2s\tremaining: 1.56s\n",
      "968:\tlearn: 0.1118759\ttotal: 47.3s\tremaining: 1.51s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "969:\tlearn: 0.1118658\ttotal: 47.3s\tremaining: 1.46s\n",
      "970:\tlearn: 0.1118504\ttotal: 47.4s\tremaining: 1.41s\n",
      "971:\tlearn: 0.1118234\ttotal: 47.4s\tremaining: 1.36s\n",
      "972:\tlearn: 0.1117969\ttotal: 47.5s\tremaining: 1.32s\n",
      "973:\tlearn: 0.1117881\ttotal: 47.5s\tremaining: 1.27s\n",
      "974:\tlearn: 0.1117657\ttotal: 47.6s\tremaining: 1.22s\n",
      "975:\tlearn: 0.1117534\ttotal: 47.6s\tremaining: 1.17s\n",
      "976:\tlearn: 0.1117357\ttotal: 47.7s\tremaining: 1.12s\n",
      "977:\tlearn: 0.1117172\ttotal: 47.7s\tremaining: 1.07s\n",
      "978:\tlearn: 0.1117103\ttotal: 47.8s\tremaining: 1.02s\n",
      "979:\tlearn: 0.1116738\ttotal: 47.8s\tremaining: 976ms\n",
      "980:\tlearn: 0.1116605\ttotal: 47.9s\tremaining: 927ms\n",
      "981:\tlearn: 0.1116368\ttotal: 47.9s\tremaining: 878ms\n",
      "982:\tlearn: 0.1116137\ttotal: 48s\tremaining: 829ms\n",
      "983:\tlearn: 0.1115968\ttotal: 48s\tremaining: 781ms\n",
      "984:\tlearn: 0.1115851\ttotal: 48.1s\tremaining: 732ms\n",
      "985:\tlearn: 0.1115591\ttotal: 48.1s\tremaining: 683ms\n",
      "986:\tlearn: 0.1115506\ttotal: 48.2s\tremaining: 634ms\n",
      "987:\tlearn: 0.1115461\ttotal: 48.2s\tremaining: 585ms\n",
      "988:\tlearn: 0.1115206\ttotal: 48.3s\tremaining: 537ms\n",
      "989:\tlearn: 0.1115097\ttotal: 48.3s\tremaining: 488ms\n",
      "990:\tlearn: 0.1114910\ttotal: 48.4s\tremaining: 439ms\n",
      "991:\tlearn: 0.1114744\ttotal: 48.4s\tremaining: 390ms\n",
      "992:\tlearn: 0.1114468\ttotal: 48.5s\tremaining: 342ms\n",
      "993:\tlearn: 0.1114267\ttotal: 48.5s\tremaining: 293ms\n",
      "994:\tlearn: 0.1114197\ttotal: 48.6s\tremaining: 244ms\n",
      "995:\tlearn: 0.1114044\ttotal: 48.6s\tremaining: 195ms\n",
      "996:\tlearn: 0.1113858\ttotal: 48.7s\tremaining: 146ms\n",
      "997:\tlearn: 0.1113685\ttotal: 48.7s\tremaining: 97.6ms\n",
      "998:\tlearn: 0.1113543\ttotal: 48.8s\tremaining: 48.8ms\n",
      "999:\tlearn: 0.1113327\ttotal: 48.8s\tremaining: 0us\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<catboost.core.CatBoostRegressor at 0x25eaaf54808>"
      ]
     },
     "execution_count": 165,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "boost.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_pred=boost.predict(X_train)\n",
    "y_test_pred=boost.predict(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "111.33271193869358"
      ]
     },
     "execution_count": 167,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1000*np.sqrt(mean_squared_error(y_train,y_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {},
   "outputs": [],
   "source": [
    "export_submission(y_test_pred,\"Boost Cat 0.01\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "parameters={'learning_rate':np.arange(0.01,0.1,0.01)}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "search=GridSearchCV(CatBoostRegressor(),parameters,scoring='neg_mean_squared_error',cv=5,verbose=2,n_jobs=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "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=10)]: Using backend LokyBackend with 10 concurrent workers.\n",
      "[Parallel(n_jobs=10)]: Done  21 tasks      | elapsed: 14.9min\n",
      "[Parallel(n_jobs=10)]: Done  45 out of  45 | elapsed: 23.2min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0:\tlearn: 0.4581315\ttotal: 41.1ms\tremaining: 41s\n",
      "1:\tlearn: 0.4549662\ttotal: 81.9ms\tremaining: 40.9s\n",
      "2:\tlearn: 0.4518408\ttotal: 124ms\tremaining: 41.3s\n",
      "3:\tlearn: 0.4487626\ttotal: 165ms\tremaining: 41.1s\n",
      "4:\tlearn: 0.4457285\ttotal: 202ms\tremaining: 40.3s\n",
      "5:\tlearn: 0.4427298\ttotal: 239ms\tremaining: 39.7s\n",
      "6:\tlearn: 0.4397585\ttotal: 279ms\tremaining: 39.6s\n",
      "7:\tlearn: 0.4368247\ttotal: 320ms\tremaining: 39.6s\n",
      "8:\tlearn: 0.4339013\ttotal: 358ms\tremaining: 39.4s\n",
      "9:\tlearn: 0.4310652\ttotal: 400ms\tremaining: 39.6s\n",
      "10:\tlearn: 0.4282187\ttotal: 440ms\tremaining: 39.6s\n",
      "11:\tlearn: 0.4254297\ttotal: 484ms\tremaining: 39.8s\n",
      "12:\tlearn: 0.4226888\ttotal: 523ms\tremaining: 39.7s\n",
      "13:\tlearn: 0.4199831\ttotal: 563ms\tremaining: 39.7s\n",
      "14:\tlearn: 0.4173058\ttotal: 604ms\tremaining: 39.7s\n",
      "15:\tlearn: 0.4146414\ttotal: 642ms\tremaining: 39.5s\n",
      "16:\tlearn: 0.4120248\ttotal: 685ms\tremaining: 39.6s\n",
      "17:\tlearn: 0.4094287\ttotal: 725ms\tremaining: 39.5s\n",
      "18:\tlearn: 0.4068415\ttotal: 763ms\tremaining: 39.4s\n",
      "19:\tlearn: 0.4043302\ttotal: 801ms\tremaining: 39.3s\n",
      "20:\tlearn: 0.4018176\ttotal: 839ms\tremaining: 39.1s\n",
      "21:\tlearn: 0.3993394\ttotal: 880ms\tremaining: 39.1s\n",
      "22:\tlearn: 0.3968977\ttotal: 917ms\tremaining: 39s\n",
      "23:\tlearn: 0.3944945\ttotal: 955ms\tremaining: 38.8s\n",
      "24:\tlearn: 0.3921242\ttotal: 994ms\tremaining: 38.8s\n",
      "25:\tlearn: 0.3897632\ttotal: 1.03s\tremaining: 38.7s\n",
      "26:\tlearn: 0.3874590\ttotal: 1.07s\tremaining: 38.7s\n",
      "27:\tlearn: 0.3851723\ttotal: 1.11s\tremaining: 38.7s\n",
      "28:\tlearn: 0.3829132\ttotal: 1.15s\tremaining: 38.7s\n",
      "29:\tlearn: 0.3806870\ttotal: 1.19s\tremaining: 38.6s\n",
      "30:\tlearn: 0.3784883\ttotal: 1.24s\tremaining: 38.6s\n",
      "31:\tlearn: 0.3763173\ttotal: 1.28s\tremaining: 38.7s\n",
      "32:\tlearn: 0.3741824\ttotal: 1.32s\tremaining: 38.8s\n",
      "33:\tlearn: 0.3720566\ttotal: 1.36s\tremaining: 38.8s\n",
      "34:\tlearn: 0.3699887\ttotal: 1.41s\tremaining: 38.8s\n",
      "35:\tlearn: 0.3679215\ttotal: 1.45s\tremaining: 38.8s\n",
      "36:\tlearn: 0.3658981\ttotal: 1.49s\tremaining: 38.9s\n",
      "37:\tlearn: 0.3638889\ttotal: 1.54s\tremaining: 39s\n",
      "38:\tlearn: 0.3619258\ttotal: 1.58s\tremaining: 39s\n",
      "39:\tlearn: 0.3599740\ttotal: 1.62s\tremaining: 38.8s\n",
      "40:\tlearn: 0.3580382\ttotal: 1.65s\tremaining: 38.7s\n",
      "41:\tlearn: 0.3561350\ttotal: 1.7s\tremaining: 38.7s\n",
      "42:\tlearn: 0.3542670\ttotal: 1.74s\tremaining: 38.7s\n",
      "43:\tlearn: 0.3523965\ttotal: 1.78s\tremaining: 38.6s\n",
      "44:\tlearn: 0.3505659\ttotal: 1.82s\tremaining: 38.6s\n",
      "45:\tlearn: 0.3487651\ttotal: 1.86s\tremaining: 38.6s\n",
      "46:\tlearn: 0.3469826\ttotal: 1.91s\tremaining: 38.6s\n",
      "47:\tlearn: 0.3452100\ttotal: 1.95s\tremaining: 38.7s\n",
      "48:\tlearn: 0.3434673\ttotal: 1.99s\tremaining: 38.6s\n",
      "49:\tlearn: 0.3417484\ttotal: 2.03s\tremaining: 38.6s\n",
      "50:\tlearn: 0.3400959\ttotal: 2.07s\tremaining: 38.5s\n",
      "51:\tlearn: 0.3384340\ttotal: 2.11s\tremaining: 38.5s\n",
      "52:\tlearn: 0.3367832\ttotal: 2.15s\tremaining: 38.5s\n",
      "53:\tlearn: 0.3351525\ttotal: 2.19s\tremaining: 38.5s\n",
      "54:\tlearn: 0.3335590\ttotal: 2.23s\tremaining: 38.4s\n",
      "55:\tlearn: 0.3319936\ttotal: 2.27s\tremaining: 38.3s\n",
      "56:\tlearn: 0.3304491\ttotal: 2.31s\tremaining: 38.3s\n",
      "57:\tlearn: 0.3289201\ttotal: 2.35s\tremaining: 38.2s\n",
      "58:\tlearn: 0.3274058\ttotal: 2.39s\tremaining: 38.1s\n",
      "59:\tlearn: 0.3259149\ttotal: 2.43s\tremaining: 38s\n",
      "60:\tlearn: 0.3244632\ttotal: 2.46s\tremaining: 38s\n",
      "61:\tlearn: 0.3230187\ttotal: 2.5s\tremaining: 37.9s\n",
      "62:\tlearn: 0.3216050\ttotal: 2.55s\tremaining: 37.9s\n",
      "63:\tlearn: 0.3202035\ttotal: 2.59s\tremaining: 37.9s\n",
      "64:\tlearn: 0.3188156\ttotal: 2.63s\tremaining: 37.8s\n",
      "65:\tlearn: 0.3174437\ttotal: 2.67s\tremaining: 37.8s\n",
      "66:\tlearn: 0.3161125\ttotal: 2.72s\tremaining: 37.8s\n",
      "67:\tlearn: 0.3147884\ttotal: 2.76s\tremaining: 37.8s\n",
      "68:\tlearn: 0.3134852\ttotal: 2.8s\tremaining: 37.8s\n",
      "69:\tlearn: 0.3122068\ttotal: 2.83s\tremaining: 37.7s\n",
      "70:\tlearn: 0.3109271\ttotal: 2.88s\tremaining: 37.6s\n",
      "71:\tlearn: 0.3096765\ttotal: 2.92s\tremaining: 37.6s\n",
      "72:\tlearn: 0.3084442\ttotal: 2.96s\tremaining: 37.5s\n",
      "73:\tlearn: 0.3072257\ttotal: 2.99s\tremaining: 37.4s\n",
      "74:\tlearn: 0.3060419\ttotal: 3.03s\tremaining: 37.4s\n",
      "75:\tlearn: 0.3048779\ttotal: 3.07s\tremaining: 37.3s\n",
      "76:\tlearn: 0.3037203\ttotal: 3.11s\tremaining: 37.2s\n",
      "77:\tlearn: 0.3025789\ttotal: 3.14s\tremaining: 37.2s\n",
      "78:\tlearn: 0.3014515\ttotal: 3.18s\tremaining: 37.1s\n",
      "79:\tlearn: 0.3003506\ttotal: 3.22s\tremaining: 37s\n",
      "80:\tlearn: 0.2992662\ttotal: 3.26s\tremaining: 36.9s\n",
      "81:\tlearn: 0.2981809\ttotal: 3.3s\tremaining: 36.9s\n",
      "82:\tlearn: 0.2971104\ttotal: 3.34s\tremaining: 36.9s\n",
      "83:\tlearn: 0.2960648\ttotal: 3.37s\tremaining: 36.8s\n",
      "84:\tlearn: 0.2950394\ttotal: 3.41s\tremaining: 36.7s\n",
      "85:\tlearn: 0.2940247\ttotal: 3.45s\tremaining: 36.6s\n",
      "86:\tlearn: 0.2930178\ttotal: 3.48s\tremaining: 36.6s\n",
      "87:\tlearn: 0.2920329\ttotal: 3.52s\tremaining: 36.5s\n",
      "88:\tlearn: 0.2910726\ttotal: 3.56s\tremaining: 36.4s\n",
      "89:\tlearn: 0.2901174\ttotal: 3.6s\tremaining: 36.4s\n",
      "90:\tlearn: 0.2891841\ttotal: 3.64s\tremaining: 36.3s\n",
      "91:\tlearn: 0.2882697\ttotal: 3.69s\tremaining: 36.4s\n",
      "92:\tlearn: 0.2873760\ttotal: 3.73s\tremaining: 36.4s\n",
      "93:\tlearn: 0.2864869\ttotal: 3.76s\tremaining: 36.3s\n",
      "94:\tlearn: 0.2856175\ttotal: 3.8s\tremaining: 36.2s\n",
      "95:\tlearn: 0.2847425\ttotal: 3.84s\tremaining: 36.1s\n",
      "96:\tlearn: 0.2838810\ttotal: 3.88s\tremaining: 36.1s\n",
      "97:\tlearn: 0.2830397\ttotal: 3.92s\tremaining: 36.1s\n",
      "98:\tlearn: 0.2822114\ttotal: 3.96s\tremaining: 36s\n",
      "99:\tlearn: 0.2813917\ttotal: 4s\tremaining: 36s\n",
      "100:\tlearn: 0.2805864\ttotal: 4.04s\tremaining: 36s\n",
      "101:\tlearn: 0.2798084\ttotal: 4.08s\tremaining: 35.9s\n",
      "102:\tlearn: 0.2790254\ttotal: 4.13s\tremaining: 36s\n",
      "103:\tlearn: 0.2782628\ttotal: 4.17s\tremaining: 35.9s\n",
      "104:\tlearn: 0.2775216\ttotal: 4.21s\tremaining: 35.9s\n",
      "105:\tlearn: 0.2767853\ttotal: 4.25s\tremaining: 35.8s\n",
      "106:\tlearn: 0.2760572\ttotal: 4.29s\tremaining: 35.8s\n",
      "107:\tlearn: 0.2753309\ttotal: 4.33s\tremaining: 35.8s\n",
      "108:\tlearn: 0.2746160\ttotal: 4.37s\tremaining: 35.7s\n",
      "109:\tlearn: 0.2739091\ttotal: 4.41s\tremaining: 35.7s\n",
      "110:\tlearn: 0.2732252\ttotal: 4.45s\tremaining: 35.6s\n",
      "111:\tlearn: 0.2725539\ttotal: 4.49s\tremaining: 35.6s\n",
      "112:\tlearn: 0.2718813\ttotal: 4.53s\tremaining: 35.6s\n",
      "113:\tlearn: 0.2712199\ttotal: 4.57s\tremaining: 35.5s\n",
      "114:\tlearn: 0.2705897\ttotal: 4.61s\tremaining: 35.5s\n",
      "115:\tlearn: 0.2699535\ttotal: 4.64s\tremaining: 35.4s\n",
      "116:\tlearn: 0.2693328\ttotal: 4.68s\tremaining: 35.4s\n",
      "117:\tlearn: 0.2687077\ttotal: 4.72s\tremaining: 35.3s\n",
      "118:\tlearn: 0.2681140\ttotal: 4.76s\tremaining: 35.3s\n",
      "119:\tlearn: 0.2675183\ttotal: 4.8s\tremaining: 35.2s\n",
      "120:\tlearn: 0.2669333\ttotal: 4.84s\tremaining: 35.2s\n",
      "121:\tlearn: 0.2663537\ttotal: 4.88s\tremaining: 35.1s\n",
      "122:\tlearn: 0.2657809\ttotal: 4.92s\tremaining: 35.1s\n",
      "123:\tlearn: 0.2652213\ttotal: 4.96s\tremaining: 35s\n",
      "124:\tlearn: 0.2646808\ttotal: 4.99s\tremaining: 35s\n",
      "125:\tlearn: 0.2641487\ttotal: 5.03s\tremaining: 34.9s\n",
      "126:\tlearn: 0.2636269\ttotal: 5.07s\tremaining: 34.8s\n",
      "127:\tlearn: 0.2631171\ttotal: 5.1s\tremaining: 34.8s\n",
      "128:\tlearn: 0.2626066\ttotal: 5.13s\tremaining: 34.7s\n",
      "129:\tlearn: 0.2620972\ttotal: 5.17s\tremaining: 34.6s\n",
      "130:\tlearn: 0.2616066\ttotal: 5.2s\tremaining: 34.5s\n",
      "131:\tlearn: 0.2611163\ttotal: 5.24s\tremaining: 34.4s\n",
      "132:\tlearn: 0.2606345\ttotal: 5.28s\tremaining: 34.4s\n",
      "133:\tlearn: 0.2601626\ttotal: 5.31s\tremaining: 34.3s\n",
      "134:\tlearn: 0.2596936\ttotal: 5.34s\tremaining: 34.2s\n",
      "135:\tlearn: 0.2592362\ttotal: 5.38s\tremaining: 34.2s\n",
      "136:\tlearn: 0.2587960\ttotal: 5.41s\tremaining: 34.1s\n",
      "137:\tlearn: 0.2583622\ttotal: 5.45s\tremaining: 34s\n",
      "138:\tlearn: 0.2579367\ttotal: 5.49s\tremaining: 34s\n",
      "139:\tlearn: 0.2575169\ttotal: 5.52s\tremaining: 33.9s\n",
      "140:\tlearn: 0.2571027\ttotal: 5.55s\tremaining: 33.8s\n",
      "141:\tlearn: 0.2566872\ttotal: 5.59s\tremaining: 33.8s\n",
      "142:\tlearn: 0.2562796\ttotal: 5.62s\tremaining: 33.7s\n",
      "143:\tlearn: 0.2558841\ttotal: 5.66s\tremaining: 33.6s\n",
      "144:\tlearn: 0.2554927\ttotal: 5.7s\tremaining: 33.6s\n",
      "145:\tlearn: 0.2551120\ttotal: 5.73s\tremaining: 33.5s\n",
      "146:\tlearn: 0.2547324\ttotal: 5.77s\tremaining: 33.5s\n",
      "147:\tlearn: 0.2543574\ttotal: 5.8s\tremaining: 33.4s\n",
      "148:\tlearn: 0.2539927\ttotal: 5.84s\tremaining: 33.3s\n",
      "149:\tlearn: 0.2536338\ttotal: 5.88s\tremaining: 33.3s\n",
      "150:\tlearn: 0.2532773\ttotal: 5.92s\tremaining: 33.3s\n",
      "151:\tlearn: 0.2529298\ttotal: 5.95s\tremaining: 33.2s\n",
      "152:\tlearn: 0.2525873\ttotal: 5.98s\tremaining: 33.1s\n",
      "153:\tlearn: 0.2522544\ttotal: 6.02s\tremaining: 33.1s\n",
      "154:\tlearn: 0.2519203\ttotal: 6.06s\tremaining: 33s\n",
      "155:\tlearn: 0.2515921\ttotal: 6.09s\tremaining: 33s\n",
      "156:\tlearn: 0.2512701\ttotal: 6.13s\tremaining: 32.9s\n",
      "157:\tlearn: 0.2509579\ttotal: 6.16s\tremaining: 32.8s\n",
      "158:\tlearn: 0.2506522\ttotal: 6.2s\tremaining: 32.8s\n",
      "159:\tlearn: 0.2503480\ttotal: 6.23s\tremaining: 32.7s\n",
      "160:\tlearn: 0.2500489\ttotal: 6.27s\tremaining: 32.7s\n",
      "161:\tlearn: 0.2497572\ttotal: 6.3s\tremaining: 32.6s\n",
      "162:\tlearn: 0.2494745\ttotal: 6.34s\tremaining: 32.5s\n",
      "163:\tlearn: 0.2491894\ttotal: 6.37s\tremaining: 32.5s\n",
      "164:\tlearn: 0.2489123\ttotal: 6.41s\tremaining: 32.4s\n",
      "165:\tlearn: 0.2486345\ttotal: 6.44s\tremaining: 32.3s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "166:\tlearn: 0.2483624\ttotal: 6.48s\tremaining: 32.3s\n",
      "167:\tlearn: 0.2480952\ttotal: 6.51s\tremaining: 32.2s\n",
      "168:\tlearn: 0.2478364\ttotal: 6.55s\tremaining: 32.2s\n",
      "169:\tlearn: 0.2475800\ttotal: 6.58s\tremaining: 32.2s\n",
      "170:\tlearn: 0.2473318\ttotal: 6.62s\tremaining: 32.1s\n",
      "171:\tlearn: 0.2470841\ttotal: 6.66s\tremaining: 32.1s\n",
      "172:\tlearn: 0.2468403\ttotal: 6.71s\tremaining: 32.1s\n",
      "173:\tlearn: 0.2465988\ttotal: 6.74s\tremaining: 32s\n",
      "174:\tlearn: 0.2463623\ttotal: 6.78s\tremaining: 32s\n",
      "175:\tlearn: 0.2461301\ttotal: 6.82s\tremaining: 31.9s\n",
      "176:\tlearn: 0.2459005\ttotal: 6.86s\tremaining: 31.9s\n",
      "177:\tlearn: 0.2456802\ttotal: 6.9s\tremaining: 31.8s\n",
      "178:\tlearn: 0.2454685\ttotal: 6.94s\tremaining: 31.8s\n",
      "179:\tlearn: 0.2452494\ttotal: 6.97s\tremaining: 31.8s\n",
      "180:\tlearn: 0.2450387\ttotal: 7.01s\tremaining: 31.7s\n",
      "181:\tlearn: 0.2448321\ttotal: 7.05s\tremaining: 31.7s\n",
      "182:\tlearn: 0.2446227\ttotal: 7.08s\tremaining: 31.6s\n",
      "183:\tlearn: 0.2444227\ttotal: 7.11s\tremaining: 31.6s\n",
      "184:\tlearn: 0.2442284\ttotal: 7.15s\tremaining: 31.5s\n",
      "185:\tlearn: 0.2440335\ttotal: 7.19s\tremaining: 31.5s\n",
      "186:\tlearn: 0.2438382\ttotal: 7.23s\tremaining: 31.4s\n",
      "187:\tlearn: 0.2436514\ttotal: 7.27s\tremaining: 31.4s\n",
      "188:\tlearn: 0.2434645\ttotal: 7.3s\tremaining: 31.3s\n",
      "189:\tlearn: 0.2432835\ttotal: 7.34s\tremaining: 31.3s\n",
      "190:\tlearn: 0.2431058\ttotal: 7.38s\tremaining: 31.2s\n",
      "191:\tlearn: 0.2429333\ttotal: 7.42s\tremaining: 31.2s\n",
      "192:\tlearn: 0.2427618\ttotal: 7.46s\tremaining: 31.2s\n",
      "193:\tlearn: 0.2425962\ttotal: 7.49s\tremaining: 31.1s\n",
      "194:\tlearn: 0.2424334\ttotal: 7.53s\tremaining: 31.1s\n",
      "195:\tlearn: 0.2422749\ttotal: 7.56s\tremaining: 31s\n",
      "196:\tlearn: 0.2421146\ttotal: 7.6s\tremaining: 31s\n",
      "197:\tlearn: 0.2419573\ttotal: 7.63s\tremaining: 30.9s\n",
      "198:\tlearn: 0.2418029\ttotal: 7.67s\tremaining: 30.9s\n",
      "199:\tlearn: 0.2416520\ttotal: 7.71s\tremaining: 30.8s\n",
      "200:\tlearn: 0.2415016\ttotal: 7.74s\tremaining: 30.8s\n",
      "201:\tlearn: 0.2413529\ttotal: 7.78s\tremaining: 30.7s\n",
      "202:\tlearn: 0.2412056\ttotal: 7.81s\tremaining: 30.7s\n",
      "203:\tlearn: 0.2410617\ttotal: 7.84s\tremaining: 30.6s\n",
      "204:\tlearn: 0.2409235\ttotal: 7.88s\tremaining: 30.6s\n",
      "205:\tlearn: 0.2407897\ttotal: 7.91s\tremaining: 30.5s\n",
      "206:\tlearn: 0.2406549\ttotal: 7.95s\tremaining: 30.5s\n",
      "207:\tlearn: 0.2405244\ttotal: 7.99s\tremaining: 30.4s\n",
      "208:\tlearn: 0.2403899\ttotal: 8.03s\tremaining: 30.4s\n",
      "209:\tlearn: 0.2402621\ttotal: 8.06s\tremaining: 30.3s\n",
      "210:\tlearn: 0.2401349\ttotal: 8.09s\tremaining: 30.3s\n",
      "211:\tlearn: 0.2400118\ttotal: 8.13s\tremaining: 30.2s\n",
      "212:\tlearn: 0.2398910\ttotal: 8.16s\tremaining: 30.2s\n",
      "213:\tlearn: 0.2397701\ttotal: 8.2s\tremaining: 30.1s\n",
      "214:\tlearn: 0.2396535\ttotal: 8.23s\tremaining: 30.1s\n",
      "215:\tlearn: 0.2395370\ttotal: 8.27s\tremaining: 30s\n",
      "216:\tlearn: 0.2394232\ttotal: 8.3s\tremaining: 29.9s\n",
      "217:\tlearn: 0.2393124\ttotal: 8.33s\tremaining: 29.9s\n",
      "218:\tlearn: 0.2392065\ttotal: 8.37s\tremaining: 29.8s\n",
      "219:\tlearn: 0.2390991\ttotal: 8.4s\tremaining: 29.8s\n",
      "220:\tlearn: 0.2389951\ttotal: 8.44s\tremaining: 29.7s\n",
      "221:\tlearn: 0.2388892\ttotal: 8.47s\tremaining: 29.7s\n",
      "222:\tlearn: 0.2387854\ttotal: 8.51s\tremaining: 29.6s\n",
      "223:\tlearn: 0.2386852\ttotal: 8.54s\tremaining: 29.6s\n",
      "224:\tlearn: 0.2385900\ttotal: 8.57s\tremaining: 29.5s\n",
      "225:\tlearn: 0.2384903\ttotal: 8.61s\tremaining: 29.5s\n",
      "226:\tlearn: 0.2383931\ttotal: 8.64s\tremaining: 29.4s\n",
      "227:\tlearn: 0.2382989\ttotal: 8.68s\tremaining: 29.4s\n",
      "228:\tlearn: 0.2382067\ttotal: 8.71s\tremaining: 29.3s\n",
      "229:\tlearn: 0.2381159\ttotal: 8.75s\tremaining: 29.3s\n",
      "230:\tlearn: 0.2380264\ttotal: 8.78s\tremaining: 29.2s\n",
      "231:\tlearn: 0.2379394\ttotal: 8.82s\tremaining: 29.2s\n",
      "232:\tlearn: 0.2378515\ttotal: 8.85s\tremaining: 29.1s\n",
      "233:\tlearn: 0.2377673\ttotal: 8.88s\tremaining: 29.1s\n",
      "234:\tlearn: 0.2376874\ttotal: 8.92s\tremaining: 29s\n",
      "235:\tlearn: 0.2376037\ttotal: 8.95s\tremaining: 29s\n",
      "236:\tlearn: 0.2375236\ttotal: 8.99s\tremaining: 28.9s\n",
      "237:\tlearn: 0.2374429\ttotal: 9.03s\tremaining: 28.9s\n",
      "238:\tlearn: 0.2373671\ttotal: 9.06s\tremaining: 28.9s\n",
      "239:\tlearn: 0.2372902\ttotal: 9.1s\tremaining: 28.8s\n",
      "240:\tlearn: 0.2372131\ttotal: 9.13s\tremaining: 28.8s\n",
      "241:\tlearn: 0.2371400\ttotal: 9.17s\tremaining: 28.7s\n",
      "242:\tlearn: 0.2370706\ttotal: 9.21s\tremaining: 28.7s\n",
      "243:\tlearn: 0.2369980\ttotal: 9.24s\tremaining: 28.6s\n",
      "244:\tlearn: 0.2369267\ttotal: 9.27s\tremaining: 28.6s\n",
      "245:\tlearn: 0.2368600\ttotal: 9.31s\tremaining: 28.5s\n",
      "246:\tlearn: 0.2367911\ttotal: 9.35s\tremaining: 28.5s\n",
      "247:\tlearn: 0.2367240\ttotal: 9.38s\tremaining: 28.5s\n",
      "248:\tlearn: 0.2366606\ttotal: 9.42s\tremaining: 28.4s\n",
      "249:\tlearn: 0.2365981\ttotal: 9.45s\tremaining: 28.4s\n",
      "250:\tlearn: 0.2365349\ttotal: 9.49s\tremaining: 28.3s\n",
      "251:\tlearn: 0.2364712\ttotal: 9.52s\tremaining: 28.3s\n",
      "252:\tlearn: 0.2364106\ttotal: 9.56s\tremaining: 28.2s\n",
      "253:\tlearn: 0.2363511\ttotal: 9.6s\tremaining: 28.2s\n",
      "254:\tlearn: 0.2362920\ttotal: 9.63s\tremaining: 28.2s\n",
      "255:\tlearn: 0.2362397\ttotal: 9.67s\tremaining: 28.1s\n",
      "256:\tlearn: 0.2361826\ttotal: 9.71s\tremaining: 28.1s\n",
      "257:\tlearn: 0.2361282\ttotal: 9.75s\tremaining: 28s\n",
      "258:\tlearn: 0.2360737\ttotal: 9.78s\tremaining: 28s\n",
      "259:\tlearn: 0.2360178\ttotal: 9.82s\tremaining: 27.9s\n",
      "260:\tlearn: 0.2359645\ttotal: 9.85s\tremaining: 27.9s\n",
      "261:\tlearn: 0.2359120\ttotal: 9.89s\tremaining: 27.9s\n",
      "262:\tlearn: 0.2358631\ttotal: 9.92s\tremaining: 27.8s\n",
      "263:\tlearn: 0.2358130\ttotal: 9.95s\tremaining: 27.7s\n",
      "264:\tlearn: 0.2357633\ttotal: 9.98s\tremaining: 27.7s\n",
      "265:\tlearn: 0.2357146\ttotal: 10s\tremaining: 27.6s\n",
      "266:\tlearn: 0.2356660\ttotal: 10.1s\tremaining: 27.6s\n",
      "267:\tlearn: 0.2356185\ttotal: 10.1s\tremaining: 27.6s\n",
      "268:\tlearn: 0.2355766\ttotal: 10.1s\tremaining: 27.5s\n",
      "269:\tlearn: 0.2355300\ttotal: 10.2s\tremaining: 27.5s\n",
      "270:\tlearn: 0.2354838\ttotal: 10.2s\tremaining: 27.4s\n",
      "271:\tlearn: 0.2354392\ttotal: 10.2s\tremaining: 27.4s\n",
      "272:\tlearn: 0.2353945\ttotal: 10.3s\tremaining: 27.4s\n",
      "273:\tlearn: 0.2353499\ttotal: 10.3s\tremaining: 27.3s\n",
      "274:\tlearn: 0.2353098\ttotal: 10.3s\tremaining: 27.3s\n",
      "275:\tlearn: 0.2352683\ttotal: 10.4s\tremaining: 27.2s\n",
      "276:\tlearn: 0.2352302\ttotal: 10.4s\tremaining: 27.2s\n",
      "277:\tlearn: 0.2351888\ttotal: 10.4s\tremaining: 27.1s\n",
      "278:\tlearn: 0.2351482\ttotal: 10.5s\tremaining: 27.1s\n",
      "279:\tlearn: 0.2351108\ttotal: 10.5s\tremaining: 27s\n",
      "280:\tlearn: 0.2350711\ttotal: 10.5s\tremaining: 27s\n",
      "281:\tlearn: 0.2350334\ttotal: 10.6s\tremaining: 26.9s\n",
      "282:\tlearn: 0.2349970\ttotal: 10.6s\tremaining: 26.9s\n",
      "283:\tlearn: 0.2349598\ttotal: 10.7s\tremaining: 26.9s\n",
      "284:\tlearn: 0.2349259\ttotal: 10.7s\tremaining: 26.8s\n",
      "285:\tlearn: 0.2348895\ttotal: 10.7s\tremaining: 26.8s\n",
      "286:\tlearn: 0.2348562\ttotal: 10.8s\tremaining: 26.8s\n",
      "287:\tlearn: 0.2348216\ttotal: 10.8s\tremaining: 26.7s\n",
      "288:\tlearn: 0.2347889\ttotal: 10.8s\tremaining: 26.7s\n",
      "289:\tlearn: 0.2347573\ttotal: 10.9s\tremaining: 26.6s\n",
      "290:\tlearn: 0.2347260\ttotal: 10.9s\tremaining: 26.6s\n",
      "291:\tlearn: 0.2346962\ttotal: 10.9s\tremaining: 26.5s\n",
      "292:\tlearn: 0.2346654\ttotal: 11s\tremaining: 26.5s\n",
      "293:\tlearn: 0.2346345\ttotal: 11s\tremaining: 26.4s\n",
      "294:\tlearn: 0.2346062\ttotal: 11.1s\tremaining: 26.4s\n",
      "295:\tlearn: 0.2345766\ttotal: 11.1s\tremaining: 26.4s\n",
      "296:\tlearn: 0.2345472\ttotal: 11.1s\tremaining: 26.3s\n",
      "297:\tlearn: 0.2345188\ttotal: 11.2s\tremaining: 26.3s\n",
      "298:\tlearn: 0.2344908\ttotal: 11.2s\tremaining: 26.2s\n",
      "299:\tlearn: 0.2344627\ttotal: 11.2s\tremaining: 26.2s\n",
      "300:\tlearn: 0.2344340\ttotal: 11.3s\tremaining: 26.2s\n",
      "301:\tlearn: 0.2344081\ttotal: 11.3s\tremaining: 26.1s\n",
      "302:\tlearn: 0.2343835\ttotal: 11.3s\tremaining: 26.1s\n",
      "303:\tlearn: 0.2343574\ttotal: 11.4s\tremaining: 26s\n",
      "304:\tlearn: 0.2343331\ttotal: 11.4s\tremaining: 26s\n",
      "305:\tlearn: 0.2343103\ttotal: 11.4s\tremaining: 25.9s\n",
      "306:\tlearn: 0.2342852\ttotal: 11.5s\tremaining: 25.9s\n",
      "307:\tlearn: 0.2342607\ttotal: 11.5s\tremaining: 25.8s\n",
      "308:\tlearn: 0.2342367\ttotal: 11.5s\tremaining: 25.8s\n",
      "309:\tlearn: 0.2342125\ttotal: 11.6s\tremaining: 25.8s\n",
      "310:\tlearn: 0.2341891\ttotal: 11.6s\tremaining: 25.7s\n",
      "311:\tlearn: 0.2341661\ttotal: 11.7s\tremaining: 25.7s\n",
      "312:\tlearn: 0.2341433\ttotal: 11.7s\tremaining: 25.6s\n",
      "313:\tlearn: 0.2341221\ttotal: 11.7s\tremaining: 25.6s\n",
      "314:\tlearn: 0.2341011\ttotal: 11.8s\tremaining: 25.6s\n",
      "315:\tlearn: 0.2340811\ttotal: 11.8s\tremaining: 25.5s\n",
      "316:\tlearn: 0.2340601\ttotal: 11.8s\tremaining: 25.5s\n",
      "317:\tlearn: 0.2340388\ttotal: 11.9s\tremaining: 25.4s\n",
      "318:\tlearn: 0.2340193\ttotal: 11.9s\tremaining: 25.4s\n",
      "319:\tlearn: 0.2339991\ttotal: 11.9s\tremaining: 25.3s\n",
      "320:\tlearn: 0.2339809\ttotal: 12s\tremaining: 25.3s\n",
      "321:\tlearn: 0.2339616\ttotal: 12s\tremaining: 25.3s\n",
      "322:\tlearn: 0.2339434\ttotal: 12s\tremaining: 25.2s\n",
      "323:\tlearn: 0.2339252\ttotal: 12.1s\tremaining: 25.2s\n",
      "324:\tlearn: 0.2339067\ttotal: 12.1s\tremaining: 25.2s\n",
      "325:\tlearn: 0.2338883\ttotal: 12.2s\tremaining: 25.1s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "326:\tlearn: 0.2338696\ttotal: 12.2s\tremaining: 25.1s\n",
      "327:\tlearn: 0.2338512\ttotal: 12.2s\tremaining: 25.1s\n",
      "328:\tlearn: 0.2338335\ttotal: 12.3s\tremaining: 25s\n",
      "329:\tlearn: 0.2338171\ttotal: 12.3s\tremaining: 25s\n",
      "330:\tlearn: 0.2338011\ttotal: 12.3s\tremaining: 24.9s\n",
      "331:\tlearn: 0.2337844\ttotal: 12.4s\tremaining: 24.9s\n",
      "332:\tlearn: 0.2337676\ttotal: 12.4s\tremaining: 24.9s\n",
      "333:\tlearn: 0.2337504\ttotal: 12.5s\tremaining: 24.8s\n",
      "334:\tlearn: 0.2337341\ttotal: 12.5s\tremaining: 24.8s\n",
      "335:\tlearn: 0.2337202\ttotal: 12.5s\tremaining: 24.7s\n",
      "336:\tlearn: 0.2337040\ttotal: 12.6s\tremaining: 24.7s\n",
      "337:\tlearn: 0.2336880\ttotal: 12.6s\tremaining: 24.7s\n",
      "338:\tlearn: 0.2336744\ttotal: 12.6s\tremaining: 24.6s\n",
      "339:\tlearn: 0.2336622\ttotal: 12.7s\tremaining: 24.6s\n",
      "340:\tlearn: 0.2336474\ttotal: 12.7s\tremaining: 24.6s\n",
      "341:\tlearn: 0.2336335\ttotal: 12.8s\tremaining: 24.5s\n",
      "342:\tlearn: 0.2336195\ttotal: 12.8s\tremaining: 24.5s\n",
      "343:\tlearn: 0.2336064\ttotal: 12.8s\tremaining: 24.5s\n",
      "344:\tlearn: 0.2335939\ttotal: 12.9s\tremaining: 24.4s\n",
      "345:\tlearn: 0.2335801\ttotal: 12.9s\tremaining: 24.4s\n",
      "346:\tlearn: 0.2335671\ttotal: 12.9s\tremaining: 24.4s\n",
      "347:\tlearn: 0.2335537\ttotal: 13s\tremaining: 24.3s\n",
      "348:\tlearn: 0.2335419\ttotal: 13s\tremaining: 24.3s\n",
      "349:\tlearn: 0.2335304\ttotal: 13s\tremaining: 24.2s\n",
      "350:\tlearn: 0.2335190\ttotal: 13.1s\tremaining: 24.2s\n",
      "351:\tlearn: 0.2335058\ttotal: 13.1s\tremaining: 24.1s\n",
      "352:\tlearn: 0.2334945\ttotal: 13.1s\tremaining: 24.1s\n",
      "353:\tlearn: 0.2334823\ttotal: 13.2s\tremaining: 24.1s\n",
      "354:\tlearn: 0.2334708\ttotal: 13.2s\tremaining: 24s\n",
      "355:\tlearn: 0.2334596\ttotal: 13.3s\tremaining: 24s\n",
      "356:\tlearn: 0.2334487\ttotal: 13.3s\tremaining: 23.9s\n",
      "357:\tlearn: 0.2334379\ttotal: 13.3s\tremaining: 23.9s\n",
      "358:\tlearn: 0.2334271\ttotal: 13.4s\tremaining: 23.9s\n",
      "359:\tlearn: 0.2334155\ttotal: 13.4s\tremaining: 23.8s\n",
      "360:\tlearn: 0.2334062\ttotal: 13.4s\tremaining: 23.8s\n",
      "361:\tlearn: 0.2333969\ttotal: 13.5s\tremaining: 23.7s\n",
      "362:\tlearn: 0.2333869\ttotal: 13.5s\tremaining: 23.7s\n",
      "363:\tlearn: 0.2333753\ttotal: 13.5s\tremaining: 23.7s\n",
      "364:\tlearn: 0.2333645\ttotal: 13.6s\tremaining: 23.6s\n",
      "365:\tlearn: 0.2333548\ttotal: 13.6s\tremaining: 23.6s\n",
      "366:\tlearn: 0.2333463\ttotal: 13.7s\tremaining: 23.5s\n",
      "367:\tlearn: 0.2333372\ttotal: 13.7s\tremaining: 23.5s\n",
      "368:\tlearn: 0.2333273\ttotal: 13.7s\tremaining: 23.5s\n",
      "369:\tlearn: 0.2333189\ttotal: 13.8s\tremaining: 23.4s\n",
      "370:\tlearn: 0.2333093\ttotal: 13.8s\tremaining: 23.4s\n",
      "371:\tlearn: 0.2333008\ttotal: 13.8s\tremaining: 23.3s\n",
      "372:\tlearn: 0.2332907\ttotal: 13.9s\tremaining: 23.3s\n",
      "373:\tlearn: 0.2332832\ttotal: 13.9s\tremaining: 23.3s\n",
      "374:\tlearn: 0.2332748\ttotal: 13.9s\tremaining: 23.2s\n",
      "375:\tlearn: 0.2332660\ttotal: 14s\tremaining: 23.2s\n",
      "376:\tlearn: 0.2332566\ttotal: 14s\tremaining: 23.1s\n",
      "377:\tlearn: 0.2332491\ttotal: 14s\tremaining: 23.1s\n",
      "378:\tlearn: 0.2332420\ttotal: 14.1s\tremaining: 23s\n",
      "379:\tlearn: 0.2332337\ttotal: 14.1s\tremaining: 23s\n",
      "380:\tlearn: 0.2332256\ttotal: 14.1s\tremaining: 23s\n",
      "381:\tlearn: 0.2332190\ttotal: 14.2s\tremaining: 22.9s\n",
      "382:\tlearn: 0.2332121\ttotal: 14.2s\tremaining: 22.9s\n",
      "383:\tlearn: 0.2332051\ttotal: 14.2s\tremaining: 22.8s\n",
      "384:\tlearn: 0.2331965\ttotal: 14.3s\tremaining: 22.8s\n",
      "385:\tlearn: 0.2331881\ttotal: 14.3s\tremaining: 22.8s\n",
      "386:\tlearn: 0.2331803\ttotal: 14.3s\tremaining: 22.7s\n",
      "387:\tlearn: 0.2331729\ttotal: 14.4s\tremaining: 22.7s\n",
      "388:\tlearn: 0.2331657\ttotal: 14.4s\tremaining: 22.6s\n",
      "389:\tlearn: 0.2331587\ttotal: 14.5s\tremaining: 22.6s\n",
      "390:\tlearn: 0.2331515\ttotal: 14.5s\tremaining: 22.6s\n",
      "391:\tlearn: 0.2331444\ttotal: 14.5s\tremaining: 22.5s\n",
      "392:\tlearn: 0.2331382\ttotal: 14.6s\tremaining: 22.5s\n",
      "393:\tlearn: 0.2331324\ttotal: 14.6s\tremaining: 22.4s\n",
      "394:\tlearn: 0.2331246\ttotal: 14.6s\tremaining: 22.4s\n",
      "395:\tlearn: 0.2331174\ttotal: 14.7s\tremaining: 22.4s\n",
      "396:\tlearn: 0.2331109\ttotal: 14.7s\tremaining: 22.3s\n",
      "397:\tlearn: 0.2331051\ttotal: 14.7s\tremaining: 22.3s\n",
      "398:\tlearn: 0.2330989\ttotal: 14.8s\tremaining: 22.2s\n",
      "399:\tlearn: 0.2330930\ttotal: 14.8s\tremaining: 22.2s\n",
      "400:\tlearn: 0.2330874\ttotal: 14.8s\tremaining: 22.2s\n",
      "401:\tlearn: 0.2330812\ttotal: 14.9s\tremaining: 22.1s\n",
      "402:\tlearn: 0.2330762\ttotal: 14.9s\tremaining: 22.1s\n",
      "403:\tlearn: 0.2330696\ttotal: 15s\tremaining: 22.1s\n",
      "404:\tlearn: 0.2330613\ttotal: 15s\tremaining: 22s\n",
      "405:\tlearn: 0.2330551\ttotal: 15s\tremaining: 22s\n",
      "406:\tlearn: 0.2330494\ttotal: 15.1s\tremaining: 22s\n",
      "407:\tlearn: 0.2330442\ttotal: 15.1s\tremaining: 21.9s\n",
      "408:\tlearn: 0.2330376\ttotal: 15.1s\tremaining: 21.9s\n",
      "409:\tlearn: 0.2330311\ttotal: 15.2s\tremaining: 21.9s\n",
      "410:\tlearn: 0.2330264\ttotal: 15.2s\tremaining: 21.8s\n",
      "411:\tlearn: 0.2330220\ttotal: 15.2s\tremaining: 21.8s\n",
      "412:\tlearn: 0.2330178\ttotal: 15.3s\tremaining: 21.7s\n",
      "413:\tlearn: 0.2330122\ttotal: 15.3s\tremaining: 21.7s\n",
      "414:\tlearn: 0.2330073\ttotal: 15.4s\tremaining: 21.6s\n",
      "415:\tlearn: 0.2330022\ttotal: 15.4s\tremaining: 21.6s\n",
      "416:\tlearn: 0.2329979\ttotal: 15.4s\tremaining: 21.6s\n",
      "417:\tlearn: 0.2329930\ttotal: 15.5s\tremaining: 21.5s\n",
      "418:\tlearn: 0.2329884\ttotal: 15.5s\tremaining: 21.5s\n",
      "419:\tlearn: 0.2329841\ttotal: 15.5s\tremaining: 21.4s\n",
      "420:\tlearn: 0.2329792\ttotal: 15.6s\tremaining: 21.4s\n",
      "421:\tlearn: 0.2329747\ttotal: 15.6s\tremaining: 21.4s\n",
      "422:\tlearn: 0.2329682\ttotal: 15.6s\tremaining: 21.3s\n",
      "423:\tlearn: 0.2329647\ttotal: 15.7s\tremaining: 21.3s\n",
      "424:\tlearn: 0.2329608\ttotal: 15.7s\tremaining: 21.3s\n",
      "425:\tlearn: 0.2329556\ttotal: 15.7s\tremaining: 21.2s\n",
      "426:\tlearn: 0.2329510\ttotal: 15.8s\tremaining: 21.2s\n",
      "427:\tlearn: 0.2329470\ttotal: 15.8s\tremaining: 21.1s\n",
      "428:\tlearn: 0.2329428\ttotal: 15.8s\tremaining: 21.1s\n",
      "429:\tlearn: 0.2329359\ttotal: 15.9s\tremaining: 21.1s\n",
      "430:\tlearn: 0.2329313\ttotal: 15.9s\tremaining: 21s\n",
      "431:\tlearn: 0.2329253\ttotal: 16s\tremaining: 21s\n",
      "432:\tlearn: 0.2329215\ttotal: 16s\tremaining: 20.9s\n",
      "433:\tlearn: 0.2329180\ttotal: 16s\tremaining: 20.9s\n",
      "434:\tlearn: 0.2329144\ttotal: 16.1s\tremaining: 20.9s\n",
      "435:\tlearn: 0.2329107\ttotal: 16.1s\tremaining: 20.8s\n",
      "436:\tlearn: 0.2329067\ttotal: 16.1s\tremaining: 20.8s\n",
      "437:\tlearn: 0.2329033\ttotal: 16.2s\tremaining: 20.7s\n",
      "438:\tlearn: 0.2328984\ttotal: 16.2s\tremaining: 20.7s\n",
      "439:\tlearn: 0.2328948\ttotal: 16.2s\tremaining: 20.7s\n",
      "440:\tlearn: 0.2328913\ttotal: 16.3s\tremaining: 20.6s\n",
      "441:\tlearn: 0.2328844\ttotal: 16.3s\tremaining: 20.6s\n",
      "442:\tlearn: 0.2328810\ttotal: 16.3s\tremaining: 20.6s\n",
      "443:\tlearn: 0.2328765\ttotal: 16.4s\tremaining: 20.5s\n",
      "444:\tlearn: 0.2328726\ttotal: 16.4s\tremaining: 20.5s\n",
      "445:\tlearn: 0.2328682\ttotal: 16.5s\tremaining: 20.4s\n",
      "446:\tlearn: 0.2328654\ttotal: 16.5s\tremaining: 20.4s\n",
      "447:\tlearn: 0.2328625\ttotal: 16.5s\tremaining: 20.3s\n",
      "448:\tlearn: 0.2328599\ttotal: 16.5s\tremaining: 20.3s\n",
      "449:\tlearn: 0.2328570\ttotal: 16.6s\tremaining: 20.3s\n",
      "450:\tlearn: 0.2328542\ttotal: 16.6s\tremaining: 20.2s\n",
      "451:\tlearn: 0.2328513\ttotal: 16.7s\tremaining: 20.2s\n",
      "452:\tlearn: 0.2328483\ttotal: 16.7s\tremaining: 20.2s\n",
      "453:\tlearn: 0.2328451\ttotal: 16.7s\tremaining: 20.1s\n",
      "454:\tlearn: 0.2328416\ttotal: 16.8s\tremaining: 20.1s\n",
      "455:\tlearn: 0.2328367\ttotal: 16.8s\tremaining: 20s\n",
      "456:\tlearn: 0.2328336\ttotal: 16.8s\tremaining: 20s\n",
      "457:\tlearn: 0.2328312\ttotal: 16.9s\tremaining: 20s\n",
      "458:\tlearn: 0.2328283\ttotal: 16.9s\tremaining: 19.9s\n",
      "459:\tlearn: 0.2328257\ttotal: 16.9s\tremaining: 19.9s\n",
      "460:\tlearn: 0.2328233\ttotal: 17s\tremaining: 19.8s\n",
      "461:\tlearn: 0.2328209\ttotal: 17s\tremaining: 19.8s\n",
      "462:\tlearn: 0.2328167\ttotal: 17s\tremaining: 19.8s\n",
      "463:\tlearn: 0.2328150\ttotal: 17.1s\tremaining: 19.7s\n",
      "464:\tlearn: 0.2328121\ttotal: 17.1s\tremaining: 19.7s\n",
      "465:\tlearn: 0.2328084\ttotal: 17.1s\tremaining: 19.6s\n",
      "466:\tlearn: 0.2328044\ttotal: 17.2s\tremaining: 19.6s\n",
      "467:\tlearn: 0.2328014\ttotal: 17.2s\tremaining: 19.6s\n",
      "468:\tlearn: 0.2327993\ttotal: 17.3s\tremaining: 19.5s\n",
      "469:\tlearn: 0.2327965\ttotal: 17.3s\tremaining: 19.5s\n",
      "470:\tlearn: 0.2327930\ttotal: 17.3s\tremaining: 19.5s\n",
      "471:\tlearn: 0.2327902\ttotal: 17.4s\tremaining: 19.4s\n",
      "472:\tlearn: 0.2327884\ttotal: 17.4s\tremaining: 19.4s\n",
      "473:\tlearn: 0.2327860\ttotal: 17.4s\tremaining: 19.4s\n",
      "474:\tlearn: 0.2327836\ttotal: 17.5s\tremaining: 19.3s\n",
      "475:\tlearn: 0.2327807\ttotal: 17.5s\tremaining: 19.3s\n",
      "476:\tlearn: 0.2327780\ttotal: 17.5s\tremaining: 19.2s\n",
      "477:\tlearn: 0.2327750\ttotal: 17.6s\tremaining: 19.2s\n",
      "478:\tlearn: 0.2327726\ttotal: 17.6s\tremaining: 19.2s\n",
      "479:\tlearn: 0.2327711\ttotal: 17.7s\tremaining: 19.1s\n",
      "480:\tlearn: 0.2327677\ttotal: 17.7s\tremaining: 19.1s\n",
      "481:\tlearn: 0.2327657\ttotal: 17.7s\tremaining: 19.1s\n",
      "482:\tlearn: 0.2327628\ttotal: 17.8s\tremaining: 19s\n",
      "483:\tlearn: 0.2327596\ttotal: 17.8s\tremaining: 19s\n",
      "484:\tlearn: 0.2327567\ttotal: 17.8s\tremaining: 19s\n",
      "485:\tlearn: 0.2327527\ttotal: 17.9s\tremaining: 18.9s\n",
      "486:\tlearn: 0.2327507\ttotal: 17.9s\tremaining: 18.9s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "487:\tlearn: 0.2327487\ttotal: 18s\tremaining: 18.8s\n",
      "488:\tlearn: 0.2327467\ttotal: 18s\tremaining: 18.8s\n",
      "489:\tlearn: 0.2327445\ttotal: 18s\tremaining: 18.8s\n",
      "490:\tlearn: 0.2327426\ttotal: 18.1s\tremaining: 18.7s\n",
      "491:\tlearn: 0.2327380\ttotal: 18.1s\tremaining: 18.7s\n",
      "492:\tlearn: 0.2327347\ttotal: 18.1s\tremaining: 18.7s\n",
      "493:\tlearn: 0.2327317\ttotal: 18.2s\tremaining: 18.6s\n",
      "494:\tlearn: 0.2327275\ttotal: 18.2s\tremaining: 18.6s\n",
      "495:\tlearn: 0.2327241\ttotal: 18.3s\tremaining: 18.6s\n",
      "496:\tlearn: 0.2327225\ttotal: 18.3s\tremaining: 18.5s\n",
      "497:\tlearn: 0.2327200\ttotal: 18.3s\tremaining: 18.5s\n",
      "498:\tlearn: 0.2327153\ttotal: 18.4s\tremaining: 18.4s\n",
      "499:\tlearn: 0.2327138\ttotal: 18.4s\tremaining: 18.4s\n",
      "500:\tlearn: 0.2327112\ttotal: 18.4s\tremaining: 18.4s\n",
      "501:\tlearn: 0.2327083\ttotal: 18.5s\tremaining: 18.3s\n",
      "502:\tlearn: 0.2327068\ttotal: 18.5s\tremaining: 18.3s\n",
      "503:\tlearn: 0.2327031\ttotal: 18.5s\tremaining: 18.3s\n",
      "504:\tlearn: 0.2327013\ttotal: 18.6s\tremaining: 18.2s\n",
      "505:\tlearn: 0.2326992\ttotal: 18.6s\tremaining: 18.2s\n",
      "506:\tlearn: 0.2326960\ttotal: 18.7s\tremaining: 18.1s\n",
      "507:\tlearn: 0.2326949\ttotal: 18.7s\tremaining: 18.1s\n",
      "508:\tlearn: 0.2326933\ttotal: 18.7s\tremaining: 18.1s\n",
      "509:\tlearn: 0.2326912\ttotal: 18.8s\tremaining: 18s\n",
      "510:\tlearn: 0.2326879\ttotal: 18.8s\tremaining: 18s\n",
      "511:\tlearn: 0.2326857\ttotal: 18.8s\tremaining: 18s\n",
      "512:\tlearn: 0.2326845\ttotal: 18.9s\tremaining: 17.9s\n",
      "513:\tlearn: 0.2326827\ttotal: 18.9s\tremaining: 17.9s\n",
      "514:\tlearn: 0.2326807\ttotal: 18.9s\tremaining: 17.8s\n",
      "515:\tlearn: 0.2326769\ttotal: 19s\tremaining: 17.8s\n",
      "516:\tlearn: 0.2326751\ttotal: 19s\tremaining: 17.8s\n",
      "517:\tlearn: 0.2326740\ttotal: 19s\tremaining: 17.7s\n",
      "518:\tlearn: 0.2326719\ttotal: 19.1s\tremaining: 17.7s\n",
      "519:\tlearn: 0.2326694\ttotal: 19.1s\tremaining: 17.6s\n",
      "520:\tlearn: 0.2326671\ttotal: 19.2s\tremaining: 17.6s\n",
      "521:\tlearn: 0.2326640\ttotal: 19.2s\tremaining: 17.6s\n",
      "522:\tlearn: 0.2326602\ttotal: 19.2s\tremaining: 17.5s\n",
      "523:\tlearn: 0.2326562\ttotal: 19.3s\tremaining: 17.5s\n",
      "524:\tlearn: 0.2326515\ttotal: 19.3s\tremaining: 17.5s\n",
      "525:\tlearn: 0.2326499\ttotal: 19.3s\tremaining: 17.4s\n",
      "526:\tlearn: 0.2326475\ttotal: 19.4s\tremaining: 17.4s\n",
      "527:\tlearn: 0.2326464\ttotal: 19.4s\tremaining: 17.3s\n",
      "528:\tlearn: 0.2326446\ttotal: 19.4s\tremaining: 17.3s\n",
      "529:\tlearn: 0.2326414\ttotal: 19.5s\tremaining: 17.3s\n",
      "530:\tlearn: 0.2326387\ttotal: 19.5s\tremaining: 17.2s\n",
      "531:\tlearn: 0.2326371\ttotal: 19.5s\tremaining: 17.2s\n",
      "532:\tlearn: 0.2326328\ttotal: 19.6s\tremaining: 17.2s\n",
      "533:\tlearn: 0.2326306\ttotal: 19.6s\tremaining: 17.1s\n",
      "534:\tlearn: 0.2326288\ttotal: 19.6s\tremaining: 17.1s\n",
      "535:\tlearn: 0.2326276\ttotal: 19.7s\tremaining: 17s\n",
      "536:\tlearn: 0.2326263\ttotal: 19.7s\tremaining: 17s\n",
      "537:\tlearn: 0.2326250\ttotal: 19.7s\tremaining: 17s\n",
      "538:\tlearn: 0.2326242\ttotal: 19.8s\tremaining: 16.9s\n",
      "539:\tlearn: 0.2326222\ttotal: 19.8s\tremaining: 16.9s\n",
      "540:\tlearn: 0.2326202\ttotal: 19.9s\tremaining: 16.8s\n",
      "541:\tlearn: 0.2326162\ttotal: 19.9s\tremaining: 16.8s\n",
      "542:\tlearn: 0.2326130\ttotal: 19.9s\tremaining: 16.8s\n",
      "543:\tlearn: 0.2326106\ttotal: 20s\tremaining: 16.7s\n",
      "544:\tlearn: 0.2326083\ttotal: 20s\tremaining: 16.7s\n",
      "545:\tlearn: 0.2326057\ttotal: 20s\tremaining: 16.7s\n",
      "546:\tlearn: 0.2326043\ttotal: 20.1s\tremaining: 16.6s\n",
      "547:\tlearn: 0.2326024\ttotal: 20.1s\tremaining: 16.6s\n",
      "548:\tlearn: 0.2325995\ttotal: 20.2s\tremaining: 16.6s\n",
      "549:\tlearn: 0.2325969\ttotal: 20.2s\tremaining: 16.5s\n",
      "550:\tlearn: 0.2325944\ttotal: 20.2s\tremaining: 16.5s\n",
      "551:\tlearn: 0.2325933\ttotal: 20.3s\tremaining: 16.4s\n",
      "552:\tlearn: 0.2325915\ttotal: 20.3s\tremaining: 16.4s\n",
      "553:\tlearn: 0.2325892\ttotal: 20.3s\tremaining: 16.4s\n",
      "554:\tlearn: 0.2325865\ttotal: 20.4s\tremaining: 16.3s\n",
      "555:\tlearn: 0.2325837\ttotal: 20.4s\tremaining: 16.3s\n",
      "556:\tlearn: 0.2325814\ttotal: 20.5s\tremaining: 16.3s\n",
      "557:\tlearn: 0.2325775\ttotal: 20.5s\tremaining: 16.2s\n",
      "558:\tlearn: 0.2325759\ttotal: 20.5s\tremaining: 16.2s\n",
      "559:\tlearn: 0.2325740\ttotal: 20.6s\tremaining: 16.2s\n",
      "560:\tlearn: 0.2325713\ttotal: 20.6s\tremaining: 16.1s\n",
      "561:\tlearn: 0.2325692\ttotal: 20.6s\tremaining: 16.1s\n",
      "562:\tlearn: 0.2325683\ttotal: 20.7s\tremaining: 16s\n",
      "563:\tlearn: 0.2325658\ttotal: 20.7s\tremaining: 16s\n",
      "564:\tlearn: 0.2325650\ttotal: 20.7s\tremaining: 16s\n",
      "565:\tlearn: 0.2325625\ttotal: 20.8s\tremaining: 15.9s\n",
      "566:\tlearn: 0.2325604\ttotal: 20.8s\tremaining: 15.9s\n",
      "567:\tlearn: 0.2325597\ttotal: 20.8s\tremaining: 15.9s\n",
      "568:\tlearn: 0.2325576\ttotal: 20.9s\tremaining: 15.8s\n",
      "569:\tlearn: 0.2325558\ttotal: 20.9s\tremaining: 15.8s\n",
      "570:\tlearn: 0.2325541\ttotal: 21s\tremaining: 15.7s\n",
      "571:\tlearn: 0.2325519\ttotal: 21s\tremaining: 15.7s\n",
      "572:\tlearn: 0.2325504\ttotal: 21s\tremaining: 15.7s\n",
      "573:\tlearn: 0.2325484\ttotal: 21.1s\tremaining: 15.6s\n",
      "574:\tlearn: 0.2325472\ttotal: 21.1s\tremaining: 15.6s\n",
      "575:\tlearn: 0.2325449\ttotal: 21.1s\tremaining: 15.6s\n",
      "576:\tlearn: 0.2325414\ttotal: 21.2s\tremaining: 15.5s\n",
      "577:\tlearn: 0.2325394\ttotal: 21.2s\tremaining: 15.5s\n",
      "578:\tlearn: 0.2325375\ttotal: 21.2s\tremaining: 15.4s\n",
      "579:\tlearn: 0.2325352\ttotal: 21.3s\tremaining: 15.4s\n",
      "580:\tlearn: 0.2325333\ttotal: 21.3s\tremaining: 15.4s\n",
      "581:\tlearn: 0.2325311\ttotal: 21.3s\tremaining: 15.3s\n",
      "582:\tlearn: 0.2325296\ttotal: 21.4s\tremaining: 15.3s\n",
      "583:\tlearn: 0.2325268\ttotal: 21.4s\tremaining: 15.3s\n",
      "584:\tlearn: 0.2325264\ttotal: 21.4s\tremaining: 15.2s\n",
      "585:\tlearn: 0.2325233\ttotal: 21.5s\tremaining: 15.2s\n",
      "586:\tlearn: 0.2325224\ttotal: 21.5s\tremaining: 15.1s\n",
      "587:\tlearn: 0.2325193\ttotal: 21.6s\tremaining: 15.1s\n",
      "588:\tlearn: 0.2325184\ttotal: 21.6s\tremaining: 15.1s\n",
      "589:\tlearn: 0.2325172\ttotal: 21.6s\tremaining: 15s\n",
      "590:\tlearn: 0.2325155\ttotal: 21.7s\tremaining: 15s\n",
      "591:\tlearn: 0.2325143\ttotal: 21.7s\tremaining: 15s\n",
      "592:\tlearn: 0.2325136\ttotal: 21.7s\tremaining: 14.9s\n",
      "593:\tlearn: 0.2325103\ttotal: 21.8s\tremaining: 14.9s\n",
      "594:\tlearn: 0.2325089\ttotal: 21.8s\tremaining: 14.8s\n",
      "595:\tlearn: 0.2325070\ttotal: 21.8s\tremaining: 14.8s\n",
      "596:\tlearn: 0.2325042\ttotal: 21.9s\tremaining: 14.8s\n",
      "597:\tlearn: 0.2325022\ttotal: 21.9s\tremaining: 14.7s\n",
      "598:\tlearn: 0.2324999\ttotal: 21.9s\tremaining: 14.7s\n",
      "599:\tlearn: 0.2324981\ttotal: 22s\tremaining: 14.7s\n",
      "600:\tlearn: 0.2324970\ttotal: 22s\tremaining: 14.6s\n",
      "601:\tlearn: 0.2324956\ttotal: 22.1s\tremaining: 14.6s\n",
      "602:\tlearn: 0.2324918\ttotal: 22.1s\tremaining: 14.5s\n",
      "603:\tlearn: 0.2324901\ttotal: 22.1s\tremaining: 14.5s\n",
      "604:\tlearn: 0.2324863\ttotal: 22.2s\tremaining: 14.5s\n",
      "605:\tlearn: 0.2324841\ttotal: 22.2s\tremaining: 14.4s\n",
      "606:\tlearn: 0.2324828\ttotal: 22.2s\tremaining: 14.4s\n",
      "607:\tlearn: 0.2324815\ttotal: 22.3s\tremaining: 14.4s\n",
      "608:\tlearn: 0.2324799\ttotal: 22.3s\tremaining: 14.3s\n",
      "609:\tlearn: 0.2324783\ttotal: 22.3s\tremaining: 14.3s\n",
      "610:\tlearn: 0.2324763\ttotal: 22.4s\tremaining: 14.2s\n",
      "611:\tlearn: 0.2324741\ttotal: 22.4s\tremaining: 14.2s\n",
      "612:\tlearn: 0.2324721\ttotal: 22.4s\tremaining: 14.2s\n",
      "613:\tlearn: 0.2324707\ttotal: 22.5s\tremaining: 14.1s\n",
      "614:\tlearn: 0.2324694\ttotal: 22.5s\tremaining: 14.1s\n",
      "615:\tlearn: 0.2324682\ttotal: 22.5s\tremaining: 14.1s\n",
      "616:\tlearn: 0.2324672\ttotal: 22.6s\tremaining: 14s\n",
      "617:\tlearn: 0.2324656\ttotal: 22.6s\tremaining: 14s\n",
      "618:\tlearn: 0.2324643\ttotal: 22.6s\tremaining: 13.9s\n",
      "619:\tlearn: 0.2324630\ttotal: 22.7s\tremaining: 13.9s\n",
      "620:\tlearn: 0.2324605\ttotal: 22.7s\tremaining: 13.9s\n",
      "621:\tlearn: 0.2324587\ttotal: 22.7s\tremaining: 13.8s\n",
      "622:\tlearn: 0.2324577\ttotal: 22.8s\tremaining: 13.8s\n",
      "623:\tlearn: 0.2324560\ttotal: 22.8s\tremaining: 13.7s\n",
      "624:\tlearn: 0.2324548\ttotal: 22.9s\tremaining: 13.7s\n",
      "625:\tlearn: 0.2324531\ttotal: 22.9s\tremaining: 13.7s\n",
      "626:\tlearn: 0.2324518\ttotal: 22.9s\tremaining: 13.6s\n",
      "627:\tlearn: 0.2324501\ttotal: 23s\tremaining: 13.6s\n",
      "628:\tlearn: 0.2324479\ttotal: 23s\tremaining: 13.6s\n",
      "629:\tlearn: 0.2324472\ttotal: 23s\tremaining: 13.5s\n",
      "630:\tlearn: 0.2324464\ttotal: 23.1s\tremaining: 13.5s\n",
      "631:\tlearn: 0.2324439\ttotal: 23.1s\tremaining: 13.5s\n",
      "632:\tlearn: 0.2324433\ttotal: 23.2s\tremaining: 13.4s\n",
      "633:\tlearn: 0.2324428\ttotal: 23.2s\tremaining: 13.4s\n",
      "634:\tlearn: 0.2324410\ttotal: 23.2s\tremaining: 13.4s\n",
      "635:\tlearn: 0.2324402\ttotal: 23.3s\tremaining: 13.3s\n",
      "636:\tlearn: 0.2324390\ttotal: 23.3s\tremaining: 13.3s\n",
      "637:\tlearn: 0.2324371\ttotal: 23.3s\tremaining: 13.2s\n",
      "638:\tlearn: 0.2324341\ttotal: 23.4s\tremaining: 13.2s\n",
      "639:\tlearn: 0.2324321\ttotal: 23.4s\tremaining: 13.2s\n",
      "640:\tlearn: 0.2324301\ttotal: 23.4s\tremaining: 13.1s\n",
      "641:\tlearn: 0.2324281\ttotal: 23.5s\tremaining: 13.1s\n",
      "642:\tlearn: 0.2324277\ttotal: 23.5s\tremaining: 13.1s\n",
      "643:\tlearn: 0.2324249\ttotal: 23.6s\tremaining: 13s\n",
      "644:\tlearn: 0.2324223\ttotal: 23.6s\tremaining: 13s\n",
      "645:\tlearn: 0.2324209\ttotal: 23.7s\tremaining: 13s\n",
      "646:\tlearn: 0.2324188\ttotal: 23.7s\tremaining: 12.9s\n",
      "647:\tlearn: 0.2324184\ttotal: 23.7s\tremaining: 12.9s\n",
      "648:\tlearn: 0.2324157\ttotal: 23.8s\tremaining: 12.9s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "649:\tlearn: 0.2324147\ttotal: 23.8s\tremaining: 12.8s\n",
      "650:\tlearn: 0.2324136\ttotal: 23.8s\tremaining: 12.8s\n",
      "651:\tlearn: 0.2324114\ttotal: 23.9s\tremaining: 12.8s\n",
      "652:\tlearn: 0.2324095\ttotal: 23.9s\tremaining: 12.7s\n",
      "653:\tlearn: 0.2324072\ttotal: 24s\tremaining: 12.7s\n",
      "654:\tlearn: 0.2324057\ttotal: 24s\tremaining: 12.6s\n",
      "655:\tlearn: 0.2324042\ttotal: 24s\tremaining: 12.6s\n",
      "656:\tlearn: 0.2324024\ttotal: 24.1s\tremaining: 12.6s\n",
      "657:\tlearn: 0.2324017\ttotal: 24.1s\tremaining: 12.5s\n",
      "658:\tlearn: 0.2324000\ttotal: 24.1s\tremaining: 12.5s\n",
      "659:\tlearn: 0.2323998\ttotal: 24.2s\tremaining: 12.5s\n",
      "660:\tlearn: 0.2323984\ttotal: 24.2s\tremaining: 12.4s\n",
      "661:\tlearn: 0.2323956\ttotal: 24.3s\tremaining: 12.4s\n",
      "662:\tlearn: 0.2323941\ttotal: 24.3s\tremaining: 12.3s\n",
      "663:\tlearn: 0.2323929\ttotal: 24.3s\tremaining: 12.3s\n",
      "664:\tlearn: 0.2323917\ttotal: 24.4s\tremaining: 12.3s\n",
      "665:\tlearn: 0.2323905\ttotal: 24.4s\tremaining: 12.2s\n",
      "666:\tlearn: 0.2323888\ttotal: 24.4s\tremaining: 12.2s\n",
      "667:\tlearn: 0.2323866\ttotal: 24.5s\tremaining: 12.2s\n",
      "668:\tlearn: 0.2323853\ttotal: 24.5s\tremaining: 12.1s\n",
      "669:\tlearn: 0.2323846\ttotal: 24.5s\tremaining: 12.1s\n",
      "670:\tlearn: 0.2323840\ttotal: 24.6s\tremaining: 12.1s\n",
      "671:\tlearn: 0.2323837\ttotal: 24.6s\tremaining: 12s\n",
      "672:\tlearn: 0.2323830\ttotal: 24.7s\tremaining: 12s\n",
      "673:\tlearn: 0.2323825\ttotal: 24.7s\tremaining: 11.9s\n",
      "674:\tlearn: 0.2323817\ttotal: 24.7s\tremaining: 11.9s\n",
      "675:\tlearn: 0.2323812\ttotal: 24.8s\tremaining: 11.9s\n",
      "676:\tlearn: 0.2323803\ttotal: 24.8s\tremaining: 11.8s\n",
      "677:\tlearn: 0.2323785\ttotal: 24.8s\tremaining: 11.8s\n",
      "678:\tlearn: 0.2323771\ttotal: 24.9s\tremaining: 11.8s\n",
      "679:\tlearn: 0.2323768\ttotal: 24.9s\tremaining: 11.7s\n",
      "680:\tlearn: 0.2323744\ttotal: 24.9s\tremaining: 11.7s\n",
      "681:\tlearn: 0.2323728\ttotal: 25s\tremaining: 11.6s\n",
      "682:\tlearn: 0.2323714\ttotal: 25s\tremaining: 11.6s\n",
      "683:\tlearn: 0.2323701\ttotal: 25.1s\tremaining: 11.6s\n",
      "684:\tlearn: 0.2323679\ttotal: 25.1s\tremaining: 11.5s\n",
      "685:\tlearn: 0.2323673\ttotal: 25.1s\tremaining: 11.5s\n",
      "686:\tlearn: 0.2323655\ttotal: 25.2s\tremaining: 11.5s\n",
      "687:\tlearn: 0.2323645\ttotal: 25.2s\tremaining: 11.4s\n",
      "688:\tlearn: 0.2323629\ttotal: 25.2s\tremaining: 11.4s\n",
      "689:\tlearn: 0.2323615\ttotal: 25.3s\tremaining: 11.4s\n",
      "690:\tlearn: 0.2323586\ttotal: 25.3s\tremaining: 11.3s\n",
      "691:\tlearn: 0.2323575\ttotal: 25.3s\tremaining: 11.3s\n",
      "692:\tlearn: 0.2323558\ttotal: 25.4s\tremaining: 11.2s\n",
      "693:\tlearn: 0.2323555\ttotal: 25.4s\tremaining: 11.2s\n",
      "694:\tlearn: 0.2323531\ttotal: 25.5s\tremaining: 11.2s\n",
      "695:\tlearn: 0.2323514\ttotal: 25.5s\tremaining: 11.1s\n",
      "696:\tlearn: 0.2323499\ttotal: 25.6s\tremaining: 11.1s\n",
      "697:\tlearn: 0.2323491\ttotal: 25.6s\tremaining: 11.1s\n",
      "698:\tlearn: 0.2323471\ttotal: 25.6s\tremaining: 11s\n",
      "699:\tlearn: 0.2323447\ttotal: 25.7s\tremaining: 11s\n",
      "700:\tlearn: 0.2323431\ttotal: 25.7s\tremaining: 11s\n",
      "701:\tlearn: 0.2323423\ttotal: 25.7s\tremaining: 10.9s\n",
      "702:\tlearn: 0.2323415\ttotal: 25.8s\tremaining: 10.9s\n",
      "703:\tlearn: 0.2323412\ttotal: 25.8s\tremaining: 10.9s\n",
      "704:\tlearn: 0.2323393\ttotal: 25.9s\tremaining: 10.8s\n",
      "705:\tlearn: 0.2323373\ttotal: 25.9s\tremaining: 10.8s\n",
      "706:\tlearn: 0.2323368\ttotal: 25.9s\tremaining: 10.7s\n",
      "707:\tlearn: 0.2323355\ttotal: 26s\tremaining: 10.7s\n",
      "708:\tlearn: 0.2323342\ttotal: 26s\tremaining: 10.7s\n",
      "709:\tlearn: 0.2323328\ttotal: 26s\tremaining: 10.6s\n",
      "710:\tlearn: 0.2323321\ttotal: 26.1s\tremaining: 10.6s\n",
      "711:\tlearn: 0.2323309\ttotal: 26.1s\tremaining: 10.6s\n",
      "712:\tlearn: 0.2323296\ttotal: 26.1s\tremaining: 10.5s\n",
      "713:\tlearn: 0.2323287\ttotal: 26.2s\tremaining: 10.5s\n",
      "714:\tlearn: 0.2323266\ttotal: 26.2s\tremaining: 10.4s\n",
      "715:\tlearn: 0.2323248\ttotal: 26.2s\tremaining: 10.4s\n",
      "716:\tlearn: 0.2323241\ttotal: 26.3s\tremaining: 10.4s\n",
      "717:\tlearn: 0.2323221\ttotal: 26.3s\tremaining: 10.3s\n",
      "718:\tlearn: 0.2323207\ttotal: 26.4s\tremaining: 10.3s\n",
      "719:\tlearn: 0.2323190\ttotal: 26.4s\tremaining: 10.3s\n",
      "720:\tlearn: 0.2323165\ttotal: 26.4s\tremaining: 10.2s\n",
      "721:\tlearn: 0.2323146\ttotal: 26.5s\tremaining: 10.2s\n",
      "722:\tlearn: 0.2323129\ttotal: 26.5s\tremaining: 10.2s\n",
      "723:\tlearn: 0.2323114\ttotal: 26.6s\tremaining: 10.1s\n",
      "724:\tlearn: 0.2323100\ttotal: 26.6s\tremaining: 10.1s\n",
      "725:\tlearn: 0.2323088\ttotal: 26.6s\tremaining: 10.1s\n",
      "726:\tlearn: 0.2323067\ttotal: 26.7s\tremaining: 10s\n",
      "727:\tlearn: 0.2323043\ttotal: 26.7s\tremaining: 9.98s\n",
      "728:\tlearn: 0.2323036\ttotal: 26.7s\tremaining: 9.94s\n",
      "729:\tlearn: 0.2323024\ttotal: 26.8s\tremaining: 9.9s\n",
      "730:\tlearn: 0.2323003\ttotal: 26.8s\tremaining: 9.87s\n",
      "731:\tlearn: 0.2322979\ttotal: 26.9s\tremaining: 9.83s\n",
      "732:\tlearn: 0.2322961\ttotal: 26.9s\tremaining: 9.79s\n",
      "733:\tlearn: 0.2322946\ttotal: 26.9s\tremaining: 9.76s\n",
      "734:\tlearn: 0.2322939\ttotal: 27s\tremaining: 9.72s\n",
      "735:\tlearn: 0.2322919\ttotal: 27s\tremaining: 9.68s\n",
      "736:\tlearn: 0.2322906\ttotal: 27s\tremaining: 9.65s\n",
      "737:\tlearn: 0.2322886\ttotal: 27.1s\tremaining: 9.61s\n",
      "738:\tlearn: 0.2322875\ttotal: 27.1s\tremaining: 9.57s\n",
      "739:\tlearn: 0.2322864\ttotal: 27.2s\tremaining: 9.54s\n",
      "740:\tlearn: 0.2322846\ttotal: 27.2s\tremaining: 9.5s\n",
      "741:\tlearn: 0.2322836\ttotal: 27.2s\tremaining: 9.46s\n",
      "742:\tlearn: 0.2322822\ttotal: 27.3s\tremaining: 9.43s\n",
      "743:\tlearn: 0.2322807\ttotal: 27.3s\tremaining: 9.39s\n",
      "744:\tlearn: 0.2322794\ttotal: 27.3s\tremaining: 9.36s\n",
      "745:\tlearn: 0.2322777\ttotal: 27.4s\tremaining: 9.32s\n",
      "746:\tlearn: 0.2322765\ttotal: 27.4s\tremaining: 9.29s\n",
      "747:\tlearn: 0.2322752\ttotal: 27.5s\tremaining: 9.25s\n",
      "748:\tlearn: 0.2322744\ttotal: 27.5s\tremaining: 9.21s\n",
      "749:\tlearn: 0.2322722\ttotal: 27.5s\tremaining: 9.18s\n",
      "750:\tlearn: 0.2322711\ttotal: 27.6s\tremaining: 9.14s\n",
      "751:\tlearn: 0.2322699\ttotal: 27.6s\tremaining: 9.1s\n",
      "752:\tlearn: 0.2322690\ttotal: 27.6s\tremaining: 9.07s\n",
      "753:\tlearn: 0.2322675\ttotal: 27.7s\tremaining: 9.03s\n",
      "754:\tlearn: 0.2322662\ttotal: 27.7s\tremaining: 8.99s\n",
      "755:\tlearn: 0.2322649\ttotal: 27.8s\tremaining: 8.96s\n",
      "756:\tlearn: 0.2322636\ttotal: 27.8s\tremaining: 8.92s\n",
      "757:\tlearn: 0.2322619\ttotal: 27.8s\tremaining: 8.88s\n",
      "758:\tlearn: 0.2322600\ttotal: 27.9s\tremaining: 8.85s\n",
      "759:\tlearn: 0.2322576\ttotal: 27.9s\tremaining: 8.81s\n",
      "760:\tlearn: 0.2322573\ttotal: 27.9s\tremaining: 8.78s\n",
      "761:\tlearn: 0.2322555\ttotal: 28s\tremaining: 8.74s\n",
      "762:\tlearn: 0.2322540\ttotal: 28s\tremaining: 8.7s\n",
      "763:\tlearn: 0.2322522\ttotal: 28s\tremaining: 8.66s\n",
      "764:\tlearn: 0.2322505\ttotal: 28.1s\tremaining: 8.63s\n",
      "765:\tlearn: 0.2322486\ttotal: 28.1s\tremaining: 8.59s\n",
      "766:\tlearn: 0.2322477\ttotal: 28.2s\tremaining: 8.55s\n",
      "767:\tlearn: 0.2322464\ttotal: 28.2s\tremaining: 8.52s\n",
      "768:\tlearn: 0.2322454\ttotal: 28.2s\tremaining: 8.48s\n",
      "769:\tlearn: 0.2322428\ttotal: 28.3s\tremaining: 8.45s\n",
      "770:\tlearn: 0.2322405\ttotal: 28.3s\tremaining: 8.41s\n",
      "771:\tlearn: 0.2322388\ttotal: 28.4s\tremaining: 8.38s\n",
      "772:\tlearn: 0.2322372\ttotal: 28.4s\tremaining: 8.34s\n",
      "773:\tlearn: 0.2322355\ttotal: 28.4s\tremaining: 8.3s\n",
      "774:\tlearn: 0.2322349\ttotal: 28.5s\tremaining: 8.27s\n",
      "775:\tlearn: 0.2322331\ttotal: 28.5s\tremaining: 8.23s\n",
      "776:\tlearn: 0.2322315\ttotal: 28.6s\tremaining: 8.2s\n",
      "777:\tlearn: 0.2322293\ttotal: 28.6s\tremaining: 8.16s\n",
      "778:\tlearn: 0.2322283\ttotal: 28.6s\tremaining: 8.13s\n",
      "779:\tlearn: 0.2322271\ttotal: 28.7s\tremaining: 8.09s\n",
      "780:\tlearn: 0.2322267\ttotal: 28.7s\tremaining: 8.05s\n",
      "781:\tlearn: 0.2322255\ttotal: 28.8s\tremaining: 8.02s\n",
      "782:\tlearn: 0.2322247\ttotal: 28.8s\tremaining: 7.98s\n",
      "783:\tlearn: 0.2322230\ttotal: 28.8s\tremaining: 7.94s\n",
      "784:\tlearn: 0.2322215\ttotal: 28.9s\tremaining: 7.91s\n",
      "785:\tlearn: 0.2322200\ttotal: 28.9s\tremaining: 7.87s\n",
      "786:\tlearn: 0.2322176\ttotal: 28.9s\tremaining: 7.83s\n",
      "787:\tlearn: 0.2322168\ttotal: 29s\tremaining: 7.8s\n",
      "788:\tlearn: 0.2322154\ttotal: 29s\tremaining: 7.76s\n",
      "789:\tlearn: 0.2322140\ttotal: 29.1s\tremaining: 7.72s\n",
      "790:\tlearn: 0.2322132\ttotal: 29.1s\tremaining: 7.68s\n",
      "791:\tlearn: 0.2322111\ttotal: 29.1s\tremaining: 7.65s\n",
      "792:\tlearn: 0.2322099\ttotal: 29.2s\tremaining: 7.61s\n",
      "793:\tlearn: 0.2322085\ttotal: 29.2s\tremaining: 7.58s\n",
      "794:\tlearn: 0.2322077\ttotal: 29.2s\tremaining: 7.54s\n",
      "795:\tlearn: 0.2322065\ttotal: 29.3s\tremaining: 7.5s\n",
      "796:\tlearn: 0.2322052\ttotal: 29.3s\tremaining: 7.46s\n",
      "797:\tlearn: 0.2322044\ttotal: 29.3s\tremaining: 7.43s\n",
      "798:\tlearn: 0.2322034\ttotal: 29.4s\tremaining: 7.39s\n",
      "799:\tlearn: 0.2322018\ttotal: 29.4s\tremaining: 7.35s\n",
      "800:\tlearn: 0.2322013\ttotal: 29.5s\tremaining: 7.32s\n",
      "801:\tlearn: 0.2322007\ttotal: 29.5s\tremaining: 7.28s\n",
      "802:\tlearn: 0.2321994\ttotal: 29.5s\tremaining: 7.25s\n",
      "803:\tlearn: 0.2321974\ttotal: 29.6s\tremaining: 7.21s\n",
      "804:\tlearn: 0.2321957\ttotal: 29.6s\tremaining: 7.17s\n",
      "805:\tlearn: 0.2321947\ttotal: 29.6s\tremaining: 7.14s\n",
      "806:\tlearn: 0.2321936\ttotal: 29.7s\tremaining: 7.1s\n",
      "807:\tlearn: 0.2321918\ttotal: 29.7s\tremaining: 7.06s\n",
      "808:\tlearn: 0.2321911\ttotal: 29.8s\tremaining: 7.03s\n",
      "809:\tlearn: 0.2321900\ttotal: 29.8s\tremaining: 6.99s\n",
      "810:\tlearn: 0.2321886\ttotal: 29.8s\tremaining: 6.95s\n",
      "811:\tlearn: 0.2321880\ttotal: 29.9s\tremaining: 6.91s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "812:\tlearn: 0.2321869\ttotal: 29.9s\tremaining: 6.88s\n",
      "813:\tlearn: 0.2321862\ttotal: 29.9s\tremaining: 6.84s\n",
      "814:\tlearn: 0.2321854\ttotal: 30s\tremaining: 6.8s\n",
      "815:\tlearn: 0.2321831\ttotal: 30s\tremaining: 6.76s\n",
      "816:\tlearn: 0.2321816\ttotal: 30s\tremaining: 6.73s\n",
      "817:\tlearn: 0.2321802\ttotal: 30.1s\tremaining: 6.69s\n",
      "818:\tlearn: 0.2321795\ttotal: 30.1s\tremaining: 6.65s\n",
      "819:\tlearn: 0.2321781\ttotal: 30.1s\tremaining: 6.62s\n",
      "820:\tlearn: 0.2321767\ttotal: 30.2s\tremaining: 6.58s\n",
      "821:\tlearn: 0.2321757\ttotal: 30.2s\tremaining: 6.54s\n",
      "822:\tlearn: 0.2321742\ttotal: 30.2s\tremaining: 6.5s\n",
      "823:\tlearn: 0.2321723\ttotal: 30.3s\tremaining: 6.47s\n",
      "824:\tlearn: 0.2321708\ttotal: 30.3s\tremaining: 6.43s\n",
      "825:\tlearn: 0.2321698\ttotal: 30.3s\tremaining: 6.39s\n",
      "826:\tlearn: 0.2321691\ttotal: 30.4s\tremaining: 6.36s\n",
      "827:\tlearn: 0.2321681\ttotal: 30.4s\tremaining: 6.32s\n",
      "828:\tlearn: 0.2321668\ttotal: 30.4s\tremaining: 6.28s\n",
      "829:\tlearn: 0.2321664\ttotal: 30.5s\tremaining: 6.24s\n",
      "830:\tlearn: 0.2321656\ttotal: 30.5s\tremaining: 6.21s\n",
      "831:\tlearn: 0.2321647\ttotal: 30.6s\tremaining: 6.17s\n",
      "832:\tlearn: 0.2321632\ttotal: 30.6s\tremaining: 6.13s\n",
      "833:\tlearn: 0.2321627\ttotal: 30.6s\tremaining: 6.09s\n",
      "834:\tlearn: 0.2321620\ttotal: 30.7s\tremaining: 6.06s\n",
      "835:\tlearn: 0.2321601\ttotal: 30.7s\tremaining: 6.02s\n",
      "836:\tlearn: 0.2321585\ttotal: 30.7s\tremaining: 5.98s\n",
      "837:\tlearn: 0.2321571\ttotal: 30.8s\tremaining: 5.95s\n",
      "838:\tlearn: 0.2321561\ttotal: 30.8s\tremaining: 5.91s\n",
      "839:\tlearn: 0.2321546\ttotal: 30.8s\tremaining: 5.87s\n",
      "840:\tlearn: 0.2321534\ttotal: 30.9s\tremaining: 5.84s\n",
      "841:\tlearn: 0.2321525\ttotal: 30.9s\tremaining: 5.8s\n",
      "842:\tlearn: 0.2321520\ttotal: 31s\tremaining: 5.76s\n",
      "843:\tlearn: 0.2321501\ttotal: 31s\tremaining: 5.73s\n",
      "844:\tlearn: 0.2321495\ttotal: 31s\tremaining: 5.69s\n",
      "845:\tlearn: 0.2321485\ttotal: 31.1s\tremaining: 5.66s\n",
      "846:\tlearn: 0.2321478\ttotal: 31.1s\tremaining: 5.62s\n",
      "847:\tlearn: 0.2321466\ttotal: 31.1s\tremaining: 5.58s\n",
      "848:\tlearn: 0.2321454\ttotal: 31.2s\tremaining: 5.55s\n",
      "849:\tlearn: 0.2321439\ttotal: 31.2s\tremaining: 5.51s\n",
      "850:\tlearn: 0.2321423\ttotal: 31.3s\tremaining: 5.47s\n",
      "851:\tlearn: 0.2321410\ttotal: 31.3s\tremaining: 5.44s\n",
      "852:\tlearn: 0.2321395\ttotal: 31.3s\tremaining: 5.4s\n",
      "853:\tlearn: 0.2321392\ttotal: 31.4s\tremaining: 5.36s\n",
      "854:\tlearn: 0.2321379\ttotal: 31.4s\tremaining: 5.33s\n",
      "855:\tlearn: 0.2321372\ttotal: 31.4s\tremaining: 5.29s\n",
      "856:\tlearn: 0.2321352\ttotal: 31.5s\tremaining: 5.25s\n",
      "857:\tlearn: 0.2321342\ttotal: 31.5s\tremaining: 5.21s\n",
      "858:\tlearn: 0.2321332\ttotal: 31.5s\tremaining: 5.18s\n",
      "859:\tlearn: 0.2321324\ttotal: 31.6s\tremaining: 5.14s\n",
      "860:\tlearn: 0.2321311\ttotal: 31.6s\tremaining: 5.1s\n",
      "861:\tlearn: 0.2321300\ttotal: 31.6s\tremaining: 5.07s\n",
      "862:\tlearn: 0.2321288\ttotal: 31.7s\tremaining: 5.03s\n",
      "863:\tlearn: 0.2321277\ttotal: 31.7s\tremaining: 4.99s\n",
      "864:\tlearn: 0.2321271\ttotal: 31.8s\tremaining: 4.96s\n",
      "865:\tlearn: 0.2321261\ttotal: 31.8s\tremaining: 4.92s\n",
      "866:\tlearn: 0.2321247\ttotal: 31.8s\tremaining: 4.88s\n",
      "867:\tlearn: 0.2321231\ttotal: 31.9s\tremaining: 4.84s\n",
      "868:\tlearn: 0.2321223\ttotal: 31.9s\tremaining: 4.81s\n",
      "869:\tlearn: 0.2321212\ttotal: 31.9s\tremaining: 4.77s\n",
      "870:\tlearn: 0.2321198\ttotal: 32s\tremaining: 4.73s\n",
      "871:\tlearn: 0.2321183\ttotal: 32s\tremaining: 4.7s\n",
      "872:\tlearn: 0.2321173\ttotal: 32s\tremaining: 4.66s\n",
      "873:\tlearn: 0.2321161\ttotal: 32.1s\tremaining: 4.62s\n",
      "874:\tlearn: 0.2321150\ttotal: 32.1s\tremaining: 4.58s\n",
      "875:\tlearn: 0.2321146\ttotal: 32.1s\tremaining: 4.55s\n",
      "876:\tlearn: 0.2321138\ttotal: 32.2s\tremaining: 4.51s\n",
      "877:\tlearn: 0.2321136\ttotal: 32.2s\tremaining: 4.47s\n",
      "878:\tlearn: 0.2321129\ttotal: 32.2s\tremaining: 4.44s\n",
      "879:\tlearn: 0.2321114\ttotal: 32.3s\tremaining: 4.4s\n",
      "880:\tlearn: 0.2321112\ttotal: 32.3s\tremaining: 4.36s\n",
      "881:\tlearn: 0.2321101\ttotal: 32.3s\tremaining: 4.33s\n",
      "882:\tlearn: 0.2321087\ttotal: 32.4s\tremaining: 4.29s\n",
      "883:\tlearn: 0.2321071\ttotal: 32.4s\tremaining: 4.25s\n",
      "884:\tlearn: 0.2321061\ttotal: 32.4s\tremaining: 4.21s\n",
      "885:\tlearn: 0.2321045\ttotal: 32.5s\tremaining: 4.18s\n",
      "886:\tlearn: 0.2321032\ttotal: 32.5s\tremaining: 4.14s\n",
      "887:\tlearn: 0.2321021\ttotal: 32.5s\tremaining: 4.1s\n",
      "888:\tlearn: 0.2321002\ttotal: 32.6s\tremaining: 4.07s\n",
      "889:\tlearn: 0.2321002\ttotal: 32.6s\tremaining: 4.03s\n",
      "890:\tlearn: 0.2320993\ttotal: 32.6s\tremaining: 3.99s\n",
      "891:\tlearn: 0.2320987\ttotal: 32.7s\tremaining: 3.96s\n",
      "892:\tlearn: 0.2320981\ttotal: 32.7s\tremaining: 3.92s\n",
      "893:\tlearn: 0.2320968\ttotal: 32.7s\tremaining: 3.88s\n",
      "894:\tlearn: 0.2320953\ttotal: 32.8s\tremaining: 3.85s\n",
      "895:\tlearn: 0.2320946\ttotal: 32.8s\tremaining: 3.81s\n",
      "896:\tlearn: 0.2320935\ttotal: 32.9s\tremaining: 3.77s\n",
      "897:\tlearn: 0.2320920\ttotal: 32.9s\tremaining: 3.73s\n",
      "898:\tlearn: 0.2320909\ttotal: 32.9s\tremaining: 3.7s\n",
      "899:\tlearn: 0.2320895\ttotal: 33s\tremaining: 3.66s\n",
      "900:\tlearn: 0.2320878\ttotal: 33s\tremaining: 3.62s\n",
      "901:\tlearn: 0.2320864\ttotal: 33s\tremaining: 3.59s\n",
      "902:\tlearn: 0.2320854\ttotal: 33.1s\tremaining: 3.55s\n",
      "903:\tlearn: 0.2320839\ttotal: 33.1s\tremaining: 3.52s\n",
      "904:\tlearn: 0.2320818\ttotal: 33.1s\tremaining: 3.48s\n",
      "905:\tlearn: 0.2320810\ttotal: 33.2s\tremaining: 3.44s\n",
      "906:\tlearn: 0.2320781\ttotal: 33.2s\tremaining: 3.4s\n",
      "907:\tlearn: 0.2320772\ttotal: 33.2s\tremaining: 3.37s\n",
      "908:\tlearn: 0.2320762\ttotal: 33.3s\tremaining: 3.33s\n",
      "909:\tlearn: 0.2320748\ttotal: 33.3s\tremaining: 3.29s\n",
      "910:\tlearn: 0.2320728\ttotal: 33.4s\tremaining: 3.26s\n",
      "911:\tlearn: 0.2320713\ttotal: 33.4s\tremaining: 3.22s\n",
      "912:\tlearn: 0.2320704\ttotal: 33.4s\tremaining: 3.18s\n",
      "913:\tlearn: 0.2320689\ttotal: 33.5s\tremaining: 3.15s\n",
      "914:\tlearn: 0.2320678\ttotal: 33.5s\tremaining: 3.11s\n",
      "915:\tlearn: 0.2320664\ttotal: 33.5s\tremaining: 3.07s\n",
      "916:\tlearn: 0.2320646\ttotal: 33.6s\tremaining: 3.04s\n",
      "917:\tlearn: 0.2320633\ttotal: 33.6s\tremaining: 3s\n",
      "918:\tlearn: 0.2320626\ttotal: 33.6s\tremaining: 2.96s\n",
      "919:\tlearn: 0.2320618\ttotal: 33.7s\tremaining: 2.93s\n",
      "920:\tlearn: 0.2320607\ttotal: 33.7s\tremaining: 2.89s\n",
      "921:\tlearn: 0.2320596\ttotal: 33.7s\tremaining: 2.85s\n",
      "922:\tlearn: 0.2320588\ttotal: 33.8s\tremaining: 2.82s\n",
      "923:\tlearn: 0.2320577\ttotal: 33.8s\tremaining: 2.78s\n",
      "924:\tlearn: 0.2320559\ttotal: 33.9s\tremaining: 2.75s\n",
      "925:\tlearn: 0.2320554\ttotal: 33.9s\tremaining: 2.71s\n",
      "926:\tlearn: 0.2320544\ttotal: 33.9s\tremaining: 2.67s\n",
      "927:\tlearn: 0.2320533\ttotal: 34s\tremaining: 2.63s\n",
      "928:\tlearn: 0.2320523\ttotal: 34s\tremaining: 2.6s\n",
      "929:\tlearn: 0.2320509\ttotal: 34s\tremaining: 2.56s\n",
      "930:\tlearn: 0.2320502\ttotal: 34.1s\tremaining: 2.52s\n",
      "931:\tlearn: 0.2320485\ttotal: 34.1s\tremaining: 2.49s\n",
      "932:\tlearn: 0.2320467\ttotal: 34.2s\tremaining: 2.45s\n",
      "933:\tlearn: 0.2320461\ttotal: 34.2s\tremaining: 2.42s\n",
      "934:\tlearn: 0.2320452\ttotal: 34.2s\tremaining: 2.38s\n",
      "935:\tlearn: 0.2320430\ttotal: 34.3s\tremaining: 2.34s\n",
      "936:\tlearn: 0.2320423\ttotal: 34.3s\tremaining: 2.31s\n",
      "937:\tlearn: 0.2320415\ttotal: 34.3s\tremaining: 2.27s\n",
      "938:\tlearn: 0.2320408\ttotal: 34.4s\tremaining: 2.23s\n",
      "939:\tlearn: 0.2320399\ttotal: 34.4s\tremaining: 2.2s\n",
      "940:\tlearn: 0.2320398\ttotal: 34.4s\tremaining: 2.16s\n",
      "941:\tlearn: 0.2320391\ttotal: 34.5s\tremaining: 2.12s\n",
      "942:\tlearn: 0.2320381\ttotal: 34.5s\tremaining: 2.09s\n",
      "943:\tlearn: 0.2320372\ttotal: 34.6s\tremaining: 2.05s\n",
      "944:\tlearn: 0.2320360\ttotal: 34.6s\tremaining: 2.01s\n",
      "945:\tlearn: 0.2320353\ttotal: 34.6s\tremaining: 1.98s\n",
      "946:\tlearn: 0.2320349\ttotal: 34.7s\tremaining: 1.94s\n",
      "947:\tlearn: 0.2320335\ttotal: 34.7s\tremaining: 1.9s\n",
      "948:\tlearn: 0.2320334\ttotal: 34.7s\tremaining: 1.87s\n",
      "949:\tlearn: 0.2320324\ttotal: 34.8s\tremaining: 1.83s\n",
      "950:\tlearn: 0.2320321\ttotal: 34.8s\tremaining: 1.79s\n",
      "951:\tlearn: 0.2320311\ttotal: 34.8s\tremaining: 1.76s\n",
      "952:\tlearn: 0.2320307\ttotal: 34.9s\tremaining: 1.72s\n",
      "953:\tlearn: 0.2320298\ttotal: 34.9s\tremaining: 1.68s\n",
      "954:\tlearn: 0.2320290\ttotal: 34.9s\tremaining: 1.65s\n",
      "955:\tlearn: 0.2320272\ttotal: 35s\tremaining: 1.61s\n",
      "956:\tlearn: 0.2320261\ttotal: 35s\tremaining: 1.57s\n",
      "957:\tlearn: 0.2320247\ttotal: 35.1s\tremaining: 1.54s\n",
      "958:\tlearn: 0.2320233\ttotal: 35.1s\tremaining: 1.5s\n",
      "959:\tlearn: 0.2320231\ttotal: 35.1s\tremaining: 1.46s\n",
      "960:\tlearn: 0.2320229\ttotal: 35.1s\tremaining: 1.43s\n",
      "961:\tlearn: 0.2320213\ttotal: 35.2s\tremaining: 1.39s\n",
      "962:\tlearn: 0.2320202\ttotal: 35.2s\tremaining: 1.35s\n",
      "963:\tlearn: 0.2320195\ttotal: 35.3s\tremaining: 1.32s\n",
      "964:\tlearn: 0.2320176\ttotal: 35.3s\tremaining: 1.28s\n",
      "965:\tlearn: 0.2320161\ttotal: 35.3s\tremaining: 1.24s\n",
      "966:\tlearn: 0.2320153\ttotal: 35.4s\tremaining: 1.21s\n",
      "967:\tlearn: 0.2320146\ttotal: 35.4s\tremaining: 1.17s\n",
      "968:\tlearn: 0.2320140\ttotal: 35.4s\tremaining: 1.13s\n",
      "969:\tlearn: 0.2320127\ttotal: 35.5s\tremaining: 1.1s\n",
      "970:\tlearn: 0.2320115\ttotal: 35.5s\tremaining: 1.06s\n",
      "971:\tlearn: 0.2320083\ttotal: 35.5s\tremaining: 1.02s\n",
      "972:\tlearn: 0.2320080\ttotal: 35.6s\tremaining: 987ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "973:\tlearn: 0.2320068\ttotal: 35.6s\tremaining: 951ms\n",
      "974:\tlearn: 0.2320050\ttotal: 35.7s\tremaining: 914ms\n",
      "975:\tlearn: 0.2320036\ttotal: 35.7s\tremaining: 878ms\n",
      "976:\tlearn: 0.2320015\ttotal: 35.7s\tremaining: 841ms\n",
      "977:\tlearn: 0.2319988\ttotal: 35.8s\tremaining: 805ms\n",
      "978:\tlearn: 0.2319965\ttotal: 35.8s\tremaining: 768ms\n",
      "979:\tlearn: 0.2319962\ttotal: 35.8s\tremaining: 731ms\n",
      "980:\tlearn: 0.2319943\ttotal: 35.9s\tremaining: 695ms\n",
      "981:\tlearn: 0.2319933\ttotal: 35.9s\tremaining: 658ms\n",
      "982:\tlearn: 0.2319918\ttotal: 35.9s\tremaining: 621ms\n",
      "983:\tlearn: 0.2319915\ttotal: 36s\tremaining: 585ms\n",
      "984:\tlearn: 0.2319908\ttotal: 36s\tremaining: 548ms\n",
      "985:\tlearn: 0.2319897\ttotal: 36.1s\tremaining: 512ms\n",
      "986:\tlearn: 0.2319895\ttotal: 36.1s\tremaining: 475ms\n",
      "987:\tlearn: 0.2319894\ttotal: 36.1s\tremaining: 439ms\n",
      "988:\tlearn: 0.2319876\ttotal: 36.1s\tremaining: 402ms\n",
      "989:\tlearn: 0.2319864\ttotal: 36.2s\tremaining: 366ms\n",
      "990:\tlearn: 0.2319854\ttotal: 36.2s\tremaining: 329ms\n",
      "991:\tlearn: 0.2319847\ttotal: 36.3s\tremaining: 293ms\n",
      "992:\tlearn: 0.2319835\ttotal: 36.3s\tremaining: 256ms\n",
      "993:\tlearn: 0.2319822\ttotal: 36.4s\tremaining: 219ms\n",
      "994:\tlearn: 0.2319808\ttotal: 36.4s\tremaining: 183ms\n",
      "995:\tlearn: 0.2319799\ttotal: 36.4s\tremaining: 146ms\n",
      "996:\tlearn: 0.2319782\ttotal: 36.5s\tremaining: 110ms\n",
      "997:\tlearn: 0.2319774\ttotal: 36.5s\tremaining: 73.2ms\n",
      "998:\tlearn: 0.2319762\ttotal: 36.6s\tremaining: 36.6ms\n",
      "999:\tlearn: 0.2319760\ttotal: 36.6s\tremaining: 0us\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score=nan,\n",
       "             estimator=<catboost.core.CatBoostRegressor object at 0x0000020897FE7EC8>,\n",
       "             iid='deprecated', n_jobs=10,\n",
       "             param_grid={'learning_rate': array([0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09])},\n",
       "             pre_dispatch='2*n_jobs', refit=True, return_train_score=False,\n",
       "             scoring='neg_mean_squared_error', verbose=2)"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'learning_rate': 0.01}"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.05416058943637192"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## XGBoost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "metadata": {},
   "outputs": [],
   "source": [
    "boost=XGBRegressor()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBRegressor(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "             colsample_bynode=1, colsample_bytree=1, gamma=0, gpu_id=-1,\n",
       "             importance_type='gain', interaction_constraints='',\n",
       "             learning_rate=0.300000012, max_delta_step=0, max_depth=6,\n",
       "             min_child_weight=1, missing=nan, monotone_constraints='()',\n",
       "             n_estimators=100, n_jobs=0, num_parallel_tree=1,\n",
       "             objective='reg:squarederror', random_state=0, reg_alpha=0,\n",
       "             reg_lambda=1, scale_pos_weight=1, subsample=1, tree_method='exact',\n",
       "             validate_parameters=1, verbosity=None)"
      ]
     },
     "execution_count": 170,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "boost.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "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>Course_ID</th>\n",
       "      <th>Course_Domain</th>\n",
       "      <th>Course_Type</th>\n",
       "      <th>Short_Promotion</th>\n",
       "      <th>Public_Holiday</th>\n",
       "      <th>Long_Promotion</th>\n",
       "      <th>Competition_Metric</th>\n",
       "      <th>Day_of_2_Weeks</th>\n",
       "      <th>Week_No</th>\n",
       "      <th>Year_No</th>\n",
       "      <th>Target_Course_ID</th>\n",
       "      <th>Group_Traffic</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.007</td>\n",
       "      <td>1</td>\n",
       "      <td>23</td>\n",
       "      <td>3</td>\n",
       "      <td>107.0</td>\n",
       "      <td>14175.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.007</td>\n",
       "      <td>2</td>\n",
       "      <td>23</td>\n",
       "      <td>3</td>\n",
       "      <td>123.0</td>\n",
       "      <td>17850.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.007</td>\n",
       "      <td>3</td>\n",
       "      <td>23</td>\n",
       "      <td>3</td>\n",
       "      <td>96.5</td>\n",
       "      <td>12988.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.007</td>\n",
       "      <td>4</td>\n",
       "      <td>23</td>\n",
       "      <td>3</td>\n",
       "      <td>93.0</td>\n",
       "      <td>12484.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.007</td>\n",
       "      <td>5</td>\n",
       "      <td>23</td>\n",
       "      <td>3</td>\n",
       "      <td>46.0</td>\n",
       "      <td>7245.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35995</th>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.070</td>\n",
       "      <td>14</td>\n",
       "      <td>30</td>\n",
       "      <td>3</td>\n",
       "      <td>149.0</td>\n",
       "      <td>10993.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35996</th>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.070</td>\n",
       "      <td>1</td>\n",
       "      <td>31</td>\n",
       "      <td>3</td>\n",
       "      <td>129.0</td>\n",
       "      <td>9796.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35997</th>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.070</td>\n",
       "      <td>2</td>\n",
       "      <td>31</td>\n",
       "      <td>3</td>\n",
       "      <td>125.0</td>\n",
       "      <td>9576.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35998</th>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.070</td>\n",
       "      <td>3</td>\n",
       "      <td>31</td>\n",
       "      <td>3</td>\n",
       "      <td>124.0</td>\n",
       "      <td>9124.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35999</th>\n",
       "      <td>600</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.070</td>\n",
       "      <td>4</td>\n",
       "      <td>31</td>\n",
       "      <td>3</td>\n",
       "      <td>129.5</td>\n",
       "      <td>9481.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>36000 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Course_ID  Course_Domain  Course_Type  Short_Promotion  Public_Holiday  \\\n",
       "0              1              1            0                1               0   \n",
       "1              1              1            0                1               0   \n",
       "2              1              1            0                1               0   \n",
       "3              1              1            0                1               0   \n",
       "4              1              1            0                0               0   \n",
       "...          ...            ...          ...              ...             ...   \n",
       "35995        600              3            2                1               0   \n",
       "35996        600              3            2                1               0   \n",
       "35997        600              3            2                1               0   \n",
       "35998        600              3            2                1               0   \n",
       "35999        600              3            2                1               0   \n",
       "\n",
       "       Long_Promotion  Competition_Metric  Day_of_2_Weeks  Week_No  Year_No  \\\n",
       "0                   1               0.007               1       23        3   \n",
       "1                   1               0.007               2       23        3   \n",
       "2                   1               0.007               3       23        3   \n",
       "3                   1               0.007               4       23        3   \n",
       "4                   1               0.007               5       23        3   \n",
       "...               ...                 ...             ...      ...      ...   \n",
       "35995               1               0.070              14       30        3   \n",
       "35996               1               0.070               1       31        3   \n",
       "35997               1               0.070               2       31        3   \n",
       "35998               1               0.070               3       31        3   \n",
       "35999               1               0.070               4       31        3   \n",
       "\n",
       "       Target_Course_ID  Group_Traffic  \n",
       "0                 107.0        14175.0  \n",
       "1                 123.0        17850.0  \n",
       "2                  96.5        12988.5  \n",
       "3                  93.0        12484.5  \n",
       "4                  46.0         7245.0  \n",
       "...                 ...            ...  \n",
       "35995             149.0        10993.5  \n",
       "35996             129.0         9796.5  \n",
       "35997             125.0         9576.0  \n",
       "35998             124.0         9124.5  \n",
       "35999             129.5         9481.5  \n",
       "\n",
       "[36000 rows x 12 columns]"
      ]
     },
     "execution_count": 171,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_pred=boost.predict(X_train)\n",
    "y_test_pred=boost.predict(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "85.66977124247153"
      ]
     },
     "execution_count": 173,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1000*np.sqrt(mean_squared_error(y_train,y_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([4.758759 , 4.9058685, 4.680691 , ..., 4.903514 , 4.936084 ,\n",
       "       4.9767456], dtype=float32)"
      ]
     },
     "execution_count": 174,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {},
   "outputs": [],
   "source": [
    "export_submission(y_test_pred,\"XGBoost\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 258,
   "metadata": {},
   "outputs": [],
   "source": [
    "parameters={'base_score':[0.2,0.3,0.4,0.5,0.6,0.7,0.8],\n",
    "           'max_depth':[2,3,4,5,6,7,8,9,10,11,12],\n",
    "           'subsample':[0.5,0.6,0.7,0.8,0.9,1.0],\n",
    "            'learning_rate':[0.1,0.2,0.3,0.4,0.5,0.6]\n",
    "            }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 259,
   "metadata": {},
   "outputs": [],
   "source": [
    "search=RandomizedSearchCV(XGBRegressor(),parameters,scoring='neg_mean_squared_error',cv=5,verbose=100,n_iter=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 260,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 10 candidates, totalling 50 fits\n",
      "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[CV] subsample=0.7, max_depth=9, learning_rate=0.3, base_score=0.3 ...\n",
      "[CV]  subsample=0.7, max_depth=9, learning_rate=0.3, base_score=0.3, score=-0.011, total=  26.4s\n",
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:   26.3s remaining:    0.0s\n",
      "[CV] subsample=0.7, max_depth=9, learning_rate=0.3, base_score=0.3 ...\n",
      "[CV]  subsample=0.7, max_depth=9, learning_rate=0.3, base_score=0.3, score=-0.008, total=  26.0s\n",
      "[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:   52.3s remaining:    0.0s\n",
      "[CV] subsample=0.7, max_depth=9, learning_rate=0.3, base_score=0.3 ...\n",
      "[CV]  subsample=0.7, max_depth=9, learning_rate=0.3, base_score=0.3, score=-0.009, total=  26.0s\n",
      "[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:  1.3min remaining:    0.0s\n",
      "[CV] subsample=0.7, max_depth=9, learning_rate=0.3, base_score=0.3 ...\n",
      "[CV]  subsample=0.7, max_depth=9, learning_rate=0.3, base_score=0.3, score=-0.008, total=  26.1s\n",
      "[Parallel(n_jobs=1)]: Done   4 out of   4 | elapsed:  1.7min remaining:    0.0s\n",
      "[CV] subsample=0.7, max_depth=9, learning_rate=0.3, base_score=0.3 ...\n",
      "[CV]  subsample=0.7, max_depth=9, learning_rate=0.3, base_score=0.3, score=-0.009, total=  26.5s\n",
      "[Parallel(n_jobs=1)]: Done   5 out of   5 | elapsed:  2.2min remaining:    0.0s\n",
      "[CV] subsample=0.5, max_depth=4, learning_rate=0.1, base_score=0.3 ...\n",
      "[CV]  subsample=0.5, max_depth=4, learning_rate=0.1, base_score=0.3, score=-0.014, total=  13.4s\n",
      "[Parallel(n_jobs=1)]: Done   6 out of   6 | elapsed:  2.4min remaining:    0.0s\n",
      "[CV] subsample=0.5, max_depth=4, learning_rate=0.1, base_score=0.3 ...\n",
      "[CV]  subsample=0.5, max_depth=4, learning_rate=0.1, base_score=0.3, score=-0.013, total=  13.6s\n",
      "[Parallel(n_jobs=1)]: Done   7 out of   7 | elapsed:  2.6min remaining:    0.0s\n",
      "[CV] subsample=0.5, max_depth=4, learning_rate=0.1, base_score=0.3 ...\n",
      "[CV]  subsample=0.5, max_depth=4, learning_rate=0.1, base_score=0.3, score=-0.014, total=  13.4s\n",
      "[Parallel(n_jobs=1)]: Done   8 out of   8 | elapsed:  2.9min remaining:    0.0s\n",
      "[CV] subsample=0.5, max_depth=4, learning_rate=0.1, base_score=0.3 ...\n",
      "[CV]  subsample=0.5, max_depth=4, learning_rate=0.1, base_score=0.3, score=-0.014, total=  11.3s\n",
      "[Parallel(n_jobs=1)]: Done   9 out of   9 | elapsed:  3.0min remaining:    0.0s\n",
      "[CV] subsample=0.5, max_depth=4, learning_rate=0.1, base_score=0.3 ...\n",
      "[CV]  subsample=0.5, max_depth=4, learning_rate=0.1, base_score=0.3, score=-0.013, total=  13.1s\n",
      "[Parallel(n_jobs=1)]: Done  10 out of  10 | elapsed:  3.3min remaining:    0.0s\n",
      "[CV] subsample=0.9, max_depth=12, learning_rate=0.3, base_score=0.7 ..\n",
      "[CV]  subsample=0.9, max_depth=12, learning_rate=0.3, base_score=0.7, score=-0.010, total=  34.0s\n",
      "[Parallel(n_jobs=1)]: Done  11 out of  11 | elapsed:  3.8min remaining:    0.0s\n",
      "[CV] subsample=0.9, max_depth=12, learning_rate=0.3, base_score=0.7 ..\n",
      "[CV]  subsample=0.9, max_depth=12, learning_rate=0.3, base_score=0.7, score=-0.008, total=  30.0s\n",
      "[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:  4.3min remaining:    0.0s\n",
      "[CV] subsample=0.9, max_depth=12, learning_rate=0.3, base_score=0.7 ..\n",
      "[CV]  subsample=0.9, max_depth=12, learning_rate=0.3, base_score=0.7, score=-0.009, total=  33.5s\n",
      "[Parallel(n_jobs=1)]: Done  13 out of  13 | elapsed:  4.9min remaining:    0.0s\n",
      "[CV] subsample=0.9, max_depth=12, learning_rate=0.3, base_score=0.7 ..\n",
      "[CV]  subsample=0.9, max_depth=12, learning_rate=0.3, base_score=0.7, score=-0.009, total=  33.6s\n",
      "[Parallel(n_jobs=1)]: Done  14 out of  14 | elapsed:  5.4min remaining:    0.0s\n",
      "[CV] subsample=0.9, max_depth=12, learning_rate=0.3, base_score=0.7 ..\n",
      "[CV]  subsample=0.9, max_depth=12, learning_rate=0.3, base_score=0.7, score=-0.010, total=  29.4s\n",
      "[Parallel(n_jobs=1)]: Done  15 out of  15 | elapsed:  5.9min remaining:    0.0s\n",
      "[CV] subsample=0.8, max_depth=6, learning_rate=0.4, base_score=0.3 ...\n",
      "[CV]  subsample=0.8, max_depth=6, learning_rate=0.4, base_score=0.3, score=-0.011, total=  16.7s\n",
      "[Parallel(n_jobs=1)]: Done  16 out of  16 | elapsed:  6.2min remaining:    0.0s\n",
      "[CV] subsample=0.8, max_depth=6, learning_rate=0.4, base_score=0.3 ...\n",
      "[CV]  subsample=0.8, max_depth=6, learning_rate=0.4, base_score=0.3, score=-0.008, total=  16.9s\n",
      "[Parallel(n_jobs=1)]: Done  17 out of  17 | elapsed:  6.5min remaining:    0.0s\n",
      "[CV] subsample=0.8, max_depth=6, learning_rate=0.4, base_score=0.3 ...\n",
      "[CV]  subsample=0.8, max_depth=6, learning_rate=0.4, base_score=0.3, score=-0.009, total=  17.7s\n",
      "[Parallel(n_jobs=1)]: Done  18 out of  18 | elapsed:  6.8min remaining:    0.0s\n",
      "[CV] subsample=0.8, max_depth=6, learning_rate=0.4, base_score=0.3 ...\n",
      "[CV]  subsample=0.8, max_depth=6, learning_rate=0.4, base_score=0.3, score=-0.009, total=  16.3s\n",
      "[Parallel(n_jobs=1)]: Done  19 out of  19 | elapsed:  7.1min remaining:    0.0s\n",
      "[CV] subsample=0.8, max_depth=6, learning_rate=0.4, base_score=0.3 ...\n",
      "[CV]  subsample=0.8, max_depth=6, learning_rate=0.4, base_score=0.3, score=-0.009, total=  16.9s\n",
      "[Parallel(n_jobs=1)]: Done  20 out of  20 | elapsed:  7.3min remaining:    0.0s\n",
      "[CV] subsample=1.0, max_depth=2, learning_rate=0.2, base_score=0.4 ...\n",
      "[CV]  subsample=1.0, max_depth=2, learning_rate=0.2, base_score=0.4, score=-0.020, total=   4.9s\n",
      "[Parallel(n_jobs=1)]: Done  21 out of  21 | elapsed:  7.4min remaining:    0.0s\n",
      "[CV] subsample=1.0, max_depth=2, learning_rate=0.2, base_score=0.4 ...\n",
      "[CV]  subsample=1.0, max_depth=2, learning_rate=0.2, base_score=0.4, score=-0.018, total=   5.4s\n",
      "[Parallel(n_jobs=1)]: Done  22 out of  22 | elapsed:  7.5min remaining:    0.0s\n",
      "[CV] subsample=1.0, max_depth=2, learning_rate=0.2, base_score=0.4 ...\n",
      "[CV]  subsample=1.0, max_depth=2, learning_rate=0.2, base_score=0.4, score=-0.020, total=   5.4s\n",
      "[Parallel(n_jobs=1)]: Done  23 out of  23 | elapsed:  7.6min remaining:    0.0s\n",
      "[CV] subsample=1.0, max_depth=2, learning_rate=0.2, base_score=0.4 ...\n",
      "[CV]  subsample=1.0, max_depth=2, learning_rate=0.2, base_score=0.4, score=-0.020, total=   4.8s\n",
      "[Parallel(n_jobs=1)]: Done  24 out of  24 | elapsed:  7.7min remaining:    0.0s\n",
      "[CV] subsample=1.0, max_depth=2, learning_rate=0.2, base_score=0.4 ...\n",
      "[CV]  subsample=1.0, max_depth=2, learning_rate=0.2, base_score=0.4, score=-0.019, total=   6.2s\n",
      "[Parallel(n_jobs=1)]: Done  25 out of  25 | elapsed:  7.8min remaining:    0.0s\n",
      "[CV] subsample=0.8, max_depth=12, learning_rate=0.4, base_score=0.3 ..\n",
      "[CV]  subsample=0.8, max_depth=12, learning_rate=0.4, base_score=0.3, score=-0.012, total=  34.3s\n",
      "[Parallel(n_jobs=1)]: Done  26 out of  26 | elapsed:  8.4min remaining:    0.0s\n",
      "[CV] subsample=0.8, max_depth=12, learning_rate=0.4, base_score=0.3 ..\n",
      "[CV]  subsample=0.8, max_depth=12, learning_rate=0.4, base_score=0.3, score=-0.009, total=  34.8s\n",
      "[Parallel(n_jobs=1)]: Done  27 out of  27 | elapsed:  8.9min remaining:    0.0s\n",
      "[CV] subsample=0.8, max_depth=12, learning_rate=0.4, base_score=0.3 ..\n",
      "[CV]  subsample=0.8, max_depth=12, learning_rate=0.4, base_score=0.3, score=-0.010, total=  35.6s\n",
      "[Parallel(n_jobs=1)]: Done  28 out of  28 | elapsed:  9.5min remaining:    0.0s\n",
      "[CV] subsample=0.8, max_depth=12, learning_rate=0.4, base_score=0.3 ..\n",
      "[CV]  subsample=0.8, max_depth=12, learning_rate=0.4, base_score=0.3, score=-0.010, total=  35.5s\n",
      "[Parallel(n_jobs=1)]: Done  29 out of  29 | elapsed: 10.1min remaining:    0.0s\n",
      "[CV] subsample=0.8, max_depth=12, learning_rate=0.4, base_score=0.3 ..\n",
      "[CV]  subsample=0.8, max_depth=12, learning_rate=0.4, base_score=0.3, score=-0.012, total=  35.3s\n",
      "[Parallel(n_jobs=1)]: Done  30 out of  30 | elapsed: 10.7min remaining:    0.0s\n",
      "[CV] subsample=1.0, max_depth=11, learning_rate=0.4, base_score=0.2 ..\n",
      "[CV]  subsample=1.0, max_depth=11, learning_rate=0.4, base_score=0.2, score=-0.012, total=  27.2s\n",
      "[Parallel(n_jobs=1)]: Done  31 out of  31 | elapsed: 11.2min remaining:    0.0s\n",
      "[CV] subsample=1.0, max_depth=11, learning_rate=0.4, base_score=0.2 ..\n",
      "[CV]  subsample=1.0, max_depth=11, learning_rate=0.4, base_score=0.2, score=-0.008, total=  28.1s\n",
      "[Parallel(n_jobs=1)]: Done  32 out of  32 | elapsed: 11.6min remaining:    0.0s\n",
      "[CV] subsample=1.0, max_depth=11, learning_rate=0.4, base_score=0.2 ..\n",
      "[CV]  subsample=1.0, max_depth=11, learning_rate=0.4, base_score=0.2, score=-0.009, total=  27.6s\n",
      "[Parallel(n_jobs=1)]: Done  33 out of  33 | elapsed: 12.1min remaining:    0.0s\n",
      "[CV] subsample=1.0, max_depth=11, learning_rate=0.4, base_score=0.2 ..\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  subsample=1.0, max_depth=11, learning_rate=0.4, base_score=0.2, score=-0.009, total=  28.5s\n",
      "[Parallel(n_jobs=1)]: Done  34 out of  34 | elapsed: 12.6min remaining:    0.0s\n",
      "[CV] subsample=1.0, max_depth=11, learning_rate=0.4, base_score=0.2 ..\n",
      "[CV]  subsample=1.0, max_depth=11, learning_rate=0.4, base_score=0.2, score=-0.010, total=  27.8s\n",
      "[Parallel(n_jobs=1)]: Done  35 out of  35 | elapsed: 13.0min remaining:    0.0s\n",
      "[CV] subsample=0.9, max_depth=12, learning_rate=0.5, base_score=0.3 ..\n",
      "[CV]  subsample=0.9, max_depth=12, learning_rate=0.5, base_score=0.3, score=-0.012, total=  35.7s\n",
      "[Parallel(n_jobs=1)]: Done  36 out of  36 | elapsed: 13.6min remaining:    0.0s\n",
      "[CV] subsample=0.9, max_depth=12, learning_rate=0.5, base_score=0.3 ..\n",
      "[CV]  subsample=0.9, max_depth=12, learning_rate=0.5, base_score=0.3, score=-0.009, total=  33.6s\n",
      "[Parallel(n_jobs=1)]: Done  37 out of  37 | elapsed: 14.2min remaining:    0.0s\n",
      "[CV] subsample=0.9, max_depth=12, learning_rate=0.5, base_score=0.3 ..\n",
      "[CV]  subsample=0.9, max_depth=12, learning_rate=0.5, base_score=0.3, score=-0.010, total=  33.5s\n",
      "[Parallel(n_jobs=1)]: Done  38 out of  38 | elapsed: 14.8min remaining:    0.0s\n",
      "[CV] subsample=0.9, max_depth=12, learning_rate=0.5, base_score=0.3 ..\n",
      "[CV]  subsample=0.9, max_depth=12, learning_rate=0.5, base_score=0.3, score=-0.010, total=  33.0s\n",
      "[Parallel(n_jobs=1)]: Done  39 out of  39 | elapsed: 15.3min remaining:    0.0s\n",
      "[CV] subsample=0.9, max_depth=12, learning_rate=0.5, base_score=0.3 ..\n",
      "[CV]  subsample=0.9, max_depth=12, learning_rate=0.5, base_score=0.3, score=-0.011, total=  33.4s\n",
      "[Parallel(n_jobs=1)]: Done  40 out of  40 | elapsed: 15.9min remaining:    0.0s\n",
      "[CV] subsample=0.7, max_depth=5, learning_rate=0.2, base_score=0.4 ...\n",
      "[CV]  subsample=0.7, max_depth=5, learning_rate=0.2, base_score=0.4, score=-0.010, total=  14.8s\n",
      "[Parallel(n_jobs=1)]: Done  41 out of  41 | elapsed: 16.1min remaining:    0.0s\n",
      "[CV] subsample=0.7, max_depth=5, learning_rate=0.2, base_score=0.4 ...\n",
      "[CV]  subsample=0.7, max_depth=5, learning_rate=0.2, base_score=0.4, score=-0.009, total=  14.9s\n",
      "[Parallel(n_jobs=1)]: Done  42 out of  42 | elapsed: 16.4min remaining:    0.0s\n",
      "[CV] subsample=0.7, max_depth=5, learning_rate=0.2, base_score=0.4 ...\n",
      "[CV]  subsample=0.7, max_depth=5, learning_rate=0.2, base_score=0.4, score=-0.010, total=  14.8s\n",
      "[Parallel(n_jobs=1)]: Done  43 out of  43 | elapsed: 16.6min remaining:    0.0s\n",
      "[CV] subsample=0.7, max_depth=5, learning_rate=0.2, base_score=0.4 ...\n",
      "[CV]  subsample=0.7, max_depth=5, learning_rate=0.2, base_score=0.4, score=-0.009, total=  14.6s\n",
      "[Parallel(n_jobs=1)]: Done  44 out of  44 | elapsed: 16.8min remaining:    0.0s\n",
      "[CV] subsample=0.7, max_depth=5, learning_rate=0.2, base_score=0.4 ...\n",
      "[CV]  subsample=0.7, max_depth=5, learning_rate=0.2, base_score=0.4, score=-0.010, total=  14.7s\n",
      "[Parallel(n_jobs=1)]: Done  45 out of  45 | elapsed: 17.1min remaining:    0.0s\n",
      "[CV] subsample=0.7, max_depth=6, learning_rate=0.3, base_score=0.6 ...\n",
      "[CV]  subsample=0.7, max_depth=6, learning_rate=0.3, base_score=0.6, score=-0.010, total=  17.0s\n",
      "[Parallel(n_jobs=1)]: Done  46 out of  46 | elapsed: 17.4min remaining:    0.0s\n",
      "[CV] subsample=0.7, max_depth=6, learning_rate=0.3, base_score=0.6 ...\n",
      "[CV]  subsample=0.7, max_depth=6, learning_rate=0.3, base_score=0.6, score=-0.008, total=  17.1s\n",
      "[Parallel(n_jobs=1)]: Done  47 out of  47 | elapsed: 17.7min remaining:    0.0s\n",
      "[CV] subsample=0.7, max_depth=6, learning_rate=0.3, base_score=0.6 ...\n",
      "[CV]  subsample=0.7, max_depth=6, learning_rate=0.3, base_score=0.6, score=-0.010, total=  16.7s\n",
      "[Parallel(n_jobs=1)]: Done  48 out of  48 | elapsed: 17.9min remaining:    0.0s\n",
      "[CV] subsample=0.7, max_depth=6, learning_rate=0.3, base_score=0.6 ...\n",
      "[CV]  subsample=0.7, max_depth=6, learning_rate=0.3, base_score=0.6, score=-0.009, total=  16.9s\n",
      "[Parallel(n_jobs=1)]: Done  49 out of  49 | elapsed: 18.2min remaining:    0.0s\n",
      "[CV] subsample=0.7, max_depth=6, learning_rate=0.3, base_score=0.6 ...\n",
      "[CV]  subsample=0.7, max_depth=6, learning_rate=0.3, base_score=0.6, score=-0.009, total=  16.9s\n",
      "[Parallel(n_jobs=1)]: Done  50 out of  50 | elapsed: 18.5min remaining:    0.0s\n",
      "[Parallel(n_jobs=1)]: Done  50 out of  50 | elapsed: 18.5min finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=5, error_score=nan,\n",
       "                   estimator=XGBRegressor(base_score=None, booster=None,\n",
       "                                          colsample_bylevel=None,\n",
       "                                          colsample_bynode=None,\n",
       "                                          colsample_bytree=None, gamma=None,\n",
       "                                          gpu_id=None, importance_type='gain',\n",
       "                                          interaction_constraints=None,\n",
       "                                          learning_rate=None,\n",
       "                                          max_delta_step=None, max_depth=None,\n",
       "                                          min_child_weight=None, missing=nan,\n",
       "                                          monotone_constraints=None,\n",
       "                                          n_...\n",
       "                                          verbosity=None),\n",
       "                   iid='deprecated', n_iter=10, n_jobs=None,\n",
       "                   param_distributions={'base_score': [0.2, 0.3, 0.4, 0.5, 0.6,\n",
       "                                                       0.7, 0.8],\n",
       "                                        'learning_rate': [0.1, 0.2, 0.3, 0.4,\n",
       "                                                          0.5, 0.6],\n",
       "                                        'max_depth': [2, 3, 4, 5, 6, 7, 8, 9,\n",
       "                                                      10, 11, 12],\n",
       "                                        'subsample': [0.5, 0.6, 0.7, 0.8, 0.9,\n",
       "                                                      1.0]},\n",
       "                   pre_dispatch='2*n_jobs', random_state=None, refit=True,\n",
       "                   return_train_score=False, scoring='neg_mean_squared_error',\n",
       "                   verbose=100)"
      ]
     },
     "execution_count": 260,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 261,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'subsample': 0.7, 'max_depth': 9, 'learning_rate': 0.3, 'base_score': 0.3}"
      ]
     },
     "execution_count": 261,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 262,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.008966443416140548"
      ]
     },
     "execution_count": 262,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 263,
   "metadata": {},
   "outputs": [],
   "source": [
    "boost=XGBRegressor(subsample=0.7,max_depth=9,learning_rate=0.3,base_score=0.3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 264,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBRegressor(base_score=0.3, booster='gbtree', colsample_bylevel=1,\n",
       "             colsample_bynode=1, colsample_bytree=1, gamma=0, gpu_id=-1,\n",
       "             importance_type='gain', interaction_constraints='',\n",
       "             learning_rate=0.3, max_delta_step=0, max_depth=9,\n",
       "             min_child_weight=1, missing=nan, monotone_constraints='()',\n",
       "             n_estimators=100, n_jobs=0, num_parallel_tree=1,\n",
       "             objective='reg:squarederror', random_state=0, reg_alpha=0,\n",
       "             reg_lambda=1, scale_pos_weight=1, subsample=0.7,\n",
       "             tree_method='exact', validate_parameters=1, verbosity=None)"
      ]
     },
     "execution_count": 264,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "boost.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 265,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_pred=boost.predict(X_train)\n",
    "y_test_pred=boost.predict(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 266,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "68.95229770074857"
      ]
     },
     "execution_count": 266,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1000*np.sqrt(mean_squared_error(y_train,y_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 267,
   "metadata": {},
   "outputs": [],
   "source": [
    "export_submission(y_test_pred,\"XGBoost More Improved\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ensembling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 491,
   "metadata": {},
   "outputs": [],
   "source": [
    "data1=pd.read_csv(\"Compiled New/Catboost.csv\")\n",
    "data2=pd.read_csv(\"Compiled New/LGBM.csv\")\n",
    "data3=pd.read_csv(\"Compiled New/Random Forest.csv\")\n",
    "data4=pd.read_csv(\"Compiled New/XGB.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 492,
   "metadata": {},
   "outputs": [],
   "source": [
    "aa=(data1[\"Sales\"]+data2[\"Sales\"]+data3[\"Sales\"]+data4[\"Sales\"])/4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 493,
   "metadata": {},
   "outputs": [],
   "source": [
    "def export_submission(predictions,filename=\"Sample.csv\"):\n",
    "    \n",
    "    ss=sample_sub.copy()\n",
    "    ss[\"Sales\"]=predictions\n",
    "    ss.to_csv(f\"{filename}.csv\",index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 494,
   "metadata": {},
   "outputs": [],
   "source": [
    "export_submission(aa,\"Simple_Mean_All\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 495,
   "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>ID</th>\n",
       "      <th>Sales</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>883</td>\n",
       "      <td>106.474328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>884</td>\n",
       "      <td>122.680358</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>885</td>\n",
       "      <td>96.400178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>886</td>\n",
       "      <td>93.214625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>887</td>\n",
       "      <td>47.277525</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35995</th>\n",
       "      <td>548083</td>\n",
       "      <td>147.101124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35996</th>\n",
       "      <td>548084</td>\n",
       "      <td>126.964361</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35997</th>\n",
       "      <td>548085</td>\n",
       "      <td>122.921234</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35998</th>\n",
       "      <td>548086</td>\n",
       "      <td>121.567769</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35999</th>\n",
       "      <td>548087</td>\n",
       "      <td>126.865331</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>36000 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           ID       Sales\n",
       "0         883  106.474328\n",
       "1         884  122.680358\n",
       "2         885   96.400178\n",
       "3         886   93.214625\n",
       "4         887   47.277525\n",
       "...       ...         ...\n",
       "35995  548083  147.101124\n",
       "35996  548084  126.964361\n",
       "35997  548085  122.921234\n",
       "35998  548086  121.567769\n",
       "35999  548087  126.865331\n",
       "\n",
       "[36000 rows x 2 columns]"
      ]
     },
     "execution_count": 495,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.read_csv(\"Simple_Mean_All.csv\")"
   ]
  }
 ],
 "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": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}