XGBoost-Model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import xgboost as xgb\n",
    "\n",
    "import pickle\n",
    "from geopy.geocoders import Nominatim\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "pd.set_option('display.max_columns', None)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1458644, 11)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Read train file\n",
    "train_df = pd.read_csv('Taxi-Trip-Duration-Data/train.csv')\n",
    "train_df.shape"
   ]
  },
  {
   "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>vendor_id</th>\n",
       "      <th>pickup_datetime</th>\n",
       "      <th>dropoff_datetime</th>\n",
       "      <th>passenger_count</th>\n",
       "      <th>pickup_longitude</th>\n",
       "      <th>pickup_latitude</th>\n",
       "      <th>dropoff_longitude</th>\n",
       "      <th>dropoff_latitude</th>\n",
       "      <th>store_and_fwd_flag</th>\n",
       "      <th>trip_duration</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>id2875421</td>\n",
       "      <td>2</td>\n",
       "      <td>2016-03-14 17:24:55</td>\n",
       "      <td>2016-03-14 17:32:30</td>\n",
       "      <td>1</td>\n",
       "      <td>-73.982155</td>\n",
       "      <td>40.767937</td>\n",
       "      <td>-73.964630</td>\n",
       "      <td>40.765602</td>\n",
       "      <td>N</td>\n",
       "      <td>455</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>id2377394</td>\n",
       "      <td>1</td>\n",
       "      <td>2016-06-12 00:43:35</td>\n",
       "      <td>2016-06-12 00:54:38</td>\n",
       "      <td>1</td>\n",
       "      <td>-73.980415</td>\n",
       "      <td>40.738564</td>\n",
       "      <td>-73.999481</td>\n",
       "      <td>40.731152</td>\n",
       "      <td>N</td>\n",
       "      <td>663</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>id3858529</td>\n",
       "      <td>2</td>\n",
       "      <td>2016-01-19 11:35:24</td>\n",
       "      <td>2016-01-19 12:10:48</td>\n",
       "      <td>1</td>\n",
       "      <td>-73.979027</td>\n",
       "      <td>40.763939</td>\n",
       "      <td>-74.005333</td>\n",
       "      <td>40.710087</td>\n",
       "      <td>N</td>\n",
       "      <td>2124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>id3504673</td>\n",
       "      <td>2</td>\n",
       "      <td>2016-04-06 19:32:31</td>\n",
       "      <td>2016-04-06 19:39:40</td>\n",
       "      <td>1</td>\n",
       "      <td>-74.010040</td>\n",
       "      <td>40.719971</td>\n",
       "      <td>-74.012268</td>\n",
       "      <td>40.706718</td>\n",
       "      <td>N</td>\n",
       "      <td>429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>id2181028</td>\n",
       "      <td>2</td>\n",
       "      <td>2016-03-26 13:30:55</td>\n",
       "      <td>2016-03-26 13:38:10</td>\n",
       "      <td>1</td>\n",
       "      <td>-73.973053</td>\n",
       "      <td>40.793209</td>\n",
       "      <td>-73.972923</td>\n",
       "      <td>40.782520</td>\n",
       "      <td>N</td>\n",
       "      <td>435</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          id  vendor_id      pickup_datetime     dropoff_datetime  \\\n",
       "0  id2875421          2  2016-03-14 17:24:55  2016-03-14 17:32:30   \n",
       "1  id2377394          1  2016-06-12 00:43:35  2016-06-12 00:54:38   \n",
       "2  id3858529          2  2016-01-19 11:35:24  2016-01-19 12:10:48   \n",
       "3  id3504673          2  2016-04-06 19:32:31  2016-04-06 19:39:40   \n",
       "4  id2181028          2  2016-03-26 13:30:55  2016-03-26 13:38:10   \n",
       "\n",
       "   passenger_count  pickup_longitude  pickup_latitude  dropoff_longitude  \\\n",
       "0                1        -73.982155        40.767937         -73.964630   \n",
       "1                1        -73.980415        40.738564         -73.999481   \n",
       "2                1        -73.979027        40.763939         -74.005333   \n",
       "3                1        -74.010040        40.719971         -74.012268   \n",
       "4                1        -73.973053        40.793209         -73.972923   \n",
       "\n",
       "   dropoff_latitude store_and_fwd_flag  trip_duration  \n",
       "0         40.765602                  N            455  \n",
       "1         40.731152                  N            663  \n",
       "2         40.710087                  N           2124  \n",
       "3         40.706718                  N            429  \n",
       "4         40.782520                  N            435  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Pre-Processing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    1450599\n",
       "1       8045\n",
       "Name: store_and_fwd_flag, dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Convert character variables to numeric \n",
    "f = lambda x: 0 if x == 'N' else 1\n",
    "train_df[\"store_and_fwd_flag\"] = train_df[\"store_and_fwd_flag\"].apply(lambda x: f(x))\n",
    "train_df[\"store_and_fwd_flag\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Engineering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# First, convert datetime strings into datetime\n",
    "train_df[\"dropoff_datetime\"] = pd.to_datetime(train_df[\"dropoff_datetime\"], format='%Y-%m-%d %H:%M:%S')\n",
    "train_df[\"pickup_datetime\"] = pd.to_datetime(train_df[\"pickup_datetime\"], format='%Y-%m-%d %H:%M:%S')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Now construct other variables, like month, date, etc.\n",
    "train_df[\"pickup_month\"] = train_df[\"pickup_datetime\"].dt.month\n",
    "train_df[\"pickup_day\"] = train_df[\"pickup_datetime\"].dt.day\n",
    "train_df[\"pickup_weekday\"] = train_df[\"pickup_datetime\"].dt.weekday\n",
    "train_df[\"pickup_hour\"] = train_df[\"pickup_datetime\"].dt.hour\n",
    "train_df[\"pickup_minute\"] = train_df[\"pickup_datetime\"].dt.minute"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Get latitude and longitude differences \n",
    "train_df[\"latitude_difference\"] = train_df[\"dropoff_latitude\"] - train_df[\"pickup_latitude\"]\n",
    "train_df[\"longitude_difference\"] = train_df[\"dropoff_longitude\"] - train_df[\"pickup_longitude\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Convert duration to minutes for easier interpretation\n",
    "train_df[\"trip_duration\"] = train_df[\"trip_duration\"].apply(lambda x: round(x/60))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Convert trip distance from longitude and latitude differences to Manhattan distance.\n",
    "train_df[\"trip_distance\"] = 0.621371 * 6371 * \\\n",
    "(abs(2 * np.arctan2(np.sqrt(np.square(np.sin((abs(train_df[\"latitude_difference\"]) * np.pi / 180) / 2))), \n",
    "np.sqrt(1-(np.square(np.sin((abs(train_df[\"latitude_difference\"]) * np.pi / 180) / 2)))))) + \\\n",
    "abs(2 * np.arctan2(np.sqrt(np.square(np.sin((abs(train_df[\"longitude_difference\"]) * np.pi / 180) / 2))), \n",
    "np.sqrt(1-(np.square(np.sin((abs(train_df[\"longitude_difference\"]) * np.pi / 180) / 2)))))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>vendor_id</th>\n",
       "      <th>pickup_datetime</th>\n",
       "      <th>dropoff_datetime</th>\n",
       "      <th>passenger_count</th>\n",
       "      <th>pickup_longitude</th>\n",
       "      <th>pickup_latitude</th>\n",
       "      <th>dropoff_longitude</th>\n",
       "      <th>dropoff_latitude</th>\n",
       "      <th>store_and_fwd_flag</th>\n",
       "      <th>trip_duration</th>\n",
       "      <th>pickup_month</th>\n",
       "      <th>pickup_day</th>\n",
       "      <th>pickup_weekday</th>\n",
       "      <th>pickup_hour</th>\n",
       "      <th>pickup_minute</th>\n",
       "      <th>latitude_difference</th>\n",
       "      <th>longitude_difference</th>\n",
       "      <th>trip_distance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>id2875421</td>\n",
       "      <td>2</td>\n",
       "      <td>2016-03-14 17:24:55</td>\n",
       "      <td>2016-03-14 17:32:30</td>\n",
       "      <td>1</td>\n",
       "      <td>-73.982155</td>\n",
       "      <td>40.767937</td>\n",
       "      <td>-73.964630</td>\n",
       "      <td>40.765602</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "      <td>14</td>\n",
       "      <td>0</td>\n",
       "      <td>17</td>\n",
       "      <td>24</td>\n",
       "      <td>-0.002335</td>\n",
       "      <td>0.017525</td>\n",
       "      <td>1.372146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>id2377394</td>\n",
       "      <td>1</td>\n",
       "      <td>2016-06-12 00:43:35</td>\n",
       "      <td>2016-06-12 00:54:38</td>\n",
       "      <td>1</td>\n",
       "      <td>-73.980415</td>\n",
       "      <td>40.738564</td>\n",
       "      <td>-73.999481</td>\n",
       "      <td>40.731152</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>6</td>\n",
       "      <td>12</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>43</td>\n",
       "      <td>-0.007412</td>\n",
       "      <td>-0.019066</td>\n",
       "      <td>1.829440</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>id3858529</td>\n",
       "      <td>2</td>\n",
       "      <td>2016-01-19 11:35:24</td>\n",
       "      <td>2016-01-19 12:10:48</td>\n",
       "      <td>1</td>\n",
       "      <td>-73.979027</td>\n",
       "      <td>40.763939</td>\n",
       "      <td>-74.005333</td>\n",
       "      <td>40.710087</td>\n",
       "      <td>0</td>\n",
       "      <td>35</td>\n",
       "      <td>1</td>\n",
       "      <td>19</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>35</td>\n",
       "      <td>-0.053852</td>\n",
       "      <td>-0.026306</td>\n",
       "      <td>5.538397</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>id3504673</td>\n",
       "      <td>2</td>\n",
       "      <td>2016-04-06 19:32:31</td>\n",
       "      <td>2016-04-06 19:39:40</td>\n",
       "      <td>1</td>\n",
       "      <td>-74.010040</td>\n",
       "      <td>40.719971</td>\n",
       "      <td>-74.012268</td>\n",
       "      <td>40.706718</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "      <td>19</td>\n",
       "      <td>32</td>\n",
       "      <td>-0.013252</td>\n",
       "      <td>-0.002228</td>\n",
       "      <td>1.069567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>id2181028</td>\n",
       "      <td>2</td>\n",
       "      <td>2016-03-26 13:30:55</td>\n",
       "      <td>2016-03-26 13:38:10</td>\n",
       "      <td>1</td>\n",
       "      <td>-73.973053</td>\n",
       "      <td>40.793209</td>\n",
       "      <td>-73.972923</td>\n",
       "      <td>40.782520</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>26</td>\n",
       "      <td>5</td>\n",
       "      <td>13</td>\n",
       "      <td>30</td>\n",
       "      <td>-0.010689</td>\n",
       "      <td>0.000130</td>\n",
       "      <td>0.747485</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          id  vendor_id     pickup_datetime    dropoff_datetime  \\\n",
       "0  id2875421          2 2016-03-14 17:24:55 2016-03-14 17:32:30   \n",
       "1  id2377394          1 2016-06-12 00:43:35 2016-06-12 00:54:38   \n",
       "2  id3858529          2 2016-01-19 11:35:24 2016-01-19 12:10:48   \n",
       "3  id3504673          2 2016-04-06 19:32:31 2016-04-06 19:39:40   \n",
       "4  id2181028          2 2016-03-26 13:30:55 2016-03-26 13:38:10   \n",
       "\n",
       "   passenger_count  pickup_longitude  pickup_latitude  dropoff_longitude  \\\n",
       "0                1        -73.982155        40.767937         -73.964630   \n",
       "1                1        -73.980415        40.738564         -73.999481   \n",
       "2                1        -73.979027        40.763939         -74.005333   \n",
       "3                1        -74.010040        40.719971         -74.012268   \n",
       "4                1        -73.973053        40.793209         -73.972923   \n",
       "\n",
       "   dropoff_latitude  store_and_fwd_flag  trip_duration  pickup_month  \\\n",
       "0         40.765602                   0              8             3   \n",
       "1         40.731152                   0             11             6   \n",
       "2         40.710087                   0             35             1   \n",
       "3         40.706718                   0              7             4   \n",
       "4         40.782520                   0              7             3   \n",
       "\n",
       "   pickup_day  pickup_weekday  pickup_hour  pickup_minute  \\\n",
       "0          14               0           17             24   \n",
       "1          12               6            0             43   \n",
       "2          19               1           11             35   \n",
       "3           6               2           19             32   \n",
       "4          26               5           13             30   \n",
       "\n",
       "   latitude_difference  longitude_difference  trip_distance  \n",
       "0            -0.002335              0.017525       1.372146  \n",
       "1            -0.007412             -0.019066       1.829440  \n",
       "2            -0.053852             -0.026306       5.538397  \n",
       "3            -0.013252             -0.002228       1.069567  \n",
       "4            -0.010689              0.000130       0.747485  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predictive Modeling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X = train_df.drop([\"trip_duration\", \"id\", \"vendor_id\", \"pickup_datetime\", \"dropoff_datetime\"], axis=1)\n",
    "y = train_df[\"trip_duration\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Split the train data into training, test, and valdiation sets\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=2019)\n",
    "X_train, X_val, y_train, y_val = train_test_split(X_train, y_train, test_size=0.2, random_state=2019)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Define evaluation metric\n",
    "def rmsle(y_true, y_pred):\n",
    "    assert len(y_true) == len(y_pred)\n",
    "    return np.square(np.log(y_pred + 1) - np.log(y_true + 1)).mean() ** 0.5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# XGBoost parameters \n",
    "params = {\n",
    "    'booster':            'gbtree',\n",
    "    'objective':          'reg:linear',\n",
    "    'learning_rate':      0.05,\n",
    "    'max_depth':          14,\n",
    "    'subsample':          0.9,\n",
    "    'colsample_bytree':   0.7,\n",
    "    'colsample_bylevel':  0.7,\n",
    "    'silent':             1,\n",
    "    'feval':              'rmsle'\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Define train and validation sets\n",
    "dtrain = xgb.DMatrix(X_train, np.log(y_train+1))\n",
    "dval = xgb.DMatrix(X_val, np.log(y_val+1))\n",
    "\n",
    "# this is for tracking the error\n",
    "watchlist = [(dval, 'eval'), (dtrain, 'train')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0]\teval-rmse:2.00944\ttrain-rmse:2.00596\n",
      "[1]\teval-rmse:1.91338\ttrain-rmse:1.90994\n",
      "[2]\teval-rmse:1.82191\ttrain-rmse:1.81841\n",
      "[3]\teval-rmse:1.73527\ttrain-rmse:1.73171\n",
      "[4]\teval-rmse:1.65351\ttrain-rmse:1.64992\n",
      "[5]\teval-rmse:1.57753\ttrain-rmse:1.5739\n",
      "[6]\teval-rmse:1.50361\ttrain-rmse:1.49983\n",
      "[7]\teval-rmse:1.4336\ttrain-rmse:1.42978\n",
      "[8]\teval-rmse:1.36829\ttrain-rmse:1.3644\n",
      "[9]\teval-rmse:1.30584\ttrain-rmse:1.30174\n",
      "[10]\teval-rmse:1.24711\ttrain-rmse:1.24293\n",
      "[11]\teval-rmse:1.19076\ttrain-rmse:1.18649\n",
      "[12]\teval-rmse:1.13898\ttrain-rmse:1.13457\n",
      "[13]\teval-rmse:1.08854\ttrain-rmse:1.0839\n",
      "[14]\teval-rmse:1.04129\ttrain-rmse:1.03637\n",
      "[15]\teval-rmse:0.996349\ttrain-rmse:0.991182\n",
      "[16]\teval-rmse:0.954038\ttrain-rmse:0.948487\n",
      "[17]\teval-rmse:0.914116\ttrain-rmse:0.908204\n",
      "[18]\teval-rmse:0.877326\ttrain-rmse:0.871156\n",
      "[19]\teval-rmse:0.841526\ttrain-rmse:0.83486\n",
      "[20]\teval-rmse:0.80813\ttrain-rmse:0.801134\n",
      "[21]\teval-rmse:0.776705\ttrain-rmse:0.76925\n",
      "[22]\teval-rmse:0.747169\ttrain-rmse:0.739316\n",
      "[23]\teval-rmse:0.719725\ttrain-rmse:0.711521\n",
      "[24]\teval-rmse:0.693754\ttrain-rmse:0.68486\n",
      "[25]\teval-rmse:0.669098\ttrain-rmse:0.659575\n",
      "[26]\teval-rmse:0.645812\ttrain-rmse:0.635751\n",
      "[27]\teval-rmse:0.623988\ttrain-rmse:0.613068\n",
      "[28]\teval-rmse:0.603994\ttrain-rmse:0.592437\n",
      "[29]\teval-rmse:0.584863\ttrain-rmse:0.572716\n",
      "[30]\teval-rmse:0.566934\ttrain-rmse:0.553934\n",
      "[31]\teval-rmse:0.551134\ttrain-rmse:0.537536\n",
      "[32]\teval-rmse:0.535852\ttrain-rmse:0.521617\n",
      "[33]\teval-rmse:0.521662\ttrain-rmse:0.506821\n",
      "[34]\teval-rmse:0.508102\ttrain-rmse:0.492401\n",
      "[35]\teval-rmse:0.495427\ttrain-rmse:0.478822\n",
      "[36]\teval-rmse:0.483727\ttrain-rmse:0.466216\n",
      "[37]\teval-rmse:0.473408\ttrain-rmse:0.45524\n",
      "[38]\teval-rmse:0.463739\ttrain-rmse:0.444816\n",
      "[39]\teval-rmse:0.454676\ttrain-rmse:0.435079\n",
      "[40]\teval-rmse:0.446144\ttrain-rmse:0.425475\n",
      "[41]\teval-rmse:0.438591\ttrain-rmse:0.417202\n",
      "[42]\teval-rmse:0.431566\ttrain-rmse:0.409505\n",
      "[43]\teval-rmse:0.424631\ttrain-rmse:0.401392\n",
      "[44]\teval-rmse:0.418386\ttrain-rmse:0.394425\n",
      "[45]\teval-rmse:0.412649\ttrain-rmse:0.387967\n",
      "[46]\teval-rmse:0.407246\ttrain-rmse:0.381519\n",
      "[47]\teval-rmse:0.402152\ttrain-rmse:0.375365\n",
      "[48]\teval-rmse:0.397553\ttrain-rmse:0.370004\n",
      "[49]\teval-rmse:0.393293\ttrain-rmse:0.365041\n",
      "[50]\teval-rmse:0.38944\ttrain-rmse:0.360284\n",
      "[51]\teval-rmse:0.385731\ttrain-rmse:0.355501\n",
      "[52]\teval-rmse:0.382478\ttrain-rmse:0.351125\n",
      "[53]\teval-rmse:0.379373\ttrain-rmse:0.346936\n",
      "[54]\teval-rmse:0.37656\ttrain-rmse:0.343052\n",
      "[55]\teval-rmse:0.374109\ttrain-rmse:0.339881\n",
      "[56]\teval-rmse:0.371781\ttrain-rmse:0.336887\n",
      "[57]\teval-rmse:0.369582\ttrain-rmse:0.33392\n",
      "[58]\teval-rmse:0.367652\ttrain-rmse:0.33126\n",
      "[59]\teval-rmse:0.365762\ttrain-rmse:0.328711\n",
      "[60]\teval-rmse:0.364039\ttrain-rmse:0.326429\n",
      "[61]\teval-rmse:0.362355\ttrain-rmse:0.323841\n",
      "[62]\teval-rmse:0.360944\ttrain-rmse:0.321625\n",
      "[63]\teval-rmse:0.359604\ttrain-rmse:0.319565\n",
      "[64]\teval-rmse:0.35838\ttrain-rmse:0.317694\n",
      "[65]\teval-rmse:0.357362\ttrain-rmse:0.316012\n",
      "[66]\teval-rmse:0.356269\ttrain-rmse:0.314146\n",
      "[67]\teval-rmse:0.355326\ttrain-rmse:0.312687\n",
      "[68]\teval-rmse:0.354399\ttrain-rmse:0.311155\n",
      "[69]\teval-rmse:0.353602\ttrain-rmse:0.309655\n",
      "[70]\teval-rmse:0.352767\ttrain-rmse:0.308043\n",
      "[71]\teval-rmse:0.352035\ttrain-rmse:0.306821\n",
      "[72]\teval-rmse:0.351358\ttrain-rmse:0.305275\n",
      "[73]\teval-rmse:0.350805\ttrain-rmse:0.304103\n",
      "[74]\teval-rmse:0.350137\ttrain-rmse:0.302793\n",
      "[75]\teval-rmse:0.34951\ttrain-rmse:0.301382\n",
      "[76]\teval-rmse:0.348903\ttrain-rmse:0.299974\n",
      "[77]\teval-rmse:0.348488\ttrain-rmse:0.299024\n",
      "[78]\teval-rmse:0.348039\ttrain-rmse:0.297959\n",
      "[79]\teval-rmse:0.347543\ttrain-rmse:0.296577\n",
      "[80]\teval-rmse:0.347122\ttrain-rmse:0.295668\n",
      "[81]\teval-rmse:0.34678\ttrain-rmse:0.294678\n",
      "[82]\teval-rmse:0.346344\ttrain-rmse:0.293551\n",
      "[83]\teval-rmse:0.345915\ttrain-rmse:0.29224\n",
      "[84]\teval-rmse:0.345545\ttrain-rmse:0.291042\n",
      "[85]\teval-rmse:0.345259\ttrain-rmse:0.290196\n",
      "[86]\teval-rmse:0.344985\ttrain-rmse:0.289262\n",
      "[87]\teval-rmse:0.344709\ttrain-rmse:0.288514\n",
      "[88]\teval-rmse:0.344362\ttrain-rmse:0.287465\n",
      "[89]\teval-rmse:0.344079\ttrain-rmse:0.28645\n",
      "[90]\teval-rmse:0.343776\ttrain-rmse:0.285531\n",
      "[91]\teval-rmse:0.343576\ttrain-rmse:0.284781\n",
      "[92]\teval-rmse:0.343407\ttrain-rmse:0.283851\n",
      "[93]\teval-rmse:0.343205\ttrain-rmse:0.283003\n",
      "[94]\teval-rmse:0.34299\ttrain-rmse:0.281995\n",
      "[95]\teval-rmse:0.342864\ttrain-rmse:0.281296\n",
      "[96]\teval-rmse:0.342567\ttrain-rmse:0.280361\n",
      "[97]\teval-rmse:0.342459\ttrain-rmse:0.279664\n",
      "[98]\teval-rmse:0.342244\ttrain-rmse:0.278751\n",
      "[99]\teval-rmse:0.342057\ttrain-rmse:0.278037\n",
      "[100]\teval-rmse:0.341804\ttrain-rmse:0.277048\n",
      "[101]\teval-rmse:0.341637\ttrain-rmse:0.276199\n",
      "[102]\teval-rmse:0.341521\ttrain-rmse:0.275397\n",
      "[103]\teval-rmse:0.341307\ttrain-rmse:0.27447\n",
      "[104]\teval-rmse:0.3411\ttrain-rmse:0.273711\n",
      "[105]\teval-rmse:0.340946\ttrain-rmse:0.273105\n",
      "[106]\teval-rmse:0.340713\ttrain-rmse:0.27234\n",
      "[107]\teval-rmse:0.340551\ttrain-rmse:0.271694\n",
      "[108]\teval-rmse:0.340331\ttrain-rmse:0.27084\n",
      "[109]\teval-rmse:0.340259\ttrain-rmse:0.270225\n",
      "[110]\teval-rmse:0.340045\ttrain-rmse:0.269475\n",
      "[111]\teval-rmse:0.339869\ttrain-rmse:0.268627\n",
      "[112]\teval-rmse:0.339807\ttrain-rmse:0.267983\n",
      "[113]\teval-rmse:0.339685\ttrain-rmse:0.267266\n",
      "[114]\teval-rmse:0.339538\ttrain-rmse:0.266563\n",
      "[115]\teval-rmse:0.339439\ttrain-rmse:0.265999\n",
      "[116]\teval-rmse:0.339275\ttrain-rmse:0.265256\n",
      "[117]\teval-rmse:0.339208\ttrain-rmse:0.264663\n",
      "[118]\teval-rmse:0.339066\ttrain-rmse:0.264058\n",
      "[119]\teval-rmse:0.338926\ttrain-rmse:0.263379\n",
      "[120]\teval-rmse:0.338815\ttrain-rmse:0.262688\n",
      "[121]\teval-rmse:0.338747\ttrain-rmse:0.262037\n",
      "[122]\teval-rmse:0.338657\ttrain-rmse:0.261437\n",
      "[123]\teval-rmse:0.338469\ttrain-rmse:0.260758\n",
      "[124]\teval-rmse:0.33839\ttrain-rmse:0.260204\n",
      "[125]\teval-rmse:0.338248\ttrain-rmse:0.259592\n",
      "[126]\teval-rmse:0.338159\ttrain-rmse:0.259209\n",
      "[127]\teval-rmse:0.338111\ttrain-rmse:0.258783\n",
      "[128]\teval-rmse:0.338034\ttrain-rmse:0.25819\n",
      "[129]\teval-rmse:0.337915\ttrain-rmse:0.257581\n",
      "[130]\teval-rmse:0.337839\ttrain-rmse:0.257079\n",
      "[131]\teval-rmse:0.337719\ttrain-rmse:0.256633\n",
      "[132]\teval-rmse:0.337575\ttrain-rmse:0.256063\n",
      "[133]\teval-rmse:0.33738\ttrain-rmse:0.255468\n",
      "[134]\teval-rmse:0.337317\ttrain-rmse:0.255006\n",
      "[135]\teval-rmse:0.337221\ttrain-rmse:0.254443\n",
      "[136]\teval-rmse:0.337192\ttrain-rmse:0.253991\n",
      "[137]\teval-rmse:0.33716\ttrain-rmse:0.253525\n",
      "[138]\teval-rmse:0.337033\ttrain-rmse:0.252815\n",
      "[139]\teval-rmse:0.336993\ttrain-rmse:0.252399\n",
      "[140]\teval-rmse:0.336971\ttrain-rmse:0.252047\n",
      "[141]\teval-rmse:0.33686\ttrain-rmse:0.251438\n",
      "[142]\teval-rmse:0.336821\ttrain-rmse:0.250938\n",
      "[143]\teval-rmse:0.336723\ttrain-rmse:0.250365\n",
      "[144]\teval-rmse:0.336611\ttrain-rmse:0.249865\n",
      "[145]\teval-rmse:0.336557\ttrain-rmse:0.24937\n",
      "[146]\teval-rmse:0.336542\ttrain-rmse:0.249056\n",
      "[147]\teval-rmse:0.336495\ttrain-rmse:0.248702\n",
      "[148]\teval-rmse:0.336453\ttrain-rmse:0.24821\n",
      "[149]\teval-rmse:0.336319\ttrain-rmse:0.24773\n",
      "[150]\teval-rmse:0.336296\ttrain-rmse:0.247433\n",
      "[151]\teval-rmse:0.336202\ttrain-rmse:0.246797\n",
      "[152]\teval-rmse:0.33619\ttrain-rmse:0.246442\n",
      "[153]\teval-rmse:0.336149\ttrain-rmse:0.245907\n",
      "[154]\teval-rmse:0.336111\ttrain-rmse:0.245573\n",
      "[155]\teval-rmse:0.33608\ttrain-rmse:0.245068\n",
      "[156]\teval-rmse:0.336038\ttrain-rmse:0.244789\n",
      "[157]\teval-rmse:0.336019\ttrain-rmse:0.244445\n",
      "[158]\teval-rmse:0.335916\ttrain-rmse:0.243881\n",
      "[159]\teval-rmse:0.335864\ttrain-rmse:0.24359\n",
      "[160]\teval-rmse:0.335774\ttrain-rmse:0.243241\n",
      "[161]\teval-rmse:0.335749\ttrain-rmse:0.242749\n",
      "[162]\teval-rmse:0.335673\ttrain-rmse:0.242361\n",
      "[163]\teval-rmse:0.335638\ttrain-rmse:0.242154\n",
      "[164]\teval-rmse:0.335586\ttrain-rmse:0.241807\n",
      "[165]\teval-rmse:0.335463\ttrain-rmse:0.241218\n",
      "[166]\teval-rmse:0.335444\ttrain-rmse:0.241039\n",
      "[167]\teval-rmse:0.33544\ttrain-rmse:0.24085\n",
      "[168]\teval-rmse:0.33541\ttrain-rmse:0.240582\n",
      "[169]\teval-rmse:0.335221\ttrain-rmse:0.239863\n",
      "[170]\teval-rmse:0.33517\ttrain-rmse:0.239517\n",
      "[171]\teval-rmse:0.335118\ttrain-rmse:0.239297\n",
      "[172]\teval-rmse:0.335076\ttrain-rmse:0.239025\n",
      "[173]\teval-rmse:0.335059\ttrain-rmse:0.238829\n",
      "[174]\teval-rmse:0.334969\ttrain-rmse:0.23833\n",
      "[175]\teval-rmse:0.334943\ttrain-rmse:0.238118\n",
      "[176]\teval-rmse:0.334905\ttrain-rmse:0.237872\n",
      "[177]\teval-rmse:0.334802\ttrain-rmse:0.237339\n",
      "[178]\teval-rmse:0.334795\ttrain-rmse:0.237176\n",
      "[179]\teval-rmse:0.334734\ttrain-rmse:0.236673\n",
      "[180]\teval-rmse:0.334725\ttrain-rmse:0.236623\n",
      "[181]\teval-rmse:0.334645\ttrain-rmse:0.236144\n",
      "[182]\teval-rmse:0.334585\ttrain-rmse:0.236003\n",
      "[183]\teval-rmse:0.334547\ttrain-rmse:0.235685\n",
      "[184]\teval-rmse:0.334549\ttrain-rmse:0.235284\n",
      "[185]\teval-rmse:0.3345\ttrain-rmse:0.234983\n",
      "[186]\teval-rmse:0.33449\ttrain-rmse:0.234916\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[187]\teval-rmse:0.334466\ttrain-rmse:0.234712\n",
      "[188]\teval-rmse:0.334385\ttrain-rmse:0.234232\n",
      "[189]\teval-rmse:0.334386\ttrain-rmse:0.234014\n",
      "[190]\teval-rmse:0.334356\ttrain-rmse:0.23377\n",
      "[191]\teval-rmse:0.334358\ttrain-rmse:0.233545\n",
      "[192]\teval-rmse:0.334352\ttrain-rmse:0.233408\n",
      "[193]\teval-rmse:0.334336\ttrain-rmse:0.233274\n",
      "[194]\teval-rmse:0.334333\ttrain-rmse:0.23318\n",
      "[195]\teval-rmse:0.33429\ttrain-rmse:0.232826\n",
      "[196]\teval-rmse:0.334235\ttrain-rmse:0.232543\n",
      "[197]\teval-rmse:0.334228\ttrain-rmse:0.23237\n",
      "[198]\teval-rmse:0.334234\ttrain-rmse:0.232278\n",
      "[199]\teval-rmse:0.334223\ttrain-rmse:0.232187\n",
      "[200]\teval-rmse:0.33421\ttrain-rmse:0.231961\n",
      "[201]\teval-rmse:0.334186\ttrain-rmse:0.231716\n",
      "[202]\teval-rmse:0.334178\ttrain-rmse:0.23152\n",
      "[203]\teval-rmse:0.334165\ttrain-rmse:0.23142\n",
      "[204]\teval-rmse:0.334117\ttrain-rmse:0.231167\n",
      "[205]\teval-rmse:0.334113\ttrain-rmse:0.231109\n",
      "[206]\teval-rmse:0.334066\ttrain-rmse:0.230908\n",
      "[207]\teval-rmse:0.334053\ttrain-rmse:0.230573\n",
      "[208]\teval-rmse:0.333972\ttrain-rmse:0.230348\n",
      "[209]\teval-rmse:0.333926\ttrain-rmse:0.230079\n",
      "[210]\teval-rmse:0.333901\ttrain-rmse:0.229838\n",
      "[211]\teval-rmse:0.333838\ttrain-rmse:0.229443\n",
      "[212]\teval-rmse:0.333806\ttrain-rmse:0.229209\n",
      "[213]\teval-rmse:0.333803\ttrain-rmse:0.229019\n",
      "[214]\teval-rmse:0.333776\ttrain-rmse:0.228565\n",
      "[215]\teval-rmse:0.333773\ttrain-rmse:0.228501\n",
      "[216]\teval-rmse:0.333754\ttrain-rmse:0.228402\n",
      "[217]\teval-rmse:0.333671\ttrain-rmse:0.228046\n",
      "[218]\teval-rmse:0.333649\ttrain-rmse:0.227936\n",
      "[219]\teval-rmse:0.333631\ttrain-rmse:0.227743\n",
      "[220]\teval-rmse:0.333586\ttrain-rmse:0.227473\n",
      "[221]\teval-rmse:0.33358\ttrain-rmse:0.227256\n",
      "[222]\teval-rmse:0.333511\ttrain-rmse:0.226818\n",
      "[223]\teval-rmse:0.333503\ttrain-rmse:0.226548\n",
      "[224]\teval-rmse:0.333452\ttrain-rmse:0.226229\n",
      "[225]\teval-rmse:0.333449\ttrain-rmse:0.226134\n",
      "[226]\teval-rmse:0.333445\ttrain-rmse:0.225938\n",
      "[227]\teval-rmse:0.333418\ttrain-rmse:0.225787\n",
      "[228]\teval-rmse:0.333414\ttrain-rmse:0.225612\n",
      "[229]\teval-rmse:0.333398\ttrain-rmse:0.225514\n",
      "[230]\teval-rmse:0.333339\ttrain-rmse:0.225272\n",
      "[231]\teval-rmse:0.333332\ttrain-rmse:0.225168\n",
      "[232]\teval-rmse:0.333333\ttrain-rmse:0.225118\n",
      "[233]\teval-rmse:0.333311\ttrain-rmse:0.225026\n",
      "[234]\teval-rmse:0.333293\ttrain-rmse:0.224899\n",
      "[235]\teval-rmse:0.333286\ttrain-rmse:0.224768\n",
      "[236]\teval-rmse:0.33327\ttrain-rmse:0.224662\n",
      "[237]\teval-rmse:0.333115\ttrain-rmse:0.224051\n",
      "[238]\teval-rmse:0.333064\ttrain-rmse:0.22364\n",
      "[239]\teval-rmse:0.333034\ttrain-rmse:0.223339\n",
      "[240]\teval-rmse:0.333037\ttrain-rmse:0.223195\n",
      "[241]\teval-rmse:0.332995\ttrain-rmse:0.222835\n",
      "[242]\teval-rmse:0.332986\ttrain-rmse:0.22261\n",
      "[243]\teval-rmse:0.332888\ttrain-rmse:0.222081\n",
      "[244]\teval-rmse:0.332835\ttrain-rmse:0.221681\n",
      "[245]\teval-rmse:0.332778\ttrain-rmse:0.221358\n",
      "[246]\teval-rmse:0.33276\ttrain-rmse:0.2211\n",
      "[247]\teval-rmse:0.33271\ttrain-rmse:0.220875\n",
      "[248]\teval-rmse:0.332711\ttrain-rmse:0.220781\n",
      "[249]\teval-rmse:0.332687\ttrain-rmse:0.220559\n",
      "[250]\teval-rmse:0.332674\ttrain-rmse:0.22039\n",
      "[251]\teval-rmse:0.332651\ttrain-rmse:0.220277\n",
      "[252]\teval-rmse:0.332632\ttrain-rmse:0.219946\n",
      "[253]\teval-rmse:0.332628\ttrain-rmse:0.219904\n",
      "[254]\teval-rmse:0.332619\ttrain-rmse:0.219823\n",
      "[255]\teval-rmse:0.332614\ttrain-rmse:0.219741\n",
      "[256]\teval-rmse:0.332584\ttrain-rmse:0.219635\n",
      "[257]\teval-rmse:0.332579\ttrain-rmse:0.219522\n",
      "[258]\teval-rmse:0.332474\ttrain-rmse:0.21903\n",
      "[259]\teval-rmse:0.332429\ttrain-rmse:0.218869\n",
      "[260]\teval-rmse:0.332425\ttrain-rmse:0.218823\n",
      "[261]\teval-rmse:0.332384\ttrain-rmse:0.218499\n",
      "[262]\teval-rmse:0.332383\ttrain-rmse:0.218355\n",
      "[263]\teval-rmse:0.33236\ttrain-rmse:0.218206\n",
      "[264]\teval-rmse:0.332357\ttrain-rmse:0.218172\n",
      "[265]\teval-rmse:0.332355\ttrain-rmse:0.218135\n",
      "[266]\teval-rmse:0.332351\ttrain-rmse:0.217948\n",
      "[267]\teval-rmse:0.332344\ttrain-rmse:0.217874\n",
      "[268]\teval-rmse:0.332316\ttrain-rmse:0.217734\n",
      "[269]\teval-rmse:0.332306\ttrain-rmse:0.217464\n",
      "[270]\teval-rmse:0.332271\ttrain-rmse:0.217269\n",
      "[271]\teval-rmse:0.332253\ttrain-rmse:0.217098\n",
      "[272]\teval-rmse:0.332249\ttrain-rmse:0.217059\n",
      "[273]\teval-rmse:0.332231\ttrain-rmse:0.216803\n",
      "[274]\teval-rmse:0.33223\ttrain-rmse:0.216688\n",
      "[275]\teval-rmse:0.332221\ttrain-rmse:0.21661\n",
      "[276]\teval-rmse:0.332207\ttrain-rmse:0.216438\n",
      "[277]\teval-rmse:0.33215\ttrain-rmse:0.216085\n",
      "[278]\teval-rmse:0.332149\ttrain-rmse:0.215833\n",
      "[279]\teval-rmse:0.332146\ttrain-rmse:0.215718\n",
      "[280]\teval-rmse:0.332119\ttrain-rmse:0.21542\n",
      "[281]\teval-rmse:0.332108\ttrain-rmse:0.215369\n",
      "[282]\teval-rmse:0.33209\ttrain-rmse:0.215127\n",
      "[283]\teval-rmse:0.33208\ttrain-rmse:0.214814\n",
      "[284]\teval-rmse:0.331983\ttrain-rmse:0.214408\n",
      "[285]\teval-rmse:0.331981\ttrain-rmse:0.214247\n",
      "[286]\teval-rmse:0.331974\ttrain-rmse:0.213953\n",
      "[287]\teval-rmse:0.33194\ttrain-rmse:0.213706\n",
      "[288]\teval-rmse:0.331941\ttrain-rmse:0.213652\n",
      "[289]\teval-rmse:0.331923\ttrain-rmse:0.213427\n",
      "[290]\teval-rmse:0.331911\ttrain-rmse:0.213292\n",
      "[291]\teval-rmse:0.331907\ttrain-rmse:0.213121\n",
      "[292]\teval-rmse:0.331883\ttrain-rmse:0.212946\n",
      "[293]\teval-rmse:0.331884\ttrain-rmse:0.212919\n",
      "[294]\teval-rmse:0.331869\ttrain-rmse:0.212734\n",
      "[295]\teval-rmse:0.331841\ttrain-rmse:0.212402\n",
      "[296]\teval-rmse:0.331836\ttrain-rmse:0.21238\n",
      "[297]\teval-rmse:0.331826\ttrain-rmse:0.212276\n",
      "[298]\teval-rmse:0.331821\ttrain-rmse:0.212174\n",
      "[299]\teval-rmse:0.331788\ttrain-rmse:0.211874\n",
      "[300]\teval-rmse:0.331787\ttrain-rmse:0.211829\n",
      "[301]\teval-rmse:0.331782\ttrain-rmse:0.211745\n",
      "[302]\teval-rmse:0.331772\ttrain-rmse:0.2115\n",
      "[303]\teval-rmse:0.331756\ttrain-rmse:0.211297\n",
      "[304]\teval-rmse:0.33172\ttrain-rmse:0.211016\n",
      "[305]\teval-rmse:0.33171\ttrain-rmse:0.210955\n",
      "[306]\teval-rmse:0.331707\ttrain-rmse:0.210891\n",
      "[307]\teval-rmse:0.331651\ttrain-rmse:0.210528\n",
      "[308]\teval-rmse:0.331625\ttrain-rmse:0.210184\n",
      "[309]\teval-rmse:0.331591\ttrain-rmse:0.209976\n",
      "[310]\teval-rmse:0.331574\ttrain-rmse:0.209866\n",
      "[311]\teval-rmse:0.331517\ttrain-rmse:0.209609\n",
      "[312]\teval-rmse:0.331507\ttrain-rmse:0.209501\n",
      "[313]\teval-rmse:0.331515\ttrain-rmse:0.209426\n",
      "[314]\teval-rmse:0.331509\ttrain-rmse:0.209311\n",
      "[315]\teval-rmse:0.331494\ttrain-rmse:0.209203\n",
      "[316]\teval-rmse:0.331492\ttrain-rmse:0.209076\n",
      "[317]\teval-rmse:0.331484\ttrain-rmse:0.208994\n",
      "[318]\teval-rmse:0.331472\ttrain-rmse:0.208851\n",
      "[319]\teval-rmse:0.331463\ttrain-rmse:0.208742\n",
      "[320]\teval-rmse:0.331464\ttrain-rmse:0.208668\n",
      "[321]\teval-rmse:0.331464\ttrain-rmse:0.208599\n",
      "[322]\teval-rmse:0.331462\ttrain-rmse:0.208418\n",
      "[323]\teval-rmse:0.331462\ttrain-rmse:0.208381\n",
      "[324]\teval-rmse:0.331463\ttrain-rmse:0.208338\n",
      "[325]\teval-rmse:0.331416\ttrain-rmse:0.20801\n",
      "[326]\teval-rmse:0.331352\ttrain-rmse:0.207661\n",
      "[327]\teval-rmse:0.331327\ttrain-rmse:0.207494\n",
      "[328]\teval-rmse:0.331326\ttrain-rmse:0.207401\n",
      "[329]\teval-rmse:0.331321\ttrain-rmse:0.207283\n",
      "[330]\teval-rmse:0.331315\ttrain-rmse:0.20718\n",
      "[331]\teval-rmse:0.331314\ttrain-rmse:0.206934\n",
      "[332]\teval-rmse:0.33131\ttrain-rmse:0.206916\n",
      "[333]\teval-rmse:0.33131\ttrain-rmse:0.206889\n",
      "[334]\teval-rmse:0.331292\ttrain-rmse:0.206721\n",
      "[335]\teval-rmse:0.331283\ttrain-rmse:0.20656\n",
      "[336]\teval-rmse:0.331277\ttrain-rmse:0.206466\n",
      "[337]\teval-rmse:0.331276\ttrain-rmse:0.206398\n",
      "[338]\teval-rmse:0.331269\ttrain-rmse:0.206317\n",
      "[339]\teval-rmse:0.33126\ttrain-rmse:0.206247\n",
      "[340]\teval-rmse:0.331248\ttrain-rmse:0.206124\n",
      "[341]\teval-rmse:0.331234\ttrain-rmse:0.205984\n",
      "[342]\teval-rmse:0.331218\ttrain-rmse:0.205782\n",
      "[343]\teval-rmse:0.331211\ttrain-rmse:0.205737\n",
      "[344]\teval-rmse:0.33121\ttrain-rmse:0.205717\n",
      "[345]\teval-rmse:0.331206\ttrain-rmse:0.205623\n",
      "[346]\teval-rmse:0.331205\ttrain-rmse:0.205524\n",
      "[347]\teval-rmse:0.331196\ttrain-rmse:0.205421\n",
      "[348]\teval-rmse:0.331194\ttrain-rmse:0.205383\n",
      "[349]\teval-rmse:0.331195\ttrain-rmse:0.2052\n",
      "[350]\teval-rmse:0.331182\ttrain-rmse:0.204875\n",
      "[351]\teval-rmse:0.331181\ttrain-rmse:0.204751\n",
      "[352]\teval-rmse:0.331172\ttrain-rmse:0.204556\n",
      "[353]\teval-rmse:0.331173\ttrain-rmse:0.204535\n",
      "[354]\teval-rmse:0.331174\ttrain-rmse:0.204416\n",
      "[355]\teval-rmse:0.331169\ttrain-rmse:0.204363\n",
      "[356]\teval-rmse:0.331163\ttrain-rmse:0.204162\n",
      "[357]\teval-rmse:0.331142\ttrain-rmse:0.203978\n",
      "[358]\teval-rmse:0.331139\ttrain-rmse:0.203821\n",
      "[359]\teval-rmse:0.331138\ttrain-rmse:0.203762\n",
      "[360]\teval-rmse:0.331135\ttrain-rmse:0.203609\n",
      "[361]\teval-rmse:0.331141\ttrain-rmse:0.203549\n",
      "[362]\teval-rmse:0.331124\ttrain-rmse:0.203326\n",
      "[363]\teval-rmse:0.33111\ttrain-rmse:0.20309\n",
      "[364]\teval-rmse:0.331079\ttrain-rmse:0.202909\n",
      "[365]\teval-rmse:0.331086\ttrain-rmse:0.202839\n",
      "[366]\teval-rmse:0.331053\ttrain-rmse:0.202619\n",
      "[367]\teval-rmse:0.330998\ttrain-rmse:0.202374\n",
      "[368]\teval-rmse:0.331002\ttrain-rmse:0.20224\n",
      "[369]\teval-rmse:0.331\ttrain-rmse:0.202184\n",
      "[370]\teval-rmse:0.330998\ttrain-rmse:0.20203\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[371]\teval-rmse:0.330998\ttrain-rmse:0.201988\n",
      "[372]\teval-rmse:0.330982\ttrain-rmse:0.201881\n",
      "[373]\teval-rmse:0.330974\ttrain-rmse:0.201633\n",
      "[374]\teval-rmse:0.330967\ttrain-rmse:0.201566\n",
      "[375]\teval-rmse:0.330967\ttrain-rmse:0.201507\n",
      "[376]\teval-rmse:0.330957\ttrain-rmse:0.201406\n",
      "[377]\teval-rmse:0.330958\ttrain-rmse:0.201389\n",
      "[378]\teval-rmse:0.330958\ttrain-rmse:0.201382\n",
      "[379]\teval-rmse:0.330954\ttrain-rmse:0.201285\n",
      "[380]\teval-rmse:0.330953\ttrain-rmse:0.201224\n",
      "[381]\teval-rmse:0.330934\ttrain-rmse:0.201098\n",
      "[382]\teval-rmse:0.330903\ttrain-rmse:0.200911\n",
      "[383]\teval-rmse:0.330884\ttrain-rmse:0.200731\n",
      "[384]\teval-rmse:0.330883\ttrain-rmse:0.200651\n",
      "[385]\teval-rmse:0.330853\ttrain-rmse:0.200408\n",
      "[386]\teval-rmse:0.330852\ttrain-rmse:0.200324\n",
      "[387]\teval-rmse:0.330849\ttrain-rmse:0.200223\n",
      "[388]\teval-rmse:0.330839\ttrain-rmse:0.200147\n",
      "[389]\teval-rmse:0.330841\ttrain-rmse:0.200123\n",
      "[390]\teval-rmse:0.330761\ttrain-rmse:0.199674\n",
      "[391]\teval-rmse:0.33076\ttrain-rmse:0.199556\n",
      "[392]\teval-rmse:0.330753\ttrain-rmse:0.199464\n",
      "[393]\teval-rmse:0.330736\ttrain-rmse:0.199359\n",
      "[394]\teval-rmse:0.330734\ttrain-rmse:0.199292\n",
      "[395]\teval-rmse:0.330724\ttrain-rmse:0.199181\n",
      "[396]\teval-rmse:0.330722\ttrain-rmse:0.19914\n",
      "[397]\teval-rmse:0.330714\ttrain-rmse:0.199029\n",
      "[398]\teval-rmse:0.330718\ttrain-rmse:0.198958\n",
      "[399]\teval-rmse:0.330714\ttrain-rmse:0.198899\n",
      "[400]\teval-rmse:0.330716\ttrain-rmse:0.198854\n",
      "[401]\teval-rmse:0.330726\ttrain-rmse:0.198754\n",
      "[402]\teval-rmse:0.330725\ttrain-rmse:0.198717\n",
      "[403]\teval-rmse:0.330713\ttrain-rmse:0.198553\n",
      "[404]\teval-rmse:0.330713\ttrain-rmse:0.198513\n",
      "[405]\teval-rmse:0.330712\ttrain-rmse:0.198478\n",
      "[406]\teval-rmse:0.330709\ttrain-rmse:0.198448\n",
      "[407]\teval-rmse:0.330685\ttrain-rmse:0.198304\n",
      "[408]\teval-rmse:0.330682\ttrain-rmse:0.19823\n",
      "[409]\teval-rmse:0.330678\ttrain-rmse:0.198187\n",
      "[410]\teval-rmse:0.330657\ttrain-rmse:0.197981\n",
      "[411]\teval-rmse:0.330657\ttrain-rmse:0.197923\n",
      "[412]\teval-rmse:0.330648\ttrain-rmse:0.197775\n",
      "[413]\teval-rmse:0.330643\ttrain-rmse:0.197737\n",
      "[414]\teval-rmse:0.330641\ttrain-rmse:0.19764\n",
      "[415]\teval-rmse:0.330617\ttrain-rmse:0.1975\n",
      "[416]\teval-rmse:0.33061\ttrain-rmse:0.197385\n",
      "[417]\teval-rmse:0.330601\ttrain-rmse:0.197288\n",
      "[418]\teval-rmse:0.33058\ttrain-rmse:0.197034\n",
      "[419]\teval-rmse:0.330578\ttrain-rmse:0.196992\n",
      "[420]\teval-rmse:0.330564\ttrain-rmse:0.196847\n",
      "[421]\teval-rmse:0.330564\ttrain-rmse:0.19678\n",
      "[422]\teval-rmse:0.330559\ttrain-rmse:0.196674\n",
      "[423]\teval-rmse:0.330553\ttrain-rmse:0.196542\n",
      "[424]\teval-rmse:0.330553\ttrain-rmse:0.196532\n",
      "[425]\teval-rmse:0.330549\ttrain-rmse:0.196434\n",
      "[426]\teval-rmse:0.330541\ttrain-rmse:0.196369\n",
      "[427]\teval-rmse:0.330535\ttrain-rmse:0.196299\n",
      "[428]\teval-rmse:0.330529\ttrain-rmse:0.196187\n",
      "[429]\teval-rmse:0.330514\ttrain-rmse:0.196057\n",
      "[430]\teval-rmse:0.330503\ttrain-rmse:0.195983\n",
      "[431]\teval-rmse:0.330504\ttrain-rmse:0.195925\n",
      "[432]\teval-rmse:0.330497\ttrain-rmse:0.195853\n",
      "[433]\teval-rmse:0.330483\ttrain-rmse:0.195693\n",
      "[434]\teval-rmse:0.330484\ttrain-rmse:0.195635\n",
      "[435]\teval-rmse:0.330468\ttrain-rmse:0.19546\n",
      "[436]\teval-rmse:0.33047\ttrain-rmse:0.195323\n",
      "[437]\teval-rmse:0.330474\ttrain-rmse:0.195211\n",
      "[438]\teval-rmse:0.330444\ttrain-rmse:0.194991\n",
      "[439]\teval-rmse:0.330441\ttrain-rmse:0.194955\n",
      "[440]\teval-rmse:0.330437\ttrain-rmse:0.194858\n",
      "[441]\teval-rmse:0.330424\ttrain-rmse:0.194637\n",
      "[442]\teval-rmse:0.330421\ttrain-rmse:0.194536\n",
      "[443]\teval-rmse:0.330409\ttrain-rmse:0.194211\n",
      "[444]\teval-rmse:0.330409\ttrain-rmse:0.194193\n",
      "[445]\teval-rmse:0.330408\ttrain-rmse:0.19395\n",
      "[446]\teval-rmse:0.330404\ttrain-rmse:0.193901\n",
      "[447]\teval-rmse:0.33038\ttrain-rmse:0.193717\n",
      "[448]\teval-rmse:0.330354\ttrain-rmse:0.19348\n",
      "[449]\teval-rmse:0.330354\ttrain-rmse:0.193419\n",
      "[450]\teval-rmse:0.330354\ttrain-rmse:0.193345\n",
      "[451]\teval-rmse:0.330341\ttrain-rmse:0.193169\n",
      "[452]\teval-rmse:0.330334\ttrain-rmse:0.192982\n",
      "[453]\teval-rmse:0.330334\ttrain-rmse:0.192933\n",
      "[454]\teval-rmse:0.330337\ttrain-rmse:0.192879\n",
      "[455]\teval-rmse:0.330337\ttrain-rmse:0.19286\n",
      "[456]\teval-rmse:0.330338\ttrain-rmse:0.192834\n",
      "[457]\teval-rmse:0.330332\ttrain-rmse:0.192716\n",
      "[458]\teval-rmse:0.330324\ttrain-rmse:0.192623\n",
      "[459]\teval-rmse:0.330321\ttrain-rmse:0.192455\n",
      "[460]\teval-rmse:0.330312\ttrain-rmse:0.192308\n",
      "[461]\teval-rmse:0.330314\ttrain-rmse:0.192252\n",
      "[462]\teval-rmse:0.330295\ttrain-rmse:0.192113\n",
      "[463]\teval-rmse:0.330295\ttrain-rmse:0.1921\n",
      "[464]\teval-rmse:0.330297\ttrain-rmse:0.192011\n",
      "[465]\teval-rmse:0.33029\ttrain-rmse:0.191811\n",
      "[466]\teval-rmse:0.330278\ttrain-rmse:0.191697\n",
      "[467]\teval-rmse:0.330278\ttrain-rmse:0.191618\n",
      "[468]\teval-rmse:0.330267\ttrain-rmse:0.191427\n",
      "[469]\teval-rmse:0.330261\ttrain-rmse:0.191224\n",
      "[470]\teval-rmse:0.330261\ttrain-rmse:0.191163\n",
      "[471]\teval-rmse:0.330259\ttrain-rmse:0.191076\n",
      "[472]\teval-rmse:0.330253\ttrain-rmse:0.190942\n",
      "[473]\teval-rmse:0.33024\ttrain-rmse:0.190859\n",
      "[474]\teval-rmse:0.330231\ttrain-rmse:0.190743\n",
      "[475]\teval-rmse:0.330229\ttrain-rmse:0.190695\n",
      "[476]\teval-rmse:0.330218\ttrain-rmse:0.190589\n",
      "[477]\teval-rmse:0.330213\ttrain-rmse:0.190424\n",
      "[478]\teval-rmse:0.330196\ttrain-rmse:0.190157\n",
      "[479]\teval-rmse:0.330191\ttrain-rmse:0.190055\n",
      "[480]\teval-rmse:0.33019\ttrain-rmse:0.189934\n",
      "[481]\teval-rmse:0.330188\ttrain-rmse:0.189805\n",
      "[482]\teval-rmse:0.33019\ttrain-rmse:0.18978\n",
      "[483]\teval-rmse:0.330193\ttrain-rmse:0.189595\n",
      "[484]\teval-rmse:0.330193\ttrain-rmse:0.189509\n",
      "[485]\teval-rmse:0.330184\ttrain-rmse:0.189375\n",
      "[486]\teval-rmse:0.33018\ttrain-rmse:0.189339\n",
      "[487]\teval-rmse:0.330162\ttrain-rmse:0.189169\n",
      "[488]\teval-rmse:0.330161\ttrain-rmse:0.189015\n",
      "[489]\teval-rmse:0.33016\ttrain-rmse:0.188937\n",
      "[490]\teval-rmse:0.330153\ttrain-rmse:0.18888\n",
      "[491]\teval-rmse:0.330151\ttrain-rmse:0.188819\n",
      "[492]\teval-rmse:0.330148\ttrain-rmse:0.188753\n",
      "[493]\teval-rmse:0.330147\ttrain-rmse:0.188545\n",
      "[494]\teval-rmse:0.330148\ttrain-rmse:0.188455\n",
      "[495]\teval-rmse:0.330149\ttrain-rmse:0.188371\n",
      "[496]\teval-rmse:0.330148\ttrain-rmse:0.188319\n",
      "[497]\teval-rmse:0.330147\ttrain-rmse:0.188291\n",
      "[498]\teval-rmse:0.330147\ttrain-rmse:0.188182\n",
      "[499]\teval-rmse:0.330144\ttrain-rmse:0.188096\n",
      "[500]\teval-rmse:0.330144\ttrain-rmse:0.188008\n",
      "[501]\teval-rmse:0.330146\ttrain-rmse:0.187962\n",
      "[502]\teval-rmse:0.330147\ttrain-rmse:0.187922\n",
      "[503]\teval-rmse:0.33014\ttrain-rmse:0.187732\n",
      "[504]\teval-rmse:0.330126\ttrain-rmse:0.187593\n",
      "[505]\teval-rmse:0.33012\ttrain-rmse:0.187533\n",
      "[506]\teval-rmse:0.330099\ttrain-rmse:0.18735\n",
      "[507]\teval-rmse:0.3301\ttrain-rmse:0.187293\n",
      "[508]\teval-rmse:0.330094\ttrain-rmse:0.187167\n",
      "[509]\teval-rmse:0.330091\ttrain-rmse:0.186955\n",
      "[510]\teval-rmse:0.330087\ttrain-rmse:0.186819\n",
      "[511]\teval-rmse:0.330079\ttrain-rmse:0.186594\n",
      "[512]\teval-rmse:0.330078\ttrain-rmse:0.186483\n",
      "[513]\teval-rmse:0.330077\ttrain-rmse:0.186423\n",
      "[514]\teval-rmse:0.330075\ttrain-rmse:0.186284\n",
      "[515]\teval-rmse:0.330079\ttrain-rmse:0.186205\n",
      "[516]\teval-rmse:0.330075\ttrain-rmse:0.186018\n",
      "[517]\teval-rmse:0.330076\ttrain-rmse:0.185915\n",
      "[518]\teval-rmse:0.33008\ttrain-rmse:0.185795\n",
      "[519]\teval-rmse:0.330074\ttrain-rmse:0.185653\n",
      "[520]\teval-rmse:0.33007\ttrain-rmse:0.185565\n",
      "[521]\teval-rmse:0.330063\ttrain-rmse:0.185289\n",
      "[522]\teval-rmse:0.330061\ttrain-rmse:0.185208\n",
      "[523]\teval-rmse:0.330059\ttrain-rmse:0.185129\n",
      "[524]\teval-rmse:0.33005\ttrain-rmse:0.185087\n",
      "[525]\teval-rmse:0.330045\ttrain-rmse:0.184926\n",
      "[526]\teval-rmse:0.330049\ttrain-rmse:0.184865\n",
      "[527]\teval-rmse:0.330029\ttrain-rmse:0.184727\n",
      "[528]\teval-rmse:0.330021\ttrain-rmse:0.184569\n",
      "[529]\teval-rmse:0.330018\ttrain-rmse:0.184393\n",
      "[530]\teval-rmse:0.330017\ttrain-rmse:0.184309\n",
      "[531]\teval-rmse:0.330015\ttrain-rmse:0.184257\n",
      "[532]\teval-rmse:0.330019\ttrain-rmse:0.18424\n",
      "[533]\teval-rmse:0.330014\ttrain-rmse:0.184019\n",
      "[534]\teval-rmse:0.330019\ttrain-rmse:0.183861\n",
      "[535]\teval-rmse:0.330015\ttrain-rmse:0.183747\n",
      "[536]\teval-rmse:0.330014\ttrain-rmse:0.183664\n",
      "[537]\teval-rmse:0.33\ttrain-rmse:0.183518\n",
      "[538]\teval-rmse:0.33\ttrain-rmse:0.183437\n",
      "[539]\teval-rmse:0.330003\ttrain-rmse:0.183382\n",
      "[540]\teval-rmse:0.33\ttrain-rmse:0.18329\n",
      "[541]\teval-rmse:0.329995\ttrain-rmse:0.183194\n",
      "[542]\teval-rmse:0.329989\ttrain-rmse:0.18314\n",
      "[543]\teval-rmse:0.329989\ttrain-rmse:0.183132\n",
      "[544]\teval-rmse:0.329987\ttrain-rmse:0.183048\n",
      "[545]\teval-rmse:0.329977\ttrain-rmse:0.1829\n",
      "[546]\teval-rmse:0.329972\ttrain-rmse:0.182812\n",
      "[547]\teval-rmse:0.329957\ttrain-rmse:0.18268\n",
      "[548]\teval-rmse:0.329953\ttrain-rmse:0.182627\n",
      "[549]\teval-rmse:0.329941\ttrain-rmse:0.182473\n",
      "[550]\teval-rmse:0.329944\ttrain-rmse:0.182358\n",
      "[551]\teval-rmse:0.329927\ttrain-rmse:0.182238\n",
      "[552]\teval-rmse:0.329926\ttrain-rmse:0.182223\n",
      "[553]\teval-rmse:0.329907\ttrain-rmse:0.182095\n",
      "[554]\teval-rmse:0.329913\ttrain-rmse:0.181961\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[555]\teval-rmse:0.329905\ttrain-rmse:0.181889\n",
      "[556]\teval-rmse:0.329907\ttrain-rmse:0.181829\n",
      "[557]\teval-rmse:0.329905\ttrain-rmse:0.181744\n",
      "[558]\teval-rmse:0.329902\ttrain-rmse:0.181682\n",
      "[559]\teval-rmse:0.3299\ttrain-rmse:0.181599\n",
      "[560]\teval-rmse:0.329896\ttrain-rmse:0.181539\n",
      "[561]\teval-rmse:0.329894\ttrain-rmse:0.181458\n",
      "[562]\teval-rmse:0.329881\ttrain-rmse:0.181253\n",
      "[563]\teval-rmse:0.329879\ttrain-rmse:0.181153\n",
      "[564]\teval-rmse:0.329869\ttrain-rmse:0.181018\n",
      "[565]\teval-rmse:0.329868\ttrain-rmse:0.180964\n",
      "[566]\teval-rmse:0.329868\ttrain-rmse:0.180865\n",
      "[567]\teval-rmse:0.329862\ttrain-rmse:0.18077\n",
      "[568]\teval-rmse:0.329861\ttrain-rmse:0.180677\n",
      "[569]\teval-rmse:0.329861\ttrain-rmse:0.180633\n",
      "[570]\teval-rmse:0.329858\ttrain-rmse:0.180482\n",
      "[571]\teval-rmse:0.329858\ttrain-rmse:0.180439\n",
      "[572]\teval-rmse:0.329857\ttrain-rmse:0.180353\n",
      "[573]\teval-rmse:0.329855\ttrain-rmse:0.180224\n",
      "[574]\teval-rmse:0.329856\ttrain-rmse:0.180217\n",
      "[575]\teval-rmse:0.329861\ttrain-rmse:0.180102\n",
      "[576]\teval-rmse:0.329861\ttrain-rmse:0.180064\n",
      "[577]\teval-rmse:0.329862\ttrain-rmse:0.180003\n",
      "[578]\teval-rmse:0.329866\ttrain-rmse:0.179887\n",
      "[579]\teval-rmse:0.32986\ttrain-rmse:0.179773\n",
      "[580]\teval-rmse:0.329856\ttrain-rmse:0.179703\n",
      "[581]\teval-rmse:0.329853\ttrain-rmse:0.179665\n",
      "[582]\teval-rmse:0.329854\ttrain-rmse:0.179655\n",
      "[583]\teval-rmse:0.329853\ttrain-rmse:0.179498\n",
      "[584]\teval-rmse:0.329853\ttrain-rmse:0.179479\n",
      "[585]\teval-rmse:0.329842\ttrain-rmse:0.179389\n",
      "[586]\teval-rmse:0.32983\ttrain-rmse:0.179254\n",
      "[587]\teval-rmse:0.32983\ttrain-rmse:0.179198\n",
      "[588]\teval-rmse:0.329823\ttrain-rmse:0.179049\n",
      "[589]\teval-rmse:0.329812\ttrain-rmse:0.17893\n",
      "[590]\teval-rmse:0.329782\ttrain-rmse:0.178739\n",
      "[591]\teval-rmse:0.329776\ttrain-rmse:0.178653\n",
      "[592]\teval-rmse:0.329775\ttrain-rmse:0.178601\n",
      "[593]\teval-rmse:0.329773\ttrain-rmse:0.178541\n",
      "[594]\teval-rmse:0.329769\ttrain-rmse:0.178386\n",
      "[595]\teval-rmse:0.32977\ttrain-rmse:0.178346\n",
      "[596]\teval-rmse:0.329771\ttrain-rmse:0.17831\n",
      "[597]\teval-rmse:0.329768\ttrain-rmse:0.178243\n",
      "[598]\teval-rmse:0.329768\ttrain-rmse:0.178192\n",
      "[599]\teval-rmse:0.329759\ttrain-rmse:0.178104\n",
      "[600]\teval-rmse:0.329756\ttrain-rmse:0.178015\n",
      "[601]\teval-rmse:0.329749\ttrain-rmse:0.177906\n",
      "[602]\teval-rmse:0.329749\ttrain-rmse:0.177893\n",
      "[603]\teval-rmse:0.329746\ttrain-rmse:0.177808\n",
      "[604]\teval-rmse:0.329737\ttrain-rmse:0.177587\n",
      "[605]\teval-rmse:0.329738\ttrain-rmse:0.17754\n",
      "[606]\teval-rmse:0.329738\ttrain-rmse:0.177527\n",
      "[607]\teval-rmse:0.329738\ttrain-rmse:0.177517\n",
      "[608]\teval-rmse:0.329738\ttrain-rmse:0.177462\n",
      "[609]\teval-rmse:0.329727\ttrain-rmse:0.177355\n",
      "[610]\teval-rmse:0.329731\ttrain-rmse:0.177283\n",
      "[611]\teval-rmse:0.329728\ttrain-rmse:0.177206\n",
      "[612]\teval-rmse:0.329722\ttrain-rmse:0.177028\n",
      "[613]\teval-rmse:0.329705\ttrain-rmse:0.17692\n",
      "[614]\teval-rmse:0.329703\ttrain-rmse:0.1768\n",
      "[615]\teval-rmse:0.329699\ttrain-rmse:0.17666\n",
      "[616]\teval-rmse:0.329698\ttrain-rmse:0.176568\n",
      "[617]\teval-rmse:0.329695\ttrain-rmse:0.176547\n",
      "[618]\teval-rmse:0.329686\ttrain-rmse:0.176394\n",
      "[619]\teval-rmse:0.329673\ttrain-rmse:0.176286\n",
      "[620]\teval-rmse:0.329667\ttrain-rmse:0.176248\n",
      "[621]\teval-rmse:0.32967\ttrain-rmse:0.176203\n",
      "[622]\teval-rmse:0.329667\ttrain-rmse:0.176143\n",
      "[623]\teval-rmse:0.329654\ttrain-rmse:0.176019\n",
      "[624]\teval-rmse:0.32966\ttrain-rmse:0.175903\n",
      "[625]\teval-rmse:0.32966\ttrain-rmse:0.175858\n",
      "[626]\teval-rmse:0.32966\ttrain-rmse:0.175759\n",
      "[627]\teval-rmse:0.329658\ttrain-rmse:0.175702\n",
      "[628]\teval-rmse:0.329651\ttrain-rmse:0.175643\n",
      "[629]\teval-rmse:0.329647\ttrain-rmse:0.175554\n",
      "[630]\teval-rmse:0.329634\ttrain-rmse:0.17538\n",
      "[631]\teval-rmse:0.329631\ttrain-rmse:0.175327\n",
      "[632]\teval-rmse:0.329631\ttrain-rmse:0.175245\n",
      "[633]\teval-rmse:0.32962\ttrain-rmse:0.17515\n",
      "[634]\teval-rmse:0.329624\ttrain-rmse:0.175081\n",
      "[635]\teval-rmse:0.329625\ttrain-rmse:0.175041\n",
      "[636]\teval-rmse:0.329618\ttrain-rmse:0.174906\n",
      "[637]\teval-rmse:0.329617\ttrain-rmse:0.17488\n",
      "[638]\teval-rmse:0.329612\ttrain-rmse:0.17478\n",
      "[639]\teval-rmse:0.329607\ttrain-rmse:0.174639\n",
      "[640]\teval-rmse:0.329606\ttrain-rmse:0.174608\n",
      "[641]\teval-rmse:0.329606\ttrain-rmse:0.174579\n",
      "[642]\teval-rmse:0.329602\ttrain-rmse:0.174504\n",
      "[643]\teval-rmse:0.329595\ttrain-rmse:0.174383\n",
      "[644]\teval-rmse:0.329594\ttrain-rmse:0.174266\n",
      "[645]\teval-rmse:0.329592\ttrain-rmse:0.17425\n",
      "[646]\teval-rmse:0.329585\ttrain-rmse:0.174126\n",
      "[647]\teval-rmse:0.329576\ttrain-rmse:0.174039\n",
      "[648]\teval-rmse:0.329566\ttrain-rmse:0.173916\n",
      "[649]\teval-rmse:0.329564\ttrain-rmse:0.173857\n",
      "[650]\teval-rmse:0.329565\ttrain-rmse:0.173741\n",
      "[651]\teval-rmse:0.32956\ttrain-rmse:0.173686\n",
      "[652]\teval-rmse:0.329553\ttrain-rmse:0.173577\n",
      "[653]\teval-rmse:0.329541\ttrain-rmse:0.173387\n",
      "[654]\teval-rmse:0.329536\ttrain-rmse:0.173311\n",
      "[655]\teval-rmse:0.329531\ttrain-rmse:0.173249\n",
      "[656]\teval-rmse:0.32953\ttrain-rmse:0.173203\n",
      "[657]\teval-rmse:0.329529\ttrain-rmse:0.173144\n",
      "[658]\teval-rmse:0.329531\ttrain-rmse:0.173085\n",
      "[659]\teval-rmse:0.329526\ttrain-rmse:0.172969\n",
      "[660]\teval-rmse:0.329526\ttrain-rmse:0.172951\n",
      "[661]\teval-rmse:0.329527\ttrain-rmse:0.172817\n",
      "[662]\teval-rmse:0.329523\ttrain-rmse:0.172795\n",
      "[663]\teval-rmse:0.329522\ttrain-rmse:0.1727\n",
      "[664]\teval-rmse:0.329521\ttrain-rmse:0.172649\n",
      "[665]\teval-rmse:0.329516\ttrain-rmse:0.172575\n",
      "[666]\teval-rmse:0.329516\ttrain-rmse:0.172557\n",
      "[667]\teval-rmse:0.329512\ttrain-rmse:0.17248\n",
      "[668]\teval-rmse:0.329511\ttrain-rmse:0.172394\n",
      "[669]\teval-rmse:0.32951\ttrain-rmse:0.172367\n",
      "[670]\teval-rmse:0.329507\ttrain-rmse:0.172343\n",
      "[671]\teval-rmse:0.329508\ttrain-rmse:0.172288\n",
      "[672]\teval-rmse:0.329509\ttrain-rmse:0.17224\n",
      "[673]\teval-rmse:0.329495\ttrain-rmse:0.172105\n",
      "[674]\teval-rmse:0.329488\ttrain-rmse:0.171961\n",
      "[675]\teval-rmse:0.329473\ttrain-rmse:0.171808\n",
      "[676]\teval-rmse:0.329471\ttrain-rmse:0.171729\n",
      "[677]\teval-rmse:0.329475\ttrain-rmse:0.171617\n",
      "[678]\teval-rmse:0.329474\ttrain-rmse:0.171545\n",
      "[679]\teval-rmse:0.329476\ttrain-rmse:0.171497\n",
      "[680]\teval-rmse:0.329474\ttrain-rmse:0.171419\n",
      "[681]\teval-rmse:0.329472\ttrain-rmse:0.171327\n",
      "[682]\teval-rmse:0.32947\ttrain-rmse:0.171271\n",
      "[683]\teval-rmse:0.329467\ttrain-rmse:0.171205\n",
      "[684]\teval-rmse:0.329467\ttrain-rmse:0.171112\n",
      "[685]\teval-rmse:0.329459\ttrain-rmse:0.17096\n",
      "[686]\teval-rmse:0.329452\ttrain-rmse:0.170881\n",
      "[687]\teval-rmse:0.329451\ttrain-rmse:0.170751\n",
      "[688]\teval-rmse:0.329446\ttrain-rmse:0.170599\n",
      "[689]\teval-rmse:0.329445\ttrain-rmse:0.170562\n",
      "[690]\teval-rmse:0.329442\ttrain-rmse:0.17047\n",
      "[691]\teval-rmse:0.329442\ttrain-rmse:0.170396\n",
      "[692]\teval-rmse:0.32944\ttrain-rmse:0.170325\n",
      "[693]\teval-rmse:0.329433\ttrain-rmse:0.170271\n",
      "[694]\teval-rmse:0.329435\ttrain-rmse:0.17019\n",
      "[695]\teval-rmse:0.329429\ttrain-rmse:0.170061\n",
      "[696]\teval-rmse:0.32942\ttrain-rmse:0.169956\n",
      "[697]\teval-rmse:0.329416\ttrain-rmse:0.169795\n",
      "[698]\teval-rmse:0.329418\ttrain-rmse:0.16976\n",
      "[699]\teval-rmse:0.329416\ttrain-rmse:0.16961\n",
      "[700]\teval-rmse:0.329402\ttrain-rmse:0.16949\n",
      "[701]\teval-rmse:0.329401\ttrain-rmse:0.169406\n",
      "[702]\teval-rmse:0.329396\ttrain-rmse:0.16934\n",
      "[703]\teval-rmse:0.329388\ttrain-rmse:0.169261\n",
      "[704]\teval-rmse:0.329386\ttrain-rmse:0.169214\n",
      "[705]\teval-rmse:0.329384\ttrain-rmse:0.169187\n",
      "[706]\teval-rmse:0.329388\ttrain-rmse:0.168988\n",
      "[707]\teval-rmse:0.329391\ttrain-rmse:0.168954\n",
      "[708]\teval-rmse:0.32939\ttrain-rmse:0.168855\n",
      "[709]\teval-rmse:0.329388\ttrain-rmse:0.168792\n",
      "[710]\teval-rmse:0.329388\ttrain-rmse:0.168764\n",
      "[711]\teval-rmse:0.32937\ttrain-rmse:0.168596\n",
      "[712]\teval-rmse:0.329371\ttrain-rmse:0.168507\n",
      "[713]\teval-rmse:0.329372\ttrain-rmse:0.168455\n",
      "[714]\teval-rmse:0.329372\ttrain-rmse:0.168387\n",
      "[715]\teval-rmse:0.329373\ttrain-rmse:0.168358\n",
      "[716]\teval-rmse:0.329373\ttrain-rmse:0.168316\n",
      "[717]\teval-rmse:0.329369\ttrain-rmse:0.16823\n",
      "[718]\teval-rmse:0.329369\ttrain-rmse:0.168167\n",
      "[719]\teval-rmse:0.329366\ttrain-rmse:0.168114\n",
      "[720]\teval-rmse:0.329365\ttrain-rmse:0.168099\n",
      "[721]\teval-rmse:0.329363\ttrain-rmse:0.168007\n",
      "[722]\teval-rmse:0.329366\ttrain-rmse:0.167969\n",
      "[723]\teval-rmse:0.329362\ttrain-rmse:0.167896\n",
      "[724]\teval-rmse:0.329363\ttrain-rmse:0.167864\n",
      "[725]\teval-rmse:0.329355\ttrain-rmse:0.167749\n",
      "[726]\teval-rmse:0.329357\ttrain-rmse:0.167649\n",
      "[727]\teval-rmse:0.329358\ttrain-rmse:0.16761\n",
      "[728]\teval-rmse:0.329356\ttrain-rmse:0.167578\n",
      "[729]\teval-rmse:0.329361\ttrain-rmse:0.167379\n",
      "[730]\teval-rmse:0.329358\ttrain-rmse:0.167315\n",
      "[731]\teval-rmse:0.329357\ttrain-rmse:0.16724\n",
      "[732]\teval-rmse:0.329358\ttrain-rmse:0.167222\n",
      "[733]\teval-rmse:0.329355\ttrain-rmse:0.167164\n",
      "[734]\teval-rmse:0.329349\ttrain-rmse:0.167101\n",
      "[735]\teval-rmse:0.329349\ttrain-rmse:0.16696\n",
      "[736]\teval-rmse:0.329349\ttrain-rmse:0.16691\n",
      "[737]\teval-rmse:0.329347\ttrain-rmse:0.166826\n",
      "[738]\teval-rmse:0.329347\ttrain-rmse:0.166741\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[739]\teval-rmse:0.329349\ttrain-rmse:0.166723\n",
      "[740]\teval-rmse:0.329349\ttrain-rmse:0.166695\n",
      "[741]\teval-rmse:0.329346\ttrain-rmse:0.166644\n",
      "[742]\teval-rmse:0.329348\ttrain-rmse:0.166493\n",
      "[743]\teval-rmse:0.329347\ttrain-rmse:0.166468\n",
      "[744]\teval-rmse:0.329344\ttrain-rmse:0.166441\n",
      "[745]\teval-rmse:0.329343\ttrain-rmse:0.16631\n",
      "[746]\teval-rmse:0.329345\ttrain-rmse:0.166206\n",
      "[747]\teval-rmse:0.329344\ttrain-rmse:0.166154\n",
      "[748]\teval-rmse:0.329344\ttrain-rmse:0.166129\n",
      "[749]\teval-rmse:0.329346\ttrain-rmse:0.166093\n",
      "[750]\teval-rmse:0.329349\ttrain-rmse:0.166016\n",
      "[751]\teval-rmse:0.329347\ttrain-rmse:0.165956\n",
      "[752]\teval-rmse:0.329349\ttrain-rmse:0.16587\n",
      "[753]\teval-rmse:0.329346\ttrain-rmse:0.165796\n",
      "[754]\teval-rmse:0.329349\ttrain-rmse:0.165692\n",
      "[755]\teval-rmse:0.329346\ttrain-rmse:0.16559\n",
      "[756]\teval-rmse:0.329345\ttrain-rmse:0.165518\n",
      "[757]\teval-rmse:0.329342\ttrain-rmse:0.165406\n",
      "[758]\teval-rmse:0.32934\ttrain-rmse:0.165371\n",
      "[759]\teval-rmse:0.329336\ttrain-rmse:0.1653\n",
      "[760]\teval-rmse:0.329324\ttrain-rmse:0.165196\n",
      "[761]\teval-rmse:0.32932\ttrain-rmse:0.165103\n",
      "[762]\teval-rmse:0.32932\ttrain-rmse:0.16505\n",
      "[763]\teval-rmse:0.329319\ttrain-rmse:0.165019\n",
      "[764]\teval-rmse:0.32932\ttrain-rmse:0.164973\n",
      "[765]\teval-rmse:0.329323\ttrain-rmse:0.16492\n",
      "[766]\teval-rmse:0.32932\ttrain-rmse:0.164821\n",
      "[767]\teval-rmse:0.329318\ttrain-rmse:0.164793\n",
      "[768]\teval-rmse:0.329318\ttrain-rmse:0.164677\n",
      "[769]\teval-rmse:0.329315\ttrain-rmse:0.164534\n",
      "[770]\teval-rmse:0.329312\ttrain-rmse:0.164458\n",
      "[771]\teval-rmse:0.329312\ttrain-rmse:0.16432\n",
      "[772]\teval-rmse:0.329314\ttrain-rmse:0.164271\n",
      "[773]\teval-rmse:0.329312\ttrain-rmse:0.164175\n",
      "[774]\teval-rmse:0.329309\ttrain-rmse:0.164113\n",
      "[775]\teval-rmse:0.329309\ttrain-rmse:0.164065\n",
      "[776]\teval-rmse:0.329303\ttrain-rmse:0.163986\n",
      "[777]\teval-rmse:0.329302\ttrain-rmse:0.163936\n",
      "[778]\teval-rmse:0.329298\ttrain-rmse:0.163786\n",
      "[779]\teval-rmse:0.329296\ttrain-rmse:0.16368\n",
      "[780]\teval-rmse:0.329297\ttrain-rmse:0.163632\n",
      "[781]\teval-rmse:0.329301\ttrain-rmse:0.163589\n",
      "[782]\teval-rmse:0.329303\ttrain-rmse:0.163546\n",
      "[783]\teval-rmse:0.329303\ttrain-rmse:0.163485\n",
      "[784]\teval-rmse:0.329299\ttrain-rmse:0.163415\n",
      "[785]\teval-rmse:0.329299\ttrain-rmse:0.163382\n",
      "[786]\teval-rmse:0.329299\ttrain-rmse:0.163313\n",
      "[787]\teval-rmse:0.3293\ttrain-rmse:0.163291\n",
      "[788]\teval-rmse:0.329291\ttrain-rmse:0.163201\n",
      "[789]\teval-rmse:0.329291\ttrain-rmse:0.163176\n",
      "[790]\teval-rmse:0.329289\ttrain-rmse:0.163118\n",
      "[791]\teval-rmse:0.329288\ttrain-rmse:0.163053\n",
      "[792]\teval-rmse:0.329291\ttrain-rmse:0.162956\n",
      "[793]\teval-rmse:0.32929\ttrain-rmse:0.162877\n",
      "[794]\teval-rmse:0.329287\ttrain-rmse:0.162821\n",
      "[795]\teval-rmse:0.329288\ttrain-rmse:0.162738\n",
      "[796]\teval-rmse:0.329288\ttrain-rmse:0.162728\n",
      "[797]\teval-rmse:0.329288\ttrain-rmse:0.162639\n",
      "[798]\teval-rmse:0.329282\ttrain-rmse:0.162536\n",
      "[799]\teval-rmse:0.32928\ttrain-rmse:0.162488\n",
      "[800]\teval-rmse:0.329285\ttrain-rmse:0.162425\n",
      "[801]\teval-rmse:0.329285\ttrain-rmse:0.162383\n",
      "[802]\teval-rmse:0.329284\ttrain-rmse:0.162326\n",
      "[803]\teval-rmse:0.329284\ttrain-rmse:0.162299\n",
      "[804]\teval-rmse:0.32928\ttrain-rmse:0.162235\n",
      "[805]\teval-rmse:0.329284\ttrain-rmse:0.162112\n",
      "[806]\teval-rmse:0.329281\ttrain-rmse:0.162065\n",
      "[807]\teval-rmse:0.329278\ttrain-rmse:0.161984\n",
      "[808]\teval-rmse:0.329271\ttrain-rmse:0.161757\n",
      "[809]\teval-rmse:0.329267\ttrain-rmse:0.161736\n",
      "[810]\teval-rmse:0.329262\ttrain-rmse:0.161624\n",
      "[811]\teval-rmse:0.329265\ttrain-rmse:0.161536\n",
      "[812]\teval-rmse:0.329248\ttrain-rmse:0.161319\n",
      "[813]\teval-rmse:0.329233\ttrain-rmse:0.16121\n",
      "[814]\teval-rmse:0.329234\ttrain-rmse:0.161172\n",
      "[815]\teval-rmse:0.329234\ttrain-rmse:0.161093\n",
      "[816]\teval-rmse:0.329234\ttrain-rmse:0.161069\n",
      "[817]\teval-rmse:0.329236\ttrain-rmse:0.160955\n",
      "[818]\teval-rmse:0.329236\ttrain-rmse:0.160886\n",
      "[819]\teval-rmse:0.329235\ttrain-rmse:0.160868\n",
      "[820]\teval-rmse:0.329228\ttrain-rmse:0.160787\n",
      "[821]\teval-rmse:0.329229\ttrain-rmse:0.160737\n",
      "[822]\teval-rmse:0.32923\ttrain-rmse:0.160618\n",
      "[823]\teval-rmse:0.329231\ttrain-rmse:0.160575\n",
      "[824]\teval-rmse:0.329237\ttrain-rmse:0.160505\n",
      "[825]\teval-rmse:0.329236\ttrain-rmse:0.160464\n",
      "[826]\teval-rmse:0.329236\ttrain-rmse:0.160394\n",
      "[827]\teval-rmse:0.329234\ttrain-rmse:0.160362\n",
      "[828]\teval-rmse:0.329235\ttrain-rmse:0.160298\n",
      "[829]\teval-rmse:0.329231\ttrain-rmse:0.160256\n",
      "[830]\teval-rmse:0.32923\ttrain-rmse:0.160146\n",
      "[831]\teval-rmse:0.329231\ttrain-rmse:0.160101\n",
      "[832]\teval-rmse:0.329231\ttrain-rmse:0.16002\n",
      "[833]\teval-rmse:0.329233\ttrain-rmse:0.159952\n",
      "[834]\teval-rmse:0.329235\ttrain-rmse:0.159888\n",
      "[835]\teval-rmse:0.329234\ttrain-rmse:0.159798\n",
      "[836]\teval-rmse:0.329228\ttrain-rmse:0.159725\n",
      "[837]\teval-rmse:0.32922\ttrain-rmse:0.159643\n",
      "[838]\teval-rmse:0.329219\ttrain-rmse:0.159624\n",
      "[839]\teval-rmse:0.329216\ttrain-rmse:0.15953\n",
      "[840]\teval-rmse:0.32922\ttrain-rmse:0.159405\n",
      "[841]\teval-rmse:0.32922\ttrain-rmse:0.159362\n",
      "[842]\teval-rmse:0.329214\ttrain-rmse:0.159257\n",
      "[843]\teval-rmse:0.329216\ttrain-rmse:0.159225\n",
      "[844]\teval-rmse:0.329209\ttrain-rmse:0.159045\n",
      "[845]\teval-rmse:0.329208\ttrain-rmse:0.158984\n",
      "[846]\teval-rmse:0.329204\ttrain-rmse:0.158952\n",
      "[847]\teval-rmse:0.329211\ttrain-rmse:0.158934\n",
      "[848]\teval-rmse:0.329211\ttrain-rmse:0.158859\n",
      "[849]\teval-rmse:0.32921\ttrain-rmse:0.158843\n",
      "[850]\teval-rmse:0.329211\ttrain-rmse:0.158783\n",
      "[851]\teval-rmse:0.329204\ttrain-rmse:0.158632\n",
      "[852]\teval-rmse:0.329206\ttrain-rmse:0.158614\n",
      "[853]\teval-rmse:0.329209\ttrain-rmse:0.158566\n",
      "[854]\teval-rmse:0.329207\ttrain-rmse:0.158477\n",
      "[855]\teval-rmse:0.329208\ttrain-rmse:0.158375\n",
      "[856]\teval-rmse:0.329208\ttrain-rmse:0.158277\n",
      "[857]\teval-rmse:0.329213\ttrain-rmse:0.158199\n",
      "[858]\teval-rmse:0.329213\ttrain-rmse:0.158182\n",
      "[859]\teval-rmse:0.329207\ttrain-rmse:0.158091\n",
      "[860]\teval-rmse:0.329209\ttrain-rmse:0.158012\n",
      "[861]\teval-rmse:0.329208\ttrain-rmse:0.157958\n",
      "[862]\teval-rmse:0.329211\ttrain-rmse:0.157876\n",
      "[863]\teval-rmse:0.329206\ttrain-rmse:0.157811\n",
      "[864]\teval-rmse:0.329203\ttrain-rmse:0.157765\n",
      "[865]\teval-rmse:0.329191\ttrain-rmse:0.157629\n",
      "[866]\teval-rmse:0.329176\ttrain-rmse:0.157397\n",
      "[867]\teval-rmse:0.329172\ttrain-rmse:0.157346\n",
      "[868]\teval-rmse:0.329171\ttrain-rmse:0.157303\n",
      "[869]\teval-rmse:0.329169\ttrain-rmse:0.157241\n",
      "[870]\teval-rmse:0.329168\ttrain-rmse:0.157214\n",
      "[871]\teval-rmse:0.329167\ttrain-rmse:0.157176\n",
      "[872]\teval-rmse:0.329167\ttrain-rmse:0.157136\n",
      "[873]\teval-rmse:0.329167\ttrain-rmse:0.15709\n",
      "[874]\teval-rmse:0.329167\ttrain-rmse:0.157011\n",
      "[875]\teval-rmse:0.329172\ttrain-rmse:0.156946\n",
      "[876]\teval-rmse:0.329171\ttrain-rmse:0.156914\n",
      "[877]\teval-rmse:0.329164\ttrain-rmse:0.156834\n",
      "[878]\teval-rmse:0.329163\ttrain-rmse:0.156768\n",
      "[879]\teval-rmse:0.329163\ttrain-rmse:0.156737\n",
      "[880]\teval-rmse:0.329167\ttrain-rmse:0.156662\n",
      "[881]\teval-rmse:0.329163\ttrain-rmse:0.156602\n",
      "[882]\teval-rmse:0.329153\ttrain-rmse:0.156474\n",
      "[883]\teval-rmse:0.329158\ttrain-rmse:0.156379\n",
      "[884]\teval-rmse:0.32915\ttrain-rmse:0.156286\n",
      "[885]\teval-rmse:0.329148\ttrain-rmse:0.156267\n",
      "[886]\teval-rmse:0.329146\ttrain-rmse:0.156237\n",
      "[887]\teval-rmse:0.329146\ttrain-rmse:0.156196\n",
      "[888]\teval-rmse:0.329139\ttrain-rmse:0.156139\n",
      "[889]\teval-rmse:0.329138\ttrain-rmse:0.156083\n",
      "[890]\teval-rmse:0.329133\ttrain-rmse:0.156014\n",
      "[891]\teval-rmse:0.329138\ttrain-rmse:0.155977\n",
      "[892]\teval-rmse:0.32913\ttrain-rmse:0.15593\n",
      "[893]\teval-rmse:0.32913\ttrain-rmse:0.155862\n",
      "[894]\teval-rmse:0.329133\ttrain-rmse:0.155822\n",
      "[895]\teval-rmse:0.329133\ttrain-rmse:0.155771\n",
      "[896]\teval-rmse:0.329133\ttrain-rmse:0.155728\n",
      "[897]\teval-rmse:0.329133\ttrain-rmse:0.155659\n",
      "[898]\teval-rmse:0.329134\ttrain-rmse:0.155632\n",
      "[899]\teval-rmse:0.329135\ttrain-rmse:0.15558\n",
      "[900]\teval-rmse:0.329137\ttrain-rmse:0.155515\n",
      "[901]\teval-rmse:0.329136\ttrain-rmse:0.155484\n",
      "[902]\teval-rmse:0.329132\ttrain-rmse:0.155352\n",
      "[903]\teval-rmse:0.329131\ttrain-rmse:0.155327\n",
      "[904]\teval-rmse:0.329128\ttrain-rmse:0.15524\n",
      "[905]\teval-rmse:0.329128\ttrain-rmse:0.15517\n",
      "[906]\teval-rmse:0.329123\ttrain-rmse:0.155123\n",
      "[907]\teval-rmse:0.329125\ttrain-rmse:0.155054\n",
      "[908]\teval-rmse:0.329125\ttrain-rmse:0.155008\n",
      "[909]\teval-rmse:0.329124\ttrain-rmse:0.154882\n",
      "[910]\teval-rmse:0.329126\ttrain-rmse:0.154834\n",
      "[911]\teval-rmse:0.329124\ttrain-rmse:0.154809\n",
      "[912]\teval-rmse:0.329124\ttrain-rmse:0.154745\n",
      "[913]\teval-rmse:0.32912\ttrain-rmse:0.154639\n",
      "[914]\teval-rmse:0.329121\ttrain-rmse:0.154611\n",
      "[915]\teval-rmse:0.32912\ttrain-rmse:0.154565\n",
      "[916]\teval-rmse:0.329117\ttrain-rmse:0.154506\n",
      "[917]\teval-rmse:0.329114\ttrain-rmse:0.154444\n",
      "[918]\teval-rmse:0.329106\ttrain-rmse:0.154319\n",
      "[919]\teval-rmse:0.329108\ttrain-rmse:0.154269\n",
      "[920]\teval-rmse:0.329106\ttrain-rmse:0.154225\n",
      "[921]\teval-rmse:0.329105\ttrain-rmse:0.154195\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[922]\teval-rmse:0.329105\ttrain-rmse:0.154125\n",
      "[923]\teval-rmse:0.329106\ttrain-rmse:0.154083\n",
      "[924]\teval-rmse:0.329106\ttrain-rmse:0.154037\n",
      "[925]\teval-rmse:0.329109\ttrain-rmse:0.153934\n",
      "[926]\teval-rmse:0.329112\ttrain-rmse:0.153884\n",
      "[927]\teval-rmse:0.329113\ttrain-rmse:0.153793\n",
      "[928]\teval-rmse:0.329112\ttrain-rmse:0.153743\n",
      "[929]\teval-rmse:0.329112\ttrain-rmse:0.153711\n",
      "[930]\teval-rmse:0.329111\ttrain-rmse:0.153648\n",
      "[931]\teval-rmse:0.329111\ttrain-rmse:0.153591\n",
      "[932]\teval-rmse:0.329115\ttrain-rmse:0.153554\n",
      "[933]\teval-rmse:0.329113\ttrain-rmse:0.153526\n",
      "[934]\teval-rmse:0.32911\ttrain-rmse:0.153405\n",
      "[935]\teval-rmse:0.329111\ttrain-rmse:0.15338\n",
      "[936]\teval-rmse:0.329111\ttrain-rmse:0.153373\n",
      "[937]\teval-rmse:0.329112\ttrain-rmse:0.15335\n",
      "[938]\teval-rmse:0.329109\ttrain-rmse:0.153257\n",
      "[939]\teval-rmse:0.329106\ttrain-rmse:0.153204\n",
      "[940]\teval-rmse:0.329102\ttrain-rmse:0.153119\n",
      "[941]\teval-rmse:0.329105\ttrain-rmse:0.153065\n",
      "[942]\teval-rmse:0.329105\ttrain-rmse:0.153005\n",
      "[943]\teval-rmse:0.329101\ttrain-rmse:0.152919\n",
      "[944]\teval-rmse:0.329103\ttrain-rmse:0.152868\n",
      "[945]\teval-rmse:0.329102\ttrain-rmse:0.152743\n",
      "[946]\teval-rmse:0.329101\ttrain-rmse:0.152674\n",
      "[947]\teval-rmse:0.329106\ttrain-rmse:0.152516\n",
      "[948]\teval-rmse:0.329098\ttrain-rmse:0.152405\n",
      "[949]\teval-rmse:0.329098\ttrain-rmse:0.152396\n",
      "[950]\teval-rmse:0.329099\ttrain-rmse:0.152354\n",
      "[951]\teval-rmse:0.329101\ttrain-rmse:0.152313\n",
      "[952]\teval-rmse:0.3291\ttrain-rmse:0.152283\n",
      "[953]\teval-rmse:0.3291\ttrain-rmse:0.15223\n",
      "[954]\teval-rmse:0.329095\ttrain-rmse:0.152194\n",
      "[955]\teval-rmse:0.329093\ttrain-rmse:0.152153\n",
      "[956]\teval-rmse:0.329093\ttrain-rmse:0.151967\n",
      "[957]\teval-rmse:0.329095\ttrain-rmse:0.151907\n",
      "[958]\teval-rmse:0.329095\ttrain-rmse:0.151858\n",
      "[959]\teval-rmse:0.329075\ttrain-rmse:0.151682\n",
      "[960]\teval-rmse:0.329075\ttrain-rmse:0.151656\n",
      "[961]\teval-rmse:0.32907\ttrain-rmse:0.151604\n",
      "[962]\teval-rmse:0.329069\ttrain-rmse:0.151535\n",
      "[963]\teval-rmse:0.329066\ttrain-rmse:0.15147\n",
      "[964]\teval-rmse:0.329063\ttrain-rmse:0.151423\n",
      "[965]\teval-rmse:0.329059\ttrain-rmse:0.151385\n",
      "[966]\teval-rmse:0.32906\ttrain-rmse:0.151351\n",
      "[967]\teval-rmse:0.32906\ttrain-rmse:0.151304\n",
      "[968]\teval-rmse:0.329059\ttrain-rmse:0.151261\n",
      "[969]\teval-rmse:0.329059\ttrain-rmse:0.151209\n",
      "[970]\teval-rmse:0.329057\ttrain-rmse:0.151147\n",
      "[971]\teval-rmse:0.329057\ttrain-rmse:0.151082\n",
      "[972]\teval-rmse:0.329055\ttrain-rmse:0.151043\n",
      "[973]\teval-rmse:0.329059\ttrain-rmse:0.150965\n",
      "[974]\teval-rmse:0.32906\ttrain-rmse:0.150898\n",
      "[975]\teval-rmse:0.329062\ttrain-rmse:0.150857\n",
      "[976]\teval-rmse:0.329062\ttrain-rmse:0.15079\n",
      "[977]\teval-rmse:0.329059\ttrain-rmse:0.150713\n",
      "[978]\teval-rmse:0.32906\ttrain-rmse:0.150693\n",
      "[979]\teval-rmse:0.32906\ttrain-rmse:0.150613\n",
      "[980]\teval-rmse:0.329059\ttrain-rmse:0.150577\n",
      "[981]\teval-rmse:0.329058\ttrain-rmse:0.150567\n",
      "[982]\teval-rmse:0.329057\ttrain-rmse:0.150536\n",
      "[983]\teval-rmse:0.329057\ttrain-rmse:0.150492\n",
      "[984]\teval-rmse:0.329057\ttrain-rmse:0.150463\n",
      "[985]\teval-rmse:0.329058\ttrain-rmse:0.150427\n",
      "[986]\teval-rmse:0.329058\ttrain-rmse:0.150408\n",
      "[987]\teval-rmse:0.329059\ttrain-rmse:0.150362\n",
      "[988]\teval-rmse:0.32906\ttrain-rmse:0.150331\n",
      "[989]\teval-rmse:0.329062\ttrain-rmse:0.150303\n",
      "[990]\teval-rmse:0.329053\ttrain-rmse:0.150221\n",
      "[991]\teval-rmse:0.329055\ttrain-rmse:0.150199\n",
      "[992]\teval-rmse:0.329056\ttrain-rmse:0.15013\n",
      "[993]\teval-rmse:0.329056\ttrain-rmse:0.150097\n",
      "[994]\teval-rmse:0.329055\ttrain-rmse:0.150009\n",
      "[995]\teval-rmse:0.329055\ttrain-rmse:0.149965\n",
      "[996]\teval-rmse:0.329054\ttrain-rmse:0.149908\n",
      "[997]\teval-rmse:0.329051\ttrain-rmse:0.149866\n",
      "[998]\teval-rmse:0.329052\ttrain-rmse:0.14978\n",
      "[999]\teval-rmse:0.329051\ttrain-rmse:0.149689\n"
     ]
    }
   ],
   "source": [
    "# Number of training rounds\n",
    "nrounds = 1000\n",
    "\n",
    "# Train model\n",
    "gbm = xgb.train(params,\n",
    "                dtrain,\n",
    "                num_boost_round = nrounds,\n",
    "                evals = watchlist,\n",
    "                verbose_eval = True\n",
    "                )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Test predictions\n",
    "y_pred = np.exp(gbm.predict(xgb.DMatrix(X_test))) - 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5.0934283983094115"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Use mean absolute error to get a basic estimate of the error\n",
    "mae = (abs(y_pred - y_test)).mean()\n",
    "mae"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'dropoff_latitude': 408961,\n",
       " 'dropoff_longitude': 398115,\n",
       " 'latitude_difference': 321800,\n",
       " 'longitude_difference': 315388,\n",
       " 'passenger_count': 138235,\n",
       " 'pickup_day': 246727,\n",
       " 'pickup_hour': 186841,\n",
       " 'pickup_latitude': 466207,\n",
       " 'pickup_longitude': 499024,\n",
       " 'pickup_minute': 247127,\n",
       " 'pickup_month': 132498,\n",
       " 'pickup_weekday': 125545,\n",
       " 'store_and_fwd_flag': 3881,\n",
       " 'trip_distance': 279769}"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Take a look at feature importance\n",
    "feature_scores = gbm.get_fscore()\n",
    "feature_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'dropoff_latitude': 0.10847432361533511,\n",
       " 'dropoff_longitude': 0.10559749058252288,\n",
       " 'latitude_difference': 0.08535541858371541,\n",
       " 'longitude_difference': 0.0836546760605371,\n",
       " 'passenger_count': 0.03666596111845836,\n",
       " 'pickup_day': 0.06544277924457537,\n",
       " 'pickup_hour': 0.04955839578495952,\n",
       " 'pickup_latitude': 0.12365846374039222,\n",
       " 'pickup_longitude': 0.1323629658275948,\n",
       " 'pickup_minute': 0.06554887671950851,\n",
       " 'pickup_month': 0.035144258084229725,\n",
       " 'pickup_weekday': 0.033300018726204327,\n",
       " 'store_and_fwd_flag': 0.0010294107505388426,\n",
       " 'trip_distance': 0.07420696116142785}"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# This is not very telling, so let's scale the features\n",
    "summ = 0\n",
    "for key in feature_scores:\n",
    "    summ = summ + feature_scores[key]\n",
    "\n",
    "for key in feature_scores:\n",
    "    feature_scores[key] = feature_scores[key] / summ\n",
    "\n",
    "feature_scores"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Saved"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "filename = \"xgb_model.sav\"\n",
    "pickle.dump(gbm, open(filename, 'wb'))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}