notebook/model_training.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# MODEL TRAINING\n",
    "\n",
    "- Import relevant libraries\n",
    "- Define features from EDA\n",
    "- Import the data\n",
    "- Obtain a suitable split for the data\n",
    "- Resample the data\n",
    "- Preprocessing pipelines for all samples\n",
    "    - Check for rare values before implementing pipelines.\n",
    "    - Encode categorical features\n",
    "    - Handle outliers\n",
    "    - Perform scaling\n",
    "- Feature selection \n",
    "    - Categorical feature selection for various sampled data\n",
    "    - Numerical feature selection for various sampled data\n",
    "- Model selection and optimization\n",
    "    - Obtain the best model and hyperparameter from the src folder"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.0) Import relevant libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "#for data manipulation\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "#for visualization \n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "#for data splitting\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "#for feature selection\n",
    "from sklearn.feature_selection import SelectKBest, f_classif, chi2, mutual_info_classif, SelectFromModel\n",
    "\n",
    "#for categorical encoding\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "\n",
    "#for scaling\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "#for model evaluation \n",
    "from sklearn.metrics import f1_score, accuracy_score, confusion_matrix, classification_report\n",
    "\n",
    "#predictive algorithms\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier, GradientBoostingClassifier\n",
    "from xgboost import XGBClassifier\n",
    "from catboost import CatBoostClassifier\n",
    "\n",
    "#sampling techniques\n",
    "from imblearn.over_sampling import RandomOverSampler\n",
    "from imblearn.under_sampling import RandomUnderSampler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.set_option('mode.chained_assignment', None)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.0) Define features from EDA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "initial_features = ['policy_tenure', 'age_of_car', 'age_of_policyholder', 'population_density', 'make', 'airbags', 'displacement', 'cylinder', 'gear_box', 'turning_radius', 'length', 'height', 'gross_weight', 'ncap_rating',\n",
    "                    'is_esc', 'is_adjustable_steering', 'is_tpms', 'is_parking_sensors', 'is_parking_camera', 'rear_brakes_type', 'transmission_type', 'is_front_fog_lights', 'is_rear_window_wiper', 'is_rear_window_defogger',\n",
    "                    'is_brake_assist', 'is_central_locking', 'is_driver_seat_height_adjustable', 'is_day_night_rear_view_mirror', 'is_ecw', 'segment', 'fuel_type', 'max_torque', 'steering_type']\n",
    "\n",
    "scaled_features = ['policy_tenure', 'age_of_car', 'age_of_policyholder']\n",
    "\n",
    "categorical_features = ['is_esc', 'is_adjustable_steering', 'is_tpms', 'is_parking_sensors', 'is_parking_camera', 'rear_brakes_type', 'transmission_type', 'is_front_fog_lights', 'is_rear_window_wiper', \n",
    "                        'is_rear_window_defogger', 'is_brake_assist', 'is_central_locking', 'is_driver_seat_height_adjustable', 'is_day_night_rear_view_mirror', 'is_ecw', 'segment', 'fuel_type', \n",
    "                        'max_torque', 'steering_type']\n",
    "\n",
    "lo_card = ['is_esc', 'is_adjustable_steering', 'is_tpms', 'is_parking_sensors', 'is_parking_camera', 'rear_brakes_type', 'transmission_type', 'is_front_fog_lights', 'is_rear_window_wiper', 'is_rear_window_defogger', \n",
    "           'is_brake_assist', 'is_central_locking', 'is_driver_seat_height_adjustable', 'is_day_night_rear_view_mirror', 'is_ecw']\n",
    "\n",
    "hi_card = ['segment', 'fuel_type', 'max_torque', 'steering_type']\n",
    "\n",
    "rare_features = ['segment', 'max_torque', 'steering_type']\n",
    "\n",
    "unscaled_features = initial_features[3:14]\n",
    "\n",
    "numerical_features = scaled_features + unscaled_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['policy_tenure',\n",
       " 'age_of_car',\n",
       " 'age_of_policyholder',\n",
       " 'population_density',\n",
       " 'make',\n",
       " 'airbags',\n",
       " 'displacement',\n",
       " 'cylinder',\n",
       " 'gear_box',\n",
       " 'turning_radius',\n",
       " 'length',\n",
       " 'height',\n",
       " 'gross_weight',\n",
       " 'ncap_rating']"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "numerical_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#confirm \n",
    "len(initial_features) == len(numerical_features) + len(categorical_features)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.0) Import the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv('dataset/data.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#set the predictors[x] and the target[y]\n",
    "x = data[initial_features]\n",
    "y = data.is_claim"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.0) Split the data\n",
    "\n",
    "- Apply split to data ingestion script."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.05, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>policy_tenure</th>\n",
       "      <th>age_of_car</th>\n",
       "      <th>age_of_policyholder</th>\n",
       "      <th>population_density</th>\n",
       "      <th>make</th>\n",
       "      <th>airbags</th>\n",
       "      <th>displacement</th>\n",
       "      <th>cylinder</th>\n",
       "      <th>gear_box</th>\n",
       "      <th>turning_radius</th>\n",
       "      <th>...</th>\n",
       "      <th>is_rear_window_defogger</th>\n",
       "      <th>is_brake_assist</th>\n",
       "      <th>is_central_locking</th>\n",
       "      <th>is_driver_seat_height_adjustable</th>\n",
       "      <th>is_day_night_rear_view_mirror</th>\n",
       "      <th>is_ecw</th>\n",
       "      <th>segment</th>\n",
       "      <th>fuel_type</th>\n",
       "      <th>max_torque</th>\n",
       "      <th>steering_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>39542</th>\n",
       "      <td>0.998082</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.490385</td>\n",
       "      <td>16206</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>1493</td>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>5.2</td>\n",
       "      <td>...</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>C2</td>\n",
       "      <td>Diesel</td>\n",
       "      <td>250Nm@2750rpm</td>\n",
       "      <td>Power</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11589</th>\n",
       "      <td>0.782329</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.490385</td>\n",
       "      <td>17804</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>796</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>4.6</td>\n",
       "      <td>...</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>A</td>\n",
       "      <td>CNG</td>\n",
       "      <td>60Nm@3500rpm</td>\n",
       "      <td>Power</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21337</th>\n",
       "      <td>0.713391</td>\n",
       "      <td>0.10</td>\n",
       "      <td>0.375000</td>\n",
       "      <td>8794</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1197</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>4.8</td>\n",
       "      <td>...</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>B2</td>\n",
       "      <td>Petrol</td>\n",
       "      <td>113Nm@4400rpm</td>\n",
       "      <td>Electric</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56297</th>\n",
       "      <td>0.339513</td>\n",
       "      <td>0.03</td>\n",
       "      <td>0.480769</td>\n",
       "      <td>6108</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>796</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>4.6</td>\n",
       "      <td>...</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>A</td>\n",
       "      <td>CNG</td>\n",
       "      <td>60Nm@3500rpm</td>\n",
       "      <td>Power</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25297</th>\n",
       "      <td>0.214826</td>\n",
       "      <td>0.03</td>\n",
       "      <td>0.538462</td>\n",
       "      <td>6112</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>796</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>4.6</td>\n",
       "      <td>...</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>A</td>\n",
       "      <td>CNG</td>\n",
       "      <td>60Nm@3500rpm</td>\n",
       "      <td>Power</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       policy_tenure  age_of_car  age_of_policyholder  population_density  \\\n",
       "39542       0.998082        0.15             0.490385               16206   \n",
       "11589       0.782329        0.00             0.490385               17804   \n",
       "21337       0.713391        0.10             0.375000                8794   \n",
       "56297       0.339513        0.03             0.480769                6108   \n",
       "25297       0.214826        0.03             0.538462                6112   \n",
       "\n",
       "       make  airbags  displacement  cylinder  gear_box  turning_radius  ...  \\\n",
       "39542     3        6          1493         4         6             5.2  ...   \n",
       "11589     1        2           796         3         5             4.6  ...   \n",
       "21337     1        2          1197         4         5             4.8  ...   \n",
       "56297     1        2           796         3         5             4.6  ...   \n",
       "25297     1        2           796         3         5             4.6  ...   \n",
       "\n",
       "       is_rear_window_defogger  is_brake_assist  is_central_locking  \\\n",
       "39542                      Yes              Yes                 Yes   \n",
       "11589                       No               No                  No   \n",
       "21337                       No              Yes                 Yes   \n",
       "56297                       No               No                  No   \n",
       "25297                       No               No                  No   \n",
       "\n",
       "       is_driver_seat_height_adjustable is_day_night_rear_view_mirror is_ecw  \\\n",
       "39542                               Yes                            No    Yes   \n",
       "11589                                No                            No     No   \n",
       "21337                               Yes                           Yes    Yes   \n",
       "56297                                No                            No     No   \n",
       "25297                                No                            No     No   \n",
       "\n",
       "      segment fuel_type     max_torque steering_type  \n",
       "39542      C2    Diesel  250Nm@2750rpm         Power  \n",
       "11589       A       CNG   60Nm@3500rpm         Power  \n",
       "21337      B2    Petrol  113Nm@4400rpm      Electric  \n",
       "56297       A       CNG   60Nm@3500rpm         Power  \n",
       "25297       A       CNG   60Nm@3500rpm         Power  \n",
       "\n",
       "[5 rows x 33 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.0) Resample the data\n",
    "\n",
    "- Implement the RandomUnderSampler and RandomOverSampler techniques\n",
    "- Determine the best sample set and implement"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "#for undersampling\n",
    "RUS = RandomUnderSampler(random_state=45)\n",
    "x_rus, y_rus = RUS.fit_resample(x_train, y_train)\n",
    "\n",
    "#for oversampling\n",
    "ROS = RandomOverSampler(random_state=45)\n",
    "x_ros, y_ros = ROS.fit_resample(x_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The unsampled set has 55662 samples and 33 features.\n",
      "The undersampled set has 7128 samples and 33 features.\n",
      "The oversampled set has 104196 samples and 33 features.\n"
     ]
    }
   ],
   "source": [
    "print('The unsampled set has {} samples and {} features.'.format(x_train.shape[0], x_train.shape[1]))\n",
    "print('The undersampled set has {} samples and {} features.'.format(x_rus.shape[0], x_rus.shape[1]))\n",
    "print('The oversampled set has {} samples and {} features.'.format(x_ros.shape[0], x_ros.shape[1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 6.0) Preprocessing pipelines\n",
    "\n",
    "- Confirm rare features are commmon in various samples.\n",
    "- Generate class to handle outliers in the preprocessing pipeline\n",
    "- Handle outliers for the already scaled data\n",
    "- Generate numerical and categorical preprocessing pipelines"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.pipeline import Pipeline, make_pipeline\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.base import TransformerMixin, BaseEstimator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import set_config\n",
    "set_config(display='diagram')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import OrdinalEncoder"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6.1) Rare features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['segment', 'max_torque', 'steering_type']"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rare_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "segment          True\n",
       "max_torque       True\n",
       "steering_type    True\n",
       "dtype: bool"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_test[rare_features].nunique() == x_train[rare_features].nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "segment          True\n",
       "max_torque       True\n",
       "steering_type    True\n",
       "dtype: bool"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_rus[rare_features].nunique() == x_train[rare_features].nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "***Insight***\n",
    "\n",
    "- The rare features from the data analysis notebook occur in both the undersampled and test splits.\n",
    "- There will be no need for rare value imputation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6.2) Outlier imputation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "#generate function for capping\n",
    "def outlier_cap(df, col, a, b=0):\n",
    "    df.loc[df[col] > a, col] = a\n",
    "    df.loc[df[col] < b, col] = b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "#display features with outliers and their respective upper and lower limits\n",
    "out_dict = {\n",
    "    'age_of_car' : [0.245, -0.115],\n",
    "    'age_of_policyholder' : [0.822, 0.091],\n",
    "    'population_density' : [58399.5, -25244.5]\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`age_of_car`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "#for unsampled set\n",
    "outlier_cap(x_train, 'age_of_car', 0.245)\n",
    "\n",
    "#for undersampled set\n",
    "outlier_cap(x_rus, 'age_of_car', 0.245)\n",
    "\n",
    "#for oversampled set\n",
    "outlier_cap(x_ros, 'age_of_car', 0.245)\n",
    "\n",
    "#for test set\n",
    "outlier_cap(x_test, 'age_of_car', 0.245)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`age_of_policyholder`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "#for unsampled set\n",
    "outlier_cap(x_train, 'age_of_policyholder', 0.822, 0.091)\n",
    "\n",
    "#for undersampled set\n",
    "outlier_cap(x_rus, 'age_of_policyholder', 0.822, 0.091)\n",
    "\n",
    "#for oversampled set\n",
    "outlier_cap(x_ros, 'age_of_policyholder', 0.822, 0.091)\n",
    "\n",
    "#for test set\n",
    "outlier_cap(x_test, 'age_of_policyholder', 0.822, 0.091)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`population_density`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "#for unsampled set\n",
    "outlier_cap(x_train, 'population_density', 58399.5)\n",
    "\n",
    "#for undersampled set\n",
    "outlier_cap(x_rus, 'population_density', 58399.5)\n",
    "\n",
    "#for oversampled set\n",
    "outlier_cap(x_ros, 'population_density', 58399.5)\n",
    "\n",
    "#for test set\n",
    "outlier_cap(x_test, 'population_density', 58399.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "58399.5"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#confirm\n",
    "x_ros.population_density.max()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6.3) Generate preprocessing pipelines"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_preprocessor = Pipeline(\n",
    "    steps=[\n",
    "           ('scaler', MinMaxScaler(feature_range=(0,1)))\n",
    "           ]\n",
    ")\n",
    "\n",
    "cat_preprocessor = Pipeline(\n",
    "    steps=[\n",
    "        ('encoder', OrdinalEncoder()),\n",
    "        ('scale', MinMaxScaler(feature_range=(0,1)))\n",
    "    ]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "preprocessor  = ColumnTransformer([\n",
    "    ('num', num_preprocessor, unscaled_features),\n",
    "    ('cat', cat_preprocessor, categorical_features),\n",
    "], remainder='passthrough'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "preprocessing_pipeline = make_pipeline(\n",
    "    preprocessor\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[(&#x27;columntransformer&#x27;,\n",
       "                 ColumnTransformer(remainder=&#x27;passthrough&#x27;,\n",
       "                                   transformers=[(&#x27;num&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;scaler&#x27;,\n",
       "                                                                   MinMaxScaler())]),\n",
       "                                                  [&#x27;population_density&#x27;, &#x27;make&#x27;,\n",
       "                                                   &#x27;airbags&#x27;, &#x27;displacement&#x27;,\n",
       "                                                   &#x27;cylinder&#x27;, &#x27;gear_box&#x27;,\n",
       "                                                   &#x27;turning_radius&#x27;, &#x27;length&#x27;,\n",
       "                                                   &#x27;height&#x27;, &#x27;gross_weight&#x27;,\n",
       "                                                   &#x27;ncap_rating&#x27;]),\n",
       "                                                 (&#x27;cat&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;encoder&#x27;,\n",
       "                                                                   OrdinalEncoder()),\n",
       "                                                                  (&#x27;scale&#x27;,\n",
       "                                                                   Min...\n",
       "                                                   &#x27;is_adjustable_steering&#x27;,\n",
       "                                                   &#x27;is_tpms&#x27;,\n",
       "                                                   &#x27;is_parking_sensors&#x27;,\n",
       "                                                   &#x27;is_parking_camera&#x27;,\n",
       "                                                   &#x27;rear_brakes_type&#x27;,\n",
       "                                                   &#x27;transmission_type&#x27;,\n",
       "                                                   &#x27;is_front_fog_lights&#x27;,\n",
       "                                                   &#x27;is_rear_window_wiper&#x27;,\n",
       "                                                   &#x27;is_rear_window_defogger&#x27;,\n",
       "                                                   &#x27;is_brake_assist&#x27;,\n",
       "                                                   &#x27;is_central_locking&#x27;,\n",
       "                                                   &#x27;is_driver_seat_height_adjustable&#x27;,\n",
       "                                                   &#x27;is_day_night_rear_view_mirror&#x27;,\n",
       "                                                   &#x27;is_ecw&#x27;, &#x27;segment&#x27;,\n",
       "                                                   &#x27;fuel_type&#x27;, &#x27;max_torque&#x27;,\n",
       "                                                   &#x27;steering_type&#x27;])]))])</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" ><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;columntransformer&#x27;,\n",
       "                 ColumnTransformer(remainder=&#x27;passthrough&#x27;,\n",
       "                                   transformers=[(&#x27;num&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;scaler&#x27;,\n",
       "                                                                   MinMaxScaler())]),\n",
       "                                                  [&#x27;population_density&#x27;, &#x27;make&#x27;,\n",
       "                                                   &#x27;airbags&#x27;, &#x27;displacement&#x27;,\n",
       "                                                   &#x27;cylinder&#x27;, &#x27;gear_box&#x27;,\n",
       "                                                   &#x27;turning_radius&#x27;, &#x27;length&#x27;,\n",
       "                                                   &#x27;height&#x27;, &#x27;gross_weight&#x27;,\n",
       "                                                   &#x27;ncap_rating&#x27;]),\n",
       "                                                 (&#x27;cat&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;encoder&#x27;,\n",
       "                                                                   OrdinalEncoder()),\n",
       "                                                                  (&#x27;scale&#x27;,\n",
       "                                                                   Min...\n",
       "                                                   &#x27;is_adjustable_steering&#x27;,\n",
       "                                                   &#x27;is_tpms&#x27;,\n",
       "                                                   &#x27;is_parking_sensors&#x27;,\n",
       "                                                   &#x27;is_parking_camera&#x27;,\n",
       "                                                   &#x27;rear_brakes_type&#x27;,\n",
       "                                                   &#x27;transmission_type&#x27;,\n",
       "                                                   &#x27;is_front_fog_lights&#x27;,\n",
       "                                                   &#x27;is_rear_window_wiper&#x27;,\n",
       "                                                   &#x27;is_rear_window_defogger&#x27;,\n",
       "                                                   &#x27;is_brake_assist&#x27;,\n",
       "                                                   &#x27;is_central_locking&#x27;,\n",
       "                                                   &#x27;is_driver_seat_height_adjustable&#x27;,\n",
       "                                                   &#x27;is_day_night_rear_view_mirror&#x27;,\n",
       "                                                   &#x27;is_ecw&#x27;, &#x27;segment&#x27;,\n",
       "                                                   &#x27;fuel_type&#x27;, &#x27;max_torque&#x27;,\n",
       "                                                   &#x27;steering_type&#x27;])]))])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" ><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">columntransformer: ColumnTransformer</label><div class=\"sk-toggleable__content\"><pre>ColumnTransformer(remainder=&#x27;passthrough&#x27;,\n",
       "                  transformers=[(&#x27;num&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;scaler&#x27;, MinMaxScaler())]),\n",
       "                                 [&#x27;population_density&#x27;, &#x27;make&#x27;, &#x27;airbags&#x27;,\n",
       "                                  &#x27;displacement&#x27;, &#x27;cylinder&#x27;, &#x27;gear_box&#x27;,\n",
       "                                  &#x27;turning_radius&#x27;, &#x27;length&#x27;, &#x27;height&#x27;,\n",
       "                                  &#x27;gross_weight&#x27;, &#x27;ncap_rating&#x27;]),\n",
       "                                (&#x27;cat&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;encoder&#x27;, OrdinalEncoder()),\n",
       "                                                 (&#x27;scale&#x27;, MinMaxScaler())]),\n",
       "                                 [&#x27;is_esc&#x27;, &#x27;is_adjustable_steering&#x27;, &#x27;is_tpms&#x27;,\n",
       "                                  &#x27;is_parking_sensors&#x27;, &#x27;is_parking_camera&#x27;,\n",
       "                                  &#x27;rear_brakes_type&#x27;, &#x27;transmission_type&#x27;,\n",
       "                                  &#x27;is_front_fog_lights&#x27;, &#x27;is_rear_window_wiper&#x27;,\n",
       "                                  &#x27;is_rear_window_defogger&#x27;, &#x27;is_brake_assist&#x27;,\n",
       "                                  &#x27;is_central_locking&#x27;,\n",
       "                                  &#x27;is_driver_seat_height_adjustable&#x27;,\n",
       "                                  &#x27;is_day_night_rear_view_mirror&#x27;, &#x27;is_ecw&#x27;,\n",
       "                                  &#x27;segment&#x27;, &#x27;fuel_type&#x27;, &#x27;max_torque&#x27;,\n",
       "                                  &#x27;steering_type&#x27;])])</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" ><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">num</label><div class=\"sk-toggleable__content\"><pre>[&#x27;population_density&#x27;, &#x27;make&#x27;, &#x27;airbags&#x27;, &#x27;displacement&#x27;, &#x27;cylinder&#x27;, &#x27;gear_box&#x27;, &#x27;turning_radius&#x27;, &#x27;length&#x27;, &#x27;height&#x27;, &#x27;gross_weight&#x27;, &#x27;ncap_rating&#x27;]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-4\" type=\"checkbox\" ><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">MinMaxScaler</label><div class=\"sk-toggleable__content\"><pre>MinMaxScaler()</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-5\" type=\"checkbox\" ><label for=\"sk-estimator-id-5\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">cat</label><div class=\"sk-toggleable__content\"><pre>[&#x27;is_esc&#x27;, &#x27;is_adjustable_steering&#x27;, &#x27;is_tpms&#x27;, &#x27;is_parking_sensors&#x27;, &#x27;is_parking_camera&#x27;, &#x27;rear_brakes_type&#x27;, &#x27;transmission_type&#x27;, &#x27;is_front_fog_lights&#x27;, &#x27;is_rear_window_wiper&#x27;, &#x27;is_rear_window_defogger&#x27;, &#x27;is_brake_assist&#x27;, &#x27;is_central_locking&#x27;, &#x27;is_driver_seat_height_adjustable&#x27;, &#x27;is_day_night_rear_view_mirror&#x27;, &#x27;is_ecw&#x27;, &#x27;segment&#x27;, &#x27;fuel_type&#x27;, &#x27;max_torque&#x27;, &#x27;steering_type&#x27;]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-6\" type=\"checkbox\" ><label for=\"sk-estimator-id-6\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">OrdinalEncoder</label><div class=\"sk-toggleable__content\"><pre>OrdinalEncoder()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-7\" type=\"checkbox\" ><label for=\"sk-estimator-id-7\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">MinMaxScaler</label><div class=\"sk-toggleable__content\"><pre>MinMaxScaler()</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-8\" type=\"checkbox\" ><label for=\"sk-estimator-id-8\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">remainder</label><div class=\"sk-toggleable__content\"><pre></pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-9\" type=\"checkbox\" ><label for=\"sk-estimator-id-9\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">passthrough</label><div class=\"sk-toggleable__content\"><pre>passthrough</pre></div></div></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "Pipeline(steps=[('columntransformer',\n",
       "                 ColumnTransformer(remainder='passthrough',\n",
       "                                   transformers=[('num',\n",
       "                                                  Pipeline(steps=[('scaler',\n",
       "                                                                   MinMaxScaler())]),\n",
       "                                                  ['population_density', 'make',\n",
       "                                                   'airbags', 'displacement',\n",
       "                                                   'cylinder', 'gear_box',\n",
       "                                                   'turning_radius', 'length',\n",
       "                                                   'height', 'gross_weight',\n",
       "                                                   'ncap_rating']),\n",
       "                                                 ('cat',\n",
       "                                                  Pipeline(steps=[('encoder',\n",
       "                                                                   OrdinalEncoder()),\n",
       "                                                                  ('scale',\n",
       "                                                                   Min...\n",
       "                                                   'is_adjustable_steering',\n",
       "                                                   'is_tpms',\n",
       "                                                   'is_parking_sensors',\n",
       "                                                   'is_parking_camera',\n",
       "                                                   'rear_brakes_type',\n",
       "                                                   'transmission_type',\n",
       "                                                   'is_front_fog_lights',\n",
       "                                                   'is_rear_window_wiper',\n",
       "                                                   'is_rear_window_defogger',\n",
       "                                                   'is_brake_assist',\n",
       "                                                   'is_central_locking',\n",
       "                                                   'is_driver_seat_height_adjustable',\n",
       "                                                   'is_day_night_rear_view_mirror',\n",
       "                                                   'is_ecw', 'segment',\n",
       "                                                   'fuel_type', 'max_torque',\n",
       "                                                   'steering_type'])]))])"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "preprocessing_pipeline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "#generate a new train set with the pipeline\n",
    "new_train = pd.DataFrame(preprocessing_pipeline.fit_transform(x_train, y_train))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.273897</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.992877</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.457143</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.8</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.500</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.998082</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.490385</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.301397</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.200299</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.625</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.782329</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.490385</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.146344</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.571225</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.428571</td>\n",
       "      <td>0.467836</td>\n",
       "      <td>0.157143</td>\n",
       "      <td>0.424514</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.713391</td>\n",
       "      <td>0.10</td>\n",
       "      <td>0.375000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.100121</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.200299</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.625</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.339513</td>\n",
       "      <td>0.03</td>\n",
       "      <td>0.480769</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.100190</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.200299</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.625</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.214826</td>\n",
       "      <td>0.03</td>\n",
       "      <td>0.538462</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         0    1    2         3    4    5         6         7         8   \\\n",
       "0  0.273897  0.5  1.0  0.992877  1.0  1.0  1.000000  1.000000  0.457143   \n",
       "1  0.301397  0.0  0.2  0.000000  0.0  0.0  0.142857  0.000000  0.000000   \n",
       "2  0.146344  0.0  0.2  0.571225  1.0  0.0  0.428571  0.467836  0.157143   \n",
       "3  0.100121  0.0  0.2  0.000000  0.0  0.0  0.142857  0.000000  0.000000   \n",
       "4  0.100190  0.0  0.2  0.000000  0.0  0.0  0.142857  0.000000  0.000000   \n",
       "\n",
       "         9   ...   23   24   25   26   27     28   29        30    31  \\\n",
       "0  1.000000  ...  1.0  0.0  1.0  0.8  0.5  0.500  1.0  0.998082  0.15   \n",
       "1  0.200299  ...  0.0  0.0  0.0  0.0  0.0  0.625  1.0  0.782329  0.00   \n",
       "2  0.424514  ...  1.0  1.0  1.0  0.4  1.0  0.000  0.0  0.713391  0.10   \n",
       "3  0.200299  ...  0.0  0.0  0.0  0.0  0.0  0.625  1.0  0.339513  0.03   \n",
       "4  0.200299  ...  0.0  0.0  0.0  0.0  0.0  0.625  1.0  0.214826  0.03   \n",
       "\n",
       "         32  \n",
       "0  0.490385  \n",
       "1  0.490385  \n",
       "2  0.375000  \n",
       "3  0.480769  \n",
       "4  0.538462  \n",
       "\n",
       "[5 rows x 33 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "pipeline_features = ['population_density', 'make', 'airbags', 'displacement', 'cylinder', 'gear_box', 'turning_radius', 'length', 'height', 'gross_weight', 'ncap_rating', 'is_esc', \n",
    "                    'is_adjustable_steering', 'is_tpms', 'is_parking_sensors', 'is_parking_camera', 'rear_brakes_type', 'transmission_type', 'is_front_fog_lights', 'is_rear_window_wiper', \n",
    "                    'is_rear_window_defogger', 'is_brake_assist', 'is_central_locking', 'is_driver_seat_height_adjustable', 'is_day_night_rear_view_mirror', 'is_ecw', 'segment', 'fuel_type', \n",
    "                    'max_torque', 'steering_type', 'policy_tenure', 'age_of_car', 'age_of_policyholder']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "new_train.columns = pipeline_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>population_density</th>\n",
       "      <th>make</th>\n",
       "      <th>airbags</th>\n",
       "      <th>displacement</th>\n",
       "      <th>cylinder</th>\n",
       "      <th>gear_box</th>\n",
       "      <th>turning_radius</th>\n",
       "      <th>length</th>\n",
       "      <th>height</th>\n",
       "      <th>gross_weight</th>\n",
       "      <th>...</th>\n",
       "      <th>is_driver_seat_height_adjustable</th>\n",
       "      <th>is_day_night_rear_view_mirror</th>\n",
       "      <th>is_ecw</th>\n",
       "      <th>segment</th>\n",
       "      <th>fuel_type</th>\n",
       "      <th>max_torque</th>\n",
       "      <th>steering_type</th>\n",
       "      <th>policy_tenure</th>\n",
       "      <th>age_of_car</th>\n",
       "      <th>age_of_policyholder</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.273897</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.992877</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.457143</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.8</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.500</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.998082</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.490385</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.301397</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.200299</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.625</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.782329</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.490385</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.146344</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.571225</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.428571</td>\n",
       "      <td>0.467836</td>\n",
       "      <td>0.157143</td>\n",
       "      <td>0.424514</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.713391</td>\n",
       "      <td>0.10</td>\n",
       "      <td>0.375000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.100121</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.200299</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.625</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.339513</td>\n",
       "      <td>0.03</td>\n",
       "      <td>0.480769</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.100190</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.200299</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.625</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.214826</td>\n",
       "      <td>0.03</td>\n",
       "      <td>0.538462</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   population_density  make  airbags  displacement  cylinder  gear_box  \\\n",
       "0            0.273897   0.5      1.0      0.992877       1.0       1.0   \n",
       "1            0.301397   0.0      0.2      0.000000       0.0       0.0   \n",
       "2            0.146344   0.0      0.2      0.571225       1.0       0.0   \n",
       "3            0.100121   0.0      0.2      0.000000       0.0       0.0   \n",
       "4            0.100190   0.0      0.2      0.000000       0.0       0.0   \n",
       "\n",
       "   turning_radius    length    height  gross_weight  ...  \\\n",
       "0        1.000000  1.000000  0.457143      1.000000  ...   \n",
       "1        0.142857  0.000000  0.000000      0.200299  ...   \n",
       "2        0.428571  0.467836  0.157143      0.424514  ...   \n",
       "3        0.142857  0.000000  0.000000      0.200299  ...   \n",
       "4        0.142857  0.000000  0.000000      0.200299  ...   \n",
       "\n",
       "   is_driver_seat_height_adjustable  is_day_night_rear_view_mirror  is_ecw  \\\n",
       "0                               1.0                            0.0     1.0   \n",
       "1                               0.0                            0.0     0.0   \n",
       "2                               1.0                            1.0     1.0   \n",
       "3                               0.0                            0.0     0.0   \n",
       "4                               0.0                            0.0     0.0   \n",
       "\n",
       "   segment  fuel_type  max_torque  steering_type  policy_tenure  age_of_car  \\\n",
       "0      0.8        0.5       0.500            1.0       0.998082        0.15   \n",
       "1      0.0        0.0       0.625            1.0       0.782329        0.00   \n",
       "2      0.4        1.0       0.000            0.0       0.713391        0.10   \n",
       "3      0.0        0.0       0.625            1.0       0.339513        0.03   \n",
       "4      0.0        0.0       0.625            1.0       0.214826        0.03   \n",
       "\n",
       "   age_of_policyholder  \n",
       "0             0.490385  \n",
       "1             0.490385  \n",
       "2             0.375000  \n",
       "3             0.480769  \n",
       "4             0.538462  \n",
       "\n",
       "[5 rows x 33 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Repeat for undersampled, oversampled and test sets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "rus_train = pd.DataFrame(preprocessing_pipeline.transform(x_rus))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "rus_train.columns = pipeline_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "ros_train = pd.DataFrame(preprocessing_pipeline.transform(x_ros))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "ros_train.columns = pipeline_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_test = pd.DataFrame(preprocessing_pipeline.transform(x_test))\n",
    "x_test.columns = pipeline_features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 7.0) Feature selection methods"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 7.1) Categorical variables\n",
    "\n",
    "- Select the best categorical features.\n",
    "- Use approximately half of the maximum feature score as the threshold for feature selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "#define function for plotting graphs\n",
    "def chi_plot(df, features, y):\n",
    "    scores = chi2(df[features], y)\n",
    "\n",
    "    p_values = pd.Series(scores[1], index=df[features].columns)\n",
    "    p_values.sort_values(ascending=False, inplace=True)\n",
    "\n",
    "    p_values.plot.bar(figsize=(20, 20))\n",
    "    print(p_values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "rear_brakes_type                    0.936573\n",
      "transmission_type                   0.926753\n",
      "is_tpms                             0.886924\n",
      "is_parking_camera                   0.875984\n",
      "is_parking_sensors                  0.668088\n",
      "is_rear_window_defogger             0.585296\n",
      "is_rear_window_wiper                0.572109\n",
      "is_esc                              0.541955\n",
      "segment                             0.436046\n",
      "is_central_locking                  0.376903\n",
      "is_ecw                              0.376903\n",
      "fuel_type                           0.163985\n",
      "steering_type                       0.126944\n",
      "is_day_night_rear_view_mirror       0.112151\n",
      "max_torque                          0.087390\n",
      "is_driver_seat_height_adjustable    0.085127\n",
      "is_brake_assist                     0.071398\n",
      "is_front_fog_lights                 0.053378\n",
      "is_adjustable_steering              0.029124\n",
      "dtype: float64\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x2000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "chi_plot(new_train, categorical_features, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "transmission_type                   0.953163\n",
      "rear_brakes_type                    0.838330\n",
      "is_parking_camera                   0.790678\n",
      "is_tpms                             0.717825\n",
      "is_parking_sensors                  0.578131\n",
      "is_rear_window_wiper                0.567499\n",
      "is_rear_window_defogger             0.521839\n",
      "is_esc                              0.433838\n",
      "segment                             0.399192\n",
      "is_central_locking                  0.344756\n",
      "is_ecw                              0.344756\n",
      "fuel_type                           0.164345\n",
      "steering_type                       0.113478\n",
      "is_day_night_rear_view_mirror       0.100394\n",
      "is_driver_seat_height_adjustable    0.084248\n",
      "is_front_fog_lights                 0.070247\n",
      "is_brake_assist                     0.062559\n",
      "max_torque                          0.059061\n",
      "is_adjustable_steering              0.046195\n",
      "dtype: float64\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x2000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#for undersampled data\n",
    "chi_plot(rus_train, categorical_features, y_rus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "is_parking_camera                   7.927508e-01\n",
      "rear_brakes_type                    5.940417e-01\n",
      "transmission_type                   5.389904e-01\n",
      "is_tpms                             3.427045e-01\n",
      "is_parking_sensors                  1.677756e-01\n",
      "is_rear_window_wiper                1.744559e-02\n",
      "segment                             1.670515e-02\n",
      "is_esc                              1.290535e-02\n",
      "is_central_locking                  1.222048e-02\n",
      "is_ecw                              1.222048e-02\n",
      "is_rear_window_defogger             1.070596e-02\n",
      "fuel_type                           6.994080e-05\n",
      "steering_type                       2.295553e-05\n",
      "is_day_night_rear_view_mirror       5.645307e-06\n",
      "max_torque                          1.580971e-07\n",
      "is_driver_seat_height_adjustable    5.444152e-08\n",
      "is_brake_assist                     4.931710e-08\n",
      "is_front_fog_lights                 1.514723e-09\n",
      "is_adjustable_steering              3.499228e-11\n",
      "dtype: float64\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x2000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "chi_plot(ros_train, categorical_features, y_ros)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "***Insight***\n",
    "\n",
    "- The unsampled and undersampled categorical features had similar feature importances.\n",
    "- The oversampled set however, had less prominent features with importance. \n",
    "- However, among the three sets, three features were prominent with high importance, these features are `is_parking_camera`, `rear_brakes_type` and `transmission_type`.\n",
    "- Create categorical feature selection variables for each set using respective thresholds."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "#for unsampled set\n",
    "un_cat = ['rear_brakes_type', 'transmission_type', 'is_tpms', 'is_parking_camera', 'is_parking_sensors', 'is_rear_window_defogger', 'is_rear_window_wiper', 'is_esc'] \n",
    "\n",
    "#for undersampled set\n",
    "under_cat = ['transmission_type','rear_brakes_type', 'is_parking_camera', 'is_tpms', 'is_parking_sensors','is_rear_window_wiper', 'is_rear_window_defogger'] \n",
    "\n",
    "#for the oversampled set\n",
    "over_cat = ['is_parking_camera', 'rear_brakes_type', 'transmission_type']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 7.2) Numerical Feature selection\n",
    "\n",
    "- The F-Classification and Mutual information methods will be used for numerical feature importance\n",
    "- These methods will also be implemeted on the various splits\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_selection import SelectKBest"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 7.2.1) F-Classification feature selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "def f_selection(df, features, y, string):\n",
    "    fs = SelectKBest(score_func=f_classif, k='all')\n",
    "    fs.fit(df[features], y)\n",
    "    fs_dict = dict(zip(df[features].columns.to_list(), fs.scores_))\n",
    "\n",
    "    plt.figure(figsize=(30,20))\n",
    "    plt.bar(list(fs_dict.keys()), fs_dict.values(), color='grey')\n",
    "    plt.xlabel('Feature')\n",
    "    plt.ylabel('Feature Importance')\n",
    "    plt.title(string)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 3000x2000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f_selection(new_train, numerical_features, y_train, 'Unsampled Set')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 3000x2000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f_selection(ros_train, numerical_features, y_ros, 'Oversampled Set')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 3000x2000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f_selection(rus_train, numerical_features, y_rus, 'Undersampled set')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "***Insight***\n",
    "\n",
    "- The `policy_tenure` feature had the most significant importance in all the sets.\n",
    "- Aside the aforementioned feature, there were four other features with quite some importance. These features namely; `age_of_car`, `age_of_policyholder`, `population_density` and `cylinder`.\n",
    "- The unsampled and oversampled sets had the same order of feature importance.\n",
    "- The undersamped set however had a slightly different order.\n",
    "- Since the top 5 features were common for the f-selection method, create a general list to represent these features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_f_classif = ['policy_tenure', 'age_of_car', 'age_of_policyholder', 'population_density', 'cylinder']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 7.2.2) Mutual Information feature selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "def mi_selection(df, features, y, string):\n",
    "    mi = SelectKBest(score_func=mutual_info_classif, k='all')\n",
    "    mi.fit(df[features], y)\n",
    "    fs_dict = dict(zip(df[features].columns.to_list(), mi.scores_))\n",
    "\n",
    "    plt.figure(figsize=(30,20))\n",
    "    plt.bar(list(fs_dict.keys()), fs_dict.values(), color='blue')\n",
    "    plt.xlabel('Feature')\n",
    "    plt.ylabel('Feature Importance')\n",
    "    plt.title(string)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 3000x2000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mi_selection(new_train, numerical_features, y_train, 'Unsampled Set')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 3000x2000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mi_selection(rus_train, numerical_features, y_rus, 'Undersampled Set')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 3000x2000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mi_selection(ros_train, numerical_features, y_ros, 'Oversampled set')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "***Insight***\n",
    "\n",
    "- Generally, the range for feature importance of all the sets were small, with the highest range being ~ 0 - 0.5. \n",
    "- Generate lists for all selected numerical features.\n",
    "\n",
    "- *Unsampled Set*:\n",
    "    - The `cylinder` feature had the most importance to the target. \n",
    "    - Only the `airbags` feature was above the desired threshold\n",
    "    - The `ncap_rating` and `height` features are slightly and should be considered to increase the feature space.\n",
    "\n",
    "- *Undersampled Set*:\n",
    "    - The `policy_tenure` had the most feature importance.\n",
    "    - Only the `cylinder` and `displacement` features were above the desired threshold.\n",
    "\n",
    "- *Oversampled Set*: \n",
    "    - This set yieled different results \n",
    "    - The `policy_tenure` feature was the only feature of significant importance. \n",
    "    - Remaining features should be discarded due to extremely low levels of importance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "#for the unsampled set\n",
    "mi_un = ['cylinder', 'airbags', 'ncap_rating', 'height']\n",
    "\n",
    "#for the undersampled set\n",
    "mi_under = ['policy_tenure', 'cylinder', 'displacement']\n",
    "\n",
    "#for the oversampled set\n",
    "mi_over = ['policy_tenure']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 8.0) Model selection and optimization\n",
    "\n",
    "- Determine predictive models;\n",
    "    - Logistic Regression\n",
    "    - Support Vector Classifier\n",
    "    - Random Forest Classifier\n",
    "\n",
    "- Generate predictions for;\n",
    "    - unfiltered features for the various splits\n",
    "    - selected features from feature selection for various splits\n",
    "\n",
    "- Evaluate predictions;\n",
    "    - Classification reports\n",
    "    - Confusion matrices\n",
    "    - f1_score and accuracy_score metrics\n",
    "\n",
    "- Select the best model and set\n",
    "\n",
    "- Model optimization\n",
    "    - Hyperparemeter tuning\n",
    "    - Decision threshold evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "base_svc = SVC(random_state=45)\n",
    "base_rfc = RandomForestClassifier(random_state=45)\n",
    "base_gbc = GradientBoostingClassifier(random_state=45)\n",
    "base_ada = AdaBoostClassifier(random_state=45)\n",
    "base_xgb = XGBClassifier(random_state=45)\n",
    "base_cat = CatBoostClassifier(random_state=45)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "def sampler(model, undersampled, oversampled, unsampled, y_undersampled, y_oversampled, y_unsampled):\n",
    "    #for undersampled\n",
    "    model.fit(undersampled,y_undersampled)\n",
    "    undersampled_preds = model.predict(x_test)\n",
    "    print('The accuracy score for the {} undersampled set is: {}'.format(model, accuracy_score(y_test, undersampled_preds)))\n",
    "    print('The f1 score for the {} undersampled set is: {}'.format(model, f1_score(y_test, undersampled_preds)))\n",
    "    print('\\n')\n",
    "\n",
    "    #for the unsampled\n",
    "    model.fit(unsampled,y_unsampled)\n",
    "    unsampled_preds = model.predict(x_test)\n",
    "    print('The accuracy score for the {} unsampled set is: {}'.format(model, accuracy_score(y_test, unsampled_preds)))\n",
    "    print('The f1 score for the {} unsampled set is: {}'.format(model, f1_score(y_test, unsampled_preds)))\n",
    "    print('\\n')\n",
    "\n",
    "    #for the oversampled\n",
    "    model.fit(oversampled, y_oversampled)\n",
    "    oversampled_preds = model.predict(x_test)\n",
    "    print('The accuracy score for the {} oversampled set is: {}'.format(model, accuracy_score(y_test, oversampled_preds)))\n",
    "    print('The f1 score for the {} oversampled set is: {}'.format(model, f1_score(y_test, oversampled_preds)))\n",
    "    print('\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The accuracy score for the SVC(random_state=45) undersampled set is: 0.5184300341296928\n",
      "The f1 score for the SVC(random_state=45) undersampled set is: 0.1359461114513166\n",
      "\n",
      "\n",
      "The accuracy score for the SVC(random_state=45) unsampled set is: 0.9372013651877134\n",
      "The f1 score for the SVC(random_state=45) unsampled set is: 0.0\n",
      "\n",
      "\n",
      "The accuracy score for the SVC(random_state=45) oversampled set is: 0.504778156996587\n",
      "The f1 score for the SVC(random_state=45) oversampled set is: 0.13989330171902786\n",
      "\n",
      "\n",
      "The accuracy score for the RandomForestClassifier(random_state=45) undersampled set is: 0.5638225255972696\n",
      "The f1 score for the RandomForestClassifier(random_state=45) undersampled set is: 0.14800000000000002\n",
      "\n",
      "\n",
      "The accuracy score for the RandomForestClassifier(random_state=45) unsampled set is: 0.9341296928327645\n",
      "The f1 score for the RandomForestClassifier(random_state=45) unsampled set is: 0.02030456852791878\n",
      "\n",
      "\n",
      "The accuracy score for the RandomForestClassifier(random_state=45) oversampled set is: 0.9197952218430034\n",
      "The f1 score for the RandomForestClassifier(random_state=45) oversampled set is: 0.03292181069958848\n",
      "\n",
      "\n",
      "The accuracy score for the GradientBoostingClassifier(random_state=45) undersampled set is: 0.5337883959044368\n",
      "The f1 score for the GradientBoostingClassifier(random_state=45) undersampled set is: 0.15365551425030977\n",
      "\n",
      "\n",
      "The accuracy score for the GradientBoostingClassifier(random_state=45) unsampled set is: 0.9372013651877134\n",
      "The f1 score for the GradientBoostingClassifier(random_state=45) unsampled set is: 0.0\n",
      "\n",
      "\n",
      "The accuracy score for the GradientBoostingClassifier(random_state=45) oversampled set is: 0.5556313993174061\n",
      "The f1 score for the GradientBoostingClassifier(random_state=45) oversampled set is: 0.15344603381014305\n",
      "\n",
      "\n",
      "The accuracy score for the AdaBoostClassifier(random_state=45) undersampled set is: 0.5320819112627987\n",
      "The f1 score for the AdaBoostClassifier(random_state=45) undersampled set is: 0.15003099814011162\n",
      "\n",
      "\n",
      "The accuracy score for the AdaBoostClassifier(random_state=45) unsampled set is: 0.9372013651877134\n",
      "The f1 score for the AdaBoostClassifier(random_state=45) unsampled set is: 0.0\n",
      "\n",
      "\n",
      "The accuracy score for the AdaBoostClassifier(random_state=45) oversampled set is: 0.532764505119454\n",
      "The f1 score for the AdaBoostClassifier(random_state=45) oversampled set is: 0.1427676894176581\n",
      "\n",
      "\n",
      "The accuracy score for the XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
      "              colsample_bylevel=None, colsample_bynode=None,\n",
      "              colsample_bytree=None, device=None, early_stopping_rounds=None,\n",
      "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
      "              gamma=None, grow_policy=None, importance_type=None,\n",
      "              interaction_constraints=None, learning_rate=None, max_bin=None,\n",
      "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
      "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
      "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
      "              multi_strategy=None, n_estimators=None, n_jobs=None,\n",
      "              num_parallel_tree=None, random_state=45, ...) undersampled set is: 0.5511945392491467\n",
      "The f1 score for the XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
      "              colsample_bylevel=None, colsample_bynode=None,\n",
      "              colsample_bytree=None, device=None, early_stopping_rounds=None,\n",
      "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
      "              gamma=None, grow_policy=None, importance_type=None,\n",
      "              interaction_constraints=None, learning_rate=None, max_bin=None,\n",
      "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
      "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
      "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
      "              multi_strategy=None, n_estimators=None, n_jobs=None,\n",
      "              num_parallel_tree=None, random_state=45, ...) undersampled set is: 0.14443721535458687\n",
      "\n",
      "\n",
      "The accuracy score for the XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
      "              colsample_bylevel=None, colsample_bynode=None,\n",
      "              colsample_bytree=None, device=None, early_stopping_rounds=None,\n",
      "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
      "              gamma=None, grow_policy=None, importance_type=None,\n",
      "              interaction_constraints=None, learning_rate=None, max_bin=None,\n",
      "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
      "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
      "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
      "              multi_strategy=None, n_estimators=None, n_jobs=None,\n",
      "              num_parallel_tree=None, random_state=45, ...) unsampled set is: 0.9361774744027304\n",
      "The f1 score for the XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
      "              colsample_bylevel=None, colsample_bynode=None,\n",
      "              colsample_bytree=None, device=None, early_stopping_rounds=None,\n",
      "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
      "              gamma=None, grow_policy=None, importance_type=None,\n",
      "              interaction_constraints=None, learning_rate=None, max_bin=None,\n",
      "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
      "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
      "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
      "              multi_strategy=None, n_estimators=None, n_jobs=None,\n",
      "              num_parallel_tree=None, random_state=45, ...) unsampled set is: 0.0\n",
      "\n",
      "\n",
      "The accuracy score for the XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
      "              colsample_bylevel=None, colsample_bynode=None,\n",
      "              colsample_bytree=None, device=None, early_stopping_rounds=None,\n",
      "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
      "              gamma=None, grow_policy=None, importance_type=None,\n",
      "              interaction_constraints=None, learning_rate=None, max_bin=None,\n",
      "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
      "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
      "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
      "              multi_strategy=None, n_estimators=None, n_jobs=None,\n",
      "              num_parallel_tree=None, random_state=45, ...) oversampled set is: 0.6996587030716723\n",
      "The f1 score for the XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
      "              colsample_bylevel=None, colsample_bynode=None,\n",
      "              colsample_bytree=None, device=None, early_stopping_rounds=None,\n",
      "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
      "              gamma=None, grow_policy=None, importance_type=None,\n",
      "              interaction_constraints=None, learning_rate=None, max_bin=None,\n",
      "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
      "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
      "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
      "              multi_strategy=None, n_estimators=None, n_jobs=None,\n",
      "              num_parallel_tree=None, random_state=45, ...) oversampled set is: 0.14563106796116507\n",
      "\n",
      "\n",
      "Learning rate set to 0.023831\n",
      "0:\tlearn: 0.6914278\ttotal: 195ms\tremaining: 3m 14s\n",
      "1:\tlearn: 0.6895753\ttotal: 203ms\tremaining: 1m 41s\n",
      "2:\tlearn: 0.6879581\ttotal: 209ms\tremaining: 1m 9s\n",
      "3:\tlearn: 0.6864764\ttotal: 216ms\tremaining: 53.8s\n",
      "4:\tlearn: 0.6851470\ttotal: 224ms\tremaining: 44.5s\n",
      "5:\tlearn: 0.6837175\ttotal: 230ms\tremaining: 38s\n",
      "6:\tlearn: 0.6823206\ttotal: 237ms\tremaining: 33.7s\n",
      "7:\tlearn: 0.6808796\ttotal: 246ms\tremaining: 30.5s\n",
      "8:\tlearn: 0.6797154\ttotal: 255ms\tremaining: 28.1s\n",
      "9:\tlearn: 0.6784719\ttotal: 262ms\tremaining: 25.9s\n",
      "10:\tlearn: 0.6773430\ttotal: 270ms\tremaining: 24.2s\n",
      "11:\tlearn: 0.6761140\ttotal: 276ms\tremaining: 22.7s\n",
      "12:\tlearn: 0.6750139\ttotal: 283ms\tremaining: 21.5s\n",
      "13:\tlearn: 0.6739328\ttotal: 290ms\tremaining: 20.4s\n",
      "14:\tlearn: 0.6730573\ttotal: 297ms\tremaining: 19.5s\n",
      "15:\tlearn: 0.6720324\ttotal: 304ms\tremaining: 18.7s\n",
      "16:\tlearn: 0.6710015\ttotal: 310ms\tremaining: 17.9s\n",
      "17:\tlearn: 0.6700374\ttotal: 318ms\tremaining: 17.4s\n",
      "18:\tlearn: 0.6691312\ttotal: 325ms\tremaining: 16.8s\n",
      "19:\tlearn: 0.6681805\ttotal: 331ms\tremaining: 16.2s\n",
      "20:\tlearn: 0.6673735\ttotal: 339ms\tremaining: 15.8s\n",
      "21:\tlearn: 0.6666408\ttotal: 346ms\tremaining: 15.4s\n",
      "22:\tlearn: 0.6658552\ttotal: 353ms\tremaining: 15s\n",
      "23:\tlearn: 0.6652324\ttotal: 359ms\tremaining: 14.6s\n",
      "24:\tlearn: 0.6646672\ttotal: 367ms\tremaining: 14.3s\n",
      "25:\tlearn: 0.6640235\ttotal: 377ms\tremaining: 14.1s\n",
      "26:\tlearn: 0.6634382\ttotal: 386ms\tremaining: 13.9s\n",
      "27:\tlearn: 0.6627443\ttotal: 392ms\tremaining: 13.6s\n",
      "28:\tlearn: 0.6621021\ttotal: 405ms\tremaining: 13.5s\n",
      "29:\tlearn: 0.6615179\ttotal: 412ms\tremaining: 13.3s\n",
      "30:\tlearn: 0.6609353\ttotal: 418ms\tremaining: 13.1s\n",
      "31:\tlearn: 0.6604193\ttotal: 426ms\tremaining: 12.9s\n",
      "32:\tlearn: 0.6598420\ttotal: 435ms\tremaining: 12.8s\n",
      "33:\tlearn: 0.6593364\ttotal: 443ms\tremaining: 12.6s\n",
      "34:\tlearn: 0.6588136\ttotal: 450ms\tremaining: 12.4s\n",
      "35:\tlearn: 0.6584164\ttotal: 456ms\tremaining: 12.2s\n",
      "36:\tlearn: 0.6579441\ttotal: 465ms\tremaining: 12.1s\n",
      "37:\tlearn: 0.6574121\ttotal: 471ms\tremaining: 11.9s\n",
      "38:\tlearn: 0.6569265\ttotal: 479ms\tremaining: 11.8s\n",
      "39:\tlearn: 0.6567320\ttotal: 485ms\tremaining: 11.6s\n",
      "40:\tlearn: 0.6562721\ttotal: 491ms\tremaining: 11.5s\n",
      "41:\tlearn: 0.6558863\ttotal: 498ms\tremaining: 11.4s\n",
      "42:\tlearn: 0.6556627\ttotal: 505ms\tremaining: 11.2s\n",
      "43:\tlearn: 0.6552826\ttotal: 513ms\tremaining: 11.1s\n",
      "44:\tlearn: 0.6549070\ttotal: 520ms\tremaining: 11s\n",
      "45:\tlearn: 0.6547316\ttotal: 528ms\tremaining: 10.9s\n",
      "46:\tlearn: 0.6543165\ttotal: 534ms\tremaining: 10.8s\n",
      "47:\tlearn: 0.6540025\ttotal: 540ms\tremaining: 10.7s\n",
      "48:\tlearn: 0.6536924\ttotal: 551ms\tremaining: 10.7s\n",
      "49:\tlearn: 0.6534284\ttotal: 559ms\tremaining: 10.6s\n",
      "50:\tlearn: 0.6531757\ttotal: 566ms\tremaining: 10.5s\n",
      "51:\tlearn: 0.6529936\ttotal: 573ms\tremaining: 10.5s\n",
      "52:\tlearn: 0.6526052\ttotal: 581ms\tremaining: 10.4s\n",
      "53:\tlearn: 0.6523401\ttotal: 587ms\tremaining: 10.3s\n",
      "54:\tlearn: 0.6520818\ttotal: 596ms\tremaining: 10.2s\n",
      "55:\tlearn: 0.6518307\ttotal: 602ms\tremaining: 10.2s\n",
      "56:\tlearn: 0.6517126\ttotal: 610ms\tremaining: 10.1s\n",
      "57:\tlearn: 0.6513629\ttotal: 618ms\tremaining: 10s\n",
      "58:\tlearn: 0.6511789\ttotal: 627ms\tremaining: 9.99s\n",
      "59:\tlearn: 0.6509457\ttotal: 633ms\tremaining: 9.91s\n",
      "60:\tlearn: 0.6506761\ttotal: 640ms\tremaining: 9.86s\n",
      "61:\tlearn: 0.6503932\ttotal: 649ms\tremaining: 9.81s\n",
      "62:\tlearn: 0.6501403\ttotal: 657ms\tremaining: 9.77s\n",
      "63:\tlearn: 0.6499573\ttotal: 663ms\tremaining: 9.69s\n",
      "64:\tlearn: 0.6496530\ttotal: 669ms\tremaining: 9.62s\n",
      "65:\tlearn: 0.6494280\ttotal: 676ms\tremaining: 9.57s\n",
      "66:\tlearn: 0.6491813\ttotal: 683ms\tremaining: 9.51s\n",
      "67:\tlearn: 0.6489203\ttotal: 691ms\tremaining: 9.47s\n",
      "68:\tlearn: 0.6486231\ttotal: 697ms\tremaining: 9.41s\n",
      "69:\tlearn: 0.6484593\ttotal: 705ms\tremaining: 9.37s\n",
      "70:\tlearn: 0.6483817\ttotal: 709ms\tremaining: 9.28s\n",
      "71:\tlearn: 0.6480853\ttotal: 716ms\tremaining: 9.22s\n",
      "72:\tlearn: 0.6479231\ttotal: 724ms\tremaining: 9.19s\n",
      "73:\tlearn: 0.6476555\ttotal: 730ms\tremaining: 9.13s\n",
      "74:\tlearn: 0.6474898\ttotal: 738ms\tremaining: 9.1s\n",
      "75:\tlearn: 0.6473696\ttotal: 744ms\tremaining: 9.05s\n",
      "76:\tlearn: 0.6471443\ttotal: 751ms\tremaining: 9s\n",
      "77:\tlearn: 0.6470554\ttotal: 760ms\tremaining: 8.98s\n",
      "78:\tlearn: 0.6469113\ttotal: 770ms\tremaining: 8.97s\n",
      "79:\tlearn: 0.6466148\ttotal: 776ms\tremaining: 8.92s\n",
      "80:\tlearn: 0.6463639\ttotal: 782ms\tremaining: 8.87s\n",
      "81:\tlearn: 0.6461589\ttotal: 790ms\tremaining: 8.84s\n",
      "82:\tlearn: 0.6460068\ttotal: 796ms\tremaining: 8.8s\n",
      "83:\tlearn: 0.6458664\ttotal: 807ms\tremaining: 8.8s\n",
      "84:\tlearn: 0.6455418\ttotal: 817ms\tremaining: 8.79s\n",
      "85:\tlearn: 0.6453434\ttotal: 825ms\tremaining: 8.77s\n",
      "86:\tlearn: 0.6450301\ttotal: 834ms\tremaining: 8.76s\n",
      "87:\tlearn: 0.6447763\ttotal: 841ms\tremaining: 8.72s\n",
      "88:\tlearn: 0.6446416\ttotal: 848ms\tremaining: 8.68s\n",
      "89:\tlearn: 0.6444121\ttotal: 857ms\tremaining: 8.66s\n",
      "90:\tlearn: 0.6442592\ttotal: 863ms\tremaining: 8.62s\n",
      "91:\tlearn: 0.6441094\ttotal: 871ms\tremaining: 8.6s\n",
      "92:\tlearn: 0.6439106\ttotal: 877ms\tremaining: 8.56s\n",
      "93:\tlearn: 0.6436917\ttotal: 886ms\tremaining: 8.54s\n",
      "94:\tlearn: 0.6436022\ttotal: 892ms\tremaining: 8.49s\n",
      "95:\tlearn: 0.6434352\ttotal: 899ms\tremaining: 8.47s\n",
      "96:\tlearn: 0.6433197\ttotal: 906ms\tremaining: 8.43s\n",
      "97:\tlearn: 0.6430938\ttotal: 912ms\tremaining: 8.4s\n",
      "98:\tlearn: 0.6428502\ttotal: 920ms\tremaining: 8.37s\n",
      "99:\tlearn: 0.6427272\ttotal: 927ms\tremaining: 8.34s\n",
      "100:\tlearn: 0.6425658\ttotal: 935ms\tremaining: 8.32s\n",
      "101:\tlearn: 0.6424200\ttotal: 941ms\tremaining: 8.28s\n",
      "102:\tlearn: 0.6422889\ttotal: 950ms\tremaining: 8.27s\n",
      "103:\tlearn: 0.6421046\ttotal: 956ms\tremaining: 8.24s\n",
      "104:\tlearn: 0.6420037\ttotal: 963ms\tremaining: 8.21s\n",
      "105:\tlearn: 0.6419130\ttotal: 971ms\tremaining: 8.19s\n",
      "106:\tlearn: 0.6416877\ttotal: 978ms\tremaining: 8.16s\n",
      "107:\tlearn: 0.6416283\ttotal: 985ms\tremaining: 8.13s\n",
      "108:\tlearn: 0.6414623\ttotal: 991ms\tremaining: 8.1s\n",
      "109:\tlearn: 0.6412829\ttotal: 1000ms\tremaining: 8.09s\n",
      "110:\tlearn: 0.6412015\ttotal: 1.01s\tremaining: 8.06s\n",
      "111:\tlearn: 0.6411024\ttotal: 1.01s\tremaining: 8.05s\n",
      "112:\tlearn: 0.6408718\ttotal: 1.02s\tremaining: 8.03s\n",
      "113:\tlearn: 0.6407994\ttotal: 1.03s\tremaining: 8.03s\n",
      "114:\tlearn: 0.6406823\ttotal: 1.04s\tremaining: 8s\n",
      "115:\tlearn: 0.6405414\ttotal: 1.05s\tremaining: 7.99s\n",
      "116:\tlearn: 0.6404830\ttotal: 1.05s\tremaining: 7.95s\n",
      "117:\tlearn: 0.6403881\ttotal: 1.06s\tremaining: 7.93s\n",
      "118:\tlearn: 0.6402326\ttotal: 1.07s\tremaining: 7.92s\n",
      "119:\tlearn: 0.6400664\ttotal: 1.08s\tremaining: 7.89s\n",
      "120:\tlearn: 0.6399250\ttotal: 1.08s\tremaining: 7.87s\n",
      "121:\tlearn: 0.6397864\ttotal: 1.09s\tremaining: 7.84s\n",
      "122:\tlearn: 0.6396989\ttotal: 1.1s\tremaining: 7.82s\n",
      "123:\tlearn: 0.6396157\ttotal: 1.1s\tremaining: 7.8s\n",
      "124:\tlearn: 0.6394670\ttotal: 1.11s\tremaining: 7.78s\n",
      "125:\tlearn: 0.6392792\ttotal: 1.12s\tremaining: 7.75s\n",
      "126:\tlearn: 0.6391023\ttotal: 1.13s\tremaining: 7.74s\n",
      "127:\tlearn: 0.6389455\ttotal: 1.13s\tremaining: 7.71s\n",
      "128:\tlearn: 0.6388032\ttotal: 1.14s\tremaining: 7.69s\n",
      "129:\tlearn: 0.6385958\ttotal: 1.15s\tremaining: 7.67s\n",
      "130:\tlearn: 0.6385064\ttotal: 1.15s\tremaining: 7.64s\n",
      "131:\tlearn: 0.6384760\ttotal: 1.16s\tremaining: 7.63s\n",
      "132:\tlearn: 0.6383636\ttotal: 1.17s\tremaining: 7.61s\n",
      "133:\tlearn: 0.6382669\ttotal: 1.18s\tremaining: 7.6s\n",
      "134:\tlearn: 0.6380896\ttotal: 1.18s\tremaining: 7.57s\n",
      "135:\tlearn: 0.6379776\ttotal: 1.19s\tremaining: 7.55s\n",
      "136:\tlearn: 0.6378074\ttotal: 1.2s\tremaining: 7.53s\n",
      "137:\tlearn: 0.6376591\ttotal: 1.2s\tremaining: 7.5s\n",
      "138:\tlearn: 0.6374187\ttotal: 1.21s\tremaining: 7.51s\n",
      "139:\tlearn: 0.6373132\ttotal: 1.22s\tremaining: 7.49s\n",
      "140:\tlearn: 0.6371524\ttotal: 1.23s\tremaining: 7.5s\n",
      "141:\tlearn: 0.6370210\ttotal: 1.24s\tremaining: 7.48s\n",
      "142:\tlearn: 0.6369134\ttotal: 1.25s\tremaining: 7.47s\n",
      "143:\tlearn: 0.6367007\ttotal: 1.25s\tremaining: 7.46s\n",
      "144:\tlearn: 0.6364978\ttotal: 1.26s\tremaining: 7.45s\n",
      "145:\tlearn: 0.6363950\ttotal: 1.27s\tremaining: 7.43s\n",
      "146:\tlearn: 0.6363189\ttotal: 1.28s\tremaining: 7.41s\n",
      "147:\tlearn: 0.6362140\ttotal: 1.28s\tremaining: 7.38s\n",
      "148:\tlearn: 0.6361104\ttotal: 1.29s\tremaining: 7.37s\n",
      "149:\tlearn: 0.6360147\ttotal: 1.3s\tremaining: 7.35s\n",
      "150:\tlearn: 0.6359285\ttotal: 1.3s\tremaining: 7.34s\n",
      "151:\tlearn: 0.6357521\ttotal: 1.31s\tremaining: 7.32s\n",
      "152:\tlearn: 0.6357103\ttotal: 1.32s\tremaining: 7.3s\n",
      "153:\tlearn: 0.6356186\ttotal: 1.32s\tremaining: 7.28s\n",
      "154:\tlearn: 0.6354985\ttotal: 1.33s\tremaining: 7.26s\n",
      "155:\tlearn: 0.6354171\ttotal: 1.34s\tremaining: 7.25s\n",
      "156:\tlearn: 0.6353021\ttotal: 1.34s\tremaining: 7.22s\n",
      "157:\tlearn: 0.6352627\ttotal: 1.35s\tremaining: 7.21s\n",
      "158:\tlearn: 0.6351401\ttotal: 1.36s\tremaining: 7.2s\n",
      "159:\tlearn: 0.6350041\ttotal: 1.37s\tremaining: 7.18s\n",
      "160:\tlearn: 0.6349056\ttotal: 1.38s\tremaining: 7.17s\n",
      "161:\tlearn: 0.6347856\ttotal: 1.38s\tremaining: 7.15s\n",
      "162:\tlearn: 0.6346812\ttotal: 1.39s\tremaining: 7.13s\n",
      "163:\tlearn: 0.6345957\ttotal: 1.4s\tremaining: 7.11s\n",
      "164:\tlearn: 0.6344653\ttotal: 1.4s\tremaining: 7.11s\n",
      "165:\tlearn: 0.6343346\ttotal: 1.41s\tremaining: 7.09s\n",
      "166:\tlearn: 0.6341355\ttotal: 1.42s\tremaining: 7.08s\n",
      "167:\tlearn: 0.6339868\ttotal: 1.43s\tremaining: 7.06s\n",
      "168:\tlearn: 0.6337789\ttotal: 1.43s\tremaining: 7.04s\n",
      "169:\tlearn: 0.6336706\ttotal: 1.44s\tremaining: 7.03s\n",
      "170:\tlearn: 0.6335983\ttotal: 1.45s\tremaining: 7.02s\n",
      "171:\tlearn: 0.6334356\ttotal: 1.46s\tremaining: 7.01s\n",
      "172:\tlearn: 0.6333441\ttotal: 1.46s\tremaining: 6.99s\n",
      "173:\tlearn: 0.6331724\ttotal: 1.47s\tremaining: 6.97s\n",
      "174:\tlearn: 0.6331236\ttotal: 1.48s\tremaining: 6.96s\n",
      "175:\tlearn: 0.6329951\ttotal: 1.48s\tremaining: 6.94s\n",
      "176:\tlearn: 0.6329472\ttotal: 1.49s\tremaining: 6.92s\n",
      "177:\tlearn: 0.6327617\ttotal: 1.5s\tremaining: 6.9s\n",
      "178:\tlearn: 0.6326674\ttotal: 1.5s\tremaining: 6.89s\n",
      "179:\tlearn: 0.6326347\ttotal: 1.51s\tremaining: 6.88s\n",
      "180:\tlearn: 0.6324688\ttotal: 1.51s\tremaining: 6.86s\n",
      "181:\tlearn: 0.6323268\ttotal: 1.52s\tremaining: 6.85s\n",
      "182:\tlearn: 0.6322788\ttotal: 1.53s\tremaining: 6.83s\n",
      "183:\tlearn: 0.6322000\ttotal: 1.54s\tremaining: 6.81s\n",
      "184:\tlearn: 0.6321162\ttotal: 1.54s\tremaining: 6.8s\n",
      "185:\tlearn: 0.6319643\ttotal: 1.55s\tremaining: 6.78s\n",
      "186:\tlearn: 0.6318209\ttotal: 1.56s\tremaining: 6.77s\n",
      "187:\tlearn: 0.6317272\ttotal: 1.56s\tremaining: 6.76s\n",
      "188:\tlearn: 0.6316241\ttotal: 1.57s\tremaining: 6.75s\n",
      "189:\tlearn: 0.6315384\ttotal: 1.58s\tremaining: 6.73s\n",
      "190:\tlearn: 0.6313849\ttotal: 1.59s\tremaining: 6.72s\n",
      "191:\tlearn: 0.6313130\ttotal: 1.59s\tremaining: 6.71s\n",
      "192:\tlearn: 0.6312457\ttotal: 1.6s\tremaining: 6.69s\n",
      "193:\tlearn: 0.6310983\ttotal: 1.61s\tremaining: 6.68s\n",
      "194:\tlearn: 0.6309769\ttotal: 1.61s\tremaining: 6.66s\n",
      "195:\tlearn: 0.6308744\ttotal: 1.63s\tremaining: 6.67s\n",
      "196:\tlearn: 0.6307683\ttotal: 1.64s\tremaining: 6.67s\n",
      "197:\tlearn: 0.6306658\ttotal: 1.64s\tremaining: 6.66s\n",
      "198:\tlearn: 0.6306650\ttotal: 1.65s\tremaining: 6.63s\n",
      "199:\tlearn: 0.6305691\ttotal: 1.66s\tremaining: 6.63s\n",
      "200:\tlearn: 0.6304780\ttotal: 1.66s\tremaining: 6.61s\n",
      "201:\tlearn: 0.6304129\ttotal: 1.67s\tremaining: 6.6s\n",
      "202:\tlearn: 0.6302631\ttotal: 1.68s\tremaining: 6.58s\n",
      "203:\tlearn: 0.6301280\ttotal: 1.69s\tremaining: 6.58s\n",
      "204:\tlearn: 0.6300078\ttotal: 1.69s\tremaining: 6.56s\n",
      "205:\tlearn: 0.6299216\ttotal: 1.7s\tremaining: 6.55s\n",
      "206:\tlearn: 0.6299014\ttotal: 1.7s\tremaining: 6.53s\n",
      "207:\tlearn: 0.6298621\ttotal: 1.71s\tremaining: 6.52s\n",
      "208:\tlearn: 0.6298033\ttotal: 1.72s\tremaining: 6.51s\n",
      "209:\tlearn: 0.6296724\ttotal: 1.73s\tremaining: 6.49s\n",
      "210:\tlearn: 0.6295148\ttotal: 1.73s\tremaining: 6.48s\n",
      "211:\tlearn: 0.6293312\ttotal: 1.74s\tremaining: 6.46s\n",
      "212:\tlearn: 0.6292442\ttotal: 1.75s\tremaining: 6.46s\n",
      "213:\tlearn: 0.6291504\ttotal: 1.75s\tremaining: 6.44s\n",
      "214:\tlearn: 0.6290883\ttotal: 1.76s\tremaining: 6.42s\n",
      "215:\tlearn: 0.6289837\ttotal: 1.77s\tremaining: 6.42s\n",
      "216:\tlearn: 0.6288123\ttotal: 1.78s\tremaining: 6.42s\n",
      "217:\tlearn: 0.6288036\ttotal: 1.78s\tremaining: 6.4s\n",
      "218:\tlearn: 0.6287342\ttotal: 1.79s\tremaining: 6.39s\n",
      "219:\tlearn: 0.6286817\ttotal: 1.8s\tremaining: 6.38s\n",
      "220:\tlearn: 0.6285771\ttotal: 1.81s\tremaining: 6.37s\n",
      "221:\tlearn: 0.6284402\ttotal: 1.82s\tremaining: 6.37s\n",
      "222:\tlearn: 0.6283661\ttotal: 1.82s\tremaining: 6.36s\n",
      "223:\tlearn: 0.6282660\ttotal: 1.83s\tremaining: 6.35s\n",
      "224:\tlearn: 0.6281302\ttotal: 1.84s\tremaining: 6.34s\n",
      "225:\tlearn: 0.6280619\ttotal: 1.85s\tremaining: 6.33s\n",
      "226:\tlearn: 0.6280057\ttotal: 1.85s\tremaining: 6.31s\n",
      "227:\tlearn: 0.6278700\ttotal: 1.86s\tremaining: 6.3s\n",
      "228:\tlearn: 0.6277654\ttotal: 1.87s\tremaining: 6.3s\n",
      "229:\tlearn: 0.6276460\ttotal: 1.88s\tremaining: 6.29s\n",
      "230:\tlearn: 0.6275189\ttotal: 1.88s\tremaining: 6.27s\n",
      "231:\tlearn: 0.6274120\ttotal: 1.89s\tremaining: 6.26s\n",
      "232:\tlearn: 0.6272978\ttotal: 1.9s\tremaining: 6.25s\n",
      "233:\tlearn: 0.6272629\ttotal: 1.9s\tremaining: 6.23s\n",
      "234:\tlearn: 0.6272215\ttotal: 1.91s\tremaining: 6.22s\n",
      "235:\tlearn: 0.6271397\ttotal: 1.92s\tremaining: 6.21s\n",
      "236:\tlearn: 0.6270455\ttotal: 1.92s\tremaining: 6.2s\n",
      "237:\tlearn: 0.6269964\ttotal: 1.93s\tremaining: 6.18s\n",
      "238:\tlearn: 0.6269004\ttotal: 1.94s\tremaining: 6.17s\n",
      "239:\tlearn: 0.6267893\ttotal: 1.94s\tremaining: 6.16s\n",
      "240:\tlearn: 0.6267386\ttotal: 1.95s\tremaining: 6.14s\n",
      "241:\tlearn: 0.6265273\ttotal: 1.96s\tremaining: 6.14s\n",
      "242:\tlearn: 0.6264342\ttotal: 1.97s\tremaining: 6.12s\n",
      "243:\tlearn: 0.6262413\ttotal: 1.97s\tremaining: 6.11s\n",
      "244:\tlearn: 0.6261981\ttotal: 1.98s\tremaining: 6.1s\n",
      "245:\tlearn: 0.6260330\ttotal: 1.99s\tremaining: 6.09s\n",
      "246:\tlearn: 0.6258408\ttotal: 1.99s\tremaining: 6.08s\n",
      "247:\tlearn: 0.6257373\ttotal: 2s\tremaining: 6.07s\n",
      "248:\tlearn: 0.6255855\ttotal: 2.01s\tremaining: 6.06s\n",
      "249:\tlearn: 0.6254939\ttotal: 2.02s\tremaining: 6.05s\n",
      "250:\tlearn: 0.6254016\ttotal: 2.03s\tremaining: 6.06s\n",
      "251:\tlearn: 0.6253807\ttotal: 2.04s\tremaining: 6.05s\n",
      "252:\tlearn: 0.6253255\ttotal: 2.04s\tremaining: 6.04s\n",
      "253:\tlearn: 0.6250751\ttotal: 2.05s\tremaining: 6.03s\n",
      "254:\tlearn: 0.6249592\ttotal: 2.06s\tremaining: 6.02s\n",
      "255:\tlearn: 0.6248408\ttotal: 2.07s\tremaining: 6.01s\n",
      "256:\tlearn: 0.6247315\ttotal: 2.07s\tremaining: 6s\n",
      "257:\tlearn: 0.6246706\ttotal: 2.08s\tremaining: 5.99s\n",
      "258:\tlearn: 0.6246551\ttotal: 2.09s\tremaining: 5.97s\n",
      "259:\tlearn: 0.6245724\ttotal: 2.09s\tremaining: 5.96s\n",
      "260:\tlearn: 0.6244416\ttotal: 2.1s\tremaining: 5.95s\n",
      "261:\tlearn: 0.6243725\ttotal: 2.11s\tremaining: 5.94s\n",
      "262:\tlearn: 0.6242868\ttotal: 2.11s\tremaining: 5.92s\n",
      "263:\tlearn: 0.6242796\ttotal: 2.12s\tremaining: 5.91s\n",
      "264:\tlearn: 0.6241976\ttotal: 2.13s\tremaining: 5.9s\n",
      "265:\tlearn: 0.6241167\ttotal: 2.13s\tremaining: 5.89s\n",
      "266:\tlearn: 0.6240101\ttotal: 2.14s\tremaining: 5.88s\n",
      "267:\tlearn: 0.6239026\ttotal: 2.15s\tremaining: 5.86s\n",
      "268:\tlearn: 0.6238088\ttotal: 2.15s\tremaining: 5.86s\n",
      "269:\tlearn: 0.6236786\ttotal: 2.16s\tremaining: 5.84s\n",
      "270:\tlearn: 0.6235785\ttotal: 2.17s\tremaining: 5.84s\n",
      "271:\tlearn: 0.6233925\ttotal: 2.18s\tremaining: 5.83s\n",
      "272:\tlearn: 0.6232796\ttotal: 2.19s\tremaining: 5.82s\n",
      "273:\tlearn: 0.6231613\ttotal: 2.19s\tremaining: 5.81s\n",
      "274:\tlearn: 0.6231022\ttotal: 2.2s\tremaining: 5.8s\n",
      "275:\tlearn: 0.6230129\ttotal: 2.21s\tremaining: 5.79s\n",
      "276:\tlearn: 0.6229460\ttotal: 2.21s\tremaining: 5.78s\n",
      "277:\tlearn: 0.6228545\ttotal: 2.22s\tremaining: 5.77s\n",
      "278:\tlearn: 0.6227267\ttotal: 2.23s\tremaining: 5.76s\n",
      "279:\tlearn: 0.6226629\ttotal: 2.24s\tremaining: 5.75s\n",
      "280:\tlearn: 0.6225783\ttotal: 2.25s\tremaining: 5.75s\n",
      "281:\tlearn: 0.6224731\ttotal: 2.25s\tremaining: 5.74s\n",
      "282:\tlearn: 0.6224137\ttotal: 2.26s\tremaining: 5.73s\n",
      "283:\tlearn: 0.6224134\ttotal: 2.27s\tremaining: 5.71s\n",
      "284:\tlearn: 0.6223199\ttotal: 2.27s\tremaining: 5.7s\n",
      "285:\tlearn: 0.6222779\ttotal: 2.28s\tremaining: 5.69s\n",
      "286:\tlearn: 0.6222010\ttotal: 2.29s\tremaining: 5.68s\n",
      "287:\tlearn: 0.6221098\ttotal: 2.29s\tremaining: 5.67s\n",
      "288:\tlearn: 0.6220681\ttotal: 2.3s\tremaining: 5.66s\n",
      "289:\tlearn: 0.6219321\ttotal: 2.31s\tremaining: 5.65s\n",
      "290:\tlearn: 0.6218171\ttotal: 2.32s\tremaining: 5.64s\n",
      "291:\tlearn: 0.6217301\ttotal: 2.32s\tremaining: 5.63s\n",
      "292:\tlearn: 0.6216688\ttotal: 2.33s\tremaining: 5.62s\n",
      "293:\tlearn: 0.6216505\ttotal: 2.33s\tremaining: 5.61s\n",
      "294:\tlearn: 0.6215949\ttotal: 2.34s\tremaining: 5.6s\n",
      "295:\tlearn: 0.6215008\ttotal: 2.35s\tremaining: 5.59s\n",
      "296:\tlearn: 0.6213677\ttotal: 2.35s\tremaining: 5.58s\n",
      "297:\tlearn: 0.6212811\ttotal: 2.36s\tremaining: 5.57s\n",
      "298:\tlearn: 0.6211415\ttotal: 2.37s\tremaining: 5.56s\n",
      "299:\tlearn: 0.6211235\ttotal: 2.38s\tremaining: 5.55s\n",
      "300:\tlearn: 0.6210659\ttotal: 2.38s\tremaining: 5.53s\n",
      "301:\tlearn: 0.6210058\ttotal: 2.39s\tremaining: 5.52s\n",
      "302:\tlearn: 0.6208870\ttotal: 2.4s\tremaining: 5.51s\n",
      "303:\tlearn: 0.6208010\ttotal: 2.4s\tremaining: 5.5s\n",
      "304:\tlearn: 0.6206916\ttotal: 2.41s\tremaining: 5.5s\n",
      "305:\tlearn: 0.6205953\ttotal: 2.42s\tremaining: 5.49s\n",
      "306:\tlearn: 0.6205749\ttotal: 2.43s\tremaining: 5.49s\n",
      "307:\tlearn: 0.6205170\ttotal: 2.44s\tremaining: 5.48s\n",
      "308:\tlearn: 0.6204394\ttotal: 2.45s\tremaining: 5.47s\n",
      "309:\tlearn: 0.6203332\ttotal: 2.45s\tremaining: 5.46s\n",
      "310:\tlearn: 0.6202588\ttotal: 2.46s\tremaining: 5.46s\n",
      "311:\tlearn: 0.6201514\ttotal: 2.47s\tremaining: 5.45s\n",
      "312:\tlearn: 0.6200673\ttotal: 2.48s\tremaining: 5.44s\n",
      "313:\tlearn: 0.6199994\ttotal: 2.48s\tremaining: 5.43s\n",
      "314:\tlearn: 0.6198967\ttotal: 2.49s\tremaining: 5.42s\n",
      "315:\tlearn: 0.6197917\ttotal: 2.5s\tremaining: 5.41s\n",
      "316:\tlearn: 0.6196358\ttotal: 2.51s\tremaining: 5.4s\n",
      "317:\tlearn: 0.6195175\ttotal: 2.51s\tremaining: 5.39s\n",
      "318:\tlearn: 0.6194498\ttotal: 2.52s\tremaining: 5.38s\n",
      "319:\tlearn: 0.6194231\ttotal: 2.53s\tremaining: 5.37s\n",
      "320:\tlearn: 0.6194137\ttotal: 2.53s\tremaining: 5.36s\n",
      "321:\tlearn: 0.6193359\ttotal: 2.54s\tremaining: 5.35s\n",
      "322:\tlearn: 0.6192076\ttotal: 2.55s\tremaining: 5.34s\n",
      "323:\tlearn: 0.6191121\ttotal: 2.56s\tremaining: 5.33s\n",
      "324:\tlearn: 0.6190536\ttotal: 2.56s\tremaining: 5.32s\n",
      "325:\tlearn: 0.6189535\ttotal: 2.57s\tremaining: 5.31s\n",
      "326:\tlearn: 0.6187909\ttotal: 2.58s\tremaining: 5.3s\n",
      "327:\tlearn: 0.6186224\ttotal: 2.58s\tremaining: 5.29s\n",
      "328:\tlearn: 0.6185279\ttotal: 2.59s\tremaining: 5.28s\n",
      "329:\tlearn: 0.6183251\ttotal: 2.6s\tremaining: 5.27s\n",
      "330:\tlearn: 0.6182004\ttotal: 2.6s\tremaining: 5.26s\n",
      "331:\tlearn: 0.6180932\ttotal: 2.61s\tremaining: 5.25s\n",
      "332:\tlearn: 0.6180088\ttotal: 2.62s\tremaining: 5.24s\n",
      "333:\tlearn: 0.6178907\ttotal: 2.63s\tremaining: 5.24s\n",
      "334:\tlearn: 0.6177296\ttotal: 2.63s\tremaining: 5.23s\n",
      "335:\tlearn: 0.6176386\ttotal: 2.64s\tremaining: 5.22s\n",
      "336:\tlearn: 0.6175408\ttotal: 2.65s\tremaining: 5.21s\n",
      "337:\tlearn: 0.6174788\ttotal: 2.66s\tremaining: 5.21s\n",
      "338:\tlearn: 0.6173116\ttotal: 2.67s\tremaining: 5.2s\n",
      "339:\tlearn: 0.6172728\ttotal: 2.67s\tremaining: 5.19s\n",
      "340:\tlearn: 0.6171461\ttotal: 2.68s\tremaining: 5.18s\n",
      "341:\tlearn: 0.6170592\ttotal: 2.69s\tremaining: 5.17s\n",
      "342:\tlearn: 0.6169582\ttotal: 2.69s\tremaining: 5.16s\n",
      "343:\tlearn: 0.6168565\ttotal: 2.7s\tremaining: 5.15s\n",
      "344:\tlearn: 0.6167847\ttotal: 2.71s\tremaining: 5.14s\n",
      "345:\tlearn: 0.6166702\ttotal: 2.72s\tremaining: 5.14s\n",
      "346:\tlearn: 0.6165645\ttotal: 2.72s\tremaining: 5.13s\n",
      "347:\tlearn: 0.6164681\ttotal: 2.73s\tremaining: 5.12s\n",
      "348:\tlearn: 0.6163773\ttotal: 2.74s\tremaining: 5.11s\n",
      "349:\tlearn: 0.6162187\ttotal: 2.75s\tremaining: 5.1s\n",
      "350:\tlearn: 0.6160961\ttotal: 2.75s\tremaining: 5.09s\n",
      "351:\tlearn: 0.6159630\ttotal: 2.76s\tremaining: 5.08s\n",
      "352:\tlearn: 0.6158225\ttotal: 2.77s\tremaining: 5.07s\n",
      "353:\tlearn: 0.6156837\ttotal: 2.77s\tremaining: 5.06s\n",
      "354:\tlearn: 0.6156034\ttotal: 2.78s\tremaining: 5.06s\n",
      "355:\tlearn: 0.6155476\ttotal: 2.79s\tremaining: 5.04s\n",
      "356:\tlearn: 0.6154550\ttotal: 2.8s\tremaining: 5.04s\n",
      "357:\tlearn: 0.6152859\ttotal: 2.8s\tremaining: 5.03s\n",
      "358:\tlearn: 0.6151809\ttotal: 2.81s\tremaining: 5.01s\n",
      "359:\tlearn: 0.6151040\ttotal: 2.82s\tremaining: 5.01s\n",
      "360:\tlearn: 0.6150587\ttotal: 2.82s\tremaining: 5s\n",
      "361:\tlearn: 0.6149607\ttotal: 2.83s\tremaining: 4.99s\n",
      "362:\tlearn: 0.6148764\ttotal: 2.84s\tremaining: 4.98s\n",
      "363:\tlearn: 0.6147701\ttotal: 2.85s\tremaining: 4.98s\n",
      "364:\tlearn: 0.6146805\ttotal: 2.86s\tremaining: 4.97s\n",
      "365:\tlearn: 0.6145336\ttotal: 2.87s\tremaining: 4.96s\n",
      "366:\tlearn: 0.6144526\ttotal: 2.88s\tremaining: 4.96s\n",
      "367:\tlearn: 0.6143978\ttotal: 2.88s\tremaining: 4.95s\n",
      "368:\tlearn: 0.6142843\ttotal: 2.89s\tremaining: 4.95s\n",
      "369:\tlearn: 0.6141285\ttotal: 2.9s\tremaining: 4.93s\n",
      "370:\tlearn: 0.6140745\ttotal: 2.9s\tremaining: 4.92s\n",
      "371:\tlearn: 0.6139924\ttotal: 2.91s\tremaining: 4.92s\n",
      "372:\tlearn: 0.6138690\ttotal: 2.92s\tremaining: 4.91s\n",
      "373:\tlearn: 0.6137845\ttotal: 2.93s\tremaining: 4.9s\n",
      "374:\tlearn: 0.6137117\ttotal: 2.93s\tremaining: 4.89s\n",
      "375:\tlearn: 0.6136478\ttotal: 2.94s\tremaining: 4.88s\n",
      "376:\tlearn: 0.6135688\ttotal: 2.95s\tremaining: 4.87s\n",
      "377:\tlearn: 0.6134568\ttotal: 2.96s\tremaining: 4.87s\n",
      "378:\tlearn: 0.6133464\ttotal: 2.96s\tremaining: 4.86s\n",
      "379:\tlearn: 0.6131907\ttotal: 2.97s\tremaining: 4.85s\n",
      "380:\tlearn: 0.6130931\ttotal: 2.98s\tremaining: 4.84s\n",
      "381:\tlearn: 0.6129574\ttotal: 2.98s\tremaining: 4.83s\n",
      "382:\tlearn: 0.6128028\ttotal: 2.99s\tremaining: 4.82s\n",
      "383:\tlearn: 0.6126726\ttotal: 3s\tremaining: 4.81s\n",
      "384:\tlearn: 0.6125651\ttotal: 3.01s\tremaining: 4.8s\n",
      "385:\tlearn: 0.6124624\ttotal: 3.01s\tremaining: 4.79s\n",
      "386:\tlearn: 0.6123262\ttotal: 3.02s\tremaining: 4.78s\n",
      "387:\tlearn: 0.6122022\ttotal: 3.03s\tremaining: 4.78s\n",
      "388:\tlearn: 0.6120932\ttotal: 3.04s\tremaining: 4.77s\n",
      "389:\tlearn: 0.6119572\ttotal: 3.04s\tremaining: 4.76s\n",
      "390:\tlearn: 0.6118064\ttotal: 3.05s\tremaining: 4.75s\n",
      "391:\tlearn: 0.6116613\ttotal: 3.06s\tremaining: 4.75s\n",
      "392:\tlearn: 0.6115714\ttotal: 3.07s\tremaining: 4.74s\n",
      "393:\tlearn: 0.6114893\ttotal: 3.07s\tremaining: 4.73s\n",
      "394:\tlearn: 0.6113329\ttotal: 3.08s\tremaining: 4.72s\n",
      "395:\tlearn: 0.6112360\ttotal: 3.09s\tremaining: 4.71s\n",
      "396:\tlearn: 0.6110534\ttotal: 3.1s\tremaining: 4.7s\n",
      "397:\tlearn: 0.6109435\ttotal: 3.1s\tremaining: 4.69s\n",
      "398:\tlearn: 0.6108353\ttotal: 3.11s\tremaining: 4.68s\n",
      "399:\tlearn: 0.6107372\ttotal: 3.12s\tremaining: 4.67s\n",
      "400:\tlearn: 0.6105570\ttotal: 3.12s\tremaining: 4.67s\n",
      "401:\tlearn: 0.6103777\ttotal: 3.13s\tremaining: 4.66s\n",
      "402:\tlearn: 0.6102974\ttotal: 3.14s\tremaining: 4.65s\n",
      "403:\tlearn: 0.6101646\ttotal: 3.15s\tremaining: 4.64s\n",
      "404:\tlearn: 0.6100188\ttotal: 3.15s\tremaining: 4.63s\n",
      "405:\tlearn: 0.6099140\ttotal: 3.16s\tremaining: 4.62s\n",
      "406:\tlearn: 0.6098218\ttotal: 3.17s\tremaining: 4.61s\n",
      "407:\tlearn: 0.6097011\ttotal: 3.17s\tremaining: 4.6s\n",
      "408:\tlearn: 0.6095641\ttotal: 3.18s\tremaining: 4.59s\n",
      "409:\tlearn: 0.6094538\ttotal: 3.19s\tremaining: 4.59s\n",
      "410:\tlearn: 0.6093418\ttotal: 3.19s\tremaining: 4.58s\n",
      "411:\tlearn: 0.6092754\ttotal: 3.2s\tremaining: 4.57s\n",
      "412:\tlearn: 0.6091584\ttotal: 3.21s\tremaining: 4.56s\n",
      "413:\tlearn: 0.6089718\ttotal: 3.21s\tremaining: 4.55s\n",
      "414:\tlearn: 0.6088441\ttotal: 3.22s\tremaining: 4.54s\n",
      "415:\tlearn: 0.6087351\ttotal: 3.23s\tremaining: 4.54s\n",
      "416:\tlearn: 0.6085909\ttotal: 3.24s\tremaining: 4.53s\n",
      "417:\tlearn: 0.6084237\ttotal: 3.24s\tremaining: 4.52s\n",
      "418:\tlearn: 0.6082488\ttotal: 3.26s\tremaining: 4.51s\n",
      "419:\tlearn: 0.6081430\ttotal: 3.26s\tremaining: 4.51s\n",
      "420:\tlearn: 0.6080223\ttotal: 3.27s\tremaining: 4.5s\n",
      "421:\tlearn: 0.6078859\ttotal: 3.28s\tremaining: 4.5s\n",
      "422:\tlearn: 0.6077926\ttotal: 3.29s\tremaining: 4.49s\n",
      "423:\tlearn: 0.6076833\ttotal: 3.3s\tremaining: 4.48s\n",
      "424:\tlearn: 0.6075478\ttotal: 3.31s\tremaining: 4.47s\n",
      "425:\tlearn: 0.6074659\ttotal: 3.32s\tremaining: 4.47s\n",
      "426:\tlearn: 0.6073458\ttotal: 3.33s\tremaining: 4.46s\n",
      "427:\tlearn: 0.6072613\ttotal: 3.33s\tremaining: 4.45s\n",
      "428:\tlearn: 0.6071261\ttotal: 3.34s\tremaining: 4.44s\n",
      "429:\tlearn: 0.6070252\ttotal: 3.35s\tremaining: 4.44s\n",
      "430:\tlearn: 0.6069480\ttotal: 3.35s\tremaining: 4.42s\n",
      "431:\tlearn: 0.6068180\ttotal: 3.36s\tremaining: 4.42s\n",
      "432:\tlearn: 0.6066813\ttotal: 3.37s\tremaining: 4.41s\n",
      "433:\tlearn: 0.6065613\ttotal: 3.37s\tremaining: 4.4s\n",
      "434:\tlearn: 0.6064283\ttotal: 3.38s\tremaining: 4.39s\n",
      "435:\tlearn: 0.6062976\ttotal: 3.39s\tremaining: 4.38s\n",
      "436:\tlearn: 0.6062151\ttotal: 3.39s\tremaining: 4.37s\n",
      "437:\tlearn: 0.6061185\ttotal: 3.4s\tremaining: 4.36s\n",
      "438:\tlearn: 0.6059799\ttotal: 3.41s\tremaining: 4.35s\n",
      "439:\tlearn: 0.6058736\ttotal: 3.41s\tremaining: 4.34s\n",
      "440:\tlearn: 0.6057262\ttotal: 3.42s\tremaining: 4.33s\n",
      "441:\tlearn: 0.6056079\ttotal: 3.43s\tremaining: 4.33s\n",
      "442:\tlearn: 0.6054761\ttotal: 3.44s\tremaining: 4.32s\n",
      "443:\tlearn: 0.6053800\ttotal: 3.44s\tremaining: 4.32s\n",
      "444:\tlearn: 0.6052545\ttotal: 3.45s\tremaining: 4.31s\n",
      "445:\tlearn: 0.6051575\ttotal: 3.46s\tremaining: 4.3s\n",
      "446:\tlearn: 0.6049968\ttotal: 3.47s\tremaining: 4.29s\n",
      "447:\tlearn: 0.6047733\ttotal: 3.48s\tremaining: 4.28s\n",
      "448:\tlearn: 0.6046634\ttotal: 3.48s\tremaining: 4.27s\n",
      "449:\tlearn: 0.6045636\ttotal: 3.49s\tremaining: 4.27s\n",
      "450:\tlearn: 0.6044651\ttotal: 3.5s\tremaining: 4.26s\n",
      "451:\tlearn: 0.6043453\ttotal: 3.5s\tremaining: 4.25s\n",
      "452:\tlearn: 0.6042341\ttotal: 3.51s\tremaining: 4.24s\n",
      "453:\tlearn: 0.6040636\ttotal: 3.52s\tremaining: 4.23s\n",
      "454:\tlearn: 0.6039605\ttotal: 3.52s\tremaining: 4.22s\n",
      "455:\tlearn: 0.6038125\ttotal: 3.53s\tremaining: 4.21s\n",
      "456:\tlearn: 0.6036541\ttotal: 3.54s\tremaining: 4.21s\n",
      "457:\tlearn: 0.6035369\ttotal: 3.54s\tremaining: 4.2s\n",
      "458:\tlearn: 0.6034329\ttotal: 3.55s\tremaining: 4.19s\n",
      "459:\tlearn: 0.6033367\ttotal: 3.56s\tremaining: 4.18s\n",
      "460:\tlearn: 0.6031978\ttotal: 3.56s\tremaining: 4.17s\n",
      "461:\tlearn: 0.6030393\ttotal: 3.57s\tremaining: 4.16s\n",
      "462:\tlearn: 0.6029497\ttotal: 3.58s\tremaining: 4.15s\n",
      "463:\tlearn: 0.6028240\ttotal: 3.59s\tremaining: 4.14s\n",
      "464:\tlearn: 0.6026933\ttotal: 3.59s\tremaining: 4.13s\n",
      "465:\tlearn: 0.6025476\ttotal: 3.6s\tremaining: 4.13s\n",
      "466:\tlearn: 0.6024418\ttotal: 3.61s\tremaining: 4.12s\n",
      "467:\tlearn: 0.6023436\ttotal: 3.61s\tremaining: 4.11s\n",
      "468:\tlearn: 0.6021534\ttotal: 3.62s\tremaining: 4.1s\n",
      "469:\tlearn: 0.6020176\ttotal: 3.63s\tremaining: 4.09s\n",
      "470:\tlearn: 0.6019534\ttotal: 3.64s\tremaining: 4.08s\n",
      "471:\tlearn: 0.6018334\ttotal: 3.65s\tremaining: 4.08s\n",
      "472:\tlearn: 0.6017368\ttotal: 3.66s\tremaining: 4.07s\n",
      "473:\tlearn: 0.6016133\ttotal: 3.66s\tremaining: 4.07s\n",
      "474:\tlearn: 0.6014353\ttotal: 3.67s\tremaining: 4.06s\n",
      "475:\tlearn: 0.6013256\ttotal: 3.68s\tremaining: 4.05s\n",
      "476:\tlearn: 0.6012334\ttotal: 3.69s\tremaining: 4.04s\n",
      "477:\tlearn: 0.6010765\ttotal: 3.7s\tremaining: 4.04s\n",
      "478:\tlearn: 0.6009982\ttotal: 3.71s\tremaining: 4.03s\n",
      "479:\tlearn: 0.6008946\ttotal: 3.71s\tremaining: 4.02s\n",
      "480:\tlearn: 0.6007719\ttotal: 3.72s\tremaining: 4.01s\n",
      "481:\tlearn: 0.6006299\ttotal: 3.73s\tremaining: 4s\n",
      "482:\tlearn: 0.6004806\ttotal: 3.73s\tremaining: 4s\n",
      "483:\tlearn: 0.6003800\ttotal: 3.74s\tremaining: 3.99s\n",
      "484:\tlearn: 0.6002206\ttotal: 3.75s\tremaining: 3.98s\n",
      "485:\tlearn: 0.6001376\ttotal: 3.75s\tremaining: 3.97s\n",
      "486:\tlearn: 0.5999877\ttotal: 3.76s\tremaining: 3.96s\n",
      "487:\tlearn: 0.5998674\ttotal: 3.77s\tremaining: 3.95s\n",
      "488:\tlearn: 0.5997307\ttotal: 3.77s\tremaining: 3.94s\n",
      "489:\tlearn: 0.5996072\ttotal: 3.78s\tremaining: 3.94s\n",
      "490:\tlearn: 0.5994834\ttotal: 3.79s\tremaining: 3.93s\n",
      "491:\tlearn: 0.5992967\ttotal: 3.79s\tremaining: 3.92s\n",
      "492:\tlearn: 0.5991764\ttotal: 3.8s\tremaining: 3.91s\n",
      "493:\tlearn: 0.5990214\ttotal: 3.81s\tremaining: 3.9s\n",
      "494:\tlearn: 0.5988895\ttotal: 3.82s\tremaining: 3.89s\n",
      "495:\tlearn: 0.5987234\ttotal: 3.82s\tremaining: 3.88s\n",
      "496:\tlearn: 0.5985755\ttotal: 3.83s\tremaining: 3.88s\n",
      "497:\tlearn: 0.5984420\ttotal: 3.84s\tremaining: 3.87s\n",
      "498:\tlearn: 0.5982980\ttotal: 3.85s\tremaining: 3.87s\n",
      "499:\tlearn: 0.5981391\ttotal: 3.86s\tremaining: 3.86s\n",
      "500:\tlearn: 0.5980544\ttotal: 3.86s\tremaining: 3.85s\n",
      "501:\tlearn: 0.5979227\ttotal: 3.87s\tremaining: 3.84s\n",
      "502:\tlearn: 0.5978205\ttotal: 3.88s\tremaining: 3.83s\n",
      "503:\tlearn: 0.5976917\ttotal: 3.88s\tremaining: 3.82s\n",
      "504:\tlearn: 0.5976124\ttotal: 3.89s\tremaining: 3.81s\n",
      "505:\tlearn: 0.5975021\ttotal: 3.9s\tremaining: 3.81s\n",
      "506:\tlearn: 0.5973946\ttotal: 3.9s\tremaining: 3.8s\n",
      "507:\tlearn: 0.5972559\ttotal: 3.92s\tremaining: 3.79s\n",
      "508:\tlearn: 0.5971173\ttotal: 3.92s\tremaining: 3.78s\n",
      "509:\tlearn: 0.5970291\ttotal: 3.93s\tremaining: 3.77s\n",
      "510:\tlearn: 0.5968682\ttotal: 3.94s\tremaining: 3.77s\n",
      "511:\tlearn: 0.5967642\ttotal: 3.94s\tremaining: 3.76s\n",
      "512:\tlearn: 0.5966291\ttotal: 3.95s\tremaining: 3.75s\n",
      "513:\tlearn: 0.5965059\ttotal: 3.96s\tremaining: 3.74s\n",
      "514:\tlearn: 0.5964390\ttotal: 3.96s\tremaining: 3.73s\n",
      "515:\tlearn: 0.5963092\ttotal: 3.97s\tremaining: 3.72s\n",
      "516:\tlearn: 0.5961658\ttotal: 3.98s\tremaining: 3.71s\n",
      "517:\tlearn: 0.5960508\ttotal: 3.98s\tremaining: 3.71s\n",
      "518:\tlearn: 0.5959315\ttotal: 3.99s\tremaining: 3.7s\n",
      "519:\tlearn: 0.5957635\ttotal: 4s\tremaining: 3.69s\n",
      "520:\tlearn: 0.5956797\ttotal: 4s\tremaining: 3.68s\n",
      "521:\tlearn: 0.5955669\ttotal: 4.01s\tremaining: 3.67s\n",
      "522:\tlearn: 0.5954556\ttotal: 4.02s\tremaining: 3.66s\n",
      "523:\tlearn: 0.5953606\ttotal: 4.03s\tremaining: 3.66s\n",
      "524:\tlearn: 0.5952042\ttotal: 4.03s\tremaining: 3.65s\n",
      "525:\tlearn: 0.5951002\ttotal: 4.04s\tremaining: 3.64s\n",
      "526:\tlearn: 0.5949886\ttotal: 4.05s\tremaining: 3.63s\n",
      "527:\tlearn: 0.5948884\ttotal: 4.06s\tremaining: 3.63s\n",
      "528:\tlearn: 0.5947605\ttotal: 4.07s\tremaining: 3.62s\n",
      "529:\tlearn: 0.5946631\ttotal: 4.07s\tremaining: 3.61s\n",
      "530:\tlearn: 0.5945524\ttotal: 4.08s\tremaining: 3.6s\n",
      "531:\tlearn: 0.5943836\ttotal: 4.09s\tremaining: 3.6s\n",
      "532:\tlearn: 0.5942627\ttotal: 4.1s\tremaining: 3.59s\n",
      "533:\tlearn: 0.5941627\ttotal: 4.11s\tremaining: 3.58s\n",
      "534:\tlearn: 0.5940683\ttotal: 4.11s\tremaining: 3.57s\n",
      "535:\tlearn: 0.5939645\ttotal: 4.12s\tremaining: 3.56s\n",
      "536:\tlearn: 0.5938485\ttotal: 4.13s\tremaining: 3.56s\n",
      "537:\tlearn: 0.5936479\ttotal: 4.13s\tremaining: 3.55s\n",
      "538:\tlearn: 0.5935016\ttotal: 4.14s\tremaining: 3.54s\n",
      "539:\tlearn: 0.5933795\ttotal: 4.15s\tremaining: 3.53s\n",
      "540:\tlearn: 0.5932745\ttotal: 4.15s\tremaining: 3.52s\n",
      "541:\tlearn: 0.5931634\ttotal: 4.16s\tremaining: 3.52s\n",
      "542:\tlearn: 0.5929716\ttotal: 4.17s\tremaining: 3.51s\n",
      "543:\tlearn: 0.5928426\ttotal: 4.17s\tremaining: 3.5s\n",
      "544:\tlearn: 0.5927103\ttotal: 4.18s\tremaining: 3.49s\n",
      "545:\tlearn: 0.5926064\ttotal: 4.19s\tremaining: 3.48s\n",
      "546:\tlearn: 0.5925052\ttotal: 4.19s\tremaining: 3.47s\n",
      "547:\tlearn: 0.5923872\ttotal: 4.2s\tremaining: 3.46s\n",
      "548:\tlearn: 0.5923369\ttotal: 4.21s\tremaining: 3.46s\n",
      "549:\tlearn: 0.5922508\ttotal: 4.21s\tremaining: 3.45s\n",
      "550:\tlearn: 0.5921948\ttotal: 4.22s\tremaining: 3.44s\n",
      "551:\tlearn: 0.5920652\ttotal: 4.23s\tremaining: 3.43s\n",
      "552:\tlearn: 0.5919931\ttotal: 4.24s\tremaining: 3.42s\n",
      "553:\tlearn: 0.5918148\ttotal: 4.25s\tremaining: 3.42s\n",
      "554:\tlearn: 0.5916174\ttotal: 4.25s\tremaining: 3.41s\n",
      "555:\tlearn: 0.5914966\ttotal: 4.26s\tremaining: 3.4s\n",
      "556:\tlearn: 0.5913527\ttotal: 4.27s\tremaining: 3.4s\n",
      "557:\tlearn: 0.5912402\ttotal: 4.28s\tremaining: 3.39s\n",
      "558:\tlearn: 0.5910697\ttotal: 4.28s\tremaining: 3.38s\n",
      "559:\tlearn: 0.5909501\ttotal: 4.29s\tremaining: 3.37s\n",
      "560:\tlearn: 0.5908563\ttotal: 4.3s\tremaining: 3.36s\n",
      "561:\tlearn: 0.5907530\ttotal: 4.3s\tremaining: 3.35s\n",
      "562:\tlearn: 0.5906303\ttotal: 4.31s\tremaining: 3.34s\n",
      "563:\tlearn: 0.5904797\ttotal: 4.32s\tremaining: 3.34s\n",
      "564:\tlearn: 0.5903670\ttotal: 4.32s\tremaining: 3.33s\n",
      "565:\tlearn: 0.5902195\ttotal: 4.33s\tremaining: 3.32s\n",
      "566:\tlearn: 0.5900741\ttotal: 4.34s\tremaining: 3.31s\n",
      "567:\tlearn: 0.5899710\ttotal: 4.34s\tremaining: 3.3s\n",
      "568:\tlearn: 0.5898379\ttotal: 4.35s\tremaining: 3.3s\n",
      "569:\tlearn: 0.5897548\ttotal: 4.36s\tremaining: 3.29s\n",
      "570:\tlearn: 0.5896173\ttotal: 4.37s\tremaining: 3.28s\n",
      "571:\tlearn: 0.5894775\ttotal: 4.37s\tremaining: 3.27s\n",
      "572:\tlearn: 0.5893771\ttotal: 4.38s\tremaining: 3.26s\n",
      "573:\tlearn: 0.5892522\ttotal: 4.39s\tremaining: 3.25s\n",
      "574:\tlearn: 0.5891324\ttotal: 4.39s\tremaining: 3.25s\n",
      "575:\tlearn: 0.5890373\ttotal: 4.4s\tremaining: 3.24s\n",
      "576:\tlearn: 0.5889818\ttotal: 4.41s\tremaining: 3.23s\n",
      "577:\tlearn: 0.5889059\ttotal: 4.41s\tremaining: 3.22s\n",
      "578:\tlearn: 0.5888383\ttotal: 4.42s\tremaining: 3.21s\n",
      "579:\tlearn: 0.5887157\ttotal: 4.43s\tremaining: 3.21s\n",
      "580:\tlearn: 0.5886433\ttotal: 4.44s\tremaining: 3.2s\n",
      "581:\tlearn: 0.5885152\ttotal: 4.45s\tremaining: 3.19s\n",
      "582:\tlearn: 0.5883709\ttotal: 4.45s\tremaining: 3.19s\n",
      "583:\tlearn: 0.5882496\ttotal: 4.46s\tremaining: 3.18s\n",
      "584:\tlearn: 0.5881249\ttotal: 4.47s\tremaining: 3.17s\n",
      "585:\tlearn: 0.5880215\ttotal: 4.47s\tremaining: 3.16s\n",
      "586:\tlearn: 0.5879502\ttotal: 4.48s\tremaining: 3.15s\n",
      "587:\tlearn: 0.5878377\ttotal: 4.49s\tremaining: 3.14s\n",
      "588:\tlearn: 0.5877214\ttotal: 4.5s\tremaining: 3.14s\n",
      "589:\tlearn: 0.5875842\ttotal: 4.5s\tremaining: 3.13s\n",
      "590:\tlearn: 0.5874585\ttotal: 4.51s\tremaining: 3.12s\n",
      "591:\tlearn: 0.5873432\ttotal: 4.52s\tremaining: 3.11s\n",
      "592:\tlearn: 0.5872577\ttotal: 4.52s\tremaining: 3.1s\n",
      "593:\tlearn: 0.5872010\ttotal: 4.53s\tremaining: 3.1s\n",
      "594:\tlearn: 0.5871041\ttotal: 4.54s\tremaining: 3.09s\n",
      "595:\tlearn: 0.5870331\ttotal: 4.54s\tremaining: 3.08s\n",
      "596:\tlearn: 0.5869309\ttotal: 4.55s\tremaining: 3.07s\n",
      "597:\tlearn: 0.5867545\ttotal: 4.56s\tremaining: 3.06s\n",
      "598:\tlearn: 0.5866190\ttotal: 4.57s\tremaining: 3.06s\n",
      "599:\tlearn: 0.5864962\ttotal: 4.57s\tremaining: 3.05s\n",
      "600:\tlearn: 0.5864005\ttotal: 4.58s\tremaining: 3.04s\n",
      "601:\tlearn: 0.5863260\ttotal: 4.59s\tremaining: 3.03s\n",
      "602:\tlearn: 0.5862158\ttotal: 4.59s\tremaining: 3.02s\n",
      "603:\tlearn: 0.5860974\ttotal: 4.6s\tremaining: 3.02s\n",
      "604:\tlearn: 0.5859857\ttotal: 4.61s\tremaining: 3.01s\n",
      "605:\tlearn: 0.5858451\ttotal: 4.61s\tremaining: 3s\n",
      "606:\tlearn: 0.5856884\ttotal: 4.62s\tremaining: 2.99s\n",
      "607:\tlearn: 0.5855699\ttotal: 4.63s\tremaining: 2.98s\n",
      "608:\tlearn: 0.5854583\ttotal: 4.64s\tremaining: 2.98s\n",
      "609:\tlearn: 0.5853941\ttotal: 4.65s\tremaining: 2.97s\n",
      "610:\tlearn: 0.5852947\ttotal: 4.66s\tremaining: 2.96s\n",
      "611:\tlearn: 0.5852085\ttotal: 4.67s\tremaining: 2.96s\n",
      "612:\tlearn: 0.5850987\ttotal: 4.68s\tremaining: 2.95s\n",
      "613:\tlearn: 0.5849761\ttotal: 4.68s\tremaining: 2.94s\n",
      "614:\tlearn: 0.5848559\ttotal: 4.69s\tremaining: 2.94s\n",
      "615:\tlearn: 0.5847378\ttotal: 4.7s\tremaining: 2.93s\n",
      "616:\tlearn: 0.5846408\ttotal: 4.7s\tremaining: 2.92s\n",
      "617:\tlearn: 0.5845696\ttotal: 4.71s\tremaining: 2.91s\n",
      "618:\tlearn: 0.5844550\ttotal: 4.72s\tremaining: 2.9s\n",
      "619:\tlearn: 0.5843256\ttotal: 4.72s\tremaining: 2.9s\n",
      "620:\tlearn: 0.5842302\ttotal: 4.73s\tremaining: 2.89s\n",
      "621:\tlearn: 0.5841151\ttotal: 4.74s\tremaining: 2.88s\n",
      "622:\tlearn: 0.5839500\ttotal: 4.75s\tremaining: 2.87s\n",
      "623:\tlearn: 0.5838703\ttotal: 4.75s\tremaining: 2.87s\n",
      "624:\tlearn: 0.5837746\ttotal: 4.76s\tremaining: 2.86s\n",
      "625:\tlearn: 0.5836320\ttotal: 4.77s\tremaining: 2.85s\n",
      "626:\tlearn: 0.5835164\ttotal: 4.78s\tremaining: 2.84s\n",
      "627:\tlearn: 0.5834569\ttotal: 4.78s\tremaining: 2.83s\n",
      "628:\tlearn: 0.5833289\ttotal: 4.79s\tremaining: 2.83s\n",
      "629:\tlearn: 0.5832418\ttotal: 4.79s\tremaining: 2.82s\n",
      "630:\tlearn: 0.5831804\ttotal: 4.8s\tremaining: 2.81s\n",
      "631:\tlearn: 0.5830793\ttotal: 4.81s\tremaining: 2.8s\n",
      "632:\tlearn: 0.5829790\ttotal: 4.82s\tremaining: 2.79s\n",
      "633:\tlearn: 0.5828694\ttotal: 4.82s\tremaining: 2.78s\n",
      "634:\tlearn: 0.5827961\ttotal: 4.83s\tremaining: 2.78s\n",
      "635:\tlearn: 0.5826980\ttotal: 4.84s\tremaining: 2.77s\n",
      "636:\tlearn: 0.5825582\ttotal: 4.84s\tremaining: 2.76s\n",
      "637:\tlearn: 0.5824415\ttotal: 4.85s\tremaining: 2.75s\n",
      "638:\tlearn: 0.5823248\ttotal: 4.86s\tremaining: 2.75s\n",
      "639:\tlearn: 0.5822094\ttotal: 4.87s\tremaining: 2.74s\n",
      "640:\tlearn: 0.5821287\ttotal: 4.88s\tremaining: 2.73s\n",
      "641:\tlearn: 0.5820547\ttotal: 4.89s\tremaining: 2.73s\n",
      "642:\tlearn: 0.5819544\ttotal: 4.9s\tremaining: 2.72s\n",
      "643:\tlearn: 0.5818029\ttotal: 4.92s\tremaining: 2.72s\n",
      "644:\tlearn: 0.5817016\ttotal: 4.93s\tremaining: 2.71s\n",
      "645:\tlearn: 0.5816104\ttotal: 4.94s\tremaining: 2.7s\n",
      "646:\tlearn: 0.5815438\ttotal: 4.95s\tremaining: 2.7s\n",
      "647:\tlearn: 0.5813847\ttotal: 4.96s\tremaining: 2.69s\n",
      "648:\tlearn: 0.5812713\ttotal: 4.97s\tremaining: 2.69s\n",
      "649:\tlearn: 0.5811613\ttotal: 4.97s\tremaining: 2.68s\n",
      "650:\tlearn: 0.5810380\ttotal: 4.98s\tremaining: 2.67s\n",
      "651:\tlearn: 0.5809747\ttotal: 4.99s\tremaining: 2.66s\n",
      "652:\tlearn: 0.5808873\ttotal: 5s\tremaining: 2.65s\n",
      "653:\tlearn: 0.5807255\ttotal: 5s\tremaining: 2.65s\n",
      "654:\tlearn: 0.5806091\ttotal: 5.01s\tremaining: 2.64s\n",
      "655:\tlearn: 0.5805400\ttotal: 5.01s\tremaining: 2.63s\n",
      "656:\tlearn: 0.5804082\ttotal: 5.02s\tremaining: 2.62s\n",
      "657:\tlearn: 0.5802467\ttotal: 5.03s\tremaining: 2.62s\n",
      "658:\tlearn: 0.5801508\ttotal: 5.04s\tremaining: 2.61s\n",
      "659:\tlearn: 0.5800394\ttotal: 5.05s\tremaining: 2.6s\n",
      "660:\tlearn: 0.5799365\ttotal: 5.05s\tremaining: 2.59s\n",
      "661:\tlearn: 0.5798567\ttotal: 5.06s\tremaining: 2.58s\n",
      "662:\tlearn: 0.5797560\ttotal: 5.07s\tremaining: 2.58s\n",
      "663:\tlearn: 0.5795873\ttotal: 5.08s\tremaining: 2.57s\n",
      "664:\tlearn: 0.5794841\ttotal: 5.09s\tremaining: 2.56s\n",
      "665:\tlearn: 0.5793946\ttotal: 5.1s\tremaining: 2.56s\n",
      "666:\tlearn: 0.5793162\ttotal: 5.11s\tremaining: 2.55s\n",
      "667:\tlearn: 0.5792161\ttotal: 5.11s\tremaining: 2.54s\n",
      "668:\tlearn: 0.5791265\ttotal: 5.12s\tremaining: 2.53s\n",
      "669:\tlearn: 0.5790096\ttotal: 5.13s\tremaining: 2.52s\n",
      "670:\tlearn: 0.5789494\ttotal: 5.13s\tremaining: 2.52s\n",
      "671:\tlearn: 0.5788326\ttotal: 5.14s\tremaining: 2.51s\n",
      "672:\tlearn: 0.5787334\ttotal: 5.15s\tremaining: 2.5s\n",
      "673:\tlearn: 0.5786283\ttotal: 5.15s\tremaining: 2.49s\n",
      "674:\tlearn: 0.5785498\ttotal: 5.16s\tremaining: 2.48s\n",
      "675:\tlearn: 0.5784741\ttotal: 5.17s\tremaining: 2.48s\n",
      "676:\tlearn: 0.5783575\ttotal: 5.17s\tremaining: 2.47s\n",
      "677:\tlearn: 0.5782678\ttotal: 5.18s\tremaining: 2.46s\n",
      "678:\tlearn: 0.5781373\ttotal: 5.19s\tremaining: 2.45s\n",
      "679:\tlearn: 0.5780470\ttotal: 5.2s\tremaining: 2.44s\n",
      "680:\tlearn: 0.5779348\ttotal: 5.2s\tremaining: 2.44s\n",
      "681:\tlearn: 0.5778166\ttotal: 5.21s\tremaining: 2.43s\n",
      "682:\tlearn: 0.5777032\ttotal: 5.22s\tremaining: 2.42s\n",
      "683:\tlearn: 0.5776160\ttotal: 5.22s\tremaining: 2.41s\n",
      "684:\tlearn: 0.5775475\ttotal: 5.23s\tremaining: 2.4s\n",
      "685:\tlearn: 0.5774518\ttotal: 5.24s\tremaining: 2.4s\n",
      "686:\tlearn: 0.5773501\ttotal: 5.25s\tremaining: 2.39s\n",
      "687:\tlearn: 0.5772194\ttotal: 5.25s\tremaining: 2.38s\n",
      "688:\tlearn: 0.5771647\ttotal: 5.26s\tremaining: 2.37s\n",
      "689:\tlearn: 0.5770358\ttotal: 5.27s\tremaining: 2.37s\n",
      "690:\tlearn: 0.5768803\ttotal: 5.28s\tremaining: 2.36s\n",
      "691:\tlearn: 0.5767548\ttotal: 5.28s\tremaining: 2.35s\n",
      "692:\tlearn: 0.5766618\ttotal: 5.29s\tremaining: 2.34s\n",
      "693:\tlearn: 0.5766137\ttotal: 5.3s\tremaining: 2.34s\n",
      "694:\tlearn: 0.5765123\ttotal: 5.31s\tremaining: 2.33s\n",
      "695:\tlearn: 0.5764031\ttotal: 5.32s\tremaining: 2.32s\n",
      "696:\tlearn: 0.5763137\ttotal: 5.32s\tremaining: 2.31s\n",
      "697:\tlearn: 0.5762380\ttotal: 5.33s\tremaining: 2.31s\n",
      "698:\tlearn: 0.5761355\ttotal: 5.34s\tremaining: 2.3s\n",
      "699:\tlearn: 0.5760014\ttotal: 5.34s\tremaining: 2.29s\n",
      "700:\tlearn: 0.5759154\ttotal: 5.35s\tremaining: 2.28s\n",
      "701:\tlearn: 0.5758208\ttotal: 5.36s\tremaining: 2.27s\n",
      "702:\tlearn: 0.5757002\ttotal: 5.37s\tremaining: 2.27s\n",
      "703:\tlearn: 0.5755925\ttotal: 5.37s\tremaining: 2.26s\n",
      "704:\tlearn: 0.5754941\ttotal: 5.38s\tremaining: 2.25s\n",
      "705:\tlearn: 0.5753918\ttotal: 5.39s\tremaining: 2.24s\n",
      "706:\tlearn: 0.5753142\ttotal: 5.39s\tremaining: 2.23s\n",
      "707:\tlearn: 0.5752282\ttotal: 5.4s\tremaining: 2.23s\n",
      "708:\tlearn: 0.5751346\ttotal: 5.41s\tremaining: 2.22s\n",
      "709:\tlearn: 0.5750134\ttotal: 5.42s\tremaining: 2.21s\n",
      "710:\tlearn: 0.5749471\ttotal: 5.42s\tremaining: 2.2s\n",
      "711:\tlearn: 0.5748562\ttotal: 5.43s\tremaining: 2.19s\n",
      "712:\tlearn: 0.5747700\ttotal: 5.43s\tremaining: 2.19s\n",
      "713:\tlearn: 0.5746488\ttotal: 5.44s\tremaining: 2.18s\n",
      "714:\tlearn: 0.5745636\ttotal: 5.45s\tremaining: 2.17s\n",
      "715:\tlearn: 0.5744619\ttotal: 5.46s\tremaining: 2.16s\n",
      "716:\tlearn: 0.5743668\ttotal: 5.46s\tremaining: 2.16s\n",
      "717:\tlearn: 0.5742099\ttotal: 5.47s\tremaining: 2.15s\n",
      "718:\tlearn: 0.5740596\ttotal: 5.49s\tremaining: 2.14s\n",
      "719:\tlearn: 0.5739826\ttotal: 5.49s\tremaining: 2.14s\n",
      "720:\tlearn: 0.5738619\ttotal: 5.5s\tremaining: 2.13s\n",
      "721:\tlearn: 0.5737804\ttotal: 5.51s\tremaining: 2.12s\n",
      "722:\tlearn: 0.5737039\ttotal: 5.52s\tremaining: 2.11s\n",
      "723:\tlearn: 0.5736068\ttotal: 5.52s\tremaining: 2.1s\n",
      "724:\tlearn: 0.5734651\ttotal: 5.53s\tremaining: 2.1s\n",
      "725:\tlearn: 0.5733302\ttotal: 5.54s\tremaining: 2.09s\n",
      "726:\tlearn: 0.5732537\ttotal: 5.55s\tremaining: 2.08s\n",
      "727:\tlearn: 0.5731570\ttotal: 5.55s\tremaining: 2.07s\n",
      "728:\tlearn: 0.5730516\ttotal: 5.56s\tremaining: 2.07s\n",
      "729:\tlearn: 0.5729378\ttotal: 5.57s\tremaining: 2.06s\n",
      "730:\tlearn: 0.5728399\ttotal: 5.57s\tremaining: 2.05s\n",
      "731:\tlearn: 0.5726551\ttotal: 5.58s\tremaining: 2.04s\n",
      "732:\tlearn: 0.5725455\ttotal: 5.59s\tremaining: 2.04s\n",
      "733:\tlearn: 0.5724912\ttotal: 5.6s\tremaining: 2.03s\n",
      "734:\tlearn: 0.5724246\ttotal: 5.6s\tremaining: 2.02s\n",
      "735:\tlearn: 0.5723705\ttotal: 5.61s\tremaining: 2.01s\n",
      "736:\tlearn: 0.5722885\ttotal: 5.62s\tremaining: 2s\n",
      "737:\tlearn: 0.5722017\ttotal: 5.62s\tremaining: 2s\n",
      "738:\tlearn: 0.5721472\ttotal: 5.63s\tremaining: 1.99s\n",
      "739:\tlearn: 0.5721019\ttotal: 5.64s\tremaining: 1.98s\n",
      "740:\tlearn: 0.5720041\ttotal: 5.64s\tremaining: 1.97s\n",
      "741:\tlearn: 0.5719527\ttotal: 5.65s\tremaining: 1.97s\n",
      "742:\tlearn: 0.5718108\ttotal: 5.66s\tremaining: 1.96s\n",
      "743:\tlearn: 0.5717351\ttotal: 5.67s\tremaining: 1.95s\n",
      "744:\tlearn: 0.5716207\ttotal: 5.67s\tremaining: 1.94s\n",
      "745:\tlearn: 0.5715327\ttotal: 5.68s\tremaining: 1.94s\n",
      "746:\tlearn: 0.5714263\ttotal: 5.69s\tremaining: 1.93s\n",
      "747:\tlearn: 0.5713303\ttotal: 5.7s\tremaining: 1.92s\n",
      "748:\tlearn: 0.5712383\ttotal: 5.71s\tremaining: 1.91s\n",
      "749:\tlearn: 0.5711214\ttotal: 5.72s\tremaining: 1.91s\n",
      "750:\tlearn: 0.5710500\ttotal: 5.73s\tremaining: 1.9s\n",
      "751:\tlearn: 0.5709588\ttotal: 5.73s\tremaining: 1.89s\n",
      "752:\tlearn: 0.5708724\ttotal: 5.74s\tremaining: 1.88s\n",
      "753:\tlearn: 0.5707973\ttotal: 5.75s\tremaining: 1.88s\n",
      "754:\tlearn: 0.5707092\ttotal: 5.75s\tremaining: 1.87s\n",
      "755:\tlearn: 0.5706074\ttotal: 5.76s\tremaining: 1.86s\n",
      "756:\tlearn: 0.5705073\ttotal: 5.77s\tremaining: 1.85s\n",
      "757:\tlearn: 0.5703970\ttotal: 5.77s\tremaining: 1.84s\n",
      "758:\tlearn: 0.5702888\ttotal: 5.78s\tremaining: 1.83s\n",
      "759:\tlearn: 0.5702157\ttotal: 5.79s\tremaining: 1.83s\n",
      "760:\tlearn: 0.5701480\ttotal: 5.79s\tremaining: 1.82s\n",
      "761:\tlearn: 0.5700484\ttotal: 5.8s\tremaining: 1.81s\n",
      "762:\tlearn: 0.5699050\ttotal: 5.81s\tremaining: 1.8s\n",
      "763:\tlearn: 0.5697929\ttotal: 5.82s\tremaining: 1.8s\n",
      "764:\tlearn: 0.5697395\ttotal: 5.82s\tremaining: 1.79s\n",
      "765:\tlearn: 0.5696890\ttotal: 5.83s\tremaining: 1.78s\n",
      "766:\tlearn: 0.5696171\ttotal: 5.84s\tremaining: 1.77s\n",
      "767:\tlearn: 0.5695223\ttotal: 5.84s\tremaining: 1.76s\n",
      "768:\tlearn: 0.5694564\ttotal: 5.85s\tremaining: 1.76s\n",
      "769:\tlearn: 0.5693653\ttotal: 5.86s\tremaining: 1.75s\n",
      "770:\tlearn: 0.5693046\ttotal: 5.86s\tremaining: 1.74s\n",
      "771:\tlearn: 0.5692338\ttotal: 5.87s\tremaining: 1.73s\n",
      "772:\tlearn: 0.5691495\ttotal: 5.88s\tremaining: 1.73s\n",
      "773:\tlearn: 0.5690396\ttotal: 5.89s\tremaining: 1.72s\n",
      "774:\tlearn: 0.5689226\ttotal: 5.89s\tremaining: 1.71s\n",
      "775:\tlearn: 0.5688215\ttotal: 5.91s\tremaining: 1.7s\n",
      "776:\tlearn: 0.5687250\ttotal: 5.91s\tremaining: 1.7s\n",
      "777:\tlearn: 0.5686569\ttotal: 5.92s\tremaining: 1.69s\n",
      "778:\tlearn: 0.5686107\ttotal: 5.93s\tremaining: 1.68s\n",
      "779:\tlearn: 0.5685305\ttotal: 5.94s\tremaining: 1.67s\n",
      "780:\tlearn: 0.5684164\ttotal: 5.94s\tremaining: 1.67s\n",
      "781:\tlearn: 0.5683259\ttotal: 5.95s\tremaining: 1.66s\n",
      "782:\tlearn: 0.5682316\ttotal: 5.96s\tremaining: 1.65s\n",
      "783:\tlearn: 0.5680973\ttotal: 5.96s\tremaining: 1.64s\n",
      "784:\tlearn: 0.5680227\ttotal: 5.97s\tremaining: 1.64s\n",
      "785:\tlearn: 0.5679132\ttotal: 5.98s\tremaining: 1.63s\n",
      "786:\tlearn: 0.5678196\ttotal: 5.98s\tremaining: 1.62s\n",
      "787:\tlearn: 0.5677333\ttotal: 5.99s\tremaining: 1.61s\n",
      "788:\tlearn: 0.5677038\ttotal: 6s\tremaining: 1.6s\n",
      "789:\tlearn: 0.5676166\ttotal: 6.01s\tremaining: 1.6s\n",
      "790:\tlearn: 0.5675472\ttotal: 6.01s\tremaining: 1.59s\n",
      "791:\tlearn: 0.5674154\ttotal: 6.02s\tremaining: 1.58s\n",
      "792:\tlearn: 0.5672872\ttotal: 6.03s\tremaining: 1.57s\n",
      "793:\tlearn: 0.5671940\ttotal: 6.03s\tremaining: 1.56s\n",
      "794:\tlearn: 0.5671049\ttotal: 6.04s\tremaining: 1.56s\n",
      "795:\tlearn: 0.5670308\ttotal: 6.05s\tremaining: 1.55s\n",
      "796:\tlearn: 0.5669324\ttotal: 6.05s\tremaining: 1.54s\n",
      "797:\tlearn: 0.5668734\ttotal: 6.06s\tremaining: 1.53s\n",
      "798:\tlearn: 0.5668039\ttotal: 6.07s\tremaining: 1.53s\n",
      "799:\tlearn: 0.5667215\ttotal: 6.08s\tremaining: 1.52s\n",
      "800:\tlearn: 0.5666229\ttotal: 6.08s\tremaining: 1.51s\n",
      "801:\tlearn: 0.5665252\ttotal: 6.09s\tremaining: 1.5s\n",
      "802:\tlearn: 0.5664129\ttotal: 6.1s\tremaining: 1.5s\n",
      "803:\tlearn: 0.5662587\ttotal: 6.1s\tremaining: 1.49s\n",
      "804:\tlearn: 0.5661946\ttotal: 6.12s\tremaining: 1.48s\n",
      "805:\tlearn: 0.5661180\ttotal: 6.12s\tremaining: 1.47s\n",
      "806:\tlearn: 0.5660169\ttotal: 6.13s\tremaining: 1.47s\n",
      "807:\tlearn: 0.5659544\ttotal: 6.14s\tremaining: 1.46s\n",
      "808:\tlearn: 0.5658421\ttotal: 6.15s\tremaining: 1.45s\n",
      "809:\tlearn: 0.5657260\ttotal: 6.15s\tremaining: 1.44s\n",
      "810:\tlearn: 0.5656074\ttotal: 6.16s\tremaining: 1.44s\n",
      "811:\tlearn: 0.5655276\ttotal: 6.17s\tremaining: 1.43s\n",
      "812:\tlearn: 0.5654438\ttotal: 6.17s\tremaining: 1.42s\n",
      "813:\tlearn: 0.5652641\ttotal: 6.18s\tremaining: 1.41s\n",
      "814:\tlearn: 0.5652195\ttotal: 6.18s\tremaining: 1.4s\n",
      "815:\tlearn: 0.5651459\ttotal: 6.19s\tremaining: 1.4s\n",
      "816:\tlearn: 0.5650485\ttotal: 6.2s\tremaining: 1.39s\n",
      "817:\tlearn: 0.5649259\ttotal: 6.21s\tremaining: 1.38s\n",
      "818:\tlearn: 0.5648078\ttotal: 6.21s\tremaining: 1.37s\n",
      "819:\tlearn: 0.5647536\ttotal: 6.22s\tremaining: 1.36s\n",
      "820:\tlearn: 0.5646602\ttotal: 6.23s\tremaining: 1.36s\n",
      "821:\tlearn: 0.5645521\ttotal: 6.24s\tremaining: 1.35s\n",
      "822:\tlearn: 0.5644796\ttotal: 6.24s\tremaining: 1.34s\n",
      "823:\tlearn: 0.5643913\ttotal: 6.25s\tremaining: 1.33s\n",
      "824:\tlearn: 0.5643480\ttotal: 6.26s\tremaining: 1.33s\n",
      "825:\tlearn: 0.5642496\ttotal: 6.26s\tremaining: 1.32s\n",
      "826:\tlearn: 0.5641064\ttotal: 6.27s\tremaining: 1.31s\n",
      "827:\tlearn: 0.5640011\ttotal: 6.28s\tremaining: 1.3s\n",
      "828:\tlearn: 0.5638726\ttotal: 6.29s\tremaining: 1.3s\n",
      "829:\tlearn: 0.5637744\ttotal: 6.29s\tremaining: 1.29s\n",
      "830:\tlearn: 0.5636939\ttotal: 6.3s\tremaining: 1.28s\n",
      "831:\tlearn: 0.5636178\ttotal: 6.31s\tremaining: 1.27s\n",
      "832:\tlearn: 0.5635342\ttotal: 6.32s\tremaining: 1.27s\n",
      "833:\tlearn: 0.5634707\ttotal: 6.32s\tremaining: 1.26s\n",
      "834:\tlearn: 0.5634251\ttotal: 6.33s\tremaining: 1.25s\n",
      "835:\tlearn: 0.5633241\ttotal: 6.33s\tremaining: 1.24s\n",
      "836:\tlearn: 0.5632027\ttotal: 6.34s\tremaining: 1.24s\n",
      "837:\tlearn: 0.5631018\ttotal: 6.35s\tremaining: 1.23s\n",
      "838:\tlearn: 0.5630044\ttotal: 6.36s\tremaining: 1.22s\n",
      "839:\tlearn: 0.5629212\ttotal: 6.36s\tremaining: 1.21s\n",
      "840:\tlearn: 0.5628598\ttotal: 6.37s\tremaining: 1.2s\n",
      "841:\tlearn: 0.5627646\ttotal: 6.38s\tremaining: 1.2s\n",
      "842:\tlearn: 0.5626594\ttotal: 6.38s\tremaining: 1.19s\n",
      "843:\tlearn: 0.5625914\ttotal: 6.39s\tremaining: 1.18s\n",
      "844:\tlearn: 0.5625449\ttotal: 6.4s\tremaining: 1.17s\n",
      "845:\tlearn: 0.5624438\ttotal: 6.41s\tremaining: 1.17s\n",
      "846:\tlearn: 0.5623294\ttotal: 6.41s\tremaining: 1.16s\n",
      "847:\tlearn: 0.5622207\ttotal: 6.42s\tremaining: 1.15s\n",
      "848:\tlearn: 0.5620809\ttotal: 6.43s\tremaining: 1.14s\n",
      "849:\tlearn: 0.5620048\ttotal: 6.44s\tremaining: 1.14s\n",
      "850:\tlearn: 0.5619354\ttotal: 6.45s\tremaining: 1.13s\n",
      "851:\tlearn: 0.5618546\ttotal: 6.45s\tremaining: 1.12s\n",
      "852:\tlearn: 0.5617569\ttotal: 6.46s\tremaining: 1.11s\n",
      "853:\tlearn: 0.5616471\ttotal: 6.47s\tremaining: 1.1s\n",
      "854:\tlearn: 0.5615853\ttotal: 6.48s\tremaining: 1.1s\n",
      "855:\tlearn: 0.5615231\ttotal: 6.48s\tremaining: 1.09s\n",
      "856:\tlearn: 0.5614855\ttotal: 6.49s\tremaining: 1.08s\n",
      "857:\tlearn: 0.5614314\ttotal: 6.5s\tremaining: 1.07s\n",
      "858:\tlearn: 0.5613512\ttotal: 6.5s\tremaining: 1.07s\n",
      "859:\tlearn: 0.5612144\ttotal: 6.51s\tremaining: 1.06s\n",
      "860:\tlearn: 0.5611338\ttotal: 6.52s\tremaining: 1.05s\n",
      "861:\tlearn: 0.5610132\ttotal: 6.53s\tremaining: 1.04s\n",
      "862:\tlearn: 0.5609246\ttotal: 6.54s\tremaining: 1.04s\n",
      "863:\tlearn: 0.5608605\ttotal: 6.54s\tremaining: 1.03s\n",
      "864:\tlearn: 0.5608160\ttotal: 6.55s\tremaining: 1.02s\n",
      "865:\tlearn: 0.5607559\ttotal: 6.56s\tremaining: 1.01s\n",
      "866:\tlearn: 0.5607099\ttotal: 6.56s\tremaining: 1.01s\n",
      "867:\tlearn: 0.5606643\ttotal: 6.57s\tremaining: 1000ms\n",
      "868:\tlearn: 0.5605892\ttotal: 6.58s\tremaining: 992ms\n",
      "869:\tlearn: 0.5605201\ttotal: 6.59s\tremaining: 984ms\n",
      "870:\tlearn: 0.5604200\ttotal: 6.59s\tremaining: 977ms\n",
      "871:\tlearn: 0.5603291\ttotal: 6.6s\tremaining: 969ms\n",
      "872:\tlearn: 0.5602641\ttotal: 6.61s\tremaining: 961ms\n",
      "873:\tlearn: 0.5601935\ttotal: 6.61s\tremaining: 953ms\n",
      "874:\tlearn: 0.5601075\ttotal: 6.62s\tremaining: 946ms\n",
      "875:\tlearn: 0.5600249\ttotal: 6.63s\tremaining: 938ms\n",
      "876:\tlearn: 0.5599295\ttotal: 6.63s\tremaining: 931ms\n",
      "877:\tlearn: 0.5597557\ttotal: 6.64s\tremaining: 923ms\n",
      "878:\tlearn: 0.5596996\ttotal: 6.65s\tremaining: 915ms\n",
      "879:\tlearn: 0.5596185\ttotal: 6.66s\tremaining: 908ms\n",
      "880:\tlearn: 0.5595392\ttotal: 6.66s\tremaining: 900ms\n",
      "881:\tlearn: 0.5594800\ttotal: 6.67s\tremaining: 892ms\n",
      "882:\tlearn: 0.5593866\ttotal: 6.68s\tremaining: 885ms\n",
      "883:\tlearn: 0.5592926\ttotal: 6.69s\tremaining: 877ms\n",
      "884:\tlearn: 0.5592010\ttotal: 6.69s\tremaining: 870ms\n",
      "885:\tlearn: 0.5590990\ttotal: 6.7s\tremaining: 862ms\n",
      "886:\tlearn: 0.5590064\ttotal: 6.71s\tremaining: 855ms\n",
      "887:\tlearn: 0.5588947\ttotal: 6.72s\tremaining: 847ms\n",
      "888:\tlearn: 0.5588430\ttotal: 6.72s\tremaining: 840ms\n",
      "889:\tlearn: 0.5587512\ttotal: 6.73s\tremaining: 832ms\n",
      "890:\tlearn: 0.5586716\ttotal: 6.74s\tremaining: 825ms\n",
      "891:\tlearn: 0.5585637\ttotal: 6.75s\tremaining: 817ms\n",
      "892:\tlearn: 0.5584554\ttotal: 6.76s\tremaining: 810ms\n",
      "893:\tlearn: 0.5583462\ttotal: 6.76s\tremaining: 802ms\n",
      "894:\tlearn: 0.5582610\ttotal: 6.77s\tremaining: 794ms\n",
      "895:\tlearn: 0.5582079\ttotal: 6.78s\tremaining: 787ms\n",
      "896:\tlearn: 0.5581491\ttotal: 6.79s\tremaining: 779ms\n",
      "897:\tlearn: 0.5580866\ttotal: 6.79s\tremaining: 772ms\n",
      "898:\tlearn: 0.5580260\ttotal: 6.8s\tremaining: 764ms\n",
      "899:\tlearn: 0.5579572\ttotal: 6.81s\tremaining: 756ms\n",
      "900:\tlearn: 0.5579165\ttotal: 6.82s\tremaining: 749ms\n",
      "901:\tlearn: 0.5578191\ttotal: 6.82s\tremaining: 741ms\n",
      "902:\tlearn: 0.5577359\ttotal: 6.83s\tremaining: 734ms\n",
      "903:\tlearn: 0.5576966\ttotal: 6.84s\tremaining: 726ms\n",
      "904:\tlearn: 0.5575974\ttotal: 6.84s\tremaining: 719ms\n",
      "905:\tlearn: 0.5575032\ttotal: 6.85s\tremaining: 711ms\n",
      "906:\tlearn: 0.5574177\ttotal: 6.86s\tremaining: 703ms\n",
      "907:\tlearn: 0.5572926\ttotal: 6.86s\tremaining: 696ms\n",
      "908:\tlearn: 0.5572033\ttotal: 6.87s\tremaining: 688ms\n",
      "909:\tlearn: 0.5570169\ttotal: 6.88s\tremaining: 680ms\n",
      "910:\tlearn: 0.5569447\ttotal: 6.88s\tremaining: 673ms\n",
      "911:\tlearn: 0.5568697\ttotal: 6.89s\tremaining: 665ms\n",
      "912:\tlearn: 0.5567830\ttotal: 6.9s\tremaining: 657ms\n",
      "913:\tlearn: 0.5566971\ttotal: 6.9s\tremaining: 650ms\n",
      "914:\tlearn: 0.5565932\ttotal: 6.91s\tremaining: 642ms\n",
      "915:\tlearn: 0.5565031\ttotal: 6.92s\tremaining: 635ms\n",
      "916:\tlearn: 0.5564464\ttotal: 6.93s\tremaining: 627ms\n",
      "917:\tlearn: 0.5563937\ttotal: 6.93s\tremaining: 619ms\n",
      "918:\tlearn: 0.5562908\ttotal: 6.94s\tremaining: 612ms\n",
      "919:\tlearn: 0.5562384\ttotal: 6.95s\tremaining: 604ms\n",
      "920:\tlearn: 0.5561117\ttotal: 6.96s\tremaining: 597ms\n",
      "921:\tlearn: 0.5560406\ttotal: 6.97s\tremaining: 589ms\n",
      "922:\tlearn: 0.5559868\ttotal: 6.97s\tremaining: 582ms\n",
      "923:\tlearn: 0.5558855\ttotal: 6.98s\tremaining: 574ms\n",
      "924:\tlearn: 0.5558369\ttotal: 6.99s\tremaining: 567ms\n",
      "925:\tlearn: 0.5557413\ttotal: 6.99s\tremaining: 559ms\n",
      "926:\tlearn: 0.5556783\ttotal: 7s\tremaining: 551ms\n",
      "927:\tlearn: 0.5555928\ttotal: 7.01s\tremaining: 544ms\n",
      "928:\tlearn: 0.5554982\ttotal: 7.02s\tremaining: 536ms\n",
      "929:\tlearn: 0.5554247\ttotal: 7.02s\tremaining: 529ms\n",
      "930:\tlearn: 0.5553791\ttotal: 7.03s\tremaining: 521ms\n",
      "931:\tlearn: 0.5553192\ttotal: 7.04s\tremaining: 513ms\n",
      "932:\tlearn: 0.5552323\ttotal: 7.04s\tremaining: 506ms\n",
      "933:\tlearn: 0.5550939\ttotal: 7.05s\tremaining: 498ms\n",
      "934:\tlearn: 0.5549687\ttotal: 7.06s\tremaining: 491ms\n",
      "935:\tlearn: 0.5548751\ttotal: 7.06s\tremaining: 483ms\n",
      "936:\tlearn: 0.5548018\ttotal: 7.07s\tremaining: 476ms\n",
      "937:\tlearn: 0.5546387\ttotal: 7.08s\tremaining: 468ms\n",
      "938:\tlearn: 0.5545537\ttotal: 7.08s\tremaining: 460ms\n",
      "939:\tlearn: 0.5544666\ttotal: 7.09s\tremaining: 453ms\n",
      "940:\tlearn: 0.5543764\ttotal: 7.1s\tremaining: 445ms\n",
      "941:\tlearn: 0.5542881\ttotal: 7.11s\tremaining: 438ms\n",
      "942:\tlearn: 0.5542167\ttotal: 7.11s\tremaining: 430ms\n",
      "943:\tlearn: 0.5541458\ttotal: 7.12s\tremaining: 422ms\n",
      "944:\tlearn: 0.5540294\ttotal: 7.13s\tremaining: 415ms\n",
      "945:\tlearn: 0.5539326\ttotal: 7.14s\tremaining: 408ms\n",
      "946:\tlearn: 0.5538290\ttotal: 7.15s\tremaining: 400ms\n",
      "947:\tlearn: 0.5537552\ttotal: 7.16s\tremaining: 393ms\n",
      "948:\tlearn: 0.5536273\ttotal: 7.16s\tremaining: 385ms\n",
      "949:\tlearn: 0.5535041\ttotal: 7.17s\tremaining: 377ms\n",
      "950:\tlearn: 0.5534536\ttotal: 7.18s\tremaining: 370ms\n",
      "951:\tlearn: 0.5533916\ttotal: 7.18s\tremaining: 362ms\n",
      "952:\tlearn: 0.5532919\ttotal: 7.19s\tremaining: 355ms\n",
      "953:\tlearn: 0.5531569\ttotal: 7.2s\tremaining: 347ms\n",
      "954:\tlearn: 0.5530668\ttotal: 7.2s\tremaining: 339ms\n",
      "955:\tlearn: 0.5530175\ttotal: 7.21s\tremaining: 332ms\n",
      "956:\tlearn: 0.5529519\ttotal: 7.22s\tremaining: 324ms\n",
      "957:\tlearn: 0.5528797\ttotal: 7.23s\tremaining: 317ms\n",
      "958:\tlearn: 0.5528017\ttotal: 7.23s\tremaining: 309ms\n",
      "959:\tlearn: 0.5527082\ttotal: 7.24s\tremaining: 302ms\n",
      "960:\tlearn: 0.5526019\ttotal: 7.25s\tremaining: 294ms\n",
      "961:\tlearn: 0.5525275\ttotal: 7.25s\tremaining: 287ms\n",
      "962:\tlearn: 0.5523837\ttotal: 7.26s\tremaining: 279ms\n",
      "963:\tlearn: 0.5523121\ttotal: 7.27s\tremaining: 271ms\n",
      "964:\tlearn: 0.5521000\ttotal: 7.27s\tremaining: 264ms\n",
      "965:\tlearn: 0.5520067\ttotal: 7.28s\tremaining: 256ms\n",
      "966:\tlearn: 0.5519002\ttotal: 7.29s\tremaining: 249ms\n",
      "967:\tlearn: 0.5518524\ttotal: 7.29s\tremaining: 241ms\n",
      "968:\tlearn: 0.5517433\ttotal: 7.3s\tremaining: 234ms\n",
      "969:\tlearn: 0.5516859\ttotal: 7.31s\tremaining: 226ms\n",
      "970:\tlearn: 0.5515618\ttotal: 7.31s\tremaining: 218ms\n",
      "971:\tlearn: 0.5514960\ttotal: 7.32s\tremaining: 211ms\n",
      "972:\tlearn: 0.5514072\ttotal: 7.33s\tremaining: 203ms\n",
      "973:\tlearn: 0.5513069\ttotal: 7.34s\tremaining: 196ms\n",
      "974:\tlearn: 0.5512380\ttotal: 7.34s\tremaining: 188ms\n",
      "975:\tlearn: 0.5512017\ttotal: 7.35s\tremaining: 181ms\n",
      "976:\tlearn: 0.5511485\ttotal: 7.36s\tremaining: 173ms\n",
      "977:\tlearn: 0.5510544\ttotal: 7.37s\tremaining: 166ms\n",
      "978:\tlearn: 0.5509511\ttotal: 7.38s\tremaining: 158ms\n",
      "979:\tlearn: 0.5509037\ttotal: 7.38s\tremaining: 151ms\n",
      "980:\tlearn: 0.5508564\ttotal: 7.39s\tremaining: 143ms\n",
      "981:\tlearn: 0.5508095\ttotal: 7.39s\tremaining: 136ms\n",
      "982:\tlearn: 0.5507533\ttotal: 7.4s\tremaining: 128ms\n",
      "983:\tlearn: 0.5506596\ttotal: 7.41s\tremaining: 120ms\n",
      "984:\tlearn: 0.5505842\ttotal: 7.42s\tremaining: 113ms\n",
      "985:\tlearn: 0.5505052\ttotal: 7.42s\tremaining: 105ms\n",
      "986:\tlearn: 0.5504029\ttotal: 7.43s\tremaining: 97.9ms\n",
      "987:\tlearn: 0.5503256\ttotal: 7.44s\tremaining: 90.3ms\n",
      "988:\tlearn: 0.5501837\ttotal: 7.44s\tremaining: 82.8ms\n",
      "989:\tlearn: 0.5500938\ttotal: 7.45s\tremaining: 75.3ms\n",
      "990:\tlearn: 0.5500349\ttotal: 7.46s\tremaining: 67.7ms\n",
      "991:\tlearn: 0.5499288\ttotal: 7.46s\tremaining: 60.2ms\n",
      "992:\tlearn: 0.5498756\ttotal: 7.47s\tremaining: 52.7ms\n",
      "993:\tlearn: 0.5498196\ttotal: 7.48s\tremaining: 45.1ms\n",
      "994:\tlearn: 0.5497489\ttotal: 7.49s\tremaining: 37.6ms\n",
      "995:\tlearn: 0.5496739\ttotal: 7.49s\tremaining: 30.1ms\n",
      "996:\tlearn: 0.5496137\ttotal: 7.5s\tremaining: 22.6ms\n",
      "997:\tlearn: 0.5495069\ttotal: 7.51s\tremaining: 15ms\n",
      "998:\tlearn: 0.5494108\ttotal: 7.51s\tremaining: 7.52ms\n",
      "999:\tlearn: 0.5493292\ttotal: 7.52s\tremaining: 0us\n",
      "The accuracy score for the <catboost.core.CatBoostClassifier object at 0x0000015AB44F3EB0> undersampled set is: 0.5477815699658704\n",
      "The f1 score for the <catboost.core.CatBoostClassifier object at 0x0000015AB44F3EB0> undersampled set is: 0.15227127319257835\n",
      "\n",
      "\n",
      "Learning rate set to 0.057318\n",
      "0:\tlearn: 0.6288349\ttotal: 25ms\tremaining: 25s\n",
      "1:\tlearn: 0.5727012\ttotal: 46.7ms\tremaining: 23.3s\n",
      "2:\tlearn: 0.5256276\ttotal: 66.8ms\tremaining: 22.2s\n",
      "3:\tlearn: 0.4839724\ttotal: 98.5ms\tremaining: 24.5s\n",
      "4:\tlearn: 0.4496219\ttotal: 115ms\tremaining: 22.9s\n",
      "5:\tlearn: 0.4202649\ttotal: 130ms\tremaining: 21.5s\n",
      "6:\tlearn: 0.3950974\ttotal: 149ms\tremaining: 21.1s\n",
      "7:\tlearn: 0.3734957\ttotal: 170ms\tremaining: 21s\n",
      "8:\tlearn: 0.3549049\ttotal: 193ms\tremaining: 21.2s\n",
      "9:\tlearn: 0.3390612\ttotal: 218ms\tremaining: 21.6s\n",
      "10:\tlearn: 0.3253623\ttotal: 236ms\tremaining: 21.2s\n",
      "11:\tlearn: 0.3130219\ttotal: 258ms\tremaining: 21.2s\n",
      "12:\tlearn: 0.3028714\ttotal: 281ms\tremaining: 21.4s\n",
      "13:\tlearn: 0.2942348\ttotal: 305ms\tremaining: 21.5s\n",
      "14:\tlearn: 0.2859543\ttotal: 332ms\tremaining: 21.8s\n",
      "15:\tlearn: 0.2795326\ttotal: 351ms\tremaining: 21.6s\n",
      "16:\tlearn: 0.2739944\ttotal: 365ms\tremaining: 21.1s\n",
      "17:\tlearn: 0.2679904\ttotal: 386ms\tremaining: 21.1s\n",
      "18:\tlearn: 0.2638629\ttotal: 406ms\tremaining: 21s\n",
      "19:\tlearn: 0.2600495\ttotal: 431ms\tremaining: 21.1s\n",
      "20:\tlearn: 0.2565758\ttotal: 447ms\tremaining: 20.8s\n",
      "21:\tlearn: 0.2534857\ttotal: 475ms\tremaining: 21.1s\n",
      "22:\tlearn: 0.2508247\ttotal: 495ms\tremaining: 21s\n",
      "23:\tlearn: 0.2488051\ttotal: 524ms\tremaining: 21.3s\n",
      "24:\tlearn: 0.2466939\ttotal: 544ms\tremaining: 21.2s\n",
      "25:\tlearn: 0.2445271\ttotal: 566ms\tremaining: 21.2s\n",
      "26:\tlearn: 0.2428802\ttotal: 591ms\tremaining: 21.3s\n",
      "27:\tlearn: 0.2414019\ttotal: 615ms\tremaining: 21.3s\n",
      "28:\tlearn: 0.2402504\ttotal: 638ms\tremaining: 21.4s\n",
      "29:\tlearn: 0.2388414\ttotal: 659ms\tremaining: 21.3s\n",
      "30:\tlearn: 0.2375364\ttotal: 682ms\tremaining: 21.3s\n",
      "31:\tlearn: 0.2368086\ttotal: 706ms\tremaining: 21.4s\n",
      "32:\tlearn: 0.2358811\ttotal: 737ms\tremaining: 21.6s\n",
      "33:\tlearn: 0.2349717\ttotal: 760ms\tremaining: 21.6s\n",
      "34:\tlearn: 0.2345420\ttotal: 775ms\tremaining: 21.4s\n",
      "35:\tlearn: 0.2338231\ttotal: 797ms\tremaining: 21.3s\n",
      "36:\tlearn: 0.2334926\ttotal: 814ms\tremaining: 21.2s\n",
      "37:\tlearn: 0.2330792\ttotal: 834ms\tremaining: 21.1s\n",
      "38:\tlearn: 0.2325305\ttotal: 857ms\tremaining: 21.1s\n",
      "39:\tlearn: 0.2321173\ttotal: 879ms\tremaining: 21.1s\n",
      "40:\tlearn: 0.2318086\ttotal: 903ms\tremaining: 21.1s\n",
      "41:\tlearn: 0.2314828\ttotal: 936ms\tremaining: 21.3s\n",
      "42:\tlearn: 0.2312561\ttotal: 956ms\tremaining: 21.3s\n",
      "43:\tlearn: 0.2309264\ttotal: 978ms\tremaining: 21.3s\n",
      "44:\tlearn: 0.2306503\ttotal: 1s\tremaining: 21.3s\n",
      "45:\tlearn: 0.2304999\ttotal: 1.02s\tremaining: 21.2s\n",
      "46:\tlearn: 0.2302208\ttotal: 1.04s\tremaining: 21.2s\n",
      "47:\tlearn: 0.2300098\ttotal: 1.07s\tremaining: 21.3s\n",
      "48:\tlearn: 0.2298033\ttotal: 1.1s\tremaining: 21.3s\n",
      "49:\tlearn: 0.2295454\ttotal: 1.12s\tremaining: 21.3s\n",
      "50:\tlearn: 0.2293595\ttotal: 1.15s\tremaining: 21.4s\n",
      "51:\tlearn: 0.2292283\ttotal: 1.17s\tremaining: 21.4s\n",
      "52:\tlearn: 0.2290214\ttotal: 1.2s\tremaining: 21.4s\n",
      "53:\tlearn: 0.2288700\ttotal: 1.22s\tremaining: 21.4s\n",
      "54:\tlearn: 0.2287574\ttotal: 1.24s\tremaining: 21.4s\n",
      "55:\tlearn: 0.2286508\ttotal: 1.27s\tremaining: 21.4s\n",
      "56:\tlearn: 0.2284805\ttotal: 1.29s\tremaining: 21.4s\n",
      "57:\tlearn: 0.2283443\ttotal: 1.32s\tremaining: 21.4s\n",
      "58:\tlearn: 0.2282469\ttotal: 1.34s\tremaining: 21.4s\n",
      "59:\tlearn: 0.2281119\ttotal: 1.37s\tremaining: 21.5s\n",
      "60:\tlearn: 0.2280328\ttotal: 1.4s\tremaining: 21.5s\n",
      "61:\tlearn: 0.2279450\ttotal: 1.42s\tremaining: 21.5s\n",
      "62:\tlearn: 0.2278466\ttotal: 1.44s\tremaining: 21.5s\n",
      "63:\tlearn: 0.2277342\ttotal: 1.47s\tremaining: 21.5s\n",
      "64:\tlearn: 0.2276034\ttotal: 1.49s\tremaining: 21.5s\n",
      "65:\tlearn: 0.2275374\ttotal: 1.52s\tremaining: 21.5s\n",
      "66:\tlearn: 0.2274683\ttotal: 1.54s\tremaining: 21.5s\n",
      "67:\tlearn: 0.2274332\ttotal: 1.56s\tremaining: 21.4s\n",
      "68:\tlearn: 0.2273613\ttotal: 1.6s\tremaining: 21.6s\n",
      "69:\tlearn: 0.2273058\ttotal: 1.62s\tremaining: 21.5s\n",
      "70:\tlearn: 0.2272383\ttotal: 1.64s\tremaining: 21.5s\n",
      "71:\tlearn: 0.2271647\ttotal: 1.66s\tremaining: 21.5s\n",
      "72:\tlearn: 0.2271151\ttotal: 1.69s\tremaining: 21.5s\n",
      "73:\tlearn: 0.2270644\ttotal: 1.71s\tremaining: 21.4s\n",
      "74:\tlearn: 0.2270418\ttotal: 1.74s\tremaining: 21.4s\n",
      "75:\tlearn: 0.2269868\ttotal: 1.76s\tremaining: 21.4s\n",
      "76:\tlearn: 0.2269385\ttotal: 1.78s\tremaining: 21.4s\n",
      "77:\tlearn: 0.2268625\ttotal: 1.82s\tremaining: 21.5s\n",
      "78:\tlearn: 0.2267960\ttotal: 1.84s\tremaining: 21.5s\n",
      "79:\tlearn: 0.2267500\ttotal: 1.87s\tremaining: 21.5s\n",
      "80:\tlearn: 0.2266835\ttotal: 1.89s\tremaining: 21.5s\n",
      "81:\tlearn: 0.2266502\ttotal: 1.92s\tremaining: 21.4s\n",
      "82:\tlearn: 0.2266005\ttotal: 1.94s\tremaining: 21.4s\n",
      "83:\tlearn: 0.2265756\ttotal: 1.96s\tremaining: 21.4s\n",
      "84:\tlearn: 0.2265251\ttotal: 1.98s\tremaining: 21.3s\n",
      "85:\tlearn: 0.2264712\ttotal: 2.01s\tremaining: 21.3s\n",
      "86:\tlearn: 0.2264217\ttotal: 2.04s\tremaining: 21.4s\n",
      "87:\tlearn: 0.2263835\ttotal: 2.06s\tremaining: 21.4s\n",
      "88:\tlearn: 0.2263172\ttotal: 2.08s\tremaining: 21.3s\n",
      "89:\tlearn: 0.2262848\ttotal: 2.1s\tremaining: 21.3s\n",
      "90:\tlearn: 0.2262513\ttotal: 2.13s\tremaining: 21.3s\n",
      "91:\tlearn: 0.2262032\ttotal: 2.15s\tremaining: 21.2s\n",
      "92:\tlearn: 0.2261784\ttotal: 2.18s\tremaining: 21.2s\n",
      "93:\tlearn: 0.2261355\ttotal: 2.2s\tremaining: 21.2s\n",
      "94:\tlearn: 0.2260829\ttotal: 2.23s\tremaining: 21.2s\n",
      "95:\tlearn: 0.2260266\ttotal: 2.26s\tremaining: 21.3s\n",
      "96:\tlearn: 0.2259873\ttotal: 2.28s\tremaining: 21.2s\n",
      "97:\tlearn: 0.2259537\ttotal: 2.3s\tremaining: 21.2s\n",
      "98:\tlearn: 0.2258851\ttotal: 2.33s\tremaining: 21.2s\n",
      "99:\tlearn: 0.2258639\ttotal: 2.35s\tremaining: 21.2s\n",
      "100:\tlearn: 0.2258115\ttotal: 2.37s\tremaining: 21.1s\n",
      "101:\tlearn: 0.2257616\ttotal: 2.4s\tremaining: 21.1s\n",
      "102:\tlearn: 0.2257274\ttotal: 2.42s\tremaining: 21.1s\n",
      "103:\tlearn: 0.2256972\ttotal: 2.45s\tremaining: 21.1s\n",
      "104:\tlearn: 0.2256626\ttotal: 2.47s\tremaining: 21.1s\n",
      "105:\tlearn: 0.2256332\ttotal: 2.5s\tremaining: 21.1s\n",
      "106:\tlearn: 0.2255853\ttotal: 2.52s\tremaining: 21s\n",
      "107:\tlearn: 0.2255674\ttotal: 2.54s\tremaining: 21s\n",
      "108:\tlearn: 0.2255426\ttotal: 2.56s\tremaining: 20.9s\n",
      "109:\tlearn: 0.2254837\ttotal: 2.59s\tremaining: 20.9s\n",
      "110:\tlearn: 0.2254514\ttotal: 2.61s\tremaining: 20.9s\n",
      "111:\tlearn: 0.2254164\ttotal: 2.64s\tremaining: 20.9s\n",
      "112:\tlearn: 0.2253586\ttotal: 2.67s\tremaining: 20.9s\n",
      "113:\tlearn: 0.2253233\ttotal: 2.69s\tremaining: 20.9s\n",
      "114:\tlearn: 0.2253140\ttotal: 2.71s\tremaining: 20.9s\n",
      "115:\tlearn: 0.2252986\ttotal: 2.74s\tremaining: 20.9s\n",
      "116:\tlearn: 0.2252768\ttotal: 2.76s\tremaining: 20.8s\n",
      "117:\tlearn: 0.2252523\ttotal: 2.78s\tremaining: 20.8s\n",
      "118:\tlearn: 0.2252192\ttotal: 2.81s\tremaining: 20.8s\n",
      "119:\tlearn: 0.2252013\ttotal: 2.83s\tremaining: 20.8s\n",
      "120:\tlearn: 0.2251838\ttotal: 2.86s\tremaining: 20.8s\n",
      "121:\tlearn: 0.2251441\ttotal: 2.88s\tremaining: 20.7s\n",
      "122:\tlearn: 0.2251133\ttotal: 2.91s\tremaining: 20.8s\n",
      "123:\tlearn: 0.2250839\ttotal: 2.94s\tremaining: 20.8s\n",
      "124:\tlearn: 0.2250199\ttotal: 2.96s\tremaining: 20.8s\n",
      "125:\tlearn: 0.2249876\ttotal: 2.99s\tremaining: 20.7s\n",
      "126:\tlearn: 0.2249527\ttotal: 3.01s\tremaining: 20.7s\n",
      "127:\tlearn: 0.2249048\ttotal: 3.03s\tremaining: 20.7s\n",
      "128:\tlearn: 0.2248697\ttotal: 3.06s\tremaining: 20.7s\n",
      "129:\tlearn: 0.2248375\ttotal: 3.09s\tremaining: 20.7s\n",
      "130:\tlearn: 0.2247949\ttotal: 3.11s\tremaining: 20.7s\n",
      "131:\tlearn: 0.2247727\ttotal: 3.18s\tremaining: 20.9s\n",
      "132:\tlearn: 0.2247444\ttotal: 3.2s\tremaining: 20.9s\n",
      "133:\tlearn: 0.2246725\ttotal: 3.23s\tremaining: 20.8s\n",
      "134:\tlearn: 0.2246089\ttotal: 3.25s\tremaining: 20.8s\n",
      "135:\tlearn: 0.2245761\ttotal: 3.27s\tremaining: 20.8s\n",
      "136:\tlearn: 0.2245417\ttotal: 3.3s\tremaining: 20.8s\n",
      "137:\tlearn: 0.2245137\ttotal: 3.32s\tremaining: 20.7s\n",
      "138:\tlearn: 0.2244827\ttotal: 3.35s\tremaining: 20.7s\n",
      "139:\tlearn: 0.2244582\ttotal: 3.37s\tremaining: 20.7s\n",
      "140:\tlearn: 0.2244379\ttotal: 3.4s\tremaining: 20.7s\n",
      "141:\tlearn: 0.2244035\ttotal: 3.42s\tremaining: 20.7s\n",
      "142:\tlearn: 0.2243404\ttotal: 3.44s\tremaining: 20.6s\n",
      "143:\tlearn: 0.2243024\ttotal: 3.47s\tremaining: 20.6s\n",
      "144:\tlearn: 0.2242812\ttotal: 3.5s\tremaining: 20.6s\n",
      "145:\tlearn: 0.2242087\ttotal: 3.52s\tremaining: 20.6s\n",
      "146:\tlearn: 0.2241852\ttotal: 3.54s\tremaining: 20.5s\n",
      "147:\tlearn: 0.2241540\ttotal: 3.56s\tremaining: 20.5s\n",
      "148:\tlearn: 0.2241156\ttotal: 3.6s\tremaining: 20.6s\n",
      "149:\tlearn: 0.2240631\ttotal: 3.62s\tremaining: 20.5s\n",
      "150:\tlearn: 0.2240177\ttotal: 3.65s\tremaining: 20.5s\n",
      "151:\tlearn: 0.2239840\ttotal: 3.67s\tremaining: 20.5s\n",
      "152:\tlearn: 0.2239570\ttotal: 3.7s\tremaining: 20.5s\n",
      "153:\tlearn: 0.2239415\ttotal: 3.72s\tremaining: 20.4s\n",
      "154:\tlearn: 0.2239310\ttotal: 3.74s\tremaining: 20.4s\n",
      "155:\tlearn: 0.2239070\ttotal: 3.76s\tremaining: 20.4s\n",
      "156:\tlearn: 0.2238924\ttotal: 3.79s\tremaining: 20.3s\n",
      "157:\tlearn: 0.2238443\ttotal: 3.82s\tremaining: 20.4s\n",
      "158:\tlearn: 0.2238371\ttotal: 3.84s\tremaining: 20.3s\n",
      "159:\tlearn: 0.2237999\ttotal: 3.86s\tremaining: 20.3s\n",
      "160:\tlearn: 0.2237690\ttotal: 3.89s\tremaining: 20.3s\n",
      "161:\tlearn: 0.2237263\ttotal: 3.91s\tremaining: 20.2s\n",
      "162:\tlearn: 0.2237051\ttotal: 3.94s\tremaining: 20.2s\n",
      "163:\tlearn: 0.2236618\ttotal: 3.96s\tremaining: 20.2s\n",
      "164:\tlearn: 0.2236328\ttotal: 3.98s\tremaining: 20.2s\n",
      "165:\tlearn: 0.2235751\ttotal: 4.01s\tremaining: 20.2s\n",
      "166:\tlearn: 0.2235577\ttotal: 4.04s\tremaining: 20.1s\n",
      "167:\tlearn: 0.2235215\ttotal: 4.06s\tremaining: 20.1s\n",
      "168:\tlearn: 0.2234882\ttotal: 4.09s\tremaining: 20.1s\n",
      "169:\tlearn: 0.2234507\ttotal: 4.11s\tremaining: 20.1s\n",
      "170:\tlearn: 0.2234269\ttotal: 4.14s\tremaining: 20.1s\n",
      "171:\tlearn: 0.2233996\ttotal: 4.16s\tremaining: 20s\n",
      "172:\tlearn: 0.2233652\ttotal: 4.18s\tremaining: 20s\n",
      "173:\tlearn: 0.2233189\ttotal: 4.2s\tremaining: 20s\n",
      "174:\tlearn: 0.2232860\ttotal: 4.22s\tremaining: 19.9s\n",
      "175:\tlearn: 0.2232736\ttotal: 4.25s\tremaining: 19.9s\n",
      "176:\tlearn: 0.2232410\ttotal: 4.28s\tremaining: 19.9s\n",
      "177:\tlearn: 0.2231965\ttotal: 4.3s\tremaining: 19.9s\n",
      "178:\tlearn: 0.2231553\ttotal: 4.33s\tremaining: 19.8s\n",
      "179:\tlearn: 0.2231453\ttotal: 4.35s\tremaining: 19.8s\n",
      "180:\tlearn: 0.2231158\ttotal: 4.37s\tremaining: 19.8s\n",
      "181:\tlearn: 0.2230725\ttotal: 4.4s\tremaining: 19.8s\n",
      "182:\tlearn: 0.2230350\ttotal: 4.42s\tremaining: 19.7s\n",
      "183:\tlearn: 0.2230080\ttotal: 4.44s\tremaining: 19.7s\n",
      "184:\tlearn: 0.2229792\ttotal: 4.47s\tremaining: 19.7s\n",
      "185:\tlearn: 0.2229570\ttotal: 4.5s\tremaining: 19.7s\n",
      "186:\tlearn: 0.2229114\ttotal: 4.52s\tremaining: 19.6s\n",
      "187:\tlearn: 0.2228793\ttotal: 4.54s\tremaining: 19.6s\n",
      "188:\tlearn: 0.2228650\ttotal: 4.57s\tremaining: 19.6s\n",
      "189:\tlearn: 0.2228124\ttotal: 4.59s\tremaining: 19.6s\n",
      "190:\tlearn: 0.2227716\ttotal: 4.61s\tremaining: 19.5s\n",
      "191:\tlearn: 0.2227324\ttotal: 4.63s\tremaining: 19.5s\n",
      "192:\tlearn: 0.2226969\ttotal: 4.66s\tremaining: 19.5s\n",
      "193:\tlearn: 0.2226580\ttotal: 4.69s\tremaining: 19.5s\n",
      "194:\tlearn: 0.2226154\ttotal: 4.78s\tremaining: 19.7s\n",
      "195:\tlearn: 0.2225619\ttotal: 4.84s\tremaining: 19.9s\n",
      "196:\tlearn: 0.2225202\ttotal: 4.87s\tremaining: 19.8s\n",
      "197:\tlearn: 0.2224915\ttotal: 4.89s\tremaining: 19.8s\n",
      "198:\tlearn: 0.2224644\ttotal: 4.92s\tremaining: 19.8s\n",
      "199:\tlearn: 0.2224196\ttotal: 4.95s\tremaining: 19.8s\n",
      "200:\tlearn: 0.2223950\ttotal: 4.97s\tremaining: 19.8s\n",
      "201:\tlearn: 0.2223546\ttotal: 4.99s\tremaining: 19.7s\n",
      "202:\tlearn: 0.2223278\ttotal: 5.01s\tremaining: 19.7s\n",
      "203:\tlearn: 0.2222862\ttotal: 5.04s\tremaining: 19.6s\n",
      "204:\tlearn: 0.2222455\ttotal: 5.06s\tremaining: 19.6s\n",
      "205:\tlearn: 0.2221965\ttotal: 5.08s\tremaining: 19.6s\n",
      "206:\tlearn: 0.2221532\ttotal: 5.11s\tremaining: 19.6s\n",
      "207:\tlearn: 0.2221097\ttotal: 5.14s\tremaining: 19.6s\n",
      "208:\tlearn: 0.2220520\ttotal: 5.17s\tremaining: 19.6s\n",
      "209:\tlearn: 0.2220022\ttotal: 5.19s\tremaining: 19.5s\n",
      "210:\tlearn: 0.2219538\ttotal: 5.21s\tremaining: 19.5s\n",
      "211:\tlearn: 0.2219121\ttotal: 5.24s\tremaining: 19.5s\n",
      "212:\tlearn: 0.2218802\ttotal: 5.26s\tremaining: 19.4s\n",
      "213:\tlearn: 0.2218428\ttotal: 5.29s\tremaining: 19.4s\n",
      "214:\tlearn: 0.2218041\ttotal: 5.31s\tremaining: 19.4s\n",
      "215:\tlearn: 0.2217632\ttotal: 5.33s\tremaining: 19.4s\n",
      "216:\tlearn: 0.2217180\ttotal: 5.37s\tremaining: 19.4s\n",
      "217:\tlearn: 0.2216662\ttotal: 5.39s\tremaining: 19.3s\n",
      "218:\tlearn: 0.2216092\ttotal: 5.42s\tremaining: 19.3s\n",
      "219:\tlearn: 0.2215556\ttotal: 5.44s\tremaining: 19.3s\n",
      "220:\tlearn: 0.2215174\ttotal: 5.46s\tremaining: 19.2s\n",
      "221:\tlearn: 0.2214691\ttotal: 5.48s\tremaining: 19.2s\n",
      "222:\tlearn: 0.2214292\ttotal: 5.51s\tremaining: 19.2s\n",
      "223:\tlearn: 0.2213872\ttotal: 5.53s\tremaining: 19.2s\n",
      "224:\tlearn: 0.2213587\ttotal: 5.55s\tremaining: 19.1s\n",
      "225:\tlearn: 0.2213255\ttotal: 5.59s\tremaining: 19.1s\n",
      "226:\tlearn: 0.2212842\ttotal: 5.61s\tremaining: 19.1s\n",
      "227:\tlearn: 0.2212248\ttotal: 5.64s\tremaining: 19.1s\n",
      "228:\tlearn: 0.2211840\ttotal: 5.66s\tremaining: 19.1s\n",
      "229:\tlearn: 0.2211682\ttotal: 5.68s\tremaining: 19s\n",
      "230:\tlearn: 0.2211221\ttotal: 5.71s\tremaining: 19s\n",
      "231:\tlearn: 0.2210697\ttotal: 5.73s\tremaining: 19s\n",
      "232:\tlearn: 0.2210313\ttotal: 5.75s\tremaining: 18.9s\n",
      "233:\tlearn: 0.2209925\ttotal: 5.78s\tremaining: 18.9s\n",
      "234:\tlearn: 0.2209418\ttotal: 5.81s\tremaining: 18.9s\n",
      "235:\tlearn: 0.2209125\ttotal: 5.83s\tremaining: 18.9s\n",
      "236:\tlearn: 0.2208784\ttotal: 5.85s\tremaining: 18.8s\n",
      "237:\tlearn: 0.2208474\ttotal: 5.88s\tremaining: 18.8s\n",
      "238:\tlearn: 0.2208101\ttotal: 5.9s\tremaining: 18.8s\n",
      "239:\tlearn: 0.2207776\ttotal: 5.92s\tremaining: 18.8s\n",
      "240:\tlearn: 0.2207362\ttotal: 5.94s\tremaining: 18.7s\n",
      "241:\tlearn: 0.2207061\ttotal: 5.97s\tremaining: 18.7s\n",
      "242:\tlearn: 0.2206767\ttotal: 5.99s\tremaining: 18.7s\n",
      "243:\tlearn: 0.2206444\ttotal: 6.02s\tremaining: 18.7s\n",
      "244:\tlearn: 0.2206017\ttotal: 6.05s\tremaining: 18.6s\n",
      "245:\tlearn: 0.2205765\ttotal: 6.07s\tremaining: 18.6s\n",
      "246:\tlearn: 0.2205489\ttotal: 6.09s\tremaining: 18.6s\n",
      "247:\tlearn: 0.2205035\ttotal: 6.12s\tremaining: 18.5s\n",
      "248:\tlearn: 0.2204664\ttotal: 6.14s\tremaining: 18.5s\n",
      "249:\tlearn: 0.2204319\ttotal: 6.17s\tremaining: 18.5s\n",
      "250:\tlearn: 0.2203959\ttotal: 6.19s\tremaining: 18.5s\n",
      "251:\tlearn: 0.2203390\ttotal: 6.21s\tremaining: 18.4s\n",
      "252:\tlearn: 0.2202995\ttotal: 6.24s\tremaining: 18.4s\n",
      "253:\tlearn: 0.2202586\ttotal: 6.27s\tremaining: 18.4s\n",
      "254:\tlearn: 0.2202183\ttotal: 6.29s\tremaining: 18.4s\n",
      "255:\tlearn: 0.2201790\ttotal: 6.31s\tremaining: 18.3s\n",
      "256:\tlearn: 0.2201351\ttotal: 6.33s\tremaining: 18.3s\n",
      "257:\tlearn: 0.2201080\ttotal: 6.36s\tremaining: 18.3s\n",
      "258:\tlearn: 0.2200752\ttotal: 6.38s\tremaining: 18.3s\n",
      "259:\tlearn: 0.2200338\ttotal: 6.41s\tremaining: 18.2s\n",
      "260:\tlearn: 0.2199889\ttotal: 6.43s\tremaining: 18.2s\n",
      "261:\tlearn: 0.2199470\ttotal: 6.47s\tremaining: 18.2s\n",
      "262:\tlearn: 0.2199097\ttotal: 6.49s\tremaining: 18.2s\n",
      "263:\tlearn: 0.2198834\ttotal: 6.51s\tremaining: 18.2s\n",
      "264:\tlearn: 0.2198273\ttotal: 6.54s\tremaining: 18.1s\n",
      "265:\tlearn: 0.2197931\ttotal: 6.56s\tremaining: 18.1s\n",
      "266:\tlearn: 0.2197600\ttotal: 6.58s\tremaining: 18.1s\n",
      "267:\tlearn: 0.2197325\ttotal: 6.61s\tremaining: 18s\n",
      "268:\tlearn: 0.2196996\ttotal: 6.63s\tremaining: 18s\n",
      "269:\tlearn: 0.2196624\ttotal: 6.66s\tremaining: 18s\n",
      "270:\tlearn: 0.2196244\ttotal: 6.68s\tremaining: 18s\n",
      "271:\tlearn: 0.2195871\ttotal: 6.71s\tremaining: 18s\n",
      "272:\tlearn: 0.2195633\ttotal: 6.73s\tremaining: 17.9s\n",
      "273:\tlearn: 0.2195288\ttotal: 6.75s\tremaining: 17.9s\n",
      "274:\tlearn: 0.2194964\ttotal: 6.78s\tremaining: 17.9s\n",
      "275:\tlearn: 0.2194369\ttotal: 6.8s\tremaining: 17.8s\n",
      "276:\tlearn: 0.2194077\ttotal: 6.82s\tremaining: 17.8s\n",
      "277:\tlearn: 0.2193560\ttotal: 6.85s\tremaining: 17.8s\n",
      "278:\tlearn: 0.2193131\ttotal: 6.88s\tremaining: 17.8s\n",
      "279:\tlearn: 0.2192804\ttotal: 6.91s\tremaining: 17.8s\n",
      "280:\tlearn: 0.2192439\ttotal: 6.93s\tremaining: 17.7s\n",
      "281:\tlearn: 0.2192222\ttotal: 6.95s\tremaining: 17.7s\n",
      "282:\tlearn: 0.2191785\ttotal: 6.98s\tremaining: 17.7s\n",
      "283:\tlearn: 0.2191271\ttotal: 7s\tremaining: 17.7s\n",
      "284:\tlearn: 0.2190918\ttotal: 7.03s\tremaining: 17.6s\n",
      "285:\tlearn: 0.2190594\ttotal: 7.05s\tremaining: 17.6s\n",
      "286:\tlearn: 0.2190370\ttotal: 7.07s\tremaining: 17.6s\n",
      "287:\tlearn: 0.2189787\ttotal: 7.1s\tremaining: 17.6s\n",
      "288:\tlearn: 0.2189441\ttotal: 7.12s\tremaining: 17.5s\n",
      "289:\tlearn: 0.2189143\ttotal: 7.15s\tremaining: 17.5s\n",
      "290:\tlearn: 0.2188803\ttotal: 7.17s\tremaining: 17.5s\n",
      "291:\tlearn: 0.2188450\ttotal: 7.19s\tremaining: 17.4s\n",
      "292:\tlearn: 0.2188183\ttotal: 7.21s\tremaining: 17.4s\n",
      "293:\tlearn: 0.2187897\ttotal: 7.24s\tremaining: 17.4s\n",
      "294:\tlearn: 0.2187631\ttotal: 7.26s\tremaining: 17.4s\n",
      "295:\tlearn: 0.2187425\ttotal: 7.29s\tremaining: 17.3s\n",
      "296:\tlearn: 0.2187058\ttotal: 7.31s\tremaining: 17.3s\n",
      "297:\tlearn: 0.2186854\ttotal: 7.33s\tremaining: 17.3s\n",
      "298:\tlearn: 0.2186476\ttotal: 7.36s\tremaining: 17.3s\n",
      "299:\tlearn: 0.2185950\ttotal: 7.38s\tremaining: 17.2s\n",
      "300:\tlearn: 0.2185522\ttotal: 7.41s\tremaining: 17.2s\n",
      "301:\tlearn: 0.2185195\ttotal: 7.43s\tremaining: 17.2s\n",
      "302:\tlearn: 0.2184937\ttotal: 7.45s\tremaining: 17.1s\n",
      "303:\tlearn: 0.2184695\ttotal: 7.48s\tremaining: 17.1s\n",
      "304:\tlearn: 0.2184400\ttotal: 7.5s\tremaining: 17.1s\n",
      "305:\tlearn: 0.2184050\ttotal: 7.53s\tremaining: 17.1s\n",
      "306:\tlearn: 0.2183699\ttotal: 7.55s\tremaining: 17.1s\n",
      "307:\tlearn: 0.2183428\ttotal: 7.58s\tremaining: 17s\n",
      "308:\tlearn: 0.2183243\ttotal: 7.6s\tremaining: 17s\n",
      "309:\tlearn: 0.2183010\ttotal: 7.62s\tremaining: 17s\n",
      "310:\tlearn: 0.2182591\ttotal: 7.64s\tremaining: 16.9s\n",
      "311:\tlearn: 0.2182227\ttotal: 7.67s\tremaining: 16.9s\n",
      "312:\tlearn: 0.2181835\ttotal: 7.69s\tremaining: 16.9s\n",
      "313:\tlearn: 0.2181519\ttotal: 7.72s\tremaining: 16.9s\n",
      "314:\tlearn: 0.2181078\ttotal: 7.75s\tremaining: 16.9s\n",
      "315:\tlearn: 0.2180653\ttotal: 7.77s\tremaining: 16.8s\n",
      "316:\tlearn: 0.2180341\ttotal: 7.79s\tremaining: 16.8s\n",
      "317:\tlearn: 0.2180067\ttotal: 7.82s\tremaining: 16.8s\n",
      "318:\tlearn: 0.2179731\ttotal: 7.84s\tremaining: 16.7s\n",
      "319:\tlearn: 0.2179283\ttotal: 7.86s\tremaining: 16.7s\n",
      "320:\tlearn: 0.2178982\ttotal: 7.88s\tremaining: 16.7s\n",
      "321:\tlearn: 0.2178603\ttotal: 7.91s\tremaining: 16.7s\n",
      "322:\tlearn: 0.2178094\ttotal: 7.94s\tremaining: 16.6s\n",
      "323:\tlearn: 0.2177802\ttotal: 7.96s\tremaining: 16.6s\n",
      "324:\tlearn: 0.2177475\ttotal: 7.99s\tremaining: 16.6s\n",
      "325:\tlearn: 0.2177181\ttotal: 8.01s\tremaining: 16.6s\n",
      "326:\tlearn: 0.2176900\ttotal: 8.03s\tremaining: 16.5s\n",
      "327:\tlearn: 0.2176521\ttotal: 8.06s\tremaining: 16.5s\n",
      "328:\tlearn: 0.2176137\ttotal: 8.08s\tremaining: 16.5s\n",
      "329:\tlearn: 0.2175834\ttotal: 8.1s\tremaining: 16.4s\n",
      "330:\tlearn: 0.2175412\ttotal: 8.13s\tremaining: 16.4s\n",
      "331:\tlearn: 0.2174982\ttotal: 8.16s\tremaining: 16.4s\n",
      "332:\tlearn: 0.2174653\ttotal: 8.18s\tremaining: 16.4s\n",
      "333:\tlearn: 0.2174378\ttotal: 8.2s\tremaining: 16.4s\n",
      "334:\tlearn: 0.2174137\ttotal: 8.22s\tremaining: 16.3s\n",
      "335:\tlearn: 0.2173856\ttotal: 8.24s\tremaining: 16.3s\n",
      "336:\tlearn: 0.2173588\ttotal: 8.27s\tremaining: 16.3s\n",
      "337:\tlearn: 0.2173219\ttotal: 8.29s\tremaining: 16.2s\n",
      "338:\tlearn: 0.2173009\ttotal: 8.31s\tremaining: 16.2s\n",
      "339:\tlearn: 0.2172596\ttotal: 8.34s\tremaining: 16.2s\n",
      "340:\tlearn: 0.2172369\ttotal: 8.37s\tremaining: 16.2s\n",
      "341:\tlearn: 0.2172123\ttotal: 8.39s\tremaining: 16.1s\n",
      "342:\tlearn: 0.2171726\ttotal: 8.41s\tremaining: 16.1s\n",
      "343:\tlearn: 0.2171361\ttotal: 8.43s\tremaining: 16.1s\n",
      "344:\tlearn: 0.2170909\ttotal: 8.46s\tremaining: 16.1s\n",
      "345:\tlearn: 0.2170549\ttotal: 8.48s\tremaining: 16s\n",
      "346:\tlearn: 0.2170301\ttotal: 8.5s\tremaining: 16s\n",
      "347:\tlearn: 0.2169859\ttotal: 8.53s\tremaining: 16s\n",
      "348:\tlearn: 0.2169627\ttotal: 8.55s\tremaining: 16s\n",
      "349:\tlearn: 0.2169172\ttotal: 8.62s\tremaining: 16s\n",
      "350:\tlearn: 0.2168759\ttotal: 8.64s\tremaining: 16s\n",
      "351:\tlearn: 0.2168302\ttotal: 8.66s\tremaining: 15.9s\n",
      "352:\tlearn: 0.2167841\ttotal: 8.69s\tremaining: 15.9s\n",
      "353:\tlearn: 0.2167487\ttotal: 8.71s\tremaining: 15.9s\n",
      "354:\tlearn: 0.2167083\ttotal: 8.73s\tremaining: 15.9s\n",
      "355:\tlearn: 0.2166599\ttotal: 8.76s\tremaining: 15.8s\n",
      "356:\tlearn: 0.2166161\ttotal: 8.79s\tremaining: 15.8s\n",
      "357:\tlearn: 0.2165614\ttotal: 8.81s\tremaining: 15.8s\n",
      "358:\tlearn: 0.2165349\ttotal: 8.83s\tremaining: 15.8s\n",
      "359:\tlearn: 0.2165062\ttotal: 8.86s\tremaining: 15.7s\n",
      "360:\tlearn: 0.2164751\ttotal: 8.88s\tremaining: 15.7s\n",
      "361:\tlearn: 0.2164363\ttotal: 8.9s\tremaining: 15.7s\n",
      "362:\tlearn: 0.2163939\ttotal: 8.93s\tremaining: 15.7s\n",
      "363:\tlearn: 0.2163228\ttotal: 8.95s\tremaining: 15.6s\n",
      "364:\tlearn: 0.2162815\ttotal: 8.98s\tremaining: 15.6s\n",
      "365:\tlearn: 0.2162447\ttotal: 9.01s\tremaining: 15.6s\n",
      "366:\tlearn: 0.2162072\ttotal: 9.03s\tremaining: 15.6s\n",
      "367:\tlearn: 0.2161682\ttotal: 9.05s\tremaining: 15.5s\n",
      "368:\tlearn: 0.2161444\ttotal: 9.07s\tremaining: 15.5s\n",
      "369:\tlearn: 0.2160951\ttotal: 9.1s\tremaining: 15.5s\n",
      "370:\tlearn: 0.2160565\ttotal: 9.13s\tremaining: 15.5s\n",
      "371:\tlearn: 0.2160195\ttotal: 9.15s\tremaining: 15.4s\n",
      "372:\tlearn: 0.2159911\ttotal: 9.17s\tremaining: 15.4s\n",
      "373:\tlearn: 0.2159502\ttotal: 9.2s\tremaining: 15.4s\n",
      "374:\tlearn: 0.2159178\ttotal: 9.22s\tremaining: 15.4s\n",
      "375:\tlearn: 0.2158943\ttotal: 9.25s\tremaining: 15.3s\n",
      "376:\tlearn: 0.2158588\ttotal: 9.27s\tremaining: 15.3s\n",
      "377:\tlearn: 0.2158166\ttotal: 9.29s\tremaining: 15.3s\n",
      "378:\tlearn: 0.2157871\ttotal: 9.32s\tremaining: 15.3s\n",
      "379:\tlearn: 0.2157628\ttotal: 9.34s\tremaining: 15.2s\n",
      "380:\tlearn: 0.2157374\ttotal: 9.36s\tremaining: 15.2s\n",
      "381:\tlearn: 0.2156811\ttotal: 9.39s\tremaining: 15.2s\n",
      "382:\tlearn: 0.2156510\ttotal: 9.42s\tremaining: 15.2s\n",
      "383:\tlearn: 0.2156258\ttotal: 9.45s\tremaining: 15.2s\n",
      "384:\tlearn: 0.2155967\ttotal: 9.47s\tremaining: 15.1s\n",
      "385:\tlearn: 0.2155746\ttotal: 9.49s\tremaining: 15.1s\n",
      "386:\tlearn: 0.2155486\ttotal: 9.52s\tremaining: 15.1s\n",
      "387:\tlearn: 0.2154968\ttotal: 9.54s\tremaining: 15.1s\n",
      "388:\tlearn: 0.2154727\ttotal: 9.56s\tremaining: 15s\n",
      "389:\tlearn: 0.2154378\ttotal: 9.59s\tremaining: 15s\n",
      "390:\tlearn: 0.2154034\ttotal: 9.61s\tremaining: 15s\n",
      "391:\tlearn: 0.2153682\ttotal: 9.64s\tremaining: 15s\n",
      "392:\tlearn: 0.2153464\ttotal: 9.66s\tremaining: 14.9s\n",
      "393:\tlearn: 0.2153215\ttotal: 9.69s\tremaining: 14.9s\n",
      "394:\tlearn: 0.2152947\ttotal: 9.71s\tremaining: 14.9s\n",
      "395:\tlearn: 0.2152597\ttotal: 9.73s\tremaining: 14.8s\n",
      "396:\tlearn: 0.2152317\ttotal: 9.76s\tremaining: 14.8s\n",
      "397:\tlearn: 0.2152038\ttotal: 9.78s\tremaining: 14.8s\n",
      "398:\tlearn: 0.2151658\ttotal: 9.8s\tremaining: 14.8s\n",
      "399:\tlearn: 0.2151313\ttotal: 9.83s\tremaining: 14.7s\n",
      "400:\tlearn: 0.2151004\ttotal: 9.86s\tremaining: 14.7s\n",
      "401:\tlearn: 0.2150740\ttotal: 9.89s\tremaining: 14.7s\n",
      "402:\tlearn: 0.2150505\ttotal: 9.91s\tremaining: 14.7s\n",
      "403:\tlearn: 0.2149912\ttotal: 9.94s\tremaining: 14.7s\n",
      "404:\tlearn: 0.2149663\ttotal: 9.96s\tremaining: 14.6s\n",
      "405:\tlearn: 0.2149312\ttotal: 9.98s\tremaining: 14.6s\n",
      "406:\tlearn: 0.2149020\ttotal: 10s\tremaining: 14.6s\n",
      "407:\tlearn: 0.2148748\ttotal: 10s\tremaining: 14.5s\n",
      "408:\tlearn: 0.2148475\ttotal: 10.1s\tremaining: 14.5s\n",
      "409:\tlearn: 0.2148062\ttotal: 10.1s\tremaining: 14.5s\n",
      "410:\tlearn: 0.2147765\ttotal: 10.1s\tremaining: 14.5s\n",
      "411:\tlearn: 0.2147542\ttotal: 10.1s\tremaining: 14.5s\n",
      "412:\tlearn: 0.2147252\ttotal: 10.2s\tremaining: 14.4s\n",
      "413:\tlearn: 0.2146943\ttotal: 10.2s\tremaining: 14.4s\n",
      "414:\tlearn: 0.2146714\ttotal: 10.2s\tremaining: 14.4s\n",
      "415:\tlearn: 0.2146341\ttotal: 10.2s\tremaining: 14.4s\n",
      "416:\tlearn: 0.2146122\ttotal: 10.2s\tremaining: 14.3s\n",
      "417:\tlearn: 0.2145828\ttotal: 10.3s\tremaining: 14.3s\n",
      "418:\tlearn: 0.2145384\ttotal: 10.3s\tremaining: 14.3s\n",
      "419:\tlearn: 0.2145089\ttotal: 10.3s\tremaining: 14.3s\n",
      "420:\tlearn: 0.2144786\ttotal: 10.4s\tremaining: 14.2s\n",
      "421:\tlearn: 0.2144513\ttotal: 10.4s\tremaining: 14.2s\n",
      "422:\tlearn: 0.2144202\ttotal: 10.4s\tremaining: 14.2s\n",
      "423:\tlearn: 0.2143849\ttotal: 10.4s\tremaining: 14.2s\n",
      "424:\tlearn: 0.2143420\ttotal: 10.4s\tremaining: 14.1s\n",
      "425:\tlearn: 0.2143084\ttotal: 10.5s\tremaining: 14.1s\n",
      "426:\tlearn: 0.2142777\ttotal: 10.5s\tremaining: 14.1s\n",
      "427:\tlearn: 0.2142438\ttotal: 10.5s\tremaining: 14.1s\n",
      "428:\tlearn: 0.2142198\ttotal: 10.6s\tremaining: 14s\n",
      "429:\tlearn: 0.2141975\ttotal: 10.6s\tremaining: 14s\n",
      "430:\tlearn: 0.2141725\ttotal: 10.6s\tremaining: 14s\n",
      "431:\tlearn: 0.2141256\ttotal: 10.6s\tremaining: 14s\n",
      "432:\tlearn: 0.2140903\ttotal: 10.7s\tremaining: 13.9s\n",
      "433:\tlearn: 0.2140619\ttotal: 10.7s\tremaining: 13.9s\n",
      "434:\tlearn: 0.2140309\ttotal: 10.7s\tremaining: 13.9s\n",
      "435:\tlearn: 0.2140040\ttotal: 10.7s\tremaining: 13.9s\n",
      "436:\tlearn: 0.2139635\ttotal: 10.8s\tremaining: 13.9s\n",
      "437:\tlearn: 0.2139322\ttotal: 10.8s\tremaining: 13.8s\n",
      "438:\tlearn: 0.2139105\ttotal: 10.8s\tremaining: 13.8s\n",
      "439:\tlearn: 0.2138909\ttotal: 10.8s\tremaining: 13.8s\n",
      "440:\tlearn: 0.2138763\ttotal: 10.8s\tremaining: 13.7s\n",
      "441:\tlearn: 0.2138384\ttotal: 10.9s\tremaining: 13.7s\n",
      "442:\tlearn: 0.2138211\ttotal: 10.9s\tremaining: 13.7s\n",
      "443:\tlearn: 0.2137918\ttotal: 10.9s\tremaining: 13.7s\n",
      "444:\tlearn: 0.2137740\ttotal: 10.9s\tremaining: 13.7s\n",
      "445:\tlearn: 0.2137428\ttotal: 11s\tremaining: 13.6s\n",
      "446:\tlearn: 0.2137142\ttotal: 11s\tremaining: 13.6s\n",
      "447:\tlearn: 0.2136724\ttotal: 11s\tremaining: 13.6s\n",
      "448:\tlearn: 0.2136320\ttotal: 11s\tremaining: 13.6s\n",
      "449:\tlearn: 0.2135662\ttotal: 11.1s\tremaining: 13.5s\n",
      "450:\tlearn: 0.2135313\ttotal: 11.1s\tremaining: 13.5s\n",
      "451:\tlearn: 0.2135058\ttotal: 11.1s\tremaining: 13.5s\n",
      "452:\tlearn: 0.2134805\ttotal: 11.1s\tremaining: 13.5s\n",
      "453:\tlearn: 0.2134596\ttotal: 11.2s\tremaining: 13.4s\n",
      "454:\tlearn: 0.2134354\ttotal: 11.2s\tremaining: 13.4s\n",
      "455:\tlearn: 0.2134120\ttotal: 11.2s\tremaining: 13.4s\n",
      "456:\tlearn: 0.2133745\ttotal: 11.2s\tremaining: 13.4s\n",
      "457:\tlearn: 0.2133392\ttotal: 11.3s\tremaining: 13.3s\n",
      "458:\tlearn: 0.2132946\ttotal: 11.3s\tremaining: 13.3s\n",
      "459:\tlearn: 0.2132667\ttotal: 11.3s\tremaining: 13.3s\n",
      "460:\tlearn: 0.2132340\ttotal: 11.3s\tremaining: 13.3s\n",
      "461:\tlearn: 0.2132000\ttotal: 11.4s\tremaining: 13.2s\n",
      "462:\tlearn: 0.2131557\ttotal: 11.4s\tremaining: 13.2s\n",
      "463:\tlearn: 0.2131289\ttotal: 11.4s\tremaining: 13.2s\n",
      "464:\tlearn: 0.2130961\ttotal: 11.4s\tremaining: 13.2s\n",
      "465:\tlearn: 0.2130625\ttotal: 11.5s\tremaining: 13.1s\n",
      "466:\tlearn: 0.2130435\ttotal: 11.5s\tremaining: 13.1s\n",
      "467:\tlearn: 0.2130135\ttotal: 11.5s\tremaining: 13.1s\n",
      "468:\tlearn: 0.2129725\ttotal: 11.5s\tremaining: 13.1s\n",
      "469:\tlearn: 0.2129398\ttotal: 11.6s\tremaining: 13s\n",
      "470:\tlearn: 0.2129056\ttotal: 11.6s\tremaining: 13s\n",
      "471:\tlearn: 0.2128783\ttotal: 11.6s\tremaining: 13s\n",
      "472:\tlearn: 0.2128603\ttotal: 11.6s\tremaining: 13s\n",
      "473:\tlearn: 0.2128267\ttotal: 11.7s\tremaining: 12.9s\n",
      "474:\tlearn: 0.2127961\ttotal: 11.7s\tremaining: 12.9s\n",
      "475:\tlearn: 0.2127725\ttotal: 11.7s\tremaining: 12.9s\n",
      "476:\tlearn: 0.2127567\ttotal: 11.7s\tremaining: 12.9s\n",
      "477:\tlearn: 0.2127260\ttotal: 11.8s\tremaining: 12.8s\n",
      "478:\tlearn: 0.2127007\ttotal: 11.8s\tremaining: 12.8s\n",
      "479:\tlearn: 0.2126699\ttotal: 11.8s\tremaining: 12.8s\n",
      "480:\tlearn: 0.2126485\ttotal: 11.8s\tremaining: 12.8s\n",
      "481:\tlearn: 0.2126307\ttotal: 11.9s\tremaining: 12.7s\n",
      "482:\tlearn: 0.2126110\ttotal: 11.9s\tremaining: 12.7s\n",
      "483:\tlearn: 0.2125678\ttotal: 11.9s\tremaining: 12.7s\n",
      "484:\tlearn: 0.2125304\ttotal: 11.9s\tremaining: 12.7s\n",
      "485:\tlearn: 0.2124955\ttotal: 12s\tremaining: 12.7s\n",
      "486:\tlearn: 0.2124507\ttotal: 12s\tremaining: 12.6s\n",
      "487:\tlearn: 0.2124216\ttotal: 12s\tremaining: 12.6s\n",
      "488:\tlearn: 0.2123847\ttotal: 12s\tremaining: 12.6s\n",
      "489:\tlearn: 0.2123639\ttotal: 12.1s\tremaining: 12.6s\n",
      "490:\tlearn: 0.2123316\ttotal: 12.1s\tremaining: 12.5s\n",
      "491:\tlearn: 0.2122965\ttotal: 12.1s\tremaining: 12.5s\n",
      "492:\tlearn: 0.2122645\ttotal: 12.1s\tremaining: 12.5s\n",
      "493:\tlearn: 0.2122232\ttotal: 12.2s\tremaining: 12.5s\n",
      "494:\tlearn: 0.2122022\ttotal: 12.2s\tremaining: 12.4s\n",
      "495:\tlearn: 0.2121743\ttotal: 12.2s\tremaining: 12.4s\n",
      "496:\tlearn: 0.2121461\ttotal: 12.2s\tremaining: 12.4s\n",
      "497:\tlearn: 0.2121058\ttotal: 12.3s\tremaining: 12.4s\n",
      "498:\tlearn: 0.2120806\ttotal: 12.3s\tremaining: 12.3s\n",
      "499:\tlearn: 0.2120554\ttotal: 12.3s\tremaining: 12.3s\n",
      "500:\tlearn: 0.2120325\ttotal: 12.3s\tremaining: 12.3s\n",
      "501:\tlearn: 0.2120083\ttotal: 12.4s\tremaining: 12.3s\n",
      "502:\tlearn: 0.2119880\ttotal: 12.4s\tremaining: 12.2s\n",
      "503:\tlearn: 0.2119701\ttotal: 12.4s\tremaining: 12.2s\n",
      "504:\tlearn: 0.2119351\ttotal: 12.4s\tremaining: 12.2s\n",
      "505:\tlearn: 0.2119040\ttotal: 12.5s\tremaining: 12.2s\n",
      "506:\tlearn: 0.2118884\ttotal: 12.5s\tremaining: 12.1s\n",
      "507:\tlearn: 0.2118594\ttotal: 12.5s\tremaining: 12.1s\n",
      "508:\tlearn: 0.2118365\ttotal: 12.5s\tremaining: 12.1s\n",
      "509:\tlearn: 0.2118052\ttotal: 12.6s\tremaining: 12.1s\n",
      "510:\tlearn: 0.2117838\ttotal: 12.6s\tremaining: 12s\n",
      "511:\tlearn: 0.2117511\ttotal: 12.6s\tremaining: 12s\n",
      "512:\tlearn: 0.2117153\ttotal: 12.6s\tremaining: 12s\n",
      "513:\tlearn: 0.2116797\ttotal: 12.7s\tremaining: 12s\n",
      "514:\tlearn: 0.2116548\ttotal: 12.7s\tremaining: 12s\n",
      "515:\tlearn: 0.2116239\ttotal: 12.7s\tremaining: 11.9s\n",
      "516:\tlearn: 0.2115928\ttotal: 12.7s\tremaining: 11.9s\n",
      "517:\tlearn: 0.2115695\ttotal: 12.8s\tremaining: 11.9s\n",
      "518:\tlearn: 0.2115345\ttotal: 12.8s\tremaining: 11.8s\n",
      "519:\tlearn: 0.2115154\ttotal: 12.8s\tremaining: 11.8s\n",
      "520:\tlearn: 0.2114922\ttotal: 12.8s\tremaining: 11.8s\n",
      "521:\tlearn: 0.2114720\ttotal: 12.9s\tremaining: 11.8s\n",
      "522:\tlearn: 0.2114457\ttotal: 12.9s\tremaining: 11.7s\n",
      "523:\tlearn: 0.2114233\ttotal: 12.9s\tremaining: 11.8s\n",
      "524:\tlearn: 0.2114071\ttotal: 13s\tremaining: 11.7s\n",
      "525:\tlearn: 0.2113901\ttotal: 13s\tremaining: 11.7s\n",
      "526:\tlearn: 0.2113695\ttotal: 13s\tremaining: 11.7s\n",
      "527:\tlearn: 0.2113381\ttotal: 13s\tremaining: 11.7s\n",
      "528:\tlearn: 0.2113026\ttotal: 13.1s\tremaining: 11.6s\n",
      "529:\tlearn: 0.2112805\ttotal: 13.1s\tremaining: 11.6s\n",
      "530:\tlearn: 0.2112675\ttotal: 13.1s\tremaining: 11.6s\n",
      "531:\tlearn: 0.2112389\ttotal: 13.1s\tremaining: 11.6s\n",
      "532:\tlearn: 0.2112158\ttotal: 13.2s\tremaining: 11.5s\n",
      "533:\tlearn: 0.2111958\ttotal: 13.2s\tremaining: 11.5s\n",
      "534:\tlearn: 0.2111739\ttotal: 13.2s\tremaining: 11.5s\n",
      "535:\tlearn: 0.2111494\ttotal: 13.2s\tremaining: 11.5s\n",
      "536:\tlearn: 0.2111243\ttotal: 13.3s\tremaining: 11.4s\n",
      "537:\tlearn: 0.2110850\ttotal: 13.3s\tremaining: 11.4s\n",
      "538:\tlearn: 0.2110622\ttotal: 13.3s\tremaining: 11.4s\n",
      "539:\tlearn: 0.2110390\ttotal: 13.3s\tremaining: 11.4s\n",
      "540:\tlearn: 0.2110269\ttotal: 13.4s\tremaining: 11.3s\n",
      "541:\tlearn: 0.2109959\ttotal: 13.4s\tremaining: 11.3s\n",
      "542:\tlearn: 0.2109737\ttotal: 13.4s\tremaining: 11.3s\n",
      "543:\tlearn: 0.2109427\ttotal: 13.4s\tremaining: 11.3s\n",
      "544:\tlearn: 0.2109229\ttotal: 13.4s\tremaining: 11.2s\n",
      "545:\tlearn: 0.2109041\ttotal: 13.5s\tremaining: 11.2s\n",
      "546:\tlearn: 0.2108927\ttotal: 13.5s\tremaining: 11.2s\n",
      "547:\tlearn: 0.2108627\ttotal: 13.5s\tremaining: 11.1s\n",
      "548:\tlearn: 0.2108381\ttotal: 13.5s\tremaining: 11.1s\n",
      "549:\tlearn: 0.2108060\ttotal: 13.6s\tremaining: 11.1s\n",
      "550:\tlearn: 0.2107936\ttotal: 13.6s\tremaining: 11.1s\n",
      "551:\tlearn: 0.2107751\ttotal: 13.6s\tremaining: 11.1s\n",
      "552:\tlearn: 0.2107492\ttotal: 13.6s\tremaining: 11s\n",
      "553:\tlearn: 0.2107324\ttotal: 13.7s\tremaining: 11s\n",
      "554:\tlearn: 0.2107038\ttotal: 13.7s\tremaining: 11s\n",
      "555:\tlearn: 0.2106768\ttotal: 13.7s\tremaining: 10.9s\n",
      "556:\tlearn: 0.2106558\ttotal: 13.7s\tremaining: 10.9s\n",
      "557:\tlearn: 0.2106110\ttotal: 13.8s\tremaining: 10.9s\n",
      "558:\tlearn: 0.2105919\ttotal: 13.8s\tremaining: 10.9s\n",
      "559:\tlearn: 0.2105676\ttotal: 13.8s\tremaining: 10.9s\n",
      "560:\tlearn: 0.2105484\ttotal: 13.8s\tremaining: 10.8s\n",
      "561:\tlearn: 0.2105214\ttotal: 13.9s\tremaining: 10.8s\n",
      "562:\tlearn: 0.2104970\ttotal: 13.9s\tremaining: 10.8s\n",
      "563:\tlearn: 0.2104713\ttotal: 13.9s\tremaining: 10.7s\n",
      "564:\tlearn: 0.2104518\ttotal: 13.9s\tremaining: 10.7s\n",
      "565:\tlearn: 0.2104198\ttotal: 13.9s\tremaining: 10.7s\n",
      "566:\tlearn: 0.2103926\ttotal: 14s\tremaining: 10.7s\n",
      "567:\tlearn: 0.2103696\ttotal: 14s\tremaining: 10.7s\n",
      "568:\tlearn: 0.2103314\ttotal: 14s\tremaining: 10.6s\n",
      "569:\tlearn: 0.2103018\ttotal: 14.1s\tremaining: 10.6s\n",
      "570:\tlearn: 0.2102565\ttotal: 14.1s\tremaining: 10.6s\n",
      "571:\tlearn: 0.2102419\ttotal: 14.1s\tremaining: 10.6s\n",
      "572:\tlearn: 0.2101997\ttotal: 14.1s\tremaining: 10.5s\n",
      "573:\tlearn: 0.2101709\ttotal: 14.1s\tremaining: 10.5s\n",
      "574:\tlearn: 0.2101460\ttotal: 14.2s\tremaining: 10.5s\n",
      "575:\tlearn: 0.2101287\ttotal: 14.2s\tremaining: 10.5s\n",
      "576:\tlearn: 0.2100962\ttotal: 14.2s\tremaining: 10.4s\n",
      "577:\tlearn: 0.2100461\ttotal: 14.3s\tremaining: 10.4s\n",
      "578:\tlearn: 0.2100226\ttotal: 14.3s\tremaining: 10.4s\n",
      "579:\tlearn: 0.2099953\ttotal: 14.3s\tremaining: 10.4s\n",
      "580:\tlearn: 0.2099756\ttotal: 14.3s\tremaining: 10.3s\n",
      "581:\tlearn: 0.2099402\ttotal: 14.3s\tremaining: 10.3s\n",
      "582:\tlearn: 0.2099126\ttotal: 14.4s\tremaining: 10.3s\n",
      "583:\tlearn: 0.2098790\ttotal: 14.4s\tremaining: 10.3s\n",
      "584:\tlearn: 0.2098721\ttotal: 14.4s\tremaining: 10.2s\n",
      "585:\tlearn: 0.2098417\ttotal: 14.5s\tremaining: 10.2s\n",
      "586:\tlearn: 0.2098162\ttotal: 14.5s\tremaining: 10.2s\n",
      "587:\tlearn: 0.2097673\ttotal: 14.5s\tremaining: 10.2s\n",
      "588:\tlearn: 0.2097421\ttotal: 14.5s\tremaining: 10.1s\n",
      "589:\tlearn: 0.2097140\ttotal: 14.6s\tremaining: 10.1s\n",
      "590:\tlearn: 0.2096879\ttotal: 14.6s\tremaining: 10.1s\n",
      "591:\tlearn: 0.2096649\ttotal: 14.6s\tremaining: 10.1s\n",
      "592:\tlearn: 0.2096254\ttotal: 14.6s\tremaining: 10s\n",
      "593:\tlearn: 0.2096051\ttotal: 14.7s\tremaining: 10s\n",
      "594:\tlearn: 0.2095619\ttotal: 14.7s\tremaining: 9.99s\n",
      "595:\tlearn: 0.2095478\ttotal: 14.7s\tremaining: 9.97s\n",
      "596:\tlearn: 0.2095194\ttotal: 14.7s\tremaining: 9.94s\n",
      "597:\tlearn: 0.2095105\ttotal: 14.8s\tremaining: 9.92s\n",
      "598:\tlearn: 0.2094770\ttotal: 14.8s\tremaining: 9.9s\n",
      "599:\tlearn: 0.2094384\ttotal: 14.8s\tremaining: 9.87s\n",
      "600:\tlearn: 0.2094018\ttotal: 14.8s\tremaining: 9.85s\n",
      "601:\tlearn: 0.2093654\ttotal: 14.9s\tremaining: 9.82s\n",
      "602:\tlearn: 0.2093408\ttotal: 14.9s\tremaining: 9.8s\n",
      "603:\tlearn: 0.2093153\ttotal: 14.9s\tremaining: 9.77s\n",
      "604:\tlearn: 0.2092955\ttotal: 14.9s\tremaining: 9.75s\n",
      "605:\tlearn: 0.2092828\ttotal: 14.9s\tremaining: 9.72s\n",
      "606:\tlearn: 0.2092599\ttotal: 15s\tremaining: 9.69s\n",
      "607:\tlearn: 0.2092377\ttotal: 15s\tremaining: 9.67s\n",
      "608:\tlearn: 0.2091942\ttotal: 15s\tremaining: 9.65s\n",
      "609:\tlearn: 0.2091551\ttotal: 15.1s\tremaining: 9.62s\n",
      "610:\tlearn: 0.2091225\ttotal: 15.1s\tremaining: 9.6s\n",
      "611:\tlearn: 0.2090948\ttotal: 15.1s\tremaining: 9.57s\n",
      "612:\tlearn: 0.2090721\ttotal: 15.1s\tremaining: 9.55s\n",
      "613:\tlearn: 0.2090620\ttotal: 15.2s\tremaining: 9.53s\n",
      "614:\tlearn: 0.2090318\ttotal: 15.2s\tremaining: 9.5s\n",
      "615:\tlearn: 0.2089925\ttotal: 15.2s\tremaining: 9.48s\n",
      "616:\tlearn: 0.2089648\ttotal: 15.2s\tremaining: 9.45s\n",
      "617:\tlearn: 0.2089189\ttotal: 15.3s\tremaining: 9.43s\n",
      "618:\tlearn: 0.2088825\ttotal: 15.3s\tremaining: 9.4s\n",
      "619:\tlearn: 0.2088679\ttotal: 15.3s\tremaining: 9.38s\n",
      "620:\tlearn: 0.2088339\ttotal: 15.3s\tremaining: 9.35s\n",
      "621:\tlearn: 0.2088141\ttotal: 15.3s\tremaining: 9.33s\n",
      "622:\tlearn: 0.2087936\ttotal: 15.4s\tremaining: 9.3s\n",
      "623:\tlearn: 0.2087610\ttotal: 15.4s\tremaining: 9.28s\n",
      "624:\tlearn: 0.2087325\ttotal: 15.4s\tremaining: 9.26s\n",
      "625:\tlearn: 0.2087135\ttotal: 15.4s\tremaining: 9.23s\n",
      "626:\tlearn: 0.2086894\ttotal: 15.5s\tremaining: 9.2s\n",
      "627:\tlearn: 0.2086547\ttotal: 15.5s\tremaining: 9.18s\n",
      "628:\tlearn: 0.2086237\ttotal: 15.5s\tremaining: 9.15s\n",
      "629:\tlearn: 0.2086110\ttotal: 15.5s\tremaining: 9.13s\n",
      "630:\tlearn: 0.2085715\ttotal: 15.6s\tremaining: 9.1s\n",
      "631:\tlearn: 0.2085655\ttotal: 15.6s\tremaining: 9.08s\n",
      "632:\tlearn: 0.2085369\ttotal: 15.6s\tremaining: 9.05s\n",
      "633:\tlearn: 0.2085217\ttotal: 15.6s\tremaining: 9.03s\n",
      "634:\tlearn: 0.2084853\ttotal: 15.7s\tremaining: 9s\n",
      "635:\tlearn: 0.2084479\ttotal: 15.7s\tremaining: 8.98s\n",
      "636:\tlearn: 0.2084207\ttotal: 15.7s\tremaining: 8.95s\n",
      "637:\tlearn: 0.2083792\ttotal: 15.7s\tremaining: 8.93s\n",
      "638:\tlearn: 0.2083497\ttotal: 15.8s\tremaining: 8.9s\n",
      "639:\tlearn: 0.2083215\ttotal: 15.8s\tremaining: 8.88s\n",
      "640:\tlearn: 0.2082983\ttotal: 15.8s\tremaining: 8.85s\n",
      "641:\tlearn: 0.2082635\ttotal: 15.8s\tremaining: 8.83s\n",
      "642:\tlearn: 0.2082429\ttotal: 15.9s\tremaining: 8.8s\n",
      "643:\tlearn: 0.2081911\ttotal: 15.9s\tremaining: 8.78s\n",
      "644:\tlearn: 0.2081615\ttotal: 15.9s\tremaining: 8.75s\n",
      "645:\tlearn: 0.2081249\ttotal: 15.9s\tremaining: 8.72s\n",
      "646:\tlearn: 0.2080862\ttotal: 15.9s\tremaining: 8.7s\n",
      "647:\tlearn: 0.2080514\ttotal: 16s\tremaining: 8.68s\n",
      "648:\tlearn: 0.2080042\ttotal: 16s\tremaining: 8.66s\n",
      "649:\tlearn: 0.2079849\ttotal: 16s\tremaining: 8.63s\n",
      "650:\tlearn: 0.2079426\ttotal: 16.1s\tremaining: 8.61s\n",
      "651:\tlearn: 0.2079158\ttotal: 16.1s\tremaining: 8.58s\n",
      "652:\tlearn: 0.2079008\ttotal: 16.1s\tremaining: 8.55s\n",
      "653:\tlearn: 0.2078793\ttotal: 16.1s\tremaining: 8.53s\n",
      "654:\tlearn: 0.2078520\ttotal: 16.1s\tremaining: 8.5s\n",
      "655:\tlearn: 0.2078278\ttotal: 16.2s\tremaining: 8.48s\n",
      "656:\tlearn: 0.2077996\ttotal: 16.2s\tremaining: 8.45s\n",
      "657:\tlearn: 0.2077762\ttotal: 16.2s\tremaining: 8.43s\n",
      "658:\tlearn: 0.2077433\ttotal: 16.2s\tremaining: 8.4s\n",
      "659:\tlearn: 0.2077286\ttotal: 16.3s\tremaining: 8.38s\n",
      "660:\tlearn: 0.2076943\ttotal: 16.3s\tremaining: 8.35s\n",
      "661:\tlearn: 0.2076743\ttotal: 16.3s\tremaining: 8.32s\n",
      "662:\tlearn: 0.2076369\ttotal: 16.3s\tremaining: 8.3s\n",
      "663:\tlearn: 0.2076190\ttotal: 16.4s\tremaining: 8.28s\n",
      "664:\tlearn: 0.2076055\ttotal: 16.4s\tremaining: 8.25s\n",
      "665:\tlearn: 0.2075720\ttotal: 16.4s\tremaining: 8.22s\n",
      "666:\tlearn: 0.2075250\ttotal: 16.4s\tremaining: 8.2s\n",
      "667:\tlearn: 0.2075037\ttotal: 16.5s\tremaining: 8.18s\n",
      "668:\tlearn: 0.2074788\ttotal: 16.5s\tremaining: 8.15s\n",
      "669:\tlearn: 0.2074560\ttotal: 16.5s\tremaining: 8.13s\n",
      "670:\tlearn: 0.2074244\ttotal: 16.5s\tremaining: 8.1s\n",
      "671:\tlearn: 0.2074075\ttotal: 16.6s\tremaining: 8.08s\n",
      "672:\tlearn: 0.2073882\ttotal: 16.6s\tremaining: 8.05s\n",
      "673:\tlearn: 0.2073636\ttotal: 16.6s\tremaining: 8.03s\n",
      "674:\tlearn: 0.2073370\ttotal: 16.6s\tremaining: 8.01s\n",
      "675:\tlearn: 0.2073258\ttotal: 16.7s\tremaining: 7.98s\n",
      "676:\tlearn: 0.2073025\ttotal: 16.7s\tremaining: 7.96s\n",
      "677:\tlearn: 0.2072856\ttotal: 16.7s\tremaining: 7.93s\n",
      "678:\tlearn: 0.2072687\ttotal: 16.7s\tremaining: 7.9s\n",
      "679:\tlearn: 0.2072318\ttotal: 16.7s\tremaining: 7.88s\n",
      "680:\tlearn: 0.2072119\ttotal: 16.8s\tremaining: 7.86s\n",
      "681:\tlearn: 0.2071996\ttotal: 16.8s\tremaining: 7.83s\n",
      "682:\tlearn: 0.2071793\ttotal: 16.8s\tremaining: 7.8s\n",
      "683:\tlearn: 0.2071364\ttotal: 16.8s\tremaining: 7.78s\n",
      "684:\tlearn: 0.2071052\ttotal: 16.9s\tremaining: 7.76s\n",
      "685:\tlearn: 0.2070831\ttotal: 16.9s\tremaining: 7.73s\n",
      "686:\tlearn: 0.2070595\ttotal: 16.9s\tremaining: 7.71s\n",
      "687:\tlearn: 0.2070362\ttotal: 16.9s\tremaining: 7.68s\n",
      "688:\tlearn: 0.2070184\ttotal: 17s\tremaining: 7.66s\n",
      "689:\tlearn: 0.2069916\ttotal: 17s\tremaining: 7.63s\n",
      "690:\tlearn: 0.2069802\ttotal: 17s\tremaining: 7.61s\n",
      "691:\tlearn: 0.2069490\ttotal: 17s\tremaining: 7.58s\n",
      "692:\tlearn: 0.2069315\ttotal: 17.1s\tremaining: 7.56s\n",
      "693:\tlearn: 0.2069045\ttotal: 17.1s\tremaining: 7.53s\n",
      "694:\tlearn: 0.2068839\ttotal: 17.1s\tremaining: 7.51s\n",
      "695:\tlearn: 0.2068474\ttotal: 17.1s\tremaining: 7.48s\n",
      "696:\tlearn: 0.2068372\ttotal: 17.2s\tremaining: 7.46s\n",
      "697:\tlearn: 0.2067977\ttotal: 17.2s\tremaining: 7.43s\n",
      "698:\tlearn: 0.2067754\ttotal: 17.2s\tremaining: 7.41s\n",
      "699:\tlearn: 0.2067446\ttotal: 17.2s\tremaining: 7.38s\n",
      "700:\tlearn: 0.2067195\ttotal: 17.3s\tremaining: 7.36s\n",
      "701:\tlearn: 0.2066995\ttotal: 17.3s\tremaining: 7.34s\n",
      "702:\tlearn: 0.2066745\ttotal: 17.3s\tremaining: 7.31s\n",
      "703:\tlearn: 0.2066542\ttotal: 17.3s\tremaining: 7.29s\n",
      "704:\tlearn: 0.2066282\ttotal: 17.4s\tremaining: 7.26s\n",
      "705:\tlearn: 0.2066037\ttotal: 17.4s\tremaining: 7.24s\n",
      "706:\tlearn: 0.2065885\ttotal: 17.4s\tremaining: 7.21s\n",
      "707:\tlearn: 0.2065554\ttotal: 17.4s\tremaining: 7.19s\n",
      "708:\tlearn: 0.2065503\ttotal: 17.4s\tremaining: 7.16s\n",
      "709:\tlearn: 0.2065218\ttotal: 17.5s\tremaining: 7.14s\n",
      "710:\tlearn: 0.2064989\ttotal: 17.5s\tremaining: 7.11s\n",
      "711:\tlearn: 0.2064813\ttotal: 17.5s\tremaining: 7.09s\n",
      "712:\tlearn: 0.2064445\ttotal: 17.5s\tremaining: 7.06s\n",
      "713:\tlearn: 0.2064234\ttotal: 17.6s\tremaining: 7.04s\n",
      "714:\tlearn: 0.2063968\ttotal: 17.6s\tremaining: 7.01s\n",
      "715:\tlearn: 0.2063692\ttotal: 17.6s\tremaining: 6.99s\n",
      "716:\tlearn: 0.2063516\ttotal: 17.6s\tremaining: 6.96s\n",
      "717:\tlearn: 0.2063374\ttotal: 17.7s\tremaining: 6.93s\n",
      "718:\tlearn: 0.2063163\ttotal: 17.7s\tremaining: 6.91s\n",
      "719:\tlearn: 0.2062860\ttotal: 17.7s\tremaining: 6.89s\n",
      "720:\tlearn: 0.2062640\ttotal: 17.7s\tremaining: 6.86s\n",
      "721:\tlearn: 0.2062230\ttotal: 17.8s\tremaining: 6.84s\n",
      "722:\tlearn: 0.2061982\ttotal: 17.8s\tremaining: 6.81s\n",
      "723:\tlearn: 0.2061836\ttotal: 17.8s\tremaining: 6.79s\n",
      "724:\tlearn: 0.2061466\ttotal: 17.8s\tremaining: 6.76s\n",
      "725:\tlearn: 0.2061315\ttotal: 17.9s\tremaining: 6.74s\n",
      "726:\tlearn: 0.2061066\ttotal: 17.9s\tremaining: 6.72s\n",
      "727:\tlearn: 0.2060883\ttotal: 17.9s\tremaining: 6.7s\n",
      "728:\tlearn: 0.2060684\ttotal: 18s\tremaining: 6.68s\n",
      "729:\tlearn: 0.2060386\ttotal: 18s\tremaining: 6.66s\n",
      "730:\tlearn: 0.2060102\ttotal: 18s\tremaining: 6.63s\n",
      "731:\tlearn: 0.2059872\ttotal: 18s\tremaining: 6.61s\n",
      "732:\tlearn: 0.2059693\ttotal: 18.1s\tremaining: 6.58s\n",
      "733:\tlearn: 0.2059597\ttotal: 18.1s\tremaining: 6.56s\n",
      "734:\tlearn: 0.2059305\ttotal: 18.1s\tremaining: 6.53s\n",
      "735:\tlearn: 0.2059079\ttotal: 18.1s\tremaining: 6.51s\n",
      "736:\tlearn: 0.2058810\ttotal: 18.2s\tremaining: 6.48s\n",
      "737:\tlearn: 0.2058616\ttotal: 18.2s\tremaining: 6.46s\n",
      "738:\tlearn: 0.2058392\ttotal: 18.2s\tremaining: 6.43s\n",
      "739:\tlearn: 0.2058224\ttotal: 18.2s\tremaining: 6.41s\n",
      "740:\tlearn: 0.2057912\ttotal: 18.3s\tremaining: 6.38s\n",
      "741:\tlearn: 0.2057710\ttotal: 18.3s\tremaining: 6.36s\n",
      "742:\tlearn: 0.2057536\ttotal: 18.3s\tremaining: 6.34s\n",
      "743:\tlearn: 0.2057341\ttotal: 18.3s\tremaining: 6.31s\n",
      "744:\tlearn: 0.2057215\ttotal: 18.4s\tremaining: 6.29s\n",
      "745:\tlearn: 0.2056969\ttotal: 18.4s\tremaining: 6.26s\n",
      "746:\tlearn: 0.2056663\ttotal: 18.4s\tremaining: 6.24s\n",
      "747:\tlearn: 0.2056489\ttotal: 18.4s\tremaining: 6.21s\n",
      "748:\tlearn: 0.2056033\ttotal: 18.5s\tremaining: 6.19s\n",
      "749:\tlearn: 0.2055826\ttotal: 18.5s\tremaining: 6.16s\n",
      "750:\tlearn: 0.2055560\ttotal: 18.5s\tremaining: 6.14s\n",
      "751:\tlearn: 0.2055290\ttotal: 18.5s\tremaining: 6.11s\n",
      "752:\tlearn: 0.2055136\ttotal: 18.6s\tremaining: 6.09s\n",
      "753:\tlearn: 0.2054724\ttotal: 18.6s\tremaining: 6.06s\n",
      "754:\tlearn: 0.2054309\ttotal: 18.6s\tremaining: 6.04s\n",
      "755:\tlearn: 0.2053838\ttotal: 18.6s\tremaining: 6.01s\n",
      "756:\tlearn: 0.2053559\ttotal: 18.7s\tremaining: 5.99s\n",
      "757:\tlearn: 0.2053146\ttotal: 18.7s\tremaining: 5.96s\n",
      "758:\tlearn: 0.2052784\ttotal: 18.7s\tremaining: 5.94s\n",
      "759:\tlearn: 0.2052559\ttotal: 18.7s\tremaining: 5.92s\n",
      "760:\tlearn: 0.2052402\ttotal: 18.8s\tremaining: 5.89s\n",
      "761:\tlearn: 0.2052183\ttotal: 18.8s\tremaining: 5.87s\n",
      "762:\tlearn: 0.2052013\ttotal: 18.8s\tremaining: 5.84s\n",
      "763:\tlearn: 0.2051915\ttotal: 18.8s\tremaining: 5.82s\n",
      "764:\tlearn: 0.2051672\ttotal: 18.9s\tremaining: 5.79s\n",
      "765:\tlearn: 0.2051444\ttotal: 18.9s\tremaining: 5.77s\n",
      "766:\tlearn: 0.2051305\ttotal: 18.9s\tremaining: 5.74s\n",
      "767:\tlearn: 0.2050982\ttotal: 18.9s\tremaining: 5.72s\n",
      "768:\tlearn: 0.2050666\ttotal: 19s\tremaining: 5.7s\n",
      "769:\tlearn: 0.2050370\ttotal: 19s\tremaining: 5.67s\n",
      "770:\tlearn: 0.2050054\ttotal: 19s\tremaining: 5.64s\n",
      "771:\tlearn: 0.2049868\ttotal: 19s\tremaining: 5.62s\n",
      "772:\tlearn: 0.2049551\ttotal: 19.1s\tremaining: 5.59s\n",
      "773:\tlearn: 0.2049226\ttotal: 19.1s\tremaining: 5.57s\n",
      "774:\tlearn: 0.2049027\ttotal: 19.1s\tremaining: 5.54s\n",
      "775:\tlearn: 0.2048688\ttotal: 19.1s\tremaining: 5.52s\n",
      "776:\tlearn: 0.2048462\ttotal: 19.2s\tremaining: 5.5s\n",
      "777:\tlearn: 0.2048366\ttotal: 19.2s\tremaining: 5.47s\n",
      "778:\tlearn: 0.2048160\ttotal: 19.2s\tremaining: 5.45s\n",
      "779:\tlearn: 0.2047976\ttotal: 19.2s\tremaining: 5.42s\n",
      "780:\tlearn: 0.2047706\ttotal: 19.3s\tremaining: 5.4s\n",
      "781:\tlearn: 0.2047525\ttotal: 19.3s\tremaining: 5.37s\n",
      "782:\tlearn: 0.2047412\ttotal: 19.3s\tremaining: 5.35s\n",
      "783:\tlearn: 0.2047154\ttotal: 19.3s\tremaining: 5.33s\n",
      "784:\tlearn: 0.2046906\ttotal: 19.4s\tremaining: 5.31s\n",
      "785:\tlearn: 0.2046702\ttotal: 19.4s\tremaining: 5.28s\n",
      "786:\tlearn: 0.2046353\ttotal: 19.4s\tremaining: 5.26s\n",
      "787:\tlearn: 0.2046141\ttotal: 19.5s\tremaining: 5.24s\n",
      "788:\tlearn: 0.2045985\ttotal: 19.5s\tremaining: 5.21s\n",
      "789:\tlearn: 0.2045689\ttotal: 19.5s\tremaining: 5.18s\n",
      "790:\tlearn: 0.2045495\ttotal: 19.5s\tremaining: 5.16s\n",
      "791:\tlearn: 0.2045134\ttotal: 19.6s\tremaining: 5.13s\n",
      "792:\tlearn: 0.2044746\ttotal: 19.6s\tremaining: 5.11s\n",
      "793:\tlearn: 0.2044370\ttotal: 19.6s\tremaining: 5.09s\n",
      "794:\tlearn: 0.2044197\ttotal: 19.6s\tremaining: 5.06s\n",
      "795:\tlearn: 0.2043934\ttotal: 19.7s\tremaining: 5.04s\n",
      "796:\tlearn: 0.2043608\ttotal: 19.7s\tremaining: 5.01s\n",
      "797:\tlearn: 0.2043354\ttotal: 19.7s\tremaining: 4.99s\n",
      "798:\tlearn: 0.2043173\ttotal: 19.7s\tremaining: 4.96s\n",
      "799:\tlearn: 0.2042907\ttotal: 19.7s\tremaining: 4.94s\n",
      "800:\tlearn: 0.2042548\ttotal: 19.8s\tremaining: 4.91s\n",
      "801:\tlearn: 0.2042382\ttotal: 19.8s\tremaining: 4.89s\n",
      "802:\tlearn: 0.2042179\ttotal: 19.8s\tremaining: 4.86s\n",
      "803:\tlearn: 0.2041941\ttotal: 19.8s\tremaining: 4.84s\n",
      "804:\tlearn: 0.2041758\ttotal: 19.9s\tremaining: 4.81s\n",
      "805:\tlearn: 0.2041580\ttotal: 19.9s\tremaining: 4.79s\n",
      "806:\tlearn: 0.2041331\ttotal: 19.9s\tremaining: 4.76s\n",
      "807:\tlearn: 0.2041166\ttotal: 19.9s\tremaining: 4.74s\n",
      "808:\tlearn: 0.2040863\ttotal: 20s\tremaining: 4.71s\n",
      "809:\tlearn: 0.2040693\ttotal: 20s\tremaining: 4.69s\n",
      "810:\tlearn: 0.2040409\ttotal: 20s\tremaining: 4.66s\n",
      "811:\tlearn: 0.2040110\ttotal: 20s\tremaining: 4.64s\n",
      "812:\tlearn: 0.2039800\ttotal: 20.1s\tremaining: 4.62s\n",
      "813:\tlearn: 0.2039536\ttotal: 20.1s\tremaining: 4.59s\n",
      "814:\tlearn: 0.2039155\ttotal: 20.1s\tremaining: 4.57s\n",
      "815:\tlearn: 0.2038907\ttotal: 20.1s\tremaining: 4.54s\n",
      "816:\tlearn: 0.2038614\ttotal: 20.2s\tremaining: 4.52s\n",
      "817:\tlearn: 0.2038375\ttotal: 20.2s\tremaining: 4.49s\n",
      "818:\tlearn: 0.2037999\ttotal: 20.2s\tremaining: 4.47s\n",
      "819:\tlearn: 0.2037761\ttotal: 20.2s\tremaining: 4.44s\n",
      "820:\tlearn: 0.2037426\ttotal: 20.3s\tremaining: 4.42s\n",
      "821:\tlearn: 0.2037153\ttotal: 20.3s\tremaining: 4.39s\n",
      "822:\tlearn: 0.2036935\ttotal: 20.3s\tremaining: 4.37s\n",
      "823:\tlearn: 0.2036731\ttotal: 20.3s\tremaining: 4.34s\n",
      "824:\tlearn: 0.2036508\ttotal: 20.4s\tremaining: 4.32s\n",
      "825:\tlearn: 0.2036252\ttotal: 20.4s\tremaining: 4.29s\n",
      "826:\tlearn: 0.2035907\ttotal: 20.4s\tremaining: 4.27s\n",
      "827:\tlearn: 0.2035747\ttotal: 20.4s\tremaining: 4.25s\n",
      "828:\tlearn: 0.2035510\ttotal: 20.5s\tremaining: 4.22s\n",
      "829:\tlearn: 0.2035324\ttotal: 20.5s\tremaining: 4.2s\n",
      "830:\tlearn: 0.2035208\ttotal: 20.5s\tremaining: 4.17s\n",
      "831:\tlearn: 0.2035055\ttotal: 20.5s\tremaining: 4.15s\n",
      "832:\tlearn: 0.2034712\ttotal: 20.6s\tremaining: 4.12s\n",
      "833:\tlearn: 0.2034418\ttotal: 20.6s\tremaining: 4.1s\n",
      "834:\tlearn: 0.2034133\ttotal: 20.6s\tremaining: 4.08s\n",
      "835:\tlearn: 0.2033912\ttotal: 20.7s\tremaining: 4.05s\n",
      "836:\tlearn: 0.2033728\ttotal: 20.7s\tremaining: 4.03s\n",
      "837:\tlearn: 0.2033525\ttotal: 20.7s\tremaining: 4s\n",
      "838:\tlearn: 0.2033230\ttotal: 20.7s\tremaining: 3.98s\n",
      "839:\tlearn: 0.2033056\ttotal: 20.8s\tremaining: 3.95s\n",
      "840:\tlearn: 0.2032654\ttotal: 20.8s\tremaining: 3.93s\n",
      "841:\tlearn: 0.2032503\ttotal: 20.8s\tremaining: 3.9s\n",
      "842:\tlearn: 0.2032267\ttotal: 20.8s\tremaining: 3.88s\n",
      "843:\tlearn: 0.2032010\ttotal: 20.9s\tremaining: 3.86s\n",
      "844:\tlearn: 0.2031692\ttotal: 20.9s\tremaining: 3.84s\n",
      "845:\tlearn: 0.2031495\ttotal: 20.9s\tremaining: 3.81s\n",
      "846:\tlearn: 0.2031319\ttotal: 21s\tremaining: 3.79s\n",
      "847:\tlearn: 0.2031080\ttotal: 21s\tremaining: 3.76s\n",
      "848:\tlearn: 0.2030838\ttotal: 21s\tremaining: 3.74s\n",
      "849:\tlearn: 0.2030703\ttotal: 21s\tremaining: 3.71s\n",
      "850:\tlearn: 0.2030446\ttotal: 21.1s\tremaining: 3.69s\n",
      "851:\tlearn: 0.2030269\ttotal: 21.1s\tremaining: 3.66s\n",
      "852:\tlearn: 0.2029978\ttotal: 21.1s\tremaining: 3.64s\n",
      "853:\tlearn: 0.2029639\ttotal: 21.1s\tremaining: 3.61s\n",
      "854:\tlearn: 0.2029281\ttotal: 21.2s\tremaining: 3.59s\n",
      "855:\tlearn: 0.2028957\ttotal: 21.2s\tremaining: 3.56s\n",
      "856:\tlearn: 0.2028697\ttotal: 21.2s\tremaining: 3.54s\n",
      "857:\tlearn: 0.2028483\ttotal: 21.2s\tremaining: 3.52s\n",
      "858:\tlearn: 0.2028318\ttotal: 21.3s\tremaining: 3.49s\n",
      "859:\tlearn: 0.2027963\ttotal: 21.3s\tremaining: 3.46s\n",
      "860:\tlearn: 0.2027801\ttotal: 21.3s\tremaining: 3.44s\n",
      "861:\tlearn: 0.2027612\ttotal: 21.3s\tremaining: 3.42s\n",
      "862:\tlearn: 0.2027391\ttotal: 21.4s\tremaining: 3.39s\n",
      "863:\tlearn: 0.2027145\ttotal: 21.4s\tremaining: 3.37s\n",
      "864:\tlearn: 0.2027030\ttotal: 21.4s\tremaining: 3.34s\n",
      "865:\tlearn: 0.2026894\ttotal: 21.4s\tremaining: 3.32s\n",
      "866:\tlearn: 0.2026597\ttotal: 21.5s\tremaining: 3.29s\n",
      "867:\tlearn: 0.2026402\ttotal: 21.5s\tremaining: 3.27s\n",
      "868:\tlearn: 0.2026137\ttotal: 21.5s\tremaining: 3.25s\n",
      "869:\tlearn: 0.2025901\ttotal: 21.6s\tremaining: 3.23s\n",
      "870:\tlearn: 0.2025767\ttotal: 21.7s\tremaining: 3.21s\n",
      "871:\tlearn: 0.2025568\ttotal: 21.7s\tremaining: 3.19s\n",
      "872:\tlearn: 0.2025169\ttotal: 21.7s\tremaining: 3.16s\n",
      "873:\tlearn: 0.2024946\ttotal: 21.8s\tremaining: 3.14s\n",
      "874:\tlearn: 0.2024644\ttotal: 21.8s\tremaining: 3.11s\n",
      "875:\tlearn: 0.2024429\ttotal: 21.8s\tremaining: 3.09s\n",
      "876:\tlearn: 0.2024192\ttotal: 21.8s\tremaining: 3.06s\n",
      "877:\tlearn: 0.2024049\ttotal: 21.8s\tremaining: 3.04s\n",
      "878:\tlearn: 0.2023907\ttotal: 21.9s\tremaining: 3.01s\n",
      "879:\tlearn: 0.2023725\ttotal: 21.9s\tremaining: 2.99s\n",
      "880:\tlearn: 0.2023399\ttotal: 21.9s\tremaining: 2.96s\n",
      "881:\tlearn: 0.2022861\ttotal: 21.9s\tremaining: 2.94s\n",
      "882:\tlearn: 0.2022610\ttotal: 22s\tremaining: 2.91s\n",
      "883:\tlearn: 0.2022548\ttotal: 22s\tremaining: 2.89s\n",
      "884:\tlearn: 0.2022310\ttotal: 22s\tremaining: 2.86s\n",
      "885:\tlearn: 0.2022083\ttotal: 22s\tremaining: 2.84s\n",
      "886:\tlearn: 0.2021865\ttotal: 22.1s\tremaining: 2.81s\n",
      "887:\tlearn: 0.2021605\ttotal: 22.1s\tremaining: 2.79s\n",
      "888:\tlearn: 0.2021303\ttotal: 22.1s\tremaining: 2.76s\n",
      "889:\tlearn: 0.2020943\ttotal: 22.2s\tremaining: 2.74s\n",
      "890:\tlearn: 0.2020764\ttotal: 22.2s\tremaining: 2.71s\n",
      "891:\tlearn: 0.2020504\ttotal: 22.2s\tremaining: 2.69s\n",
      "892:\tlearn: 0.2020251\ttotal: 22.2s\tremaining: 2.66s\n",
      "893:\tlearn: 0.2020144\ttotal: 22.2s\tremaining: 2.64s\n",
      "894:\tlearn: 0.2019836\ttotal: 22.3s\tremaining: 2.61s\n",
      "895:\tlearn: 0.2019731\ttotal: 22.3s\tremaining: 2.59s\n",
      "896:\tlearn: 0.2019587\ttotal: 22.3s\tremaining: 2.56s\n",
      "897:\tlearn: 0.2019321\ttotal: 22.4s\tremaining: 2.54s\n",
      "898:\tlearn: 0.2019104\ttotal: 22.4s\tremaining: 2.51s\n",
      "899:\tlearn: 0.2018757\ttotal: 22.4s\tremaining: 2.49s\n",
      "900:\tlearn: 0.2018528\ttotal: 22.4s\tremaining: 2.46s\n",
      "901:\tlearn: 0.2018278\ttotal: 22.4s\tremaining: 2.44s\n",
      "902:\tlearn: 0.2018116\ttotal: 22.5s\tremaining: 2.41s\n",
      "903:\tlearn: 0.2017902\ttotal: 22.5s\tremaining: 2.39s\n",
      "904:\tlearn: 0.2017659\ttotal: 22.5s\tremaining: 2.36s\n",
      "905:\tlearn: 0.2017436\ttotal: 22.5s\tremaining: 2.34s\n",
      "906:\tlearn: 0.2017282\ttotal: 22.6s\tremaining: 2.31s\n",
      "907:\tlearn: 0.2016993\ttotal: 22.6s\tremaining: 2.29s\n",
      "908:\tlearn: 0.2016800\ttotal: 22.6s\tremaining: 2.26s\n",
      "909:\tlearn: 0.2016760\ttotal: 22.6s\tremaining: 2.24s\n",
      "910:\tlearn: 0.2016457\ttotal: 22.7s\tremaining: 2.21s\n",
      "911:\tlearn: 0.2016256\ttotal: 22.7s\tremaining: 2.19s\n",
      "912:\tlearn: 0.2015932\ttotal: 22.7s\tremaining: 2.16s\n",
      "913:\tlearn: 0.2015776\ttotal: 22.7s\tremaining: 2.14s\n",
      "914:\tlearn: 0.2015600\ttotal: 22.8s\tremaining: 2.12s\n",
      "915:\tlearn: 0.2015400\ttotal: 22.8s\tremaining: 2.09s\n",
      "916:\tlearn: 0.2014975\ttotal: 22.8s\tremaining: 2.06s\n",
      "917:\tlearn: 0.2014701\ttotal: 22.8s\tremaining: 2.04s\n",
      "918:\tlearn: 0.2014534\ttotal: 22.9s\tremaining: 2.01s\n",
      "919:\tlearn: 0.2014305\ttotal: 22.9s\tremaining: 1.99s\n",
      "920:\tlearn: 0.2014059\ttotal: 22.9s\tremaining: 1.97s\n",
      "921:\tlearn: 0.2013661\ttotal: 22.9s\tremaining: 1.94s\n",
      "922:\tlearn: 0.2013608\ttotal: 23s\tremaining: 1.92s\n",
      "923:\tlearn: 0.2013335\ttotal: 23s\tremaining: 1.89s\n",
      "924:\tlearn: 0.2013205\ttotal: 23s\tremaining: 1.86s\n",
      "925:\tlearn: 0.2012924\ttotal: 23s\tremaining: 1.84s\n",
      "926:\tlearn: 0.2012832\ttotal: 23.1s\tremaining: 1.81s\n",
      "927:\tlearn: 0.2012578\ttotal: 23.1s\tremaining: 1.79s\n",
      "928:\tlearn: 0.2012301\ttotal: 23.1s\tremaining: 1.76s\n",
      "929:\tlearn: 0.2011980\ttotal: 23.1s\tremaining: 1.74s\n",
      "930:\tlearn: 0.2011687\ttotal: 23.1s\tremaining: 1.72s\n",
      "931:\tlearn: 0.2011499\ttotal: 23.2s\tremaining: 1.69s\n",
      "932:\tlearn: 0.2011293\ttotal: 23.2s\tremaining: 1.67s\n",
      "933:\tlearn: 0.2011107\ttotal: 23.2s\tremaining: 1.64s\n",
      "934:\tlearn: 0.2010904\ttotal: 23.2s\tremaining: 1.62s\n",
      "935:\tlearn: 0.2010643\ttotal: 23.3s\tremaining: 1.59s\n",
      "936:\tlearn: 0.2010342\ttotal: 23.3s\tremaining: 1.57s\n",
      "937:\tlearn: 0.2010152\ttotal: 23.3s\tremaining: 1.54s\n",
      "938:\tlearn: 0.2009989\ttotal: 23.3s\tremaining: 1.52s\n",
      "939:\tlearn: 0.2009798\ttotal: 23.4s\tremaining: 1.49s\n",
      "940:\tlearn: 0.2009498\ttotal: 23.4s\tremaining: 1.47s\n",
      "941:\tlearn: 0.2009296\ttotal: 23.4s\tremaining: 1.44s\n",
      "942:\tlearn: 0.2009043\ttotal: 23.5s\tremaining: 1.42s\n",
      "943:\tlearn: 0.2008619\ttotal: 23.5s\tremaining: 1.39s\n",
      "944:\tlearn: 0.2008361\ttotal: 23.5s\tremaining: 1.37s\n",
      "945:\tlearn: 0.2008047\ttotal: 23.5s\tremaining: 1.34s\n",
      "946:\tlearn: 0.2007826\ttotal: 23.5s\tremaining: 1.32s\n",
      "947:\tlearn: 0.2007632\ttotal: 23.6s\tremaining: 1.29s\n",
      "948:\tlearn: 0.2007475\ttotal: 23.6s\tremaining: 1.27s\n",
      "949:\tlearn: 0.2007178\ttotal: 23.6s\tremaining: 1.24s\n",
      "950:\tlearn: 0.2006919\ttotal: 23.6s\tremaining: 1.22s\n",
      "951:\tlearn: 0.2006488\ttotal: 23.7s\tremaining: 1.19s\n",
      "952:\tlearn: 0.2006206\ttotal: 23.7s\tremaining: 1.17s\n",
      "953:\tlearn: 0.2006062\ttotal: 23.7s\tremaining: 1.14s\n",
      "954:\tlearn: 0.2005743\ttotal: 23.7s\tremaining: 1.12s\n",
      "955:\tlearn: 0.2005447\ttotal: 23.8s\tremaining: 1.09s\n",
      "956:\tlearn: 0.2005280\ttotal: 23.8s\tremaining: 1.07s\n",
      "957:\tlearn: 0.2005125\ttotal: 23.8s\tremaining: 1.04s\n",
      "958:\tlearn: 0.2005001\ttotal: 23.8s\tremaining: 1.02s\n",
      "959:\tlearn: 0.2004835\ttotal: 23.9s\tremaining: 995ms\n",
      "960:\tlearn: 0.2004593\ttotal: 23.9s\tremaining: 970ms\n",
      "961:\tlearn: 0.2004340\ttotal: 23.9s\tremaining: 945ms\n",
      "962:\tlearn: 0.2004012\ttotal: 23.9s\tremaining: 920ms\n",
      "963:\tlearn: 0.2003651\ttotal: 24s\tremaining: 895ms\n",
      "964:\tlearn: 0.2003493\ttotal: 24s\tremaining: 870ms\n",
      "965:\tlearn: 0.2003302\ttotal: 24s\tremaining: 845ms\n",
      "966:\tlearn: 0.2003111\ttotal: 24s\tremaining: 820ms\n",
      "967:\tlearn: 0.2002891\ttotal: 24.1s\tremaining: 795ms\n",
      "968:\tlearn: 0.2002697\ttotal: 24.1s\tremaining: 771ms\n",
      "969:\tlearn: 0.2002536\ttotal: 24.1s\tremaining: 746ms\n",
      "970:\tlearn: 0.2002321\ttotal: 24.1s\tremaining: 721ms\n",
      "971:\tlearn: 0.2001980\ttotal: 24.2s\tremaining: 696ms\n",
      "972:\tlearn: 0.2001732\ttotal: 24.2s\tremaining: 671ms\n",
      "973:\tlearn: 0.2001599\ttotal: 24.2s\tremaining: 646ms\n",
      "974:\tlearn: 0.2001429\ttotal: 24.3s\tremaining: 622ms\n",
      "975:\tlearn: 0.2001151\ttotal: 24.3s\tremaining: 598ms\n",
      "976:\tlearn: 0.2000834\ttotal: 24.3s\tremaining: 573ms\n",
      "977:\tlearn: 0.2000696\ttotal: 24.4s\tremaining: 548ms\n",
      "978:\tlearn: 0.2000410\ttotal: 24.4s\tremaining: 523ms\n",
      "979:\tlearn: 0.2000207\ttotal: 24.4s\tremaining: 498ms\n",
      "980:\tlearn: 0.2000062\ttotal: 24.4s\tremaining: 473ms\n",
      "981:\tlearn: 0.1999903\ttotal: 24.5s\tremaining: 448ms\n",
      "982:\tlearn: 0.1999583\ttotal: 24.5s\tremaining: 423ms\n",
      "983:\tlearn: 0.1999389\ttotal: 24.5s\tremaining: 398ms\n",
      "984:\tlearn: 0.1999240\ttotal: 24.5s\tremaining: 374ms\n",
      "985:\tlearn: 0.1999089\ttotal: 24.6s\tremaining: 349ms\n",
      "986:\tlearn: 0.1998884\ttotal: 24.6s\tremaining: 324ms\n",
      "987:\tlearn: 0.1998548\ttotal: 24.6s\tremaining: 299ms\n",
      "988:\tlearn: 0.1998355\ttotal: 24.6s\tremaining: 274ms\n",
      "989:\tlearn: 0.1998077\ttotal: 24.7s\tremaining: 249ms\n",
      "990:\tlearn: 0.1997975\ttotal: 24.7s\tremaining: 224ms\n",
      "991:\tlearn: 0.1997739\ttotal: 24.7s\tremaining: 199ms\n",
      "992:\tlearn: 0.1997594\ttotal: 24.7s\tremaining: 174ms\n",
      "993:\tlearn: 0.1997394\ttotal: 24.7s\tremaining: 149ms\n",
      "994:\tlearn: 0.1997213\ttotal: 24.8s\tremaining: 124ms\n",
      "995:\tlearn: 0.1996979\ttotal: 24.8s\tremaining: 99.6ms\n",
      "996:\tlearn: 0.1996624\ttotal: 24.8s\tremaining: 74.7ms\n",
      "997:\tlearn: 0.1996479\ttotal: 24.9s\tremaining: 49.8ms\n",
      "998:\tlearn: 0.1996258\ttotal: 24.9s\tremaining: 24.9ms\n",
      "999:\tlearn: 0.1996080\ttotal: 24.9s\tremaining: 0us\n",
      "The accuracy score for the <catboost.core.CatBoostClassifier object at 0x0000015AB44F3EB0> unsampled set is: 0.9372013651877134\n",
      "The f1 score for the <catboost.core.CatBoostClassifier object at 0x0000015AB44F3EB0> unsampled set is: 0.0\n",
      "\n",
      "\n",
      "Learning rate set to 0.074913\n",
      "0:\tlearn: 0.6871015\ttotal: 32ms\tremaining: 32s\n",
      "1:\tlearn: 0.6818465\ttotal: 67.6ms\tremaining: 33.7s\n",
      "2:\tlearn: 0.6773593\ttotal: 105ms\tremaining: 35s\n",
      "3:\tlearn: 0.6735384\ttotal: 142ms\tremaining: 35.4s\n",
      "4:\tlearn: 0.6702980\ttotal: 178ms\tremaining: 35.3s\n",
      "5:\tlearn: 0.6674864\ttotal: 224ms\tremaining: 37.1s\n",
      "6:\tlearn: 0.6645829\ttotal: 260ms\tremaining: 36.9s\n",
      "7:\tlearn: 0.6622458\ttotal: 299ms\tremaining: 37.1s\n",
      "8:\tlearn: 0.6602320\ttotal: 336ms\tremaining: 37s\n",
      "9:\tlearn: 0.6584818\ttotal: 368ms\tremaining: 36.4s\n",
      "10:\tlearn: 0.6570333\ttotal: 406ms\tremaining: 36.5s\n",
      "11:\tlearn: 0.6557162\ttotal: 442ms\tremaining: 36.4s\n",
      "12:\tlearn: 0.6543389\ttotal: 487ms\tremaining: 36.9s\n",
      "13:\tlearn: 0.6531932\ttotal: 524ms\tremaining: 36.9s\n",
      "14:\tlearn: 0.6523433\ttotal: 565ms\tremaining: 37.1s\n",
      "15:\tlearn: 0.6512424\ttotal: 600ms\tremaining: 36.9s\n",
      "16:\tlearn: 0.6503604\ttotal: 638ms\tremaining: 36.9s\n",
      "17:\tlearn: 0.6495495\ttotal: 683ms\tremaining: 37.3s\n",
      "18:\tlearn: 0.6487249\ttotal: 721ms\tremaining: 37.2s\n",
      "19:\tlearn: 0.6481267\ttotal: 755ms\tremaining: 37s\n",
      "20:\tlearn: 0.6471208\ttotal: 788ms\tremaining: 36.7s\n",
      "21:\tlearn: 0.6466221\ttotal: 822ms\tremaining: 36.6s\n",
      "22:\tlearn: 0.6460774\ttotal: 869ms\tremaining: 36.9s\n",
      "23:\tlearn: 0.6452422\ttotal: 904ms\tremaining: 36.8s\n",
      "24:\tlearn: 0.6443489\ttotal: 945ms\tremaining: 36.8s\n",
      "25:\tlearn: 0.6438649\ttotal: 982ms\tremaining: 36.8s\n",
      "26:\tlearn: 0.6432051\ttotal: 1.02s\tremaining: 36.7s\n",
      "27:\tlearn: 0.6425165\ttotal: 1.06s\tremaining: 36.8s\n",
      "28:\tlearn: 0.6420604\ttotal: 1.11s\tremaining: 37.2s\n",
      "29:\tlearn: 0.6416733\ttotal: 1.15s\tremaining: 37.2s\n",
      "30:\tlearn: 0.6410999\ttotal: 1.19s\tremaining: 37.1s\n",
      "31:\tlearn: 0.6406184\ttotal: 1.22s\tremaining: 36.9s\n",
      "32:\tlearn: 0.6400647\ttotal: 1.26s\tremaining: 36.9s\n",
      "33:\tlearn: 0.6395686\ttotal: 1.29s\tremaining: 36.8s\n",
      "34:\tlearn: 0.6392507\ttotal: 1.33s\tremaining: 36.6s\n",
      "35:\tlearn: 0.6387995\ttotal: 1.38s\tremaining: 36.9s\n",
      "36:\tlearn: 0.6382731\ttotal: 1.41s\tremaining: 36.8s\n",
      "37:\tlearn: 0.6378182\ttotal: 1.45s\tremaining: 36.7s\n",
      "38:\tlearn: 0.6374010\ttotal: 1.49s\tremaining: 36.6s\n",
      "39:\tlearn: 0.6370084\ttotal: 1.52s\tremaining: 36.5s\n",
      "40:\tlearn: 0.6366600\ttotal: 1.56s\tremaining: 36.6s\n",
      "41:\tlearn: 0.6359670\ttotal: 1.6s\tremaining: 36.5s\n",
      "42:\tlearn: 0.6354470\ttotal: 1.63s\tremaining: 36.4s\n",
      "43:\tlearn: 0.6351687\ttotal: 1.67s\tremaining: 36.3s\n",
      "44:\tlearn: 0.6348435\ttotal: 1.71s\tremaining: 36.2s\n",
      "45:\tlearn: 0.6345368\ttotal: 1.74s\tremaining: 36s\n",
      "46:\tlearn: 0.6340933\ttotal: 1.77s\tremaining: 36s\n",
      "47:\tlearn: 0.6336894\ttotal: 1.82s\tremaining: 36.1s\n",
      "48:\tlearn: 0.6332735\ttotal: 1.86s\tremaining: 36s\n",
      "49:\tlearn: 0.6328578\ttotal: 1.89s\tremaining: 35.9s\n",
      "50:\tlearn: 0.6324434\ttotal: 1.93s\tremaining: 35.9s\n",
      "51:\tlearn: 0.6321860\ttotal: 1.96s\tremaining: 35.8s\n",
      "52:\tlearn: 0.6317729\ttotal: 2s\tremaining: 35.8s\n",
      "53:\tlearn: 0.6314453\ttotal: 2.04s\tremaining: 35.7s\n",
      "54:\tlearn: 0.6311190\ttotal: 2.07s\tremaining: 35.6s\n",
      "55:\tlearn: 0.6307752\ttotal: 2.11s\tremaining: 35.6s\n",
      "56:\tlearn: 0.6304935\ttotal: 2.15s\tremaining: 35.5s\n",
      "57:\tlearn: 0.6301579\ttotal: 2.18s\tremaining: 35.4s\n",
      "58:\tlearn: 0.6298384\ttotal: 2.22s\tremaining: 35.4s\n",
      "59:\tlearn: 0.6295572\ttotal: 2.26s\tremaining: 35.4s\n",
      "60:\tlearn: 0.6292760\ttotal: 2.29s\tremaining: 35.3s\n",
      "61:\tlearn: 0.6289363\ttotal: 2.33s\tremaining: 35.2s\n",
      "62:\tlearn: 0.6285809\ttotal: 2.37s\tremaining: 35.2s\n",
      "63:\tlearn: 0.6281512\ttotal: 2.4s\tremaining: 35.2s\n",
      "64:\tlearn: 0.6279665\ttotal: 2.48s\tremaining: 35.7s\n",
      "65:\tlearn: 0.6277529\ttotal: 2.52s\tremaining: 35.7s\n",
      "66:\tlearn: 0.6274386\ttotal: 2.55s\tremaining: 35.6s\n",
      "67:\tlearn: 0.6272130\ttotal: 2.59s\tremaining: 35.5s\n",
      "68:\tlearn: 0.6269253\ttotal: 2.63s\tremaining: 35.5s\n",
      "69:\tlearn: 0.6265921\ttotal: 2.67s\tremaining: 35.5s\n",
      "70:\tlearn: 0.6263458\ttotal: 2.7s\tremaining: 35.4s\n",
      "71:\tlearn: 0.6260433\ttotal: 2.74s\tremaining: 35.3s\n",
      "72:\tlearn: 0.6256953\ttotal: 2.77s\tremaining: 35.2s\n",
      "73:\tlearn: 0.6253883\ttotal: 2.81s\tremaining: 35.1s\n",
      "74:\tlearn: 0.6249658\ttotal: 2.85s\tremaining: 35.2s\n",
      "75:\tlearn: 0.6246468\ttotal: 2.89s\tremaining: 35.1s\n",
      "76:\tlearn: 0.6243094\ttotal: 2.93s\tremaining: 35.1s\n",
      "77:\tlearn: 0.6238601\ttotal: 2.96s\tremaining: 35s\n",
      "78:\tlearn: 0.6236600\ttotal: 3s\tremaining: 34.9s\n",
      "79:\tlearn: 0.6233524\ttotal: 3.03s\tremaining: 34.9s\n",
      "80:\tlearn: 0.6231924\ttotal: 3.07s\tremaining: 34.8s\n",
      "81:\tlearn: 0.6228542\ttotal: 3.11s\tremaining: 34.9s\n",
      "82:\tlearn: 0.6226161\ttotal: 3.15s\tremaining: 34.8s\n",
      "83:\tlearn: 0.6222939\ttotal: 3.18s\tremaining: 34.7s\n",
      "84:\tlearn: 0.6219311\ttotal: 3.22s\tremaining: 34.7s\n",
      "85:\tlearn: 0.6216408\ttotal: 3.25s\tremaining: 34.6s\n",
      "86:\tlearn: 0.6212484\ttotal: 3.29s\tremaining: 34.6s\n",
      "87:\tlearn: 0.6208242\ttotal: 3.34s\tremaining: 34.6s\n",
      "88:\tlearn: 0.6205790\ttotal: 3.37s\tremaining: 34.5s\n",
      "89:\tlearn: 0.6202331\ttotal: 3.41s\tremaining: 34.5s\n",
      "90:\tlearn: 0.6198669\ttotal: 3.44s\tremaining: 34.4s\n",
      "91:\tlearn: 0.6196146\ttotal: 3.48s\tremaining: 34.4s\n",
      "92:\tlearn: 0.6193718\ttotal: 3.52s\tremaining: 34.3s\n",
      "93:\tlearn: 0.6190632\ttotal: 3.56s\tremaining: 34.3s\n",
      "94:\tlearn: 0.6188232\ttotal: 3.59s\tremaining: 34.2s\n",
      "95:\tlearn: 0.6185553\ttotal: 3.63s\tremaining: 34.2s\n",
      "96:\tlearn: 0.6183116\ttotal: 3.67s\tremaining: 34.1s\n",
      "97:\tlearn: 0.6180497\ttotal: 3.7s\tremaining: 34.1s\n",
      "98:\tlearn: 0.6177741\ttotal: 3.73s\tremaining: 34s\n",
      "99:\tlearn: 0.6175133\ttotal: 3.78s\tremaining: 34s\n",
      "100:\tlearn: 0.6172378\ttotal: 3.82s\tremaining: 34s\n",
      "101:\tlearn: 0.6168853\ttotal: 3.85s\tremaining: 33.9s\n",
      "102:\tlearn: 0.6165827\ttotal: 3.89s\tremaining: 33.9s\n",
      "103:\tlearn: 0.6162882\ttotal: 3.93s\tremaining: 33.8s\n",
      "104:\tlearn: 0.6159974\ttotal: 3.97s\tremaining: 33.9s\n",
      "105:\tlearn: 0.6157255\ttotal: 4.01s\tremaining: 33.8s\n",
      "106:\tlearn: 0.6152250\ttotal: 4.04s\tremaining: 33.7s\n",
      "107:\tlearn: 0.6149081\ttotal: 4.08s\tremaining: 33.7s\n",
      "108:\tlearn: 0.6145563\ttotal: 4.12s\tremaining: 33.6s\n",
      "109:\tlearn: 0.6143243\ttotal: 4.16s\tremaining: 33.7s\n",
      "110:\tlearn: 0.6140303\ttotal: 4.2s\tremaining: 33.6s\n",
      "111:\tlearn: 0.6137486\ttotal: 4.23s\tremaining: 33.5s\n",
      "112:\tlearn: 0.6135906\ttotal: 4.31s\tremaining: 33.8s\n",
      "113:\tlearn: 0.6134164\ttotal: 4.38s\tremaining: 34s\n",
      "114:\tlearn: 0.6131525\ttotal: 4.47s\tremaining: 34.4s\n",
      "115:\tlearn: 0.6130119\ttotal: 4.52s\tremaining: 34.5s\n",
      "116:\tlearn: 0.6126979\ttotal: 4.58s\tremaining: 34.5s\n",
      "117:\tlearn: 0.6124193\ttotal: 4.64s\tremaining: 34.7s\n",
      "118:\tlearn: 0.6122628\ttotal: 4.7s\tremaining: 34.8s\n",
      "119:\tlearn: 0.6120166\ttotal: 4.76s\tremaining: 34.9s\n",
      "120:\tlearn: 0.6117580\ttotal: 4.8s\tremaining: 34.9s\n",
      "121:\tlearn: 0.6115318\ttotal: 4.85s\tremaining: 34.9s\n",
      "122:\tlearn: 0.6112390\ttotal: 4.91s\tremaining: 35s\n",
      "123:\tlearn: 0.6109956\ttotal: 4.95s\tremaining: 35s\n",
      "124:\tlearn: 0.6107177\ttotal: 5s\tremaining: 35s\n",
      "125:\tlearn: 0.6104523\ttotal: 5.05s\tremaining: 35.1s\n",
      "126:\tlearn: 0.6101491\ttotal: 5.11s\tremaining: 35.1s\n",
      "127:\tlearn: 0.6099205\ttotal: 5.15s\tremaining: 35.1s\n",
      "128:\tlearn: 0.6096819\ttotal: 5.2s\tremaining: 35.1s\n",
      "129:\tlearn: 0.6095196\ttotal: 5.25s\tremaining: 35.2s\n",
      "130:\tlearn: 0.6091553\ttotal: 5.32s\tremaining: 35.3s\n",
      "131:\tlearn: 0.6089067\ttotal: 5.37s\tremaining: 35.3s\n",
      "132:\tlearn: 0.6086700\ttotal: 5.41s\tremaining: 35.3s\n",
      "133:\tlearn: 0.6083854\ttotal: 5.46s\tremaining: 35.3s\n",
      "134:\tlearn: 0.6081109\ttotal: 5.51s\tremaining: 35.3s\n",
      "135:\tlearn: 0.6079345\ttotal: 5.57s\tremaining: 35.4s\n",
      "136:\tlearn: 0.6076742\ttotal: 5.65s\tremaining: 35.6s\n",
      "137:\tlearn: 0.6074369\ttotal: 5.69s\tremaining: 35.5s\n",
      "138:\tlearn: 0.6071593\ttotal: 5.74s\tremaining: 35.5s\n",
      "139:\tlearn: 0.6069199\ttotal: 5.77s\tremaining: 35.4s\n",
      "140:\tlearn: 0.6065442\ttotal: 5.81s\tremaining: 35.4s\n",
      "141:\tlearn: 0.6063051\ttotal: 5.85s\tremaining: 35.3s\n",
      "142:\tlearn: 0.6060558\ttotal: 5.88s\tremaining: 35.2s\n",
      "143:\tlearn: 0.6056920\ttotal: 5.92s\tremaining: 35.2s\n",
      "144:\tlearn: 0.6054459\ttotal: 5.96s\tremaining: 35.1s\n",
      "145:\tlearn: 0.6052023\ttotal: 5.99s\tremaining: 35.1s\n",
      "146:\tlearn: 0.6050041\ttotal: 6.03s\tremaining: 35s\n",
      "147:\tlearn: 0.6048091\ttotal: 6.07s\tremaining: 34.9s\n",
      "148:\tlearn: 0.6046050\ttotal: 6.1s\tremaining: 34.8s\n",
      "149:\tlearn: 0.6043510\ttotal: 6.14s\tremaining: 34.8s\n",
      "150:\tlearn: 0.6041222\ttotal: 6.18s\tremaining: 34.7s\n",
      "151:\tlearn: 0.6038808\ttotal: 6.21s\tremaining: 34.6s\n",
      "152:\tlearn: 0.6037176\ttotal: 6.24s\tremaining: 34.6s\n",
      "153:\tlearn: 0.6033883\ttotal: 6.29s\tremaining: 34.5s\n",
      "154:\tlearn: 0.6032531\ttotal: 6.32s\tremaining: 34.4s\n",
      "155:\tlearn: 0.6029437\ttotal: 6.36s\tremaining: 34.4s\n",
      "156:\tlearn: 0.6027467\ttotal: 6.39s\tremaining: 34.3s\n",
      "157:\tlearn: 0.6025431\ttotal: 6.43s\tremaining: 34.2s\n",
      "158:\tlearn: 0.6023020\ttotal: 6.47s\tremaining: 34.2s\n",
      "159:\tlearn: 0.6020560\ttotal: 6.51s\tremaining: 34.2s\n",
      "160:\tlearn: 0.6018571\ttotal: 6.54s\tremaining: 34.1s\n",
      "161:\tlearn: 0.6013359\ttotal: 6.58s\tremaining: 34s\n",
      "162:\tlearn: 0.6011355\ttotal: 6.61s\tremaining: 33.9s\n",
      "163:\tlearn: 0.6007285\ttotal: 6.64s\tremaining: 33.9s\n",
      "164:\tlearn: 0.6004837\ttotal: 6.68s\tremaining: 33.8s\n",
      "165:\tlearn: 0.6001363\ttotal: 6.72s\tremaining: 33.8s\n",
      "166:\tlearn: 0.5999445\ttotal: 6.76s\tremaining: 33.7s\n",
      "167:\tlearn: 0.5996547\ttotal: 6.8s\tremaining: 33.7s\n",
      "168:\tlearn: 0.5993937\ttotal: 6.83s\tremaining: 33.6s\n",
      "169:\tlearn: 0.5992663\ttotal: 6.86s\tremaining: 33.5s\n",
      "170:\tlearn: 0.5991457\ttotal: 6.91s\tremaining: 33.5s\n",
      "171:\tlearn: 0.5988026\ttotal: 6.95s\tremaining: 33.4s\n",
      "172:\tlearn: 0.5985429\ttotal: 6.99s\tremaining: 33.4s\n",
      "173:\tlearn: 0.5983001\ttotal: 7.02s\tremaining: 33.3s\n",
      "174:\tlearn: 0.5978029\ttotal: 7.05s\tremaining: 33.2s\n",
      "175:\tlearn: 0.5975233\ttotal: 7.08s\tremaining: 33.2s\n",
      "176:\tlearn: 0.5972594\ttotal: 7.12s\tremaining: 33.1s\n",
      "177:\tlearn: 0.5970588\ttotal: 7.16s\tremaining: 33.1s\n",
      "178:\tlearn: 0.5966790\ttotal: 7.2s\tremaining: 33s\n",
      "179:\tlearn: 0.5964361\ttotal: 7.24s\tremaining: 33s\n",
      "180:\tlearn: 0.5961997\ttotal: 7.27s\tremaining: 32.9s\n",
      "181:\tlearn: 0.5959043\ttotal: 7.3s\tremaining: 32.8s\n",
      "182:\tlearn: 0.5956360\ttotal: 7.34s\tremaining: 32.8s\n",
      "183:\tlearn: 0.5953868\ttotal: 7.37s\tremaining: 32.7s\n",
      "184:\tlearn: 0.5951808\ttotal: 7.42s\tremaining: 32.7s\n",
      "185:\tlearn: 0.5949027\ttotal: 7.45s\tremaining: 32.6s\n",
      "186:\tlearn: 0.5947402\ttotal: 7.48s\tremaining: 32.5s\n",
      "187:\tlearn: 0.5944162\ttotal: 7.52s\tremaining: 32.5s\n",
      "188:\tlearn: 0.5942326\ttotal: 7.55s\tremaining: 32.4s\n",
      "189:\tlearn: 0.5940388\ttotal: 7.59s\tremaining: 32.4s\n",
      "190:\tlearn: 0.5938216\ttotal: 7.63s\tremaining: 32.3s\n",
      "191:\tlearn: 0.5936457\ttotal: 7.67s\tremaining: 32.3s\n",
      "192:\tlearn: 0.5934752\ttotal: 7.7s\tremaining: 32.2s\n",
      "193:\tlearn: 0.5932657\ttotal: 7.74s\tremaining: 32.1s\n",
      "194:\tlearn: 0.5930609\ttotal: 7.77s\tremaining: 32.1s\n",
      "195:\tlearn: 0.5928005\ttotal: 7.8s\tremaining: 32s\n",
      "196:\tlearn: 0.5925647\ttotal: 7.84s\tremaining: 32s\n",
      "197:\tlearn: 0.5922585\ttotal: 7.88s\tremaining: 31.9s\n",
      "198:\tlearn: 0.5920525\ttotal: 7.92s\tremaining: 31.9s\n",
      "199:\tlearn: 0.5918685\ttotal: 7.95s\tremaining: 31.8s\n",
      "200:\tlearn: 0.5915387\ttotal: 7.98s\tremaining: 31.7s\n",
      "201:\tlearn: 0.5913412\ttotal: 8.02s\tremaining: 31.7s\n",
      "202:\tlearn: 0.5910819\ttotal: 8.06s\tremaining: 31.7s\n",
      "203:\tlearn: 0.5908182\ttotal: 8.1s\tremaining: 31.6s\n",
      "204:\tlearn: 0.5905745\ttotal: 8.14s\tremaining: 31.6s\n",
      "205:\tlearn: 0.5903757\ttotal: 8.17s\tremaining: 31.5s\n",
      "206:\tlearn: 0.5899956\ttotal: 8.21s\tremaining: 31.5s\n",
      "207:\tlearn: 0.5897980\ttotal: 8.25s\tremaining: 31.4s\n",
      "208:\tlearn: 0.5895361\ttotal: 8.29s\tremaining: 31.4s\n",
      "209:\tlearn: 0.5893104\ttotal: 8.33s\tremaining: 31.3s\n",
      "210:\tlearn: 0.5890596\ttotal: 8.37s\tremaining: 31.3s\n",
      "211:\tlearn: 0.5887630\ttotal: 8.4s\tremaining: 31.2s\n",
      "212:\tlearn: 0.5885276\ttotal: 8.44s\tremaining: 31.2s\n",
      "213:\tlearn: 0.5882688\ttotal: 8.48s\tremaining: 31.1s\n",
      "214:\tlearn: 0.5879203\ttotal: 8.51s\tremaining: 31.1s\n",
      "215:\tlearn: 0.5875949\ttotal: 8.56s\tremaining: 31.1s\n",
      "216:\tlearn: 0.5872903\ttotal: 8.59s\tremaining: 31s\n",
      "217:\tlearn: 0.5870093\ttotal: 8.63s\tremaining: 30.9s\n",
      "218:\tlearn: 0.5866816\ttotal: 8.66s\tremaining: 30.9s\n",
      "219:\tlearn: 0.5864367\ttotal: 8.7s\tremaining: 30.8s\n",
      "220:\tlearn: 0.5860980\ttotal: 8.73s\tremaining: 30.8s\n",
      "221:\tlearn: 0.5858068\ttotal: 8.81s\tremaining: 30.9s\n",
      "222:\tlearn: 0.5856083\ttotal: 8.84s\tremaining: 30.8s\n",
      "223:\tlearn: 0.5853713\ttotal: 8.88s\tremaining: 30.8s\n",
      "224:\tlearn: 0.5851342\ttotal: 8.92s\tremaining: 30.7s\n",
      "225:\tlearn: 0.5848535\ttotal: 8.97s\tremaining: 30.7s\n",
      "226:\tlearn: 0.5846103\ttotal: 9.01s\tremaining: 30.7s\n",
      "227:\tlearn: 0.5843642\ttotal: 9.04s\tremaining: 30.6s\n",
      "228:\tlearn: 0.5840804\ttotal: 9.08s\tremaining: 30.6s\n",
      "229:\tlearn: 0.5838580\ttotal: 9.12s\tremaining: 30.5s\n",
      "230:\tlearn: 0.5835834\ttotal: 9.15s\tremaining: 30.5s\n",
      "231:\tlearn: 0.5833706\ttotal: 9.18s\tremaining: 30.4s\n",
      "232:\tlearn: 0.5831684\ttotal: 9.22s\tremaining: 30.4s\n",
      "233:\tlearn: 0.5829479\ttotal: 9.26s\tremaining: 30.3s\n",
      "234:\tlearn: 0.5827342\ttotal: 9.29s\tremaining: 30.2s\n",
      "235:\tlearn: 0.5825056\ttotal: 9.33s\tremaining: 30.2s\n",
      "236:\tlearn: 0.5822863\ttotal: 9.37s\tremaining: 30.2s\n",
      "237:\tlearn: 0.5820868\ttotal: 9.4s\tremaining: 30.1s\n",
      "238:\tlearn: 0.5818731\ttotal: 9.45s\tremaining: 30.1s\n",
      "239:\tlearn: 0.5816060\ttotal: 9.48s\tremaining: 30s\n",
      "240:\tlearn: 0.5813762\ttotal: 9.52s\tremaining: 30s\n",
      "241:\tlearn: 0.5811308\ttotal: 9.55s\tremaining: 29.9s\n",
      "242:\tlearn: 0.5808446\ttotal: 9.58s\tremaining: 29.9s\n",
      "243:\tlearn: 0.5806464\ttotal: 9.62s\tremaining: 29.8s\n",
      "244:\tlearn: 0.5803769\ttotal: 9.66s\tremaining: 29.8s\n",
      "245:\tlearn: 0.5801955\ttotal: 9.7s\tremaining: 29.7s\n",
      "246:\tlearn: 0.5799620\ttotal: 9.74s\tremaining: 29.7s\n",
      "247:\tlearn: 0.5797572\ttotal: 9.78s\tremaining: 29.6s\n",
      "248:\tlearn: 0.5796164\ttotal: 9.81s\tremaining: 29.6s\n",
      "249:\tlearn: 0.5794376\ttotal: 9.85s\tremaining: 29.5s\n",
      "250:\tlearn: 0.5792604\ttotal: 9.88s\tremaining: 29.5s\n",
      "251:\tlearn: 0.5790287\ttotal: 9.92s\tremaining: 29.4s\n",
      "252:\tlearn: 0.5788610\ttotal: 9.95s\tremaining: 29.4s\n",
      "253:\tlearn: 0.5784503\ttotal: 9.99s\tremaining: 29.3s\n",
      "254:\tlearn: 0.5781879\ttotal: 10s\tremaining: 29.3s\n",
      "255:\tlearn: 0.5779917\ttotal: 10.1s\tremaining: 29.2s\n",
      "256:\tlearn: 0.5777722\ttotal: 10.1s\tremaining: 29.2s\n",
      "257:\tlearn: 0.5775435\ttotal: 10.1s\tremaining: 29.1s\n",
      "258:\tlearn: 0.5772945\ttotal: 10.2s\tremaining: 29.1s\n",
      "259:\tlearn: 0.5771521\ttotal: 10.2s\tremaining: 29s\n",
      "260:\tlearn: 0.5768713\ttotal: 10.2s\tremaining: 29s\n",
      "261:\tlearn: 0.5767350\ttotal: 10.3s\tremaining: 28.9s\n",
      "262:\tlearn: 0.5764845\ttotal: 10.3s\tremaining: 28.9s\n",
      "263:\tlearn: 0.5763154\ttotal: 10.3s\tremaining: 28.8s\n",
      "264:\tlearn: 0.5761285\ttotal: 10.4s\tremaining: 28.8s\n",
      "265:\tlearn: 0.5759661\ttotal: 10.4s\tremaining: 28.7s\n",
      "266:\tlearn: 0.5756967\ttotal: 10.4s\tremaining: 28.7s\n",
      "267:\tlearn: 0.5754887\ttotal: 10.5s\tremaining: 28.7s\n",
      "268:\tlearn: 0.5752684\ttotal: 10.5s\tremaining: 28.6s\n",
      "269:\tlearn: 0.5750004\ttotal: 10.6s\tremaining: 28.6s\n",
      "270:\tlearn: 0.5748066\ttotal: 10.6s\tremaining: 28.5s\n",
      "271:\tlearn: 0.5745528\ttotal: 10.6s\tremaining: 28.5s\n",
      "272:\tlearn: 0.5742966\ttotal: 10.7s\tremaining: 28.4s\n",
      "273:\tlearn: 0.5741362\ttotal: 10.7s\tremaining: 28.4s\n",
      "274:\tlearn: 0.5739816\ttotal: 10.7s\tremaining: 28.3s\n",
      "275:\tlearn: 0.5738051\ttotal: 10.8s\tremaining: 28.3s\n",
      "276:\tlearn: 0.5735771\ttotal: 10.8s\tremaining: 28.2s\n",
      "277:\tlearn: 0.5734184\ttotal: 10.8s\tremaining: 28.2s\n",
      "278:\tlearn: 0.5732221\ttotal: 10.9s\tremaining: 28.1s\n",
      "279:\tlearn: 0.5730816\ttotal: 10.9s\tremaining: 28.1s\n",
      "280:\tlearn: 0.5728113\ttotal: 11s\tremaining: 28s\n",
      "281:\tlearn: 0.5726238\ttotal: 11s\tremaining: 28s\n",
      "282:\tlearn: 0.5723600\ttotal: 11s\tremaining: 28s\n",
      "283:\tlearn: 0.5721651\ttotal: 11.1s\tremaining: 27.9s\n",
      "284:\tlearn: 0.5719276\ttotal: 11.1s\tremaining: 27.8s\n",
      "285:\tlearn: 0.5717037\ttotal: 11.1s\tremaining: 27.8s\n",
      "286:\tlearn: 0.5714624\ttotal: 11.2s\tremaining: 27.8s\n",
      "287:\tlearn: 0.5713653\ttotal: 11.2s\tremaining: 27.8s\n",
      "288:\tlearn: 0.5711185\ttotal: 11.3s\tremaining: 27.7s\n",
      "289:\tlearn: 0.5709374\ttotal: 11.3s\tremaining: 27.7s\n",
      "290:\tlearn: 0.5707553\ttotal: 11.3s\tremaining: 27.6s\n",
      "291:\tlearn: 0.5705222\ttotal: 11.4s\tremaining: 27.6s\n",
      "292:\tlearn: 0.5702498\ttotal: 11.4s\tremaining: 27.5s\n",
      "293:\tlearn: 0.5701155\ttotal: 11.4s\tremaining: 27.5s\n",
      "294:\tlearn: 0.5699990\ttotal: 11.5s\tremaining: 27.5s\n",
      "295:\tlearn: 0.5698106\ttotal: 11.5s\tremaining: 27.4s\n",
      "296:\tlearn: 0.5695800\ttotal: 11.6s\tremaining: 27.4s\n",
      "297:\tlearn: 0.5694017\ttotal: 11.6s\tremaining: 27.3s\n",
      "298:\tlearn: 0.5692556\ttotal: 11.6s\tremaining: 27.3s\n",
      "299:\tlearn: 0.5690281\ttotal: 11.7s\tremaining: 27.2s\n",
      "300:\tlearn: 0.5688306\ttotal: 11.7s\tremaining: 27.2s\n",
      "301:\tlearn: 0.5686364\ttotal: 11.7s\tremaining: 27.1s\n",
      "302:\tlearn: 0.5684658\ttotal: 11.8s\tremaining: 27.1s\n",
      "303:\tlearn: 0.5681728\ttotal: 11.8s\tremaining: 27.1s\n",
      "304:\tlearn: 0.5679562\ttotal: 11.9s\tremaining: 27s\n",
      "305:\tlearn: 0.5677134\ttotal: 11.9s\tremaining: 27s\n",
      "306:\tlearn: 0.5673615\ttotal: 11.9s\tremaining: 26.9s\n",
      "307:\tlearn: 0.5671789\ttotal: 12s\tremaining: 26.9s\n",
      "308:\tlearn: 0.5670118\ttotal: 12s\tremaining: 26.8s\n",
      "309:\tlearn: 0.5668358\ttotal: 12s\tremaining: 26.8s\n",
      "310:\tlearn: 0.5666590\ttotal: 12.1s\tremaining: 26.8s\n",
      "311:\tlearn: 0.5663878\ttotal: 12.1s\tremaining: 26.7s\n",
      "312:\tlearn: 0.5661442\ttotal: 12.2s\tremaining: 26.7s\n",
      "313:\tlearn: 0.5658375\ttotal: 12.2s\tremaining: 26.6s\n",
      "314:\tlearn: 0.5656243\ttotal: 12.2s\tremaining: 26.6s\n",
      "315:\tlearn: 0.5653960\ttotal: 12.3s\tremaining: 26.5s\n",
      "316:\tlearn: 0.5651876\ttotal: 12.3s\tremaining: 26.5s\n",
      "317:\tlearn: 0.5650483\ttotal: 12.3s\tremaining: 26.5s\n",
      "318:\tlearn: 0.5647089\ttotal: 12.4s\tremaining: 26.4s\n",
      "319:\tlearn: 0.5645167\ttotal: 12.4s\tremaining: 26.4s\n",
      "320:\tlearn: 0.5643327\ttotal: 12.4s\tremaining: 26.3s\n",
      "321:\tlearn: 0.5641695\ttotal: 12.5s\tremaining: 26.3s\n",
      "322:\tlearn: 0.5640323\ttotal: 12.5s\tremaining: 26.2s\n",
      "323:\tlearn: 0.5638291\ttotal: 12.5s\tremaining: 26.2s\n",
      "324:\tlearn: 0.5636699\ttotal: 12.6s\tremaining: 26.1s\n",
      "325:\tlearn: 0.5635081\ttotal: 12.6s\tremaining: 26.1s\n",
      "326:\tlearn: 0.5633332\ttotal: 12.7s\tremaining: 26.1s\n",
      "327:\tlearn: 0.5630895\ttotal: 12.7s\tremaining: 26s\n",
      "328:\tlearn: 0.5626486\ttotal: 12.8s\tremaining: 26.1s\n",
      "329:\tlearn: 0.5623516\ttotal: 12.8s\tremaining: 26s\n",
      "330:\tlearn: 0.5621255\ttotal: 12.9s\tremaining: 26s\n",
      "331:\tlearn: 0.5618953\ttotal: 12.9s\tremaining: 25.9s\n",
      "332:\tlearn: 0.5616924\ttotal: 13s\tremaining: 26s\n",
      "333:\tlearn: 0.5614669\ttotal: 13.1s\tremaining: 26.1s\n",
      "334:\tlearn: 0.5612939\ttotal: 13.1s\tremaining: 26.1s\n",
      "335:\tlearn: 0.5610364\ttotal: 13.2s\tremaining: 26s\n",
      "336:\tlearn: 0.5607586\ttotal: 13.2s\tremaining: 26s\n",
      "337:\tlearn: 0.5605210\ttotal: 13.3s\tremaining: 26s\n",
      "338:\tlearn: 0.5603020\ttotal: 13.3s\tremaining: 25.9s\n",
      "339:\tlearn: 0.5601709\ttotal: 13.3s\tremaining: 25.9s\n",
      "340:\tlearn: 0.5598967\ttotal: 13.4s\tremaining: 25.9s\n",
      "341:\tlearn: 0.5598122\ttotal: 13.4s\tremaining: 25.8s\n",
      "342:\tlearn: 0.5596353\ttotal: 13.4s\tremaining: 25.8s\n",
      "343:\tlearn: 0.5594433\ttotal: 13.5s\tremaining: 25.7s\n",
      "344:\tlearn: 0.5592773\ttotal: 13.5s\tremaining: 25.7s\n",
      "345:\tlearn: 0.5590598\ttotal: 13.6s\tremaining: 25.6s\n",
      "346:\tlearn: 0.5588715\ttotal: 13.6s\tremaining: 25.6s\n",
      "347:\tlearn: 0.5586964\ttotal: 13.6s\tremaining: 25.5s\n",
      "348:\tlearn: 0.5585790\ttotal: 13.7s\tremaining: 25.5s\n",
      "349:\tlearn: 0.5583657\ttotal: 13.7s\tremaining: 25.5s\n",
      "350:\tlearn: 0.5582570\ttotal: 13.7s\tremaining: 25.4s\n",
      "351:\tlearn: 0.5580988\ttotal: 13.8s\tremaining: 25.4s\n",
      "352:\tlearn: 0.5578810\ttotal: 13.8s\tremaining: 25.3s\n",
      "353:\tlearn: 0.5577182\ttotal: 13.9s\tremaining: 25.3s\n",
      "354:\tlearn: 0.5575567\ttotal: 13.9s\tremaining: 25.3s\n",
      "355:\tlearn: 0.5573667\ttotal: 14s\tremaining: 25.3s\n",
      "356:\tlearn: 0.5571974\ttotal: 14s\tremaining: 25.2s\n",
      "357:\tlearn: 0.5569990\ttotal: 14s\tremaining: 25.2s\n",
      "358:\tlearn: 0.5567157\ttotal: 14.1s\tremaining: 25.2s\n",
      "359:\tlearn: 0.5565631\ttotal: 14.1s\tremaining: 25.1s\n",
      "360:\tlearn: 0.5563342\ttotal: 14.2s\tremaining: 25.1s\n",
      "361:\tlearn: 0.5561064\ttotal: 14.2s\tremaining: 25.1s\n",
      "362:\tlearn: 0.5558738\ttotal: 14.3s\tremaining: 25s\n",
      "363:\tlearn: 0.5557122\ttotal: 14.3s\tremaining: 25s\n",
      "364:\tlearn: 0.5555969\ttotal: 14.3s\tremaining: 24.9s\n",
      "365:\tlearn: 0.5553396\ttotal: 14.4s\tremaining: 24.9s\n",
      "366:\tlearn: 0.5551722\ttotal: 14.4s\tremaining: 24.8s\n",
      "367:\tlearn: 0.5550610\ttotal: 14.4s\tremaining: 24.8s\n",
      "368:\tlearn: 0.5549127\ttotal: 14.5s\tremaining: 24.8s\n",
      "369:\tlearn: 0.5547452\ttotal: 14.5s\tremaining: 24.7s\n",
      "370:\tlearn: 0.5545052\ttotal: 14.6s\tremaining: 24.7s\n",
      "371:\tlearn: 0.5543854\ttotal: 14.6s\tremaining: 24.6s\n",
      "372:\tlearn: 0.5540838\ttotal: 14.6s\tremaining: 24.6s\n",
      "373:\tlearn: 0.5539310\ttotal: 14.7s\tremaining: 24.6s\n",
      "374:\tlearn: 0.5537189\ttotal: 14.7s\tremaining: 24.5s\n",
      "375:\tlearn: 0.5534908\ttotal: 14.8s\tremaining: 24.5s\n",
      "376:\tlearn: 0.5533448\ttotal: 14.8s\tremaining: 24.5s\n",
      "377:\tlearn: 0.5531509\ttotal: 14.8s\tremaining: 24.4s\n",
      "378:\tlearn: 0.5529425\ttotal: 14.9s\tremaining: 24.4s\n",
      "379:\tlearn: 0.5527529\ttotal: 14.9s\tremaining: 24.3s\n",
      "380:\tlearn: 0.5526259\ttotal: 14.9s\tremaining: 24.3s\n",
      "381:\tlearn: 0.5524586\ttotal: 15s\tremaining: 24.2s\n",
      "382:\tlearn: 0.5523109\ttotal: 15s\tremaining: 24.2s\n",
      "383:\tlearn: 0.5520623\ttotal: 15s\tremaining: 24.1s\n",
      "384:\tlearn: 0.5518867\ttotal: 15.1s\tremaining: 24.1s\n",
      "385:\tlearn: 0.5517835\ttotal: 15.1s\tremaining: 24.1s\n",
      "386:\tlearn: 0.5516713\ttotal: 15.2s\tremaining: 24s\n",
      "387:\tlearn: 0.5515821\ttotal: 15.2s\tremaining: 24s\n",
      "388:\tlearn: 0.5513564\ttotal: 15.2s\tremaining: 23.9s\n",
      "389:\tlearn: 0.5511996\ttotal: 15.3s\tremaining: 23.9s\n",
      "390:\tlearn: 0.5510138\ttotal: 15.3s\tremaining: 23.8s\n",
      "391:\tlearn: 0.5508490\ttotal: 15.3s\tremaining: 23.8s\n",
      "392:\tlearn: 0.5506879\ttotal: 15.4s\tremaining: 23.7s\n",
      "393:\tlearn: 0.5504146\ttotal: 15.4s\tremaining: 23.7s\n",
      "394:\tlearn: 0.5502535\ttotal: 15.5s\tremaining: 23.7s\n",
      "395:\tlearn: 0.5500853\ttotal: 15.5s\tremaining: 23.6s\n",
      "396:\tlearn: 0.5497702\ttotal: 15.5s\tremaining: 23.6s\n",
      "397:\tlearn: 0.5496928\ttotal: 15.6s\tremaining: 23.5s\n",
      "398:\tlearn: 0.5494799\ttotal: 15.6s\tremaining: 23.5s\n",
      "399:\tlearn: 0.5492530\ttotal: 15.6s\tremaining: 23.4s\n",
      "400:\tlearn: 0.5490738\ttotal: 15.7s\tremaining: 23.5s\n",
      "401:\tlearn: 0.5489514\ttotal: 15.7s\tremaining: 23.4s\n",
      "402:\tlearn: 0.5488179\ttotal: 15.8s\tremaining: 23.4s\n",
      "403:\tlearn: 0.5486758\ttotal: 15.8s\tremaining: 23.3s\n",
      "404:\tlearn: 0.5484934\ttotal: 15.8s\tremaining: 23.3s\n",
      "405:\tlearn: 0.5482334\ttotal: 15.9s\tremaining: 23.2s\n",
      "406:\tlearn: 0.5480847\ttotal: 15.9s\tremaining: 23.2s\n",
      "407:\tlearn: 0.5478595\ttotal: 16s\tremaining: 23.2s\n",
      "408:\tlearn: 0.5477232\ttotal: 16s\tremaining: 23.1s\n",
      "409:\tlearn: 0.5475388\ttotal: 16s\tremaining: 23.1s\n",
      "410:\tlearn: 0.5474077\ttotal: 16.1s\tremaining: 23s\n",
      "411:\tlearn: 0.5472225\ttotal: 16.1s\tremaining: 23s\n",
      "412:\tlearn: 0.5470229\ttotal: 16.2s\tremaining: 23s\n",
      "413:\tlearn: 0.5468342\ttotal: 16.2s\tremaining: 22.9s\n",
      "414:\tlearn: 0.5467015\ttotal: 16.2s\tremaining: 22.9s\n",
      "415:\tlearn: 0.5465780\ttotal: 16.3s\tremaining: 22.8s\n",
      "416:\tlearn: 0.5464127\ttotal: 16.3s\tremaining: 22.8s\n",
      "417:\tlearn: 0.5462638\ttotal: 16.3s\tremaining: 22.7s\n",
      "418:\tlearn: 0.5461015\ttotal: 16.4s\tremaining: 22.7s\n",
      "419:\tlearn: 0.5458936\ttotal: 16.4s\tremaining: 22.6s\n",
      "420:\tlearn: 0.5457054\ttotal: 16.4s\tremaining: 22.6s\n",
      "421:\tlearn: 0.5455451\ttotal: 16.5s\tremaining: 22.6s\n",
      "422:\tlearn: 0.5453738\ttotal: 16.5s\tremaining: 22.5s\n",
      "423:\tlearn: 0.5452443\ttotal: 16.5s\tremaining: 22.5s\n",
      "424:\tlearn: 0.5450160\ttotal: 16.6s\tremaining: 22.4s\n",
      "425:\tlearn: 0.5448853\ttotal: 16.6s\tremaining: 22.4s\n",
      "426:\tlearn: 0.5446359\ttotal: 16.7s\tremaining: 22.3s\n",
      "427:\tlearn: 0.5444003\ttotal: 16.7s\tremaining: 22.3s\n",
      "428:\tlearn: 0.5442018\ttotal: 16.7s\tremaining: 22.3s\n",
      "429:\tlearn: 0.5439937\ttotal: 16.8s\tremaining: 22.2s\n",
      "430:\tlearn: 0.5438125\ttotal: 16.8s\tremaining: 22.2s\n",
      "431:\tlearn: 0.5436162\ttotal: 16.8s\tremaining: 22.1s\n",
      "432:\tlearn: 0.5433861\ttotal: 16.9s\tremaining: 22.1s\n",
      "433:\tlearn: 0.5431786\ttotal: 16.9s\tremaining: 22.1s\n",
      "434:\tlearn: 0.5430375\ttotal: 17s\tremaining: 22s\n",
      "435:\tlearn: 0.5428000\ttotal: 17s\tremaining: 22s\n",
      "436:\tlearn: 0.5426355\ttotal: 17s\tremaining: 21.9s\n",
      "437:\tlearn: 0.5424918\ttotal: 17.1s\tremaining: 21.9s\n",
      "438:\tlearn: 0.5422908\ttotal: 17.1s\tremaining: 21.9s\n",
      "439:\tlearn: 0.5421207\ttotal: 17.1s\tremaining: 21.8s\n",
      "440:\tlearn: 0.5419579\ttotal: 17.2s\tremaining: 21.8s\n",
      "441:\tlearn: 0.5417708\ttotal: 17.2s\tremaining: 21.7s\n",
      "442:\tlearn: 0.5415822\ttotal: 17.3s\tremaining: 21.7s\n",
      "443:\tlearn: 0.5414067\ttotal: 17.3s\tremaining: 21.6s\n",
      "444:\tlearn: 0.5412608\ttotal: 17.3s\tremaining: 21.6s\n",
      "445:\tlearn: 0.5411261\ttotal: 17.4s\tremaining: 21.6s\n",
      "446:\tlearn: 0.5410053\ttotal: 17.4s\tremaining: 21.5s\n",
      "447:\tlearn: 0.5408242\ttotal: 17.4s\tremaining: 21.5s\n",
      "448:\tlearn: 0.5406977\ttotal: 17.5s\tremaining: 21.4s\n",
      "449:\tlearn: 0.5405122\ttotal: 17.5s\tremaining: 21.4s\n",
      "450:\tlearn: 0.5403985\ttotal: 17.5s\tremaining: 21.4s\n",
      "451:\tlearn: 0.5402587\ttotal: 17.6s\tremaining: 21.3s\n",
      "452:\tlearn: 0.5401418\ttotal: 17.6s\tremaining: 21.3s\n",
      "453:\tlearn: 0.5399950\ttotal: 17.7s\tremaining: 21.2s\n",
      "454:\tlearn: 0.5398226\ttotal: 17.7s\tremaining: 21.2s\n",
      "455:\tlearn: 0.5396096\ttotal: 17.7s\tremaining: 21.1s\n",
      "456:\tlearn: 0.5394774\ttotal: 17.8s\tremaining: 21.1s\n",
      "457:\tlearn: 0.5393033\ttotal: 17.8s\tremaining: 21.1s\n",
      "458:\tlearn: 0.5391704\ttotal: 17.8s\tremaining: 21s\n",
      "459:\tlearn: 0.5388551\ttotal: 17.9s\tremaining: 21s\n",
      "460:\tlearn: 0.5386812\ttotal: 17.9s\tremaining: 20.9s\n",
      "461:\tlearn: 0.5384896\ttotal: 17.9s\tremaining: 20.9s\n",
      "462:\tlearn: 0.5383211\ttotal: 18s\tremaining: 20.9s\n",
      "463:\tlearn: 0.5381780\ttotal: 18s\tremaining: 20.8s\n",
      "464:\tlearn: 0.5378756\ttotal: 18.1s\tremaining: 20.8s\n",
      "465:\tlearn: 0.5376604\ttotal: 18.1s\tremaining: 20.7s\n",
      "466:\tlearn: 0.5375356\ttotal: 18.1s\tremaining: 20.7s\n",
      "467:\tlearn: 0.5374232\ttotal: 18.2s\tremaining: 20.6s\n",
      "468:\tlearn: 0.5372144\ttotal: 18.2s\tremaining: 20.6s\n",
      "469:\tlearn: 0.5370375\ttotal: 18.2s\tremaining: 20.6s\n",
      "470:\tlearn: 0.5369080\ttotal: 18.3s\tremaining: 20.5s\n",
      "471:\tlearn: 0.5367736\ttotal: 18.3s\tremaining: 20.5s\n",
      "472:\tlearn: 0.5366151\ttotal: 18.4s\tremaining: 20.4s\n",
      "473:\tlearn: 0.5365050\ttotal: 18.4s\tremaining: 20.4s\n",
      "474:\tlearn: 0.5362773\ttotal: 18.4s\tremaining: 20.4s\n",
      "475:\tlearn: 0.5360021\ttotal: 18.5s\tremaining: 20.3s\n",
      "476:\tlearn: 0.5358026\ttotal: 18.5s\tremaining: 20.3s\n",
      "477:\tlearn: 0.5356585\ttotal: 18.6s\tremaining: 20.3s\n",
      "478:\tlearn: 0.5354923\ttotal: 18.6s\tremaining: 20.2s\n",
      "479:\tlearn: 0.5353392\ttotal: 18.6s\tremaining: 20.2s\n",
      "480:\tlearn: 0.5351831\ttotal: 18.7s\tremaining: 20.1s\n",
      "481:\tlearn: 0.5350577\ttotal: 18.7s\tremaining: 20.1s\n",
      "482:\tlearn: 0.5348145\ttotal: 18.7s\tremaining: 20.1s\n",
      "483:\tlearn: 0.5347150\ttotal: 18.8s\tremaining: 20s\n",
      "484:\tlearn: 0.5345675\ttotal: 18.8s\tremaining: 20s\n",
      "485:\tlearn: 0.5344369\ttotal: 18.9s\tremaining: 20s\n",
      "486:\tlearn: 0.5343095\ttotal: 18.9s\tremaining: 19.9s\n",
      "487:\tlearn: 0.5341852\ttotal: 18.9s\tremaining: 19.9s\n",
      "488:\tlearn: 0.5340285\ttotal: 19s\tremaining: 19.8s\n",
      "489:\tlearn: 0.5338918\ttotal: 19s\tremaining: 19.8s\n",
      "490:\tlearn: 0.5336794\ttotal: 19s\tremaining: 19.7s\n",
      "491:\tlearn: 0.5335756\ttotal: 19.1s\tremaining: 19.7s\n",
      "492:\tlearn: 0.5334597\ttotal: 19.1s\tremaining: 19.7s\n",
      "493:\tlearn: 0.5332806\ttotal: 19.2s\tremaining: 19.6s\n",
      "494:\tlearn: 0.5331442\ttotal: 19.2s\tremaining: 19.6s\n",
      "495:\tlearn: 0.5330693\ttotal: 19.2s\tremaining: 19.5s\n",
      "496:\tlearn: 0.5329162\ttotal: 19.3s\tremaining: 19.5s\n",
      "497:\tlearn: 0.5327768\ttotal: 19.3s\tremaining: 19.5s\n",
      "498:\tlearn: 0.5326704\ttotal: 19.3s\tremaining: 19.4s\n",
      "499:\tlearn: 0.5325175\ttotal: 19.4s\tremaining: 19.4s\n",
      "500:\tlearn: 0.5323599\ttotal: 19.4s\tremaining: 19.3s\n",
      "501:\tlearn: 0.5320900\ttotal: 19.5s\tremaining: 19.3s\n",
      "502:\tlearn: 0.5318494\ttotal: 19.5s\tremaining: 19.3s\n",
      "503:\tlearn: 0.5317499\ttotal: 19.5s\tremaining: 19.2s\n",
      "504:\tlearn: 0.5315170\ttotal: 19.6s\tremaining: 19.2s\n",
      "505:\tlearn: 0.5313683\ttotal: 19.6s\tremaining: 19.1s\n",
      "506:\tlearn: 0.5312356\ttotal: 19.6s\tremaining: 19.1s\n",
      "507:\tlearn: 0.5309895\ttotal: 19.7s\tremaining: 19.1s\n",
      "508:\tlearn: 0.5308106\ttotal: 19.7s\tremaining: 19s\n",
      "509:\tlearn: 0.5306228\ttotal: 19.7s\tremaining: 19s\n",
      "510:\tlearn: 0.5305606\ttotal: 19.8s\tremaining: 18.9s\n",
      "511:\tlearn: 0.5303501\ttotal: 19.8s\tremaining: 18.9s\n",
      "512:\tlearn: 0.5301226\ttotal: 19.9s\tremaining: 18.8s\n",
      "513:\tlearn: 0.5298896\ttotal: 19.9s\tremaining: 18.8s\n",
      "514:\tlearn: 0.5297066\ttotal: 19.9s\tremaining: 18.8s\n",
      "515:\tlearn: 0.5295675\ttotal: 20s\tremaining: 18.7s\n",
      "516:\tlearn: 0.5293480\ttotal: 20s\tremaining: 18.7s\n",
      "517:\tlearn: 0.5291332\ttotal: 20s\tremaining: 18.7s\n",
      "518:\tlearn: 0.5289801\ttotal: 20.1s\tremaining: 18.6s\n",
      "519:\tlearn: 0.5288530\ttotal: 20.1s\tremaining: 18.6s\n",
      "520:\tlearn: 0.5285676\ttotal: 20.2s\tremaining: 18.5s\n",
      "521:\tlearn: 0.5284175\ttotal: 20.2s\tremaining: 18.5s\n",
      "522:\tlearn: 0.5282849\ttotal: 20.2s\tremaining: 18.5s\n",
      "523:\tlearn: 0.5281610\ttotal: 20.3s\tremaining: 18.4s\n",
      "524:\tlearn: 0.5280130\ttotal: 20.3s\tremaining: 18.4s\n",
      "525:\tlearn: 0.5277786\ttotal: 20.3s\tremaining: 18.3s\n",
      "526:\tlearn: 0.5276068\ttotal: 20.4s\tremaining: 18.3s\n",
      "527:\tlearn: 0.5274663\ttotal: 20.4s\tremaining: 18.2s\n",
      "528:\tlearn: 0.5273544\ttotal: 20.5s\tremaining: 18.2s\n",
      "529:\tlearn: 0.5272624\ttotal: 20.5s\tremaining: 18.2s\n",
      "530:\tlearn: 0.5271067\ttotal: 20.5s\tremaining: 18.1s\n",
      "531:\tlearn: 0.5269011\ttotal: 20.6s\tremaining: 18.1s\n",
      "532:\tlearn: 0.5267671\ttotal: 20.6s\tremaining: 18s\n",
      "533:\tlearn: 0.5265383\ttotal: 20.6s\tremaining: 18s\n",
      "534:\tlearn: 0.5264465\ttotal: 20.7s\tremaining: 18s\n",
      "535:\tlearn: 0.5262646\ttotal: 20.7s\tremaining: 17.9s\n",
      "536:\tlearn: 0.5261096\ttotal: 20.7s\tremaining: 17.9s\n",
      "537:\tlearn: 0.5259252\ttotal: 20.8s\tremaining: 17.8s\n",
      "538:\tlearn: 0.5257158\ttotal: 20.8s\tremaining: 17.8s\n",
      "539:\tlearn: 0.5255531\ttotal: 20.9s\tremaining: 17.8s\n",
      "540:\tlearn: 0.5253463\ttotal: 20.9s\tremaining: 17.7s\n",
      "541:\tlearn: 0.5251892\ttotal: 20.9s\tremaining: 17.7s\n",
      "542:\tlearn: 0.5250725\ttotal: 21s\tremaining: 17.6s\n",
      "543:\tlearn: 0.5249484\ttotal: 21s\tremaining: 17.6s\n",
      "544:\tlearn: 0.5248039\ttotal: 21s\tremaining: 17.6s\n",
      "545:\tlearn: 0.5246751\ttotal: 21.1s\tremaining: 17.5s\n",
      "546:\tlearn: 0.5245158\ttotal: 21.1s\tremaining: 17.5s\n",
      "547:\tlearn: 0.5243510\ttotal: 21.1s\tremaining: 17.4s\n",
      "548:\tlearn: 0.5242213\ttotal: 21.2s\tremaining: 17.4s\n",
      "549:\tlearn: 0.5239047\ttotal: 21.2s\tremaining: 17.4s\n",
      "550:\tlearn: 0.5237408\ttotal: 21.3s\tremaining: 17.3s\n",
      "551:\tlearn: 0.5235681\ttotal: 21.3s\tremaining: 17.3s\n",
      "552:\tlearn: 0.5232036\ttotal: 21.4s\tremaining: 17.3s\n",
      "553:\tlearn: 0.5230569\ttotal: 21.4s\tremaining: 17.2s\n",
      "554:\tlearn: 0.5229700\ttotal: 21.4s\tremaining: 17.2s\n",
      "555:\tlearn: 0.5226897\ttotal: 21.5s\tremaining: 17.2s\n",
      "556:\tlearn: 0.5225277\ttotal: 21.5s\tremaining: 17.1s\n",
      "557:\tlearn: 0.5223776\ttotal: 21.6s\tremaining: 17.1s\n",
      "558:\tlearn: 0.5222470\ttotal: 21.6s\tremaining: 17s\n",
      "559:\tlearn: 0.5221052\ttotal: 21.6s\tremaining: 17s\n",
      "560:\tlearn: 0.5219649\ttotal: 21.7s\tremaining: 16.9s\n",
      "561:\tlearn: 0.5218501\ttotal: 21.7s\tremaining: 16.9s\n",
      "562:\tlearn: 0.5215849\ttotal: 21.7s\tremaining: 16.9s\n",
      "563:\tlearn: 0.5212949\ttotal: 21.8s\tremaining: 16.8s\n",
      "564:\tlearn: 0.5211091\ttotal: 21.8s\tremaining: 16.8s\n",
      "565:\tlearn: 0.5209247\ttotal: 21.9s\tremaining: 16.8s\n",
      "566:\tlearn: 0.5207274\ttotal: 21.9s\tremaining: 16.7s\n",
      "567:\tlearn: 0.5205559\ttotal: 21.9s\tremaining: 16.7s\n",
      "568:\tlearn: 0.5204462\ttotal: 22s\tremaining: 16.6s\n",
      "569:\tlearn: 0.5202890\ttotal: 22s\tremaining: 16.6s\n",
      "570:\tlearn: 0.5201346\ttotal: 22s\tremaining: 16.6s\n",
      "571:\tlearn: 0.5200073\ttotal: 22.1s\tremaining: 16.5s\n",
      "572:\tlearn: 0.5198614\ttotal: 22.1s\tremaining: 16.5s\n",
      "573:\tlearn: 0.5196223\ttotal: 22.1s\tremaining: 16.4s\n",
      "574:\tlearn: 0.5194324\ttotal: 22.2s\tremaining: 16.4s\n",
      "575:\tlearn: 0.5192823\ttotal: 22.2s\tremaining: 16.4s\n",
      "576:\tlearn: 0.5191128\ttotal: 22.3s\tremaining: 16.3s\n",
      "577:\tlearn: 0.5190025\ttotal: 22.3s\tremaining: 16.3s\n",
      "578:\tlearn: 0.5188788\ttotal: 22.3s\tremaining: 16.2s\n",
      "579:\tlearn: 0.5187882\ttotal: 22.4s\tremaining: 16.2s\n",
      "580:\tlearn: 0.5186462\ttotal: 22.4s\tremaining: 16.2s\n",
      "581:\tlearn: 0.5184222\ttotal: 22.5s\tremaining: 16.1s\n",
      "582:\tlearn: 0.5183060\ttotal: 22.5s\tremaining: 16.1s\n",
      "583:\tlearn: 0.5181970\ttotal: 22.5s\tremaining: 16.1s\n",
      "584:\tlearn: 0.5179524\ttotal: 22.6s\tremaining: 16s\n",
      "585:\tlearn: 0.5178085\ttotal: 22.6s\tremaining: 16s\n",
      "586:\tlearn: 0.5176925\ttotal: 22.7s\tremaining: 15.9s\n",
      "587:\tlearn: 0.5175132\ttotal: 22.7s\tremaining: 15.9s\n",
      "588:\tlearn: 0.5173611\ttotal: 22.7s\tremaining: 15.9s\n",
      "589:\tlearn: 0.5172570\ttotal: 22.8s\tremaining: 15.8s\n",
      "590:\tlearn: 0.5171361\ttotal: 22.8s\tremaining: 15.8s\n",
      "591:\tlearn: 0.5170358\ttotal: 22.9s\tremaining: 15.8s\n",
      "592:\tlearn: 0.5168770\ttotal: 22.9s\tremaining: 15.7s\n",
      "593:\tlearn: 0.5167267\ttotal: 22.9s\tremaining: 15.7s\n",
      "594:\tlearn: 0.5166277\ttotal: 23s\tremaining: 15.6s\n",
      "595:\tlearn: 0.5164683\ttotal: 23s\tremaining: 15.6s\n",
      "596:\tlearn: 0.5163251\ttotal: 23s\tremaining: 15.6s\n",
      "597:\tlearn: 0.5161427\ttotal: 23.1s\tremaining: 15.5s\n",
      "598:\tlearn: 0.5160190\ttotal: 23.1s\tremaining: 15.5s\n",
      "599:\tlearn: 0.5157478\ttotal: 23.2s\tremaining: 15.4s\n",
      "600:\tlearn: 0.5155236\ttotal: 23.2s\tremaining: 15.4s\n",
      "601:\tlearn: 0.5154046\ttotal: 23.2s\tremaining: 15.4s\n",
      "602:\tlearn: 0.5152925\ttotal: 23.3s\tremaining: 15.3s\n",
      "603:\tlearn: 0.5151558\ttotal: 23.3s\tremaining: 15.3s\n",
      "604:\tlearn: 0.5150458\ttotal: 23.3s\tremaining: 15.2s\n",
      "605:\tlearn: 0.5149266\ttotal: 23.4s\tremaining: 15.2s\n",
      "606:\tlearn: 0.5147529\ttotal: 23.4s\tremaining: 15.2s\n",
      "607:\tlearn: 0.5145560\ttotal: 23.4s\tremaining: 15.1s\n",
      "608:\tlearn: 0.5144449\ttotal: 23.5s\tremaining: 15.1s\n",
      "609:\tlearn: 0.5143378\ttotal: 23.5s\tremaining: 15s\n",
      "610:\tlearn: 0.5141725\ttotal: 23.6s\tremaining: 15s\n",
      "611:\tlearn: 0.5140577\ttotal: 23.6s\tremaining: 15s\n",
      "612:\tlearn: 0.5139155\ttotal: 23.6s\tremaining: 14.9s\n",
      "613:\tlearn: 0.5137604\ttotal: 23.7s\tremaining: 14.9s\n",
      "614:\tlearn: 0.5135943\ttotal: 23.7s\tremaining: 14.8s\n",
      "615:\tlearn: 0.5135053\ttotal: 23.7s\tremaining: 14.8s\n",
      "616:\tlearn: 0.5133816\ttotal: 23.8s\tremaining: 14.8s\n",
      "617:\tlearn: 0.5132861\ttotal: 23.8s\tremaining: 14.7s\n",
      "618:\tlearn: 0.5131831\ttotal: 23.9s\tremaining: 14.7s\n",
      "619:\tlearn: 0.5129674\ttotal: 23.9s\tremaining: 14.6s\n",
      "620:\tlearn: 0.5128455\ttotal: 24s\tremaining: 14.6s\n",
      "621:\tlearn: 0.5127478\ttotal: 24s\tremaining: 14.6s\n",
      "622:\tlearn: 0.5125605\ttotal: 24s\tremaining: 14.5s\n",
      "623:\tlearn: 0.5123808\ttotal: 24.1s\tremaining: 14.5s\n",
      "624:\tlearn: 0.5121946\ttotal: 24.1s\tremaining: 14.5s\n",
      "625:\tlearn: 0.5120875\ttotal: 24.2s\tremaining: 14.4s\n",
      "626:\tlearn: 0.5119414\ttotal: 24.2s\tremaining: 14.4s\n",
      "627:\tlearn: 0.5117608\ttotal: 24.2s\tremaining: 14.4s\n",
      "628:\tlearn: 0.5115655\ttotal: 24.3s\tremaining: 14.3s\n",
      "629:\tlearn: 0.5114120\ttotal: 24.3s\tremaining: 14.3s\n",
      "630:\tlearn: 0.5112232\ttotal: 24.4s\tremaining: 14.2s\n",
      "631:\tlearn: 0.5110015\ttotal: 24.4s\tremaining: 14.2s\n",
      "632:\tlearn: 0.5108261\ttotal: 24.4s\tremaining: 14.2s\n",
      "633:\tlearn: 0.5106850\ttotal: 24.5s\tremaining: 14.1s\n",
      "634:\tlearn: 0.5105379\ttotal: 24.5s\tremaining: 14.1s\n",
      "635:\tlearn: 0.5104314\ttotal: 24.6s\tremaining: 14.1s\n",
      "636:\tlearn: 0.5102590\ttotal: 24.6s\tremaining: 14s\n",
      "637:\tlearn: 0.5100628\ttotal: 24.7s\tremaining: 14s\n",
      "638:\tlearn: 0.5099192\ttotal: 24.7s\tremaining: 14s\n",
      "639:\tlearn: 0.5097894\ttotal: 24.7s\tremaining: 13.9s\n",
      "640:\tlearn: 0.5095939\ttotal: 24.8s\tremaining: 13.9s\n",
      "641:\tlearn: 0.5093662\ttotal: 24.8s\tremaining: 13.8s\n",
      "642:\tlearn: 0.5091854\ttotal: 24.9s\tremaining: 13.8s\n",
      "643:\tlearn: 0.5088707\ttotal: 24.9s\tremaining: 13.8s\n",
      "644:\tlearn: 0.5087839\ttotal: 24.9s\tremaining: 13.7s\n",
      "645:\tlearn: 0.5085516\ttotal: 25s\tremaining: 13.7s\n",
      "646:\tlearn: 0.5084822\ttotal: 25s\tremaining: 13.6s\n",
      "647:\tlearn: 0.5083534\ttotal: 25s\tremaining: 13.6s\n",
      "648:\tlearn: 0.5082031\ttotal: 25.1s\tremaining: 13.6s\n",
      "649:\tlearn: 0.5079823\ttotal: 25.1s\tremaining: 13.5s\n",
      "650:\tlearn: 0.5078220\ttotal: 25.2s\tremaining: 13.5s\n",
      "651:\tlearn: 0.5077007\ttotal: 25.2s\tremaining: 13.4s\n",
      "652:\tlearn: 0.5075763\ttotal: 25.2s\tremaining: 13.4s\n",
      "653:\tlearn: 0.5073997\ttotal: 25.3s\tremaining: 13.4s\n",
      "654:\tlearn: 0.5073066\ttotal: 25.3s\tremaining: 13.3s\n",
      "655:\tlearn: 0.5072043\ttotal: 25.4s\tremaining: 13.3s\n",
      "656:\tlearn: 0.5070891\ttotal: 25.4s\tremaining: 13.3s\n",
      "657:\tlearn: 0.5068268\ttotal: 25.4s\tremaining: 13.2s\n",
      "658:\tlearn: 0.5066021\ttotal: 25.5s\tremaining: 13.2s\n",
      "659:\tlearn: 0.5064630\ttotal: 25.5s\tremaining: 13.1s\n",
      "660:\tlearn: 0.5063843\ttotal: 25.5s\tremaining: 13.1s\n",
      "661:\tlearn: 0.5062002\ttotal: 25.6s\tremaining: 13.1s\n",
      "662:\tlearn: 0.5060272\ttotal: 25.6s\tremaining: 13s\n",
      "663:\tlearn: 0.5058740\ttotal: 25.6s\tremaining: 13s\n",
      "664:\tlearn: 0.5057343\ttotal: 25.7s\tremaining: 12.9s\n",
      "665:\tlearn: 0.5055610\ttotal: 25.7s\tremaining: 12.9s\n",
      "666:\tlearn: 0.5053879\ttotal: 25.8s\tremaining: 12.9s\n",
      "667:\tlearn: 0.5052923\ttotal: 25.8s\tremaining: 12.8s\n",
      "668:\tlearn: 0.5050941\ttotal: 25.8s\tremaining: 12.8s\n",
      "669:\tlearn: 0.5047823\ttotal: 25.9s\tremaining: 12.7s\n",
      "670:\tlearn: 0.5046100\ttotal: 25.9s\tremaining: 12.7s\n",
      "671:\tlearn: 0.5043891\ttotal: 25.9s\tremaining: 12.7s\n",
      "672:\tlearn: 0.5042855\ttotal: 26s\tremaining: 12.6s\n",
      "673:\tlearn: 0.5041305\ttotal: 26s\tremaining: 12.6s\n",
      "674:\tlearn: 0.5040146\ttotal: 26s\tremaining: 12.5s\n",
      "675:\tlearn: 0.5038978\ttotal: 26.1s\tremaining: 12.5s\n",
      "676:\tlearn: 0.5036961\ttotal: 26.1s\tremaining: 12.5s\n",
      "677:\tlearn: 0.5035319\ttotal: 26.2s\tremaining: 12.4s\n",
      "678:\tlearn: 0.5033789\ttotal: 26.2s\tremaining: 12.4s\n",
      "679:\tlearn: 0.5032200\ttotal: 26.3s\tremaining: 12.4s\n",
      "680:\tlearn: 0.5030875\ttotal: 26.4s\tremaining: 12.4s\n",
      "681:\tlearn: 0.5028982\ttotal: 26.4s\tremaining: 12.3s\n",
      "682:\tlearn: 0.5027515\ttotal: 26.5s\tremaining: 12.3s\n",
      "683:\tlearn: 0.5026094\ttotal: 26.5s\tremaining: 12.2s\n",
      "684:\tlearn: 0.5025033\ttotal: 26.5s\tremaining: 12.2s\n",
      "685:\tlearn: 0.5023600\ttotal: 26.6s\tremaining: 12.2s\n",
      "686:\tlearn: 0.5022165\ttotal: 26.6s\tremaining: 12.1s\n",
      "687:\tlearn: 0.5020837\ttotal: 26.6s\tremaining: 12.1s\n",
      "688:\tlearn: 0.5018731\ttotal: 26.7s\tremaining: 12s\n",
      "689:\tlearn: 0.5017232\ttotal: 26.7s\tremaining: 12s\n",
      "690:\tlearn: 0.5016773\ttotal: 26.8s\tremaining: 12s\n",
      "691:\tlearn: 0.5016077\ttotal: 26.8s\tremaining: 11.9s\n",
      "692:\tlearn: 0.5014444\ttotal: 26.8s\tremaining: 11.9s\n",
      "693:\tlearn: 0.5012307\ttotal: 26.9s\tremaining: 11.8s\n",
      "694:\tlearn: 0.5010905\ttotal: 26.9s\tremaining: 11.8s\n",
      "695:\tlearn: 0.5009609\ttotal: 26.9s\tremaining: 11.8s\n",
      "696:\tlearn: 0.5007827\ttotal: 27s\tremaining: 11.7s\n",
      "697:\tlearn: 0.5005616\ttotal: 27s\tremaining: 11.7s\n",
      "698:\tlearn: 0.5005106\ttotal: 27s\tremaining: 11.6s\n",
      "699:\tlearn: 0.5003074\ttotal: 27.1s\tremaining: 11.6s\n",
      "700:\tlearn: 0.5001568\ttotal: 27.2s\tremaining: 11.6s\n",
      "701:\tlearn: 0.4999594\ttotal: 27.2s\tremaining: 11.5s\n",
      "702:\tlearn: 0.4998398\ttotal: 27.2s\tremaining: 11.5s\n",
      "703:\tlearn: 0.4996571\ttotal: 27.3s\tremaining: 11.5s\n",
      "704:\tlearn: 0.4995411\ttotal: 27.3s\tremaining: 11.4s\n",
      "705:\tlearn: 0.4994126\ttotal: 27.4s\tremaining: 11.4s\n",
      "706:\tlearn: 0.4992125\ttotal: 27.4s\tremaining: 11.4s\n",
      "707:\tlearn: 0.4990323\ttotal: 27.4s\tremaining: 11.3s\n",
      "708:\tlearn: 0.4989053\ttotal: 27.5s\tremaining: 11.3s\n",
      "709:\tlearn: 0.4987870\ttotal: 27.5s\tremaining: 11.2s\n",
      "710:\tlearn: 0.4984928\ttotal: 27.5s\tremaining: 11.2s\n",
      "711:\tlearn: 0.4983962\ttotal: 27.6s\tremaining: 11.2s\n",
      "712:\tlearn: 0.4982982\ttotal: 27.6s\tremaining: 11.1s\n",
      "713:\tlearn: 0.4981391\ttotal: 27.6s\tremaining: 11.1s\n",
      "714:\tlearn: 0.4980142\ttotal: 27.7s\tremaining: 11s\n",
      "715:\tlearn: 0.4979069\ttotal: 27.7s\tremaining: 11s\n",
      "716:\tlearn: 0.4977684\ttotal: 27.8s\tremaining: 11s\n",
      "717:\tlearn: 0.4976246\ttotal: 27.8s\tremaining: 10.9s\n",
      "718:\tlearn: 0.4974954\ttotal: 27.8s\tremaining: 10.9s\n",
      "719:\tlearn: 0.4973733\ttotal: 27.9s\tremaining: 10.8s\n",
      "720:\tlearn: 0.4971181\ttotal: 27.9s\tremaining: 10.8s\n",
      "721:\tlearn: 0.4969942\ttotal: 27.9s\tremaining: 10.8s\n",
      "722:\tlearn: 0.4968932\ttotal: 28s\tremaining: 10.7s\n",
      "723:\tlearn: 0.4967923\ttotal: 28s\tremaining: 10.7s\n",
      "724:\tlearn: 0.4965445\ttotal: 28.1s\tremaining: 10.6s\n",
      "725:\tlearn: 0.4964213\ttotal: 28.1s\tremaining: 10.6s\n",
      "726:\tlearn: 0.4962607\ttotal: 28.1s\tremaining: 10.6s\n",
      "727:\tlearn: 0.4962026\ttotal: 28.2s\tremaining: 10.5s\n",
      "728:\tlearn: 0.4960800\ttotal: 28.2s\tremaining: 10.5s\n",
      "729:\tlearn: 0.4958510\ttotal: 28.3s\tremaining: 10.4s\n",
      "730:\tlearn: 0.4956869\ttotal: 28.3s\tremaining: 10.4s\n",
      "731:\tlearn: 0.4955989\ttotal: 28.3s\tremaining: 10.4s\n",
      "732:\tlearn: 0.4955103\ttotal: 28.4s\tremaining: 10.3s\n",
      "733:\tlearn: 0.4953935\ttotal: 28.4s\tremaining: 10.3s\n",
      "734:\tlearn: 0.4953146\ttotal: 28.4s\tremaining: 10.3s\n",
      "735:\tlearn: 0.4951226\ttotal: 28.5s\tremaining: 10.2s\n",
      "736:\tlearn: 0.4949111\ttotal: 28.5s\tremaining: 10.2s\n",
      "737:\tlearn: 0.4948237\ttotal: 28.5s\tremaining: 10.1s\n",
      "738:\tlearn: 0.4947300\ttotal: 28.6s\tremaining: 10.1s\n",
      "739:\tlearn: 0.4945716\ttotal: 28.6s\tremaining: 10.1s\n",
      "740:\tlearn: 0.4943350\ttotal: 28.7s\tremaining: 10s\n",
      "741:\tlearn: 0.4940690\ttotal: 28.7s\tremaining: 9.98s\n",
      "742:\tlearn: 0.4938807\ttotal: 28.7s\tremaining: 9.94s\n",
      "743:\tlearn: 0.4937482\ttotal: 28.8s\tremaining: 9.9s\n",
      "744:\tlearn: 0.4935787\ttotal: 28.8s\tremaining: 9.87s\n",
      "745:\tlearn: 0.4934737\ttotal: 28.9s\tremaining: 9.83s\n",
      "746:\tlearn: 0.4933387\ttotal: 28.9s\tremaining: 9.79s\n",
      "747:\tlearn: 0.4931628\ttotal: 28.9s\tremaining: 9.75s\n",
      "748:\tlearn: 0.4929909\ttotal: 29s\tremaining: 9.71s\n",
      "749:\tlearn: 0.4928854\ttotal: 29s\tremaining: 9.67s\n",
      "750:\tlearn: 0.4928020\ttotal: 29s\tremaining: 9.63s\n",
      "751:\tlearn: 0.4926830\ttotal: 29.1s\tremaining: 9.59s\n",
      "752:\tlearn: 0.4925373\ttotal: 29.1s\tremaining: 9.55s\n",
      "753:\tlearn: 0.4924083\ttotal: 29.2s\tremaining: 9.51s\n",
      "754:\tlearn: 0.4922804\ttotal: 29.2s\tremaining: 9.47s\n",
      "755:\tlearn: 0.4921491\ttotal: 29.2s\tremaining: 9.44s\n",
      "756:\tlearn: 0.4920339\ttotal: 29.3s\tremaining: 9.4s\n",
      "757:\tlearn: 0.4918210\ttotal: 29.3s\tremaining: 9.36s\n",
      "758:\tlearn: 0.4916701\ttotal: 29.4s\tremaining: 9.32s\n",
      "759:\tlearn: 0.4915854\ttotal: 29.4s\tremaining: 9.28s\n",
      "760:\tlearn: 0.4913596\ttotal: 29.4s\tremaining: 9.24s\n",
      "761:\tlearn: 0.4911234\ttotal: 29.5s\tremaining: 9.21s\n",
      "762:\tlearn: 0.4909630\ttotal: 29.5s\tremaining: 9.17s\n",
      "763:\tlearn: 0.4908128\ttotal: 29.6s\tremaining: 9.13s\n",
      "764:\tlearn: 0.4907354\ttotal: 29.6s\tremaining: 9.09s\n",
      "765:\tlearn: 0.4906394\ttotal: 29.6s\tremaining: 9.05s\n",
      "766:\tlearn: 0.4904326\ttotal: 29.7s\tremaining: 9.01s\n",
      "767:\tlearn: 0.4903435\ttotal: 29.7s\tremaining: 8.99s\n",
      "768:\tlearn: 0.4900855\ttotal: 29.8s\tremaining: 8.95s\n",
      "769:\tlearn: 0.4899525\ttotal: 29.8s\tremaining: 8.91s\n",
      "770:\tlearn: 0.4898588\ttotal: 29.9s\tremaining: 8.87s\n",
      "771:\tlearn: 0.4896803\ttotal: 29.9s\tremaining: 8.83s\n",
      "772:\tlearn: 0.4895600\ttotal: 29.9s\tremaining: 8.79s\n",
      "773:\tlearn: 0.4894116\ttotal: 30s\tremaining: 8.75s\n",
      "774:\tlearn: 0.4892531\ttotal: 30s\tremaining: 8.71s\n",
      "775:\tlearn: 0.4891149\ttotal: 30s\tremaining: 8.67s\n",
      "776:\tlearn: 0.4889871\ttotal: 30.1s\tremaining: 8.63s\n",
      "777:\tlearn: 0.4888399\ttotal: 30.1s\tremaining: 8.6s\n",
      "778:\tlearn: 0.4886826\ttotal: 30.2s\tremaining: 8.56s\n",
      "779:\tlearn: 0.4885756\ttotal: 30.2s\tremaining: 8.53s\n",
      "780:\tlearn: 0.4883908\ttotal: 30.3s\tremaining: 8.48s\n",
      "781:\tlearn: 0.4882812\ttotal: 30.3s\tremaining: 8.45s\n",
      "782:\tlearn: 0.4881460\ttotal: 30.3s\tremaining: 8.41s\n",
      "783:\tlearn: 0.4879921\ttotal: 30.4s\tremaining: 8.37s\n",
      "784:\tlearn: 0.4878887\ttotal: 30.4s\tremaining: 8.33s\n",
      "785:\tlearn: 0.4877462\ttotal: 30.4s\tremaining: 8.29s\n",
      "786:\tlearn: 0.4876017\ttotal: 30.5s\tremaining: 8.25s\n",
      "787:\tlearn: 0.4875093\ttotal: 30.5s\tremaining: 8.21s\n",
      "788:\tlearn: 0.4873649\ttotal: 30.6s\tremaining: 8.17s\n",
      "789:\tlearn: 0.4871991\ttotal: 30.6s\tremaining: 8.13s\n",
      "790:\tlearn: 0.4870799\ttotal: 30.6s\tremaining: 8.09s\n",
      "791:\tlearn: 0.4869377\ttotal: 30.7s\tremaining: 8.06s\n",
      "792:\tlearn: 0.4868043\ttotal: 30.7s\tremaining: 8.02s\n",
      "793:\tlearn: 0.4866809\ttotal: 30.8s\tremaining: 7.98s\n",
      "794:\tlearn: 0.4865576\ttotal: 30.8s\tremaining: 7.94s\n",
      "795:\tlearn: 0.4863372\ttotal: 30.8s\tremaining: 7.9s\n",
      "796:\tlearn: 0.4861977\ttotal: 30.9s\tremaining: 7.86s\n",
      "797:\tlearn: 0.4860855\ttotal: 30.9s\tremaining: 7.82s\n",
      "798:\tlearn: 0.4860219\ttotal: 30.9s\tremaining: 7.78s\n",
      "799:\tlearn: 0.4858960\ttotal: 31s\tremaining: 7.74s\n",
      "800:\tlearn: 0.4857990\ttotal: 31s\tremaining: 7.7s\n",
      "801:\tlearn: 0.4856848\ttotal: 31s\tremaining: 7.66s\n",
      "802:\tlearn: 0.4855134\ttotal: 31.1s\tremaining: 7.63s\n",
      "803:\tlearn: 0.4853968\ttotal: 31.1s\tremaining: 7.59s\n",
      "804:\tlearn: 0.4852500\ttotal: 31.2s\tremaining: 7.55s\n",
      "805:\tlearn: 0.4850183\ttotal: 31.2s\tremaining: 7.51s\n",
      "806:\tlearn: 0.4847447\ttotal: 31.2s\tremaining: 7.47s\n",
      "807:\tlearn: 0.4846556\ttotal: 31.3s\tremaining: 7.43s\n",
      "808:\tlearn: 0.4844639\ttotal: 31.3s\tremaining: 7.39s\n",
      "809:\tlearn: 0.4842979\ttotal: 31.3s\tremaining: 7.35s\n",
      "810:\tlearn: 0.4842109\ttotal: 31.4s\tremaining: 7.31s\n",
      "811:\tlearn: 0.4840512\ttotal: 31.4s\tremaining: 7.27s\n",
      "812:\tlearn: 0.4839347\ttotal: 31.5s\tremaining: 7.24s\n",
      "813:\tlearn: 0.4838132\ttotal: 31.5s\tremaining: 7.2s\n",
      "814:\tlearn: 0.4836428\ttotal: 31.5s\tremaining: 7.16s\n",
      "815:\tlearn: 0.4835360\ttotal: 31.6s\tremaining: 7.12s\n",
      "816:\tlearn: 0.4834728\ttotal: 31.6s\tremaining: 7.08s\n",
      "817:\tlearn: 0.4831904\ttotal: 31.6s\tremaining: 7.04s\n",
      "818:\tlearn: 0.4831068\ttotal: 31.7s\tremaining: 7s\n",
      "819:\tlearn: 0.4829751\ttotal: 31.7s\tremaining: 6.96s\n",
      "820:\tlearn: 0.4828196\ttotal: 31.8s\tremaining: 6.92s\n",
      "821:\tlearn: 0.4826832\ttotal: 31.8s\tremaining: 6.88s\n",
      "822:\tlearn: 0.4825058\ttotal: 31.8s\tremaining: 6.84s\n",
      "823:\tlearn: 0.4823878\ttotal: 31.9s\tremaining: 6.81s\n",
      "824:\tlearn: 0.4823382\ttotal: 31.9s\tremaining: 6.77s\n",
      "825:\tlearn: 0.4822253\ttotal: 31.9s\tremaining: 6.73s\n",
      "826:\tlearn: 0.4821032\ttotal: 32s\tremaining: 6.69s\n",
      "827:\tlearn: 0.4820297\ttotal: 32s\tremaining: 6.65s\n",
      "828:\tlearn: 0.4819400\ttotal: 32s\tremaining: 6.61s\n",
      "829:\tlearn: 0.4818157\ttotal: 32.1s\tremaining: 6.57s\n",
      "830:\tlearn: 0.4816688\ttotal: 32.1s\tremaining: 6.53s\n",
      "831:\tlearn: 0.4815033\ttotal: 32.2s\tremaining: 6.5s\n",
      "832:\tlearn: 0.4812871\ttotal: 32.2s\tremaining: 6.46s\n",
      "833:\tlearn: 0.4810040\ttotal: 32.2s\tremaining: 6.42s\n",
      "834:\tlearn: 0.4808327\ttotal: 32.3s\tremaining: 6.38s\n",
      "835:\tlearn: 0.4806090\ttotal: 32.3s\tremaining: 6.34s\n",
      "836:\tlearn: 0.4804659\ttotal: 32.3s\tremaining: 6.3s\n",
      "837:\tlearn: 0.4803658\ttotal: 32.4s\tremaining: 6.27s\n",
      "838:\tlearn: 0.4801324\ttotal: 32.5s\tremaining: 6.23s\n",
      "839:\tlearn: 0.4799316\ttotal: 32.5s\tremaining: 6.19s\n",
      "840:\tlearn: 0.4797666\ttotal: 32.5s\tremaining: 6.15s\n",
      "841:\tlearn: 0.4796173\ttotal: 32.6s\tremaining: 6.12s\n",
      "842:\tlearn: 0.4795342\ttotal: 32.6s\tremaining: 6.08s\n",
      "843:\tlearn: 0.4794423\ttotal: 32.7s\tremaining: 6.04s\n",
      "844:\tlearn: 0.4792927\ttotal: 32.7s\tremaining: 6s\n",
      "845:\tlearn: 0.4791638\ttotal: 32.7s\tremaining: 5.96s\n",
      "846:\tlearn: 0.4790672\ttotal: 32.8s\tremaining: 5.92s\n",
      "847:\tlearn: 0.4789877\ttotal: 32.8s\tremaining: 5.88s\n",
      "848:\tlearn: 0.4788764\ttotal: 32.8s\tremaining: 5.84s\n",
      "849:\tlearn: 0.4787523\ttotal: 32.9s\tremaining: 5.8s\n",
      "850:\tlearn: 0.4785481\ttotal: 32.9s\tremaining: 5.76s\n",
      "851:\tlearn: 0.4784016\ttotal: 33s\tremaining: 5.72s\n",
      "852:\tlearn: 0.4782199\ttotal: 33s\tremaining: 5.69s\n",
      "853:\tlearn: 0.4780539\ttotal: 33s\tremaining: 5.65s\n",
      "854:\tlearn: 0.4779006\ttotal: 33.1s\tremaining: 5.61s\n",
      "855:\tlearn: 0.4777234\ttotal: 33.1s\tremaining: 5.57s\n",
      "856:\tlearn: 0.4775377\ttotal: 33.1s\tremaining: 5.53s\n",
      "857:\tlearn: 0.4773849\ttotal: 33.2s\tremaining: 5.49s\n",
      "858:\tlearn: 0.4772380\ttotal: 33.2s\tremaining: 5.45s\n",
      "859:\tlearn: 0.4770252\ttotal: 33.3s\tremaining: 5.42s\n",
      "860:\tlearn: 0.4768173\ttotal: 33.3s\tremaining: 5.38s\n",
      "861:\tlearn: 0.4767265\ttotal: 33.3s\tremaining: 5.34s\n",
      "862:\tlearn: 0.4765804\ttotal: 33.4s\tremaining: 5.3s\n",
      "863:\tlearn: 0.4763956\ttotal: 33.4s\tremaining: 5.26s\n",
      "864:\tlearn: 0.4762207\ttotal: 33.4s\tremaining: 5.22s\n",
      "865:\tlearn: 0.4761387\ttotal: 33.5s\tremaining: 5.18s\n",
      "866:\tlearn: 0.4759849\ttotal: 33.5s\tremaining: 5.14s\n",
      "867:\tlearn: 0.4758734\ttotal: 33.6s\tremaining: 5.1s\n",
      "868:\tlearn: 0.4757484\ttotal: 33.6s\tremaining: 5.06s\n",
      "869:\tlearn: 0.4756353\ttotal: 33.6s\tremaining: 5.02s\n",
      "870:\tlearn: 0.4755245\ttotal: 33.7s\tremaining: 4.98s\n",
      "871:\tlearn: 0.4754271\ttotal: 33.7s\tremaining: 4.95s\n",
      "872:\tlearn: 0.4752820\ttotal: 33.7s\tremaining: 4.91s\n",
      "873:\tlearn: 0.4751924\ttotal: 33.8s\tremaining: 4.87s\n",
      "874:\tlearn: 0.4751110\ttotal: 33.8s\tremaining: 4.83s\n",
      "875:\tlearn: 0.4748869\ttotal: 33.8s\tremaining: 4.79s\n",
      "876:\tlearn: 0.4747498\ttotal: 33.9s\tremaining: 4.75s\n",
      "877:\tlearn: 0.4746687\ttotal: 33.9s\tremaining: 4.71s\n",
      "878:\tlearn: 0.4745908\ttotal: 34s\tremaining: 4.67s\n",
      "879:\tlearn: 0.4744281\ttotal: 34s\tremaining: 4.63s\n",
      "880:\tlearn: 0.4743418\ttotal: 34s\tremaining: 4.59s\n",
      "881:\tlearn: 0.4741819\ttotal: 34.1s\tremaining: 4.56s\n",
      "882:\tlearn: 0.4740096\ttotal: 34.1s\tremaining: 4.52s\n",
      "883:\tlearn: 0.4739095\ttotal: 34.1s\tremaining: 4.48s\n",
      "884:\tlearn: 0.4737599\ttotal: 34.2s\tremaining: 4.44s\n",
      "885:\tlearn: 0.4736578\ttotal: 34.2s\tremaining: 4.4s\n",
      "886:\tlearn: 0.4735607\ttotal: 34.2s\tremaining: 4.36s\n",
      "887:\tlearn: 0.4734603\ttotal: 34.3s\tremaining: 4.32s\n",
      "888:\tlearn: 0.4733252\ttotal: 34.3s\tremaining: 4.28s\n",
      "889:\tlearn: 0.4731938\ttotal: 34.4s\tremaining: 4.25s\n",
      "890:\tlearn: 0.4730426\ttotal: 34.4s\tremaining: 4.21s\n",
      "891:\tlearn: 0.4729413\ttotal: 34.4s\tremaining: 4.17s\n",
      "892:\tlearn: 0.4728144\ttotal: 34.5s\tremaining: 4.13s\n",
      "893:\tlearn: 0.4726637\ttotal: 34.5s\tremaining: 4.09s\n",
      "894:\tlearn: 0.4725830\ttotal: 34.5s\tremaining: 4.05s\n",
      "895:\tlearn: 0.4723805\ttotal: 34.6s\tremaining: 4.01s\n",
      "896:\tlearn: 0.4722322\ttotal: 34.6s\tremaining: 3.98s\n",
      "897:\tlearn: 0.4720646\ttotal: 34.7s\tremaining: 3.94s\n",
      "898:\tlearn: 0.4719585\ttotal: 34.7s\tremaining: 3.9s\n",
      "899:\tlearn: 0.4718786\ttotal: 34.7s\tremaining: 3.86s\n",
      "900:\tlearn: 0.4717093\ttotal: 34.8s\tremaining: 3.83s\n",
      "901:\tlearn: 0.4715050\ttotal: 34.9s\tremaining: 3.79s\n",
      "902:\tlearn: 0.4713082\ttotal: 34.9s\tremaining: 3.75s\n",
      "903:\tlearn: 0.4710950\ttotal: 34.9s\tremaining: 3.71s\n",
      "904:\tlearn: 0.4709582\ttotal: 35s\tremaining: 3.67s\n",
      "905:\tlearn: 0.4707787\ttotal: 35s\tremaining: 3.63s\n",
      "906:\tlearn: 0.4706434\ttotal: 35s\tremaining: 3.59s\n",
      "907:\tlearn: 0.4704745\ttotal: 35.1s\tremaining: 3.55s\n",
      "908:\tlearn: 0.4703826\ttotal: 35.1s\tremaining: 3.52s\n",
      "909:\tlearn: 0.4702777\ttotal: 35.2s\tremaining: 3.48s\n",
      "910:\tlearn: 0.4701839\ttotal: 35.2s\tremaining: 3.44s\n",
      "911:\tlearn: 0.4700476\ttotal: 35.2s\tremaining: 3.4s\n",
      "912:\tlearn: 0.4699761\ttotal: 35.3s\tremaining: 3.36s\n",
      "913:\tlearn: 0.4698725\ttotal: 35.3s\tremaining: 3.32s\n",
      "914:\tlearn: 0.4696631\ttotal: 35.3s\tremaining: 3.28s\n",
      "915:\tlearn: 0.4696014\ttotal: 35.4s\tremaining: 3.24s\n",
      "916:\tlearn: 0.4694730\ttotal: 35.4s\tremaining: 3.2s\n",
      "917:\tlearn: 0.4693920\ttotal: 35.4s\tremaining: 3.17s\n",
      "918:\tlearn: 0.4692291\ttotal: 35.5s\tremaining: 3.13s\n",
      "919:\tlearn: 0.4690765\ttotal: 35.5s\tremaining: 3.09s\n",
      "920:\tlearn: 0.4689399\ttotal: 35.6s\tremaining: 3.05s\n",
      "921:\tlearn: 0.4688421\ttotal: 35.6s\tremaining: 3.01s\n",
      "922:\tlearn: 0.4687363\ttotal: 35.6s\tremaining: 2.97s\n",
      "923:\tlearn: 0.4686208\ttotal: 35.7s\tremaining: 2.93s\n",
      "924:\tlearn: 0.4684979\ttotal: 35.7s\tremaining: 2.89s\n",
      "925:\tlearn: 0.4684212\ttotal: 35.7s\tremaining: 2.86s\n",
      "926:\tlearn: 0.4683262\ttotal: 35.8s\tremaining: 2.82s\n",
      "927:\tlearn: 0.4680842\ttotal: 35.8s\tremaining: 2.78s\n",
      "928:\tlearn: 0.4679091\ttotal: 35.9s\tremaining: 2.74s\n",
      "929:\tlearn: 0.4677861\ttotal: 35.9s\tremaining: 2.7s\n",
      "930:\tlearn: 0.4676413\ttotal: 35.9s\tremaining: 2.66s\n",
      "931:\tlearn: 0.4675256\ttotal: 36s\tremaining: 2.62s\n",
      "932:\tlearn: 0.4674451\ttotal: 36s\tremaining: 2.58s\n",
      "933:\tlearn: 0.4673370\ttotal: 36s\tremaining: 2.55s\n",
      "934:\tlearn: 0.4672285\ttotal: 36.1s\tremaining: 2.51s\n",
      "935:\tlearn: 0.4671089\ttotal: 36.1s\tremaining: 2.47s\n",
      "936:\tlearn: 0.4669291\ttotal: 36.2s\tremaining: 2.43s\n",
      "937:\tlearn: 0.4668033\ttotal: 36.2s\tremaining: 2.39s\n",
      "938:\tlearn: 0.4666764\ttotal: 36.2s\tremaining: 2.35s\n",
      "939:\tlearn: 0.4665409\ttotal: 36.3s\tremaining: 2.31s\n",
      "940:\tlearn: 0.4664130\ttotal: 36.3s\tremaining: 2.28s\n",
      "941:\tlearn: 0.4662662\ttotal: 36.3s\tremaining: 2.24s\n",
      "942:\tlearn: 0.4661731\ttotal: 36.4s\tremaining: 2.2s\n",
      "943:\tlearn: 0.4660897\ttotal: 36.4s\tremaining: 2.16s\n",
      "944:\tlearn: 0.4659198\ttotal: 36.5s\tremaining: 2.12s\n",
      "945:\tlearn: 0.4658123\ttotal: 36.5s\tremaining: 2.08s\n",
      "946:\tlearn: 0.4655792\ttotal: 36.5s\tremaining: 2.04s\n",
      "947:\tlearn: 0.4654187\ttotal: 36.6s\tremaining: 2s\n",
      "948:\tlearn: 0.4653415\ttotal: 36.6s\tremaining: 1.97s\n",
      "949:\tlearn: 0.4652374\ttotal: 36.6s\tremaining: 1.93s\n",
      "950:\tlearn: 0.4650149\ttotal: 36.7s\tremaining: 1.89s\n",
      "951:\tlearn: 0.4648662\ttotal: 36.7s\tremaining: 1.85s\n",
      "952:\tlearn: 0.4646795\ttotal: 36.7s\tremaining: 1.81s\n",
      "953:\tlearn: 0.4645788\ttotal: 36.8s\tremaining: 1.77s\n",
      "954:\tlearn: 0.4644712\ttotal: 36.8s\tremaining: 1.73s\n",
      "955:\tlearn: 0.4643018\ttotal: 36.9s\tremaining: 1.7s\n",
      "956:\tlearn: 0.4641355\ttotal: 36.9s\tremaining: 1.66s\n",
      "957:\tlearn: 0.4640200\ttotal: 36.9s\tremaining: 1.62s\n",
      "958:\tlearn: 0.4638976\ttotal: 37s\tremaining: 1.58s\n",
      "959:\tlearn: 0.4637538\ttotal: 37s\tremaining: 1.54s\n",
      "960:\tlearn: 0.4636502\ttotal: 37s\tremaining: 1.5s\n",
      "961:\tlearn: 0.4635267\ttotal: 37.1s\tremaining: 1.47s\n",
      "962:\tlearn: 0.4633551\ttotal: 37.1s\tremaining: 1.43s\n",
      "963:\tlearn: 0.4632830\ttotal: 37.2s\tremaining: 1.39s\n",
      "964:\tlearn: 0.4631403\ttotal: 37.2s\tremaining: 1.35s\n",
      "965:\tlearn: 0.4629927\ttotal: 37.2s\tremaining: 1.31s\n",
      "966:\tlearn: 0.4627746\ttotal: 37.3s\tremaining: 1.27s\n",
      "967:\tlearn: 0.4626778\ttotal: 37.3s\tremaining: 1.23s\n",
      "968:\tlearn: 0.4625412\ttotal: 37.4s\tremaining: 1.2s\n",
      "969:\tlearn: 0.4624270\ttotal: 37.4s\tremaining: 1.16s\n",
      "970:\tlearn: 0.4623315\ttotal: 37.4s\tremaining: 1.12s\n",
      "971:\tlearn: 0.4622049\ttotal: 37.5s\tremaining: 1.08s\n",
      "972:\tlearn: 0.4621057\ttotal: 37.5s\tremaining: 1.04s\n",
      "973:\tlearn: 0.4619750\ttotal: 37.6s\tremaining: 1s\n",
      "974:\tlearn: 0.4618137\ttotal: 37.6s\tremaining: 964ms\n",
      "975:\tlearn: 0.4615878\ttotal: 37.6s\tremaining: 925ms\n",
      "976:\tlearn: 0.4613595\ttotal: 37.7s\tremaining: 887ms\n",
      "977:\tlearn: 0.4612842\ttotal: 37.7s\tremaining: 848ms\n",
      "978:\tlearn: 0.4611178\ttotal: 37.7s\tremaining: 810ms\n",
      "979:\tlearn: 0.4609935\ttotal: 37.8s\tremaining: 771ms\n",
      "980:\tlearn: 0.4608742\ttotal: 37.8s\tremaining: 733ms\n",
      "981:\tlearn: 0.4607649\ttotal: 37.9s\tremaining: 694ms\n",
      "982:\tlearn: 0.4606389\ttotal: 37.9s\tremaining: 655ms\n",
      "983:\tlearn: 0.4605357\ttotal: 37.9s\tremaining: 617ms\n",
      "984:\tlearn: 0.4603989\ttotal: 38s\tremaining: 578ms\n",
      "985:\tlearn: 0.4601675\ttotal: 38s\tremaining: 540ms\n",
      "986:\tlearn: 0.4599494\ttotal: 38s\tremaining: 501ms\n",
      "987:\tlearn: 0.4598042\ttotal: 38.1s\tremaining: 462ms\n",
      "988:\tlearn: 0.4597079\ttotal: 38.1s\tremaining: 424ms\n",
      "989:\tlearn: 0.4596135\ttotal: 38.1s\tremaining: 385ms\n",
      "990:\tlearn: 0.4594517\ttotal: 38.2s\tremaining: 347ms\n",
      "991:\tlearn: 0.4593834\ttotal: 38.2s\tremaining: 308ms\n",
      "992:\tlearn: 0.4592774\ttotal: 38.3s\tremaining: 270ms\n",
      "993:\tlearn: 0.4591685\ttotal: 38.3s\tremaining: 231ms\n",
      "994:\tlearn: 0.4590823\ttotal: 38.3s\tremaining: 193ms\n",
      "995:\tlearn: 0.4589880\ttotal: 38.4s\tremaining: 154ms\n",
      "996:\tlearn: 0.4588357\ttotal: 38.4s\tremaining: 116ms\n",
      "997:\tlearn: 0.4586742\ttotal: 38.5s\tremaining: 77.1ms\n",
      "998:\tlearn: 0.4584835\ttotal: 38.5s\tremaining: 38.5ms\n",
      "999:\tlearn: 0.4583702\ttotal: 38.5s\tremaining: 0us\n",
      "The accuracy score for the <catboost.core.CatBoostClassifier object at 0x0000015AB44F3EB0> oversampled set is: 0.6993174061433447\n",
      "The f1 score for the <catboost.core.CatBoostClassifier object at 0x0000015AB44F3EB0> oversampled set is: 0.15043394406943106\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# for i in [base_svc, base_rfc, base_gbc, base_ada, base_xgb, base_cat]:\n",
    "#     sampler(i, rus_train, ros_train, new_train, y_rus, y_ros, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Running the above cell shows the base gradient boosting classifier performed best. Confirm with hyperparameter tuning and select the best model and corresponding hyperparameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "#generate function to generate classification reports\n",
    "#generate a function to display confusion matrices and classification reports\n",
    "def report(true, predicted, string):\n",
    "    matrix = confusion_matrix(true, predicted)\n",
    "    print(string + ' ' + 'Confusion Matrix')\n",
    "    print(matrix)\n",
    "    print('\\n')\n",
    "    report = classification_report(true, predicted)\n",
    "    print(string + ' ' + 'Classification Report')\n",
    "    print(report)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "gbc = GradientBoostingClassifier(min_samples_split=4, n_estimators=32, random_state=45, subsample=0.6)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "#generate function for logistic regression predictions\n",
    "def gb_model(sample_df, y_df, cols, f_string):\n",
    "\n",
    "    gbc.fit(sample_df[cols], y_df)\n",
    "\n",
    "    preds = gbc.predict(x_test[cols])\n",
    "\n",
    "    report(y_test, preds, string=f_string)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "#for the unsampled set\n",
    "f_features_un = num_f_classif + un_cat      #f-feature selection\n",
    "mi_features_un = mi_un + un_cat      #mutual info feature selection\n",
    "\n",
    "#for the undersampled set\n",
    "f_features_under = num_f_classif + under_cat\n",
    "mi_features_under = mi_under + under_cat\n",
    "\n",
    "#for the oversampled set\n",
    "f_features_over = num_f_classif + over_cat\n",
    "mi_features_over = mi_over + over_cat"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 8.1) Unsampled split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 8.1.1) Features using F-classification methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "f_classif features Confusion Matrix\n",
      "[[2746    0]\n",
      " [ 184    0]]\n",
      "\n",
      "\n",
      "f_classif features Classification Report\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.94      1.00      0.97      2746\n",
      "           1       0.00      0.00      0.00       184\n",
      "\n",
      "    accuracy                           0.94      2930\n",
      "   macro avg       0.47      0.50      0.48      2930\n",
      "weighted avg       0.88      0.94      0.91      2930\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "d:\\insurance_claims\\venv\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1469: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "d:\\insurance_claims\\venv\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1469: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "d:\\insurance_claims\\venv\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1469: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    }
   ],
   "source": [
    "#for svc:\n",
    "gb_model(new_train, y_train, f_features_un, 'f_classif features')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 8.1.2) Features using mutual information methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mutual info Confusion Matrix\n",
      "[[2746    0]\n",
      " [ 184    0]]\n",
      "\n",
      "\n",
      "mutual info Classification Report\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.94      1.00      0.97      2746\n",
      "           1       0.00      0.00      0.00       184\n",
      "\n",
      "    accuracy                           0.94      2930\n",
      "   macro avg       0.47      0.50      0.48      2930\n",
      "weighted avg       0.88      0.94      0.91      2930\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "d:\\insurance_claims\\venv\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1469: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "d:\\insurance_claims\\venv\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1469: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "d:\\insurance_claims\\venv\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1469: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    }
   ],
   "source": [
    "#for svc:\n",
    "gb_model(new_train, y_train, mi_features_un, 'mutual info')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "***Insight***\n",
    "\n",
    "- For the unsampled split, the gradient boosting model performed poorly for both sets of features\n",
    "- The model failed to generate any positive occurences which led to a very high accuracy score.\n",
    "- As suspected, the unsampled data did not relay necessary information to generate predictions."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 8.2) Undersampled split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 8.2.1) Features using F-classification methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "f_classif features Confusion Matrix\n",
      "[[1362 1384]\n",
      " [  54  130]]\n",
      "\n",
      "\n",
      "f_classif features Classification Report\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.50      0.65      2746\n",
      "           1       0.09      0.71      0.15       184\n",
      "\n",
      "    accuracy                           0.51      2930\n",
      "   macro avg       0.52      0.60      0.40      2930\n",
      "weighted avg       0.91      0.51      0.62      2930\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#for svc:\n",
    "gb_model(rus_train, y_rus, f_features_under, 'f_classif features')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 8.2.2) Features using mutual information methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mutual info Confusion Matrix\n",
      "[[1271 1475]\n",
      " [  63  121]]\n",
      "\n",
      "\n",
      "mutual info Classification Report\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      0.46      0.62      2746\n",
      "           1       0.08      0.66      0.14       184\n",
      "\n",
      "    accuracy                           0.48      2930\n",
      "   macro avg       0.51      0.56      0.38      2930\n",
      "weighted avg       0.90      0.48      0.59      2930\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#for svc:\n",
    "gb_model(rus_train, y_rus, mi_features_under, 'mutual info')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "***Insight***\n",
    "\n",
    "- The undersampled set led to better performance of the model.\n",
    "- For the f_classif feature selection method:\n",
    "    - Out of 184 **true** positive occurences, the model correctly predicted 130. However, 1438 predictions were mismatched to achieve this.\n",
    "    - 1362 matched samples for the majority class led to a high precision of 96%.\n",
    "    - The precision and recall parameters for the minority class were better, with an accuracy of 65%.\n",
    "    - The model was able to generate 1514 predictions for the minority class.\n",
    "\n",
    "- For mutual information set:\n",
    "    - The model correctly predicted 121 positive occurences-slightly lower than the f_classif set, with a hundred more mismatched predictions\n",
    "    - The precision and recall parameters for both classes were lower, with 48%.\n",
    "\n",
    "- For the undersampled set, the f_classif set performed better."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 8.3) Oversampled split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 8.3.1) Features using F-classification methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "f_classif features Confusion Matrix\n",
      "[[1415 1331]\n",
      " [  59  125]]\n",
      "\n",
      "\n",
      "f_classif features Classification Report\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.52      0.67      2746\n",
      "           1       0.09      0.68      0.15       184\n",
      "\n",
      "    accuracy                           0.53      2930\n",
      "   macro avg       0.52      0.60      0.41      2930\n",
      "weighted avg       0.91      0.53      0.64      2930\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#for svc:\n",
    "gb_model(ros_train, y_ros, f_features_over, 'f_classif features')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 8.3.2) Features using mutual information methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mutual info features Confusion Matrix\n",
      "[[1385 1361]\n",
      " [  66  118]]\n",
      "\n",
      "\n",
      "mutual info features Classification Report\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      0.50      0.66      2746\n",
      "           1       0.08      0.64      0.14       184\n",
      "\n",
      "    accuracy                           0.51      2930\n",
      "   macro avg       0.52      0.57      0.40      2930\n",
      "weighted avg       0.90      0.51      0.63      2930\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#for svc:\n",
    "gb_model(ros_train, y_ros, mi_features_over, 'mutual info features')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "***Insight***\n",
    "\n",
    "- The oversampled set led to better model performance (as compared to the unsampled set) for both feature selection methods\n",
    "- However, the undersampled set performed better than oversampled set\n",
    "- The undersampled set will be used for model training"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 8.4) Model optimization\n",
    "\n",
    "- Determine top features for the selected model\n",
    "    - mutual information\n",
    "    - f_classif\n",
    "- Select the best features from the gradient boosting model\n",
    "- Generate optimized results using decision threshold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "def best(df, y_df, scorefunc, cols, k_value):\n",
    "    SelectKBest(score_func=scorefunc, k=k_value).fit_transform(df[cols], y_df)\n",
    "    print(df[cols].columns[SelectKBest(score_func=scorefunc, k=k_value).fit(df[cols], y_df).get_support()])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 8.4.1) Determine top features for the model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Mutual Information**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['policy_tenure', 'age_of_policyholder', 'population_density',\n",
      "       'gear_box', 'is_adjustable_steering', 'is_tpms', 'is_parking_sensors',\n",
      "       'is_rear_window_wiper', 'is_day_night_rear_view_mirror', 'is_ecw'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "best(rus_train, y_rus, mutual_info_classif, initial_features, 10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['policy_tenure', 'age_of_car', 'displacement', 'is_esc',\n",
      "       'is_rear_window_defogger'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "best(rus_train, y_rus, mutual_info_classif, initial_features, 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['policy_tenure', 'cylinder', 'displacement', 'is_parking_camera',\n",
      "       'is_tpms'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "best(rus_train, y_rus, mutual_info_classif, mi_features_under, 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['policy_tenure', 'cylinder', 'displacement', 'transmission_type',\n",
      "       'rear_brakes_type', 'is_parking_camera', 'is_tpms',\n",
      "       'is_parking_sensors', 'is_rear_window_wiper',\n",
      "       'is_rear_window_defogger'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "best(rus_train, y_rus, mutual_info_classif, mi_features_under, 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**F_classif**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['policy_tenure', 'age_of_car', 'age_of_policyholder',\n",
      "       'population_density', 'cylinder', 'is_adjustable_steering',\n",
      "       'is_parking_sensors', 'is_front_fog_lights', 'is_brake_assist',\n",
      "       'max_torque'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "best(rus_train, y_rus, f_classif, initial_features, 10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['policy_tenure', 'age_of_car', 'population_density', 'cylinder',\n",
      "       'max_torque'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "best(rus_train, y_rus, f_classif, initial_features, 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['policy_tenure', 'age_of_car', 'age_of_policyholder',\n",
      "       'population_density', 'cylinder', 'is_parking_sensors'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "best(rus_train, y_rus, f_classif, f_features_under, 6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['policy_tenure', 'age_of_car', 'age_of_policyholder',\n",
      "       'population_density', 'cylinder', 'rear_brakes_type', 'is_tpms',\n",
      "       'is_parking_sensors', 'is_rear_window_wiper',\n",
      "       'is_rear_window_defogger'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "best(rus_train, y_rus, f_classif, f_features_under, 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 8.4.2) Best features from the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "selector = SelectFromModel(estimator=gbc).fit(rus_train, y_rus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['population_density', 'policy_tenure', 'age_of_car',\n",
       "       'age_of_policyholder'], dtype=object)"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "selector.get_feature_names_out()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "top_features = ['population_density', 'policy_tenure', 'age_of_car', 'age_of_policyholder']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "selector2 = SelectFromModel(estimator=gbc, threshold='0.25*mean').fit(rus_train, y_rus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "f_set2 = list(selector2.get_feature_names_out())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "selector3 = SelectFromModel(estimator=gbc, threshold='0.5*mean').fit(rus_train, y_rus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "f_set3 = list(selector3.get_feature_names_out())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(top_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(f_set2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(f_set3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "selector4 = SelectFromModel(estimator=gbc, threshold='0.75*mean').fit(rus_train, y_rus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "f_set4 = list(selector4.get_feature_names_out())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "7"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(f_set4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['population_density',\n",
       " 'displacement',\n",
       " 'height',\n",
       " 'gross_weight',\n",
       " 'policy_tenure',\n",
       " 'age_of_car',\n",
       " 'age_of_policyholder']"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f_set4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['population_density',\n",
       " 'displacement',\n",
       " 'length',\n",
       " 'height',\n",
       " 'gross_weight',\n",
       " 'policy_tenure',\n",
       " 'age_of_car',\n",
       " 'age_of_policyholder']"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f_set3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['population_density',\n",
       " 'displacement',\n",
       " 'length',\n",
       " 'height',\n",
       " 'gross_weight',\n",
       " 'max_torque',\n",
       " 'steering_type',\n",
       " 'policy_tenure',\n",
       " 'age_of_car',\n",
       " 'age_of_policyholder']"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f_set2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['population_density', 'policy_tenure', 'age_of_car', 'age_of_policyholder']"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "top_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "gbc.fit(rus_train[top_features], y_rus)\n",
    "\n",
    "preds2 = gbc.predict(x_test[top_features])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top features for gbc Confusion Matrix\n",
      "[[1296 1450]\n",
      " [  55  129]]\n",
      "\n",
      "\n",
      "Top features for gbc Classification Report\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.47      0.63      2746\n",
      "           1       0.08      0.70      0.15       184\n",
      "\n",
      "    accuracy                           0.49      2930\n",
      "   macro avg       0.52      0.59      0.39      2930\n",
      "weighted avg       0.90      0.49      0.60      2930\n",
      "\n"
     ]
    }
   ],
   "source": [
    "report(y_test, preds2, 'Top features for gbc')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top2 Confusion Matrix\n",
      "[[1388 1358]\n",
      " [  54  130]]\n",
      "\n",
      "\n",
      "Top2 Classification Report\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.51      0.66      2746\n",
      "           1       0.09      0.71      0.16       184\n",
      "\n",
      "    accuracy                           0.52      2930\n",
      "   macro avg       0.52      0.61      0.41      2930\n",
      "weighted avg       0.91      0.52      0.63      2930\n",
      "\n"
     ]
    }
   ],
   "source": [
    "gbc.fit(rus_train[f_set2], y_rus)\n",
    "\n",
    "preds3 = gbc.predict(x_test[f_set2])\n",
    "\n",
    "report(y_test, preds3, 'Top2')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top3 Confusion Matrix\n",
      "[[1400 1346]\n",
      " [  58  126]]\n",
      "\n",
      "\n",
      "Top3 Classification Report\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.51      0.67      2746\n",
      "           1       0.09      0.68      0.15       184\n",
      "\n",
      "    accuracy                           0.52      2930\n",
      "   macro avg       0.52      0.60      0.41      2930\n",
      "weighted avg       0.91      0.52      0.63      2930\n",
      "\n"
     ]
    }
   ],
   "source": [
    "gbc.fit(rus_train[f_set3], y_rus)\n",
    "\n",
    "preds4 = gbc.predict(x_test[f_set3])\n",
    "\n",
    "report(y_test, preds4, 'Top3')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top4 Confusion Matrix\n",
      "[[1381 1365]\n",
      " [  58  126]]\n",
      "\n",
      "\n",
      "Top4 Classification Report\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.50      0.66      2746\n",
      "           1       0.08      0.68      0.15       184\n",
      "\n",
      "    accuracy                           0.51      2930\n",
      "   macro avg       0.52      0.59      0.41      2930\n",
      "weighted avg       0.90      0.51      0.63      2930\n",
      "\n"
     ]
    }
   ],
   "source": [
    "gbc.fit(rus_train[f_set4], y_rus)\n",
    "\n",
    "preds5 = gbc.predict(x_test[f_set4])\n",
    "\n",
    "report(y_test, preds5, 'Top4')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['population_density',\n",
       " 'displacement',\n",
       " 'length',\n",
       " 'height',\n",
       " 'gross_weight',\n",
       " 'max_torque',\n",
       " 'steering_type',\n",
       " 'policy_tenure',\n",
       " 'age_of_car',\n",
       " 'age_of_policyholder']"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f_set2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "user_features = ['age_of_car', 'age_of_policyholder', 'policy_tenure', 'population_density', 'steering_type']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "User features Confusion Matrix\n",
      "[[1332 1414]\n",
      " [  54  130]]\n",
      "\n",
      "\n",
      "User features Classification Report\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.49      0.64      2746\n",
      "           1       0.08      0.71      0.15       184\n",
      "\n",
      "    accuracy                           0.50      2930\n",
      "   macro avg       0.52      0.60      0.40      2930\n",
      "weighted avg       0.91      0.50      0.61      2930\n",
      "\n"
     ]
    }
   ],
   "source": [
    "gbc.fit(rus_train[user_features], y_rus)\n",
    "\n",
    "user_preds = gbc.predict(x_test[user_features])\n",
    "\n",
    "report(y_test, user_preds, 'User features')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 8.4.3) Decision Threshold\n",
    "- Generate classification report for:\n",
    "    - base gradient boosting model\n",
    "    - tuned gradient boosting model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For the base gradient boosting model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-3 {color: black;}#sk-container-id-3 pre{padding: 0;}#sk-container-id-3 div.sk-toggleable {background-color: white;}#sk-container-id-3 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-3 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-3 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-3 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-3 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-3 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-3 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-3 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-3 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-3 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-3 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-3 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-3 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-3 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-3 div.sk-item {position: relative;z-index: 1;}#sk-container-id-3 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-3 div.sk-item::before, #sk-container-id-3 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-3 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-3 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-3 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-3 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-3 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-3 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-3 div.sk-label-container {text-align: center;}#sk-container-id-3 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-3 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-3\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GradientBoostingClassifier(random_state=45)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-11\" type=\"checkbox\" checked><label for=\"sk-estimator-id-11\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GradientBoostingClassifier</label><div class=\"sk-toggleable__content\"><pre>GradientBoostingClassifier(random_state=45)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "GradientBoostingClassifier(random_state=45)"
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "base_gbc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [],
   "source": [
    "#threshold\n",
    "base_gbc.fit(rus_train[user_features], y_rus)\n",
    "\n",
    "threshold = base_gbc.predict_proba(x_test[user_features])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.45374963, 0.54625037],\n",
       "       [0.6192474 , 0.3807526 ],\n",
       "       [0.42472148, 0.57527852],\n",
       "       ...,\n",
       "       [0.42464319, 0.57535681],\n",
       "       [0.44417899, 0.55582101],\n",
       "       [0.55228195, 0.44771805]])"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "threshold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [],
   "source": [
    "threshold = threshold[:, 1]\n",
    "threshold_probs = pd.Series(threshold)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7909442500117208"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "threshold_probs.max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Threshold=0.548, F-Score=0.16081\n"
     ]
    }
   ],
   "source": [
    "# apply threshold to the probabilities to create labels\n",
    "def to_labels(probs, threshold):\n",
    " return (probs >= threshold).astype('int')\n",
    " \n",
    "# define thresholds\n",
    "thresholds = np.arange(0, 1, 0.001)\n",
    "# evaluate each threshold\n",
    "scores = [f1_score(y_test, to_labels(threshold_probs, t)) for t in thresholds]\n",
    "# get best threshold\n",
    "ix = np.argmax(scores)\n",
    "print('Threshold=%.3f, F-Score=%.5f' % (thresholds[ix], scores[ix]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       0\n",
       "1       0\n",
       "2       1\n",
       "3       1\n",
       "4       0\n",
       "       ..\n",
       "2925    1\n",
       "2926    0\n",
       "2927    1\n",
       "2928    1\n",
       "2929    0\n",
       "Length: 2930, dtype: int32"
      ]
     },
     "execution_count": 126,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "threshold_preds = pd.Series(np.where(threshold_probs < 0.548, 0, 1))\n",
    "threshold_preds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Threshold predictions Confusion Matrix\n",
      "[[1752  994]\n",
      " [  81  103]]\n",
      "\n",
      "\n",
      "Threshold predictions Classification Report\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.64      0.77      2746\n",
      "           1       0.09      0.56      0.16       184\n",
      "\n",
      "    accuracy                           0.63      2930\n",
      "   macro avg       0.52      0.60      0.46      2930\n",
      "weighted avg       0.90      0.63      0.73      2930\n",
      "\n"
     ]
    }
   ],
   "source": [
    "report(y_test, threshold_preds, 'Threshold predictions')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For the hyperparemter tuned gradient boosting model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [],
   "source": [
    "new_gbc = GradientBoostingClassifier(n_estimators=16, random_state=45, subsample=0.6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Updated GBC Confusion Matrix\n",
      "[[1317 1429]\n",
      " [  53  131]]\n",
      "\n",
      "\n",
      "Updated GBC Classification Report\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.48      0.64      2746\n",
      "           1       0.08      0.71      0.15       184\n",
      "\n",
      "    accuracy                           0.49      2930\n",
      "   macro avg       0.52      0.60      0.40      2930\n",
      "weighted avg       0.91      0.49      0.61      2930\n",
      "\n"
     ]
    }
   ],
   "source": [
    "new_gbc.fit(rus_train[user_features], y_rus)\n",
    "\n",
    "report(y_test, new_gbc.predict(x_test[user_features]), \"Updated GBC\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Generate decision threshold for optimized model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Threshold=0.533, F-Score=0.16085\n"
     ]
    }
   ],
   "source": [
    "new_thresh = new_gbc.predict_proba(x_test[user_features])[:, 1]\n",
    "new_thresh = pd.Series(new_thresh)\n",
    "\n",
    "# define thresholds\n",
    "thresholds = np.arange(0, 1, 0.001)\n",
    "# evaluate each threshold\n",
    "scores = [f1_score(y_test, to_labels(new_thresh, t)) for t in thresholds]\n",
    "# get best threshold\n",
    "ix = np.argmax(scores)\n",
    "print('Threshold=%.3f, F-Score=%.5f' % (thresholds[ix], scores[ix]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [],
   "source": [
    "new_thresh = np.where(new_thresh < 0.533, 0, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Updated threshold Confusion Matrix\n",
      "[[1638 1108]\n",
      " [  71  113]]\n",
      "\n",
      "\n",
      "Updated threshold Classification Report\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.60      0.74      2746\n",
      "           1       0.09      0.61      0.16       184\n",
      "\n",
      "    accuracy                           0.60      2930\n",
      "   macro avg       0.53      0.61      0.45      2930\n",
      "weighted avg       0.90      0.60      0.70      2930\n",
      "\n"
     ]
    }
   ],
   "source": [
    "report(y_test, new_thresh, \"Updated threshold\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "***Insight***\n",
    "\n",
    "- The decision threshold increased performance for both models.\n",
    "- The threshold implementation with the optimized model provided the best results."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# CONCLUSION\n",
    "\n",
    "- Due to the imbalanced target variables, sampling techniques had to be implemented\n",
    "- The RandomUnderSampler provided the best results for all predictive models.\n",
    "- The random_state was set to 45 to ensure even distribution of data\n",
    "- The GradientBoostingClassifier was the best performing model, with a few hyperparameters tuned.\n",
    "- These hyperparameters were: n_estimators=16 and subsample=0.6\n",
    "- The remaining parameters were set to their default values.\n",
    "- The decision threshold was set to 0.533."
   ]
  }
 ],
 "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": "3.8.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}