FInal  Solution.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Reading File\n",
    "1)Convert the date into datetimes format"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['reservation_id', 'booking_date', 'checkin_date', 'checkout_date',\n",
       "       'channel_code', 'main_product_code', 'numberofadults',\n",
       "       'numberofchildren', 'persontravellingid', 'resort_region_code',\n",
       "       'resort_type_code', 'room_type_booked_code', 'roomnights',\n",
       "       'season_holidayed_code', 'state_code_residence', 'state_code_resort',\n",
       "       'total_pax', 'member_age_buckets', 'booking_type_code', 'memberid',\n",
       "       'cluster_code', 'reservationstatusid_code', 'resort_id',\n",
       "       'amount_spent_per_room_night_scaled'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "df = pd.read_csv(\"train_5CLrC8b/train.csv\")\n",
    "test = pd.read_csv('test.csv')\n",
    "import pandas as pd\n",
    "pd.set_option('display.max_columns', 500)\n",
    "\n",
    "df.checkin_date = pd.to_datetime(df.checkin_date, format=\"%d/%m/%y\")\n",
    "\n",
    "df.checkout_date = pd.to_datetime(df.checkout_date, format=\"%d/%m/%y\")\n",
    "\n",
    "df.booking_date= pd.to_datetime(df.booking_date, format=\"%d/%m/%y\")\n",
    "\n",
    "test.checkin_date = pd.to_datetime(test.checkin_date, format=\"%d/%m/%y\")\n",
    "\n",
    "test.checkout_date = pd.to_datetime(test.checkout_date, format=\"%d/%m/%y\")\n",
    "\n",
    "test.booking_date= pd.to_datetime(test.booking_date, format=\"%d/%m/%y\")\n",
    "\n",
    "df.columns\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Making DateTime Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(341424, 24)\n",
      "(341424, 31)\n",
      "(341424, 38)\n",
      "(146765, 23)\n",
      "(146765, 30)\n",
      "(146765, 37)\n"
     ]
    }
   ],
   "source": [
    "\n",
    "def dt_feat(train,date_col):\n",
    "    print(train.shape)\n",
    "    train[date_col+'week_day'] = train[date_col].dt.dayofweek\n",
    "    train[date_col+'month'] = train[date_col].dt.month\n",
    "    train[date_col+'is_month_end'] = train[date_col].dt.is_month_end\n",
    "    train[date_col+'dayofyear'] = train[date_col].dt.dayofyear\n",
    "    train[date_col+'day'] = train[date_col].dt.day\n",
    "    train[date_col+'year'] = train[date_col].dt.year\n",
    "    train[date_col+'weekyear'] = train[date_col].dt.weekofyear\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "    return train\n",
    "\n",
    "train1 =dt_feat(df,'checkin_date')\n",
    "\n",
    "train1=dt_feat(train1,'checkout_date')\n",
    "\n",
    "train1=dt_feat(train1,'booking_date')\n",
    "df = train1.copy()\n",
    "\n",
    "train1 =dt_feat(test,'checkin_date')\n",
    "\n",
    "train1=dt_feat(train1,'checkout_date')\n",
    "\n",
    "train1=dt_feat(train1,'booking_date')\n",
    "test = train1.copy()\n",
    "\n",
    "df.columns\n",
    "\n",
    "cat_cols = [f for f in df.columns if (df[f].dtype == 'object' and f not in ['reservation_id'])]\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = df.copy()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### LabelEncodig the categorical Variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "for col in cat_cols:\n",
    "    lbl = LabelEncoder()\n",
    "    lbl.fit(list(train[col].values.astype('str')) + list(test[col].values.astype('str')))\n",
    "    train[col] = lbl.transform(list(train[col].values.astype('str')))\n",
    "    test[col] = lbl.transform(list(test[col].values.astype('str')))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Making very Important Features like \n",
    "1 Number of days stayed<br>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "train['in_book']=train.checkin_date-train.booking_date\n",
    "test['in_book']=test.checkin_date-test.booking_date\n",
    "train['in_out'] = train.checkout_date-train.checkin_date\n",
    "test['in_out'] = test.checkout_date-test.checkin_date\n",
    "def day_get(x):\n",
    "    return x.days\n",
    "train.in_out = train.in_out.apply(day_get)\n",
    "test.in_out = test.in_out.apply(day_get)\n",
    "train.in_book = train.in_book.apply(day_get)\n",
    "test.in_book = test.in_book.apply(day_get)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_bkp=train.copy()\n",
    "test_bkp=test.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(341424, 47)\n",
      "(146765, 46)\n"
     ]
    }
   ],
   "source": [
    "print(train.shape)\n",
    "print(test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Making Groupby feat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:1: FutureWarning: Sorting because non-concatenation axis is not aligned. A future version\n",
      "of pandas will change to not sort by default.\n",
      "\n",
      "To accept the future behavior, pass 'sort=False'.\n",
      "\n",
      "To retain the current behavior and silence the warning, pass 'sort=True'.\n",
      "\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n"
     ]
    }
   ],
   "source": [
    "merge = pd.concat([train_bkp,test_bkp])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','memberid']).reservation_id.count()).reset_index(),suffixes=('','res_mem'),on=['resort_id','memberid'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id']).reservation_id.count()).reset_index(),suffixes=('','_x_res'),on='resort_id',how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','checkout_dateyear','checkout_datemonth']).reservation_id.count()).reset_index(),suffixes=('','_res_month'),on=['resort_id','checkout_dateyear','checkout_datemonth'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid']).reservation_id.count()).reset_index(),suffixes=('','mem'),on=['memberid'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid','checkout_dateyear']).reservation_id.count()).reset_index(),suffixes=('','mem_year'),on=['memberid','checkout_dateyear'],how='left')\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id']).roomnights.median()).reset_index(),suffixes=('','res_amount_median'),on=['resort_id'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id']).roomnights.max()).reset_index(),suffixes=('','res_amount_max'),on=['resort_id'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id']).roomnights.min()).reset_index(),suffixes=('','res_amount_min'),on=['resort_id'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id']).roomnights.std()).reset_index(),suffixes=('','res_amount_std'),on=['resort_id'],how='left')\n",
    "\n",
    "#0.9651884690956651\n",
    "\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','memberid']).roomnights.median()).reset_index(),suffixes=('','res_mem_amount_median'),on=['resort_id','memberid'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','memberid']).roomnights.max()).reset_index(),suffixes=('','res_mem_amount_max'),on=['resort_id','memberid'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','memberid']).roomnights.min()).reset_index(),suffixes=('','res_mem_amount_min'),on=['resort_id','memberid'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','memberid']).roomnights.std()).reset_index(),suffixes=('','res_mem_amount_std'),on=['resort_id','memberid'],how='left')\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid']).roomnights.median()).reset_index(),suffixes=('','mem_amount_median'),on=['memberid'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid']).roomnights.max()).reset_index(),suffixes=('','mem_amount_max'),on=['memberid'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid']).roomnights.min()).reset_index(),suffixes=('','mem_amount_min'),on=['memberid'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid']).roomnights.std()).reset_index(),suffixes=('','mem_amount_std'),on=['memberid'],how='left')\n",
    "\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','room_type_booked_code']).reservation_id.count()).reset_index(),suffixes=('','_res_type'),on=['resort_id','room_type_booked_code'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','checkout_datemonth','room_type_booked_code']).reservation_id.count()).reset_index(),suffixes=('','_res_code_month_x'),on=['resort_id','checkout_datemonth','room_type_booked_code'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','main_product_code']).reservation_id.count()).reset_index(),suffixes=('','_res_main_code'),on=['resort_id','main_product_code'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['channel_code','checkout_dateyear','checkout_datemonth']).reservation_id.count()).reset_index(),suffixes=('','channel_check_yearmonth'),on=['channel_code','checkout_dateyear','checkout_datemonth'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['checkout_dateyear','checkout_datemonth']).reservation_id.count()).reset_index(),suffixes=('','check_Date_month'),on=['checkout_dateyear','checkout_datemonth'],how='left')\n",
    "\n",
    "#0.9651884690956651\n",
    "\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','memberid']).in_out.median()).reset_index(),suffixes=('','res_meminoutt_median'),on=['resort_id','memberid'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','memberid']).in_out.max()).reset_index(),suffixes=('','res_mem_inout_max'),on=['resort_id','memberid'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','memberid']).in_out.min()).reset_index(),suffixes=('','res_mem_inout_min'),on=['resort_id','memberid'],how='left')\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','memberid']).in_out.std()).reset_index(),suffixes=('','res_meminoutt_std'),on=['resort_id','memberid'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid']).in_out.median()).reset_index(),suffixes=('','mem_inout_median'),on=['memberid'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid']).in_out.max()).reset_index(),suffixes=('','mem_inout_max'),on=['memberid'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid']).in_out.min()).reset_index(),suffixes=('','mem_inout_min'),on=['memberid'],how='left')\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid']).in_out.std()).reset_index(),suffixes=('','mem_inout_std'),on=['memberid'],how='left')\n",
    "\n",
    "\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid']).total_pax.median()).reset_index(),suffixes=('','mem_inout_median_pax'),on=['memberid'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid']).total_pax.max()).reset_index(),suffixes=('','mem_inout_max_pax'),on=['memberid'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid']).total_pax.min()).reset_index(),suffixes=('','mem_inout_min_pax'),on=['memberid'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid']).total_pax.std()).reset_index(),suffixes=('','mem_inout_std_pax'),on=['memberid'],how='left')\n",
    "\n",
    "\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id']).total_pax.median()).reset_index(),suffixes=('','res_inout_median_pax'),on=['resort_id'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id']).total_pax.max()).reset_index(),suffixes=('','res_inout_max_pax'),on=['resort_id'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id']).total_pax.min()).reset_index(),suffixes=('','res_inout_min_pax'),on=['resort_id'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id']).total_pax.std()).reset_index(),suffixes=('','res_inout_std_pax'),on=['resort_id'],how='left')\n",
    "\n",
    "\n",
    "\n",
    "# merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','memberid']).total_pax.median()).reset_index(),suffixes=('','mem_res_inout_median_pax'),on=['resort_id','memberid'],how='left')\n",
    "# merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','memberid']).total_pax.max()).reset_index(),suffixes=('','mem_res_inout_max_pax'),on=['resort_id','memberid'],how='left')\n",
    "# merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','memberid']).total_pax.min()).reset_index(),suffixes=('','mem_res_inout_min_pax'),on=['resort_id','memberid'],how='left')\n",
    "# merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','memberid']).total_pax.std()).reset_index(),suffixes=('','mem_res_inout_std_pax'),on=['resort_id','memberid'],how='left')\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','main_product_code']).reservation_id.count()).reset_index(),suffixes=('','res_memmain_product_code'),on=['resort_id','main_product_code'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['main_product_code']).reservation_id.count()).reset_index(),suffixes=('','_x_resmain_product_code'),on='main_product_code',how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid','resort_id','main_product_code']).reservation_id.count()).reset_index(),suffixes=('','memjknknk'),on=['memberid','resort_id','main_product_code'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid','main_product_code']).reservation_id.count()).reset_index(),suffixes=('','mem_yearmain_product_code'),on=['memberid','main_product_code'],how='left')\n",
    "\n",
    "\n",
    "\n",
    "#merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','checkout_dateyear','checkin_dateweekyear']).total_pax.count()).reset_index(),suffixes=('','res_inout_std_pax','week_year'),on=['resort_id','checkout_dateyear','checkin_dateweekyear'],how='left')\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','checkout_dateyear','checkin_dateweekyear']).reservation_id.count()).reset_index(),suffixes=('','_res_month_week_year'),on=['resort_id','checkout_dateyear','checkin_dateweekyear'],how='left')\n",
    "\n",
    "\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['memberid','checkout_dateyear','checkin_dateweekyear']).reservation_id.count()).reset_index(),suffixes=('','_res_month_week_year_member'),on=['memberid','checkout_dateyear','checkin_dateweekyear'],how='left')\n",
    "\n",
    "\n",
    "\n",
    "import datetime\n",
    "\n",
    "merge['booking_dayes'] = (datetime.datetime.today()-merge['booking_date']).dt.days\n",
    "merge['checkin_dayes'] = (datetime.datetime.today()-merge['checkin_date']).dt.days\n",
    "merge['checkout_dayes'] = (datetime.datetime.today()-merge['checkout_date']).dt.days\n",
    "\n",
    "\n",
    "\n",
    "merge['rooms_booked_per_night'] = merge.in_out/merge.roomnights\n",
    "merge['rooms_bookinf_per_night'] = merge.in_book/merge.roomnights\n",
    "\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['state_code_residence']).total_pax.median()).reset_index(),suffixes=('','mem_inout_median_pax_state_code'),on=['state_code_residence'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['state_code_residence']).total_pax.max()).reset_index(),suffixes=('','mem_inout_max_pax_state_code'),on=['state_code_residence'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['state_code_residence']).total_pax.min()).reset_index(),suffixes=('','mem_inout_min_pax_state_code'),on=['state_code_residence'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['state_code_residence']).total_pax.std()).reset_index(),suffixes=('','mem_inout_std_pax_state_code'),on=['state_code_residence'],how='left')\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','state_code_residence']).roomnights.median()).reset_index(),suffixes=('','res_mem_amount_medianstate_code_resid'),on=['resort_id','state_code_residence'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','state_code_residence']).roomnights.max()).reset_index(),suffixes=('','res_mem_amount_maxstate_code_resid'),on=['resort_id','state_code_residence'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','state_code_residence']).roomnights.min()).reset_index(),suffixes=('','res_mem_amount_minstate_code_resid'),on=['resort_id','state_code_residence'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','state_code_residence']).roomnights.std()).reset_index(),suffixes=('','res_mem_amount_stdstate_code_resid'),on=['resort_id','state_code_residence'],how='left')\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['state_code_residence']).roomnights.median()).reset_index(),suffixes=('','mem_amount_medianstate_code_resid'),on=['state_code_residence'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['state_code_residence']).roomnights.max()).reset_index(),suffixes=('','mem_amount_maxstate_code_resid'),on=['state_code_residence'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['state_code_residence']).roomnights.min()).reset_index(),suffixes=('','mem_amount_minstate_code_resid'),on=['state_code_residence'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['state_code_residence']).roomnights.std()).reset_index(),suffixes=('','mem_amount_stdstate_code_resid'),on=['state_code_residence'],how='left')\n",
    "\n",
    "\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','state_code_residence']).in_out.median()).reset_index(),suffixes=('','res_meminoutt_medianstate_code_resid'),on=['resort_id','state_code_residence'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','state_code_residence']).in_out.max()).reset_index(),suffixes=('','res_mem_inout_maxstate_code_resid'),on=['resort_id','state_code_residence'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','state_code_residence']).in_out.min()).reset_index(),suffixes=('','res_mem_inout_minstate_code_resid'),on=['resort_id','state_code_residence'],how='left')\n",
    "\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','state_code_residence']).in_out.std()).reset_index(),suffixes=('','res_meminoutt_stdstate_code_resid'),on=['resort_id','state_code_residence'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['state_code_residence']).in_out.median()).reset_index(),suffixes=('','mem_inout_medianstate_code_resid'),on=['state_code_residence'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['state_code_residence']).in_out.max()).reset_index(),suffixes=('','mem_inout_maxstate_code_resid'),on=['state_code_residence'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['state_code_residence']).in_out.min()).reset_index(),suffixes=('','mem_inout_minstate_code_resid'),on=['state_code_residence'],how='left')\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##  Difference between time when resort was previously/next booked <br>\n",
    "1) When person A books the time of last past booking of that resort\n",
    "2) When person A books the time when next booking will made. and so on....."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/lightgbm/__init__.py:46: UserWarning: Starting from version 2.2.1, the library file in distribution wheels for macOS is built by the Apple Clang (Xcode_9.4.1) compiler.\n",
      "This means that in case of installing LightGBM from PyPI via the ``pip install lightgbm`` command, you don't need to install the gcc compiler anymore.\n",
      "Instead of that, you need to install the OpenMP library, which is required for running LightGBM on the system with the Apple Clang compiler.\n",
      "You can install the OpenMP library by the following command: ``brew install libomp``.\n",
      "  \"You can install the OpenMP library by the following command: ``brew install libomp``.\", UserWarning)\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import time\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "import lightgbm as lgb\n",
    "import gc\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "\n",
    "predictors=[]\n",
    "\n",
    "def do_prev_Click( df,agg_suffix='prevClick', agg_type='float32'):\n",
    "\n",
    "    print(f\">> \\nExtracting {agg_suffix} time calculation features...\\n\")\n",
    "    \n",
    "    GROUP_BY_NEXT_CLICKS = [\n",
    "    \n",
    "     {'groupby': ['resort_id', 'memberid']},\n",
    "     {'groupby': ['resort_id']},\n",
    "        \n",
    "    ]\n",
    "\n",
    "    # Calculate the time to next click for each group\n",
    "    for spec in GROUP_BY_NEXT_CLICKS:\n",
    "    \n",
    "       # Name of new feature\n",
    "        new_feature = '{}_{}'.format('_'.join(spec['groupby']),agg_suffix)    \n",
    "    \n",
    "        # Unique list of features to select\n",
    "        all_features = spec['groupby'] + ['booking_date']\n",
    "\n",
    "        # Run calculation\n",
    "        print(f\">> Grouping by {spec['groupby']}, and saving time to {agg_suffix} in: {new_feature}\")\n",
    "        df[new_feature] = (df.booking_date - df[all_features].groupby(spec[\n",
    "                'groupby']).booking_date.shift(+1) ).dt.days.astype(agg_type)\n",
    "        \n",
    "        predictors.append(new_feature)\n",
    "        gc.collect()\n",
    "    return (df)    \n",
    "\n",
    "\n",
    "def do_next_Click( df,agg_suffix='nextClick', agg_type='float32'):\n",
    "    \n",
    "    print(f\">> \\nExtracting {agg_suffix} time calculation features...\\n\")\n",
    "    \n",
    "    GROUP_BY_NEXT_CLICKS = [\n",
    "    \n",
    "    # V3\n",
    "   \n",
    "     {'groupby': ['resort_id', 'memberid']},\n",
    "     {'groupby': ['resort_id']},\n",
    "    # {'groupby': ['memberid']}\n",
    "        ]\n",
    "\n",
    "    # Calculate the time to next click for each group\n",
    "    for spec in GROUP_BY_NEXT_CLICKS:\n",
    "    \n",
    "       # Name of new feature\n",
    "        new_feature = '{}_{}'.format('_'.join(spec['groupby']),agg_suffix)    \n",
    "    \n",
    "        # Unique list of features to select\n",
    "        all_features = spec['groupby'] + ['booking_date']\n",
    "\n",
    "        # Run calculation\n",
    "        print(f\">> Grouping by {spec['groupby']}, and saving time to {agg_suffix} in: {new_feature}\")\n",
    "        df[new_feature] = (df[all_features].groupby(spec[\n",
    "            'groupby']).booking_date.shift(-1) - df.booking_date).dt.days.astype(agg_type)\n",
    "        \n",
    "        #predictors.webpage_idend(new_feature)\n",
    "        gc.collect()\n",
    "    return (df)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">> \n",
      "Extracting prevClick time calculation features...\n",
      "\n",
      ">> Grouping by ['resort_id', 'memberid'], and saving time to prevClick in: resort_id_memberid_prevClick\n",
      ">> Grouping by ['resort_id'], and saving time to prevClick in: resort_id_prevClick\n"
     ]
    }
   ],
   "source": [
    "merge = do_prev_Click(merge)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">> \n",
      "Extracting nextClick time calculation features...\n",
      "\n",
      ">> Grouping by ['resort_id', 'memberid'], and saving time to nextClick in: resort_id_memberid_nextClick\n",
      ">> Grouping by ['resort_id'], and saving time to nextClick in: resort_id_nextClick\n"
     ]
    }
   ],
   "source": [
    "merge = do_next_Click(merge)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(488189, 119)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merge.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Making the categories of columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "member_age_buckets 10\n",
      "memberid 101327\n",
      "resort_id 32\n",
      "cluster_code 6\n",
      "reservationstatusid_code 4\n",
      "resort_id 32\n",
      "channel_code 3\n",
      "main_product_code 5\n",
      "persontravellingid 6\n",
      "resort_region_code 3\n",
      "resort_type_code 7\n",
      "room_type_booked_code 6\n",
      "season_holidayed_code 4\n",
      "state_code_residence 37\n",
      "state_code_resort 11\n",
      "member_age_buckets 10\n",
      "booking_type_code 2\n"
     ]
    }
   ],
   "source": [
    "categor_col =['member_age_buckets','memberid','resort_id',\n",
    " 'cluster_code',\n",
    " 'reservationstatusid_code',\n",
    " 'resort_id',\n",
    " 'channel_code',\n",
    " 'main_product_code',\n",
    " 'persontravellingid',\n",
    " 'resort_region_code',\n",
    " 'resort_type_code',\n",
    " 'room_type_booked_code',\n",
    " 'season_holidayed_code',\n",
    " 'state_code_residence',\n",
    " 'state_code_resort',\n",
    " 'member_age_buckets',\n",
    " 'booking_type_code']\n",
    "\n",
    "for i in categor_col:\n",
    "    print(i,train[i].nunique())\n",
    "\n",
    "for i in categor_col:\n",
    "    merge[i] = merge[i].astype('category')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['state_code_residence']).reservation_id.count()).reset_index(),suffixes=('','_x_res_mem_state_score'),on=['state_code_residence'],how='left')\n",
    "merge = merge.merge(pd.DataFrame(merge.groupby(by=['resort_id','state_code_residence']).reservation_id.count()).reset_index(),suffixes=('','_x_res_mem_state_scoreresort_id'),on=['resort_id','state_code_residence'],how='left')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Getting train and Test out of Merged File"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = merge.loc[merge.amount_spent_per_room_night_scaled.notnull()]\n",
    "test = merge.loc[merge.amount_spent_per_room_night_scaled.isnull()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "X= train.drop(['reservation_id','booking_date','checkin_date','checkout_date','amount_spent_per_room_night_scaled','memberid'],axis=1)\n",
    "\n",
    "y=train.amount_spent_per_room_night_scaled"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dividing on the Basis of \n",
    "1)Random<br>\n",
    "2)Unique MemberIds<br>\n",
    "I cross - validated my model on both the cross validation technniques which resulted in very robust Model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "train_x,test_x,train_y,test_y = train_test_split(X,y,test_size=0.2,random_state=3)\n",
    "\n",
    "from lightgbm import LGBMRegressor\n",
    "\n",
    "clf=LGBMRegressor(n_estimators=100)\n",
    "\n",
    "from catboost import CatBoostRegressor\n",
    "\n",
    "#clf=CatBoostRegressor()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf=LGBMRegressor(n_estimators=100,learning_rate=.12,subsample=1,reg_alpha=25,min_child_weight=49)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LGBMRegressor(boosting_type='gbdt', class_weight=None, colsample_bytree=1.0,\n",
       "       importance_type='split', learning_rate=0.12, max_depth=-1,\n",
       "       min_child_samples=20, min_child_weight=49, min_split_gain=0.0,\n",
       "       n_estimators=100, n_jobs=-1, num_leaves=31, objective=None,\n",
       "       random_state=None, reg_alpha=25, reg_lambda=0.0, silent=True,\n",
       "       subsample=1, subsample_for_bin=200000, subsample_freq=0)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.fit(train_x,train_y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### RMSE score comes out to be ~ 96.1 for 115 features on default params"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9611527041325448\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "import numpy as np\n",
    "\n",
    "print(np.sqrt(mean_squared_error(test_y,clf.predict(test_x))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Further tuning for LGB model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training until validation scores don't improve for 100 rounds.\n",
      "[100]\tvalid_0's l2: 0.924557\tvalid_0's rmse: 0.961539\n",
      "[200]\tvalid_0's l2: 0.923346\tvalid_0's rmse: 0.960909\n",
      "[300]\tvalid_0's l2: 0.923464\tvalid_0's rmse: 0.96097\n",
      "Early stopping, best iteration is:\n",
      "[208]\tvalid_0's l2: 0.923284\tvalid_0's rmse: 0.960877\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LGBMRegressor(boosting_type='gbdt', class_weight=None, colsample_bytree=1.0,\n",
       "       importance_type='split', learning_rate=0.12, max_depth=-1,\n",
       "       min_child_samples=20, min_child_weight=1, min_split_gain=0.0,\n",
       "       n_estimators=1000, n_jobs=-1, num_leaves=31, objective=None,\n",
       "       random_state=None, reg_alpha=5, reg_lambda=0.0, silent=True,\n",
       "       subsample=1, subsample_for_bin=200000, subsample_freq=0)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf=LGBMRegressor(n_estimators=1000,learning_rate=.06,subsample=0.9,reg_alpha=25,min_child_weight=49)\n",
    "clf=LGBMRegressor()\n",
    "clf=LGBMRegressor(n_estimators=1000,learning_rate=.12,subsample=1,reg_alpha=5,min_child_weight=1)\n",
    "\n",
    "\n",
    "clf.fit(train_x,train_y,eval_set=[(test_x,test_y)],eval_metric='rmse',verbose=100,early_stopping_rounds=100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Testing Cross Validation Startegy-2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_x = train.iloc[:137148].drop(['reservation_id','booking_date','checkin_date','checkout_date','amount_spent_per_room_night_scaled','memberid'],axis=1)\n",
    "train_x = train.iloc[137148:].drop(['reservation_id','booking_date','checkin_date','checkout_date','amount_spent_per_room_night_scaled','memberid'],axis=1)\n",
    "test_y = train.iloc[:137148].amount_spent_per_room_night_scaled\n",
    "train_y = train.iloc[137148:].amount_spent_per_room_night_scaled"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training until validation scores don't improve for 100 rounds.\n",
      "[100]\tvalid_0's l2: 0.887848\tvalid_0's rmse: 0.942257\n",
      "[200]\tvalid_0's l2: 0.881577\tvalid_0's rmse: 0.938924\n",
      "[300]\tvalid_0's l2: 0.880875\tvalid_0's rmse: 0.93855\n",
      "[400]\tvalid_0's l2: 0.880975\tvalid_0's rmse: 0.938603\n",
      "Early stopping, best iteration is:\n",
      "[350]\tvalid_0's l2: 0.880857\tvalid_0's rmse: 0.93854\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LGBMRegressor(boosting_type='gbdt', class_weight=None, colsample_bytree=1.0,\n",
       "       importance_type='split', learning_rate=0.05, max_depth=-1,\n",
       "       min_child_samples=20, min_child_weight=49, min_split_gain=0.0,\n",
       "       n_estimators=717, n_jobs=-1, num_leaves=31, objective=None,\n",
       "       random_state=None, reg_alpha=25, reg_lambda=0.0, silent=True,\n",
       "       subsample=0.9, subsample_for_bin=200000, subsample_freq=0)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf=LGBMRegressor(n_estimators=717,learning_rate=.05,subsample=0.9,reg_alpha=25,min_child_weight=49)\n",
    "clf.fit(train_x,train_y,eval_set=[(test_x,test_y)],eval_metric='rmse',verbose=100,early_stopping_rounds=100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Fitting the tuned model on Final training Set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LGBMRegressor(boosting_type='gbdt', class_weight=None, colsample_bytree=1.0,\n",
       "       importance_type='split', learning_rate=0.05, max_depth=-1,\n",
       "       min_child_samples=20, min_child_weight=49, min_split_gain=0.0,\n",
       "       n_estimators=717, n_jobs=-1, num_leaves=31, objective=None,\n",
       "       random_state=None, reg_alpha=25, reg_lambda=0.0, silent=True,\n",
       "       subsample=0.9, subsample_for_bin=200000, subsample_freq=0)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf=LGBMRegressor(n_estimators=717,learning_rate=.05,subsample=0.9,reg_alpha=25,min_child_weight=49)\n",
    "clf.fit(X,y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Prediction 1 : Single Model LGB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_test_lgb_single_model = clf.predict(test[X.columns])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([8.44966797, 7.40593963, 7.55029077, ..., 8.63509013, 8.36026112,\n",
       "       7.47302073])"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_test_lgb_single_model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Single Best Light GBM Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "sub = pd.read_csv('sample_submission.csv')\n",
    "\n",
    "sub.amount_spent_per_room_night_scaled=pred_test_lgb_single_model\n",
    "sub.to_csv('single_lgb_model.csv',index=None)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Fitting Single Best  Catboost Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Run Next Cell\n"
     ]
    }
   ],
   "source": [
    "try:\n",
    "    X.drop('target',axis=1,inplace=True)\n",
    "except:\n",
    "    print('Run Next Cell')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "MAX_ROUNDS = 4000\n",
    "OPTIMIZE_ROUNDS = True\n",
    "LEARNING_RATE = 0.05\n",
    "model = CatBoostRegressor(verbose=100,\n",
    "    learning_rate=LEARNING_RATE, \n",
    "    l2_leaf_reg = 8, \n",
    "    iterations = MAX_ROUNDS,\n",
    "#    verbose = True,\n",
    "    loss_function='RMSE',\n",
    ")\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### validating and finding the optimal number of iterations for the Catboost Model\n",
    "###### ~1231 is optimal for Catboost model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0:\tlearn: 7.4396353\ttest: 7.3779558\tbest: 7.3779558 (0)\ttotal: 93.6ms\tremaining: 6m 14s\n",
      "100:\tlearn: 1.0038490\ttest: 0.9597464\tbest: 0.9597464 (100)\ttotal: 8.34s\tremaining: 5m 21s\n",
      "200:\tlearn: 0.9891052\ttest: 0.9458932\tbest: 0.9458932 (200)\ttotal: 16.3s\tremaining: 5m 8s\n",
      "300:\tlearn: 0.9843892\ttest: 0.9430126\tbest: 0.9430126 (300)\ttotal: 23.8s\tremaining: 4m 52s\n",
      "400:\tlearn: 0.9811663\ttest: 0.9416773\tbest: 0.9416723 (399)\ttotal: 33.1s\tremaining: 4m 57s\n",
      "500:\tlearn: 0.9787529\ttest: 0.9411042\tbest: 0.9411042 (500)\ttotal: 42.8s\tremaining: 4m 58s\n",
      "600:\tlearn: 0.9763226\ttest: 0.9405696\tbest: 0.9405696 (600)\ttotal: 50.6s\tremaining: 4m 46s\n",
      "700:\tlearn: 0.9744956\ttest: 0.9403179\tbest: 0.9403166 (699)\ttotal: 1m\tremaining: 4m 44s\n",
      "800:\tlearn: 0.9726677\ttest: 0.9401079\tbest: 0.9400628 (792)\ttotal: 1m 11s\tremaining: 4m 44s\n",
      "900:\tlearn: 0.9709539\ttest: 0.9400918\tbest: 0.9400370 (841)\ttotal: 1m 19s\tremaining: 4m 34s\n",
      "1000:\tlearn: 0.9692047\ttest: 0.9399658\tbest: 0.9399658 (1000)\ttotal: 1m 28s\tremaining: 4m 23s\n",
      "1100:\tlearn: 0.9674948\ttest: 0.9398849\tbest: 0.9398664 (1096)\ttotal: 1m 35s\tremaining: 4m 12s\n",
      "1200:\tlearn: 0.9658825\ttest: 0.9398069\tbest: 0.9398069 (1200)\ttotal: 1m 43s\tremaining: 4m\n",
      "1300:\tlearn: 0.9643021\ttest: 0.9398164\tbest: 0.9397821 (1231)\ttotal: 1m 50s\tremaining: 3m 48s\n",
      "1400:\tlearn: 0.9627637\ttest: 0.9398035\tbest: 0.9397821 (1231)\ttotal: 1m 57s\tremaining: 3m 37s\n",
      "1500:\tlearn: 0.9613045\ttest: 0.9398232\tbest: 0.9397821 (1231)\ttotal: 2m 5s\tremaining: 3m 28s\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-88-409775bf9924>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      1\u001b[0m fit_model = model.fit(train_x, train_y, \n\u001b[1;32m      2\u001b[0m                                \u001b[0meval_set\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtest_x\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mtest_y\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m                                \u001b[0muse_best_model\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      4\u001b[0m                              )\n",
      "\u001b[0;32m/anaconda3/lib/python3.6/site-packages/catboost/core.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, cat_features, sample_weight, baseline, use_best_model, eval_set, verbose, logging_level, plot, column_description, verbose_eval, metric_period, silent, early_stopping_rounds, save_snapshot, snapshot_file, snapshot_interval)\u001b[0m\n\u001b[1;32m   2547\u001b[0m                          \u001b[0muse_best_model\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0meval_set\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlogging_level\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mplot\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcolumn_description\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2548\u001b[0m                          \u001b[0mverbose_eval\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmetric_period\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msilent\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mearly_stopping_rounds\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2549\u001b[0;31m                          save_snapshot, snapshot_file, snapshot_interval)\n\u001b[0m\u001b[1;32m   2550\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2551\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mpredict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mntree_start\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mntree_end\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mthread_count\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/anaconda3/lib/python3.6/site-packages/catboost/core.py\u001b[0m in \u001b[0;36m_fit\u001b[0;34m(self, X, y, cat_features, pairs, sample_weight, group_id, group_weight, subgroup_id, pairs_weight, baseline, use_best_model, eval_set, verbose, logging_level, plot, column_description, verbose_eval, metric_period, silent, early_stopping_rounds, save_snapshot, snapshot_file, snapshot_interval)\u001b[0m\n\u001b[1;32m   1124\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1125\u001b[0m         \u001b[0;32mwith\u001b[0m \u001b[0mlog_fixup\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mplot_wrapper\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mplot\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_params\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1126\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_train\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_pool\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0meval_sets\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mparams\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mallow_clear_pool\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1127\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1128\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0;32mnot\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_object\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_has_leaf_weights_in_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mallow_clear_pool\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/anaconda3/lib/python3.6/site-packages/catboost/core.py\u001b[0m in \u001b[0;36m_train\u001b[0;34m(self, train_pool, test_pool, params, allow_clear_pool)\u001b[0m\n\u001b[1;32m    826\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    827\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_train\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtrain_pool\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtest_pool\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mparams\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mallow_clear_pool\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 828\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_object\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_train\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_pool\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtest_pool\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mparams\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mallow_clear_pool\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    829\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_set_trained_model_attributes\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    830\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m_catboost.pyx\u001b[0m in \u001b[0;36m_catboost._CatBoost._train\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;32m_catboost.pyx\u001b[0m in \u001b[0;36m_catboost._CatBoost._train\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "fit_model = model.fit(train_x, train_y, \n",
    "                               eval_set=[(test_x,test_y)],\n",
    "                               use_best_model=True\n",
    "                             )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Single Catboost Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0:\tlearn: 7.4155573\ttotal: 198ms\tremaining: 4m 2s\n",
      "100:\tlearn: 0.9842473\ttotal: 14.8s\tremaining: 2m 45s\n",
      "200:\tlearn: 0.9709161\ttotal: 30s\tremaining: 2m 33s\n",
      "300:\tlearn: 0.9668858\ttotal: 42.4s\tremaining: 2m 11s\n",
      "400:\tlearn: 0.9644321\ttotal: 54.6s\tremaining: 1m 52s\n",
      "500:\tlearn: 0.9624945\ttotal: 1m 6s\tremaining: 1m 36s\n",
      "600:\tlearn: 0.9608910\ttotal: 1m 19s\tremaining: 1m 23s\n",
      "700:\tlearn: 0.9595212\ttotal: 1m 32s\tremaining: 1m 9s\n",
      "800:\tlearn: 0.9582312\ttotal: 1m 45s\tremaining: 56.4s\n",
      "900:\tlearn: 0.9569728\ttotal: 1m 57s\tremaining: 43s\n",
      "1000:\tlearn: 0.9557936\ttotal: 2m 11s\tremaining: 30.1s\n",
      "1100:\tlearn: 0.9546238\ttotal: 2m 23s\tremaining: 17s\n",
      "1200:\tlearn: 0.9534425\ttotal: 2m 37s\tremaining: 3.92s\n",
      "1230:\tlearn: 0.9531279\ttotal: 2m 40s\tremaining: 0us\n"
     ]
    }
   ],
   "source": [
    "MAX_ROUNDS = 1231\n",
    "OPTIMIZE_ROUNDS = True\n",
    "LEARNING_RATE = 0.05\n",
    "model = CatBoostRegressor(verbose=100,\n",
    "    learning_rate=LEARNING_RATE, \n",
    "    l2_leaf_reg = 8, \n",
    "    iterations = MAX_ROUNDS,\n",
    "#    verbose = True,\n",
    "    loss_function='RMSE',\n",
    ")\n",
    "\n",
    "fit_model = model.fit(X,y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_catboost_model=model.predict(test[X.columns])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([8.38006997, 7.42584487, 7.52780001, ..., 8.53768314, 8.04933145,\n",
       "       7.29158108])"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_catboost_model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Making 5 FoldPredictions for Catboost Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 189,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Fold  0\n",
      "0:\tlearn: 7.4141836\ttest: 7.4214446\tbest: 7.4214446 (0)\ttotal: 110ms\tremaining: 3m 39s\n",
      "100:\tlearn: 0.9865132\ttest: 0.9896582\tbest: 0.9896582 (100)\ttotal: 9.74s\tremaining: 3m 3s\n",
      "200:\tlearn: 0.9718449\ttest: 0.9764771\tbest: 0.9764771 (200)\ttotal: 18.2s\tremaining: 2m 43s\n",
      "300:\tlearn: 0.9673192\ttest: 0.9730701\tbest: 0.9730701 (300)\ttotal: 26.1s\tremaining: 2m 27s\n",
      "400:\tlearn: 0.9648425\ttest: 0.9717942\tbest: 0.9717942 (400)\ttotal: 33.8s\tremaining: 2m 14s\n",
      "500:\tlearn: 0.9629156\ttest: 0.9709697\tbest: 0.9709697 (500)\ttotal: 41.5s\tremaining: 2m 4s\n",
      "600:\tlearn: 0.9613595\ttest: 0.9703420\tbest: 0.9703420 (600)\ttotal: 49s\tremaining: 1m 54s\n",
      "700:\tlearn: 0.9598764\ttest: 0.9699402\tbest: 0.9699391 (699)\ttotal: 56.8s\tremaining: 1m 45s\n",
      "800:\tlearn: 0.9584390\ttest: 0.9695649\tbest: 0.9695649 (800)\ttotal: 1m 4s\tremaining: 1m 36s\n",
      "900:\tlearn: 0.9572336\ttest: 0.9692826\tbest: 0.9692802 (895)\ttotal: 1m 12s\tremaining: 1m 28s\n",
      "1000:\tlearn: 0.9559975\ttest: 0.9690703\tbest: 0.9690703 (1000)\ttotal: 1m 19s\tremaining: 1m 19s\n",
      "1100:\tlearn: 0.9547470\ttest: 0.9688426\tbest: 0.9688426 (1100)\ttotal: 1m 27s\tremaining: 1m 11s\n",
      "1200:\tlearn: 0.9535969\ttest: 0.9686927\tbest: 0.9686878 (1196)\ttotal: 1m 35s\tremaining: 1m 3s\n",
      "1300:\tlearn: 0.9524989\ttest: 0.9685862\tbest: 0.9685818 (1295)\ttotal: 1m 44s\tremaining: 56.3s\n",
      "1400:\tlearn: 0.9512648\ttest: 0.9684531\tbest: 0.9684509 (1398)\ttotal: 1m 52s\tremaining: 48.3s\n",
      "1500:\tlearn: 0.9501490\ttest: 0.9683145\tbest: 0.9683133 (1497)\ttotal: 2m 1s\tremaining: 40.5s\n",
      "1600:\tlearn: 0.9489446\ttest: 0.9682404\tbest: 0.9682404 (1600)\ttotal: 2m 9s\tremaining: 32.4s\n",
      "1700:\tlearn: 0.9478363\ttest: 0.9681222\tbest: 0.9681222 (1700)\ttotal: 2m 17s\tremaining: 24.2s\n",
      "1800:\tlearn: 0.9467978\ttest: 0.9680257\tbest: 0.9680166 (1796)\ttotal: 2m 25s\tremaining: 16.1s\n",
      "1900:\tlearn: 0.9457274\ttest: 0.9679642\tbest: 0.9679642 (1900)\ttotal: 2m 34s\tremaining: 8.02s\n",
      "1999:\tlearn: 0.9446911\ttest: 0.9679231\tbest: 0.9679071 (1963)\ttotal: 2m 41s\tremaining: 0us\n",
      "\n",
      "bestTest = 0.967907128\n",
      "bestIteration = 1963\n",
      "\n",
      "Shrink model to first 1964 iterations.\n",
      "  N trees =  1964\n",
      "  Gini =  0.9679071268013583\n",
      "\n",
      "Fold  1\n",
      "0:\tlearn: 7.4161920\ttest: 7.4128362\tbest: 7.4128362 (0)\ttotal: 79.4ms\tremaining: 2m 38s\n",
      "100:\tlearn: 0.9868669\ttest: 0.9856278\tbest: 0.9856278 (100)\ttotal: 9.18s\tremaining: 2m 52s\n",
      "200:\tlearn: 0.9725338\ttest: 0.9731711\tbest: 0.9731711 (200)\ttotal: 17.6s\tremaining: 2m 37s\n",
      "300:\tlearn: 0.9678554\ttest: 0.9698327\tbest: 0.9698327 (300)\ttotal: 25.7s\tremaining: 2m 24s\n",
      "400:\tlearn: 0.9652255\ttest: 0.9684401\tbest: 0.9684393 (399)\ttotal: 33.4s\tremaining: 2m 13s\n",
      "500:\tlearn: 0.9633303\ttest: 0.9676468\tbest: 0.9676468 (500)\ttotal: 41s\tremaining: 2m 2s\n",
      "600:\tlearn: 0.9616318\ttest: 0.9670924\tbest: 0.9670923 (599)\ttotal: 50.8s\tremaining: 1m 58s\n",
      "700:\tlearn: 0.9601830\ttest: 0.9666539\tbest: 0.9666539 (700)\ttotal: 1m 1s\tremaining: 1m 53s\n",
      "800:\tlearn: 0.9589139\ttest: 0.9663933\tbest: 0.9663933 (800)\ttotal: 1m 9s\tremaining: 1m 43s\n",
      "900:\tlearn: 0.9577132\ttest: 0.9661361\tbest: 0.9661361 (900)\ttotal: 1m 16s\tremaining: 1m 33s\n",
      "1000:\tlearn: 0.9564062\ttest: 0.9658573\tbest: 0.9658514 (997)\ttotal: 1m 24s\tremaining: 1m 24s\n",
      "1100:\tlearn: 0.9551124\ttest: 0.9656415\tbest: 0.9656415 (1100)\ttotal: 1m 32s\tremaining: 1m 15s\n",
      "1200:\tlearn: 0.9538201\ttest: 0.9654364\tbest: 0.9654339 (1197)\ttotal: 1m 41s\tremaining: 1m 7s\n",
      "1300:\tlearn: 0.9526645\ttest: 0.9652599\tbest: 0.9652531 (1298)\ttotal: 1m 49s\tremaining: 58.8s\n",
      "1400:\tlearn: 0.9515410\ttest: 0.9651231\tbest: 0.9651231 (1400)\ttotal: 1m 57s\tremaining: 50.2s\n",
      "1500:\tlearn: 0.9504631\ttest: 0.9650212\tbest: 0.9650212 (1500)\ttotal: 2m 5s\tremaining: 41.6s\n",
      "1600:\tlearn: 0.9493618\ttest: 0.9649564\tbest: 0.9649542 (1599)\ttotal: 2m 13s\tremaining: 33.2s\n",
      "1700:\tlearn: 0.9482355\ttest: 0.9648417\tbest: 0.9648417 (1700)\ttotal: 2m 21s\tremaining: 24.8s\n",
      "1800:\tlearn: 0.9471905\ttest: 0.9647651\tbest: 0.9647651 (1800)\ttotal: 2m 28s\tremaining: 16.5s\n",
      "1900:\tlearn: 0.9460804\ttest: 0.9646622\tbest: 0.9646616 (1895)\ttotal: 2m 36s\tremaining: 8.16s\n",
      "1999:\tlearn: 0.9450385\ttest: 0.9645593\tbest: 0.9645593 (1999)\ttotal: 2m 44s\tremaining: 0us\n",
      "\n",
      "bestTest = 0.9645592588\n",
      "bestIteration = 1999\n",
      "\n",
      "  N trees =  2000\n",
      "  Gini =  0.9645592575286951\n",
      "\n",
      "Fold  2\n",
      "0:\tlearn: 7.4145663\ttest: 7.4197391\tbest: 7.4197391 (0)\ttotal: 77.1ms\tremaining: 2m 34s\n",
      "100:\tlearn: 0.9881923\ttest: 0.9806821\tbest: 0.9806821 (100)\ttotal: 9.12s\tremaining: 2m 51s\n",
      "200:\tlearn: 0.9738722\ttest: 0.9675383\tbest: 0.9675383 (200)\ttotal: 17.5s\tremaining: 2m 36s\n",
      "300:\tlearn: 0.9693059\ttest: 0.9643741\tbest: 0.9643741 (300)\ttotal: 25.5s\tremaining: 2m 24s\n",
      "400:\tlearn: 0.9664974\ttest: 0.9630026\tbest: 0.9630026 (400)\ttotal: 33.5s\tremaining: 2m 13s\n",
      "500:\tlearn: 0.9644676\ttest: 0.9622526\tbest: 0.9622526 (500)\ttotal: 41.3s\tremaining: 2m 3s\n",
      "600:\tlearn: 0.9626959\ttest: 0.9617687\tbest: 0.9617687 (600)\ttotal: 49.6s\tremaining: 1m 55s\n",
      "700:\tlearn: 0.9611510\ttest: 0.9614305\tbest: 0.9614293 (699)\ttotal: 57.9s\tremaining: 1m 47s\n",
      "800:\tlearn: 0.9596375\ttest: 0.9609933\tbest: 0.9609933 (800)\ttotal: 1m 6s\tremaining: 1m 39s\n",
      "900:\tlearn: 0.9582951\ttest: 0.9607552\tbest: 0.9607552 (900)\ttotal: 1m 14s\tremaining: 1m 30s\n",
      "1000:\tlearn: 0.9569590\ttest: 0.9605615\tbest: 0.9605615 (1000)\ttotal: 1m 22s\tremaining: 1m 22s\n",
      "1100:\tlearn: 0.9556870\ttest: 0.9604111\tbest: 0.9604048 (1097)\ttotal: 1m 30s\tremaining: 1m 13s\n",
      "1200:\tlearn: 0.9544699\ttest: 0.9602216\tbest: 0.9602216 (1200)\ttotal: 1m 38s\tremaining: 1m 5s\n",
      "1300:\tlearn: 0.9533036\ttest: 0.9601120\tbest: 0.9600998 (1265)\ttotal: 1m 46s\tremaining: 57.2s\n",
      "1400:\tlearn: 0.9521198\ttest: 0.9600119\tbest: 0.9600107 (1392)\ttotal: 1m 54s\tremaining: 49s\n",
      "1500:\tlearn: 0.9509329\ttest: 0.9598937\tbest: 0.9598865 (1492)\ttotal: 2m 2s\tremaining: 40.7s\n",
      "1600:\tlearn: 0.9498940\ttest: 0.9598416\tbest: 0.9598405 (1599)\ttotal: 2m 10s\tremaining: 32.5s\n",
      "1700:\tlearn: 0.9488926\ttest: 0.9598400\tbest: 0.9598218 (1686)\ttotal: 2m 18s\tremaining: 24.3s\n",
      "1800:\tlearn: 0.9477804\ttest: 0.9597759\tbest: 0.9597703 (1790)\ttotal: 2m 26s\tremaining: 16.1s\n",
      "1900:\tlearn: 0.9466716\ttest: 0.9597062\tbest: 0.9597005 (1867)\ttotal: 2m 33s\tremaining: 8.02s\n",
      "1999:\tlearn: 0.9456289\ttest: 0.9596609\tbest: 0.9596609 (1999)\ttotal: 2m 41s\tremaining: 0us\n",
      "\n",
      "bestTest = 0.959660936\n",
      "bestIteration = 1999\n",
      "\n",
      "  N trees =  2000\n",
      "  Gini =  0.9596609353939728\n",
      "\n",
      "Fold  3\n",
      "0:\tlearn: 7.4159594\ttest: 7.4138101\tbest: 7.4138101 (0)\ttotal: 86.4ms\tremaining: 2m 52s\n",
      "100:\tlearn: 0.9864668\ttest: 0.9919584\tbest: 0.9919584 (100)\ttotal: 9.37s\tremaining: 2m 56s\n",
      "200:\tlearn: 0.9717886\ttest: 0.9778758\tbest: 0.9778758 (200)\ttotal: 17.8s\tremaining: 2m 38s\n",
      "300:\tlearn: 0.9673833\ttest: 0.9743094\tbest: 0.9743094 (300)\ttotal: 25.8s\tremaining: 2m 25s\n",
      "400:\tlearn: 0.9648536\ttest: 0.9728874\tbest: 0.9728874 (400)\ttotal: 33.6s\tremaining: 2m 13s\n",
      "500:\tlearn: 0.9628081\ttest: 0.9719615\tbest: 0.9719615 (500)\ttotal: 41.5s\tremaining: 2m 4s\n",
      "600:\tlearn: 0.9611903\ttest: 0.9712762\tbest: 0.9712762 (600)\ttotal: 49.4s\tremaining: 1m 54s\n",
      "700:\tlearn: 0.9597353\ttest: 0.9708299\tbest: 0.9708290 (699)\ttotal: 57.5s\tremaining: 1m 46s\n",
      "800:\tlearn: 0.9583420\ttest: 0.9703897\tbest: 0.9703897 (800)\ttotal: 1m 5s\tremaining: 1m 37s\n",
      "900:\tlearn: 0.9570200\ttest: 0.9700710\tbest: 0.9700684 (897)\ttotal: 1m 12s\tremaining: 1m 28s\n",
      "1000:\tlearn: 0.9557625\ttest: 0.9697254\tbest: 0.9697254 (1000)\ttotal: 1m 20s\tremaining: 1m 20s\n",
      "1100:\tlearn: 0.9544332\ttest: 0.9694778\tbest: 0.9694712 (1099)\ttotal: 1m 28s\tremaining: 1m 12s\n",
      "1200:\tlearn: 0.9532920\ttest: 0.9692887\tbest: 0.9692887 (1200)\ttotal: 1m 36s\tremaining: 1m 4s\n",
      "1300:\tlearn: 0.9522094\ttest: 0.9692194\tbest: 0.9692064 (1270)\ttotal: 1m 44s\tremaining: 56.1s\n",
      "1400:\tlearn: 0.9510898\ttest: 0.9690965\tbest: 0.9690900 (1395)\ttotal: 1m 52s\tremaining: 48s\n",
      "1500:\tlearn: 0.9499583\ttest: 0.9689752\tbest: 0.9689728 (1489)\ttotal: 2m\tremaining: 40s\n",
      "1600:\tlearn: 0.9488197\ttest: 0.9688000\tbest: 0.9688000 (1600)\ttotal: 2m 8s\tremaining: 32s\n",
      "1700:\tlearn: 0.9477035\ttest: 0.9686407\tbest: 0.9686403 (1699)\ttotal: 2m 16s\tremaining: 23.9s\n",
      "1800:\tlearn: 0.9466340\ttest: 0.9685410\tbest: 0.9685406 (1799)\ttotal: 2m 24s\tremaining: 15.9s\n",
      "1900:\tlearn: 0.9455818\ttest: 0.9684940\tbest: 0.9684922 (1896)\ttotal: 2m 32s\tremaining: 7.92s\n",
      "1999:\tlearn: 0.9445294\ttest: 0.9684234\tbest: 0.9684234 (1999)\ttotal: 2m 40s\tremaining: 0us\n",
      "\n",
      "bestTest = 0.9684234459\n",
      "bestIteration = 1999\n",
      "\n",
      "  N trees =  2000\n",
      "  Gini =  0.968423445005046\n",
      "\n",
      "Fold  4\n",
      "0:\tlearn: 7.4170383\ttest: 7.4102204\tbest: 7.4102204 (0)\ttotal: 159ms\tremaining: 5m 18s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100:\tlearn: 0.9862354\ttest: 0.9919600\tbest: 0.9919600 (100)\ttotal: 9.65s\tremaining: 3m 1s\n",
      "200:\tlearn: 0.9716066\ttest: 0.9778498\tbest: 0.9778498 (200)\ttotal: 18.1s\tremaining: 2m 42s\n",
      "300:\tlearn: 0.9671280\ttest: 0.9741392\tbest: 0.9741392 (300)\ttotal: 26.2s\tremaining: 2m 27s\n",
      "400:\tlearn: 0.9643527\ttest: 0.9725154\tbest: 0.9725154 (400)\ttotal: 34s\tremaining: 2m 15s\n",
      "500:\tlearn: 0.9622759\ttest: 0.9715804\tbest: 0.9715804 (500)\ttotal: 42s\tremaining: 2m 5s\n",
      "600:\tlearn: 0.9605007\ttest: 0.9708873\tbest: 0.9708860 (599)\ttotal: 49.8s\tremaining: 1m 55s\n",
      "700:\tlearn: 0.9590097\ttest: 0.9705038\tbest: 0.9704969 (697)\ttotal: 57.3s\tremaining: 1m 46s\n",
      "800:\tlearn: 0.9575842\ttest: 0.9701045\tbest: 0.9701045 (800)\ttotal: 1m 5s\tremaining: 1m 37s\n",
      "900:\tlearn: 0.9562200\ttest: 0.9698063\tbest: 0.9698027 (896)\ttotal: 1m 13s\tremaining: 1m 29s\n",
      "1000:\tlearn: 0.9549968\ttest: 0.9696009\tbest: 0.9696009 (1000)\ttotal: 1m 21s\tremaining: 1m 21s\n",
      "1100:\tlearn: 0.9537004\ttest: 0.9693528\tbest: 0.9693528 (1100)\ttotal: 1m 29s\tremaining: 1m 12s\n",
      "1200:\tlearn: 0.9524045\ttest: 0.9691845\tbest: 0.9691784 (1184)\ttotal: 1m 37s\tremaining: 1m 4s\n",
      "1300:\tlearn: 0.9512567\ttest: 0.9690575\tbest: 0.9690575 (1300)\ttotal: 1m 44s\tremaining: 56.4s\n",
      "1400:\tlearn: 0.9501704\ttest: 0.9689204\tbest: 0.9689128 (1393)\ttotal: 1m 53s\tremaining: 48.4s\n",
      "1500:\tlearn: 0.9490271\ttest: 0.9688447\tbest: 0.9688361 (1493)\ttotal: 2m 1s\tremaining: 40.3s\n",
      "1600:\tlearn: 0.9478950\ttest: 0.9687146\tbest: 0.9687090 (1598)\ttotal: 2m 9s\tremaining: 32.2s\n",
      "1700:\tlearn: 0.9467519\ttest: 0.9685896\tbest: 0.9685812 (1694)\ttotal: 2m 17s\tremaining: 24.1s\n",
      "1800:\tlearn: 0.9457302\ttest: 0.9685090\tbest: 0.9684975 (1789)\ttotal: 2m 25s\tremaining: 16s\n",
      "1900:\tlearn: 0.9446397\ttest: 0.9683898\tbest: 0.9683888 (1896)\ttotal: 2m 33s\tremaining: 7.97s\n",
      "1999:\tlearn: 0.9435873\ttest: 0.9683209\tbest: 0.9683209 (1999)\ttotal: 2m 42s\tremaining: 0us\n",
      "\n",
      "bestTest = 0.968320894\n",
      "bestIteration = 1999\n",
      "\n",
      "  N trees =  2000\n",
      "  Gini =  0.9683208935233364\n"
     ]
    }
   ],
   "source": [
    "K = 5\n",
    "kf = KFold(n_splits = K, random_state = 1, shuffle = True)\n",
    "y_valid_pred = 0*y\n",
    "y_test_pred = 0\n",
    "for i, (train_index, test_index) in enumerate(kf.split(X)):\n",
    "    \n",
    "    # Create data for this fold\n",
    "    y_train, y_valid = y.iloc[train_index], y.iloc[test_index]\n",
    "    X_train, X_valid = X.iloc[train_index,:], X.iloc[test_index,:]\n",
    "    print( \"\\nFold \", i)\n",
    "    \n",
    "    # Run model for this fold\n",
    "    if OPTIMIZE_ROUNDS:\n",
    "        fit_model = model.fit(X_train, y_train, \n",
    "                               eval_set=[(X_valid, y_valid)],\n",
    "                               use_best_model=True\n",
    "                             )\n",
    "        print( \"  N trees = \", model.tree_count_ )\n",
    "    else:\n",
    "        fit_model = model.fit( X_train, y_train )\n",
    "        \n",
    "    # Generate validation predictions for this fold\n",
    "    pred = fit_model.predict(X_valid)\n",
    "    print( \"  Gini = \", np.sqrt(mean_squared_error(y_valid, pred) ))\n",
    "    y_valid_pred.iloc[test_index] = pred\n",
    "    \n",
    "    # Accumulate test set predictions\n",
    "    y_test_pred += fit_model.predict(x_test)\n",
    "    \n",
    "y_test_pred /= K  # Average test set predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 190,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([8.35435413, 7.40192963, 7.50145624, ..., 8.50680044, 8.12733681,\n",
       "       7.22221566])"
      ]
     },
     "execution_count": 190,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test_fold_cat = y_test_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "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",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>roomnightsmem_amount_minstate_code_resid</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>in_outmem_inout_medianstate_code_resid</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_paxres_inout_min_pax</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reservation_idres_memmain_product_code</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>roomnightsres_amount_min</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>in_outres_mem_inout_minstate_code_resid</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_paxmem_inout_min_pax_state_code</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>checkout_dateis_month_end</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>roomnightsmem_amount_medianstate_code_resid</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>resort_type_code</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_paxres_inout_median_pax</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_paxmem_inout_median_pax_state_code</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>in_outmem_inout_minstate_code_resid</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>roomnightsres_amount_max</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>checkin_dateis_month_end</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cluster_code</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_paxres_inout_max_pax</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>checkout_dateyear</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>roomnightsres_amount_std</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>resort_region_code</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>checkin_dateyear</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>booking_datemonth</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>checkout_datemonth</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>roomnightsres_mem_amount_minstate_code_resid</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_paxres_inout_std_pax</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>checkin_datemonth</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reservation_id_x_res</th>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>roomnightsres_amount_median</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>booking_dateyear</th>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_paxmem_inout_max_pax_state_code</th>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reservation_id_x_res_mem_state_scoreresort_id</th>\n",
       "      <td>213</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>roomnightsmem_amount_median</th>\n",
       "      <td>220</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>booking_dateday</th>\n",
       "      <td>226</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>checkin_dayes</th>\n",
       "      <td>240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>in_outres_meminoutt_stdstate_code_resid</th>\n",
       "      <td>242</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reservation_idcheck_Date_month</th>\n",
       "      <td>243</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>checkout_datedayofyear</th>\n",
       "      <td>244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>booking_datedayofyear</th>\n",
       "      <td>250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>checkout_dateday</th>\n",
       "      <td>255</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>resort_id_memberid_nextClick</th>\n",
       "      <td>256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>main_product_code</th>\n",
       "      <td>258</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>checkin_datedayofyear</th>\n",
       "      <td>263</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>in_out</th>\n",
       "      <td>268</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>numberofadults</th>\n",
       "      <td>295</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>booking_dayes</th>\n",
       "      <td>309</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reservation_idchannel_check_yearmonth</th>\n",
       "      <td>309</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>resort_id_memberid_prevClick</th>\n",
       "      <td>335</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rooms_booked_per_night</th>\n",
       "      <td>355</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>resort_id_nextClick</th>\n",
       "      <td>358</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>persontravellingid</th>\n",
       "      <td>378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_pax</th>\n",
       "      <td>380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>resort_id_prevClick</th>\n",
       "      <td>380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rooms_bookinf_per_night</th>\n",
       "      <td>411</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>in_book</th>\n",
       "      <td>425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_paxmem_inout_std_pax</th>\n",
       "      <td>508</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>in_outmem_inout_std</th>\n",
       "      <td>545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>roomnightsmem_amount_std</th>\n",
       "      <td>568</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>member_age_buckets</th>\n",
       "      <td>1023</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>state_code_residence</th>\n",
       "      <td>2406</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>resort_id</th>\n",
       "      <td>4366</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>115 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                  0\n",
       "roomnightsmem_amount_minstate_code_resid          0\n",
       "in_outmem_inout_medianstate_code_resid            0\n",
       "total_paxres_inout_min_pax                        0\n",
       "reservation_idres_memmain_product_code            0\n",
       "roomnightsres_amount_min                          0\n",
       "in_outres_mem_inout_minstate_code_resid           0\n",
       "total_paxmem_inout_min_pax_state_code             0\n",
       "checkout_dateis_month_end                         0\n",
       "roomnightsmem_amount_medianstate_code_resid       0\n",
       "resort_type_code                                  0\n",
       "total_paxres_inout_median_pax                     0\n",
       "total_paxmem_inout_median_pax_state_code          0\n",
       "in_outmem_inout_minstate_code_resid               0\n",
       "roomnightsres_amount_max                          0\n",
       "checkin_dateis_month_end                          1\n",
       "cluster_code                                      1\n",
       "total_paxres_inout_max_pax                        1\n",
       "checkout_dateyear                                 1\n",
       "roomnightsres_amount_std                          1\n",
       "resort_region_code                                3\n",
       "checkin_dateyear                                  3\n",
       "booking_datemonth                                 3\n",
       "checkout_datemonth                                3\n",
       "roomnightsres_mem_amount_minstate_code_resid      4\n",
       "total_paxres_inout_std_pax                        5\n",
       "checkin_datemonth                                 6\n",
       "reservation_id_x_res                              7\n",
       "roomnightsres_amount_median                       8\n",
       "booking_dateyear                                  9\n",
       "total_paxmem_inout_max_pax_state_code             9\n",
       "...                                             ...\n",
       "reservation_id_x_res_mem_state_scoreresort_id   213\n",
       "roomnightsmem_amount_median                     220\n",
       "booking_dateday                                 226\n",
       "checkin_dayes                                   240\n",
       "in_outres_meminoutt_stdstate_code_resid         242\n",
       "reservation_idcheck_Date_month                  243\n",
       "checkout_datedayofyear                          244\n",
       "booking_datedayofyear                           250\n",
       "checkout_dateday                                255\n",
       "resort_id_memberid_nextClick                    256\n",
       "main_product_code                               258\n",
       "checkin_datedayofyear                           263\n",
       "in_out                                          268\n",
       "numberofadults                                  295\n",
       "booking_dayes                                   309\n",
       "reservation_idchannel_check_yearmonth           309\n",
       "resort_id_memberid_prevClick                    335\n",
       "rooms_booked_per_night                          355\n",
       "resort_id_nextClick                             358\n",
       "persontravellingid                              378\n",
       "total_pax                                       380\n",
       "resort_id_prevClick                             380\n",
       "rooms_bookinf_per_night                         411\n",
       "in_book                                         425\n",
       "total_paxmem_inout_std_pax                      508\n",
       "in_outmem_inout_std                             545\n",
       "roomnightsmem_amount_std                        568\n",
       "member_age_buckets                             1023\n",
       "state_code_residence                           2406\n",
       "resort_id                                      4366\n",
       "\n",
       "[115 rows x 1 columns]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(clf.feature_importances_,index=train_x.columns).sort_values(by=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_test = test[X.columns]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Making 5Fold prediction Model for LGB dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "FEATS_EXCLUDED=['target']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Not there\n"
     ]
    }
   ],
   "source": [
    "try:\n",
    "    X.drop('target',axis=1,inplace=True)\n",
    "except:\n",
    "    print(\"Not there\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "categor_col =cat_cols+[\n",
    "       'channel_code',\n",
    "            'main_product_code', \n",
    "      'persontravellingid', 'resort_region_code',\n",
    "       'resort_type_code', 'room_type_booked_code', \n",
    "       'season_holidayed_code', 'state_code_residence', 'state_code_resort',\n",
    "       'member_age_buckets', 'booking_type_code', 'memberid',\n",
    "       ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "categor_col.remove('memberid')\n",
    "categor_col.remove('memberid')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "import datetime\n",
    "import gc\n",
    "import lightgbm as lgb\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import os\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import time\n",
    "import warnings\n",
    "\n",
    "from contextlib import contextmanager\n",
    "from pandas.core.common import SettingWithCopyWarning\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.model_selection import KFold, StratifiedKFold\n",
    "\n",
    "def kfold_lightgbm(train_df, test_df, num_folds, y,stratified = False, debug= False):\n",
    "    print(\"Starting LightGBM. Train shape: {}, test shape: {}\".format(train_df.shape, test_df.shape))\n",
    "    train_df['target']=y\n",
    "    # Cross validation model\n",
    "    if stratified:\n",
    "        folds = StratifiedKFold(n_splits= num_folds, shuffle=True, random_state=326)\n",
    "    else:\n",
    "        folds = KFold(n_splits= num_folds, shuffle=True, random_state=326)\n",
    "\n",
    "    # Create arrays and dataframes to store results\n",
    "    oof_preds = np.zeros(train_df.shape[0])\n",
    "    sub_preds = np.zeros(test_df.shape[0])\n",
    "    feature_importance_df = pd.DataFrame()\n",
    "    feats = [f for f in train_df.columns if f not in FEATS_EXCLUDED]\n",
    "\n",
    "    # k-fold\n",
    "    for n_fold, (train_idx, valid_idx) in enumerate(folds.split(train_df[feats])):\n",
    "        train_x, train_y = train_df[feats].iloc[train_idx], train_df['target'].iloc[train_idx]\n",
    "        valid_x, valid_y = train_df[feats].iloc[valid_idx], train_df['target'].iloc[valid_idx]\n",
    "\n",
    "        # set data structure\n",
    "        lgb_train = lgb.Dataset(train_x,\n",
    "                                label=train_y,\n",
    "                                free_raw_data=False)\n",
    "        lgb_test = lgb.Dataset(valid_x,\n",
    "                               label=valid_y,\n",
    "                               free_raw_data=False)\n",
    "        params = {\"objective\" : \"regression\", \"metric\" : \"rmse\", 'n_estimators':3000, 'early_stopping_rounds':200,\n",
    "              \"num_leaves\" : 31, \"learning_rate\" : 0.05, \"bagging_fraction\" : 0.9,\n",
    "               \"bagging_seed\" : 0, \"num_threads\" : 4,\"colsample_bytree\" : 0.8,\"lambda_l1\": 25,\"min_child_weight\" : 44.97\n",
    "          ,\"min_child_samples\": 5,\n",
    "             }\n",
    "\n",
    "\n",
    "\n",
    "        reg = lgb.train(\n",
    "                        params,\n",
    "                        lgb_train,\n",
    "                        valid_sets=[lgb_train, lgb_test],\n",
    "                        valid_names=['train', 'test'],\n",
    "                        num_boost_round=10000,\n",
    "                        early_stopping_rounds= 200,\n",
    "                        verbose_eval=100,\n",
    "            categorical_feature=categor_col\n",
    "                        )\n",
    "\n",
    "        oof_preds[valid_idx] = reg.predict(valid_x, num_iteration=reg.best_iteration)\n",
    "        sub_preds += reg.predict(test_df[feats], num_iteration=reg.best_iteration) / folds.n_splits\n",
    "\n",
    "        fold_importance_df = pd.DataFrame()\n",
    "        fold_importance_df[\"feature\"] = feats\n",
    "        fold_importance_df[\"importance\"] = np.log1p(reg.feature_importance(importance_type='gain', iteration=reg.best_iteration))\n",
    "        fold_importance_df[\"fold\"] = n_fold + 1\n",
    "        feature_importance_df = pd.concat([feature_importance_df, fold_importance_df], axis=0)\n",
    "        print('Fold %2d RMSE : %.6f' % (n_fold + 1, rmse(valid_y, oof_preds[valid_idx])))\n",
    "        del reg, train_x, train_y, valid_x, valid_y\n",
    "        gc.collect()\n",
    "\n",
    "    # display importances\n",
    "    display_importances(feature_importance_df)\n",
    "\n",
    "    if not debug:\n",
    "        # save submission file\n",
    "        return sub_preds\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Starting LightGBM. Train shape: (341424, 115), test shape: (146765, 121)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/lightgbm/engine.py:113: UserWarning: Found `n_estimators` in params. Will use it instead of argument\n",
      "  warnings.warn(\"Found `{}` in params. Will use it instead of argument\".format(alias))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/engine.py:118: UserWarning: Found `early_stopping_rounds` in params. Will use it instead of argument\n",
      "  warnings.warn(\"Found `{}` in params. Will use it instead of argument\".format(alias))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/basic.py:1188: UserWarning: categorical_feature in Dataset is overridden.\n",
      "New categorical_feature is ['booking_type_code', 'channel_code', 'cluster_code', 'main_product_code', 'member_age_buckets', 'member_age_buckets', 'persontravellingid', 'reservationstatusid_code', 'resort_id', 'resort_region_code', 'resort_type_code', 'room_type_booked_code', 'season_holidayed_code', 'state_code_residence', 'state_code_resort']\n",
      "  'New categorical_feature is {}'.format(sorted(list(categorical_feature))))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/basic.py:1188: UserWarning: categorical_feature in Dataset is overridden.\n",
      "New categorical_feature is ['booking_type_code', 'channel_code', 'cluster_code', 'main_product_code', 'member_age_buckets', 'member_age_buckets', 'persontravellingid', 'reservationstatusid_code', 'resort_id', 'resort_region_code', 'resort_type_code', 'room_type_booked_code', 'season_holidayed_code', 'state_code_residence', 'state_code_resort']\n",
      "  'New categorical_feature is {}'.format(sorted(list(categorical_feature))))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/basic.py:742: UserWarning: categorical_feature in param dict is overridden.\n",
      "  warnings.warn('categorical_feature in param dict is overridden.')\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training until validation scores don't improve for 200 rounds.\n",
      "[100]\ttrain's rmse: 0.962849\ttest's rmse: 0.972401\n",
      "[200]\ttrain's rmse: 0.953202\ttest's rmse: 0.96843\n",
      "[300]\ttrain's rmse: 0.9473\ttest's rmse: 0.967665\n",
      "[400]\ttrain's rmse: 0.942733\ttest's rmse: 0.967395\n",
      "[500]\ttrain's rmse: 0.938358\ttest's rmse: 0.967282\n",
      "[600]\ttrain's rmse: 0.934283\ttest's rmse: 0.967164\n",
      "[700]\ttrain's rmse: 0.930437\ttest's rmse: 0.967097\n",
      "[800]\ttrain's rmse: 0.926429\ttest's rmse: 0.967073\n",
      "[900]\ttrain's rmse: 0.922624\ttest's rmse: 0.967024\n",
      "[1000]\ttrain's rmse: 0.918783\ttest's rmse: 0.967028\n",
      "Early stopping, best iteration is:\n",
      "[876]\ttrain's rmse: 0.923496\ttest's rmse: 0.966996\n",
      "Fold  1 RMSE : 0.966996\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/lightgbm/engine.py:113: UserWarning: Found `n_estimators` in params. Will use it instead of argument\n",
      "  warnings.warn(\"Found `{}` in params. Will use it instead of argument\".format(alias))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/engine.py:118: UserWarning: Found `early_stopping_rounds` in params. Will use it instead of argument\n",
      "  warnings.warn(\"Found `{}` in params. Will use it instead of argument\".format(alias))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/basic.py:1188: UserWarning: categorical_feature in Dataset is overridden.\n",
      "New categorical_feature is ['booking_type_code', 'channel_code', 'cluster_code', 'main_product_code', 'member_age_buckets', 'member_age_buckets', 'persontravellingid', 'reservationstatusid_code', 'resort_id', 'resort_region_code', 'resort_type_code', 'room_type_booked_code', 'season_holidayed_code', 'state_code_residence', 'state_code_resort']\n",
      "  'New categorical_feature is {}'.format(sorted(list(categorical_feature))))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/basic.py:1188: UserWarning: categorical_feature in Dataset is overridden.\n",
      "New categorical_feature is ['booking_type_code', 'channel_code', 'cluster_code', 'main_product_code', 'member_age_buckets', 'member_age_buckets', 'persontravellingid', 'reservationstatusid_code', 'resort_id', 'resort_region_code', 'resort_type_code', 'room_type_booked_code', 'season_holidayed_code', 'state_code_residence', 'state_code_resort']\n",
      "  'New categorical_feature is {}'.format(sorted(list(categorical_feature))))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/basic.py:742: UserWarning: categorical_feature in param dict is overridden.\n",
      "  warnings.warn('categorical_feature in param dict is overridden.')\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training until validation scores don't improve for 200 rounds.\n",
      "[100]\ttrain's rmse: 0.963517\ttest's rmse: 0.970017\n",
      "[200]\ttrain's rmse: 0.953886\ttest's rmse: 0.966038\n",
      "[300]\ttrain's rmse: 0.947987\ttest's rmse: 0.965212\n",
      "[400]\ttrain's rmse: 0.943171\ttest's rmse: 0.964731\n",
      "[500]\ttrain's rmse: 0.938856\ttest's rmse: 0.964568\n",
      "[600]\ttrain's rmse: 0.934806\ttest's rmse: 0.964424\n",
      "[700]\ttrain's rmse: 0.930923\ttest's rmse: 0.964285\n",
      "[800]\ttrain's rmse: 0.92698\ttest's rmse: 0.964171\n",
      "[900]\ttrain's rmse: 0.923232\ttest's rmse: 0.964184\n",
      "[1000]\ttrain's rmse: 0.919596\ttest's rmse: 0.964121\n",
      "[1100]\ttrain's rmse: 0.91611\ttest's rmse: 0.964152\n",
      "[1200]\ttrain's rmse: 0.912483\ttest's rmse: 0.964316\n",
      "Early stopping, best iteration is:\n",
      "[1036]\ttrain's rmse: 0.918254\ttest's rmse: 0.964075\n",
      "Fold  2 RMSE : 0.964075\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/lightgbm/engine.py:113: UserWarning: Found `n_estimators` in params. Will use it instead of argument\n",
      "  warnings.warn(\"Found `{}` in params. Will use it instead of argument\".format(alias))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/engine.py:118: UserWarning: Found `early_stopping_rounds` in params. Will use it instead of argument\n",
      "  warnings.warn(\"Found `{}` in params. Will use it instead of argument\".format(alias))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/basic.py:1188: UserWarning: categorical_feature in Dataset is overridden.\n",
      "New categorical_feature is ['booking_type_code', 'channel_code', 'cluster_code', 'main_product_code', 'member_age_buckets', 'member_age_buckets', 'persontravellingid', 'reservationstatusid_code', 'resort_id', 'resort_region_code', 'resort_type_code', 'room_type_booked_code', 'season_holidayed_code', 'state_code_residence', 'state_code_resort']\n",
      "  'New categorical_feature is {}'.format(sorted(list(categorical_feature))))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/basic.py:1188: UserWarning: categorical_feature in Dataset is overridden.\n",
      "New categorical_feature is ['booking_type_code', 'channel_code', 'cluster_code', 'main_product_code', 'member_age_buckets', 'member_age_buckets', 'persontravellingid', 'reservationstatusid_code', 'resort_id', 'resort_region_code', 'resort_type_code', 'room_type_booked_code', 'season_holidayed_code', 'state_code_residence', 'state_code_resort']\n",
      "  'New categorical_feature is {}'.format(sorted(list(categorical_feature))))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/basic.py:742: UserWarning: categorical_feature in param dict is overridden.\n",
      "  warnings.warn('categorical_feature in param dict is overridden.')\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training until validation scores don't improve for 200 rounds.\n",
      "[100]\ttrain's rmse: 0.963451\ttest's rmse: 0.970643\n",
      "[200]\ttrain's rmse: 0.953847\ttest's rmse: 0.966227\n",
      "[300]\ttrain's rmse: 0.948094\ttest's rmse: 0.965242\n",
      "[400]\ttrain's rmse: 0.943574\ttest's rmse: 0.964886\n",
      "[500]\ttrain's rmse: 0.939363\ttest's rmse: 0.964613\n",
      "[600]\ttrain's rmse: 0.935208\ttest's rmse: 0.964449\n",
      "[700]\ttrain's rmse: 0.931385\ttest's rmse: 0.964348\n",
      "[800]\ttrain's rmse: 0.92754\ttest's rmse: 0.964251\n",
      "[900]\ttrain's rmse: 0.923856\ttest's rmse: 0.964174\n",
      "[1000]\ttrain's rmse: 0.920196\ttest's rmse: 0.964155\n",
      "[1100]\ttrain's rmse: 0.91657\ttest's rmse: 0.964155\n",
      "[1200]\ttrain's rmse: 0.913079\ttest's rmse: 0.964123\n",
      "[1300]\ttrain's rmse: 0.909815\ttest's rmse: 0.96415\n",
      "Early stopping, best iteration is:\n",
      "[1197]\ttrain's rmse: 0.913203\ttest's rmse: 0.964104\n",
      "Fold  3 RMSE : 0.964104\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/lightgbm/engine.py:113: UserWarning: Found `n_estimators` in params. Will use it instead of argument\n",
      "  warnings.warn(\"Found `{}` in params. Will use it instead of argument\".format(alias))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/engine.py:118: UserWarning: Found `early_stopping_rounds` in params. Will use it instead of argument\n",
      "  warnings.warn(\"Found `{}` in params. Will use it instead of argument\".format(alias))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/basic.py:1188: UserWarning: categorical_feature in Dataset is overridden.\n",
      "New categorical_feature is ['booking_type_code', 'channel_code', 'cluster_code', 'main_product_code', 'member_age_buckets', 'member_age_buckets', 'persontravellingid', 'reservationstatusid_code', 'resort_id', 'resort_region_code', 'resort_type_code', 'room_type_booked_code', 'season_holidayed_code', 'state_code_residence', 'state_code_resort']\n",
      "  'New categorical_feature is {}'.format(sorted(list(categorical_feature))))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/basic.py:1188: UserWarning: categorical_feature in Dataset is overridden.\n",
      "New categorical_feature is ['booking_type_code', 'channel_code', 'cluster_code', 'main_product_code', 'member_age_buckets', 'member_age_buckets', 'persontravellingid', 'reservationstatusid_code', 'resort_id', 'resort_region_code', 'resort_type_code', 'room_type_booked_code', 'season_holidayed_code', 'state_code_residence', 'state_code_resort']\n",
      "  'New categorical_feature is {}'.format(sorted(list(categorical_feature))))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/basic.py:742: UserWarning: categorical_feature in param dict is overridden.\n",
      "  warnings.warn('categorical_feature in param dict is overridden.')\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training until validation scores don't improve for 200 rounds.\n",
      "[100]\ttrain's rmse: 0.964655\ttest's rmse: 0.964894\n",
      "[200]\ttrain's rmse: 0.955037\ttest's rmse: 0.96124\n",
      "[300]\ttrain's rmse: 0.948954\ttest's rmse: 0.960401\n",
      "[400]\ttrain's rmse: 0.944343\ttest's rmse: 0.960121\n",
      "[500]\ttrain's rmse: 0.940032\ttest's rmse: 0.960034\n",
      "[600]\ttrain's rmse: 0.93596\ttest's rmse: 0.959911\n",
      "[700]\ttrain's rmse: 0.931916\ttest's rmse: 0.959993\n",
      "[800]\ttrain's rmse: 0.928072\ttest's rmse: 0.960029\n",
      "Early stopping, best iteration is:\n",
      "[600]\ttrain's rmse: 0.93596\ttest's rmse: 0.959911\n",
      "Fold  4 RMSE : 0.959911\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/lightgbm/engine.py:113: UserWarning: Found `n_estimators` in params. Will use it instead of argument\n",
      "  warnings.warn(\"Found `{}` in params. Will use it instead of argument\".format(alias))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/engine.py:118: UserWarning: Found `early_stopping_rounds` in params. Will use it instead of argument\n",
      "  warnings.warn(\"Found `{}` in params. Will use it instead of argument\".format(alias))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/basic.py:1188: UserWarning: categorical_feature in Dataset is overridden.\n",
      "New categorical_feature is ['booking_type_code', 'channel_code', 'cluster_code', 'main_product_code', 'member_age_buckets', 'member_age_buckets', 'persontravellingid', 'reservationstatusid_code', 'resort_id', 'resort_region_code', 'resort_type_code', 'room_type_booked_code', 'season_holidayed_code', 'state_code_residence', 'state_code_resort']\n",
      "  'New categorical_feature is {}'.format(sorted(list(categorical_feature))))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/basic.py:1188: UserWarning: categorical_feature in Dataset is overridden.\n",
      "New categorical_feature is ['booking_type_code', 'channel_code', 'cluster_code', 'main_product_code', 'member_age_buckets', 'member_age_buckets', 'persontravellingid', 'reservationstatusid_code', 'resort_id', 'resort_region_code', 'resort_type_code', 'room_type_booked_code', 'season_holidayed_code', 'state_code_residence', 'state_code_resort']\n",
      "  'New categorical_feature is {}'.format(sorted(list(categorical_feature))))\n",
      "/anaconda3/lib/python3.6/site-packages/lightgbm/basic.py:742: UserWarning: categorical_feature in param dict is overridden.\n",
      "  warnings.warn('categorical_feature in param dict is overridden.')\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training until validation scores don't improve for 200 rounds.\n",
      "[100]\ttrain's rmse: 0.964043\ttest's rmse: 0.969136\n",
      "[200]\ttrain's rmse: 0.954405\ttest's rmse: 0.964275\n",
      "[300]\ttrain's rmse: 0.948702\ttest's rmse: 0.963143\n",
      "[400]\ttrain's rmse: 0.944159\ttest's rmse: 0.962782\n",
      "[500]\ttrain's rmse: 0.940001\ttest's rmse: 0.962604\n",
      "[600]\ttrain's rmse: 0.935859\ttest's rmse: 0.96235\n",
      "[700]\ttrain's rmse: 0.932019\ttest's rmse: 0.962266\n",
      "[800]\ttrain's rmse: 0.928233\ttest's rmse: 0.962139\n",
      "[900]\ttrain's rmse: 0.92456\ttest's rmse: 0.962139\n",
      "Early stopping, best iteration is:\n",
      "[785]\ttrain's rmse: 0.928789\ttest's rmse: 0.962117\n",
      "Fold  5 RMSE : 0.962117\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def display_importances(feature_importance_df_):\n",
    "    cols = feature_importance_df_[[\"feature\", \"importance\"]].groupby(\"feature\").mean().sort_values(by=\"importance\", ascending=False)[:40].index\n",
    "    best_features = feature_importance_df_.loc[feature_importance_df_.feature.isin(cols)]\n",
    "\n",
    "    plt.figure(figsize=(8, 10))\n",
    "    sns.barplot(x=\"importance\", y=\"feature\", data=best_features.sort_values(by=\"importance\", ascending=False))\n",
    "    plt.title('LightGBM Features (avg over folds)')\n",
    "    plt.tight_layout()\n",
    "    plt.savefig('lgbm_importances.png')\n",
    "\n",
    "\n",
    "def rmse(y_true, y_pred):\n",
    "    return np.sqrt(mean_squared_error(y_true, y_pred))\n",
    "\n",
    "FEATS_EXCLUDED=['target']\n",
    "pred_test = kfold_lightgbm(X,test,5,train.amount_spent_per_room_night_scaled)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_lgb_5_fold = pred_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_ensem= pred_test_lgb_single_model*0.3+pred_catboost_model*0.2+pred_lgb_5_fold*0.3+y_test_fold_cat*0.2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {},
   "outputs": [],
   "source": [
    "sub = pd.read_csv('sample_submission.csv')\n",
    "\n",
    "sub.amount_spent_per_room_night_scaled=pred_ensem\n",
    "\n",
    "sub.to_csv('submission_18_rewrite_parallel.csv',index=None)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Feature Selection Code :\n",
    "Taken from TEARTH's Notebook ~ Amexpert Solution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tqdm import*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "def permutation_feature_selection(model, X_val, y_val, pred_proba=False, rep=3, max_delta_score=0.0):\n",
    "    from sklearn.metrics import roc_auc_score\n",
    "\n",
    "    # to do: predict_proba from Booster\n",
    "\n",
    "    \"\"\"\"\"\n",
    "    Perform permutation feature importance calculation for trained LightGBM model. \n",
    "    Scorer - ROC AUC. \n",
    "    The lower score with permuted feature - the more important feature is. \n",
    "    \n",
    "    Parameters\n",
    "    ----------\n",
    "    model : lightGBM.Booster\n",
    "        Trained model to perform feature importance calculation \n",
    "        \n",
    "    X_val : pandas.DataFrame\n",
    "        Validation dataset\n",
    "            \n",
    "    y_val : pandas.Series \n",
    "        Targets for validation dataset\n",
    "       \n",
    "    rep : integer (default = 3)\n",
    "        Number of permutations. More permutations lead to more robust results, but requires more calculation time.\n",
    "        It is recommended to use values in the range [3; 10].\n",
    "        \n",
    "    min_delta_score : float (default = 0.0)\n",
    "        Minimum delta ins score to keep feature \n",
    "        It is recommended to use values in the range [-0.0001; 0].\n",
    "        \n",
    "    Output\n",
    "    ----------\n",
    "    selected_features : list\n",
    "        Important features \n",
    "        \n",
    "    importance_df : pandas.DataFrame \n",
    "        Records of calculations \n",
    "    \"\"\"\"\"\n",
    "\n",
    "    print('Permutation feature importance is calculating...')\n",
    "\n",
    "    columns = []\n",
    "    scores = []\n",
    "    stds = []\n",
    "    score_max = []\n",
    "    score_min = []\n",
    "\n",
    "    # calculate the score of model with no permuted features, this is our baseline\n",
    "    y_hat_nopert = model.predict(X_val)\n",
    "    score_init = -mean_squared_error(y_val, y_hat_nopert)\n",
    "\n",
    "    for cc in tqdm(X_val.columns):\n",
    "        scores_cc = []\n",
    "        for seed in range(rep):\n",
    "            # shuffle single column of dataset\n",
    "            data_temp = X_val.copy()\n",
    "            data_temp[cc] = data_temp[cc].sample(n=X_val.shape[0], random_state=seed).reset_index(drop=True)\n",
    "\n",
    "            # make prediction on 'shuffled' dataset and score it\n",
    "            y_hat = model.predict(data_temp)\n",
    "            score = -mean_squared_error(y_val, y_hat)\n",
    "            scores_cc.append(score)\n",
    "\n",
    "        columns.append(cc)\n",
    "        scores.append(np.mean(scores_cc) - score_init)\n",
    "        stds.append(np.std(scores_cc))\n",
    "        score_max.append(np.max(scores_cc) - score_init)\n",
    "        score_min.append(np.min(scores_cc) - score_init)\n",
    "\n",
    "    importance_df = pd.DataFrame({'delta_score_mean': scores,\n",
    "                                  'delta_score_std': stds,\n",
    "                                  'delta_score_max': score_max,\n",
    "                                  'delta_score_min': score_min,\n",
    "                                  'feature': columns\n",
    "                                  })\n",
    "    importance_df = importance_df.sort_values(by='delta_score_mean')\n",
    "    selected_features = list(importance_df.loc[importance_df['delta_score_mean'] <= max_delta_score, 'feature'])\n",
    "\n",
    "    print('Permutation feature importance calculation is done. Overall number of features: ', importance_df.shape[0],\n",
    "          'Number of selected features:', len(selected_features))\n",
    "\n",
    "    return selected_features, importance_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_x = test_x.astype(float)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Permutation feature importance is calculating...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  0%|          | 0/88 [00:00<?, ?it/s]\u001b[A\n",
      "  1%|          | 1/88 [00:23<34:11, 23.58s/it]\u001b[A\n",
      "  2%|▏         | 2/88 [00:47<34:02, 23.75s/it]\u001b[A\n",
      "  3%|▎         | 3/88 [01:12<34:07, 24.08s/it]\u001b[A\n",
      "  5%|▍         | 4/88 [01:37<34:09, 24.40s/it]\u001b[A\n",
      "  6%|▌         | 5/88 [02:01<33:27, 24.18s/it]\u001b[A\n",
      "  7%|▋         | 6/88 [02:27<33:41, 24.65s/it]\u001b[A\n",
      "  8%|▊         | 7/88 [02:54<34:26, 25.52s/it]\u001b[A\n",
      "  9%|▉         | 8/88 [03:18<33:19, 25.00s/it]\u001b[A\n",
      " 10%|█         | 9/88 [03:42<32:25, 24.63s/it]\u001b[A\n",
      " 11%|█▏        | 10/88 [04:07<32:16, 24.83s/it]\u001b[A\n",
      " 12%|█▎        | 11/88 [04:31<31:26, 24.50s/it]\u001b[A\n",
      " 14%|█▎        | 12/88 [04:55<30:45, 24.28s/it]\u001b[A\n",
      " 15%|█▍        | 13/88 [05:18<30:11, 24.15s/it]\u001b[A\n",
      " 16%|█▌        | 14/88 [05:43<30:04, 24.38s/it]\u001b[A\n",
      " 17%|█▋        | 15/88 [06:16<32:33, 26.77s/it]\u001b[A\n",
      " 18%|█▊        | 16/88 [06:45<33:00, 27.51s/it]\u001b[A\n",
      " 19%|█▉        | 17/88 [07:12<32:29, 27.46s/it]\u001b[A\n",
      " 20%|██        | 18/88 [07:43<33:01, 28.31s/it]\u001b[A\n",
      " 22%|██▏       | 19/88 [08:22<36:25, 31.67s/it]\u001b[A\n",
      " 23%|██▎       | 20/88 [08:59<37:37, 33.20s/it]\u001b[A\n",
      " 24%|██▍       | 21/88 [09:35<37:56, 33.97s/it]\u001b[A\n",
      " 25%|██▌       | 22/88 [10:10<37:51, 34.42s/it]\u001b[A\n",
      " 26%|██▌       | 23/88 [10:46<37:43, 34.83s/it]\u001b[A\n",
      " 27%|██▋       | 24/88 [11:26<38:53, 36.46s/it]\u001b[A\n",
      " 28%|██▊       | 25/88 [11:59<37:11, 35.41s/it]\u001b[A\n",
      " 30%|██▉       | 26/88 [12:21<32:25, 31.38s/it]\u001b[A\n",
      " 31%|███       | 27/88 [12:48<30:31, 30.03s/it]\u001b[A\n",
      " 32%|███▏      | 28/88 [13:15<29:02, 29.04s/it]\u001b[A\n",
      " 33%|███▎      | 29/88 [13:42<27:55, 28.40s/it]\u001b[A\n",
      " 34%|███▍      | 30/88 [14:08<26:52, 27.80s/it]\u001b[A\n",
      " 35%|███▌      | 31/88 [14:34<25:49, 27.18s/it]\u001b[A\n",
      " 36%|███▋      | 32/88 [15:03<26:04, 27.94s/it]\u001b[A\n",
      " 38%|███▊      | 33/88 [15:31<25:29, 27.81s/it]\u001b[A\n",
      " 39%|███▊      | 34/88 [15:58<24:47, 27.55s/it]\u001b[A\n",
      " 40%|███▉      | 35/88 [16:25<24:08, 27.33s/it]\u001b[A\n",
      " 41%|████      | 36/88 [16:52<23:38, 27.28s/it]\u001b[A\n",
      " 42%|████▏     | 37/88 [17:19<23:06, 27.19s/it]\u001b[A\n",
      " 43%|████▎     | 38/88 [17:45<22:30, 27.01s/it]\u001b[A\n",
      " 44%|████▍     | 39/88 [18:11<21:43, 26.60s/it]\u001b[A\n",
      " 45%|████▌     | 40/88 [18:38<21:26, 26.80s/it]\u001b[A\n",
      " 47%|████▋     | 41/88 [19:07<21:30, 27.46s/it]\u001b[A\n",
      " 48%|████▊     | 42/88 [19:35<21:04, 27.49s/it]\u001b[A\n",
      " 49%|████▉     | 43/88 [20:03<20:40, 27.58s/it]\u001b[A\n",
      " 50%|█████     | 44/88 [20:31<20:22, 27.80s/it]\u001b[A\n",
      " 51%|█████     | 45/88 [20:58<19:48, 27.65s/it]\u001b[A\n",
      " 52%|█████▏    | 46/88 [21:25<19:10, 27.40s/it]\u001b[A\n",
      " 53%|█████▎    | 47/88 [21:52<18:31, 27.12s/it]\u001b[A\n",
      " 55%|█████▍    | 48/88 [22:18<17:58, 26.97s/it]\u001b[A\n",
      " 56%|█████▌    | 49/88 [22:45<17:26, 26.82s/it]\u001b[A\n",
      " 57%|█████▋    | 50/88 [23:13<17:19, 27.36s/it]\u001b[A\n",
      " 58%|█████▊    | 51/88 [23:41<16:58, 27.52s/it]\u001b[A\n",
      " 59%|█████▉    | 52/88 [24:08<16:20, 27.24s/it]\u001b[A\n",
      " 60%|██████    | 53/88 [24:35<15:50, 27.17s/it]\u001b[A\n",
      " 61%|██████▏   | 54/88 [25:02<15:20, 27.07s/it]\u001b[A\n",
      " 62%|██████▎   | 55/88 [25:28<14:45, 26.84s/it]\u001b[A\n",
      " 64%|██████▎   | 56/88 [25:59<14:56, 28.01s/it]\u001b[A\n",
      " 65%|██████▍   | 57/88 [26:25<14:17, 27.66s/it]\u001b[A\n",
      " 66%|██████▌   | 58/88 [26:50<13:17, 26.59s/it]\u001b[A\n",
      " 67%|██████▋   | 59/88 [27:14<12:31, 25.93s/it]\u001b[A\n",
      " 68%|██████▊   | 60/88 [27:38<11:48, 25.31s/it]\u001b[A\n",
      " 69%|██████▉   | 61/88 [28:02<11:12, 24.91s/it]\u001b[A\n",
      " 70%|███████   | 62/88 [28:26<10:45, 24.84s/it]\u001b[A\n",
      " 72%|███████▏  | 63/88 [28:50<10:14, 24.58s/it]\u001b[A\n",
      " 73%|███████▎  | 64/88 [29:15<09:47, 24.46s/it]\u001b[A\n",
      " 74%|███████▍  | 65/88 [29:39<09:23, 24.48s/it]\u001b[A\n",
      " 75%|███████▌  | 66/88 [30:03<08:55, 24.35s/it]\u001b[A\n",
      " 76%|███████▌  | 67/88 [30:27<08:26, 24.11s/it]\u001b[A\n",
      " 77%|███████▋  | 68/88 [30:50<07:58, 23.92s/it]\u001b[A\n",
      " 78%|███████▊  | 69/88 [31:14<07:34, 23.91s/it]\u001b[A\n",
      " 80%|███████▉  | 70/88 [31:37<07:07, 23.74s/it]\u001b[A\n",
      " 81%|████████  | 71/88 [32:01<06:42, 23.65s/it]\u001b[A\n",
      " 82%|████████▏ | 72/88 [32:25<06:18, 23.68s/it]\u001b[A\n",
      " 83%|████████▎ | 73/88 [32:48<05:54, 23.63s/it]\u001b[A\n",
      " 84%|████████▍ | 74/88 [33:12<05:31, 23.65s/it]\u001b[A\n",
      " 85%|████████▌ | 75/88 [33:36<05:08, 23.76s/it]\u001b[A\n",
      " 86%|████████▋ | 76/88 [33:59<04:44, 23.70s/it]\u001b[A\n",
      " 88%|████████▊ | 77/88 [34:24<04:22, 23.86s/it]\u001b[A\n",
      " 89%|████████▊ | 78/88 [34:48<04:01, 24.10s/it]\u001b[A\n",
      " 90%|████████▉ | 79/88 [35:12<03:36, 24.04s/it]\u001b[A\n",
      " 91%|█████████ | 80/88 [35:36<03:11, 23.88s/it]\u001b[A\n",
      " 92%|█████████▏| 81/88 [36:00<02:47, 23.89s/it]\u001b[A\n",
      " 93%|█████████▎| 82/88 [36:23<02:22, 23.79s/it]\u001b[A\n",
      " 94%|█████████▍| 83/88 [36:47<01:58, 23.72s/it]\u001b[A\n",
      " 95%|█████████▌| 84/88 [37:11<01:35, 23.75s/it]\u001b[A\n",
      " 97%|█████████▋| 85/88 [37:36<01:12, 24.19s/it]\u001b[A\n",
      " 98%|█████████▊| 86/88 [38:02<00:49, 24.89s/it]\u001b[A\n",
      " 99%|█████████▉| 87/88 [38:31<00:26, 26.14s/it]\u001b[A\n",
      "100%|██████████| 88/88 [38:58<00:00, 26.41s/it]\u001b[A\n",
      "\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Permutation feature importance calculation is done. Overall number of features:  88 Number of selected features: 74\n"
     ]
    }
   ],
   "source": [
    "selected_features, importance_df = permutation_feature_selection(model, test_x, test_y, rep=6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['in_out',\n",
       " 'total_pax',\n",
       " 'resort_id',\n",
       " 'roomnights',\n",
       " 'main_product_code',\n",
       " 'checkout_datedayofyear',\n",
       " 'persontravellingid',\n",
       " 'numberofadults',\n",
       " 'resort_id_memberid_prevClick',\n",
       " 'in_book',\n",
       " 'state_code_residence',\n",
       " 'in_outres_meminoutt_mean',\n",
       " 'resort_id_memberid_nextClick',\n",
       " 'total_paxmem_inout_mean_pax',\n",
       " 'checkin_dayes',\n",
       " 'roomnightsmem_amount_mean',\n",
       " 'reservation_idmem_year',\n",
       " 'roomnightsres_mem_amount_std',\n",
       " 'room_type_booked_code',\n",
       " 'reservation_id_res_main_code',\n",
       " 'in_outmem_inout_mean',\n",
       " 'checkin_datedayofyear',\n",
       " 'season_holidayed_code',\n",
       " 'channel_code',\n",
       " 'booking_type_code',\n",
       " 'member_age_buckets',\n",
       " 'reservation_idmem',\n",
       " 'total_paxmem_res_inout_min_pax',\n",
       " 'checkout_dayes',\n",
       " 'numberofchildren',\n",
       " 'total_paxmem_res_inout_mean_pax',\n",
       " 'booking_dayes',\n",
       " 'memberid',\n",
       " 'reservation_id_x_res',\n",
       " 'checkout_dateday',\n",
       " 'in_outmem_inout_std',\n",
       " 'total_paxmem_inout_std_pax',\n",
       " 'in_outres_mem_inout_max',\n",
       " 'reservation_idcheck_Date_month',\n",
       " 'roomnightsres_mem_amount_mean',\n",
       " 'checkout_dateweek_day',\n",
       " 'roomnightsres_mem_amount_min',\n",
       " 'roomnightsmem_amount_std',\n",
       " 'checkin_dateweek_day',\n",
       " 'reservation_id_res_month',\n",
       " 'reservation_idres_mem',\n",
       " 'reservation_id_res_code_month_x',\n",
       " 'roomnightsres_amount_mean',\n",
       " 'booking_dateday',\n",
       " 'roomnightsmem_amount_max',\n",
       " 'reservation_idchannel_check_yearmonth',\n",
       " 'roomnightsres_mem_amount_max',\n",
       " 'cluster_code',\n",
       " 'roomnightsmem_amount_min',\n",
       " 'reservationstatusid_code',\n",
       " 'checkin_dateday',\n",
       " 'booking_dateweek_day',\n",
       " 'total_paxmem_res_inout_max_pax',\n",
       " 'total_paxres_inout_std_pax',\n",
       " 'resort_region_code',\n",
       " 'state_code_resort',\n",
       " 'in_outmem_inout_max',\n",
       " 'resort_id_nextClick',\n",
       " 'booking_datemonth',\n",
       " 'checkout_datemonth',\n",
       " 'total_paxmem_res_inout_std_pax',\n",
       " 'total_paxres_inout_max_pax',\n",
       " 'checkin_datemonth',\n",
       " 'total_paxres_inout_min_pax',\n",
       " 'total_paxmem_inout_max_pax',\n",
       " 'in_outres_mem_inout_min',\n",
       " 'resort_type_code',\n",
       " 'roomnightsres_amount_min',\n",
       " 'checkin_dateis_month_end']"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "selected_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "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>delta_score_mean</th>\n",
       "      <th>delta_score_std</th>\n",
       "      <th>delta_score_max</th>\n",
       "      <th>delta_score_min</th>\n",
       "      <th>feature</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>-0.463994</td>\n",
       "      <td>0.000613</td>\n",
       "      <td>-4.631294e-01</td>\n",
       "      <td>-4.648738e-01</td>\n",
       "      <td>in_out</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>-0.129431</td>\n",
       "      <td>0.000246</td>\n",
       "      <td>-1.292121e-01</td>\n",
       "      <td>-1.297751e-01</td>\n",
       "      <td>total_pax</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>-0.055005</td>\n",
       "      <td>0.000737</td>\n",
       "      <td>-5.398173e-02</td>\n",
       "      <td>-5.619149e-02</td>\n",
       "      <td>resort_id</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>-0.045337</td>\n",
       "      <td>0.000594</td>\n",
       "      <td>-4.477726e-02</td>\n",
       "      <td>-4.630805e-02</td>\n",
       "      <td>roomnights</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>-0.023719</td>\n",
       "      <td>0.000294</td>\n",
       "      <td>-2.326946e-02</td>\n",
       "      <td>-2.420608e-02</td>\n",
       "      <td>main_product_code</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>-0.019990</td>\n",
       "      <td>0.000041</td>\n",
       "      <td>-1.994635e-02</td>\n",
       "      <td>-2.006684e-02</td>\n",
       "      <td>checkout_datedayofyear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>-0.017676</td>\n",
       "      <td>0.000416</td>\n",
       "      <td>-1.695882e-02</td>\n",
       "      <td>-1.816312e-02</td>\n",
       "      <td>persontravellingid</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>-0.010939</td>\n",
       "      <td>0.000141</td>\n",
       "      <td>-1.075202e-02</td>\n",
       "      <td>-1.112467e-02</td>\n",
       "      <td>numberofadults</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>84</th>\n",
       "      <td>-0.009089</td>\n",
       "      <td>0.000143</td>\n",
       "      <td>-8.837406e-03</td>\n",
       "      <td>-9.289125e-03</td>\n",
       "      <td>resort_id_memberid_prevClick</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>-0.007380</td>\n",
       "      <td>0.000053</td>\n",
       "      <td>-7.281405e-03</td>\n",
       "      <td>-7.448653e-03</td>\n",
       "      <td>in_book</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>-0.006113</td>\n",
       "      <td>0.000156</td>\n",
       "      <td>-5.935646e-03</td>\n",
       "      <td>-6.403043e-03</td>\n",
       "      <td>state_code_residence</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>61</th>\n",
       "      <td>-0.005652</td>\n",
       "      <td>0.000065</td>\n",
       "      <td>-5.552603e-03</td>\n",
       "      <td>-5.727515e-03</td>\n",
       "      <td>in_outres_meminoutt_mean</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86</th>\n",
       "      <td>-0.005111</td>\n",
       "      <td>0.000114</td>\n",
       "      <td>-4.983243e-03</td>\n",
       "      <td>-5.259947e-03</td>\n",
       "      <td>resort_id_memberid_nextClick</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>69</th>\n",
       "      <td>-0.004176</td>\n",
       "      <td>0.000043</td>\n",
       "      <td>-4.111305e-03</td>\n",
       "      <td>-4.238803e-03</td>\n",
       "      <td>total_paxmem_inout_mean_pax</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82</th>\n",
       "      <td>-0.003827</td>\n",
       "      <td>0.000112</td>\n",
       "      <td>-3.627562e-03</td>\n",
       "      <td>-3.997082e-03</td>\n",
       "      <td>checkin_dayes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>-0.003491</td>\n",
       "      <td>0.000126</td>\n",
       "      <td>-3.319022e-03</td>\n",
       "      <td>-3.693094e-03</td>\n",
       "      <td>roomnightsmem_amount_mean</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>-0.002975</td>\n",
       "      <td>0.000125</td>\n",
       "      <td>-2.786152e-03</td>\n",
       "      <td>-3.141810e-03</td>\n",
       "      <td>reservation_idmem_year</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>-0.002387</td>\n",
       "      <td>0.000094</td>\n",
       "      <td>-2.223221e-03</td>\n",
       "      <td>-2.495479e-03</td>\n",
       "      <td>roomnightsres_mem_amount_std</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>-0.002215</td>\n",
       "      <td>0.000069</td>\n",
       "      <td>-2.122139e-03</td>\n",
       "      <td>-2.307540e-03</td>\n",
       "      <td>room_type_booked_code</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58</th>\n",
       "      <td>-0.002195</td>\n",
       "      <td>0.000027</td>\n",
       "      <td>-2.168685e-03</td>\n",
       "      <td>-2.252244e-03</td>\n",
       "      <td>reservation_id_res_main_code</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>65</th>\n",
       "      <td>-0.001829</td>\n",
       "      <td>0.000045</td>\n",
       "      <td>-1.776335e-03</td>\n",
       "      <td>-1.898529e-03</td>\n",
       "      <td>in_outmem_inout_mean</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>-0.001764</td>\n",
       "      <td>0.000155</td>\n",
       "      <td>-1.493990e-03</td>\n",
       "      <td>-2.013583e-03</td>\n",
       "      <td>checkin_datedayofyear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>-0.001736</td>\n",
       "      <td>0.000034</td>\n",
       "      <td>-1.693881e-03</td>\n",
       "      <td>-1.796772e-03</td>\n",
       "      <td>season_holidayed_code</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>-0.001703</td>\n",
       "      <td>0.000055</td>\n",
       "      <td>-1.636482e-03</td>\n",
       "      <td>-1.783183e-03</td>\n",
       "      <td>channel_code</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>-0.001655</td>\n",
       "      <td>0.000177</td>\n",
       "      <td>-1.470420e-03</td>\n",
       "      <td>-1.968604e-03</td>\n",
       "      <td>booking_type_code</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>-0.001468</td>\n",
       "      <td>0.000101</td>\n",
       "      <td>-1.333005e-03</td>\n",
       "      <td>-1.633689e-03</td>\n",
       "      <td>member_age_buckets</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>-0.001303</td>\n",
       "      <td>0.000029</td>\n",
       "      <td>-1.275979e-03</td>\n",
       "      <td>-1.365482e-03</td>\n",
       "      <td>reservation_idmem</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>79</th>\n",
       "      <td>-0.001150</td>\n",
       "      <td>0.000070</td>\n",
       "      <td>-1.078454e-03</td>\n",
       "      <td>-1.259854e-03</td>\n",
       "      <td>total_paxmem_res_inout_min_pax</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83</th>\n",
       "      <td>-0.001142</td>\n",
       "      <td>0.000069</td>\n",
       "      <td>-1.045451e-03</td>\n",
       "      <td>-1.269904e-03</td>\n",
       "      <td>checkout_dayes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>-0.001053</td>\n",
       "      <td>0.000134</td>\n",
       "      <td>-9.133899e-04</td>\n",
       "      <td>-1.292227e-03</td>\n",
       "      <td>numberofchildren</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>-0.000049</td>\n",
       "      <td>0.000007</td>\n",
       "      <td>-3.866645e-05</td>\n",
       "      <td>-6.010982e-05</td>\n",
       "      <td>total_paxres_inout_std_pax</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>-0.000041</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>-3.718851e-05</td>\n",
       "      <td>-5.261584e-05</td>\n",
       "      <td>resort_region_code</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>-0.000035</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>-2.817473e-05</td>\n",
       "      <td>-4.354087e-05</td>\n",
       "      <td>state_code_resort</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>-0.000024</td>\n",
       "      <td>0.000024</td>\n",
       "      <td>1.975609e-05</td>\n",
       "      <td>-6.182811e-05</td>\n",
       "      <td>in_outmem_inout_max</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>87</th>\n",
       "      <td>-0.000022</td>\n",
       "      <td>0.000043</td>\n",
       "      <td>3.405922e-05</td>\n",
       "      <td>-9.730907e-05</td>\n",
       "      <td>resort_id_nextClick</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.000018</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>-1.198635e-05</td>\n",
       "      <td>-2.736961e-05</td>\n",
       "      <td>booking_datemonth</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>-0.000012</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>-4.637959e-06</td>\n",
       "      <td>-1.881964e-05</td>\n",
       "      <td>checkout_datemonth</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80</th>\n",
       "      <td>-0.000011</td>\n",
       "      <td>0.000044</td>\n",
       "      <td>5.506130e-05</td>\n",
       "      <td>-6.984380e-05</td>\n",
       "      <td>total_paxmem_res_inout_std_pax</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>74</th>\n",
       "      <td>-0.000011</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>-5.464448e-06</td>\n",
       "      <td>-1.495638e-05</td>\n",
       "      <td>total_paxres_inout_max_pax</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>-0.000010</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>-1.348388e-06</td>\n",
       "      <td>-2.076148e-05</td>\n",
       "      <td>checkin_datemonth</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>-0.000009</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>-1.556335e-06</td>\n",
       "      <td>-1.792878e-05</td>\n",
       "      <td>total_paxres_inout_min_pax</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>70</th>\n",
       "      <td>-0.000008</td>\n",
       "      <td>0.000013</td>\n",
       "      <td>1.720204e-05</td>\n",
       "      <td>-2.139633e-05</td>\n",
       "      <td>total_paxmem_inout_max_pax</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>63</th>\n",
       "      <td>-0.000008</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>7.902975e-07</td>\n",
       "      <td>-1.530265e-05</td>\n",
       "      <td>in_outres_mem_inout_min</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>-0.000005</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>-3.953454e-06</td>\n",
       "      <td>-1.106431e-05</td>\n",
       "      <td>resort_type_code</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>-0.000003</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>5.865176e-07</td>\n",
       "      <td>-7.212169e-06</td>\n",
       "      <td>roomnightsres_amount_min</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>checkin_dateis_month_end</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>8.347111e-06</td>\n",
       "      <td>-5.602088e-07</td>\n",
       "      <td>booking_dateis_month_end</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>0.000005</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>6.893355e-06</td>\n",
       "      <td>2.423832e-06</td>\n",
       "      <td>checkout_dateyear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.000006</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>1.511294e-05</td>\n",
       "      <td>-4.708714e-07</td>\n",
       "      <td>checkin_dateyear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.000008</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>1.197568e-05</td>\n",
       "      <td>5.414997e-06</td>\n",
       "      <td>booking_dateyear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>0.000011</td>\n",
       "      <td>0.000027</td>\n",
       "      <td>4.944004e-05</td>\n",
       "      <td>-2.171268e-05</td>\n",
       "      <td>reservation_id_res_type</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>64</th>\n",
       "      <td>0.000014</td>\n",
       "      <td>0.000034</td>\n",
       "      <td>6.298918e-05</td>\n",
       "      <td>-2.528627e-05</td>\n",
       "      <td>in_outres_meminoutt_std</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.000014</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>2.107943e-05</td>\n",
       "      <td>8.595403e-06</td>\n",
       "      <td>checkout_dateis_month_end</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>0.000016</td>\n",
       "      <td>0.000010</td>\n",
       "      <td>2.887211e-05</td>\n",
       "      <td>1.631448e-06</td>\n",
       "      <td>in_outmem_inout_min</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>2.458534e-05</td>\n",
       "      <td>8.038468e-06</td>\n",
       "      <td>roomnightsres_amount_std</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>71</th>\n",
       "      <td>0.000048</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>7.416311e-05</td>\n",
       "      <td>3.432612e-05</td>\n",
       "      <td>total_paxmem_inout_min_pax</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>0.000068</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>8.891171e-05</td>\n",
       "      <td>4.696805e-05</td>\n",
       "      <td>roomnightsres_amount_max</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>0.000072</td>\n",
       "      <td>0.000022</td>\n",
       "      <td>1.059994e-04</td>\n",
       "      <td>3.970062e-05</td>\n",
       "      <td>total_paxres_inout_mean_pax</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.000135</td>\n",
       "      <td>0.000022</td>\n",
       "      <td>1.560491e-04</td>\n",
       "      <td>9.014925e-05</td>\n",
       "      <td>booking_datedayofyear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85</th>\n",
       "      <td>0.000148</td>\n",
       "      <td>0.000047</td>\n",
       "      <td>2.140120e-04</td>\n",
       "      <td>9.178760e-05</td>\n",
       "      <td>resort_id_prevClick</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>88 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    delta_score_mean  delta_score_std  delta_score_max  delta_score_min  \\\n",
       "22         -0.463994         0.000613    -4.631294e-01    -4.648738e-01   \n",
       "38         -0.129431         0.000246    -1.292121e-01    -1.297751e-01   \n",
       "30         -0.055005         0.000737    -5.398173e-02    -5.619149e-02   \n",
       "34         -0.045337         0.000594    -4.477726e-02    -4.630805e-02   \n",
       "23         -0.023719         0.000294    -2.326946e-02    -2.420608e-02   \n",
       "15         -0.019990         0.000041    -1.994635e-02    -2.006684e-02   \n",
       "28         -0.017676         0.000416    -1.695882e-02    -1.816312e-02   \n",
       "26         -0.010939         0.000141    -1.075202e-02    -1.112467e-02   \n",
       "84         -0.009089         0.000143    -8.837406e-03    -9.289125e-03   \n",
       "21         -0.007380         0.000053    -7.281405e-03    -7.448653e-03   \n",
       "36         -0.006113         0.000156    -5.935646e-03    -6.403043e-03   \n",
       "61         -0.005652         0.000065    -5.552603e-03    -5.727515e-03   \n",
       "86         -0.005111         0.000114    -4.983243e-03    -5.259947e-03   \n",
       "69         -0.004176         0.000043    -4.111305e-03    -4.238803e-03   \n",
       "82         -0.003827         0.000112    -3.627562e-03    -3.997082e-03   \n",
       "52         -0.003491         0.000126    -3.319022e-03    -3.693094e-03   \n",
       "43         -0.002975         0.000125    -2.786152e-03    -3.141810e-03   \n",
       "51         -0.002387         0.000094    -2.223221e-03    -2.495479e-03   \n",
       "33         -0.002215         0.000069    -2.122139e-03    -2.307540e-03   \n",
       "58         -0.002195         0.000027    -2.168685e-03    -2.252244e-03   \n",
       "65         -0.001829         0.000045    -1.776335e-03    -1.898529e-03   \n",
       "9          -0.001764         0.000155    -1.493990e-03    -2.013583e-03   \n",
       "35         -0.001736         0.000034    -1.693881e-03    -1.796772e-03   \n",
       "7          -0.001703         0.000055    -1.636482e-03    -1.783183e-03   \n",
       "6          -0.001655         0.000177    -1.470420e-03    -1.968604e-03   \n",
       "24         -0.001468         0.000101    -1.333005e-03    -1.633689e-03   \n",
       "42         -0.001303         0.000029    -1.275979e-03    -1.365482e-03   \n",
       "79         -0.001150         0.000070    -1.078454e-03    -1.259854e-03   \n",
       "83         -0.001142         0.000069    -1.045451e-03    -1.269904e-03   \n",
       "27         -0.001053         0.000134    -9.133899e-04    -1.292227e-03   \n",
       "..               ...              ...              ...              ...   \n",
       "76         -0.000049         0.000007    -3.866645e-05    -6.010982e-05   \n",
       "31         -0.000041         0.000005    -3.718851e-05    -5.261584e-05   \n",
       "37         -0.000035         0.000005    -2.817473e-05    -4.354087e-05   \n",
       "66         -0.000024         0.000024     1.975609e-05    -6.182811e-05   \n",
       "87         -0.000022         0.000043     3.405922e-05    -9.730907e-05   \n",
       "3          -0.000018         0.000006    -1.198635e-05    -2.736961e-05   \n",
       "17         -0.000012         0.000005    -4.637959e-06    -1.881964e-05   \n",
       "80         -0.000011         0.000044     5.506130e-05    -6.984380e-05   \n",
       "74         -0.000011         0.000003    -5.464448e-06    -1.495638e-05   \n",
       "11         -0.000010         0.000008    -1.348388e-06    -2.076148e-05   \n",
       "75         -0.000009         0.000006    -1.556335e-06    -1.792878e-05   \n",
       "70         -0.000008         0.000013     1.720204e-05    -2.139633e-05   \n",
       "63         -0.000008         0.000005     7.902975e-07    -1.530265e-05   \n",
       "32         -0.000005         0.000003    -3.953454e-06    -1.106431e-05   \n",
       "46         -0.000003         0.000002     5.865176e-07    -7.212169e-06   \n",
       "10          0.000000         0.000000     0.000000e+00     0.000000e+00   \n",
       "2           0.000002         0.000003     8.347111e-06    -5.602088e-07   \n",
       "19          0.000005         0.000002     6.893355e-06     2.423832e-06   \n",
       "13          0.000006         0.000005     1.511294e-05    -4.708714e-07   \n",
       "5           0.000008         0.000002     1.197568e-05     5.414997e-06   \n",
       "56          0.000011         0.000027     4.944004e-05    -2.171268e-05   \n",
       "64          0.000014         0.000034     6.298918e-05    -2.528627e-05   \n",
       "16          0.000014         0.000004     2.107943e-05     8.595403e-06   \n",
       "67          0.000016         0.000010     2.887211e-05     1.631448e-06   \n",
       "47          0.000019         0.000006     2.458534e-05     8.038468e-06   \n",
       "71          0.000048         0.000014     7.416311e-05     3.432612e-05   \n",
       "45          0.000068         0.000014     8.891171e-05     4.696805e-05   \n",
       "73          0.000072         0.000022     1.059994e-04     3.970062e-05   \n",
       "1           0.000135         0.000022     1.560491e-04     9.014925e-05   \n",
       "85          0.000148         0.000047     2.140120e-04     9.178760e-05   \n",
       "\n",
       "                           feature  \n",
       "22                          in_out  \n",
       "38                       total_pax  \n",
       "30                       resort_id  \n",
       "34                      roomnights  \n",
       "23               main_product_code  \n",
       "15          checkout_datedayofyear  \n",
       "28              persontravellingid  \n",
       "26                  numberofadults  \n",
       "84    resort_id_memberid_prevClick  \n",
       "21                         in_book  \n",
       "36            state_code_residence  \n",
       "61        in_outres_meminoutt_mean  \n",
       "86    resort_id_memberid_nextClick  \n",
       "69     total_paxmem_inout_mean_pax  \n",
       "82                   checkin_dayes  \n",
       "52       roomnightsmem_amount_mean  \n",
       "43          reservation_idmem_year  \n",
       "51    roomnightsres_mem_amount_std  \n",
       "33           room_type_booked_code  \n",
       "58    reservation_id_res_main_code  \n",
       "65            in_outmem_inout_mean  \n",
       "9            checkin_datedayofyear  \n",
       "35           season_holidayed_code  \n",
       "7                     channel_code  \n",
       "6                booking_type_code  \n",
       "24              member_age_buckets  \n",
       "42               reservation_idmem  \n",
       "79  total_paxmem_res_inout_min_pax  \n",
       "83                  checkout_dayes  \n",
       "27                numberofchildren  \n",
       "..                             ...  \n",
       "76      total_paxres_inout_std_pax  \n",
       "31              resort_region_code  \n",
       "37               state_code_resort  \n",
       "66             in_outmem_inout_max  \n",
       "87             resort_id_nextClick  \n",
       "3                booking_datemonth  \n",
       "17              checkout_datemonth  \n",
       "80  total_paxmem_res_inout_std_pax  \n",
       "74      total_paxres_inout_max_pax  \n",
       "11               checkin_datemonth  \n",
       "75      total_paxres_inout_min_pax  \n",
       "70      total_paxmem_inout_max_pax  \n",
       "63         in_outres_mem_inout_min  \n",
       "32                resort_type_code  \n",
       "46        roomnightsres_amount_min  \n",
       "10        checkin_dateis_month_end  \n",
       "2         booking_dateis_month_end  \n",
       "19               checkout_dateyear  \n",
       "13                checkin_dateyear  \n",
       "5                 booking_dateyear  \n",
       "56         reservation_id_res_type  \n",
       "64         in_outres_meminoutt_std  \n",
       "16       checkout_dateis_month_end  \n",
       "67             in_outmem_inout_min  \n",
       "47        roomnightsres_amount_std  \n",
       "71      total_paxmem_inout_min_pax  \n",
       "45        roomnightsres_amount_max  \n",
       "73     total_paxres_inout_mean_pax  \n",
       "1            booking_datedayofyear  \n",
       "85             resort_id_prevClick  \n",
       "\n",
       "[88 rows x 5 columns]"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "importance_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Run Stacking of Various ML Algos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from scipy.stats import skew\n",
    "import xgboost as xgb\n",
    "from sklearn.ensemble import ExtraTreesRegressor\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from math import sqrt\n",
    "from sklearn.metrics import roc_auc_score\n",
    "from sklearn.model_selection import KFold\n",
    "from catboost import CatBoostClassifier\n",
    "from lightgbm import LGBMClassifier\n",
    "import gc\n",
    "\n",
    "NFOLDS = 3\n",
    "SEED = 0\n",
    "NROWS = None\n",
    "categorical_feats =categor_col\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "list.remove(x): x not in list",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-120-b95da8fc877d>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mcategor_col\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mremove\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'memberid'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m: list.remove(x): x not in list"
     ]
    }
   ],
   "source": [
    "categor_col.remove('memberid')\n",
    "X.drop('target',axis=1,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:33: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# for f_ in categorical_feats:\n",
    "#     data[f_], indexer = pd.factorize(data[f_])\n",
    "#     test[f_] = indexer.get_indexer(test[f_])\n",
    "    \n",
    "# gc.enable()\n",
    "\n",
    "# y_train = data['TARGET']\n",
    "# del data['TARGET']\n",
    "\n",
    "# ###################################\n",
    "# # PLEASE DON'T DO THIS AT HOME LOL\n",
    "# # Averaging factorized categorical features defeats my own reasoning\n",
    "# ################################### \n",
    "# prev_cat_features = [\n",
    "#     f_ for f_ in prev.columns if prev[f_].dtype == 'object'\n",
    "# ]\n",
    "# for f_ in prev_cat_features:\n",
    "#     prev[f_], _ = pd.factorize(prev[f_])\n",
    "    \n",
    "# avg_prev = prev.groupby('SK_ID_CURR').mean()\n",
    "# cnt_prev = prev[['SK_ID_CURR', 'SK_ID_PREV']].groupby('SK_ID_CURR').count()\n",
    "# avg_prev['nb_app'] = cnt_prev['SK_ID_PREV']\n",
    "# del avg_prev['SK_ID_PREV']\n",
    "\n",
    "# x_train = data.merge(right=avg_prev.reset_index(), how='left', on='SK_ID_CURR')\n",
    "# x_test = test.merge(right=avg_prev.reset_index(), how='left', on='SK_ID_CURR')\n",
    "\n",
    "# x_train = x_train.fillna(0)\n",
    "# x_test= x_test.fillna(0)\n",
    "for i in categor_col:\n",
    "    X[i]=X[i].astype(int)\n",
    "    test[i]=test[i].astype(int)    \n",
    "\n",
    "# excluded_feats = ['SK_ID_CURR']\n",
    "features = X.columns\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train = X[features]\n",
    "x_test = test[features]\n",
    "ntrain = x_train.shape[0]\n",
    "ntest = x_test.shape[0]\n",
    "\n",
    "\n",
    "y_train=y\n",
    "kf = KFold(n_splits = NFOLDS, shuffle=True, random_state=SEED)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "class SklearnWrapper(object):\n",
    "    def __init__(self, clf, seed=0, params=None):\n",
    "        params['random_state'] = seed\n",
    "        self.clf = clf(**params)\n",
    "\n",
    "    def train(self, x_train, y_train):\n",
    "        print(x_train.isnull().sum().sum())\n",
    "        self.clf.fit(x_train.fillna(-999).astype(float), y_train)\n",
    "\n",
    "    def predict(self, x):\n",
    "        return self.clf.predict(x.fillna(-999).astype(float))\n",
    "\n",
    "class CatboostWrapper(object):\n",
    "    def __init__(self, clf, seed=0, params=None):\n",
    "        params['random_seed'] = seed\n",
    "        self.clf = clf(**params)\n",
    "\n",
    "    def train(self, x_train, y_train):\n",
    "        self.clf.fit(x_train, y_train)\n",
    "\n",
    "    def predict(self, x):\n",
    "        return self.clf.predict(x)\n",
    "        \n",
    "class LightGBMWrapper(object):\n",
    "    def __init__(self, clf, seed=0, params=None):\n",
    "        params['feature_fraction_seed'] = seed\n",
    "        params['bagging_seed'] = seed\n",
    "        self.clf = clf(**params)\n",
    "\n",
    "    def train(self, x_train, y_train):\n",
    "        self.clf.fit(x_train, y_train)\n",
    "\n",
    "    def predict(self, x):\n",
    "        return self.clf.predict(x)\n",
    "\n",
    "\n",
    "class XgbWrapper(object):\n",
    "    def __init__(self, seed=0, params=None):\n",
    "        self.param = params\n",
    "        self.param['seed'] = seed\n",
    "        self.nrounds = params.pop('nrounds', 250)\n",
    "\n",
    "    def train(self, x_train, y_train):\n",
    "        dtrain = xgb.DMatrix(x_train, label=y_train)\n",
    "        self.gbdt = xgb.train(self.param, dtrain, self.nrounds)\n",
    "\n",
    "    def predict(self, x):\n",
    "        return self.gbdt.predict(xgb.DMatrix(x))\n",
    "\n",
    "\n",
    "def get_oof(clf):\n",
    "    oof_train = np.zeros((ntrain,))\n",
    "    oof_test = np.zeros((ntest,))\n",
    "    oof_test_skf = np.empty((NFOLDS, ntest))\n",
    "\n",
    "    for i, (train_index, test_index) in enumerate(kf.split(x_train)):\n",
    "        x_tr = x_train.loc[train_index]\n",
    "        y_tr = y_train.loc[train_index]\n",
    "        x_te = x_train.loc[test_index]\n",
    "\n",
    "        clf.train(x_tr, y_tr)\n",
    "\n",
    "        oof_train[test_index] = clf.predict(x_te)\n",
    "        oof_test_skf[i, :] = clf.predict(x_test)\n",
    "\n",
    "    oof_test[:] = oof_test_skf.mean(axis=0)\n",
    "    return oof_train.reshape(-1, 1), oof_test.reshape(-1, 1)\n",
    "\n",
    "def get_oof_et(clf):\n",
    "    oof_train = np.zeros((ntrain,))\n",
    "    oof_test = np.zeros((ntest,))\n",
    "    oof_test_skf = np.empty((NFOLDS, ntest))\n",
    "\n",
    "    for i, (train_index, test_index) in enumerate(kf.split(x_train)):\n",
    "        x_tr = x_train.loc[train_index]\n",
    "        y_tr = y_train.loc[train_index]\n",
    "        x_te = x_train.loc[test_index]\n",
    "\n",
    "        clf.train(x_tr, y_tr)\n",
    "\n",
    "        oof_train[test_index] = clf.predict(x_te)\n",
    "        oof_test_skf[i, :] = clf.predict(x_test.fillna(-999))\n",
    "\n",
    "    oof_test[:] = oof_test_skf.mean(axis=0)\n",
    "    return oof_train.reshape(-1, 1), oof_test.reshape(-1, 1)\n",
    "\n",
    "\n",
    "\n",
    "et_params = {\n",
    "    'n_jobs': 16,\n",
    "    'n_estimators': 200,\n",
    "    'max_features': 0.5,\n",
    "    'max_depth': 12,\n",
    "    'min_samples_leaf': 2,\n",
    "}\n",
    "\n",
    "rf_params = {\n",
    "    'n_jobs': 16,\n",
    "    'n_estimators': 200,\n",
    "    'max_features': 0.2,\n",
    "    'max_depth': 12,\n",
    "    'min_samples_leaf': 2,\n",
    "}\n",
    "\n",
    "xgb_params = {\n",
    "    'seed': 0,\n",
    "    'colsample_bytree': 0.7,\n",
    "    'silent': 1,\n",
    "    'subsample': 0.7,\n",
    "    'learning_rate': 0.075,\n",
    "    'objective': 'reg:linear',\n",
    "    'max_depth': 4,\n",
    "    'num_parallel_tree': 1,\n",
    "    'min_child_weight': 1,\n",
    "    'nrounds': 400\n",
    "}\n",
    "\n",
    "catboost_params = {\n",
    "    'iterations': 1400,\n",
    "    'learning_rate': 0.05,\n",
    "    'depth': 8,\n",
    "    'l2_leaf_reg': 17,\n",
    "\n",
    "    'eval_metric': 'RMSE',\n",
    "    'od_type': 'Iter',\n",
    "    'allow_writing_files': False\n",
    "}\n",
    "\n",
    "lightgbm_params  = {\"objective\" : \"regression\", \"metric\" : \"rmse\", \n",
    "                    \n",
    "              \"num_leaves\" : 31, \"learning_rate\" : 0.05, \"bagging_fraction\" : 0.9,\n",
    "               \"bagging_seed\" : 0, \"num_threads\" : 4,\"colsample_bytree\" : 0.8,\"lambda_l1\": 25,\"min_child_weight\" : 44.97\n",
    "          ,\"min_child_samples\": 5,'iterations': 500,\n",
    "             }\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "xg = XgbWrapper(seed=SEED, params=xgb_params)\n",
    "#et = SklearnWrapper(clf=ExtraTreesRegressor, seed=SEED, params=et_params)\n",
    "rf = SklearnWrapper(clf=RandomForestRegressor, seed=SEED, params=rf_params)\n",
    "cb = CatboostWrapper(clf= CatBoostRegressor, seed = SEED, params=catboost_params)\n",
    "lg = LightGBMWrapper(clf = LGBMRegressor, seed = SEED, params = lightgbm_params)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2\n",
      "0:\tlearn: 7.4163577\ttotal: 107ms\tremaining: 2m 29s\n",
      "1:\tlearn: 7.0568571\ttotal: 263ms\tremaining: 3m 3s\n",
      "2:\tlearn: 6.7138837\ttotal: 412ms\tremaining: 3m 11s\n",
      "3:\tlearn: 6.3880818\ttotal: 528ms\tremaining: 3m 4s\n",
      "4:\tlearn: 6.0792619\ttotal: 661ms\tremaining: 3m 4s\n",
      "5:\tlearn: 5.7859463\ttotal: 798ms\tremaining: 3m 5s\n",
      "6:\tlearn: 5.5080233\ttotal: 968ms\tremaining: 3m 12s\n",
      "7:\tlearn: 5.2454829\ttotal: 1.1s\tremaining: 3m 11s\n",
      "8:\tlearn: 4.9952675\ttotal: 1.27s\tremaining: 3m 17s\n",
      "9:\tlearn: 4.7581194\ttotal: 1.38s\tremaining: 3m 11s\n",
      "10:\tlearn: 4.5343603\ttotal: 1.55s\tremaining: 3m 15s\n",
      "11:\tlearn: 4.3210761\ttotal: 1.66s\tremaining: 3m 11s\n",
      "12:\tlearn: 4.1191321\ttotal: 1.82s\tremaining: 3m 14s\n",
      "13:\tlearn: 3.9287337\ttotal: 2s\tremaining: 3m 17s\n",
      "14:\tlearn: 3.7473058\ttotal: 2.21s\tremaining: 3m 23s\n",
      "15:\tlearn: 3.5754783\ttotal: 2.36s\tremaining: 3m 23s\n",
      "16:\tlearn: 3.4140547\ttotal: 2.58s\tremaining: 3m 29s\n",
      "17:\tlearn: 3.2613450\ttotal: 2.83s\tremaining: 3m 37s\n",
      "18:\tlearn: 3.1164968\ttotal: 3.21s\tremaining: 3m 53s\n",
      "19:\tlearn: 2.9801494\ttotal: 3.4s\tremaining: 3m 54s\n",
      "20:\tlearn: 2.8506703\ttotal: 3.54s\tremaining: 3m 52s\n",
      "21:\tlearn: 2.7291379\ttotal: 3.71s\tremaining: 3m 52s\n",
      "22:\tlearn: 2.6140701\ttotal: 3.94s\tremaining: 3m 55s\n",
      "23:\tlearn: 2.5057097\ttotal: 4.06s\tremaining: 3m 52s\n",
      "24:\tlearn: 2.4030912\ttotal: 4.26s\tremaining: 3m 54s\n",
      "25:\tlearn: 2.3066093\ttotal: 4.47s\tremaining: 3m 56s\n",
      "26:\tlearn: 2.2162047\ttotal: 4.66s\tremaining: 3m 57s\n",
      "27:\tlearn: 2.1305401\ttotal: 4.83s\tremaining: 3m 56s\n",
      "28:\tlearn: 2.0500906\ttotal: 4.96s\tremaining: 3m 54s\n",
      "29:\tlearn: 1.9741504\ttotal: 5.16s\tremaining: 3m 55s\n",
      "30:\tlearn: 1.9030762\ttotal: 5.3s\tremaining: 3m 54s\n",
      "31:\tlearn: 1.8369038\ttotal: 5.5s\tremaining: 3m 55s\n",
      "32:\tlearn: 1.7751097\ttotal: 5.71s\tremaining: 3m 56s\n",
      "33:\tlearn: 1.7167544\ttotal: 5.9s\tremaining: 3m 57s\n",
      "34:\tlearn: 1.6621984\ttotal: 6.06s\tremaining: 3m 56s\n",
      "35:\tlearn: 1.6111341\ttotal: 6.22s\tremaining: 3m 55s\n",
      "36:\tlearn: 1.5635078\ttotal: 6.43s\tremaining: 3m 56s\n",
      "37:\tlearn: 1.5189801\ttotal: 6.57s\tremaining: 3m 55s\n",
      "38:\tlearn: 1.4776752\ttotal: 6.82s\tremaining: 3m 57s\n",
      "39:\tlearn: 1.4392596\ttotal: 6.98s\tremaining: 3m 57s\n",
      "40:\tlearn: 1.4040316\ttotal: 7.12s\tremaining: 3m 56s\n",
      "41:\tlearn: 1.3706291\ttotal: 7.28s\tremaining: 3m 55s\n",
      "42:\tlearn: 1.3398449\ttotal: 7.43s\tremaining: 3m 54s\n",
      "43:\tlearn: 1.3120075\ttotal: 7.58s\tremaining: 3m 53s\n",
      "44:\tlearn: 1.2856060\ttotal: 7.76s\tremaining: 3m 53s\n",
      "45:\tlearn: 1.2614392\ttotal: 7.91s\tremaining: 3m 52s\n",
      "46:\tlearn: 1.2389150\ttotal: 8.11s\tremaining: 3m 53s\n",
      "47:\tlearn: 1.2179741\ttotal: 8.26s\tremaining: 3m 52s\n",
      "48:\tlearn: 1.1985796\ttotal: 8.46s\tremaining: 3m 53s\n",
      "49:\tlearn: 1.1804869\ttotal: 8.64s\tremaining: 3m 53s\n",
      "50:\tlearn: 1.1647039\ttotal: 8.77s\tremaining: 3m 52s\n",
      "51:\tlearn: 1.1499587\ttotal: 8.91s\tremaining: 3m 51s\n",
      "52:\tlearn: 1.1360628\ttotal: 9.09s\tremaining: 3m 51s\n",
      "53:\tlearn: 1.1235748\ttotal: 9.25s\tremaining: 3m 50s\n",
      "54:\tlearn: 1.1117478\ttotal: 9.46s\tremaining: 3m 51s\n",
      "55:\tlearn: 1.1011962\ttotal: 9.75s\tremaining: 3m 53s\n",
      "56:\tlearn: 1.0914711\ttotal: 10s\tremaining: 3m 55s\n",
      "57:\tlearn: 1.0824260\ttotal: 10.2s\tremaining: 3m 57s\n",
      "58:\tlearn: 1.0741608\ttotal: 10.6s\tremaining: 4m 1s\n",
      "59:\tlearn: 1.0663464\ttotal: 10.8s\tremaining: 4m 1s\n",
      "60:\tlearn: 1.0595319\ttotal: 11s\tremaining: 4m 1s\n",
      "61:\tlearn: 1.0530120\ttotal: 11.2s\tremaining: 4m 1s\n",
      "62:\tlearn: 1.0470890\ttotal: 11.3s\tremaining: 4m\n",
      "63:\tlearn: 1.0416400\ttotal: 11.5s\tremaining: 4m\n",
      "64:\tlearn: 1.0365561\ttotal: 11.7s\tremaining: 4m\n",
      "65:\tlearn: 1.0318763\ttotal: 11.9s\tremaining: 4m\n",
      "66:\tlearn: 1.0275350\ttotal: 12.1s\tremaining: 4m\n",
      "67:\tlearn: 1.0238022\ttotal: 12.3s\tremaining: 4m\n",
      "68:\tlearn: 1.0203947\ttotal: 12.4s\tremaining: 4m\n",
      "69:\tlearn: 1.0171970\ttotal: 12.6s\tremaining: 3m 58s\n",
      "70:\tlearn: 1.0140609\ttotal: 12.7s\tremaining: 3m 58s\n",
      "71:\tlearn: 1.0113266\ttotal: 12.8s\tremaining: 3m 56s\n",
      "72:\tlearn: 1.0086814\ttotal: 13s\tremaining: 3m 55s\n",
      "73:\tlearn: 1.0063506\ttotal: 13.1s\tremaining: 3m 54s\n",
      "74:\tlearn: 1.0042629\ttotal: 13.2s\tremaining: 3m 53s\n",
      "75:\tlearn: 1.0021773\ttotal: 13.4s\tremaining: 3m 53s\n",
      "76:\tlearn: 1.0003243\ttotal: 13.5s\tremaining: 3m 52s\n",
      "77:\tlearn: 0.9987576\ttotal: 13.7s\tremaining: 3m 52s\n",
      "78:\tlearn: 0.9972587\ttotal: 13.8s\tremaining: 3m 51s\n",
      "79:\tlearn: 0.9957278\ttotal: 14s\tremaining: 3m 51s\n",
      "80:\tlearn: 0.9943949\ttotal: 14.2s\tremaining: 3m 50s\n",
      "81:\tlearn: 0.9930108\ttotal: 14.4s\tremaining: 3m 50s\n",
      "82:\tlearn: 0.9919102\ttotal: 14.5s\tremaining: 3m 50s\n",
      "83:\tlearn: 0.9907966\ttotal: 14.7s\tremaining: 3m 50s\n",
      "84:\tlearn: 0.9898802\ttotal: 14.8s\tremaining: 3m 49s\n",
      "85:\tlearn: 0.9889393\ttotal: 15s\tremaining: 3m 48s\n",
      "86:\tlearn: 0.9880934\ttotal: 15.3s\tremaining: 3m 50s\n",
      "87:\tlearn: 0.9872522\ttotal: 15.6s\tremaining: 3m 52s\n",
      "88:\tlearn: 0.9864351\ttotal: 15.8s\tremaining: 3m 52s\n",
      "89:\tlearn: 0.9857407\ttotal: 16s\tremaining: 3m 53s\n",
      "90:\tlearn: 0.9849645\ttotal: 16.3s\tremaining: 3m 54s\n",
      "91:\tlearn: 0.9843598\ttotal: 16.4s\tremaining: 3m 53s\n",
      "92:\tlearn: 0.9837801\ttotal: 16.6s\tremaining: 3m 53s\n",
      "93:\tlearn: 0.9832278\ttotal: 16.7s\tremaining: 3m 52s\n",
      "94:\tlearn: 0.9826102\ttotal: 16.8s\tremaining: 3m 51s\n",
      "95:\tlearn: 0.9821540\ttotal: 17.4s\tremaining: 3m 56s\n",
      "96:\tlearn: 0.9816233\ttotal: 17.6s\tremaining: 3m 55s\n",
      "97:\tlearn: 0.9811737\ttotal: 17.7s\tremaining: 3m 55s\n",
      "98:\tlearn: 0.9808040\ttotal: 17.9s\tremaining: 3m 54s\n",
      "99:\tlearn: 0.9803625\ttotal: 18s\tremaining: 3m 54s\n",
      "100:\tlearn: 0.9799932\ttotal: 18.1s\tremaining: 3m 53s\n",
      "101:\tlearn: 0.9796524\ttotal: 18.3s\tremaining: 3m 52s\n",
      "102:\tlearn: 0.9793280\ttotal: 18.4s\tremaining: 3m 51s\n",
      "103:\tlearn: 0.9789901\ttotal: 18.5s\tremaining: 3m 50s\n",
      "104:\tlearn: 0.9786724\ttotal: 18.6s\tremaining: 3m 49s\n",
      "105:\tlearn: 0.9783721\ttotal: 18.7s\tremaining: 3m 48s\n",
      "106:\tlearn: 0.9780382\ttotal: 18.9s\tremaining: 3m 48s\n",
      "107:\tlearn: 0.9777532\ttotal: 19s\tremaining: 3m 47s\n",
      "108:\tlearn: 0.9774560\ttotal: 19.2s\tremaining: 3m 47s\n",
      "109:\tlearn: 0.9772118\ttotal: 19.3s\tremaining: 3m 46s\n",
      "110:\tlearn: 0.9769831\ttotal: 19.5s\tremaining: 3m 46s\n",
      "111:\tlearn: 0.9767552\ttotal: 19.6s\tremaining: 3m 45s\n",
      "112:\tlearn: 0.9764664\ttotal: 19.8s\tremaining: 3m 45s\n",
      "113:\tlearn: 0.9762273\ttotal: 20.1s\tremaining: 3m 47s\n",
      "114:\tlearn: 0.9760012\ttotal: 20.4s\tremaining: 3m 47s\n",
      "115:\tlearn: 0.9757455\ttotal: 20.5s\tremaining: 3m 47s\n",
      "116:\tlearn: 0.9755130\ttotal: 20.8s\tremaining: 3m 48s\n",
      "117:\tlearn: 0.9752625\ttotal: 21s\tremaining: 3m 48s\n",
      "118:\tlearn: 0.9750172\ttotal: 21.2s\tremaining: 3m 48s\n",
      "119:\tlearn: 0.9748290\ttotal: 21.3s\tremaining: 3m 47s\n",
      "120:\tlearn: 0.9746101\ttotal: 21.5s\tremaining: 3m 47s\n",
      "121:\tlearn: 0.9744380\ttotal: 21.6s\tremaining: 3m 46s\n",
      "122:\tlearn: 0.9742782\ttotal: 21.8s\tremaining: 3m 46s\n",
      "123:\tlearn: 0.9741046\ttotal: 22.1s\tremaining: 3m 46s\n",
      "124:\tlearn: 0.9739394\ttotal: 22.2s\tremaining: 3m 46s\n",
      "125:\tlearn: 0.9737322\ttotal: 22.4s\tremaining: 3m 46s\n",
      "126:\tlearn: 0.9735361\ttotal: 22.7s\tremaining: 3m 47s\n",
      "127:\tlearn: 0.9733485\ttotal: 22.9s\tremaining: 3m 47s\n",
      "128:\tlearn: 0.9732247\ttotal: 23.1s\tremaining: 3m 47s\n",
      "129:\tlearn: 0.9731125\ttotal: 23.3s\tremaining: 3m 47s\n",
      "130:\tlearn: 0.9730041\ttotal: 23.5s\tremaining: 3m 47s\n",
      "131:\tlearn: 0.9728461\ttotal: 23.9s\tremaining: 3m 49s\n",
      "132:\tlearn: 0.9726446\ttotal: 24.1s\tremaining: 3m 49s\n",
      "133:\tlearn: 0.9724812\ttotal: 24.3s\tremaining: 3m 49s\n",
      "134:\tlearn: 0.9723286\ttotal: 24.4s\tremaining: 3m 49s\n",
      "135:\tlearn: 0.9722169\ttotal: 24.6s\tremaining: 3m 49s\n",
      "136:\tlearn: 0.9720863\ttotal: 24.7s\tremaining: 3m 48s\n",
      "137:\tlearn: 0.9719725\ttotal: 24.8s\tremaining: 3m 47s\n",
      "138:\tlearn: 0.9718167\ttotal: 25s\tremaining: 3m 46s\n",
      "139:\tlearn: 0.9717238\ttotal: 25.1s\tremaining: 3m 45s\n",
      "140:\tlearn: 0.9716241\ttotal: 25.2s\tremaining: 3m 45s\n",
      "141:\tlearn: 0.9715009\ttotal: 25.4s\tremaining: 3m 44s\n",
      "142:\tlearn: 0.9713737\ttotal: 25.5s\tremaining: 3m 44s\n",
      "143:\tlearn: 0.9712626\ttotal: 25.6s\tremaining: 3m 43s\n",
      "144:\tlearn: 0.9711379\ttotal: 25.8s\tremaining: 3m 43s\n",
      "145:\tlearn: 0.9710053\ttotal: 25.9s\tremaining: 3m 42s\n",
      "146:\tlearn: 0.9708949\ttotal: 26.1s\tremaining: 3m 42s\n",
      "147:\tlearn: 0.9708151\ttotal: 26.2s\tremaining: 3m 41s\n",
      "148:\tlearn: 0.9706781\ttotal: 26.3s\tremaining: 3m 41s\n",
      "149:\tlearn: 0.9706041\ttotal: 26.4s\tremaining: 3m 40s\n",
      "150:\tlearn: 0.9704872\ttotal: 26.6s\tremaining: 3m 39s\n",
      "151:\tlearn: 0.9703902\ttotal: 26.7s\tremaining: 3m 39s\n",
      "152:\tlearn: 0.9702714\ttotal: 26.8s\tremaining: 3m 38s\n",
      "153:\tlearn: 0.9701229\ttotal: 26.9s\tremaining: 3m 37s\n",
      "154:\tlearn: 0.9700150\ttotal: 27s\tremaining: 3m 36s\n",
      "155:\tlearn: 0.9698584\ttotal: 27.1s\tremaining: 3m 36s\n",
      "156:\tlearn: 0.9697371\ttotal: 27.2s\tremaining: 3m 35s\n",
      "157:\tlearn: 0.9696395\ttotal: 27.3s\tremaining: 3m 34s\n",
      "158:\tlearn: 0.9695890\ttotal: 27.4s\tremaining: 3m 34s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "159:\tlearn: 0.9694790\ttotal: 27.6s\tremaining: 3m 33s\n",
      "160:\tlearn: 0.9694154\ttotal: 27.6s\tremaining: 3m 32s\n",
      "161:\tlearn: 0.9693470\ttotal: 27.7s\tremaining: 3m 32s\n",
      "162:\tlearn: 0.9692606\ttotal: 27.9s\tremaining: 3m 31s\n",
      "163:\tlearn: 0.9691457\ttotal: 28s\tremaining: 3m 30s\n",
      "164:\tlearn: 0.9690642\ttotal: 28.1s\tremaining: 3m 30s\n",
      "165:\tlearn: 0.9689946\ttotal: 28.2s\tremaining: 3m 29s\n",
      "166:\tlearn: 0.9688896\ttotal: 28.3s\tremaining: 3m 29s\n",
      "167:\tlearn: 0.9688050\ttotal: 28.5s\tremaining: 3m 28s\n",
      "168:\tlearn: 0.9686896\ttotal: 28.7s\tremaining: 3m 29s\n",
      "169:\tlearn: 0.9685779\ttotal: 29s\tremaining: 3m 29s\n",
      "170:\tlearn: 0.9685099\ttotal: 29.1s\tremaining: 3m 29s\n",
      "171:\tlearn: 0.9683734\ttotal: 29.3s\tremaining: 3m 28s\n",
      "172:\tlearn: 0.9682800\ttotal: 29.4s\tremaining: 3m 28s\n",
      "173:\tlearn: 0.9681907\ttotal: 29.5s\tremaining: 3m 27s\n",
      "174:\tlearn: 0.9681271\ttotal: 29.7s\tremaining: 3m 27s\n",
      "175:\tlearn: 0.9680165\ttotal: 29.8s\tremaining: 3m 27s\n",
      "176:\tlearn: 0.9679387\ttotal: 29.9s\tremaining: 3m 26s\n",
      "177:\tlearn: 0.9679106\ttotal: 30s\tremaining: 3m 25s\n",
      "178:\tlearn: 0.9678768\ttotal: 30.1s\tremaining: 3m 25s\n",
      "179:\tlearn: 0.9677675\ttotal: 30.2s\tremaining: 3m 24s\n",
      "180:\tlearn: 0.9676733\ttotal: 30.3s\tremaining: 3m 23s\n",
      "181:\tlearn: 0.9676169\ttotal: 30.4s\tremaining: 3m 23s\n",
      "182:\tlearn: 0.9675064\ttotal: 30.5s\tremaining: 3m 22s\n",
      "183:\tlearn: 0.9674441\ttotal: 30.6s\tremaining: 3m 22s\n",
      "184:\tlearn: 0.9673548\ttotal: 30.7s\tremaining: 3m 21s\n",
      "185:\tlearn: 0.9672555\ttotal: 30.8s\tremaining: 3m 21s\n",
      "186:\tlearn: 0.9671743\ttotal: 30.9s\tremaining: 3m 20s\n",
      "187:\tlearn: 0.9671164\ttotal: 31.1s\tremaining: 3m 20s\n",
      "188:\tlearn: 0.9670399\ttotal: 31.2s\tremaining: 3m 19s\n",
      "189:\tlearn: 0.9669682\ttotal: 31.3s\tremaining: 3m 19s\n",
      "190:\tlearn: 0.9669204\ttotal: 31.4s\tremaining: 3m 18s\n",
      "191:\tlearn: 0.9668466\ttotal: 31.5s\tremaining: 3m 18s\n",
      "192:\tlearn: 0.9667953\ttotal: 31.6s\tremaining: 3m 17s\n",
      "193:\tlearn: 0.9667575\ttotal: 31.7s\tremaining: 3m 17s\n",
      "194:\tlearn: 0.9666495\ttotal: 31.9s\tremaining: 3m 16s\n",
      "195:\tlearn: 0.9665714\ttotal: 32s\tremaining: 3m 16s\n",
      "196:\tlearn: 0.9664800\ttotal: 32.1s\tremaining: 3m 15s\n",
      "197:\tlearn: 0.9664291\ttotal: 32.2s\tremaining: 3m 15s\n",
      "198:\tlearn: 0.9663642\ttotal: 32.3s\tremaining: 3m 14s\n",
      "199:\tlearn: 0.9663087\ttotal: 32.4s\tremaining: 3m 14s\n",
      "200:\tlearn: 0.9662441\ttotal: 32.5s\tremaining: 3m 13s\n",
      "201:\tlearn: 0.9661963\ttotal: 32.6s\tremaining: 3m 13s\n",
      "202:\tlearn: 0.9661036\ttotal: 32.7s\tremaining: 3m 13s\n",
      "203:\tlearn: 0.9660517\ttotal: 32.9s\tremaining: 3m 12s\n",
      "204:\tlearn: 0.9659857\ttotal: 33s\tremaining: 3m 12s\n",
      "205:\tlearn: 0.9659029\ttotal: 33.1s\tremaining: 3m 11s\n",
      "206:\tlearn: 0.9658451\ttotal: 33.2s\tremaining: 3m 11s\n",
      "207:\tlearn: 0.9657793\ttotal: 33.3s\tremaining: 3m 10s\n",
      "208:\tlearn: 0.9657384\ttotal: 33.4s\tremaining: 3m 10s\n",
      "209:\tlearn: 0.9656831\ttotal: 33.5s\tremaining: 3m 10s\n",
      "210:\tlearn: 0.9656107\ttotal: 33.7s\tremaining: 3m 9s\n",
      "211:\tlearn: 0.9655455\ttotal: 33.8s\tremaining: 3m 9s\n",
      "212:\tlearn: 0.9654720\ttotal: 33.9s\tremaining: 3m 9s\n",
      "213:\tlearn: 0.9654233\ttotal: 34s\tremaining: 3m 8s\n",
      "214:\tlearn: 0.9653405\ttotal: 34.2s\tremaining: 3m 8s\n",
      "215:\tlearn: 0.9652812\ttotal: 34.3s\tremaining: 3m 8s\n",
      "216:\tlearn: 0.9652241\ttotal: 34.4s\tremaining: 3m 7s\n",
      "217:\tlearn: 0.9651802\ttotal: 34.5s\tremaining: 3m 7s\n",
      "218:\tlearn: 0.9651101\ttotal: 34.7s\tremaining: 3m 6s\n",
      "219:\tlearn: 0.9650394\ttotal: 34.8s\tremaining: 3m 6s\n",
      "220:\tlearn: 0.9649636\ttotal: 34.9s\tremaining: 3m 6s\n",
      "221:\tlearn: 0.9649150\ttotal: 35s\tremaining: 3m 5s\n",
      "222:\tlearn: 0.9648218\ttotal: 35.2s\tremaining: 3m 6s\n",
      "223:\tlearn: 0.9647765\ttotal: 35.4s\tremaining: 3m 5s\n",
      "224:\tlearn: 0.9647394\ttotal: 35.5s\tremaining: 3m 5s\n",
      "225:\tlearn: 0.9646552\ttotal: 35.6s\tremaining: 3m 4s\n",
      "226:\tlearn: 0.9645849\ttotal: 35.7s\tremaining: 3m 4s\n",
      "227:\tlearn: 0.9644892\ttotal: 35.8s\tremaining: 3m 4s\n",
      "228:\tlearn: 0.9644283\ttotal: 35.9s\tremaining: 3m 3s\n",
      "229:\tlearn: 0.9644046\ttotal: 36s\tremaining: 3m 3s\n",
      "230:\tlearn: 0.9643624\ttotal: 36.1s\tremaining: 3m 2s\n",
      "231:\tlearn: 0.9643087\ttotal: 36.2s\tremaining: 3m 2s\n",
      "232:\tlearn: 0.9642388\ttotal: 36.3s\tremaining: 3m 1s\n",
      "233:\tlearn: 0.9641866\ttotal: 36.4s\tremaining: 3m 1s\n",
      "234:\tlearn: 0.9641113\ttotal: 36.6s\tremaining: 3m 1s\n",
      "235:\tlearn: 0.9640539\ttotal: 36.7s\tremaining: 3m\n",
      "236:\tlearn: 0.9639804\ttotal: 36.8s\tremaining: 3m\n",
      "237:\tlearn: 0.9639055\ttotal: 36.9s\tremaining: 3m\n",
      "238:\tlearn: 0.9638781\ttotal: 37s\tremaining: 2m 59s\n",
      "239:\tlearn: 0.9638175\ttotal: 37.1s\tremaining: 2m 59s\n",
      "240:\tlearn: 0.9637539\ttotal: 37.2s\tremaining: 2m 59s\n",
      "241:\tlearn: 0.9637491\ttotal: 37.3s\tremaining: 2m 58s\n",
      "242:\tlearn: 0.9636880\ttotal: 37.4s\tremaining: 2m 58s\n",
      "243:\tlearn: 0.9636509\ttotal: 37.5s\tremaining: 2m 57s\n",
      "244:\tlearn: 0.9636118\ttotal: 37.6s\tremaining: 2m 57s\n",
      "245:\tlearn: 0.9635663\ttotal: 37.7s\tremaining: 2m 56s\n",
      "246:\tlearn: 0.9634947\ttotal: 37.8s\tremaining: 2m 56s\n",
      "247:\tlearn: 0.9634103\ttotal: 38s\tremaining: 2m 56s\n",
      "248:\tlearn: 0.9633458\ttotal: 38.1s\tremaining: 2m 55s\n",
      "249:\tlearn: 0.9633094\ttotal: 38.2s\tremaining: 2m 55s\n",
      "250:\tlearn: 0.9632726\ttotal: 38.3s\tremaining: 2m 55s\n",
      "251:\tlearn: 0.9632436\ttotal: 38.4s\tremaining: 2m 54s\n",
      "252:\tlearn: 0.9631639\ttotal: 38.5s\tremaining: 2m 54s\n",
      "253:\tlearn: 0.9630944\ttotal: 38.7s\tremaining: 2m 54s\n",
      "254:\tlearn: 0.9630502\ttotal: 38.7s\tremaining: 2m 53s\n",
      "255:\tlearn: 0.9629802\ttotal: 38.9s\tremaining: 2m 53s\n",
      "256:\tlearn: 0.9629319\ttotal: 39s\tremaining: 2m 53s\n",
      "257:\tlearn: 0.9628648\ttotal: 39.1s\tremaining: 2m 53s\n",
      "258:\tlearn: 0.9627785\ttotal: 39.2s\tremaining: 2m 52s\n",
      "259:\tlearn: 0.9627147\ttotal: 39.3s\tremaining: 2m 52s\n",
      "260:\tlearn: 0.9626646\ttotal: 39.4s\tremaining: 2m 52s\n",
      "261:\tlearn: 0.9626632\ttotal: 39.5s\tremaining: 2m 51s\n",
      "262:\tlearn: 0.9625845\ttotal: 39.7s\tremaining: 2m 51s\n",
      "263:\tlearn: 0.9625385\ttotal: 39.8s\tremaining: 2m 51s\n",
      "264:\tlearn: 0.9624695\ttotal: 39.9s\tremaining: 2m 50s\n",
      "265:\tlearn: 0.9624031\ttotal: 40s\tremaining: 2m 50s\n",
      "266:\tlearn: 0.9623905\ttotal: 40.1s\tremaining: 2m 50s\n",
      "267:\tlearn: 0.9623349\ttotal: 40.2s\tremaining: 2m 49s\n",
      "268:\tlearn: 0.9622821\ttotal: 40.3s\tremaining: 2m 49s\n",
      "269:\tlearn: 0.9622119\ttotal: 40.4s\tremaining: 2m 49s\n",
      "270:\tlearn: 0.9621516\ttotal: 40.5s\tremaining: 2m 48s\n",
      "271:\tlearn: 0.9621401\ttotal: 40.6s\tremaining: 2m 48s\n",
      "272:\tlearn: 0.9620754\ttotal: 40.7s\tremaining: 2m 48s\n",
      "273:\tlearn: 0.9620746\ttotal: 40.8s\tremaining: 2m 47s\n",
      "274:\tlearn: 0.9620018\ttotal: 40.9s\tremaining: 2m 47s\n",
      "275:\tlearn: 0.9619272\ttotal: 41.1s\tremaining: 2m 47s\n",
      "276:\tlearn: 0.9618541\ttotal: 41.2s\tremaining: 2m 46s\n",
      "277:\tlearn: 0.9618143\ttotal: 41.3s\tremaining: 2m 46s\n",
      "278:\tlearn: 0.9618015\ttotal: 41.4s\tremaining: 2m 46s\n",
      "279:\tlearn: 0.9617605\ttotal: 41.4s\tremaining: 2m 45s\n",
      "280:\tlearn: 0.9617019\ttotal: 41.6s\tremaining: 2m 45s\n",
      "281:\tlearn: 0.9616990\ttotal: 41.6s\tremaining: 2m 45s\n",
      "282:\tlearn: 0.9616649\ttotal: 41.7s\tremaining: 2m 44s\n",
      "283:\tlearn: 0.9616126\ttotal: 41.8s\tremaining: 2m 44s\n",
      "284:\tlearn: 0.9615340\ttotal: 41.9s\tremaining: 2m 44s\n",
      "285:\tlearn: 0.9614861\ttotal: 42s\tremaining: 2m 43s\n",
      "286:\tlearn: 0.9614490\ttotal: 42.1s\tremaining: 2m 43s\n",
      "287:\tlearn: 0.9613966\ttotal: 42.3s\tremaining: 2m 43s\n",
      "288:\tlearn: 0.9613432\ttotal: 42.4s\tremaining: 2m 42s\n",
      "289:\tlearn: 0.9612895\ttotal: 42.5s\tremaining: 2m 42s\n",
      "290:\tlearn: 0.9612831\ttotal: 42.6s\tremaining: 2m 42s\n",
      "291:\tlearn: 0.9612609\ttotal: 42.7s\tremaining: 2m 41s\n",
      "292:\tlearn: 0.9612233\ttotal: 42.8s\tremaining: 2m 41s\n",
      "293:\tlearn: 0.9611842\ttotal: 42.9s\tremaining: 2m 41s\n",
      "294:\tlearn: 0.9611249\ttotal: 43s\tremaining: 2m 41s\n",
      "295:\tlearn: 0.9611218\ttotal: 43.1s\tremaining: 2m 40s\n",
      "296:\tlearn: 0.9610804\ttotal: 43.2s\tremaining: 2m 40s\n",
      "297:\tlearn: 0.9610377\ttotal: 43.3s\tremaining: 2m 40s\n",
      "298:\tlearn: 0.9609617\ttotal: 43.5s\tremaining: 2m 40s\n",
      "299:\tlearn: 0.9609230\ttotal: 43.6s\tremaining: 2m 39s\n",
      "300:\tlearn: 0.9608662\ttotal: 43.7s\tremaining: 2m 39s\n",
      "301:\tlearn: 0.9608638\ttotal: 43.8s\tremaining: 2m 39s\n",
      "302:\tlearn: 0.9608185\ttotal: 43.9s\tremaining: 2m 38s\n",
      "303:\tlearn: 0.9607874\ttotal: 44s\tremaining: 2m 38s\n",
      "304:\tlearn: 0.9607524\ttotal: 44.1s\tremaining: 2m 38s\n",
      "305:\tlearn: 0.9606961\ttotal: 44.3s\tremaining: 2m 38s\n",
      "306:\tlearn: 0.9606376\ttotal: 44.4s\tremaining: 2m 38s\n",
      "307:\tlearn: 0.9605905\ttotal: 44.5s\tremaining: 2m 37s\n",
      "308:\tlearn: 0.9605553\ttotal: 44.6s\tremaining: 2m 37s\n",
      "309:\tlearn: 0.9605234\ttotal: 44.7s\tremaining: 2m 37s\n",
      "310:\tlearn: 0.9604706\ttotal: 44.8s\tremaining: 2m 37s\n",
      "311:\tlearn: 0.9604663\ttotal: 44.9s\tremaining: 2m 36s\n",
      "312:\tlearn: 0.9604171\ttotal: 45s\tremaining: 2m 36s\n",
      "313:\tlearn: 0.9603910\ttotal: 45.1s\tremaining: 2m 36s\n",
      "314:\tlearn: 0.9603058\ttotal: 45.3s\tremaining: 2m 35s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "315:\tlearn: 0.9603018\ttotal: 45.4s\tremaining: 2m 35s\n",
      "316:\tlearn: 0.9602526\ttotal: 45.5s\tremaining: 2m 35s\n",
      "317:\tlearn: 0.9602073\ttotal: 45.6s\tremaining: 2m 35s\n",
      "318:\tlearn: 0.9601657\ttotal: 45.7s\tremaining: 2m 34s\n",
      "319:\tlearn: 0.9601261\ttotal: 45.8s\tremaining: 2m 34s\n",
      "320:\tlearn: 0.9600734\ttotal: 45.9s\tremaining: 2m 34s\n",
      "321:\tlearn: 0.9600023\ttotal: 46s\tremaining: 2m 34s\n",
      "322:\tlearn: 0.9599284\ttotal: 46.1s\tremaining: 2m 33s\n",
      "323:\tlearn: 0.9598618\ttotal: 46.2s\tremaining: 2m 33s\n",
      "324:\tlearn: 0.9598201\ttotal: 46.3s\tremaining: 2m 33s\n",
      "325:\tlearn: 0.9597675\ttotal: 46.4s\tremaining: 2m 33s\n",
      "326:\tlearn: 0.9597604\ttotal: 46.5s\tremaining: 2m 32s\n",
      "327:\tlearn: 0.9597297\ttotal: 46.6s\tremaining: 2m 32s\n",
      "328:\tlearn: 0.9596973\ttotal: 46.7s\tremaining: 2m 32s\n",
      "329:\tlearn: 0.9596272\ttotal: 46.8s\tremaining: 2m 31s\n",
      "330:\tlearn: 0.9596035\ttotal: 46.9s\tremaining: 2m 31s\n",
      "331:\tlearn: 0.9595492\ttotal: 47s\tremaining: 2m 31s\n",
      "332:\tlearn: 0.9595398\ttotal: 47.1s\tremaining: 2m 31s\n",
      "333:\tlearn: 0.9594886\ttotal: 47.3s\tremaining: 2m 30s\n",
      "334:\tlearn: 0.9594276\ttotal: 47.4s\tremaining: 2m 30s\n",
      "335:\tlearn: 0.9593722\ttotal: 47.5s\tremaining: 2m 30s\n",
      "336:\tlearn: 0.9593201\ttotal: 47.6s\tremaining: 2m 30s\n",
      "337:\tlearn: 0.9592963\ttotal: 47.7s\tremaining: 2m 30s\n",
      "338:\tlearn: 0.9592423\ttotal: 47.9s\tremaining: 2m 29s\n",
      "339:\tlearn: 0.9591885\ttotal: 48s\tremaining: 2m 29s\n",
      "340:\tlearn: 0.9591763\ttotal: 48.1s\tremaining: 2m 29s\n",
      "341:\tlearn: 0.9591170\ttotal: 48.2s\tremaining: 2m 29s\n",
      "342:\tlearn: 0.9590774\ttotal: 48.3s\tremaining: 2m 28s\n",
      "343:\tlearn: 0.9590151\ttotal: 48.4s\tremaining: 2m 28s\n",
      "344:\tlearn: 0.9589637\ttotal: 48.5s\tremaining: 2m 28s\n",
      "345:\tlearn: 0.9589216\ttotal: 48.6s\tremaining: 2m 28s\n",
      "346:\tlearn: 0.9588724\ttotal: 48.7s\tremaining: 2m 27s\n",
      "347:\tlearn: 0.9588345\ttotal: 48.8s\tremaining: 2m 27s\n",
      "348:\tlearn: 0.9587814\ttotal: 49s\tremaining: 2m 27s\n",
      "349:\tlearn: 0.9587432\ttotal: 49.1s\tremaining: 2m 27s\n",
      "350:\tlearn: 0.9587013\ttotal: 49.2s\tremaining: 2m 26s\n",
      "351:\tlearn: 0.9586545\ttotal: 49.3s\tremaining: 2m 26s\n",
      "352:\tlearn: 0.9586221\ttotal: 49.4s\tremaining: 2m 26s\n",
      "353:\tlearn: 0.9585841\ttotal: 49.5s\tremaining: 2m 26s\n",
      "354:\tlearn: 0.9585463\ttotal: 49.6s\tremaining: 2m 26s\n",
      "355:\tlearn: 0.9585009\ttotal: 49.7s\tremaining: 2m 25s\n",
      "356:\tlearn: 0.9584320\ttotal: 49.8s\tremaining: 2m 25s\n",
      "357:\tlearn: 0.9584095\ttotal: 49.9s\tremaining: 2m 25s\n",
      "358:\tlearn: 0.9583918\ttotal: 50s\tremaining: 2m 25s\n",
      "359:\tlearn: 0.9583318\ttotal: 50.1s\tremaining: 2m 24s\n",
      "360:\tlearn: 0.9582927\ttotal: 50.2s\tremaining: 2m 24s\n",
      "361:\tlearn: 0.9582574\ttotal: 50.3s\tremaining: 2m 24s\n",
      "362:\tlearn: 0.9581986\ttotal: 50.4s\tremaining: 2m 24s\n",
      "363:\tlearn: 0.9581504\ttotal: 50.5s\tremaining: 2m 23s\n",
      "364:\tlearn: 0.9581205\ttotal: 50.6s\tremaining: 2m 23s\n",
      "365:\tlearn: 0.9580583\ttotal: 50.7s\tremaining: 2m 23s\n",
      "366:\tlearn: 0.9580216\ttotal: 50.8s\tremaining: 2m 23s\n",
      "367:\tlearn: 0.9579569\ttotal: 51s\tremaining: 2m 22s\n",
      "368:\tlearn: 0.9579187\ttotal: 51.1s\tremaining: 2m 22s\n",
      "369:\tlearn: 0.9579168\ttotal: 51.2s\tremaining: 2m 22s\n",
      "370:\tlearn: 0.9578810\ttotal: 51.3s\tremaining: 2m 22s\n",
      "371:\tlearn: 0.9578381\ttotal: 51.4s\tremaining: 2m 22s\n",
      "372:\tlearn: 0.9577740\ttotal: 51.5s\tremaining: 2m 21s\n",
      "373:\tlearn: 0.9577462\ttotal: 51.7s\tremaining: 2m 21s\n",
      "374:\tlearn: 0.9577161\ttotal: 51.8s\tremaining: 2m 21s\n",
      "375:\tlearn: 0.9576518\ttotal: 51.9s\tremaining: 2m 21s\n",
      "376:\tlearn: 0.9575895\ttotal: 52s\tremaining: 2m 21s\n",
      "377:\tlearn: 0.9575287\ttotal: 52.1s\tremaining: 2m 20s\n",
      "378:\tlearn: 0.9574979\ttotal: 52.2s\tremaining: 2m 20s\n",
      "379:\tlearn: 0.9574580\ttotal: 52.3s\tremaining: 2m 20s\n",
      "380:\tlearn: 0.9574237\ttotal: 52.4s\tremaining: 2m 20s\n",
      "381:\tlearn: 0.9573823\ttotal: 52.5s\tremaining: 2m 19s\n",
      "382:\tlearn: 0.9573387\ttotal: 52.6s\tremaining: 2m 19s\n",
      "383:\tlearn: 0.9572993\ttotal: 52.7s\tremaining: 2m 19s\n",
      "384:\tlearn: 0.9572941\ttotal: 52.8s\tremaining: 2m 19s\n",
      "385:\tlearn: 0.9572388\ttotal: 52.9s\tremaining: 2m 18s\n",
      "386:\tlearn: 0.9572329\ttotal: 53s\tremaining: 2m 18s\n",
      "387:\tlearn: 0.9571732\ttotal: 53.1s\tremaining: 2m 18s\n",
      "388:\tlearn: 0.9571285\ttotal: 53.2s\tremaining: 2m 18s\n",
      "389:\tlearn: 0.9570938\ttotal: 53.3s\tremaining: 2m 18s\n",
      "390:\tlearn: 0.9570565\ttotal: 53.4s\tremaining: 2m 17s\n",
      "391:\tlearn: 0.9570027\ttotal: 53.5s\tremaining: 2m 17s\n",
      "392:\tlearn: 0.9569624\ttotal: 53.7s\tremaining: 2m 17s\n",
      "393:\tlearn: 0.9569336\ttotal: 53.8s\tremaining: 2m 17s\n",
      "394:\tlearn: 0.9568977\ttotal: 53.9s\tremaining: 2m 17s\n",
      "395:\tlearn: 0.9568453\ttotal: 54s\tremaining: 2m 17s\n",
      "396:\tlearn: 0.9568228\ttotal: 54.1s\tremaining: 2m 16s\n",
      "397:\tlearn: 0.9567788\ttotal: 54.3s\tremaining: 2m 16s\n",
      "398:\tlearn: 0.9567321\ttotal: 54.4s\tremaining: 2m 16s\n",
      "399:\tlearn: 0.9566944\ttotal: 54.5s\tremaining: 2m 16s\n",
      "400:\tlearn: 0.9566385\ttotal: 54.7s\tremaining: 2m 16s\n",
      "401:\tlearn: 0.9565919\ttotal: 54.8s\tremaining: 2m 16s\n",
      "402:\tlearn: 0.9565678\ttotal: 54.9s\tremaining: 2m 15s\n",
      "403:\tlearn: 0.9565391\ttotal: 55.1s\tremaining: 2m 15s\n",
      "404:\tlearn: 0.9564869\ttotal: 55.2s\tremaining: 2m 15s\n",
      "405:\tlearn: 0.9564535\ttotal: 55.3s\tremaining: 2m 15s\n",
      "406:\tlearn: 0.9563953\ttotal: 55.4s\tremaining: 2m 15s\n",
      "407:\tlearn: 0.9563537\ttotal: 55.5s\tremaining: 2m 15s\n",
      "408:\tlearn: 0.9563228\ttotal: 55.6s\tremaining: 2m 14s\n",
      "409:\tlearn: 0.9563023\ttotal: 55.7s\tremaining: 2m 14s\n",
      "410:\tlearn: 0.9562752\ttotal: 55.8s\tremaining: 2m 14s\n",
      "411:\tlearn: 0.9562170\ttotal: 55.9s\tremaining: 2m 14s\n",
      "412:\tlearn: 0.9561800\ttotal: 56s\tremaining: 2m 13s\n",
      "413:\tlearn: 0.9561515\ttotal: 56.1s\tremaining: 2m 13s\n",
      "414:\tlearn: 0.9560958\ttotal: 56.3s\tremaining: 2m 13s\n",
      "415:\tlearn: 0.9560052\ttotal: 56.4s\tremaining: 2m 13s\n",
      "416:\tlearn: 0.9559590\ttotal: 56.5s\tremaining: 2m 13s\n",
      "417:\tlearn: 0.9558974\ttotal: 56.7s\tremaining: 2m 13s\n",
      "418:\tlearn: 0.9558652\ttotal: 56.8s\tremaining: 2m 12s\n",
      "419:\tlearn: 0.9558098\ttotal: 56.9s\tremaining: 2m 12s\n",
      "420:\tlearn: 0.9557731\ttotal: 57s\tremaining: 2m 12s\n",
      "421:\tlearn: 0.9557448\ttotal: 57.1s\tremaining: 2m 12s\n",
      "422:\tlearn: 0.9557441\ttotal: 57.2s\tremaining: 2m 12s\n",
      "423:\tlearn: 0.9556935\ttotal: 57.3s\tremaining: 2m 11s\n",
      "424:\tlearn: 0.9556599\ttotal: 57.4s\tremaining: 2m 11s\n",
      "425:\tlearn: 0.9556364\ttotal: 57.5s\tremaining: 2m 11s\n",
      "426:\tlearn: 0.9556345\ttotal: 57.6s\tremaining: 2m 11s\n",
      "427:\tlearn: 0.9556309\ttotal: 57.7s\tremaining: 2m 10s\n",
      "428:\tlearn: 0.9556177\ttotal: 57.8s\tremaining: 2m 10s\n",
      "429:\tlearn: 0.9555596\ttotal: 57.9s\tremaining: 2m 10s\n",
      "430:\tlearn: 0.9555239\ttotal: 58s\tremaining: 2m 10s\n",
      "431:\tlearn: 0.9555021\ttotal: 58.1s\tremaining: 2m 10s\n",
      "432:\tlearn: 0.9554585\ttotal: 58.2s\tremaining: 2m 9s\n",
      "433:\tlearn: 0.9554575\ttotal: 58.3s\tremaining: 2m 9s\n",
      "434:\tlearn: 0.9554536\ttotal: 58.4s\tremaining: 2m 9s\n",
      "435:\tlearn: 0.9554507\ttotal: 58.5s\tremaining: 2m 9s\n",
      "436:\tlearn: 0.9554346\ttotal: 58.6s\tremaining: 2m 9s\n",
      "437:\tlearn: 0.9553727\ttotal: 58.7s\tremaining: 2m 8s\n",
      "438:\tlearn: 0.9553182\ttotal: 58.9s\tremaining: 2m 8s\n",
      "439:\tlearn: 0.9553121\ttotal: 59s\tremaining: 2m 8s\n",
      "440:\tlearn: 0.9552630\ttotal: 59.1s\tremaining: 2m 8s\n",
      "441:\tlearn: 0.9552297\ttotal: 59.2s\tremaining: 2m 8s\n",
      "442:\tlearn: 0.9551672\ttotal: 59.3s\tremaining: 2m 8s\n",
      "443:\tlearn: 0.9551361\ttotal: 59.4s\tremaining: 2m 7s\n",
      "444:\tlearn: 0.9551061\ttotal: 59.5s\tremaining: 2m 7s\n",
      "445:\tlearn: 0.9551050\ttotal: 59.6s\tremaining: 2m 7s\n",
      "446:\tlearn: 0.9550676\ttotal: 59.7s\tremaining: 2m 7s\n",
      "447:\tlearn: 0.9550271\ttotal: 59.9s\tremaining: 2m 7s\n",
      "448:\tlearn: 0.9550248\ttotal: 59.9s\tremaining: 2m 6s\n",
      "449:\tlearn: 0.9549889\ttotal: 1m\tremaining: 2m 6s\n",
      "450:\tlearn: 0.9549712\ttotal: 1m\tremaining: 2m 6s\n",
      "451:\tlearn: 0.9549548\ttotal: 1m\tremaining: 2m 6s\n",
      "452:\tlearn: 0.9549255\ttotal: 1m\tremaining: 2m 6s\n",
      "453:\tlearn: 0.9548591\ttotal: 1m\tremaining: 2m 6s\n",
      "454:\tlearn: 0.9548584\ttotal: 1m\tremaining: 2m 5s\n",
      "455:\tlearn: 0.9548329\ttotal: 1m\tremaining: 2m 5s\n",
      "456:\tlearn: 0.9547801\ttotal: 1m\tremaining: 2m 5s\n",
      "457:\tlearn: 0.9547600\ttotal: 1m\tremaining: 2m 5s\n",
      "458:\tlearn: 0.9547220\ttotal: 1m 1s\tremaining: 2m 5s\n",
      "459:\tlearn: 0.9546929\ttotal: 1m 1s\tremaining: 2m 4s\n",
      "460:\tlearn: 0.9546876\ttotal: 1m 1s\tremaining: 2m 4s\n",
      "461:\tlearn: 0.9546568\ttotal: 1m 1s\tremaining: 2m 4s\n",
      "462:\tlearn: 0.9546376\ttotal: 1m 1s\tremaining: 2m 4s\n",
      "463:\tlearn: 0.9546090\ttotal: 1m 1s\tremaining: 2m 4s\n",
      "464:\tlearn: 0.9545467\ttotal: 1m 1s\tremaining: 2m 4s\n",
      "465:\tlearn: 0.9545057\ttotal: 1m 1s\tremaining: 2m 3s\n",
      "466:\tlearn: 0.9544608\ttotal: 1m 1s\tremaining: 2m 3s\n",
      "467:\tlearn: 0.9544581\ttotal: 1m 2s\tremaining: 2m 3s\n",
      "468:\tlearn: 0.9544094\ttotal: 1m 2s\tremaining: 2m 3s\n",
      "469:\tlearn: 0.9543737\ttotal: 1m 2s\tremaining: 2m 3s\n",
      "470:\tlearn: 0.9543502\ttotal: 1m 2s\tremaining: 2m 3s\n",
      "471:\tlearn: 0.9543415\ttotal: 1m 2s\tremaining: 2m 3s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "472:\tlearn: 0.9542810\ttotal: 1m 2s\tremaining: 2m 3s\n",
      "473:\tlearn: 0.9542777\ttotal: 1m 2s\tremaining: 2m 2s\n",
      "474:\tlearn: 0.9542704\ttotal: 1m 2s\tremaining: 2m 2s\n",
      "475:\tlearn: 0.9542459\ttotal: 1m 3s\tremaining: 2m 2s\n",
      "476:\tlearn: 0.9542133\ttotal: 1m 3s\tremaining: 2m 2s\n",
      "477:\tlearn: 0.9541701\ttotal: 1m 3s\tremaining: 2m 2s\n",
      "478:\tlearn: 0.9541390\ttotal: 1m 3s\tremaining: 2m 2s\n",
      "479:\tlearn: 0.9540874\ttotal: 1m 3s\tremaining: 2m 1s\n",
      "480:\tlearn: 0.9540317\ttotal: 1m 3s\tremaining: 2m 1s\n",
      "481:\tlearn: 0.9539962\ttotal: 1m 3s\tremaining: 2m 1s\n",
      "482:\tlearn: 0.9539920\ttotal: 1m 3s\tremaining: 2m 1s\n",
      "483:\tlearn: 0.9539693\ttotal: 1m 4s\tremaining: 2m 1s\n",
      "484:\tlearn: 0.9539461\ttotal: 1m 4s\tremaining: 2m 1s\n",
      "485:\tlearn: 0.9538878\ttotal: 1m 4s\tremaining: 2m 1s\n",
      "486:\tlearn: 0.9538512\ttotal: 1m 4s\tremaining: 2m\n",
      "487:\tlearn: 0.9538173\ttotal: 1m 4s\tremaining: 2m\n",
      "488:\tlearn: 0.9537354\ttotal: 1m 4s\tremaining: 2m\n",
      "489:\tlearn: 0.9536999\ttotal: 1m 4s\tremaining: 2m\n",
      "490:\tlearn: 0.9536752\ttotal: 1m 5s\tremaining: 2m\n",
      "491:\tlearn: 0.9536455\ttotal: 1m 5s\tremaining: 2m\n",
      "492:\tlearn: 0.9535976\ttotal: 1m 5s\tremaining: 2m\n",
      "493:\tlearn: 0.9535569\ttotal: 1m 5s\tremaining: 1m 59s\n",
      "494:\tlearn: 0.9535210\ttotal: 1m 5s\tremaining: 1m 59s\n",
      "495:\tlearn: 0.9534851\ttotal: 1m 5s\tremaining: 1m 59s\n",
      "496:\tlearn: 0.9534517\ttotal: 1m 5s\tremaining: 1m 59s\n",
      "497:\tlearn: 0.9534262\ttotal: 1m 5s\tremaining: 1m 59s\n",
      "498:\tlearn: 0.9533980\ttotal: 1m 5s\tremaining: 1m 59s\n",
      "499:\tlearn: 0.9533946\ttotal: 1m 6s\tremaining: 1m 58s\n",
      "500:\tlearn: 0.9533640\ttotal: 1m 6s\tremaining: 1m 58s\n",
      "501:\tlearn: 0.9533300\ttotal: 1m 6s\tremaining: 1m 58s\n",
      "502:\tlearn: 0.9533216\ttotal: 1m 6s\tremaining: 1m 58s\n",
      "503:\tlearn: 0.9532844\ttotal: 1m 6s\tremaining: 1m 58s\n",
      "504:\tlearn: 0.9532556\ttotal: 1m 6s\tremaining: 1m 57s\n",
      "505:\tlearn: 0.9531966\ttotal: 1m 6s\tremaining: 1m 57s\n",
      "506:\tlearn: 0.9531576\ttotal: 1m 6s\tremaining: 1m 57s\n",
      "507:\tlearn: 0.9531567\ttotal: 1m 6s\tremaining: 1m 57s\n",
      "508:\tlearn: 0.9531268\ttotal: 1m 6s\tremaining: 1m 57s\n",
      "509:\tlearn: 0.9530993\ttotal: 1m 7s\tremaining: 1m 57s\n",
      "510:\tlearn: 0.9530755\ttotal: 1m 7s\tremaining: 1m 56s\n",
      "511:\tlearn: 0.9530613\ttotal: 1m 7s\tremaining: 1m 56s\n",
      "512:\tlearn: 0.9530415\ttotal: 1m 7s\tremaining: 1m 56s\n",
      "513:\tlearn: 0.9530088\ttotal: 1m 7s\tremaining: 1m 56s\n",
      "514:\tlearn: 0.9529886\ttotal: 1m 7s\tremaining: 1m 56s\n",
      "515:\tlearn: 0.9529676\ttotal: 1m 7s\tremaining: 1m 55s\n",
      "516:\tlearn: 0.9529152\ttotal: 1m 7s\tremaining: 1m 55s\n",
      "517:\tlearn: 0.9529054\ttotal: 1m 7s\tremaining: 1m 55s\n",
      "518:\tlearn: 0.9528963\ttotal: 1m 7s\tremaining: 1m 55s\n",
      "519:\tlearn: 0.9528803\ttotal: 1m 8s\tremaining: 1m 55s\n",
      "520:\tlearn: 0.9528390\ttotal: 1m 8s\tremaining: 1m 55s\n",
      "521:\tlearn: 0.9527973\ttotal: 1m 8s\tremaining: 1m 54s\n",
      "522:\tlearn: 0.9527589\ttotal: 1m 8s\tremaining: 1m 54s\n",
      "523:\tlearn: 0.9527543\ttotal: 1m 8s\tremaining: 1m 54s\n",
      "524:\tlearn: 0.9527434\ttotal: 1m 8s\tremaining: 1m 54s\n",
      "525:\tlearn: 0.9527144\ttotal: 1m 8s\tremaining: 1m 54s\n",
      "526:\tlearn: 0.9526844\ttotal: 1m 8s\tremaining: 1m 53s\n",
      "527:\tlearn: 0.9526534\ttotal: 1m 8s\tremaining: 1m 53s\n",
      "528:\tlearn: 0.9526139\ttotal: 1m 8s\tremaining: 1m 53s\n",
      "529:\tlearn: 0.9525808\ttotal: 1m 9s\tremaining: 1m 53s\n",
      "530:\tlearn: 0.9525515\ttotal: 1m 9s\tremaining: 1m 53s\n",
      "531:\tlearn: 0.9525117\ttotal: 1m 9s\tremaining: 1m 53s\n",
      "532:\tlearn: 0.9524589\ttotal: 1m 9s\tremaining: 1m 52s\n",
      "533:\tlearn: 0.9524204\ttotal: 1m 9s\tremaining: 1m 52s\n",
      "534:\tlearn: 0.9523863\ttotal: 1m 9s\tremaining: 1m 52s\n",
      "535:\tlearn: 0.9523522\ttotal: 1m 9s\tremaining: 1m 52s\n",
      "536:\tlearn: 0.9523431\ttotal: 1m 9s\tremaining: 1m 52s\n",
      "537:\tlearn: 0.9523353\ttotal: 1m 9s\tremaining: 1m 52s\n",
      "538:\tlearn: 0.9522934\ttotal: 1m 10s\tremaining: 1m 51s\n",
      "539:\tlearn: 0.9522822\ttotal: 1m 10s\tremaining: 1m 51s\n",
      "540:\tlearn: 0.9522396\ttotal: 1m 10s\tremaining: 1m 51s\n",
      "541:\tlearn: 0.9522306\ttotal: 1m 10s\tremaining: 1m 51s\n",
      "542:\tlearn: 0.9522084\ttotal: 1m 10s\tremaining: 1m 51s\n",
      "543:\tlearn: 0.9521689\ttotal: 1m 10s\tremaining: 1m 51s\n",
      "544:\tlearn: 0.9521397\ttotal: 1m 10s\tremaining: 1m 50s\n",
      "545:\tlearn: 0.9521224\ttotal: 1m 10s\tremaining: 1m 50s\n",
      "546:\tlearn: 0.9520777\ttotal: 1m 10s\tremaining: 1m 50s\n",
      "547:\tlearn: 0.9520294\ttotal: 1m 10s\tremaining: 1m 50s\n",
      "548:\tlearn: 0.9519940\ttotal: 1m 11s\tremaining: 1m 50s\n",
      "549:\tlearn: 0.9519810\ttotal: 1m 11s\tremaining: 1m 50s\n",
      "550:\tlearn: 0.9519487\ttotal: 1m 11s\tremaining: 1m 49s\n",
      "551:\tlearn: 0.9519214\ttotal: 1m 11s\tremaining: 1m 49s\n",
      "552:\tlearn: 0.9519166\ttotal: 1m 11s\tremaining: 1m 49s\n",
      "553:\tlearn: 0.9519004\ttotal: 1m 11s\tremaining: 1m 49s\n",
      "554:\tlearn: 0.9518883\ttotal: 1m 11s\tremaining: 1m 49s\n",
      "555:\tlearn: 0.9518380\ttotal: 1m 11s\tremaining: 1m 49s\n",
      "556:\tlearn: 0.9517855\ttotal: 1m 11s\tremaining: 1m 48s\n",
      "557:\tlearn: 0.9517827\ttotal: 1m 12s\tremaining: 1m 48s\n",
      "558:\tlearn: 0.9517602\ttotal: 1m 12s\tremaining: 1m 48s\n",
      "559:\tlearn: 0.9517177\ttotal: 1m 12s\tremaining: 1m 48s\n",
      "560:\tlearn: 0.9516981\ttotal: 1m 12s\tremaining: 1m 48s\n",
      "561:\tlearn: 0.9516937\ttotal: 1m 12s\tremaining: 1m 48s\n",
      "562:\tlearn: 0.9516831\ttotal: 1m 12s\tremaining: 1m 47s\n",
      "563:\tlearn: 0.9516726\ttotal: 1m 12s\tremaining: 1m 47s\n",
      "564:\tlearn: 0.9516015\ttotal: 1m 12s\tremaining: 1m 47s\n",
      "565:\tlearn: 0.9515578\ttotal: 1m 12s\tremaining: 1m 47s\n",
      "566:\tlearn: 0.9515208\ttotal: 1m 13s\tremaining: 1m 47s\n",
      "567:\tlearn: 0.9514740\ttotal: 1m 13s\tremaining: 1m 47s\n",
      "568:\tlearn: 0.9514578\ttotal: 1m 13s\tremaining: 1m 47s\n",
      "569:\tlearn: 0.9514192\ttotal: 1m 13s\tremaining: 1m 46s\n",
      "570:\tlearn: 0.9513887\ttotal: 1m 13s\tremaining: 1m 46s\n",
      "571:\tlearn: 0.9513828\ttotal: 1m 13s\tremaining: 1m 46s\n",
      "572:\tlearn: 0.9513814\ttotal: 1m 13s\tremaining: 1m 46s\n",
      "573:\tlearn: 0.9513193\ttotal: 1m 13s\tremaining: 1m 46s\n",
      "574:\tlearn: 0.9512936\ttotal: 1m 14s\tremaining: 1m 46s\n",
      "575:\tlearn: 0.9512575\ttotal: 1m 14s\tremaining: 1m 46s\n",
      "576:\tlearn: 0.9512495\ttotal: 1m 14s\tremaining: 1m 45s\n",
      "577:\tlearn: 0.9512112\ttotal: 1m 14s\tremaining: 1m 45s\n",
      "578:\tlearn: 0.9511724\ttotal: 1m 14s\tremaining: 1m 45s\n",
      "579:\tlearn: 0.9511546\ttotal: 1m 14s\tremaining: 1m 45s\n",
      "580:\tlearn: 0.9511222\ttotal: 1m 14s\tremaining: 1m 45s\n",
      "581:\tlearn: 0.9510694\ttotal: 1m 14s\tremaining: 1m 45s\n",
      "582:\tlearn: 0.9510432\ttotal: 1m 14s\tremaining: 1m 44s\n",
      "583:\tlearn: 0.9510249\ttotal: 1m 14s\tremaining: 1m 44s\n",
      "584:\tlearn: 0.9509886\ttotal: 1m 15s\tremaining: 1m 44s\n",
      "585:\tlearn: 0.9509676\ttotal: 1m 15s\tremaining: 1m 44s\n",
      "586:\tlearn: 0.9509465\ttotal: 1m 15s\tremaining: 1m 44s\n",
      "587:\tlearn: 0.9509265\ttotal: 1m 15s\tremaining: 1m 44s\n",
      "588:\tlearn: 0.9509209\ttotal: 1m 15s\tremaining: 1m 43s\n",
      "589:\tlearn: 0.9509037\ttotal: 1m 15s\tremaining: 1m 43s\n",
      "590:\tlearn: 0.9508761\ttotal: 1m 15s\tremaining: 1m 43s\n",
      "591:\tlearn: 0.9508335\ttotal: 1m 15s\tremaining: 1m 43s\n",
      "592:\tlearn: 0.9507825\ttotal: 1m 15s\tremaining: 1m 43s\n",
      "593:\tlearn: 0.9507378\ttotal: 1m 16s\tremaining: 1m 43s\n",
      "594:\tlearn: 0.9507136\ttotal: 1m 16s\tremaining: 1m 43s\n",
      "595:\tlearn: 0.9506902\ttotal: 1m 16s\tremaining: 1m 42s\n",
      "596:\tlearn: 0.9506673\ttotal: 1m 16s\tremaining: 1m 42s\n",
      "597:\tlearn: 0.9506435\ttotal: 1m 16s\tremaining: 1m 42s\n",
      "598:\tlearn: 0.9506180\ttotal: 1m 16s\tremaining: 1m 42s\n",
      "599:\tlearn: 0.9506093\ttotal: 1m 16s\tremaining: 1m 42s\n",
      "600:\tlearn: 0.9505838\ttotal: 1m 16s\tremaining: 1m 42s\n",
      "601:\tlearn: 0.9505806\ttotal: 1m 16s\tremaining: 1m 41s\n",
      "602:\tlearn: 0.9505541\ttotal: 1m 16s\tremaining: 1m 41s\n",
      "603:\tlearn: 0.9505139\ttotal: 1m 17s\tremaining: 1m 41s\n",
      "604:\tlearn: 0.9505100\ttotal: 1m 17s\tremaining: 1m 41s\n",
      "605:\tlearn: 0.9505074\ttotal: 1m 17s\tremaining: 1m 41s\n",
      "606:\tlearn: 0.9505063\ttotal: 1m 17s\tremaining: 1m 40s\n",
      "607:\tlearn: 0.9504740\ttotal: 1m 17s\tremaining: 1m 40s\n",
      "608:\tlearn: 0.9504449\ttotal: 1m 17s\tremaining: 1m 40s\n",
      "609:\tlearn: 0.9504116\ttotal: 1m 17s\tremaining: 1m 40s\n",
      "610:\tlearn: 0.9503432\ttotal: 1m 17s\tremaining: 1m 40s\n",
      "611:\tlearn: 0.9502789\ttotal: 1m 17s\tremaining: 1m 40s\n",
      "612:\tlearn: 0.9502420\ttotal: 1m 17s\tremaining: 1m 40s\n",
      "613:\tlearn: 0.9502391\ttotal: 1m 18s\tremaining: 1m 39s\n",
      "614:\tlearn: 0.9502145\ttotal: 1m 18s\tremaining: 1m 39s\n",
      "615:\tlearn: 0.9502141\ttotal: 1m 18s\tremaining: 1m 39s\n",
      "616:\tlearn: 0.9501667\ttotal: 1m 18s\tremaining: 1m 39s\n",
      "617:\tlearn: 0.9501276\ttotal: 1m 18s\tremaining: 1m 39s\n",
      "618:\tlearn: 0.9501079\ttotal: 1m 18s\tremaining: 1m 39s\n",
      "619:\tlearn: 0.9500891\ttotal: 1m 18s\tremaining: 1m 38s\n",
      "620:\tlearn: 0.9500569\ttotal: 1m 18s\tremaining: 1m 38s\n",
      "621:\tlearn: 0.9500460\ttotal: 1m 18s\tremaining: 1m 38s\n",
      "622:\tlearn: 0.9500456\ttotal: 1m 18s\tremaining: 1m 38s\n",
      "623:\tlearn: 0.9500296\ttotal: 1m 18s\tremaining: 1m 38s\n",
      "624:\tlearn: 0.9499986\ttotal: 1m 19s\tremaining: 1m 38s\n",
      "625:\tlearn: 0.9499901\ttotal: 1m 19s\tremaining: 1m 37s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "626:\tlearn: 0.9499802\ttotal: 1m 19s\tremaining: 1m 37s\n",
      "627:\tlearn: 0.9499703\ttotal: 1m 19s\tremaining: 1m 37s\n",
      "628:\tlearn: 0.9499446\ttotal: 1m 19s\tremaining: 1m 37s\n",
      "629:\tlearn: 0.9499010\ttotal: 1m 19s\tremaining: 1m 37s\n",
      "630:\tlearn: 0.9498730\ttotal: 1m 19s\tremaining: 1m 37s\n",
      "631:\tlearn: 0.9498550\ttotal: 1m 19s\tremaining: 1m 36s\n",
      "632:\tlearn: 0.9498265\ttotal: 1m 19s\tremaining: 1m 36s\n",
      "633:\tlearn: 0.9497925\ttotal: 1m 19s\tremaining: 1m 36s\n",
      "634:\tlearn: 0.9497559\ttotal: 1m 20s\tremaining: 1m 36s\n",
      "635:\tlearn: 0.9497308\ttotal: 1m 20s\tremaining: 1m 36s\n",
      "636:\tlearn: 0.9496666\ttotal: 1m 20s\tremaining: 1m 36s\n",
      "637:\tlearn: 0.9496661\ttotal: 1m 20s\tremaining: 1m 36s\n",
      "638:\tlearn: 0.9496466\ttotal: 1m 20s\tremaining: 1m 35s\n",
      "639:\tlearn: 0.9496167\ttotal: 1m 20s\tremaining: 1m 35s\n",
      "640:\tlearn: 0.9495755\ttotal: 1m 20s\tremaining: 1m 35s\n",
      "641:\tlearn: 0.9495718\ttotal: 1m 20s\tremaining: 1m 35s\n",
      "642:\tlearn: 0.9495464\ttotal: 1m 20s\tremaining: 1m 35s\n",
      "643:\tlearn: 0.9495039\ttotal: 1m 21s\tremaining: 1m 35s\n",
      "644:\tlearn: 0.9494675\ttotal: 1m 21s\tremaining: 1m 34s\n",
      "645:\tlearn: 0.9494330\ttotal: 1m 21s\tremaining: 1m 34s\n",
      "646:\tlearn: 0.9493846\ttotal: 1m 21s\tremaining: 1m 34s\n",
      "647:\tlearn: 0.9493804\ttotal: 1m 21s\tremaining: 1m 34s\n",
      "648:\tlearn: 0.9493438\ttotal: 1m 21s\tremaining: 1m 34s\n",
      "649:\tlearn: 0.9493016\ttotal: 1m 21s\tremaining: 1m 34s\n",
      "650:\tlearn: 0.9492980\ttotal: 1m 21s\tremaining: 1m 34s\n",
      "651:\tlearn: 0.9492545\ttotal: 1m 21s\tremaining: 1m 33s\n",
      "652:\tlearn: 0.9492090\ttotal: 1m 22s\tremaining: 1m 33s\n",
      "653:\tlearn: 0.9492071\ttotal: 1m 22s\tremaining: 1m 33s\n",
      "654:\tlearn: 0.9491750\ttotal: 1m 22s\tremaining: 1m 33s\n",
      "655:\tlearn: 0.9491740\ttotal: 1m 22s\tremaining: 1m 33s\n",
      "656:\tlearn: 0.9491431\ttotal: 1m 22s\tremaining: 1m 33s\n",
      "657:\tlearn: 0.9491291\ttotal: 1m 22s\tremaining: 1m 32s\n",
      "658:\tlearn: 0.9490888\ttotal: 1m 22s\tremaining: 1m 32s\n",
      "659:\tlearn: 0.9490834\ttotal: 1m 22s\tremaining: 1m 32s\n",
      "660:\tlearn: 0.9490689\ttotal: 1m 22s\tremaining: 1m 32s\n",
      "661:\tlearn: 0.9490417\ttotal: 1m 22s\tremaining: 1m 32s\n",
      "662:\tlearn: 0.9490165\ttotal: 1m 22s\tremaining: 1m 32s\n",
      "663:\tlearn: 0.9489734\ttotal: 1m 23s\tremaining: 1m 32s\n",
      "664:\tlearn: 0.9489536\ttotal: 1m 23s\tremaining: 1m 31s\n",
      "665:\tlearn: 0.9489531\ttotal: 1m 23s\tremaining: 1m 31s\n",
      "666:\tlearn: 0.9489355\ttotal: 1m 23s\tremaining: 1m 31s\n",
      "667:\tlearn: 0.9489178\ttotal: 1m 23s\tremaining: 1m 31s\n",
      "668:\tlearn: 0.9488650\ttotal: 1m 23s\tremaining: 1m 31s\n",
      "669:\tlearn: 0.9488633\ttotal: 1m 23s\tremaining: 1m 31s\n",
      "670:\tlearn: 0.9488346\ttotal: 1m 23s\tremaining: 1m 30s\n",
      "671:\tlearn: 0.9488272\ttotal: 1m 23s\tremaining: 1m 30s\n",
      "672:\tlearn: 0.9487902\ttotal: 1m 23s\tremaining: 1m 30s\n",
      "673:\tlearn: 0.9487853\ttotal: 1m 24s\tremaining: 1m 30s\n",
      "674:\tlearn: 0.9487321\ttotal: 1m 24s\tremaining: 1m 30s\n",
      "675:\tlearn: 0.9487047\ttotal: 1m 24s\tremaining: 1m 30s\n",
      "676:\tlearn: 0.9486867\ttotal: 1m 24s\tremaining: 1m 30s\n",
      "677:\tlearn: 0.9486336\ttotal: 1m 24s\tremaining: 1m 29s\n",
      "678:\tlearn: 0.9486309\ttotal: 1m 24s\tremaining: 1m 29s\n",
      "679:\tlearn: 0.9485898\ttotal: 1m 24s\tremaining: 1m 29s\n",
      "680:\tlearn: 0.9485639\ttotal: 1m 24s\tremaining: 1m 29s\n",
      "681:\tlearn: 0.9485159\ttotal: 1m 24s\tremaining: 1m 29s\n",
      "682:\tlearn: 0.9484724\ttotal: 1m 24s\tremaining: 1m 29s\n",
      "683:\tlearn: 0.9484668\ttotal: 1m 25s\tremaining: 1m 29s\n",
      "684:\tlearn: 0.9484298\ttotal: 1m 25s\tremaining: 1m 28s\n",
      "685:\tlearn: 0.9483936\ttotal: 1m 25s\tremaining: 1m 28s\n",
      "686:\tlearn: 0.9483755\ttotal: 1m 25s\tremaining: 1m 28s\n",
      "687:\tlearn: 0.9483587\ttotal: 1m 25s\tremaining: 1m 28s\n",
      "688:\tlearn: 0.9483339\ttotal: 1m 25s\tremaining: 1m 28s\n",
      "689:\tlearn: 0.9483034\ttotal: 1m 25s\tremaining: 1m 28s\n",
      "690:\tlearn: 0.9482997\ttotal: 1m 25s\tremaining: 1m 27s\n",
      "691:\tlearn: 0.9482498\ttotal: 1m 25s\tremaining: 1m 27s\n",
      "692:\tlearn: 0.9482050\ttotal: 1m 26s\tremaining: 1m 27s\n",
      "693:\tlearn: 0.9481704\ttotal: 1m 26s\tremaining: 1m 27s\n",
      "694:\tlearn: 0.9481523\ttotal: 1m 26s\tremaining: 1m 27s\n",
      "695:\tlearn: 0.9480999\ttotal: 1m 26s\tremaining: 1m 27s\n",
      "696:\tlearn: 0.9480697\ttotal: 1m 26s\tremaining: 1m 27s\n",
      "697:\tlearn: 0.9480549\ttotal: 1m 26s\tremaining: 1m 27s\n",
      "698:\tlearn: 0.9480168\ttotal: 1m 26s\tremaining: 1m 26s\n",
      "699:\tlearn: 0.9480150\ttotal: 1m 26s\tremaining: 1m 26s\n",
      "700:\tlearn: 0.9479869\ttotal: 1m 26s\tremaining: 1m 26s\n",
      "701:\tlearn: 0.9479591\ttotal: 1m 26s\tremaining: 1m 26s\n",
      "702:\tlearn: 0.9479273\ttotal: 1m 27s\tremaining: 1m 26s\n",
      "703:\tlearn: 0.9478761\ttotal: 1m 27s\tremaining: 1m 26s\n",
      "704:\tlearn: 0.9478248\ttotal: 1m 27s\tremaining: 1m 26s\n",
      "705:\tlearn: 0.9478231\ttotal: 1m 27s\tremaining: 1m 25s\n",
      "706:\tlearn: 0.9478171\ttotal: 1m 27s\tremaining: 1m 25s\n",
      "707:\tlearn: 0.9477921\ttotal: 1m 27s\tremaining: 1m 25s\n",
      "708:\tlearn: 0.9477458\ttotal: 1m 27s\tremaining: 1m 25s\n",
      "709:\tlearn: 0.9477201\ttotal: 1m 27s\tremaining: 1m 25s\n",
      "710:\tlearn: 0.9476980\ttotal: 1m 27s\tremaining: 1m 25s\n",
      "711:\tlearn: 0.9476951\ttotal: 1m 27s\tremaining: 1m 24s\n",
      "712:\tlearn: 0.9476886\ttotal: 1m 28s\tremaining: 1m 24s\n",
      "713:\tlearn: 0.9476667\ttotal: 1m 28s\tremaining: 1m 24s\n",
      "714:\tlearn: 0.9476488\ttotal: 1m 28s\tremaining: 1m 24s\n",
      "715:\tlearn: 0.9476113\ttotal: 1m 28s\tremaining: 1m 24s\n",
      "716:\tlearn: 0.9475752\ttotal: 1m 28s\tremaining: 1m 24s\n",
      "717:\tlearn: 0.9475394\ttotal: 1m 28s\tremaining: 1m 24s\n",
      "718:\tlearn: 0.9475030\ttotal: 1m 28s\tremaining: 1m 23s\n",
      "719:\tlearn: 0.9474750\ttotal: 1m 28s\tremaining: 1m 23s\n",
      "720:\tlearn: 0.9474396\ttotal: 1m 28s\tremaining: 1m 23s\n",
      "721:\tlearn: 0.9474326\ttotal: 1m 28s\tremaining: 1m 23s\n",
      "722:\tlearn: 0.9473858\ttotal: 1m 29s\tremaining: 1m 23s\n",
      "723:\tlearn: 0.9473430\ttotal: 1m 29s\tremaining: 1m 23s\n",
      "724:\tlearn: 0.9473134\ttotal: 1m 29s\tremaining: 1m 23s\n",
      "725:\tlearn: 0.9472573\ttotal: 1m 29s\tremaining: 1m 23s\n",
      "726:\tlearn: 0.9472294\ttotal: 1m 29s\tremaining: 1m 22s\n",
      "727:\tlearn: 0.9472250\ttotal: 1m 29s\tremaining: 1m 22s\n",
      "728:\tlearn: 0.9472077\ttotal: 1m 29s\tremaining: 1m 22s\n",
      "729:\tlearn: 0.9471863\ttotal: 1m 29s\tremaining: 1m 22s\n",
      "730:\tlearn: 0.9471515\ttotal: 1m 29s\tremaining: 1m 22s\n",
      "731:\tlearn: 0.9470901\ttotal: 1m 30s\tremaining: 1m 22s\n",
      "732:\tlearn: 0.9470348\ttotal: 1m 30s\tremaining: 1m 22s\n",
      "733:\tlearn: 0.9470016\ttotal: 1m 30s\tremaining: 1m 21s\n",
      "734:\tlearn: 0.9469633\ttotal: 1m 30s\tremaining: 1m 21s\n",
      "735:\tlearn: 0.9469255\ttotal: 1m 30s\tremaining: 1m 21s\n",
      "736:\tlearn: 0.9469153\ttotal: 1m 30s\tremaining: 1m 21s\n",
      "737:\tlearn: 0.9468889\ttotal: 1m 30s\tremaining: 1m 21s\n",
      "738:\tlearn: 0.9468658\ttotal: 1m 30s\tremaining: 1m 21s\n",
      "739:\tlearn: 0.9468413\ttotal: 1m 30s\tremaining: 1m 21s\n",
      "740:\tlearn: 0.9467802\ttotal: 1m 31s\tremaining: 1m 20s\n",
      "741:\tlearn: 0.9467099\ttotal: 1m 31s\tremaining: 1m 20s\n",
      "742:\tlearn: 0.9466755\ttotal: 1m 31s\tremaining: 1m 20s\n",
      "743:\tlearn: 0.9466672\ttotal: 1m 31s\tremaining: 1m 20s\n",
      "744:\tlearn: 0.9466633\ttotal: 1m 31s\tremaining: 1m 20s\n",
      "745:\tlearn: 0.9466495\ttotal: 1m 31s\tremaining: 1m 20s\n",
      "746:\tlearn: 0.9465956\ttotal: 1m 31s\tremaining: 1m 20s\n",
      "747:\tlearn: 0.9465582\ttotal: 1m 31s\tremaining: 1m 20s\n",
      "748:\tlearn: 0.9465160\ttotal: 1m 31s\tremaining: 1m 19s\n",
      "749:\tlearn: 0.9465065\ttotal: 1m 31s\tremaining: 1m 19s\n",
      "750:\tlearn: 0.9464578\ttotal: 1m 32s\tremaining: 1m 19s\n",
      "751:\tlearn: 0.9464257\ttotal: 1m 32s\tremaining: 1m 19s\n",
      "752:\tlearn: 0.9464099\ttotal: 1m 32s\tremaining: 1m 19s\n",
      "753:\tlearn: 0.9464080\ttotal: 1m 32s\tremaining: 1m 19s\n",
      "754:\tlearn: 0.9463853\ttotal: 1m 32s\tremaining: 1m 19s\n",
      "755:\tlearn: 0.9463823\ttotal: 1m 32s\tremaining: 1m 18s\n",
      "756:\tlearn: 0.9463581\ttotal: 1m 32s\tremaining: 1m 18s\n",
      "757:\tlearn: 0.9463034\ttotal: 1m 32s\tremaining: 1m 18s\n",
      "758:\tlearn: 0.9462658\ttotal: 1m 32s\tremaining: 1m 18s\n",
      "759:\tlearn: 0.9462445\ttotal: 1m 33s\tremaining: 1m 18s\n",
      "760:\tlearn: 0.9461992\ttotal: 1m 33s\tremaining: 1m 18s\n",
      "761:\tlearn: 0.9461784\ttotal: 1m 33s\tremaining: 1m 18s\n",
      "762:\tlearn: 0.9461423\ttotal: 1m 33s\tremaining: 1m 17s\n",
      "763:\tlearn: 0.9461120\ttotal: 1m 33s\tremaining: 1m 17s\n",
      "764:\tlearn: 0.9461047\ttotal: 1m 33s\tremaining: 1m 17s\n",
      "765:\tlearn: 0.9460650\ttotal: 1m 33s\tremaining: 1m 17s\n",
      "766:\tlearn: 0.9460165\ttotal: 1m 33s\tremaining: 1m 17s\n",
      "767:\tlearn: 0.9459853\ttotal: 1m 33s\tremaining: 1m 17s\n",
      "768:\tlearn: 0.9459825\ttotal: 1m 33s\tremaining: 1m 17s\n",
      "769:\tlearn: 0.9459820\ttotal: 1m 34s\tremaining: 1m 16s\n",
      "770:\tlearn: 0.9459599\ttotal: 1m 34s\tremaining: 1m 16s\n",
      "771:\tlearn: 0.9459348\ttotal: 1m 34s\tremaining: 1m 16s\n",
      "772:\tlearn: 0.9458949\ttotal: 1m 34s\tremaining: 1m 16s\n",
      "773:\tlearn: 0.9458682\ttotal: 1m 34s\tremaining: 1m 16s\n",
      "774:\tlearn: 0.9458306\ttotal: 1m 34s\tremaining: 1m 16s\n",
      "775:\tlearn: 0.9457962\ttotal: 1m 34s\tremaining: 1m 16s\n",
      "776:\tlearn: 0.9457809\ttotal: 1m 34s\tremaining: 1m 15s\n",
      "777:\tlearn: 0.9457167\ttotal: 1m 34s\tremaining: 1m 15s\n",
      "778:\tlearn: 0.9457151\ttotal: 1m 34s\tremaining: 1m 15s\n",
      "779:\tlearn: 0.9456828\ttotal: 1m 35s\tremaining: 1m 15s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "780:\tlearn: 0.9456788\ttotal: 1m 35s\tremaining: 1m 15s\n",
      "781:\tlearn: 0.9456769\ttotal: 1m 35s\tremaining: 1m 15s\n",
      "782:\tlearn: 0.9456651\ttotal: 1m 35s\tremaining: 1m 15s\n",
      "783:\tlearn: 0.9456376\ttotal: 1m 35s\tremaining: 1m 14s\n",
      "784:\tlearn: 0.9456331\ttotal: 1m 35s\tremaining: 1m 14s\n",
      "785:\tlearn: 0.9456043\ttotal: 1m 35s\tremaining: 1m 14s\n",
      "786:\tlearn: 0.9455681\ttotal: 1m 35s\tremaining: 1m 14s\n",
      "787:\tlearn: 0.9455505\ttotal: 1m 35s\tremaining: 1m 14s\n",
      "788:\tlearn: 0.9455325\ttotal: 1m 35s\tremaining: 1m 14s\n",
      "789:\tlearn: 0.9455129\ttotal: 1m 36s\tremaining: 1m 14s\n",
      "790:\tlearn: 0.9454804\ttotal: 1m 36s\tremaining: 1m 13s\n",
      "791:\tlearn: 0.9454546\ttotal: 1m 36s\tremaining: 1m 13s\n",
      "792:\tlearn: 0.9454306\ttotal: 1m 36s\tremaining: 1m 13s\n",
      "793:\tlearn: 0.9453898\ttotal: 1m 36s\tremaining: 1m 13s\n",
      "794:\tlearn: 0.9453842\ttotal: 1m 36s\tremaining: 1m 13s\n",
      "795:\tlearn: 0.9453593\ttotal: 1m 36s\tremaining: 1m 13s\n",
      "796:\tlearn: 0.9453467\ttotal: 1m 36s\tremaining: 1m 13s\n",
      "797:\tlearn: 0.9452920\ttotal: 1m 36s\tremaining: 1m 13s\n",
      "798:\tlearn: 0.9452536\ttotal: 1m 36s\tremaining: 1m 12s\n",
      "799:\tlearn: 0.9452330\ttotal: 1m 37s\tremaining: 1m 12s\n",
      "800:\tlearn: 0.9452091\ttotal: 1m 37s\tremaining: 1m 12s\n",
      "801:\tlearn: 0.9451910\ttotal: 1m 37s\tremaining: 1m 12s\n",
      "802:\tlearn: 0.9451820\ttotal: 1m 37s\tremaining: 1m 12s\n",
      "803:\tlearn: 0.9451296\ttotal: 1m 37s\tremaining: 1m 12s\n",
      "804:\tlearn: 0.9450944\ttotal: 1m 37s\tremaining: 1m 12s\n",
      "805:\tlearn: 0.9450510\ttotal: 1m 37s\tremaining: 1m 11s\n",
      "806:\tlearn: 0.9450161\ttotal: 1m 37s\tremaining: 1m 11s\n",
      "807:\tlearn: 0.9449968\ttotal: 1m 37s\tremaining: 1m 11s\n",
      "808:\tlearn: 0.9449651\ttotal: 1m 38s\tremaining: 1m 11s\n",
      "809:\tlearn: 0.9449151\ttotal: 1m 38s\tremaining: 1m 11s\n",
      "810:\tlearn: 0.9448950\ttotal: 1m 38s\tremaining: 1m 11s\n",
      "811:\tlearn: 0.9448523\ttotal: 1m 38s\tremaining: 1m 11s\n",
      "812:\tlearn: 0.9448110\ttotal: 1m 38s\tremaining: 1m 11s\n",
      "813:\tlearn: 0.9447879\ttotal: 1m 38s\tremaining: 1m 10s\n",
      "814:\tlearn: 0.9447313\ttotal: 1m 38s\tremaining: 1m 10s\n",
      "815:\tlearn: 0.9447086\ttotal: 1m 38s\tremaining: 1m 10s\n",
      "816:\tlearn: 0.9446767\ttotal: 1m 38s\tremaining: 1m 10s\n",
      "817:\tlearn: 0.9446630\ttotal: 1m 38s\tremaining: 1m 10s\n",
      "818:\tlearn: 0.9446180\ttotal: 1m 39s\tremaining: 1m 10s\n",
      "819:\tlearn: 0.9445933\ttotal: 1m 39s\tremaining: 1m 10s\n",
      "820:\tlearn: 0.9445679\ttotal: 1m 39s\tremaining: 1m 10s\n",
      "821:\tlearn: 0.9445436\ttotal: 1m 39s\tremaining: 1m 9s\n",
      "822:\tlearn: 0.9445360\ttotal: 1m 39s\tremaining: 1m 9s\n",
      "823:\tlearn: 0.9444921\ttotal: 1m 39s\tremaining: 1m 9s\n",
      "824:\tlearn: 0.9444555\ttotal: 1m 39s\tremaining: 1m 9s\n",
      "825:\tlearn: 0.9444540\ttotal: 1m 39s\tremaining: 1m 9s\n",
      "826:\tlearn: 0.9444330\ttotal: 1m 39s\tremaining: 1m 9s\n",
      "827:\tlearn: 0.9443967\ttotal: 1m 39s\tremaining: 1m 9s\n",
      "828:\tlearn: 0.9443842\ttotal: 1m 40s\tremaining: 1m 8s\n",
      "829:\tlearn: 0.9443823\ttotal: 1m 40s\tremaining: 1m 8s\n",
      "830:\tlearn: 0.9443724\ttotal: 1m 40s\tremaining: 1m 8s\n",
      "831:\tlearn: 0.9442988\ttotal: 1m 40s\tremaining: 1m 8s\n",
      "832:\tlearn: 0.9442463\ttotal: 1m 40s\tremaining: 1m 8s\n",
      "833:\tlearn: 0.9442142\ttotal: 1m 40s\tremaining: 1m 8s\n",
      "834:\tlearn: 0.9441938\ttotal: 1m 40s\tremaining: 1m 8s\n",
      "835:\tlearn: 0.9441648\ttotal: 1m 40s\tremaining: 1m 7s\n",
      "836:\tlearn: 0.9440978\ttotal: 1m 40s\tremaining: 1m 7s\n",
      "837:\tlearn: 0.9440966\ttotal: 1m 40s\tremaining: 1m 7s\n",
      "838:\tlearn: 0.9440961\ttotal: 1m 41s\tremaining: 1m 7s\n",
      "839:\tlearn: 0.9440560\ttotal: 1m 41s\tremaining: 1m 7s\n",
      "840:\tlearn: 0.9440533\ttotal: 1m 41s\tremaining: 1m 7s\n",
      "841:\tlearn: 0.9440523\ttotal: 1m 41s\tremaining: 1m 7s\n",
      "842:\tlearn: 0.9440369\ttotal: 1m 41s\tremaining: 1m 6s\n",
      "843:\tlearn: 0.9440000\ttotal: 1m 41s\tremaining: 1m 6s\n",
      "844:\tlearn: 0.9439876\ttotal: 1m 41s\tremaining: 1m 6s\n",
      "845:\tlearn: 0.9439737\ttotal: 1m 41s\tremaining: 1m 6s\n",
      "846:\tlearn: 0.9439513\ttotal: 1m 41s\tremaining: 1m 6s\n",
      "847:\tlearn: 0.9439183\ttotal: 1m 41s\tremaining: 1m 6s\n",
      "848:\tlearn: 0.9439023\ttotal: 1m 41s\tremaining: 1m 6s\n",
      "849:\tlearn: 0.9438799\ttotal: 1m 42s\tremaining: 1m 6s\n",
      "850:\tlearn: 0.9438566\ttotal: 1m 42s\tremaining: 1m 5s\n",
      "851:\tlearn: 0.9438020\ttotal: 1m 42s\tremaining: 1m 5s\n",
      "852:\tlearn: 0.9437745\ttotal: 1m 42s\tremaining: 1m 5s\n",
      "853:\tlearn: 0.9437486\ttotal: 1m 42s\tremaining: 1m 5s\n",
      "854:\tlearn: 0.9436982\ttotal: 1m 42s\tremaining: 1m 5s\n",
      "855:\tlearn: 0.9436658\ttotal: 1m 42s\tremaining: 1m 5s\n",
      "856:\tlearn: 0.9436455\ttotal: 1m 42s\tremaining: 1m 5s\n",
      "857:\tlearn: 0.9436437\ttotal: 1m 42s\tremaining: 1m 5s\n",
      "858:\tlearn: 0.9436017\ttotal: 1m 43s\tremaining: 1m 4s\n",
      "859:\tlearn: 0.9435758\ttotal: 1m 43s\tremaining: 1m 4s\n",
      "860:\tlearn: 0.9435592\ttotal: 1m 43s\tremaining: 1m 4s\n",
      "861:\tlearn: 0.9435353\ttotal: 1m 43s\tremaining: 1m 4s\n",
      "862:\tlearn: 0.9435123\ttotal: 1m 43s\tremaining: 1m 4s\n",
      "863:\tlearn: 0.9434989\ttotal: 1m 43s\tremaining: 1m 4s\n",
      "864:\tlearn: 0.9434443\ttotal: 1m 43s\tremaining: 1m 4s\n",
      "865:\tlearn: 0.9434209\ttotal: 1m 43s\tremaining: 1m 3s\n",
      "866:\tlearn: 0.9434105\ttotal: 1m 43s\tremaining: 1m 3s\n",
      "867:\tlearn: 0.9434004\ttotal: 1m 43s\tremaining: 1m 3s\n",
      "868:\tlearn: 0.9433491\ttotal: 1m 44s\tremaining: 1m 3s\n",
      "869:\tlearn: 0.9433021\ttotal: 1m 44s\tremaining: 1m 3s\n",
      "870:\tlearn: 0.9432903\ttotal: 1m 44s\tremaining: 1m 3s\n",
      "871:\tlearn: 0.9432865\ttotal: 1m 44s\tremaining: 1m 3s\n",
      "872:\tlearn: 0.9432341\ttotal: 1m 44s\tremaining: 1m 3s\n",
      "873:\tlearn: 0.9432143\ttotal: 1m 44s\tremaining: 1m 2s\n",
      "874:\tlearn: 0.9431693\ttotal: 1m 44s\tremaining: 1m 2s\n",
      "875:\tlearn: 0.9431659\ttotal: 1m 44s\tremaining: 1m 2s\n",
      "876:\tlearn: 0.9431437\ttotal: 1m 44s\tremaining: 1m 2s\n",
      "877:\tlearn: 0.9431155\ttotal: 1m 44s\tremaining: 1m 2s\n",
      "878:\tlearn: 0.9430979\ttotal: 1m 45s\tremaining: 1m 2s\n",
      "879:\tlearn: 0.9430547\ttotal: 1m 45s\tremaining: 1m 2s\n",
      "880:\tlearn: 0.9430181\ttotal: 1m 45s\tremaining: 1m 2s\n",
      "881:\tlearn: 0.9429979\ttotal: 1m 45s\tremaining: 1m 1s\n",
      "882:\tlearn: 0.9429774\ttotal: 1m 45s\tremaining: 1m 1s\n",
      "883:\tlearn: 0.9429593\ttotal: 1m 45s\tremaining: 1m 1s\n",
      "884:\tlearn: 0.9429208\ttotal: 1m 45s\tremaining: 1m 1s\n",
      "885:\tlearn: 0.9428773\ttotal: 1m 45s\tremaining: 1m 1s\n",
      "886:\tlearn: 0.9428743\ttotal: 1m 45s\tremaining: 1m 1s\n",
      "887:\tlearn: 0.9428506\ttotal: 1m 46s\tremaining: 1m 1s\n",
      "888:\tlearn: 0.9428222\ttotal: 1m 46s\tremaining: 1m 1s\n",
      "889:\tlearn: 0.9427858\ttotal: 1m 46s\tremaining: 1m\n",
      "890:\tlearn: 0.9427838\ttotal: 1m 46s\tremaining: 1m\n",
      "891:\tlearn: 0.9427531\ttotal: 1m 46s\tremaining: 1m\n",
      "892:\tlearn: 0.9427491\ttotal: 1m 46s\tremaining: 1m\n",
      "893:\tlearn: 0.9427188\ttotal: 1m 46s\tremaining: 1m\n",
      "894:\tlearn: 0.9427164\ttotal: 1m 46s\tremaining: 1m\n",
      "895:\tlearn: 0.9426791\ttotal: 1m 46s\tremaining: 1m\n",
      "896:\tlearn: 0.9426419\ttotal: 1m 46s\tremaining: 60s\n",
      "897:\tlearn: 0.9426087\ttotal: 1m 47s\tremaining: 59.9s\n",
      "898:\tlearn: 0.9425975\ttotal: 1m 47s\tremaining: 59.7s\n",
      "899:\tlearn: 0.9425653\ttotal: 1m 47s\tremaining: 59.6s\n",
      "900:\tlearn: 0.9425349\ttotal: 1m 47s\tremaining: 59.5s\n",
      "901:\tlearn: 0.9424860\ttotal: 1m 47s\tremaining: 59.4s\n",
      "902:\tlearn: 0.9424566\ttotal: 1m 47s\tremaining: 59.2s\n",
      "903:\tlearn: 0.9424134\ttotal: 1m 47s\tremaining: 59.1s\n",
      "904:\tlearn: 0.9423848\ttotal: 1m 47s\tremaining: 59s\n",
      "905:\tlearn: 0.9423615\ttotal: 1m 47s\tremaining: 58.9s\n",
      "906:\tlearn: 0.9423141\ttotal: 1m 48s\tremaining: 58.7s\n",
      "907:\tlearn: 0.9422774\ttotal: 1m 48s\tremaining: 58.6s\n",
      "908:\tlearn: 0.9422437\ttotal: 1m 48s\tremaining: 58.5s\n",
      "909:\tlearn: 0.9422061\ttotal: 1m 48s\tremaining: 58.4s\n",
      "910:\tlearn: 0.9421925\ttotal: 1m 48s\tremaining: 58.2s\n",
      "911:\tlearn: 0.9421718\ttotal: 1m 48s\tremaining: 58.1s\n",
      "912:\tlearn: 0.9421436\ttotal: 1m 48s\tremaining: 58s\n",
      "913:\tlearn: 0.9421121\ttotal: 1m 48s\tremaining: 57.9s\n",
      "914:\tlearn: 0.9421072\ttotal: 1m 48s\tremaining: 57.7s\n",
      "915:\tlearn: 0.9420717\ttotal: 1m 49s\tremaining: 57.6s\n",
      "916:\tlearn: 0.9420440\ttotal: 1m 49s\tremaining: 57.5s\n",
      "917:\tlearn: 0.9420421\ttotal: 1m 49s\tremaining: 57.4s\n",
      "918:\tlearn: 0.9420400\ttotal: 1m 49s\tremaining: 57.2s\n",
      "919:\tlearn: 0.9420226\ttotal: 1m 49s\tremaining: 57.1s\n",
      "920:\tlearn: 0.9420185\ttotal: 1m 49s\tremaining: 56.9s\n",
      "921:\tlearn: 0.9419846\ttotal: 1m 49s\tremaining: 56.8s\n",
      "922:\tlearn: 0.9419611\ttotal: 1m 49s\tremaining: 56.7s\n",
      "923:\tlearn: 0.9419336\ttotal: 1m 49s\tremaining: 56.6s\n",
      "924:\tlearn: 0.9418835\ttotal: 1m 49s\tremaining: 56.4s\n",
      "925:\tlearn: 0.9418451\ttotal: 1m 50s\tremaining: 56.3s\n",
      "926:\tlearn: 0.9418172\ttotal: 1m 50s\tremaining: 56.2s\n",
      "927:\tlearn: 0.9418158\ttotal: 1m 50s\tremaining: 56.1s\n",
      "928:\tlearn: 0.9417626\ttotal: 1m 50s\tremaining: 55.9s\n",
      "929:\tlearn: 0.9417586\ttotal: 1m 50s\tremaining: 55.8s\n",
      "930:\tlearn: 0.9417297\ttotal: 1m 50s\tremaining: 55.7s\n",
      "931:\tlearn: 0.9417287\ttotal: 1m 50s\tremaining: 55.5s\n",
      "932:\tlearn: 0.9417173\ttotal: 1m 50s\tremaining: 55.4s\n",
      "933:\tlearn: 0.9416799\ttotal: 1m 50s\tremaining: 55.3s\n",
      "934:\tlearn: 0.9416211\ttotal: 1m 50s\tremaining: 55.2s\n",
      "935:\tlearn: 0.9416169\ttotal: 1m 51s\tremaining: 55s\n",
      "936:\tlearn: 0.9415925\ttotal: 1m 51s\tremaining: 54.9s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "937:\tlearn: 0.9415606\ttotal: 1m 51s\tremaining: 54.8s\n",
      "938:\tlearn: 0.9415254\ttotal: 1m 51s\tremaining: 54.6s\n",
      "939:\tlearn: 0.9414940\ttotal: 1m 51s\tremaining: 54.5s\n",
      "940:\tlearn: 0.9414751\ttotal: 1m 51s\tremaining: 54.4s\n",
      "941:\tlearn: 0.9414471\ttotal: 1m 51s\tremaining: 54.3s\n",
      "942:\tlearn: 0.9414264\ttotal: 1m 51s\tremaining: 54.1s\n",
      "943:\tlearn: 0.9414249\ttotal: 1m 51s\tremaining: 54s\n",
      "944:\tlearn: 0.9413756\ttotal: 1m 51s\tremaining: 53.9s\n",
      "945:\tlearn: 0.9413223\ttotal: 1m 52s\tremaining: 53.8s\n",
      "946:\tlearn: 0.9413021\ttotal: 1m 52s\tremaining: 53.7s\n",
      "947:\tlearn: 0.9412424\ttotal: 1m 52s\tremaining: 53.5s\n",
      "948:\tlearn: 0.9412105\ttotal: 1m 52s\tremaining: 53.4s\n",
      "949:\tlearn: 0.9411867\ttotal: 1m 52s\tremaining: 53.3s\n",
      "950:\tlearn: 0.9411830\ttotal: 1m 52s\tremaining: 53.2s\n",
      "951:\tlearn: 0.9411670\ttotal: 1m 52s\tremaining: 53s\n",
      "952:\tlearn: 0.9411320\ttotal: 1m 52s\tremaining: 52.9s\n",
      "953:\tlearn: 0.9410894\ttotal: 1m 52s\tremaining: 52.8s\n",
      "954:\tlearn: 0.9410877\ttotal: 1m 52s\tremaining: 52.7s\n",
      "955:\tlearn: 0.9410872\ttotal: 1m 53s\tremaining: 52.5s\n",
      "956:\tlearn: 0.9410667\ttotal: 1m 53s\tremaining: 52.4s\n",
      "957:\tlearn: 0.9410533\ttotal: 1m 53s\tremaining: 52.3s\n",
      "958:\tlearn: 0.9410173\ttotal: 1m 53s\tremaining: 52.2s\n",
      "959:\tlearn: 0.9409887\ttotal: 1m 53s\tremaining: 52s\n",
      "960:\tlearn: 0.9409698\ttotal: 1m 53s\tremaining: 51.9s\n",
      "961:\tlearn: 0.9409417\ttotal: 1m 53s\tremaining: 51.8s\n",
      "962:\tlearn: 0.9408997\ttotal: 1m 53s\tremaining: 51.7s\n",
      "963:\tlearn: 0.9408973\ttotal: 1m 53s\tremaining: 51.6s\n",
      "964:\tlearn: 0.9408656\ttotal: 1m 54s\tremaining: 51.4s\n",
      "965:\tlearn: 0.9408357\ttotal: 1m 54s\tremaining: 51.3s\n",
      "966:\tlearn: 0.9408024\ttotal: 1m 54s\tremaining: 51.2s\n",
      "967:\tlearn: 0.9407730\ttotal: 1m 54s\tremaining: 51.1s\n",
      "968:\tlearn: 0.9407509\ttotal: 1m 54s\tremaining: 51s\n",
      "969:\tlearn: 0.9407226\ttotal: 1m 54s\tremaining: 50.8s\n",
      "970:\tlearn: 0.9406862\ttotal: 1m 54s\tremaining: 50.7s\n",
      "971:\tlearn: 0.9406545\ttotal: 1m 54s\tremaining: 50.6s\n",
      "972:\tlearn: 0.9406267\ttotal: 1m 55s\tremaining: 50.5s\n",
      "973:\tlearn: 0.9405955\ttotal: 1m 55s\tremaining: 50.4s\n",
      "974:\tlearn: 0.9405735\ttotal: 1m 55s\tremaining: 50.2s\n",
      "975:\tlearn: 0.9405156\ttotal: 1m 55s\tremaining: 50.1s\n",
      "976:\tlearn: 0.9405026\ttotal: 1m 55s\tremaining: 50s\n",
      "977:\tlearn: 0.9404653\ttotal: 1m 55s\tremaining: 49.9s\n",
      "978:\tlearn: 0.9404465\ttotal: 1m 55s\tremaining: 49.7s\n",
      "979:\tlearn: 0.9404139\ttotal: 1m 55s\tremaining: 49.6s\n",
      "980:\tlearn: 0.9403795\ttotal: 1m 55s\tremaining: 49.5s\n",
      "981:\tlearn: 0.9403287\ttotal: 1m 55s\tremaining: 49.4s\n",
      "982:\tlearn: 0.9402895\ttotal: 1m 56s\tremaining: 49.2s\n",
      "983:\tlearn: 0.9402649\ttotal: 1m 56s\tremaining: 49.1s\n",
      "984:\tlearn: 0.9402325\ttotal: 1m 56s\tremaining: 49s\n",
      "985:\tlearn: 0.9402259\ttotal: 1m 56s\tremaining: 48.9s\n",
      "986:\tlearn: 0.9402141\ttotal: 1m 56s\tremaining: 48.7s\n",
      "987:\tlearn: 0.9401902\ttotal: 1m 56s\tremaining: 48.6s\n",
      "988:\tlearn: 0.9401611\ttotal: 1m 56s\tremaining: 48.5s\n",
      "989:\tlearn: 0.9401162\ttotal: 1m 56s\tremaining: 48.4s\n",
      "990:\tlearn: 0.9400800\ttotal: 1m 56s\tremaining: 48.3s\n",
      "991:\tlearn: 0.9400627\ttotal: 1m 57s\tremaining: 48.1s\n",
      "992:\tlearn: 0.9400339\ttotal: 1m 57s\tremaining: 48s\n",
      "993:\tlearn: 0.9400003\ttotal: 1m 57s\tremaining: 47.9s\n",
      "994:\tlearn: 0.9399978\ttotal: 1m 57s\tremaining: 47.8s\n",
      "995:\tlearn: 0.9399862\ttotal: 1m 57s\tremaining: 47.6s\n",
      "996:\tlearn: 0.9399524\ttotal: 1m 57s\tremaining: 47.5s\n",
      "997:\tlearn: 0.9399432\ttotal: 1m 57s\tremaining: 47.4s\n",
      "998:\tlearn: 0.9399256\ttotal: 1m 57s\tremaining: 47.2s\n",
      "999:\tlearn: 0.9398841\ttotal: 1m 57s\tremaining: 47.1s\n",
      "1000:\tlearn: 0.9398353\ttotal: 1m 57s\tremaining: 47s\n",
      "1001:\tlearn: 0.9398131\ttotal: 1m 58s\tremaining: 46.9s\n",
      "1002:\tlearn: 0.9398039\ttotal: 1m 58s\tremaining: 46.8s\n",
      "1003:\tlearn: 0.9397581\ttotal: 1m 58s\tremaining: 46.6s\n",
      "1004:\tlearn: 0.9397322\ttotal: 1m 58s\tremaining: 46.5s\n",
      "1005:\tlearn: 0.9397050\ttotal: 1m 58s\tremaining: 46.4s\n",
      "1006:\tlearn: 0.9396797\ttotal: 1m 58s\tremaining: 46.3s\n",
      "1007:\tlearn: 0.9396518\ttotal: 1m 58s\tremaining: 46.2s\n",
      "1008:\tlearn: 0.9396417\ttotal: 1m 58s\tremaining: 46s\n",
      "1009:\tlearn: 0.9396189\ttotal: 1m 58s\tremaining: 45.9s\n",
      "1010:\tlearn: 0.9395848\ttotal: 1m 59s\tremaining: 45.8s\n",
      "1011:\tlearn: 0.9395276\ttotal: 1m 59s\tremaining: 45.7s\n",
      "1012:\tlearn: 0.9394850\ttotal: 1m 59s\tremaining: 45.6s\n",
      "1013:\tlearn: 0.9394461\ttotal: 1m 59s\tremaining: 45.5s\n",
      "1014:\tlearn: 0.9394291\ttotal: 1m 59s\tremaining: 45.3s\n",
      "1015:\tlearn: 0.9394158\ttotal: 1m 59s\tremaining: 45.2s\n",
      "1016:\tlearn: 0.9393806\ttotal: 1m 59s\tremaining: 45.1s\n",
      "1017:\tlearn: 0.9393472\ttotal: 1m 59s\tremaining: 45s\n",
      "1018:\tlearn: 0.9393084\ttotal: 1m 59s\tremaining: 44.8s\n",
      "1019:\tlearn: 0.9392901\ttotal: 2m\tremaining: 44.7s\n",
      "1020:\tlearn: 0.9392422\ttotal: 2m\tremaining: 44.6s\n",
      "1021:\tlearn: 0.9392204\ttotal: 2m\tremaining: 44.5s\n",
      "1022:\tlearn: 0.9392189\ttotal: 2m\tremaining: 44.4s\n",
      "1023:\tlearn: 0.9392072\ttotal: 2m\tremaining: 44.2s\n",
      "1024:\tlearn: 0.9391972\ttotal: 2m\tremaining: 44.1s\n",
      "1025:\tlearn: 0.9391803\ttotal: 2m\tremaining: 44s\n",
      "1026:\tlearn: 0.9391787\ttotal: 2m\tremaining: 43.8s\n",
      "1027:\tlearn: 0.9391517\ttotal: 2m\tremaining: 43.7s\n",
      "1028:\tlearn: 0.9391217\ttotal: 2m\tremaining: 43.6s\n",
      "1029:\tlearn: 0.9390954\ttotal: 2m 1s\tremaining: 43.5s\n",
      "1030:\tlearn: 0.9390618\ttotal: 2m 1s\tremaining: 43.4s\n",
      "1031:\tlearn: 0.9390506\ttotal: 2m 1s\tremaining: 43.2s\n",
      "1032:\tlearn: 0.9390150\ttotal: 2m 1s\tremaining: 43.1s\n",
      "1033:\tlearn: 0.9389963\ttotal: 2m 1s\tremaining: 43s\n",
      "1034:\tlearn: 0.9389621\ttotal: 2m 1s\tremaining: 42.9s\n",
      "1035:\tlearn: 0.9389257\ttotal: 2m 1s\tremaining: 42.7s\n",
      "1036:\tlearn: 0.9388974\ttotal: 2m 1s\tremaining: 42.6s\n",
      "1037:\tlearn: 0.9388559\ttotal: 2m 1s\tremaining: 42.5s\n",
      "1038:\tlearn: 0.9388396\ttotal: 2m 1s\tremaining: 42.4s\n",
      "1039:\tlearn: 0.9388041\ttotal: 2m 2s\tremaining: 42.3s\n",
      "1040:\tlearn: 0.9387653\ttotal: 2m 2s\tremaining: 42.2s\n",
      "1041:\tlearn: 0.9387126\ttotal: 2m 2s\tremaining: 42s\n",
      "1042:\tlearn: 0.9386794\ttotal: 2m 2s\tremaining: 41.9s\n",
      "1043:\tlearn: 0.9386525\ttotal: 2m 2s\tremaining: 41.8s\n",
      "1044:\tlearn: 0.9386507\ttotal: 2m 2s\tremaining: 41.7s\n",
      "1045:\tlearn: 0.9386124\ttotal: 2m 2s\tremaining: 41.6s\n",
      "1046:\tlearn: 0.9386027\ttotal: 2m 2s\tremaining: 41.5s\n",
      "1047:\tlearn: 0.9385434\ttotal: 2m 3s\tremaining: 41.4s\n",
      "1048:\tlearn: 0.9385296\ttotal: 2m 3s\tremaining: 41.2s\n",
      "1049:\tlearn: 0.9385284\ttotal: 2m 3s\tremaining: 41.1s\n",
      "1050:\tlearn: 0.9385075\ttotal: 2m 3s\tremaining: 41s\n",
      "1051:\tlearn: 0.9384792\ttotal: 2m 3s\tremaining: 40.9s\n",
      "1052:\tlearn: 0.9384315\ttotal: 2m 3s\tremaining: 40.8s\n",
      "1053:\tlearn: 0.9384100\ttotal: 2m 3s\tremaining: 40.6s\n",
      "1054:\tlearn: 0.9384000\ttotal: 2m 3s\tremaining: 40.5s\n",
      "1055:\tlearn: 0.9383579\ttotal: 2m 4s\tremaining: 40.4s\n",
      "1056:\tlearn: 0.9383338\ttotal: 2m 4s\tremaining: 40.3s\n",
      "1057:\tlearn: 0.9383098\ttotal: 2m 4s\tremaining: 40.2s\n",
      "1058:\tlearn: 0.9382925\ttotal: 2m 4s\tremaining: 40.1s\n",
      "1059:\tlearn: 0.9382500\ttotal: 2m 4s\tremaining: 39.9s\n",
      "1060:\tlearn: 0.9382068\ttotal: 2m 4s\tremaining: 39.8s\n",
      "1061:\tlearn: 0.9381870\ttotal: 2m 4s\tremaining: 39.7s\n",
      "1062:\tlearn: 0.9381646\ttotal: 2m 4s\tremaining: 39.6s\n",
      "1063:\tlearn: 0.9381282\ttotal: 2m 4s\tremaining: 39.5s\n",
      "1064:\tlearn: 0.9381045\ttotal: 2m 5s\tremaining: 39.3s\n",
      "1065:\tlearn: 0.9380565\ttotal: 2m 5s\tremaining: 39.2s\n",
      "1066:\tlearn: 0.9380273\ttotal: 2m 5s\tremaining: 39.1s\n",
      "1067:\tlearn: 0.9380010\ttotal: 2m 5s\tremaining: 39s\n",
      "1068:\tlearn: 0.9379641\ttotal: 2m 5s\tremaining: 38.9s\n",
      "1069:\tlearn: 0.9379374\ttotal: 2m 5s\tremaining: 38.8s\n",
      "1070:\tlearn: 0.9379261\ttotal: 2m 5s\tremaining: 38.6s\n",
      "1071:\tlearn: 0.9378995\ttotal: 2m 5s\tremaining: 38.5s\n",
      "1072:\tlearn: 0.9378774\ttotal: 2m 5s\tremaining: 38.4s\n",
      "1073:\tlearn: 0.9378242\ttotal: 2m 6s\tremaining: 38.3s\n",
      "1074:\tlearn: 0.9377830\ttotal: 2m 6s\tremaining: 38.1s\n",
      "1075:\tlearn: 0.9377825\ttotal: 2m 6s\tremaining: 38s\n",
      "1076:\tlearn: 0.9377453\ttotal: 2m 6s\tremaining: 37.9s\n",
      "1077:\tlearn: 0.9377326\ttotal: 2m 6s\tremaining: 37.8s\n",
      "1078:\tlearn: 0.9377102\ttotal: 2m 6s\tremaining: 37.6s\n",
      "1079:\tlearn: 0.9376770\ttotal: 2m 6s\tremaining: 37.5s\n",
      "1080:\tlearn: 0.9376475\ttotal: 2m 6s\tremaining: 37.4s\n",
      "1081:\tlearn: 0.9376121\ttotal: 2m 6s\tremaining: 37.3s\n",
      "1082:\tlearn: 0.9375765\ttotal: 2m 6s\tremaining: 37.2s\n",
      "1083:\tlearn: 0.9375487\ttotal: 2m 7s\tremaining: 37s\n",
      "1084:\tlearn: 0.9375056\ttotal: 2m 7s\tremaining: 36.9s\n",
      "1085:\tlearn: 0.9374731\ttotal: 2m 7s\tremaining: 36.8s\n",
      "1086:\tlearn: 0.9374533\ttotal: 2m 7s\tremaining: 36.7s\n",
      "1087:\tlearn: 0.9374206\ttotal: 2m 7s\tremaining: 36.6s\n",
      "1088:\tlearn: 0.9374065\ttotal: 2m 7s\tremaining: 36.4s\n",
      "1089:\tlearn: 0.9373768\ttotal: 2m 7s\tremaining: 36.3s\n",
      "1090:\tlearn: 0.9373528\ttotal: 2m 7s\tremaining: 36.2s\n",
      "1091:\tlearn: 0.9373483\ttotal: 2m 7s\tremaining: 36.1s\n",
      "1092:\tlearn: 0.9373162\ttotal: 2m 7s\tremaining: 35.9s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1093:\tlearn: 0.9372903\ttotal: 2m 8s\tremaining: 35.8s\n",
      "1094:\tlearn: 0.9372581\ttotal: 2m 8s\tremaining: 35.7s\n",
      "1095:\tlearn: 0.9372273\ttotal: 2m 8s\tremaining: 35.6s\n",
      "1096:\tlearn: 0.9372241\ttotal: 2m 8s\tremaining: 35.5s\n",
      "1097:\tlearn: 0.9371988\ttotal: 2m 8s\tremaining: 35.3s\n",
      "1098:\tlearn: 0.9371944\ttotal: 2m 8s\tremaining: 35.2s\n",
      "1099:\tlearn: 0.9371721\ttotal: 2m 8s\tremaining: 35.1s\n",
      "1100:\tlearn: 0.9371715\ttotal: 2m 8s\tremaining: 35s\n",
      "1101:\tlearn: 0.9371415\ttotal: 2m 8s\tremaining: 34.8s\n",
      "1102:\tlearn: 0.9370983\ttotal: 2m 8s\tremaining: 34.7s\n",
      "1103:\tlearn: 0.9370615\ttotal: 2m 9s\tremaining: 34.6s\n",
      "1104:\tlearn: 0.9370266\ttotal: 2m 9s\tremaining: 34.5s\n",
      "1105:\tlearn: 0.9370222\ttotal: 2m 9s\tremaining: 34.4s\n",
      "1106:\tlearn: 0.9370093\ttotal: 2m 9s\tremaining: 34.2s\n",
      "1107:\tlearn: 0.9370063\ttotal: 2m 9s\tremaining: 34.1s\n",
      "1108:\tlearn: 0.9369837\ttotal: 2m 9s\tremaining: 34s\n",
      "1109:\tlearn: 0.9369621\ttotal: 2m 9s\tremaining: 33.9s\n",
      "1110:\tlearn: 0.9369345\ttotal: 2m 9s\tremaining: 33.8s\n",
      "1111:\tlearn: 0.9369324\ttotal: 2m 9s\tremaining: 33.6s\n",
      "1112:\tlearn: 0.9369183\ttotal: 2m 9s\tremaining: 33.5s\n",
      "1113:\tlearn: 0.9369005\ttotal: 2m 10s\tremaining: 33.4s\n",
      "1114:\tlearn: 0.9368612\ttotal: 2m 10s\tremaining: 33.3s\n",
      "1115:\tlearn: 0.9368409\ttotal: 2m 10s\tremaining: 33.2s\n",
      "1116:\tlearn: 0.9368023\ttotal: 2m 10s\tremaining: 33s\n",
      "1117:\tlearn: 0.9367819\ttotal: 2m 10s\tremaining: 32.9s\n",
      "1118:\tlearn: 0.9367440\ttotal: 2m 10s\tremaining: 32.8s\n",
      "1119:\tlearn: 0.9367218\ttotal: 2m 10s\tremaining: 32.7s\n",
      "1120:\tlearn: 0.9366835\ttotal: 2m 10s\tremaining: 32.6s\n",
      "1121:\tlearn: 0.9366676\ttotal: 2m 10s\tremaining: 32.4s\n",
      "1122:\tlearn: 0.9366302\ttotal: 2m 11s\tremaining: 32.3s\n",
      "1123:\tlearn: 0.9366295\ttotal: 2m 11s\tremaining: 32.2s\n",
      "1124:\tlearn: 0.9366132\ttotal: 2m 11s\tremaining: 32.1s\n",
      "1125:\tlearn: 0.9365766\ttotal: 2m 11s\tremaining: 32s\n",
      "1126:\tlearn: 0.9365589\ttotal: 2m 11s\tremaining: 31.8s\n",
      "1127:\tlearn: 0.9365528\ttotal: 2m 11s\tremaining: 31.7s\n",
      "1128:\tlearn: 0.9365193\ttotal: 2m 11s\tremaining: 31.6s\n",
      "1129:\tlearn: 0.9365008\ttotal: 2m 11s\tremaining: 31.5s\n",
      "1130:\tlearn: 0.9364973\ttotal: 2m 11s\tremaining: 31.3s\n",
      "1131:\tlearn: 0.9364821\ttotal: 2m 11s\tremaining: 31.2s\n",
      "1132:\tlearn: 0.9364512\ttotal: 2m 11s\tremaining: 31.1s\n",
      "1133:\tlearn: 0.9364322\ttotal: 2m 12s\tremaining: 31s\n",
      "1134:\tlearn: 0.9364023\ttotal: 2m 12s\tremaining: 30.9s\n",
      "1135:\tlearn: 0.9364014\ttotal: 2m 12s\tremaining: 30.7s\n",
      "1136:\tlearn: 0.9363592\ttotal: 2m 12s\tremaining: 30.6s\n",
      "1137:\tlearn: 0.9363458\ttotal: 2m 12s\tremaining: 30.5s\n",
      "1138:\tlearn: 0.9363188\ttotal: 2m 12s\tremaining: 30.4s\n",
      "1139:\tlearn: 0.9362980\ttotal: 2m 12s\tremaining: 30.3s\n",
      "1140:\tlearn: 0.9362655\ttotal: 2m 12s\tremaining: 30.1s\n",
      "1141:\tlearn: 0.9362394\ttotal: 2m 12s\tremaining: 30s\n",
      "1142:\tlearn: 0.9362131\ttotal: 2m 13s\tremaining: 29.9s\n",
      "1143:\tlearn: 0.9361854\ttotal: 2m 13s\tremaining: 29.8s\n",
      "1144:\tlearn: 0.9361488\ttotal: 2m 13s\tremaining: 29.7s\n",
      "1145:\tlearn: 0.9361256\ttotal: 2m 13s\tremaining: 29.6s\n",
      "1146:\tlearn: 0.9361028\ttotal: 2m 13s\tremaining: 29.4s\n",
      "1147:\tlearn: 0.9360998\ttotal: 2m 13s\tremaining: 29.3s\n",
      "1148:\tlearn: 0.9360878\ttotal: 2m 13s\tremaining: 29.2s\n",
      "1149:\tlearn: 0.9360613\ttotal: 2m 13s\tremaining: 29.1s\n",
      "1150:\tlearn: 0.9360531\ttotal: 2m 13s\tremaining: 29s\n",
      "1151:\tlearn: 0.9359990\ttotal: 2m 13s\tremaining: 28.8s\n",
      "1152:\tlearn: 0.9359863\ttotal: 2m 14s\tremaining: 28.7s\n",
      "1153:\tlearn: 0.9359620\ttotal: 2m 14s\tremaining: 28.6s\n",
      "1154:\tlearn: 0.9359516\ttotal: 2m 14s\tremaining: 28.5s\n",
      "1155:\tlearn: 0.9359266\ttotal: 2m 14s\tremaining: 28.4s\n",
      "1156:\tlearn: 0.9359193\ttotal: 2m 14s\tremaining: 28.2s\n",
      "1157:\tlearn: 0.9359170\ttotal: 2m 14s\tremaining: 28.1s\n",
      "1158:\tlearn: 0.9359071\ttotal: 2m 14s\tremaining: 28s\n",
      "1159:\tlearn: 0.9358570\ttotal: 2m 14s\tremaining: 27.9s\n",
      "1160:\tlearn: 0.9358484\ttotal: 2m 14s\tremaining: 27.8s\n",
      "1161:\tlearn: 0.9358202\ttotal: 2m 14s\tremaining: 27.6s\n",
      "1162:\tlearn: 0.9358003\ttotal: 2m 15s\tremaining: 27.5s\n",
      "1163:\tlearn: 0.9357594\ttotal: 2m 15s\tremaining: 27.4s\n",
      "1164:\tlearn: 0.9357403\ttotal: 2m 15s\tremaining: 27.3s\n",
      "1165:\tlearn: 0.9356842\ttotal: 2m 15s\tremaining: 27.2s\n",
      "1166:\tlearn: 0.9356470\ttotal: 2m 15s\tremaining: 27.1s\n",
      "1167:\tlearn: 0.9356254\ttotal: 2m 15s\tremaining: 26.9s\n",
      "1168:\tlearn: 0.9355856\ttotal: 2m 15s\tremaining: 26.8s\n",
      "1169:\tlearn: 0.9355535\ttotal: 2m 15s\tremaining: 26.7s\n",
      "1170:\tlearn: 0.9355407\ttotal: 2m 15s\tremaining: 26.6s\n",
      "1171:\tlearn: 0.9355325\ttotal: 2m 16s\tremaining: 26.5s\n",
      "1172:\tlearn: 0.9354900\ttotal: 2m 16s\tremaining: 26.3s\n",
      "1173:\tlearn: 0.9354312\ttotal: 2m 16s\tremaining: 26.2s\n",
      "1174:\tlearn: 0.9354149\ttotal: 2m 16s\tremaining: 26.1s\n",
      "1175:\tlearn: 0.9354009\ttotal: 2m 16s\tremaining: 26s\n",
      "1176:\tlearn: 0.9353520\ttotal: 2m 16s\tremaining: 25.9s\n",
      "1177:\tlearn: 0.9353340\ttotal: 2m 16s\tremaining: 25.8s\n",
      "1178:\tlearn: 0.9352981\ttotal: 2m 16s\tremaining: 25.6s\n",
      "1179:\tlearn: 0.9352895\ttotal: 2m 16s\tremaining: 25.5s\n",
      "1180:\tlearn: 0.9352433\ttotal: 2m 17s\tremaining: 25.4s\n",
      "1181:\tlearn: 0.9352014\ttotal: 2m 17s\tremaining: 25.3s\n",
      "1182:\tlearn: 0.9351751\ttotal: 2m 17s\tremaining: 25.2s\n",
      "1183:\tlearn: 0.9351442\ttotal: 2m 17s\tremaining: 25.1s\n",
      "1184:\tlearn: 0.9351157\ttotal: 2m 17s\tremaining: 24.9s\n",
      "1185:\tlearn: 0.9350901\ttotal: 2m 17s\tremaining: 24.8s\n",
      "1186:\tlearn: 0.9350610\ttotal: 2m 17s\tremaining: 24.7s\n",
      "1187:\tlearn: 0.9350340\ttotal: 2m 17s\tremaining: 24.6s\n",
      "1188:\tlearn: 0.9350157\ttotal: 2m 17s\tremaining: 24.5s\n",
      "1189:\tlearn: 0.9350087\ttotal: 2m 17s\tremaining: 24.3s\n",
      "1190:\tlearn: 0.9349742\ttotal: 2m 18s\tremaining: 24.2s\n",
      "1191:\tlearn: 0.9349487\ttotal: 2m 18s\tremaining: 24.1s\n",
      "1192:\tlearn: 0.9349194\ttotal: 2m 18s\tremaining: 24s\n",
      "1193:\tlearn: 0.9348983\ttotal: 2m 18s\tremaining: 23.9s\n",
      "1194:\tlearn: 0.9348674\ttotal: 2m 18s\tremaining: 23.8s\n",
      "1195:\tlearn: 0.9348501\ttotal: 2m 18s\tremaining: 23.6s\n",
      "1196:\tlearn: 0.9347902\ttotal: 2m 18s\tremaining: 23.5s\n",
      "1197:\tlearn: 0.9347768\ttotal: 2m 18s\tremaining: 23.4s\n",
      "1198:\tlearn: 0.9347429\ttotal: 2m 18s\tremaining: 23.3s\n",
      "1199:\tlearn: 0.9347204\ttotal: 2m 18s\tremaining: 23.2s\n",
      "1200:\tlearn: 0.9346931\ttotal: 2m 19s\tremaining: 23.1s\n",
      "1201:\tlearn: 0.9346692\ttotal: 2m 19s\tremaining: 22.9s\n",
      "1202:\tlearn: 0.9346441\ttotal: 2m 19s\tremaining: 22.8s\n",
      "1203:\tlearn: 0.9346298\ttotal: 2m 19s\tremaining: 22.7s\n",
      "1204:\tlearn: 0.9346124\ttotal: 2m 19s\tremaining: 22.6s\n",
      "1205:\tlearn: 0.9345796\ttotal: 2m 19s\tremaining: 22.5s\n",
      "1206:\tlearn: 0.9345120\ttotal: 2m 19s\tremaining: 22.4s\n",
      "1207:\tlearn: 0.9344945\ttotal: 2m 19s\tremaining: 22.2s\n",
      "1208:\tlearn: 0.9344713\ttotal: 2m 19s\tremaining: 22.1s\n",
      "1209:\tlearn: 0.9344586\ttotal: 2m 20s\tremaining: 22s\n",
      "1210:\tlearn: 0.9344270\ttotal: 2m 20s\tremaining: 21.9s\n",
      "1211:\tlearn: 0.9343841\ttotal: 2m 20s\tremaining: 21.8s\n",
      "1212:\tlearn: 0.9343635\ttotal: 2m 20s\tremaining: 21.6s\n",
      "1213:\tlearn: 0.9343381\ttotal: 2m 20s\tremaining: 21.5s\n",
      "1214:\tlearn: 0.9343067\ttotal: 2m 20s\tremaining: 21.4s\n",
      "1215:\tlearn: 0.9342952\ttotal: 2m 20s\tremaining: 21.3s\n",
      "1216:\tlearn: 0.9342670\ttotal: 2m 20s\tremaining: 21.2s\n",
      "1217:\tlearn: 0.9342362\ttotal: 2m 20s\tremaining: 21.1s\n",
      "1218:\tlearn: 0.9342186\ttotal: 2m 21s\tremaining: 20.9s\n",
      "1219:\tlearn: 0.9342054\ttotal: 2m 21s\tremaining: 20.8s\n",
      "1220:\tlearn: 0.9341672\ttotal: 2m 21s\tremaining: 20.7s\n",
      "1221:\tlearn: 0.9341514\ttotal: 2m 21s\tremaining: 20.6s\n",
      "1222:\tlearn: 0.9341269\ttotal: 2m 21s\tremaining: 20.5s\n",
      "1223:\tlearn: 0.9341044\ttotal: 2m 21s\tremaining: 20.4s\n",
      "1224:\tlearn: 0.9340781\ttotal: 2m 21s\tremaining: 20.2s\n",
      "1225:\tlearn: 0.9340705\ttotal: 2m 21s\tremaining: 20.1s\n",
      "1226:\tlearn: 0.9340523\ttotal: 2m 21s\tremaining: 20s\n",
      "1227:\tlearn: 0.9340081\ttotal: 2m 21s\tremaining: 19.9s\n",
      "1228:\tlearn: 0.9339811\ttotal: 2m 22s\tremaining: 19.8s\n",
      "1229:\tlearn: 0.9339778\ttotal: 2m 22s\tremaining: 19.6s\n",
      "1230:\tlearn: 0.9339498\ttotal: 2m 22s\tremaining: 19.5s\n",
      "1231:\tlearn: 0.9339267\ttotal: 2m 22s\tremaining: 19.4s\n",
      "1232:\tlearn: 0.9338820\ttotal: 2m 22s\tremaining: 19.3s\n",
      "1233:\tlearn: 0.9338456\ttotal: 2m 22s\tremaining: 19.2s\n",
      "1234:\tlearn: 0.9338177\ttotal: 2m 22s\tremaining: 19.1s\n",
      "1235:\tlearn: 0.9338156\ttotal: 2m 22s\tremaining: 18.9s\n",
      "1236:\tlearn: 0.9338135\ttotal: 2m 22s\tremaining: 18.8s\n",
      "1237:\tlearn: 0.9337661\ttotal: 2m 22s\tremaining: 18.7s\n",
      "1238:\tlearn: 0.9337335\ttotal: 2m 23s\tremaining: 18.6s\n",
      "1239:\tlearn: 0.9337321\ttotal: 2m 23s\tremaining: 18.5s\n",
      "1240:\tlearn: 0.9337268\ttotal: 2m 23s\tremaining: 18.4s\n",
      "1241:\tlearn: 0.9337110\ttotal: 2m 23s\tremaining: 18.2s\n",
      "1242:\tlearn: 0.9336934\ttotal: 2m 23s\tremaining: 18.1s\n",
      "1243:\tlearn: 0.9336668\ttotal: 2m 23s\tremaining: 18s\n",
      "1244:\tlearn: 0.9336481\ttotal: 2m 23s\tremaining: 17.9s\n",
      "1245:\tlearn: 0.9336467\ttotal: 2m 23s\tremaining: 17.8s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1246:\tlearn: 0.9336229\ttotal: 2m 23s\tremaining: 17.6s\n",
      "1247:\tlearn: 0.9335946\ttotal: 2m 23s\tremaining: 17.5s\n",
      "1248:\tlearn: 0.9335552\ttotal: 2m 24s\tremaining: 17.4s\n",
      "1249:\tlearn: 0.9335282\ttotal: 2m 24s\tremaining: 17.3s\n",
      "1250:\tlearn: 0.9335081\ttotal: 2m 24s\tremaining: 17.2s\n",
      "1251:\tlearn: 0.9334476\ttotal: 2m 24s\tremaining: 17.1s\n",
      "1252:\tlearn: 0.9333945\ttotal: 2m 24s\tremaining: 17s\n",
      "1253:\tlearn: 0.9333622\ttotal: 2m 24s\tremaining: 16.8s\n",
      "1254:\tlearn: 0.9333256\ttotal: 2m 24s\tremaining: 16.7s\n",
      "1255:\tlearn: 0.9333238\ttotal: 2m 24s\tremaining: 16.6s\n",
      "1256:\tlearn: 0.9332736\ttotal: 2m 24s\tremaining: 16.5s\n",
      "1257:\tlearn: 0.9332513\ttotal: 2m 25s\tremaining: 16.4s\n",
      "1258:\tlearn: 0.9332219\ttotal: 2m 25s\tremaining: 16.3s\n",
      "1259:\tlearn: 0.9331827\ttotal: 2m 25s\tremaining: 16.1s\n",
      "1260:\tlearn: 0.9331515\ttotal: 2m 25s\tremaining: 16s\n",
      "1261:\tlearn: 0.9331162\ttotal: 2m 25s\tremaining: 15.9s\n",
      "1262:\tlearn: 0.9330971\ttotal: 2m 25s\tremaining: 15.8s\n",
      "1263:\tlearn: 0.9330592\ttotal: 2m 25s\tremaining: 15.7s\n",
      "1264:\tlearn: 0.9330461\ttotal: 2m 25s\tremaining: 15.6s\n",
      "1265:\tlearn: 0.9330160\ttotal: 2m 25s\tremaining: 15.4s\n",
      "1266:\tlearn: 0.9329917\ttotal: 2m 25s\tremaining: 15.3s\n",
      "1267:\tlearn: 0.9329652\ttotal: 2m 26s\tremaining: 15.2s\n",
      "1268:\tlearn: 0.9329379\ttotal: 2m 26s\tremaining: 15.1s\n",
      "1269:\tlearn: 0.9328935\ttotal: 2m 26s\tremaining: 15s\n",
      "1270:\tlearn: 0.9328647\ttotal: 2m 26s\tremaining: 14.9s\n",
      "1271:\tlearn: 0.9328326\ttotal: 2m 26s\tremaining: 14.7s\n",
      "1272:\tlearn: 0.9327894\ttotal: 2m 26s\tremaining: 14.6s\n",
      "1273:\tlearn: 0.9327590\ttotal: 2m 26s\tremaining: 14.5s\n",
      "1274:\tlearn: 0.9327285\ttotal: 2m 26s\tremaining: 14.4s\n",
      "1275:\tlearn: 0.9326929\ttotal: 2m 26s\tremaining: 14.3s\n",
      "1276:\tlearn: 0.9326400\ttotal: 2m 27s\tremaining: 14.2s\n",
      "1277:\tlearn: 0.9326042\ttotal: 2m 27s\tremaining: 14s\n",
      "1278:\tlearn: 0.9325829\ttotal: 2m 27s\tremaining: 13.9s\n",
      "1279:\tlearn: 0.9325517\ttotal: 2m 27s\tremaining: 13.8s\n",
      "1280:\tlearn: 0.9325321\ttotal: 2m 27s\tremaining: 13.7s\n",
      "1281:\tlearn: 0.9325012\ttotal: 2m 27s\tremaining: 13.6s\n",
      "1282:\tlearn: 0.9324871\ttotal: 2m 27s\tremaining: 13.5s\n",
      "1283:\tlearn: 0.9324573\ttotal: 2m 27s\tremaining: 13.3s\n",
      "1284:\tlearn: 0.9324390\ttotal: 2m 27s\tremaining: 13.2s\n",
      "1285:\tlearn: 0.9324223\ttotal: 2m 27s\tremaining: 13.1s\n",
      "1286:\tlearn: 0.9324002\ttotal: 2m 28s\tremaining: 13s\n",
      "1287:\tlearn: 0.9323973\ttotal: 2m 28s\tremaining: 12.9s\n",
      "1288:\tlearn: 0.9323613\ttotal: 2m 28s\tremaining: 12.8s\n",
      "1289:\tlearn: 0.9323415\ttotal: 2m 28s\tremaining: 12.7s\n",
      "1290:\tlearn: 0.9323129\ttotal: 2m 28s\tremaining: 12.5s\n",
      "1291:\tlearn: 0.9322914\ttotal: 2m 28s\tremaining: 12.4s\n",
      "1292:\tlearn: 0.9322653\ttotal: 2m 28s\tremaining: 12.3s\n",
      "1293:\tlearn: 0.9322635\ttotal: 2m 28s\tremaining: 12.2s\n",
      "1294:\tlearn: 0.9322443\ttotal: 2m 28s\tremaining: 12.1s\n",
      "1295:\tlearn: 0.9322004\ttotal: 2m 28s\tremaining: 12s\n",
      "1296:\tlearn: 0.9321761\ttotal: 2m 29s\tremaining: 11.8s\n",
      "1297:\tlearn: 0.9321500\ttotal: 2m 29s\tremaining: 11.7s\n",
      "1298:\tlearn: 0.9321465\ttotal: 2m 29s\tremaining: 11.6s\n",
      "1299:\tlearn: 0.9321142\ttotal: 2m 29s\tremaining: 11.5s\n",
      "1300:\tlearn: 0.9320865\ttotal: 2m 29s\tremaining: 11.4s\n",
      "1301:\tlearn: 0.9320815\ttotal: 2m 29s\tremaining: 11.3s\n",
      "1302:\tlearn: 0.9320579\ttotal: 2m 29s\tremaining: 11.1s\n",
      "1303:\tlearn: 0.9320382\ttotal: 2m 29s\tremaining: 11s\n",
      "1304:\tlearn: 0.9320220\ttotal: 2m 29s\tremaining: 10.9s\n",
      "1305:\tlearn: 0.9320204\ttotal: 2m 30s\tremaining: 10.8s\n",
      "1306:\tlearn: 0.9319859\ttotal: 2m 30s\tremaining: 10.7s\n",
      "1307:\tlearn: 0.9319844\ttotal: 2m 30s\tremaining: 10.6s\n",
      "1308:\tlearn: 0.9319812\ttotal: 2m 30s\tremaining: 10.4s\n",
      "1309:\tlearn: 0.9319782\ttotal: 2m 30s\tremaining: 10.3s\n",
      "1310:\tlearn: 0.9319746\ttotal: 2m 30s\tremaining: 10.2s\n",
      "1311:\tlearn: 0.9319584\ttotal: 2m 30s\tremaining: 10.1s\n",
      "1312:\tlearn: 0.9319333\ttotal: 2m 30s\tremaining: 9.98s\n",
      "1313:\tlearn: 0.9319257\ttotal: 2m 30s\tremaining: 9.87s\n",
      "1314:\tlearn: 0.9318806\ttotal: 2m 30s\tremaining: 9.75s\n",
      "1315:\tlearn: 0.9318548\ttotal: 2m 30s\tremaining: 9.64s\n",
      "1316:\tlearn: 0.9318261\ttotal: 2m 31s\tremaining: 9.52s\n",
      "1317:\tlearn: 0.9317911\ttotal: 2m 31s\tremaining: 9.41s\n",
      "1318:\tlearn: 0.9317589\ttotal: 2m 31s\tremaining: 9.29s\n",
      "1319:\tlearn: 0.9317318\ttotal: 2m 31s\tremaining: 9.18s\n",
      "1320:\tlearn: 0.9317140\ttotal: 2m 31s\tremaining: 9.06s\n",
      "1321:\tlearn: 0.9316813\ttotal: 2m 31s\tremaining: 8.95s\n",
      "1322:\tlearn: 0.9316519\ttotal: 2m 31s\tremaining: 8.83s\n",
      "1323:\tlearn: 0.9316257\ttotal: 2m 31s\tremaining: 8.72s\n",
      "1324:\tlearn: 0.9316034\ttotal: 2m 31s\tremaining: 8.6s\n",
      "1325:\tlearn: 0.9315918\ttotal: 2m 32s\tremaining: 8.49s\n",
      "1326:\tlearn: 0.9315906\ttotal: 2m 32s\tremaining: 8.37s\n",
      "1327:\tlearn: 0.9315745\ttotal: 2m 32s\tremaining: 8.26s\n",
      "1328:\tlearn: 0.9315406\ttotal: 2m 32s\tremaining: 8.14s\n",
      "1329:\tlearn: 0.9315226\ttotal: 2m 32s\tremaining: 8.02s\n",
      "1330:\tlearn: 0.9315073\ttotal: 2m 32s\tremaining: 7.91s\n",
      "1331:\tlearn: 0.9314830\ttotal: 2m 32s\tremaining: 7.79s\n",
      "1332:\tlearn: 0.9314610\ttotal: 2m 32s\tremaining: 7.68s\n",
      "1333:\tlearn: 0.9314540\ttotal: 2m 32s\tremaining: 7.56s\n",
      "1334:\tlearn: 0.9314275\ttotal: 2m 32s\tremaining: 7.45s\n",
      "1335:\tlearn: 0.9313912\ttotal: 2m 33s\tremaining: 7.33s\n",
      "1336:\tlearn: 0.9313600\ttotal: 2m 33s\tremaining: 7.21s\n",
      "1337:\tlearn: 0.9313515\ttotal: 2m 33s\tremaining: 7.1s\n",
      "1338:\tlearn: 0.9313270\ttotal: 2m 33s\tremaining: 6.99s\n",
      "1339:\tlearn: 0.9312943\ttotal: 2m 33s\tremaining: 6.87s\n",
      "1340:\tlearn: 0.9312590\ttotal: 2m 33s\tremaining: 6.75s\n",
      "1341:\tlearn: 0.9312274\ttotal: 2m 33s\tremaining: 6.64s\n",
      "1342:\tlearn: 0.9312081\ttotal: 2m 33s\tremaining: 6.53s\n",
      "1343:\tlearn: 0.9311702\ttotal: 2m 33s\tremaining: 6.41s\n",
      "1344:\tlearn: 0.9311426\ttotal: 2m 33s\tremaining: 6.3s\n",
      "1345:\tlearn: 0.9311174\ttotal: 2m 34s\tremaining: 6.18s\n",
      "1346:\tlearn: 0.9311102\ttotal: 2m 34s\tremaining: 6.07s\n",
      "1347:\tlearn: 0.9311062\ttotal: 2m 34s\tremaining: 5.95s\n",
      "1348:\tlearn: 0.9310800\ttotal: 2m 34s\tremaining: 5.83s\n",
      "1349:\tlearn: 0.9310149\ttotal: 2m 34s\tremaining: 5.72s\n",
      "1350:\tlearn: 0.9309806\ttotal: 2m 34s\tremaining: 5.61s\n",
      "1351:\tlearn: 0.9309502\ttotal: 2m 34s\tremaining: 5.49s\n",
      "1352:\tlearn: 0.9309252\ttotal: 2m 34s\tremaining: 5.38s\n",
      "1353:\tlearn: 0.9308915\ttotal: 2m 34s\tremaining: 5.26s\n",
      "1354:\tlearn: 0.9308651\ttotal: 2m 35s\tremaining: 5.15s\n",
      "1355:\tlearn: 0.9308336\ttotal: 2m 35s\tremaining: 5.03s\n",
      "1356:\tlearn: 0.9308232\ttotal: 2m 35s\tremaining: 4.92s\n",
      "1357:\tlearn: 0.9307996\ttotal: 2m 35s\tremaining: 4.8s\n",
      "1358:\tlearn: 0.9307806\ttotal: 2m 35s\tremaining: 4.69s\n",
      "1359:\tlearn: 0.9307482\ttotal: 2m 35s\tremaining: 4.58s\n",
      "1360:\tlearn: 0.9307161\ttotal: 2m 35s\tremaining: 4.46s\n",
      "1361:\tlearn: 0.9306851\ttotal: 2m 35s\tremaining: 4.35s\n",
      "1362:\tlearn: 0.9306814\ttotal: 2m 35s\tremaining: 4.23s\n",
      "1363:\tlearn: 0.9306496\ttotal: 2m 35s\tremaining: 4.12s\n",
      "1364:\tlearn: 0.9306297\ttotal: 2m 36s\tremaining: 4s\n",
      "1365:\tlearn: 0.9305929\ttotal: 2m 36s\tremaining: 3.89s\n",
      "1366:\tlearn: 0.9305603\ttotal: 2m 36s\tremaining: 3.77s\n",
      "1367:\tlearn: 0.9305430\ttotal: 2m 36s\tremaining: 3.66s\n",
      "1368:\tlearn: 0.9305299\ttotal: 2m 36s\tremaining: 3.54s\n",
      "1369:\tlearn: 0.9304814\ttotal: 2m 36s\tremaining: 3.43s\n",
      "1370:\tlearn: 0.9304574\ttotal: 2m 36s\tremaining: 3.31s\n",
      "1371:\tlearn: 0.9304306\ttotal: 2m 36s\tremaining: 3.2s\n",
      "1372:\tlearn: 0.9303964\ttotal: 2m 36s\tremaining: 3.09s\n",
      "1373:\tlearn: 0.9303927\ttotal: 2m 37s\tremaining: 2.97s\n",
      "1374:\tlearn: 0.9303608\ttotal: 2m 37s\tremaining: 2.86s\n",
      "1375:\tlearn: 0.9303528\ttotal: 2m 37s\tremaining: 2.74s\n",
      "1376:\tlearn: 0.9303486\ttotal: 2m 37s\tremaining: 2.63s\n",
      "1377:\tlearn: 0.9303288\ttotal: 2m 37s\tremaining: 2.51s\n",
      "1378:\tlearn: 0.9303017\ttotal: 2m 37s\tremaining: 2.4s\n",
      "1379:\tlearn: 0.9302859\ttotal: 2m 37s\tremaining: 2.28s\n",
      "1380:\tlearn: 0.9302554\ttotal: 2m 37s\tremaining: 2.17s\n",
      "1381:\tlearn: 0.9302396\ttotal: 2m 37s\tremaining: 2.06s\n",
      "1382:\tlearn: 0.9302173\ttotal: 2m 37s\tremaining: 1.94s\n",
      "1383:\tlearn: 0.9301694\ttotal: 2m 38s\tremaining: 1.83s\n",
      "1384:\tlearn: 0.9301370\ttotal: 2m 38s\tremaining: 1.71s\n",
      "1385:\tlearn: 0.9301357\ttotal: 2m 38s\tremaining: 1.6s\n",
      "1386:\tlearn: 0.9301122\ttotal: 2m 38s\tremaining: 1.48s\n",
      "1387:\tlearn: 0.9300779\ttotal: 2m 38s\tremaining: 1.37s\n",
      "1388:\tlearn: 0.9300423\ttotal: 2m 38s\tremaining: 1.25s\n",
      "1389:\tlearn: 0.9300118\ttotal: 2m 38s\tremaining: 1.14s\n",
      "1390:\tlearn: 0.9299843\ttotal: 2m 38s\tremaining: 1.03s\n",
      "1391:\tlearn: 0.9299804\ttotal: 2m 38s\tremaining: 913ms\n",
      "1392:\tlearn: 0.9299442\ttotal: 2m 39s\tremaining: 799ms\n",
      "1393:\tlearn: 0.9299151\ttotal: 2m 39s\tremaining: 685ms\n",
      "1394:\tlearn: 0.9298744\ttotal: 2m 39s\tremaining: 571ms\n",
      "1395:\tlearn: 0.9298444\ttotal: 2m 39s\tremaining: 457ms\n",
      "1396:\tlearn: 0.9298234\ttotal: 2m 39s\tremaining: 342ms\n",
      "1397:\tlearn: 0.9298149\ttotal: 2m 39s\tremaining: 228ms\n",
      "1398:\tlearn: 0.9297895\ttotal: 2m 39s\tremaining: 114ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1399:\tlearn: 0.9297602\ttotal: 2m 39s\tremaining: 0us\n",
      "0:\tlearn: 7.4160622\ttotal: 96.2ms\tremaining: 2m 14s\n",
      "1:\tlearn: 7.0565687\ttotal: 241ms\tremaining: 2m 48s\n",
      "2:\tlearn: 6.7135847\ttotal: 360ms\tremaining: 2m 47s\n",
      "3:\tlearn: 6.3877870\ttotal: 457ms\tremaining: 2m 39s\n",
      "4:\tlearn: 6.0789982\ttotal: 551ms\tremaining: 2m 33s\n",
      "5:\tlearn: 5.7856925\ttotal: 668ms\tremaining: 2m 35s\n",
      "6:\tlearn: 5.5077616\ttotal: 785ms\tremaining: 2m 36s\n",
      "7:\tlearn: 5.2451972\ttotal: 892ms\tremaining: 2m 35s\n",
      "8:\tlearn: 4.9949879\ttotal: 1.01s\tremaining: 2m 36s\n",
      "9:\tlearn: 4.7578493\ttotal: 1.09s\tremaining: 2m 31s\n",
      "10:\tlearn: 4.5340679\ttotal: 1.22s\tremaining: 2m 33s\n",
      "11:\tlearn: 4.3208096\ttotal: 1.28s\tremaining: 2m 28s\n",
      "12:\tlearn: 4.1190661\ttotal: 1.39s\tremaining: 2m 28s\n",
      "13:\tlearn: 3.9286973\ttotal: 1.53s\tremaining: 2m 31s\n",
      "14:\tlearn: 3.7472798\ttotal: 1.61s\tremaining: 2m 29s\n",
      "15:\tlearn: 3.5754979\ttotal: 1.7s\tremaining: 2m 27s\n",
      "16:\tlearn: 3.4143234\ttotal: 1.84s\tremaining: 2m 29s\n",
      "17:\tlearn: 3.2615820\ttotal: 1.94s\tremaining: 2m 29s\n",
      "18:\tlearn: 3.1167531\ttotal: 2.04s\tremaining: 2m 28s\n",
      "19:\tlearn: 2.9804108\ttotal: 2.17s\tremaining: 2m 29s\n",
      "20:\tlearn: 2.8509589\ttotal: 2.26s\tremaining: 2m 28s\n",
      "21:\tlearn: 2.7294461\ttotal: 2.4s\tremaining: 2m 30s\n",
      "22:\tlearn: 2.6142231\ttotal: 2.48s\tremaining: 2m 28s\n",
      "23:\tlearn: 2.5059180\ttotal: 2.58s\tremaining: 2m 27s\n",
      "24:\tlearn: 2.4033053\ttotal: 2.69s\tremaining: 2m 28s\n",
      "25:\tlearn: 2.3066802\ttotal: 2.83s\tremaining: 2m 29s\n",
      "26:\tlearn: 2.2162845\ttotal: 2.94s\tremaining: 2m 29s\n",
      "27:\tlearn: 2.1306643\ttotal: 3.03s\tremaining: 2m 28s\n",
      "28:\tlearn: 2.0505521\ttotal: 3.13s\tremaining: 2m 28s\n",
      "29:\tlearn: 1.9748355\ttotal: 3.26s\tremaining: 2m 28s\n",
      "30:\tlearn: 1.9037347\ttotal: 3.36s\tremaining: 2m 28s\n",
      "31:\tlearn: 1.8375952\ttotal: 3.5s\tremaining: 2m 29s\n",
      "32:\tlearn: 1.7756947\ttotal: 3.63s\tremaining: 2m 30s\n",
      "33:\tlearn: 1.7176328\ttotal: 3.74s\tremaining: 2m 30s\n",
      "34:\tlearn: 1.6633119\ttotal: 3.83s\tremaining: 2m 29s\n",
      "35:\tlearn: 1.6121378\ttotal: 3.93s\tremaining: 2m 28s\n",
      "36:\tlearn: 1.5645093\ttotal: 4.06s\tremaining: 2m 29s\n",
      "37:\tlearn: 1.5197395\ttotal: 4.17s\tremaining: 2m 29s\n",
      "38:\tlearn: 1.4780685\ttotal: 4.28s\tremaining: 2m 29s\n",
      "39:\tlearn: 1.4395646\ttotal: 4.41s\tremaining: 2m 29s\n",
      "40:\tlearn: 1.4040091\ttotal: 4.5s\tremaining: 2m 29s\n",
      "41:\tlearn: 1.3703095\ttotal: 4.63s\tremaining: 2m 29s\n",
      "42:\tlearn: 1.3399096\ttotal: 4.75s\tremaining: 2m 29s\n",
      "43:\tlearn: 1.3115397\ttotal: 4.87s\tremaining: 2m 30s\n",
      "44:\tlearn: 1.2849276\ttotal: 5s\tremaining: 2m 30s\n",
      "45:\tlearn: 1.2601503\ttotal: 5.15s\tremaining: 2m 31s\n",
      "46:\tlearn: 1.2377217\ttotal: 5.28s\tremaining: 2m 32s\n",
      "47:\tlearn: 1.2167585\ttotal: 5.39s\tremaining: 2m 31s\n",
      "48:\tlearn: 1.1975785\ttotal: 5.54s\tremaining: 2m 32s\n",
      "49:\tlearn: 1.1794988\ttotal: 5.67s\tremaining: 2m 32s\n",
      "50:\tlearn: 1.1636760\ttotal: 5.76s\tremaining: 2m 32s\n",
      "51:\tlearn: 1.1490961\ttotal: 5.88s\tremaining: 2m 32s\n",
      "52:\tlearn: 1.1351590\ttotal: 6s\tremaining: 2m 32s\n",
      "53:\tlearn: 1.1223999\ttotal: 6.13s\tremaining: 2m 32s\n",
      "54:\tlearn: 1.1107354\ttotal: 6.26s\tremaining: 2m 33s\n",
      "55:\tlearn: 1.1002004\ttotal: 6.38s\tremaining: 2m 33s\n",
      "56:\tlearn: 1.0903427\ttotal: 6.5s\tremaining: 2m 33s\n",
      "57:\tlearn: 1.0814792\ttotal: 6.62s\tremaining: 2m 33s\n",
      "58:\tlearn: 1.0732301\ttotal: 6.77s\tremaining: 2m 33s\n",
      "59:\tlearn: 1.0657705\ttotal: 6.92s\tremaining: 2m 34s\n",
      "60:\tlearn: 1.0587431\ttotal: 7.05s\tremaining: 2m 34s\n",
      "61:\tlearn: 1.0522164\ttotal: 7.17s\tremaining: 2m 34s\n",
      "62:\tlearn: 1.0463031\ttotal: 7.31s\tremaining: 2m 35s\n",
      "63:\tlearn: 1.0408449\ttotal: 7.43s\tremaining: 2m 35s\n",
      "64:\tlearn: 1.0359133\ttotal: 7.54s\tremaining: 2m 34s\n",
      "65:\tlearn: 1.0312494\ttotal: 7.66s\tremaining: 2m 34s\n",
      "66:\tlearn: 1.0269600\ttotal: 7.79s\tremaining: 2m 34s\n",
      "67:\tlearn: 1.0232300\ttotal: 7.9s\tremaining: 2m 34s\n",
      "68:\tlearn: 1.0198775\ttotal: 8.01s\tremaining: 2m 34s\n",
      "69:\tlearn: 1.0165272\ttotal: 8.13s\tremaining: 2m 34s\n",
      "70:\tlearn: 1.0133299\ttotal: 8.27s\tremaining: 2m 34s\n",
      "71:\tlearn: 1.0106227\ttotal: 8.38s\tremaining: 2m 34s\n",
      "72:\tlearn: 1.0081172\ttotal: 8.5s\tremaining: 2m 34s\n",
      "73:\tlearn: 1.0058875\ttotal: 8.63s\tremaining: 2m 34s\n",
      "74:\tlearn: 1.0037801\ttotal: 8.75s\tremaining: 2m 34s\n",
      "75:\tlearn: 1.0018171\ttotal: 8.85s\tremaining: 2m 34s\n",
      "76:\tlearn: 0.9999508\ttotal: 8.96s\tremaining: 2m 33s\n",
      "77:\tlearn: 0.9982058\ttotal: 9.08s\tremaining: 2m 33s\n",
      "78:\tlearn: 0.9966629\ttotal: 9.18s\tremaining: 2m 33s\n",
      "79:\tlearn: 0.9952421\ttotal: 9.29s\tremaining: 2m 33s\n",
      "80:\tlearn: 0.9939096\ttotal: 9.41s\tremaining: 2m 33s\n",
      "81:\tlearn: 0.9926335\ttotal: 9.56s\tremaining: 2m 33s\n",
      "82:\tlearn: 0.9914941\ttotal: 9.66s\tremaining: 2m 33s\n",
      "83:\tlearn: 0.9902803\ttotal: 9.78s\tremaining: 2m 33s\n",
      "84:\tlearn: 0.9891835\ttotal: 9.9s\tremaining: 2m 33s\n",
      "85:\tlearn: 0.9882793\ttotal: 10s\tremaining: 2m 32s\n",
      "86:\tlearn: 0.9874324\ttotal: 10.1s\tremaining: 2m 32s\n",
      "87:\tlearn: 0.9866911\ttotal: 10.2s\tremaining: 2m 32s\n",
      "88:\tlearn: 0.9859430\ttotal: 10.3s\tremaining: 2m 32s\n",
      "89:\tlearn: 0.9851546\ttotal: 10.5s\tremaining: 2m 32s\n",
      "90:\tlearn: 0.9844452\ttotal: 10.6s\tremaining: 2m 32s\n",
      "91:\tlearn: 0.9838387\ttotal: 10.7s\tremaining: 2m 32s\n",
      "92:\tlearn: 0.9832087\ttotal: 10.9s\tremaining: 2m 32s\n",
      "93:\tlearn: 0.9826140\ttotal: 11s\tremaining: 2m 32s\n",
      "94:\tlearn: 0.9819444\ttotal: 11.1s\tremaining: 2m 32s\n",
      "95:\tlearn: 0.9814016\ttotal: 11.2s\tremaining: 2m 32s\n",
      "96:\tlearn: 0.9809369\ttotal: 11.3s\tremaining: 2m 31s\n",
      "97:\tlearn: 0.9804350\ttotal: 11.4s\tremaining: 2m 31s\n",
      "98:\tlearn: 0.9800142\ttotal: 11.5s\tremaining: 2m 31s\n",
      "99:\tlearn: 0.9795855\ttotal: 11.7s\tremaining: 2m 31s\n",
      "100:\tlearn: 0.9792043\ttotal: 11.8s\tremaining: 2m 31s\n",
      "101:\tlearn: 0.9788089\ttotal: 11.9s\tremaining: 2m 31s\n",
      "102:\tlearn: 0.9784995\ttotal: 12s\tremaining: 2m 31s\n",
      "103:\tlearn: 0.9780989\ttotal: 12.1s\tremaining: 2m 30s\n",
      "104:\tlearn: 0.9777433\ttotal: 12.2s\tremaining: 2m 30s\n",
      "105:\tlearn: 0.9774836\ttotal: 12.3s\tremaining: 2m 30s\n",
      "106:\tlearn: 0.9770866\ttotal: 12.4s\tremaining: 2m 30s\n",
      "107:\tlearn: 0.9767872\ttotal: 12.6s\tremaining: 2m 30s\n",
      "108:\tlearn: 0.9765398\ttotal: 12.7s\tremaining: 2m 30s\n",
      "109:\tlearn: 0.9762724\ttotal: 12.8s\tremaining: 2m 30s\n",
      "110:\tlearn: 0.9759002\ttotal: 12.9s\tremaining: 2m 30s\n",
      "111:\tlearn: 0.9755884\ttotal: 13.1s\tremaining: 2m 30s\n",
      "112:\tlearn: 0.9753398\ttotal: 13.2s\tremaining: 2m 29s\n",
      "113:\tlearn: 0.9751262\ttotal: 13.3s\tremaining: 2m 29s\n",
      "114:\tlearn: 0.9748951\ttotal: 13.4s\tremaining: 2m 29s\n",
      "115:\tlearn: 0.9746617\ttotal: 13.5s\tremaining: 2m 29s\n",
      "116:\tlearn: 0.9744600\ttotal: 13.6s\tremaining: 2m 29s\n",
      "117:\tlearn: 0.9743075\ttotal: 13.7s\tremaining: 2m 29s\n",
      "118:\tlearn: 0.9741104\ttotal: 13.8s\tremaining: 2m 28s\n",
      "119:\tlearn: 0.9738917\ttotal: 13.9s\tremaining: 2m 28s\n",
      "120:\tlearn: 0.9736493\ttotal: 14.1s\tremaining: 2m 28s\n",
      "121:\tlearn: 0.9733934\ttotal: 14.2s\tremaining: 2m 28s\n",
      "122:\tlearn: 0.9731756\ttotal: 14.3s\tremaining: 2m 28s\n",
      "123:\tlearn: 0.9729175\ttotal: 14.4s\tremaining: 2m 28s\n",
      "124:\tlearn: 0.9727116\ttotal: 14.5s\tremaining: 2m 28s\n",
      "125:\tlearn: 0.9725219\ttotal: 14.6s\tremaining: 2m 27s\n",
      "126:\tlearn: 0.9723088\ttotal: 14.7s\tremaining: 2m 27s\n",
      "127:\tlearn: 0.9721740\ttotal: 14.9s\tremaining: 2m 27s\n",
      "128:\tlearn: 0.9720522\ttotal: 15s\tremaining: 2m 27s\n",
      "129:\tlearn: 0.9719593\ttotal: 15.1s\tremaining: 2m 27s\n",
      "130:\tlearn: 0.9718188\ttotal: 15.2s\tremaining: 2m 27s\n",
      "131:\tlearn: 0.9716140\ttotal: 15.3s\tremaining: 2m 27s\n",
      "132:\tlearn: 0.9714003\ttotal: 15.4s\tremaining: 2m 26s\n",
      "133:\tlearn: 0.9712311\ttotal: 15.5s\tremaining: 2m 26s\n",
      "134:\tlearn: 0.9710725\ttotal: 15.6s\tremaining: 2m 26s\n",
      "135:\tlearn: 0.9708769\ttotal: 15.8s\tremaining: 2m 26s\n",
      "136:\tlearn: 0.9707199\ttotal: 15.9s\tremaining: 2m 26s\n",
      "137:\tlearn: 0.9705722\ttotal: 16.1s\tremaining: 2m 27s\n",
      "138:\tlearn: 0.9704606\ttotal: 16.2s\tremaining: 2m 27s\n",
      "139:\tlearn: 0.9703253\ttotal: 16.4s\tremaining: 2m 27s\n",
      "140:\tlearn: 0.9701832\ttotal: 16.5s\tremaining: 2m 27s\n",
      "141:\tlearn: 0.9700393\ttotal: 16.6s\tremaining: 2m 27s\n",
      "142:\tlearn: 0.9699367\ttotal: 16.7s\tremaining: 2m 26s\n",
      "143:\tlearn: 0.9698284\ttotal: 16.8s\tremaining: 2m 26s\n",
      "144:\tlearn: 0.9697405\ttotal: 16.9s\tremaining: 2m 26s\n",
      "145:\tlearn: 0.9696096\ttotal: 17s\tremaining: 2m 26s\n",
      "146:\tlearn: 0.9694994\ttotal: 17.2s\tremaining: 2m 26s\n",
      "147:\tlearn: 0.9694244\ttotal: 17.3s\tremaining: 2m 26s\n",
      "148:\tlearn: 0.9693017\ttotal: 17.4s\tremaining: 2m 26s\n",
      "149:\tlearn: 0.9692024\ttotal: 17.5s\tremaining: 2m 25s\n",
      "150:\tlearn: 0.9690920\ttotal: 17.6s\tremaining: 2m 25s\n",
      "151:\tlearn: 0.9689998\ttotal: 17.7s\tremaining: 2m 25s\n",
      "152:\tlearn: 0.9688868\ttotal: 17.8s\tremaining: 2m 25s\n",
      "153:\tlearn: 0.9687786\ttotal: 17.9s\tremaining: 2m 25s\n",
      "154:\tlearn: 0.9686457\ttotal: 18.1s\tremaining: 2m 25s\n",
      "155:\tlearn: 0.9685316\ttotal: 18.2s\tremaining: 2m 24s\n",
      "156:\tlearn: 0.9684085\ttotal: 18.3s\tremaining: 2m 24s\n",
      "157:\tlearn: 0.9683161\ttotal: 18.4s\tremaining: 2m 24s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "158:\tlearn: 0.9682391\ttotal: 18.5s\tremaining: 2m 24s\n",
      "159:\tlearn: 0.9681510\ttotal: 18.6s\tremaining: 2m 24s\n",
      "160:\tlearn: 0.9680701\ttotal: 18.7s\tremaining: 2m 24s\n",
      "161:\tlearn: 0.9679724\ttotal: 18.8s\tremaining: 2m 23s\n",
      "162:\tlearn: 0.9679150\ttotal: 19s\tremaining: 2m 23s\n",
      "163:\tlearn: 0.9678156\ttotal: 19.1s\tremaining: 2m 23s\n",
      "164:\tlearn: 0.9676985\ttotal: 19.2s\tremaining: 2m 23s\n",
      "165:\tlearn: 0.9675946\ttotal: 19.3s\tremaining: 2m 23s\n",
      "166:\tlearn: 0.9675138\ttotal: 19.4s\tremaining: 2m 23s\n",
      "167:\tlearn: 0.9674149\ttotal: 19.5s\tremaining: 2m 23s\n",
      "168:\tlearn: 0.9673259\ttotal: 19.6s\tremaining: 2m 23s\n",
      "169:\tlearn: 0.9672392\ttotal: 19.8s\tremaining: 2m 22s\n",
      "170:\tlearn: 0.9671886\ttotal: 19.9s\tremaining: 2m 22s\n",
      "171:\tlearn: 0.9670715\ttotal: 20s\tremaining: 2m 22s\n",
      "172:\tlearn: 0.9670005\ttotal: 20.1s\tremaining: 2m 22s\n",
      "173:\tlearn: 0.9668929\ttotal: 20.2s\tremaining: 2m 22s\n",
      "174:\tlearn: 0.9667861\ttotal: 20.3s\tremaining: 2m 22s\n",
      "175:\tlearn: 0.9667048\ttotal: 20.4s\tremaining: 2m 22s\n",
      "176:\tlearn: 0.9665894\ttotal: 20.5s\tremaining: 2m 21s\n",
      "177:\tlearn: 0.9664968\ttotal: 20.6s\tremaining: 2m 21s\n",
      "178:\tlearn: 0.9664303\ttotal: 20.7s\tremaining: 2m 21s\n",
      "179:\tlearn: 0.9663467\ttotal: 20.8s\tremaining: 2m 21s\n",
      "180:\tlearn: 0.9662972\ttotal: 20.9s\tremaining: 2m 20s\n",
      "181:\tlearn: 0.9662397\ttotal: 21s\tremaining: 2m 20s\n",
      "182:\tlearn: 0.9661713\ttotal: 21.1s\tremaining: 2m 20s\n",
      "183:\tlearn: 0.9660899\ttotal: 21.2s\tremaining: 2m 20s\n",
      "184:\tlearn: 0.9660121\ttotal: 21.3s\tremaining: 2m 20s\n",
      "185:\tlearn: 0.9659106\ttotal: 21.4s\tremaining: 2m 19s\n",
      "186:\tlearn: 0.9658435\ttotal: 21.5s\tremaining: 2m 19s\n",
      "187:\tlearn: 0.9657701\ttotal: 21.7s\tremaining: 2m 19s\n",
      "188:\tlearn: 0.9656918\ttotal: 21.8s\tremaining: 2m 19s\n",
      "189:\tlearn: 0.9656216\ttotal: 21.9s\tremaining: 2m 19s\n",
      "190:\tlearn: 0.9655464\ttotal: 22s\tremaining: 2m 19s\n",
      "191:\tlearn: 0.9654507\ttotal: 22.1s\tremaining: 2m 19s\n",
      "192:\tlearn: 0.9653486\ttotal: 22.2s\tremaining: 2m 18s\n",
      "193:\tlearn: 0.9652851\ttotal: 22.3s\tremaining: 2m 18s\n",
      "194:\tlearn: 0.9651862\ttotal: 22.5s\tremaining: 2m 18s\n",
      "195:\tlearn: 0.9651203\ttotal: 22.6s\tremaining: 2m 18s\n",
      "196:\tlearn: 0.9650541\ttotal: 22.7s\tremaining: 2m 18s\n",
      "197:\tlearn: 0.9650083\ttotal: 22.8s\tremaining: 2m 18s\n",
      "198:\tlearn: 0.9649316\ttotal: 22.9s\tremaining: 2m 17s\n",
      "199:\tlearn: 0.9648694\ttotal: 23s\tremaining: 2m 17s\n",
      "200:\tlearn: 0.9647796\ttotal: 23.1s\tremaining: 2m 17s\n",
      "201:\tlearn: 0.9647028\ttotal: 23.2s\tremaining: 2m 17s\n",
      "202:\tlearn: 0.9646400\ttotal: 23.3s\tremaining: 2m 17s\n",
      "203:\tlearn: 0.9646038\ttotal: 23.4s\tremaining: 2m 17s\n",
      "204:\tlearn: 0.9645305\ttotal: 23.5s\tremaining: 2m 17s\n",
      "205:\tlearn: 0.9644516\ttotal: 23.6s\tremaining: 2m 17s\n",
      "206:\tlearn: 0.9644066\ttotal: 23.7s\tremaining: 2m 16s\n",
      "207:\tlearn: 0.9643767\ttotal: 23.8s\tremaining: 2m 16s\n",
      "208:\tlearn: 0.9643015\ttotal: 24s\tremaining: 2m 16s\n",
      "209:\tlearn: 0.9642265\ttotal: 24.1s\tremaining: 2m 16s\n",
      "210:\tlearn: 0.9641681\ttotal: 24.2s\tremaining: 2m 16s\n",
      "211:\tlearn: 0.9641016\ttotal: 24.3s\tremaining: 2m 15s\n",
      "212:\tlearn: 0.9640506\ttotal: 24.4s\tremaining: 2m 15s\n",
      "213:\tlearn: 0.9639668\ttotal: 24.5s\tremaining: 2m 15s\n",
      "214:\tlearn: 0.9638687\ttotal: 24.6s\tremaining: 2m 15s\n",
      "215:\tlearn: 0.9638026\ttotal: 24.7s\tremaining: 2m 15s\n",
      "216:\tlearn: 0.9637312\ttotal: 24.8s\tremaining: 2m 15s\n",
      "217:\tlearn: 0.9636648\ttotal: 25s\tremaining: 2m 15s\n",
      "218:\tlearn: 0.9636113\ttotal: 25s\tremaining: 2m 15s\n",
      "219:\tlearn: 0.9635529\ttotal: 25.1s\tremaining: 2m 14s\n",
      "220:\tlearn: 0.9635000\ttotal: 25.2s\tremaining: 2m 14s\n",
      "221:\tlearn: 0.9634461\ttotal: 25.3s\tremaining: 2m 14s\n",
      "222:\tlearn: 0.9633865\ttotal: 25.4s\tremaining: 2m 14s\n",
      "223:\tlearn: 0.9632994\ttotal: 25.6s\tremaining: 2m 14s\n",
      "224:\tlearn: 0.9632557\ttotal: 25.7s\tremaining: 2m 14s\n",
      "225:\tlearn: 0.9632011\ttotal: 25.8s\tremaining: 2m 13s\n",
      "226:\tlearn: 0.9631373\ttotal: 25.9s\tremaining: 2m 13s\n",
      "227:\tlearn: 0.9631014\ttotal: 26s\tremaining: 2m 13s\n",
      "228:\tlearn: 0.9630361\ttotal: 26.1s\tremaining: 2m 13s\n",
      "229:\tlearn: 0.9629893\ttotal: 26.2s\tremaining: 2m 13s\n",
      "230:\tlearn: 0.9629405\ttotal: 26.3s\tremaining: 2m 13s\n",
      "231:\tlearn: 0.9629043\ttotal: 26.4s\tremaining: 2m 12s\n",
      "232:\tlearn: 0.9628482\ttotal: 26.5s\tremaining: 2m 12s\n",
      "233:\tlearn: 0.9628086\ttotal: 26.6s\tremaining: 2m 12s\n",
      "234:\tlearn: 0.9627471\ttotal: 26.7s\tremaining: 2m 12s\n",
      "235:\tlearn: 0.9626615\ttotal: 26.8s\tremaining: 2m 12s\n",
      "236:\tlearn: 0.9625816\ttotal: 27s\tremaining: 2m 12s\n",
      "237:\tlearn: 0.9625239\ttotal: 27.1s\tremaining: 2m 12s\n",
      "238:\tlearn: 0.9624359\ttotal: 27.2s\tremaining: 2m 12s\n",
      "239:\tlearn: 0.9623824\ttotal: 27.3s\tremaining: 2m 12s\n",
      "240:\tlearn: 0.9622961\ttotal: 27.5s\tremaining: 2m 12s\n",
      "241:\tlearn: 0.9622145\ttotal: 27.6s\tremaining: 2m 11s\n",
      "242:\tlearn: 0.9621425\ttotal: 27.7s\tremaining: 2m 11s\n",
      "243:\tlearn: 0.9620842\ttotal: 27.8s\tremaining: 2m 11s\n",
      "244:\tlearn: 0.9620359\ttotal: 27.9s\tremaining: 2m 11s\n",
      "245:\tlearn: 0.9619813\ttotal: 28s\tremaining: 2m 11s\n",
      "246:\tlearn: 0.9619121\ttotal: 28.1s\tremaining: 2m 11s\n",
      "247:\tlearn: 0.9618444\ttotal: 28.2s\tremaining: 2m 11s\n",
      "248:\tlearn: 0.9618168\ttotal: 28.3s\tremaining: 2m 10s\n",
      "249:\tlearn: 0.9617631\ttotal: 28.4s\tremaining: 2m 10s\n",
      "250:\tlearn: 0.9616797\ttotal: 28.5s\tremaining: 2m 10s\n",
      "251:\tlearn: 0.9616402\ttotal: 28.6s\tremaining: 2m 10s\n",
      "252:\tlearn: 0.9615733\ttotal: 28.7s\tremaining: 2m 10s\n",
      "253:\tlearn: 0.9615140\ttotal: 28.8s\tremaining: 2m 10s\n",
      "254:\tlearn: 0.9614748\ttotal: 28.9s\tremaining: 2m 9s\n",
      "255:\tlearn: 0.9614151\ttotal: 29.1s\tremaining: 2m 9s\n",
      "256:\tlearn: 0.9613745\ttotal: 29.2s\tremaining: 2m 9s\n",
      "257:\tlearn: 0.9613309\ttotal: 29.3s\tremaining: 2m 9s\n",
      "258:\tlearn: 0.9612591\ttotal: 29.4s\tremaining: 2m 9s\n",
      "259:\tlearn: 0.9612097\ttotal: 29.5s\tremaining: 2m 9s\n",
      "260:\tlearn: 0.9611785\ttotal: 29.6s\tremaining: 2m 9s\n",
      "261:\tlearn: 0.9611216\ttotal: 29.7s\tremaining: 2m 8s\n",
      "262:\tlearn: 0.9610752\ttotal: 29.8s\tremaining: 2m 8s\n",
      "263:\tlearn: 0.9610148\ttotal: 29.9s\tremaining: 2m 8s\n",
      "264:\tlearn: 0.9609591\ttotal: 30s\tremaining: 2m 8s\n",
      "265:\tlearn: 0.9609007\ttotal: 30.1s\tremaining: 2m 8s\n",
      "266:\tlearn: 0.9608551\ttotal: 30.2s\tremaining: 2m 8s\n",
      "267:\tlearn: 0.9607932\ttotal: 30.3s\tremaining: 2m 7s\n",
      "268:\tlearn: 0.9607777\ttotal: 30.4s\tremaining: 2m 7s\n",
      "269:\tlearn: 0.9607224\ttotal: 30.5s\tremaining: 2m 7s\n",
      "270:\tlearn: 0.9606683\ttotal: 30.6s\tremaining: 2m 7s\n",
      "271:\tlearn: 0.9606059\ttotal: 30.7s\tremaining: 2m 7s\n",
      "272:\tlearn: 0.9605654\ttotal: 30.8s\tremaining: 2m 7s\n",
      "273:\tlearn: 0.9605631\ttotal: 30.9s\tremaining: 2m 6s\n",
      "274:\tlearn: 0.9605280\ttotal: 31s\tremaining: 2m 6s\n",
      "275:\tlearn: 0.9604604\ttotal: 31.1s\tremaining: 2m 6s\n",
      "276:\tlearn: 0.9604175\ttotal: 31.2s\tremaining: 2m 6s\n",
      "277:\tlearn: 0.9603500\ttotal: 31.3s\tremaining: 2m 6s\n",
      "278:\tlearn: 0.9603483\ttotal: 31.4s\tremaining: 2m 6s\n",
      "279:\tlearn: 0.9602976\ttotal: 31.5s\tremaining: 2m 6s\n",
      "280:\tlearn: 0.9602477\ttotal: 31.6s\tremaining: 2m 5s\n",
      "281:\tlearn: 0.9602136\ttotal: 31.7s\tremaining: 2m 5s\n",
      "282:\tlearn: 0.9601459\ttotal: 31.9s\tremaining: 2m 5s\n",
      "283:\tlearn: 0.9601031\ttotal: 32s\tremaining: 2m 5s\n",
      "284:\tlearn: 0.9600357\ttotal: 32.1s\tremaining: 2m 5s\n",
      "285:\tlearn: 0.9599892\ttotal: 32.2s\tremaining: 2m 5s\n",
      "286:\tlearn: 0.9599386\ttotal: 32.3s\tremaining: 2m 5s\n",
      "287:\tlearn: 0.9599358\ttotal: 32.4s\tremaining: 2m 4s\n",
      "288:\tlearn: 0.9598697\ttotal: 32.5s\tremaining: 2m 4s\n",
      "289:\tlearn: 0.9598187\ttotal: 32.6s\tremaining: 2m 4s\n",
      "290:\tlearn: 0.9598173\ttotal: 32.7s\tremaining: 2m 4s\n",
      "291:\tlearn: 0.9597459\ttotal: 32.8s\tremaining: 2m 4s\n",
      "292:\tlearn: 0.9597227\ttotal: 32.9s\tremaining: 2m 4s\n",
      "293:\tlearn: 0.9596514\ttotal: 33s\tremaining: 2m 4s\n",
      "294:\tlearn: 0.9596302\ttotal: 33.1s\tremaining: 2m 3s\n",
      "295:\tlearn: 0.9596268\ttotal: 33.2s\tremaining: 2m 3s\n",
      "296:\tlearn: 0.9595860\ttotal: 33.3s\tremaining: 2m 3s\n",
      "297:\tlearn: 0.9595397\ttotal: 33.4s\tremaining: 2m 3s\n",
      "298:\tlearn: 0.9595079\ttotal: 33.5s\tremaining: 2m 3s\n",
      "299:\tlearn: 0.9594308\ttotal: 33.6s\tremaining: 2m 3s\n",
      "300:\tlearn: 0.9593937\ttotal: 33.7s\tremaining: 2m 3s\n",
      "301:\tlearn: 0.9593515\ttotal: 33.8s\tremaining: 2m 2s\n",
      "302:\tlearn: 0.9593136\ttotal: 33.9s\tremaining: 2m 2s\n",
      "303:\tlearn: 0.9592741\ttotal: 34s\tremaining: 2m 2s\n",
      "304:\tlearn: 0.9592141\ttotal: 34.1s\tremaining: 2m 2s\n",
      "305:\tlearn: 0.9591741\ttotal: 34.2s\tremaining: 2m 2s\n",
      "306:\tlearn: 0.9591123\ttotal: 34.3s\tremaining: 2m 2s\n",
      "307:\tlearn: 0.9590557\ttotal: 34.4s\tremaining: 2m 1s\n",
      "308:\tlearn: 0.9590174\ttotal: 34.5s\tremaining: 2m 1s\n",
      "309:\tlearn: 0.9589808\ttotal: 34.6s\tremaining: 2m 1s\n",
      "310:\tlearn: 0.9589549\ttotal: 34.7s\tremaining: 2m 1s\n",
      "311:\tlearn: 0.9588713\ttotal: 34.8s\tremaining: 2m 1s\n",
      "312:\tlearn: 0.9588290\ttotal: 34.9s\tremaining: 2m 1s\n",
      "313:\tlearn: 0.9587829\ttotal: 35.1s\tremaining: 2m 1s\n",
      "314:\tlearn: 0.9587182\ttotal: 35.2s\tremaining: 2m 1s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "315:\tlearn: 0.9586673\ttotal: 35.3s\tremaining: 2m 1s\n",
      "316:\tlearn: 0.9585932\ttotal: 35.4s\tremaining: 2m 1s\n",
      "317:\tlearn: 0.9585336\ttotal: 35.5s\tremaining: 2m\n",
      "318:\tlearn: 0.9584406\ttotal: 35.7s\tremaining: 2m\n",
      "319:\tlearn: 0.9584025\ttotal: 35.8s\tremaining: 2m\n",
      "320:\tlearn: 0.9583302\ttotal: 35.9s\tremaining: 2m\n",
      "321:\tlearn: 0.9582860\ttotal: 36s\tremaining: 2m\n",
      "322:\tlearn: 0.9582284\ttotal: 36.1s\tremaining: 2m\n",
      "323:\tlearn: 0.9581666\ttotal: 36.2s\tremaining: 2m\n",
      "324:\tlearn: 0.9581128\ttotal: 36.3s\tremaining: 2m\n",
      "325:\tlearn: 0.9581115\ttotal: 36.5s\tremaining: 2m\n",
      "326:\tlearn: 0.9580288\ttotal: 36.6s\tremaining: 2m\n",
      "327:\tlearn: 0.9579289\ttotal: 37.2s\tremaining: 2m 1s\n",
      "328:\tlearn: 0.9578996\ttotal: 37.3s\tremaining: 2m 1s\n",
      "329:\tlearn: 0.9578621\ttotal: 37.4s\tremaining: 2m 1s\n",
      "330:\tlearn: 0.9578422\ttotal: 37.5s\tremaining: 2m 1s\n",
      "331:\tlearn: 0.9578086\ttotal: 37.6s\tremaining: 2m\n",
      "332:\tlearn: 0.9577704\ttotal: 37.7s\tremaining: 2m\n",
      "333:\tlearn: 0.9577210\ttotal: 37.8s\tremaining: 2m\n",
      "334:\tlearn: 0.9576794\ttotal: 37.9s\tremaining: 2m\n",
      "335:\tlearn: 0.9576473\ttotal: 38s\tremaining: 2m\n",
      "336:\tlearn: 0.9576002\ttotal: 38.1s\tremaining: 2m\n",
      "337:\tlearn: 0.9575566\ttotal: 38.2s\tremaining: 1m 59s\n",
      "338:\tlearn: 0.9575249\ttotal: 38.3s\tremaining: 1m 59s\n",
      "339:\tlearn: 0.9574764\ttotal: 38.4s\tremaining: 1m 59s\n",
      "340:\tlearn: 0.9574197\ttotal: 38.5s\tremaining: 1m 59s\n",
      "341:\tlearn: 0.9573689\ttotal: 38.6s\tremaining: 1m 59s\n",
      "342:\tlearn: 0.9573397\ttotal: 38.7s\tremaining: 1m 59s\n",
      "343:\tlearn: 0.9572892\ttotal: 38.8s\tremaining: 1m 59s\n",
      "344:\tlearn: 0.9572635\ttotal: 38.9s\tremaining: 1m 59s\n",
      "345:\tlearn: 0.9572458\ttotal: 39s\tremaining: 1m 58s\n",
      "346:\tlearn: 0.9572220\ttotal: 39.1s\tremaining: 1m 58s\n",
      "347:\tlearn: 0.9571757\ttotal: 39.2s\tremaining: 1m 58s\n",
      "348:\tlearn: 0.9571283\ttotal: 39.4s\tremaining: 1m 58s\n",
      "349:\tlearn: 0.9570897\ttotal: 39.5s\tremaining: 1m 58s\n",
      "350:\tlearn: 0.9570230\ttotal: 39.5s\tremaining: 1m 58s\n",
      "351:\tlearn: 0.9569789\ttotal: 39.6s\tremaining: 1m 58s\n",
      "352:\tlearn: 0.9569416\ttotal: 39.7s\tremaining: 1m 57s\n",
      "353:\tlearn: 0.9569107\ttotal: 39.8s\tremaining: 1m 57s\n",
      "354:\tlearn: 0.9569041\ttotal: 39.9s\tremaining: 1m 57s\n",
      "355:\tlearn: 0.9568345\ttotal: 40s\tremaining: 1m 57s\n",
      "356:\tlearn: 0.9567712\ttotal: 40.2s\tremaining: 1m 57s\n",
      "357:\tlearn: 0.9567371\ttotal: 40.2s\tremaining: 1m 57s\n",
      "358:\tlearn: 0.9567292\ttotal: 40.3s\tremaining: 1m 56s\n",
      "359:\tlearn: 0.9566760\ttotal: 40.4s\tremaining: 1m 56s\n",
      "360:\tlearn: 0.9566157\ttotal: 40.5s\tremaining: 1m 56s\n",
      "361:\tlearn: 0.9565752\ttotal: 40.6s\tremaining: 1m 56s\n",
      "362:\tlearn: 0.9565068\ttotal: 40.8s\tremaining: 1m 56s\n",
      "363:\tlearn: 0.9564702\ttotal: 40.9s\tremaining: 1m 56s\n",
      "364:\tlearn: 0.9564126\ttotal: 41s\tremaining: 1m 56s\n",
      "365:\tlearn: 0.9563381\ttotal: 41.1s\tremaining: 1m 56s\n",
      "366:\tlearn: 0.9563193\ttotal: 41.2s\tremaining: 1m 55s\n",
      "367:\tlearn: 0.9562513\ttotal: 41.3s\tremaining: 1m 55s\n",
      "368:\tlearn: 0.9562235\ttotal: 41.4s\tremaining: 1m 55s\n",
      "369:\tlearn: 0.9561722\ttotal: 41.5s\tremaining: 1m 55s\n",
      "370:\tlearn: 0.9561451\ttotal: 41.6s\tremaining: 1m 55s\n",
      "371:\tlearn: 0.9561425\ttotal: 41.7s\tremaining: 1m 55s\n",
      "372:\tlearn: 0.9560995\ttotal: 41.8s\tremaining: 1m 55s\n",
      "373:\tlearn: 0.9560500\ttotal: 41.9s\tremaining: 1m 54s\n",
      "374:\tlearn: 0.9560223\ttotal: 42s\tremaining: 1m 54s\n",
      "375:\tlearn: 0.9559756\ttotal: 42.1s\tremaining: 1m 54s\n",
      "376:\tlearn: 0.9559117\ttotal: 42.2s\tremaining: 1m 54s\n",
      "377:\tlearn: 0.9558802\ttotal: 42.3s\tremaining: 1m 54s\n",
      "378:\tlearn: 0.9558318\ttotal: 42.4s\tremaining: 1m 54s\n",
      "379:\tlearn: 0.9557635\ttotal: 42.5s\tremaining: 1m 54s\n",
      "380:\tlearn: 0.9557402\ttotal: 42.6s\tremaining: 1m 53s\n",
      "381:\tlearn: 0.9557005\ttotal: 42.7s\tremaining: 1m 53s\n",
      "382:\tlearn: 0.9556919\ttotal: 42.8s\tremaining: 1m 53s\n",
      "383:\tlearn: 0.9556629\ttotal: 42.9s\tremaining: 1m 53s\n",
      "384:\tlearn: 0.9556197\ttotal: 43s\tremaining: 1m 53s\n",
      "385:\tlearn: 0.9555931\ttotal: 43.1s\tremaining: 1m 53s\n",
      "386:\tlearn: 0.9555429\ttotal: 43.2s\tremaining: 1m 53s\n",
      "387:\tlearn: 0.9555037\ttotal: 43.3s\tremaining: 1m 53s\n",
      "388:\tlearn: 0.9554581\ttotal: 43.4s\tremaining: 1m 52s\n",
      "389:\tlearn: 0.9554193\ttotal: 43.5s\tremaining: 1m 52s\n",
      "390:\tlearn: 0.9553481\ttotal: 43.7s\tremaining: 1m 52s\n",
      "391:\tlearn: 0.9553468\ttotal: 43.7s\tremaining: 1m 52s\n",
      "392:\tlearn: 0.9553300\ttotal: 43.8s\tremaining: 1m 52s\n",
      "393:\tlearn: 0.9552978\ttotal: 43.9s\tremaining: 1m 52s\n",
      "394:\tlearn: 0.9552358\ttotal: 44s\tremaining: 1m 51s\n",
      "395:\tlearn: 0.9552102\ttotal: 44.1s\tremaining: 1m 51s\n",
      "396:\tlearn: 0.9551738\ttotal: 44.2s\tremaining: 1m 51s\n",
      "397:\tlearn: 0.9551309\ttotal: 44.3s\tremaining: 1m 51s\n",
      "398:\tlearn: 0.9550971\ttotal: 44.4s\tremaining: 1m 51s\n",
      "399:\tlearn: 0.9550518\ttotal: 44.6s\tremaining: 1m 51s\n",
      "400:\tlearn: 0.9550484\ttotal: 44.6s\tremaining: 1m 51s\n",
      "401:\tlearn: 0.9550210\ttotal: 44.8s\tremaining: 1m 51s\n",
      "402:\tlearn: 0.9549895\ttotal: 44.8s\tremaining: 1m 50s\n",
      "403:\tlearn: 0.9549326\ttotal: 45s\tremaining: 1m 50s\n",
      "404:\tlearn: 0.9548831\ttotal: 45s\tremaining: 1m 50s\n",
      "405:\tlearn: 0.9548457\ttotal: 45.1s\tremaining: 1m 50s\n",
      "406:\tlearn: 0.9548060\ttotal: 45.2s\tremaining: 1m 50s\n",
      "407:\tlearn: 0.9547610\ttotal: 45.4s\tremaining: 1m 50s\n",
      "408:\tlearn: 0.9547597\ttotal: 45.4s\tremaining: 1m 50s\n",
      "409:\tlearn: 0.9547434\ttotal: 45.5s\tremaining: 1m 49s\n",
      "410:\tlearn: 0.9546982\ttotal: 45.6s\tremaining: 1m 49s\n",
      "411:\tlearn: 0.9546692\ttotal: 45.7s\tremaining: 1m 49s\n",
      "412:\tlearn: 0.9546120\ttotal: 45.8s\tremaining: 1m 49s\n",
      "413:\tlearn: 0.9545715\ttotal: 45.9s\tremaining: 1m 49s\n",
      "414:\tlearn: 0.9545404\ttotal: 46s\tremaining: 1m 49s\n",
      "415:\tlearn: 0.9544921\ttotal: 46.2s\tremaining: 1m 49s\n",
      "416:\tlearn: 0.9544419\ttotal: 46.3s\tremaining: 1m 49s\n",
      "417:\tlearn: 0.9544026\ttotal: 46.4s\tremaining: 1m 48s\n",
      "418:\tlearn: 0.9543643\ttotal: 46.5s\tremaining: 1m 48s\n",
      "419:\tlearn: 0.9543631\ttotal: 46.6s\tremaining: 1m 48s\n",
      "420:\tlearn: 0.9543336\ttotal: 46.6s\tremaining: 1m 48s\n",
      "421:\tlearn: 0.9542668\ttotal: 46.8s\tremaining: 1m 48s\n",
      "422:\tlearn: 0.9542271\ttotal: 46.9s\tremaining: 1m 48s\n",
      "423:\tlearn: 0.9541858\ttotal: 47s\tremaining: 1m 48s\n",
      "424:\tlearn: 0.9541449\ttotal: 47.1s\tremaining: 1m 48s\n",
      "425:\tlearn: 0.9541036\ttotal: 47.2s\tremaining: 1m 47s\n",
      "426:\tlearn: 0.9540596\ttotal: 47.3s\tremaining: 1m 47s\n",
      "427:\tlearn: 0.9540062\ttotal: 47.4s\tremaining: 1m 47s\n",
      "428:\tlearn: 0.9539502\ttotal: 47.6s\tremaining: 1m 47s\n",
      "429:\tlearn: 0.9539259\ttotal: 47.6s\tremaining: 1m 47s\n",
      "430:\tlearn: 0.9538924\ttotal: 47.7s\tremaining: 1m 47s\n",
      "431:\tlearn: 0.9538429\ttotal: 47.9s\tremaining: 1m 47s\n",
      "432:\tlearn: 0.9537780\ttotal: 48s\tremaining: 1m 47s\n",
      "433:\tlearn: 0.9537565\ttotal: 48.1s\tremaining: 1m 47s\n",
      "434:\tlearn: 0.9536927\ttotal: 48.2s\tremaining: 1m 46s\n",
      "435:\tlearn: 0.9536671\ttotal: 48.3s\tremaining: 1m 46s\n",
      "436:\tlearn: 0.9536161\ttotal: 48.4s\tremaining: 1m 46s\n",
      "437:\tlearn: 0.9535713\ttotal: 48.5s\tremaining: 1m 46s\n",
      "438:\tlearn: 0.9535366\ttotal: 48.6s\tremaining: 1m 46s\n",
      "439:\tlearn: 0.9534868\ttotal: 48.7s\tremaining: 1m 46s\n",
      "440:\tlearn: 0.9534813\ttotal: 48.8s\tremaining: 1m 46s\n",
      "441:\tlearn: 0.9534421\ttotal: 48.9s\tremaining: 1m 45s\n",
      "442:\tlearn: 0.9534207\ttotal: 49s\tremaining: 1m 45s\n",
      "443:\tlearn: 0.9533638\ttotal: 49.1s\tremaining: 1m 45s\n",
      "444:\tlearn: 0.9533169\ttotal: 49.2s\tremaining: 1m 45s\n",
      "445:\tlearn: 0.9532972\ttotal: 49.3s\tremaining: 1m 45s\n",
      "446:\tlearn: 0.9532956\ttotal: 49.4s\tremaining: 1m 45s\n",
      "447:\tlearn: 0.9532711\ttotal: 49.5s\tremaining: 1m 45s\n",
      "448:\tlearn: 0.9532413\ttotal: 49.6s\tremaining: 1m 45s\n",
      "449:\tlearn: 0.9531961\ttotal: 49.7s\tremaining: 1m 44s\n",
      "450:\tlearn: 0.9531625\ttotal: 49.8s\tremaining: 1m 44s\n",
      "451:\tlearn: 0.9531197\ttotal: 49.9s\tremaining: 1m 44s\n",
      "452:\tlearn: 0.9530803\ttotal: 50s\tremaining: 1m 44s\n",
      "453:\tlearn: 0.9530521\ttotal: 50.1s\tremaining: 1m 44s\n",
      "454:\tlearn: 0.9529943\ttotal: 50.2s\tremaining: 1m 44s\n",
      "455:\tlearn: 0.9529518\ttotal: 50.3s\tremaining: 1m 44s\n",
      "456:\tlearn: 0.9529174\ttotal: 50.4s\tremaining: 1m 44s\n",
      "457:\tlearn: 0.9529165\ttotal: 50.5s\tremaining: 1m 43s\n",
      "458:\tlearn: 0.9528736\ttotal: 50.6s\tremaining: 1m 43s\n",
      "459:\tlearn: 0.9528070\ttotal: 50.7s\tremaining: 1m 43s\n",
      "460:\tlearn: 0.9527710\ttotal: 50.8s\tremaining: 1m 43s\n",
      "461:\tlearn: 0.9527350\ttotal: 50.9s\tremaining: 1m 43s\n",
      "462:\tlearn: 0.9526828\ttotal: 51s\tremaining: 1m 43s\n",
      "463:\tlearn: 0.9526560\ttotal: 51.1s\tremaining: 1m 43s\n",
      "464:\tlearn: 0.9526336\ttotal: 51.2s\tremaining: 1m 42s\n",
      "465:\tlearn: 0.9526129\ttotal: 51.3s\tremaining: 1m 42s\n",
      "466:\tlearn: 0.9525926\ttotal: 51.4s\tremaining: 1m 42s\n",
      "467:\tlearn: 0.9525405\ttotal: 51.5s\tremaining: 1m 42s\n",
      "468:\tlearn: 0.9524896\ttotal: 51.6s\tremaining: 1m 42s\n",
      "469:\tlearn: 0.9524332\ttotal: 51.7s\tremaining: 1m 42s\n",
      "470:\tlearn: 0.9523797\ttotal: 51.8s\tremaining: 1m 42s\n",
      "471:\tlearn: 0.9523328\ttotal: 51.9s\tremaining: 1m 42s\n",
      "472:\tlearn: 0.9523025\ttotal: 52s\tremaining: 1m 41s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "473:\tlearn: 0.9522862\ttotal: 52.1s\tremaining: 1m 41s\n",
      "474:\tlearn: 0.9522565\ttotal: 52.2s\tremaining: 1m 41s\n",
      "475:\tlearn: 0.9522221\ttotal: 52.3s\tremaining: 1m 41s\n",
      "476:\tlearn: 0.9521747\ttotal: 52.4s\tremaining: 1m 41s\n",
      "477:\tlearn: 0.9521432\ttotal: 52.5s\tremaining: 1m 41s\n",
      "478:\tlearn: 0.9521413\ttotal: 52.6s\tremaining: 1m 41s\n",
      "479:\tlearn: 0.9521098\ttotal: 52.7s\tremaining: 1m 40s\n",
      "480:\tlearn: 0.9520613\ttotal: 52.8s\tremaining: 1m 40s\n",
      "481:\tlearn: 0.9520012\ttotal: 52.9s\tremaining: 1m 40s\n",
      "482:\tlearn: 0.9519772\ttotal: 53s\tremaining: 1m 40s\n",
      "483:\tlearn: 0.9519255\ttotal: 53.1s\tremaining: 1m 40s\n",
      "484:\tlearn: 0.9519240\ttotal: 53.2s\tremaining: 1m 40s\n",
      "485:\tlearn: 0.9518794\ttotal: 53.3s\tremaining: 1m 40s\n",
      "486:\tlearn: 0.9518586\ttotal: 53.4s\tremaining: 1m 40s\n",
      "487:\tlearn: 0.9518180\ttotal: 53.5s\tremaining: 1m 40s\n",
      "488:\tlearn: 0.9518171\ttotal: 53.6s\tremaining: 1m 39s\n",
      "489:\tlearn: 0.9517735\ttotal: 53.7s\tremaining: 1m 39s\n",
      "490:\tlearn: 0.9517397\ttotal: 53.8s\tremaining: 1m 39s\n",
      "491:\tlearn: 0.9517078\ttotal: 53.9s\tremaining: 1m 39s\n",
      "492:\tlearn: 0.9516878\ttotal: 54s\tremaining: 1m 39s\n",
      "493:\tlearn: 0.9516404\ttotal: 54.1s\tremaining: 1m 39s\n",
      "494:\tlearn: 0.9516174\ttotal: 54.2s\tremaining: 1m 39s\n",
      "495:\tlearn: 0.9515839\ttotal: 54.3s\tremaining: 1m 38s\n",
      "496:\tlearn: 0.9515520\ttotal: 54.4s\tremaining: 1m 38s\n",
      "497:\tlearn: 0.9515181\ttotal: 54.5s\tremaining: 1m 38s\n",
      "498:\tlearn: 0.9514938\ttotal: 54.6s\tremaining: 1m 38s\n",
      "499:\tlearn: 0.9514557\ttotal: 54.7s\tremaining: 1m 38s\n",
      "500:\tlearn: 0.9514345\ttotal: 54.8s\tremaining: 1m 38s\n",
      "501:\tlearn: 0.9513868\ttotal: 54.9s\tremaining: 1m 38s\n",
      "502:\tlearn: 0.9513364\ttotal: 55s\tremaining: 1m 38s\n",
      "503:\tlearn: 0.9513123\ttotal: 55.1s\tremaining: 1m 38s\n",
      "504:\tlearn: 0.9512746\ttotal: 55.2s\tremaining: 1m 37s\n",
      "505:\tlearn: 0.9512502\ttotal: 55.3s\tremaining: 1m 37s\n",
      "506:\tlearn: 0.9512249\ttotal: 55.4s\tremaining: 1m 37s\n",
      "507:\tlearn: 0.9511972\ttotal: 55.5s\tremaining: 1m 37s\n",
      "508:\tlearn: 0.9511680\ttotal: 55.6s\tremaining: 1m 37s\n",
      "509:\tlearn: 0.9511313\ttotal: 55.7s\tremaining: 1m 37s\n",
      "510:\tlearn: 0.9511226\ttotal: 55.8s\tremaining: 1m 37s\n",
      "511:\tlearn: 0.9510950\ttotal: 55.9s\tremaining: 1m 36s\n",
      "512:\tlearn: 0.9510488\ttotal: 56s\tremaining: 1m 36s\n",
      "513:\tlearn: 0.9509923\ttotal: 56.1s\tremaining: 1m 36s\n",
      "514:\tlearn: 0.9509551\ttotal: 56.2s\tremaining: 1m 36s\n",
      "515:\tlearn: 0.9509539\ttotal: 56.3s\tremaining: 1m 36s\n",
      "516:\tlearn: 0.9509510\ttotal: 56.4s\tremaining: 1m 36s\n",
      "517:\tlearn: 0.9509212\ttotal: 56.4s\tremaining: 1m 36s\n",
      "518:\tlearn: 0.9508954\ttotal: 56.5s\tremaining: 1m 35s\n",
      "519:\tlearn: 0.9508684\ttotal: 56.6s\tremaining: 1m 35s\n",
      "520:\tlearn: 0.9508497\ttotal: 56.7s\tremaining: 1m 35s\n",
      "521:\tlearn: 0.9508473\ttotal: 56.8s\tremaining: 1m 35s\n",
      "522:\tlearn: 0.9508467\ttotal: 56.9s\tremaining: 1m 35s\n",
      "523:\tlearn: 0.9508044\ttotal: 57s\tremaining: 1m 35s\n",
      "524:\tlearn: 0.9507744\ttotal: 57.1s\tremaining: 1m 35s\n",
      "525:\tlearn: 0.9507727\ttotal: 57.2s\tremaining: 1m 35s\n",
      "526:\tlearn: 0.9507290\ttotal: 57.3s\tremaining: 1m 34s\n",
      "527:\tlearn: 0.9506892\ttotal: 57.4s\tremaining: 1m 34s\n",
      "528:\tlearn: 0.9506880\ttotal: 57.5s\tremaining: 1m 34s\n",
      "529:\tlearn: 0.9506406\ttotal: 57.6s\tremaining: 1m 34s\n",
      "530:\tlearn: 0.9506120\ttotal: 57.7s\tremaining: 1m 34s\n",
      "531:\tlearn: 0.9505860\ttotal: 57.8s\tremaining: 1m 34s\n",
      "532:\tlearn: 0.9505618\ttotal: 57.9s\tremaining: 1m 34s\n",
      "533:\tlearn: 0.9505156\ttotal: 58s\tremaining: 1m 34s\n",
      "534:\tlearn: 0.9504728\ttotal: 58.1s\tremaining: 1m 33s\n",
      "535:\tlearn: 0.9504355\ttotal: 58.2s\tremaining: 1m 33s\n",
      "536:\tlearn: 0.9504057\ttotal: 58.3s\tremaining: 1m 33s\n",
      "537:\tlearn: 0.9503888\ttotal: 58.4s\tremaining: 1m 33s\n",
      "538:\tlearn: 0.9503874\ttotal: 58.5s\tremaining: 1m 33s\n",
      "539:\tlearn: 0.9503155\ttotal: 58.6s\tremaining: 1m 33s\n",
      "540:\tlearn: 0.9502866\ttotal: 58.7s\tremaining: 1m 33s\n",
      "541:\tlearn: 0.9502478\ttotal: 58.8s\tremaining: 1m 33s\n",
      "542:\tlearn: 0.9501821\ttotal: 58.9s\tremaining: 1m 32s\n",
      "543:\tlearn: 0.9501317\ttotal: 59s\tremaining: 1m 32s\n",
      "544:\tlearn: 0.9500921\ttotal: 59.1s\tremaining: 1m 32s\n",
      "545:\tlearn: 0.9500871\ttotal: 59.2s\tremaining: 1m 32s\n",
      "546:\tlearn: 0.9500492\ttotal: 59.3s\tremaining: 1m 32s\n",
      "547:\tlearn: 0.9500179\ttotal: 59.4s\tremaining: 1m 32s\n",
      "548:\tlearn: 0.9499866\ttotal: 59.5s\tremaining: 1m 32s\n",
      "549:\tlearn: 0.9499398\ttotal: 59.6s\tremaining: 1m 32s\n",
      "550:\tlearn: 0.9499127\ttotal: 59.7s\tremaining: 1m 32s\n",
      "551:\tlearn: 0.9498733\ttotal: 59.8s\tremaining: 1m 31s\n",
      "552:\tlearn: 0.9498347\ttotal: 59.9s\tremaining: 1m 31s\n",
      "553:\tlearn: 0.9497566\ttotal: 1m\tremaining: 1m 31s\n",
      "554:\tlearn: 0.9497121\ttotal: 1m\tremaining: 1m 31s\n",
      "555:\tlearn: 0.9496850\ttotal: 1m\tremaining: 1m 31s\n",
      "556:\tlearn: 0.9496627\ttotal: 1m\tremaining: 1m 31s\n",
      "557:\tlearn: 0.9496372\ttotal: 1m\tremaining: 1m 31s\n",
      "558:\tlearn: 0.9496121\ttotal: 1m\tremaining: 1m 31s\n",
      "559:\tlearn: 0.9495865\ttotal: 1m\tremaining: 1m 30s\n",
      "560:\tlearn: 0.9495399\ttotal: 1m\tremaining: 1m 30s\n",
      "561:\tlearn: 0.9495360\ttotal: 1m\tremaining: 1m 30s\n",
      "562:\tlearn: 0.9494999\ttotal: 1m\tremaining: 1m 30s\n",
      "563:\tlearn: 0.9494811\ttotal: 1m 1s\tremaining: 1m 30s\n",
      "564:\tlearn: 0.9494352\ttotal: 1m 1s\tremaining: 1m 30s\n",
      "565:\tlearn: 0.9493763\ttotal: 1m 1s\tremaining: 1m 30s\n",
      "566:\tlearn: 0.9493229\ttotal: 1m 1s\tremaining: 1m 30s\n",
      "567:\tlearn: 0.9492652\ttotal: 1m 1s\tremaining: 1m 30s\n",
      "568:\tlearn: 0.9492089\ttotal: 1m 1s\tremaining: 1m 30s\n",
      "569:\tlearn: 0.9491667\ttotal: 1m 1s\tremaining: 1m 29s\n",
      "570:\tlearn: 0.9491269\ttotal: 1m 1s\tremaining: 1m 29s\n",
      "571:\tlearn: 0.9490793\ttotal: 1m 1s\tremaining: 1m 29s\n",
      "572:\tlearn: 0.9490544\ttotal: 1m 2s\tremaining: 1m 29s\n",
      "573:\tlearn: 0.9489963\ttotal: 1m 2s\tremaining: 1m 29s\n",
      "574:\tlearn: 0.9489618\ttotal: 1m 2s\tremaining: 1m 29s\n",
      "575:\tlearn: 0.9489455\ttotal: 1m 2s\tremaining: 1m 29s\n",
      "576:\tlearn: 0.9489146\ttotal: 1m 2s\tremaining: 1m 29s\n",
      "577:\tlearn: 0.9488641\ttotal: 1m 2s\tremaining: 1m 29s\n",
      "578:\tlearn: 0.9488185\ttotal: 1m 2s\tremaining: 1m 28s\n",
      "579:\tlearn: 0.9488111\ttotal: 1m 2s\tremaining: 1m 28s\n",
      "580:\tlearn: 0.9487845\ttotal: 1m 2s\tremaining: 1m 28s\n",
      "581:\tlearn: 0.9487514\ttotal: 1m 3s\tremaining: 1m 28s\n",
      "582:\tlearn: 0.9486964\ttotal: 1m 3s\tremaining: 1m 28s\n",
      "583:\tlearn: 0.9486699\ttotal: 1m 3s\tremaining: 1m 28s\n",
      "584:\tlearn: 0.9486427\ttotal: 1m 3s\tremaining: 1m 28s\n",
      "585:\tlearn: 0.9485927\ttotal: 1m 3s\tremaining: 1m 28s\n",
      "586:\tlearn: 0.9485735\ttotal: 1m 3s\tremaining: 1m 28s\n",
      "587:\tlearn: 0.9485547\ttotal: 1m 3s\tremaining: 1m 27s\n",
      "588:\tlearn: 0.9485124\ttotal: 1m 3s\tremaining: 1m 27s\n",
      "589:\tlearn: 0.9484708\ttotal: 1m 3s\tremaining: 1m 27s\n",
      "590:\tlearn: 0.9484435\ttotal: 1m 3s\tremaining: 1m 27s\n",
      "591:\tlearn: 0.9483787\ttotal: 1m 4s\tremaining: 1m 27s\n",
      "592:\tlearn: 0.9483532\ttotal: 1m 4s\tremaining: 1m 27s\n",
      "593:\tlearn: 0.9483377\ttotal: 1m 4s\tremaining: 1m 27s\n",
      "594:\tlearn: 0.9482986\ttotal: 1m 4s\tremaining: 1m 27s\n",
      "595:\tlearn: 0.9482684\ttotal: 1m 4s\tremaining: 1m 27s\n",
      "596:\tlearn: 0.9482447\ttotal: 1m 4s\tremaining: 1m 26s\n",
      "597:\tlearn: 0.9482217\ttotal: 1m 4s\tremaining: 1m 26s\n",
      "598:\tlearn: 0.9481832\ttotal: 1m 4s\tremaining: 1m 26s\n",
      "599:\tlearn: 0.9481262\ttotal: 1m 4s\tremaining: 1m 26s\n",
      "600:\tlearn: 0.9480877\ttotal: 1m 5s\tremaining: 1m 26s\n",
      "601:\tlearn: 0.9480662\ttotal: 1m 5s\tremaining: 1m 26s\n",
      "602:\tlearn: 0.9480389\ttotal: 1m 5s\tremaining: 1m 26s\n",
      "603:\tlearn: 0.9479886\ttotal: 1m 5s\tremaining: 1m 26s\n",
      "604:\tlearn: 0.9479468\ttotal: 1m 5s\tremaining: 1m 25s\n",
      "605:\tlearn: 0.9479212\ttotal: 1m 5s\tremaining: 1m 25s\n",
      "606:\tlearn: 0.9478942\ttotal: 1m 5s\tremaining: 1m 25s\n",
      "607:\tlearn: 0.9478402\ttotal: 1m 5s\tremaining: 1m 25s\n",
      "608:\tlearn: 0.9478119\ttotal: 1m 5s\tremaining: 1m 25s\n",
      "609:\tlearn: 0.9477982\ttotal: 1m 5s\tremaining: 1m 25s\n",
      "610:\tlearn: 0.9477698\ttotal: 1m 6s\tremaining: 1m 25s\n",
      "611:\tlearn: 0.9477209\ttotal: 1m 6s\tremaining: 1m 25s\n",
      "612:\tlearn: 0.9476818\ttotal: 1m 6s\tremaining: 1m 25s\n",
      "613:\tlearn: 0.9476370\ttotal: 1m 6s\tremaining: 1m 24s\n",
      "614:\tlearn: 0.9476252\ttotal: 1m 6s\tremaining: 1m 24s\n",
      "615:\tlearn: 0.9475959\ttotal: 1m 6s\tremaining: 1m 24s\n",
      "616:\tlearn: 0.9475548\ttotal: 1m 6s\tremaining: 1m 24s\n",
      "617:\tlearn: 0.9475235\ttotal: 1m 6s\tremaining: 1m 24s\n",
      "618:\tlearn: 0.9474823\ttotal: 1m 6s\tremaining: 1m 24s\n",
      "619:\tlearn: 0.9474382\ttotal: 1m 6s\tremaining: 1m 24s\n",
      "620:\tlearn: 0.9474031\ttotal: 1m 7s\tremaining: 1m 24s\n",
      "621:\tlearn: 0.9473495\ttotal: 1m 7s\tremaining: 1m 24s\n",
      "622:\tlearn: 0.9473061\ttotal: 1m 7s\tremaining: 1m 23s\n",
      "623:\tlearn: 0.9472758\ttotal: 1m 7s\tremaining: 1m 23s\n",
      "624:\tlearn: 0.9472530\ttotal: 1m 7s\tremaining: 1m 23s\n",
      "625:\tlearn: 0.9472204\ttotal: 1m 7s\tremaining: 1m 23s\n",
      "626:\tlearn: 0.9471749\ttotal: 1m 7s\tremaining: 1m 23s\n",
      "627:\tlearn: 0.9471542\ttotal: 1m 7s\tremaining: 1m 23s\n",
      "628:\tlearn: 0.9471179\ttotal: 1m 8s\tremaining: 1m 23s\n",
      "629:\tlearn: 0.9470744\ttotal: 1m 8s\tremaining: 1m 23s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "630:\tlearn: 0.9470360\ttotal: 1m 8s\tremaining: 1m 23s\n",
      "631:\tlearn: 0.9470206\ttotal: 1m 8s\tremaining: 1m 23s\n",
      "632:\tlearn: 0.9470072\ttotal: 1m 8s\tremaining: 1m 23s\n",
      "633:\tlearn: 0.9470053\ttotal: 1m 8s\tremaining: 1m 23s\n",
      "634:\tlearn: 0.9469418\ttotal: 1m 9s\tremaining: 1m 24s\n",
      "635:\tlearn: 0.9469005\ttotal: 1m 10s\tremaining: 1m 24s\n",
      "636:\tlearn: 0.9468689\ttotal: 1m 10s\tremaining: 1m 24s\n",
      "637:\tlearn: 0.9468204\ttotal: 1m 10s\tremaining: 1m 24s\n",
      "638:\tlearn: 0.9468180\ttotal: 1m 11s\tremaining: 1m 24s\n",
      "639:\tlearn: 0.9467833\ttotal: 1m 12s\tremaining: 1m 25s\n",
      "640:\tlearn: 0.9467563\ttotal: 1m 12s\tremaining: 1m 26s\n",
      "641:\tlearn: 0.9467373\ttotal: 1m 12s\tremaining: 1m 26s\n",
      "642:\tlearn: 0.9466938\ttotal: 1m 13s\tremaining: 1m 26s\n",
      "643:\tlearn: 0.9466646\ttotal: 1m 13s\tremaining: 1m 26s\n",
      "644:\tlearn: 0.9466395\ttotal: 1m 14s\tremaining: 1m 26s\n",
      "645:\tlearn: 0.9466183\ttotal: 1m 14s\tremaining: 1m 27s\n",
      "646:\tlearn: 0.9465964\ttotal: 1m 14s\tremaining: 1m 27s\n",
      "647:\tlearn: 0.9465544\ttotal: 1m 14s\tremaining: 1m 27s\n",
      "648:\tlearn: 0.9465179\ttotal: 1m 15s\tremaining: 1m 27s\n",
      "649:\tlearn: 0.9465120\ttotal: 1m 15s\tremaining: 1m 27s\n",
      "650:\tlearn: 0.9464778\ttotal: 1m 15s\tremaining: 1m 27s\n",
      "651:\tlearn: 0.9464450\ttotal: 1m 15s\tremaining: 1m 27s\n",
      "652:\tlearn: 0.9464135\ttotal: 1m 15s\tremaining: 1m 26s\n",
      "653:\tlearn: 0.9463650\ttotal: 1m 16s\tremaining: 1m 26s\n",
      "654:\tlearn: 0.9463403\ttotal: 1m 16s\tremaining: 1m 26s\n",
      "655:\tlearn: 0.9463064\ttotal: 1m 16s\tremaining: 1m 26s\n",
      "656:\tlearn: 0.9462830\ttotal: 1m 16s\tremaining: 1m 26s\n",
      "657:\tlearn: 0.9462796\ttotal: 1m 16s\tremaining: 1m 26s\n",
      "658:\tlearn: 0.9462731\ttotal: 1m 17s\tremaining: 1m 26s\n",
      "659:\tlearn: 0.9462277\ttotal: 1m 17s\tremaining: 1m 26s\n",
      "660:\tlearn: 0.9462001\ttotal: 1m 17s\tremaining: 1m 26s\n",
      "661:\tlearn: 0.9461687\ttotal: 1m 17s\tremaining: 1m 26s\n",
      "662:\tlearn: 0.9461382\ttotal: 1m 18s\tremaining: 1m 27s\n",
      "663:\tlearn: 0.9461097\ttotal: 1m 18s\tremaining: 1m 27s\n",
      "664:\tlearn: 0.9460907\ttotal: 1m 18s\tremaining: 1m 26s\n",
      "665:\tlearn: 0.9460682\ttotal: 1m 18s\tremaining: 1m 26s\n",
      "666:\tlearn: 0.9460333\ttotal: 1m 19s\tremaining: 1m 26s\n",
      "667:\tlearn: 0.9460305\ttotal: 1m 19s\tremaining: 1m 26s\n",
      "668:\tlearn: 0.9459912\ttotal: 1m 19s\tremaining: 1m 26s\n",
      "669:\tlearn: 0.9459556\ttotal: 1m 19s\tremaining: 1m 26s\n",
      "670:\tlearn: 0.9459024\ttotal: 1m 19s\tremaining: 1m 26s\n",
      "671:\tlearn: 0.9458644\ttotal: 1m 20s\tremaining: 1m 26s\n",
      "672:\tlearn: 0.9458367\ttotal: 1m 20s\tremaining: 1m 27s\n",
      "673:\tlearn: 0.9458354\ttotal: 1m 20s\tremaining: 1m 27s\n",
      "674:\tlearn: 0.9458194\ttotal: 1m 20s\tremaining: 1m 27s\n",
      "675:\tlearn: 0.9457769\ttotal: 1m 21s\tremaining: 1m 26s\n",
      "676:\tlearn: 0.9457159\ttotal: 1m 21s\tremaining: 1m 26s\n",
      "677:\tlearn: 0.9456963\ttotal: 1m 21s\tremaining: 1m 26s\n",
      "678:\tlearn: 0.9456629\ttotal: 1m 21s\tremaining: 1m 27s\n",
      "679:\tlearn: 0.9456446\ttotal: 1m 22s\tremaining: 1m 27s\n",
      "680:\tlearn: 0.9456386\ttotal: 1m 22s\tremaining: 1m 27s\n",
      "681:\tlearn: 0.9456110\ttotal: 1m 22s\tremaining: 1m 27s\n",
      "682:\tlearn: 0.9455511\ttotal: 1m 22s\tremaining: 1m 27s\n",
      "683:\tlearn: 0.9455140\ttotal: 1m 22s\tremaining: 1m 26s\n",
      "684:\tlearn: 0.9454868\ttotal: 1m 23s\tremaining: 1m 26s\n",
      "685:\tlearn: 0.9454496\ttotal: 1m 23s\tremaining: 1m 26s\n",
      "686:\tlearn: 0.9454163\ttotal: 1m 23s\tremaining: 1m 26s\n",
      "687:\tlearn: 0.9453934\ttotal: 1m 23s\tremaining: 1m 26s\n",
      "688:\tlearn: 0.9453924\ttotal: 1m 24s\tremaining: 1m 26s\n",
      "689:\tlearn: 0.9453674\ttotal: 1m 24s\tremaining: 1m 26s\n",
      "690:\tlearn: 0.9453280\ttotal: 1m 24s\tremaining: 1m 26s\n",
      "691:\tlearn: 0.9452987\ttotal: 1m 24s\tremaining: 1m 27s\n",
      "692:\tlearn: 0.9452464\ttotal: 1m 25s\tremaining: 1m 27s\n",
      "693:\tlearn: 0.9451903\ttotal: 1m 25s\tremaining: 1m 27s\n",
      "694:\tlearn: 0.9451719\ttotal: 1m 25s\tremaining: 1m 27s\n",
      "695:\tlearn: 0.9451522\ttotal: 1m 26s\tremaining: 1m 27s\n",
      "696:\tlearn: 0.9451188\ttotal: 1m 27s\tremaining: 1m 28s\n",
      "697:\tlearn: 0.9450857\ttotal: 1m 27s\tremaining: 1m 28s\n",
      "698:\tlearn: 0.9450851\ttotal: 1m 28s\tremaining: 1m 28s\n",
      "699:\tlearn: 0.9450164\ttotal: 1m 29s\tremaining: 1m 29s\n",
      "700:\tlearn: 0.9449860\ttotal: 1m 29s\tremaining: 1m 29s\n",
      "701:\tlearn: 0.9449701\ttotal: 1m 30s\tremaining: 1m 29s\n",
      "702:\tlearn: 0.9449425\ttotal: 1m 30s\tremaining: 1m 29s\n",
      "703:\tlearn: 0.9449216\ttotal: 1m 30s\tremaining: 1m 29s\n",
      "704:\tlearn: 0.9448760\ttotal: 1m 30s\tremaining: 1m 29s\n",
      "705:\tlearn: 0.9448514\ttotal: 1m 31s\tremaining: 1m 29s\n",
      "706:\tlearn: 0.9448180\ttotal: 1m 31s\tremaining: 1m 29s\n",
      "707:\tlearn: 0.9447724\ttotal: 1m 31s\tremaining: 1m 29s\n",
      "708:\tlearn: 0.9447720\ttotal: 1m 31s\tremaining: 1m 29s\n",
      "709:\tlearn: 0.9447561\ttotal: 1m 32s\tremaining: 1m 29s\n",
      "710:\tlearn: 0.9447200\ttotal: 1m 32s\tremaining: 1m 29s\n",
      "711:\tlearn: 0.9446996\ttotal: 1m 32s\tremaining: 1m 29s\n",
      "712:\tlearn: 0.9446688\ttotal: 1m 32s\tremaining: 1m 29s\n",
      "713:\tlearn: 0.9446531\ttotal: 1m 32s\tremaining: 1m 29s\n",
      "714:\tlearn: 0.9446118\ttotal: 1m 32s\tremaining: 1m 29s\n",
      "715:\tlearn: 0.9445703\ttotal: 1m 33s\tremaining: 1m 29s\n",
      "716:\tlearn: 0.9445135\ttotal: 1m 33s\tremaining: 1m 29s\n",
      "717:\tlearn: 0.9444893\ttotal: 1m 33s\tremaining: 1m 28s\n",
      "718:\tlearn: 0.9444549\ttotal: 1m 33s\tremaining: 1m 28s\n",
      "719:\tlearn: 0.9444323\ttotal: 1m 34s\tremaining: 1m 29s\n",
      "720:\tlearn: 0.9444289\ttotal: 1m 34s\tremaining: 1m 29s\n",
      "721:\tlearn: 0.9443813\ttotal: 1m 34s\tremaining: 1m 28s\n",
      "722:\tlearn: 0.9443436\ttotal: 1m 34s\tremaining: 1m 28s\n",
      "723:\tlearn: 0.9443173\ttotal: 1m 34s\tremaining: 1m 28s\n",
      "724:\tlearn: 0.9442608\ttotal: 1m 34s\tremaining: 1m 28s\n",
      "725:\tlearn: 0.9442363\ttotal: 1m 34s\tremaining: 1m 28s\n",
      "726:\tlearn: 0.9442302\ttotal: 1m 35s\tremaining: 1m 28s\n",
      "727:\tlearn: 0.9441773\ttotal: 1m 35s\tremaining: 1m 28s\n",
      "728:\tlearn: 0.9441378\ttotal: 1m 35s\tremaining: 1m 28s\n",
      "729:\tlearn: 0.9441092\ttotal: 1m 35s\tremaining: 1m 28s\n",
      "730:\tlearn: 0.9440605\ttotal: 1m 36s\tremaining: 1m 28s\n",
      "731:\tlearn: 0.9440342\ttotal: 1m 36s\tremaining: 1m 28s\n",
      "732:\tlearn: 0.9439900\ttotal: 1m 36s\tremaining: 1m 28s\n",
      "733:\tlearn: 0.9439647\ttotal: 1m 36s\tremaining: 1m 27s\n",
      "734:\tlearn: 0.9439204\ttotal: 1m 36s\tremaining: 1m 27s\n",
      "735:\tlearn: 0.9438827\ttotal: 1m 36s\tremaining: 1m 27s\n",
      "736:\tlearn: 0.9438330\ttotal: 1m 37s\tremaining: 1m 27s\n",
      "737:\tlearn: 0.9438094\ttotal: 1m 37s\tremaining: 1m 27s\n",
      "738:\tlearn: 0.9437823\ttotal: 1m 37s\tremaining: 1m 27s\n",
      "739:\tlearn: 0.9437580\ttotal: 1m 37s\tremaining: 1m 27s\n",
      "740:\tlearn: 0.9437218\ttotal: 1m 37s\tremaining: 1m 26s\n",
      "741:\tlearn: 0.9437209\ttotal: 1m 37s\tremaining: 1m 26s\n",
      "742:\tlearn: 0.9437014\ttotal: 1m 37s\tremaining: 1m 26s\n",
      "743:\tlearn: 0.9436680\ttotal: 1m 37s\tremaining: 1m 26s\n",
      "744:\tlearn: 0.9436418\ttotal: 1m 37s\tremaining: 1m 26s\n",
      "745:\tlearn: 0.9436062\ttotal: 1m 38s\tremaining: 1m 26s\n",
      "746:\tlearn: 0.9435764\ttotal: 1m 38s\tremaining: 1m 25s\n",
      "747:\tlearn: 0.9435445\ttotal: 1m 38s\tremaining: 1m 25s\n",
      "748:\tlearn: 0.9435193\ttotal: 1m 38s\tremaining: 1m 25s\n",
      "749:\tlearn: 0.9434895\ttotal: 1m 38s\tremaining: 1m 25s\n",
      "750:\tlearn: 0.9434882\ttotal: 1m 38s\tremaining: 1m 25s\n",
      "751:\tlearn: 0.9434710\ttotal: 1m 38s\tremaining: 1m 25s\n",
      "752:\tlearn: 0.9434581\ttotal: 1m 38s\tremaining: 1m 25s\n",
      "753:\tlearn: 0.9434166\ttotal: 1m 38s\tremaining: 1m 24s\n",
      "754:\tlearn: 0.9433796\ttotal: 1m 39s\tremaining: 1m 24s\n",
      "755:\tlearn: 0.9433557\ttotal: 1m 39s\tremaining: 1m 24s\n",
      "756:\tlearn: 0.9432985\ttotal: 1m 39s\tremaining: 1m 24s\n",
      "757:\tlearn: 0.9432791\ttotal: 1m 39s\tremaining: 1m 24s\n",
      "758:\tlearn: 0.9432519\ttotal: 1m 39s\tremaining: 1m 24s\n",
      "759:\tlearn: 0.9432020\ttotal: 1m 39s\tremaining: 1m 24s\n",
      "760:\tlearn: 0.9431631\ttotal: 1m 39s\tremaining: 1m 23s\n",
      "761:\tlearn: 0.9431493\ttotal: 1m 39s\tremaining: 1m 23s\n",
      "762:\tlearn: 0.9431217\ttotal: 1m 40s\tremaining: 1m 23s\n",
      "763:\tlearn: 0.9430761\ttotal: 1m 40s\tremaining: 1m 23s\n",
      "764:\tlearn: 0.9430171\ttotal: 1m 40s\tremaining: 1m 23s\n",
      "765:\tlearn: 0.9429683\ttotal: 1m 40s\tremaining: 1m 23s\n",
      "766:\tlearn: 0.9429407\ttotal: 1m 40s\tremaining: 1m 23s\n",
      "767:\tlearn: 0.9428937\ttotal: 1m 40s\tremaining: 1m 22s\n",
      "768:\tlearn: 0.9428596\ttotal: 1m 40s\tremaining: 1m 22s\n",
      "769:\tlearn: 0.9428210\ttotal: 1m 40s\tremaining: 1m 22s\n",
      "770:\tlearn: 0.9428038\ttotal: 1m 40s\tremaining: 1m 22s\n",
      "771:\tlearn: 0.9427803\ttotal: 1m 40s\tremaining: 1m 22s\n",
      "772:\tlearn: 0.9427726\ttotal: 1m 41s\tremaining: 1m 22s\n",
      "773:\tlearn: 0.9427209\ttotal: 1m 41s\tremaining: 1m 21s\n",
      "774:\tlearn: 0.9426679\ttotal: 1m 41s\tremaining: 1m 21s\n",
      "775:\tlearn: 0.9426076\ttotal: 1m 41s\tremaining: 1m 21s\n",
      "776:\tlearn: 0.9425755\ttotal: 1m 41s\tremaining: 1m 21s\n",
      "777:\tlearn: 0.9425546\ttotal: 1m 41s\tremaining: 1m 21s\n",
      "778:\tlearn: 0.9425129\ttotal: 1m 41s\tremaining: 1m 21s\n",
      "779:\tlearn: 0.9424874\ttotal: 1m 41s\tremaining: 1m 21s\n",
      "780:\tlearn: 0.9424703\ttotal: 1m 41s\tremaining: 1m 20s\n",
      "781:\tlearn: 0.9424687\ttotal: 1m 42s\tremaining: 1m 20s\n",
      "782:\tlearn: 0.9424251\ttotal: 1m 42s\tremaining: 1m 20s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "783:\tlearn: 0.9424059\ttotal: 1m 42s\tremaining: 1m 20s\n",
      "784:\tlearn: 0.9423568\ttotal: 1m 42s\tremaining: 1m 20s\n",
      "785:\tlearn: 0.9423273\ttotal: 1m 42s\tremaining: 1m 20s\n",
      "786:\tlearn: 0.9422992\ttotal: 1m 42s\tremaining: 1m 19s\n",
      "787:\tlearn: 0.9422901\ttotal: 1m 42s\tremaining: 1m 19s\n",
      "788:\tlearn: 0.9422682\ttotal: 1m 42s\tremaining: 1m 19s\n",
      "789:\tlearn: 0.9421931\ttotal: 1m 42s\tremaining: 1m 19s\n",
      "790:\tlearn: 0.9421373\ttotal: 1m 43s\tremaining: 1m 19s\n",
      "791:\tlearn: 0.9421165\ttotal: 1m 43s\tremaining: 1m 19s\n",
      "792:\tlearn: 0.9420818\ttotal: 1m 43s\tremaining: 1m 19s\n",
      "793:\tlearn: 0.9420561\ttotal: 1m 43s\tremaining: 1m 18s\n",
      "794:\tlearn: 0.9419979\ttotal: 1m 43s\tremaining: 1m 18s\n",
      "795:\tlearn: 0.9419645\ttotal: 1m 43s\tremaining: 1m 18s\n",
      "796:\tlearn: 0.9419316\ttotal: 1m 43s\tremaining: 1m 18s\n",
      "797:\tlearn: 0.9418714\ttotal: 1m 43s\tremaining: 1m 18s\n",
      "798:\tlearn: 0.9418446\ttotal: 1m 43s\tremaining: 1m 18s\n",
      "799:\tlearn: 0.9418127\ttotal: 1m 43s\tremaining: 1m 18s\n",
      "800:\tlearn: 0.9418120\ttotal: 1m 44s\tremaining: 1m 17s\n",
      "801:\tlearn: 0.9417858\ttotal: 1m 44s\tremaining: 1m 17s\n",
      "802:\tlearn: 0.9417547\ttotal: 1m 44s\tremaining: 1m 17s\n",
      "803:\tlearn: 0.9417173\ttotal: 1m 44s\tremaining: 1m 17s\n",
      "804:\tlearn: 0.9416845\ttotal: 1m 44s\tremaining: 1m 17s\n",
      "805:\tlearn: 0.9416434\ttotal: 1m 44s\tremaining: 1m 17s\n",
      "806:\tlearn: 0.9416061\ttotal: 1m 44s\tremaining: 1m 17s\n",
      "807:\tlearn: 0.9415446\ttotal: 1m 44s\tremaining: 1m 16s\n",
      "808:\tlearn: 0.9415348\ttotal: 1m 44s\tremaining: 1m 16s\n",
      "809:\tlearn: 0.9415245\ttotal: 1m 45s\tremaining: 1m 16s\n",
      "810:\tlearn: 0.9415233\ttotal: 1m 45s\tremaining: 1m 16s\n",
      "811:\tlearn: 0.9415020\ttotal: 1m 45s\tremaining: 1m 16s\n",
      "812:\tlearn: 0.9414747\ttotal: 1m 45s\tremaining: 1m 16s\n",
      "813:\tlearn: 0.9414334\ttotal: 1m 45s\tremaining: 1m 16s\n",
      "814:\tlearn: 0.9414047\ttotal: 1m 45s\tremaining: 1m 15s\n",
      "815:\tlearn: 0.9414021\ttotal: 1m 45s\tremaining: 1m 15s\n",
      "816:\tlearn: 0.9413802\ttotal: 1m 45s\tremaining: 1m 15s\n",
      "817:\tlearn: 0.9413486\ttotal: 1m 45s\tremaining: 1m 15s\n",
      "818:\tlearn: 0.9413037\ttotal: 1m 46s\tremaining: 1m 15s\n",
      "819:\tlearn: 0.9412826\ttotal: 1m 46s\tremaining: 1m 15s\n",
      "820:\tlearn: 0.9412439\ttotal: 1m 46s\tremaining: 1m 15s\n",
      "821:\tlearn: 0.9412133\ttotal: 1m 46s\tremaining: 1m 14s\n",
      "822:\tlearn: 0.9411858\ttotal: 1m 46s\tremaining: 1m 14s\n",
      "823:\tlearn: 0.9411706\ttotal: 1m 46s\tremaining: 1m 14s\n",
      "824:\tlearn: 0.9411465\ttotal: 1m 46s\tremaining: 1m 14s\n",
      "825:\tlearn: 0.9411091\ttotal: 1m 46s\tremaining: 1m 14s\n",
      "826:\tlearn: 0.9411076\ttotal: 1m 46s\tremaining: 1m 14s\n",
      "827:\tlearn: 0.9410469\ttotal: 1m 47s\tremaining: 1m 14s\n",
      "828:\tlearn: 0.9410096\ttotal: 1m 47s\tremaining: 1m 13s\n",
      "829:\tlearn: 0.9409491\ttotal: 1m 47s\tremaining: 1m 13s\n",
      "830:\tlearn: 0.9409183\ttotal: 1m 47s\tremaining: 1m 13s\n",
      "831:\tlearn: 0.9408865\ttotal: 1m 47s\tremaining: 1m 13s\n",
      "832:\tlearn: 0.9408560\ttotal: 1m 47s\tremaining: 1m 13s\n",
      "833:\tlearn: 0.9408123\ttotal: 1m 47s\tremaining: 1m 13s\n",
      "834:\tlearn: 0.9407920\ttotal: 1m 47s\tremaining: 1m 13s\n",
      "835:\tlearn: 0.9407812\ttotal: 1m 47s\tremaining: 1m 12s\n",
      "836:\tlearn: 0.9407364\ttotal: 1m 47s\tremaining: 1m 12s\n",
      "837:\tlearn: 0.9407258\ttotal: 1m 48s\tremaining: 1m 12s\n",
      "838:\tlearn: 0.9406879\ttotal: 1m 48s\tremaining: 1m 12s\n",
      "839:\tlearn: 0.9406551\ttotal: 1m 48s\tremaining: 1m 12s\n",
      "840:\tlearn: 0.9406388\ttotal: 1m 48s\tremaining: 1m 12s\n",
      "841:\tlearn: 0.9406135\ttotal: 1m 48s\tremaining: 1m 12s\n",
      "842:\tlearn: 0.9405977\ttotal: 1m 48s\tremaining: 1m 11s\n",
      "843:\tlearn: 0.9405748\ttotal: 1m 48s\tremaining: 1m 11s\n",
      "844:\tlearn: 0.9405587\ttotal: 1m 48s\tremaining: 1m 11s\n",
      "845:\tlearn: 0.9405308\ttotal: 1m 48s\tremaining: 1m 11s\n",
      "846:\tlearn: 0.9404999\ttotal: 1m 49s\tremaining: 1m 11s\n",
      "847:\tlearn: 0.9404620\ttotal: 1m 49s\tremaining: 1m 11s\n",
      "848:\tlearn: 0.9404109\ttotal: 1m 49s\tremaining: 1m 10s\n",
      "849:\tlearn: 0.9403705\ttotal: 1m 49s\tremaining: 1m 10s\n",
      "850:\tlearn: 0.9403347\ttotal: 1m 49s\tremaining: 1m 10s\n",
      "851:\tlearn: 0.9403095\ttotal: 1m 49s\tremaining: 1m 10s\n",
      "852:\tlearn: 0.9402797\ttotal: 1m 49s\tremaining: 1m 10s\n",
      "853:\tlearn: 0.9402632\ttotal: 1m 49s\tremaining: 1m 10s\n",
      "854:\tlearn: 0.9402197\ttotal: 1m 49s\tremaining: 1m 10s\n",
      "855:\tlearn: 0.9401830\ttotal: 1m 50s\tremaining: 1m 10s\n",
      "856:\tlearn: 0.9401404\ttotal: 1m 50s\tremaining: 1m 9s\n",
      "857:\tlearn: 0.9401054\ttotal: 1m 50s\tremaining: 1m 9s\n",
      "858:\tlearn: 0.9400753\ttotal: 1m 50s\tremaining: 1m 9s\n",
      "859:\tlearn: 0.9400358\ttotal: 1m 50s\tremaining: 1m 9s\n",
      "860:\tlearn: 0.9400050\ttotal: 1m 50s\tremaining: 1m 9s\n",
      "861:\tlearn: 0.9399914\ttotal: 1m 50s\tremaining: 1m 9s\n",
      "862:\tlearn: 0.9399910\ttotal: 1m 50s\tremaining: 1m 8s\n",
      "863:\tlearn: 0.9399541\ttotal: 1m 50s\tremaining: 1m 8s\n",
      "864:\tlearn: 0.9399327\ttotal: 1m 50s\tremaining: 1m 8s\n",
      "865:\tlearn: 0.9398894\ttotal: 1m 51s\tremaining: 1m 8s\n",
      "866:\tlearn: 0.9398576\ttotal: 1m 51s\tremaining: 1m 8s\n",
      "867:\tlearn: 0.9398554\ttotal: 1m 51s\tremaining: 1m 8s\n",
      "868:\tlearn: 0.9398255\ttotal: 1m 51s\tremaining: 1m 8s\n",
      "869:\tlearn: 0.9397737\ttotal: 1m 51s\tremaining: 1m 7s\n",
      "870:\tlearn: 0.9397175\ttotal: 1m 51s\tremaining: 1m 7s\n",
      "871:\tlearn: 0.9396778\ttotal: 1m 51s\tremaining: 1m 7s\n",
      "872:\tlearn: 0.9396431\ttotal: 1m 51s\tremaining: 1m 7s\n",
      "873:\tlearn: 0.9396059\ttotal: 1m 51s\tremaining: 1m 7s\n",
      "874:\tlearn: 0.9395563\ttotal: 1m 52s\tremaining: 1m 7s\n",
      "875:\tlearn: 0.9395533\ttotal: 1m 52s\tremaining: 1m 7s\n",
      "876:\tlearn: 0.9395183\ttotal: 1m 52s\tremaining: 1m 7s\n",
      "877:\tlearn: 0.9394841\ttotal: 1m 52s\tremaining: 1m 6s\n",
      "878:\tlearn: 0.9394500\ttotal: 1m 52s\tremaining: 1m 6s\n",
      "879:\tlearn: 0.9394246\ttotal: 1m 52s\tremaining: 1m 6s\n",
      "880:\tlearn: 0.9394223\ttotal: 1m 52s\tremaining: 1m 6s\n",
      "881:\tlearn: 0.9394106\ttotal: 1m 52s\tremaining: 1m 6s\n",
      "882:\tlearn: 0.9393854\ttotal: 1m 52s\tremaining: 1m 6s\n",
      "883:\tlearn: 0.9393605\ttotal: 1m 53s\tremaining: 1m 6s\n",
      "884:\tlearn: 0.9393327\ttotal: 1m 53s\tremaining: 1m 5s\n",
      "885:\tlearn: 0.9393320\ttotal: 1m 53s\tremaining: 1m 5s\n",
      "886:\tlearn: 0.9392825\ttotal: 1m 53s\tremaining: 1m 5s\n",
      "887:\tlearn: 0.9392500\ttotal: 1m 53s\tremaining: 1m 5s\n",
      "888:\tlearn: 0.9392409\ttotal: 1m 53s\tremaining: 1m 5s\n",
      "889:\tlearn: 0.9392201\ttotal: 1m 53s\tremaining: 1m 5s\n",
      "890:\tlearn: 0.9392193\ttotal: 1m 53s\tremaining: 1m 5s\n",
      "891:\tlearn: 0.9392156\ttotal: 1m 53s\tremaining: 1m 4s\n",
      "892:\tlearn: 0.9392148\ttotal: 1m 53s\tremaining: 1m 4s\n",
      "893:\tlearn: 0.9391904\ttotal: 1m 53s\tremaining: 1m 4s\n",
      "894:\tlearn: 0.9391648\ttotal: 1m 54s\tremaining: 1m 4s\n",
      "895:\tlearn: 0.9391638\ttotal: 1m 54s\tremaining: 1m 4s\n",
      "896:\tlearn: 0.9391178\ttotal: 1m 54s\tremaining: 1m 4s\n",
      "897:\tlearn: 0.9390886\ttotal: 1m 54s\tremaining: 1m 3s\n",
      "898:\tlearn: 0.9390600\ttotal: 1m 54s\tremaining: 1m 3s\n",
      "899:\tlearn: 0.9390141\ttotal: 1m 54s\tremaining: 1m 3s\n",
      "900:\tlearn: 0.9389842\ttotal: 1m 54s\tremaining: 1m 3s\n",
      "901:\tlearn: 0.9389709\ttotal: 1m 54s\tremaining: 1m 3s\n",
      "902:\tlearn: 0.9389303\ttotal: 1m 54s\tremaining: 1m 3s\n",
      "903:\tlearn: 0.9388972\ttotal: 1m 54s\tremaining: 1m 3s\n",
      "904:\tlearn: 0.9388627\ttotal: 1m 55s\tremaining: 1m 2s\n",
      "905:\tlearn: 0.9388336\ttotal: 1m 55s\tremaining: 1m 2s\n",
      "906:\tlearn: 0.9387830\ttotal: 1m 55s\tremaining: 1m 2s\n",
      "907:\tlearn: 0.9387441\ttotal: 1m 55s\tremaining: 1m 2s\n",
      "908:\tlearn: 0.9387082\ttotal: 1m 55s\tremaining: 1m 2s\n",
      "909:\tlearn: 0.9386779\ttotal: 1m 55s\tremaining: 1m 2s\n",
      "910:\tlearn: 0.9386427\ttotal: 1m 55s\tremaining: 1m 2s\n",
      "911:\tlearn: 0.9385897\ttotal: 1m 55s\tremaining: 1m 2s\n",
      "912:\tlearn: 0.9385819\ttotal: 1m 56s\tremaining: 1m 1s\n",
      "913:\tlearn: 0.9385659\ttotal: 1m 56s\tremaining: 1m 1s\n",
      "914:\tlearn: 0.9385612\ttotal: 1m 56s\tremaining: 1m 1s\n",
      "915:\tlearn: 0.9385267\ttotal: 1m 56s\tremaining: 1m 1s\n",
      "916:\tlearn: 0.9384945\ttotal: 1m 56s\tremaining: 1m 1s\n",
      "917:\tlearn: 0.9384719\ttotal: 1m 56s\tremaining: 1m 1s\n",
      "918:\tlearn: 0.9384283\ttotal: 1m 56s\tremaining: 1m 1s\n",
      "919:\tlearn: 0.9383684\ttotal: 1m 56s\tremaining: 1m 1s\n",
      "920:\tlearn: 0.9383473\ttotal: 1m 57s\tremaining: 1m\n",
      "921:\tlearn: 0.9383355\ttotal: 1m 57s\tremaining: 1m\n",
      "922:\tlearn: 0.9382832\ttotal: 1m 57s\tremaining: 1m\n",
      "923:\tlearn: 0.9382703\ttotal: 1m 57s\tremaining: 1m\n",
      "924:\tlearn: 0.9382399\ttotal: 1m 57s\tremaining: 1m\n",
      "925:\tlearn: 0.9382104\ttotal: 1m 57s\tremaining: 1m\n",
      "926:\tlearn: 0.9381886\ttotal: 1m 57s\tremaining: 1m\n",
      "927:\tlearn: 0.9381647\ttotal: 1m 58s\tremaining: 1m\n",
      "928:\tlearn: 0.9381486\ttotal: 1m 58s\tremaining: 1m\n",
      "929:\tlearn: 0.9381283\ttotal: 1m 58s\tremaining: 59.9s\n",
      "930:\tlearn: 0.9380993\ttotal: 1m 58s\tremaining: 59.8s\n",
      "931:\tlearn: 0.9380625\ttotal: 1m 58s\tremaining: 59.7s\n",
      "932:\tlearn: 0.9380324\ttotal: 1m 58s\tremaining: 59.6s\n",
      "933:\tlearn: 0.9379906\ttotal: 1m 59s\tremaining: 59.5s\n",
      "934:\tlearn: 0.9379155\ttotal: 1m 59s\tremaining: 59.4s\n",
      "935:\tlearn: 0.9379075\ttotal: 1m 59s\tremaining: 59.2s\n",
      "936:\tlearn: 0.9378563\ttotal: 1m 59s\tremaining: 59.1s\n",
      "937:\tlearn: 0.9378384\ttotal: 1m 59s\tremaining: 58.9s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "938:\tlearn: 0.9377981\ttotal: 1m 59s\tremaining: 58.8s\n",
      "939:\tlearn: 0.9377714\ttotal: 1m 59s\tremaining: 58.7s\n",
      "940:\tlearn: 0.9377707\ttotal: 1m 59s\tremaining: 58.5s\n",
      "941:\tlearn: 0.9377385\ttotal: 1m 59s\tremaining: 58.4s\n",
      "942:\tlearn: 0.9377063\ttotal: 2m\tremaining: 58.2s\n",
      "943:\tlearn: 0.9376825\ttotal: 2m\tremaining: 58.1s\n",
      "944:\tlearn: 0.9376357\ttotal: 2m\tremaining: 58s\n",
      "945:\tlearn: 0.9376070\ttotal: 2m\tremaining: 57.8s\n",
      "946:\tlearn: 0.9375769\ttotal: 2m\tremaining: 57.7s\n",
      "947:\tlearn: 0.9375506\ttotal: 2m\tremaining: 57.6s\n",
      "948:\tlearn: 0.9375031\ttotal: 2m\tremaining: 57.4s\n",
      "949:\tlearn: 0.9374735\ttotal: 2m\tremaining: 57.3s\n",
      "950:\tlearn: 0.9374399\ttotal: 2m\tremaining: 57.1s\n",
      "951:\tlearn: 0.9374009\ttotal: 2m 1s\tremaining: 57s\n",
      "952:\tlearn: 0.9373992\ttotal: 2m 1s\tremaining: 56.9s\n",
      "953:\tlearn: 0.9373718\ttotal: 2m 1s\tremaining: 56.7s\n",
      "954:\tlearn: 0.9373621\ttotal: 2m 1s\tremaining: 56.6s\n",
      "955:\tlearn: 0.9373029\ttotal: 2m 1s\tremaining: 56.5s\n",
      "956:\tlearn: 0.9372745\ttotal: 2m 1s\tremaining: 56.3s\n",
      "957:\tlearn: 0.9372406\ttotal: 2m 1s\tremaining: 56.2s\n",
      "958:\tlearn: 0.9371990\ttotal: 2m 1s\tremaining: 56s\n",
      "959:\tlearn: 0.9371755\ttotal: 2m 1s\tremaining: 55.9s\n",
      "960:\tlearn: 0.9371344\ttotal: 2m 1s\tremaining: 55.8s\n",
      "961:\tlearn: 0.9371153\ttotal: 2m 2s\tremaining: 55.6s\n",
      "962:\tlearn: 0.9370851\ttotal: 2m 2s\tremaining: 55.5s\n",
      "963:\tlearn: 0.9370639\ttotal: 2m 2s\tremaining: 55.3s\n",
      "964:\tlearn: 0.9370620\ttotal: 2m 2s\tremaining: 55.2s\n",
      "965:\tlearn: 0.9370474\ttotal: 2m 2s\tremaining: 55.1s\n",
      "966:\tlearn: 0.9370464\ttotal: 2m 2s\tremaining: 54.9s\n",
      "967:\tlearn: 0.9370450\ttotal: 2m 2s\tremaining: 54.7s\n",
      "968:\tlearn: 0.9370443\ttotal: 2m 2s\tremaining: 54.6s\n",
      "969:\tlearn: 0.9370264\ttotal: 2m 2s\tremaining: 54.5s\n",
      "970:\tlearn: 0.9369893\ttotal: 2m 2s\tremaining: 54.3s\n",
      "971:\tlearn: 0.9369443\ttotal: 2m 2s\tremaining: 54.2s\n",
      "972:\tlearn: 0.9368943\ttotal: 2m 3s\tremaining: 54.1s\n",
      "973:\tlearn: 0.9368825\ttotal: 2m 3s\tremaining: 53.9s\n",
      "974:\tlearn: 0.9368329\ttotal: 2m 3s\tremaining: 53.8s\n",
      "975:\tlearn: 0.9367854\ttotal: 2m 3s\tremaining: 53.7s\n",
      "976:\tlearn: 0.9367540\ttotal: 2m 3s\tremaining: 53.5s\n",
      "977:\tlearn: 0.9367283\ttotal: 2m 3s\tremaining: 53.4s\n",
      "978:\tlearn: 0.9366935\ttotal: 2m 3s\tremaining: 53.3s\n",
      "979:\tlearn: 0.9366756\ttotal: 2m 3s\tremaining: 53.1s\n",
      "980:\tlearn: 0.9366493\ttotal: 2m 3s\tremaining: 53s\n",
      "981:\tlearn: 0.9365782\ttotal: 2m 4s\tremaining: 52.9s\n",
      "982:\tlearn: 0.9365483\ttotal: 2m 4s\tremaining: 52.8s\n",
      "983:\tlearn: 0.9365151\ttotal: 2m 4s\tremaining: 52.6s\n",
      "984:\tlearn: 0.9364919\ttotal: 2m 4s\tremaining: 52.5s\n",
      "985:\tlearn: 0.9364894\ttotal: 2m 4s\tremaining: 52.3s\n",
      "986:\tlearn: 0.9364545\ttotal: 2m 4s\tremaining: 52.2s\n",
      "987:\tlearn: 0.9364532\ttotal: 2m 4s\tremaining: 52.1s\n",
      "988:\tlearn: 0.9364277\ttotal: 2m 4s\tremaining: 51.9s\n",
      "989:\tlearn: 0.9364118\ttotal: 2m 4s\tremaining: 51.8s\n",
      "990:\tlearn: 0.9363791\ttotal: 2m 5s\tremaining: 51.7s\n",
      "991:\tlearn: 0.9363514\ttotal: 2m 5s\tremaining: 51.5s\n",
      "992:\tlearn: 0.9363503\ttotal: 2m 5s\tremaining: 51.4s\n",
      "993:\tlearn: 0.9363232\ttotal: 2m 5s\tremaining: 51.2s\n",
      "994:\tlearn: 0.9363036\ttotal: 2m 5s\tremaining: 51.1s\n",
      "995:\tlearn: 0.9362806\ttotal: 2m 5s\tremaining: 51s\n",
      "996:\tlearn: 0.9362497\ttotal: 2m 5s\tremaining: 50.8s\n",
      "997:\tlearn: 0.9362238\ttotal: 2m 5s\tremaining: 50.7s\n",
      "998:\tlearn: 0.9361879\ttotal: 2m 5s\tremaining: 50.6s\n",
      "999:\tlearn: 0.9361459\ttotal: 2m 5s\tremaining: 50.4s\n",
      "1000:\tlearn: 0.9361119\ttotal: 2m 6s\tremaining: 50.3s\n",
      "1001:\tlearn: 0.9361028\ttotal: 2m 6s\tremaining: 50.2s\n",
      "1002:\tlearn: 0.9360577\ttotal: 2m 6s\tremaining: 50s\n",
      "1003:\tlearn: 0.9360093\ttotal: 2m 6s\tremaining: 49.9s\n",
      "1004:\tlearn: 0.9359703\ttotal: 2m 6s\tremaining: 49.8s\n",
      "1005:\tlearn: 0.9359441\ttotal: 2m 6s\tremaining: 49.6s\n",
      "1006:\tlearn: 0.9359069\ttotal: 2m 6s\tremaining: 49.5s\n",
      "1007:\tlearn: 0.9358832\ttotal: 2m 6s\tremaining: 49.4s\n",
      "1008:\tlearn: 0.9358568\ttotal: 2m 6s\tremaining: 49.2s\n",
      "1009:\tlearn: 0.9358373\ttotal: 2m 7s\tremaining: 49.1s\n",
      "1010:\tlearn: 0.9357943\ttotal: 2m 7s\tremaining: 49s\n",
      "1011:\tlearn: 0.9357381\ttotal: 2m 7s\tremaining: 48.8s\n",
      "1012:\tlearn: 0.9357226\ttotal: 2m 7s\tremaining: 48.7s\n",
      "1013:\tlearn: 0.9357044\ttotal: 2m 7s\tremaining: 48.6s\n",
      "1014:\tlearn: 0.9356789\ttotal: 2m 7s\tremaining: 48.4s\n",
      "1015:\tlearn: 0.9356738\ttotal: 2m 7s\tremaining: 48.3s\n",
      "1016:\tlearn: 0.9356298\ttotal: 2m 7s\tremaining: 48.2s\n",
      "1017:\tlearn: 0.9356093\ttotal: 2m 7s\tremaining: 48s\n",
      "1018:\tlearn: 0.9356085\ttotal: 2m 7s\tremaining: 47.9s\n",
      "1019:\tlearn: 0.9355756\ttotal: 2m 8s\tremaining: 47.8s\n",
      "1020:\tlearn: 0.9355314\ttotal: 2m 8s\tremaining: 47.6s\n",
      "1021:\tlearn: 0.9355053\ttotal: 2m 8s\tremaining: 47.5s\n",
      "1022:\tlearn: 0.9354853\ttotal: 2m 8s\tremaining: 47.4s\n",
      "1023:\tlearn: 0.9354780\ttotal: 2m 8s\tremaining: 47.2s\n",
      "1024:\tlearn: 0.9354433\ttotal: 2m 8s\tremaining: 47.1s\n",
      "1025:\tlearn: 0.9354004\ttotal: 2m 8s\tremaining: 47s\n",
      "1026:\tlearn: 0.9353995\ttotal: 2m 8s\tremaining: 46.8s\n",
      "1027:\tlearn: 0.9353611\ttotal: 2m 9s\tremaining: 46.7s\n",
      "1028:\tlearn: 0.9353542\ttotal: 2m 9s\tremaining: 46.6s\n",
      "1029:\tlearn: 0.9353536\ttotal: 2m 9s\tremaining: 46.4s\n",
      "1030:\tlearn: 0.9353201\ttotal: 2m 9s\tremaining: 46.3s\n",
      "1031:\tlearn: 0.9353196\ttotal: 2m 9s\tremaining: 46.2s\n",
      "1032:\tlearn: 0.9352955\ttotal: 2m 9s\tremaining: 46.1s\n",
      "1033:\tlearn: 0.9352620\ttotal: 2m 9s\tremaining: 45.9s\n",
      "1034:\tlearn: 0.9352341\ttotal: 2m 9s\tremaining: 45.8s\n",
      "1035:\tlearn: 0.9351845\ttotal: 2m 9s\tremaining: 45.7s\n",
      "1036:\tlearn: 0.9351566\ttotal: 2m 10s\tremaining: 45.6s\n",
      "1037:\tlearn: 0.9351558\ttotal: 2m 10s\tremaining: 45.5s\n",
      "1038:\tlearn: 0.9351012\ttotal: 2m 10s\tremaining: 45.3s\n",
      "1039:\tlearn: 0.9350667\ttotal: 2m 10s\tremaining: 45.2s\n",
      "1040:\tlearn: 0.9350271\ttotal: 2m 10s\tremaining: 45.1s\n",
      "1041:\tlearn: 0.9349968\ttotal: 2m 10s\tremaining: 45s\n",
      "1042:\tlearn: 0.9349961\ttotal: 2m 10s\tremaining: 44.8s\n",
      "1043:\tlearn: 0.9349569\ttotal: 2m 11s\tremaining: 44.7s\n",
      "1044:\tlearn: 0.9349023\ttotal: 2m 11s\tremaining: 44.6s\n",
      "1045:\tlearn: 0.9349016\ttotal: 2m 11s\tremaining: 44.5s\n",
      "1046:\tlearn: 0.9348792\ttotal: 2m 11s\tremaining: 44.4s\n",
      "1047:\tlearn: 0.9348417\ttotal: 2m 11s\tremaining: 44.2s\n",
      "1048:\tlearn: 0.9348128\ttotal: 2m 11s\tremaining: 44.1s\n",
      "1049:\tlearn: 0.9347963\ttotal: 2m 11s\tremaining: 44s\n",
      "1050:\tlearn: 0.9347609\ttotal: 2m 11s\tremaining: 43.8s\n",
      "1051:\tlearn: 0.9347206\ttotal: 2m 12s\tremaining: 43.7s\n",
      "1052:\tlearn: 0.9346891\ttotal: 2m 12s\tremaining: 43.6s\n",
      "1053:\tlearn: 0.9346504\ttotal: 2m 12s\tremaining: 43.5s\n",
      "1054:\tlearn: 0.9346051\ttotal: 2m 12s\tremaining: 43.3s\n",
      "1055:\tlearn: 0.9345521\ttotal: 2m 12s\tremaining: 43.2s\n",
      "1056:\tlearn: 0.9345271\ttotal: 2m 12s\tremaining: 43.1s\n",
      "1057:\tlearn: 0.9344841\ttotal: 2m 12s\tremaining: 43s\n",
      "1058:\tlearn: 0.9344449\ttotal: 2m 12s\tremaining: 42.8s\n",
      "1059:\tlearn: 0.9344263\ttotal: 2m 13s\tremaining: 42.7s\n",
      "1060:\tlearn: 0.9343928\ttotal: 2m 13s\tremaining: 42.6s\n",
      "1061:\tlearn: 0.9343814\ttotal: 2m 13s\tremaining: 42.4s\n",
      "1062:\tlearn: 0.9343626\ttotal: 2m 13s\tremaining: 42.3s\n",
      "1063:\tlearn: 0.9343238\ttotal: 2m 13s\tremaining: 42.2s\n",
      "1064:\tlearn: 0.9342942\ttotal: 2m 13s\tremaining: 42.1s\n",
      "1065:\tlearn: 0.9342781\ttotal: 2m 13s\tremaining: 41.9s\n",
      "1066:\tlearn: 0.9342551\ttotal: 2m 13s\tremaining: 41.8s\n",
      "1067:\tlearn: 0.9342194\ttotal: 2m 14s\tremaining: 41.7s\n",
      "1068:\tlearn: 0.9342184\ttotal: 2m 14s\tremaining: 41.6s\n",
      "1069:\tlearn: 0.9341763\ttotal: 2m 14s\tremaining: 41.4s\n",
      "1070:\tlearn: 0.9341605\ttotal: 2m 14s\tremaining: 41.3s\n",
      "1071:\tlearn: 0.9341253\ttotal: 2m 14s\tremaining: 41.2s\n",
      "1072:\tlearn: 0.9340959\ttotal: 2m 14s\tremaining: 41s\n",
      "1073:\tlearn: 0.9340929\ttotal: 2m 14s\tremaining: 40.9s\n",
      "1074:\tlearn: 0.9340636\ttotal: 2m 14s\tremaining: 40.8s\n",
      "1075:\tlearn: 0.9340102\ttotal: 2m 14s\tremaining: 40.7s\n",
      "1076:\tlearn: 0.9339798\ttotal: 2m 15s\tremaining: 40.5s\n",
      "1077:\tlearn: 0.9339503\ttotal: 2m 15s\tremaining: 40.4s\n",
      "1078:\tlearn: 0.9339183\ttotal: 2m 15s\tremaining: 40.3s\n",
      "1079:\tlearn: 0.9339091\ttotal: 2m 15s\tremaining: 40.2s\n",
      "1080:\tlearn: 0.9338590\ttotal: 2m 15s\tremaining: 40.1s\n",
      "1081:\tlearn: 0.9338122\ttotal: 2m 15s\tremaining: 39.9s\n",
      "1082:\tlearn: 0.9337635\ttotal: 2m 15s\tremaining: 39.8s\n",
      "1083:\tlearn: 0.9337265\ttotal: 2m 16s\tremaining: 39.7s\n",
      "1084:\tlearn: 0.9336928\ttotal: 2m 16s\tremaining: 39.6s\n",
      "1085:\tlearn: 0.9336445\ttotal: 2m 16s\tremaining: 39.5s\n",
      "1086:\tlearn: 0.9336040\ttotal: 2m 16s\tremaining: 39.3s\n",
      "1087:\tlearn: 0.9335760\ttotal: 2m 16s\tremaining: 39.2s\n",
      "1088:\tlearn: 0.9335525\ttotal: 2m 16s\tremaining: 39.1s\n",
      "1089:\tlearn: 0.9335197\ttotal: 2m 16s\tremaining: 39s\n",
      "1090:\tlearn: 0.9334579\ttotal: 2m 17s\tremaining: 38.8s\n",
      "1091:\tlearn: 0.9334309\ttotal: 2m 17s\tremaining: 38.7s\n",
      "1092:\tlearn: 0.9333859\ttotal: 2m 17s\tremaining: 38.6s\n",
      "1093:\tlearn: 0.9333575\ttotal: 2m 17s\tremaining: 38.5s\n",
      "1094:\tlearn: 0.9333320\ttotal: 2m 17s\tremaining: 38.3s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1095:\tlearn: 0.9333036\ttotal: 2m 17s\tremaining: 38.2s\n",
      "1096:\tlearn: 0.9332741\ttotal: 2m 17s\tremaining: 38.1s\n",
      "1097:\tlearn: 0.9332484\ttotal: 2m 17s\tremaining: 37.9s\n",
      "1098:\tlearn: 0.9332077\ttotal: 2m 17s\tremaining: 37.8s\n",
      "1099:\tlearn: 0.9331716\ttotal: 2m 18s\tremaining: 37.7s\n",
      "1100:\tlearn: 0.9331361\ttotal: 2m 18s\tremaining: 37.6s\n",
      "1101:\tlearn: 0.9331111\ttotal: 2m 18s\tremaining: 37.4s\n",
      "1102:\tlearn: 0.9330924\ttotal: 2m 18s\tremaining: 37.3s\n",
      "1103:\tlearn: 0.9330572\ttotal: 2m 18s\tremaining: 37.2s\n",
      "1104:\tlearn: 0.9330143\ttotal: 2m 18s\tremaining: 37.1s\n",
      "1105:\tlearn: 0.9329817\ttotal: 2m 18s\tremaining: 37s\n",
      "1106:\tlearn: 0.9329335\ttotal: 2m 19s\tremaining: 36.8s\n",
      "1107:\tlearn: 0.9329053\ttotal: 2m 19s\tremaining: 36.7s\n",
      "1108:\tlearn: 0.9328769\ttotal: 2m 19s\tremaining: 36.6s\n",
      "1109:\tlearn: 0.9328741\ttotal: 2m 19s\tremaining: 36.5s\n",
      "1110:\tlearn: 0.9328525\ttotal: 2m 19s\tremaining: 36.3s\n",
      "1111:\tlearn: 0.9328208\ttotal: 2m 19s\tremaining: 36.2s\n",
      "1112:\tlearn: 0.9327765\ttotal: 2m 19s\tremaining: 36.1s\n",
      "1113:\tlearn: 0.9327750\ttotal: 2m 19s\tremaining: 36s\n",
      "1114:\tlearn: 0.9327687\ttotal: 2m 20s\tremaining: 35.8s\n",
      "1115:\tlearn: 0.9327671\ttotal: 2m 20s\tremaining: 35.7s\n",
      "1116:\tlearn: 0.9327515\ttotal: 2m 20s\tremaining: 35.6s\n",
      "1117:\tlearn: 0.9327263\ttotal: 2m 20s\tremaining: 35.4s\n",
      "1118:\tlearn: 0.9327250\ttotal: 2m 20s\tremaining: 35.3s\n",
      "1119:\tlearn: 0.9327017\ttotal: 2m 20s\tremaining: 35.2s\n",
      "1120:\tlearn: 0.9326780\ttotal: 2m 20s\tremaining: 35.1s\n",
      "1121:\tlearn: 0.9326570\ttotal: 2m 20s\tremaining: 34.9s\n",
      "1122:\tlearn: 0.9326317\ttotal: 2m 21s\tremaining: 34.8s\n",
      "1123:\tlearn: 0.9326148\ttotal: 2m 21s\tremaining: 34.7s\n",
      "1124:\tlearn: 0.9326104\ttotal: 2m 21s\tremaining: 34.6s\n",
      "1125:\tlearn: 0.9325946\ttotal: 2m 21s\tremaining: 34.4s\n",
      "1126:\tlearn: 0.9325579\ttotal: 2m 21s\tremaining: 34.3s\n",
      "1127:\tlearn: 0.9325335\ttotal: 2m 21s\tremaining: 34.2s\n",
      "1128:\tlearn: 0.9325043\ttotal: 2m 21s\tremaining: 34.1s\n",
      "1129:\tlearn: 0.9324815\ttotal: 2m 21s\tremaining: 33.9s\n",
      "1130:\tlearn: 0.9324473\ttotal: 2m 22s\tremaining: 33.8s\n",
      "1131:\tlearn: 0.9324082\ttotal: 2m 22s\tremaining: 33.7s\n",
      "1132:\tlearn: 0.9323678\ttotal: 2m 22s\tremaining: 33.6s\n",
      "1133:\tlearn: 0.9323398\ttotal: 2m 22s\tremaining: 33.5s\n",
      "1134:\tlearn: 0.9322950\ttotal: 2m 22s\tremaining: 33.3s\n",
      "1135:\tlearn: 0.9322470\ttotal: 2m 22s\tremaining: 33.2s\n",
      "1136:\tlearn: 0.9322208\ttotal: 2m 22s\tremaining: 33.1s\n",
      "1137:\tlearn: 0.9321751\ttotal: 2m 23s\tremaining: 33s\n",
      "1138:\tlearn: 0.9321476\ttotal: 2m 23s\tremaining: 32.8s\n",
      "1139:\tlearn: 0.9321358\ttotal: 2m 23s\tremaining: 32.7s\n",
      "1140:\tlearn: 0.9321201\ttotal: 2m 23s\tremaining: 32.6s\n",
      "1141:\tlearn: 0.9321169\ttotal: 2m 23s\tremaining: 32.4s\n",
      "1142:\tlearn: 0.9320965\ttotal: 2m 23s\tremaining: 32.3s\n",
      "1143:\tlearn: 0.9320653\ttotal: 2m 23s\tremaining: 32.2s\n",
      "1144:\tlearn: 0.9320545\ttotal: 2m 23s\tremaining: 32s\n",
      "1145:\tlearn: 0.9320345\ttotal: 2m 23s\tremaining: 31.9s\n",
      "1146:\tlearn: 0.9320193\ttotal: 2m 24s\tremaining: 31.8s\n",
      "1147:\tlearn: 0.9320176\ttotal: 2m 24s\tremaining: 31.7s\n",
      "1148:\tlearn: 0.9319598\ttotal: 2m 24s\tremaining: 31.5s\n",
      "1149:\tlearn: 0.9319252\ttotal: 2m 24s\tremaining: 31.4s\n",
      "1150:\tlearn: 0.9318777\ttotal: 2m 24s\tremaining: 31.3s\n",
      "1151:\tlearn: 0.9318740\ttotal: 2m 24s\tremaining: 31.2s\n",
      "1152:\tlearn: 0.9318411\ttotal: 2m 24s\tremaining: 31s\n",
      "1153:\tlearn: 0.9318248\ttotal: 2m 24s\tremaining: 30.9s\n",
      "1154:\tlearn: 0.9317861\ttotal: 2m 24s\tremaining: 30.8s\n",
      "1155:\tlearn: 0.9317592\ttotal: 2m 25s\tremaining: 30.6s\n",
      "1156:\tlearn: 0.9317214\ttotal: 2m 25s\tremaining: 30.5s\n",
      "1157:\tlearn: 0.9316828\ttotal: 2m 25s\tremaining: 30.4s\n",
      "1158:\tlearn: 0.9316532\ttotal: 2m 25s\tremaining: 30.3s\n",
      "1159:\tlearn: 0.9316159\ttotal: 2m 25s\tremaining: 30.2s\n",
      "1160:\tlearn: 0.9315855\ttotal: 2m 25s\tremaining: 30.1s\n",
      "1161:\tlearn: 0.9315546\ttotal: 2m 26s\tremaining: 30s\n",
      "1162:\tlearn: 0.9315184\ttotal: 2m 26s\tremaining: 29.9s\n",
      "1163:\tlearn: 0.9314858\ttotal: 2m 26s\tremaining: 29.7s\n",
      "1164:\tlearn: 0.9314353\ttotal: 2m 26s\tremaining: 29.6s\n",
      "1165:\tlearn: 0.9314094\ttotal: 2m 26s\tremaining: 29.5s\n",
      "1166:\tlearn: 0.9313871\ttotal: 2m 26s\tremaining: 29.4s\n",
      "1167:\tlearn: 0.9313693\ttotal: 2m 27s\tremaining: 29.2s\n",
      "1168:\tlearn: 0.9313270\ttotal: 2m 27s\tremaining: 29.1s\n",
      "1169:\tlearn: 0.9312930\ttotal: 2m 27s\tremaining: 29s\n",
      "1170:\tlearn: 0.9312565\ttotal: 2m 27s\tremaining: 28.9s\n",
      "1171:\tlearn: 0.9312300\ttotal: 2m 27s\tremaining: 28.7s\n",
      "1172:\tlearn: 0.9312058\ttotal: 2m 27s\tremaining: 28.6s\n",
      "1173:\tlearn: 0.9311768\ttotal: 2m 28s\tremaining: 28.5s\n",
      "1174:\tlearn: 0.9311567\ttotal: 2m 28s\tremaining: 28.4s\n",
      "1175:\tlearn: 0.9311267\ttotal: 2m 28s\tremaining: 28.3s\n",
      "1176:\tlearn: 0.9310754\ttotal: 2m 28s\tremaining: 28.2s\n",
      "1177:\tlearn: 0.9310390\ttotal: 2m 28s\tremaining: 28s\n",
      "1178:\tlearn: 0.9310236\ttotal: 2m 28s\tremaining: 27.9s\n",
      "1179:\tlearn: 0.9310002\ttotal: 2m 29s\tremaining: 27.8s\n",
      "1180:\tlearn: 0.9309834\ttotal: 2m 29s\tremaining: 27.7s\n",
      "1181:\tlearn: 0.9309811\ttotal: 2m 29s\tremaining: 27.6s\n",
      "1182:\tlearn: 0.9309302\ttotal: 2m 29s\tremaining: 27.5s\n",
      "1183:\tlearn: 0.9309104\ttotal: 2m 29s\tremaining: 27.4s\n",
      "1184:\tlearn: 0.9308776\ttotal: 2m 30s\tremaining: 27.3s\n",
      "1185:\tlearn: 0.9308341\ttotal: 2m 30s\tremaining: 27.2s\n",
      "1186:\tlearn: 0.9307982\ttotal: 2m 30s\tremaining: 27.1s\n",
      "1187:\tlearn: 0.9307669\ttotal: 2m 30s\tremaining: 26.9s\n",
      "1188:\tlearn: 0.9307272\ttotal: 2m 30s\tremaining: 26.8s\n",
      "1189:\tlearn: 0.9306914\ttotal: 2m 31s\tremaining: 26.7s\n",
      "1190:\tlearn: 0.9306611\ttotal: 2m 31s\tremaining: 26.6s\n",
      "1191:\tlearn: 0.9306229\ttotal: 2m 31s\tremaining: 26.5s\n",
      "1192:\tlearn: 0.9306126\ttotal: 2m 31s\tremaining: 26.4s\n",
      "1193:\tlearn: 0.9305804\ttotal: 2m 32s\tremaining: 26.3s\n",
      "1194:\tlearn: 0.9305453\ttotal: 2m 32s\tremaining: 26.2s\n",
      "1195:\tlearn: 0.9305435\ttotal: 2m 32s\tremaining: 26s\n",
      "1196:\tlearn: 0.9305268\ttotal: 2m 32s\tremaining: 25.9s\n",
      "1197:\tlearn: 0.9304785\ttotal: 2m 32s\tremaining: 25.8s\n",
      "1198:\tlearn: 0.9304534\ttotal: 2m 32s\tremaining: 25.7s\n",
      "1199:\tlearn: 0.9304110\ttotal: 2m 33s\tremaining: 25.5s\n",
      "1200:\tlearn: 0.9303845\ttotal: 2m 33s\tremaining: 25.4s\n",
      "1201:\tlearn: 0.9303568\ttotal: 2m 33s\tremaining: 25.3s\n",
      "1202:\tlearn: 0.9303448\ttotal: 2m 33s\tremaining: 25.2s\n",
      "1203:\tlearn: 0.9303331\ttotal: 2m 33s\tremaining: 25.1s\n",
      "1204:\tlearn: 0.9303233\ttotal: 2m 33s\tremaining: 24.9s\n",
      "1205:\tlearn: 0.9303118\ttotal: 2m 33s\tremaining: 24.8s\n",
      "1206:\tlearn: 0.9302928\ttotal: 2m 34s\tremaining: 24.7s\n",
      "1207:\tlearn: 0.9302443\ttotal: 2m 34s\tremaining: 24.5s\n",
      "1208:\tlearn: 0.9302091\ttotal: 2m 34s\tremaining: 24.4s\n",
      "1209:\tlearn: 0.9301704\ttotal: 2m 34s\tremaining: 24.3s\n",
      "1210:\tlearn: 0.9301321\ttotal: 2m 34s\tremaining: 24.2s\n",
      "1211:\tlearn: 0.9301166\ttotal: 2m 35s\tremaining: 24.1s\n",
      "1212:\tlearn: 0.9301035\ttotal: 2m 35s\tremaining: 23.9s\n",
      "1213:\tlearn: 0.9300760\ttotal: 2m 35s\tremaining: 23.8s\n",
      "1214:\tlearn: 0.9300268\ttotal: 2m 35s\tremaining: 23.7s\n",
      "1215:\tlearn: 0.9300078\ttotal: 2m 35s\tremaining: 23.6s\n",
      "1216:\tlearn: 0.9299820\ttotal: 2m 35s\tremaining: 23.4s\n",
      "1217:\tlearn: 0.9299705\ttotal: 2m 35s\tremaining: 23.3s\n",
      "1218:\tlearn: 0.9299519\ttotal: 2m 35s\tremaining: 23.2s\n",
      "1219:\tlearn: 0.9299329\ttotal: 2m 36s\tremaining: 23s\n",
      "1220:\tlearn: 0.9299079\ttotal: 2m 36s\tremaining: 22.9s\n",
      "1221:\tlearn: 0.9298485\ttotal: 2m 36s\tremaining: 22.8s\n",
      "1222:\tlearn: 0.9298322\ttotal: 2m 36s\tremaining: 22.7s\n",
      "1223:\tlearn: 0.9297975\ttotal: 2m 36s\tremaining: 22.6s\n",
      "1224:\tlearn: 0.9297733\ttotal: 2m 36s\tremaining: 22.4s\n",
      "1225:\tlearn: 0.9297325\ttotal: 2m 37s\tremaining: 22.3s\n",
      "1226:\tlearn: 0.9297148\ttotal: 2m 37s\tremaining: 22.2s\n",
      "1227:\tlearn: 0.9296875\ttotal: 2m 37s\tremaining: 22.1s\n",
      "1228:\tlearn: 0.9296509\ttotal: 2m 37s\tremaining: 21.9s\n",
      "1229:\tlearn: 0.9296156\ttotal: 2m 37s\tremaining: 21.8s\n",
      "1230:\tlearn: 0.9296085\ttotal: 2m 37s\tremaining: 21.7s\n",
      "1231:\tlearn: 0.9295633\ttotal: 2m 38s\tremaining: 21.6s\n",
      "1232:\tlearn: 0.9295587\ttotal: 2m 38s\tremaining: 21.4s\n",
      "1233:\tlearn: 0.9295575\ttotal: 2m 38s\tremaining: 21.3s\n",
      "1234:\tlearn: 0.9295058\ttotal: 2m 38s\tremaining: 21.2s\n",
      "1235:\tlearn: 0.9294894\ttotal: 2m 38s\tremaining: 21s\n",
      "1236:\tlearn: 0.9294627\ttotal: 2m 38s\tremaining: 20.9s\n",
      "1237:\tlearn: 0.9294614\ttotal: 2m 38s\tremaining: 20.8s\n",
      "1238:\tlearn: 0.9294345\ttotal: 2m 38s\tremaining: 20.7s\n",
      "1239:\tlearn: 0.9293866\ttotal: 2m 39s\tremaining: 20.5s\n",
      "1240:\tlearn: 0.9293250\ttotal: 2m 39s\tremaining: 20.4s\n",
      "1241:\tlearn: 0.9292807\ttotal: 2m 39s\tremaining: 20.3s\n",
      "1242:\tlearn: 0.9292402\ttotal: 2m 39s\tremaining: 20.2s\n",
      "1243:\tlearn: 0.9292148\ttotal: 2m 39s\tremaining: 20s\n",
      "1244:\tlearn: 0.9291994\ttotal: 2m 39s\tremaining: 19.9s\n",
      "1245:\tlearn: 0.9291579\ttotal: 2m 39s\tremaining: 19.8s\n",
      "1246:\tlearn: 0.9291249\ttotal: 2m 39s\tremaining: 19.6s\n",
      "1247:\tlearn: 0.9290894\ttotal: 2m 40s\tremaining: 19.5s\n",
      "1248:\tlearn: 0.9290710\ttotal: 2m 40s\tremaining: 19.4s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1249:\tlearn: 0.9290551\ttotal: 2m 40s\tremaining: 19.3s\n",
      "1250:\tlearn: 0.9290306\ttotal: 2m 40s\tremaining: 19.1s\n",
      "1251:\tlearn: 0.9289919\ttotal: 2m 40s\tremaining: 19s\n",
      "1252:\tlearn: 0.9289676\ttotal: 2m 40s\tremaining: 18.9s\n",
      "1253:\tlearn: 0.9289408\ttotal: 2m 41s\tremaining: 18.8s\n",
      "1254:\tlearn: 0.9289137\ttotal: 2m 41s\tremaining: 18.6s\n",
      "1255:\tlearn: 0.9288870\ttotal: 2m 41s\tremaining: 18.5s\n",
      "1256:\tlearn: 0.9288863\ttotal: 2m 41s\tremaining: 18.4s\n",
      "1257:\tlearn: 0.9288566\ttotal: 2m 41s\tremaining: 18.3s\n",
      "1258:\tlearn: 0.9288160\ttotal: 2m 41s\tremaining: 18.1s\n",
      "1259:\tlearn: 0.9287861\ttotal: 2m 41s\tremaining: 18s\n",
      "1260:\tlearn: 0.9287570\ttotal: 2m 42s\tremaining: 17.9s\n",
      "1261:\tlearn: 0.9287341\ttotal: 2m 42s\tremaining: 17.8s\n",
      "1262:\tlearn: 0.9287110\ttotal: 2m 42s\tremaining: 17.6s\n",
      "1263:\tlearn: 0.9286629\ttotal: 2m 42s\tremaining: 17.5s\n",
      "1264:\tlearn: 0.9286010\ttotal: 2m 42s\tremaining: 17.4s\n",
      "1265:\tlearn: 0.9285838\ttotal: 2m 42s\tremaining: 17.3s\n",
      "1266:\tlearn: 0.9285804\ttotal: 2m 43s\tremaining: 17.1s\n",
      "1267:\tlearn: 0.9285488\ttotal: 2m 43s\tremaining: 17s\n",
      "1268:\tlearn: 0.9285471\ttotal: 2m 43s\tremaining: 16.9s\n",
      "1269:\tlearn: 0.9285296\ttotal: 2m 43s\tremaining: 16.7s\n",
      "1270:\tlearn: 0.9285193\ttotal: 2m 43s\tremaining: 16.6s\n",
      "1271:\tlearn: 0.9285151\ttotal: 2m 43s\tremaining: 16.5s\n",
      "1272:\tlearn: 0.9285141\ttotal: 2m 43s\tremaining: 16.4s\n",
      "1273:\tlearn: 0.9285137\ttotal: 2m 43s\tremaining: 16.2s\n",
      "1274:\tlearn: 0.9284806\ttotal: 2m 44s\tremaining: 16.1s\n",
      "1275:\tlearn: 0.9284790\ttotal: 2m 44s\tremaining: 16s\n",
      "1276:\tlearn: 0.9284769\ttotal: 2m 44s\tremaining: 15.8s\n",
      "1277:\tlearn: 0.9284511\ttotal: 2m 44s\tremaining: 15.7s\n",
      "1278:\tlearn: 0.9284131\ttotal: 2m 44s\tremaining: 15.6s\n",
      "1279:\tlearn: 0.9283822\ttotal: 2m 44s\tremaining: 15.4s\n",
      "1280:\tlearn: 0.9283559\ttotal: 2m 44s\tremaining: 15.3s\n",
      "1281:\tlearn: 0.9283553\ttotal: 2m 44s\tremaining: 15.2s\n",
      "1282:\tlearn: 0.9283406\ttotal: 2m 44s\tremaining: 15.1s\n",
      "1283:\tlearn: 0.9283275\ttotal: 2m 45s\tremaining: 14.9s\n",
      "1284:\tlearn: 0.9283056\ttotal: 2m 45s\tremaining: 14.8s\n",
      "1285:\tlearn: 0.9282529\ttotal: 2m 45s\tremaining: 14.7s\n",
      "1286:\tlearn: 0.9282307\ttotal: 2m 45s\tremaining: 14.5s\n",
      "1287:\tlearn: 0.9281919\ttotal: 2m 45s\tremaining: 14.4s\n",
      "1288:\tlearn: 0.9281592\ttotal: 2m 45s\tremaining: 14.3s\n",
      "1289:\tlearn: 0.9281470\ttotal: 2m 45s\tremaining: 14.2s\n",
      "1290:\tlearn: 0.9281355\ttotal: 2m 46s\tremaining: 14s\n",
      "1291:\tlearn: 0.9281189\ttotal: 2m 46s\tremaining: 13.9s\n",
      "1292:\tlearn: 0.9280894\ttotal: 2m 46s\tremaining: 13.8s\n",
      "1293:\tlearn: 0.9280559\ttotal: 2m 46s\tremaining: 13.6s\n",
      "1294:\tlearn: 0.9280489\ttotal: 2m 46s\tremaining: 13.5s\n",
      "1295:\tlearn: 0.9280431\ttotal: 2m 46s\tremaining: 13.4s\n",
      "1296:\tlearn: 0.9280083\ttotal: 2m 46s\tremaining: 13.2s\n",
      "1297:\tlearn: 0.9279863\ttotal: 2m 46s\tremaining: 13.1s\n",
      "1298:\tlearn: 0.9279507\ttotal: 2m 46s\tremaining: 13s\n",
      "1299:\tlearn: 0.9279294\ttotal: 2m 47s\tremaining: 12.9s\n",
      "1300:\tlearn: 0.9279108\ttotal: 2m 47s\tremaining: 12.7s\n",
      "1301:\tlearn: 0.9278860\ttotal: 2m 47s\tremaining: 12.6s\n",
      "1302:\tlearn: 0.9278661\ttotal: 2m 47s\tremaining: 12.5s\n",
      "1303:\tlearn: 0.9278484\ttotal: 2m 47s\tremaining: 12.3s\n",
      "1304:\tlearn: 0.9278254\ttotal: 2m 47s\tremaining: 12.2s\n",
      "1305:\tlearn: 0.9277785\ttotal: 2m 47s\tremaining: 12.1s\n",
      "1306:\tlearn: 0.9277687\ttotal: 2m 48s\tremaining: 12s\n",
      "1307:\tlearn: 0.9277499\ttotal: 2m 48s\tremaining: 11.8s\n",
      "1308:\tlearn: 0.9277098\ttotal: 2m 48s\tremaining: 11.7s\n",
      "1309:\tlearn: 0.9276630\ttotal: 2m 48s\tremaining: 11.6s\n",
      "1310:\tlearn: 0.9276277\ttotal: 2m 48s\tremaining: 11.5s\n",
      "1311:\tlearn: 0.9275975\ttotal: 2m 48s\tremaining: 11.3s\n",
      "1312:\tlearn: 0.9275640\ttotal: 2m 48s\tremaining: 11.2s\n",
      "1313:\tlearn: 0.9275052\ttotal: 2m 48s\tremaining: 11.1s\n",
      "1314:\tlearn: 0.9274754\ttotal: 2m 49s\tremaining: 10.9s\n",
      "1315:\tlearn: 0.9274460\ttotal: 2m 49s\tremaining: 10.8s\n",
      "1316:\tlearn: 0.9274155\ttotal: 2m 49s\tremaining: 10.7s\n",
      "1317:\tlearn: 0.9273739\ttotal: 2m 49s\tremaining: 10.6s\n",
      "1318:\tlearn: 0.9273337\ttotal: 2m 49s\tremaining: 10.4s\n",
      "1319:\tlearn: 0.9273188\ttotal: 2m 49s\tremaining: 10.3s\n",
      "1320:\tlearn: 0.9273021\ttotal: 2m 50s\tremaining: 10.2s\n",
      "1321:\tlearn: 0.9272774\ttotal: 2m 50s\tremaining: 10s\n",
      "1322:\tlearn: 0.9272506\ttotal: 2m 50s\tremaining: 9.91s\n",
      "1323:\tlearn: 0.9272084\ttotal: 2m 50s\tremaining: 9.79s\n",
      "1324:\tlearn: 0.9271605\ttotal: 2m 50s\tremaining: 9.66s\n",
      "1325:\tlearn: 0.9271531\ttotal: 2m 50s\tremaining: 9.53s\n",
      "1326:\tlearn: 0.9271343\ttotal: 2m 50s\tremaining: 9.4s\n",
      "1327:\tlearn: 0.9270985\ttotal: 2m 50s\tremaining: 9.27s\n",
      "1328:\tlearn: 0.9270571\ttotal: 2m 51s\tremaining: 9.14s\n",
      "1329:\tlearn: 0.9270084\ttotal: 2m 51s\tremaining: 9.02s\n",
      "1330:\tlearn: 0.9269885\ttotal: 2m 51s\tremaining: 8.89s\n",
      "1331:\tlearn: 0.9269696\ttotal: 2m 51s\tremaining: 8.76s\n",
      "1332:\tlearn: 0.9269469\ttotal: 2m 51s\tremaining: 8.63s\n",
      "1333:\tlearn: 0.9269217\ttotal: 2m 51s\tremaining: 8.5s\n",
      "1334:\tlearn: 0.9269049\ttotal: 2m 51s\tremaining: 8.38s\n",
      "1335:\tlearn: 0.9268820\ttotal: 2m 51s\tremaining: 8.24s\n",
      "1336:\tlearn: 0.9268679\ttotal: 2m 52s\tremaining: 8.12s\n",
      "1337:\tlearn: 0.9268536\ttotal: 2m 52s\tremaining: 7.99s\n",
      "1338:\tlearn: 0.9268317\ttotal: 2m 52s\tremaining: 7.86s\n",
      "1339:\tlearn: 0.9268159\ttotal: 2m 52s\tremaining: 7.73s\n",
      "1340:\tlearn: 0.9268153\ttotal: 2m 52s\tremaining: 7.6s\n",
      "1341:\tlearn: 0.9267967\ttotal: 2m 52s\tremaining: 7.47s\n",
      "1342:\tlearn: 0.9267772\ttotal: 2m 52s\tremaining: 7.34s\n",
      "1343:\tlearn: 0.9267538\ttotal: 2m 53s\tremaining: 7.22s\n",
      "1344:\tlearn: 0.9267258\ttotal: 2m 53s\tremaining: 7.09s\n",
      "1345:\tlearn: 0.9267053\ttotal: 2m 53s\tremaining: 6.96s\n",
      "1346:\tlearn: 0.9266808\ttotal: 2m 53s\tremaining: 6.83s\n",
      "1347:\tlearn: 0.9266742\ttotal: 2m 53s\tremaining: 6.7s\n",
      "1348:\tlearn: 0.9266525\ttotal: 2m 53s\tremaining: 6.57s\n",
      "1349:\tlearn: 0.9266183\ttotal: 2m 53s\tremaining: 6.44s\n",
      "1350:\tlearn: 0.9266067\ttotal: 2m 53s\tremaining: 6.31s\n",
      "1351:\tlearn: 0.9265737\ttotal: 2m 54s\tremaining: 6.18s\n",
      "1352:\tlearn: 0.9265402\ttotal: 2m 54s\tremaining: 6.06s\n",
      "1353:\tlearn: 0.9265167\ttotal: 2m 54s\tremaining: 5.93s\n",
      "1354:\tlearn: 0.9264907\ttotal: 2m 54s\tremaining: 5.8s\n",
      "1355:\tlearn: 0.9264487\ttotal: 2m 54s\tremaining: 5.67s\n",
      "1356:\tlearn: 0.9264194\ttotal: 2m 54s\tremaining: 5.54s\n",
      "1357:\tlearn: 0.9263995\ttotal: 2m 54s\tremaining: 5.42s\n",
      "1358:\tlearn: 0.9263768\ttotal: 2m 55s\tremaining: 5.29s\n",
      "1359:\tlearn: 0.9263744\ttotal: 2m 55s\tremaining: 5.16s\n",
      "1360:\tlearn: 0.9263493\ttotal: 2m 55s\tremaining: 5.03s\n",
      "1361:\tlearn: 0.9263138\ttotal: 2m 55s\tremaining: 4.9s\n",
      "1362:\tlearn: 0.9262867\ttotal: 2m 55s\tremaining: 4.77s\n",
      "1363:\tlearn: 0.9262368\ttotal: 2m 55s\tremaining: 4.64s\n",
      "1364:\tlearn: 0.9262113\ttotal: 2m 55s\tremaining: 4.51s\n",
      "1365:\tlearn: 0.9262101\ttotal: 2m 55s\tremaining: 4.38s\n",
      "1366:\tlearn: 0.9262094\ttotal: 2m 56s\tremaining: 4.25s\n",
      "1367:\tlearn: 0.9261865\ttotal: 2m 56s\tremaining: 4.12s\n",
      "1368:\tlearn: 0.9261479\ttotal: 2m 56s\tremaining: 3.99s\n",
      "1369:\tlearn: 0.9261040\ttotal: 2m 56s\tremaining: 3.87s\n",
      "1370:\tlearn: 0.9260740\ttotal: 2m 56s\tremaining: 3.74s\n",
      "1371:\tlearn: 0.9260489\ttotal: 2m 56s\tremaining: 3.61s\n",
      "1372:\tlearn: 0.9260142\ttotal: 2m 56s\tremaining: 3.48s\n",
      "1373:\tlearn: 0.9259983\ttotal: 2m 56s\tremaining: 3.35s\n",
      "1374:\tlearn: 0.9259653\ttotal: 2m 57s\tremaining: 3.22s\n",
      "1375:\tlearn: 0.9259227\ttotal: 2m 57s\tremaining: 3.09s\n",
      "1376:\tlearn: 0.9258997\ttotal: 2m 57s\tremaining: 2.96s\n",
      "1377:\tlearn: 0.9258727\ttotal: 2m 57s\tremaining: 2.84s\n",
      "1378:\tlearn: 0.9258543\ttotal: 2m 57s\tremaining: 2.71s\n",
      "1379:\tlearn: 0.9258331\ttotal: 2m 57s\tremaining: 2.58s\n",
      "1380:\tlearn: 0.9257954\ttotal: 2m 58s\tremaining: 2.45s\n",
      "1381:\tlearn: 0.9257947\ttotal: 2m 58s\tremaining: 2.32s\n",
      "1382:\tlearn: 0.9257709\ttotal: 2m 58s\tremaining: 2.19s\n",
      "1383:\tlearn: 0.9257445\ttotal: 2m 58s\tremaining: 2.06s\n",
      "1384:\tlearn: 0.9257408\ttotal: 2m 58s\tremaining: 1.94s\n",
      "1385:\tlearn: 0.9257042\ttotal: 2m 58s\tremaining: 1.81s\n",
      "1386:\tlearn: 0.9256657\ttotal: 2m 58s\tremaining: 1.68s\n",
      "1387:\tlearn: 0.9256274\ttotal: 2m 59s\tremaining: 1.55s\n",
      "1388:\tlearn: 0.9256213\ttotal: 2m 59s\tremaining: 1.42s\n",
      "1389:\tlearn: 0.9255990\ttotal: 2m 59s\tremaining: 1.29s\n",
      "1390:\tlearn: 0.9255718\ttotal: 2m 59s\tremaining: 1.16s\n",
      "1391:\tlearn: 0.9255698\ttotal: 2m 59s\tremaining: 1.03s\n",
      "1392:\tlearn: 0.9255634\ttotal: 2m 59s\tremaining: 904ms\n",
      "1393:\tlearn: 0.9255328\ttotal: 2m 59s\tremaining: 775ms\n",
      "1394:\tlearn: 0.9255304\ttotal: 2m 59s\tremaining: 645ms\n",
      "1395:\tlearn: 0.9255147\ttotal: 3m\tremaining: 516ms\n",
      "1396:\tlearn: 0.9254811\ttotal: 3m\tremaining: 387ms\n",
      "1397:\tlearn: 0.9254351\ttotal: 3m\tremaining: 258ms\n",
      "1398:\tlearn: 0.9254150\ttotal: 3m\tremaining: 129ms\n",
      "1399:\tlearn: 0.9253918\ttotal: 3m\tremaining: 0us\n",
      "0:\tlearn: 7.4153667\ttotal: 99.2ms\tremaining: 2m 18s\n",
      "1:\tlearn: 7.0559227\ttotal: 235ms\tremaining: 2m 44s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2:\tlearn: 6.7129928\ttotal: 361ms\tremaining: 2m 48s\n",
      "3:\tlearn: 6.3872302\ttotal: 457ms\tremaining: 2m 39s\n",
      "4:\tlearn: 6.0784280\ttotal: 550ms\tremaining: 2m 33s\n",
      "5:\tlearn: 5.7851677\ttotal: 671ms\tremaining: 2m 35s\n",
      "6:\tlearn: 5.5072734\ttotal: 784ms\tremaining: 2m 35s\n",
      "7:\tlearn: 5.2447532\ttotal: 894ms\tremaining: 2m 35s\n",
      "8:\tlearn: 4.9949541\ttotal: 1.03s\tremaining: 2m 39s\n",
      "9:\tlearn: 4.7578237\ttotal: 1.11s\tremaining: 2m 35s\n",
      "10:\tlearn: 4.5340715\ttotal: 1.25s\tremaining: 2m 37s\n",
      "11:\tlearn: 4.3208227\ttotal: 1.31s\tremaining: 2m 32s\n",
      "12:\tlearn: 4.1191915\ttotal: 1.44s\tremaining: 2m 33s\n",
      "13:\tlearn: 3.9284647\ttotal: 1.56s\tremaining: 2m 34s\n",
      "14:\tlearn: 3.7470824\ttotal: 1.66s\tremaining: 2m 32s\n",
      "15:\tlearn: 3.5754124\ttotal: 1.76s\tremaining: 2m 32s\n",
      "16:\tlearn: 3.4142355\ttotal: 1.9s\tremaining: 2m 34s\n",
      "17:\tlearn: 3.2615125\ttotal: 2s\tremaining: 2m 33s\n",
      "18:\tlearn: 3.1166899\ttotal: 2.11s\tremaining: 2m 33s\n",
      "19:\tlearn: 2.9803263\ttotal: 2.25s\tremaining: 2m 35s\n",
      "20:\tlearn: 2.8509198\ttotal: 2.35s\tremaining: 2m 34s\n",
      "21:\tlearn: 2.7294544\ttotal: 2.48s\tremaining: 2m 35s\n",
      "22:\tlearn: 2.6143909\ttotal: 2.58s\tremaining: 2m 34s\n",
      "23:\tlearn: 2.5057396\ttotal: 2.69s\tremaining: 2m 34s\n",
      "24:\tlearn: 2.4031617\ttotal: 2.84s\tremaining: 2m 36s\n",
      "25:\tlearn: 2.3067195\ttotal: 3.04s\tremaining: 2m 40s\n",
      "26:\tlearn: 2.2167460\ttotal: 3.19s\tremaining: 2m 42s\n",
      "27:\tlearn: 2.1311593\ttotal: 3.3s\tremaining: 2m 41s\n",
      "28:\tlearn: 2.0507741\ttotal: 3.38s\tremaining: 2m 39s\n",
      "29:\tlearn: 1.9748753\ttotal: 3.58s\tremaining: 2m 43s\n",
      "30:\tlearn: 1.9038127\ttotal: 3.7s\tremaining: 2m 43s\n",
      "31:\tlearn: 1.8375771\ttotal: 3.84s\tremaining: 2m 44s\n",
      "32:\tlearn: 1.7754985\ttotal: 4.01s\tremaining: 2m 46s\n",
      "33:\tlearn: 1.7175820\ttotal: 4.2s\tremaining: 2m 48s\n",
      "34:\tlearn: 1.6636179\ttotal: 4.37s\tremaining: 2m 50s\n",
      "35:\tlearn: 1.6127440\ttotal: 4.53s\tremaining: 2m 51s\n",
      "36:\tlearn: 1.5650835\ttotal: 4.66s\tremaining: 2m 51s\n",
      "37:\tlearn: 1.5205907\ttotal: 4.81s\tremaining: 2m 52s\n",
      "38:\tlearn: 1.4792973\ttotal: 5s\tremaining: 2m 54s\n",
      "39:\tlearn: 1.4408463\ttotal: 5.16s\tremaining: 2m 55s\n",
      "40:\tlearn: 1.4050928\ttotal: 5.29s\tremaining: 2m 55s\n",
      "41:\tlearn: 1.3718929\ttotal: 5.47s\tremaining: 2m 56s\n",
      "42:\tlearn: 1.3412241\ttotal: 5.62s\tremaining: 2m 57s\n",
      "43:\tlearn: 1.3129729\ttotal: 5.75s\tremaining: 2m 57s\n",
      "44:\tlearn: 1.2862231\ttotal: 5.95s\tremaining: 2m 59s\n",
      "45:\tlearn: 1.2616127\ttotal: 6.12s\tremaining: 3m\n",
      "46:\tlearn: 1.2386081\ttotal: 6.32s\tremaining: 3m 1s\n",
      "47:\tlearn: 1.2181338\ttotal: 6.49s\tremaining: 3m 2s\n",
      "48:\tlearn: 1.1990672\ttotal: 6.64s\tremaining: 3m 3s\n",
      "49:\tlearn: 1.1814464\ttotal: 6.83s\tremaining: 3m 4s\n",
      "50:\tlearn: 1.1651275\ttotal: 7.04s\tremaining: 3m 6s\n",
      "51:\tlearn: 1.1499859\ttotal: 7.21s\tremaining: 3m 6s\n",
      "52:\tlearn: 1.1359408\ttotal: 7.34s\tremaining: 3m 6s\n",
      "53:\tlearn: 1.1232314\ttotal: 7.47s\tremaining: 3m 6s\n",
      "54:\tlearn: 1.1117963\ttotal: 7.62s\tremaining: 3m 6s\n",
      "55:\tlearn: 1.1010864\ttotal: 7.76s\tremaining: 3m 6s\n",
      "56:\tlearn: 1.0916859\ttotal: 7.88s\tremaining: 3m 5s\n",
      "57:\tlearn: 1.0827675\ttotal: 8s\tremaining: 3m 5s\n",
      "58:\tlearn: 1.0744360\ttotal: 8.14s\tremaining: 3m 4s\n",
      "59:\tlearn: 1.0669924\ttotal: 8.26s\tremaining: 3m 4s\n",
      "60:\tlearn: 1.0599500\ttotal: 8.45s\tremaining: 3m 5s\n",
      "61:\tlearn: 1.0535950\ttotal: 8.56s\tremaining: 3m 4s\n",
      "62:\tlearn: 1.0477213\ttotal: 8.67s\tremaining: 3m 3s\n",
      "63:\tlearn: 1.0425358\ttotal: 8.78s\tremaining: 3m 3s\n",
      "64:\tlearn: 1.0375848\ttotal: 8.91s\tremaining: 3m 2s\n",
      "65:\tlearn: 1.0330419\ttotal: 9.02s\tremaining: 3m 2s\n",
      "66:\tlearn: 1.0288181\ttotal: 9.16s\tremaining: 3m 2s\n",
      "67:\tlearn: 1.0251151\ttotal: 9.3s\tremaining: 3m 2s\n",
      "68:\tlearn: 1.0215233\ttotal: 9.42s\tremaining: 3m 1s\n",
      "69:\tlearn: 1.0183654\ttotal: 9.56s\tremaining: 3m 1s\n",
      "70:\tlearn: 1.0152030\ttotal: 9.7s\tremaining: 3m 1s\n",
      "71:\tlearn: 1.0124611\ttotal: 9.82s\tremaining: 3m 1s\n",
      "72:\tlearn: 1.0100562\ttotal: 9.95s\tremaining: 3m\n",
      "73:\tlearn: 1.0076658\ttotal: 10.1s\tremaining: 3m\n",
      "74:\tlearn: 1.0055645\ttotal: 10.2s\tremaining: 2m 59s\n",
      "75:\tlearn: 1.0037028\ttotal: 10.3s\tremaining: 2m 59s\n",
      "76:\tlearn: 1.0017729\ttotal: 10.4s\tremaining: 2m 59s\n",
      "77:\tlearn: 1.0000404\ttotal: 10.6s\tremaining: 2m 58s\n",
      "78:\tlearn: 0.9984303\ttotal: 10.7s\tremaining: 2m 58s\n",
      "79:\tlearn: 0.9968802\ttotal: 10.8s\tremaining: 2m 58s\n",
      "80:\tlearn: 0.9955470\ttotal: 10.9s\tremaining: 2m 57s\n",
      "81:\tlearn: 0.9943358\ttotal: 11s\tremaining: 2m 57s\n",
      "82:\tlearn: 0.9932575\ttotal: 11.2s\tremaining: 2m 57s\n",
      "83:\tlearn: 0.9921718\ttotal: 11.3s\tremaining: 2m 56s\n",
      "84:\tlearn: 0.9910685\ttotal: 11.4s\tremaining: 2m 56s\n",
      "85:\tlearn: 0.9901317\ttotal: 11.5s\tremaining: 2m 56s\n",
      "86:\tlearn: 0.9891976\ttotal: 11.7s\tremaining: 2m 56s\n",
      "87:\tlearn: 0.9883727\ttotal: 11.8s\tremaining: 2m 55s\n",
      "88:\tlearn: 0.9875733\ttotal: 11.9s\tremaining: 2m 55s\n",
      "89:\tlearn: 0.9868397\ttotal: 12s\tremaining: 2m 55s\n",
      "90:\tlearn: 0.9860120\ttotal: 12.2s\tremaining: 2m 55s\n",
      "91:\tlearn: 0.9854231\ttotal: 12.3s\tremaining: 2m 54s\n",
      "92:\tlearn: 0.9847694\ttotal: 12.4s\tremaining: 2m 54s\n",
      "93:\tlearn: 0.9841153\ttotal: 12.5s\tremaining: 2m 53s\n",
      "94:\tlearn: 0.9835195\ttotal: 12.6s\tremaining: 2m 53s\n",
      "95:\tlearn: 0.9829718\ttotal: 12.8s\tremaining: 2m 53s\n",
      "96:\tlearn: 0.9825221\ttotal: 12.9s\tremaining: 2m 52s\n",
      "97:\tlearn: 0.9819879\ttotal: 13s\tremaining: 2m 52s\n",
      "98:\tlearn: 0.9815486\ttotal: 13.1s\tremaining: 2m 52s\n",
      "99:\tlearn: 0.9811086\ttotal: 13.2s\tremaining: 2m 51s\n",
      "100:\tlearn: 0.9805966\ttotal: 13.4s\tremaining: 2m 51s\n",
      "101:\tlearn: 0.9802170\ttotal: 13.5s\tremaining: 2m 51s\n",
      "102:\tlearn: 0.9798440\ttotal: 13.6s\tremaining: 2m 50s\n",
      "103:\tlearn: 0.9795164\ttotal: 13.7s\tremaining: 2m 50s\n",
      "104:\tlearn: 0.9792070\ttotal: 13.8s\tremaining: 2m 50s\n",
      "105:\tlearn: 0.9789153\ttotal: 13.9s\tremaining: 2m 49s\n",
      "106:\tlearn: 0.9785615\ttotal: 14s\tremaining: 2m 49s\n",
      "107:\tlearn: 0.9782454\ttotal: 14.2s\tremaining: 2m 49s\n",
      "108:\tlearn: 0.9779681\ttotal: 14.3s\tremaining: 2m 49s\n",
      "109:\tlearn: 0.9776983\ttotal: 14.4s\tremaining: 2m 48s\n",
      "110:\tlearn: 0.9774467\ttotal: 14.5s\tremaining: 2m 48s\n",
      "111:\tlearn: 0.9771948\ttotal: 14.6s\tremaining: 2m 48s\n",
      "112:\tlearn: 0.9769011\ttotal: 14.7s\tremaining: 2m 47s\n",
      "113:\tlearn: 0.9766970\ttotal: 14.9s\tremaining: 2m 47s\n",
      "114:\tlearn: 0.9764135\ttotal: 15s\tremaining: 2m 47s\n",
      "115:\tlearn: 0.9762133\ttotal: 15.1s\tremaining: 2m 47s\n",
      "116:\tlearn: 0.9759895\ttotal: 15.2s\tremaining: 2m 47s\n",
      "117:\tlearn: 0.9758078\ttotal: 15.3s\tremaining: 2m 46s\n",
      "118:\tlearn: 0.9755545\ttotal: 15.5s\tremaining: 2m 46s\n",
      "119:\tlearn: 0.9753226\ttotal: 15.6s\tremaining: 2m 46s\n",
      "120:\tlearn: 0.9751664\ttotal: 15.7s\tremaining: 2m 45s\n",
      "121:\tlearn: 0.9749941\ttotal: 15.8s\tremaining: 2m 45s\n",
      "122:\tlearn: 0.9748200\ttotal: 15.9s\tremaining: 2m 45s\n",
      "123:\tlearn: 0.9745801\ttotal: 16s\tremaining: 2m 44s\n",
      "124:\tlearn: 0.9743879\ttotal: 16.2s\tremaining: 2m 44s\n",
      "125:\tlearn: 0.9742552\ttotal: 16.3s\tremaining: 2m 44s\n",
      "126:\tlearn: 0.9740881\ttotal: 16.4s\tremaining: 2m 44s\n",
      "127:\tlearn: 0.9739850\ttotal: 16.5s\tremaining: 2m 43s\n",
      "128:\tlearn: 0.9738099\ttotal: 16.6s\tremaining: 2m 43s\n",
      "129:\tlearn: 0.9736457\ttotal: 16.7s\tremaining: 2m 43s\n",
      "130:\tlearn: 0.9735260\ttotal: 16.8s\tremaining: 2m 42s\n",
      "131:\tlearn: 0.9734066\ttotal: 16.9s\tremaining: 2m 42s\n",
      "132:\tlearn: 0.9732438\ttotal: 17.1s\tremaining: 2m 42s\n",
      "133:\tlearn: 0.9731050\ttotal: 17.2s\tremaining: 2m 42s\n",
      "134:\tlearn: 0.9729811\ttotal: 17.3s\tremaining: 2m 41s\n",
      "135:\tlearn: 0.9728590\ttotal: 17.4s\tremaining: 2m 41s\n",
      "136:\tlearn: 0.9727850\ttotal: 17.5s\tremaining: 2m 41s\n",
      "137:\tlearn: 0.9726434\ttotal: 17.6s\tremaining: 2m 40s\n",
      "138:\tlearn: 0.9724688\ttotal: 17.7s\tremaining: 2m 40s\n",
      "139:\tlearn: 0.9723541\ttotal: 17.8s\tremaining: 2m 40s\n",
      "140:\tlearn: 0.9722203\ttotal: 17.9s\tremaining: 2m 40s\n",
      "141:\tlearn: 0.9720764\ttotal: 18.1s\tremaining: 2m 40s\n",
      "142:\tlearn: 0.9719582\ttotal: 18.2s\tremaining: 2m 40s\n",
      "143:\tlearn: 0.9718295\ttotal: 18.3s\tremaining: 2m 39s\n",
      "144:\tlearn: 0.9717271\ttotal: 18.5s\tremaining: 2m 39s\n",
      "145:\tlearn: 0.9716147\ttotal: 18.6s\tremaining: 2m 39s\n",
      "146:\tlearn: 0.9715215\ttotal: 18.7s\tremaining: 2m 39s\n",
      "147:\tlearn: 0.9714152\ttotal: 18.8s\tremaining: 2m 38s\n",
      "148:\tlearn: 0.9712924\ttotal: 18.9s\tremaining: 2m 38s\n",
      "149:\tlearn: 0.9712076\ttotal: 19s\tremaining: 2m 38s\n",
      "150:\tlearn: 0.9710861\ttotal: 19.1s\tremaining: 2m 38s\n",
      "151:\tlearn: 0.9709667\ttotal: 19.3s\tremaining: 2m 38s\n",
      "152:\tlearn: 0.9708285\ttotal: 19.4s\tremaining: 2m 38s\n",
      "153:\tlearn: 0.9707497\ttotal: 19.5s\tremaining: 2m 37s\n",
      "154:\tlearn: 0.9706677\ttotal: 19.6s\tremaining: 2m 37s\n",
      "155:\tlearn: 0.9705771\ttotal: 19.8s\tremaining: 2m 37s\n",
      "156:\tlearn: 0.9704510\ttotal: 19.9s\tremaining: 2m 37s\n",
      "157:\tlearn: 0.9703349\ttotal: 20s\tremaining: 2m 37s\n",
      "158:\tlearn: 0.9702339\ttotal: 20.1s\tremaining: 2m 37s\n",
      "159:\tlearn: 0.9700861\ttotal: 20.3s\tremaining: 2m 37s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "160:\tlearn: 0.9699792\ttotal: 20.4s\tremaining: 2m 36s\n",
      "161:\tlearn: 0.9698973\ttotal: 20.5s\tremaining: 2m 36s\n",
      "162:\tlearn: 0.9698107\ttotal: 20.6s\tremaining: 2m 36s\n",
      "163:\tlearn: 0.9697017\ttotal: 20.8s\tremaining: 2m 36s\n",
      "164:\tlearn: 0.9696011\ttotal: 20.8s\tremaining: 2m 36s\n",
      "165:\tlearn: 0.9695443\ttotal: 21s\tremaining: 2m 35s\n",
      "166:\tlearn: 0.9694685\ttotal: 21.1s\tremaining: 2m 35s\n",
      "167:\tlearn: 0.9693597\ttotal: 21.2s\tremaining: 2m 35s\n",
      "168:\tlearn: 0.9692571\ttotal: 21.4s\tremaining: 2m 35s\n",
      "169:\tlearn: 0.9691636\ttotal: 21.5s\tremaining: 2m 35s\n",
      "170:\tlearn: 0.9690580\ttotal: 21.7s\tremaining: 2m 36s\n",
      "171:\tlearn: 0.9689587\ttotal: 21.8s\tremaining: 2m 35s\n",
      "172:\tlearn: 0.9689059\ttotal: 21.9s\tremaining: 2m 35s\n",
      "173:\tlearn: 0.9688108\ttotal: 22.1s\tremaining: 2m 35s\n",
      "174:\tlearn: 0.9687037\ttotal: 22.3s\tremaining: 2m 35s\n",
      "175:\tlearn: 0.9686331\ttotal: 22.4s\tremaining: 2m 35s\n",
      "176:\tlearn: 0.9685348\ttotal: 22.6s\tremaining: 2m 36s\n",
      "177:\tlearn: 0.9684680\ttotal: 22.8s\tremaining: 2m 36s\n",
      "178:\tlearn: 0.9684145\ttotal: 22.9s\tremaining: 2m 36s\n",
      "179:\tlearn: 0.9683613\ttotal: 23.1s\tremaining: 2m 36s\n",
      "180:\tlearn: 0.9682649\ttotal: 23.2s\tremaining: 2m 36s\n",
      "181:\tlearn: 0.9681858\ttotal: 23.4s\tremaining: 2m 36s\n",
      "182:\tlearn: 0.9681230\ttotal: 23.6s\tremaining: 2m 36s\n",
      "183:\tlearn: 0.9680153\ttotal: 23.8s\tremaining: 2m 37s\n",
      "184:\tlearn: 0.9679463\ttotal: 23.9s\tremaining: 2m 37s\n",
      "185:\tlearn: 0.9678608\ttotal: 24s\tremaining: 2m 36s\n",
      "186:\tlearn: 0.9678134\ttotal: 24.2s\tremaining: 2m 36s\n",
      "187:\tlearn: 0.9677667\ttotal: 24.3s\tremaining: 2m 36s\n",
      "188:\tlearn: 0.9676797\ttotal: 24.5s\tremaining: 2m 36s\n",
      "189:\tlearn: 0.9676243\ttotal: 24.6s\tremaining: 2m 36s\n",
      "190:\tlearn: 0.9675362\ttotal: 24.8s\tremaining: 2m 36s\n",
      "191:\tlearn: 0.9674743\ttotal: 24.9s\tremaining: 2m 36s\n",
      "192:\tlearn: 0.9674030\ttotal: 25s\tremaining: 2m 36s\n",
      "193:\tlearn: 0.9673446\ttotal: 25.1s\tremaining: 2m 36s\n",
      "194:\tlearn: 0.9672808\ttotal: 25.2s\tremaining: 2m 35s\n",
      "195:\tlearn: 0.9672407\ttotal: 25.3s\tremaining: 2m 35s\n",
      "196:\tlearn: 0.9671582\ttotal: 25.5s\tremaining: 2m 35s\n",
      "197:\tlearn: 0.9670892\ttotal: 25.7s\tremaining: 2m 35s\n",
      "198:\tlearn: 0.9669965\ttotal: 25.8s\tremaining: 2m 35s\n",
      "199:\tlearn: 0.9669487\ttotal: 25.9s\tremaining: 2m 35s\n",
      "200:\tlearn: 0.9668217\ttotal: 26.1s\tremaining: 2m 35s\n",
      "201:\tlearn: 0.9667546\ttotal: 26.2s\tremaining: 2m 35s\n",
      "202:\tlearn: 0.9666679\ttotal: 26.4s\tremaining: 2m 35s\n",
      "203:\tlearn: 0.9666081\ttotal: 26.6s\tremaining: 2m 35s\n",
      "204:\tlearn: 0.9665484\ttotal: 26.7s\tremaining: 2m 35s\n",
      "205:\tlearn: 0.9664412\ttotal: 26.8s\tremaining: 2m 35s\n",
      "206:\tlearn: 0.9663772\ttotal: 27s\tremaining: 2m 35s\n",
      "207:\tlearn: 0.9662925\ttotal: 27.2s\tremaining: 2m 35s\n",
      "208:\tlearn: 0.9662343\ttotal: 27.3s\tremaining: 2m 35s\n",
      "209:\tlearn: 0.9661685\ttotal: 27.4s\tremaining: 2m 35s\n",
      "210:\tlearn: 0.9661215\ttotal: 27.6s\tremaining: 2m 35s\n",
      "211:\tlearn: 0.9660575\ttotal: 27.7s\tremaining: 2m 35s\n",
      "212:\tlearn: 0.9659892\ttotal: 27.8s\tremaining: 2m 34s\n",
      "213:\tlearn: 0.9659779\ttotal: 27.9s\tremaining: 2m 34s\n",
      "214:\tlearn: 0.9658937\ttotal: 28.1s\tremaining: 2m 34s\n",
      "215:\tlearn: 0.9658025\ttotal: 28.2s\tremaining: 2m 34s\n",
      "216:\tlearn: 0.9657538\ttotal: 28.3s\tremaining: 2m 34s\n",
      "217:\tlearn: 0.9656988\ttotal: 28.4s\tremaining: 2m 33s\n",
      "218:\tlearn: 0.9656062\ttotal: 28.5s\tremaining: 2m 33s\n",
      "219:\tlearn: 0.9655378\ttotal: 28.6s\tremaining: 2m 33s\n",
      "220:\tlearn: 0.9654739\ttotal: 28.7s\tremaining: 2m 33s\n",
      "221:\tlearn: 0.9654045\ttotal: 28.8s\tremaining: 2m 33s\n",
      "222:\tlearn: 0.9653498\ttotal: 28.9s\tremaining: 2m 32s\n",
      "223:\tlearn: 0.9653070\ttotal: 29s\tremaining: 2m 32s\n",
      "224:\tlearn: 0.9652757\ttotal: 29.1s\tremaining: 2m 32s\n",
      "225:\tlearn: 0.9652433\ttotal: 29.2s\tremaining: 2m 31s\n",
      "226:\tlearn: 0.9652064\ttotal: 29.3s\tremaining: 2m 31s\n",
      "227:\tlearn: 0.9651425\ttotal: 29.5s\tremaining: 2m 31s\n",
      "228:\tlearn: 0.9650566\ttotal: 29.6s\tremaining: 2m 31s\n",
      "229:\tlearn: 0.9649837\ttotal: 29.7s\tremaining: 2m 31s\n",
      "230:\tlearn: 0.9649021\ttotal: 29.9s\tremaining: 2m 31s\n",
      "231:\tlearn: 0.9648574\ttotal: 30s\tremaining: 2m 30s\n",
      "232:\tlearn: 0.9647666\ttotal: 30.1s\tremaining: 2m 30s\n",
      "233:\tlearn: 0.9647272\ttotal: 30.2s\tremaining: 2m 30s\n",
      "234:\tlearn: 0.9646669\ttotal: 30.3s\tremaining: 2m 30s\n",
      "235:\tlearn: 0.9646156\ttotal: 30.4s\tremaining: 2m 30s\n",
      "236:\tlearn: 0.9645566\ttotal: 30.6s\tremaining: 2m 29s\n",
      "237:\tlearn: 0.9645050\ttotal: 30.7s\tremaining: 2m 29s\n",
      "238:\tlearn: 0.9644127\ttotal: 30.8s\tremaining: 2m 29s\n",
      "239:\tlearn: 0.9642960\ttotal: 31s\tremaining: 2m 29s\n",
      "240:\tlearn: 0.9642616\ttotal: 31.1s\tremaining: 2m 29s\n",
      "241:\tlearn: 0.9641771\ttotal: 31.2s\tremaining: 2m 29s\n",
      "242:\tlearn: 0.9641161\ttotal: 31.3s\tremaining: 2m 28s\n",
      "243:\tlearn: 0.9640143\ttotal: 31.4s\tremaining: 2m 28s\n",
      "244:\tlearn: 0.9639495\ttotal: 31.6s\tremaining: 2m 28s\n",
      "245:\tlearn: 0.9638891\ttotal: 31.7s\tremaining: 2m 28s\n",
      "246:\tlearn: 0.9638302\ttotal: 31.8s\tremaining: 2m 28s\n",
      "247:\tlearn: 0.9637980\ttotal: 31.9s\tremaining: 2m 28s\n",
      "248:\tlearn: 0.9637414\ttotal: 32s\tremaining: 2m 28s\n",
      "249:\tlearn: 0.9636954\ttotal: 32.2s\tremaining: 2m 27s\n",
      "250:\tlearn: 0.9636473\ttotal: 32.2s\tremaining: 2m 27s\n",
      "251:\tlearn: 0.9635988\ttotal: 32.3s\tremaining: 2m 27s\n",
      "252:\tlearn: 0.9635440\ttotal: 32.5s\tremaining: 2m 27s\n",
      "253:\tlearn: 0.9634819\ttotal: 32.6s\tremaining: 2m 26s\n",
      "254:\tlearn: 0.9634433\ttotal: 32.7s\tremaining: 2m 26s\n",
      "255:\tlearn: 0.9633873\ttotal: 32.8s\tremaining: 2m 26s\n",
      "256:\tlearn: 0.9633140\ttotal: 32.9s\tremaining: 2m 26s\n",
      "257:\tlearn: 0.9632573\ttotal: 33.1s\tremaining: 2m 26s\n",
      "258:\tlearn: 0.9632079\ttotal: 33.2s\tremaining: 2m 26s\n",
      "259:\tlearn: 0.9631452\ttotal: 33.3s\tremaining: 2m 25s\n",
      "260:\tlearn: 0.9630933\ttotal: 33.4s\tremaining: 2m 25s\n",
      "261:\tlearn: 0.9630341\ttotal: 33.5s\tremaining: 2m 25s\n",
      "262:\tlearn: 0.9629324\ttotal: 33.6s\tremaining: 2m 25s\n",
      "263:\tlearn: 0.9628808\ttotal: 33.7s\tremaining: 2m 25s\n",
      "264:\tlearn: 0.9628121\ttotal: 33.8s\tremaining: 2m 24s\n",
      "265:\tlearn: 0.9627622\ttotal: 33.9s\tremaining: 2m 24s\n",
      "266:\tlearn: 0.9627073\ttotal: 34.1s\tremaining: 2m 24s\n",
      "267:\tlearn: 0.9626356\ttotal: 34.2s\tremaining: 2m 24s\n",
      "268:\tlearn: 0.9626082\ttotal: 34.3s\tremaining: 2m 24s\n",
      "269:\tlearn: 0.9625742\ttotal: 34.4s\tremaining: 2m 23s\n",
      "270:\tlearn: 0.9625446\ttotal: 34.5s\tremaining: 2m 23s\n",
      "271:\tlearn: 0.9624827\ttotal: 34.6s\tremaining: 2m 23s\n",
      "272:\tlearn: 0.9624496\ttotal: 34.7s\tremaining: 2m 23s\n",
      "273:\tlearn: 0.9623849\ttotal: 34.8s\tremaining: 2m 23s\n",
      "274:\tlearn: 0.9623203\ttotal: 35s\tremaining: 2m 23s\n",
      "275:\tlearn: 0.9622630\ttotal: 35.1s\tremaining: 2m 22s\n",
      "276:\tlearn: 0.9622520\ttotal: 35.2s\tremaining: 2m 22s\n",
      "277:\tlearn: 0.9622022\ttotal: 35.3s\tremaining: 2m 22s\n",
      "278:\tlearn: 0.9621366\ttotal: 35.4s\tremaining: 2m 22s\n",
      "279:\tlearn: 0.9620762\ttotal: 35.5s\tremaining: 2m 21s\n",
      "280:\tlearn: 0.9620065\ttotal: 35.6s\tremaining: 2m 21s\n",
      "281:\tlearn: 0.9619736\ttotal: 35.7s\tremaining: 2m 21s\n",
      "282:\tlearn: 0.9619559\ttotal: 35.8s\tremaining: 2m 21s\n",
      "283:\tlearn: 0.9618909\ttotal: 35.9s\tremaining: 2m 21s\n",
      "284:\tlearn: 0.9618442\ttotal: 36.1s\tremaining: 2m 21s\n",
      "285:\tlearn: 0.9617974\ttotal: 36.2s\tremaining: 2m 21s\n",
      "286:\tlearn: 0.9617342\ttotal: 36.4s\tremaining: 2m 21s\n",
      "287:\tlearn: 0.9616905\ttotal: 36.5s\tremaining: 2m 20s\n",
      "288:\tlearn: 0.9616389\ttotal: 36.6s\tremaining: 2m 20s\n",
      "289:\tlearn: 0.9615820\ttotal: 36.7s\tremaining: 2m 20s\n",
      "290:\tlearn: 0.9615408\ttotal: 36.9s\tremaining: 2m 20s\n",
      "291:\tlearn: 0.9614910\ttotal: 37s\tremaining: 2m 20s\n",
      "292:\tlearn: 0.9614328\ttotal: 37.2s\tremaining: 2m 20s\n",
      "293:\tlearn: 0.9613862\ttotal: 37.3s\tremaining: 2m 20s\n",
      "294:\tlearn: 0.9613415\ttotal: 37.4s\tremaining: 2m 20s\n",
      "295:\tlearn: 0.9613030\ttotal: 37.6s\tremaining: 2m 20s\n",
      "296:\tlearn: 0.9612192\ttotal: 37.8s\tremaining: 2m 20s\n",
      "297:\tlearn: 0.9611690\ttotal: 37.9s\tremaining: 2m 20s\n",
      "298:\tlearn: 0.9610972\ttotal: 38.1s\tremaining: 2m 20s\n",
      "299:\tlearn: 0.9610795\ttotal: 38.2s\tremaining: 2m 20s\n",
      "300:\tlearn: 0.9610084\ttotal: 38.3s\tremaining: 2m 20s\n",
      "301:\tlearn: 0.9609516\ttotal: 38.5s\tremaining: 2m 19s\n",
      "302:\tlearn: 0.9608821\ttotal: 38.6s\tremaining: 2m 19s\n",
      "303:\tlearn: 0.9608370\ttotal: 38.7s\tremaining: 2m 19s\n",
      "304:\tlearn: 0.9608170\ttotal: 38.8s\tremaining: 2m 19s\n",
      "305:\tlearn: 0.9607949\ttotal: 39s\tremaining: 2m 19s\n",
      "306:\tlearn: 0.9607470\ttotal: 39.1s\tremaining: 2m 19s\n",
      "307:\tlearn: 0.9607148\ttotal: 39.3s\tremaining: 2m 19s\n",
      "308:\tlearn: 0.9606793\ttotal: 39.3s\tremaining: 2m 18s\n",
      "309:\tlearn: 0.9606420\ttotal: 39.5s\tremaining: 2m 18s\n",
      "310:\tlearn: 0.9605938\ttotal: 39.6s\tremaining: 2m 18s\n",
      "311:\tlearn: 0.9605288\ttotal: 39.8s\tremaining: 2m 18s\n",
      "312:\tlearn: 0.9604979\ttotal: 39.9s\tremaining: 2m 18s\n",
      "313:\tlearn: 0.9604558\ttotal: 40.1s\tremaining: 2m 18s\n",
      "314:\tlearn: 0.9604052\ttotal: 40.2s\tremaining: 2m 18s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "315:\tlearn: 0.9603285\ttotal: 40.3s\tremaining: 2m 18s\n",
      "316:\tlearn: 0.9603020\ttotal: 40.4s\tremaining: 2m 18s\n",
      "317:\tlearn: 0.9602846\ttotal: 40.6s\tremaining: 2m 18s\n",
      "318:\tlearn: 0.9602236\ttotal: 40.8s\tremaining: 2m 18s\n",
      "319:\tlearn: 0.9601915\ttotal: 41s\tremaining: 2m 18s\n",
      "320:\tlearn: 0.9601703\ttotal: 41.1s\tremaining: 2m 17s\n",
      "321:\tlearn: 0.9600822\ttotal: 41.2s\tremaining: 2m 17s\n",
      "322:\tlearn: 0.9600650\ttotal: 41.4s\tremaining: 2m 17s\n",
      "323:\tlearn: 0.9599980\ttotal: 41.5s\tremaining: 2m 17s\n",
      "324:\tlearn: 0.9599213\ttotal: 41.7s\tremaining: 2m 17s\n",
      "325:\tlearn: 0.9598603\ttotal: 41.8s\tremaining: 2m 17s\n",
      "326:\tlearn: 0.9598345\ttotal: 41.9s\tremaining: 2m 17s\n",
      "327:\tlearn: 0.9598152\ttotal: 42s\tremaining: 2m 17s\n",
      "328:\tlearn: 0.9597856\ttotal: 42.1s\tremaining: 2m 17s\n",
      "329:\tlearn: 0.9597379\ttotal: 42.2s\tremaining: 2m 16s\n",
      "330:\tlearn: 0.9596850\ttotal: 42.4s\tremaining: 2m 16s\n",
      "331:\tlearn: 0.9596453\ttotal: 42.6s\tremaining: 2m 16s\n",
      "332:\tlearn: 0.9595991\ttotal: 42.7s\tremaining: 2m 16s\n",
      "333:\tlearn: 0.9595358\ttotal: 42.9s\tremaining: 2m 16s\n",
      "334:\tlearn: 0.9595086\ttotal: 43.1s\tremaining: 2m 16s\n",
      "335:\tlearn: 0.9594513\ttotal: 43.2s\tremaining: 2m 16s\n",
      "336:\tlearn: 0.9593967\ttotal: 43.4s\tremaining: 2m 16s\n",
      "337:\tlearn: 0.9593640\ttotal: 43.4s\tremaining: 2m 16s\n",
      "338:\tlearn: 0.9593137\ttotal: 43.5s\tremaining: 2m 16s\n",
      "339:\tlearn: 0.9593047\ttotal: 43.7s\tremaining: 2m 16s\n",
      "340:\tlearn: 0.9592543\ttotal: 43.8s\tremaining: 2m 15s\n",
      "341:\tlearn: 0.9591987\ttotal: 43.9s\tremaining: 2m 15s\n",
      "342:\tlearn: 0.9591451\ttotal: 44s\tremaining: 2m 15s\n",
      "343:\tlearn: 0.9590977\ttotal: 44.2s\tremaining: 2m 15s\n",
      "344:\tlearn: 0.9590774\ttotal: 44.3s\tremaining: 2m 15s\n",
      "345:\tlearn: 0.9590296\ttotal: 44.3s\tremaining: 2m 15s\n",
      "346:\tlearn: 0.9590097\ttotal: 44.4s\tremaining: 2m 14s\n",
      "347:\tlearn: 0.9589731\ttotal: 44.6s\tremaining: 2m 14s\n",
      "348:\tlearn: 0.9589286\ttotal: 44.7s\tremaining: 2m 14s\n",
      "349:\tlearn: 0.9589258\ttotal: 44.8s\tremaining: 2m 14s\n",
      "350:\tlearn: 0.9588642\ttotal: 44.9s\tremaining: 2m 14s\n",
      "351:\tlearn: 0.9588238\ttotal: 45s\tremaining: 2m 13s\n",
      "352:\tlearn: 0.9587992\ttotal: 45.1s\tremaining: 2m 13s\n",
      "353:\tlearn: 0.9587710\ttotal: 45.2s\tremaining: 2m 13s\n",
      "354:\tlearn: 0.9587060\ttotal: 45.3s\tremaining: 2m 13s\n",
      "355:\tlearn: 0.9586384\ttotal: 45.4s\tremaining: 2m 13s\n",
      "356:\tlearn: 0.9586064\ttotal: 45.5s\tremaining: 2m 12s\n",
      "357:\tlearn: 0.9585892\ttotal: 45.6s\tremaining: 2m 12s\n",
      "358:\tlearn: 0.9585636\ttotal: 45.7s\tremaining: 2m 12s\n",
      "359:\tlearn: 0.9585371\ttotal: 45.8s\tremaining: 2m 12s\n",
      "360:\tlearn: 0.9585272\ttotal: 45.9s\tremaining: 2m 12s\n",
      "361:\tlearn: 0.9584515\ttotal: 46s\tremaining: 2m 12s\n",
      "362:\tlearn: 0.9583960\ttotal: 46.2s\tremaining: 2m 11s\n",
      "363:\tlearn: 0.9583242\ttotal: 46.3s\tremaining: 2m 11s\n",
      "364:\tlearn: 0.9582672\ttotal: 46.4s\tremaining: 2m 11s\n",
      "365:\tlearn: 0.9582138\ttotal: 46.5s\tremaining: 2m 11s\n",
      "366:\tlearn: 0.9581798\ttotal: 46.6s\tremaining: 2m 11s\n",
      "367:\tlearn: 0.9581407\ttotal: 46.7s\tremaining: 2m 11s\n",
      "368:\tlearn: 0.9581083\ttotal: 46.9s\tremaining: 2m 10s\n",
      "369:\tlearn: 0.9580743\ttotal: 47s\tremaining: 2m 10s\n",
      "370:\tlearn: 0.9580172\ttotal: 47.1s\tremaining: 2m 10s\n",
      "371:\tlearn: 0.9579747\ttotal: 47.2s\tremaining: 2m 10s\n",
      "372:\tlearn: 0.9579492\ttotal: 47.3s\tremaining: 2m 10s\n",
      "373:\tlearn: 0.9579064\ttotal: 47.4s\tremaining: 2m 10s\n",
      "374:\tlearn: 0.9578976\ttotal: 47.5s\tremaining: 2m 9s\n",
      "375:\tlearn: 0.9578952\ttotal: 47.6s\tremaining: 2m 9s\n",
      "376:\tlearn: 0.9578353\ttotal: 47.7s\tremaining: 2m 9s\n",
      "377:\tlearn: 0.9578058\ttotal: 47.8s\tremaining: 2m 9s\n",
      "378:\tlearn: 0.9577541\ttotal: 47.9s\tremaining: 2m 9s\n",
      "379:\tlearn: 0.9577167\ttotal: 48s\tremaining: 2m 8s\n",
      "380:\tlearn: 0.9576751\ttotal: 48.1s\tremaining: 2m 8s\n",
      "381:\tlearn: 0.9576294\ttotal: 48.2s\tremaining: 2m 8s\n",
      "382:\tlearn: 0.9575783\ttotal: 48.4s\tremaining: 2m 8s\n",
      "383:\tlearn: 0.9575248\ttotal: 48.5s\tremaining: 2m 8s\n",
      "384:\tlearn: 0.9574949\ttotal: 48.6s\tremaining: 2m 8s\n",
      "385:\tlearn: 0.9574630\ttotal: 48.7s\tremaining: 2m 7s\n",
      "386:\tlearn: 0.9574312\ttotal: 48.8s\tremaining: 2m 7s\n",
      "387:\tlearn: 0.9573940\ttotal: 48.9s\tremaining: 2m 7s\n",
      "388:\tlearn: 0.9573596\ttotal: 49s\tremaining: 2m 7s\n",
      "389:\tlearn: 0.9573096\ttotal: 49.2s\tremaining: 2m 7s\n",
      "390:\tlearn: 0.9572739\ttotal: 49.3s\tremaining: 2m 7s\n",
      "391:\tlearn: 0.9572282\ttotal: 49.4s\tremaining: 2m 7s\n",
      "392:\tlearn: 0.9571701\ttotal: 49.5s\tremaining: 2m 6s\n",
      "393:\tlearn: 0.9571117\ttotal: 49.6s\tremaining: 2m 6s\n",
      "394:\tlearn: 0.9571061\ttotal: 49.7s\tremaining: 2m 6s\n",
      "395:\tlearn: 0.9570657\ttotal: 49.8s\tremaining: 2m 6s\n",
      "396:\tlearn: 0.9570353\ttotal: 49.9s\tremaining: 2m 6s\n",
      "397:\tlearn: 0.9569967\ttotal: 50s\tremaining: 2m 5s\n",
      "398:\tlearn: 0.9569670\ttotal: 50.2s\tremaining: 2m 5s\n",
      "399:\tlearn: 0.9569386\ttotal: 50.3s\tremaining: 2m 5s\n",
      "400:\tlearn: 0.9569153\ttotal: 50.4s\tremaining: 2m 5s\n",
      "401:\tlearn: 0.9568744\ttotal: 50.5s\tremaining: 2m 5s\n",
      "402:\tlearn: 0.9568278\ttotal: 50.6s\tremaining: 2m 5s\n",
      "403:\tlearn: 0.9567828\ttotal: 50.7s\tremaining: 2m 5s\n",
      "404:\tlearn: 0.9567400\ttotal: 50.8s\tremaining: 2m 4s\n",
      "405:\tlearn: 0.9567355\ttotal: 50.9s\tremaining: 2m 4s\n",
      "406:\tlearn: 0.9567079\ttotal: 51s\tremaining: 2m 4s\n",
      "407:\tlearn: 0.9566790\ttotal: 51.1s\tremaining: 2m 4s\n",
      "408:\tlearn: 0.9566099\ttotal: 51.3s\tremaining: 2m 4s\n",
      "409:\tlearn: 0.9565632\ttotal: 51.4s\tremaining: 2m 4s\n",
      "410:\tlearn: 0.9565232\ttotal: 51.5s\tremaining: 2m 3s\n",
      "411:\tlearn: 0.9564785\ttotal: 51.6s\tremaining: 2m 3s\n",
      "412:\tlearn: 0.9564576\ttotal: 51.7s\tremaining: 2m 3s\n",
      "413:\tlearn: 0.9564173\ttotal: 51.8s\tremaining: 2m 3s\n",
      "414:\tlearn: 0.9563739\ttotal: 51.9s\tremaining: 2m 3s\n",
      "415:\tlearn: 0.9563378\ttotal: 52s\tremaining: 2m 3s\n",
      "416:\tlearn: 0.9563200\ttotal: 52.1s\tremaining: 2m 2s\n",
      "417:\tlearn: 0.9562708\ttotal: 52.2s\tremaining: 2m 2s\n",
      "418:\tlearn: 0.9562586\ttotal: 52.3s\tremaining: 2m 2s\n",
      "419:\tlearn: 0.9562564\ttotal: 52.4s\tremaining: 2m 2s\n",
      "420:\tlearn: 0.9561974\ttotal: 52.6s\tremaining: 2m 2s\n",
      "421:\tlearn: 0.9561901\ttotal: 52.7s\tremaining: 2m 2s\n",
      "422:\tlearn: 0.9561846\ttotal: 52.8s\tremaining: 2m 1s\n",
      "423:\tlearn: 0.9561657\ttotal: 52.9s\tremaining: 2m 1s\n",
      "424:\tlearn: 0.9561468\ttotal: 52.9s\tremaining: 2m 1s\n",
      "425:\tlearn: 0.9561081\ttotal: 53s\tremaining: 2m 1s\n",
      "426:\tlearn: 0.9560987\ttotal: 53.1s\tremaining: 2m 1s\n",
      "427:\tlearn: 0.9560935\ttotal: 53.2s\tremaining: 2m\n",
      "428:\tlearn: 0.9560469\ttotal: 53.3s\tremaining: 2m\n",
      "429:\tlearn: 0.9560124\ttotal: 53.4s\tremaining: 2m\n",
      "430:\tlearn: 0.9559720\ttotal: 53.5s\tremaining: 2m\n",
      "431:\tlearn: 0.9559261\ttotal: 53.7s\tremaining: 2m\n",
      "432:\tlearn: 0.9559056\ttotal: 53.8s\tremaining: 2m\n",
      "433:\tlearn: 0.9558441\ttotal: 53.9s\tremaining: 1m 59s\n",
      "434:\tlearn: 0.9558290\ttotal: 54s\tremaining: 1m 59s\n",
      "435:\tlearn: 0.9558139\ttotal: 54.2s\tremaining: 1m 59s\n",
      "436:\tlearn: 0.9557840\ttotal: 54.3s\tremaining: 1m 59s\n",
      "437:\tlearn: 0.9557763\ttotal: 54.4s\tremaining: 1m 59s\n",
      "438:\tlearn: 0.9557429\ttotal: 54.6s\tremaining: 1m 59s\n",
      "439:\tlearn: 0.9557371\ttotal: 54.7s\tremaining: 1m 59s\n",
      "440:\tlearn: 0.9556930\ttotal: 54.8s\tremaining: 1m 59s\n",
      "441:\tlearn: 0.9556499\ttotal: 54.9s\tremaining: 1m 59s\n",
      "442:\tlearn: 0.9556447\ttotal: 55s\tremaining: 1m 58s\n",
      "443:\tlearn: 0.9555938\ttotal: 55.2s\tremaining: 1m 58s\n",
      "444:\tlearn: 0.9555685\ttotal: 55.3s\tremaining: 1m 58s\n",
      "445:\tlearn: 0.9555232\ttotal: 55.4s\tremaining: 1m 58s\n",
      "446:\tlearn: 0.9554600\ttotal: 55.6s\tremaining: 1m 58s\n",
      "447:\tlearn: 0.9554189\ttotal: 55.8s\tremaining: 1m 58s\n",
      "448:\tlearn: 0.9553824\ttotal: 55.9s\tremaining: 1m 58s\n",
      "449:\tlearn: 0.9553344\ttotal: 56s\tremaining: 1m 58s\n",
      "450:\tlearn: 0.9553183\ttotal: 56.2s\tremaining: 1m 58s\n",
      "451:\tlearn: 0.9552847\ttotal: 56.4s\tremaining: 1m 58s\n",
      "452:\tlearn: 0.9552748\ttotal: 56.5s\tremaining: 1m 58s\n",
      "453:\tlearn: 0.9552021\ttotal: 56.6s\tremaining: 1m 57s\n",
      "454:\tlearn: 0.9551947\ttotal: 56.7s\tremaining: 1m 57s\n",
      "455:\tlearn: 0.9551400\ttotal: 56.8s\tremaining: 1m 57s\n",
      "456:\tlearn: 0.9550806\ttotal: 56.9s\tremaining: 1m 57s\n",
      "457:\tlearn: 0.9550436\ttotal: 57.1s\tremaining: 1m 57s\n",
      "458:\tlearn: 0.9549961\ttotal: 57.3s\tremaining: 1m 57s\n",
      "459:\tlearn: 0.9549666\ttotal: 57.4s\tremaining: 1m 57s\n",
      "460:\tlearn: 0.9549540\ttotal: 57.6s\tremaining: 1m 57s\n",
      "461:\tlearn: 0.9549279\ttotal: 57.7s\tremaining: 1m 57s\n",
      "462:\tlearn: 0.9548847\ttotal: 57.9s\tremaining: 1m 57s\n",
      "463:\tlearn: 0.9548307\ttotal: 58.1s\tremaining: 1m 57s\n",
      "464:\tlearn: 0.9547857\ttotal: 58.2s\tremaining: 1m 56s\n",
      "465:\tlearn: 0.9547635\ttotal: 58.3s\tremaining: 1m 56s\n",
      "466:\tlearn: 0.9547215\ttotal: 58.4s\tremaining: 1m 56s\n",
      "467:\tlearn: 0.9546841\ttotal: 58.5s\tremaining: 1m 56s\n",
      "468:\tlearn: 0.9546609\ttotal: 58.6s\tremaining: 1m 56s\n",
      "469:\tlearn: 0.9546261\ttotal: 58.8s\tremaining: 1m 56s\n",
      "470:\tlearn: 0.9545988\ttotal: 58.9s\tremaining: 1m 56s\n",
      "471:\tlearn: 0.9545359\ttotal: 59s\tremaining: 1m 56s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "472:\tlearn: 0.9545057\ttotal: 59.2s\tremaining: 1m 55s\n",
      "473:\tlearn: 0.9544537\ttotal: 59.3s\tremaining: 1m 55s\n",
      "474:\tlearn: 0.9544262\ttotal: 59.4s\tremaining: 1m 55s\n",
      "475:\tlearn: 0.9544100\ttotal: 59.5s\tremaining: 1m 55s\n",
      "476:\tlearn: 0.9543854\ttotal: 59.7s\tremaining: 1m 55s\n",
      "477:\tlearn: 0.9543393\ttotal: 59.9s\tremaining: 1m 55s\n",
      "478:\tlearn: 0.9543353\ttotal: 60s\tremaining: 1m 55s\n",
      "479:\tlearn: 0.9542999\ttotal: 1m\tremaining: 1m 55s\n",
      "480:\tlearn: 0.9542667\ttotal: 1m\tremaining: 1m 55s\n",
      "481:\tlearn: 0.9542640\ttotal: 1m\tremaining: 1m 54s\n",
      "482:\tlearn: 0.9542198\ttotal: 1m\tremaining: 1m 54s\n",
      "483:\tlearn: 0.9541949\ttotal: 1m\tremaining: 1m 54s\n",
      "484:\tlearn: 0.9541654\ttotal: 1m\tremaining: 1m 54s\n",
      "485:\tlearn: 0.9541080\ttotal: 1m\tremaining: 1m 54s\n",
      "486:\tlearn: 0.9540559\ttotal: 1m 1s\tremaining: 1m 54s\n",
      "487:\tlearn: 0.9540281\ttotal: 1m 1s\tremaining: 1m 54s\n",
      "488:\tlearn: 0.9540210\ttotal: 1m 1s\tremaining: 1m 54s\n",
      "489:\tlearn: 0.9539750\ttotal: 1m 1s\tremaining: 1m 54s\n",
      "490:\tlearn: 0.9539171\ttotal: 1m 1s\tremaining: 1m 54s\n",
      "491:\tlearn: 0.9539006\ttotal: 1m 1s\tremaining: 1m 54s\n",
      "492:\tlearn: 0.9538626\ttotal: 1m 1s\tremaining: 1m 53s\n",
      "493:\tlearn: 0.9538530\ttotal: 1m 1s\tremaining: 1m 53s\n",
      "494:\tlearn: 0.9538256\ttotal: 1m 2s\tremaining: 1m 53s\n",
      "495:\tlearn: 0.9537902\ttotal: 1m 2s\tremaining: 1m 53s\n",
      "496:\tlearn: 0.9537847\ttotal: 1m 2s\tremaining: 1m 53s\n",
      "497:\tlearn: 0.9537415\ttotal: 1m 2s\tremaining: 1m 53s\n",
      "498:\tlearn: 0.9536953\ttotal: 1m 2s\tremaining: 1m 53s\n",
      "499:\tlearn: 0.9536624\ttotal: 1m 2s\tremaining: 1m 52s\n",
      "500:\tlearn: 0.9536335\ttotal: 1m 2s\tremaining: 1m 52s\n",
      "501:\tlearn: 0.9535929\ttotal: 1m 3s\tremaining: 1m 52s\n",
      "502:\tlearn: 0.9535460\ttotal: 1m 3s\tremaining: 1m 52s\n",
      "503:\tlearn: 0.9535055\ttotal: 1m 3s\tremaining: 1m 52s\n",
      "504:\tlearn: 0.9534762\ttotal: 1m 3s\tremaining: 1m 52s\n",
      "505:\tlearn: 0.9534291\ttotal: 1m 3s\tremaining: 1m 52s\n",
      "506:\tlearn: 0.9534169\ttotal: 1m 3s\tremaining: 1m 52s\n",
      "507:\tlearn: 0.9533720\ttotal: 1m 3s\tremaining: 1m 51s\n",
      "508:\tlearn: 0.9533382\ttotal: 1m 3s\tremaining: 1m 51s\n",
      "509:\tlearn: 0.9532831\ttotal: 1m 4s\tremaining: 1m 51s\n",
      "510:\tlearn: 0.9532401\ttotal: 1m 4s\tremaining: 1m 51s\n",
      "511:\tlearn: 0.9532001\ttotal: 1m 4s\tremaining: 1m 51s\n",
      "512:\tlearn: 0.9531626\ttotal: 1m 4s\tremaining: 1m 51s\n",
      "513:\tlearn: 0.9531365\ttotal: 1m 4s\tremaining: 1m 51s\n",
      "514:\tlearn: 0.9531333\ttotal: 1m 4s\tremaining: 1m 51s\n",
      "515:\tlearn: 0.9531071\ttotal: 1m 4s\tremaining: 1m 51s\n",
      "516:\tlearn: 0.9530545\ttotal: 1m 5s\tremaining: 1m 51s\n",
      "517:\tlearn: 0.9530238\ttotal: 1m 5s\tremaining: 1m 51s\n",
      "518:\tlearn: 0.9530072\ttotal: 1m 5s\tremaining: 1m 50s\n",
      "519:\tlearn: 0.9529883\ttotal: 1m 5s\tremaining: 1m 50s\n",
      "520:\tlearn: 0.9529443\ttotal: 1m 5s\tremaining: 1m 50s\n",
      "521:\tlearn: 0.9529368\ttotal: 1m 5s\tremaining: 1m 50s\n",
      "522:\tlearn: 0.9528827\ttotal: 1m 5s\tremaining: 1m 50s\n",
      "523:\tlearn: 0.9528456\ttotal: 1m 6s\tremaining: 1m 50s\n",
      "524:\tlearn: 0.9528121\ttotal: 1m 6s\tremaining: 1m 50s\n",
      "525:\tlearn: 0.9527593\ttotal: 1m 6s\tremaining: 1m 50s\n",
      "526:\tlearn: 0.9527300\ttotal: 1m 6s\tremaining: 1m 50s\n",
      "527:\tlearn: 0.9526977\ttotal: 1m 6s\tremaining: 1m 50s\n",
      "528:\tlearn: 0.9526834\ttotal: 1m 6s\tremaining: 1m 49s\n",
      "529:\tlearn: 0.9526225\ttotal: 1m 6s\tremaining: 1m 49s\n",
      "530:\tlearn: 0.9525928\ttotal: 1m 6s\tremaining: 1m 49s\n",
      "531:\tlearn: 0.9525492\ttotal: 1m 7s\tremaining: 1m 49s\n",
      "532:\tlearn: 0.9525289\ttotal: 1m 7s\tremaining: 1m 49s\n",
      "533:\tlearn: 0.9524650\ttotal: 1m 7s\tremaining: 1m 49s\n",
      "534:\tlearn: 0.9524352\ttotal: 1m 7s\tremaining: 1m 49s\n",
      "535:\tlearn: 0.9524094\ttotal: 1m 7s\tremaining: 1m 49s\n",
      "536:\tlearn: 0.9523923\ttotal: 1m 7s\tremaining: 1m 49s\n",
      "537:\tlearn: 0.9523526\ttotal: 1m 7s\tremaining: 1m 48s\n",
      "538:\tlearn: 0.9523280\ttotal: 1m 8s\tremaining: 1m 48s\n",
      "539:\tlearn: 0.9522919\ttotal: 1m 8s\tremaining: 1m 48s\n",
      "540:\tlearn: 0.9522568\ttotal: 1m 8s\tremaining: 1m 48s\n",
      "541:\tlearn: 0.9522419\ttotal: 1m 8s\tremaining: 1m 48s\n",
      "542:\tlearn: 0.9522002\ttotal: 1m 8s\tremaining: 1m 48s\n",
      "543:\tlearn: 0.9521502\ttotal: 1m 8s\tremaining: 1m 48s\n",
      "544:\tlearn: 0.9521199\ttotal: 1m 8s\tremaining: 1m 48s\n",
      "545:\tlearn: 0.9521177\ttotal: 1m 9s\tremaining: 1m 47s\n",
      "546:\tlearn: 0.9521083\ttotal: 1m 9s\tremaining: 1m 47s\n",
      "547:\tlearn: 0.9520559\ttotal: 1m 9s\tremaining: 1m 47s\n",
      "548:\tlearn: 0.9520263\ttotal: 1m 9s\tremaining: 1m 47s\n",
      "549:\tlearn: 0.9519832\ttotal: 1m 9s\tremaining: 1m 47s\n",
      "550:\tlearn: 0.9519492\ttotal: 1m 9s\tremaining: 1m 47s\n",
      "551:\tlearn: 0.9519047\ttotal: 1m 9s\tremaining: 1m 47s\n",
      "552:\tlearn: 0.9518798\ttotal: 1m 9s\tremaining: 1m 46s\n",
      "553:\tlearn: 0.9518358\ttotal: 1m 9s\tremaining: 1m 46s\n",
      "554:\tlearn: 0.9517498\ttotal: 1m 10s\tremaining: 1m 46s\n",
      "555:\tlearn: 0.9517135\ttotal: 1m 10s\tremaining: 1m 46s\n",
      "556:\tlearn: 0.9517061\ttotal: 1m 10s\tremaining: 1m 46s\n",
      "557:\tlearn: 0.9516828\ttotal: 1m 10s\tremaining: 1m 46s\n",
      "558:\tlearn: 0.9516205\ttotal: 1m 10s\tremaining: 1m 46s\n",
      "559:\tlearn: 0.9515842\ttotal: 1m 10s\tremaining: 1m 45s\n",
      "560:\tlearn: 0.9515363\ttotal: 1m 10s\tremaining: 1m 45s\n",
      "561:\tlearn: 0.9514861\ttotal: 1m 10s\tremaining: 1m 45s\n",
      "562:\tlearn: 0.9514363\ttotal: 1m 11s\tremaining: 1m 45s\n",
      "563:\tlearn: 0.9513923\ttotal: 1m 11s\tremaining: 1m 45s\n",
      "564:\tlearn: 0.9513362\ttotal: 1m 11s\tremaining: 1m 45s\n",
      "565:\tlearn: 0.9512985\ttotal: 1m 11s\tremaining: 1m 45s\n",
      "566:\tlearn: 0.9512597\ttotal: 1m 11s\tremaining: 1m 45s\n",
      "567:\tlearn: 0.9512336\ttotal: 1m 11s\tremaining: 1m 45s\n",
      "568:\tlearn: 0.9511868\ttotal: 1m 11s\tremaining: 1m 44s\n",
      "569:\tlearn: 0.9511599\ttotal: 1m 11s\tremaining: 1m 44s\n",
      "570:\tlearn: 0.9511271\ttotal: 1m 12s\tremaining: 1m 44s\n",
      "571:\tlearn: 0.9511229\ttotal: 1m 12s\tremaining: 1m 44s\n",
      "572:\tlearn: 0.9511186\ttotal: 1m 12s\tremaining: 1m 44s\n",
      "573:\tlearn: 0.9510718\ttotal: 1m 12s\tremaining: 1m 44s\n",
      "574:\tlearn: 0.9510492\ttotal: 1m 12s\tremaining: 1m 43s\n",
      "575:\tlearn: 0.9510155\ttotal: 1m 12s\tremaining: 1m 43s\n",
      "576:\tlearn: 0.9509903\ttotal: 1m 12s\tremaining: 1m 43s\n",
      "577:\tlearn: 0.9509633\ttotal: 1m 12s\tremaining: 1m 43s\n",
      "578:\tlearn: 0.9509258\ttotal: 1m 12s\tremaining: 1m 43s\n",
      "579:\tlearn: 0.9508919\ttotal: 1m 13s\tremaining: 1m 43s\n",
      "580:\tlearn: 0.9508491\ttotal: 1m 13s\tremaining: 1m 43s\n",
      "581:\tlearn: 0.9508273\ttotal: 1m 13s\tremaining: 1m 42s\n",
      "582:\tlearn: 0.9507718\ttotal: 1m 13s\tremaining: 1m 42s\n",
      "583:\tlearn: 0.9507356\ttotal: 1m 13s\tremaining: 1m 42s\n",
      "584:\tlearn: 0.9506851\ttotal: 1m 13s\tremaining: 1m 42s\n",
      "585:\tlearn: 0.9506611\ttotal: 1m 13s\tremaining: 1m 42s\n",
      "586:\tlearn: 0.9506154\ttotal: 1m 13s\tremaining: 1m 42s\n",
      "587:\tlearn: 0.9505748\ttotal: 1m 14s\tremaining: 1m 42s\n",
      "588:\tlearn: 0.9505358\ttotal: 1m 14s\tremaining: 1m 42s\n",
      "589:\tlearn: 0.9504972\ttotal: 1m 14s\tremaining: 1m 42s\n",
      "590:\tlearn: 0.9504577\ttotal: 1m 14s\tremaining: 1m 42s\n",
      "591:\tlearn: 0.9504446\ttotal: 1m 14s\tremaining: 1m 41s\n",
      "592:\tlearn: 0.9504011\ttotal: 1m 14s\tremaining: 1m 41s\n",
      "593:\tlearn: 0.9503272\ttotal: 1m 15s\tremaining: 1m 41s\n",
      "594:\tlearn: 0.9502898\ttotal: 1m 15s\tremaining: 1m 41s\n",
      "595:\tlearn: 0.9502657\ttotal: 1m 15s\tremaining: 1m 41s\n",
      "596:\tlearn: 0.9502223\ttotal: 1m 15s\tremaining: 1m 41s\n",
      "597:\tlearn: 0.9501982\ttotal: 1m 15s\tremaining: 1m 41s\n",
      "598:\tlearn: 0.9501551\ttotal: 1m 15s\tremaining: 1m 41s\n",
      "599:\tlearn: 0.9501031\ttotal: 1m 16s\tremaining: 1m 41s\n",
      "600:\tlearn: 0.9500741\ttotal: 1m 16s\tremaining: 1m 41s\n",
      "601:\tlearn: 0.9500502\ttotal: 1m 16s\tremaining: 1m 41s\n",
      "602:\tlearn: 0.9499690\ttotal: 1m 16s\tremaining: 1m 40s\n",
      "603:\tlearn: 0.9499401\ttotal: 1m 16s\tremaining: 1m 40s\n",
      "604:\tlearn: 0.9498993\ttotal: 1m 16s\tremaining: 1m 40s\n",
      "605:\tlearn: 0.9498535\ttotal: 1m 16s\tremaining: 1m 40s\n",
      "606:\tlearn: 0.9498077\ttotal: 1m 16s\tremaining: 1m 40s\n",
      "607:\tlearn: 0.9497697\ttotal: 1m 16s\tremaining: 1m 40s\n",
      "608:\tlearn: 0.9497122\ttotal: 1m 17s\tremaining: 1m 40s\n",
      "609:\tlearn: 0.9497056\ttotal: 1m 17s\tremaining: 1m 40s\n",
      "610:\tlearn: 0.9496703\ttotal: 1m 17s\tremaining: 1m 39s\n",
      "611:\tlearn: 0.9496233\ttotal: 1m 17s\tremaining: 1m 39s\n",
      "612:\tlearn: 0.9495626\ttotal: 1m 17s\tremaining: 1m 39s\n",
      "613:\tlearn: 0.9495092\ttotal: 1m 17s\tremaining: 1m 39s\n",
      "614:\tlearn: 0.9494754\ttotal: 1m 18s\tremaining: 1m 39s\n",
      "615:\tlearn: 0.9494360\ttotal: 1m 18s\tremaining: 1m 39s\n",
      "616:\tlearn: 0.9494327\ttotal: 1m 18s\tremaining: 1m 39s\n",
      "617:\tlearn: 0.9493939\ttotal: 1m 18s\tremaining: 1m 39s\n",
      "618:\tlearn: 0.9493575\ttotal: 1m 18s\tremaining: 1m 39s\n",
      "619:\tlearn: 0.9493566\ttotal: 1m 18s\tremaining: 1m 39s\n",
      "620:\tlearn: 0.9493536\ttotal: 1m 18s\tremaining: 1m 38s\n",
      "621:\tlearn: 0.9493492\ttotal: 1m 18s\tremaining: 1m 38s\n",
      "622:\tlearn: 0.9492947\ttotal: 1m 19s\tremaining: 1m 38s\n",
      "623:\tlearn: 0.9492918\ttotal: 1m 19s\tremaining: 1m 38s\n",
      "624:\tlearn: 0.9492423\ttotal: 1m 19s\tremaining: 1m 38s\n",
      "625:\tlearn: 0.9492216\ttotal: 1m 19s\tremaining: 1m 38s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "626:\tlearn: 0.9491758\ttotal: 1m 19s\tremaining: 1m 38s\n",
      "627:\tlearn: 0.9491722\ttotal: 1m 19s\tremaining: 1m 37s\n",
      "628:\tlearn: 0.9491302\ttotal: 1m 19s\tremaining: 1m 37s\n",
      "629:\tlearn: 0.9491069\ttotal: 1m 19s\tremaining: 1m 37s\n",
      "630:\tlearn: 0.9490606\ttotal: 1m 20s\tremaining: 1m 37s\n",
      "631:\tlearn: 0.9490262\ttotal: 1m 20s\tremaining: 1m 37s\n",
      "632:\tlearn: 0.9490087\ttotal: 1m 20s\tremaining: 1m 37s\n",
      "633:\tlearn: 0.9489888\ttotal: 1m 20s\tremaining: 1m 37s\n",
      "634:\tlearn: 0.9489644\ttotal: 1m 20s\tremaining: 1m 37s\n",
      "635:\tlearn: 0.9489387\ttotal: 1m 20s\tremaining: 1m 36s\n",
      "636:\tlearn: 0.9489033\ttotal: 1m 20s\tremaining: 1m 36s\n",
      "637:\tlearn: 0.9488700\ttotal: 1m 20s\tremaining: 1m 36s\n",
      "638:\tlearn: 0.9488404\ttotal: 1m 21s\tremaining: 1m 36s\n",
      "639:\tlearn: 0.9488169\ttotal: 1m 21s\tremaining: 1m 36s\n",
      "640:\tlearn: 0.9487915\ttotal: 1m 21s\tremaining: 1m 36s\n",
      "641:\tlearn: 0.9487826\ttotal: 1m 21s\tremaining: 1m 36s\n",
      "642:\tlearn: 0.9487432\ttotal: 1m 21s\tremaining: 1m 35s\n",
      "643:\tlearn: 0.9486855\ttotal: 1m 21s\tremaining: 1m 35s\n",
      "644:\tlearn: 0.9486484\ttotal: 1m 21s\tremaining: 1m 35s\n",
      "645:\tlearn: 0.9486284\ttotal: 1m 22s\tremaining: 1m 35s\n",
      "646:\tlearn: 0.9485942\ttotal: 1m 22s\tremaining: 1m 35s\n",
      "647:\tlearn: 0.9485738\ttotal: 1m 22s\tremaining: 1m 35s\n",
      "648:\tlearn: 0.9485219\ttotal: 1m 22s\tremaining: 1m 35s\n",
      "649:\tlearn: 0.9484856\ttotal: 1m 22s\tremaining: 1m 35s\n",
      "650:\tlearn: 0.9484495\ttotal: 1m 22s\tremaining: 1m 35s\n",
      "651:\tlearn: 0.9484159\ttotal: 1m 22s\tremaining: 1m 35s\n",
      "652:\tlearn: 0.9483811\ttotal: 1m 23s\tremaining: 1m 35s\n",
      "653:\tlearn: 0.9483411\ttotal: 1m 23s\tremaining: 1m 34s\n",
      "654:\tlearn: 0.9483159\ttotal: 1m 23s\tremaining: 1m 34s\n",
      "655:\tlearn: 0.9482887\ttotal: 1m 23s\tremaining: 1m 34s\n",
      "656:\tlearn: 0.9482570\ttotal: 1m 23s\tremaining: 1m 34s\n",
      "657:\tlearn: 0.9482162\ttotal: 1m 23s\tremaining: 1m 34s\n",
      "658:\tlearn: 0.9481852\ttotal: 1m 23s\tremaining: 1m 34s\n",
      "659:\tlearn: 0.9481454\ttotal: 1m 24s\tremaining: 1m 34s\n",
      "660:\tlearn: 0.9481006\ttotal: 1m 24s\tremaining: 1m 34s\n",
      "661:\tlearn: 0.9480553\ttotal: 1m 24s\tremaining: 1m 34s\n",
      "662:\tlearn: 0.9480301\ttotal: 1m 24s\tremaining: 1m 33s\n",
      "663:\tlearn: 0.9479842\ttotal: 1m 24s\tremaining: 1m 33s\n",
      "664:\tlearn: 0.9479654\ttotal: 1m 24s\tremaining: 1m 33s\n",
      "665:\tlearn: 0.9479607\ttotal: 1m 24s\tremaining: 1m 33s\n",
      "666:\tlearn: 0.9479004\ttotal: 1m 25s\tremaining: 1m 33s\n",
      "667:\tlearn: 0.9478731\ttotal: 1m 25s\tremaining: 1m 33s\n",
      "668:\tlearn: 0.9478468\ttotal: 1m 25s\tremaining: 1m 33s\n",
      "669:\tlearn: 0.9478104\ttotal: 1m 25s\tremaining: 1m 33s\n",
      "670:\tlearn: 0.9477706\ttotal: 1m 25s\tremaining: 1m 33s\n",
      "671:\tlearn: 0.9477512\ttotal: 1m 25s\tremaining: 1m 32s\n",
      "672:\tlearn: 0.9477246\ttotal: 1m 25s\tremaining: 1m 32s\n",
      "673:\tlearn: 0.9477043\ttotal: 1m 26s\tremaining: 1m 32s\n",
      "674:\tlearn: 0.9476719\ttotal: 1m 26s\tremaining: 1m 32s\n",
      "675:\tlearn: 0.9476438\ttotal: 1m 26s\tremaining: 1m 32s\n",
      "676:\tlearn: 0.9476396\ttotal: 1m 26s\tremaining: 1m 32s\n",
      "677:\tlearn: 0.9476209\ttotal: 1m 26s\tremaining: 1m 32s\n",
      "678:\tlearn: 0.9475946\ttotal: 1m 26s\tremaining: 1m 32s\n",
      "679:\tlearn: 0.9475696\ttotal: 1m 26s\tremaining: 1m 31s\n",
      "680:\tlearn: 0.9475551\ttotal: 1m 26s\tremaining: 1m 31s\n",
      "681:\tlearn: 0.9475265\ttotal: 1m 27s\tremaining: 1m 31s\n",
      "682:\tlearn: 0.9475088\ttotal: 1m 27s\tremaining: 1m 31s\n",
      "683:\tlearn: 0.9474718\ttotal: 1m 27s\tremaining: 1m 31s\n",
      "684:\tlearn: 0.9474193\ttotal: 1m 27s\tremaining: 1m 31s\n",
      "685:\tlearn: 0.9473858\ttotal: 1m 27s\tremaining: 1m 31s\n",
      "686:\tlearn: 0.9473564\ttotal: 1m 27s\tremaining: 1m 31s\n",
      "687:\tlearn: 0.9473215\ttotal: 1m 28s\tremaining: 1m 31s\n",
      "688:\tlearn: 0.9472876\ttotal: 1m 28s\tremaining: 1m 31s\n",
      "689:\tlearn: 0.9472418\ttotal: 1m 28s\tremaining: 1m 31s\n",
      "690:\tlearn: 0.9472126\ttotal: 1m 28s\tremaining: 1m 30s\n",
      "691:\tlearn: 0.9472017\ttotal: 1m 28s\tremaining: 1m 30s\n",
      "692:\tlearn: 0.9471646\ttotal: 1m 28s\tremaining: 1m 30s\n",
      "693:\tlearn: 0.9471473\ttotal: 1m 28s\tremaining: 1m 30s\n",
      "694:\tlearn: 0.9471164\ttotal: 1m 29s\tremaining: 1m 30s\n",
      "695:\tlearn: 0.9470837\ttotal: 1m 29s\tremaining: 1m 30s\n",
      "696:\tlearn: 0.9470750\ttotal: 1m 29s\tremaining: 1m 30s\n",
      "697:\tlearn: 0.9470131\ttotal: 1m 29s\tremaining: 1m 30s\n",
      "698:\tlearn: 0.9469989\ttotal: 1m 29s\tremaining: 1m 29s\n",
      "699:\tlearn: 0.9469707\ttotal: 1m 29s\tremaining: 1m 29s\n",
      "700:\tlearn: 0.9469678\ttotal: 1m 29s\tremaining: 1m 29s\n",
      "701:\tlearn: 0.9469358\ttotal: 1m 29s\tremaining: 1m 29s\n",
      "702:\tlearn: 0.9468910\ttotal: 1m 30s\tremaining: 1m 29s\n",
      "703:\tlearn: 0.9468401\ttotal: 1m 30s\tremaining: 1m 29s\n",
      "704:\tlearn: 0.9468091\ttotal: 1m 30s\tremaining: 1m 29s\n",
      "705:\tlearn: 0.9467744\ttotal: 1m 30s\tremaining: 1m 29s\n",
      "706:\tlearn: 0.9467441\ttotal: 1m 30s\tremaining: 1m 28s\n",
      "707:\tlearn: 0.9467412\ttotal: 1m 30s\tremaining: 1m 28s\n",
      "708:\tlearn: 0.9466965\ttotal: 1m 31s\tremaining: 1m 28s\n",
      "709:\tlearn: 0.9466577\ttotal: 1m 31s\tremaining: 1m 28s\n",
      "710:\tlearn: 0.9466076\ttotal: 1m 31s\tremaining: 1m 28s\n",
      "711:\tlearn: 0.9465550\ttotal: 1m 31s\tremaining: 1m 28s\n",
      "712:\tlearn: 0.9465347\ttotal: 1m 31s\tremaining: 1m 28s\n",
      "713:\tlearn: 0.9465099\ttotal: 1m 31s\tremaining: 1m 28s\n",
      "714:\tlearn: 0.9464677\ttotal: 1m 31s\tremaining: 1m 28s\n",
      "715:\tlearn: 0.9464304\ttotal: 1m 32s\tremaining: 1m 27s\n",
      "716:\tlearn: 0.9463786\ttotal: 1m 32s\tremaining: 1m 27s\n",
      "717:\tlearn: 0.9463351\ttotal: 1m 32s\tremaining: 1m 27s\n",
      "718:\tlearn: 0.9463159\ttotal: 1m 32s\tremaining: 1m 27s\n",
      "719:\tlearn: 0.9462747\ttotal: 1m 32s\tremaining: 1m 27s\n",
      "720:\tlearn: 0.9462484\ttotal: 1m 32s\tremaining: 1m 27s\n",
      "721:\tlearn: 0.9462193\ttotal: 1m 32s\tremaining: 1m 27s\n",
      "722:\tlearn: 0.9462155\ttotal: 1m 33s\tremaining: 1m 27s\n",
      "723:\tlearn: 0.9461726\ttotal: 1m 33s\tremaining: 1m 27s\n",
      "724:\tlearn: 0.9461258\ttotal: 1m 33s\tremaining: 1m 26s\n",
      "725:\tlearn: 0.9461154\ttotal: 1m 33s\tremaining: 1m 26s\n",
      "726:\tlearn: 0.9460892\ttotal: 1m 33s\tremaining: 1m 26s\n",
      "727:\tlearn: 0.9460615\ttotal: 1m 33s\tremaining: 1m 26s\n",
      "728:\tlearn: 0.9460478\ttotal: 1m 33s\tremaining: 1m 26s\n",
      "729:\tlearn: 0.9460007\ttotal: 1m 34s\tremaining: 1m 26s\n",
      "730:\tlearn: 0.9459396\ttotal: 1m 34s\tremaining: 1m 26s\n",
      "731:\tlearn: 0.9459023\ttotal: 1m 34s\tremaining: 1m 26s\n",
      "732:\tlearn: 0.9458588\ttotal: 1m 34s\tremaining: 1m 25s\n",
      "733:\tlearn: 0.9458330\ttotal: 1m 34s\tremaining: 1m 25s\n",
      "734:\tlearn: 0.9458077\ttotal: 1m 34s\tremaining: 1m 25s\n",
      "735:\tlearn: 0.9457437\ttotal: 1m 34s\tremaining: 1m 25s\n",
      "736:\tlearn: 0.9456928\ttotal: 1m 35s\tremaining: 1m 25s\n",
      "737:\tlearn: 0.9456635\ttotal: 1m 35s\tremaining: 1m 25s\n",
      "738:\tlearn: 0.9456212\ttotal: 1m 35s\tremaining: 1m 25s\n",
      "739:\tlearn: 0.9456168\ttotal: 1m 35s\tremaining: 1m 25s\n",
      "740:\tlearn: 0.9455955\ttotal: 1m 35s\tremaining: 1m 25s\n",
      "741:\tlearn: 0.9455334\ttotal: 1m 35s\tremaining: 1m 24s\n",
      "742:\tlearn: 0.9455098\ttotal: 1m 35s\tremaining: 1m 24s\n",
      "743:\tlearn: 0.9454768\ttotal: 1m 36s\tremaining: 1m 24s\n",
      "744:\tlearn: 0.9454431\ttotal: 1m 36s\tremaining: 1m 24s\n",
      "745:\tlearn: 0.9454396\ttotal: 1m 36s\tremaining: 1m 24s\n",
      "746:\tlearn: 0.9454160\ttotal: 1m 36s\tremaining: 1m 24s\n",
      "747:\tlearn: 0.9453748\ttotal: 1m 36s\tremaining: 1m 24s\n",
      "748:\tlearn: 0.9453388\ttotal: 1m 36s\tremaining: 1m 24s\n",
      "749:\tlearn: 0.9453365\ttotal: 1m 36s\tremaining: 1m 24s\n",
      "750:\tlearn: 0.9453045\ttotal: 1m 37s\tremaining: 1m 23s\n",
      "751:\tlearn: 0.9452985\ttotal: 1m 37s\tremaining: 1m 23s\n",
      "752:\tlearn: 0.9452712\ttotal: 1m 37s\tremaining: 1m 23s\n",
      "753:\tlearn: 0.9452410\ttotal: 1m 37s\tremaining: 1m 23s\n",
      "754:\tlearn: 0.9452303\ttotal: 1m 37s\tremaining: 1m 23s\n",
      "755:\tlearn: 0.9452206\ttotal: 1m 37s\tremaining: 1m 23s\n",
      "756:\tlearn: 0.9451888\ttotal: 1m 37s\tremaining: 1m 23s\n",
      "757:\tlearn: 0.9451375\ttotal: 1m 38s\tremaining: 1m 23s\n",
      "758:\tlearn: 0.9451193\ttotal: 1m 38s\tremaining: 1m 23s\n",
      "759:\tlearn: 0.9450805\ttotal: 1m 38s\tremaining: 1m 22s\n",
      "760:\tlearn: 0.9450440\ttotal: 1m 38s\tremaining: 1m 22s\n",
      "761:\tlearn: 0.9450178\ttotal: 1m 38s\tremaining: 1m 22s\n",
      "762:\tlearn: 0.9449713\ttotal: 1m 38s\tremaining: 1m 22s\n",
      "763:\tlearn: 0.9449347\ttotal: 1m 39s\tremaining: 1m 22s\n",
      "764:\tlearn: 0.9449077\ttotal: 1m 39s\tremaining: 1m 22s\n",
      "765:\tlearn: 0.9449034\ttotal: 1m 39s\tremaining: 1m 22s\n",
      "766:\tlearn: 0.9448966\ttotal: 1m 39s\tremaining: 1m 22s\n",
      "767:\tlearn: 0.9448797\ttotal: 1m 39s\tremaining: 1m 21s\n",
      "768:\tlearn: 0.9448515\ttotal: 1m 39s\tremaining: 1m 21s\n",
      "769:\tlearn: 0.9447998\ttotal: 1m 39s\tremaining: 1m 21s\n",
      "770:\tlearn: 0.9447710\ttotal: 1m 40s\tremaining: 1m 21s\n",
      "771:\tlearn: 0.9447476\ttotal: 1m 40s\tremaining: 1m 21s\n",
      "772:\tlearn: 0.9447456\ttotal: 1m 40s\tremaining: 1m 21s\n",
      "773:\tlearn: 0.9446894\ttotal: 1m 40s\tremaining: 1m 21s\n",
      "774:\tlearn: 0.9446863\ttotal: 1m 40s\tremaining: 1m 21s\n",
      "775:\tlearn: 0.9446591\ttotal: 1m 40s\tremaining: 1m 21s\n",
      "776:\tlearn: 0.9446338\ttotal: 1m 40s\tremaining: 1m 20s\n",
      "777:\tlearn: 0.9446182\ttotal: 1m 40s\tremaining: 1m 20s\n",
      "778:\tlearn: 0.9445925\ttotal: 1m 41s\tremaining: 1m 20s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "779:\tlearn: 0.9445638\ttotal: 1m 41s\tremaining: 1m 20s\n",
      "780:\tlearn: 0.9445627\ttotal: 1m 41s\tremaining: 1m 20s\n",
      "781:\tlearn: 0.9445110\ttotal: 1m 41s\tremaining: 1m 20s\n",
      "782:\tlearn: 0.9444963\ttotal: 1m 41s\tremaining: 1m 20s\n",
      "783:\tlearn: 0.9444592\ttotal: 1m 41s\tremaining: 1m 19s\n",
      "784:\tlearn: 0.9444360\ttotal: 1m 41s\tremaining: 1m 19s\n",
      "785:\tlearn: 0.9444054\ttotal: 1m 41s\tremaining: 1m 19s\n",
      "786:\tlearn: 0.9443479\ttotal: 1m 42s\tremaining: 1m 19s\n",
      "787:\tlearn: 0.9443258\ttotal: 1m 42s\tremaining: 1m 19s\n",
      "788:\tlearn: 0.9442881\ttotal: 1m 42s\tremaining: 1m 19s\n",
      "789:\tlearn: 0.9442439\ttotal: 1m 42s\tremaining: 1m 19s\n",
      "790:\tlearn: 0.9442047\ttotal: 1m 42s\tremaining: 1m 19s\n",
      "791:\tlearn: 0.9441676\ttotal: 1m 42s\tremaining: 1m 19s\n",
      "792:\tlearn: 0.9441419\ttotal: 1m 43s\tremaining: 1m 18s\n",
      "793:\tlearn: 0.9441074\ttotal: 1m 43s\tremaining: 1m 18s\n",
      "794:\tlearn: 0.9440817\ttotal: 1m 43s\tremaining: 1m 18s\n",
      "795:\tlearn: 0.9440640\ttotal: 1m 43s\tremaining: 1m 18s\n",
      "796:\tlearn: 0.9440312\ttotal: 1m 43s\tremaining: 1m 18s\n",
      "797:\tlearn: 0.9439971\ttotal: 1m 43s\tremaining: 1m 18s\n",
      "798:\tlearn: 0.9439728\ttotal: 1m 44s\tremaining: 1m 18s\n",
      "799:\tlearn: 0.9439391\ttotal: 1m 44s\tremaining: 1m 18s\n",
      "800:\tlearn: 0.9438979\ttotal: 1m 44s\tremaining: 1m 17s\n",
      "801:\tlearn: 0.9438949\ttotal: 1m 44s\tremaining: 1m 17s\n",
      "802:\tlearn: 0.9438497\ttotal: 1m 44s\tremaining: 1m 17s\n",
      "803:\tlearn: 0.9438291\ttotal: 1m 44s\tremaining: 1m 17s\n",
      "804:\tlearn: 0.9437943\ttotal: 1m 44s\tremaining: 1m 17s\n",
      "805:\tlearn: 0.9437925\ttotal: 1m 45s\tremaining: 1m 17s\n",
      "806:\tlearn: 0.9437668\ttotal: 1m 45s\tremaining: 1m 17s\n",
      "807:\tlearn: 0.9437637\ttotal: 1m 45s\tremaining: 1m 17s\n",
      "808:\tlearn: 0.9437277\ttotal: 1m 45s\tremaining: 1m 16s\n",
      "809:\tlearn: 0.9437025\ttotal: 1m 45s\tremaining: 1m 16s\n",
      "810:\tlearn: 0.9436806\ttotal: 1m 45s\tremaining: 1m 16s\n",
      "811:\tlearn: 0.9436753\ttotal: 1m 45s\tremaining: 1m 16s\n",
      "812:\tlearn: 0.9436436\ttotal: 1m 45s\tremaining: 1m 16s\n",
      "813:\tlearn: 0.9436125\ttotal: 1m 46s\tremaining: 1m 16s\n",
      "814:\tlearn: 0.9436055\ttotal: 1m 46s\tremaining: 1m 16s\n",
      "815:\tlearn: 0.9436009\ttotal: 1m 46s\tremaining: 1m 16s\n",
      "816:\tlearn: 0.9435570\ttotal: 1m 46s\tremaining: 1m 15s\n",
      "817:\tlearn: 0.9435293\ttotal: 1m 46s\tremaining: 1m 15s\n",
      "818:\tlearn: 0.9434876\ttotal: 1m 46s\tremaining: 1m 15s\n",
      "819:\tlearn: 0.9434581\ttotal: 1m 46s\tremaining: 1m 15s\n",
      "820:\tlearn: 0.9434129\ttotal: 1m 47s\tremaining: 1m 15s\n",
      "821:\tlearn: 0.9433947\ttotal: 1m 47s\tremaining: 1m 15s\n",
      "822:\tlearn: 0.9433709\ttotal: 1m 47s\tremaining: 1m 15s\n",
      "823:\tlearn: 0.9433471\ttotal: 1m 47s\tremaining: 1m 15s\n",
      "824:\tlearn: 0.9433287\ttotal: 1m 47s\tremaining: 1m 15s\n",
      "825:\tlearn: 0.9432765\ttotal: 1m 48s\tremaining: 1m 15s\n",
      "826:\tlearn: 0.9432580\ttotal: 1m 48s\tremaining: 1m 14s\n",
      "827:\tlearn: 0.9432137\ttotal: 1m 48s\tremaining: 1m 14s\n",
      "828:\tlearn: 0.9432026\ttotal: 1m 48s\tremaining: 1m 14s\n",
      "829:\tlearn: 0.9431653\ttotal: 1m 48s\tremaining: 1m 14s\n",
      "830:\tlearn: 0.9431372\ttotal: 1m 48s\tremaining: 1m 14s\n",
      "831:\tlearn: 0.9431353\ttotal: 1m 48s\tremaining: 1m 14s\n",
      "832:\tlearn: 0.9431335\ttotal: 1m 49s\tremaining: 1m 14s\n",
      "833:\tlearn: 0.9431030\ttotal: 1m 49s\tremaining: 1m 14s\n",
      "834:\tlearn: 0.9430597\ttotal: 1m 49s\tremaining: 1m 13s\n",
      "835:\tlearn: 0.9430149\ttotal: 1m 49s\tremaining: 1m 13s\n",
      "836:\tlearn: 0.9429697\ttotal: 1m 49s\tremaining: 1m 13s\n",
      "837:\tlearn: 0.9429520\ttotal: 1m 49s\tremaining: 1m 13s\n",
      "838:\tlearn: 0.9429089\ttotal: 1m 49s\tremaining: 1m 13s\n",
      "839:\tlearn: 0.9428632\ttotal: 1m 50s\tremaining: 1m 13s\n",
      "840:\tlearn: 0.9428475\ttotal: 1m 50s\tremaining: 1m 13s\n",
      "841:\tlearn: 0.9428234\ttotal: 1m 50s\tremaining: 1m 13s\n",
      "842:\tlearn: 0.9427913\ttotal: 1m 50s\tremaining: 1m 13s\n",
      "843:\tlearn: 0.9427441\ttotal: 1m 50s\tremaining: 1m 12s\n",
      "844:\tlearn: 0.9427229\ttotal: 1m 50s\tremaining: 1m 12s\n",
      "845:\tlearn: 0.9426893\ttotal: 1m 50s\tremaining: 1m 12s\n",
      "846:\tlearn: 0.9426851\ttotal: 1m 51s\tremaining: 1m 12s\n",
      "847:\tlearn: 0.9426556\ttotal: 1m 51s\tremaining: 1m 12s\n",
      "848:\tlearn: 0.9425898\ttotal: 1m 51s\tremaining: 1m 12s\n",
      "849:\tlearn: 0.9425747\ttotal: 1m 51s\tremaining: 1m 12s\n",
      "850:\tlearn: 0.9425554\ttotal: 1m 51s\tremaining: 1m 11s\n",
      "851:\tlearn: 0.9425523\ttotal: 1m 51s\tremaining: 1m 11s\n",
      "852:\tlearn: 0.9425295\ttotal: 1m 51s\tremaining: 1m 11s\n",
      "853:\tlearn: 0.9425019\ttotal: 1m 51s\tremaining: 1m 11s\n",
      "854:\tlearn: 0.9424810\ttotal: 1m 52s\tremaining: 1m 11s\n",
      "855:\tlearn: 0.9424519\ttotal: 1m 52s\tremaining: 1m 11s\n",
      "856:\tlearn: 0.9424120\ttotal: 1m 52s\tremaining: 1m 11s\n",
      "857:\tlearn: 0.9423984\ttotal: 1m 52s\tremaining: 1m 11s\n",
      "858:\tlearn: 0.9423519\ttotal: 1m 52s\tremaining: 1m 10s\n",
      "859:\tlearn: 0.9423314\ttotal: 1m 52s\tremaining: 1m 10s\n",
      "860:\tlearn: 0.9423036\ttotal: 1m 52s\tremaining: 1m 10s\n",
      "861:\tlearn: 0.9422716\ttotal: 1m 53s\tremaining: 1m 10s\n",
      "862:\tlearn: 0.9422204\ttotal: 1m 53s\tremaining: 1m 10s\n",
      "863:\tlearn: 0.9422003\ttotal: 1m 53s\tremaining: 1m 10s\n",
      "864:\tlearn: 0.9421996\ttotal: 1m 53s\tremaining: 1m 10s\n",
      "865:\tlearn: 0.9421724\ttotal: 1m 53s\tremaining: 1m 10s\n",
      "866:\tlearn: 0.9421193\ttotal: 1m 53s\tremaining: 1m 9s\n",
      "867:\tlearn: 0.9421002\ttotal: 1m 53s\tremaining: 1m 9s\n",
      "868:\tlearn: 0.9420631\ttotal: 1m 54s\tremaining: 1m 9s\n",
      "869:\tlearn: 0.9420321\ttotal: 1m 54s\tremaining: 1m 9s\n",
      "870:\tlearn: 0.9419800\ttotal: 1m 54s\tremaining: 1m 9s\n",
      "871:\tlearn: 0.9419279\ttotal: 1m 54s\tremaining: 1m 9s\n",
      "872:\tlearn: 0.9418899\ttotal: 1m 54s\tremaining: 1m 9s\n",
      "873:\tlearn: 0.9418420\ttotal: 1m 54s\tremaining: 1m 9s\n",
      "874:\tlearn: 0.9418007\ttotal: 1m 54s\tremaining: 1m 8s\n",
      "875:\tlearn: 0.9417569\ttotal: 1m 55s\tremaining: 1m 8s\n",
      "876:\tlearn: 0.9417328\ttotal: 1m 55s\tremaining: 1m 8s\n",
      "877:\tlearn: 0.9416941\ttotal: 1m 55s\tremaining: 1m 8s\n",
      "878:\tlearn: 0.9416572\ttotal: 1m 55s\tremaining: 1m 8s\n",
      "879:\tlearn: 0.9416293\ttotal: 1m 55s\tremaining: 1m 8s\n",
      "880:\tlearn: 0.9416172\ttotal: 1m 55s\tremaining: 1m 8s\n",
      "881:\tlearn: 0.9416020\ttotal: 1m 55s\tremaining: 1m 8s\n",
      "882:\tlearn: 0.9415687\ttotal: 1m 56s\tremaining: 1m 7s\n",
      "883:\tlearn: 0.9415302\ttotal: 1m 56s\tremaining: 1m 7s\n",
      "884:\tlearn: 0.9415067\ttotal: 1m 56s\tremaining: 1m 7s\n",
      "885:\tlearn: 0.9414665\ttotal: 1m 56s\tremaining: 1m 7s\n",
      "886:\tlearn: 0.9414253\ttotal: 1m 56s\tremaining: 1m 7s\n",
      "887:\tlearn: 0.9414174\ttotal: 1m 56s\tremaining: 1m 7s\n",
      "888:\tlearn: 0.9413733\ttotal: 1m 56s\tremaining: 1m 7s\n",
      "889:\tlearn: 0.9413495\ttotal: 1m 56s\tremaining: 1m 7s\n",
      "890:\tlearn: 0.9412989\ttotal: 1m 57s\tremaining: 1m 6s\n",
      "891:\tlearn: 0.9412806\ttotal: 1m 57s\tremaining: 1m 6s\n",
      "892:\tlearn: 0.9412739\ttotal: 1m 57s\tremaining: 1m 6s\n",
      "893:\tlearn: 0.9412493\ttotal: 1m 57s\tremaining: 1m 6s\n",
      "894:\tlearn: 0.9412379\ttotal: 1m 57s\tremaining: 1m 6s\n",
      "895:\tlearn: 0.9412154\ttotal: 1m 57s\tremaining: 1m 6s\n",
      "896:\tlearn: 0.9411763\ttotal: 1m 57s\tremaining: 1m 6s\n",
      "897:\tlearn: 0.9411705\ttotal: 1m 57s\tremaining: 1m 5s\n",
      "898:\tlearn: 0.9411306\ttotal: 1m 58s\tremaining: 1m 5s\n",
      "899:\tlearn: 0.9411238\ttotal: 1m 58s\tremaining: 1m 5s\n",
      "900:\tlearn: 0.9411071\ttotal: 1m 58s\tremaining: 1m 5s\n",
      "901:\tlearn: 0.9410644\ttotal: 1m 58s\tremaining: 1m 5s\n",
      "902:\tlearn: 0.9410616\ttotal: 1m 58s\tremaining: 1m 5s\n",
      "903:\tlearn: 0.9410485\ttotal: 1m 58s\tremaining: 1m 5s\n",
      "904:\tlearn: 0.9409805\ttotal: 1m 58s\tremaining: 1m 5s\n",
      "905:\tlearn: 0.9409469\ttotal: 1m 59s\tremaining: 1m 4s\n",
      "906:\tlearn: 0.9409426\ttotal: 1m 59s\tremaining: 1m 4s\n",
      "907:\tlearn: 0.9408985\ttotal: 1m 59s\tremaining: 1m 4s\n",
      "908:\tlearn: 0.9408500\ttotal: 1m 59s\tremaining: 1m 4s\n",
      "909:\tlearn: 0.9408309\ttotal: 1m 59s\tremaining: 1m 4s\n",
      "910:\tlearn: 0.9408244\ttotal: 1m 59s\tremaining: 1m 4s\n",
      "911:\tlearn: 0.9408239\ttotal: 1m 59s\tremaining: 1m 4s\n",
      "912:\tlearn: 0.9407906\ttotal: 2m\tremaining: 1m 4s\n",
      "913:\tlearn: 0.9407481\ttotal: 2m\tremaining: 1m 3s\n",
      "914:\tlearn: 0.9406994\ttotal: 2m\tremaining: 1m 3s\n",
      "915:\tlearn: 0.9406797\ttotal: 2m\tremaining: 1m 3s\n",
      "916:\tlearn: 0.9406712\ttotal: 2m\tremaining: 1m 3s\n",
      "917:\tlearn: 0.9406324\ttotal: 2m\tremaining: 1m 3s\n",
      "918:\tlearn: 0.9406138\ttotal: 2m\tremaining: 1m 3s\n",
      "919:\tlearn: 0.9405892\ttotal: 2m\tremaining: 1m 3s\n",
      "920:\tlearn: 0.9405668\ttotal: 2m 1s\tremaining: 1m 2s\n",
      "921:\tlearn: 0.9405243\ttotal: 2m 1s\tremaining: 1m 2s\n",
      "922:\tlearn: 0.9404950\ttotal: 2m 1s\tremaining: 1m 2s\n",
      "923:\tlearn: 0.9404697\ttotal: 2m 1s\tremaining: 1m 2s\n",
      "924:\tlearn: 0.9404417\ttotal: 2m 1s\tremaining: 1m 2s\n",
      "925:\tlearn: 0.9404387\ttotal: 2m 1s\tremaining: 1m 2s\n",
      "926:\tlearn: 0.9403979\ttotal: 2m 1s\tremaining: 1m 2s\n",
      "927:\tlearn: 0.9403682\ttotal: 2m 1s\tremaining: 1m 2s\n",
      "928:\tlearn: 0.9403483\ttotal: 2m 2s\tremaining: 1m 1s\n",
      "929:\tlearn: 0.9403336\ttotal: 2m 2s\tremaining: 1m 1s\n",
      "930:\tlearn: 0.9403074\ttotal: 2m 2s\tremaining: 1m 1s\n",
      "931:\tlearn: 0.9402660\ttotal: 2m 2s\tremaining: 1m 1s\n",
      "932:\tlearn: 0.9402359\ttotal: 2m 2s\tremaining: 1m 1s\n",
      "933:\tlearn: 0.9402021\ttotal: 2m 2s\tremaining: 1m 1s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "934:\tlearn: 0.9401943\ttotal: 2m 2s\tremaining: 1m 1s\n",
      "935:\tlearn: 0.9401457\ttotal: 2m 2s\tremaining: 1m\n",
      "936:\tlearn: 0.9401219\ttotal: 2m 3s\tremaining: 1m\n",
      "937:\tlearn: 0.9400598\ttotal: 2m 3s\tremaining: 1m\n",
      "938:\tlearn: 0.9400217\ttotal: 2m 3s\tremaining: 1m\n",
      "939:\tlearn: 0.9399953\ttotal: 2m 3s\tremaining: 1m\n",
      "940:\tlearn: 0.9399825\ttotal: 2m 3s\tremaining: 1m\n",
      "941:\tlearn: 0.9399428\ttotal: 2m 3s\tremaining: 1m\n",
      "942:\tlearn: 0.9399180\ttotal: 2m 3s\tremaining: 1m\n",
      "943:\tlearn: 0.9398996\ttotal: 2m 3s\tremaining: 59.9s\n",
      "944:\tlearn: 0.9398733\ttotal: 2m 4s\tremaining: 59.7s\n",
      "945:\tlearn: 0.9398520\ttotal: 2m 4s\tremaining: 59.6s\n",
      "946:\tlearn: 0.9398173\ttotal: 2m 4s\tremaining: 59.5s\n",
      "947:\tlearn: 0.9397996\ttotal: 2m 4s\tremaining: 59.3s\n",
      "948:\tlearn: 0.9397765\ttotal: 2m 4s\tremaining: 59.2s\n",
      "949:\tlearn: 0.9397528\ttotal: 2m 4s\tremaining: 59s\n",
      "950:\tlearn: 0.9397397\ttotal: 2m 4s\tremaining: 58.9s\n",
      "951:\tlearn: 0.9396947\ttotal: 2m 4s\tremaining: 58.8s\n",
      "952:\tlearn: 0.9396891\ttotal: 2m 4s\tremaining: 58.6s\n",
      "953:\tlearn: 0.9396541\ttotal: 2m 5s\tremaining: 58.5s\n",
      "954:\tlearn: 0.9396127\ttotal: 2m 5s\tremaining: 58.4s\n",
      "955:\tlearn: 0.9395925\ttotal: 2m 5s\tremaining: 58.2s\n",
      "956:\tlearn: 0.9395670\ttotal: 2m 5s\tremaining: 58.1s\n",
      "957:\tlearn: 0.9395480\ttotal: 2m 5s\tremaining: 58s\n",
      "958:\tlearn: 0.9395301\ttotal: 2m 5s\tremaining: 57.8s\n",
      "959:\tlearn: 0.9394855\ttotal: 2m 5s\tremaining: 57.7s\n",
      "960:\tlearn: 0.9394833\ttotal: 2m 6s\tremaining: 57.6s\n",
      "961:\tlearn: 0.9394671\ttotal: 2m 6s\tremaining: 57.4s\n",
      "962:\tlearn: 0.9394257\ttotal: 2m 6s\tremaining: 57.3s\n",
      "963:\tlearn: 0.9394061\ttotal: 2m 6s\tremaining: 57.2s\n",
      "964:\tlearn: 0.9393921\ttotal: 2m 6s\tremaining: 57s\n",
      "965:\tlearn: 0.9393533\ttotal: 2m 6s\tremaining: 56.9s\n",
      "966:\tlearn: 0.9393425\ttotal: 2m 6s\tremaining: 56.7s\n",
      "967:\tlearn: 0.9393298\ttotal: 2m 6s\tremaining: 56.6s\n",
      "968:\tlearn: 0.9393270\ttotal: 2m 6s\tremaining: 56.5s\n",
      "969:\tlearn: 0.9393257\ttotal: 2m 7s\tremaining: 56.3s\n",
      "970:\tlearn: 0.9392893\ttotal: 2m 7s\tremaining: 56.2s\n",
      "971:\tlearn: 0.9392519\ttotal: 2m 7s\tremaining: 56s\n",
      "972:\tlearn: 0.9392191\ttotal: 2m 7s\tremaining: 55.9s\n",
      "973:\tlearn: 0.9391646\ttotal: 2m 7s\tremaining: 55.8s\n",
      "974:\tlearn: 0.9391350\ttotal: 2m 7s\tremaining: 55.6s\n",
      "975:\tlearn: 0.9391019\ttotal: 2m 7s\tremaining: 55.5s\n",
      "976:\tlearn: 0.9390855\ttotal: 2m 7s\tremaining: 55.4s\n",
      "977:\tlearn: 0.9390576\ttotal: 2m 7s\tremaining: 55.2s\n",
      "978:\tlearn: 0.9390314\ttotal: 2m 8s\tremaining: 55.1s\n",
      "979:\tlearn: 0.9389764\ttotal: 2m 8s\tremaining: 55s\n",
      "980:\tlearn: 0.9389317\ttotal: 2m 8s\tremaining: 54.8s\n",
      "981:\tlearn: 0.9388847\ttotal: 2m 8s\tremaining: 54.7s\n",
      "982:\tlearn: 0.9388477\ttotal: 2m 8s\tremaining: 54.6s\n",
      "983:\tlearn: 0.9388175\ttotal: 2m 8s\tremaining: 54.4s\n",
      "984:\tlearn: 0.9387814\ttotal: 2m 8s\tremaining: 54.3s\n",
      "985:\tlearn: 0.9387599\ttotal: 2m 9s\tremaining: 54.2s\n",
      "986:\tlearn: 0.9387479\ttotal: 2m 9s\tremaining: 54s\n",
      "987:\tlearn: 0.9387238\ttotal: 2m 9s\tremaining: 53.9s\n",
      "988:\tlearn: 0.9386874\ttotal: 2m 9s\tremaining: 53.8s\n",
      "989:\tlearn: 0.9386421\ttotal: 2m 9s\tremaining: 53.7s\n",
      "990:\tlearn: 0.9386150\ttotal: 2m 9s\tremaining: 53.6s\n",
      "991:\tlearn: 0.9385898\ttotal: 2m 9s\tremaining: 53.4s\n",
      "992:\tlearn: 0.9385700\ttotal: 2m 10s\tremaining: 53.3s\n",
      "993:\tlearn: 0.9385690\ttotal: 2m 10s\tremaining: 53.2s\n",
      "994:\tlearn: 0.9385346\ttotal: 2m 10s\tremaining: 53s\n",
      "995:\tlearn: 0.9384852\ttotal: 2m 10s\tremaining: 52.9s\n",
      "996:\tlearn: 0.9384485\ttotal: 2m 10s\tremaining: 52.8s\n",
      "997:\tlearn: 0.9383890\ttotal: 2m 10s\tremaining: 52.7s\n",
      "998:\tlearn: 0.9383594\ttotal: 2m 10s\tremaining: 52.5s\n",
      "999:\tlearn: 0.9383319\ttotal: 2m 11s\tremaining: 52.4s\n",
      "1000:\tlearn: 0.9383102\ttotal: 2m 11s\tremaining: 52.3s\n",
      "1001:\tlearn: 0.9382862\ttotal: 2m 11s\tremaining: 52.2s\n",
      "1002:\tlearn: 0.9382681\ttotal: 2m 11s\tremaining: 52s\n",
      "1003:\tlearn: 0.9382259\ttotal: 2m 11s\tremaining: 51.9s\n",
      "1004:\tlearn: 0.9381887\ttotal: 2m 11s\tremaining: 51.8s\n",
      "1005:\tlearn: 0.9381358\ttotal: 2m 11s\tremaining: 51.7s\n",
      "1006:\tlearn: 0.9381128\ttotal: 2m 12s\tremaining: 51.6s\n",
      "1007:\tlearn: 0.9380723\ttotal: 2m 12s\tremaining: 51.4s\n",
      "1008:\tlearn: 0.9380472\ttotal: 2m 12s\tremaining: 51.3s\n",
      "1009:\tlearn: 0.9380270\ttotal: 2m 12s\tremaining: 51.2s\n",
      "1010:\tlearn: 0.9379907\ttotal: 2m 12s\tremaining: 51.1s\n",
      "1011:\tlearn: 0.9379570\ttotal: 2m 12s\tremaining: 51s\n",
      "1012:\tlearn: 0.9379024\ttotal: 2m 13s\tremaining: 50.8s\n",
      "1013:\tlearn: 0.9378839\ttotal: 2m 13s\tremaining: 50.7s\n",
      "1014:\tlearn: 0.9378415\ttotal: 2m 13s\tremaining: 50.6s\n",
      "1015:\tlearn: 0.9377972\ttotal: 2m 13s\tremaining: 50.5s\n",
      "1016:\tlearn: 0.9377863\ttotal: 2m 13s\tremaining: 50.4s\n",
      "1017:\tlearn: 0.9377573\ttotal: 2m 13s\tremaining: 50.2s\n",
      "1018:\tlearn: 0.9377271\ttotal: 2m 14s\tremaining: 50.1s\n",
      "1019:\tlearn: 0.9376866\ttotal: 2m 14s\tremaining: 50s\n",
      "1020:\tlearn: 0.9376392\ttotal: 2m 14s\tremaining: 49.9s\n",
      "1021:\tlearn: 0.9376023\ttotal: 2m 14s\tremaining: 49.8s\n",
      "1022:\tlearn: 0.9376004\ttotal: 2m 14s\tremaining: 49.6s\n",
      "1023:\tlearn: 0.9375624\ttotal: 2m 14s\tremaining: 49.5s\n",
      "1024:\tlearn: 0.9375479\ttotal: 2m 14s\tremaining: 49.4s\n",
      "1025:\tlearn: 0.9375029\ttotal: 2m 15s\tremaining: 49.2s\n",
      "1026:\tlearn: 0.9374683\ttotal: 2m 15s\tremaining: 49.1s\n",
      "1027:\tlearn: 0.9374239\ttotal: 2m 15s\tremaining: 49s\n",
      "1028:\tlearn: 0.9374024\ttotal: 2m 15s\tremaining: 48.9s\n",
      "1029:\tlearn: 0.9373588\ttotal: 2m 15s\tremaining: 48.7s\n",
      "1030:\tlearn: 0.9373045\ttotal: 2m 15s\tremaining: 48.6s\n",
      "1031:\tlearn: 0.9372677\ttotal: 2m 15s\tremaining: 48.4s\n",
      "1032:\tlearn: 0.9372497\ttotal: 2m 15s\tremaining: 48.3s\n",
      "1033:\tlearn: 0.9372477\ttotal: 2m 16s\tremaining: 48.2s\n",
      "1034:\tlearn: 0.9372220\ttotal: 2m 16s\tremaining: 48s\n",
      "1035:\tlearn: 0.9371737\ttotal: 2m 16s\tremaining: 47.9s\n",
      "1036:\tlearn: 0.9371378\ttotal: 2m 16s\tremaining: 47.8s\n",
      "1037:\tlearn: 0.9370876\ttotal: 2m 16s\tremaining: 47.6s\n",
      "1038:\tlearn: 0.9370490\ttotal: 2m 16s\tremaining: 47.5s\n",
      "1039:\tlearn: 0.9369870\ttotal: 2m 16s\tremaining: 47.4s\n",
      "1040:\tlearn: 0.9369644\ttotal: 2m 16s\tremaining: 47.2s\n",
      "1041:\tlearn: 0.9369343\ttotal: 2m 17s\tremaining: 47.1s\n",
      "1042:\tlearn: 0.9369118\ttotal: 2m 17s\tremaining: 47s\n",
      "1043:\tlearn: 0.9368990\ttotal: 2m 17s\tremaining: 46.8s\n",
      "1044:\tlearn: 0.9368483\ttotal: 2m 17s\tremaining: 46.7s\n",
      "1045:\tlearn: 0.9368213\ttotal: 2m 17s\tremaining: 46.5s\n",
      "1046:\tlearn: 0.9367784\ttotal: 2m 17s\tremaining: 46.4s\n",
      "1047:\tlearn: 0.9367499\ttotal: 2m 17s\tremaining: 46.3s\n",
      "1048:\tlearn: 0.9367234\ttotal: 2m 17s\tremaining: 46.1s\n",
      "1049:\tlearn: 0.9366941\ttotal: 2m 18s\tremaining: 46s\n",
      "1050:\tlearn: 0.9366794\ttotal: 2m 18s\tremaining: 45.9s\n",
      "1051:\tlearn: 0.9366567\ttotal: 2m 18s\tremaining: 45.7s\n",
      "1052:\tlearn: 0.9366432\ttotal: 2m 18s\tremaining: 45.6s\n",
      "1053:\tlearn: 0.9366171\ttotal: 2m 18s\tremaining: 45.5s\n",
      "1054:\tlearn: 0.9366024\ttotal: 2m 18s\tremaining: 45.3s\n",
      "1055:\tlearn: 0.9365969\ttotal: 2m 18s\tremaining: 45.2s\n",
      "1056:\tlearn: 0.9365710\ttotal: 2m 18s\tremaining: 45s\n",
      "1057:\tlearn: 0.9365645\ttotal: 2m 18s\tremaining: 44.9s\n",
      "1058:\tlearn: 0.9365333\ttotal: 2m 18s\tremaining: 44.8s\n",
      "1059:\tlearn: 0.9364879\ttotal: 2m 19s\tremaining: 44.6s\n",
      "1060:\tlearn: 0.9364630\ttotal: 2m 19s\tremaining: 44.5s\n",
      "1061:\tlearn: 0.9364272\ttotal: 2m 19s\tremaining: 44.3s\n",
      "1062:\tlearn: 0.9363969\ttotal: 2m 19s\tremaining: 44.2s\n",
      "1063:\tlearn: 0.9363899\ttotal: 2m 19s\tremaining: 44.1s\n",
      "1064:\tlearn: 0.9363635\ttotal: 2m 19s\tremaining: 43.9s\n",
      "1065:\tlearn: 0.9363321\ttotal: 2m 19s\tremaining: 43.8s\n",
      "1066:\tlearn: 0.9363064\ttotal: 2m 19s\tremaining: 43.6s\n",
      "1067:\tlearn: 0.9362847\ttotal: 2m 19s\tremaining: 43.5s\n",
      "1068:\tlearn: 0.9362728\ttotal: 2m 20s\tremaining: 43.4s\n",
      "1069:\tlearn: 0.9362430\ttotal: 2m 20s\tremaining: 43.3s\n",
      "1070:\tlearn: 0.9362231\ttotal: 2m 20s\tremaining: 43.1s\n",
      "1071:\tlearn: 0.9362139\ttotal: 2m 20s\tremaining: 43s\n",
      "1072:\tlearn: 0.9362053\ttotal: 2m 20s\tremaining: 42.8s\n",
      "1073:\tlearn: 0.9361745\ttotal: 2m 20s\tremaining: 42.7s\n",
      "1074:\tlearn: 0.9361448\ttotal: 2m 20s\tremaining: 42.5s\n",
      "1075:\tlearn: 0.9361271\ttotal: 2m 20s\tremaining: 42.4s\n",
      "1076:\tlearn: 0.9360963\ttotal: 2m 20s\tremaining: 42.3s\n",
      "1077:\tlearn: 0.9360822\ttotal: 2m 21s\tremaining: 42.1s\n",
      "1078:\tlearn: 0.9360704\ttotal: 2m 21s\tremaining: 42s\n",
      "1079:\tlearn: 0.9360293\ttotal: 2m 21s\tremaining: 41.9s\n",
      "1080:\tlearn: 0.9360059\ttotal: 2m 21s\tremaining: 41.7s\n",
      "1081:\tlearn: 0.9359757\ttotal: 2m 21s\tremaining: 41.6s\n",
      "1082:\tlearn: 0.9359499\ttotal: 2m 21s\tremaining: 41.5s\n",
      "1083:\tlearn: 0.9359049\ttotal: 2m 21s\tremaining: 41.3s\n",
      "1084:\tlearn: 0.9358693\ttotal: 2m 21s\tremaining: 41.2s\n",
      "1085:\tlearn: 0.9358410\ttotal: 2m 21s\tremaining: 41.1s\n",
      "1086:\tlearn: 0.9358243\ttotal: 2m 22s\tremaining: 40.9s\n",
      "1087:\tlearn: 0.9358065\ttotal: 2m 22s\tremaining: 40.8s\n",
      "1088:\tlearn: 0.9357866\ttotal: 2m 22s\tremaining: 40.6s\n",
      "1089:\tlearn: 0.9357572\ttotal: 2m 22s\tremaining: 40.5s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1090:\tlearn: 0.9357269\ttotal: 2m 22s\tremaining: 40.4s\n",
      "1091:\tlearn: 0.9357021\ttotal: 2m 22s\tremaining: 40.2s\n",
      "1092:\tlearn: 0.9356715\ttotal: 2m 22s\tremaining: 40.1s\n",
      "1093:\tlearn: 0.9356464\ttotal: 2m 22s\tremaining: 40s\n",
      "1094:\tlearn: 0.9356130\ttotal: 2m 23s\tremaining: 39.9s\n",
      "1095:\tlearn: 0.9355733\ttotal: 2m 23s\tremaining: 39.7s\n",
      "1096:\tlearn: 0.9355512\ttotal: 2m 23s\tremaining: 39.6s\n",
      "1097:\tlearn: 0.9354981\ttotal: 2m 23s\tremaining: 39.5s\n",
      "1098:\tlearn: 0.9354672\ttotal: 2m 23s\tremaining: 39.3s\n",
      "1099:\tlearn: 0.9354123\ttotal: 2m 23s\tremaining: 39.2s\n",
      "1100:\tlearn: 0.9353786\ttotal: 2m 23s\tremaining: 39.1s\n",
      "1101:\tlearn: 0.9353312\ttotal: 2m 24s\tremaining: 39s\n",
      "1102:\tlearn: 0.9352847\ttotal: 2m 24s\tremaining: 38.8s\n",
      "1103:\tlearn: 0.9352807\ttotal: 2m 24s\tremaining: 38.7s\n",
      "1104:\tlearn: 0.9352343\ttotal: 2m 24s\tremaining: 38.6s\n",
      "1105:\tlearn: 0.9351979\ttotal: 2m 24s\tremaining: 38.5s\n",
      "1106:\tlearn: 0.9351846\ttotal: 2m 24s\tremaining: 38.3s\n",
      "1107:\tlearn: 0.9351506\ttotal: 2m 24s\tremaining: 38.2s\n",
      "1108:\tlearn: 0.9351121\ttotal: 2m 25s\tremaining: 38.1s\n",
      "1109:\tlearn: 0.9350877\ttotal: 2m 25s\tremaining: 37.9s\n",
      "1110:\tlearn: 0.9350592\ttotal: 2m 25s\tremaining: 37.8s\n",
      "1111:\tlearn: 0.9350294\ttotal: 2m 25s\tremaining: 37.7s\n",
      "1112:\tlearn: 0.9350272\ttotal: 2m 25s\tremaining: 37.5s\n",
      "1113:\tlearn: 0.9349975\ttotal: 2m 25s\tremaining: 37.4s\n",
      "1114:\tlearn: 0.9349757\ttotal: 2m 25s\tremaining: 37.3s\n",
      "1115:\tlearn: 0.9349672\ttotal: 2m 25s\tremaining: 37.1s\n",
      "1116:\tlearn: 0.9349455\ttotal: 2m 25s\tremaining: 37s\n",
      "1117:\tlearn: 0.9349211\ttotal: 2m 26s\tremaining: 36.8s\n",
      "1118:\tlearn: 0.9348722\ttotal: 2m 26s\tremaining: 36.7s\n",
      "1119:\tlearn: 0.9348242\ttotal: 2m 26s\tremaining: 36.6s\n",
      "1120:\tlearn: 0.9347983\ttotal: 2m 26s\tremaining: 36.4s\n",
      "1121:\tlearn: 0.9347705\ttotal: 2m 26s\tremaining: 36.3s\n",
      "1122:\tlearn: 0.9347272\ttotal: 2m 26s\tremaining: 36.2s\n",
      "1123:\tlearn: 0.9346973\ttotal: 2m 26s\tremaining: 36s\n",
      "1124:\tlearn: 0.9346793\ttotal: 2m 26s\tremaining: 35.9s\n",
      "1125:\tlearn: 0.9346455\ttotal: 2m 27s\tremaining: 35.8s\n",
      "1126:\tlearn: 0.9346070\ttotal: 2m 27s\tremaining: 35.6s\n",
      "1127:\tlearn: 0.9345701\ttotal: 2m 27s\tremaining: 35.5s\n",
      "1128:\tlearn: 0.9345295\ttotal: 2m 27s\tremaining: 35.4s\n",
      "1129:\tlearn: 0.9344908\ttotal: 2m 27s\tremaining: 35.3s\n",
      "1130:\tlearn: 0.9344546\ttotal: 2m 27s\tremaining: 35.1s\n",
      "1131:\tlearn: 0.9344261\ttotal: 2m 27s\tremaining: 35s\n",
      "1132:\tlearn: 0.9344222\ttotal: 2m 27s\tremaining: 34.9s\n",
      "1133:\tlearn: 0.9344028\ttotal: 2m 28s\tremaining: 34.7s\n",
      "1134:\tlearn: 0.9343778\ttotal: 2m 28s\tremaining: 34.6s\n",
      "1135:\tlearn: 0.9343430\ttotal: 2m 28s\tremaining: 34.5s\n",
      "1136:\tlearn: 0.9343312\ttotal: 2m 28s\tremaining: 34.3s\n",
      "1137:\tlearn: 0.9342981\ttotal: 2m 28s\tremaining: 34.2s\n",
      "1138:\tlearn: 0.9342963\ttotal: 2m 28s\tremaining: 34.1s\n",
      "1139:\tlearn: 0.9342555\ttotal: 2m 28s\tremaining: 34s\n",
      "1140:\tlearn: 0.9342344\ttotal: 2m 29s\tremaining: 33.8s\n",
      "1141:\tlearn: 0.9342184\ttotal: 2m 29s\tremaining: 33.7s\n",
      "1142:\tlearn: 0.9341775\ttotal: 2m 29s\tremaining: 33.6s\n",
      "1143:\tlearn: 0.9341305\ttotal: 2m 29s\tremaining: 33.5s\n",
      "1144:\tlearn: 0.9341102\ttotal: 2m 29s\tremaining: 33.3s\n",
      "1145:\tlearn: 0.9340866\ttotal: 2m 29s\tremaining: 33.2s\n",
      "1146:\tlearn: 0.9340754\ttotal: 2m 29s\tremaining: 33.1s\n",
      "1147:\tlearn: 0.9340583\ttotal: 2m 30s\tremaining: 33s\n",
      "1148:\tlearn: 0.9340174\ttotal: 2m 30s\tremaining: 32.8s\n",
      "1149:\tlearn: 0.9339731\ttotal: 2m 30s\tremaining: 32.7s\n",
      "1150:\tlearn: 0.9339422\ttotal: 2m 30s\tremaining: 32.6s\n",
      "1151:\tlearn: 0.9339088\ttotal: 2m 30s\tremaining: 32.5s\n",
      "1152:\tlearn: 0.9338898\ttotal: 2m 30s\tremaining: 32.3s\n",
      "1153:\tlearn: 0.9338609\ttotal: 2m 31s\tremaining: 32.2s\n",
      "1154:\tlearn: 0.9338354\ttotal: 2m 31s\tremaining: 32.1s\n",
      "1155:\tlearn: 0.9338046\ttotal: 2m 31s\tremaining: 31.9s\n",
      "1156:\tlearn: 0.9337890\ttotal: 2m 31s\tremaining: 31.8s\n",
      "1157:\tlearn: 0.9337479\ttotal: 2m 31s\tremaining: 31.7s\n",
      "1158:\tlearn: 0.9337302\ttotal: 2m 31s\tremaining: 31.5s\n",
      "1159:\tlearn: 0.9336948\ttotal: 2m 31s\tremaining: 31.4s\n",
      "1160:\tlearn: 0.9336807\ttotal: 2m 31s\tremaining: 31.3s\n",
      "1161:\tlearn: 0.9336340\ttotal: 2m 32s\tremaining: 31.1s\n",
      "1162:\tlearn: 0.9335971\ttotal: 2m 32s\tremaining: 31s\n",
      "1163:\tlearn: 0.9335778\ttotal: 2m 32s\tremaining: 30.9s\n",
      "1164:\tlearn: 0.9335582\ttotal: 2m 32s\tremaining: 30.7s\n",
      "1165:\tlearn: 0.9335180\ttotal: 2m 32s\tremaining: 30.6s\n",
      "1166:\tlearn: 0.9334909\ttotal: 2m 32s\tremaining: 30.5s\n",
      "1167:\tlearn: 0.9334499\ttotal: 2m 32s\tremaining: 30.4s\n",
      "1168:\tlearn: 0.9334225\ttotal: 2m 32s\tremaining: 30.2s\n",
      "1169:\tlearn: 0.9333917\ttotal: 2m 33s\tremaining: 30.1s\n",
      "1170:\tlearn: 0.9333506\ttotal: 2m 33s\tremaining: 30s\n",
      "1171:\tlearn: 0.9333405\ttotal: 2m 33s\tremaining: 29.8s\n",
      "1172:\tlearn: 0.9333102\ttotal: 2m 33s\tremaining: 29.7s\n",
      "1173:\tlearn: 0.9332828\ttotal: 2m 33s\tremaining: 29.6s\n",
      "1174:\tlearn: 0.9332458\ttotal: 2m 33s\tremaining: 29.5s\n",
      "1175:\tlearn: 0.9332153\ttotal: 2m 33s\tremaining: 29.3s\n",
      "1176:\tlearn: 0.9331812\ttotal: 2m 34s\tremaining: 29.2s\n",
      "1177:\tlearn: 0.9331274\ttotal: 2m 34s\tremaining: 29.1s\n",
      "1178:\tlearn: 0.9330940\ttotal: 2m 34s\tremaining: 28.9s\n",
      "1179:\tlearn: 0.9330629\ttotal: 2m 34s\tremaining: 28.8s\n",
      "1180:\tlearn: 0.9330584\ttotal: 2m 34s\tremaining: 28.7s\n",
      "1181:\tlearn: 0.9330395\ttotal: 2m 34s\tremaining: 28.6s\n",
      "1182:\tlearn: 0.9330210\ttotal: 2m 35s\tremaining: 28.4s\n",
      "1183:\tlearn: 0.9329678\ttotal: 2m 35s\tremaining: 28.3s\n",
      "1184:\tlearn: 0.9329300\ttotal: 2m 35s\tremaining: 28.2s\n",
      "1185:\tlearn: 0.9329098\ttotal: 2m 35s\tremaining: 28.1s\n",
      "1186:\tlearn: 0.9329045\ttotal: 2m 35s\tremaining: 27.9s\n",
      "1187:\tlearn: 0.9328435\ttotal: 2m 35s\tremaining: 27.8s\n",
      "1188:\tlearn: 0.9328224\ttotal: 2m 36s\tremaining: 27.7s\n",
      "1189:\tlearn: 0.9327967\ttotal: 2m 36s\tremaining: 27.6s\n",
      "1190:\tlearn: 0.9327939\ttotal: 2m 36s\tremaining: 27.4s\n",
      "1191:\tlearn: 0.9327643\ttotal: 2m 36s\tremaining: 27.3s\n",
      "1192:\tlearn: 0.9327534\ttotal: 2m 36s\tremaining: 27.2s\n",
      "1193:\tlearn: 0.9327153\ttotal: 2m 36s\tremaining: 27s\n",
      "1194:\tlearn: 0.9326915\ttotal: 2m 36s\tremaining: 26.9s\n",
      "1195:\tlearn: 0.9326636\ttotal: 2m 36s\tremaining: 26.8s\n",
      "1196:\tlearn: 0.9326522\ttotal: 2m 37s\tremaining: 26.6s\n",
      "1197:\tlearn: 0.9326004\ttotal: 2m 37s\tremaining: 26.5s\n",
      "1198:\tlearn: 0.9325814\ttotal: 2m 37s\tremaining: 26.4s\n",
      "1199:\tlearn: 0.9325533\ttotal: 2m 37s\tremaining: 26.3s\n",
      "1200:\tlearn: 0.9325212\ttotal: 2m 37s\tremaining: 26.1s\n",
      "1201:\tlearn: 0.9324831\ttotal: 2m 37s\tremaining: 26s\n",
      "1202:\tlearn: 0.9324480\ttotal: 2m 38s\tremaining: 25.9s\n",
      "1203:\tlearn: 0.9324154\ttotal: 2m 38s\tremaining: 25.7s\n",
      "1204:\tlearn: 0.9323783\ttotal: 2m 38s\tremaining: 25.6s\n",
      "1205:\tlearn: 0.9323314\ttotal: 2m 38s\tremaining: 25.5s\n",
      "1206:\tlearn: 0.9323068\ttotal: 2m 38s\tremaining: 25.4s\n",
      "1207:\tlearn: 0.9322843\ttotal: 2m 38s\tremaining: 25.2s\n",
      "1208:\tlearn: 0.9322535\ttotal: 2m 38s\tremaining: 25.1s\n",
      "1209:\tlearn: 0.9322418\ttotal: 2m 38s\tremaining: 25s\n",
      "1210:\tlearn: 0.9322353\ttotal: 2m 39s\tremaining: 24.8s\n",
      "1211:\tlearn: 0.9322335\ttotal: 2m 39s\tremaining: 24.7s\n",
      "1212:\tlearn: 0.9322040\ttotal: 2m 39s\tremaining: 24.6s\n",
      "1213:\tlearn: 0.9321903\ttotal: 2m 39s\tremaining: 24.4s\n",
      "1214:\tlearn: 0.9321624\ttotal: 2m 39s\tremaining: 24.3s\n",
      "1215:\tlearn: 0.9321230\ttotal: 2m 39s\tremaining: 24.2s\n",
      "1216:\tlearn: 0.9321085\ttotal: 2m 39s\tremaining: 24s\n",
      "1217:\tlearn: 0.9321046\ttotal: 2m 39s\tremaining: 23.9s\n",
      "1218:\tlearn: 0.9320651\ttotal: 2m 40s\tremaining: 23.8s\n",
      "1219:\tlearn: 0.9320412\ttotal: 2m 40s\tremaining: 23.6s\n",
      "1220:\tlearn: 0.9320181\ttotal: 2m 40s\tremaining: 23.5s\n",
      "1221:\tlearn: 0.9320009\ttotal: 2m 40s\tremaining: 23.4s\n",
      "1222:\tlearn: 0.9319719\ttotal: 2m 40s\tremaining: 23.2s\n",
      "1223:\tlearn: 0.9319389\ttotal: 2m 40s\tremaining: 23.1s\n",
      "1224:\tlearn: 0.9319168\ttotal: 2m 40s\tremaining: 23s\n",
      "1225:\tlearn: 0.9318967\ttotal: 2m 41s\tremaining: 22.9s\n",
      "1226:\tlearn: 0.9318709\ttotal: 2m 41s\tremaining: 22.7s\n",
      "1227:\tlearn: 0.9318224\ttotal: 2m 41s\tremaining: 22.6s\n",
      "1228:\tlearn: 0.9318053\ttotal: 2m 41s\tremaining: 22.5s\n",
      "1229:\tlearn: 0.9317807\ttotal: 2m 41s\tremaining: 22.3s\n",
      "1230:\tlearn: 0.9317535\ttotal: 2m 41s\tremaining: 22.2s\n",
      "1231:\tlearn: 0.9317109\ttotal: 2m 41s\tremaining: 22.1s\n",
      "1232:\tlearn: 0.9316766\ttotal: 2m 41s\tremaining: 21.9s\n",
      "1233:\tlearn: 0.9316518\ttotal: 2m 42s\tremaining: 21.8s\n",
      "1234:\tlearn: 0.9316219\ttotal: 2m 42s\tremaining: 21.7s\n",
      "1235:\tlearn: 0.9315959\ttotal: 2m 42s\tremaining: 21.5s\n",
      "1236:\tlearn: 0.9315753\ttotal: 2m 42s\tremaining: 21.4s\n",
      "1237:\tlearn: 0.9315586\ttotal: 2m 42s\tremaining: 21.3s\n",
      "1238:\tlearn: 0.9315286\ttotal: 2m 42s\tremaining: 21.1s\n",
      "1239:\tlearn: 0.9314931\ttotal: 2m 42s\tremaining: 21s\n",
      "1240:\tlearn: 0.9314608\ttotal: 2m 42s\tremaining: 20.9s\n",
      "1241:\tlearn: 0.9314434\ttotal: 2m 42s\tremaining: 20.7s\n",
      "1242:\tlearn: 0.9314240\ttotal: 2m 43s\tremaining: 20.6s\n",
      "1243:\tlearn: 0.9313891\ttotal: 2m 43s\tremaining: 20.5s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1244:\tlearn: 0.9313843\ttotal: 2m 43s\tremaining: 20.3s\n",
      "1245:\tlearn: 0.9313723\ttotal: 2m 43s\tremaining: 20.2s\n",
      "1246:\tlearn: 0.9313638\ttotal: 2m 43s\tremaining: 20.1s\n",
      "1247:\tlearn: 0.9313503\ttotal: 2m 43s\tremaining: 19.9s\n",
      "1248:\tlearn: 0.9312996\ttotal: 2m 43s\tremaining: 19.8s\n",
      "1249:\tlearn: 0.9312877\ttotal: 2m 43s\tremaining: 19.7s\n",
      "1250:\tlearn: 0.9312475\ttotal: 2m 43s\tremaining: 19.5s\n",
      "1251:\tlearn: 0.9312243\ttotal: 2m 44s\tremaining: 19.4s\n",
      "1252:\tlearn: 0.9312035\ttotal: 2m 44s\tremaining: 19.3s\n",
      "1253:\tlearn: 0.9311780\ttotal: 2m 44s\tremaining: 19.1s\n",
      "1254:\tlearn: 0.9311257\ttotal: 2m 44s\tremaining: 19s\n",
      "1255:\tlearn: 0.9310948\ttotal: 2m 44s\tremaining: 18.9s\n",
      "1256:\tlearn: 0.9310393\ttotal: 2m 44s\tremaining: 18.7s\n",
      "1257:\tlearn: 0.9310223\ttotal: 2m 44s\tremaining: 18.6s\n",
      "1258:\tlearn: 0.9310001\ttotal: 2m 44s\tremaining: 18.5s\n",
      "1259:\tlearn: 0.9309760\ttotal: 2m 44s\tremaining: 18.3s\n",
      "1260:\tlearn: 0.9309725\ttotal: 2m 45s\tremaining: 18.2s\n",
      "1261:\tlearn: 0.9309688\ttotal: 2m 45s\tremaining: 18.1s\n",
      "1262:\tlearn: 0.9309526\ttotal: 2m 45s\tremaining: 17.9s\n",
      "1263:\tlearn: 0.9309013\ttotal: 2m 45s\tremaining: 17.8s\n",
      "1264:\tlearn: 0.9308771\ttotal: 2m 45s\tremaining: 17.7s\n",
      "1265:\tlearn: 0.9308579\ttotal: 2m 45s\tremaining: 17.5s\n",
      "1266:\tlearn: 0.9308505\ttotal: 2m 45s\tremaining: 17.4s\n",
      "1267:\tlearn: 0.9308295\ttotal: 2m 45s\tremaining: 17.3s\n",
      "1268:\tlearn: 0.9308137\ttotal: 2m 45s\tremaining: 17.1s\n",
      "1269:\tlearn: 0.9307717\ttotal: 2m 46s\tremaining: 17s\n",
      "1270:\tlearn: 0.9307128\ttotal: 2m 46s\tremaining: 16.9s\n",
      "1271:\tlearn: 0.9306895\ttotal: 2m 46s\tremaining: 16.7s\n",
      "1272:\tlearn: 0.9306626\ttotal: 2m 46s\tremaining: 16.6s\n",
      "1273:\tlearn: 0.9306319\ttotal: 2m 46s\tremaining: 16.5s\n",
      "1274:\tlearn: 0.9305983\ttotal: 2m 46s\tremaining: 16.3s\n",
      "1275:\tlearn: 0.9305732\ttotal: 2m 46s\tremaining: 16.2s\n",
      "1276:\tlearn: 0.9305404\ttotal: 2m 46s\tremaining: 16.1s\n",
      "1277:\tlearn: 0.9305381\ttotal: 2m 46s\tremaining: 15.9s\n",
      "1278:\tlearn: 0.9305023\ttotal: 2m 47s\tremaining: 15.8s\n",
      "1279:\tlearn: 0.9304972\ttotal: 2m 47s\tremaining: 15.7s\n",
      "1280:\tlearn: 0.9304555\ttotal: 2m 47s\tremaining: 15.5s\n",
      "1281:\tlearn: 0.9304236\ttotal: 2m 47s\tremaining: 15.4s\n",
      "1282:\tlearn: 0.9303774\ttotal: 2m 47s\tremaining: 15.3s\n",
      "1283:\tlearn: 0.9303699\ttotal: 2m 47s\tremaining: 15.1s\n",
      "1284:\tlearn: 0.9303348\ttotal: 2m 47s\tremaining: 15s\n",
      "1285:\tlearn: 0.9302984\ttotal: 2m 47s\tremaining: 14.9s\n",
      "1286:\tlearn: 0.9302718\ttotal: 2m 48s\tremaining: 14.8s\n",
      "1287:\tlearn: 0.9302266\ttotal: 2m 48s\tremaining: 14.6s\n",
      "1288:\tlearn: 0.9302104\ttotal: 2m 48s\tremaining: 14.5s\n",
      "1289:\tlearn: 0.9301735\ttotal: 2m 48s\tremaining: 14.4s\n",
      "1290:\tlearn: 0.9301478\ttotal: 2m 48s\tremaining: 14.2s\n",
      "1291:\tlearn: 0.9301116\ttotal: 2m 48s\tremaining: 14.1s\n",
      "1292:\tlearn: 0.9300916\ttotal: 2m 48s\tremaining: 14s\n",
      "1293:\tlearn: 0.9300437\ttotal: 2m 48s\tremaining: 13.8s\n",
      "1294:\tlearn: 0.9300176\ttotal: 2m 49s\tremaining: 13.7s\n",
      "1295:\tlearn: 0.9299951\ttotal: 2m 49s\tremaining: 13.6s\n",
      "1296:\tlearn: 0.9299657\ttotal: 2m 49s\tremaining: 13.4s\n",
      "1297:\tlearn: 0.9299346\ttotal: 2m 49s\tremaining: 13.3s\n",
      "1298:\tlearn: 0.9299165\ttotal: 2m 49s\tremaining: 13.2s\n",
      "1299:\tlearn: 0.9298618\ttotal: 2m 49s\tremaining: 13s\n",
      "1300:\tlearn: 0.9298486\ttotal: 2m 49s\tremaining: 12.9s\n",
      "1301:\tlearn: 0.9298238\ttotal: 2m 49s\tremaining: 12.8s\n",
      "1302:\tlearn: 0.9298054\ttotal: 2m 49s\tremaining: 12.6s\n",
      "1303:\tlearn: 0.9297635\ttotal: 2m 50s\tremaining: 12.5s\n",
      "1304:\tlearn: 0.9297433\ttotal: 2m 50s\tremaining: 12.4s\n",
      "1305:\tlearn: 0.9297091\ttotal: 2m 50s\tremaining: 12.3s\n",
      "1306:\tlearn: 0.9296675\ttotal: 2m 50s\tremaining: 12.1s\n",
      "1307:\tlearn: 0.9296570\ttotal: 2m 50s\tremaining: 12s\n",
      "1308:\tlearn: 0.9296245\ttotal: 2m 50s\tremaining: 11.9s\n",
      "1309:\tlearn: 0.9295835\ttotal: 2m 50s\tremaining: 11.7s\n",
      "1310:\tlearn: 0.9295486\ttotal: 2m 50s\tremaining: 11.6s\n",
      "1311:\tlearn: 0.9295147\ttotal: 2m 50s\tremaining: 11.5s\n",
      "1312:\tlearn: 0.9295079\ttotal: 2m 51s\tremaining: 11.3s\n",
      "1313:\tlearn: 0.9294739\ttotal: 2m 51s\tremaining: 11.2s\n",
      "1314:\tlearn: 0.9294484\ttotal: 2m 51s\tremaining: 11.1s\n",
      "1315:\tlearn: 0.9294115\ttotal: 2m 51s\tremaining: 10.9s\n",
      "1316:\tlearn: 0.9293681\ttotal: 2m 51s\tremaining: 10.8s\n",
      "1317:\tlearn: 0.9293278\ttotal: 2m 51s\tremaining: 10.7s\n",
      "1318:\tlearn: 0.9293043\ttotal: 2m 51s\tremaining: 10.5s\n",
      "1319:\tlearn: 0.9292550\ttotal: 2m 51s\tremaining: 10.4s\n",
      "1320:\tlearn: 0.9292180\ttotal: 2m 52s\tremaining: 10.3s\n",
      "1321:\tlearn: 0.9291784\ttotal: 2m 52s\tremaining: 10.2s\n",
      "1322:\tlearn: 0.9291426\ttotal: 2m 52s\tremaining: 10s\n",
      "1323:\tlearn: 0.9291122\ttotal: 2m 52s\tremaining: 9.9s\n",
      "1324:\tlearn: 0.9291071\ttotal: 2m 52s\tremaining: 9.77s\n",
      "1325:\tlearn: 0.9290689\ttotal: 2m 52s\tremaining: 9.64s\n",
      "1326:\tlearn: 0.9290119\ttotal: 2m 52s\tremaining: 9.51s\n",
      "1327:\tlearn: 0.9289693\ttotal: 2m 53s\tremaining: 9.39s\n",
      "1328:\tlearn: 0.9289194\ttotal: 2m 53s\tremaining: 9.26s\n",
      "1329:\tlearn: 0.9288854\ttotal: 2m 53s\tremaining: 9.13s\n",
      "1330:\tlearn: 0.9288611\ttotal: 2m 53s\tremaining: 9s\n",
      "1331:\tlearn: 0.9288290\ttotal: 2m 53s\tremaining: 8.87s\n",
      "1332:\tlearn: 0.9288007\ttotal: 2m 53s\tremaining: 8.74s\n",
      "1333:\tlearn: 0.9287696\ttotal: 2m 54s\tremaining: 8.61s\n",
      "1334:\tlearn: 0.9287543\ttotal: 2m 54s\tremaining: 8.48s\n",
      "1335:\tlearn: 0.9287300\ttotal: 2m 54s\tremaining: 8.35s\n",
      "1336:\tlearn: 0.9287107\ttotal: 2m 54s\tremaining: 8.22s\n",
      "1337:\tlearn: 0.9286760\ttotal: 2m 54s\tremaining: 8.09s\n",
      "1338:\tlearn: 0.9286481\ttotal: 2m 54s\tremaining: 7.96s\n",
      "1339:\tlearn: 0.9286155\ttotal: 2m 54s\tremaining: 7.83s\n",
      "1340:\tlearn: 0.9285921\ttotal: 2m 55s\tremaining: 7.7s\n",
      "1341:\tlearn: 0.9285447\ttotal: 2m 55s\tremaining: 7.57s\n",
      "1342:\tlearn: 0.9285242\ttotal: 2m 55s\tremaining: 7.44s\n",
      "1343:\tlearn: 0.9285177\ttotal: 2m 55s\tremaining: 7.31s\n",
      "1344:\tlearn: 0.9284865\ttotal: 2m 55s\tremaining: 7.18s\n",
      "1345:\tlearn: 0.9284433\ttotal: 2m 55s\tremaining: 7.05s\n",
      "1346:\tlearn: 0.9284090\ttotal: 2m 55s\tremaining: 6.92s\n",
      "1347:\tlearn: 0.9283724\ttotal: 2m 56s\tremaining: 6.79s\n",
      "1348:\tlearn: 0.9283503\ttotal: 2m 56s\tremaining: 6.66s\n",
      "1349:\tlearn: 0.9283278\ttotal: 2m 56s\tremaining: 6.53s\n",
      "1350:\tlearn: 0.9283081\ttotal: 2m 56s\tremaining: 6.4s\n",
      "1351:\tlearn: 0.9282965\ttotal: 2m 56s\tremaining: 6.27s\n",
      "1352:\tlearn: 0.9282706\ttotal: 2m 56s\tremaining: 6.14s\n",
      "1353:\tlearn: 0.9282499\ttotal: 2m 56s\tremaining: 6.01s\n",
      "1354:\tlearn: 0.9282306\ttotal: 2m 56s\tremaining: 5.88s\n",
      "1355:\tlearn: 0.9281723\ttotal: 2m 57s\tremaining: 5.75s\n",
      "1356:\tlearn: 0.9281259\ttotal: 2m 57s\tremaining: 5.62s\n",
      "1357:\tlearn: 0.9281188\ttotal: 2m 57s\tremaining: 5.49s\n",
      "1358:\tlearn: 0.9281119\ttotal: 2m 57s\tremaining: 5.36s\n",
      "1359:\tlearn: 0.9281031\ttotal: 2m 57s\tremaining: 5.22s\n",
      "1360:\tlearn: 0.9280834\ttotal: 2m 57s\tremaining: 5.09s\n",
      "1361:\tlearn: 0.9280584\ttotal: 2m 57s\tremaining: 4.96s\n",
      "1362:\tlearn: 0.9280158\ttotal: 2m 58s\tremaining: 4.83s\n",
      "1363:\tlearn: 0.9280062\ttotal: 2m 58s\tremaining: 4.7s\n",
      "1364:\tlearn: 0.9279675\ttotal: 2m 58s\tremaining: 4.57s\n",
      "1365:\tlearn: 0.9279263\ttotal: 2m 58s\tremaining: 4.44s\n",
      "1366:\tlearn: 0.9279072\ttotal: 2m 58s\tremaining: 4.31s\n",
      "1367:\tlearn: 0.9278888\ttotal: 2m 58s\tremaining: 4.18s\n",
      "1368:\tlearn: 0.9278715\ttotal: 2m 58s\tremaining: 4.05s\n",
      "1369:\tlearn: 0.9278484\ttotal: 2m 58s\tremaining: 3.92s\n",
      "1370:\tlearn: 0.9278194\ttotal: 2m 58s\tremaining: 3.79s\n",
      "1371:\tlearn: 0.9277788\ttotal: 2m 59s\tremaining: 3.65s\n",
      "1372:\tlearn: 0.9277561\ttotal: 2m 59s\tremaining: 3.52s\n",
      "1373:\tlearn: 0.9277036\ttotal: 2m 59s\tremaining: 3.39s\n",
      "1374:\tlearn: 0.9276734\ttotal: 2m 59s\tremaining: 3.26s\n",
      "1375:\tlearn: 0.9276520\ttotal: 2m 59s\tremaining: 3.13s\n",
      "1376:\tlearn: 0.9276316\ttotal: 2m 59s\tremaining: 3s\n",
      "1377:\tlearn: 0.9275883\ttotal: 2m 59s\tremaining: 2.87s\n",
      "1378:\tlearn: 0.9275613\ttotal: 3m\tremaining: 2.74s\n",
      "1379:\tlearn: 0.9275436\ttotal: 3m\tremaining: 2.61s\n",
      "1380:\tlearn: 0.9275069\ttotal: 3m\tremaining: 2.48s\n",
      "1381:\tlearn: 0.9275015\ttotal: 3m\tremaining: 2.35s\n",
      "1382:\tlearn: 0.9274560\ttotal: 3m\tremaining: 2.22s\n",
      "1383:\tlearn: 0.9274547\ttotal: 3m\tremaining: 2.09s\n",
      "1384:\tlearn: 0.9274197\ttotal: 3m\tremaining: 1.96s\n",
      "1385:\tlearn: 0.9273795\ttotal: 3m\tremaining: 1.83s\n",
      "1386:\tlearn: 0.9273547\ttotal: 3m 1s\tremaining: 1.7s\n",
      "1387:\tlearn: 0.9273117\ttotal: 3m 1s\tremaining: 1.57s\n",
      "1388:\tlearn: 0.9272850\ttotal: 3m 1s\tremaining: 1.44s\n",
      "1389:\tlearn: 0.9272453\ttotal: 3m 1s\tremaining: 1.3s\n",
      "1390:\tlearn: 0.9272232\ttotal: 3m 1s\tremaining: 1.17s\n",
      "1391:\tlearn: 0.9272032\ttotal: 3m 1s\tremaining: 1.04s\n",
      "1392:\tlearn: 0.9271998\ttotal: 3m 1s\tremaining: 914ms\n",
      "1393:\tlearn: 0.9271738\ttotal: 3m 1s\tremaining: 783ms\n",
      "1394:\tlearn: 0.9271304\ttotal: 3m 2s\tremaining: 653ms\n",
      "1395:\tlearn: 0.9271008\ttotal: 3m 2s\tremaining: 522ms\n",
      "1396:\tlearn: 0.9270635\ttotal: 3m 2s\tremaining: 392ms\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1397:\tlearn: 0.9270605\ttotal: 3m 2s\tremaining: 261ms\n",
      "1398:\tlearn: 0.9270310\ttotal: 3m 2s\tremaining: 131ms\n",
      "1399:\tlearn: 0.9270011\ttotal: 3m 2s\tremaining: 0us\n",
      "3\n",
      "4\n"
     ]
    }
   ],
   "source": [
    "xg_oof_train, xg_oof_test = get_oof(xg)\n",
    "# et_oof_train, et_oof_test = get_oof_et(et)\n",
    "# print(1)\n",
    "# rf_oof_train, rf_oof_test = get_oof_et(rf)\n",
    "print(2)\n",
    "\n",
    "\n",
    "cb_oof_train, cb_oof_test = get_oof(cb)\n",
    "print(3)\n",
    "\n",
    "\n",
    "\n",
    "lg_oof_train, lg_oof_test = get_oof(lg)\n",
    "print(4)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "XG-CV: 0.9664293052563901\n",
      "CB-CV: 0.9656478784721583\n",
      "LG-CV: 0.9719571115493982\n",
      "(341424, 3),(146765, 3)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None,\n",
       "         normalize=False)"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"XG-CV: {}\".format(sqrt(mean_squared_error(y_train, xg_oof_train))))\n",
    "#print(\"ET-CV: {}\".format(sqrt(mean_squared_error(y_train.head(100), et_oof_train))))\n",
    "#print(\"RF-CV: {}\".format(sqrt(mean_squared_error(y_train.head(100), rf_oof_train))))\n",
    "print(\"CB-CV: {}\".format(sqrt(mean_squared_error(y_train, cb_oof_train))))\n",
    "print(\"LG-CV: {}\".format(sqrt(mean_squared_error(y_train, lg_oof_train))))\n",
    "\n",
    "\n",
    "x_train = np.concatenate(( xg_oof_train,lg_oof_train, cb_oof_train), axis=1)\n",
    "x_test = np.concatenate((xg_oof_test,lg_oof_test, cb_oof_test), axis=1)\n",
    "\n",
    "print(\"{},{}\".format(x_train.shape, x_test.shape))\n",
    "\n",
    "logistic_regression = LinearRegression()\n",
    "logistic_regression.fit(x_train,y_train)\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred = logistic_regression.predict(x_test)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.38462835, 0.02879436, 0.60768349])"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "logistic_regression.coef_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ensembling it with previous algos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {},
   "outputs": [],
   "source": [
    "sub = pd.read_csv('sample_submission.csv')\n",
    "\n",
    "sub.amount_spent_per_room_night_scaled=pred_ensem*0.8+pred*0.2\n",
    "sub.to_csv('xgb_cat_ensemb.csv',index=None)\n",
    "\n",
    "\n",
    "\n",
    "\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}