covid_19_modeling.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# About this notebook\n",
    "- This notebook is to model the Covid-19 cases for all counties in US based on the county-level demographic and economic data\n",
    "- The Covid-19 data are downloaded from JHU-CSSEGIS (https://github.com/CSSEGISandData/COVID-19)\n",
    "- The county-level demographics and economy datasets are from US Census Bureau"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os, findspark\n",
    "\n",
    "spark_home = os.getenv('SPARK_HOME')\n",
    "findspark.init(spark_home)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "            <div>\n",
       "                <p><b>SparkSession - in-memory</b></p>\n",
       "                \n",
       "        <div>\n",
       "            <p><b>SparkContext</b></p>\n",
       "\n",
       "            <p><a href=\"http://localhost:4040\">Spark UI</a></p>\n",
       "\n",
       "            <dl>\n",
       "              <dt>Version</dt>\n",
       "                <dd><code>v2.4.6</code></dd>\n",
       "              <dt>Master</dt>\n",
       "                <dd><code>local[*]</code></dd>\n",
       "              <dt>AppName</dt>\n",
       "                <dd><code>pyspark-shell</code></dd>\n",
       "            </dl>\n",
       "        </div>\n",
       "        \n",
       "            </div>\n",
       "        "
      ],
      "text/plain": [
       "<pyspark.sql.session.SparkSession at 0x7fa5f8723810>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from pyspark import SparkConf\n",
    "sc = SparkConf()\n",
    "\n",
    "from pyspark.sql import SparkSession\n",
    "spark = SparkSession.builder.config(conf=sc).getOrCreate()\n",
    "spark"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "import json\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\", category=FutureWarning)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyspark import SparkContext\n",
    "from pyspark.sql import Row\n",
    "from pyspark.sql import functions as F\n",
    "from pyspark.sql import types as T\n",
    "from pyspark.sql import Window\n",
    "from pyspark.sql import DataFrame"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Get data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# US Census Bureau data\n",
    "\n",
    "df_population = spark.read.csv('data/PopulationEstimates.csv', header=True, inferSchema=True)\n",
    "df_poverty = spark.read.csv('data/PovertyEstimates.csv', header=True, inferSchema=True)\n",
    "df_unemployment = spark.read.csv('data/Unemployment.csv', header=True, inferSchema=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# US county data\n",
    "\n",
    "df_counties = spark.read.csv('data/Counties.csv', header=True) \\\n",
    "    .select(F.col('X1').alias('state'), F.concat(F.col('X2'), F.col('X3')).alias('FIPS'), F.col('X4').alias('county'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Covid-19 related data, snapshot_date = 2020-08-01\n",
    "\n",
    "df_confirmed = spark.read.csv('data/time_series_covid19_confirmed_US.csv', header=True, inferSchema=True) \\\n",
    "    .fillna(0) \\\n",
    "    .withColumn('FIPS', F.format_string('%05d', F.col('FIPS').cast(T.IntegerType())))\n",
    "\n",
    "df_death = spark.read.csv('data/time_series_covid19_deaths_US.csv', header=True, inferSchema=True) \\\n",
    "    .fillna(0) \\\n",
    "    .filter('Population > 0') \\\n",
    "    .withColumn('FIPS', F.format_string('%05d', F.col('FIPS').cast(T.IntegerType())))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Section 1: Process the Covid-19 and County-Level data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Section 1a: Process the Census data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def process_census_data(df):\n",
    "    \"\"\" process the census data (population, poverty, and unemployment) \"\"\"\n",
    "    \n",
    "    import re\n",
    "    \n",
    "    # for poverty table, need to correct the schema\n",
    "    if 'FIPStxt' in df.columns:\n",
    "        df = df.withColumnRenamed('FIPStxt', 'FIPS')\n",
    "    \n",
    "    # filter only the latest (2018) data\n",
    "    cols_to_keep = [col for col in df.columns if '2018' in col] + ['FIPS']\n",
    "    df = df.select(cols_to_keep)\n",
    "\n",
    "    # fillna\n",
    "    df = df.fillna(0)\n",
    "\n",
    "    # format FIPS column\n",
    "    df = df.withColumn('FIPS', F.format_string('%05d', F.col('FIPS').cast(T.IntegerType())))\n",
    "\n",
    "    # replace comma and dollar-sign in the columns\n",
    "    udf_remove_comma = F.udf(lambda x: re.sub(',', '', str(x)), T.StringType())\n",
    "    udf_remove_dollar = F.udf(lambda x: re.sub('\\$', '',x), T.StringType())\n",
    "\n",
    "    return df.select(*[udf_remove_dollar(udf_remove_comma(col)).alias(col) for col in df.columns])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_population = process_census_data(df_population)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+-----------------+--------------+-----------+-----------+----------------+----------------------+-----------------+------------+-------------+-----------------+------------+------------+------------------+------------------------+-------------------+--------------+-----+\n",
      "|POP_ESTIMATE_2018|N_POP_CHG_2018|Births_2018|Deaths_2018|NATURAL_INC_2018|INTERNATIONAL_MIG_2018|DOMESTIC_MIG_2018|NET_MIG_2018|RESIDUAL_2018|GQ_ESTIMATES_2018|R_birth_2018|R_death_2018|R_NATURAL_INC_2018|R_INTERNATIONAL_MIG_2018|R_DOMESTIC_MIG_2018|R_NET_MIG_2018| FIPS|\n",
      "+-----------------+--------------+-----------+-----------+----------------+----------------------+-----------------+------------+-------------+-----------------+------------+------------+------------------+------------------------+-------------------+--------------+-----+\n",
      "|        327167434|       2020313|    3855500|    2814013|         1041487|                978826|                0|      978826|            0|          8091609|         0.0|         0.0|               0.0|                     0.0|                0.0|           0.0|00000|\n",
      "|          4887871|         12751|      57216|      53425|            3791|                  3344|             5718|        9062|         -102|           117443|        11.7|        10.9|               0.8|                     0.7|                1.2|           1.9|01000|\n",
      "|            55601|           158|        655|        532|             123|                    -2|               37|          35|            0|              455|        11.8|         9.6|               2.2|                     0.0|                0.7|           0.6|01001|\n",
      "|           218022|          5403|       2254|       2228|              26|                   118|             5232|        5350|           27|             2190|        10.5|        10.3|               0.1|                     0.5|               24.3|          24.8|01003|\n",
      "|            24881|          -277|        261|        324|             -63|                    12|             -227|        -215|            1|             2820|        10.4|        12.9|              -2.5|                     0.5|               -9.1|          -8.6|01005|\n",
      "+-----------------+--------------+-----------+-----------+----------------+----------------------+-----------------+------------+-------------+-----------------+------------+------------+------------------+------------------------+-------------------+--------------+-----+\n",
      "only showing top 5 rows\n",
      "\n"
     ]
    }
   ],
   "source": [
    "df_population.show(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_poverty = process_census_data(df_poverty)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+-----------+--------------+--------------+--------------+---------------+---------------+-----------+--------------+--------------+--------------+---------------+---------------+-----------+--------------+--------------+--------------+---------------+---------------+-------------+--------------+--------------+----------+-------------+-------------+-------------+--------------+--------------+-----+\n",
      "|POVALL_2018|CI90LBAll_2018|CI90UBALL_2018|PCTPOVALL_2018|CI90LBALLP_2018|CI90UBALLP_2018|POV017_2018|CI90LB017_2018|CI90UB017_2018|PCTPOV017_2018|CI90LB017P_2018|CI90UB017P_2018|POV517_2018|CI90LB517_2018|CI90UB517_2018|PCTPOV517_2018|CI90LB517P_2018|CI90UB517P_2018|MEDHHINC_2018|CI90LBINC_2018|CI90UBINC_2018|POV04_2018|CI90LB04_2018|CI90UB04_2018|PCTPOV04_2018|CI90LB04P_2018|CI90UB04P_2018| FIPS|\n",
      "+-----------+--------------+--------------+--------------+---------------+---------------+-----------+--------------+--------------+--------------+---------------+---------------+-----------+--------------+--------------+--------------+---------------+---------------+-------------+--------------+--------------+----------+-------------+-------------+-------------+--------------+--------------+-----+\n",
      "|   41852315|      41619366|      42085264|          13.1|           13.0|           13.2|   12997532|      12873127|      13121937|          18.0|           17.8|           18.2|    8930152|       8834521|       9025783|          17.0|           16.8|           17.2|        61937|         61843|         62031|   3758704|      3714862|      3802546|         19.5|          19.3|          19.7|00000|\n",
      "|     801758|        785668|        817848|          16.8|           16.5|           17.1|     255613|        247744|        263482|          23.9|           23.2|           24.6|     178175|        171349|        185001|          22.8|           21.9|           23.7|        49881|         49123|         50639|     73915|        69990|        77840|         26.0|          24.6|          27.4|01000|\n",
      "|       7587|          6334|          8840|          13.8|           11.5|           16.1|       2509|          1965|          3053|          19.3|           15.1|           23.5|       1891|          1469|          2313|          19.5|           15.1|           23.9|        59338|         53628|         65048|      None|         None|         None|          0.0|           0.0|           0.0|01001|\n",
      "|      21069|         17390|         24748|           9.8|            8.1|           11.5|       6442|          4723|          8161|          13.9|           10.2|           17.6|       4534|          3223|          5845|          13.1|            9.3|           16.9|        57588|         54437|         60739|      None|         None|         None|          0.0|           0.0|           0.0|01003|\n",
      "|       6788|          5662|          7914|          30.9|           25.8|           36.0|       2242|          1788|          2696|          43.9|           35.0|           52.8|       1411|          1056|          1766|          36.7|           27.5|           45.9|        34382|         31157|         37607|      None|         None|         None|          0.0|           0.0|           0.0|01005|\n",
      "+-----------+--------------+--------------+--------------+---------------+---------------+-----------+--------------+--------------+--------------+---------------+---------------+-----------+--------------+--------------+--------------+---------------+---------------+-------------+--------------+--------------+----------+-------------+-------------+-------------+--------------+--------------+-----+\n",
      "only showing top 5 rows\n",
      "\n"
     ]
    }
   ],
   "source": [
    "df_poverty.show(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_unemployment = process_census_data(df_unemployment)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+-------------------------+-------------+---------------+----------------------+----------------------------+-----------------------------------------+-----+\n",
      "|Civilian_labor_force_2018|Employed_2018|Unemployed_2018|Unemployment_rate_2018|Median_Household_Income_2018|Med_HH_Income_Percent_of_State_Total_2018| FIPS|\n",
      "+-------------------------+-------------+---------------+----------------------+----------------------------+-----------------------------------------+-----+\n",
      "|                161640488|    155344043|        6296445|                   3.9|                      61937 |                                      0.0|00000|\n",
      "|                  2198837|      2112347|          86490|                   3.9|                      49881 |                                    100.0|01000|\n",
      "|                    25957|        25015|            942|                   3.6|                      59338 |                                    119.0|01001|\n",
      "|                    93849|        90456|           3393|                   3.6|                      57588 |                                    115.5|01003|\n",
      "|                     8373|         7940|            433|                   5.2|                      34382 |                                     68.9|01005|\n",
      "+-------------------------+-------------+---------------+----------------------+----------------------------+-----------------------------------------+-----+\n",
      "only showing top 5 rows\n",
      "\n"
     ]
    }
   ],
   "source": [
    "df_unemployment.show(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Section 1b: Process the Covid-19 data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# join df_confirmed and df_death dataframes to make a master table\n",
    "\n",
    "df_us = df_confirmed.select('FIPS', 'Combined_Key', F.col(df_confirmed.columns[-1]).alias('confirmed_cases')) \\\n",
    ".join(df_death.select('FIPS', 'Combined_Key', 'Population', F.col(df_death.columns[-1]).alias('death_cases')),\n",
    "      on=['FIPS', 'Combined_Key'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# compute infection rates\n",
    "df_us = df_us \\\n",
    "    .withColumn('Infection_Rate', F.round(F.col('confirmed_cases')/F.col('Population'), 5)) \\\n",
    "    .withColumn('Fatality_Rate', F.round(F.col('death_cases')/F.col('confirmed_cases'), 5)) \\\n",
    "    .withColumnRenamed('Combined_Key', 'County')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# join with df_counties to get the state column\n",
    "\n",
    "df_us = df_us \\\n",
    "    .join(df_counties.select('FIPS', 'state'), on='FIPS')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "root\n",
      " |-- FIPS: string (nullable = false)\n",
      " |-- County: string (nullable = true)\n",
      " |-- confirmed_cases: integer (nullable = true)\n",
      " |-- Population: integer (nullable = true)\n",
      " |-- death_cases: integer (nullable = true)\n",
      " |-- Infection_Rate: double (nullable = true)\n",
      " |-- Fatality_Rate: double (nullable = true)\n",
      " |-- state: string (nullable = true)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "df_us.printSchema()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Section 1c: Join the Covid-19 table and the County-level data table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_us = df_us \\\n",
    "    .join(df_population, on='FIPS') \\\n",
    "    .join(df_poverty, on='FIPS') \\\n",
    "    .join(df_unemployment, on='FIPS')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "root\n",
      " |-- FIPS: string (nullable = false)\n",
      " |-- County: string (nullable = true)\n",
      " |-- confirmed_cases: integer (nullable = true)\n",
      " |-- Population: integer (nullable = true)\n",
      " |-- death_cases: integer (nullable = true)\n",
      " |-- Infection_Rate: double (nullable = true)\n",
      " |-- Fatality_Rate: double (nullable = true)\n",
      " |-- state: string (nullable = true)\n",
      " |-- POP_ESTIMATE_2018: string (nullable = true)\n",
      " |-- N_POP_CHG_2018: string (nullable = true)\n",
      " |-- Births_2018: string (nullable = true)\n",
      " |-- Deaths_2018: string (nullable = true)\n",
      " |-- NATURAL_INC_2018: string (nullable = true)\n",
      " |-- INTERNATIONAL_MIG_2018: string (nullable = true)\n",
      " |-- DOMESTIC_MIG_2018: string (nullable = true)\n",
      " |-- NET_MIG_2018: string (nullable = true)\n",
      " |-- RESIDUAL_2018: string (nullable = true)\n",
      " |-- GQ_ESTIMATES_2018: string (nullable = true)\n",
      " |-- R_birth_2018: string (nullable = true)\n",
      " |-- R_death_2018: string (nullable = true)\n",
      " |-- R_NATURAL_INC_2018: string (nullable = true)\n",
      " |-- R_INTERNATIONAL_MIG_2018: string (nullable = true)\n",
      " |-- R_DOMESTIC_MIG_2018: string (nullable = true)\n",
      " |-- R_NET_MIG_2018: string (nullable = true)\n",
      " |-- POVALL_2018: string (nullable = true)\n",
      " |-- CI90LBAll_2018: string (nullable = true)\n",
      " |-- CI90UBALL_2018: string (nullable = true)\n",
      " |-- PCTPOVALL_2018: string (nullable = true)\n",
      " |-- CI90LBALLP_2018: string (nullable = true)\n",
      " |-- CI90UBALLP_2018: string (nullable = true)\n",
      " |-- POV017_2018: string (nullable = true)\n",
      " |-- CI90LB017_2018: string (nullable = true)\n",
      " |-- CI90UB017_2018: string (nullable = true)\n",
      " |-- PCTPOV017_2018: string (nullable = true)\n",
      " |-- CI90LB017P_2018: string (nullable = true)\n",
      " |-- CI90UB017P_2018: string (nullable = true)\n",
      " |-- POV517_2018: string (nullable = true)\n",
      " |-- CI90LB517_2018: string (nullable = true)\n",
      " |-- CI90UB517_2018: string (nullable = true)\n",
      " |-- PCTPOV517_2018: string (nullable = true)\n",
      " |-- CI90LB517P_2018: string (nullable = true)\n",
      " |-- CI90UB517P_2018: string (nullable = true)\n",
      " |-- MEDHHINC_2018: string (nullable = true)\n",
      " |-- CI90LBINC_2018: string (nullable = true)\n",
      " |-- CI90UBINC_2018: string (nullable = true)\n",
      " |-- POV04_2018: string (nullable = true)\n",
      " |-- CI90LB04_2018: string (nullable = true)\n",
      " |-- CI90UB04_2018: string (nullable = true)\n",
      " |-- PCTPOV04_2018: string (nullable = true)\n",
      " |-- CI90LB04P_2018: string (nullable = true)\n",
      " |-- CI90UB04P_2018: string (nullable = true)\n",
      " |-- Civilian_labor_force_2018: string (nullable = true)\n",
      " |-- Employed_2018: string (nullable = true)\n",
      " |-- Unemployed_2018: string (nullable = true)\n",
      " |-- Unemployment_rate_2018: string (nullable = true)\n",
      " |-- Median_Household_Income_2018: string (nullable = true)\n",
      " |-- Med_HH_Income_Percent_of_State_Total_2018: string (nullable = true)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "df_us.printSchema()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# convert this master table to Pandas for futher modeling\n",
    "\n",
    "pd_us = df_us.toPandas()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Section 2: Feature Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# replace 'None' to 0\n",
    "\n",
    "pd_us = pd_us.replace('None', 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# fillna\n",
    "\n",
    "pd_us = pd_us.fillna(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# convert the data type into float\n",
    "\n",
    "X = pd_us.iloc[:, 8:].astype('double')\n",
    "Y = pd_us['confirmed_cases']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>POP_ESTIMATE_2018</th>\n",
       "      <th>N_POP_CHG_2018</th>\n",
       "      <th>Births_2018</th>\n",
       "      <th>Deaths_2018</th>\n",
       "      <th>NATURAL_INC_2018</th>\n",
       "      <th>INTERNATIONAL_MIG_2018</th>\n",
       "      <th>DOMESTIC_MIG_2018</th>\n",
       "      <th>NET_MIG_2018</th>\n",
       "      <th>RESIDUAL_2018</th>\n",
       "      <th>GQ_ESTIMATES_2018</th>\n",
       "      <th>...</th>\n",
       "      <th>CI90UB04_2018</th>\n",
       "      <th>PCTPOV04_2018</th>\n",
       "      <th>CI90LB04P_2018</th>\n",
       "      <th>CI90UB04P_2018</th>\n",
       "      <th>Civilian_labor_force_2018</th>\n",
       "      <th>Employed_2018</th>\n",
       "      <th>Unemployed_2018</th>\n",
       "      <th>Unemployment_rate_2018</th>\n",
       "      <th>Median_Household_Income_2018</th>\n",
       "      <th>Med_HH_Income_Percent_of_State_Total_2018</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>55601.0</td>\n",
       "      <td>158.0</td>\n",
       "      <td>655.0</td>\n",
       "      <td>532.0</td>\n",
       "      <td>123.0</td>\n",
       "      <td>-2.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>455.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25957.0</td>\n",
       "      <td>25015.0</td>\n",
       "      <td>942.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>59338.0</td>\n",
       "      <td>119.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>218022.0</td>\n",
       "      <td>5403.0</td>\n",
       "      <td>2254.0</td>\n",
       "      <td>2228.0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>118.0</td>\n",
       "      <td>5232.0</td>\n",
       "      <td>5350.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>2190.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>93849.0</td>\n",
       "      <td>90456.0</td>\n",
       "      <td>3393.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>57588.0</td>\n",
       "      <td>115.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>24881.0</td>\n",
       "      <td>-277.0</td>\n",
       "      <td>261.0</td>\n",
       "      <td>324.0</td>\n",
       "      <td>-63.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>-227.0</td>\n",
       "      <td>-215.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2820.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>8373.0</td>\n",
       "      <td>7940.0</td>\n",
       "      <td>433.0</td>\n",
       "      <td>5.2</td>\n",
       "      <td>34382.0</td>\n",
       "      <td>68.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>22400.0</td>\n",
       "      <td>-155.0</td>\n",
       "      <td>250.0</td>\n",
       "      <td>256.0</td>\n",
       "      <td>-6.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>-158.0</td>\n",
       "      <td>-148.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>2151.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>8661.0</td>\n",
       "      <td>8317.0</td>\n",
       "      <td>344.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>46064.0</td>\n",
       "      <td>92.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>57840.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>681.0</td>\n",
       "      <td>662.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>-10.0</td>\n",
       "      <td>-2.0</td>\n",
       "      <td>-4.0</td>\n",
       "      <td>489.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25006.0</td>\n",
       "      <td>24128.0</td>\n",
       "      <td>878.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>50412.0</td>\n",
       "      <td>101.1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 49 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   POP_ESTIMATE_2018  N_POP_CHG_2018  Births_2018  Deaths_2018  \\\n",
       "0            55601.0           158.0        655.0        532.0   \n",
       "1           218022.0          5403.0       2254.0       2228.0   \n",
       "2            24881.0          -277.0        261.0        324.0   \n",
       "3            22400.0          -155.0        250.0        256.0   \n",
       "4            57840.0            13.0        681.0        662.0   \n",
       "\n",
       "   NATURAL_INC_2018  INTERNATIONAL_MIG_2018  DOMESTIC_MIG_2018  NET_MIG_2018  \\\n",
       "0             123.0                    -2.0               37.0          35.0   \n",
       "1              26.0                   118.0             5232.0        5350.0   \n",
       "2             -63.0                    12.0             -227.0        -215.0   \n",
       "3              -6.0                    10.0             -158.0        -148.0   \n",
       "4              19.0                     8.0              -10.0          -2.0   \n",
       "\n",
       "   RESIDUAL_2018  GQ_ESTIMATES_2018  ...  CI90UB04_2018  PCTPOV04_2018  \\\n",
       "0            0.0              455.0  ...            0.0            0.0   \n",
       "1           27.0             2190.0  ...            0.0            0.0   \n",
       "2            1.0             2820.0  ...            0.0            0.0   \n",
       "3           -1.0             2151.0  ...            0.0            0.0   \n",
       "4           -4.0              489.0  ...            0.0            0.0   \n",
       "\n",
       "   CI90LB04P_2018  CI90UB04P_2018  Civilian_labor_force_2018  Employed_2018  \\\n",
       "0             0.0             0.0                    25957.0        25015.0   \n",
       "1             0.0             0.0                    93849.0        90456.0   \n",
       "2             0.0             0.0                     8373.0         7940.0   \n",
       "3             0.0             0.0                     8661.0         8317.0   \n",
       "4             0.0             0.0                    25006.0        24128.0   \n",
       "\n",
       "   Unemployed_2018  Unemployment_rate_2018  Median_Household_Income_2018  \\\n",
       "0            942.0                     3.6                       59338.0   \n",
       "1           3393.0                     3.6                       57588.0   \n",
       "2            433.0                     5.2                       34382.0   \n",
       "3            344.0                     4.0                       46064.0   \n",
       "4            878.0                     3.5                       50412.0   \n",
       "\n",
       "   Med_HH_Income_Percent_of_State_Total_2018  \n",
       "0                                      119.0  \n",
       "1                                      115.5  \n",
       "2                                       68.9  \n",
       "3                                       92.3  \n",
       "4                                      101.1  \n",
       "\n",
       "[5 rows x 49 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     972\n",
       "1    3056\n",
       "2     550\n",
       "3     355\n",
       "4     685\n",
       "Name: confirmed_cases, dtype: int32"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# feature correlation\n",
    "\n",
    "corr = X.corr()\n",
    "\n",
    "fig, ax = plt.subplots(1, 1, figsize=(15, 15))\n",
    "ax = sns.heatmap(corr, vmin=-1, vmax=1, center=0, cmap=sns.diverging_palette(20, 200, n=200), square=True);\n",
    "ax.set_xticklabels(ax.get_xticklabels(), rotation=45, horizontalalignment='right');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Section 3: Feature Selection via LASSO Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import Lasso\n",
    "from sklearn.feature_selection import SelectFromModel\n",
    "from sklearn.preprocessing import StandardScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# scale the data\n",
    "\n",
    "scaler = StandardScaler()\n",
    "X[X.columns] = pd.DataFrame(scaler.fit_transform(X[X.columns]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2466, 49) (617, 49)\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X[Y>0], Y[Y>0], test_size=0.2, random_state=42)\n",
    "\n",
    "print(X_train.shape, X_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SelectFromModel(estimator=Lasso(alpha=0.1))"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# feature selection via Lasso regression\n",
    "\n",
    "sel_ = SelectFromModel(Lasso(alpha=0.1))\n",
    "sel_.fit(X_train.values, np.log10(y_train.values))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "total features: 49\n",
      "selected features: 9\n",
      "features with coefficients shrank to zero: 40\n"
     ]
    }
   ],
   "source": [
    "# get features that are kept by Lasso regression\n",
    "\n",
    "selected_feat = X_train.columns[(sel_.get_support())]\n",
    "print('total features: {}'.format((X_train.shape[1])))\n",
    "print('selected features: {}'.format(len(selected_feat)))\n",
    "print('features with coefficients shrank to zero: {}'.format(np.sum(sel_.estimator_.coef_ == 0)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['POP_ESTIMATE_2018', 'Births_2018', 'NATURAL_INC_2018',\n",
       "       'INTERNATIONAL_MIG_2018', 'DOMESTIC_MIG_2018', 'NET_MIG_2018',\n",
       "       'RESIDUAL_2018', 'R_death_2018', 'R_INTERNATIONAL_MIG_2018',\n",
       "       'R_DOMESTIC_MIG_2018', 'R_NET_MIG_2018', 'POVALL_2018',\n",
       "       'CI90LBAll_2018', 'CI90UBALL_2018', 'PCTPOVALL_2018', 'CI90UBALLP_2018',\n",
       "       'POV017_2018', 'CI90LB017_2018', 'CI90UB017_2018', 'PCTPOV017_2018',\n",
       "       'CI90LB017P_2018', 'CI90UB017P_2018', 'POV517_2018', 'CI90LB517_2018',\n",
       "       'CI90UB517_2018', 'PCTPOV517_2018', 'CI90UB517P_2018', 'MEDHHINC_2018',\n",
       "       'CI90UBINC_2018', 'POV04_2018', 'CI90LB04_2018', 'CI90UB04_2018',\n",
       "       'PCTPOV04_2018', 'CI90LB04P_2018', 'CI90UB04P_2018',\n",
       "       'Civilian_labor_force_2018', 'Employed_2018', 'Unemployed_2018',\n",
       "       'Unemployment_rate_2018', 'Median_Household_Income_2018'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "removed_feats = X_train.columns[(sel_.estimator_.coef_ == 0).ravel().tolist()]\n",
    "removed_feats"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Section 5: XGBoost with selected features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# select features\n",
    "\n",
    "X_train_sel = X_train[selected_feat]\n",
    "X_test_sel = X_test[selected_feat]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "from xgboost import XGBRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "# set up the XGBRegressor and the parameter space\n",
    "\n",
    "reg = XGBRegressor(n_estimators=500)\n",
    "\n",
    "param_dist = {\n",
    "    'max_depth' : np.arange(5, 10),\n",
    "    'colsample_bytree' : np.arange(0.6, 1.0, 1),\n",
    "    'subsample' : np.arange(0.6, 1.0, 1),\n",
    "    'eta' : np.logspace(-2, 0, 10)\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "# set up how many random searches to do\n",
    "\n",
    "n_iter_search = 20"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "# set up the random search\n",
    "\n",
    "random_search = RandomizedSearchCV(\n",
    "    reg,\n",
    "    param_distributions=param_dist,\n",
    "    n_iter=n_iter_search,\n",
    "    cv=5,\n",
    "    scoring='neg_mean_squared_error'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[08:51:58] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79633\tvalidation_1-rmse:1.78481\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11267\tvalidation_1-rmse:1.12478\n",
      "[10]\tvalidation_0-rmse:0.719584\tvalidation_1-rmse:0.747908\n",
      "[15]\tvalidation_0-rmse:0.502846\tvalidation_1-rmse:0.553392\n",
      "[20]\tvalidation_0-rmse:0.381153\tvalidation_1-rmse:0.45222\n",
      "[25]\tvalidation_0-rmse:0.320808\tvalidation_1-rmse:0.404822\n",
      "[30]\tvalidation_0-rmse:0.288117\tvalidation_1-rmse:0.383866\n",
      "[35]\tvalidation_0-rmse:0.267896\tvalidation_1-rmse:0.375284\n",
      "[40]\tvalidation_0-rmse:0.257185\tvalidation_1-rmse:0.37179\n",
      "[45]\tvalidation_0-rmse:0.247455\tvalidation_1-rmse:0.369188\n",
      "[50]\tvalidation_0-rmse:0.239644\tvalidation_1-rmse:0.367232\n",
      "[55]\tvalidation_0-rmse:0.233112\tvalidation_1-rmse:0.366494\n",
      "[60]\tvalidation_0-rmse:0.226115\tvalidation_1-rmse:0.367319\n",
      "Stopping. Best iteration:\n",
      "[53]\tvalidation_0-rmse:0.236435\tvalidation_1-rmse:0.366117\n",
      "\n",
      "[08:51:59] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79802\tvalidation_1-rmse:1.78666\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12703\tvalidation_1-rmse:1.12841\n",
      "[10]\tvalidation_0-rmse:0.731216\tvalidation_1-rmse:0.752816\n",
      "[15]\tvalidation_0-rmse:0.509245\tvalidation_1-rmse:0.553976\n",
      "[20]\tvalidation_0-rmse:0.392708\tvalidation_1-rmse:0.454472\n",
      "[25]\tvalidation_0-rmse:0.32713\tvalidation_1-rmse:0.405447\n",
      "[30]\tvalidation_0-rmse:0.290562\tvalidation_1-rmse:0.385488\n",
      "[35]\tvalidation_0-rmse:0.268829\tvalidation_1-rmse:0.375696\n",
      "[40]\tvalidation_0-rmse:0.255212\tvalidation_1-rmse:0.372033\n",
      "[45]\tvalidation_0-rmse:0.245665\tvalidation_1-rmse:0.370214\n",
      "[50]\tvalidation_0-rmse:0.235106\tvalidation_1-rmse:0.368119\n",
      "[55]\tvalidation_0-rmse:0.227811\tvalidation_1-rmse:0.368036\n",
      "[60]\tvalidation_0-rmse:0.221184\tvalidation_1-rmse:0.366143\n",
      "[65]\tvalidation_0-rmse:0.215561\tvalidation_1-rmse:0.365786\n",
      "[70]\tvalidation_0-rmse:0.210836\tvalidation_1-rmse:0.366036\n",
      "[75]\tvalidation_0-rmse:0.206637\tvalidation_1-rmse:0.364676\n",
      "[80]\tvalidation_0-rmse:0.202277\tvalidation_1-rmse:0.365791\n",
      "[85]\tvalidation_0-rmse:0.198729\tvalidation_1-rmse:0.365465\n",
      "Stopping. Best iteration:\n",
      "[76]\tvalidation_0-rmse:0.205995\tvalidation_1-rmse:0.36447\n",
      "\n",
      "[08:51:59] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79793\tvalidation_1-rmse:1.78615\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.1285\tvalidation_1-rmse:1.13328\n",
      "[10]\tvalidation_0-rmse:0.734452\tvalidation_1-rmse:0.75505\n",
      "[15]\tvalidation_0-rmse:0.512026\tvalidation_1-rmse:0.553501\n",
      "[20]\tvalidation_0-rmse:0.392906\tvalidation_1-rmse:0.453201\n",
      "[25]\tvalidation_0-rmse:0.329231\tvalidation_1-rmse:0.405877\n",
      "[30]\tvalidation_0-rmse:0.292959\tvalidation_1-rmse:0.383965\n",
      "[35]\tvalidation_0-rmse:0.272138\tvalidation_1-rmse:0.374618\n",
      "[40]\tvalidation_0-rmse:0.258356\tvalidation_1-rmse:0.370649\n",
      "[45]\tvalidation_0-rmse:0.247574\tvalidation_1-rmse:0.369283\n",
      "[50]\tvalidation_0-rmse:0.238362\tvalidation_1-rmse:0.368488\n",
      "[55]\tvalidation_0-rmse:0.231988\tvalidation_1-rmse:0.368137\n",
      "[60]\tvalidation_0-rmse:0.226547\tvalidation_1-rmse:0.367806\n",
      "[65]\tvalidation_0-rmse:0.222468\tvalidation_1-rmse:0.368153\n",
      "[70]\tvalidation_0-rmse:0.217389\tvalidation_1-rmse:0.368405\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.223594\tvalidation_1-rmse:0.367794\n",
      "\n",
      "[08:51:59] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79626\tvalidation_1-rmse:1.78503\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12897\tvalidation_1-rmse:1.13456\n",
      "[10]\tvalidation_0-rmse:0.734761\tvalidation_1-rmse:0.757676\n",
      "[15]\tvalidation_0-rmse:0.513857\tvalidation_1-rmse:0.559246\n",
      "[20]\tvalidation_0-rmse:0.394566\tvalidation_1-rmse:0.461915\n",
      "[25]\tvalidation_0-rmse:0.328424\tvalidation_1-rmse:0.415648\n",
      "[30]\tvalidation_0-rmse:0.29186\tvalidation_1-rmse:0.393342\n",
      "[35]\tvalidation_0-rmse:0.2698\tvalidation_1-rmse:0.383848\n",
      "[40]\tvalidation_0-rmse:0.254487\tvalidation_1-rmse:0.379983\n",
      "[45]\tvalidation_0-rmse:0.245195\tvalidation_1-rmse:0.378546\n",
      "[50]\tvalidation_0-rmse:0.235738\tvalidation_1-rmse:0.377624\n",
      "[55]\tvalidation_0-rmse:0.229972\tvalidation_1-rmse:0.377193\n",
      "[60]\tvalidation_0-rmse:0.224341\tvalidation_1-rmse:0.376848\n",
      "[65]\tvalidation_0-rmse:0.218174\tvalidation_1-rmse:0.375843\n",
      "[70]\tvalidation_0-rmse:0.213275\tvalidation_1-rmse:0.376371\n",
      "[75]\tvalidation_0-rmse:0.209558\tvalidation_1-rmse:0.376197\n",
      "[80]\tvalidation_0-rmse:0.205212\tvalidation_1-rmse:0.376944\n",
      "Stopping. Best iteration:\n",
      "[73]\tvalidation_0-rmse:0.210994\tvalidation_1-rmse:0.375669\n",
      "\n",
      "[08:51:59] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.796\tvalidation_1-rmse:1.78471\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12713\tvalidation_1-rmse:1.13236\n",
      "[10]\tvalidation_0-rmse:0.731847\tvalidation_1-rmse:0.754436\n",
      "[15]\tvalidation_0-rmse:0.509207\tvalidation_1-rmse:0.55537\n",
      "[20]\tvalidation_0-rmse:0.391151\tvalidation_1-rmse:0.45954\n",
      "[25]\tvalidation_0-rmse:0.327464\tvalidation_1-rmse:0.414048\n",
      "[30]\tvalidation_0-rmse:0.292292\tvalidation_1-rmse:0.392056\n",
      "[35]\tvalidation_0-rmse:0.273116\tvalidation_1-rmse:0.384421\n",
      "[40]\tvalidation_0-rmse:0.260679\tvalidation_1-rmse:0.381161\n",
      "[45]\tvalidation_0-rmse:0.251198\tvalidation_1-rmse:0.380135\n",
      "[50]\tvalidation_0-rmse:0.243661\tvalidation_1-rmse:0.379609\n",
      "[55]\tvalidation_0-rmse:0.237119\tvalidation_1-rmse:0.378695\n",
      "[60]\tvalidation_0-rmse:0.232269\tvalidation_1-rmse:0.379214\n",
      "Stopping. Best iteration:\n",
      "[54]\tvalidation_0-rmse:0.238061\tvalidation_1-rmse:0.378574\n",
      "\n",
      "[08:52:00] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79614\tvalidation_1-rmse:1.78407\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11374\tvalidation_1-rmse:1.11994\n",
      "[10]\tvalidation_0-rmse:0.724631\tvalidation_1-rmse:0.743554\n",
      "[15]\tvalidation_0-rmse:0.514239\tvalidation_1-rmse:0.549126\n",
      "[20]\tvalidation_0-rmse:0.400268\tvalidation_1-rmse:0.449426\n",
      "[25]\tvalidation_0-rmse:0.345923\tvalidation_1-rmse:0.404649\n",
      "[30]\tvalidation_0-rmse:0.317994\tvalidation_1-rmse:0.386144\n",
      "[35]\tvalidation_0-rmse:0.30252\tvalidation_1-rmse:0.376388\n",
      "[40]\tvalidation_0-rmse:0.294863\tvalidation_1-rmse:0.373909\n",
      "[45]\tvalidation_0-rmse:0.286019\tvalidation_1-rmse:0.371415\n",
      "[50]\tvalidation_0-rmse:0.279818\tvalidation_1-rmse:0.368649\n",
      "[55]\tvalidation_0-rmse:0.274012\tvalidation_1-rmse:0.369104\n",
      "[60]\tvalidation_0-rmse:0.268911\tvalidation_1-rmse:0.368751\n",
      "[65]\tvalidation_0-rmse:0.26435\tvalidation_1-rmse:0.370181\n",
      "Stopping. Best iteration:\n",
      "[58]\tvalidation_0-rmse:0.271137\tvalidation_1-rmse:0.368415\n",
      "\n",
      "[08:52:00] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79818\tvalidation_1-rmse:1.78572\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12911\tvalidation_1-rmse:1.12532\n",
      "[10]\tvalidation_0-rmse:0.736762\tvalidation_1-rmse:0.745403\n",
      "[15]\tvalidation_0-rmse:0.521134\tvalidation_1-rmse:0.545752\n",
      "[20]\tvalidation_0-rmse:0.411941\tvalidation_1-rmse:0.44852\n",
      "[25]\tvalidation_0-rmse:0.353254\tvalidation_1-rmse:0.401825\n",
      "[30]\tvalidation_0-rmse:0.32202\tvalidation_1-rmse:0.380904\n",
      "[35]\tvalidation_0-rmse:0.303861\tvalidation_1-rmse:0.370174\n",
      "[40]\tvalidation_0-rmse:0.292761\tvalidation_1-rmse:0.365527\n",
      "[45]\tvalidation_0-rmse:0.285199\tvalidation_1-rmse:0.362647\n",
      "[50]\tvalidation_0-rmse:0.276477\tvalidation_1-rmse:0.360668\n",
      "[55]\tvalidation_0-rmse:0.270243\tvalidation_1-rmse:0.360342\n",
      "[60]\tvalidation_0-rmse:0.26348\tvalidation_1-rmse:0.358898\n",
      "[65]\tvalidation_0-rmse:0.258888\tvalidation_1-rmse:0.358917\n",
      "[70]\tvalidation_0-rmse:0.254146\tvalidation_1-rmse:0.359126\n",
      "Stopping. Best iteration:\n",
      "[60]\tvalidation_0-rmse:0.26348\tvalidation_1-rmse:0.358898\n",
      "\n",
      "[08:52:00] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79776\tvalidation_1-rmse:1.7853\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13043\tvalidation_1-rmse:1.1321\n",
      "[10]\tvalidation_0-rmse:0.74032\tvalidation_1-rmse:0.753576\n",
      "[15]\tvalidation_0-rmse:0.523896\tvalidation_1-rmse:0.550943\n",
      "[20]\tvalidation_0-rmse:0.41136\tvalidation_1-rmse:0.453043\n",
      "[25]\tvalidation_0-rmse:0.355327\tvalidation_1-rmse:0.405661\n",
      "[30]\tvalidation_0-rmse:0.32538\tvalidation_1-rmse:0.386155\n",
      "[35]\tvalidation_0-rmse:0.308408\tvalidation_1-rmse:0.374369\n",
      "[40]\tvalidation_0-rmse:0.297905\tvalidation_1-rmse:0.370118\n",
      "[45]\tvalidation_0-rmse:0.288829\tvalidation_1-rmse:0.369271\n",
      "[50]\tvalidation_0-rmse:0.280391\tvalidation_1-rmse:0.367418\n",
      "[55]\tvalidation_0-rmse:0.27408\tvalidation_1-rmse:0.3656\n",
      "[60]\tvalidation_0-rmse:0.267961\tvalidation_1-rmse:0.36435\n",
      "[65]\tvalidation_0-rmse:0.2635\tvalidation_1-rmse:0.364854\n",
      "[70]\tvalidation_0-rmse:0.258711\tvalidation_1-rmse:0.364802\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.264737\tvalidation_1-rmse:0.364205\n",
      "\n",
      "[08:52:00] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79618\tvalidation_1-rmse:1.78469\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13165\tvalidation_1-rmse:1.1325\n",
      "[10]\tvalidation_0-rmse:0.740555\tvalidation_1-rmse:0.753282\n",
      "[15]\tvalidation_0-rmse:0.524422\tvalidation_1-rmse:0.553985\n",
      "[20]\tvalidation_0-rmse:0.4125\tvalidation_1-rmse:0.455445\n",
      "[25]\tvalidation_0-rmse:0.354068\tvalidation_1-rmse:0.407642\n",
      "[30]\tvalidation_0-rmse:0.323752\tvalidation_1-rmse:0.387289\n",
      "[35]\tvalidation_0-rmse:0.30806\tvalidation_1-rmse:0.378454\n",
      "[40]\tvalidation_0-rmse:0.2962\tvalidation_1-rmse:0.373639\n",
      "[45]\tvalidation_0-rmse:0.288136\tvalidation_1-rmse:0.375997\n",
      "[50]\tvalidation_0-rmse:0.278865\tvalidation_1-rmse:0.375086\n",
      "Stopping. Best iteration:\n",
      "[41]\tvalidation_0-rmse:0.294502\tvalidation_1-rmse:0.373169\n",
      "\n",
      "[08:52:00] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79603\tvalidation_1-rmse:1.78455\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12859\tvalidation_1-rmse:1.12862\n",
      "[10]\tvalidation_0-rmse:0.736152\tvalidation_1-rmse:0.74919\n",
      "[15]\tvalidation_0-rmse:0.51845\tvalidation_1-rmse:0.546284\n",
      "[20]\tvalidation_0-rmse:0.409148\tvalidation_1-rmse:0.449888\n",
      "[25]\tvalidation_0-rmse:0.353562\tvalidation_1-rmse:0.406002\n",
      "[30]\tvalidation_0-rmse:0.325976\tvalidation_1-rmse:0.38747\n",
      "[35]\tvalidation_0-rmse:0.308978\tvalidation_1-rmse:0.380799\n",
      "[40]\tvalidation_0-rmse:0.298563\tvalidation_1-rmse:0.376311\n",
      "[45]\tvalidation_0-rmse:0.291733\tvalidation_1-rmse:0.374472\n",
      "[50]\tvalidation_0-rmse:0.284421\tvalidation_1-rmse:0.373514\n",
      "[55]\tvalidation_0-rmse:0.278785\tvalidation_1-rmse:0.37215\n",
      "[60]\tvalidation_0-rmse:0.273449\tvalidation_1-rmse:0.372837\n",
      "[65]\tvalidation_0-rmse:0.270085\tvalidation_1-rmse:0.37385\n",
      "Stopping. Best iteration:\n",
      "[55]\tvalidation_0-rmse:0.278785\tvalidation_1-rmse:0.37215\n",
      "\n",
      "[08:52:00] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79633\tvalidation_1-rmse:1.78481\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11267\tvalidation_1-rmse:1.12478\n",
      "[10]\tvalidation_0-rmse:0.719584\tvalidation_1-rmse:0.747908\n",
      "[15]\tvalidation_0-rmse:0.502846\tvalidation_1-rmse:0.553392\n",
      "[20]\tvalidation_0-rmse:0.381153\tvalidation_1-rmse:0.45222\n",
      "[25]\tvalidation_0-rmse:0.320808\tvalidation_1-rmse:0.404822\n",
      "[30]\tvalidation_0-rmse:0.288117\tvalidation_1-rmse:0.383866\n",
      "[35]\tvalidation_0-rmse:0.267896\tvalidation_1-rmse:0.375284\n",
      "[40]\tvalidation_0-rmse:0.257185\tvalidation_1-rmse:0.37179\n",
      "[45]\tvalidation_0-rmse:0.247455\tvalidation_1-rmse:0.369188\n",
      "[50]\tvalidation_0-rmse:0.239644\tvalidation_1-rmse:0.367232\n",
      "[55]\tvalidation_0-rmse:0.233112\tvalidation_1-rmse:0.366494\n",
      "[60]\tvalidation_0-rmse:0.226115\tvalidation_1-rmse:0.367319\n",
      "Stopping. Best iteration:\n",
      "[53]\tvalidation_0-rmse:0.236435\tvalidation_1-rmse:0.366117\n",
      "\n",
      "[08:52:01] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79802\tvalidation_1-rmse:1.78666\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12703\tvalidation_1-rmse:1.12841\n",
      "[10]\tvalidation_0-rmse:0.731216\tvalidation_1-rmse:0.752816\n",
      "[15]\tvalidation_0-rmse:0.509245\tvalidation_1-rmse:0.553976\n",
      "[20]\tvalidation_0-rmse:0.392708\tvalidation_1-rmse:0.454472\n",
      "[25]\tvalidation_0-rmse:0.32713\tvalidation_1-rmse:0.405447\n",
      "[30]\tvalidation_0-rmse:0.290562\tvalidation_1-rmse:0.385488\n",
      "[35]\tvalidation_0-rmse:0.268829\tvalidation_1-rmse:0.375696\n",
      "[40]\tvalidation_0-rmse:0.255212\tvalidation_1-rmse:0.372033\n",
      "[45]\tvalidation_0-rmse:0.245665\tvalidation_1-rmse:0.370214\n",
      "[50]\tvalidation_0-rmse:0.235106\tvalidation_1-rmse:0.368119\n",
      "[55]\tvalidation_0-rmse:0.227811\tvalidation_1-rmse:0.368036\n",
      "[60]\tvalidation_0-rmse:0.221184\tvalidation_1-rmse:0.366143\n",
      "[65]\tvalidation_0-rmse:0.215561\tvalidation_1-rmse:0.365786\n",
      "[70]\tvalidation_0-rmse:0.210836\tvalidation_1-rmse:0.366036\n",
      "[75]\tvalidation_0-rmse:0.206637\tvalidation_1-rmse:0.364676\n",
      "[80]\tvalidation_0-rmse:0.202277\tvalidation_1-rmse:0.365791\n",
      "[85]\tvalidation_0-rmse:0.198729\tvalidation_1-rmse:0.365465\n",
      "Stopping. Best iteration:\n",
      "[76]\tvalidation_0-rmse:0.205995\tvalidation_1-rmse:0.36447\n",
      "\n",
      "[08:52:01] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79793\tvalidation_1-rmse:1.78615\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.1285\tvalidation_1-rmse:1.13328\n",
      "[10]\tvalidation_0-rmse:0.734452\tvalidation_1-rmse:0.75505\n",
      "[15]\tvalidation_0-rmse:0.512026\tvalidation_1-rmse:0.553501\n",
      "[20]\tvalidation_0-rmse:0.392906\tvalidation_1-rmse:0.453201\n",
      "[25]\tvalidation_0-rmse:0.329231\tvalidation_1-rmse:0.405877\n",
      "[30]\tvalidation_0-rmse:0.292959\tvalidation_1-rmse:0.383965\n",
      "[35]\tvalidation_0-rmse:0.272138\tvalidation_1-rmse:0.374618\n",
      "[40]\tvalidation_0-rmse:0.258356\tvalidation_1-rmse:0.370649\n",
      "[45]\tvalidation_0-rmse:0.247574\tvalidation_1-rmse:0.369283\n",
      "[50]\tvalidation_0-rmse:0.238362\tvalidation_1-rmse:0.368488\n",
      "[55]\tvalidation_0-rmse:0.231988\tvalidation_1-rmse:0.368137\n",
      "[60]\tvalidation_0-rmse:0.226547\tvalidation_1-rmse:0.367806\n",
      "[65]\tvalidation_0-rmse:0.222468\tvalidation_1-rmse:0.368153\n",
      "[70]\tvalidation_0-rmse:0.217389\tvalidation_1-rmse:0.368405\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.223594\tvalidation_1-rmse:0.367794\n",
      "\n",
      "[08:52:01] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79626\tvalidation_1-rmse:1.78503\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12897\tvalidation_1-rmse:1.13456\n",
      "[10]\tvalidation_0-rmse:0.734761\tvalidation_1-rmse:0.757676\n",
      "[15]\tvalidation_0-rmse:0.513857\tvalidation_1-rmse:0.559246\n",
      "[20]\tvalidation_0-rmse:0.394566\tvalidation_1-rmse:0.461915\n",
      "[25]\tvalidation_0-rmse:0.328424\tvalidation_1-rmse:0.415648\n",
      "[30]\tvalidation_0-rmse:0.29186\tvalidation_1-rmse:0.393342\n",
      "[35]\tvalidation_0-rmse:0.2698\tvalidation_1-rmse:0.383848\n",
      "[40]\tvalidation_0-rmse:0.254487\tvalidation_1-rmse:0.379983\n",
      "[45]\tvalidation_0-rmse:0.245195\tvalidation_1-rmse:0.378546\n",
      "[50]\tvalidation_0-rmse:0.235738\tvalidation_1-rmse:0.377624\n",
      "[55]\tvalidation_0-rmse:0.229972\tvalidation_1-rmse:0.377193\n",
      "[60]\tvalidation_0-rmse:0.224341\tvalidation_1-rmse:0.376848\n",
      "[65]\tvalidation_0-rmse:0.218174\tvalidation_1-rmse:0.375843\n",
      "[70]\tvalidation_0-rmse:0.213275\tvalidation_1-rmse:0.376371\n",
      "[75]\tvalidation_0-rmse:0.209558\tvalidation_1-rmse:0.376197\n",
      "[80]\tvalidation_0-rmse:0.205212\tvalidation_1-rmse:0.376944\n",
      "Stopping. Best iteration:\n",
      "[73]\tvalidation_0-rmse:0.210994\tvalidation_1-rmse:0.375669\n",
      "\n",
      "[08:52:01] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.796\tvalidation_1-rmse:1.78471\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12713\tvalidation_1-rmse:1.13236\n",
      "[10]\tvalidation_0-rmse:0.731847\tvalidation_1-rmse:0.754436\n",
      "[15]\tvalidation_0-rmse:0.509207\tvalidation_1-rmse:0.55537\n",
      "[20]\tvalidation_0-rmse:0.391151\tvalidation_1-rmse:0.45954\n",
      "[25]\tvalidation_0-rmse:0.327464\tvalidation_1-rmse:0.414048\n",
      "[30]\tvalidation_0-rmse:0.292292\tvalidation_1-rmse:0.392056\n",
      "[35]\tvalidation_0-rmse:0.273116\tvalidation_1-rmse:0.384421\n",
      "[40]\tvalidation_0-rmse:0.260679\tvalidation_1-rmse:0.381161\n",
      "[45]\tvalidation_0-rmse:0.251198\tvalidation_1-rmse:0.380135\n",
      "[50]\tvalidation_0-rmse:0.243661\tvalidation_1-rmse:0.379609\n",
      "[55]\tvalidation_0-rmse:0.237119\tvalidation_1-rmse:0.378695\n",
      "[60]\tvalidation_0-rmse:0.232269\tvalidation_1-rmse:0.379214\n",
      "Stopping. Best iteration:\n",
      "[54]\tvalidation_0-rmse:0.238061\tvalidation_1-rmse:0.378574\n",
      "\n",
      "[08:52:02] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79614\tvalidation_1-rmse:1.78407\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11374\tvalidation_1-rmse:1.11994\n",
      "[10]\tvalidation_0-rmse:0.724631\tvalidation_1-rmse:0.743554\n",
      "[15]\tvalidation_0-rmse:0.514239\tvalidation_1-rmse:0.549126\n",
      "[20]\tvalidation_0-rmse:0.400268\tvalidation_1-rmse:0.449426\n",
      "[25]\tvalidation_0-rmse:0.345923\tvalidation_1-rmse:0.404649\n",
      "[30]\tvalidation_0-rmse:0.317994\tvalidation_1-rmse:0.386144\n",
      "[35]\tvalidation_0-rmse:0.30252\tvalidation_1-rmse:0.376388\n",
      "[40]\tvalidation_0-rmse:0.294863\tvalidation_1-rmse:0.373909\n",
      "[45]\tvalidation_0-rmse:0.286019\tvalidation_1-rmse:0.371415\n",
      "[50]\tvalidation_0-rmse:0.279818\tvalidation_1-rmse:0.368649\n",
      "[55]\tvalidation_0-rmse:0.274012\tvalidation_1-rmse:0.369104\n",
      "[60]\tvalidation_0-rmse:0.268911\tvalidation_1-rmse:0.368751\n",
      "[65]\tvalidation_0-rmse:0.26435\tvalidation_1-rmse:0.370181\n",
      "Stopping. Best iteration:\n",
      "[58]\tvalidation_0-rmse:0.271137\tvalidation_1-rmse:0.368415\n",
      "\n",
      "[08:52:02] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79818\tvalidation_1-rmse:1.78572\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12911\tvalidation_1-rmse:1.12532\n",
      "[10]\tvalidation_0-rmse:0.736762\tvalidation_1-rmse:0.745403\n",
      "[15]\tvalidation_0-rmse:0.521134\tvalidation_1-rmse:0.545752\n",
      "[20]\tvalidation_0-rmse:0.411941\tvalidation_1-rmse:0.44852\n",
      "[25]\tvalidation_0-rmse:0.353254\tvalidation_1-rmse:0.401825\n",
      "[30]\tvalidation_0-rmse:0.32202\tvalidation_1-rmse:0.380904\n",
      "[35]\tvalidation_0-rmse:0.303861\tvalidation_1-rmse:0.370174\n",
      "[40]\tvalidation_0-rmse:0.292761\tvalidation_1-rmse:0.365527\n",
      "[45]\tvalidation_0-rmse:0.285199\tvalidation_1-rmse:0.362647\n",
      "[50]\tvalidation_0-rmse:0.276477\tvalidation_1-rmse:0.360668\n",
      "[55]\tvalidation_0-rmse:0.270243\tvalidation_1-rmse:0.360342\n",
      "[60]\tvalidation_0-rmse:0.26348\tvalidation_1-rmse:0.358898\n",
      "[65]\tvalidation_0-rmse:0.258888\tvalidation_1-rmse:0.358917\n",
      "[70]\tvalidation_0-rmse:0.254146\tvalidation_1-rmse:0.359126\n",
      "Stopping. Best iteration:\n",
      "[60]\tvalidation_0-rmse:0.26348\tvalidation_1-rmse:0.358898\n",
      "\n",
      "[08:52:02] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79776\tvalidation_1-rmse:1.7853\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13043\tvalidation_1-rmse:1.1321\n",
      "[10]\tvalidation_0-rmse:0.74032\tvalidation_1-rmse:0.753576\n",
      "[15]\tvalidation_0-rmse:0.523896\tvalidation_1-rmse:0.550943\n",
      "[20]\tvalidation_0-rmse:0.41136\tvalidation_1-rmse:0.453043\n",
      "[25]\tvalidation_0-rmse:0.355327\tvalidation_1-rmse:0.405661\n",
      "[30]\tvalidation_0-rmse:0.32538\tvalidation_1-rmse:0.386155\n",
      "[35]\tvalidation_0-rmse:0.308408\tvalidation_1-rmse:0.374369\n",
      "[40]\tvalidation_0-rmse:0.297905\tvalidation_1-rmse:0.370118\n",
      "[45]\tvalidation_0-rmse:0.288829\tvalidation_1-rmse:0.369271\n",
      "[50]\tvalidation_0-rmse:0.280391\tvalidation_1-rmse:0.367418\n",
      "[55]\tvalidation_0-rmse:0.27408\tvalidation_1-rmse:0.3656\n",
      "[60]\tvalidation_0-rmse:0.267961\tvalidation_1-rmse:0.36435\n",
      "[65]\tvalidation_0-rmse:0.2635\tvalidation_1-rmse:0.364854\n",
      "[70]\tvalidation_0-rmse:0.258711\tvalidation_1-rmse:0.364802\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.264737\tvalidation_1-rmse:0.364205\n",
      "\n",
      "[08:52:02] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79618\tvalidation_1-rmse:1.78469\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13165\tvalidation_1-rmse:1.1325\n",
      "[10]\tvalidation_0-rmse:0.740555\tvalidation_1-rmse:0.753282\n",
      "[15]\tvalidation_0-rmse:0.524422\tvalidation_1-rmse:0.553985\n",
      "[20]\tvalidation_0-rmse:0.4125\tvalidation_1-rmse:0.455445\n",
      "[25]\tvalidation_0-rmse:0.354068\tvalidation_1-rmse:0.407642\n",
      "[30]\tvalidation_0-rmse:0.323752\tvalidation_1-rmse:0.387289\n",
      "[35]\tvalidation_0-rmse:0.30806\tvalidation_1-rmse:0.378454\n",
      "[40]\tvalidation_0-rmse:0.2962\tvalidation_1-rmse:0.373639\n",
      "[45]\tvalidation_0-rmse:0.288136\tvalidation_1-rmse:0.375997\n",
      "[50]\tvalidation_0-rmse:0.278865\tvalidation_1-rmse:0.375086\n",
      "Stopping. Best iteration:\n",
      "[41]\tvalidation_0-rmse:0.294502\tvalidation_1-rmse:0.373169\n",
      "\n",
      "[08:52:02] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79603\tvalidation_1-rmse:1.78455\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12859\tvalidation_1-rmse:1.12862\n",
      "[10]\tvalidation_0-rmse:0.736152\tvalidation_1-rmse:0.74919\n",
      "[15]\tvalidation_0-rmse:0.51845\tvalidation_1-rmse:0.546284\n",
      "[20]\tvalidation_0-rmse:0.409148\tvalidation_1-rmse:0.449888\n",
      "[25]\tvalidation_0-rmse:0.353562\tvalidation_1-rmse:0.406002\n",
      "[30]\tvalidation_0-rmse:0.325976\tvalidation_1-rmse:0.38747\n",
      "[35]\tvalidation_0-rmse:0.308978\tvalidation_1-rmse:0.380799\n",
      "[40]\tvalidation_0-rmse:0.298563\tvalidation_1-rmse:0.376311\n",
      "[45]\tvalidation_0-rmse:0.291733\tvalidation_1-rmse:0.374472\n",
      "[50]\tvalidation_0-rmse:0.284421\tvalidation_1-rmse:0.373514\n",
      "[55]\tvalidation_0-rmse:0.278785\tvalidation_1-rmse:0.37215\n",
      "[60]\tvalidation_0-rmse:0.273449\tvalidation_1-rmse:0.372837\n",
      "[65]\tvalidation_0-rmse:0.270085\tvalidation_1-rmse:0.37385\n",
      "Stopping. Best iteration:\n",
      "[55]\tvalidation_0-rmse:0.278785\tvalidation_1-rmse:0.37215\n",
      "\n",
      "[08:52:03] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79614\tvalidation_1-rmse:1.78407\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11374\tvalidation_1-rmse:1.11994\n",
      "[10]\tvalidation_0-rmse:0.724631\tvalidation_1-rmse:0.743554\n",
      "[15]\tvalidation_0-rmse:0.514239\tvalidation_1-rmse:0.549126\n",
      "[20]\tvalidation_0-rmse:0.400268\tvalidation_1-rmse:0.449426\n",
      "[25]\tvalidation_0-rmse:0.345923\tvalidation_1-rmse:0.404649\n",
      "[30]\tvalidation_0-rmse:0.317994\tvalidation_1-rmse:0.386144\n",
      "[35]\tvalidation_0-rmse:0.30252\tvalidation_1-rmse:0.376388\n",
      "[40]\tvalidation_0-rmse:0.294863\tvalidation_1-rmse:0.373909\n",
      "[45]\tvalidation_0-rmse:0.286019\tvalidation_1-rmse:0.371415\n",
      "[50]\tvalidation_0-rmse:0.279818\tvalidation_1-rmse:0.368649\n",
      "[55]\tvalidation_0-rmse:0.274012\tvalidation_1-rmse:0.369104\n",
      "[60]\tvalidation_0-rmse:0.268911\tvalidation_1-rmse:0.368751\n",
      "[65]\tvalidation_0-rmse:0.26435\tvalidation_1-rmse:0.370181\n",
      "Stopping. Best iteration:\n",
      "[58]\tvalidation_0-rmse:0.271137\tvalidation_1-rmse:0.368415\n",
      "\n",
      "[08:52:03] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79818\tvalidation_1-rmse:1.78572\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12911\tvalidation_1-rmse:1.12532\n",
      "[10]\tvalidation_0-rmse:0.736762\tvalidation_1-rmse:0.745403\n",
      "[15]\tvalidation_0-rmse:0.521134\tvalidation_1-rmse:0.545752\n",
      "[20]\tvalidation_0-rmse:0.411941\tvalidation_1-rmse:0.44852\n",
      "[25]\tvalidation_0-rmse:0.353254\tvalidation_1-rmse:0.401825\n",
      "[30]\tvalidation_0-rmse:0.32202\tvalidation_1-rmse:0.380904\n",
      "[35]\tvalidation_0-rmse:0.303861\tvalidation_1-rmse:0.370174\n",
      "[40]\tvalidation_0-rmse:0.292761\tvalidation_1-rmse:0.365527\n",
      "[45]\tvalidation_0-rmse:0.285199\tvalidation_1-rmse:0.362647\n",
      "[50]\tvalidation_0-rmse:0.276477\tvalidation_1-rmse:0.360668\n",
      "[55]\tvalidation_0-rmse:0.270243\tvalidation_1-rmse:0.360342\n",
      "[60]\tvalidation_0-rmse:0.26348\tvalidation_1-rmse:0.358898\n",
      "[65]\tvalidation_0-rmse:0.258888\tvalidation_1-rmse:0.358917\n",
      "[70]\tvalidation_0-rmse:0.254146\tvalidation_1-rmse:0.359126\n",
      "Stopping. Best iteration:\n",
      "[60]\tvalidation_0-rmse:0.26348\tvalidation_1-rmse:0.358898\n",
      "\n",
      "[08:52:03] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79776\tvalidation_1-rmse:1.7853\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13043\tvalidation_1-rmse:1.1321\n",
      "[10]\tvalidation_0-rmse:0.74032\tvalidation_1-rmse:0.753576\n",
      "[15]\tvalidation_0-rmse:0.523896\tvalidation_1-rmse:0.550943\n",
      "[20]\tvalidation_0-rmse:0.41136\tvalidation_1-rmse:0.453043\n",
      "[25]\tvalidation_0-rmse:0.355327\tvalidation_1-rmse:0.405661\n",
      "[30]\tvalidation_0-rmse:0.32538\tvalidation_1-rmse:0.386155\n",
      "[35]\tvalidation_0-rmse:0.308408\tvalidation_1-rmse:0.374369\n",
      "[40]\tvalidation_0-rmse:0.297905\tvalidation_1-rmse:0.370118\n",
      "[45]\tvalidation_0-rmse:0.288829\tvalidation_1-rmse:0.369271\n",
      "[50]\tvalidation_0-rmse:0.280391\tvalidation_1-rmse:0.367418\n",
      "[55]\tvalidation_0-rmse:0.27408\tvalidation_1-rmse:0.3656\n",
      "[60]\tvalidation_0-rmse:0.267961\tvalidation_1-rmse:0.36435\n",
      "[65]\tvalidation_0-rmse:0.2635\tvalidation_1-rmse:0.364854\n",
      "[70]\tvalidation_0-rmse:0.258711\tvalidation_1-rmse:0.364802\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.264737\tvalidation_1-rmse:0.364205\n",
      "\n",
      "[08:52:03] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79618\tvalidation_1-rmse:1.78469\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13165\tvalidation_1-rmse:1.1325\n",
      "[10]\tvalidation_0-rmse:0.740555\tvalidation_1-rmse:0.753282\n",
      "[15]\tvalidation_0-rmse:0.524422\tvalidation_1-rmse:0.553985\n",
      "[20]\tvalidation_0-rmse:0.4125\tvalidation_1-rmse:0.455445\n",
      "[25]\tvalidation_0-rmse:0.354068\tvalidation_1-rmse:0.407642\n",
      "[30]\tvalidation_0-rmse:0.323752\tvalidation_1-rmse:0.387289\n",
      "[35]\tvalidation_0-rmse:0.30806\tvalidation_1-rmse:0.378454\n",
      "[40]\tvalidation_0-rmse:0.2962\tvalidation_1-rmse:0.373639\n",
      "[45]\tvalidation_0-rmse:0.288136\tvalidation_1-rmse:0.375997\n",
      "[50]\tvalidation_0-rmse:0.278865\tvalidation_1-rmse:0.375086\n",
      "Stopping. Best iteration:\n",
      "[41]\tvalidation_0-rmse:0.294502\tvalidation_1-rmse:0.373169\n",
      "\n",
      "[08:52:03] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79603\tvalidation_1-rmse:1.78455\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12859\tvalidation_1-rmse:1.12862\n",
      "[10]\tvalidation_0-rmse:0.736152\tvalidation_1-rmse:0.74919\n",
      "[15]\tvalidation_0-rmse:0.51845\tvalidation_1-rmse:0.546284\n",
      "[20]\tvalidation_0-rmse:0.409148\tvalidation_1-rmse:0.449888\n",
      "[25]\tvalidation_0-rmse:0.353562\tvalidation_1-rmse:0.406002\n",
      "[30]\tvalidation_0-rmse:0.325976\tvalidation_1-rmse:0.38747\n",
      "[35]\tvalidation_0-rmse:0.308978\tvalidation_1-rmse:0.380799\n",
      "[40]\tvalidation_0-rmse:0.298563\tvalidation_1-rmse:0.376311\n",
      "[45]\tvalidation_0-rmse:0.291733\tvalidation_1-rmse:0.374472\n",
      "[50]\tvalidation_0-rmse:0.284421\tvalidation_1-rmse:0.373514\n",
      "[55]\tvalidation_0-rmse:0.278785\tvalidation_1-rmse:0.37215\n",
      "[60]\tvalidation_0-rmse:0.273449\tvalidation_1-rmse:0.372837\n",
      "[65]\tvalidation_0-rmse:0.270085\tvalidation_1-rmse:0.37385\n",
      "Stopping. Best iteration:\n",
      "[55]\tvalidation_0-rmse:0.278785\tvalidation_1-rmse:0.37215\n",
      "\n",
      "[08:52:03] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79616\tvalidation_1-rmse:1.78428\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11569\tvalidation_1-rmse:1.12213\n",
      "[10]\tvalidation_0-rmse:0.729602\tvalidation_1-rmse:0.745648\n",
      "[15]\tvalidation_0-rmse:0.523083\tvalidation_1-rmse:0.550521\n",
      "[20]\tvalidation_0-rmse:0.411329\tvalidation_1-rmse:0.448002\n",
      "[25]\tvalidation_0-rmse:0.35974\tvalidation_1-rmse:0.400908\n",
      "[30]\tvalidation_0-rmse:0.334303\tvalidation_1-rmse:0.381822\n",
      "[35]\tvalidation_0-rmse:0.322187\tvalidation_1-rmse:0.372105\n",
      "[40]\tvalidation_0-rmse:0.314885\tvalidation_1-rmse:0.367836\n",
      "[45]\tvalidation_0-rmse:0.306699\tvalidation_1-rmse:0.364114\n",
      "[50]\tvalidation_0-rmse:0.300616\tvalidation_1-rmse:0.36271\n",
      "[55]\tvalidation_0-rmse:0.295377\tvalidation_1-rmse:0.362687\n",
      "[60]\tvalidation_0-rmse:0.291197\tvalidation_1-rmse:0.362315\n",
      "[65]\tvalidation_0-rmse:0.288119\tvalidation_1-rmse:0.361996\n",
      "[70]\tvalidation_0-rmse:0.284895\tvalidation_1-rmse:0.360439\n",
      "[75]\tvalidation_0-rmse:0.281633\tvalidation_1-rmse:0.359178\n",
      "[80]\tvalidation_0-rmse:0.277768\tvalidation_1-rmse:0.358541\n",
      "[85]\tvalidation_0-rmse:0.27506\tvalidation_1-rmse:0.359466\n",
      "[90]\tvalidation_0-rmse:0.271096\tvalidation_1-rmse:0.359299\n",
      "Stopping. Best iteration:\n",
      "[81]\tvalidation_0-rmse:0.2774\tvalidation_1-rmse:0.358536\n",
      "\n",
      "[08:52:04] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79835\tvalidation_1-rmse:1.78556\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13143\tvalidation_1-rmse:1.12643\n",
      "[10]\tvalidation_0-rmse:0.741529\tvalidation_1-rmse:0.746895\n",
      "[15]\tvalidation_0-rmse:0.530547\tvalidation_1-rmse:0.54735\n",
      "[20]\tvalidation_0-rmse:0.426076\tvalidation_1-rmse:0.449866\n",
      "[25]\tvalidation_0-rmse:0.370466\tvalidation_1-rmse:0.402303\n",
      "[30]\tvalidation_0-rmse:0.341876\tvalidation_1-rmse:0.38303\n",
      "[35]\tvalidation_0-rmse:0.326404\tvalidation_1-rmse:0.373672\n",
      "[40]\tvalidation_0-rmse:0.316251\tvalidation_1-rmse:0.368675\n",
      "[45]\tvalidation_0-rmse:0.309503\tvalidation_1-rmse:0.365661\n",
      "[50]\tvalidation_0-rmse:0.301739\tvalidation_1-rmse:0.361998\n",
      "[55]\tvalidation_0-rmse:0.29654\tvalidation_1-rmse:0.36137\n",
      "[60]\tvalidation_0-rmse:0.29137\tvalidation_1-rmse:0.359809\n",
      "[65]\tvalidation_0-rmse:0.287162\tvalidation_1-rmse:0.359409\n",
      "[70]\tvalidation_0-rmse:0.283067\tvalidation_1-rmse:0.359979\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.288503\tvalidation_1-rmse:0.359351\n",
      "\n",
      "[08:52:04] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79808\tvalidation_1-rmse:1.78526\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13268\tvalidation_1-rmse:1.12972\n",
      "[10]\tvalidation_0-rmse:0.744353\tvalidation_1-rmse:0.75026\n",
      "[15]\tvalidation_0-rmse:0.529972\tvalidation_1-rmse:0.545997\n",
      "[20]\tvalidation_0-rmse:0.421538\tvalidation_1-rmse:0.448715\n",
      "[25]\tvalidation_0-rmse:0.368352\tvalidation_1-rmse:0.403255\n",
      "[30]\tvalidation_0-rmse:0.340252\tvalidation_1-rmse:0.385096\n",
      "[35]\tvalidation_0-rmse:0.325761\tvalidation_1-rmse:0.375385\n",
      "[40]\tvalidation_0-rmse:0.316819\tvalidation_1-rmse:0.37195\n",
      "[45]\tvalidation_0-rmse:0.310316\tvalidation_1-rmse:0.369912\n",
      "[50]\tvalidation_0-rmse:0.303157\tvalidation_1-rmse:0.367805\n",
      "[55]\tvalidation_0-rmse:0.298504\tvalidation_1-rmse:0.367183\n",
      "[60]\tvalidation_0-rmse:0.293795\tvalidation_1-rmse:0.365981\n",
      "[65]\tvalidation_0-rmse:0.29012\tvalidation_1-rmse:0.366457\n",
      "[70]\tvalidation_0-rmse:0.286213\tvalidation_1-rmse:0.366467\n",
      "[75]\tvalidation_0-rmse:0.282531\tvalidation_1-rmse:0.366707\n",
      "Stopping. Best iteration:\n",
      "[68]\tvalidation_0-rmse:0.287279\tvalidation_1-rmse:0.365616\n",
      "\n",
      "[08:52:04] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79632\tvalidation_1-rmse:1.78427\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13411\tvalidation_1-rmse:1.13278\n",
      "[10]\tvalidation_0-rmse:0.745413\tvalidation_1-rmse:0.754151\n",
      "[15]\tvalidation_0-rmse:0.53288\tvalidation_1-rmse:0.552319\n",
      "[20]\tvalidation_0-rmse:0.425337\tvalidation_1-rmse:0.455724\n",
      "[25]\tvalidation_0-rmse:0.370999\tvalidation_1-rmse:0.412178\n",
      "[30]\tvalidation_0-rmse:0.342934\tvalidation_1-rmse:0.391715\n",
      "[35]\tvalidation_0-rmse:0.327894\tvalidation_1-rmse:0.382788\n",
      "[40]\tvalidation_0-rmse:0.318003\tvalidation_1-rmse:0.376789\n",
      "[45]\tvalidation_0-rmse:0.311635\tvalidation_1-rmse:0.375004\n",
      "[50]\tvalidation_0-rmse:0.303366\tvalidation_1-rmse:0.373218\n",
      "[55]\tvalidation_0-rmse:0.299068\tvalidation_1-rmse:0.372632\n",
      "[60]\tvalidation_0-rmse:0.294875\tvalidation_1-rmse:0.371845\n",
      "[65]\tvalidation_0-rmse:0.290275\tvalidation_1-rmse:0.371069\n",
      "[70]\tvalidation_0-rmse:0.285797\tvalidation_1-rmse:0.370264\n",
      "[75]\tvalidation_0-rmse:0.28249\tvalidation_1-rmse:0.368171\n",
      "[80]\tvalidation_0-rmse:0.279087\tvalidation_1-rmse:0.36756\n",
      "[85]\tvalidation_0-rmse:0.274939\tvalidation_1-rmse:0.367402\n",
      "[90]\tvalidation_0-rmse:0.270781\tvalidation_1-rmse:0.366771\n",
      "[95]\tvalidation_0-rmse:0.267177\tvalidation_1-rmse:0.366857\n",
      "Stopping. Best iteration:\n",
      "[89]\tvalidation_0-rmse:0.27173\tvalidation_1-rmse:0.36647\n",
      "\n",
      "[08:52:04] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79584\tvalidation_1-rmse:1.78382\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.1308\tvalidation_1-rmse:1.1312\n",
      "[10]\tvalidation_0-rmse:0.740274\tvalidation_1-rmse:0.750156\n",
      "[15]\tvalidation_0-rmse:0.527001\tvalidation_1-rmse:0.547723\n",
      "[20]\tvalidation_0-rmse:0.421592\tvalidation_1-rmse:0.44875\n",
      "[25]\tvalidation_0-rmse:0.368178\tvalidation_1-rmse:0.405226\n",
      "[30]\tvalidation_0-rmse:0.341753\tvalidation_1-rmse:0.384767\n",
      "[35]\tvalidation_0-rmse:0.32823\tvalidation_1-rmse:0.378304\n",
      "[40]\tvalidation_0-rmse:0.320174\tvalidation_1-rmse:0.374885\n",
      "[45]\tvalidation_0-rmse:0.314184\tvalidation_1-rmse:0.372483\n",
      "[50]\tvalidation_0-rmse:0.307949\tvalidation_1-rmse:0.369845\n",
      "[55]\tvalidation_0-rmse:0.303283\tvalidation_1-rmse:0.369507\n",
      "[60]\tvalidation_0-rmse:0.297863\tvalidation_1-rmse:0.368905\n",
      "[65]\tvalidation_0-rmse:0.29443\tvalidation_1-rmse:0.368643\n",
      "[70]\tvalidation_0-rmse:0.290082\tvalidation_1-rmse:0.370353\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.296155\tvalidation_1-rmse:0.368294\n",
      "\n",
      "[08:52:04] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79614\tvalidation_1-rmse:1.78407\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11374\tvalidation_1-rmse:1.11994\n",
      "[10]\tvalidation_0-rmse:0.724631\tvalidation_1-rmse:0.743554\n",
      "[15]\tvalidation_0-rmse:0.514239\tvalidation_1-rmse:0.549126\n",
      "[20]\tvalidation_0-rmse:0.400268\tvalidation_1-rmse:0.449426\n",
      "[25]\tvalidation_0-rmse:0.345923\tvalidation_1-rmse:0.404649\n",
      "[30]\tvalidation_0-rmse:0.317994\tvalidation_1-rmse:0.386144\n",
      "[35]\tvalidation_0-rmse:0.30252\tvalidation_1-rmse:0.376388\n",
      "[40]\tvalidation_0-rmse:0.294863\tvalidation_1-rmse:0.373909\n",
      "[45]\tvalidation_0-rmse:0.286019\tvalidation_1-rmse:0.371415\n",
      "[50]\tvalidation_0-rmse:0.279818\tvalidation_1-rmse:0.368649\n",
      "[55]\tvalidation_0-rmse:0.274012\tvalidation_1-rmse:0.369104\n",
      "[60]\tvalidation_0-rmse:0.268911\tvalidation_1-rmse:0.368751\n",
      "[65]\tvalidation_0-rmse:0.26435\tvalidation_1-rmse:0.370181\n",
      "Stopping. Best iteration:\n",
      "[58]\tvalidation_0-rmse:0.271137\tvalidation_1-rmse:0.368415\n",
      "\n",
      "[08:52:05] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79818\tvalidation_1-rmse:1.78572\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12911\tvalidation_1-rmse:1.12532\n",
      "[10]\tvalidation_0-rmse:0.736762\tvalidation_1-rmse:0.745403\n",
      "[15]\tvalidation_0-rmse:0.521134\tvalidation_1-rmse:0.545752\n",
      "[20]\tvalidation_0-rmse:0.411941\tvalidation_1-rmse:0.44852\n",
      "[25]\tvalidation_0-rmse:0.353254\tvalidation_1-rmse:0.401825\n",
      "[30]\tvalidation_0-rmse:0.32202\tvalidation_1-rmse:0.380904\n",
      "[35]\tvalidation_0-rmse:0.303861\tvalidation_1-rmse:0.370174\n",
      "[40]\tvalidation_0-rmse:0.292761\tvalidation_1-rmse:0.365527\n",
      "[45]\tvalidation_0-rmse:0.285199\tvalidation_1-rmse:0.362647\n",
      "[50]\tvalidation_0-rmse:0.276477\tvalidation_1-rmse:0.360668\n",
      "[55]\tvalidation_0-rmse:0.270243\tvalidation_1-rmse:0.360342\n",
      "[60]\tvalidation_0-rmse:0.26348\tvalidation_1-rmse:0.358898\n",
      "[65]\tvalidation_0-rmse:0.258888\tvalidation_1-rmse:0.358917\n",
      "[70]\tvalidation_0-rmse:0.254146\tvalidation_1-rmse:0.359126\n",
      "Stopping. Best iteration:\n",
      "[60]\tvalidation_0-rmse:0.26348\tvalidation_1-rmse:0.358898\n",
      "\n",
      "[08:52:05] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79776\tvalidation_1-rmse:1.7853\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13043\tvalidation_1-rmse:1.1321\n",
      "[10]\tvalidation_0-rmse:0.74032\tvalidation_1-rmse:0.753576\n",
      "[15]\tvalidation_0-rmse:0.523896\tvalidation_1-rmse:0.550943\n",
      "[20]\tvalidation_0-rmse:0.41136\tvalidation_1-rmse:0.453043\n",
      "[25]\tvalidation_0-rmse:0.355327\tvalidation_1-rmse:0.405661\n",
      "[30]\tvalidation_0-rmse:0.32538\tvalidation_1-rmse:0.386155\n",
      "[35]\tvalidation_0-rmse:0.308408\tvalidation_1-rmse:0.374369\n",
      "[40]\tvalidation_0-rmse:0.297905\tvalidation_1-rmse:0.370118\n",
      "[45]\tvalidation_0-rmse:0.288829\tvalidation_1-rmse:0.369271\n",
      "[50]\tvalidation_0-rmse:0.280391\tvalidation_1-rmse:0.367418\n",
      "[55]\tvalidation_0-rmse:0.27408\tvalidation_1-rmse:0.3656\n",
      "[60]\tvalidation_0-rmse:0.267961\tvalidation_1-rmse:0.36435\n",
      "[65]\tvalidation_0-rmse:0.2635\tvalidation_1-rmse:0.364854\n",
      "[70]\tvalidation_0-rmse:0.258711\tvalidation_1-rmse:0.364802\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.264737\tvalidation_1-rmse:0.364205\n",
      "\n",
      "[08:52:05] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79618\tvalidation_1-rmse:1.78469\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13165\tvalidation_1-rmse:1.1325\n",
      "[10]\tvalidation_0-rmse:0.740555\tvalidation_1-rmse:0.753282\n",
      "[15]\tvalidation_0-rmse:0.524422\tvalidation_1-rmse:0.553985\n",
      "[20]\tvalidation_0-rmse:0.4125\tvalidation_1-rmse:0.455445\n",
      "[25]\tvalidation_0-rmse:0.354068\tvalidation_1-rmse:0.407642\n",
      "[30]\tvalidation_0-rmse:0.323752\tvalidation_1-rmse:0.387289\n",
      "[35]\tvalidation_0-rmse:0.30806\tvalidation_1-rmse:0.378454\n",
      "[40]\tvalidation_0-rmse:0.2962\tvalidation_1-rmse:0.373639\n",
      "[45]\tvalidation_0-rmse:0.288136\tvalidation_1-rmse:0.375997\n",
      "[50]\tvalidation_0-rmse:0.278865\tvalidation_1-rmse:0.375086\n",
      "Stopping. Best iteration:\n",
      "[41]\tvalidation_0-rmse:0.294502\tvalidation_1-rmse:0.373169\n",
      "\n",
      "[08:52:05] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79603\tvalidation_1-rmse:1.78455\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12859\tvalidation_1-rmse:1.12862\n",
      "[10]\tvalidation_0-rmse:0.736152\tvalidation_1-rmse:0.74919\n",
      "[15]\tvalidation_0-rmse:0.51845\tvalidation_1-rmse:0.546284\n",
      "[20]\tvalidation_0-rmse:0.409148\tvalidation_1-rmse:0.449888\n",
      "[25]\tvalidation_0-rmse:0.353562\tvalidation_1-rmse:0.406002\n",
      "[30]\tvalidation_0-rmse:0.325976\tvalidation_1-rmse:0.38747\n",
      "[35]\tvalidation_0-rmse:0.308978\tvalidation_1-rmse:0.380799\n",
      "[40]\tvalidation_0-rmse:0.298563\tvalidation_1-rmse:0.376311\n",
      "[45]\tvalidation_0-rmse:0.291733\tvalidation_1-rmse:0.374472\n",
      "[50]\tvalidation_0-rmse:0.284421\tvalidation_1-rmse:0.373514\n",
      "[55]\tvalidation_0-rmse:0.278785\tvalidation_1-rmse:0.37215\n",
      "[60]\tvalidation_0-rmse:0.273449\tvalidation_1-rmse:0.372837\n",
      "[65]\tvalidation_0-rmse:0.270085\tvalidation_1-rmse:0.37385\n",
      "Stopping. Best iteration:\n",
      "[55]\tvalidation_0-rmse:0.278785\tvalidation_1-rmse:0.37215\n",
      "\n",
      "[08:52:05] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79633\tvalidation_1-rmse:1.78481\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11267\tvalidation_1-rmse:1.12478\n",
      "[10]\tvalidation_0-rmse:0.719584\tvalidation_1-rmse:0.747908\n",
      "[15]\tvalidation_0-rmse:0.502846\tvalidation_1-rmse:0.553392\n",
      "[20]\tvalidation_0-rmse:0.381153\tvalidation_1-rmse:0.45222\n",
      "[25]\tvalidation_0-rmse:0.320808\tvalidation_1-rmse:0.404822\n",
      "[30]\tvalidation_0-rmse:0.288117\tvalidation_1-rmse:0.383866\n",
      "[35]\tvalidation_0-rmse:0.267896\tvalidation_1-rmse:0.375284\n",
      "[40]\tvalidation_0-rmse:0.257185\tvalidation_1-rmse:0.37179\n",
      "[45]\tvalidation_0-rmse:0.247455\tvalidation_1-rmse:0.369188\n",
      "[50]\tvalidation_0-rmse:0.239644\tvalidation_1-rmse:0.367232\n",
      "[55]\tvalidation_0-rmse:0.233112\tvalidation_1-rmse:0.366494\n",
      "[60]\tvalidation_0-rmse:0.226115\tvalidation_1-rmse:0.367319\n",
      "Stopping. Best iteration:\n",
      "[53]\tvalidation_0-rmse:0.236435\tvalidation_1-rmse:0.366117\n",
      "\n",
      "[08:52:05] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79802\tvalidation_1-rmse:1.78666\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12703\tvalidation_1-rmse:1.12841\n",
      "[10]\tvalidation_0-rmse:0.731216\tvalidation_1-rmse:0.752816\n",
      "[15]\tvalidation_0-rmse:0.509245\tvalidation_1-rmse:0.553976\n",
      "[20]\tvalidation_0-rmse:0.392708\tvalidation_1-rmse:0.454472\n",
      "[25]\tvalidation_0-rmse:0.32713\tvalidation_1-rmse:0.405447\n",
      "[30]\tvalidation_0-rmse:0.290562\tvalidation_1-rmse:0.385488\n",
      "[35]\tvalidation_0-rmse:0.268829\tvalidation_1-rmse:0.375696\n",
      "[40]\tvalidation_0-rmse:0.255212\tvalidation_1-rmse:0.372033\n",
      "[45]\tvalidation_0-rmse:0.245665\tvalidation_1-rmse:0.370214\n",
      "[50]\tvalidation_0-rmse:0.235106\tvalidation_1-rmse:0.368119\n",
      "[55]\tvalidation_0-rmse:0.227811\tvalidation_1-rmse:0.368036\n",
      "[60]\tvalidation_0-rmse:0.221184\tvalidation_1-rmse:0.366143\n",
      "[65]\tvalidation_0-rmse:0.215561\tvalidation_1-rmse:0.365786\n",
      "[70]\tvalidation_0-rmse:0.210836\tvalidation_1-rmse:0.366036\n",
      "[75]\tvalidation_0-rmse:0.206637\tvalidation_1-rmse:0.364676\n",
      "[80]\tvalidation_0-rmse:0.202277\tvalidation_1-rmse:0.365791\n",
      "[85]\tvalidation_0-rmse:0.198729\tvalidation_1-rmse:0.365465\n",
      "Stopping. Best iteration:\n",
      "[76]\tvalidation_0-rmse:0.205995\tvalidation_1-rmse:0.36447\n",
      "\n",
      "[08:52:06] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79793\tvalidation_1-rmse:1.78615\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.1285\tvalidation_1-rmse:1.13328\n",
      "[10]\tvalidation_0-rmse:0.734452\tvalidation_1-rmse:0.75505\n",
      "[15]\tvalidation_0-rmse:0.512026\tvalidation_1-rmse:0.553501\n",
      "[20]\tvalidation_0-rmse:0.392906\tvalidation_1-rmse:0.453201\n",
      "[25]\tvalidation_0-rmse:0.329231\tvalidation_1-rmse:0.405877\n",
      "[30]\tvalidation_0-rmse:0.292959\tvalidation_1-rmse:0.383965\n",
      "[35]\tvalidation_0-rmse:0.272138\tvalidation_1-rmse:0.374618\n",
      "[40]\tvalidation_0-rmse:0.258356\tvalidation_1-rmse:0.370649\n",
      "[45]\tvalidation_0-rmse:0.247574\tvalidation_1-rmse:0.369283\n",
      "[50]\tvalidation_0-rmse:0.238362\tvalidation_1-rmse:0.368488\n",
      "[55]\tvalidation_0-rmse:0.231988\tvalidation_1-rmse:0.368137\n",
      "[60]\tvalidation_0-rmse:0.226547\tvalidation_1-rmse:0.367806\n",
      "[65]\tvalidation_0-rmse:0.222468\tvalidation_1-rmse:0.368153\n",
      "[70]\tvalidation_0-rmse:0.217389\tvalidation_1-rmse:0.368405\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.223594\tvalidation_1-rmse:0.367794\n",
      "\n",
      "[08:52:06] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79626\tvalidation_1-rmse:1.78503\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12897\tvalidation_1-rmse:1.13456\n",
      "[10]\tvalidation_0-rmse:0.734761\tvalidation_1-rmse:0.757676\n",
      "[15]\tvalidation_0-rmse:0.513857\tvalidation_1-rmse:0.559246\n",
      "[20]\tvalidation_0-rmse:0.394566\tvalidation_1-rmse:0.461915\n",
      "[25]\tvalidation_0-rmse:0.328424\tvalidation_1-rmse:0.415648\n",
      "[30]\tvalidation_0-rmse:0.29186\tvalidation_1-rmse:0.393342\n",
      "[35]\tvalidation_0-rmse:0.2698\tvalidation_1-rmse:0.383848\n",
      "[40]\tvalidation_0-rmse:0.254487\tvalidation_1-rmse:0.379983\n",
      "[45]\tvalidation_0-rmse:0.245195\tvalidation_1-rmse:0.378546\n",
      "[50]\tvalidation_0-rmse:0.235738\tvalidation_1-rmse:0.377624\n",
      "[55]\tvalidation_0-rmse:0.229972\tvalidation_1-rmse:0.377193\n",
      "[60]\tvalidation_0-rmse:0.224341\tvalidation_1-rmse:0.376848\n",
      "[65]\tvalidation_0-rmse:0.218174\tvalidation_1-rmse:0.375843\n",
      "[70]\tvalidation_0-rmse:0.213275\tvalidation_1-rmse:0.376371\n",
      "[75]\tvalidation_0-rmse:0.209558\tvalidation_1-rmse:0.376197\n",
      "[80]\tvalidation_0-rmse:0.205212\tvalidation_1-rmse:0.376944\n",
      "Stopping. Best iteration:\n",
      "[73]\tvalidation_0-rmse:0.210994\tvalidation_1-rmse:0.375669\n",
      "\n",
      "[08:52:06] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.796\tvalidation_1-rmse:1.78471\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12713\tvalidation_1-rmse:1.13236\n",
      "[10]\tvalidation_0-rmse:0.731847\tvalidation_1-rmse:0.754436\n",
      "[15]\tvalidation_0-rmse:0.509207\tvalidation_1-rmse:0.55537\n",
      "[20]\tvalidation_0-rmse:0.391151\tvalidation_1-rmse:0.45954\n",
      "[25]\tvalidation_0-rmse:0.327464\tvalidation_1-rmse:0.414048\n",
      "[30]\tvalidation_0-rmse:0.292292\tvalidation_1-rmse:0.392056\n",
      "[35]\tvalidation_0-rmse:0.273116\tvalidation_1-rmse:0.384421\n",
      "[40]\tvalidation_0-rmse:0.260679\tvalidation_1-rmse:0.381161\n",
      "[45]\tvalidation_0-rmse:0.251198\tvalidation_1-rmse:0.380135\n",
      "[50]\tvalidation_0-rmse:0.243661\tvalidation_1-rmse:0.379609\n",
      "[55]\tvalidation_0-rmse:0.237119\tvalidation_1-rmse:0.378695\n",
      "[60]\tvalidation_0-rmse:0.232269\tvalidation_1-rmse:0.379214\n",
      "Stopping. Best iteration:\n",
      "[54]\tvalidation_0-rmse:0.238061\tvalidation_1-rmse:0.378574\n",
      "\n",
      "[08:52:06] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79642\tvalidation_1-rmse:1.7846\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11316\tvalidation_1-rmse:1.12223\n",
      "[10]\tvalidation_0-rmse:0.721741\tvalidation_1-rmse:0.745172\n",
      "[15]\tvalidation_0-rmse:0.5071\tvalidation_1-rmse:0.551192\n",
      "[20]\tvalidation_0-rmse:0.388798\tvalidation_1-rmse:0.450634\n",
      "[25]\tvalidation_0-rmse:0.329746\tvalidation_1-rmse:0.405946\n",
      "[30]\tvalidation_0-rmse:0.299598\tvalidation_1-rmse:0.386954\n",
      "[35]\tvalidation_0-rmse:0.282761\tvalidation_1-rmse:0.377792\n",
      "[40]\tvalidation_0-rmse:0.274318\tvalidation_1-rmse:0.374764\n",
      "[45]\tvalidation_0-rmse:0.26423\tvalidation_1-rmse:0.373965\n",
      "[50]\tvalidation_0-rmse:0.257918\tvalidation_1-rmse:0.371235\n",
      "[55]\tvalidation_0-rmse:0.250489\tvalidation_1-rmse:0.370673\n",
      "[60]\tvalidation_0-rmse:0.244453\tvalidation_1-rmse:0.369002\n",
      "[65]\tvalidation_0-rmse:0.240554\tvalidation_1-rmse:0.369352\n",
      "[70]\tvalidation_0-rmse:0.236514\tvalidation_1-rmse:0.36985\n",
      "Stopping. Best iteration:\n",
      "[61]\tvalidation_0-rmse:0.243274\tvalidation_1-rmse:0.368939\n",
      "\n",
      "[08:52:07] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79802\tvalidation_1-rmse:1.78587\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12789\tvalidation_1-rmse:1.12716\n",
      "[10]\tvalidation_0-rmse:0.733492\tvalidation_1-rmse:0.747764\n",
      "[15]\tvalidation_0-rmse:0.514339\tvalidation_1-rmse:0.548014\n",
      "[20]\tvalidation_0-rmse:0.400729\tvalidation_1-rmse:0.450011\n",
      "[25]\tvalidation_0-rmse:0.33875\tvalidation_1-rmse:0.399496\n",
      "[30]\tvalidation_0-rmse:0.304376\tvalidation_1-rmse:0.378026\n",
      "[35]\tvalidation_0-rmse:0.285189\tvalidation_1-rmse:0.369022\n",
      "[40]\tvalidation_0-rmse:0.272586\tvalidation_1-rmse:0.364174\n",
      "[45]\tvalidation_0-rmse:0.264009\tvalidation_1-rmse:0.361922\n",
      "[50]\tvalidation_0-rmse:0.255542\tvalidation_1-rmse:0.360527\n",
      "[55]\tvalidation_0-rmse:0.248554\tvalidation_1-rmse:0.360459\n",
      "[60]\tvalidation_0-rmse:0.241987\tvalidation_1-rmse:0.358443\n",
      "[65]\tvalidation_0-rmse:0.237326\tvalidation_1-rmse:0.357847\n",
      "[70]\tvalidation_0-rmse:0.232239\tvalidation_1-rmse:0.357498\n",
      "[75]\tvalidation_0-rmse:0.227911\tvalidation_1-rmse:0.357595\n",
      "[80]\tvalidation_0-rmse:0.223704\tvalidation_1-rmse:0.358013\n",
      "[85]\tvalidation_0-rmse:0.219842\tvalidation_1-rmse:0.357949\n",
      "Stopping. Best iteration:\n",
      "[77]\tvalidation_0-rmse:0.226019\tvalidation_1-rmse:0.357149\n",
      "\n",
      "[08:52:07] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79793\tvalidation_1-rmse:1.78591\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12917\tvalidation_1-rmse:1.13408\n",
      "[10]\tvalidation_0-rmse:0.736497\tvalidation_1-rmse:0.754251\n",
      "[15]\tvalidation_0-rmse:0.51694\tvalidation_1-rmse:0.553667\n",
      "[20]\tvalidation_0-rmse:0.400577\tvalidation_1-rmse:0.45585\n",
      "[25]\tvalidation_0-rmse:0.339449\tvalidation_1-rmse:0.408296\n",
      "[30]\tvalidation_0-rmse:0.305147\tvalidation_1-rmse:0.387752\n",
      "[35]\tvalidation_0-rmse:0.286103\tvalidation_1-rmse:0.37819\n",
      "[40]\tvalidation_0-rmse:0.275916\tvalidation_1-rmse:0.374516\n",
      "[45]\tvalidation_0-rmse:0.266851\tvalidation_1-rmse:0.372745\n",
      "[50]\tvalidation_0-rmse:0.256993\tvalidation_1-rmse:0.371241\n",
      "[55]\tvalidation_0-rmse:0.251081\tvalidation_1-rmse:0.370043\n",
      "[60]\tvalidation_0-rmse:0.245548\tvalidation_1-rmse:0.369649\n",
      "[65]\tvalidation_0-rmse:0.239857\tvalidation_1-rmse:0.36919\n",
      "[70]\tvalidation_0-rmse:0.235005\tvalidation_1-rmse:0.368912\n",
      "[75]\tvalidation_0-rmse:0.229798\tvalidation_1-rmse:0.370284\n",
      "Stopping. Best iteration:\n",
      "[69]\tvalidation_0-rmse:0.235846\tvalidation_1-rmse:0.368421\n",
      "\n",
      "[08:52:07] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79615\tvalidation_1-rmse:1.78485\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13005\tvalidation_1-rmse:1.1341\n",
      "[10]\tvalidation_0-rmse:0.737756\tvalidation_1-rmse:0.75612\n",
      "[15]\tvalidation_0-rmse:0.519381\tvalidation_1-rmse:0.555522\n",
      "[20]\tvalidation_0-rmse:0.404825\tvalidation_1-rmse:0.457393\n",
      "[25]\tvalidation_0-rmse:0.342741\tvalidation_1-rmse:0.408563\n",
      "[30]\tvalidation_0-rmse:0.309118\tvalidation_1-rmse:0.389082\n",
      "[35]\tvalidation_0-rmse:0.289916\tvalidation_1-rmse:0.379144\n",
      "[40]\tvalidation_0-rmse:0.27636\tvalidation_1-rmse:0.375386\n",
      "[45]\tvalidation_0-rmse:0.265307\tvalidation_1-rmse:0.373768\n",
      "[50]\tvalidation_0-rmse:0.255225\tvalidation_1-rmse:0.373328\n",
      "[55]\tvalidation_0-rmse:0.249794\tvalidation_1-rmse:0.372279\n",
      "[60]\tvalidation_0-rmse:0.24366\tvalidation_1-rmse:0.370641\n",
      "[65]\tvalidation_0-rmse:0.238738\tvalidation_1-rmse:0.370554\n",
      "[70]\tvalidation_0-rmse:0.23317\tvalidation_1-rmse:0.370545\n",
      "[75]\tvalidation_0-rmse:0.229069\tvalidation_1-rmse:0.369436\n",
      "[80]\tvalidation_0-rmse:0.225605\tvalidation_1-rmse:0.369867\n",
      "Stopping. Best iteration:\n",
      "[73]\tvalidation_0-rmse:0.230383\tvalidation_1-rmse:0.369368\n",
      "\n",
      "[08:52:07] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79602\tvalidation_1-rmse:1.78473\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12756\tvalidation_1-rmse:1.1315\n",
      "[10]\tvalidation_0-rmse:0.732466\tvalidation_1-rmse:0.753535\n",
      "[15]\tvalidation_0-rmse:0.511994\tvalidation_1-rmse:0.551604\n",
      "[20]\tvalidation_0-rmse:0.398393\tvalidation_1-rmse:0.45364\n",
      "[25]\tvalidation_0-rmse:0.337384\tvalidation_1-rmse:0.408026\n",
      "[30]\tvalidation_0-rmse:0.305423\tvalidation_1-rmse:0.387491\n",
      "[35]\tvalidation_0-rmse:0.286767\tvalidation_1-rmse:0.38059\n",
      "[40]\tvalidation_0-rmse:0.275528\tvalidation_1-rmse:0.376201\n",
      "[45]\tvalidation_0-rmse:0.267359\tvalidation_1-rmse:0.375647\n",
      "[50]\tvalidation_0-rmse:0.259591\tvalidation_1-rmse:0.374532\n",
      "[55]\tvalidation_0-rmse:0.253667\tvalidation_1-rmse:0.374237\n",
      "[60]\tvalidation_0-rmse:0.24844\tvalidation_1-rmse:0.373773\n",
      "[65]\tvalidation_0-rmse:0.244824\tvalidation_1-rmse:0.374254\n",
      "[70]\tvalidation_0-rmse:0.239737\tvalidation_1-rmse:0.375263\n",
      "Stopping. Best iteration:\n",
      "[60]\tvalidation_0-rmse:0.24844\tvalidation_1-rmse:0.373773\n",
      "\n",
      "[08:52:08] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79633\tvalidation_1-rmse:1.78481\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11267\tvalidation_1-rmse:1.12478\n",
      "[10]\tvalidation_0-rmse:0.719584\tvalidation_1-rmse:0.747908\n",
      "[15]\tvalidation_0-rmse:0.502846\tvalidation_1-rmse:0.553392\n",
      "[20]\tvalidation_0-rmse:0.381153\tvalidation_1-rmse:0.45222\n",
      "[25]\tvalidation_0-rmse:0.320808\tvalidation_1-rmse:0.404822\n",
      "[30]\tvalidation_0-rmse:0.288117\tvalidation_1-rmse:0.383866\n",
      "[35]\tvalidation_0-rmse:0.267896\tvalidation_1-rmse:0.375284\n",
      "[40]\tvalidation_0-rmse:0.257185\tvalidation_1-rmse:0.37179\n",
      "[45]\tvalidation_0-rmse:0.247455\tvalidation_1-rmse:0.369188\n",
      "[50]\tvalidation_0-rmse:0.239644\tvalidation_1-rmse:0.367232\n",
      "[55]\tvalidation_0-rmse:0.233112\tvalidation_1-rmse:0.366494\n",
      "[60]\tvalidation_0-rmse:0.226115\tvalidation_1-rmse:0.367319\n",
      "Stopping. Best iteration:\n",
      "[53]\tvalidation_0-rmse:0.236435\tvalidation_1-rmse:0.366117\n",
      "\n",
      "[08:52:08] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79802\tvalidation_1-rmse:1.78666\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12703\tvalidation_1-rmse:1.12841\n",
      "[10]\tvalidation_0-rmse:0.731216\tvalidation_1-rmse:0.752816\n",
      "[15]\tvalidation_0-rmse:0.509245\tvalidation_1-rmse:0.553976\n",
      "[20]\tvalidation_0-rmse:0.392708\tvalidation_1-rmse:0.454472\n",
      "[25]\tvalidation_0-rmse:0.32713\tvalidation_1-rmse:0.405447\n",
      "[30]\tvalidation_0-rmse:0.290562\tvalidation_1-rmse:0.385488\n",
      "[35]\tvalidation_0-rmse:0.268829\tvalidation_1-rmse:0.375696\n",
      "[40]\tvalidation_0-rmse:0.255212\tvalidation_1-rmse:0.372033\n",
      "[45]\tvalidation_0-rmse:0.245665\tvalidation_1-rmse:0.370214\n",
      "[50]\tvalidation_0-rmse:0.235106\tvalidation_1-rmse:0.368119\n",
      "[55]\tvalidation_0-rmse:0.227811\tvalidation_1-rmse:0.368036\n",
      "[60]\tvalidation_0-rmse:0.221184\tvalidation_1-rmse:0.366143\n",
      "[65]\tvalidation_0-rmse:0.215561\tvalidation_1-rmse:0.365786\n",
      "[70]\tvalidation_0-rmse:0.210836\tvalidation_1-rmse:0.366036\n",
      "[75]\tvalidation_0-rmse:0.206637\tvalidation_1-rmse:0.364676\n",
      "[80]\tvalidation_0-rmse:0.202277\tvalidation_1-rmse:0.365791\n",
      "[85]\tvalidation_0-rmse:0.198729\tvalidation_1-rmse:0.365465\n",
      "Stopping. Best iteration:\n",
      "[76]\tvalidation_0-rmse:0.205995\tvalidation_1-rmse:0.36447\n",
      "\n",
      "[08:52:08] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79793\tvalidation_1-rmse:1.78615\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.1285\tvalidation_1-rmse:1.13328\n",
      "[10]\tvalidation_0-rmse:0.734452\tvalidation_1-rmse:0.75505\n",
      "[15]\tvalidation_0-rmse:0.512026\tvalidation_1-rmse:0.553501\n",
      "[20]\tvalidation_0-rmse:0.392906\tvalidation_1-rmse:0.453201\n",
      "[25]\tvalidation_0-rmse:0.329231\tvalidation_1-rmse:0.405877\n",
      "[30]\tvalidation_0-rmse:0.292959\tvalidation_1-rmse:0.383965\n",
      "[35]\tvalidation_0-rmse:0.272138\tvalidation_1-rmse:0.374618\n",
      "[40]\tvalidation_0-rmse:0.258356\tvalidation_1-rmse:0.370649\n",
      "[45]\tvalidation_0-rmse:0.247574\tvalidation_1-rmse:0.369283\n",
      "[50]\tvalidation_0-rmse:0.238362\tvalidation_1-rmse:0.368488\n",
      "[55]\tvalidation_0-rmse:0.231988\tvalidation_1-rmse:0.368137\n",
      "[60]\tvalidation_0-rmse:0.226547\tvalidation_1-rmse:0.367806\n",
      "[65]\tvalidation_0-rmse:0.222468\tvalidation_1-rmse:0.368153\n",
      "[70]\tvalidation_0-rmse:0.217389\tvalidation_1-rmse:0.368405\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.223594\tvalidation_1-rmse:0.367794\n",
      "\n",
      "[08:52:08] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79626\tvalidation_1-rmse:1.78503\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12897\tvalidation_1-rmse:1.13456\n",
      "[10]\tvalidation_0-rmse:0.734761\tvalidation_1-rmse:0.757676\n",
      "[15]\tvalidation_0-rmse:0.513857\tvalidation_1-rmse:0.559246\n",
      "[20]\tvalidation_0-rmse:0.394566\tvalidation_1-rmse:0.461915\n",
      "[25]\tvalidation_0-rmse:0.328424\tvalidation_1-rmse:0.415648\n",
      "[30]\tvalidation_0-rmse:0.29186\tvalidation_1-rmse:0.393342\n",
      "[35]\tvalidation_0-rmse:0.2698\tvalidation_1-rmse:0.383848\n",
      "[40]\tvalidation_0-rmse:0.254487\tvalidation_1-rmse:0.379983\n",
      "[45]\tvalidation_0-rmse:0.245195\tvalidation_1-rmse:0.378546\n",
      "[50]\tvalidation_0-rmse:0.235738\tvalidation_1-rmse:0.377624\n",
      "[55]\tvalidation_0-rmse:0.229972\tvalidation_1-rmse:0.377193\n",
      "[60]\tvalidation_0-rmse:0.224341\tvalidation_1-rmse:0.376848\n",
      "[65]\tvalidation_0-rmse:0.218174\tvalidation_1-rmse:0.375843\n",
      "[70]\tvalidation_0-rmse:0.213275\tvalidation_1-rmse:0.376371\n",
      "[75]\tvalidation_0-rmse:0.209558\tvalidation_1-rmse:0.376197\n",
      "[80]\tvalidation_0-rmse:0.205212\tvalidation_1-rmse:0.376944\n",
      "Stopping. Best iteration:\n",
      "[73]\tvalidation_0-rmse:0.210994\tvalidation_1-rmse:0.375669\n",
      "\n",
      "[08:52:08] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.796\tvalidation_1-rmse:1.78471\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12713\tvalidation_1-rmse:1.13236\n",
      "[10]\tvalidation_0-rmse:0.731847\tvalidation_1-rmse:0.754436\n",
      "[15]\tvalidation_0-rmse:0.509207\tvalidation_1-rmse:0.55537\n",
      "[20]\tvalidation_0-rmse:0.391151\tvalidation_1-rmse:0.45954\n",
      "[25]\tvalidation_0-rmse:0.327464\tvalidation_1-rmse:0.414048\n",
      "[30]\tvalidation_0-rmse:0.292292\tvalidation_1-rmse:0.392056\n",
      "[35]\tvalidation_0-rmse:0.273116\tvalidation_1-rmse:0.384421\n",
      "[40]\tvalidation_0-rmse:0.260679\tvalidation_1-rmse:0.381161\n",
      "[45]\tvalidation_0-rmse:0.251198\tvalidation_1-rmse:0.380135\n",
      "[50]\tvalidation_0-rmse:0.243661\tvalidation_1-rmse:0.379609\n",
      "[55]\tvalidation_0-rmse:0.237119\tvalidation_1-rmse:0.378695\n",
      "[60]\tvalidation_0-rmse:0.232269\tvalidation_1-rmse:0.379214\n",
      "Stopping. Best iteration:\n",
      "[54]\tvalidation_0-rmse:0.238061\tvalidation_1-rmse:0.378574\n",
      "\n",
      "[08:52:09] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79642\tvalidation_1-rmse:1.7846\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11316\tvalidation_1-rmse:1.12223\n",
      "[10]\tvalidation_0-rmse:0.721741\tvalidation_1-rmse:0.745172\n",
      "[15]\tvalidation_0-rmse:0.5071\tvalidation_1-rmse:0.551192\n",
      "[20]\tvalidation_0-rmse:0.388798\tvalidation_1-rmse:0.450634\n",
      "[25]\tvalidation_0-rmse:0.329746\tvalidation_1-rmse:0.405946\n",
      "[30]\tvalidation_0-rmse:0.299598\tvalidation_1-rmse:0.386954\n",
      "[35]\tvalidation_0-rmse:0.282761\tvalidation_1-rmse:0.377792\n",
      "[40]\tvalidation_0-rmse:0.274318\tvalidation_1-rmse:0.374764\n",
      "[45]\tvalidation_0-rmse:0.26423\tvalidation_1-rmse:0.373965\n",
      "[50]\tvalidation_0-rmse:0.257918\tvalidation_1-rmse:0.371235\n",
      "[55]\tvalidation_0-rmse:0.250489\tvalidation_1-rmse:0.370673\n",
      "[60]\tvalidation_0-rmse:0.244453\tvalidation_1-rmse:0.369002\n",
      "[65]\tvalidation_0-rmse:0.240554\tvalidation_1-rmse:0.369352\n",
      "[70]\tvalidation_0-rmse:0.236514\tvalidation_1-rmse:0.36985\n",
      "Stopping. Best iteration:\n",
      "[61]\tvalidation_0-rmse:0.243274\tvalidation_1-rmse:0.368939\n",
      "\n",
      "[08:52:09] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79802\tvalidation_1-rmse:1.78587\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12789\tvalidation_1-rmse:1.12716\n",
      "[10]\tvalidation_0-rmse:0.733492\tvalidation_1-rmse:0.747764\n",
      "[15]\tvalidation_0-rmse:0.514339\tvalidation_1-rmse:0.548014\n",
      "[20]\tvalidation_0-rmse:0.400729\tvalidation_1-rmse:0.450011\n",
      "[25]\tvalidation_0-rmse:0.33875\tvalidation_1-rmse:0.399496\n",
      "[30]\tvalidation_0-rmse:0.304376\tvalidation_1-rmse:0.378026\n",
      "[35]\tvalidation_0-rmse:0.285189\tvalidation_1-rmse:0.369022\n",
      "[40]\tvalidation_0-rmse:0.272586\tvalidation_1-rmse:0.364174\n",
      "[45]\tvalidation_0-rmse:0.264009\tvalidation_1-rmse:0.361922\n",
      "[50]\tvalidation_0-rmse:0.255542\tvalidation_1-rmse:0.360527\n",
      "[55]\tvalidation_0-rmse:0.248554\tvalidation_1-rmse:0.360459\n",
      "[60]\tvalidation_0-rmse:0.241987\tvalidation_1-rmse:0.358443\n",
      "[65]\tvalidation_0-rmse:0.237326\tvalidation_1-rmse:0.357847\n",
      "[70]\tvalidation_0-rmse:0.232239\tvalidation_1-rmse:0.357498\n",
      "[75]\tvalidation_0-rmse:0.227911\tvalidation_1-rmse:0.357595\n",
      "[80]\tvalidation_0-rmse:0.223704\tvalidation_1-rmse:0.358013\n",
      "[85]\tvalidation_0-rmse:0.219842\tvalidation_1-rmse:0.357949\n",
      "Stopping. Best iteration:\n",
      "[77]\tvalidation_0-rmse:0.226019\tvalidation_1-rmse:0.357149\n",
      "\n",
      "[08:52:09] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79793\tvalidation_1-rmse:1.78591\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12917\tvalidation_1-rmse:1.13408\n",
      "[10]\tvalidation_0-rmse:0.736497\tvalidation_1-rmse:0.754251\n",
      "[15]\tvalidation_0-rmse:0.51694\tvalidation_1-rmse:0.553667\n",
      "[20]\tvalidation_0-rmse:0.400577\tvalidation_1-rmse:0.45585\n",
      "[25]\tvalidation_0-rmse:0.339449\tvalidation_1-rmse:0.408296\n",
      "[30]\tvalidation_0-rmse:0.305147\tvalidation_1-rmse:0.387752\n",
      "[35]\tvalidation_0-rmse:0.286103\tvalidation_1-rmse:0.37819\n",
      "[40]\tvalidation_0-rmse:0.275916\tvalidation_1-rmse:0.374516\n",
      "[45]\tvalidation_0-rmse:0.266851\tvalidation_1-rmse:0.372745\n",
      "[50]\tvalidation_0-rmse:0.256993\tvalidation_1-rmse:0.371241\n",
      "[55]\tvalidation_0-rmse:0.251081\tvalidation_1-rmse:0.370043\n",
      "[60]\tvalidation_0-rmse:0.245548\tvalidation_1-rmse:0.369649\n",
      "[65]\tvalidation_0-rmse:0.239857\tvalidation_1-rmse:0.36919\n",
      "[70]\tvalidation_0-rmse:0.235005\tvalidation_1-rmse:0.368912\n",
      "[75]\tvalidation_0-rmse:0.229798\tvalidation_1-rmse:0.370284\n",
      "Stopping. Best iteration:\n",
      "[69]\tvalidation_0-rmse:0.235846\tvalidation_1-rmse:0.368421\n",
      "\n",
      "[08:52:09] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79615\tvalidation_1-rmse:1.78485\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13005\tvalidation_1-rmse:1.1341\n",
      "[10]\tvalidation_0-rmse:0.737756\tvalidation_1-rmse:0.75612\n",
      "[15]\tvalidation_0-rmse:0.519381\tvalidation_1-rmse:0.555522\n",
      "[20]\tvalidation_0-rmse:0.404825\tvalidation_1-rmse:0.457393\n",
      "[25]\tvalidation_0-rmse:0.342741\tvalidation_1-rmse:0.408563\n",
      "[30]\tvalidation_0-rmse:0.309118\tvalidation_1-rmse:0.389082\n",
      "[35]\tvalidation_0-rmse:0.289916\tvalidation_1-rmse:0.379144\n",
      "[40]\tvalidation_0-rmse:0.27636\tvalidation_1-rmse:0.375386\n",
      "[45]\tvalidation_0-rmse:0.265307\tvalidation_1-rmse:0.373768\n",
      "[50]\tvalidation_0-rmse:0.255225\tvalidation_1-rmse:0.373328\n",
      "[55]\tvalidation_0-rmse:0.249794\tvalidation_1-rmse:0.372279\n",
      "[60]\tvalidation_0-rmse:0.24366\tvalidation_1-rmse:0.370641\n",
      "[65]\tvalidation_0-rmse:0.238738\tvalidation_1-rmse:0.370554\n",
      "[70]\tvalidation_0-rmse:0.23317\tvalidation_1-rmse:0.370545\n",
      "[75]\tvalidation_0-rmse:0.229069\tvalidation_1-rmse:0.369436\n",
      "[80]\tvalidation_0-rmse:0.225605\tvalidation_1-rmse:0.369867\n",
      "Stopping. Best iteration:\n",
      "[73]\tvalidation_0-rmse:0.230383\tvalidation_1-rmse:0.369368\n",
      "\n",
      "[08:52:10] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79602\tvalidation_1-rmse:1.78473\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12756\tvalidation_1-rmse:1.1315\n",
      "[10]\tvalidation_0-rmse:0.732466\tvalidation_1-rmse:0.753535\n",
      "[15]\tvalidation_0-rmse:0.511994\tvalidation_1-rmse:0.551604\n",
      "[20]\tvalidation_0-rmse:0.398393\tvalidation_1-rmse:0.45364\n",
      "[25]\tvalidation_0-rmse:0.337384\tvalidation_1-rmse:0.408026\n",
      "[30]\tvalidation_0-rmse:0.305423\tvalidation_1-rmse:0.387491\n",
      "[35]\tvalidation_0-rmse:0.286767\tvalidation_1-rmse:0.38059\n",
      "[40]\tvalidation_0-rmse:0.275528\tvalidation_1-rmse:0.376201\n",
      "[45]\tvalidation_0-rmse:0.267359\tvalidation_1-rmse:0.375647\n",
      "[50]\tvalidation_0-rmse:0.259591\tvalidation_1-rmse:0.374532\n",
      "[55]\tvalidation_0-rmse:0.253667\tvalidation_1-rmse:0.374237\n",
      "[60]\tvalidation_0-rmse:0.24844\tvalidation_1-rmse:0.373773\n",
      "[65]\tvalidation_0-rmse:0.244824\tvalidation_1-rmse:0.374254\n",
      "[70]\tvalidation_0-rmse:0.239737\tvalidation_1-rmse:0.375263\n",
      "Stopping. Best iteration:\n",
      "[60]\tvalidation_0-rmse:0.24844\tvalidation_1-rmse:0.373773\n",
      "\n",
      "[08:52:10] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79622\tvalidation_1-rmse:1.78433\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11153\tvalidation_1-rmse:1.12333\n",
      "[10]\tvalidation_0-rmse:0.718458\tvalidation_1-rmse:0.749993\n",
      "[15]\tvalidation_0-rmse:0.499255\tvalidation_1-rmse:0.556273\n",
      "[20]\tvalidation_0-rmse:0.374837\tvalidation_1-rmse:0.455274\n",
      "[25]\tvalidation_0-rmse:0.311229\tvalidation_1-rmse:0.408113\n",
      "[30]\tvalidation_0-rmse:0.275637\tvalidation_1-rmse:0.387859\n",
      "[35]\tvalidation_0-rmse:0.254947\tvalidation_1-rmse:0.379731\n",
      "[40]\tvalidation_0-rmse:0.242595\tvalidation_1-rmse:0.374518\n",
      "[45]\tvalidation_0-rmse:0.232355\tvalidation_1-rmse:0.372552\n",
      "[50]\tvalidation_0-rmse:0.225854\tvalidation_1-rmse:0.370406\n",
      "[55]\tvalidation_0-rmse:0.218441\tvalidation_1-rmse:0.370233\n",
      "[60]\tvalidation_0-rmse:0.212013\tvalidation_1-rmse:0.370685\n",
      "[65]\tvalidation_0-rmse:0.207482\tvalidation_1-rmse:0.370536\n",
      "[70]\tvalidation_0-rmse:0.203434\tvalidation_1-rmse:0.371416\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.209029\tvalidation_1-rmse:0.370211\n",
      "\n",
      "[08:52:10] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79797\tvalidation_1-rmse:1.78676\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12678\tvalidation_1-rmse:1.12959\n",
      "[10]\tvalidation_0-rmse:0.730607\tvalidation_1-rmse:0.757089\n",
      "[15]\tvalidation_0-rmse:0.506067\tvalidation_1-rmse:0.556352\n",
      "[20]\tvalidation_0-rmse:0.38461\tvalidation_1-rmse:0.45764\n",
      "[25]\tvalidation_0-rmse:0.31688\tvalidation_1-rmse:0.410529\n",
      "[30]\tvalidation_0-rmse:0.278814\tvalidation_1-rmse:0.390416\n",
      "[35]\tvalidation_0-rmse:0.255447\tvalidation_1-rmse:0.381389\n",
      "[40]\tvalidation_0-rmse:0.241887\tvalidation_1-rmse:0.376428\n",
      "[45]\tvalidation_0-rmse:0.232805\tvalidation_1-rmse:0.373314\n",
      "[50]\tvalidation_0-rmse:0.22367\tvalidation_1-rmse:0.371759\n",
      "[55]\tvalidation_0-rmse:0.216736\tvalidation_1-rmse:0.371392\n",
      "[60]\tvalidation_0-rmse:0.210673\tvalidation_1-rmse:0.371749\n",
      "Stopping. Best iteration:\n",
      "[53]\tvalidation_0-rmse:0.218544\tvalidation_1-rmse:0.371154\n",
      "\n",
      "[08:52:10] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79789\tvalidation_1-rmse:1.78613\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12794\tvalidation_1-rmse:1.13374\n",
      "[10]\tvalidation_0-rmse:0.733675\tvalidation_1-rmse:0.755319\n",
      "[15]\tvalidation_0-rmse:0.509854\tvalidation_1-rmse:0.55434\n",
      "[20]\tvalidation_0-rmse:0.386643\tvalidation_1-rmse:0.454524\n",
      "[25]\tvalidation_0-rmse:0.317034\tvalidation_1-rmse:0.404037\n",
      "[30]\tvalidation_0-rmse:0.278749\tvalidation_1-rmse:0.384248\n",
      "[35]\tvalidation_0-rmse:0.256712\tvalidation_1-rmse:0.375802\n",
      "[40]\tvalidation_0-rmse:0.242546\tvalidation_1-rmse:0.372336\n",
      "[45]\tvalidation_0-rmse:0.232125\tvalidation_1-rmse:0.371233\n",
      "[50]\tvalidation_0-rmse:0.223019\tvalidation_1-rmse:0.370527\n",
      "[55]\tvalidation_0-rmse:0.217198\tvalidation_1-rmse:0.370612\n",
      "[60]\tvalidation_0-rmse:0.21022\tvalidation_1-rmse:0.370876\n",
      "Stopping. Best iteration:\n",
      "[53]\tvalidation_0-rmse:0.219714\tvalidation_1-rmse:0.370271\n",
      "\n",
      "[08:52:11] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79611\tvalidation_1-rmse:1.78489\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12727\tvalidation_1-rmse:1.13675\n",
      "[10]\tvalidation_0-rmse:0.73105\tvalidation_1-rmse:0.76181\n",
      "[15]\tvalidation_0-rmse:0.507903\tvalidation_1-rmse:0.563567\n",
      "[20]\tvalidation_0-rmse:0.38647\tvalidation_1-rmse:0.466489\n",
      "[25]\tvalidation_0-rmse:0.316879\tvalidation_1-rmse:0.418283\n",
      "[30]\tvalidation_0-rmse:0.27868\tvalidation_1-rmse:0.398044\n",
      "[35]\tvalidation_0-rmse:0.256244\tvalidation_1-rmse:0.388055\n",
      "[40]\tvalidation_0-rmse:0.241163\tvalidation_1-rmse:0.381425\n",
      "[45]\tvalidation_0-rmse:0.230794\tvalidation_1-rmse:0.379983\n",
      "[50]\tvalidation_0-rmse:0.221156\tvalidation_1-rmse:0.379438\n",
      "[55]\tvalidation_0-rmse:0.213652\tvalidation_1-rmse:0.379408\n",
      "[60]\tvalidation_0-rmse:0.209144\tvalidation_1-rmse:0.378471\n",
      "[65]\tvalidation_0-rmse:0.20397\tvalidation_1-rmse:0.376969\n",
      "[70]\tvalidation_0-rmse:0.19967\tvalidation_1-rmse:0.376657\n",
      "[75]\tvalidation_0-rmse:0.196093\tvalidation_1-rmse:0.376218\n",
      "[80]\tvalidation_0-rmse:0.193499\tvalidation_1-rmse:0.375494\n",
      "[85]\tvalidation_0-rmse:0.190532\tvalidation_1-rmse:0.375343\n",
      "[90]\tvalidation_0-rmse:0.187125\tvalidation_1-rmse:0.375768\n",
      "Stopping. Best iteration:\n",
      "[84]\tvalidation_0-rmse:0.190951\tvalidation_1-rmse:0.375289\n",
      "\n",
      "[08:52:11] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.7959\tvalidation_1-rmse:1.785\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12677\tvalidation_1-rmse:1.13462\n",
      "[10]\tvalidation_0-rmse:0.73071\tvalidation_1-rmse:0.758096\n",
      "[15]\tvalidation_0-rmse:0.504315\tvalidation_1-rmse:0.556977\n",
      "[20]\tvalidation_0-rmse:0.382501\tvalidation_1-rmse:0.45949\n",
      "[25]\tvalidation_0-rmse:0.315611\tvalidation_1-rmse:0.415192\n",
      "[30]\tvalidation_0-rmse:0.278726\tvalidation_1-rmse:0.394977\n",
      "[35]\tvalidation_0-rmse:0.256999\tvalidation_1-rmse:0.386806\n",
      "[40]\tvalidation_0-rmse:0.243641\tvalidation_1-rmse:0.383256\n",
      "[45]\tvalidation_0-rmse:0.23382\tvalidation_1-rmse:0.380981\n",
      "[50]\tvalidation_0-rmse:0.225649\tvalidation_1-rmse:0.379231\n",
      "[55]\tvalidation_0-rmse:0.220167\tvalidation_1-rmse:0.3785\n",
      "[60]\tvalidation_0-rmse:0.215563\tvalidation_1-rmse:0.377997\n",
      "[65]\tvalidation_0-rmse:0.210713\tvalidation_1-rmse:0.37836\n",
      "Stopping. Best iteration:\n",
      "[59]\tvalidation_0-rmse:0.216456\tvalidation_1-rmse:0.377886\n",
      "\n",
      "[08:52:11] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79616\tvalidation_1-rmse:1.78428\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11569\tvalidation_1-rmse:1.12213\n",
      "[10]\tvalidation_0-rmse:0.729602\tvalidation_1-rmse:0.745648\n",
      "[15]\tvalidation_0-rmse:0.523083\tvalidation_1-rmse:0.550521\n",
      "[20]\tvalidation_0-rmse:0.411329\tvalidation_1-rmse:0.448002\n",
      "[25]\tvalidation_0-rmse:0.35974\tvalidation_1-rmse:0.400908\n",
      "[30]\tvalidation_0-rmse:0.334303\tvalidation_1-rmse:0.381822\n",
      "[35]\tvalidation_0-rmse:0.322187\tvalidation_1-rmse:0.372105\n",
      "[40]\tvalidation_0-rmse:0.314885\tvalidation_1-rmse:0.367836\n",
      "[45]\tvalidation_0-rmse:0.306699\tvalidation_1-rmse:0.364114\n",
      "[50]\tvalidation_0-rmse:0.300616\tvalidation_1-rmse:0.36271\n",
      "[55]\tvalidation_0-rmse:0.295377\tvalidation_1-rmse:0.362687\n",
      "[60]\tvalidation_0-rmse:0.291197\tvalidation_1-rmse:0.362315\n",
      "[65]\tvalidation_0-rmse:0.288119\tvalidation_1-rmse:0.361996\n",
      "[70]\tvalidation_0-rmse:0.284895\tvalidation_1-rmse:0.360439\n",
      "[75]\tvalidation_0-rmse:0.281633\tvalidation_1-rmse:0.359178\n",
      "[80]\tvalidation_0-rmse:0.277768\tvalidation_1-rmse:0.358541\n",
      "[85]\tvalidation_0-rmse:0.27506\tvalidation_1-rmse:0.359466\n",
      "[90]\tvalidation_0-rmse:0.271096\tvalidation_1-rmse:0.359299\n",
      "Stopping. Best iteration:\n",
      "[81]\tvalidation_0-rmse:0.2774\tvalidation_1-rmse:0.358536\n",
      "\n",
      "[08:52:11] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79835\tvalidation_1-rmse:1.78556\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13143\tvalidation_1-rmse:1.12643\n",
      "[10]\tvalidation_0-rmse:0.741529\tvalidation_1-rmse:0.746895\n",
      "[15]\tvalidation_0-rmse:0.530547\tvalidation_1-rmse:0.54735\n",
      "[20]\tvalidation_0-rmse:0.426076\tvalidation_1-rmse:0.449866\n",
      "[25]\tvalidation_0-rmse:0.370466\tvalidation_1-rmse:0.402303\n",
      "[30]\tvalidation_0-rmse:0.341876\tvalidation_1-rmse:0.38303\n",
      "[35]\tvalidation_0-rmse:0.326404\tvalidation_1-rmse:0.373672\n",
      "[40]\tvalidation_0-rmse:0.316251\tvalidation_1-rmse:0.368675\n",
      "[45]\tvalidation_0-rmse:0.309503\tvalidation_1-rmse:0.365661\n",
      "[50]\tvalidation_0-rmse:0.301739\tvalidation_1-rmse:0.361998\n",
      "[55]\tvalidation_0-rmse:0.29654\tvalidation_1-rmse:0.36137\n",
      "[60]\tvalidation_0-rmse:0.29137\tvalidation_1-rmse:0.359809\n",
      "[65]\tvalidation_0-rmse:0.287162\tvalidation_1-rmse:0.359409\n",
      "[70]\tvalidation_0-rmse:0.283067\tvalidation_1-rmse:0.359979\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.288503\tvalidation_1-rmse:0.359351\n",
      "\n",
      "[08:52:12] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79808\tvalidation_1-rmse:1.78526\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13268\tvalidation_1-rmse:1.12972\n",
      "[10]\tvalidation_0-rmse:0.744353\tvalidation_1-rmse:0.75026\n",
      "[15]\tvalidation_0-rmse:0.529972\tvalidation_1-rmse:0.545997\n",
      "[20]\tvalidation_0-rmse:0.421538\tvalidation_1-rmse:0.448715\n",
      "[25]\tvalidation_0-rmse:0.368352\tvalidation_1-rmse:0.403255\n",
      "[30]\tvalidation_0-rmse:0.340252\tvalidation_1-rmse:0.385096\n",
      "[35]\tvalidation_0-rmse:0.325761\tvalidation_1-rmse:0.375385\n",
      "[40]\tvalidation_0-rmse:0.316819\tvalidation_1-rmse:0.37195\n",
      "[45]\tvalidation_0-rmse:0.310316\tvalidation_1-rmse:0.369912\n",
      "[50]\tvalidation_0-rmse:0.303157\tvalidation_1-rmse:0.367805\n",
      "[55]\tvalidation_0-rmse:0.298504\tvalidation_1-rmse:0.367183\n",
      "[60]\tvalidation_0-rmse:0.293795\tvalidation_1-rmse:0.365981\n",
      "[65]\tvalidation_0-rmse:0.29012\tvalidation_1-rmse:0.366457\n",
      "[70]\tvalidation_0-rmse:0.286213\tvalidation_1-rmse:0.366467\n",
      "[75]\tvalidation_0-rmse:0.282531\tvalidation_1-rmse:0.366707\n",
      "Stopping. Best iteration:\n",
      "[68]\tvalidation_0-rmse:0.287279\tvalidation_1-rmse:0.365616\n",
      "\n",
      "[08:52:12] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79632\tvalidation_1-rmse:1.78427\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13411\tvalidation_1-rmse:1.13278\n",
      "[10]\tvalidation_0-rmse:0.745413\tvalidation_1-rmse:0.754151\n",
      "[15]\tvalidation_0-rmse:0.53288\tvalidation_1-rmse:0.552319\n",
      "[20]\tvalidation_0-rmse:0.425337\tvalidation_1-rmse:0.455724\n",
      "[25]\tvalidation_0-rmse:0.370999\tvalidation_1-rmse:0.412178\n",
      "[30]\tvalidation_0-rmse:0.342934\tvalidation_1-rmse:0.391715\n",
      "[35]\tvalidation_0-rmse:0.327894\tvalidation_1-rmse:0.382788\n",
      "[40]\tvalidation_0-rmse:0.318003\tvalidation_1-rmse:0.376789\n",
      "[45]\tvalidation_0-rmse:0.311635\tvalidation_1-rmse:0.375004\n",
      "[50]\tvalidation_0-rmse:0.303366\tvalidation_1-rmse:0.373218\n",
      "[55]\tvalidation_0-rmse:0.299068\tvalidation_1-rmse:0.372632\n",
      "[60]\tvalidation_0-rmse:0.294875\tvalidation_1-rmse:0.371845\n",
      "[65]\tvalidation_0-rmse:0.290275\tvalidation_1-rmse:0.371069\n",
      "[70]\tvalidation_0-rmse:0.285797\tvalidation_1-rmse:0.370264\n",
      "[75]\tvalidation_0-rmse:0.28249\tvalidation_1-rmse:0.368171\n",
      "[80]\tvalidation_0-rmse:0.279087\tvalidation_1-rmse:0.36756\n",
      "[85]\tvalidation_0-rmse:0.274939\tvalidation_1-rmse:0.367402\n",
      "[90]\tvalidation_0-rmse:0.270781\tvalidation_1-rmse:0.366771\n",
      "[95]\tvalidation_0-rmse:0.267177\tvalidation_1-rmse:0.366857\n",
      "Stopping. Best iteration:\n",
      "[89]\tvalidation_0-rmse:0.27173\tvalidation_1-rmse:0.36647\n",
      "\n",
      "[08:52:12] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79584\tvalidation_1-rmse:1.78382\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.1308\tvalidation_1-rmse:1.1312\n",
      "[10]\tvalidation_0-rmse:0.740274\tvalidation_1-rmse:0.750156\n",
      "[15]\tvalidation_0-rmse:0.527001\tvalidation_1-rmse:0.547723\n",
      "[20]\tvalidation_0-rmse:0.421592\tvalidation_1-rmse:0.44875\n",
      "[25]\tvalidation_0-rmse:0.368178\tvalidation_1-rmse:0.405226\n",
      "[30]\tvalidation_0-rmse:0.341753\tvalidation_1-rmse:0.384767\n",
      "[35]\tvalidation_0-rmse:0.32823\tvalidation_1-rmse:0.378304\n",
      "[40]\tvalidation_0-rmse:0.320174\tvalidation_1-rmse:0.374885\n",
      "[45]\tvalidation_0-rmse:0.314184\tvalidation_1-rmse:0.372483\n",
      "[50]\tvalidation_0-rmse:0.307949\tvalidation_1-rmse:0.369845\n",
      "[55]\tvalidation_0-rmse:0.303283\tvalidation_1-rmse:0.369507\n",
      "[60]\tvalidation_0-rmse:0.297863\tvalidation_1-rmse:0.368905\n",
      "[65]\tvalidation_0-rmse:0.29443\tvalidation_1-rmse:0.368643\n",
      "[70]\tvalidation_0-rmse:0.290082\tvalidation_1-rmse:0.370353\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.296155\tvalidation_1-rmse:0.368294\n",
      "\n",
      "[08:52:12] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79622\tvalidation_1-rmse:1.78433\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11153\tvalidation_1-rmse:1.12333\n",
      "[10]\tvalidation_0-rmse:0.718458\tvalidation_1-rmse:0.749993\n",
      "[15]\tvalidation_0-rmse:0.499255\tvalidation_1-rmse:0.556273\n",
      "[20]\tvalidation_0-rmse:0.374837\tvalidation_1-rmse:0.455274\n",
      "[25]\tvalidation_0-rmse:0.311229\tvalidation_1-rmse:0.408113\n",
      "[30]\tvalidation_0-rmse:0.275637\tvalidation_1-rmse:0.387859\n",
      "[35]\tvalidation_0-rmse:0.254947\tvalidation_1-rmse:0.379731\n",
      "[40]\tvalidation_0-rmse:0.242595\tvalidation_1-rmse:0.374518\n",
      "[45]\tvalidation_0-rmse:0.232355\tvalidation_1-rmse:0.372552\n",
      "[50]\tvalidation_0-rmse:0.225854\tvalidation_1-rmse:0.370406\n",
      "[55]\tvalidation_0-rmse:0.218441\tvalidation_1-rmse:0.370233\n",
      "[60]\tvalidation_0-rmse:0.212013\tvalidation_1-rmse:0.370685\n",
      "[65]\tvalidation_0-rmse:0.207482\tvalidation_1-rmse:0.370536\n",
      "[70]\tvalidation_0-rmse:0.203434\tvalidation_1-rmse:0.371416\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.209029\tvalidation_1-rmse:0.370211\n",
      "\n",
      "[08:52:12] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79797\tvalidation_1-rmse:1.78676\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12678\tvalidation_1-rmse:1.12959\n",
      "[10]\tvalidation_0-rmse:0.730607\tvalidation_1-rmse:0.757089\n",
      "[15]\tvalidation_0-rmse:0.506067\tvalidation_1-rmse:0.556352\n",
      "[20]\tvalidation_0-rmse:0.38461\tvalidation_1-rmse:0.45764\n",
      "[25]\tvalidation_0-rmse:0.31688\tvalidation_1-rmse:0.410529\n",
      "[30]\tvalidation_0-rmse:0.278814\tvalidation_1-rmse:0.390416\n",
      "[35]\tvalidation_0-rmse:0.255447\tvalidation_1-rmse:0.381389\n",
      "[40]\tvalidation_0-rmse:0.241887\tvalidation_1-rmse:0.376428\n",
      "[45]\tvalidation_0-rmse:0.232805\tvalidation_1-rmse:0.373314\n",
      "[50]\tvalidation_0-rmse:0.22367\tvalidation_1-rmse:0.371759\n",
      "[55]\tvalidation_0-rmse:0.216736\tvalidation_1-rmse:0.371392\n",
      "[60]\tvalidation_0-rmse:0.210673\tvalidation_1-rmse:0.371749\n",
      "Stopping. Best iteration:\n",
      "[53]\tvalidation_0-rmse:0.218544\tvalidation_1-rmse:0.371154\n",
      "\n",
      "[08:52:13] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79789\tvalidation_1-rmse:1.78613\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12794\tvalidation_1-rmse:1.13374\n",
      "[10]\tvalidation_0-rmse:0.733675\tvalidation_1-rmse:0.755319\n",
      "[15]\tvalidation_0-rmse:0.509854\tvalidation_1-rmse:0.55434\n",
      "[20]\tvalidation_0-rmse:0.386643\tvalidation_1-rmse:0.454524\n",
      "[25]\tvalidation_0-rmse:0.317034\tvalidation_1-rmse:0.404037\n",
      "[30]\tvalidation_0-rmse:0.278749\tvalidation_1-rmse:0.384248\n",
      "[35]\tvalidation_0-rmse:0.256712\tvalidation_1-rmse:0.375802\n",
      "[40]\tvalidation_0-rmse:0.242546\tvalidation_1-rmse:0.372336\n",
      "[45]\tvalidation_0-rmse:0.232125\tvalidation_1-rmse:0.371233\n",
      "[50]\tvalidation_0-rmse:0.223019\tvalidation_1-rmse:0.370527\n",
      "[55]\tvalidation_0-rmse:0.217198\tvalidation_1-rmse:0.370612\n",
      "[60]\tvalidation_0-rmse:0.21022\tvalidation_1-rmse:0.370876\n",
      "Stopping. Best iteration:\n",
      "[53]\tvalidation_0-rmse:0.219714\tvalidation_1-rmse:0.370271\n",
      "\n",
      "[08:52:13] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79611\tvalidation_1-rmse:1.78489\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12727\tvalidation_1-rmse:1.13675\n",
      "[10]\tvalidation_0-rmse:0.73105\tvalidation_1-rmse:0.76181\n",
      "[15]\tvalidation_0-rmse:0.507903\tvalidation_1-rmse:0.563567\n",
      "[20]\tvalidation_0-rmse:0.38647\tvalidation_1-rmse:0.466489\n",
      "[25]\tvalidation_0-rmse:0.316879\tvalidation_1-rmse:0.418283\n",
      "[30]\tvalidation_0-rmse:0.27868\tvalidation_1-rmse:0.398044\n",
      "[35]\tvalidation_0-rmse:0.256244\tvalidation_1-rmse:0.388055\n",
      "[40]\tvalidation_0-rmse:0.241163\tvalidation_1-rmse:0.381425\n",
      "[45]\tvalidation_0-rmse:0.230794\tvalidation_1-rmse:0.379983\n",
      "[50]\tvalidation_0-rmse:0.221156\tvalidation_1-rmse:0.379438\n",
      "[55]\tvalidation_0-rmse:0.213652\tvalidation_1-rmse:0.379408\n",
      "[60]\tvalidation_0-rmse:0.209144\tvalidation_1-rmse:0.378471\n",
      "[65]\tvalidation_0-rmse:0.20397\tvalidation_1-rmse:0.376969\n",
      "[70]\tvalidation_0-rmse:0.19967\tvalidation_1-rmse:0.376657\n",
      "[75]\tvalidation_0-rmse:0.196093\tvalidation_1-rmse:0.376218\n",
      "[80]\tvalidation_0-rmse:0.193499\tvalidation_1-rmse:0.375494\n",
      "[85]\tvalidation_0-rmse:0.190532\tvalidation_1-rmse:0.375343\n",
      "[90]\tvalidation_0-rmse:0.187125\tvalidation_1-rmse:0.375768\n",
      "Stopping. Best iteration:\n",
      "[84]\tvalidation_0-rmse:0.190951\tvalidation_1-rmse:0.375289\n",
      "\n",
      "[08:52:13] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.7959\tvalidation_1-rmse:1.785\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12677\tvalidation_1-rmse:1.13462\n",
      "[10]\tvalidation_0-rmse:0.73071\tvalidation_1-rmse:0.758096\n",
      "[15]\tvalidation_0-rmse:0.504315\tvalidation_1-rmse:0.556977\n",
      "[20]\tvalidation_0-rmse:0.382501\tvalidation_1-rmse:0.45949\n",
      "[25]\tvalidation_0-rmse:0.315611\tvalidation_1-rmse:0.415192\n",
      "[30]\tvalidation_0-rmse:0.278726\tvalidation_1-rmse:0.394977\n",
      "[35]\tvalidation_0-rmse:0.256999\tvalidation_1-rmse:0.386806\n",
      "[40]\tvalidation_0-rmse:0.243641\tvalidation_1-rmse:0.383256\n",
      "[45]\tvalidation_0-rmse:0.23382\tvalidation_1-rmse:0.380981\n",
      "[50]\tvalidation_0-rmse:0.225649\tvalidation_1-rmse:0.379231\n",
      "[55]\tvalidation_0-rmse:0.220167\tvalidation_1-rmse:0.3785\n",
      "[60]\tvalidation_0-rmse:0.215563\tvalidation_1-rmse:0.377997\n",
      "[65]\tvalidation_0-rmse:0.210713\tvalidation_1-rmse:0.37836\n",
      "Stopping. Best iteration:\n",
      "[59]\tvalidation_0-rmse:0.216456\tvalidation_1-rmse:0.377886\n",
      "\n",
      "[08:52:14] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79642\tvalidation_1-rmse:1.7846\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11316\tvalidation_1-rmse:1.12223\n",
      "[10]\tvalidation_0-rmse:0.721741\tvalidation_1-rmse:0.745172\n",
      "[15]\tvalidation_0-rmse:0.5071\tvalidation_1-rmse:0.551192\n",
      "[20]\tvalidation_0-rmse:0.388798\tvalidation_1-rmse:0.450634\n",
      "[25]\tvalidation_0-rmse:0.329746\tvalidation_1-rmse:0.405946\n",
      "[30]\tvalidation_0-rmse:0.299598\tvalidation_1-rmse:0.386954\n",
      "[35]\tvalidation_0-rmse:0.282761\tvalidation_1-rmse:0.377792\n",
      "[40]\tvalidation_0-rmse:0.274318\tvalidation_1-rmse:0.374764\n",
      "[45]\tvalidation_0-rmse:0.26423\tvalidation_1-rmse:0.373965\n",
      "[50]\tvalidation_0-rmse:0.257918\tvalidation_1-rmse:0.371235\n",
      "[55]\tvalidation_0-rmse:0.250489\tvalidation_1-rmse:0.370673\n",
      "[60]\tvalidation_0-rmse:0.244453\tvalidation_1-rmse:0.369002\n",
      "[65]\tvalidation_0-rmse:0.240554\tvalidation_1-rmse:0.369352\n",
      "[70]\tvalidation_0-rmse:0.236514\tvalidation_1-rmse:0.36985\n",
      "Stopping. Best iteration:\n",
      "[61]\tvalidation_0-rmse:0.243274\tvalidation_1-rmse:0.368939\n",
      "\n",
      "[08:52:14] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79802\tvalidation_1-rmse:1.78587\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12789\tvalidation_1-rmse:1.12716\n",
      "[10]\tvalidation_0-rmse:0.733492\tvalidation_1-rmse:0.747764\n",
      "[15]\tvalidation_0-rmse:0.514339\tvalidation_1-rmse:0.548014\n",
      "[20]\tvalidation_0-rmse:0.400729\tvalidation_1-rmse:0.450011\n",
      "[25]\tvalidation_0-rmse:0.33875\tvalidation_1-rmse:0.399496\n",
      "[30]\tvalidation_0-rmse:0.304376\tvalidation_1-rmse:0.378026\n",
      "[35]\tvalidation_0-rmse:0.285189\tvalidation_1-rmse:0.369022\n",
      "[40]\tvalidation_0-rmse:0.272586\tvalidation_1-rmse:0.364174\n",
      "[45]\tvalidation_0-rmse:0.264009\tvalidation_1-rmse:0.361922\n",
      "[50]\tvalidation_0-rmse:0.255542\tvalidation_1-rmse:0.360527\n",
      "[55]\tvalidation_0-rmse:0.248554\tvalidation_1-rmse:0.360459\n",
      "[60]\tvalidation_0-rmse:0.241987\tvalidation_1-rmse:0.358443\n",
      "[65]\tvalidation_0-rmse:0.237326\tvalidation_1-rmse:0.357847\n",
      "[70]\tvalidation_0-rmse:0.232239\tvalidation_1-rmse:0.357498\n",
      "[75]\tvalidation_0-rmse:0.227911\tvalidation_1-rmse:0.357595\n",
      "[80]\tvalidation_0-rmse:0.223704\tvalidation_1-rmse:0.358013\n",
      "[85]\tvalidation_0-rmse:0.219842\tvalidation_1-rmse:0.357949\n",
      "Stopping. Best iteration:\n",
      "[77]\tvalidation_0-rmse:0.226019\tvalidation_1-rmse:0.357149\n",
      "\n",
      "[08:52:14] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79793\tvalidation_1-rmse:1.78591\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12917\tvalidation_1-rmse:1.13408\n",
      "[10]\tvalidation_0-rmse:0.736497\tvalidation_1-rmse:0.754251\n",
      "[15]\tvalidation_0-rmse:0.51694\tvalidation_1-rmse:0.553667\n",
      "[20]\tvalidation_0-rmse:0.400577\tvalidation_1-rmse:0.45585\n",
      "[25]\tvalidation_0-rmse:0.339449\tvalidation_1-rmse:0.408296\n",
      "[30]\tvalidation_0-rmse:0.305147\tvalidation_1-rmse:0.387752\n",
      "[35]\tvalidation_0-rmse:0.286103\tvalidation_1-rmse:0.37819\n",
      "[40]\tvalidation_0-rmse:0.275916\tvalidation_1-rmse:0.374516\n",
      "[45]\tvalidation_0-rmse:0.266851\tvalidation_1-rmse:0.372745\n",
      "[50]\tvalidation_0-rmse:0.256993\tvalidation_1-rmse:0.371241\n",
      "[55]\tvalidation_0-rmse:0.251081\tvalidation_1-rmse:0.370043\n",
      "[60]\tvalidation_0-rmse:0.245548\tvalidation_1-rmse:0.369649\n",
      "[65]\tvalidation_0-rmse:0.239857\tvalidation_1-rmse:0.36919\n",
      "[70]\tvalidation_0-rmse:0.235005\tvalidation_1-rmse:0.368912\n",
      "[75]\tvalidation_0-rmse:0.229798\tvalidation_1-rmse:0.370284\n",
      "Stopping. Best iteration:\n",
      "[69]\tvalidation_0-rmse:0.235846\tvalidation_1-rmse:0.368421\n",
      "\n",
      "[08:52:14] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79615\tvalidation_1-rmse:1.78485\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13005\tvalidation_1-rmse:1.1341\n",
      "[10]\tvalidation_0-rmse:0.737756\tvalidation_1-rmse:0.75612\n",
      "[15]\tvalidation_0-rmse:0.519381\tvalidation_1-rmse:0.555522\n",
      "[20]\tvalidation_0-rmse:0.404825\tvalidation_1-rmse:0.457393\n",
      "[25]\tvalidation_0-rmse:0.342741\tvalidation_1-rmse:0.408563\n",
      "[30]\tvalidation_0-rmse:0.309118\tvalidation_1-rmse:0.389082\n",
      "[35]\tvalidation_0-rmse:0.289916\tvalidation_1-rmse:0.379144\n",
      "[40]\tvalidation_0-rmse:0.27636\tvalidation_1-rmse:0.375386\n",
      "[45]\tvalidation_0-rmse:0.265307\tvalidation_1-rmse:0.373768\n",
      "[50]\tvalidation_0-rmse:0.255225\tvalidation_1-rmse:0.373328\n",
      "[55]\tvalidation_0-rmse:0.249794\tvalidation_1-rmse:0.372279\n",
      "[60]\tvalidation_0-rmse:0.24366\tvalidation_1-rmse:0.370641\n",
      "[65]\tvalidation_0-rmse:0.238738\tvalidation_1-rmse:0.370554\n",
      "[70]\tvalidation_0-rmse:0.23317\tvalidation_1-rmse:0.370545\n",
      "[75]\tvalidation_0-rmse:0.229069\tvalidation_1-rmse:0.369436\n",
      "[80]\tvalidation_0-rmse:0.225605\tvalidation_1-rmse:0.369867\n",
      "Stopping. Best iteration:\n",
      "[73]\tvalidation_0-rmse:0.230383\tvalidation_1-rmse:0.369368\n",
      "\n",
      "[08:52:14] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79602\tvalidation_1-rmse:1.78473\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12756\tvalidation_1-rmse:1.1315\n",
      "[10]\tvalidation_0-rmse:0.732466\tvalidation_1-rmse:0.753535\n",
      "[15]\tvalidation_0-rmse:0.511994\tvalidation_1-rmse:0.551604\n",
      "[20]\tvalidation_0-rmse:0.398393\tvalidation_1-rmse:0.45364\n",
      "[25]\tvalidation_0-rmse:0.337384\tvalidation_1-rmse:0.408026\n",
      "[30]\tvalidation_0-rmse:0.305423\tvalidation_1-rmse:0.387491\n",
      "[35]\tvalidation_0-rmse:0.286767\tvalidation_1-rmse:0.38059\n",
      "[40]\tvalidation_0-rmse:0.275528\tvalidation_1-rmse:0.376201\n",
      "[45]\tvalidation_0-rmse:0.267359\tvalidation_1-rmse:0.375647\n",
      "[50]\tvalidation_0-rmse:0.259591\tvalidation_1-rmse:0.374532\n",
      "[55]\tvalidation_0-rmse:0.253667\tvalidation_1-rmse:0.374237\n",
      "[60]\tvalidation_0-rmse:0.24844\tvalidation_1-rmse:0.373773\n",
      "[65]\tvalidation_0-rmse:0.244824\tvalidation_1-rmse:0.374254\n",
      "[70]\tvalidation_0-rmse:0.239737\tvalidation_1-rmse:0.375263\n",
      "Stopping. Best iteration:\n",
      "[60]\tvalidation_0-rmse:0.24844\tvalidation_1-rmse:0.373773\n",
      "\n",
      "[08:52:15] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79633\tvalidation_1-rmse:1.78481\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11267\tvalidation_1-rmse:1.12478\n",
      "[10]\tvalidation_0-rmse:0.719584\tvalidation_1-rmse:0.747908\n",
      "[15]\tvalidation_0-rmse:0.502846\tvalidation_1-rmse:0.553392\n",
      "[20]\tvalidation_0-rmse:0.381153\tvalidation_1-rmse:0.45222\n",
      "[25]\tvalidation_0-rmse:0.320808\tvalidation_1-rmse:0.404822\n",
      "[30]\tvalidation_0-rmse:0.288117\tvalidation_1-rmse:0.383866\n",
      "[35]\tvalidation_0-rmse:0.267896\tvalidation_1-rmse:0.375284\n",
      "[40]\tvalidation_0-rmse:0.257185\tvalidation_1-rmse:0.37179\n",
      "[45]\tvalidation_0-rmse:0.247455\tvalidation_1-rmse:0.369188\n",
      "[50]\tvalidation_0-rmse:0.239644\tvalidation_1-rmse:0.367232\n",
      "[55]\tvalidation_0-rmse:0.233112\tvalidation_1-rmse:0.366494\n",
      "[60]\tvalidation_0-rmse:0.226115\tvalidation_1-rmse:0.367319\n",
      "Stopping. Best iteration:\n",
      "[53]\tvalidation_0-rmse:0.236435\tvalidation_1-rmse:0.366117\n",
      "\n",
      "[08:52:15] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79802\tvalidation_1-rmse:1.78666\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12703\tvalidation_1-rmse:1.12841\n",
      "[10]\tvalidation_0-rmse:0.731216\tvalidation_1-rmse:0.752816\n",
      "[15]\tvalidation_0-rmse:0.509245\tvalidation_1-rmse:0.553976\n",
      "[20]\tvalidation_0-rmse:0.392708\tvalidation_1-rmse:0.454472\n",
      "[25]\tvalidation_0-rmse:0.32713\tvalidation_1-rmse:0.405447\n",
      "[30]\tvalidation_0-rmse:0.290562\tvalidation_1-rmse:0.385488\n",
      "[35]\tvalidation_0-rmse:0.268829\tvalidation_1-rmse:0.375696\n",
      "[40]\tvalidation_0-rmse:0.255212\tvalidation_1-rmse:0.372033\n",
      "[45]\tvalidation_0-rmse:0.245665\tvalidation_1-rmse:0.370214\n",
      "[50]\tvalidation_0-rmse:0.235106\tvalidation_1-rmse:0.368119\n",
      "[55]\tvalidation_0-rmse:0.227811\tvalidation_1-rmse:0.368036\n",
      "[60]\tvalidation_0-rmse:0.221184\tvalidation_1-rmse:0.366143\n",
      "[65]\tvalidation_0-rmse:0.215561\tvalidation_1-rmse:0.365786\n",
      "[70]\tvalidation_0-rmse:0.210836\tvalidation_1-rmse:0.366036\n",
      "[75]\tvalidation_0-rmse:0.206637\tvalidation_1-rmse:0.364676\n",
      "[80]\tvalidation_0-rmse:0.202277\tvalidation_1-rmse:0.365791\n",
      "[85]\tvalidation_0-rmse:0.198729\tvalidation_1-rmse:0.365465\n",
      "Stopping. Best iteration:\n",
      "[76]\tvalidation_0-rmse:0.205995\tvalidation_1-rmse:0.36447\n",
      "\n",
      "[08:52:15] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79793\tvalidation_1-rmse:1.78615\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.1285\tvalidation_1-rmse:1.13328\n",
      "[10]\tvalidation_0-rmse:0.734452\tvalidation_1-rmse:0.75505\n",
      "[15]\tvalidation_0-rmse:0.512026\tvalidation_1-rmse:0.553501\n",
      "[20]\tvalidation_0-rmse:0.392906\tvalidation_1-rmse:0.453201\n",
      "[25]\tvalidation_0-rmse:0.329231\tvalidation_1-rmse:0.405877\n",
      "[30]\tvalidation_0-rmse:0.292959\tvalidation_1-rmse:0.383965\n",
      "[35]\tvalidation_0-rmse:0.272138\tvalidation_1-rmse:0.374618\n",
      "[40]\tvalidation_0-rmse:0.258356\tvalidation_1-rmse:0.370649\n",
      "[45]\tvalidation_0-rmse:0.247574\tvalidation_1-rmse:0.369283\n",
      "[50]\tvalidation_0-rmse:0.238362\tvalidation_1-rmse:0.368488\n",
      "[55]\tvalidation_0-rmse:0.231988\tvalidation_1-rmse:0.368137\n",
      "[60]\tvalidation_0-rmse:0.226547\tvalidation_1-rmse:0.367806\n",
      "[65]\tvalidation_0-rmse:0.222468\tvalidation_1-rmse:0.368153\n",
      "[70]\tvalidation_0-rmse:0.217389\tvalidation_1-rmse:0.368405\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.223594\tvalidation_1-rmse:0.367794\n",
      "\n",
      "[08:52:15] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79626\tvalidation_1-rmse:1.78503\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12897\tvalidation_1-rmse:1.13456\n",
      "[10]\tvalidation_0-rmse:0.734761\tvalidation_1-rmse:0.757676\n",
      "[15]\tvalidation_0-rmse:0.513857\tvalidation_1-rmse:0.559246\n",
      "[20]\tvalidation_0-rmse:0.394566\tvalidation_1-rmse:0.461915\n",
      "[25]\tvalidation_0-rmse:0.328424\tvalidation_1-rmse:0.415648\n",
      "[30]\tvalidation_0-rmse:0.29186\tvalidation_1-rmse:0.393342\n",
      "[35]\tvalidation_0-rmse:0.2698\tvalidation_1-rmse:0.383848\n",
      "[40]\tvalidation_0-rmse:0.254487\tvalidation_1-rmse:0.379983\n",
      "[45]\tvalidation_0-rmse:0.245195\tvalidation_1-rmse:0.378546\n",
      "[50]\tvalidation_0-rmse:0.235738\tvalidation_1-rmse:0.377624\n",
      "[55]\tvalidation_0-rmse:0.229972\tvalidation_1-rmse:0.377193\n",
      "[60]\tvalidation_0-rmse:0.224341\tvalidation_1-rmse:0.376848\n",
      "[65]\tvalidation_0-rmse:0.218174\tvalidation_1-rmse:0.375843\n",
      "[70]\tvalidation_0-rmse:0.213275\tvalidation_1-rmse:0.376371\n",
      "[75]\tvalidation_0-rmse:0.209558\tvalidation_1-rmse:0.376197\n",
      "[80]\tvalidation_0-rmse:0.205212\tvalidation_1-rmse:0.376944\n",
      "Stopping. Best iteration:\n",
      "[73]\tvalidation_0-rmse:0.210994\tvalidation_1-rmse:0.375669\n",
      "\n",
      "[08:52:16] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.796\tvalidation_1-rmse:1.78471\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12713\tvalidation_1-rmse:1.13236\n",
      "[10]\tvalidation_0-rmse:0.731847\tvalidation_1-rmse:0.754436\n",
      "[15]\tvalidation_0-rmse:0.509207\tvalidation_1-rmse:0.55537\n",
      "[20]\tvalidation_0-rmse:0.391151\tvalidation_1-rmse:0.45954\n",
      "[25]\tvalidation_0-rmse:0.327464\tvalidation_1-rmse:0.414048\n",
      "[30]\tvalidation_0-rmse:0.292292\tvalidation_1-rmse:0.392056\n",
      "[35]\tvalidation_0-rmse:0.273116\tvalidation_1-rmse:0.384421\n",
      "[40]\tvalidation_0-rmse:0.260679\tvalidation_1-rmse:0.381161\n",
      "[45]\tvalidation_0-rmse:0.251198\tvalidation_1-rmse:0.380135\n",
      "[50]\tvalidation_0-rmse:0.243661\tvalidation_1-rmse:0.379609\n",
      "[55]\tvalidation_0-rmse:0.237119\tvalidation_1-rmse:0.378695\n",
      "[60]\tvalidation_0-rmse:0.232269\tvalidation_1-rmse:0.379214\n",
      "Stopping. Best iteration:\n",
      "[54]\tvalidation_0-rmse:0.238061\tvalidation_1-rmse:0.378574\n",
      "\n",
      "[08:52:16] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79622\tvalidation_1-rmse:1.78433\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11153\tvalidation_1-rmse:1.12333\n",
      "[10]\tvalidation_0-rmse:0.718458\tvalidation_1-rmse:0.749993\n",
      "[15]\tvalidation_0-rmse:0.499255\tvalidation_1-rmse:0.556273\n",
      "[20]\tvalidation_0-rmse:0.374837\tvalidation_1-rmse:0.455274\n",
      "[25]\tvalidation_0-rmse:0.311229\tvalidation_1-rmse:0.408113\n",
      "[30]\tvalidation_0-rmse:0.275637\tvalidation_1-rmse:0.387859\n",
      "[35]\tvalidation_0-rmse:0.254947\tvalidation_1-rmse:0.379731\n",
      "[40]\tvalidation_0-rmse:0.242595\tvalidation_1-rmse:0.374518\n",
      "[45]\tvalidation_0-rmse:0.232355\tvalidation_1-rmse:0.372552\n",
      "[50]\tvalidation_0-rmse:0.225854\tvalidation_1-rmse:0.370406\n",
      "[55]\tvalidation_0-rmse:0.218441\tvalidation_1-rmse:0.370233\n",
      "[60]\tvalidation_0-rmse:0.212013\tvalidation_1-rmse:0.370685\n",
      "[65]\tvalidation_0-rmse:0.207482\tvalidation_1-rmse:0.370536\n",
      "[70]\tvalidation_0-rmse:0.203434\tvalidation_1-rmse:0.371416\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.209029\tvalidation_1-rmse:0.370211\n",
      "\n",
      "[08:52:16] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79797\tvalidation_1-rmse:1.78676\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12678\tvalidation_1-rmse:1.12959\n",
      "[10]\tvalidation_0-rmse:0.730607\tvalidation_1-rmse:0.757089\n",
      "[15]\tvalidation_0-rmse:0.506067\tvalidation_1-rmse:0.556352\n",
      "[20]\tvalidation_0-rmse:0.38461\tvalidation_1-rmse:0.45764\n",
      "[25]\tvalidation_0-rmse:0.31688\tvalidation_1-rmse:0.410529\n",
      "[30]\tvalidation_0-rmse:0.278814\tvalidation_1-rmse:0.390416\n",
      "[35]\tvalidation_0-rmse:0.255447\tvalidation_1-rmse:0.381389\n",
      "[40]\tvalidation_0-rmse:0.241887\tvalidation_1-rmse:0.376428\n",
      "[45]\tvalidation_0-rmse:0.232805\tvalidation_1-rmse:0.373314\n",
      "[50]\tvalidation_0-rmse:0.22367\tvalidation_1-rmse:0.371759\n",
      "[55]\tvalidation_0-rmse:0.216736\tvalidation_1-rmse:0.371392\n",
      "[60]\tvalidation_0-rmse:0.210673\tvalidation_1-rmse:0.371749\n",
      "Stopping. Best iteration:\n",
      "[53]\tvalidation_0-rmse:0.218544\tvalidation_1-rmse:0.371154\n",
      "\n",
      "[08:52:16] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79789\tvalidation_1-rmse:1.78613\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12794\tvalidation_1-rmse:1.13374\n",
      "[10]\tvalidation_0-rmse:0.733675\tvalidation_1-rmse:0.755319\n",
      "[15]\tvalidation_0-rmse:0.509854\tvalidation_1-rmse:0.55434\n",
      "[20]\tvalidation_0-rmse:0.386643\tvalidation_1-rmse:0.454524\n",
      "[25]\tvalidation_0-rmse:0.317034\tvalidation_1-rmse:0.404037\n",
      "[30]\tvalidation_0-rmse:0.278749\tvalidation_1-rmse:0.384248\n",
      "[35]\tvalidation_0-rmse:0.256712\tvalidation_1-rmse:0.375802\n",
      "[40]\tvalidation_0-rmse:0.242546\tvalidation_1-rmse:0.372336\n",
      "[45]\tvalidation_0-rmse:0.232125\tvalidation_1-rmse:0.371233\n",
      "[50]\tvalidation_0-rmse:0.223019\tvalidation_1-rmse:0.370527\n",
      "[55]\tvalidation_0-rmse:0.217198\tvalidation_1-rmse:0.370612\n",
      "[60]\tvalidation_0-rmse:0.21022\tvalidation_1-rmse:0.370876\n",
      "Stopping. Best iteration:\n",
      "[53]\tvalidation_0-rmse:0.219714\tvalidation_1-rmse:0.370271\n",
      "\n",
      "[08:52:17] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79611\tvalidation_1-rmse:1.78489\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12727\tvalidation_1-rmse:1.13675\n",
      "[10]\tvalidation_0-rmse:0.73105\tvalidation_1-rmse:0.76181\n",
      "[15]\tvalidation_0-rmse:0.507903\tvalidation_1-rmse:0.563567\n",
      "[20]\tvalidation_0-rmse:0.38647\tvalidation_1-rmse:0.466489\n",
      "[25]\tvalidation_0-rmse:0.316879\tvalidation_1-rmse:0.418283\n",
      "[30]\tvalidation_0-rmse:0.27868\tvalidation_1-rmse:0.398044\n",
      "[35]\tvalidation_0-rmse:0.256244\tvalidation_1-rmse:0.388055\n",
      "[40]\tvalidation_0-rmse:0.241163\tvalidation_1-rmse:0.381425\n",
      "[45]\tvalidation_0-rmse:0.230794\tvalidation_1-rmse:0.379983\n",
      "[50]\tvalidation_0-rmse:0.221156\tvalidation_1-rmse:0.379438\n",
      "[55]\tvalidation_0-rmse:0.213652\tvalidation_1-rmse:0.379408\n",
      "[60]\tvalidation_0-rmse:0.209144\tvalidation_1-rmse:0.378471\n",
      "[65]\tvalidation_0-rmse:0.20397\tvalidation_1-rmse:0.376969\n",
      "[70]\tvalidation_0-rmse:0.19967\tvalidation_1-rmse:0.376657\n",
      "[75]\tvalidation_0-rmse:0.196093\tvalidation_1-rmse:0.376218\n",
      "[80]\tvalidation_0-rmse:0.193499\tvalidation_1-rmse:0.375494\n",
      "[85]\tvalidation_0-rmse:0.190532\tvalidation_1-rmse:0.375343\n",
      "[90]\tvalidation_0-rmse:0.187125\tvalidation_1-rmse:0.375768\n",
      "Stopping. Best iteration:\n",
      "[84]\tvalidation_0-rmse:0.190951\tvalidation_1-rmse:0.375289\n",
      "\n",
      "[08:52:17] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.7959\tvalidation_1-rmse:1.785\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12677\tvalidation_1-rmse:1.13462\n",
      "[10]\tvalidation_0-rmse:0.73071\tvalidation_1-rmse:0.758096\n",
      "[15]\tvalidation_0-rmse:0.504315\tvalidation_1-rmse:0.556977\n",
      "[20]\tvalidation_0-rmse:0.382501\tvalidation_1-rmse:0.45949\n",
      "[25]\tvalidation_0-rmse:0.315611\tvalidation_1-rmse:0.415192\n",
      "[30]\tvalidation_0-rmse:0.278726\tvalidation_1-rmse:0.394977\n",
      "[35]\tvalidation_0-rmse:0.256999\tvalidation_1-rmse:0.386806\n",
      "[40]\tvalidation_0-rmse:0.243641\tvalidation_1-rmse:0.383256\n",
      "[45]\tvalidation_0-rmse:0.23382\tvalidation_1-rmse:0.380981\n",
      "[50]\tvalidation_0-rmse:0.225649\tvalidation_1-rmse:0.379231\n",
      "[55]\tvalidation_0-rmse:0.220167\tvalidation_1-rmse:0.3785\n",
      "[60]\tvalidation_0-rmse:0.215563\tvalidation_1-rmse:0.377997\n",
      "[65]\tvalidation_0-rmse:0.210713\tvalidation_1-rmse:0.37836\n",
      "Stopping. Best iteration:\n",
      "[59]\tvalidation_0-rmse:0.216456\tvalidation_1-rmse:0.377886\n",
      "\n",
      "[08:52:17] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79633\tvalidation_1-rmse:1.78481\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11267\tvalidation_1-rmse:1.12478\n",
      "[10]\tvalidation_0-rmse:0.719584\tvalidation_1-rmse:0.747908\n",
      "[15]\tvalidation_0-rmse:0.502846\tvalidation_1-rmse:0.553392\n",
      "[20]\tvalidation_0-rmse:0.381153\tvalidation_1-rmse:0.45222\n",
      "[25]\tvalidation_0-rmse:0.320808\tvalidation_1-rmse:0.404822\n",
      "[30]\tvalidation_0-rmse:0.288117\tvalidation_1-rmse:0.383866\n",
      "[35]\tvalidation_0-rmse:0.267896\tvalidation_1-rmse:0.375284\n",
      "[40]\tvalidation_0-rmse:0.257185\tvalidation_1-rmse:0.37179\n",
      "[45]\tvalidation_0-rmse:0.247455\tvalidation_1-rmse:0.369188\n",
      "[50]\tvalidation_0-rmse:0.239644\tvalidation_1-rmse:0.367232\n",
      "[55]\tvalidation_0-rmse:0.233112\tvalidation_1-rmse:0.366494\n",
      "[60]\tvalidation_0-rmse:0.226115\tvalidation_1-rmse:0.367319\n",
      "Stopping. Best iteration:\n",
      "[53]\tvalidation_0-rmse:0.236435\tvalidation_1-rmse:0.366117\n",
      "\n",
      "[08:52:17] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79802\tvalidation_1-rmse:1.78666\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12703\tvalidation_1-rmse:1.12841\n",
      "[10]\tvalidation_0-rmse:0.731216\tvalidation_1-rmse:0.752816\n",
      "[15]\tvalidation_0-rmse:0.509245\tvalidation_1-rmse:0.553976\n",
      "[20]\tvalidation_0-rmse:0.392708\tvalidation_1-rmse:0.454472\n",
      "[25]\tvalidation_0-rmse:0.32713\tvalidation_1-rmse:0.405447\n",
      "[30]\tvalidation_0-rmse:0.290562\tvalidation_1-rmse:0.385488\n",
      "[35]\tvalidation_0-rmse:0.268829\tvalidation_1-rmse:0.375696\n",
      "[40]\tvalidation_0-rmse:0.255212\tvalidation_1-rmse:0.372033\n",
      "[45]\tvalidation_0-rmse:0.245665\tvalidation_1-rmse:0.370214\n",
      "[50]\tvalidation_0-rmse:0.235106\tvalidation_1-rmse:0.368119\n",
      "[55]\tvalidation_0-rmse:0.227811\tvalidation_1-rmse:0.368036\n",
      "[60]\tvalidation_0-rmse:0.221184\tvalidation_1-rmse:0.366143\n",
      "[65]\tvalidation_0-rmse:0.215561\tvalidation_1-rmse:0.365786\n",
      "[70]\tvalidation_0-rmse:0.210836\tvalidation_1-rmse:0.366036\n",
      "[75]\tvalidation_0-rmse:0.206637\tvalidation_1-rmse:0.364676\n",
      "[80]\tvalidation_0-rmse:0.202277\tvalidation_1-rmse:0.365791\n",
      "[85]\tvalidation_0-rmse:0.198729\tvalidation_1-rmse:0.365465\n",
      "Stopping. Best iteration:\n",
      "[76]\tvalidation_0-rmse:0.205995\tvalidation_1-rmse:0.36447\n",
      "\n",
      "[08:52:18] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79793\tvalidation_1-rmse:1.78615\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.1285\tvalidation_1-rmse:1.13328\n",
      "[10]\tvalidation_0-rmse:0.734452\tvalidation_1-rmse:0.75505\n",
      "[15]\tvalidation_0-rmse:0.512026\tvalidation_1-rmse:0.553501\n",
      "[20]\tvalidation_0-rmse:0.392906\tvalidation_1-rmse:0.453201\n",
      "[25]\tvalidation_0-rmse:0.329231\tvalidation_1-rmse:0.405877\n",
      "[30]\tvalidation_0-rmse:0.292959\tvalidation_1-rmse:0.383965\n",
      "[35]\tvalidation_0-rmse:0.272138\tvalidation_1-rmse:0.374618\n",
      "[40]\tvalidation_0-rmse:0.258356\tvalidation_1-rmse:0.370649\n",
      "[45]\tvalidation_0-rmse:0.247574\tvalidation_1-rmse:0.369283\n",
      "[50]\tvalidation_0-rmse:0.238362\tvalidation_1-rmse:0.368488\n",
      "[55]\tvalidation_0-rmse:0.231988\tvalidation_1-rmse:0.368137\n",
      "[60]\tvalidation_0-rmse:0.226547\tvalidation_1-rmse:0.367806\n",
      "[65]\tvalidation_0-rmse:0.222468\tvalidation_1-rmse:0.368153\n",
      "[70]\tvalidation_0-rmse:0.217389\tvalidation_1-rmse:0.368405\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.223594\tvalidation_1-rmse:0.367794\n",
      "\n",
      "[08:52:18] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79626\tvalidation_1-rmse:1.78503\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12897\tvalidation_1-rmse:1.13456\n",
      "[10]\tvalidation_0-rmse:0.734761\tvalidation_1-rmse:0.757676\n",
      "[15]\tvalidation_0-rmse:0.513857\tvalidation_1-rmse:0.559246\n",
      "[20]\tvalidation_0-rmse:0.394566\tvalidation_1-rmse:0.461915\n",
      "[25]\tvalidation_0-rmse:0.328424\tvalidation_1-rmse:0.415648\n",
      "[30]\tvalidation_0-rmse:0.29186\tvalidation_1-rmse:0.393342\n",
      "[35]\tvalidation_0-rmse:0.2698\tvalidation_1-rmse:0.383848\n",
      "[40]\tvalidation_0-rmse:0.254487\tvalidation_1-rmse:0.379983\n",
      "[45]\tvalidation_0-rmse:0.245195\tvalidation_1-rmse:0.378546\n",
      "[50]\tvalidation_0-rmse:0.235738\tvalidation_1-rmse:0.377624\n",
      "[55]\tvalidation_0-rmse:0.229972\tvalidation_1-rmse:0.377193\n",
      "[60]\tvalidation_0-rmse:0.224341\tvalidation_1-rmse:0.376848\n",
      "[65]\tvalidation_0-rmse:0.218174\tvalidation_1-rmse:0.375843\n",
      "[70]\tvalidation_0-rmse:0.213275\tvalidation_1-rmse:0.376371\n",
      "[75]\tvalidation_0-rmse:0.209558\tvalidation_1-rmse:0.376197\n",
      "[80]\tvalidation_0-rmse:0.205212\tvalidation_1-rmse:0.376944\n",
      "Stopping. Best iteration:\n",
      "[73]\tvalidation_0-rmse:0.210994\tvalidation_1-rmse:0.375669\n",
      "\n",
      "[08:52:18] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.796\tvalidation_1-rmse:1.78471\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12713\tvalidation_1-rmse:1.13236\n",
      "[10]\tvalidation_0-rmse:0.731847\tvalidation_1-rmse:0.754436\n",
      "[15]\tvalidation_0-rmse:0.509207\tvalidation_1-rmse:0.55537\n",
      "[20]\tvalidation_0-rmse:0.391151\tvalidation_1-rmse:0.45954\n",
      "[25]\tvalidation_0-rmse:0.327464\tvalidation_1-rmse:0.414048\n",
      "[30]\tvalidation_0-rmse:0.292292\tvalidation_1-rmse:0.392056\n",
      "[35]\tvalidation_0-rmse:0.273116\tvalidation_1-rmse:0.384421\n",
      "[40]\tvalidation_0-rmse:0.260679\tvalidation_1-rmse:0.381161\n",
      "[45]\tvalidation_0-rmse:0.251198\tvalidation_1-rmse:0.380135\n",
      "[50]\tvalidation_0-rmse:0.243661\tvalidation_1-rmse:0.379609\n",
      "[55]\tvalidation_0-rmse:0.237119\tvalidation_1-rmse:0.378695\n",
      "[60]\tvalidation_0-rmse:0.232269\tvalidation_1-rmse:0.379214\n",
      "Stopping. Best iteration:\n",
      "[54]\tvalidation_0-rmse:0.238061\tvalidation_1-rmse:0.378574\n",
      "\n",
      "[08:52:18] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79616\tvalidation_1-rmse:1.78428\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11569\tvalidation_1-rmse:1.12213\n",
      "[10]\tvalidation_0-rmse:0.729602\tvalidation_1-rmse:0.745648\n",
      "[15]\tvalidation_0-rmse:0.523083\tvalidation_1-rmse:0.550521\n",
      "[20]\tvalidation_0-rmse:0.411329\tvalidation_1-rmse:0.448002\n",
      "[25]\tvalidation_0-rmse:0.35974\tvalidation_1-rmse:0.400908\n",
      "[30]\tvalidation_0-rmse:0.334303\tvalidation_1-rmse:0.381822\n",
      "[35]\tvalidation_0-rmse:0.322187\tvalidation_1-rmse:0.372105\n",
      "[40]\tvalidation_0-rmse:0.314885\tvalidation_1-rmse:0.367836\n",
      "[45]\tvalidation_0-rmse:0.306699\tvalidation_1-rmse:0.364114\n",
      "[50]\tvalidation_0-rmse:0.300616\tvalidation_1-rmse:0.36271\n",
      "[55]\tvalidation_0-rmse:0.295377\tvalidation_1-rmse:0.362687\n",
      "[60]\tvalidation_0-rmse:0.291197\tvalidation_1-rmse:0.362315\n",
      "[65]\tvalidation_0-rmse:0.288119\tvalidation_1-rmse:0.361996\n",
      "[70]\tvalidation_0-rmse:0.284895\tvalidation_1-rmse:0.360439\n",
      "[75]\tvalidation_0-rmse:0.281633\tvalidation_1-rmse:0.359178\n",
      "[80]\tvalidation_0-rmse:0.277768\tvalidation_1-rmse:0.358541\n",
      "[85]\tvalidation_0-rmse:0.27506\tvalidation_1-rmse:0.359466\n",
      "[90]\tvalidation_0-rmse:0.271096\tvalidation_1-rmse:0.359299\n",
      "Stopping. Best iteration:\n",
      "[81]\tvalidation_0-rmse:0.2774\tvalidation_1-rmse:0.358536\n",
      "\n",
      "[08:52:19] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79835\tvalidation_1-rmse:1.78556\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13143\tvalidation_1-rmse:1.12643\n",
      "[10]\tvalidation_0-rmse:0.741529\tvalidation_1-rmse:0.746895\n",
      "[15]\tvalidation_0-rmse:0.530547\tvalidation_1-rmse:0.54735\n",
      "[20]\tvalidation_0-rmse:0.426076\tvalidation_1-rmse:0.449866\n",
      "[25]\tvalidation_0-rmse:0.370466\tvalidation_1-rmse:0.402303\n",
      "[30]\tvalidation_0-rmse:0.341876\tvalidation_1-rmse:0.38303\n",
      "[35]\tvalidation_0-rmse:0.326404\tvalidation_1-rmse:0.373672\n",
      "[40]\tvalidation_0-rmse:0.316251\tvalidation_1-rmse:0.368675\n",
      "[45]\tvalidation_0-rmse:0.309503\tvalidation_1-rmse:0.365661\n",
      "[50]\tvalidation_0-rmse:0.301739\tvalidation_1-rmse:0.361998\n",
      "[55]\tvalidation_0-rmse:0.29654\tvalidation_1-rmse:0.36137\n",
      "[60]\tvalidation_0-rmse:0.29137\tvalidation_1-rmse:0.359809\n",
      "[65]\tvalidation_0-rmse:0.287162\tvalidation_1-rmse:0.359409\n",
      "[70]\tvalidation_0-rmse:0.283067\tvalidation_1-rmse:0.359979\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.288503\tvalidation_1-rmse:0.359351\n",
      "\n",
      "[08:52:19] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79808\tvalidation_1-rmse:1.78526\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13268\tvalidation_1-rmse:1.12972\n",
      "[10]\tvalidation_0-rmse:0.744353\tvalidation_1-rmse:0.75026\n",
      "[15]\tvalidation_0-rmse:0.529972\tvalidation_1-rmse:0.545997\n",
      "[20]\tvalidation_0-rmse:0.421538\tvalidation_1-rmse:0.448715\n",
      "[25]\tvalidation_0-rmse:0.368352\tvalidation_1-rmse:0.403255\n",
      "[30]\tvalidation_0-rmse:0.340252\tvalidation_1-rmse:0.385096\n",
      "[35]\tvalidation_0-rmse:0.325761\tvalidation_1-rmse:0.375385\n",
      "[40]\tvalidation_0-rmse:0.316819\tvalidation_1-rmse:0.37195\n",
      "[45]\tvalidation_0-rmse:0.310316\tvalidation_1-rmse:0.369912\n",
      "[50]\tvalidation_0-rmse:0.303157\tvalidation_1-rmse:0.367805\n",
      "[55]\tvalidation_0-rmse:0.298504\tvalidation_1-rmse:0.367183\n",
      "[60]\tvalidation_0-rmse:0.293795\tvalidation_1-rmse:0.365981\n",
      "[65]\tvalidation_0-rmse:0.29012\tvalidation_1-rmse:0.366457\n",
      "[70]\tvalidation_0-rmse:0.286213\tvalidation_1-rmse:0.366467\n",
      "[75]\tvalidation_0-rmse:0.282531\tvalidation_1-rmse:0.366707\n",
      "Stopping. Best iteration:\n",
      "[68]\tvalidation_0-rmse:0.287279\tvalidation_1-rmse:0.365616\n",
      "\n",
      "[08:52:19] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79632\tvalidation_1-rmse:1.78427\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13411\tvalidation_1-rmse:1.13278\n",
      "[10]\tvalidation_0-rmse:0.745413\tvalidation_1-rmse:0.754151\n",
      "[15]\tvalidation_0-rmse:0.53288\tvalidation_1-rmse:0.552319\n",
      "[20]\tvalidation_0-rmse:0.425337\tvalidation_1-rmse:0.455724\n",
      "[25]\tvalidation_0-rmse:0.370999\tvalidation_1-rmse:0.412178\n",
      "[30]\tvalidation_0-rmse:0.342934\tvalidation_1-rmse:0.391715\n",
      "[35]\tvalidation_0-rmse:0.327894\tvalidation_1-rmse:0.382788\n",
      "[40]\tvalidation_0-rmse:0.318003\tvalidation_1-rmse:0.376789\n",
      "[45]\tvalidation_0-rmse:0.311635\tvalidation_1-rmse:0.375004\n",
      "[50]\tvalidation_0-rmse:0.303366\tvalidation_1-rmse:0.373218\n",
      "[55]\tvalidation_0-rmse:0.299068\tvalidation_1-rmse:0.372632\n",
      "[60]\tvalidation_0-rmse:0.294875\tvalidation_1-rmse:0.371845\n",
      "[65]\tvalidation_0-rmse:0.290275\tvalidation_1-rmse:0.371069\n",
      "[70]\tvalidation_0-rmse:0.285797\tvalidation_1-rmse:0.370264\n",
      "[75]\tvalidation_0-rmse:0.28249\tvalidation_1-rmse:0.368171\n",
      "[80]\tvalidation_0-rmse:0.279087\tvalidation_1-rmse:0.36756\n",
      "[85]\tvalidation_0-rmse:0.274939\tvalidation_1-rmse:0.367402\n",
      "[90]\tvalidation_0-rmse:0.270781\tvalidation_1-rmse:0.366771\n",
      "[95]\tvalidation_0-rmse:0.267177\tvalidation_1-rmse:0.366857\n",
      "Stopping. Best iteration:\n",
      "[89]\tvalidation_0-rmse:0.27173\tvalidation_1-rmse:0.36647\n",
      "\n",
      "[08:52:19] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79584\tvalidation_1-rmse:1.78382\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.1308\tvalidation_1-rmse:1.1312\n",
      "[10]\tvalidation_0-rmse:0.740274\tvalidation_1-rmse:0.750156\n",
      "[15]\tvalidation_0-rmse:0.527001\tvalidation_1-rmse:0.547723\n",
      "[20]\tvalidation_0-rmse:0.421592\tvalidation_1-rmse:0.44875\n",
      "[25]\tvalidation_0-rmse:0.368178\tvalidation_1-rmse:0.405226\n",
      "[30]\tvalidation_0-rmse:0.341753\tvalidation_1-rmse:0.384767\n",
      "[35]\tvalidation_0-rmse:0.32823\tvalidation_1-rmse:0.378304\n",
      "[40]\tvalidation_0-rmse:0.320174\tvalidation_1-rmse:0.374885\n",
      "[45]\tvalidation_0-rmse:0.314184\tvalidation_1-rmse:0.372483\n",
      "[50]\tvalidation_0-rmse:0.307949\tvalidation_1-rmse:0.369845\n",
      "[55]\tvalidation_0-rmse:0.303283\tvalidation_1-rmse:0.369507\n",
      "[60]\tvalidation_0-rmse:0.297863\tvalidation_1-rmse:0.368905\n",
      "[65]\tvalidation_0-rmse:0.29443\tvalidation_1-rmse:0.368643\n",
      "[70]\tvalidation_0-rmse:0.290082\tvalidation_1-rmse:0.370353\n",
      "Stopping. Best iteration:\n",
      "[63]\tvalidation_0-rmse:0.296155\tvalidation_1-rmse:0.368294\n",
      "\n",
      "[08:52:19] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79642\tvalidation_1-rmse:1.7846\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.11316\tvalidation_1-rmse:1.12223\n",
      "[10]\tvalidation_0-rmse:0.721741\tvalidation_1-rmse:0.745172\n",
      "[15]\tvalidation_0-rmse:0.5071\tvalidation_1-rmse:0.551192\n",
      "[20]\tvalidation_0-rmse:0.388798\tvalidation_1-rmse:0.450634\n",
      "[25]\tvalidation_0-rmse:0.329746\tvalidation_1-rmse:0.405946\n",
      "[30]\tvalidation_0-rmse:0.299598\tvalidation_1-rmse:0.386954\n",
      "[35]\tvalidation_0-rmse:0.282761\tvalidation_1-rmse:0.377792\n",
      "[40]\tvalidation_0-rmse:0.274318\tvalidation_1-rmse:0.374764\n",
      "[45]\tvalidation_0-rmse:0.26423\tvalidation_1-rmse:0.373965\n",
      "[50]\tvalidation_0-rmse:0.257918\tvalidation_1-rmse:0.371235\n",
      "[55]\tvalidation_0-rmse:0.250489\tvalidation_1-rmse:0.370673\n",
      "[60]\tvalidation_0-rmse:0.244453\tvalidation_1-rmse:0.369002\n",
      "[65]\tvalidation_0-rmse:0.240554\tvalidation_1-rmse:0.369352\n",
      "[70]\tvalidation_0-rmse:0.236514\tvalidation_1-rmse:0.36985\n",
      "Stopping. Best iteration:\n",
      "[61]\tvalidation_0-rmse:0.243274\tvalidation_1-rmse:0.368939\n",
      "\n",
      "[08:52:20] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79802\tvalidation_1-rmse:1.78587\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12789\tvalidation_1-rmse:1.12716\n",
      "[10]\tvalidation_0-rmse:0.733492\tvalidation_1-rmse:0.747764\n",
      "[15]\tvalidation_0-rmse:0.514339\tvalidation_1-rmse:0.548014\n",
      "[20]\tvalidation_0-rmse:0.400729\tvalidation_1-rmse:0.450011\n",
      "[25]\tvalidation_0-rmse:0.33875\tvalidation_1-rmse:0.399496\n",
      "[30]\tvalidation_0-rmse:0.304376\tvalidation_1-rmse:0.378026\n",
      "[35]\tvalidation_0-rmse:0.285189\tvalidation_1-rmse:0.369022\n",
      "[40]\tvalidation_0-rmse:0.272586\tvalidation_1-rmse:0.364174\n",
      "[45]\tvalidation_0-rmse:0.264009\tvalidation_1-rmse:0.361922\n",
      "[50]\tvalidation_0-rmse:0.255542\tvalidation_1-rmse:0.360527\n",
      "[55]\tvalidation_0-rmse:0.248554\tvalidation_1-rmse:0.360459\n",
      "[60]\tvalidation_0-rmse:0.241987\tvalidation_1-rmse:0.358443\n",
      "[65]\tvalidation_0-rmse:0.237326\tvalidation_1-rmse:0.357847\n",
      "[70]\tvalidation_0-rmse:0.232239\tvalidation_1-rmse:0.357498\n",
      "[75]\tvalidation_0-rmse:0.227911\tvalidation_1-rmse:0.357595\n",
      "[80]\tvalidation_0-rmse:0.223704\tvalidation_1-rmse:0.358013\n",
      "[85]\tvalidation_0-rmse:0.219842\tvalidation_1-rmse:0.357949\n",
      "Stopping. Best iteration:\n",
      "[77]\tvalidation_0-rmse:0.226019\tvalidation_1-rmse:0.357149\n",
      "\n",
      "[08:52:20] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79793\tvalidation_1-rmse:1.78591\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12917\tvalidation_1-rmse:1.13408\n",
      "[10]\tvalidation_0-rmse:0.736497\tvalidation_1-rmse:0.754251\n",
      "[15]\tvalidation_0-rmse:0.51694\tvalidation_1-rmse:0.553667\n",
      "[20]\tvalidation_0-rmse:0.400577\tvalidation_1-rmse:0.45585\n",
      "[25]\tvalidation_0-rmse:0.339449\tvalidation_1-rmse:0.408296\n",
      "[30]\tvalidation_0-rmse:0.305147\tvalidation_1-rmse:0.387752\n",
      "[35]\tvalidation_0-rmse:0.286103\tvalidation_1-rmse:0.37819\n",
      "[40]\tvalidation_0-rmse:0.275916\tvalidation_1-rmse:0.374516\n",
      "[45]\tvalidation_0-rmse:0.266851\tvalidation_1-rmse:0.372745\n",
      "[50]\tvalidation_0-rmse:0.256993\tvalidation_1-rmse:0.371241\n",
      "[55]\tvalidation_0-rmse:0.251081\tvalidation_1-rmse:0.370043\n",
      "[60]\tvalidation_0-rmse:0.245548\tvalidation_1-rmse:0.369649\n",
      "[65]\tvalidation_0-rmse:0.239857\tvalidation_1-rmse:0.36919\n",
      "[70]\tvalidation_0-rmse:0.235005\tvalidation_1-rmse:0.368912\n",
      "[75]\tvalidation_0-rmse:0.229798\tvalidation_1-rmse:0.370284\n",
      "Stopping. Best iteration:\n",
      "[69]\tvalidation_0-rmse:0.235846\tvalidation_1-rmse:0.368421\n",
      "\n",
      "[08:52:20] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79615\tvalidation_1-rmse:1.78485\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.13005\tvalidation_1-rmse:1.1341\n",
      "[10]\tvalidation_0-rmse:0.737756\tvalidation_1-rmse:0.75612\n",
      "[15]\tvalidation_0-rmse:0.519381\tvalidation_1-rmse:0.555522\n",
      "[20]\tvalidation_0-rmse:0.404825\tvalidation_1-rmse:0.457393\n",
      "[25]\tvalidation_0-rmse:0.342741\tvalidation_1-rmse:0.408563\n",
      "[30]\tvalidation_0-rmse:0.309118\tvalidation_1-rmse:0.389082\n",
      "[35]\tvalidation_0-rmse:0.289916\tvalidation_1-rmse:0.379144\n",
      "[40]\tvalidation_0-rmse:0.27636\tvalidation_1-rmse:0.375386\n",
      "[45]\tvalidation_0-rmse:0.265307\tvalidation_1-rmse:0.373768\n",
      "[50]\tvalidation_0-rmse:0.255225\tvalidation_1-rmse:0.373328\n",
      "[55]\tvalidation_0-rmse:0.249794\tvalidation_1-rmse:0.372279\n",
      "[60]\tvalidation_0-rmse:0.24366\tvalidation_1-rmse:0.370641\n",
      "[65]\tvalidation_0-rmse:0.238738\tvalidation_1-rmse:0.370554\n",
      "[70]\tvalidation_0-rmse:0.23317\tvalidation_1-rmse:0.370545\n",
      "[75]\tvalidation_0-rmse:0.229069\tvalidation_1-rmse:0.369436\n",
      "[80]\tvalidation_0-rmse:0.225605\tvalidation_1-rmse:0.369867\n",
      "Stopping. Best iteration:\n",
      "[73]\tvalidation_0-rmse:0.230383\tvalidation_1-rmse:0.369368\n",
      "\n",
      "[08:52:20] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79602\tvalidation_1-rmse:1.78473\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12756\tvalidation_1-rmse:1.1315\n",
      "[10]\tvalidation_0-rmse:0.732466\tvalidation_1-rmse:0.753535\n",
      "[15]\tvalidation_0-rmse:0.511994\tvalidation_1-rmse:0.551604\n",
      "[20]\tvalidation_0-rmse:0.398393\tvalidation_1-rmse:0.45364\n",
      "[25]\tvalidation_0-rmse:0.337384\tvalidation_1-rmse:0.408026\n",
      "[30]\tvalidation_0-rmse:0.305423\tvalidation_1-rmse:0.387491\n",
      "[35]\tvalidation_0-rmse:0.286767\tvalidation_1-rmse:0.38059\n",
      "[40]\tvalidation_0-rmse:0.275528\tvalidation_1-rmse:0.376201\n",
      "[45]\tvalidation_0-rmse:0.267359\tvalidation_1-rmse:0.375647\n",
      "[50]\tvalidation_0-rmse:0.259591\tvalidation_1-rmse:0.374532\n",
      "[55]\tvalidation_0-rmse:0.253667\tvalidation_1-rmse:0.374237\n",
      "[60]\tvalidation_0-rmse:0.24844\tvalidation_1-rmse:0.373773\n",
      "[65]\tvalidation_0-rmse:0.244824\tvalidation_1-rmse:0.374254\n",
      "[70]\tvalidation_0-rmse:0.239737\tvalidation_1-rmse:0.375263\n",
      "Stopping. Best iteration:\n",
      "[60]\tvalidation_0-rmse:0.24844\tvalidation_1-rmse:0.373773\n",
      "\n",
      "[08:52:21] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[0]\tvalidation_0-rmse:1.79563\tvalidation_1-rmse:1.7828\n",
      "Multiple eval metrics have been passed: 'validation_1-rmse' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-rmse hasn't improved in 10 rounds.\n",
      "[5]\tvalidation_0-rmse:1.12937\tvalidation_1-rmse:1.12539\n",
      "[10]\tvalidation_0-rmse:0.754882\tvalidation_1-rmse:0.761458\n",
      "[15]\tvalidation_0-rmse:0.541874\tvalidation_1-rmse:0.560318\n",
      "[20]\tvalidation_0-rmse:0.425857\tvalidation_1-rmse:0.455667\n",
      "[25]\tvalidation_0-rmse:0.369391\tvalidation_1-rmse:0.41057\n",
      "[30]\tvalidation_0-rmse:0.33831\tvalidation_1-rmse:0.389896\n",
      "[35]\tvalidation_0-rmse:0.319348\tvalidation_1-rmse:0.377981\n",
      "[40]\tvalidation_0-rmse:0.308115\tvalidation_1-rmse:0.374137\n",
      "[45]\tvalidation_0-rmse:0.300237\tvalidation_1-rmse:0.370865\n",
      "[50]\tvalidation_0-rmse:0.293578\tvalidation_1-rmse:0.370038\n",
      "[55]\tvalidation_0-rmse:0.286101\tvalidation_1-rmse:0.36805\n",
      "[60]\tvalidation_0-rmse:0.279839\tvalidation_1-rmse:0.366656\n",
      "[65]\tvalidation_0-rmse:0.275241\tvalidation_1-rmse:0.36661\n",
      "Stopping. Best iteration:\n",
      "[58]\tvalidation_0-rmse:0.28196\tvalidation_1-rmse:0.366468\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=5, estimator=XGBRegressor(n_estimators=500), n_iter=20,\n",
       "                   param_distributions={'colsample_bytree': array([0.6]),\n",
       "                                        'eta': array([0.01      , 0.01668101, 0.02782559, 0.04641589, 0.07742637,\n",
       "       0.12915497, 0.21544347, 0.35938137, 0.59948425, 1.        ]),\n",
       "                                        'max_depth': array([5, 6, 7, 8, 9]),\n",
       "                                        'subsample': array([0.6])},\n",
       "                   scoring='neg_mean_squared_error')"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# search!\n",
    "\n",
    "random_search.fit(\n",
    "    X_train_sel,\n",
    "    np.log10(y_train.values),\n",
    "    eval_metric='rmse',\n",
    "    eval_set=[(X_train_sel, np.log10(y_train.values)), (X_test_sel, np.log10(y_test.values))],\n",
    "    verbose=5,\n",
    "    early_stopping_rounds=10\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best MSE is 0.13913418408158776\n"
     ]
    }
   ],
   "source": [
    "# the best RMSE from the search\n",
    "\n",
    "print(f\"Best MSE is {-random_search.best_score_}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "# the best estimator from the search\n",
    "\n",
    "reg_best = random_search.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBRegressor(colsample_bytree=0.6, eta=0.01, max_depth=5, n_estimators=500,\n",
       "             subsample=0.6)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# look at the hyperparameters of the best model\n",
    "\n",
    "reg_best"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Section 6: Visualize feature importance from the XGBoost model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import xgboost as xgb\n",
    "\n",
    "fig, ax = plt.subplots(1, 1, figsize=(12, 7))\n",
    "xgb.plot_importance(reg_best, ax=ax);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Insight thru feature importance\n",
    "- DEATHS_2018 is one critical feature, which makes sense because it implies the age of the population living in the county. It is known that it is easier for old people to be infected by Covid-19\n",
    "- N_POP_CHG_2018 means the population change in 2018, which is important in this case because it implies the population liquidity in the county -- more people moving around can lead to more people infected by Covid-19"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}