6-XGB-CoreLib-Clean-Synthesis-Copy2.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import sklearn as sk\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "game_csv = pd.read_csv('./data/Training_dataset_Original.csv',na_values=[\"NA\", \"na\", \"missing\"])\n",
    "gameDf = pd.DataFrame(game_csv)\n",
    "\n",
    "dataDict = pd.read_csv('./data/Data_Dictionary.csv',na_values=\"NA\")\n",
    "\n",
    "#Leaderboard\n",
    "leader_csv = pd.read_csv('./data/Leaderboard_dataset.csv',na_values=[\"NA\", \"na\", \"missing\"])\n",
    "leaderDf = pd.DataFrame(leader_csv)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(80000, 49)\n",
      "(49, 2)\n"
     ]
    }
   ],
   "source": [
    "print(gameDf.shape)\n",
    "print(dataDict.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "dic = dict(zip(dataDict.values[:, 1], dataDict.values[:, 0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'mvar25'"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dic['Average tenure of active revolving credit cards (in days)']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#remove ID column\n",
    "del gameDf[dic['Application ID (primary key)']]\n",
    "del dic['Application ID (primary key)']\n",
    "\n",
    "leaderAppID = leaderDf['application_key']\n",
    "del leaderDf['application_key']\n",
    "leaderX = leaderDf.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "gameX = gameDf.iloc[:,:-1]\n",
    "gameY = gameDf.iloc[:,-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "range(0, 80000)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "range(gameDf[\"mvar47\"].shape[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'C'"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gameX[\"mvar47\"][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#one hot encoding C=0 L=1\n",
    "gameX['mvar47'] = pd.get_dummies(gameX['mvar47'])\n",
    "\n",
    "leaderX['mvar47'] = pd.get_dummies(leaderX['mvar47'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype('uint8')"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gameX[\"mvar47\"].dtype"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Synthesys begins"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(80000, 47)\n",
      "(25000, 47)\n"
     ]
    }
   ],
   "source": [
    "print(gameX.shape)\n",
    "print(leaderX.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "gameX = gameX.assign(var345 = (1+gameX['mvar3'])*(1+gameX['mvar4'])*(1+gameX['mvar5']))\n",
    "leaderX = leaderX.assign(var345 = (1+leaderX['mvar3'])*(1+leaderX['mvar4'])*(1+leaderX['mvar5']) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(80000, 48)\n",
      "(25000, 48)\n"
     ]
    }
   ],
   "source": [
    "print(gameX.shape)\n",
    "print(leaderX.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "gameX  = gameX.assign(mvar678 = (gameX['mvar7'] + gameX['mvar8'])/(1+gameX['mvar6']))\n",
    "leaderX = leaderX.assign(mvar678 = (leaderX['mvar7'] + leaderX['mvar8'])/(1+leaderX['mvar6']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(80000, 49)\n",
      "(25000, 49)\n"
     ]
    }
   ],
   "source": [
    "print(gameX.shape)\n",
    "print(leaderX.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#mvar36, 37, 38"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = gameX.copy()\n",
    "\n",
    "temp37 = df['mvar37'].copy()\n",
    "\n",
    "temp37[temp37.isnull()] = -1\n",
    "\n",
    "temp37[temp37<0] = 0.458344\n",
    "temp37[temp37==0] = 0.381028\n",
    "temp37[temp37==1] = 0.309995\n",
    "temp37[temp37==2] = 0.239528\n",
    "temp37[temp37==3] = 0.186923\n",
    "temp37[temp37==4] = 0.159852\n",
    "temp37[temp37==5] = 0.14485\n",
    "temp37[temp37==6] = 0.131449\n",
    "temp37[temp37==7] = 0.118592\n",
    "temp37[temp37==8] = 0.111612\n",
    "temp37[temp37==9] = 0.102641\n",
    "temp37[temp37>=10] = 0.0904\n",
    "\n",
    "df = df.assign(var37 = temp37)\n",
    "gameX = df.copy()\n",
    "\n",
    "\n",
    "\n",
    "df = leaderX.copy()\n",
    "\n",
    "temp37 = df['mvar37'].copy()\n",
    "\n",
    "temp37[temp37.isnull()] = -1\n",
    "\n",
    "temp37[temp37<0] = 0.458344\n",
    "temp37[temp37==0] = 0.381028\n",
    "temp37[temp37==1] = 0.309995\n",
    "temp37[temp37==2] = 0.239528\n",
    "temp37[temp37==3] = 0.186923\n",
    "temp37[temp37==4] = 0.159852\n",
    "temp37[temp37==5] = 0.14485\n",
    "temp37[temp37==6] = 0.131449\n",
    "temp37[temp37==7] = 0.118592\n",
    "temp37[temp37==8] = 0.111612\n",
    "temp37[temp37==9] = 0.102641\n",
    "temp37[temp37>=10] = 0.0904\n",
    "\n",
    "df = df.assign(var37 = temp37)\n",
    "leaderX = df.copy()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(80000, 50)\n",
      "(25000, 50)\n"
     ]
    }
   ],
   "source": [
    "print(gameX.shape)\n",
    "print(leaderX.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "gameX = gameX.assign(var363738 = (gameX['mvar36']*gameX['mvar37']/(1+gameX['mvar38'])) )\n",
    "leaderX = leaderX.assign(var363738 = (leaderX['mvar36']*leaderX['mvar37']/(1+leaderX['mvar38']) ) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(80000, 51)\n",
      "(25000, 51)\n"
     ]
    }
   ],
   "source": [
    "print(gameX.shape)\n",
    "print(leaderX.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#drop\n",
    "df = gameX.copy() \n",
    "leader_df = leaderX.copy()\n",
    "drop_target = ['mvar27']\n",
    "#number of misssing values\n",
    "# for i in df.columns:\n",
    "#     if (100*(80000-gameDf[i].describe()['count'])/80000) > "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "for i in drop_target:\n",
    "    del df[i]\n",
    "    del leader_df[i]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "leader_df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "gameX = df\n",
    "leaderX = leader_df "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Synthesis Ends"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(80000, 51)\n",
      "(25000, 51)\n"
     ]
    }
   ],
   "source": [
    "print(gameX.shape)\n",
    "print(leaderX.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mvar1</th>\n",
       "      <th>mvar2</th>\n",
       "      <th>mvar3</th>\n",
       "      <th>mvar4</th>\n",
       "      <th>mvar5</th>\n",
       "      <th>mvar6</th>\n",
       "      <th>mvar7</th>\n",
       "      <th>mvar8</th>\n",
       "      <th>mvar9</th>\n",
       "      <th>mvar10</th>\n",
       "      <th>...</th>\n",
       "      <th>mvar42</th>\n",
       "      <th>mvar43</th>\n",
       "      <th>mvar44</th>\n",
       "      <th>mvar45</th>\n",
       "      <th>mvar46</th>\n",
       "      <th>mvar47</th>\n",
       "      <th>var345</th>\n",
       "      <th>mvar678</th>\n",
       "      <th>var37</th>\n",
       "      <th>var363738</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1753.0</td>\n",
       "      <td>0.500100</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>206.000000</td>\n",
       "      <td>19179.000000</td>\n",
       "      <td>206.00000</td>\n",
       "      <td>14221.000000</td>\n",
       "      <td>19438.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.42105</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.856610</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>93.647343</td>\n",
       "      <td>0.090400</td>\n",
       "      <td>16.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1894.0</td>\n",
       "      <td>1.970100</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>297.000000</td>\n",
       "      <td>19820.000000</td>\n",
       "      <td>19820.00000</td>\n",
       "      <td>176895.000000</td>\n",
       "      <td>66346.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>15.0</td>\n",
       "      <td>0.943910</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>133.020134</td>\n",
       "      <td>0.090400</td>\n",
       "      <td>9.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1774.0</td>\n",
       "      <td>0.171800</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>197.000000</td>\n",
       "      <td>2563.000000</td>\n",
       "      <td>877.00000</td>\n",
       "      <td>8869.000000</td>\n",
       "      <td>3637.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.33333</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.764670</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>17.373737</td>\n",
       "      <td>0.159852</td>\n",
       "      <td>6.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1586.0</td>\n",
       "      <td>0.112300</td>\n",
       "      <td>5.299</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1606.702372</td>\n",
       "      <td>17595.807537</td>\n",
       "      <td>6773.68666</td>\n",
       "      <td>34097.436013</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.50000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.705285</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.096135</td>\n",
       "      <td>1</td>\n",
       "      <td>6.299000</td>\n",
       "      <td>2583.839092</td>\n",
       "      <td>0.381028</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1832.0</td>\n",
       "      <td>1.444200</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>179.000000</td>\n",
       "      <td>7577.000000</td>\n",
       "      <td>179.00000</td>\n",
       "      <td>21059.000000</td>\n",
       "      <td>20306.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>23.0</td>\n",
       "      <td>0.791900</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>43.088889</td>\n",
       "      <td>0.381028</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1838.0</td>\n",
       "      <td>0.054400</td>\n",
       "      <td>0.241</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>882.000000</td>\n",
       "      <td>14039.000000</td>\n",
       "      <td>10976.00000</td>\n",
       "      <td>278941.000000</td>\n",
       "      <td>74030.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>21.0</td>\n",
       "      <td>0.924620</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>1.241000</td>\n",
       "      <td>28.329558</td>\n",
       "      <td>0.090400</td>\n",
       "      <td>53.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1743.0</td>\n",
       "      <td>1.075300</td>\n",
       "      <td>0.272</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>397.000000</td>\n",
       "      <td>16141.000000</td>\n",
       "      <td>3568.00000</td>\n",
       "      <td>69482.000000</td>\n",
       "      <td>56755.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>14.0</td>\n",
       "      <td>0.713470</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>1.272000</td>\n",
       "      <td>49.520101</td>\n",
       "      <td>0.102641</td>\n",
       "      <td>10.285714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1935.0</td>\n",
       "      <td>0.560100</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>496.000000</td>\n",
       "      <td>44919.000000</td>\n",
       "      <td>17115.00000</td>\n",
       "      <td>124814.000000</td>\n",
       "      <td>124167.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>15.0</td>\n",
       "      <td>0.924030</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>124.816901</td>\n",
       "      <td>0.090400</td>\n",
       "      <td>7.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1885.0</td>\n",
       "      <td>0.081100</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>7369.000000</td>\n",
       "      <td>19264.000000</td>\n",
       "      <td>19264.00000</td>\n",
       "      <td>59956.000000</td>\n",
       "      <td>26633.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.897600</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.096135</td>\n",
       "      <td>0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>5.227680</td>\n",
       "      <td>0.186923</td>\n",
       "      <td>3.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1730.0</td>\n",
       "      <td>0.744400</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2103.000000</td>\n",
       "      <td>582.00000</td>\n",
       "      <td>16429.000000</td>\n",
       "      <td>9197.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.05882</td>\n",
       "      <td>14.0</td>\n",
       "      <td>0.405290</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2685.000000</td>\n",
       "      <td>0.309995</td>\n",
       "      <td>0.600000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1731.0</td>\n",
       "      <td>0.509700</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>2433.000000</td>\n",
       "      <td>8673.000000</td>\n",
       "      <td>8673.00000</td>\n",
       "      <td>6739.000000</td>\n",
       "      <td>7388.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.23529</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.642420</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.096135</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>7.126541</td>\n",
       "      <td>0.239528</td>\n",
       "      <td>0.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1755.0</td>\n",
       "      <td>0.301300</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>261.000000</td>\n",
       "      <td>4051.000000</td>\n",
       "      <td>261.00000</td>\n",
       "      <td>297.000000</td>\n",
       "      <td>261.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.33333</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.991420</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>16.458015</td>\n",
       "      <td>0.381028</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1739.0</td>\n",
       "      <td>2.290900</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7620.000000</td>\n",
       "      <td>7620.00000</td>\n",
       "      <td>64983.000000</td>\n",
       "      <td>14067.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.05882</td>\n",
       "      <td>16.0</td>\n",
       "      <td>0.441670</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>15240.000000</td>\n",
       "      <td>0.090400</td>\n",
       "      <td>15.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1562.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>34.611</td>\n",
       "      <td>3.7845</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1606.702372</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>6773.68666</td>\n",
       "      <td>198.000000</td>\n",
       "      <td>18.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.87500</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.845920</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>170.380829</td>\n",
       "      <td>2583.839092</td>\n",
       "      <td>0.381028</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>1798.0</td>\n",
       "      <td>5.082500</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>259.000000</td>\n",
       "      <td>3359.000000</td>\n",
       "      <td>2973.00000</td>\n",
       "      <td>22099.000000</td>\n",
       "      <td>13507.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.05882</td>\n",
       "      <td>13.0</td>\n",
       "      <td>0.433650</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>24.353846</td>\n",
       "      <td>0.239528</td>\n",
       "      <td>3.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>1684.0</td>\n",
       "      <td>1.547900</td>\n",
       "      <td>1.338</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1606.702372</td>\n",
       "      <td>17595.807537</td>\n",
       "      <td>6773.68666</td>\n",
       "      <td>34097.436013</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.50000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.705285</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.096135</td>\n",
       "      <td>1</td>\n",
       "      <td>2.338000</td>\n",
       "      <td>2583.839092</td>\n",
       "      <td>0.381028</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>1789.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.428</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>2478.000000</td>\n",
       "      <td>13874.000000</td>\n",
       "      <td>13874.00000</td>\n",
       "      <td>98011.000000</td>\n",
       "      <td>56200.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.05000</td>\n",
       "      <td>13.0</td>\n",
       "      <td>0.512310</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.096135</td>\n",
       "      <td>0</td>\n",
       "      <td>2.428000</td>\n",
       "      <td>11.193223</td>\n",
       "      <td>0.090400</td>\n",
       "      <td>6.600000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>1862.0</td>\n",
       "      <td>0.240100</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1487.000000</td>\n",
       "      <td>12310.000000</td>\n",
       "      <td>3964.00000</td>\n",
       "      <td>7928.000000</td>\n",
       "      <td>26564.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.566850</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>10.936828</td>\n",
       "      <td>0.309995</td>\n",
       "      <td>0.428571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>1767.0</td>\n",
       "      <td>0.300100</td>\n",
       "      <td>1.514</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>991.000000</td>\n",
       "      <td>2478.000000</td>\n",
       "      <td>2478.00000</td>\n",
       "      <td>8790.000000</td>\n",
       "      <td>6065.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.10000</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.705285</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.096135</td>\n",
       "      <td>1</td>\n",
       "      <td>2.514000</td>\n",
       "      <td>4.995968</td>\n",
       "      <td>0.381028</td>\n",
       "      <td>7.256161</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>1802.0</td>\n",
       "      <td>0.636400</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>96.000000</td>\n",
       "      <td>18759.000000</td>\n",
       "      <td>10905.00000</td>\n",
       "      <td>51037.000000</td>\n",
       "      <td>58544.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.594090</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>305.814433</td>\n",
       "      <td>0.131449</td>\n",
       "      <td>10.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>1780.0</td>\n",
       "      <td>1.955900</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>191.000000</td>\n",
       "      <td>12525.000000</td>\n",
       "      <td>12525.00000</td>\n",
       "      <td>18932.000000</td>\n",
       "      <td>14602.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.927520</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>130.468750</td>\n",
       "      <td>0.381028</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>1714.0</td>\n",
       "      <td>2.266400</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>31.000000</td>\n",
       "      <td>1671.000000</td>\n",
       "      <td>57.00000</td>\n",
       "      <td>396.000000</td>\n",
       "      <td>2582.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.37500</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.348360</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>54.000000</td>\n",
       "      <td>0.309995</td>\n",
       "      <td>0.428571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>1923.0</td>\n",
       "      <td>0.088000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>5946.000000</td>\n",
       "      <td>53893.000000</td>\n",
       "      <td>24739.00000</td>\n",
       "      <td>128435.000000</td>\n",
       "      <td>125679.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.885540</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>13.222129</td>\n",
       "      <td>0.090400</td>\n",
       "      <td>16.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>1743.0</td>\n",
       "      <td>1.720800</td>\n",
       "      <td>2.295</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>159.000000</td>\n",
       "      <td>3621.000000</td>\n",
       "      <td>2323.00000</td>\n",
       "      <td>9018.000000</td>\n",
       "      <td>9649.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.10000</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.594420</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>3.295000</td>\n",
       "      <td>37.150000</td>\n",
       "      <td>0.309995</td>\n",
       "      <td>0.555556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>1700.0</td>\n",
       "      <td>1.660500</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1037.000000</td>\n",
       "      <td>46.00000</td>\n",
       "      <td>892.000000</td>\n",
       "      <td>1235.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.20000</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.948260</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1083.000000</td>\n",
       "      <td>0.381028</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>1922.0</td>\n",
       "      <td>1.105629</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>4524.000000</td>\n",
       "      <td>15530.000000</td>\n",
       "      <td>14819.00000</td>\n",
       "      <td>116154.000000</td>\n",
       "      <td>100684.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.915500</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>6.706961</td>\n",
       "      <td>0.118592</td>\n",
       "      <td>15.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>1777.0</td>\n",
       "      <td>0.412000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>369.000000</td>\n",
       "      <td>6201.000000</td>\n",
       "      <td>6201.00000</td>\n",
       "      <td>66477.000000</td>\n",
       "      <td>34053.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.03846</td>\n",
       "      <td>23.0</td>\n",
       "      <td>0.151590</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.096135</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>33.518919</td>\n",
       "      <td>0.144850</td>\n",
       "      <td>7.307692</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>1811.0</td>\n",
       "      <td>0.911700</td>\n",
       "      <td>0.410</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.410</td>\n",
       "      <td>297.000000</td>\n",
       "      <td>27717.000000</td>\n",
       "      <td>9815.00000</td>\n",
       "      <td>110001.000000</td>\n",
       "      <td>69683.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>15.0</td>\n",
       "      <td>0.566710</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.988100</td>\n",
       "      <td>125.946309</td>\n",
       "      <td>0.102641</td>\n",
       "      <td>18.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>1658.0</td>\n",
       "      <td>2.382900</td>\n",
       "      <td>37.127</td>\n",
       "      <td>10.3678</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>396.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>6566.000000</td>\n",
       "      <td>4068.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.57692</td>\n",
       "      <td>8.0</td>\n",
       "      <td>0.353390</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>433.420111</td>\n",
       "      <td>396.000000</td>\n",
       "      <td>0.186923</td>\n",
       "      <td>1.400000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>1827.0</td>\n",
       "      <td>0.280100</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>678.000000</td>\n",
       "      <td>1189.000000</td>\n",
       "      <td>1189.00000</td>\n",
       "      <td>892.000000</td>\n",
       "      <td>2065.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.096135</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.502209</td>\n",
       "      <td>0.381028</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24970</th>\n",
       "      <td>1734.0</td>\n",
       "      <td>0.927900</td>\n",
       "      <td>0.520</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>256.000000</td>\n",
       "      <td>11643.000000</td>\n",
       "      <td>1524.00000</td>\n",
       "      <td>16748.000000</td>\n",
       "      <td>25359.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0.380780</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>1.520000</td>\n",
       "      <td>51.233463</td>\n",
       "      <td>0.381028</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24971</th>\n",
       "      <td>1849.0</td>\n",
       "      <td>1.830500</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>2251.000000</td>\n",
       "      <td>33275.000000</td>\n",
       "      <td>7505.00000</td>\n",
       "      <td>44843.000000</td>\n",
       "      <td>47453.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.921110</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>18.108348</td>\n",
       "      <td>0.159852</td>\n",
       "      <td>5.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24972</th>\n",
       "      <td>1739.0</td>\n",
       "      <td>1.525100</td>\n",
       "      <td>2.727</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>2.727</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>2129.000000</td>\n",
       "      <td>720.00000</td>\n",
       "      <td>5896.000000</td>\n",
       "      <td>5206.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.36842</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.529030</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>13.890529</td>\n",
       "      <td>316.555556</td>\n",
       "      <td>0.309995</td>\n",
       "      <td>1.285714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24973</th>\n",
       "      <td>1642.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.353</td>\n",
       "      <td>1.2114</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1606.702372</td>\n",
       "      <td>6290.000000</td>\n",
       "      <td>6773.68666</td>\n",
       "      <td>34097.436013</td>\n",
       "      <td>6290.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.83333</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>9.626224</td>\n",
       "      <td>2583.839092</td>\n",
       "      <td>0.458344</td>\n",
       "      <td>7.256161</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24974</th>\n",
       "      <td>1709.0</td>\n",
       "      <td>3.680100</td>\n",
       "      <td>7.583</td>\n",
       "      <td>3.1371</td>\n",
       "      <td>0.000</td>\n",
       "      <td>297.000000</td>\n",
       "      <td>8132.000000</td>\n",
       "      <td>297.00000</td>\n",
       "      <td>1041.000000</td>\n",
       "      <td>6960.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.59459</td>\n",
       "      <td>11.0</td>\n",
       "      <td>0.426800</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>35.508729</td>\n",
       "      <td>28.285235</td>\n",
       "      <td>0.381028</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24975</th>\n",
       "      <td>1890.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1982.000000</td>\n",
       "      <td>15162.000000</td>\n",
       "      <td>15162.00000</td>\n",
       "      <td>68181.000000</td>\n",
       "      <td>43334.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>15.0</td>\n",
       "      <td>0.271430</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>15.291982</td>\n",
       "      <td>0.102641</td>\n",
       "      <td>21.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24976</th>\n",
       "      <td>1727.0</td>\n",
       "      <td>1.844800</td>\n",
       "      <td>15.181</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>2234.000000</td>\n",
       "      <td>2754.000000</td>\n",
       "      <td>2234.00000</td>\n",
       "      <td>2676.000000</td>\n",
       "      <td>6539.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.66667</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.878900</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>16.181000</td>\n",
       "      <td>2.231767</td>\n",
       "      <td>0.186923</td>\n",
       "      <td>4.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24977</th>\n",
       "      <td>1789.0</td>\n",
       "      <td>1.299400</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>364.000000</td>\n",
       "      <td>17838.000000</td>\n",
       "      <td>17838.00000</td>\n",
       "      <td>8424.000000</td>\n",
       "      <td>22307.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.15385</td>\n",
       "      <td>8.0</td>\n",
       "      <td>0.696940</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>97.742466</td>\n",
       "      <td>0.144850</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24978</th>\n",
       "      <td>1706.0</td>\n",
       "      <td>0.499700</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>98.000000</td>\n",
       "      <td>297.000000</td>\n",
       "      <td>98.00000</td>\n",
       "      <td>546.000000</td>\n",
       "      <td>643.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.66667</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.989899</td>\n",
       "      <td>0.381028</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24979</th>\n",
       "      <td>1638.0</td>\n",
       "      <td>4.680200</td>\n",
       "      <td>16.751</td>\n",
       "      <td>2.6642</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1606.702372</td>\n",
       "      <td>7937.000000</td>\n",
       "      <td>6773.68666</td>\n",
       "      <td>3032.000000</td>\n",
       "      <td>10489.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.50000</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.726400</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>65.043214</td>\n",
       "      <td>2583.839092</td>\n",
       "      <td>0.381028</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24980</th>\n",
       "      <td>1855.0</td>\n",
       "      <td>1.996500</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>350.000000</td>\n",
       "      <td>30203.000000</td>\n",
       "      <td>22656.00000</td>\n",
       "      <td>168999.000000</td>\n",
       "      <td>166720.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.741060</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.096135</td>\n",
       "      <td>0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>150.595442</td>\n",
       "      <td>0.090400</td>\n",
       "      <td>40.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24981</th>\n",
       "      <td>1790.0</td>\n",
       "      <td>0.205000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>237.000000</td>\n",
       "      <td>18018.000000</td>\n",
       "      <td>13297.00000</td>\n",
       "      <td>3914.000000</td>\n",
       "      <td>1305.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.20000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.415820</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>131.575630</td>\n",
       "      <td>0.239528</td>\n",
       "      <td>0.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24982</th>\n",
       "      <td>1546.0</td>\n",
       "      <td>0.349100</td>\n",
       "      <td>50.803</td>\n",
       "      <td>4.2695</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1606.702372</td>\n",
       "      <td>17595.807537</td>\n",
       "      <td>6773.68666</td>\n",
       "      <td>3330.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.70370</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>272.975909</td>\n",
       "      <td>2583.839092</td>\n",
       "      <td>0.309995</td>\n",
       "      <td>1.888889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24983</th>\n",
       "      <td>1777.0</td>\n",
       "      <td>3.962900</td>\n",
       "      <td>0.968</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>14.000000</td>\n",
       "      <td>23707.000000</td>\n",
       "      <td>23707.00000</td>\n",
       "      <td>33432.000000</td>\n",
       "      <td>19640.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.04000</td>\n",
       "      <td>15.0</td>\n",
       "      <td>0.647520</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>1.968000</td>\n",
       "      <td>3160.933333</td>\n",
       "      <td>0.090400</td>\n",
       "      <td>11.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24984</th>\n",
       "      <td>1943.0</td>\n",
       "      <td>1.105629</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>24775.000000</td>\n",
       "      <td>24775.00000</td>\n",
       "      <td>80772.000000</td>\n",
       "      <td>69483.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.970680</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.096135</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>49550.000000</td>\n",
       "      <td>0.159852</td>\n",
       "      <td>16.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24985</th>\n",
       "      <td>1691.0</td>\n",
       "      <td>0.166300</td>\n",
       "      <td>1.979</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9591.000000</td>\n",
       "      <td>2627.00000</td>\n",
       "      <td>46903.000000</td>\n",
       "      <td>30402.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.03448</td>\n",
       "      <td>14.0</td>\n",
       "      <td>0.434410</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>2.979000</td>\n",
       "      <td>12218.000000</td>\n",
       "      <td>0.090400</td>\n",
       "      <td>28.166667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24986</th>\n",
       "      <td>1681.0</td>\n",
       "      <td>0.173900</td>\n",
       "      <td>9.006</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>6.568</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3004.000000</td>\n",
       "      <td>2251.00000</td>\n",
       "      <td>6818.000000</td>\n",
       "      <td>7783.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.71429</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.889080</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>75.725408</td>\n",
       "      <td>5255.000000</td>\n",
       "      <td>0.159852</td>\n",
       "      <td>16.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24987</th>\n",
       "      <td>1723.0</td>\n",
       "      <td>1.057000</td>\n",
       "      <td>0.285</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>248.000000</td>\n",
       "      <td>28479.000000</td>\n",
       "      <td>28479.00000</td>\n",
       "      <td>16352.000000</td>\n",
       "      <td>3117.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.723440</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.285000</td>\n",
       "      <td>228.746988</td>\n",
       "      <td>0.309995</td>\n",
       "      <td>0.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24988</th>\n",
       "      <td>1693.0</td>\n",
       "      <td>1.214500</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.00000</td>\n",
       "      <td>991.000000</td>\n",
       "      <td>118.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.366190</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.096135</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>10.526316</td>\n",
       "      <td>0.381028</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24989</th>\n",
       "      <td>1592.0</td>\n",
       "      <td>6.480500</td>\n",
       "      <td>5.250</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>1.204</td>\n",
       "      <td>1606.702372</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6773.68666</td>\n",
       "      <td>34097.436013</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.506490</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.096135</td>\n",
       "      <td>1</td>\n",
       "      <td>13.775000</td>\n",
       "      <td>2583.839092</td>\n",
       "      <td>0.458344</td>\n",
       "      <td>7.256161</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24990</th>\n",
       "      <td>1641.0</td>\n",
       "      <td>0.660100</td>\n",
       "      <td>8.614</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1606.702372</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6773.68666</td>\n",
       "      <td>34097.436013</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.096135</td>\n",
       "      <td>0</td>\n",
       "      <td>9.614000</td>\n",
       "      <td>2583.839092</td>\n",
       "      <td>0.458344</td>\n",
       "      <td>7.256161</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24991</th>\n",
       "      <td>1740.0</td>\n",
       "      <td>0.390100</td>\n",
       "      <td>4.322</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>490.000000</td>\n",
       "      <td>29460.000000</td>\n",
       "      <td>9449.00000</td>\n",
       "      <td>42047.000000</td>\n",
       "      <td>31583.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.20000</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.889570</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>5.322000</td>\n",
       "      <td>79.244399</td>\n",
       "      <td>0.102641</td>\n",
       "      <td>6.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24992</th>\n",
       "      <td>1651.0</td>\n",
       "      <td>2.490900</td>\n",
       "      <td>1.335</td>\n",
       "      <td>0.9108</td>\n",
       "      <td>0.000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>3163.000000</td>\n",
       "      <td>4.00000</td>\n",
       "      <td>1982.000000</td>\n",
       "      <td>3167.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.61538</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.937450</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>4.461718</td>\n",
       "      <td>633.400000</td>\n",
       "      <td>0.381028</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24993</th>\n",
       "      <td>1729.0</td>\n",
       "      <td>1.360100</td>\n",
       "      <td>0.218</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5675.000000</td>\n",
       "      <td>167.00000</td>\n",
       "      <td>15071.000000</td>\n",
       "      <td>10305.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.23810</td>\n",
       "      <td>13.0</td>\n",
       "      <td>0.718660</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.218000</td>\n",
       "      <td>5842.000000</td>\n",
       "      <td>0.309995</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24994</th>\n",
       "      <td>1730.0</td>\n",
       "      <td>1.670100</td>\n",
       "      <td>5.554</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>104.000000</td>\n",
       "      <td>5692.000000</td>\n",
       "      <td>496.00000</td>\n",
       "      <td>4410.000000</td>\n",
       "      <td>8125.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.30769</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.933120</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>6.554000</td>\n",
       "      <td>58.933333</td>\n",
       "      <td>0.239528</td>\n",
       "      <td>1.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24995</th>\n",
       "      <td>1715.0</td>\n",
       "      <td>12.000200</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1606.702372</td>\n",
       "      <td>991.000000</td>\n",
       "      <td>6773.68666</td>\n",
       "      <td>8751.000000</td>\n",
       "      <td>1814.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.07692</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.705285</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2583.839092</td>\n",
       "      <td>0.309995</td>\n",
       "      <td>7.256161</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24996</th>\n",
       "      <td>1800.0</td>\n",
       "      <td>0.055100</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>5946.000000</td>\n",
       "      <td>14867.000000</td>\n",
       "      <td>9613.00000</td>\n",
       "      <td>46490.000000</td>\n",
       "      <td>44722.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.05882</td>\n",
       "      <td>8.0</td>\n",
       "      <td>0.998590</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4.116361</td>\n",
       "      <td>0.309995</td>\n",
       "      <td>0.400000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24997</th>\n",
       "      <td>1686.0</td>\n",
       "      <td>1.330100</td>\n",
       "      <td>2.025</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>2.025</td>\n",
       "      <td>1606.702372</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6773.68666</td>\n",
       "      <td>4582.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.30000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.775190</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.096135</td>\n",
       "      <td>0</td>\n",
       "      <td>9.150625</td>\n",
       "      <td>2583.839092</td>\n",
       "      <td>0.144850</td>\n",
       "      <td>20.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24998</th>\n",
       "      <td>1826.0</td>\n",
       "      <td>1.105629</td>\n",
       "      <td>0.428</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>3568.000000</td>\n",
       "      <td>26277.000000</td>\n",
       "      <td>9058.00000</td>\n",
       "      <td>72739.000000</td>\n",
       "      <td>86797.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.03333</td>\n",
       "      <td>11.0</td>\n",
       "      <td>0.403430</td>\n",
       "      <td>0.070736</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.428000</td>\n",
       "      <td>9.900532</td>\n",
       "      <td>0.102641</td>\n",
       "      <td>21.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24999</th>\n",
       "      <td>1916.0</td>\n",
       "      <td>0.139400</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>18889.000000</td>\n",
       "      <td>18889.000000</td>\n",
       "      <td>18889.00000</td>\n",
       "      <td>28343.000000</td>\n",
       "      <td>51174.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.996630</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.999894</td>\n",
       "      <td>0.159852</td>\n",
       "      <td>8.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>25000 rows × 51 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        mvar1      mvar2   mvar3    mvar4  mvar5         mvar6         mvar7  \\\n",
       "0      1753.0   0.500100   0.000   0.0000  0.000    206.000000  19179.000000   \n",
       "1      1894.0   1.970100   0.000   0.0000  0.000    297.000000  19820.000000   \n",
       "2      1774.0   0.171800   0.000   0.0000  0.000    197.000000   2563.000000   \n",
       "3      1586.0   0.112300   5.299   0.0000  0.000   1606.702372  17595.807537   \n",
       "4      1832.0   1.444200   0.000   0.0000  0.000    179.000000   7577.000000   \n",
       "5      1838.0   0.054400   0.241   0.0000  0.000    882.000000  14039.000000   \n",
       "6      1743.0   1.075300   0.272   0.0000  0.000    397.000000  16141.000000   \n",
       "7      1935.0   0.560100   0.000   0.0000  0.000    496.000000  44919.000000   \n",
       "8      1885.0   0.081100   0.000   0.0000  0.000   7369.000000  19264.000000   \n",
       "9      1730.0   0.744400   0.000   0.0000  0.000      0.000000   2103.000000   \n",
       "10     1731.0   0.509700   0.000   0.0000  0.000   2433.000000   8673.000000   \n",
       "11     1755.0   0.301300   0.000   0.0000  0.000    261.000000   4051.000000   \n",
       "12     1739.0   2.290900   0.000   0.0000  0.000      0.000000   7620.000000   \n",
       "13     1562.0   0.000000  34.611   3.7845  0.000   1606.702372     18.000000   \n",
       "14     1798.0   5.082500   0.000   0.0000  0.000    259.000000   3359.000000   \n",
       "15     1684.0   1.547900   1.338   0.0000  0.000   1606.702372  17595.807537   \n",
       "16     1789.0   0.000000   1.428   0.0000  0.000   2478.000000  13874.000000   \n",
       "17     1862.0   0.240100   0.000   0.0000  0.000   1487.000000  12310.000000   \n",
       "18     1767.0   0.300100   1.514   0.0000  0.000    991.000000   2478.000000   \n",
       "19     1802.0   0.636400   0.000   0.0000  0.000     96.000000  18759.000000   \n",
       "20     1780.0   1.955900   0.000   0.0000  0.000    191.000000  12525.000000   \n",
       "21     1714.0   2.266400   0.000   0.0000  0.000     31.000000   1671.000000   \n",
       "22     1923.0   0.088000   0.000   0.0000  0.000   5946.000000  53893.000000   \n",
       "23     1743.0   1.720800   2.295   0.0000  0.000    159.000000   3621.000000   \n",
       "24     1700.0   1.660500   0.000   0.0000  0.000      0.000000   1037.000000   \n",
       "25     1922.0   1.105629   0.000   0.0000  0.000   4524.000000  15530.000000   \n",
       "26     1777.0   0.412000   0.000   0.0000  0.000    369.000000   6201.000000   \n",
       "27     1811.0   0.911700   0.410   0.0000  0.410    297.000000  27717.000000   \n",
       "28     1658.0   2.382900  37.127  10.3678  0.000      0.000000    396.000000   \n",
       "29     1827.0   0.280100   0.000   0.0000  0.000    678.000000   1189.000000   \n",
       "...       ...        ...     ...      ...    ...           ...           ...   \n",
       "24970  1734.0   0.927900   0.520   0.0000  0.000    256.000000  11643.000000   \n",
       "24971  1849.0   1.830500   0.000   0.0000  0.000   2251.000000  33275.000000   \n",
       "24972  1739.0   1.525100   2.727   0.0000  2.727      8.000000   2129.000000   \n",
       "24973  1642.0   0.000000   3.353   1.2114  0.000   1606.702372   6290.000000   \n",
       "24974  1709.0   3.680100   7.583   3.1371  0.000    297.000000   8132.000000   \n",
       "24975  1890.0   0.000000   0.000   0.0000  0.000   1982.000000  15162.000000   \n",
       "24976  1727.0   1.844800  15.181   0.0000  0.000   2234.000000   2754.000000   \n",
       "24977  1789.0   1.299400   0.000   0.0000  0.000    364.000000  17838.000000   \n",
       "24978  1706.0   0.499700   0.000   0.0000  0.000     98.000000    297.000000   \n",
       "24979  1638.0   4.680200  16.751   2.6642  0.000   1606.702372   7937.000000   \n",
       "24980  1855.0   1.996500   0.000   0.0000  0.000    350.000000  30203.000000   \n",
       "24981  1790.0   0.205000   0.000   0.0000  0.000    237.000000  18018.000000   \n",
       "24982  1546.0   0.349100  50.803   4.2695  0.000   1606.702372  17595.807537   \n",
       "24983  1777.0   3.962900   0.968   0.0000  0.000     14.000000  23707.000000   \n",
       "24984  1943.0   1.105629   0.000   0.0000  0.000      0.000000  24775.000000   \n",
       "24985  1691.0   0.166300   1.979   0.0000  0.000      0.000000   9591.000000   \n",
       "24986  1681.0   0.173900   9.006   0.0000  6.568      0.000000   3004.000000   \n",
       "24987  1723.0   1.057000   0.285   0.0000  0.000    248.000000  28479.000000   \n",
       "24988  1693.0   1.214500   0.000   0.0000  0.000     18.000000    100.000000   \n",
       "24989  1592.0   6.480500   5.250   0.0000  1.204   1606.702372      0.000000   \n",
       "24990  1641.0   0.660100   8.614   0.0000  0.000   1606.702372      0.000000   \n",
       "24991  1740.0   0.390100   4.322   0.0000  0.000    490.000000  29460.000000   \n",
       "24992  1651.0   2.490900   1.335   0.9108  0.000      4.000000   3163.000000   \n",
       "24993  1729.0   1.360100   0.218   0.0000  0.000      0.000000   5675.000000   \n",
       "24994  1730.0   1.670100   5.554   0.0000  0.000    104.000000   5692.000000   \n",
       "24995  1715.0  12.000200   0.000   0.0000  0.000   1606.702372    991.000000   \n",
       "24996  1800.0   0.055100   0.000   0.0000  0.000   5946.000000  14867.000000   \n",
       "24997  1686.0   1.330100   2.025   0.0000  2.025   1606.702372      0.000000   \n",
       "24998  1826.0   1.105629   0.428   0.0000  0.000   3568.000000  26277.000000   \n",
       "24999  1916.0   0.139400   0.000   0.0000  0.000  18889.000000  18889.000000   \n",
       "\n",
       "             mvar8          mvar9    mvar10    ...       mvar42  mvar43  \\\n",
       "0        206.00000   14221.000000   19438.0    ...      0.42105     3.0   \n",
       "1      19820.00000  176895.000000   66346.0    ...      0.00000    15.0   \n",
       "2        877.00000    8869.000000    3637.0    ...      0.33333     3.0   \n",
       "3       6773.68666   34097.436013       0.0    ...      1.50000     0.0   \n",
       "4        179.00000   21059.000000   20306.0    ...      0.00000    23.0   \n",
       "5      10976.00000  278941.000000   74030.0    ...      0.00000    21.0   \n",
       "6       3568.00000   69482.000000   56755.0    ...      0.00000    14.0   \n",
       "7      17115.00000  124814.000000  124167.0    ...      0.00000    15.0   \n",
       "8      19264.00000   59956.000000   26633.0    ...      0.00000     5.0   \n",
       "9        582.00000   16429.000000    9197.0    ...      0.05882    14.0   \n",
       "10      8673.00000    6739.000000    7388.0    ...      0.23529     1.0   \n",
       "11       261.00000     297.000000     261.0    ...      0.33333     2.0   \n",
       "12      7620.00000   64983.000000   14067.0    ...      0.05882    16.0   \n",
       "13      6773.68666     198.000000      18.0    ...      0.87500     1.0   \n",
       "14      2973.00000   22099.000000   13507.0    ...      0.05882    13.0   \n",
       "15      6773.68666   34097.436013       0.0    ...      1.50000     0.0   \n",
       "16     13874.00000   98011.000000   56200.0    ...      0.05000    13.0   \n",
       "17      3964.00000    7928.000000   26564.0    ...      0.00000     7.0   \n",
       "18      2478.00000    8790.000000    6065.0    ...      0.10000     3.0   \n",
       "19     10905.00000   51037.000000   58544.0    ...      0.00000     9.0   \n",
       "20     12525.00000   18932.000000   14602.0    ...      0.00000     7.0   \n",
       "21        57.00000     396.000000    2582.0    ...      0.37500     5.0   \n",
       "22     24739.00000  128435.000000  125679.0    ...      0.00000     9.0   \n",
       "23      2323.00000    9018.000000    9649.0    ...      0.10000     9.0   \n",
       "24        46.00000     892.000000    1235.0    ...      0.20000     5.0   \n",
       "25     14819.00000  116154.000000  100684.0    ...      0.00000    12.0   \n",
       "26      6201.00000   66477.000000   34053.0    ...      0.03846    23.0   \n",
       "27      9815.00000  110001.000000   69683.0    ...      0.00000    15.0   \n",
       "28         0.00000    6566.000000    4068.0    ...      0.57692     8.0   \n",
       "29      1189.00000     892.000000    2065.0    ...      0.00000     2.0   \n",
       "...            ...            ...       ...    ...          ...     ...   \n",
       "24970   1524.00000   16748.000000   25359.0    ...      0.00000    10.0   \n",
       "24971   7505.00000   44843.000000   47453.0    ...      0.00000     4.0   \n",
       "24972    720.00000    5896.000000    5206.0    ...      0.36842    12.0   \n",
       "24973   6773.68666   34097.436013    6290.0    ...      0.83333     0.0   \n",
       "24974    297.00000    1041.000000    6960.0    ...      0.59459    11.0   \n",
       "24975  15162.00000   68181.000000   43334.0    ...      0.00000    15.0   \n",
       "24976   2234.00000    2676.000000    6539.0    ...      0.66667     5.0   \n",
       "24977  17838.00000    8424.000000   22307.0    ...      0.15385     8.0   \n",
       "24978     98.00000     546.000000     643.0    ...      0.66667     4.0   \n",
       "24979   6773.68666    3032.000000   10489.0    ...      0.50000     4.0   \n",
       "24980  22656.00000  168999.000000  166720.0    ...      0.00000    12.0   \n",
       "24981  13297.00000    3914.000000    1305.0    ...      0.20000     2.0   \n",
       "24982   6773.68666    3330.000000       0.0    ...      0.70370     2.0   \n",
       "24983  23707.00000   33432.000000   19640.0    ...      0.04000    15.0   \n",
       "24984  24775.00000   80772.000000   69483.0    ...      0.00000     6.0   \n",
       "24985   2627.00000   46903.000000   30402.0    ...      0.03448    14.0   \n",
       "24986   2251.00000    6818.000000    7783.0    ...      0.71429     2.0   \n",
       "24987  28479.00000   16352.000000    3117.0    ...      0.00000     6.0   \n",
       "24988    100.00000     991.000000     118.0    ...      0.00000     3.0   \n",
       "24989   6773.68666   34097.436013       0.0    ...      1.00000     1.0   \n",
       "24990   6773.68666   34097.436013       0.0    ...      1.00000     1.0   \n",
       "24991   9449.00000   42047.000000   31583.0    ...      0.20000     4.0   \n",
       "24992      4.00000    1982.000000    3167.0    ...      0.61538     3.0   \n",
       "24993    167.00000   15071.000000   10305.0    ...      0.23810    13.0   \n",
       "24994    496.00000    4410.000000    8125.0    ...      0.30769     4.0   \n",
       "24995   6773.68666    8751.000000    1814.0    ...      0.07692     4.0   \n",
       "24996   9613.00000   46490.000000   44722.0    ...      0.05882     8.0   \n",
       "24997   6773.68666    4582.000000       0.0    ...      0.30000     2.0   \n",
       "24998   9058.00000   72739.000000   86797.0    ...      0.03333    11.0   \n",
       "24999  18889.00000   28343.000000   51174.0    ...      0.00000     5.0   \n",
       "\n",
       "         mvar44    mvar45    mvar46  mvar47      var345       mvar678  \\\n",
       "0      0.856610  0.070736  0.000000       0    1.000000     93.647343   \n",
       "1      0.943910  0.000000  0.000000       1    1.000000    133.020134   \n",
       "2      0.764670  0.000000  0.000000       1    1.000000     17.373737   \n",
       "3      0.705285  0.070736  0.096135       1    6.299000   2583.839092   \n",
       "4      0.791900  0.000000  0.000000       0    1.000000     43.088889   \n",
       "5      0.924620  0.000000  0.000000       0    1.241000     28.329558   \n",
       "6      0.713470  0.000000  0.000000       0    1.272000     49.520101   \n",
       "7      0.924030  0.000000  0.000000       0    1.000000    124.816901   \n",
       "8      0.897600  0.000000  0.096135       0    1.000000      5.227680   \n",
       "9      0.405290  0.070736  0.000000       1    1.000000   2685.000000   \n",
       "10     0.642420  0.000000  0.096135       1    1.000000      7.126541   \n",
       "11     0.991420  0.070736  0.000000       1    1.000000     16.458015   \n",
       "12     0.441670  0.070736  0.000000       1    1.000000  15240.000000   \n",
       "13     0.845920  1.000000  1.000000       1  170.380829   2583.839092   \n",
       "14     0.433650  0.000000  0.000000       1    1.000000     24.353846   \n",
       "15     0.705285  0.070736  0.096135       1    2.338000   2583.839092   \n",
       "16     0.512310  0.070736  0.096135       0    2.428000     11.193223   \n",
       "17     0.566850  0.000000  0.000000       1    1.000000     10.936828   \n",
       "18     0.705285  0.070736  0.096135       1    2.514000      4.995968   \n",
       "19     0.594090  0.000000  0.000000       1    1.000000    305.814433   \n",
       "20     0.927520  0.000000  0.000000       1    1.000000    130.468750   \n",
       "21     0.348360  0.070736  0.000000       1    1.000000     54.000000   \n",
       "22     0.885540  0.000000  0.000000       1    1.000000     13.222129   \n",
       "23     0.594420  0.070736  0.000000       1    3.295000     37.150000   \n",
       "24     0.948260  0.070736  0.000000       0    1.000000   1083.000000   \n",
       "25     0.915500  0.000000  0.000000       0    1.000000      6.706961   \n",
       "26     0.151590  0.070736  0.096135       1    1.000000     33.518919   \n",
       "27     0.566710  0.000000  0.000000       1    1.988100    125.946309   \n",
       "28     0.353390  0.070736  2.000000       1  433.420111    396.000000   \n",
       "29     1.000000  0.070736  0.096135       1    1.000000      3.502209   \n",
       "...         ...       ...       ...     ...         ...           ...   \n",
       "24970  0.380780  0.070736  0.000000       0    1.520000     51.233463   \n",
       "24971  0.921110  0.000000  0.000000       0    1.000000     18.108348   \n",
       "24972  0.529030  0.070736  0.000000       1   13.890529    316.555556   \n",
       "24973  1.000000  0.070736  0.000000       1    9.626224   2583.839092   \n",
       "24974  0.426800  0.070736  2.000000       1   35.508729     28.285235   \n",
       "24975  0.271430  0.070736  0.000000       1    1.000000     15.291982   \n",
       "24976  0.878900  0.000000  0.000000       1   16.181000      2.231767   \n",
       "24977  0.696940  0.000000  0.000000       0    1.000000     97.742466   \n",
       "24978  1.000000  0.070736  0.000000       1    1.000000      3.989899   \n",
       "24979  0.726400  0.070736  0.000000       1   65.043214   2583.839092   \n",
       "24980  0.741060  0.000000  0.096135       0    1.000000    150.595442   \n",
       "24981  0.415820  0.070736  0.000000       1    1.000000    131.575630   \n",
       "24982  1.000000  1.000000  1.000000       1  272.975909   2583.839092   \n",
       "24983  0.647520  0.070736  0.000000       0    1.968000   3160.933333   \n",
       "24984  0.970680  0.000000  0.096135       1    1.000000  49550.000000   \n",
       "24985  0.434410  0.070736  0.000000       1    2.979000  12218.000000   \n",
       "24986  0.889080  0.000000  0.000000       1   75.725408   5255.000000   \n",
       "24987  0.723440  0.070736  0.000000       1    1.285000    228.746988   \n",
       "24988  0.366190  0.070736  0.096135       1    1.000000     10.526316   \n",
       "24989  0.506490  0.070736  0.096135       1   13.775000   2583.839092   \n",
       "24990  1.000000  0.070736  0.096135       0    9.614000   2583.839092   \n",
       "24991  0.889570  1.000000  0.000000       0    5.322000     79.244399   \n",
       "24992  0.937450  0.070736  0.000000       0    4.461718    633.400000   \n",
       "24993  0.718660  0.070736  0.000000       1    1.218000   5842.000000   \n",
       "24994  0.933120  0.070736  0.000000       1    6.554000     58.933333   \n",
       "24995  0.705285  0.070736  0.000000       1    1.000000   2583.839092   \n",
       "24996  0.998590  0.000000  0.000000       0    1.000000      4.116361   \n",
       "24997  0.775190  0.070736  0.096135       0    9.150625   2583.839092   \n",
       "24998  0.403430  0.070736  0.000000       1    1.428000      9.900532   \n",
       "24999  0.996630  0.000000  0.000000       1    1.000000      1.999894   \n",
       "\n",
       "          var37  var363738  \n",
       "0      0.090400  16.500000  \n",
       "1      0.090400   9.000000  \n",
       "2      0.159852   6.000000  \n",
       "3      0.381028   0.000000  \n",
       "4      0.381028   0.000000  \n",
       "5      0.090400  53.200000  \n",
       "6      0.102641  10.285714  \n",
       "7      0.090400   7.000000  \n",
       "8      0.186923   3.000000  \n",
       "9      0.309995   0.600000  \n",
       "10     0.239528   0.666667  \n",
       "11     0.381028   0.000000  \n",
       "12     0.090400  15.000000  \n",
       "13     0.381028   0.000000  \n",
       "14     0.239528   3.000000  \n",
       "15     0.381028   0.000000  \n",
       "16     0.090400   6.600000  \n",
       "17     0.309995   0.428571  \n",
       "18     0.381028   7.256161  \n",
       "19     0.131449  10.500000  \n",
       "20     0.381028   0.000000  \n",
       "21     0.309995   0.428571  \n",
       "22     0.090400  16.666667  \n",
       "23     0.309995   0.555556  \n",
       "24     0.381028   0.000000  \n",
       "25     0.118592  15.750000  \n",
       "26     0.144850   7.307692  \n",
       "27     0.102641  18.000000  \n",
       "28     0.186923   1.400000  \n",
       "29     0.381028   0.000000  \n",
       "...         ...        ...  \n",
       "24970  0.381028   0.000000  \n",
       "24971  0.159852   5.333333  \n",
       "24972  0.309995   1.285714  \n",
       "24973  0.458344   7.256161  \n",
       "24974  0.381028   0.000000  \n",
       "24975  0.102641  21.000000  \n",
       "24976  0.186923   4.200000  \n",
       "24977  0.144850   5.000000  \n",
       "24978  0.381028   0.000000  \n",
       "24979  0.381028   0.000000  \n",
       "24980  0.090400  40.000000  \n",
       "24981  0.239528   0.666667  \n",
       "24982  0.309995   1.888889  \n",
       "24983  0.090400  11.000000  \n",
       "24984  0.159852  16.000000  \n",
       "24985  0.090400  28.166667  \n",
       "24986  0.159852  16.000000  \n",
       "24987  0.309995   0.500000  \n",
       "24988  0.381028   0.000000  \n",
       "24989  0.458344   7.256161  \n",
       "24990  0.458344   7.256161  \n",
       "24991  0.102641   6.750000  \n",
       "24992  0.381028   0.000000  \n",
       "24993  0.309995   1.000000  \n",
       "24994  0.239528   1.333333  \n",
       "24995  0.309995   7.256161  \n",
       "24996  0.309995   0.400000  \n",
       "24997  0.144850  20.000000  \n",
       "24998  0.102641  21.000000  \n",
       "24999  0.159852   8.000000  \n",
       "\n",
       "[25000 rows x 51 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Missing values median\n",
    "gameX.fillna(gameX.mean(),inplace = True)\n",
    "leaderX.fillna(gameX.mean(),inplace = True)\n",
    "\n",
    "# gameX.fillna(0,inplace = True)\n",
    "# leaderX.fillna(0,inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#BREAK OPEN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Method 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#current"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "running cross validation\n",
      "running cross validation, with preprocessing function\n",
      "running cross validation, with cutomsized loss function\n",
      "[0]\ttrain-error:0.753787\ttest-error:0.753787\n",
      "Multiple eval metrics have been passed: 'test-error' will be used for early stopping.\n",
      "\n",
      "Will train until test-error hasn't improved in 50 rounds.\n",
      "[1]\ttrain-error:0.753787\ttest-error:0.753787\n",
      "[2]\ttrain-error:0.36955\ttest-error:0.371713\n",
      "[3]\ttrain-error:0.284816\ttest-error:0.286575\n",
      "[4]\ttrain-error:0.245163\ttest-error:0.247437\n",
      "[5]\ttrain-error:0.212703\ttest-error:0.215738\n",
      "[6]\ttrain-error:0.207472\ttest-error:0.21025\n",
      "[7]\ttrain-error:0.204878\ttest-error:0.207387\n",
      "[8]\ttrain-error:0.203591\ttest-error:0.205487\n",
      "[9]\ttrain-error:0.202441\ttest-error:0.204887\n",
      "[10]\ttrain-error:0.201412\ttest-error:0.203575\n",
      "[11]\ttrain-error:0.200553\ttest-error:0.2029\n",
      "[12]\ttrain-error:0.200172\ttest-error:0.202337\n",
      "[13]\ttrain-error:0.200022\ttest-error:0.201687\n",
      "[14]\ttrain-error:0.199413\ttest-error:0.201137\n",
      "[15]\ttrain-error:0.199253\ttest-error:0.201187\n",
      "[16]\ttrain-error:0.198778\ttest-error:0.200388\n",
      "[17]\ttrain-error:0.198515\ttest-error:0.200462\n",
      "[18]\ttrain-error:0.198319\ttest-error:0.2005\n",
      "[19]\ttrain-error:0.197688\ttest-error:0.200187\n",
      "[20]\ttrain-error:0.197438\ttest-error:0.199825\n",
      "[21]\ttrain-error:0.197147\ttest-error:0.199775\n",
      "[22]\ttrain-error:0.196863\ttest-error:0.199413\n",
      "[23]\ttrain-error:0.196853\ttest-error:0.199225\n",
      "[24]\ttrain-error:0.196344\ttest-error:0.198788\n",
      "[25]\ttrain-error:0.195819\ttest-error:0.198837\n",
      "[26]\ttrain-error:0.195425\ttest-error:0.1983\n",
      "[27]\ttrain-error:0.195206\ttest-error:0.198525\n",
      "[28]\ttrain-error:0.194906\ttest-error:0.198187\n",
      "[29]\ttrain-error:0.194647\ttest-error:0.198175\n",
      "[30]\ttrain-error:0.194316\ttest-error:0.19785\n",
      "[31]\ttrain-error:0.194041\ttest-error:0.1978\n",
      "[32]\ttrain-error:0.193688\ttest-error:0.197612\n",
      "[33]\ttrain-error:0.193378\ttest-error:0.197475\n",
      "[34]\ttrain-error:0.193112\ttest-error:0.197288\n",
      "[35]\ttrain-error:0.19285\ttest-error:0.197012\n",
      "[36]\ttrain-error:0.192606\ttest-error:0.196662\n",
      "[37]\ttrain-error:0.192322\ttest-error:0.196538\n",
      "[38]\ttrain-error:0.192019\ttest-error:0.196388\n",
      "[39]\ttrain-error:0.191709\ttest-error:0.195988\n",
      "[40]\ttrain-error:0.191391\ttest-error:0.195862\n",
      "[41]\ttrain-error:0.191275\ttest-error:0.19595\n",
      "[42]\ttrain-error:0.190988\ttest-error:0.195825\n",
      "[43]\ttrain-error:0.190691\ttest-error:0.195863\n",
      "[44]\ttrain-error:0.190369\ttest-error:0.195663\n",
      "[45]\ttrain-error:0.190266\ttest-error:0.19505\n",
      "[46]\ttrain-error:0.189866\ttest-error:0.194925\n",
      "[47]\ttrain-error:0.189625\ttest-error:0.1947\n",
      "[48]\ttrain-error:0.189463\ttest-error:0.194938\n",
      "[49]\ttrain-error:0.1892\ttest-error:0.19465\n",
      "[50]\ttrain-error:0.188944\ttest-error:0.194512\n",
      "[51]\ttrain-error:0.188787\ttest-error:0.19455\n",
      "[52]\ttrain-error:0.188581\ttest-error:0.194538\n",
      "[53]\ttrain-error:0.188428\ttest-error:0.194462\n",
      "[54]\ttrain-error:0.188128\ttest-error:0.194188\n",
      "[55]\ttrain-error:0.187922\ttest-error:0.1942\n",
      "[56]\ttrain-error:0.1878\ttest-error:0.194\n",
      "[57]\ttrain-error:0.187612\ttest-error:0.193962\n",
      "[58]\ttrain-error:0.187447\ttest-error:0.193912\n",
      "[59]\ttrain-error:0.187297\ttest-error:0.193925\n",
      "[60]\ttrain-error:0.187066\ttest-error:0.194037\n",
      "[61]\ttrain-error:0.186881\ttest-error:0.193737\n",
      "[62]\ttrain-error:0.186863\ttest-error:0.19375\n",
      "[63]\ttrain-error:0.18671\ttest-error:0.193587\n",
      "[64]\ttrain-error:0.186362\ttest-error:0.1935\n",
      "[65]\ttrain-error:0.18614\ttest-error:0.19325\n",
      "[66]\ttrain-error:0.185922\ttest-error:0.19335\n",
      "[67]\ttrain-error:0.185734\ttest-error:0.193125\n",
      "[68]\ttrain-error:0.185575\ttest-error:0.193125\n",
      "[69]\ttrain-error:0.185372\ttest-error:0.19315\n",
      "[70]\ttrain-error:0.185284\ttest-error:0.193213\n",
      "[71]\ttrain-error:0.1852\ttest-error:0.19315\n",
      "[72]\ttrain-error:0.185016\ttest-error:0.193125\n",
      "[73]\ttrain-error:0.184897\ttest-error:0.193038\n",
      "[74]\ttrain-error:0.184619\ttest-error:0.192988\n",
      "[75]\ttrain-error:0.184572\ttest-error:0.192813\n",
      "[76]\ttrain-error:0.184466\ttest-error:0.192588\n",
      "[77]\ttrain-error:0.184359\ttest-error:0.192513\n",
      "[78]\ttrain-error:0.184128\ttest-error:0.19265\n",
      "[79]\ttrain-error:0.184078\ttest-error:0.192675\n",
      "[80]\ttrain-error:0.183966\ttest-error:0.192775\n",
      "[81]\ttrain-error:0.183878\ttest-error:0.1928\n",
      "[82]\ttrain-error:0.183659\ttest-error:0.19265\n",
      "[83]\ttrain-error:0.183588\ttest-error:0.1925\n",
      "[84]\ttrain-error:0.183472\ttest-error:0.192487\n",
      "[85]\ttrain-error:0.183409\ttest-error:0.192225\n",
      "[86]\ttrain-error:0.183397\ttest-error:0.1922\n",
      "[87]\ttrain-error:0.183262\ttest-error:0.192287\n",
      "[88]\ttrain-error:0.183122\ttest-error:0.19225\n",
      "[89]\ttrain-error:0.1829\ttest-error:0.191988\n",
      "[90]\ttrain-error:0.182722\ttest-error:0.191837\n",
      "[91]\ttrain-error:0.182594\ttest-error:0.191887\n",
      "[92]\ttrain-error:0.182612\ttest-error:0.191725\n",
      "[93]\ttrain-error:0.182456\ttest-error:0.191725\n",
      "[94]\ttrain-error:0.182266\ttest-error:0.191725\n",
      "[95]\ttrain-error:0.182228\ttest-error:0.191687\n",
      "[96]\ttrain-error:0.182222\ttest-error:0.191775\n",
      "[97]\ttrain-error:0.182106\ttest-error:0.19175\n",
      "[98]\ttrain-error:0.181978\ttest-error:0.191787\n",
      "[99]\ttrain-error:0.181934\ttest-error:0.1919\n",
      "[100]\ttrain-error:0.181894\ttest-error:0.191862\n",
      "[101]\ttrain-error:0.181734\ttest-error:0.191962\n",
      "[102]\ttrain-error:0.1816\ttest-error:0.191937\n",
      "[103]\ttrain-error:0.181522\ttest-error:0.191938\n",
      "[104]\ttrain-error:0.181269\ttest-error:0.191888\n",
      "[105]\ttrain-error:0.181159\ttest-error:0.191475\n",
      "[106]\ttrain-error:0.181125\ttest-error:0.191563\n",
      "[107]\ttrain-error:0.181078\ttest-error:0.191738\n",
      "[108]\ttrain-error:0.180972\ttest-error:0.191512\n",
      "[109]\ttrain-error:0.180788\ttest-error:0.191438\n",
      "[110]\ttrain-error:0.180784\ttest-error:0.191313\n",
      "[111]\ttrain-error:0.180606\ttest-error:0.191275\n",
      "[112]\ttrain-error:0.180581\ttest-error:0.19125\n",
      "[113]\ttrain-error:0.180444\ttest-error:0.191325\n",
      "[114]\ttrain-error:0.180397\ttest-error:0.191362\n",
      "[115]\ttrain-error:0.180259\ttest-error:0.191325\n",
      "[116]\ttrain-error:0.180163\ttest-error:0.191175\n",
      "[117]\ttrain-error:0.180053\ttest-error:0.19115\n",
      "[118]\ttrain-error:0.179981\ttest-error:0.191112\n",
      "[119]\ttrain-error:0.179956\ttest-error:0.1913\n",
      "[120]\ttrain-error:0.179912\ttest-error:0.191137\n",
      "[121]\ttrain-error:0.179865\ttest-error:0.191187\n",
      "[122]\ttrain-error:0.179788\ttest-error:0.191262\n",
      "[123]\ttrain-error:0.179719\ttest-error:0.191313\n",
      "[124]\ttrain-error:0.179653\ttest-error:0.191388\n",
      "[125]\ttrain-error:0.179584\ttest-error:0.191288\n",
      "[126]\ttrain-error:0.179547\ttest-error:0.191238\n",
      "[127]\ttrain-error:0.179406\ttest-error:0.191338\n",
      "[128]\ttrain-error:0.179312\ttest-error:0.191325\n",
      "[129]\ttrain-error:0.179169\ttest-error:0.191312\n",
      "[130]\ttrain-error:0.179013\ttest-error:0.191225\n",
      "[131]\ttrain-error:0.178928\ttest-error:0.191175\n",
      "[132]\ttrain-error:0.178897\ttest-error:0.191062\n",
      "[133]\ttrain-error:0.178881\ttest-error:0.191012\n",
      "[134]\ttrain-error:0.178756\ttest-error:0.190962\n",
      "[135]\ttrain-error:0.178769\ttest-error:0.191025\n",
      "[136]\ttrain-error:0.178647\ttest-error:0.191075\n",
      "[137]\ttrain-error:0.178478\ttest-error:0.191113\n",
      "[138]\ttrain-error:0.178428\ttest-error:0.190987\n",
      "[139]\ttrain-error:0.178409\ttest-error:0.190988\n",
      "[140]\ttrain-error:0.1784\ttest-error:0.190975\n",
      "[141]\ttrain-error:0.178256\ttest-error:0.19105\n",
      "[142]\ttrain-error:0.178181\ttest-error:0.190962\n",
      "[143]\ttrain-error:0.178125\ttest-error:0.190887\n",
      "[144]\ttrain-error:0.178047\ttest-error:0.191063\n",
      "[145]\ttrain-error:0.177925\ttest-error:0.191138\n",
      "[146]\ttrain-error:0.177844\ttest-error:0.191038\n",
      "[147]\ttrain-error:0.177644\ttest-error:0.19105\n",
      "[148]\ttrain-error:0.177641\ttest-error:0.190963\n",
      "[149]\ttrain-error:0.177537\ttest-error:0.190963\n",
      "[150]\ttrain-error:0.177535\ttest-error:0.1911\n",
      "[151]\ttrain-error:0.17755\ttest-error:0.191175\n",
      "[152]\ttrain-error:0.177475\ttest-error:0.191112\n",
      "[153]\ttrain-error:0.177372\ttest-error:0.191037\n",
      "[154]\ttrain-error:0.177262\ttest-error:0.190875\n",
      "[155]\ttrain-error:0.177187\ttest-error:0.190925\n",
      "[156]\ttrain-error:0.177147\ttest-error:0.190813\n",
      "[157]\ttrain-error:0.177009\ttest-error:0.19065\n",
      "[158]\ttrain-error:0.176916\ttest-error:0.190638\n",
      "[159]\ttrain-error:0.176828\ttest-error:0.190738\n",
      "[160]\ttrain-error:0.176778\ttest-error:0.190812\n",
      "[161]\ttrain-error:0.176687\ttest-error:0.190837\n",
      "[162]\ttrain-error:0.176528\ttest-error:0.190725\n",
      "[163]\ttrain-error:0.17655\ttest-error:0.1907\n",
      "[164]\ttrain-error:0.176472\ttest-error:0.190663\n",
      "[165]\ttrain-error:0.176325\ttest-error:0.19055\n",
      "[166]\ttrain-error:0.176206\ttest-error:0.190688\n",
      "[167]\ttrain-error:0.176162\ttest-error:0.190812\n",
      "[168]\ttrain-error:0.175988\ttest-error:0.190713\n",
      "[169]\ttrain-error:0.175953\ttest-error:0.190763\n",
      "[170]\ttrain-error:0.175866\ttest-error:0.190775\n",
      "[171]\ttrain-error:0.175831\ttest-error:0.190625\n",
      "[172]\ttrain-error:0.175813\ttest-error:0.1907\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[173]\ttrain-error:0.175697\ttest-error:0.190688\n",
      "[174]\ttrain-error:0.175644\ttest-error:0.190588\n",
      "[175]\ttrain-error:0.175528\ttest-error:0.190725\n",
      "[176]\ttrain-error:0.175387\ttest-error:0.190775\n",
      "[177]\ttrain-error:0.175337\ttest-error:0.190787\n",
      "[178]\ttrain-error:0.175275\ttest-error:0.19095\n",
      "[179]\ttrain-error:0.175141\ttest-error:0.1908\n",
      "[180]\ttrain-error:0.175025\ttest-error:0.190787\n",
      "[181]\ttrain-error:0.174984\ttest-error:0.1909\n",
      "[182]\ttrain-error:0.175041\ttest-error:0.190862\n",
      "[183]\ttrain-error:0.174965\ttest-error:0.19085\n",
      "[184]\ttrain-error:0.174937\ttest-error:0.190712\n",
      "[185]\ttrain-error:0.174797\ttest-error:0.190825\n",
      "[186]\ttrain-error:0.174741\ttest-error:0.190938\n",
      "[187]\ttrain-error:0.174666\ttest-error:0.191\n",
      "[188]\ttrain-error:0.174609\ttest-error:0.190937\n",
      "[189]\ttrain-error:0.174613\ttest-error:0.190887\n",
      "[190]\ttrain-error:0.174525\ttest-error:0.190937\n",
      "[191]\ttrain-error:0.174428\ttest-error:0.191025\n",
      "[192]\ttrain-error:0.174303\ttest-error:0.1911\n",
      "[193]\ttrain-error:0.174294\ttest-error:0.19125\n",
      "[194]\ttrain-error:0.174194\ttest-error:0.191262\n",
      "[195]\ttrain-error:0.174088\ttest-error:0.191088\n",
      "[196]\ttrain-error:0.17395\ttest-error:0.191075\n",
      "[197]\ttrain-error:0.173934\ttest-error:0.191025\n",
      "[198]\ttrain-error:0.173869\ttest-error:0.1911\n",
      "[199]\ttrain-error:0.173747\ttest-error:0.191188\n",
      "[200]\ttrain-error:0.173697\ttest-error:0.190975\n",
      "[201]\ttrain-error:0.173569\ttest-error:0.191087\n",
      "[202]\ttrain-error:0.173425\ttest-error:0.19105\n",
      "[203]\ttrain-error:0.173381\ttest-error:0.191037\n",
      "[204]\ttrain-error:0.173272\ttest-error:0.190963\n",
      "[205]\ttrain-error:0.173244\ttest-error:0.190975\n",
      "[206]\ttrain-error:0.173156\ttest-error:0.190962\n",
      "[207]\ttrain-error:0.173159\ttest-error:0.191013\n",
      "[208]\ttrain-error:0.173025\ttest-error:0.19095\n",
      "[209]\ttrain-error:0.172928\ttest-error:0.190962\n",
      "[210]\ttrain-error:0.172878\ttest-error:0.191125\n",
      "[211]\ttrain-error:0.172819\ttest-error:0.191162\n",
      "[212]\ttrain-error:0.17265\ttest-error:0.191137\n",
      "[213]\ttrain-error:0.172644\ttest-error:0.191063\n",
      "[214]\ttrain-error:0.172625\ttest-error:0.191225\n",
      "[215]\ttrain-error:0.172635\ttest-error:0.19115\n",
      "Stopping. Best iteration:\n",
      "[165]\ttrain-error:0.176325+0.000166989\ttest-error:0.19055+0.00149102\n",
      "\n",
      "[0]\teval-error:0.753788\ttrain-error:0.753788\n",
      "[1]\teval-error:0.753788\ttrain-error:0.753788\n",
      "[2]\teval-error:0.371225\ttrain-error:0.371225\n",
      "[3]\teval-error:0.297088\ttrain-error:0.297088\n",
      "[4]\teval-error:0.224587\ttrain-error:0.224587\n",
      "[5]\teval-error:0.212413\ttrain-error:0.212413\n",
      "[6]\teval-error:0.210037\ttrain-error:0.210037\n",
      "[7]\teval-error:0.206838\ttrain-error:0.206838\n",
      "[8]\teval-error:0.202725\ttrain-error:0.202725\n",
      "[9]\teval-error:0.202487\ttrain-error:0.202487\n",
      "[10]\teval-error:0.201063\ttrain-error:0.201063\n",
      "[11]\teval-error:0.200412\ttrain-error:0.200412\n",
      "[12]\teval-error:0.199687\ttrain-error:0.199687\n",
      "[13]\teval-error:0.199738\ttrain-error:0.199738\n",
      "[14]\teval-error:0.1992\ttrain-error:0.1992\n",
      "[15]\teval-error:0.1979\ttrain-error:0.1979\n",
      "[16]\teval-error:0.198763\ttrain-error:0.198763\n",
      "[17]\teval-error:0.198662\ttrain-error:0.198662\n",
      "[18]\teval-error:0.198362\ttrain-error:0.198362\n",
      "[19]\teval-error:0.197963\ttrain-error:0.197963\n",
      "[20]\teval-error:0.197275\ttrain-error:0.197275\n",
      "[21]\teval-error:0.197162\ttrain-error:0.197162\n",
      "[22]\teval-error:0.196337\ttrain-error:0.196337\n",
      "[23]\teval-error:0.1969\ttrain-error:0.1969\n",
      "[24]\teval-error:0.196637\ttrain-error:0.196637\n",
      "[25]\teval-error:0.19595\ttrain-error:0.19595\n",
      "[26]\teval-error:0.195725\ttrain-error:0.195725\n",
      "[27]\teval-error:0.195775\ttrain-error:0.195775\n",
      "[28]\teval-error:0.195312\ttrain-error:0.195312\n",
      "[29]\teval-error:0.195175\ttrain-error:0.195175\n",
      "[30]\teval-error:0.194637\ttrain-error:0.194637\n",
      "[31]\teval-error:0.194125\ttrain-error:0.194125\n",
      "[32]\teval-error:0.1941\ttrain-error:0.1941\n",
      "[33]\teval-error:0.193862\ttrain-error:0.193862\n",
      "[34]\teval-error:0.193775\ttrain-error:0.193775\n",
      "[35]\teval-error:0.19335\ttrain-error:0.19335\n",
      "[36]\teval-error:0.193425\ttrain-error:0.193425\n",
      "[37]\teval-error:0.192925\ttrain-error:0.192925\n",
      "[38]\teval-error:0.192525\ttrain-error:0.192525\n",
      "[39]\teval-error:0.192225\ttrain-error:0.192225\n",
      "[40]\teval-error:0.191725\ttrain-error:0.191725\n",
      "[41]\teval-error:0.191337\ttrain-error:0.191337\n",
      "[42]\teval-error:0.191075\ttrain-error:0.191075\n",
      "[43]\teval-error:0.191\ttrain-error:0.191\n",
      "[44]\teval-error:0.19085\ttrain-error:0.19085\n",
      "[45]\teval-error:0.191025\ttrain-error:0.191025\n",
      "[46]\teval-error:0.190837\ttrain-error:0.190837\n",
      "[47]\teval-error:0.19015\ttrain-error:0.19015\n",
      "[48]\teval-error:0.1899\ttrain-error:0.1899\n",
      "[49]\teval-error:0.189525\ttrain-error:0.189525\n",
      "[50]\teval-error:0.189438\ttrain-error:0.189438\n",
      "[51]\teval-error:0.189125\ttrain-error:0.189125\n",
      "[52]\teval-error:0.188925\ttrain-error:0.188925\n",
      "[53]\teval-error:0.18885\ttrain-error:0.18885\n",
      "[54]\teval-error:0.188512\ttrain-error:0.188512\n",
      "[55]\teval-error:0.188112\ttrain-error:0.188112\n",
      "[56]\teval-error:0.187837\ttrain-error:0.187837\n",
      "[57]\teval-error:0.187663\ttrain-error:0.187663\n",
      "[58]\teval-error:0.187575\ttrain-error:0.187575\n",
      "[59]\teval-error:0.18735\ttrain-error:0.18735\n",
      "[60]\teval-error:0.1872\ttrain-error:0.1872\n",
      "[61]\teval-error:0.1869\ttrain-error:0.1869\n",
      "[62]\teval-error:0.186675\ttrain-error:0.186675\n",
      "[63]\teval-error:0.18665\ttrain-error:0.18665\n",
      "[64]\teval-error:0.186837\ttrain-error:0.186837\n",
      "[65]\teval-error:0.186738\ttrain-error:0.186738\n",
      "[66]\teval-error:0.186413\ttrain-error:0.186413\n",
      "[67]\teval-error:0.186238\ttrain-error:0.186238\n",
      "[68]\teval-error:0.1862\ttrain-error:0.1862\n",
      "[69]\teval-error:0.18615\ttrain-error:0.18615\n",
      "[70]\teval-error:0.186112\ttrain-error:0.186112\n",
      "[71]\teval-error:0.186125\ttrain-error:0.186125\n",
      "[72]\teval-error:0.186087\ttrain-error:0.186087\n",
      "[73]\teval-error:0.18585\ttrain-error:0.18585\n",
      "[74]\teval-error:0.185725\ttrain-error:0.185725\n",
      "[75]\teval-error:0.18565\ttrain-error:0.18565\n",
      "[76]\teval-error:0.185537\ttrain-error:0.185537\n",
      "[77]\teval-error:0.185425\ttrain-error:0.185425\n",
      "[78]\teval-error:0.1853\ttrain-error:0.1853\n",
      "[79]\teval-error:0.185375\ttrain-error:0.185375\n",
      "[80]\teval-error:0.185025\ttrain-error:0.185025\n",
      "[81]\teval-error:0.185087\ttrain-error:0.185087\n",
      "[82]\teval-error:0.184888\ttrain-error:0.184888\n",
      "[83]\teval-error:0.184762\ttrain-error:0.184762\n",
      "[84]\teval-error:0.184537\ttrain-error:0.184537\n",
      "[85]\teval-error:0.184337\ttrain-error:0.184337\n",
      "[86]\teval-error:0.1841\ttrain-error:0.1841\n",
      "[87]\teval-error:0.184138\ttrain-error:0.184138\n",
      "[88]\teval-error:0.183987\ttrain-error:0.183987\n",
      "[89]\teval-error:0.184062\ttrain-error:0.184062\n",
      "[90]\teval-error:0.184\ttrain-error:0.184\n",
      "[91]\teval-error:0.18385\ttrain-error:0.18385\n",
      "[92]\teval-error:0.1839\ttrain-error:0.1839\n",
      "[93]\teval-error:0.184062\ttrain-error:0.184062\n",
      "[94]\teval-error:0.1839\ttrain-error:0.1839\n",
      "[95]\teval-error:0.183562\ttrain-error:0.183562\n",
      "[96]\teval-error:0.183375\ttrain-error:0.183375\n",
      "[97]\teval-error:0.18335\ttrain-error:0.18335\n",
      "[98]\teval-error:0.183388\ttrain-error:0.183388\n",
      "[99]\teval-error:0.183025\ttrain-error:0.183025\n",
      "[100]\teval-error:0.183163\ttrain-error:0.183163\n",
      "[101]\teval-error:0.182875\ttrain-error:0.182875\n",
      "[102]\teval-error:0.182938\ttrain-error:0.182938\n",
      "[103]\teval-error:0.1829\ttrain-error:0.1829\n",
      "[104]\teval-error:0.182863\ttrain-error:0.182863\n",
      "[105]\teval-error:0.182825\ttrain-error:0.182825\n",
      "[106]\teval-error:0.182762\ttrain-error:0.182762\n",
      "[107]\teval-error:0.1828\ttrain-error:0.1828\n",
      "[108]\teval-error:0.182875\ttrain-error:0.182875\n",
      "[109]\teval-error:0.1828\ttrain-error:0.1828\n",
      "[110]\teval-error:0.182525\ttrain-error:0.182525\n",
      "[111]\teval-error:0.182588\ttrain-error:0.182588\n",
      "[112]\teval-error:0.18235\ttrain-error:0.18235\n",
      "[113]\teval-error:0.182312\ttrain-error:0.182312\n",
      "[114]\teval-error:0.1822\ttrain-error:0.1822\n",
      "[115]\teval-error:0.182188\ttrain-error:0.182188\n",
      "[116]\teval-error:0.182125\ttrain-error:0.182125\n",
      "[117]\teval-error:0.182225\ttrain-error:0.182225\n",
      "[118]\teval-error:0.182287\ttrain-error:0.182287\n",
      "[119]\teval-error:0.1822\ttrain-error:0.1822\n",
      "[120]\teval-error:0.181987\ttrain-error:0.181987\n",
      "[121]\teval-error:0.181775\ttrain-error:0.181775\n",
      "[122]\teval-error:0.181775\ttrain-error:0.181775\n",
      "[123]\teval-error:0.181838\ttrain-error:0.181838\n",
      "[124]\teval-error:0.181725\ttrain-error:0.181725\n",
      "[125]\teval-error:0.1815\ttrain-error:0.1815\n",
      "[126]\teval-error:0.181388\ttrain-error:0.181388\n",
      "[127]\teval-error:0.1813\ttrain-error:0.1813\n",
      "[128]\teval-error:0.1812\ttrain-error:0.1812\n",
      "[129]\teval-error:0.181287\ttrain-error:0.181287\n",
      "[130]\teval-error:0.181312\ttrain-error:0.181312\n",
      "[131]\teval-error:0.181262\ttrain-error:0.181262\n",
      "[132]\teval-error:0.18095\ttrain-error:0.18095\n",
      "[133]\teval-error:0.1808\ttrain-error:0.1808\n",
      "[134]\teval-error:0.180875\ttrain-error:0.180875\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[135]\teval-error:0.18075\ttrain-error:0.18075\n",
      "[136]\teval-error:0.180613\ttrain-error:0.180613\n",
      "[137]\teval-error:0.180537\ttrain-error:0.180537\n",
      "[138]\teval-error:0.180663\ttrain-error:0.180663\n",
      "[139]\teval-error:0.180537\ttrain-error:0.180537\n",
      "[140]\teval-error:0.18045\ttrain-error:0.18045\n",
      "[141]\teval-error:0.180537\ttrain-error:0.180537\n",
      "[142]\teval-error:0.180512\ttrain-error:0.180512\n",
      "[143]\teval-error:0.18045\ttrain-error:0.18045\n",
      "[144]\teval-error:0.180537\ttrain-error:0.180537\n",
      "[145]\teval-error:0.1804\ttrain-error:0.1804\n",
      "[146]\teval-error:0.18035\ttrain-error:0.18035\n",
      "[147]\teval-error:0.180388\ttrain-error:0.180388\n",
      "[148]\teval-error:0.1804\ttrain-error:0.1804\n",
      "[149]\teval-error:0.18025\ttrain-error:0.18025\n",
      "[150]\teval-error:0.18015\ttrain-error:0.18015\n",
      "[151]\teval-error:0.180025\ttrain-error:0.180025\n",
      "[152]\teval-error:0.179838\ttrain-error:0.179838\n",
      "[153]\teval-error:0.179762\ttrain-error:0.179762\n",
      "[154]\teval-error:0.179613\ttrain-error:0.179613\n",
      "[155]\teval-error:0.17965\ttrain-error:0.17965\n",
      "[156]\teval-error:0.179675\ttrain-error:0.179675\n",
      "[157]\teval-error:0.179512\ttrain-error:0.179512\n",
      "[158]\teval-error:0.1794\ttrain-error:0.1794\n",
      "[159]\teval-error:0.179338\ttrain-error:0.179338\n",
      "[160]\teval-error:0.179237\ttrain-error:0.179237\n",
      "[161]\teval-error:0.179163\ttrain-error:0.179163\n",
      "[162]\teval-error:0.178975\ttrain-error:0.178975\n",
      "[163]\teval-error:0.178925\ttrain-error:0.178925\n",
      "[164]\teval-error:0.178875\ttrain-error:0.178875\n",
      "[165]\teval-error:0.178875\ttrain-error:0.178875\n",
      "time taken: 405.32858872413635\n"
     ]
    }
   ],
   "source": [
    "import xgboost as xgb\n",
    "import time\n",
    "\n",
    "\n",
    "dtrain = xgb.DMatrix(gameX, gameY)\n",
    "num_round = 1000\n",
    "early_stopping_rounds = 50\n",
    "n_fold = 5\n",
    "watchlist = [(dtrain, 'eval'), (dtrain, 'train')]\n",
    "\n",
    "print('running cross validation')\n",
    "\n",
    "print('running cross validation, with preprocessing function')\n",
    "# define the preprocessing function\n",
    "# used to return the preprocessed training, test data, and parameter\n",
    "# we can use this to do weight rescale, etc.\n",
    "# as a example, we try to set scale_pos_weight\n",
    "def fpreproc(dtrain, dtest, param):\n",
    "    label = dtrain.get_label()\n",
    "    ratio = float(np.sum(label == 0)) / np.sum(label == 1)\n",
    "    param['scale_pos_weight'] = ratio\n",
    "    return (dtrain, dtest, param)\n",
    "\n",
    "# do cross validation, for each fold\n",
    "# the dtrain, dtest, param will be passed into fpreproc\n",
    "# then the return value of fpreproc will be used to generate\n",
    "# results of that fold\n",
    "\n",
    "###\n",
    "# you can also do cross validation with customized loss function\n",
    "# See custom_objective.py\n",
    "##\n",
    "print('running cross validation, with cutomsized loss function')\n",
    "#Custom\n",
    "# wt = 0.5\n",
    "# def logregobj(preds, dtrain):\n",
    "#     labels = dtrain.get_label()\n",
    "#     preds = 1.0 / (1.0 + np.exp(-preds))\n",
    "#     grad = wt*(1.0-labels)*preds - (1-wt)*labels*(1-preds)\n",
    "#     hess = preds * (1.0 - preds) * (wt*(1.0-labels)+(1.0-wt)*labels)\n",
    "#     return grad, hess\n",
    "\n",
    "def logregobj(preds, dtrain):\n",
    "    labels = dtrain.get_label()\n",
    "    preds = 1.0 / (1.0 + np.exp(-preds))\n",
    "    grad = preds - labels\n",
    "    hess = preds * (1.0 - preds)\n",
    "    return grad, hess\n",
    "def evalerror(preds, dtrain):\n",
    "    labels = dtrain.get_label()\n",
    "    return 'error', float(sum(labels != (preds > 0.0))) / len(labels)\n",
    "\n",
    "param = {\n",
    "#     'max_depth': 3,#3 \n",
    "    'eta': 0.1, #0.1\n",
    "    'silent': 1, \n",
    "    'gamma': 0, \n",
    "    'max_depth': 4, #4\n",
    "    'min_child_weight': 1, #1\n",
    "    'subsample': 0.8, #0.8\n",
    "    'colsample_bytree': 0.8, #0,8\n",
    "    'reg_lambda': 1, #L2 reg default:1\n",
    "    'reg_alpha': 0.05, #L1 reg default:0\n",
    "    'scale_pos_weight': 1, \n",
    "#     'missing': -999\n",
    "#     'objective': 'binary:logistic', #for custom loss define it in xgb.train(obj=custom_loss)\n",
    "#     'eval_metric': 'error'\n",
    "}\n",
    "\n",
    "t = time.time()\n",
    "# CV to find optimum number of boosted trees (num_round)\n",
    "cvresult = xgb.cv(\n",
    "            param, #param set by user\n",
    "            dtrain, #train set\n",
    "            num_round, #max number of boosted tree, >early_stopping_round\n",
    "            nfold=n_fold, #n fold CV\n",
    "            seed=0, \n",
    "            obj=logregobj, #custom loss func\n",
    "            feval=evalerror, #custom eval metric\n",
    "            callbacks=[xgb.callback.print_evaluation(show_stdv=False),\n",
    "                       xgb.callback.early_stop(early_stopping_rounds)], #decides optimum number of boosted trees, <num_round\n",
    "#             fpreproc=fpreproc, #function for preprocessing\n",
    "            )\n",
    "\n",
    "num_round = cvresult.shape[0] #optimum number of boosted trees\n",
    "# bst = xgb.train(param, dtrain, num_round, watchlist, obj=logregobj, feval=evalerror, fpreproc=fpreproc)\n",
    "bst = xgb.train(param, dtrain, num_round, watchlist, obj=logregobj, feval=evalerror)\n",
    "\n",
    "print(\"time taken:\", time.time()-t)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ekka/.local/lib/python3.6/site-packages/sklearn/cross_validation.py:44: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "#910k\n",
    "import xgboost as xgb\n",
    "import time\n",
    "\n",
    "\n",
    "dtrain = xgb.DMatrix(gameX, gameY)\n",
    "num_round = 1000\n",
    "early_stopping_rounds = 50\n",
    "n_fold = 5\n",
    "watchlist = [(dtrain, 'eval'), (dtrain, 'train')]\n",
    "\n",
    "\n",
    "def logregobj(preds, dtrain):\n",
    "    labels = dtrain.get_label()\n",
    "    preds = 1.0 / (1.0 + np.exp(-preds))\n",
    "    grad = preds - labels\n",
    "    hess = preds * (1.0 - preds)\n",
    "    return grad, hess\n",
    "def evalerror(preds, dtrain):\n",
    "    labels = dtrain.get_label()\n",
    "    return 'error', float(sum(labels != (preds > 0.0))) / len(labels)\n",
    "\n",
    "\n",
    "param = {\n",
    "#     'max_depth': 3,#3 \n",
    "    'eta': 0.1, #0.1\n",
    "    'silent': 1, \n",
    "    'gamma': 0, \n",
    "    'max_depth': 4, #4\n",
    "    'min_child_weight': 1.5, #1\n",
    "    'subsample': 0.8, #0.8\n",
    "    'colsample_bytree': 0.8, #0,8\n",
    "    'reg_lambda': 1, #L2 reg default:1\n",
    "    'reg_alpha': 0.05, #L1 reg default:0\n",
    "    'scale_pos_weight': 0.25, \n",
    "#     'missing': -999\n",
    "#     'objective': 'binary:logistic', #for custom loss define it in xgb.train(obj=custom_loss)\n",
    "#     'eval_metric': 'error'\n",
    "}\n",
    "\n",
    "num_round = 231\n",
    "\n",
    "bst = xgb.train(param, dtrain, num_round, obj=logregobj, feval=evalerror)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAosAAAJcCAYAAACR0e6AAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nqZPNoiRJkjrZLEqSJKnTfwFHs0/uYo//rwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fc4dd81c0b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#xgb feature imp\n",
    "from xgboost import plot_importance\n",
    "ax = plot_importance(bst, height=0.5)\n",
    "fig = ax.figure\n",
    "fig.set_size_inches(10,10)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnsAAAJcCAYAAACBnWIZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nF0lEQVRjGuyJiIiI5DEN9kRERETy2P8HZzvFaJhMld0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f30864a2d68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#xgb feature imp\n",
    "from xgboost import plot_importance\n",
    "ax = plot_importance(bst, height=0.5)\n",
    "fig = ax.figure\n",
    "fig.set_size_inches(10,10)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "166"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cvresult.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-0.98256779, -2.59675956, -1.10583878, ...,  0.86568141,\n",
       "       -1.33436608, -0.63021302], dtype=float32)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bst.predict(dtrain)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0.27238256,  0.06934726,  0.24864748, ...,  0.70384634,\n",
       "        0.20843808,  0.34746221], dtype=float32)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1/(1+np.exp(-bst.predict(dtrain)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "dtest = xgb.DMatrix(leaderX)\n",
    "# pred = bst.predict(leaderXxgb)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "pred = 1/(1+np.exp(-bst.predict(dtest)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(25000, 2)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred = pd.DataFrame(pred)\n",
    "pred['application_key'] = leaderAppID\n",
    "pred.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>application_key</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.055446</td>\n",
       "      <td>350053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.042235</td>\n",
       "      <td>350054</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.103503</td>\n",
       "      <td>350055</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.670476</td>\n",
       "      <td>350056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.102000</td>\n",
       "      <td>350057</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.011206</td>\n",
       "      <td>350058</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.051739</td>\n",
       "      <td>350059</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.003864</td>\n",
       "      <td>350060</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.005794</td>\n",
       "      <td>350061</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.229367</td>\n",
       "      <td>350062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.061044</td>\n",
       "      <td>350063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.222500</td>\n",
       "      <td>350064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.482598</td>\n",
       "      <td>350065</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.853469</td>\n",
       "      <td>350066</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.338168</td>\n",
       "      <td>350067</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.570538</td>\n",
       "      <td>350068</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.013102</td>\n",
       "      <td>350069</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0.095431</td>\n",
       "      <td>350070</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>0.171840</td>\n",
       "      <td>350071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>0.124250</td>\n",
       "      <td>350072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>0.274146</td>\n",
       "      <td>350073</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>0.347207</td>\n",
       "      <td>350074</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>0.023190</td>\n",
       "      <td>350075</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>0.226310</td>\n",
       "      <td>350076</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>0.302242</td>\n",
       "      <td>350077</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>0.003819</td>\n",
       "      <td>350078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>0.108027</td>\n",
       "      <td>350079</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>0.069393</td>\n",
       "      <td>350080</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>0.894724</td>\n",
       "      <td>350081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>0.090553</td>\n",
       "      <td>350082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24970</th>\n",
       "      <td>0.134032</td>\n",
       "      <td>375023</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24971</th>\n",
       "      <td>0.024675</td>\n",
       "      <td>375024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24972</th>\n",
       "      <td>0.519717</td>\n",
       "      <td>375025</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24973</th>\n",
       "      <td>0.683209</td>\n",
       "      <td>375026</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24974</th>\n",
       "      <td>0.736853</td>\n",
       "      <td>375027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24975</th>\n",
       "      <td>0.014909</td>\n",
       "      <td>375028</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24976</th>\n",
       "      <td>0.221899</td>\n",
       "      <td>375029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24977</th>\n",
       "      <td>0.042979</td>\n",
       "      <td>375030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24978</th>\n",
       "      <td>0.361933</td>\n",
       "      <td>375031</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24979</th>\n",
       "      <td>0.539754</td>\n",
       "      <td>375032</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24980</th>\n",
       "      <td>0.012268</td>\n",
       "      <td>375033</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24981</th>\n",
       "      <td>0.097815</td>\n",
       "      <td>375034</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24982</th>\n",
       "      <td>0.758810</td>\n",
       "      <td>375035</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24983</th>\n",
       "      <td>0.250850</td>\n",
       "      <td>375036</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24984</th>\n",
       "      <td>0.012913</td>\n",
       "      <td>375037</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24985</th>\n",
       "      <td>0.371133</td>\n",
       "      <td>375038</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24986</th>\n",
       "      <td>0.250193</td>\n",
       "      <td>375039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24987</th>\n",
       "      <td>0.252646</td>\n",
       "      <td>375040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24988</th>\n",
       "      <td>0.492332</td>\n",
       "      <td>375041</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24989</th>\n",
       "      <td>0.876774</td>\n",
       "      <td>375042</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24990</th>\n",
       "      <td>0.434198</td>\n",
       "      <td>375043</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24991</th>\n",
       "      <td>0.042867</td>\n",
       "      <td>375044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24992</th>\n",
       "      <td>0.474856</td>\n",
       "      <td>375045</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24993</th>\n",
       "      <td>0.371138</td>\n",
       "      <td>375046</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24994</th>\n",
       "      <td>0.260519</td>\n",
       "      <td>375047</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24995</th>\n",
       "      <td>0.506411</td>\n",
       "      <td>375048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24996</th>\n",
       "      <td>0.015603</td>\n",
       "      <td>375049</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24997</th>\n",
       "      <td>0.174075</td>\n",
       "      <td>375050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24998</th>\n",
       "      <td>0.036970</td>\n",
       "      <td>375051</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24999</th>\n",
       "      <td>0.010573</td>\n",
       "      <td>375052</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>25000 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              0  application_key\n",
       "0      0.055446           350053\n",
       "1      0.042235           350054\n",
       "2      0.103503           350055\n",
       "3      0.670476           350056\n",
       "4      0.102000           350057\n",
       "5      0.011206           350058\n",
       "6      0.051739           350059\n",
       "7      0.003864           350060\n",
       "8      0.005794           350061\n",
       "9      0.229367           350062\n",
       "10     0.061044           350063\n",
       "11     0.222500           350064\n",
       "12     0.482598           350065\n",
       "13     0.853469           350066\n",
       "14     0.338168           350067\n",
       "15     0.570538           350068\n",
       "16     0.013102           350069\n",
       "17     0.095431           350070\n",
       "18     0.171840           350071\n",
       "19     0.124250           350072\n",
       "20     0.274146           350073\n",
       "21     0.347207           350074\n",
       "22     0.023190           350075\n",
       "23     0.226310           350076\n",
       "24     0.302242           350077\n",
       "25     0.003819           350078\n",
       "26     0.108027           350079\n",
       "27     0.069393           350080\n",
       "28     0.894724           350081\n",
       "29     0.090553           350082\n",
       "...         ...              ...\n",
       "24970  0.134032           375023\n",
       "24971  0.024675           375024\n",
       "24972  0.519717           375025\n",
       "24973  0.683209           375026\n",
       "24974  0.736853           375027\n",
       "24975  0.014909           375028\n",
       "24976  0.221899           375029\n",
       "24977  0.042979           375030\n",
       "24978  0.361933           375031\n",
       "24979  0.539754           375032\n",
       "24980  0.012268           375033\n",
       "24981  0.097815           375034\n",
       "24982  0.758810           375035\n",
       "24983  0.250850           375036\n",
       "24984  0.012913           375037\n",
       "24985  0.371133           375038\n",
       "24986  0.250193           375039\n",
       "24987  0.252646           375040\n",
       "24988  0.492332           375041\n",
       "24989  0.876774           375042\n",
       "24990  0.434198           375043\n",
       "24991  0.042867           375044\n",
       "24992  0.474856           375045\n",
       "24993  0.371138           375046\n",
       "24994  0.260519           375047\n",
       "24995  0.506411           375048\n",
       "24996  0.015603           375049\n",
       "24997  0.174075           375050\n",
       "24998  0.036970           375051\n",
       "24999  0.010573           375052\n",
       "\n",
       "[25000 rows x 2 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "pred = pred.sort_values(by = 0,kind = 'mergesort',ascending = True) #true for xgb method3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "cols = pred.columns.tolist()\n",
    "cols = cols[-1:] + cols[:-1]\n",
    "pred = pred[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "pred = pred.loc[:,'application_key':]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "submit = pred['application_key'][:10000]\n",
    "submit = pd.DataFrame(submit)\n",
    "submit[0] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# submit.iloc[-400:, 1] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>application_key</th>\n",
       "      <th>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>24172</th>\n",
       "      <td>374225</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22292</th>\n",
       "      <td>372345</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10071</th>\n",
       "      <td>360124</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19117</th>\n",
       "      <td>369170</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4599</th>\n",
       "      <td>354652</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3708</th>\n",
       "      <td>353761</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22263</th>\n",
       "      <td>372316</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1837</th>\n",
       "      <td>351890</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16751</th>\n",
       "      <td>366804</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2469</th>\n",
       "      <td>352522</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11214</th>\n",
       "      <td>361267</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21592</th>\n",
       "      <td>371645</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15849</th>\n",
       "      <td>365902</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23797</th>\n",
       "      <td>373850</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12148</th>\n",
       "      <td>362201</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22865</th>\n",
       "      <td>372918</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18842</th>\n",
       "      <td>368895</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>193</th>\n",
       "      <td>350246</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20330</th>\n",
       "      <td>370383</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3323</th>\n",
       "      <td>353376</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11521</th>\n",
       "      <td>361574</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13522</th>\n",
       "      <td>363575</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18109</th>\n",
       "      <td>368162</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11180</th>\n",
       "      <td>361233</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24825</th>\n",
       "      <td>374878</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>190</th>\n",
       "      <td>350243</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5738</th>\n",
       "      <td>355791</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10881</th>\n",
       "      <td>360934</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>124</th>\n",
       "      <td>350177</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13648</th>\n",
       "      <td>363701</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16730</th>\n",
       "      <td>366783</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3076</th>\n",
       "      <td>353129</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20976</th>\n",
       "      <td>371029</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9451</th>\n",
       "      <td>359504</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17386</th>\n",
       "      <td>367439</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5429</th>\n",
       "      <td>355482</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18081</th>\n",
       "      <td>368134</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2680</th>\n",
       "      <td>352733</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22389</th>\n",
       "      <td>372442</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11605</th>\n",
       "      <td>361658</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12381</th>\n",
       "      <td>362434</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16985</th>\n",
       "      <td>367038</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>993</th>\n",
       "      <td>351046</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4282</th>\n",
       "      <td>354335</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12968</th>\n",
       "      <td>363021</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16663</th>\n",
       "      <td>366716</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23061</th>\n",
       "      <td>373114</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12344</th>\n",
       "      <td>362397</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21941</th>\n",
       "      <td>371994</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11147</th>\n",
       "      <td>361200</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19621</th>\n",
       "      <td>369674</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7274</th>\n",
       "      <td>357327</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8285</th>\n",
       "      <td>358338</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7860</th>\n",
       "      <td>357913</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2417</th>\n",
       "      <td>352470</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17899</th>\n",
       "      <td>367952</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15383</th>\n",
       "      <td>365436</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9335</th>\n",
       "      <td>359388</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5023</th>\n",
       "      <td>355076</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8487</th>\n",
       "      <td>358540</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10000 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       application_key  0\n",
       "24172           374225  0\n",
       "22292           372345  0\n",
       "10071           360124  0\n",
       "19117           369170  0\n",
       "4599            354652  0\n",
       "3708            353761  0\n",
       "22263           372316  0\n",
       "1837            351890  0\n",
       "16751           366804  0\n",
       "2469            352522  0\n",
       "11214           361267  0\n",
       "21592           371645  0\n",
       "15849           365902  0\n",
       "23797           373850  0\n",
       "12148           362201  0\n",
       "22865           372918  0\n",
       "18842           368895  0\n",
       "193             350246  0\n",
       "20330           370383  0\n",
       "3323            353376  0\n",
       "11521           361574  0\n",
       "13522           363575  0\n",
       "18109           368162  0\n",
       "11180           361233  0\n",
       "24825           374878  0\n",
       "190             350243  0\n",
       "5738            355791  0\n",
       "10881           360934  0\n",
       "124             350177  0\n",
       "13648           363701  0\n",
       "...                ... ..\n",
       "16730           366783  0\n",
       "3076            353129  0\n",
       "20976           371029  0\n",
       "9451            359504  0\n",
       "17386           367439  0\n",
       "5429            355482  0\n",
       "18081           368134  0\n",
       "2680            352733  0\n",
       "22389           372442  0\n",
       "11605           361658  0\n",
       "12381           362434  0\n",
       "16985           367038  0\n",
       "993             351046  0\n",
       "4282            354335  0\n",
       "12968           363021  0\n",
       "16663           366716  0\n",
       "23061           373114  0\n",
       "12344           362397  0\n",
       "21941           371994  0\n",
       "11147           361200  0\n",
       "19621           369674  0\n",
       "7274            357327  0\n",
       "8285            358338  0\n",
       "7860            357913  0\n",
       "2417            352470  0\n",
       "17899           367952  0\n",
       "15383           365436  0\n",
       "9335            359388  0\n",
       "5023            355076  0\n",
       "8487            358540  0\n",
       "\n",
       "[10000 rows x 2 columns]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "submit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "submit.to_csv('flabbergasted_IITGuwahati_281.csv',header = False, index = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}