Round-2/Cascade-Cup-2020-Round2.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "KIPJBLtZaQ8X"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "hOmZs9cFaQ8a",
    "outputId": "a01ac509-ce4e-4a39-932b-8ccbf15b870f"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "488877"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train = pd.read_csv(\"./Data/train_age_dataset.csv\")\n",
    "len(train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "QrWPzUIWaQ8c",
    "outputId": "09d95288-5079-4bc1-b9c5-4904bc1a77d4"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "54320"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test = pd.read_csv(\"./Data/test_age_dataset.csv\")\n",
    "len(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "wKNaQf9caQ8c",
    "outputId": "4cb802ab-ab9b-4aac-9925-9f743965ed83"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>prediction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   prediction\n",
       "0           1\n",
       "1           1\n",
       "2           1\n",
       "3           1\n",
       "4           1"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample = pd.read_csv(\"./Data/sample_submission.csv\")\n",
    "sample.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "m3Bw0IWeaQ8d"
   },
   "source": [
    "## Cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "4iIwl3EcaQ8d",
    "outputId": "637ec3fd-e208-43ee-d9b4-076c39ee7ea6",
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Unnamed: 0                         0\n",
       "userId                             0\n",
       "tier                               0\n",
       "gender                             0\n",
       "following_rate                     0\n",
       "followers_avg_age                  0\n",
       "following_avg_age                  0\n",
       "max_repetitive_punc                0\n",
       "num_of_hashtags_per_action         0\n",
       "emoji_count_per_action             0\n",
       "punctuations_per_action            0\n",
       "number_of_words_per_action         0\n",
       "avgCompletion                      0\n",
       "avgTimeSpent                       0\n",
       "avgDuration                        0\n",
       "avgComments                        0\n",
       "creations                          0\n",
       "content_views                      0\n",
       "num_of_comments                    0\n",
       "weekends_trails_watched_per_day    0\n",
       "weekdays_trails_watched_per_day    0\n",
       "slot1_trails_watched_per_day       0\n",
       "slot2_trails_watched_per_day       0\n",
       "slot3_trails_watched_per_day       0\n",
       "slot4_trails_watched_per_day       0\n",
       "avgt2                              0\n",
       "age_group                          0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "lzM4VNA3aQ8d",
    "outputId": "3cffccf5-0881-457f-d62a-ba1e4da496bf",
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Unnamed: 0                         0\n",
       "userId                             0\n",
       "tier                               0\n",
       "gender                             0\n",
       "following_rate                     0\n",
       "followers_avg_age                  0\n",
       "following_avg_age                  0\n",
       "max_repetitive_punc                0\n",
       "num_of_hashtags_per_action         0\n",
       "emoji_count_per_action             0\n",
       "punctuations_per_action            0\n",
       "number_of_words_per_action         0\n",
       "avgCompletion                      0\n",
       "avgTimeSpent                       0\n",
       "avgDuration                        0\n",
       "avgComments                        0\n",
       "creations                          0\n",
       "content_views                      0\n",
       "num_of_comments                    0\n",
       "weekends_trails_watched_per_day    0\n",
       "weekdays_trails_watched_per_day    0\n",
       "slot1_trails_watched_per_day       0\n",
       "slot2_trails_watched_per_day       0\n",
       "slot3_trails_watched_per_day       0\n",
       "slot4_trails_watched_per_day       0\n",
       "avgt2                              0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "m6BjfFi4aQ8e"
   },
   "source": [
    "# WHY ARE THERE NO NULL VALUES"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 243
    },
    "id": "KTmTteE_aQ8e",
    "outputId": "a9a00e38-d627-43b9-956b-3accc42602b4"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>userId</th>\n",
       "      <th>tier</th>\n",
       "      <th>gender</th>\n",
       "      <th>following_rate</th>\n",
       "      <th>followers_avg_age</th>\n",
       "      <th>following_avg_age</th>\n",
       "      <th>max_repetitive_punc</th>\n",
       "      <th>num_of_hashtags_per_action</th>\n",
       "      <th>emoji_count_per_action</th>\n",
       "      <th>...</th>\n",
       "      <th>content_views</th>\n",
       "      <th>num_of_comments</th>\n",
       "      <th>weekends_trails_watched_per_day</th>\n",
       "      <th>weekdays_trails_watched_per_day</th>\n",
       "      <th>slot1_trails_watched_per_day</th>\n",
       "      <th>slot2_trails_watched_per_day</th>\n",
       "      <th>slot3_trails_watched_per_day</th>\n",
       "      <th>slot4_trails_watched_per_day</th>\n",
       "      <th>avgt2</th>\n",
       "      <th>age_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>265153</td>\n",
       "      <td>48958844</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.041667</td>\n",
       "      <td>0.025000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.175000</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>405231</td>\n",
       "      <td>51100441</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.093220</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.012712</td>\n",
       "      <td>0.018644</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.084746</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.033898</td>\n",
       "      <td>82.5</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>57867</td>\n",
       "      <td>6887426</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.002786</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000557</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.002786</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>272618</td>\n",
       "      <td>50742404</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.008403</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.001681</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.008403</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>251123</td>\n",
       "      <td>45589200</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.204918</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.049180</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.008197</td>\n",
       "      <td>0.057377</td>\n",
       "      <td>0.180328</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 27 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0    userId  tier  gender  following_rate  followers_avg_age  \\\n",
       "0      265153  48958844     2       1             0.0                0.0   \n",
       "1      405231  51100441     2       2             0.0                0.0   \n",
       "2       57867   6887426     2       1             0.0                0.0   \n",
       "3      272618  50742404     2       1             0.0                0.0   \n",
       "4      251123  45589200     2       2             0.0                0.0   \n",
       "\n",
       "   following_avg_age  max_repetitive_punc  num_of_hashtags_per_action  \\\n",
       "0                0.0                    0                         0.0   \n",
       "1                0.0                    0                         0.0   \n",
       "2                0.0                    0                         0.0   \n",
       "3                0.0                    0                         0.0   \n",
       "4                0.0                    0                         0.0   \n",
       "\n",
       "   emoji_count_per_action  ...  content_views  num_of_comments  \\\n",
       "0                     0.0  ...       0.200000              0.0   \n",
       "1                     0.0  ...       0.093220              0.0   \n",
       "2                     0.0  ...       0.002786              0.0   \n",
       "3                     0.0  ...       0.008403              0.0   \n",
       "4                     0.0  ...       0.204918              0.0   \n",
       "\n",
       "   weekends_trails_watched_per_day  weekdays_trails_watched_per_day  \\\n",
       "0                         0.041667                         0.025000   \n",
       "1                         0.012712                         0.018644   \n",
       "2                         0.000000                         0.000557   \n",
       "3                         0.000000                         0.001681   \n",
       "4                         0.000000                         0.049180   \n",
       "\n",
       "   slot1_trails_watched_per_day  slot2_trails_watched_per_day  \\\n",
       "0                           0.0                      0.000000   \n",
       "1                           0.0                      0.084746   \n",
       "2                           0.0                      0.002786   \n",
       "3                           0.0                      0.000000   \n",
       "4                           0.0                      0.008197   \n",
       "\n",
       "   slot3_trails_watched_per_day  slot4_trails_watched_per_day  avgt2  \\\n",
       "0                      0.175000                      0.033333    0.0   \n",
       "1                      0.000000                      0.033898   82.5   \n",
       "2                      0.000000                      0.000000    0.0   \n",
       "3                      0.000000                      0.008403    0.0   \n",
       "4                      0.057377                      0.180328    0.0   \n",
       "\n",
       "   age_group  \n",
       "0          1  \n",
       "1          2  \n",
       "2          1  \n",
       "3          1  \n",
       "4          1  \n",
       "\n",
       "[5 rows x 27 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "nwgPgbIYaQ8e"
   },
   "source": [
    "Oh looks like the null values have been filled with zeros"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "TpbjFouSaQ8f"
   },
   "source": [
    "Checking the zero null thing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 461
    },
    "id": "rKjE_Q2paQ8f",
    "outputId": "c66fa7ee-8de3-4d35-9456-05c31b643f23"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>userId</th>\n",
       "      <th>tier</th>\n",
       "      <th>gender</th>\n",
       "      <th>following_rate</th>\n",
       "      <th>followers_avg_age</th>\n",
       "      <th>following_avg_age</th>\n",
       "      <th>max_repetitive_punc</th>\n",
       "      <th>num_of_hashtags_per_action</th>\n",
       "      <th>emoji_count_per_action</th>\n",
       "      <th>...</th>\n",
       "      <th>content_views</th>\n",
       "      <th>num_of_comments</th>\n",
       "      <th>weekends_trails_watched_per_day</th>\n",
       "      <th>weekdays_trails_watched_per_day</th>\n",
       "      <th>slot1_trails_watched_per_day</th>\n",
       "      <th>slot2_trails_watched_per_day</th>\n",
       "      <th>slot3_trails_watched_per_day</th>\n",
       "      <th>slot4_trails_watched_per_day</th>\n",
       "      <th>avgt2</th>\n",
       "      <th>age_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>265153</td>\n",
       "      <td>48958844</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.041667</td>\n",
       "      <td>0.025000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.175000</td>\n",
       "      <td>0.033333</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>405231</td>\n",
       "      <td>51100441</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.093220</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.012712</td>\n",
       "      <td>0.018644</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.084746</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.033898</td>\n",
       "      <td>82.500</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>57867</td>\n",
       "      <td>6887426</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.002786</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000557</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.002786</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>272618</td>\n",
       "      <td>50742404</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.008403</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.001681</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.008403</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>251123</td>\n",
       "      <td>45589200</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.204918</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.049180</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.008197</td>\n",
       "      <td>0.057377</td>\n",
       "      <td>0.180328</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1.0</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>54315</th>\n",
       "      <td>301257</td>\n",
       "      <td>56503645</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.523364</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.108411</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.140187</td>\n",
       "      <td>0.401869</td>\n",
       "      <td>0.000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54316</th>\n",
       "      <td>39167</td>\n",
       "      <td>39840927</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.177419</td>\n",
       "      <td>1.666667</td>\n",
       "      <td>2.210526</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.435484</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.072581</td>\n",
       "      <td>0.058065</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.040323</td>\n",
       "      <td>0.104839</td>\n",
       "      <td>0.290323</td>\n",
       "      <td>140.325</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54317</th>\n",
       "      <td>471826</td>\n",
       "      <td>68398375</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.048387</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.935484</td>\n",
       "      <td>0.048387</td>\n",
       "      <td>0.209677</td>\n",
       "      <td>0.158065</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.435484</td>\n",
       "      <td>0.048387</td>\n",
       "      <td>0.725806</td>\n",
       "      <td>186.000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54318</th>\n",
       "      <td>13839</td>\n",
       "      <td>51709440</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.177966</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.088983</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.177966</td>\n",
       "      <td>0.000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54319</th>\n",
       "      <td>142595</td>\n",
       "      <td>50995322</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.271186</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.135593</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.271186</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>543197 rows × 27 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Unnamed: 0    userId  tier  gender  following_rate  followers_avg_age  \\\n",
       "0          265153  48958844     2       1        0.000000           0.000000   \n",
       "1          405231  51100441     2       2        0.000000           0.000000   \n",
       "2           57867   6887426     2       1        0.000000           0.000000   \n",
       "3          272618  50742404     2       1        0.000000           0.000000   \n",
       "4          251123  45589200     2       2        0.000000           0.000000   \n",
       "...           ...       ...   ...     ...             ...                ...   \n",
       "54315      301257  56503645     2       1        0.000000           0.000000   \n",
       "54316       39167  39840927     1       1        0.177419           1.666667   \n",
       "54317      471826  68398375     2       1        0.048387           0.000000   \n",
       "54318       13839  51709440     1       2        0.000000           0.000000   \n",
       "54319      142595  50995322     3       1        0.000000           0.000000   \n",
       "\n",
       "       following_avg_age  max_repetitive_punc  num_of_hashtags_per_action  \\\n",
       "0               0.000000                    0                         0.0   \n",
       "1               0.000000                    0                         0.0   \n",
       "2               0.000000                    0                         0.0   \n",
       "3               0.000000                    0                         0.0   \n",
       "4               0.000000                    0                         0.0   \n",
       "...                  ...                  ...                         ...   \n",
       "54315           0.000000                    0                         0.0   \n",
       "54316           2.210526                    0                         0.0   \n",
       "54317           0.000000                    3                         0.0   \n",
       "54318           0.000000                    0                         0.0   \n",
       "54319           0.000000                    0                         0.0   \n",
       "\n",
       "       emoji_count_per_action  ...  content_views  num_of_comments  \\\n",
       "0                         0.0  ...       0.200000         0.000000   \n",
       "1                         0.0  ...       0.093220         0.000000   \n",
       "2                         0.0  ...       0.002786         0.000000   \n",
       "3                         0.0  ...       0.008403         0.000000   \n",
       "4                         0.0  ...       0.204918         0.000000   \n",
       "...                       ...  ...            ...              ...   \n",
       "54315                     0.0  ...       0.523364         0.000000   \n",
       "54316                     0.0  ...       0.435484         0.000000   \n",
       "54317                     0.0  ...       0.935484         0.048387   \n",
       "54318                     0.0  ...       0.177966         0.000000   \n",
       "54319                     0.0  ...       0.271186         0.000000   \n",
       "\n",
       "       weekends_trails_watched_per_day  weekdays_trails_watched_per_day  \\\n",
       "0                             0.041667                         0.025000   \n",
       "1                             0.012712                         0.018644   \n",
       "2                             0.000000                         0.000557   \n",
       "3                             0.000000                         0.001681   \n",
       "4                             0.000000                         0.049180   \n",
       "...                                ...                              ...   \n",
       "54315                         0.000000                         0.108411   \n",
       "54316                         0.072581                         0.058065   \n",
       "54317                         0.209677                         0.158065   \n",
       "54318                         0.088983                         0.000000   \n",
       "54319                         0.135593                         0.000000   \n",
       "\n",
       "       slot1_trails_watched_per_day  slot2_trails_watched_per_day  \\\n",
       "0                          0.000000                      0.000000   \n",
       "1                          0.000000                      0.084746   \n",
       "2                          0.000000                      0.002786   \n",
       "3                          0.000000                      0.000000   \n",
       "4                          0.000000                      0.008197   \n",
       "...                             ...                           ...   \n",
       "54315                      0.000000                      0.000000   \n",
       "54316                      0.000000                      0.040323   \n",
       "54317                      0.000000                      0.435484   \n",
       "54318                      0.000000                      0.000000   \n",
       "54319                      0.271186                      0.000000   \n",
       "\n",
       "       slot3_trails_watched_per_day  slot4_trails_watched_per_day    avgt2  \\\n",
       "0                          0.175000                      0.033333    0.000   \n",
       "1                          0.000000                      0.033898   82.500   \n",
       "2                          0.000000                      0.000000    0.000   \n",
       "3                          0.000000                      0.008403    0.000   \n",
       "4                          0.057377                      0.180328    0.000   \n",
       "...                             ...                           ...      ...   \n",
       "54315                      0.140187                      0.401869    0.000   \n",
       "54316                      0.104839                      0.290323  140.325   \n",
       "54317                      0.048387                      0.725806  186.000   \n",
       "54318                      0.000000                      0.177966    0.000   \n",
       "54319                      0.000000                      0.000000    0.000   \n",
       "\n",
       "       age_group  \n",
       "0            1.0  \n",
       "1            2.0  \n",
       "2            1.0  \n",
       "3            1.0  \n",
       "4            1.0  \n",
       "...          ...  \n",
       "54315        NaN  \n",
       "54316        NaN  \n",
       "54317        NaN  \n",
       "54318        NaN  \n",
       "54319        NaN  \n",
       "\n",
       "[543197 rows x 27 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "traintest = pd.concat([train, test])\n",
    "traintest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "n1VD96Y96h6V",
    "outputId": "cc39acb8-5216-4d3a-a080-a1fd2d8fca2b"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "weekends_trails_watched_per_day\n",
       "0.000000     255819\n",
       "0.000313          1\n",
       "0.000332          1\n",
       "0.000385          1\n",
       "0.000413          1\n",
       "              ...  \n",
       "10.039683         1\n",
       "10.144628         1\n",
       "11.391129         1\n",
       "13.110294         1\n",
       "17.201754         1\n",
       "Length: 21446, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "traintest.groupby(['weekends_trails_watched_per_day']).size()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Engineering and Log Scaling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "cTGA37CsCGKy"
   },
   "outputs": [],
   "source": [
    "### Just some approximate function to calculate the age from the age groups\n",
    "traintest['followers_actual_avg'] = 30*traintest['followers_avg_age'] - 20\n",
    "traintest['following_actual_avg'] = 30*traintest['following_avg_age'] - 20\n",
    "\n",
    "### Log Scaling a few features\n",
    "traintest['rep_punc_log'] = np.log(traintest['max_repetitive_punc']+0.000000001)\n",
    "traintest['hpa_log'] = np.log(traintest['num_of_hashtags_per_action']+0.000000001)\n",
    "traintest['avgt2_log'] = np.log(traintest['avgt2']+0.000000001) \n",
    "traintest['ppa_log'] = np.log(traintest['punctuations_per_action']+0.000000001)\n",
    "traintest['epa_log'] = np.log(traintest['emoji_count_per_action']+0.000000001)\n",
    "traintest['wend_mult_log'] = np.log(traintest['punctuations_per_action']*100 + 0.000000001)\n",
    "traintest['avgCompletion_log'] = np.log(np.array(traintest['avgCompletion']).reshape(-1,1)+0.000000001)\n",
    "traintest['avgTimeSpent_log'] = np.log(np.array(traintest['avgTimeSpent']).reshape(-1,1)+0.000000001)\n",
    "\n",
    "### Punctuations per action\n",
    "ppa_bins = [0,0.0000001,0.01,0.025,0.05,0.1,0.15,0.2,0.4,0.6,0.8,1.0,1.2,1.6,2.0,3.0,4.0,5.0,100.0]\n",
    "ppa_labels = list(range(len(ppa_bins)-1))\n",
    "traintest['ppa_binned'] =  pd.cut(traintest['punctuations_per_action'], bins=ppa_bins, labels=ppa_labels)\n",
    "\n",
    "### Words per action\n",
    "wpa_bins = [0,0.0000001,1,2,3,4,5,7,10,50,1000]\n",
    "wpa_labels = [0,1,2,3,4,5,6,7,8,9]\n",
    "traintest['wpa_binned'] =  pd.cut(traintest['number_of_words_per_action'], bins=wpa_bins, labels=wpa_labels)\n",
    "traintest['wpa_log'] = np.log(traintest['number_of_words_per_action']+0.000000001)\n",
    "\n",
    "### Emoji per action\n",
    "epa_bins = [0,0.0000001,0.005,0.01,0.02,0.04,0.08,0.2,0.4,0.6,0.8,1.0,2.0,3.0]\n",
    "epa_labels = list(range(len(epa_bins)-1))\n",
    "traintest['epa_binned'] =  pd.cut(traintest['emoji_count_per_action'], bins=epa_bins, labels=epa_labels)\n",
    "\n",
    "### Creation to consumption ratio\n",
    "traintest['create_to_consume'] = np.divide(traintest['creations'], traintest['content_views'])\n",
    "\n",
    "### Total Watch time\n",
    "traintest['total_watch_time'] = np.multiply(traintest['avgTimeSpent'], traintest['content_views'])\n",
    "\n",
    "### Getting the number of videos watched in each slot on a weekday and weekend\n",
    "traintest['s1_wday'] = np.multiply(traintest['slot1_trails_watched_per_day'], traintest['weekdays_trails_watched_per_day'])\n",
    "traintest['s2_wday'] = np.multiply(traintest['slot2_trails_watched_per_day'], traintest['weekdays_trails_watched_per_day'])\n",
    "traintest['s3_wday'] = np.multiply(traintest['slot3_trails_watched_per_day'], traintest['weekdays_trails_watched_per_day'])\n",
    "traintest['s4_wday'] = np.multiply(traintest['slot4_trails_watched_per_day'], traintest['weekdays_trails_watched_per_day'])\n",
    "traintest['s1_wend'] = np.multiply(traintest['slot1_trails_watched_per_day'], traintest['weekends_trails_watched_per_day'])\n",
    "traintest['s2_wend'] = np.multiply(traintest['slot2_trails_watched_per_day'], traintest['weekends_trails_watched_per_day'])\n",
    "traintest['s3_wend'] = np.multiply(traintest['slot3_trails_watched_per_day'], traintest['weekends_trails_watched_per_day'])\n",
    "traintest['s4_wend'] = np.multiply(traintest['slot4_trails_watched_per_day'], traintest['weekends_trails_watched_per_day'])\n",
    "\n",
    "### Total = avg*num_comments\n",
    "traintest['emoji_total'] = traintest['emoji_count_per_action']*traintest['num_of_comments']\n",
    "traintest['words_total'] = traintest['number_of_words_per_action']*traintest['num_of_comments']\n",
    "traintest['punc_total'] = traintest['punctuations_per_action']*traintest['num_of_comments']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Sp4RKuOED1zN",
    "outputId": "c8943171-a4b3-4b5e-b6cc-c225ea7b78a4"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "num_of_hashtags_per_action\n",
       "0.000000    541965\n",
       "0.000119         1\n",
       "0.000238         1\n",
       "0.000688         1\n",
       "0.000920         1\n",
       "             ...  \n",
       "1.500000         2\n",
       "1.666667         1\n",
       "1.800000         1\n",
       "2.000000         2\n",
       "2.333333         3\n",
       "Length: 444, dtype: int64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "traintest.groupby(['num_of_hashtags_per_action']).size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Vg8lxx4uEIrc",
    "outputId": "1d3ea7b6-7d7a-4b98-df9a-1d9d0c015c2a"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "num_of_comments\n",
       "0.000000    494006\n",
       "0.000631         1\n",
       "0.000639         1\n",
       "0.000660         1\n",
       "0.000664         1\n",
       "             ...  \n",
       "4.344549         1\n",
       "4.516279         1\n",
       "4.575000         1\n",
       "5.296000         1\n",
       "8.196850         1\n",
       "Length: 4089, dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "traintest.groupby(['num_of_comments']).size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "8v1H-QilEb69",
    "outputId": "87565c2a-b049-4212-9194-5aae04c60fcb"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "num_hash\n",
       "0.000000    542892\n",
       "0.000001         1\n",
       "0.000006         1\n",
       "0.000007         1\n",
       "0.000013         1\n",
       "             ...  \n",
       "0.004571         1\n",
       "0.005026         1\n",
       "0.005028         1\n",
       "0.006053         1\n",
       "0.015340         1\n",
       "Length: 303, dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### Creating total number of hashtags\n",
    "traintest['num_hash'] = traintest['num_of_hashtags_per_action']*traintest['num_of_comments']\n",
    "traintest.groupby(['num_hash']).size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Vl36cWyYEs4C",
    "outputId": "7d218392-2c48-49dc-cde0-56d675e8564d"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "num_hash_rescaled\n",
       "0.0        542892\n",
       "1.0             1\n",
       "6.0             1\n",
       "7.0             1\n",
       "13.0            1\n",
       "            ...  \n",
       "4571.0          1\n",
       "5026.0          1\n",
       "5028.0          1\n",
       "6053.0          1\n",
       "15340.0         1\n",
       "Length: 249, dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "traintest['num_hash_rescaled'] = np.round(traintest['num_hash']/0.000001)\n",
    "traintest.groupby(['num_hash_rescaled']).size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "K1owFFSgFRS7",
    "outputId": "1b3ce968-ed10-4dc0-b7a4-d8f377311b99"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "hash_binned\n",
       "-1    542892\n",
       " 1       222\n",
       " 2        59\n",
       " 3        24\n",
       "dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### Binning the number of hashtags\n",
    "def hash_bins(val):\n",
    "  if val == 0:\n",
    "    return -1\n",
    "  elif val<500:\n",
    "    return 1\n",
    "  elif val<1500:\n",
    "    return 2\n",
    "  else:\n",
    "    return 3\n",
    "\n",
    "traintest['hash_binned'] = traintest['num_hash_rescaled'].apply(lambda x: hash_bins(x))\n",
    "traintest.groupby(['hash_binned']).size()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "4Lx0shGQaQ8f",
    "outputId": "dddd5990-0f65-468c-f0cb-ef38d72f58a1",
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "following_rate\n",
       "0.000000      345156\n",
       "0.000686           1\n",
       "0.000689           1\n",
       "0.000704           1\n",
       "0.000772           2\n",
       "               ...  \n",
       "426.504000         1\n",
       "488.362205         1\n",
       "563.631147         1\n",
       "685.277778         1\n",
       "895.304000         1\n",
       "Length: 16377, dtype: int64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "traintest.groupby(['following_rate']).size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "id": "98MkVW6haQ8g"
   },
   "outputs": [],
   "source": [
    "traintest['Majority Followers'] = np.round(traintest['followers_avg_age'])\n",
    "traintest['Majority Following'] = np.round(traintest['following_avg_age'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "AJFCFQthaQ8g",
    "outputId": "66291bbc-95d3-4cf9-f83b-cf79454f0762"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "content_views\n",
       "0.000873     1\n",
       "0.000883     1\n",
       "0.000887     1\n",
       "0.000895     1\n",
       "0.000909     1\n",
       "            ..\n",
       "35.801587    1\n",
       "37.206349    1\n",
       "39.966387    1\n",
       "39.991935    1\n",
       "75.662281    1\n",
       "Length: 33941, dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "traintest.groupby(['content_views']).size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "5QsjmoC7aQ8g",
    "outputId": "4f32eac4-8468-4cf1-f4b8-ffc1d60cdb0c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "view_binned\n",
       "(0.0, 0.01]       45002\n",
       "(0.01, 0.1]      198281\n",
       "(0.1, 1.0]       250142\n",
       "(1.0, 10.0]       49215\n",
       "(10.0, 100.0]       557\n",
       "dtype: int64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bins = pd.IntervalIndex.from_tuples([(0, 0.01), (0.01, 0.1), (0.1, 1), (1,10), (10,100)])\n",
    "traintest['view_binned'] = pd.cut(train['content_views'], bins = bins)\n",
    "traintest.groupby(['view_binned']).size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "kRuOIgc-aQ8h",
    "outputId": "65ca8179-f821-4a16-8b45-5722fe1fa98a"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "view_binned\n",
       "0     45002\n",
       "1    198281\n",
       "2    250142\n",
       "3     49215\n",
       "4       557\n",
       "dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Label Encoding the Interval type columns\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "view_le = LabelEncoder()\n",
    "traintest['view_binned'] = view_le.fit_transform(traintest['view_binned'])\n",
    "wpa_le = LabelEncoder()\n",
    "traintest['wpa_binned'] = wpa_le.fit_transform(traintest['wpa_binned'])\n",
    "epa_le = LabelEncoder()\n",
    "traintest['epa_binned'] = epa_le.fit_transform(traintest['epa_binned'])\n",
    "ppa_le = LabelEncoder()\n",
    "traintest['ppa_binned'] = ppa_le.fit_transform(traintest['ppa_binned'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "52uUxP8SaQ8h"
   },
   "source": [
    "## Model Testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "id": "VGBaqLo_aQ8h"
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import f1_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "id": "YuOjLQymaQ8h"
   },
   "outputs": [],
   "source": [
    "train = traintest[:488877]\n",
    "test = traintest[488877:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "id": "6coLMx-maQ8i"
   },
   "outputs": [],
   "source": [
    "X = train.drop(['age_group'], axis = 1)\n",
    "Y = train['age_group']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "id": "Q83fKQbJaQ8i"
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_Train, X_CV, Y_Train, Y_CV = train_test_split(X, Y, test_size = 0.1, random_state = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "UjuNY8y0aQ8j",
    "outputId": "94f0374f-460b-4132-b090-fd52eeae998e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0]\tvalidation_0-merror:0.24227\tvalidation_1-merror:0.24321\n",
      "Multiple eval metrics have been passed: 'validation_1-merror' will be used for early stopping.\n",
      "\n",
      "Will train until validation_1-merror hasn't improved in 200 rounds.\n",
      "[1]\tvalidation_0-merror:0.24139\tvalidation_1-merror:0.24084\n",
      "[2]\tvalidation_0-merror:0.24047\tvalidation_1-merror:0.24022\n",
      "[3]\tvalidation_0-merror:0.24033\tvalidation_1-merror:0.24004\n",
      "[4]\tvalidation_0-merror:0.23987\tvalidation_1-merror:0.23879\n",
      "[5]\tvalidation_0-merror:0.23919\tvalidation_1-merror:0.23940\n",
      "[6]\tvalidation_0-merror:0.23854\tvalidation_1-merror:0.23902\n",
      "[7]\tvalidation_0-merror:0.23815\tvalidation_1-merror:0.23844\n",
      "[8]\tvalidation_0-merror:0.23693\tvalidation_1-merror:0.23756\n",
      "[9]\tvalidation_0-merror:0.23597\tvalidation_1-merror:0.23679\n",
      "[10]\tvalidation_0-merror:0.23547\tvalidation_1-merror:0.23640\n",
      "[11]\tvalidation_0-merror:0.23480\tvalidation_1-merror:0.23591\n",
      "[12]\tvalidation_0-merror:0.23443\tvalidation_1-merror:0.23535\n",
      "[13]\tvalidation_0-merror:0.23369\tvalidation_1-merror:0.23462\n",
      "[14]\tvalidation_0-merror:0.23323\tvalidation_1-merror:0.23462\n",
      "[15]\tvalidation_0-merror:0.23284\tvalidation_1-merror:0.23470\n",
      "[16]\tvalidation_0-merror:0.23239\tvalidation_1-merror:0.23445\n",
      "[17]\tvalidation_0-merror:0.23176\tvalidation_1-merror:0.23411\n",
      "[18]\tvalidation_0-merror:0.23102\tvalidation_1-merror:0.23359\n",
      "[19]\tvalidation_0-merror:0.23038\tvalidation_1-merror:0.23351\n",
      "[20]\tvalidation_0-merror:0.22920\tvalidation_1-merror:0.23276\n",
      "[21]\tvalidation_0-merror:0.22900\tvalidation_1-merror:0.23276\n",
      "[22]\tvalidation_0-merror:0.22737\tvalidation_1-merror:0.23169\n",
      "[23]\tvalidation_0-merror:0.22691\tvalidation_1-merror:0.23134\n",
      "[24]\tvalidation_0-merror:0.22640\tvalidation_1-merror:0.23118\n",
      "[25]\tvalidation_0-merror:0.22593\tvalidation_1-merror:0.23092\n",
      "[26]\tvalidation_0-merror:0.22556\tvalidation_1-merror:0.23038\n",
      "[27]\tvalidation_0-merror:0.22513\tvalidation_1-merror:0.23004\n",
      "[28]\tvalidation_0-merror:0.22453\tvalidation_1-merror:0.22967\n",
      "[29]\tvalidation_0-merror:0.22434\tvalidation_1-merror:0.22969\n",
      "[30]\tvalidation_0-merror:0.22401\tvalidation_1-merror:0.22922\n",
      "[31]\tvalidation_0-merror:0.22382\tvalidation_1-merror:0.22887\n",
      "[32]\tvalidation_0-merror:0.22332\tvalidation_1-merror:0.22881\n",
      "[33]\tvalidation_0-merror:0.22237\tvalidation_1-merror:0.22793\n",
      "[34]\tvalidation_0-merror:0.22211\tvalidation_1-merror:0.22783\n",
      "[35]\tvalidation_0-merror:0.22157\tvalidation_1-merror:0.22760\n",
      "[36]\tvalidation_0-merror:0.22096\tvalidation_1-merror:0.22775\n",
      "[37]\tvalidation_0-merror:0.22072\tvalidation_1-merror:0.22732\n",
      "[38]\tvalidation_0-merror:0.22049\tvalidation_1-merror:0.22723\n",
      "[39]\tvalidation_0-merror:0.22014\tvalidation_1-merror:0.22666\n",
      "[40]\tvalidation_0-merror:0.21958\tvalidation_1-merror:0.22662\n",
      "[41]\tvalidation_0-merror:0.21908\tvalidation_1-merror:0.22611\n",
      "[42]\tvalidation_0-merror:0.21872\tvalidation_1-merror:0.22550\n",
      "[43]\tvalidation_0-merror:0.21844\tvalidation_1-merror:0.22568\n",
      "[44]\tvalidation_0-merror:0.21813\tvalidation_1-merror:0.22519\n",
      "[45]\tvalidation_0-merror:0.21778\tvalidation_1-merror:0.22496\n",
      "[46]\tvalidation_0-merror:0.21677\tvalidation_1-merror:0.22435\n",
      "[47]\tvalidation_0-merror:0.21588\tvalidation_1-merror:0.22378\n",
      "[48]\tvalidation_0-merror:0.21472\tvalidation_1-merror:0.22292\n",
      "[49]\tvalidation_0-merror:0.21379\tvalidation_1-merror:0.22232\n",
      "[50]\tvalidation_0-merror:0.21283\tvalidation_1-merror:0.22140\n",
      "[51]\tvalidation_0-merror:0.21200\tvalidation_1-merror:0.22087\n",
      "[52]\tvalidation_0-merror:0.21171\tvalidation_1-merror:0.22099\n",
      "[53]\tvalidation_0-merror:0.21123\tvalidation_1-merror:0.22118\n",
      "[54]\tvalidation_0-merror:0.21068\tvalidation_1-merror:0.22099\n",
      "[55]\tvalidation_0-merror:0.20990\tvalidation_1-merror:0.22079\n",
      "[56]\tvalidation_0-merror:0.20954\tvalidation_1-merror:0.22057\n",
      "[57]\tvalidation_0-merror:0.20880\tvalidation_1-merror:0.21966\n",
      "[58]\tvalidation_0-merror:0.20844\tvalidation_1-merror:0.21958\n",
      "[59]\tvalidation_0-merror:0.20803\tvalidation_1-merror:0.21965\n",
      "[60]\tvalidation_0-merror:0.20761\tvalidation_1-merror:0.21913\n",
      "[61]\tvalidation_0-merror:0.20714\tvalidation_1-merror:0.21887\n",
      "[62]\tvalidation_0-merror:0.20612\tvalidation_1-merror:0.21829\n",
      "[63]\tvalidation_0-merror:0.20568\tvalidation_1-merror:0.21836\n",
      "[64]\tvalidation_0-merror:0.20510\tvalidation_1-merror:0.21803\n",
      "[65]\tvalidation_0-merror:0.20458\tvalidation_1-merror:0.21731\n",
      "[66]\tvalidation_0-merror:0.20407\tvalidation_1-merror:0.21715\n",
      "[67]\tvalidation_0-merror:0.20358\tvalidation_1-merror:0.21680\n",
      "[68]\tvalidation_0-merror:0.20332\tvalidation_1-merror:0.21674\n",
      "[69]\tvalidation_0-merror:0.20299\tvalidation_1-merror:0.21629\n",
      "[70]\tvalidation_0-merror:0.20285\tvalidation_1-merror:0.21639\n",
      "[71]\tvalidation_0-merror:0.20263\tvalidation_1-merror:0.21586\n",
      "[72]\tvalidation_0-merror:0.20233\tvalidation_1-merror:0.21619\n",
      "[73]\tvalidation_0-merror:0.20212\tvalidation_1-merror:0.21637\n",
      "[74]\tvalidation_0-merror:0.20173\tvalidation_1-merror:0.21594\n",
      "[75]\tvalidation_0-merror:0.20137\tvalidation_1-merror:0.21533\n",
      "[76]\tvalidation_0-merror:0.20110\tvalidation_1-merror:0.21508\n",
      "[77]\tvalidation_0-merror:0.20078\tvalidation_1-merror:0.21498\n",
      "[78]\tvalidation_0-merror:0.20049\tvalidation_1-merror:0.21467\n",
      "[79]\tvalidation_0-merror:0.20023\tvalidation_1-merror:0.21467\n",
      "[80]\tvalidation_0-merror:0.19989\tvalidation_1-merror:0.21478\n",
      "[81]\tvalidation_0-merror:0.19941\tvalidation_1-merror:0.21445\n",
      "[82]\tvalidation_0-merror:0.19903\tvalidation_1-merror:0.21427\n",
      "[83]\tvalidation_0-merror:0.19875\tvalidation_1-merror:0.21416\n",
      "[84]\tvalidation_0-merror:0.19815\tvalidation_1-merror:0.21384\n",
      "[85]\tvalidation_0-merror:0.19772\tvalidation_1-merror:0.21373\n",
      "[86]\tvalidation_0-merror:0.19742\tvalidation_1-merror:0.21375\n",
      "[87]\tvalidation_0-merror:0.19719\tvalidation_1-merror:0.21373\n",
      "[88]\tvalidation_0-merror:0.19698\tvalidation_1-merror:0.21357\n",
      "[89]\tvalidation_0-merror:0.19666\tvalidation_1-merror:0.21361\n",
      "[90]\tvalidation_0-merror:0.19610\tvalidation_1-merror:0.21341\n",
      "[91]\tvalidation_0-merror:0.19581\tvalidation_1-merror:0.21324\n",
      "[92]\tvalidation_0-merror:0.19551\tvalidation_1-merror:0.21316\n",
      "[93]\tvalidation_0-merror:0.19478\tvalidation_1-merror:0.21255\n",
      "[94]\tvalidation_0-merror:0.19453\tvalidation_1-merror:0.21238\n",
      "[95]\tvalidation_0-merror:0.19420\tvalidation_1-merror:0.21216\n",
      "[96]\tvalidation_0-merror:0.19393\tvalidation_1-merror:0.21183\n",
      "[97]\tvalidation_0-merror:0.19361\tvalidation_1-merror:0.21159\n",
      "[98]\tvalidation_0-merror:0.19348\tvalidation_1-merror:0.21134\n",
      "[99]\tvalidation_0-merror:0.19320\tvalidation_1-merror:0.21163\n",
      "[100]\tvalidation_0-merror:0.19291\tvalidation_1-merror:0.21148\n",
      "[101]\tvalidation_0-merror:0.19255\tvalidation_1-merror:0.21095\n",
      "[102]\tvalidation_0-merror:0.19223\tvalidation_1-merror:0.21093\n",
      "[103]\tvalidation_0-merror:0.19181\tvalidation_1-merror:0.21058\n",
      "[104]\tvalidation_0-merror:0.19164\tvalidation_1-merror:0.21036\n",
      "[105]\tvalidation_0-merror:0.19141\tvalidation_1-merror:0.21022\n",
      "[106]\tvalidation_0-merror:0.19110\tvalidation_1-merror:0.21032\n",
      "[107]\tvalidation_0-merror:0.19088\tvalidation_1-merror:0.21050\n",
      "[108]\tvalidation_0-merror:0.19060\tvalidation_1-merror:0.21026\n",
      "[109]\tvalidation_0-merror:0.19001\tvalidation_1-merror:0.21005\n",
      "[110]\tvalidation_0-merror:0.18977\tvalidation_1-merror:0.21007\n",
      "[111]\tvalidation_0-merror:0.18970\tvalidation_1-merror:0.20985\n",
      "[112]\tvalidation_0-merror:0.18890\tvalidation_1-merror:0.20948\n",
      "[113]\tvalidation_0-merror:0.18776\tvalidation_1-merror:0.20897\n",
      "[114]\tvalidation_0-merror:0.18768\tvalidation_1-merror:0.20893\n",
      "[115]\tvalidation_0-merror:0.18744\tvalidation_1-merror:0.20886\n",
      "[116]\tvalidation_0-merror:0.18699\tvalidation_1-merror:0.20856\n",
      "[117]\tvalidation_0-merror:0.18675\tvalidation_1-merror:0.20862\n",
      "[118]\tvalidation_0-merror:0.18653\tvalidation_1-merror:0.20866\n",
      "[119]\tvalidation_0-merror:0.18625\tvalidation_1-merror:0.20868\n",
      "[120]\tvalidation_0-merror:0.18604\tvalidation_1-merror:0.20862\n",
      "[121]\tvalidation_0-merror:0.18575\tvalidation_1-merror:0.20803\n",
      "[122]\tvalidation_0-merror:0.18547\tvalidation_1-merror:0.20792\n",
      "[123]\tvalidation_0-merror:0.18511\tvalidation_1-merror:0.20776\n",
      "[124]\tvalidation_0-merror:0.18493\tvalidation_1-merror:0.20747\n",
      "[125]\tvalidation_0-merror:0.18468\tvalidation_1-merror:0.20735\n",
      "[126]\tvalidation_0-merror:0.18423\tvalidation_1-merror:0.20743\n",
      "[127]\tvalidation_0-merror:0.18407\tvalidation_1-merror:0.20725\n",
      "[128]\tvalidation_0-merror:0.18382\tvalidation_1-merror:0.20717\n",
      "[129]\tvalidation_0-merror:0.18358\tvalidation_1-merror:0.20725\n",
      "[130]\tvalidation_0-merror:0.18330\tvalidation_1-merror:0.20709\n",
      "[131]\tvalidation_0-merror:0.18303\tvalidation_1-merror:0.20686\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[132]\tvalidation_0-merror:0.18287\tvalidation_1-merror:0.20682\n",
      "[133]\tvalidation_0-merror:0.18269\tvalidation_1-merror:0.20696\n",
      "[134]\tvalidation_0-merror:0.18233\tvalidation_1-merror:0.20680\n",
      "[135]\tvalidation_0-merror:0.18195\tvalidation_1-merror:0.20668\n",
      "[136]\tvalidation_0-merror:0.18177\tvalidation_1-merror:0.20653\n",
      "[137]\tvalidation_0-merror:0.18149\tvalidation_1-merror:0.20672\n",
      "[138]\tvalidation_0-merror:0.18128\tvalidation_1-merror:0.20664\n",
      "[139]\tvalidation_0-merror:0.18112\tvalidation_1-merror:0.20676\n",
      "[140]\tvalidation_0-merror:0.18092\tvalidation_1-merror:0.20670\n",
      "[141]\tvalidation_0-merror:0.18032\tvalidation_1-merror:0.20645\n",
      "[142]\tvalidation_0-merror:0.18013\tvalidation_1-merror:0.20651\n",
      "[143]\tvalidation_0-merror:0.17983\tvalidation_1-merror:0.20627\n",
      "[144]\tvalidation_0-merror:0.17934\tvalidation_1-merror:0.20565\n",
      "[145]\tvalidation_0-merror:0.17893\tvalidation_1-merror:0.20533\n",
      "[146]\tvalidation_0-merror:0.17847\tvalidation_1-merror:0.20490\n",
      "[147]\tvalidation_0-merror:0.17825\tvalidation_1-merror:0.20449\n",
      "[148]\tvalidation_0-merror:0.17777\tvalidation_1-merror:0.20434\n",
      "[149]\tvalidation_0-merror:0.17753\tvalidation_1-merror:0.20424\n",
      "[150]\tvalidation_0-merror:0.17738\tvalidation_1-merror:0.20416\n",
      "[151]\tvalidation_0-merror:0.17716\tvalidation_1-merror:0.20436\n",
      "[152]\tvalidation_0-merror:0.17673\tvalidation_1-merror:0.20414\n",
      "[153]\tvalidation_0-merror:0.17661\tvalidation_1-merror:0.20424\n",
      "[154]\tvalidation_0-merror:0.17643\tvalidation_1-merror:0.20420\n",
      "[155]\tvalidation_0-merror:0.17577\tvalidation_1-merror:0.20396\n",
      "[156]\tvalidation_0-merror:0.17513\tvalidation_1-merror:0.20349\n",
      "[157]\tvalidation_0-merror:0.17492\tvalidation_1-merror:0.20340\n",
      "[158]\tvalidation_0-merror:0.17479\tvalidation_1-merror:0.20332\n",
      "[159]\tvalidation_0-merror:0.17466\tvalidation_1-merror:0.20338\n",
      "[160]\tvalidation_0-merror:0.17431\tvalidation_1-merror:0.20312\n",
      "[161]\tvalidation_0-merror:0.17422\tvalidation_1-merror:0.20312\n",
      "[162]\tvalidation_0-merror:0.17405\tvalidation_1-merror:0.20310\n",
      "[163]\tvalidation_0-merror:0.17383\tvalidation_1-merror:0.20328\n",
      "[164]\tvalidation_0-merror:0.17368\tvalidation_1-merror:0.20312\n",
      "[165]\tvalidation_0-merror:0.17361\tvalidation_1-merror:0.20289\n",
      "[166]\tvalidation_0-merror:0.17328\tvalidation_1-merror:0.20287\n",
      "[167]\tvalidation_0-merror:0.17283\tvalidation_1-merror:0.20271\n",
      "[168]\tvalidation_0-merror:0.17264\tvalidation_1-merror:0.20285\n",
      "[169]\tvalidation_0-merror:0.17246\tvalidation_1-merror:0.20273\n",
      "[170]\tvalidation_0-merror:0.17218\tvalidation_1-merror:0.20275\n",
      "[171]\tvalidation_0-merror:0.17200\tvalidation_1-merror:0.20261\n",
      "[172]\tvalidation_0-merror:0.17171\tvalidation_1-merror:0.20209\n",
      "[173]\tvalidation_0-merror:0.17121\tvalidation_1-merror:0.20160\n",
      "[174]\tvalidation_0-merror:0.17069\tvalidation_1-merror:0.20120\n",
      "[175]\tvalidation_0-merror:0.17039\tvalidation_1-merror:0.20075\n",
      "[176]\tvalidation_0-merror:0.17025\tvalidation_1-merror:0.20060\n",
      "[177]\tvalidation_0-merror:0.16997\tvalidation_1-merror:0.20048\n",
      "[178]\tvalidation_0-merror:0.16980\tvalidation_1-merror:0.20066\n",
      "[179]\tvalidation_0-merror:0.16964\tvalidation_1-merror:0.20081\n",
      "[180]\tvalidation_0-merror:0.16943\tvalidation_1-merror:0.20075\n",
      "[181]\tvalidation_0-merror:0.16900\tvalidation_1-merror:0.20072\n",
      "[182]\tvalidation_0-merror:0.16859\tvalidation_1-merror:0.20044\n",
      "[183]\tvalidation_0-merror:0.16845\tvalidation_1-merror:0.20048\n",
      "[184]\tvalidation_0-merror:0.16822\tvalidation_1-merror:0.20052\n",
      "[185]\tvalidation_0-merror:0.16787\tvalidation_1-merror:0.20031\n",
      "[186]\tvalidation_0-merror:0.16779\tvalidation_1-merror:0.20030\n",
      "[187]\tvalidation_0-merror:0.16761\tvalidation_1-merror:0.20052\n",
      "[188]\tvalidation_0-merror:0.16739\tvalidation_1-merror:0.20050\n",
      "[189]\tvalidation_0-merror:0.16730\tvalidation_1-merror:0.20054\n",
      "[190]\tvalidation_0-merror:0.16715\tvalidation_1-merror:0.20060\n",
      "[191]\tvalidation_0-merror:0.16706\tvalidation_1-merror:0.20058\n",
      "[192]\tvalidation_0-merror:0.16694\tvalidation_1-merror:0.20062\n",
      "[193]\tvalidation_0-merror:0.16679\tvalidation_1-merror:0.20076\n",
      "[194]\tvalidation_0-merror:0.16657\tvalidation_1-merror:0.20066\n",
      "[195]\tvalidation_0-merror:0.16615\tvalidation_1-merror:0.20044\n",
      "[196]\tvalidation_0-merror:0.16587\tvalidation_1-merror:0.20031\n",
      "[197]\tvalidation_0-merror:0.16572\tvalidation_1-merror:0.20017\n",
      "[198]\tvalidation_0-merror:0.16554\tvalidation_1-merror:0.19991\n",
      "[199]\tvalidation_0-merror:0.16518\tvalidation_1-merror:0.19962\n",
      "[200]\tvalidation_0-merror:0.16508\tvalidation_1-merror:0.19962\n",
      "[201]\tvalidation_0-merror:0.16484\tvalidation_1-merror:0.19952\n",
      "[202]\tvalidation_0-merror:0.16448\tvalidation_1-merror:0.19950\n",
      "[203]\tvalidation_0-merror:0.16426\tvalidation_1-merror:0.19925\n",
      "[204]\tvalidation_0-merror:0.16404\tvalidation_1-merror:0.19901\n",
      "[205]\tvalidation_0-merror:0.16376\tvalidation_1-merror:0.19860\n",
      "[206]\tvalidation_0-merror:0.16366\tvalidation_1-merror:0.19872\n",
      "[207]\tvalidation_0-merror:0.16350\tvalidation_1-merror:0.19856\n",
      "[208]\tvalidation_0-merror:0.16338\tvalidation_1-merror:0.19860\n",
      "[209]\tvalidation_0-merror:0.16310\tvalidation_1-merror:0.19835\n",
      "[210]\tvalidation_0-merror:0.16272\tvalidation_1-merror:0.19802\n",
      "[211]\tvalidation_0-merror:0.16256\tvalidation_1-merror:0.19800\n",
      "[212]\tvalidation_0-merror:0.16245\tvalidation_1-merror:0.19788\n",
      "[213]\tvalidation_0-merror:0.16228\tvalidation_1-merror:0.19786\n",
      "[214]\tvalidation_0-merror:0.16200\tvalidation_1-merror:0.19784\n",
      "[215]\tvalidation_0-merror:0.16184\tvalidation_1-merror:0.19780\n",
      "[216]\tvalidation_0-merror:0.16159\tvalidation_1-merror:0.19770\n",
      "[217]\tvalidation_0-merror:0.16135\tvalidation_1-merror:0.19764\n",
      "[218]\tvalidation_0-merror:0.16121\tvalidation_1-merror:0.19762\n",
      "[219]\tvalidation_0-merror:0.16093\tvalidation_1-merror:0.19784\n",
      "[220]\tvalidation_0-merror:0.16084\tvalidation_1-merror:0.19780\n",
      "[221]\tvalidation_0-merror:0.16067\tvalidation_1-merror:0.19780\n",
      "[222]\tvalidation_0-merror:0.16057\tvalidation_1-merror:0.19794\n",
      "[223]\tvalidation_0-merror:0.16030\tvalidation_1-merror:0.19782\n",
      "[224]\tvalidation_0-merror:0.16006\tvalidation_1-merror:0.19772\n",
      "[225]\tvalidation_0-merror:0.15995\tvalidation_1-merror:0.19776\n",
      "[226]\tvalidation_0-merror:0.15971\tvalidation_1-merror:0.19772\n",
      "[227]\tvalidation_0-merror:0.15954\tvalidation_1-merror:0.19770\n",
      "[228]\tvalidation_0-merror:0.15935\tvalidation_1-merror:0.19796\n",
      "[229]\tvalidation_0-merror:0.15910\tvalidation_1-merror:0.19800\n",
      "[230]\tvalidation_0-merror:0.15895\tvalidation_1-merror:0.19780\n",
      "[231]\tvalidation_0-merror:0.15867\tvalidation_1-merror:0.19778\n",
      "[232]\tvalidation_0-merror:0.15861\tvalidation_1-merror:0.19792\n",
      "[233]\tvalidation_0-merror:0.15840\tvalidation_1-merror:0.19786\n",
      "[234]\tvalidation_0-merror:0.15832\tvalidation_1-merror:0.19778\n",
      "[235]\tvalidation_0-merror:0.15819\tvalidation_1-merror:0.19774\n",
      "[236]\tvalidation_0-merror:0.15799\tvalidation_1-merror:0.19766\n",
      "[237]\tvalidation_0-merror:0.15787\tvalidation_1-merror:0.19774\n",
      "[238]\tvalidation_0-merror:0.15781\tvalidation_1-merror:0.19759\n",
      "[239]\tvalidation_0-merror:0.15754\tvalidation_1-merror:0.19747\n",
      "[240]\tvalidation_0-merror:0.15739\tvalidation_1-merror:0.19751\n",
      "[241]\tvalidation_0-merror:0.15721\tvalidation_1-merror:0.19753\n",
      "[242]\tvalidation_0-merror:0.15716\tvalidation_1-merror:0.19751\n",
      "[243]\tvalidation_0-merror:0.15698\tvalidation_1-merror:0.19741\n",
      "[244]\tvalidation_0-merror:0.15667\tvalidation_1-merror:0.19729\n",
      "[245]\tvalidation_0-merror:0.15651\tvalidation_1-merror:0.19747\n",
      "[246]\tvalidation_0-merror:0.15609\tvalidation_1-merror:0.19717\n",
      "[247]\tvalidation_0-merror:0.15587\tvalidation_1-merror:0.19696\n",
      "[248]\tvalidation_0-merror:0.15543\tvalidation_1-merror:0.19676\n",
      "[249]\tvalidation_0-merror:0.15543\tvalidation_1-merror:0.19692\n",
      "[250]\tvalidation_0-merror:0.15514\tvalidation_1-merror:0.19698\n",
      "[251]\tvalidation_0-merror:0.15498\tvalidation_1-merror:0.19688\n",
      "[252]\tvalidation_0-merror:0.15485\tvalidation_1-merror:0.19706\n",
      "[253]\tvalidation_0-merror:0.15470\tvalidation_1-merror:0.19714\n",
      "[254]\tvalidation_0-merror:0.15438\tvalidation_1-merror:0.19688\n",
      "[255]\tvalidation_0-merror:0.15417\tvalidation_1-merror:0.19680\n",
      "[256]\tvalidation_0-merror:0.15393\tvalidation_1-merror:0.19696\n",
      "[257]\tvalidation_0-merror:0.15382\tvalidation_1-merror:0.19692\n",
      "[258]\tvalidation_0-merror:0.15362\tvalidation_1-merror:0.19702\n",
      "[259]\tvalidation_0-merror:0.15354\tvalidation_1-merror:0.19692\n",
      "[260]\tvalidation_0-merror:0.15336\tvalidation_1-merror:0.19694\n",
      "[261]\tvalidation_0-merror:0.15323\tvalidation_1-merror:0.19682\n",
      "[262]\tvalidation_0-merror:0.15303\tvalidation_1-merror:0.19673\n",
      "[263]\tvalidation_0-merror:0.15292\tvalidation_1-merror:0.19657\n",
      "[264]\tvalidation_0-merror:0.15263\tvalidation_1-merror:0.19604\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[265]\tvalidation_0-merror:0.15252\tvalidation_1-merror:0.19606\n",
      "[266]\tvalidation_0-merror:0.15231\tvalidation_1-merror:0.19620\n",
      "[267]\tvalidation_0-merror:0.15215\tvalidation_1-merror:0.19620\n",
      "[268]\tvalidation_0-merror:0.15185\tvalidation_1-merror:0.19590\n",
      "[269]\tvalidation_0-merror:0.15178\tvalidation_1-merror:0.19592\n",
      "[270]\tvalidation_0-merror:0.15169\tvalidation_1-merror:0.19588\n",
      "[271]\tvalidation_0-merror:0.15160\tvalidation_1-merror:0.19575\n",
      "[272]\tvalidation_0-merror:0.15125\tvalidation_1-merror:0.19545\n",
      "[273]\tvalidation_0-merror:0.15103\tvalidation_1-merror:0.19561\n",
      "[274]\tvalidation_0-merror:0.15080\tvalidation_1-merror:0.19569\n",
      "[275]\tvalidation_0-merror:0.15075\tvalidation_1-merror:0.19565\n",
      "[276]\tvalidation_0-merror:0.15049\tvalidation_1-merror:0.19549\n",
      "[277]\tvalidation_0-merror:0.15036\tvalidation_1-merror:0.19559\n",
      "[278]\tvalidation_0-merror:0.15025\tvalidation_1-merror:0.19551\n",
      "[279]\tvalidation_0-merror:0.14996\tvalidation_1-merror:0.19541\n",
      "[280]\tvalidation_0-merror:0.14973\tvalidation_1-merror:0.19514\n",
      "[281]\tvalidation_0-merror:0.14961\tvalidation_1-merror:0.19524\n",
      "[282]\tvalidation_0-merror:0.14947\tvalidation_1-merror:0.19530\n",
      "[283]\tvalidation_0-merror:0.14938\tvalidation_1-merror:0.19559\n",
      "[284]\tvalidation_0-merror:0.14925\tvalidation_1-merror:0.19553\n",
      "[285]\tvalidation_0-merror:0.14915\tvalidation_1-merror:0.19555\n",
      "[286]\tvalidation_0-merror:0.14905\tvalidation_1-merror:0.19551\n",
      "[287]\tvalidation_0-merror:0.14893\tvalidation_1-merror:0.19516\n",
      "[288]\tvalidation_0-merror:0.14884\tvalidation_1-merror:0.19518\n",
      "[289]\tvalidation_0-merror:0.14863\tvalidation_1-merror:0.19510\n",
      "[290]\tvalidation_0-merror:0.14852\tvalidation_1-merror:0.19518\n",
      "[291]\tvalidation_0-merror:0.14836\tvalidation_1-merror:0.19500\n",
      "[292]\tvalidation_0-merror:0.14822\tvalidation_1-merror:0.19526\n",
      "[293]\tvalidation_0-merror:0.14812\tvalidation_1-merror:0.19526\n",
      "[294]\tvalidation_0-merror:0.14789\tvalidation_1-merror:0.19492\n",
      "[295]\tvalidation_0-merror:0.14762\tvalidation_1-merror:0.19475\n",
      "[296]\tvalidation_0-merror:0.14752\tvalidation_1-merror:0.19471\n",
      "[297]\tvalidation_0-merror:0.14749\tvalidation_1-merror:0.19477\n",
      "[298]\tvalidation_0-merror:0.14731\tvalidation_1-merror:0.19473\n",
      "[299]\tvalidation_0-merror:0.14712\tvalidation_1-merror:0.19469\n",
      "[300]\tvalidation_0-merror:0.14696\tvalidation_1-merror:0.19457\n",
      "[301]\tvalidation_0-merror:0.14664\tvalidation_1-merror:0.19446\n",
      "[302]\tvalidation_0-merror:0.14648\tvalidation_1-merror:0.19473\n",
      "[303]\tvalidation_0-merror:0.14607\tvalidation_1-merror:0.19467\n",
      "[304]\tvalidation_0-merror:0.14587\tvalidation_1-merror:0.19446\n",
      "[305]\tvalidation_0-merror:0.14586\tvalidation_1-merror:0.19471\n",
      "[306]\tvalidation_0-merror:0.14575\tvalidation_1-merror:0.19477\n",
      "[307]\tvalidation_0-merror:0.14560\tvalidation_1-merror:0.19477\n",
      "[308]\tvalidation_0-merror:0.14530\tvalidation_1-merror:0.19442\n",
      "[309]\tvalidation_0-merror:0.14514\tvalidation_1-merror:0.19457\n",
      "[310]\tvalidation_0-merror:0.14500\tvalidation_1-merror:0.19444\n",
      "[311]\tvalidation_0-merror:0.14473\tvalidation_1-merror:0.19442\n",
      "[312]\tvalidation_0-merror:0.14454\tvalidation_1-merror:0.19426\n",
      "[313]\tvalidation_0-merror:0.14421\tvalidation_1-merror:0.19424\n",
      "[314]\tvalidation_0-merror:0.14396\tvalidation_1-merror:0.19397\n",
      "[315]\tvalidation_0-merror:0.14385\tvalidation_1-merror:0.19383\n",
      "[316]\tvalidation_0-merror:0.14369\tvalidation_1-merror:0.19377\n",
      "[317]\tvalidation_0-merror:0.14358\tvalidation_1-merror:0.19365\n",
      "[318]\tvalidation_0-merror:0.14345\tvalidation_1-merror:0.19373\n",
      "[319]\tvalidation_0-merror:0.14336\tvalidation_1-merror:0.19379\n",
      "[320]\tvalidation_0-merror:0.14326\tvalidation_1-merror:0.19371\n",
      "[321]\tvalidation_0-merror:0.14309\tvalidation_1-merror:0.19363\n",
      "[322]\tvalidation_0-merror:0.14302\tvalidation_1-merror:0.19369\n",
      "[323]\tvalidation_0-merror:0.14284\tvalidation_1-merror:0.19387\n",
      "[324]\tvalidation_0-merror:0.14280\tvalidation_1-merror:0.19387\n",
      "[325]\tvalidation_0-merror:0.14260\tvalidation_1-merror:0.19391\n",
      "[326]\tvalidation_0-merror:0.14248\tvalidation_1-merror:0.19395\n",
      "[327]\tvalidation_0-merror:0.14239\tvalidation_1-merror:0.19397\n",
      "[328]\tvalidation_0-merror:0.14227\tvalidation_1-merror:0.19391\n",
      "[329]\tvalidation_0-merror:0.14218\tvalidation_1-merror:0.19387\n",
      "[330]\tvalidation_0-merror:0.14207\tvalidation_1-merror:0.19371\n",
      "[331]\tvalidation_0-merror:0.14196\tvalidation_1-merror:0.19365\n",
      "[332]\tvalidation_0-merror:0.14190\tvalidation_1-merror:0.19361\n",
      "[333]\tvalidation_0-merror:0.14185\tvalidation_1-merror:0.19356\n",
      "[334]\tvalidation_0-merror:0.14144\tvalidation_1-merror:0.19328\n",
      "[335]\tvalidation_0-merror:0.14128\tvalidation_1-merror:0.19316\n",
      "[336]\tvalidation_0-merror:0.14100\tvalidation_1-merror:0.19334\n",
      "[337]\tvalidation_0-merror:0.14058\tvalidation_1-merror:0.19332\n",
      "[338]\tvalidation_0-merror:0.14038\tvalidation_1-merror:0.19338\n",
      "[339]\tvalidation_0-merror:0.14023\tvalidation_1-merror:0.19318\n",
      "[340]\tvalidation_0-merror:0.14008\tvalidation_1-merror:0.19303\n",
      "[341]\tvalidation_0-merror:0.13986\tvalidation_1-merror:0.19299\n",
      "[342]\tvalidation_0-merror:0.13973\tvalidation_1-merror:0.19309\n",
      "[343]\tvalidation_0-merror:0.13960\tvalidation_1-merror:0.19303\n",
      "[344]\tvalidation_0-merror:0.13944\tvalidation_1-merror:0.19303\n",
      "[345]\tvalidation_0-merror:0.13931\tvalidation_1-merror:0.19316\n",
      "[346]\tvalidation_0-merror:0.13925\tvalidation_1-merror:0.19312\n",
      "[347]\tvalidation_0-merror:0.13906\tvalidation_1-merror:0.19301\n",
      "[348]\tvalidation_0-merror:0.13888\tvalidation_1-merror:0.19281\n",
      "[349]\tvalidation_0-merror:0.13878\tvalidation_1-merror:0.19281\n",
      "[350]\tvalidation_0-merror:0.13865\tvalidation_1-merror:0.19271\n",
      "[351]\tvalidation_0-merror:0.13849\tvalidation_1-merror:0.19287\n",
      "[352]\tvalidation_0-merror:0.13828\tvalidation_1-merror:0.19281\n",
      "[353]\tvalidation_0-merror:0.13819\tvalidation_1-merror:0.19273\n",
      "[354]\tvalidation_0-merror:0.13802\tvalidation_1-merror:0.19271\n",
      "[355]\tvalidation_0-merror:0.13776\tvalidation_1-merror:0.19242\n",
      "[356]\tvalidation_0-merror:0.13756\tvalidation_1-merror:0.19250\n",
      "[357]\tvalidation_0-merror:0.13737\tvalidation_1-merror:0.19256\n",
      "[358]\tvalidation_0-merror:0.13719\tvalidation_1-merror:0.19258\n",
      "[359]\tvalidation_0-merror:0.13706\tvalidation_1-merror:0.19256\n",
      "[360]\tvalidation_0-merror:0.13697\tvalidation_1-merror:0.19260\n",
      "[361]\tvalidation_0-merror:0.13690\tvalidation_1-merror:0.19262\n",
      "[362]\tvalidation_0-merror:0.13674\tvalidation_1-merror:0.19260\n",
      "[363]\tvalidation_0-merror:0.13671\tvalidation_1-merror:0.19248\n",
      "[364]\tvalidation_0-merror:0.13669\tvalidation_1-merror:0.19244\n",
      "[365]\tvalidation_0-merror:0.13661\tvalidation_1-merror:0.19226\n",
      "[366]\tvalidation_0-merror:0.13649\tvalidation_1-merror:0.19242\n",
      "[367]\tvalidation_0-merror:0.13643\tvalidation_1-merror:0.19242\n",
      "[368]\tvalidation_0-merror:0.13623\tvalidation_1-merror:0.19242\n",
      "[369]\tvalidation_0-merror:0.13613\tvalidation_1-merror:0.19234\n",
      "[370]\tvalidation_0-merror:0.13598\tvalidation_1-merror:0.19230\n",
      "[371]\tvalidation_0-merror:0.13584\tvalidation_1-merror:0.19226\n",
      "[372]\tvalidation_0-merror:0.13567\tvalidation_1-merror:0.19209\n",
      "[373]\tvalidation_0-merror:0.13559\tvalidation_1-merror:0.19232\n",
      "[374]\tvalidation_0-merror:0.13540\tvalidation_1-merror:0.19258\n",
      "[375]\tvalidation_0-merror:0.13516\tvalidation_1-merror:0.19244\n",
      "[376]\tvalidation_0-merror:0.13504\tvalidation_1-merror:0.19246\n",
      "[377]\tvalidation_0-merror:0.13481\tvalidation_1-merror:0.19242\n",
      "[378]\tvalidation_0-merror:0.13470\tvalidation_1-merror:0.19211\n",
      "[379]\tvalidation_0-merror:0.13462\tvalidation_1-merror:0.19211\n",
      "[380]\tvalidation_0-merror:0.13443\tvalidation_1-merror:0.19211\n",
      "[381]\tvalidation_0-merror:0.13426\tvalidation_1-merror:0.19183\n",
      "[382]\tvalidation_0-merror:0.13412\tvalidation_1-merror:0.19174\n",
      "[383]\tvalidation_0-merror:0.13394\tvalidation_1-merror:0.19183\n",
      "[384]\tvalidation_0-merror:0.13382\tvalidation_1-merror:0.19179\n",
      "[385]\tvalidation_0-merror:0.13380\tvalidation_1-merror:0.19195\n",
      "[386]\tvalidation_0-merror:0.13373\tvalidation_1-merror:0.19197\n",
      "[387]\tvalidation_0-merror:0.13365\tvalidation_1-merror:0.19205\n",
      "[388]\tvalidation_0-merror:0.13346\tvalidation_1-merror:0.19197\n",
      "[389]\tvalidation_0-merror:0.13326\tvalidation_1-merror:0.19199\n",
      "[390]\tvalidation_0-merror:0.13301\tvalidation_1-merror:0.19176\n",
      "[391]\tvalidation_0-merror:0.13273\tvalidation_1-merror:0.19156\n",
      "[392]\tvalidation_0-merror:0.13256\tvalidation_1-merror:0.19187\n",
      "[393]\tvalidation_0-merror:0.13241\tvalidation_1-merror:0.19181\n",
      "[394]\tvalidation_0-merror:0.13232\tvalidation_1-merror:0.19176\n",
      "[395]\tvalidation_0-merror:0.13221\tvalidation_1-merror:0.19168\n",
      "[396]\tvalidation_0-merror:0.13208\tvalidation_1-merror:0.19187\n",
      "[397]\tvalidation_0-merror:0.13188\tvalidation_1-merror:0.19170\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[398]\tvalidation_0-merror:0.13173\tvalidation_1-merror:0.19176\n",
      "[399]\tvalidation_0-merror:0.13160\tvalidation_1-merror:0.19172\n",
      "[400]\tvalidation_0-merror:0.13151\tvalidation_1-merror:0.19160\n",
      "[401]\tvalidation_0-merror:0.13138\tvalidation_1-merror:0.19129\n",
      "[402]\tvalidation_0-merror:0.13128\tvalidation_1-merror:0.19111\n",
      "[403]\tvalidation_0-merror:0.13118\tvalidation_1-merror:0.19103\n",
      "[404]\tvalidation_0-merror:0.13113\tvalidation_1-merror:0.19109\n",
      "[405]\tvalidation_0-merror:0.13110\tvalidation_1-merror:0.19105\n",
      "[406]\tvalidation_0-merror:0.13092\tvalidation_1-merror:0.19101\n",
      "[407]\tvalidation_0-merror:0.13072\tvalidation_1-merror:0.19142\n",
      "[408]\tvalidation_0-merror:0.13067\tvalidation_1-merror:0.19146\n",
      "[409]\tvalidation_0-merror:0.13055\tvalidation_1-merror:0.19142\n",
      "[410]\tvalidation_0-merror:0.13045\tvalidation_1-merror:0.19148\n",
      "[411]\tvalidation_0-merror:0.13034\tvalidation_1-merror:0.19115\n",
      "[412]\tvalidation_0-merror:0.13021\tvalidation_1-merror:0.19091\n",
      "[413]\tvalidation_0-merror:0.13003\tvalidation_1-merror:0.19084\n",
      "[414]\tvalidation_0-merror:0.12992\tvalidation_1-merror:0.19105\n",
      "[415]\tvalidation_0-merror:0.12981\tvalidation_1-merror:0.19109\n",
      "[416]\tvalidation_0-merror:0.12966\tvalidation_1-merror:0.19119\n",
      "[417]\tvalidation_0-merror:0.12954\tvalidation_1-merror:0.19101\n",
      "[418]\tvalidation_0-merror:0.12942\tvalidation_1-merror:0.19099\n",
      "[419]\tvalidation_0-merror:0.12932\tvalidation_1-merror:0.19093\n",
      "[420]\tvalidation_0-merror:0.12914\tvalidation_1-merror:0.19088\n",
      "[421]\tvalidation_0-merror:0.12905\tvalidation_1-merror:0.19095\n",
      "[422]\tvalidation_0-merror:0.12877\tvalidation_1-merror:0.19095\n",
      "[423]\tvalidation_0-merror:0.12873\tvalidation_1-merror:0.19107\n",
      "[424]\tvalidation_0-merror:0.12862\tvalidation_1-merror:0.19111\n",
      "[425]\tvalidation_0-merror:0.12848\tvalidation_1-merror:0.19119\n",
      "[426]\tvalidation_0-merror:0.12837\tvalidation_1-merror:0.19125\n",
      "[427]\tvalidation_0-merror:0.12830\tvalidation_1-merror:0.19134\n",
      "[428]\tvalidation_0-merror:0.12809\tvalidation_1-merror:0.19132\n",
      "[429]\tvalidation_0-merror:0.12806\tvalidation_1-merror:0.19138\n",
      "[430]\tvalidation_0-merror:0.12787\tvalidation_1-merror:0.19144\n",
      "[431]\tvalidation_0-merror:0.12761\tvalidation_1-merror:0.19119\n",
      "[432]\tvalidation_0-merror:0.12739\tvalidation_1-merror:0.19119\n",
      "[433]\tvalidation_0-merror:0.12731\tvalidation_1-merror:0.19127\n",
      "[434]\tvalidation_0-merror:0.12720\tvalidation_1-merror:0.19119\n",
      "[435]\tvalidation_0-merror:0.12704\tvalidation_1-merror:0.19123\n",
      "[436]\tvalidation_0-merror:0.12685\tvalidation_1-merror:0.19136\n",
      "[437]\tvalidation_0-merror:0.12674\tvalidation_1-merror:0.19148\n",
      "[438]\tvalidation_0-merror:0.12667\tvalidation_1-merror:0.19134\n",
      "[439]\tvalidation_0-merror:0.12661\tvalidation_1-merror:0.19123\n",
      "[440]\tvalidation_0-merror:0.12650\tvalidation_1-merror:0.19125\n",
      "[441]\tvalidation_0-merror:0.12644\tvalidation_1-merror:0.19119\n",
      "[442]\tvalidation_0-merror:0.12625\tvalidation_1-merror:0.19105\n",
      "[443]\tvalidation_0-merror:0.12614\tvalidation_1-merror:0.19111\n",
      "[444]\tvalidation_0-merror:0.12602\tvalidation_1-merror:0.19119\n",
      "[445]\tvalidation_0-merror:0.12592\tvalidation_1-merror:0.19115\n",
      "[446]\tvalidation_0-merror:0.12586\tvalidation_1-merror:0.19107\n",
      "[447]\tvalidation_0-merror:0.12573\tvalidation_1-merror:0.19111\n",
      "[448]\tvalidation_0-merror:0.12561\tvalidation_1-merror:0.19111\n",
      "[449]\tvalidation_0-merror:0.12555\tvalidation_1-merror:0.19107\n",
      "[450]\tvalidation_0-merror:0.12545\tvalidation_1-merror:0.19103\n",
      "[451]\tvalidation_0-merror:0.12545\tvalidation_1-merror:0.19093\n",
      "[452]\tvalidation_0-merror:0.12527\tvalidation_1-merror:0.19113\n",
      "[453]\tvalidation_0-merror:0.12514\tvalidation_1-merror:0.19115\n",
      "[454]\tvalidation_0-merror:0.12499\tvalidation_1-merror:0.19099\n",
      "[455]\tvalidation_0-merror:0.12482\tvalidation_1-merror:0.19097\n",
      "[456]\tvalidation_0-merror:0.12469\tvalidation_1-merror:0.19084\n",
      "[457]\tvalidation_0-merror:0.12463\tvalidation_1-merror:0.19087\n",
      "[458]\tvalidation_0-merror:0.12455\tvalidation_1-merror:0.19078\n",
      "[459]\tvalidation_0-merror:0.12448\tvalidation_1-merror:0.19072\n",
      "[460]\tvalidation_0-merror:0.12445\tvalidation_1-merror:0.19074\n",
      "[461]\tvalidation_0-merror:0.12427\tvalidation_1-merror:0.19078\n",
      "[462]\tvalidation_0-merror:0.12414\tvalidation_1-merror:0.19068\n",
      "[463]\tvalidation_0-merror:0.12397\tvalidation_1-merror:0.19050\n",
      "[464]\tvalidation_0-merror:0.12383\tvalidation_1-merror:0.19060\n",
      "[465]\tvalidation_0-merror:0.12354\tvalidation_1-merror:0.19066\n",
      "[466]\tvalidation_0-merror:0.12353\tvalidation_1-merror:0.19068\n",
      "[467]\tvalidation_0-merror:0.12345\tvalidation_1-merror:0.19060\n",
      "[468]\tvalidation_0-merror:0.12336\tvalidation_1-merror:0.19050\n",
      "[469]\tvalidation_0-merror:0.12330\tvalidation_1-merror:0.19052\n",
      "[470]\tvalidation_0-merror:0.12320\tvalidation_1-merror:0.19054\n",
      "[471]\tvalidation_0-merror:0.12313\tvalidation_1-merror:0.19066\n",
      "[472]\tvalidation_0-merror:0.12304\tvalidation_1-merror:0.19080\n",
      "[473]\tvalidation_0-merror:0.12295\tvalidation_1-merror:0.19078\n",
      "[474]\tvalidation_0-merror:0.12285\tvalidation_1-merror:0.19091\n",
      "[475]\tvalidation_0-merror:0.12270\tvalidation_1-merror:0.19097\n",
      "[476]\tvalidation_0-merror:0.12257\tvalidation_1-merror:0.19093\n",
      "[477]\tvalidation_0-merror:0.12246\tvalidation_1-merror:0.19087\n",
      "[478]\tvalidation_0-merror:0.12234\tvalidation_1-merror:0.19078\n",
      "[479]\tvalidation_0-merror:0.12225\tvalidation_1-merror:0.19084\n",
      "[480]\tvalidation_0-merror:0.12210\tvalidation_1-merror:0.19082\n",
      "[481]\tvalidation_0-merror:0.12203\tvalidation_1-merror:0.19091\n",
      "[482]\tvalidation_0-merror:0.12192\tvalidation_1-merror:0.19093\n",
      "[483]\tvalidation_0-merror:0.12179\tvalidation_1-merror:0.19078\n",
      "[484]\tvalidation_0-merror:0.12159\tvalidation_1-merror:0.19078\n",
      "[485]\tvalidation_0-merror:0.12138\tvalidation_1-merror:0.19076\n",
      "[486]\tvalidation_0-merror:0.12134\tvalidation_1-merror:0.19082\n",
      "[487]\tvalidation_0-merror:0.12118\tvalidation_1-merror:0.19087\n",
      "[488]\tvalidation_0-merror:0.12103\tvalidation_1-merror:0.19082\n",
      "[489]\tvalidation_0-merror:0.12080\tvalidation_1-merror:0.19066\n",
      "[490]\tvalidation_0-merror:0.12061\tvalidation_1-merror:0.19070\n",
      "[491]\tvalidation_0-merror:0.12053\tvalidation_1-merror:0.19078\n",
      "[492]\tvalidation_0-merror:0.12042\tvalidation_1-merror:0.19072\n",
      "[493]\tvalidation_0-merror:0.12031\tvalidation_1-merror:0.19068\n",
      "[494]\tvalidation_0-merror:0.12020\tvalidation_1-merror:0.19054\n",
      "[495]\tvalidation_0-merror:0.12016\tvalidation_1-merror:0.19050\n",
      "[496]\tvalidation_0-merror:0.12007\tvalidation_1-merror:0.19046\n",
      "[497]\tvalidation_0-merror:0.11996\tvalidation_1-merror:0.19039\n",
      "[498]\tvalidation_0-merror:0.11992\tvalidation_1-merror:0.19035\n",
      "[499]\tvalidation_0-merror:0.11986\tvalidation_1-merror:0.19050\n",
      "[500]\tvalidation_0-merror:0.11968\tvalidation_1-merror:0.19064\n",
      "[501]\tvalidation_0-merror:0.11958\tvalidation_1-merror:0.19070\n",
      "[502]\tvalidation_0-merror:0.11947\tvalidation_1-merror:0.19060\n",
      "[503]\tvalidation_0-merror:0.11946\tvalidation_1-merror:0.19070\n",
      "[504]\tvalidation_0-merror:0.11927\tvalidation_1-merror:0.19070\n",
      "[505]\tvalidation_0-merror:0.11916\tvalidation_1-merror:0.19062\n",
      "[506]\tvalidation_0-merror:0.11906\tvalidation_1-merror:0.19054\n",
      "[507]\tvalidation_0-merror:0.11898\tvalidation_1-merror:0.19035\n",
      "[508]\tvalidation_0-merror:0.11886\tvalidation_1-merror:0.19039\n",
      "[509]\tvalidation_0-merror:0.11884\tvalidation_1-merror:0.19023\n",
      "[510]\tvalidation_0-merror:0.11868\tvalidation_1-merror:0.19021\n",
      "[511]\tvalidation_0-merror:0.11861\tvalidation_1-merror:0.19025\n",
      "[512]\tvalidation_0-merror:0.11848\tvalidation_1-merror:0.19042\n",
      "[513]\tvalidation_0-merror:0.11835\tvalidation_1-merror:0.19048\n",
      "[514]\tvalidation_0-merror:0.11824\tvalidation_1-merror:0.19035\n",
      "[515]\tvalidation_0-merror:0.11812\tvalidation_1-merror:0.19037\n",
      "[516]\tvalidation_0-merror:0.11793\tvalidation_1-merror:0.19046\n",
      "[517]\tvalidation_0-merror:0.11774\tvalidation_1-merror:0.19052\n",
      "[518]\tvalidation_0-merror:0.11758\tvalidation_1-merror:0.19021\n",
      "[519]\tvalidation_0-merror:0.11745\tvalidation_1-merror:0.19023\n",
      "[520]\tvalidation_0-merror:0.11728\tvalidation_1-merror:0.19023\n",
      "[521]\tvalidation_0-merror:0.11719\tvalidation_1-merror:0.19027\n",
      "[522]\tvalidation_0-merror:0.11704\tvalidation_1-merror:0.19031\n",
      "[523]\tvalidation_0-merror:0.11688\tvalidation_1-merror:0.19013\n",
      "[524]\tvalidation_0-merror:0.11674\tvalidation_1-merror:0.19005\n",
      "[525]\tvalidation_0-merror:0.11661\tvalidation_1-merror:0.18994\n",
      "[526]\tvalidation_0-merror:0.11649\tvalidation_1-merror:0.19015\n",
      "[527]\tvalidation_0-merror:0.11633\tvalidation_1-merror:0.19011\n",
      "[528]\tvalidation_0-merror:0.11620\tvalidation_1-merror:0.18974\n",
      "[529]\tvalidation_0-merror:0.11616\tvalidation_1-merror:0.18972\n",
      "[530]\tvalidation_0-merror:0.11600\tvalidation_1-merror:0.18972\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[531]\tvalidation_0-merror:0.11595\tvalidation_1-merror:0.18978\n",
      "[532]\tvalidation_0-merror:0.11584\tvalidation_1-merror:0.18966\n",
      "[533]\tvalidation_0-merror:0.11566\tvalidation_1-merror:0.18958\n",
      "[534]\tvalidation_0-merror:0.11551\tvalidation_1-merror:0.18931\n",
      "[535]\tvalidation_0-merror:0.11543\tvalidation_1-merror:0.18935\n",
      "[536]\tvalidation_0-merror:0.11516\tvalidation_1-merror:0.18931\n",
      "[537]\tvalidation_0-merror:0.11507\tvalidation_1-merror:0.18933\n",
      "[538]\tvalidation_0-merror:0.11490\tvalidation_1-merror:0.18937\n",
      "[539]\tvalidation_0-merror:0.11480\tvalidation_1-merror:0.18929\n",
      "[540]\tvalidation_0-merror:0.11475\tvalidation_1-merror:0.18925\n",
      "[541]\tvalidation_0-merror:0.11463\tvalidation_1-merror:0.18949\n",
      "[542]\tvalidation_0-merror:0.11461\tvalidation_1-merror:0.18949\n",
      "[543]\tvalidation_0-merror:0.11451\tvalidation_1-merror:0.18951\n",
      "[544]\tvalidation_0-merror:0.11438\tvalidation_1-merror:0.18939\n",
      "[545]\tvalidation_0-merror:0.11428\tvalidation_1-merror:0.18937\n",
      "[546]\tvalidation_0-merror:0.11411\tvalidation_1-merror:0.18945\n",
      "[547]\tvalidation_0-merror:0.11402\tvalidation_1-merror:0.18943\n",
      "[548]\tvalidation_0-merror:0.11392\tvalidation_1-merror:0.18949\n",
      "[549]\tvalidation_0-merror:0.11384\tvalidation_1-merror:0.18958\n",
      "[550]\tvalidation_0-merror:0.11370\tvalidation_1-merror:0.18958\n",
      "[551]\tvalidation_0-merror:0.11361\tvalidation_1-merror:0.18941\n",
      "[552]\tvalidation_0-merror:0.11347\tvalidation_1-merror:0.18943\n",
      "[553]\tvalidation_0-merror:0.11336\tvalidation_1-merror:0.18960\n",
      "[554]\tvalidation_0-merror:0.11329\tvalidation_1-merror:0.18951\n",
      "[555]\tvalidation_0-merror:0.11317\tvalidation_1-merror:0.18968\n",
      "[556]\tvalidation_0-merror:0.11310\tvalidation_1-merror:0.18964\n",
      "[557]\tvalidation_0-merror:0.11305\tvalidation_1-merror:0.18972\n",
      "[558]\tvalidation_0-merror:0.11303\tvalidation_1-merror:0.18964\n",
      "[559]\tvalidation_0-merror:0.11296\tvalidation_1-merror:0.18954\n",
      "[560]\tvalidation_0-merror:0.11288\tvalidation_1-merror:0.18951\n",
      "[561]\tvalidation_0-merror:0.11280\tvalidation_1-merror:0.18954\n",
      "[562]\tvalidation_0-merror:0.11275\tvalidation_1-merror:0.18958\n",
      "[563]\tvalidation_0-merror:0.11262\tvalidation_1-merror:0.18968\n",
      "[564]\tvalidation_0-merror:0.11249\tvalidation_1-merror:0.18960\n",
      "[565]\tvalidation_0-merror:0.11245\tvalidation_1-merror:0.18970\n",
      "[566]\tvalidation_0-merror:0.11241\tvalidation_1-merror:0.18972\n",
      "[567]\tvalidation_0-merror:0.11235\tvalidation_1-merror:0.18970\n",
      "[568]\tvalidation_0-merror:0.11216\tvalidation_1-merror:0.18956\n",
      "[569]\tvalidation_0-merror:0.11201\tvalidation_1-merror:0.18960\n",
      "[570]\tvalidation_0-merror:0.11196\tvalidation_1-merror:0.18958\n",
      "[571]\tvalidation_0-merror:0.11189\tvalidation_1-merror:0.18956\n",
      "[572]\tvalidation_0-merror:0.11178\tvalidation_1-merror:0.18956\n",
      "[573]\tvalidation_0-merror:0.11171\tvalidation_1-merror:0.18951\n",
      "[574]\tvalidation_0-merror:0.11164\tvalidation_1-merror:0.18956\n",
      "[575]\tvalidation_0-merror:0.11155\tvalidation_1-merror:0.18970\n",
      "[576]\tvalidation_0-merror:0.11144\tvalidation_1-merror:0.18990\n",
      "[577]\tvalidation_0-merror:0.11133\tvalidation_1-merror:0.18992\n",
      "[578]\tvalidation_0-merror:0.11123\tvalidation_1-merror:0.18984\n",
      "[579]\tvalidation_0-merror:0.11112\tvalidation_1-merror:0.18984\n",
      "[580]\tvalidation_0-merror:0.11097\tvalidation_1-merror:0.18980\n",
      "[581]\tvalidation_0-merror:0.11092\tvalidation_1-merror:0.18978\n",
      "[582]\tvalidation_0-merror:0.11081\tvalidation_1-merror:0.18972\n",
      "[583]\tvalidation_0-merror:0.11075\tvalidation_1-merror:0.18960\n",
      "[584]\tvalidation_0-merror:0.11067\tvalidation_1-merror:0.18949\n",
      "[585]\tvalidation_0-merror:0.11058\tvalidation_1-merror:0.18937\n",
      "[586]\tvalidation_0-merror:0.11050\tvalidation_1-merror:0.18919\n",
      "[587]\tvalidation_0-merror:0.11045\tvalidation_1-merror:0.18919\n",
      "[588]\tvalidation_0-merror:0.11041\tvalidation_1-merror:0.18917\n",
      "[589]\tvalidation_0-merror:0.11027\tvalidation_1-merror:0.18919\n",
      "[590]\tvalidation_0-merror:0.11014\tvalidation_1-merror:0.18913\n",
      "[591]\tvalidation_0-merror:0.11007\tvalidation_1-merror:0.18927\n",
      "[592]\tvalidation_0-merror:0.11000\tvalidation_1-merror:0.18915\n",
      "[593]\tvalidation_0-merror:0.10989\tvalidation_1-merror:0.18913\n",
      "[594]\tvalidation_0-merror:0.10978\tvalidation_1-merror:0.18913\n",
      "[595]\tvalidation_0-merror:0.10972\tvalidation_1-merror:0.18900\n",
      "[596]\tvalidation_0-merror:0.10968\tvalidation_1-merror:0.18898\n",
      "[597]\tvalidation_0-merror:0.10963\tvalidation_1-merror:0.18907\n",
      "[598]\tvalidation_0-merror:0.10959\tvalidation_1-merror:0.18904\n",
      "[599]\tvalidation_0-merror:0.10944\tvalidation_1-merror:0.18892\n",
      "[600]\tvalidation_0-merror:0.10935\tvalidation_1-merror:0.18878\n",
      "[601]\tvalidation_0-merror:0.10926\tvalidation_1-merror:0.18884\n",
      "[602]\tvalidation_0-merror:0.10923\tvalidation_1-merror:0.18900\n",
      "[603]\tvalidation_0-merror:0.10917\tvalidation_1-merror:0.18888\n",
      "[604]\tvalidation_0-merror:0.10909\tvalidation_1-merror:0.18874\n",
      "[605]\tvalidation_0-merror:0.10904\tvalidation_1-merror:0.18878\n",
      "[606]\tvalidation_0-merror:0.10895\tvalidation_1-merror:0.18872\n",
      "[607]\tvalidation_0-merror:0.10889\tvalidation_1-merror:0.18868\n",
      "[608]\tvalidation_0-merror:0.10881\tvalidation_1-merror:0.18870\n",
      "[609]\tvalidation_0-merror:0.10870\tvalidation_1-merror:0.18862\n",
      "[610]\tvalidation_0-merror:0.10861\tvalidation_1-merror:0.18872\n",
      "[611]\tvalidation_0-merror:0.10844\tvalidation_1-merror:0.18880\n",
      "[612]\tvalidation_0-merror:0.10827\tvalidation_1-merror:0.18862\n",
      "[613]\tvalidation_0-merror:0.10820\tvalidation_1-merror:0.18868\n",
      "[614]\tvalidation_0-merror:0.10806\tvalidation_1-merror:0.18859\n",
      "[615]\tvalidation_0-merror:0.10799\tvalidation_1-merror:0.18859\n",
      "[616]\tvalidation_0-merror:0.10788\tvalidation_1-merror:0.18857\n",
      "[617]\tvalidation_0-merror:0.10775\tvalidation_1-merror:0.18849\n",
      "[618]\tvalidation_0-merror:0.10766\tvalidation_1-merror:0.18837\n",
      "[619]\tvalidation_0-merror:0.10752\tvalidation_1-merror:0.18855\n",
      "[620]\tvalidation_0-merror:0.10745\tvalidation_1-merror:0.18857\n",
      "[621]\tvalidation_0-merror:0.10736\tvalidation_1-merror:0.18845\n",
      "[622]\tvalidation_0-merror:0.10726\tvalidation_1-merror:0.18831\n",
      "[623]\tvalidation_0-merror:0.10710\tvalidation_1-merror:0.18827\n",
      "[624]\tvalidation_0-merror:0.10702\tvalidation_1-merror:0.18833\n",
      "[625]\tvalidation_0-merror:0.10698\tvalidation_1-merror:0.18833\n",
      "[626]\tvalidation_0-merror:0.10694\tvalidation_1-merror:0.18833\n",
      "[627]\tvalidation_0-merror:0.10687\tvalidation_1-merror:0.18837\n",
      "[628]\tvalidation_0-merror:0.10679\tvalidation_1-merror:0.18841\n",
      "[629]\tvalidation_0-merror:0.10670\tvalidation_1-merror:0.18839\n",
      "[630]\tvalidation_0-merror:0.10666\tvalidation_1-merror:0.18843\n",
      "[631]\tvalidation_0-merror:0.10664\tvalidation_1-merror:0.18843\n",
      "[632]\tvalidation_0-merror:0.10651\tvalidation_1-merror:0.18847\n",
      "[633]\tvalidation_0-merror:0.10644\tvalidation_1-merror:0.18859\n",
      "[634]\tvalidation_0-merror:0.10637\tvalidation_1-merror:0.18853\n",
      "[635]\tvalidation_0-merror:0.10627\tvalidation_1-merror:0.18870\n",
      "[636]\tvalidation_0-merror:0.10616\tvalidation_1-merror:0.18874\n",
      "[637]\tvalidation_0-merror:0.10608\tvalidation_1-merror:0.18863\n",
      "[638]\tvalidation_0-merror:0.10598\tvalidation_1-merror:0.18862\n",
      "[639]\tvalidation_0-merror:0.10588\tvalidation_1-merror:0.18866\n",
      "[640]\tvalidation_0-merror:0.10577\tvalidation_1-merror:0.18863\n",
      "[641]\tvalidation_0-merror:0.10564\tvalidation_1-merror:0.18870\n",
      "[642]\tvalidation_0-merror:0.10552\tvalidation_1-merror:0.18862\n",
      "[643]\tvalidation_0-merror:0.10538\tvalidation_1-merror:0.18872\n",
      "[644]\tvalidation_0-merror:0.10524\tvalidation_1-merror:0.18872\n",
      "[645]\tvalidation_0-merror:0.10517\tvalidation_1-merror:0.18872\n",
      "[646]\tvalidation_0-merror:0.10512\tvalidation_1-merror:0.18870\n",
      "[647]\tvalidation_0-merror:0.10508\tvalidation_1-merror:0.18872\n",
      "[648]\tvalidation_0-merror:0.10496\tvalidation_1-merror:0.18872\n",
      "[649]\tvalidation_0-merror:0.10485\tvalidation_1-merror:0.18872\n",
      "[650]\tvalidation_0-merror:0.10474\tvalidation_1-merror:0.18872\n",
      "[651]\tvalidation_0-merror:0.10467\tvalidation_1-merror:0.18884\n",
      "[652]\tvalidation_0-merror:0.10454\tvalidation_1-merror:0.18874\n",
      "[653]\tvalidation_0-merror:0.10452\tvalidation_1-merror:0.18878\n",
      "[654]\tvalidation_0-merror:0.10445\tvalidation_1-merror:0.18876\n",
      "[655]\tvalidation_0-merror:0.10441\tvalidation_1-merror:0.18892\n",
      "[656]\tvalidation_0-merror:0.10430\tvalidation_1-merror:0.18896\n",
      "[657]\tvalidation_0-merror:0.10418\tvalidation_1-merror:0.18890\n",
      "[658]\tvalidation_0-merror:0.10411\tvalidation_1-merror:0.18898\n",
      "[659]\tvalidation_0-merror:0.10400\tvalidation_1-merror:0.18902\n",
      "[660]\tvalidation_0-merror:0.10386\tvalidation_1-merror:0.18888\n",
      "[661]\tvalidation_0-merror:0.10378\tvalidation_1-merror:0.18898\n",
      "[662]\tvalidation_0-merror:0.10367\tvalidation_1-merror:0.18896\n",
      "[663]\tvalidation_0-merror:0.10353\tvalidation_1-merror:0.18904\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[664]\tvalidation_0-merror:0.10344\tvalidation_1-merror:0.18904\n",
      "[665]\tvalidation_0-merror:0.10339\tvalidation_1-merror:0.18913\n",
      "[666]\tvalidation_0-merror:0.10331\tvalidation_1-merror:0.18919\n",
      "[667]\tvalidation_0-merror:0.10320\tvalidation_1-merror:0.18904\n",
      "[668]\tvalidation_0-merror:0.10314\tvalidation_1-merror:0.18892\n",
      "[669]\tvalidation_0-merror:0.10309\tvalidation_1-merror:0.18896\n",
      "[670]\tvalidation_0-merror:0.10305\tvalidation_1-merror:0.18894\n",
      "[671]\tvalidation_0-merror:0.10290\tvalidation_1-merror:0.18868\n",
      "[672]\tvalidation_0-merror:0.10270\tvalidation_1-merror:0.18862\n",
      "[673]\tvalidation_0-merror:0.10266\tvalidation_1-merror:0.18857\n",
      "[674]\tvalidation_0-merror:0.10257\tvalidation_1-merror:0.18851\n",
      "[675]\tvalidation_0-merror:0.10250\tvalidation_1-merror:0.18837\n",
      "[676]\tvalidation_0-merror:0.10238\tvalidation_1-merror:0.18845\n",
      "[677]\tvalidation_0-merror:0.10228\tvalidation_1-merror:0.18857\n",
      "[678]\tvalidation_0-merror:0.10205\tvalidation_1-merror:0.18849\n",
      "[679]\tvalidation_0-merror:0.10195\tvalidation_1-merror:0.18839\n",
      "[680]\tvalidation_0-merror:0.10192\tvalidation_1-merror:0.18837\n",
      "[681]\tvalidation_0-merror:0.10180\tvalidation_1-merror:0.18831\n",
      "[682]\tvalidation_0-merror:0.10174\tvalidation_1-merror:0.18831\n",
      "[683]\tvalidation_0-merror:0.10163\tvalidation_1-merror:0.18812\n",
      "[684]\tvalidation_0-merror:0.10154\tvalidation_1-merror:0.18835\n",
      "[685]\tvalidation_0-merror:0.10152\tvalidation_1-merror:0.18833\n",
      "[686]\tvalidation_0-merror:0.10137\tvalidation_1-merror:0.18833\n",
      "[687]\tvalidation_0-merror:0.10126\tvalidation_1-merror:0.18833\n",
      "[688]\tvalidation_0-merror:0.10115\tvalidation_1-merror:0.18841\n",
      "[689]\tvalidation_0-merror:0.10111\tvalidation_1-merror:0.18857\n",
      "[690]\tvalidation_0-merror:0.10093\tvalidation_1-merror:0.18857\n",
      "[691]\tvalidation_0-merror:0.10085\tvalidation_1-merror:0.18859\n",
      "[692]\tvalidation_0-merror:0.10075\tvalidation_1-merror:0.18868\n",
      "[693]\tvalidation_0-merror:0.10071\tvalidation_1-merror:0.18859\n",
      "[694]\tvalidation_0-merror:0.10059\tvalidation_1-merror:0.18847\n",
      "[695]\tvalidation_0-merror:0.10051\tvalidation_1-merror:0.18851\n",
      "[696]\tvalidation_0-merror:0.10043\tvalidation_1-merror:0.18857\n",
      "[697]\tvalidation_0-merror:0.10034\tvalidation_1-merror:0.18849\n",
      "[698]\tvalidation_0-merror:0.10029\tvalidation_1-merror:0.18849\n",
      "[699]\tvalidation_0-merror:0.10017\tvalidation_1-merror:0.18847\n",
      "[700]\tvalidation_0-merror:0.10007\tvalidation_1-merror:0.18849\n",
      "[701]\tvalidation_0-merror:0.09994\tvalidation_1-merror:0.18839\n",
      "[702]\tvalidation_0-merror:0.09979\tvalidation_1-merror:0.18841\n",
      "[703]\tvalidation_0-merror:0.09971\tvalidation_1-merror:0.18851\n",
      "[704]\tvalidation_0-merror:0.09956\tvalidation_1-merror:0.18855\n",
      "[705]\tvalidation_0-merror:0.09950\tvalidation_1-merror:0.18857\n",
      "[706]\tvalidation_0-merror:0.09935\tvalidation_1-merror:0.18853\n",
      "[707]\tvalidation_0-merror:0.09929\tvalidation_1-merror:0.18849\n",
      "[708]\tvalidation_0-merror:0.09923\tvalidation_1-merror:0.18839\n",
      "[709]\tvalidation_0-merror:0.09915\tvalidation_1-merror:0.18843\n",
      "[710]\tvalidation_0-merror:0.09910\tvalidation_1-merror:0.18851\n",
      "[711]\tvalidation_0-merror:0.09905\tvalidation_1-merror:0.18855\n",
      "[712]\tvalidation_0-merror:0.09897\tvalidation_1-merror:0.18855\n",
      "[713]\tvalidation_0-merror:0.09887\tvalidation_1-merror:0.18863\n",
      "[714]\tvalidation_0-merror:0.09879\tvalidation_1-merror:0.18859\n",
      "[715]\tvalidation_0-merror:0.09866\tvalidation_1-merror:0.18862\n",
      "[716]\tvalidation_0-merror:0.09855\tvalidation_1-merror:0.18853\n",
      "[717]\tvalidation_0-merror:0.09848\tvalidation_1-merror:0.18853\n",
      "[718]\tvalidation_0-merror:0.09838\tvalidation_1-merror:0.18843\n",
      "[719]\tvalidation_0-merror:0.09835\tvalidation_1-merror:0.18845\n",
      "[720]\tvalidation_0-merror:0.09821\tvalidation_1-merror:0.18839\n",
      "[721]\tvalidation_0-merror:0.09811\tvalidation_1-merror:0.18827\n",
      "[722]\tvalidation_0-merror:0.09804\tvalidation_1-merror:0.18831\n",
      "[723]\tvalidation_0-merror:0.09796\tvalidation_1-merror:0.18829\n",
      "[724]\tvalidation_0-merror:0.09779\tvalidation_1-merror:0.18818\n",
      "[725]\tvalidation_0-merror:0.09771\tvalidation_1-merror:0.18823\n",
      "[726]\tvalidation_0-merror:0.09763\tvalidation_1-merror:0.18814\n",
      "[727]\tvalidation_0-merror:0.09748\tvalidation_1-merror:0.18804\n",
      "[728]\tvalidation_0-merror:0.09742\tvalidation_1-merror:0.18808\n",
      "[729]\tvalidation_0-merror:0.09732\tvalidation_1-merror:0.18827\n",
      "[730]\tvalidation_0-merror:0.09720\tvalidation_1-merror:0.18827\n",
      "[731]\tvalidation_0-merror:0.09710\tvalidation_1-merror:0.18835\n",
      "[732]\tvalidation_0-merror:0.09705\tvalidation_1-merror:0.18833\n",
      "[733]\tvalidation_0-merror:0.09697\tvalidation_1-merror:0.18829\n",
      "[734]\tvalidation_0-merror:0.09688\tvalidation_1-merror:0.18827\n",
      "[735]\tvalidation_0-merror:0.09680\tvalidation_1-merror:0.18825\n",
      "[736]\tvalidation_0-merror:0.09673\tvalidation_1-merror:0.18829\n",
      "[737]\tvalidation_0-merror:0.09668\tvalidation_1-merror:0.18829\n",
      "[738]\tvalidation_0-merror:0.09661\tvalidation_1-merror:0.18821\n",
      "[739]\tvalidation_0-merror:0.09652\tvalidation_1-merror:0.18829\n",
      "[740]\tvalidation_0-merror:0.09645\tvalidation_1-merror:0.18831\n",
      "[741]\tvalidation_0-merror:0.09633\tvalidation_1-merror:0.18837\n",
      "[742]\tvalidation_0-merror:0.09623\tvalidation_1-merror:0.18829\n",
      "[743]\tvalidation_0-merror:0.09611\tvalidation_1-merror:0.18816\n",
      "[744]\tvalidation_0-merror:0.09607\tvalidation_1-merror:0.18816\n",
      "[745]\tvalidation_0-merror:0.09600\tvalidation_1-merror:0.18812\n",
      "[746]\tvalidation_0-merror:0.09593\tvalidation_1-merror:0.18812\n",
      "[747]\tvalidation_0-merror:0.09577\tvalidation_1-merror:0.18812\n",
      "[748]\tvalidation_0-merror:0.09570\tvalidation_1-merror:0.18808\n",
      "[749]\tvalidation_0-merror:0.09567\tvalidation_1-merror:0.18800\n",
      "[750]\tvalidation_0-merror:0.09557\tvalidation_1-merror:0.18821\n",
      "[751]\tvalidation_0-merror:0.09540\tvalidation_1-merror:0.18816\n",
      "[752]\tvalidation_0-merror:0.09534\tvalidation_1-merror:0.18816\n",
      "[753]\tvalidation_0-merror:0.09531\tvalidation_1-merror:0.18800\n",
      "[754]\tvalidation_0-merror:0.09527\tvalidation_1-merror:0.18784\n",
      "[755]\tvalidation_0-merror:0.09521\tvalidation_1-merror:0.18778\n",
      "[756]\tvalidation_0-merror:0.09516\tvalidation_1-merror:0.18792\n",
      "[757]\tvalidation_0-merror:0.09504\tvalidation_1-merror:0.18806\n",
      "[758]\tvalidation_0-merror:0.09492\tvalidation_1-merror:0.18792\n",
      "[759]\tvalidation_0-merror:0.09483\tvalidation_1-merror:0.18792\n",
      "[760]\tvalidation_0-merror:0.09467\tvalidation_1-merror:0.18788\n",
      "[761]\tvalidation_0-merror:0.09458\tvalidation_1-merror:0.18782\n",
      "[762]\tvalidation_0-merror:0.09449\tvalidation_1-merror:0.18776\n",
      "[763]\tvalidation_0-merror:0.09442\tvalidation_1-merror:0.18774\n",
      "[764]\tvalidation_0-merror:0.09436\tvalidation_1-merror:0.18776\n",
      "[765]\tvalidation_0-merror:0.09429\tvalidation_1-merror:0.18765\n",
      "[766]\tvalidation_0-merror:0.09419\tvalidation_1-merror:0.18767\n",
      "[767]\tvalidation_0-merror:0.09410\tvalidation_1-merror:0.18774\n",
      "[768]\tvalidation_0-merror:0.09405\tvalidation_1-merror:0.18776\n",
      "[769]\tvalidation_0-merror:0.09404\tvalidation_1-merror:0.18765\n",
      "[770]\tvalidation_0-merror:0.09394\tvalidation_1-merror:0.18751\n",
      "[771]\tvalidation_0-merror:0.09385\tvalidation_1-merror:0.18763\n",
      "[772]\tvalidation_0-merror:0.09376\tvalidation_1-merror:0.18765\n",
      "[773]\tvalidation_0-merror:0.09372\tvalidation_1-merror:0.18753\n",
      "[774]\tvalidation_0-merror:0.09362\tvalidation_1-merror:0.18735\n",
      "[775]\tvalidation_0-merror:0.09350\tvalidation_1-merror:0.18733\n",
      "[776]\tvalidation_0-merror:0.09343\tvalidation_1-merror:0.18731\n",
      "[777]\tvalidation_0-merror:0.09334\tvalidation_1-merror:0.18745\n",
      "[778]\tvalidation_0-merror:0.09326\tvalidation_1-merror:0.18753\n",
      "[779]\tvalidation_0-merror:0.09319\tvalidation_1-merror:0.18751\n",
      "[780]\tvalidation_0-merror:0.09315\tvalidation_1-merror:0.18753\n",
      "[781]\tvalidation_0-merror:0.09308\tvalidation_1-merror:0.18749\n",
      "[782]\tvalidation_0-merror:0.09298\tvalidation_1-merror:0.18751\n",
      "[783]\tvalidation_0-merror:0.09292\tvalidation_1-merror:0.18739\n",
      "[784]\tvalidation_0-merror:0.09280\tvalidation_1-merror:0.18755\n",
      "[785]\tvalidation_0-merror:0.09275\tvalidation_1-merror:0.18757\n",
      "[786]\tvalidation_0-merror:0.09277\tvalidation_1-merror:0.18729\n",
      "[787]\tvalidation_0-merror:0.09273\tvalidation_1-merror:0.18733\n",
      "[788]\tvalidation_0-merror:0.09267\tvalidation_1-merror:0.18726\n",
      "[789]\tvalidation_0-merror:0.09261\tvalidation_1-merror:0.18729\n",
      "[790]\tvalidation_0-merror:0.09253\tvalidation_1-merror:0.18722\n",
      "[791]\tvalidation_0-merror:0.09248\tvalidation_1-merror:0.18716\n",
      "[792]\tvalidation_0-merror:0.09239\tvalidation_1-merror:0.18718\n",
      "[793]\tvalidation_0-merror:0.09237\tvalidation_1-merror:0.18726\n",
      "[794]\tvalidation_0-merror:0.09224\tvalidation_1-merror:0.18716\n",
      "[795]\tvalidation_0-merror:0.09216\tvalidation_1-merror:0.18710\n",
      "[796]\tvalidation_0-merror:0.09207\tvalidation_1-merror:0.18724\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[797]\tvalidation_0-merror:0.09201\tvalidation_1-merror:0.18716\n",
      "[798]\tvalidation_0-merror:0.09199\tvalidation_1-merror:0.18716\n",
      "[799]\tvalidation_0-merror:0.09191\tvalidation_1-merror:0.18716\n",
      "[800]\tvalidation_0-merror:0.09180\tvalidation_1-merror:0.18716\n",
      "[801]\tvalidation_0-merror:0.09170\tvalidation_1-merror:0.18718\n",
      "[802]\tvalidation_0-merror:0.09165\tvalidation_1-merror:0.18710\n",
      "[803]\tvalidation_0-merror:0.09159\tvalidation_1-merror:0.18698\n",
      "[804]\tvalidation_0-merror:0.09144\tvalidation_1-merror:0.18690\n",
      "[805]\tvalidation_0-merror:0.09139\tvalidation_1-merror:0.18692\n",
      "[806]\tvalidation_0-merror:0.09127\tvalidation_1-merror:0.18694\n",
      "[807]\tvalidation_0-merror:0.09112\tvalidation_1-merror:0.18671\n",
      "[808]\tvalidation_0-merror:0.09101\tvalidation_1-merror:0.18671\n",
      "[809]\tvalidation_0-merror:0.09091\tvalidation_1-merror:0.18665\n",
      "[810]\tvalidation_0-merror:0.09086\tvalidation_1-merror:0.18659\n",
      "[811]\tvalidation_0-merror:0.09080\tvalidation_1-merror:0.18653\n",
      "[812]\tvalidation_0-merror:0.09067\tvalidation_1-merror:0.18647\n",
      "[813]\tvalidation_0-merror:0.09057\tvalidation_1-merror:0.18653\n",
      "[814]\tvalidation_0-merror:0.09046\tvalidation_1-merror:0.18643\n",
      "[815]\tvalidation_0-merror:0.09038\tvalidation_1-merror:0.18653\n",
      "[816]\tvalidation_0-merror:0.09027\tvalidation_1-merror:0.18661\n",
      "[817]\tvalidation_0-merror:0.09014\tvalidation_1-merror:0.18673\n",
      "[818]\tvalidation_0-merror:0.09006\tvalidation_1-merror:0.18675\n",
      "[819]\tvalidation_0-merror:0.08991\tvalidation_1-merror:0.18684\n",
      "[820]\tvalidation_0-merror:0.08982\tvalidation_1-merror:0.18677\n",
      "[821]\tvalidation_0-merror:0.08967\tvalidation_1-merror:0.18677\n",
      "[822]\tvalidation_0-merror:0.08960\tvalidation_1-merror:0.18673\n",
      "[823]\tvalidation_0-merror:0.08952\tvalidation_1-merror:0.18677\n",
      "[824]\tvalidation_0-merror:0.08946\tvalidation_1-merror:0.18675\n",
      "[825]\tvalidation_0-merror:0.08934\tvalidation_1-merror:0.18690\n",
      "[826]\tvalidation_0-merror:0.08928\tvalidation_1-merror:0.18692\n",
      "[827]\tvalidation_0-merror:0.08920\tvalidation_1-merror:0.18706\n",
      "[828]\tvalidation_0-merror:0.08913\tvalidation_1-merror:0.18710\n",
      "[829]\tvalidation_0-merror:0.08907\tvalidation_1-merror:0.18712\n",
      "[830]\tvalidation_0-merror:0.08902\tvalidation_1-merror:0.18716\n",
      "[831]\tvalidation_0-merror:0.08894\tvalidation_1-merror:0.18714\n",
      "[832]\tvalidation_0-merror:0.08885\tvalidation_1-merror:0.18710\n",
      "[833]\tvalidation_0-merror:0.08879\tvalidation_1-merror:0.18708\n",
      "[834]\tvalidation_0-merror:0.08877\tvalidation_1-merror:0.18708\n",
      "[835]\tvalidation_0-merror:0.08872\tvalidation_1-merror:0.18700\n",
      "[836]\tvalidation_0-merror:0.08869\tvalidation_1-merror:0.18706\n",
      "[837]\tvalidation_0-merror:0.08861\tvalidation_1-merror:0.18712\n",
      "[838]\tvalidation_0-merror:0.08855\tvalidation_1-merror:0.18726\n",
      "[839]\tvalidation_0-merror:0.08848\tvalidation_1-merror:0.18735\n",
      "[840]\tvalidation_0-merror:0.08844\tvalidation_1-merror:0.18733\n",
      "[841]\tvalidation_0-merror:0.08834\tvalidation_1-merror:0.18745\n",
      "[842]\tvalidation_0-merror:0.08828\tvalidation_1-merror:0.18743\n",
      "[843]\tvalidation_0-merror:0.08815\tvalidation_1-merror:0.18747\n",
      "[844]\tvalidation_0-merror:0.08809\tvalidation_1-merror:0.18759\n",
      "[845]\tvalidation_0-merror:0.08793\tvalidation_1-merror:0.18759\n",
      "[846]\tvalidation_0-merror:0.08784\tvalidation_1-merror:0.18759\n",
      "[847]\tvalidation_0-merror:0.08776\tvalidation_1-merror:0.18761\n",
      "[848]\tvalidation_0-merror:0.08766\tvalidation_1-merror:0.18763\n",
      "[849]\tvalidation_0-merror:0.08756\tvalidation_1-merror:0.18761\n",
      "[850]\tvalidation_0-merror:0.08739\tvalidation_1-merror:0.18761\n",
      "[851]\tvalidation_0-merror:0.08730\tvalidation_1-merror:0.18735\n",
      "[852]\tvalidation_0-merror:0.08727\tvalidation_1-merror:0.18735\n",
      "[853]\tvalidation_0-merror:0.08722\tvalidation_1-merror:0.18733\n",
      "[854]\tvalidation_0-merror:0.08714\tvalidation_1-merror:0.18731\n",
      "[855]\tvalidation_0-merror:0.08701\tvalidation_1-merror:0.18749\n",
      "[856]\tvalidation_0-merror:0.08693\tvalidation_1-merror:0.18759\n",
      "[857]\tvalidation_0-merror:0.08686\tvalidation_1-merror:0.18747\n",
      "[858]\tvalidation_0-merror:0.08682\tvalidation_1-merror:0.18741\n",
      "[859]\tvalidation_0-merror:0.08680\tvalidation_1-merror:0.18737\n",
      "[860]\tvalidation_0-merror:0.08678\tvalidation_1-merror:0.18741\n",
      "[861]\tvalidation_0-merror:0.08668\tvalidation_1-merror:0.18733\n",
      "[862]\tvalidation_0-merror:0.08663\tvalidation_1-merror:0.18726\n",
      "[863]\tvalidation_0-merror:0.08656\tvalidation_1-merror:0.18722\n",
      "[864]\tvalidation_0-merror:0.08643\tvalidation_1-merror:0.18729\n",
      "[865]\tvalidation_0-merror:0.08634\tvalidation_1-merror:0.18716\n",
      "[866]\tvalidation_0-merror:0.08629\tvalidation_1-merror:0.18712\n",
      "[867]\tvalidation_0-merror:0.08619\tvalidation_1-merror:0.18708\n",
      "[868]\tvalidation_0-merror:0.08610\tvalidation_1-merror:0.18712\n",
      "[869]\tvalidation_0-merror:0.08598\tvalidation_1-merror:0.18718\n",
      "[870]\tvalidation_0-merror:0.08595\tvalidation_1-merror:0.18726\n",
      "[871]\tvalidation_0-merror:0.08594\tvalidation_1-merror:0.18726\n",
      "[872]\tvalidation_0-merror:0.08594\tvalidation_1-merror:0.18722\n",
      "[873]\tvalidation_0-merror:0.08587\tvalidation_1-merror:0.18731\n",
      "[874]\tvalidation_0-merror:0.08578\tvalidation_1-merror:0.18737\n",
      "[875]\tvalidation_0-merror:0.08571\tvalidation_1-merror:0.18735\n",
      "[876]\tvalidation_0-merror:0.08567\tvalidation_1-merror:0.18743\n",
      "[877]\tvalidation_0-merror:0.08556\tvalidation_1-merror:0.18751\n",
      "[878]\tvalidation_0-merror:0.08548\tvalidation_1-merror:0.18751\n",
      "[879]\tvalidation_0-merror:0.08544\tvalidation_1-merror:0.18761\n",
      "[880]\tvalidation_0-merror:0.08535\tvalidation_1-merror:0.18763\n",
      "[881]\tvalidation_0-merror:0.08525\tvalidation_1-merror:0.18759\n",
      "[882]\tvalidation_0-merror:0.08516\tvalidation_1-merror:0.18759\n",
      "[883]\tvalidation_0-merror:0.08515\tvalidation_1-merror:0.18765\n",
      "[884]\tvalidation_0-merror:0.08508\tvalidation_1-merror:0.18755\n",
      "[885]\tvalidation_0-merror:0.08500\tvalidation_1-merror:0.18751\n",
      "[886]\tvalidation_0-merror:0.08495\tvalidation_1-merror:0.18751\n",
      "[887]\tvalidation_0-merror:0.08486\tvalidation_1-merror:0.18755\n",
      "[888]\tvalidation_0-merror:0.08484\tvalidation_1-merror:0.18755\n",
      "[889]\tvalidation_0-merror:0.08478\tvalidation_1-merror:0.18751\n",
      "[890]\tvalidation_0-merror:0.08471\tvalidation_1-merror:0.18751\n",
      "[891]\tvalidation_0-merror:0.08466\tvalidation_1-merror:0.18761\n",
      "[892]\tvalidation_0-merror:0.08464\tvalidation_1-merror:0.18761\n",
      "[893]\tvalidation_0-merror:0.08461\tvalidation_1-merror:0.18767\n",
      "[894]\tvalidation_0-merror:0.08450\tvalidation_1-merror:0.18755\n",
      "[895]\tvalidation_0-merror:0.08444\tvalidation_1-merror:0.18755\n",
      "[896]\tvalidation_0-merror:0.08443\tvalidation_1-merror:0.18755\n",
      "[897]\tvalidation_0-merror:0.08435\tvalidation_1-merror:0.18757\n",
      "[898]\tvalidation_0-merror:0.08433\tvalidation_1-merror:0.18753\n",
      "[899]\tvalidation_0-merror:0.08432\tvalidation_1-merror:0.18753\n",
      "[900]\tvalidation_0-merror:0.08424\tvalidation_1-merror:0.18759\n",
      "[901]\tvalidation_0-merror:0.08421\tvalidation_1-merror:0.18751\n",
      "[902]\tvalidation_0-merror:0.08413\tvalidation_1-merror:0.18747\n",
      "[903]\tvalidation_0-merror:0.08406\tvalidation_1-merror:0.18755\n",
      "[904]\tvalidation_0-merror:0.08392\tvalidation_1-merror:0.18749\n",
      "[905]\tvalidation_0-merror:0.08379\tvalidation_1-merror:0.18755\n",
      "[906]\tvalidation_0-merror:0.08370\tvalidation_1-merror:0.18761\n",
      "[907]\tvalidation_0-merror:0.08362\tvalidation_1-merror:0.18763\n",
      "[908]\tvalidation_0-merror:0.08357\tvalidation_1-merror:0.18761\n",
      "[909]\tvalidation_0-merror:0.08341\tvalidation_1-merror:0.18771\n",
      "[910]\tvalidation_0-merror:0.08333\tvalidation_1-merror:0.18765\n",
      "[911]\tvalidation_0-merror:0.08324\tvalidation_1-merror:0.18767\n",
      "[912]\tvalidation_0-merror:0.08317\tvalidation_1-merror:0.18778\n",
      "[913]\tvalidation_0-merror:0.08307\tvalidation_1-merror:0.18794\n",
      "[914]\tvalidation_0-merror:0.08300\tvalidation_1-merror:0.18794\n",
      "[915]\tvalidation_0-merror:0.08290\tvalidation_1-merror:0.18800\n",
      "[916]\tvalidation_0-merror:0.08273\tvalidation_1-merror:0.18823\n",
      "[917]\tvalidation_0-merror:0.08264\tvalidation_1-merror:0.18818\n",
      "[918]\tvalidation_0-merror:0.08251\tvalidation_1-merror:0.18794\n",
      "[919]\tvalidation_0-merror:0.08242\tvalidation_1-merror:0.18806\n",
      "[920]\tvalidation_0-merror:0.08239\tvalidation_1-merror:0.18794\n",
      "[921]\tvalidation_0-merror:0.08227\tvalidation_1-merror:0.18802\n",
      "[922]\tvalidation_0-merror:0.08222\tvalidation_1-merror:0.18802\n",
      "[923]\tvalidation_0-merror:0.08212\tvalidation_1-merror:0.18804\n",
      "[924]\tvalidation_0-merror:0.08206\tvalidation_1-merror:0.18794\n",
      "[925]\tvalidation_0-merror:0.08193\tvalidation_1-merror:0.18796\n",
      "[926]\tvalidation_0-merror:0.08177\tvalidation_1-merror:0.18788\n",
      "[927]\tvalidation_0-merror:0.08175\tvalidation_1-merror:0.18794\n",
      "[928]\tvalidation_0-merror:0.08168\tvalidation_1-merror:0.18790\n",
      "[929]\tvalidation_0-merror:0.08156\tvalidation_1-merror:0.18780\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[930]\tvalidation_0-merror:0.08151\tvalidation_1-merror:0.18776\n",
      "[931]\tvalidation_0-merror:0.08147\tvalidation_1-merror:0.18767\n",
      "[932]\tvalidation_0-merror:0.08139\tvalidation_1-merror:0.18769\n",
      "[933]\tvalidation_0-merror:0.08129\tvalidation_1-merror:0.18765\n",
      "[934]\tvalidation_0-merror:0.08122\tvalidation_1-merror:0.18757\n",
      "[935]\tvalidation_0-merror:0.08116\tvalidation_1-merror:0.18743\n",
      "[936]\tvalidation_0-merror:0.08110\tvalidation_1-merror:0.18763\n",
      "[937]\tvalidation_0-merror:0.08104\tvalidation_1-merror:0.18769\n",
      "[938]\tvalidation_0-merror:0.08097\tvalidation_1-merror:0.18763\n",
      "[939]\tvalidation_0-merror:0.08085\tvalidation_1-merror:0.18763\n",
      "[940]\tvalidation_0-merror:0.08082\tvalidation_1-merror:0.18749\n",
      "[941]\tvalidation_0-merror:0.08071\tvalidation_1-merror:0.18753\n",
      "[942]\tvalidation_0-merror:0.08066\tvalidation_1-merror:0.18749\n",
      "[943]\tvalidation_0-merror:0.08062\tvalidation_1-merror:0.18741\n",
      "[944]\tvalidation_0-merror:0.08051\tvalidation_1-merror:0.18741\n",
      "[945]\tvalidation_0-merror:0.08038\tvalidation_1-merror:0.18726\n",
      "[946]\tvalidation_0-merror:0.08033\tvalidation_1-merror:0.18733\n",
      "[947]\tvalidation_0-merror:0.08021\tvalidation_1-merror:0.18726\n"
     ]
    }
   ],
   "source": [
    "## Best results with this combination, the output of the cell just shows the results from the latest experiment we tried,\n",
    "## which gave a score of 80.91 on the private leaderboard which wasn't our best (81.0346)\n",
    "from xgboost import XGBClassifier\n",
    "xgb = XGBClassifier(n_estimators=10000, max_depth = 6, reg_lambda = 1,\n",
    "                    learning_rate=0.2, n_jobs=-1, objective = 'multiclassova')\n",
    "xgb.fit(X_Train, Y_Train, early_stopping_rounds = 200, eval_set = [(X_Train, Y_Train), (X_CV, Y_CV)],\n",
    "        eval_metric = 'mlogloss', verbose = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "Iv6otehFaQ8j"
   },
   "outputs": [],
   "source": [
    "pred = xgb.predict(X_CV)\n",
    "print(f1_score(Y_CV, pred, average = 'weighted'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 516,
   "metadata": {
    "id": "SOExDJFzaQ8j"
   },
   "outputs": [],
   "source": [
    "xgb_pred = xgb.predict_proba(test.drop(['age_group'], axis = 1))\n",
    "test_pred = pd.DataFrame(xgb_pred)\n",
    "test_pred = test_pred.idxmax(axis = 1) + 1\n",
    "test_pred.columns = ['prediction']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 517,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "a3cIBi2waQ8j",
    "outputId": "8ab505ac-5fd4-4b9e-c387-974eedb0e357"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    1\n",
       "1    1\n",
       "2    1\n",
       "3    3\n",
       "4    1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 517,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_pred.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 518,
   "metadata": {
    "id": "6o1XUpl-aQ8j"
   },
   "outputs": [],
   "source": [
    "test_pred.to_csv('./Data/Preds/Submission18.csv', index = False, header = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 519,
   "metadata": {
    "id": "JlKSk6KBaQ8j"
   },
   "outputs": [],
   "source": [
    "import eli5 \n",
    "from eli5.sklearn import PermutationImportance\n",
    "\n",
    "\n",
    "perm = PermutationImportance(xgb, random_state=1).fit(X_CV, Y_CV)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 224,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "    <style>\n",
       "    table.eli5-weights tr:hover {\n",
       "        filter: brightness(85%);\n",
       "    }\n",
       "</style>\n",
       "\n",
       "\n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "        <table class=\"eli5-weights eli5-feature-importances\" style=\"border-collapse: collapse; border: none; margin-top: 0em; table-layout: auto;\">\n",
       "    <thead>\n",
       "    <tr style=\"border: none;\">\n",
       "        <th style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">Weight</th>\n",
       "        <th style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">Feature</th>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tbody>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 80.00%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.2987\n",
       "                \n",
       "                    &plusmn; 0.0035\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                creations\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 86.60%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.1686\n",
       "                \n",
       "                    &plusmn; 0.0021\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                userId\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 97.53%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0150\n",
       "                \n",
       "                    &plusmn; 0.0020\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                content_views\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 98.65%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0064\n",
       "                \n",
       "                    &plusmn; 0.0014\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                avgCompletion\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 98.91%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0047\n",
       "                \n",
       "                    &plusmn; 0.0008\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                slot4_trails_watched_per_day\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 98.92%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0046\n",
       "                \n",
       "                    &plusmn; 0.0004\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                slot2_trails_watched_per_day\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 98.94%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0045\n",
       "                \n",
       "                    &plusmn; 0.0018\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                weekdays_trails_watched_per_day\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.00%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0041\n",
       "                \n",
       "                    &plusmn; 0.0008\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                avgTimeSpent\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.07%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0037\n",
       "                \n",
       "                    &plusmn; 0.0006\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                number_of_words_per_action\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.08%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0037\n",
       "                \n",
       "                    &plusmn; 0.0007\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                followers_avg_age\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.10%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0035\n",
       "                \n",
       "                    &plusmn; 0.0003\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                Unnamed: 0\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.12%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0034\n",
       "                \n",
       "                    &plusmn; 0.0007\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                slot3_trails_watched_per_day\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.15%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0033\n",
       "                \n",
       "                    &plusmn; 0.0009\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                avgDuration\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.24%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0028\n",
       "                \n",
       "                    &plusmn; 0.0005\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                following_avg_age\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.29%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0025\n",
       "                \n",
       "                    &plusmn; 0.0008\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                following_rate\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.34%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0023\n",
       "                \n",
       "                    &plusmn; 0.0009\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                slot1_trails_watched_per_day\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.39%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0020\n",
       "                \n",
       "                    &plusmn; 0.0009\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                avgt2\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.47%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0017\n",
       "                \n",
       "                    &plusmn; 0.0010\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                weekends_trails_watched_per_day\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.52%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0015\n",
       "                \n",
       "                    &plusmn; 0.0005\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                wpa_binned\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.52%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0014\n",
       "                \n",
       "                    &plusmn; 0.0001\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                max_repetitive_punc\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.54%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0014\n",
       "                \n",
       "                    &plusmn; 0.0004\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                gender\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.60%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0011\n",
       "                \n",
       "                    &plusmn; 0.0007\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                num_of_comments\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.85%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0003\n",
       "                \n",
       "                    &plusmn; 0.0004\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                punctuations_per_action\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.86%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0002\n",
       "                \n",
       "                    &plusmn; 0.0007\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                words_total\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.91%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0001\n",
       "                \n",
       "                    &plusmn; 0.0001\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                avgCompletion_log\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.91%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0001\n",
       "                \n",
       "                    &plusmn; 0.0000\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                num_hash\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.94%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0001\n",
       "                \n",
       "                    &plusmn; 0.0001\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                emoji_total\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.94%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0001\n",
       "                \n",
       "                    &plusmn; 0.0002\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                tier\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(120, 100.00%, 99.96%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0.0000\n",
       "                \n",
       "                    &plusmn; 0.0001\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                punc_total\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(0, 100.00%, 100.00%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0\n",
       "                \n",
       "                    &plusmn; 0.0000\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                hash_binned\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(0, 100.00%, 100.00%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0\n",
       "                \n",
       "                    &plusmn; 0.0000\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                Majority Followers\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(0, 100.00%, 100.00%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0\n",
       "                \n",
       "                    &plusmn; 0.0000\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                num_hash_rescaled\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(0, 100.00%, 100.00%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0\n",
       "                \n",
       "                    &plusmn; 0.0000\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                epa_log\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(0, 100.00%, 100.00%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0\n",
       "                \n",
       "                    &plusmn; 0.0000\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                avgTimeSpent_log\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(0, 100.00%, 100.00%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0\n",
       "                \n",
       "                    &plusmn; 0.0000\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                wpa_log\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(0, 100.00%, 100.00%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0\n",
       "                \n",
       "                    &plusmn; 0.0000\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                Majority Following\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(0, 100.00%, 100.00%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0\n",
       "                \n",
       "                    &plusmn; 0.0000\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                rep_punc_log\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(0, 100.00%, 100.00%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0\n",
       "                \n",
       "                    &plusmn; 0.0000\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                hpa_log\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(0, 100.00%, 100.00%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0\n",
       "                \n",
       "                    &plusmn; 0.0000\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                avgt2_log\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(0, 100.00%, 100.00%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0\n",
       "                \n",
       "                    &plusmn; 0.0000\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                ppa_binned\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(0, 100.00%, 100.00%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0\n",
       "                \n",
       "                    &plusmn; 0.0000\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                ppa_log\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(0, 100.00%, 100.00%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                0\n",
       "                \n",
       "                    &plusmn; 0.0000\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                view_binned\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(0, 100.00%, 99.93%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                -0.0001\n",
       "                \n",
       "                    &plusmn; 0.0001\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                num_of_hashtags_per_action\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(0, 100.00%, 99.92%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                -0.0001\n",
       "                \n",
       "                    &plusmn; 0.0002\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                emoji_count_per_action\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "        <tr style=\"background-color: hsl(0, 100.00%, 99.91%); border: none;\">\n",
       "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
       "                -0.0001\n",
       "                \n",
       "                    &plusmn; 0.0002\n",
       "                \n",
       "            </td>\n",
       "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
       "                avgComments\n",
       "            </td>\n",
       "        </tr>\n",
       "    \n",
       "    \n",
       "    </tbody>\n",
       "</table>\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "    \n",
       "\n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": 224,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "eli5.show_weights(perm, feature_names = X_CV.columns.tolist(), top=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Cascade Cup 2020.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}