main32G -copy.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# coding=utf-8\n",
    "# @author:bryan\n",
    "# blog: https://blog.csdn.net/bryan__\n",
    "# github: https://github.com/YouChouNoBB/2018-tencent-ad-competition-baseline\n",
    "import pandas as pd\n",
    "import lightgbm as lgb\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.preprocessing import OneHotEncoder,LabelEncoder\n",
    "from scipy import sparse\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "ad_feature=pd.read_csv('./Oridata/tc2018/adFeature.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>aid</th>\n",
       "      <th>advertiserId</th>\n",
       "      <th>campaignId</th>\n",
       "      <th>creativeId</th>\n",
       "      <th>creativeSize</th>\n",
       "      <th>adCategoryId</th>\n",
       "      <th>productId</th>\n",
       "      <th>productType</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>177</td>\n",
       "      <td>8203</td>\n",
       "      <td>76104</td>\n",
       "      <td>1500666</td>\n",
       "      <td>59</td>\n",
       "      <td>282</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2050</td>\n",
       "      <td>19441</td>\n",
       "      <td>178687</td>\n",
       "      <td>245165</td>\n",
       "      <td>53</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1716</td>\n",
       "      <td>5552</td>\n",
       "      <td>158101</td>\n",
       "      <td>1080850</td>\n",
       "      <td>35</td>\n",
       "      <td>27</td>\n",
       "      <td>113</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>336</td>\n",
       "      <td>370</td>\n",
       "      <td>4833</td>\n",
       "      <td>119845</td>\n",
       "      <td>22</td>\n",
       "      <td>67</td>\n",
       "      <td>113</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>671</td>\n",
       "      <td>45705</td>\n",
       "      <td>352827</td>\n",
       "      <td>660519</td>\n",
       "      <td>42</td>\n",
       "      <td>67</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    aid  advertiserId  campaignId  creativeId  creativeSize  adCategoryId  \\\n",
       "0   177          8203       76104     1500666            59           282   \n",
       "1  2050         19441      178687      245165            53             1   \n",
       "2  1716          5552      158101     1080850            35            27   \n",
       "3   336           370        4833      119845            22            67   \n",
       "4   671         45705      352827      660519            42            67   \n",
       "\n",
       "   productId  productType  \n",
       "0          0            6  \n",
       "1          0            6  \n",
       "2        113            9  \n",
       "3        113            9  \n",
       "4          0            4  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ad_feature.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(173, 8)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ad_feature.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "if os.path.exists('./Oridata/tc2018/userFeature.csv'):\n",
    "    user_feature=pd.read_csv('./Oridata/tc2018/userFeature.csv')\n",
    "else:\n",
    "    print('No user_feature')\n",
    "    userFeature_data = []\n",
    "    with open('./Oridata/tc2018/userFeature.data', 'r') as f:\n",
    "        for i, line in enumerate(f):\n",
    "            line = line.strip().split('|')\n",
    "            userFeature_dict = {}\n",
    "            for each in line:\n",
    "                each_list = each.split(' ')\n",
    "                userFeature_dict[each_list[0]] = ' '.join(each_list[1:])\n",
    "            userFeature_data.append(userFeature_dict)\n",
    "            if i % 100000 == 0:\n",
    "                print(i)\n",
    "        user_feature = pd.DataFrame(userFeature_data)\n",
    "        user_feature.to_csv('../data/userFeature.csv', index=False)\n",
    "        del userFeature_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": true
   },
   "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>LBS</th>\n",
       "      <th>age</th>\n",
       "      <th>appIdAction</th>\n",
       "      <th>appIdInstall</th>\n",
       "      <th>carrier</th>\n",
       "      <th>consumptionAbility</th>\n",
       "      <th>ct</th>\n",
       "      <th>education</th>\n",
       "      <th>gender</th>\n",
       "      <th>house</th>\n",
       "      <th>interest1</th>\n",
       "      <th>interest2</th>\n",
       "      <th>interest3</th>\n",
       "      <th>interest4</th>\n",
       "      <th>interest5</th>\n",
       "      <th>kw1</th>\n",
       "      <th>kw2</th>\n",
       "      <th>kw3</th>\n",
       "      <th>marriageStatus</th>\n",
       "      <th>os</th>\n",
       "      <th>topic1</th>\n",
       "      <th>topic2</th>\n",
       "      <th>topic3</th>\n",
       "      <th>uid</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>950.0</td>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3 1</td>\n",
       "      <td>7</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>93 70 77 86 109 47 75 69 45 8 29 49 83 6 46 36...</td>\n",
       "      <td>46 19 13 29</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>52 100 72 131 116 11 71 12 8 113 28 73 6 132 9...</td>\n",
       "      <td>664359 276966 734911 103617 562294</td>\n",
       "      <td>11395 79112 115065 77033 36176</td>\n",
       "      <td>NaN</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>9826 105 8525 5488 7281</td>\n",
       "      <td>9708 5553 6745 7477 7150</td>\n",
       "      <td>NaN</td>\n",
       "      <td>26325489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>803.0</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3 1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>75 29</td>\n",
       "      <td>33</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>338851 361151 542834 496283 229952</td>\n",
       "      <td>80263 39618 53539 180 38163</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5 13</td>\n",
       "      <td>1</td>\n",
       "      <td>4391 9140 5669 1348 4388</td>\n",
       "      <td>9401 7724 1380 8890 7153</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1184123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>927.0</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3 1</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>70 12 28 106 59 49 41 6 42 115 35 116 36 11 96</td>\n",
       "      <td>51 22 79 81 70 6 21 4 41 35</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>77 72 80 116 101 13 1 109 8 50 6 42 76 9 46 36...</td>\n",
       "      <td>746140 695808 126355 771775 411511</td>\n",
       "      <td>105115 71378 41409 74061 44005</td>\n",
       "      <td>NaN</td>\n",
       "      <td>13 10</td>\n",
       "      <td>1</td>\n",
       "      <td>1502 5488 9826 2187 8088</td>\n",
       "      <td>5005 9154 2756 5612 4209</td>\n",
       "      <td>NaN</td>\n",
       "      <td>76072711</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>486.0</td>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1 3</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>93 70 77 47 76 75 69 29 59 49 122 42 46 23 17 ...</td>\n",
       "      <td>24 70 19 13 30 10</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>100 80 92 37 116 13 47 4 71 8 50 28 98 115 6 4...</td>\n",
       "      <td>283399 402245 734509 654027 32061</td>\n",
       "      <td>74010 32918 67882 116802 20957</td>\n",
       "      <td>NaN</td>\n",
       "      <td>11</td>\n",
       "      <td>1 2</td>\n",
       "      <td>1619 7342 3064 9213 8525</td>\n",
       "      <td>810 2438 5659 1844 9262</td>\n",
       "      <td>NaN</td>\n",
       "      <td>63071413</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>112.0</td>\n",
       "      <td>5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>3 1 4 2</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>75 29 49 122 36 11</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>131 116 13 8 6 132 42 9 59 18 58 64 129 103</td>\n",
       "      <td>562294 157603 589706 657719 495672</td>\n",
       "      <td>62764 97803 89066 55545 74061</td>\n",
       "      <td>NaN</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>477 9826 5808 644 2747</td>\n",
       "      <td>5483 2199 5424 1511 7751</td>\n",
       "      <td>NaN</td>\n",
       "      <td>81294159</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     LBS  age appIdAction appIdInstall  carrier  consumptionAbility       ct  \\\n",
       "0  950.0    4         NaN          NaN        1                   2      3 1   \n",
       "1  803.0    2         NaN          NaN        1                   1      3 1   \n",
       "2  927.0    1         NaN          NaN        3                   1      3 1   \n",
       "3  486.0    4         NaN          NaN        3                   1      1 3   \n",
       "4  112.0    5         NaN          NaN        3                   2  3 1 4 2   \n",
       "\n",
       "   education  gender  house  \\\n",
       "0          7       2    NaN   \n",
       "1          2       1    NaN   \n",
       "2          5       1    NaN   \n",
       "3          7       1    NaN   \n",
       "4          6       1    NaN   \n",
       "\n",
       "                                           interest1  \\\n",
       "0  93 70 77 86 109 47 75 69 45 8 29 49 83 6 46 36...   \n",
       "1                                              75 29   \n",
       "2     70 12 28 106 59 49 41 6 42 115 35 116 36 11 96   \n",
       "3  93 70 77 47 76 75 69 29 59 49 122 42 46 23 17 ...   \n",
       "4                                 75 29 49 122 36 11   \n",
       "\n",
       "                     interest2 interest3 interest4  \\\n",
       "0                  46 19 13 29       NaN       NaN   \n",
       "1                           33       NaN       NaN   \n",
       "2  51 22 79 81 70 6 21 4 41 35       NaN       NaN   \n",
       "3            24 70 19 13 30 10       NaN       NaN   \n",
       "4                          NaN       NaN       NaN   \n",
       "\n",
       "                                           interest5  \\\n",
       "0  52 100 72 131 116 11 71 12 8 113 28 73 6 132 9...   \n",
       "1                                                NaN   \n",
       "2  77 72 80 116 101 13 1 109 8 50 6 42 76 9 46 36...   \n",
       "3  100 80 92 37 116 13 47 4 71 8 50 28 98 115 6 4...   \n",
       "4        131 116 13 8 6 132 42 9 59 18 58 64 129 103   \n",
       "\n",
       "                                  kw1                             kw2  kw3  \\\n",
       "0  664359 276966 734911 103617 562294  11395 79112 115065 77033 36176  NaN   \n",
       "1  338851 361151 542834 496283 229952     80263 39618 53539 180 38163  NaN   \n",
       "2  746140 695808 126355 771775 411511  105115 71378 41409 74061 44005  NaN   \n",
       "3   283399 402245 734509 654027 32061  74010 32918 67882 116802 20957  NaN   \n",
       "4  562294 157603 589706 657719 495672   62764 97803 89066 55545 74061  NaN   \n",
       "\n",
       "  marriageStatus   os                    topic1                    topic2  \\\n",
       "0             11    2   9826 105 8525 5488 7281  9708 5553 6745 7477 7150   \n",
       "1           5 13    1  4391 9140 5669 1348 4388  9401 7724 1380 8890 7153   \n",
       "2          13 10    1  1502 5488 9826 2187 8088  5005 9154 2756 5612 4209   \n",
       "3             11  1 2  1619 7342 3064 9213 8525   810 2438 5659 1844 9262   \n",
       "4             11    1    477 9826 5808 644 2747  5483 2199 5424 1511 7751   \n",
       "\n",
       "  topic3       uid  \n",
       "0    NaN  26325489  \n",
       "1    NaN   1184123  \n",
       "2    NaN  76072711  \n",
       "3    NaN  63071413  \n",
       "4    NaN  81294159  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#显示所有列\n",
    "pd.set_option('display.max_columns', None)\n",
    "#显示所有行\n",
    "pd.set_option('display.max_rows', None)\n",
    "user_feature.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(11420039, 24)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_feature.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['LBS',\n",
       " 'age',\n",
       " 'appIdAction',\n",
       " 'appIdInstall',\n",
       " 'carrier',\n",
       " 'consumptionAbility',\n",
       " 'ct',\n",
       " 'education',\n",
       " 'gender',\n",
       " 'house',\n",
       " 'interest1',\n",
       " 'interest2',\n",
       " 'interest3',\n",
       " 'interest4',\n",
       " 'interest5',\n",
       " 'kw1',\n",
       " 'kw2',\n",
       " 'kw3',\n",
       " 'marriageStatus',\n",
       " 'os',\n",
       " 'topic1',\n",
       " 'topic2',\n",
       " 'topic3',\n",
       " 'uid']"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(user_feature) #打印每列的名字"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "train=pd.read_csv('./Oridata/tc2018/train.csv')\n",
    "predict=pd.read_csv('./Oridata/tc2018/test1.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>aid</th>\n",
       "      <th>uid</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>699</td>\n",
       "      <td>78508957</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1991</td>\n",
       "      <td>3637295</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1119</td>\n",
       "      <td>19229018</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2013</td>\n",
       "      <td>79277120</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    aid       uid  label\n",
       "0   699  78508957      0\n",
       "1  1991   3637295      0\n",
       "2  1119  19229018      0\n",
       "3  2013  79277120      0"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train[:4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8798814, 3)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>aid</th>\n",
       "      <th>uid</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2118</td>\n",
       "      <td>64355836</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>692</td>\n",
       "      <td>45051997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>692</td>\n",
       "      <td>10869198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1918</td>\n",
       "      <td>75929554</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1596</td>\n",
       "      <td>5790162</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    aid       uid\n",
       "0  2118  64355836\n",
       "1   692  45051997\n",
       "2   692  10869198\n",
       "3  1918  75929554\n",
       "4  1596   5790162"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predict.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2265989, 2)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predict.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "a = train.loc[train['label']==-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>aid</th>\n",
       "      <th>uid</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [aid, uid, label]\n",
       "Index: []"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "b = train.loc[train['label']==-1,'label']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Series([], Name: label, dtype: int64)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "train.loc[train['label']==-1,'label']=0\n",
    "predict['label']=-1 # 运行有点慢"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "data=pd.concat([train,predict])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(11064803, 3)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>aid</th>\n",
       "      <th>uid</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>699</td>\n",
       "      <td>78508957</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1991</td>\n",
       "      <td>3637295</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1119</td>\n",
       "      <td>19229018</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2013</td>\n",
       "      <td>79277120</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>692</td>\n",
       "      <td>41528441</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    aid       uid  label\n",
       "0   699  78508957      0\n",
       "1  1991   3637295      0\n",
       "2  1119  19229018      0\n",
       "3  2013  79277120      0\n",
       "4   692  41528441      0"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "data=pd.merge(data,ad_feature,on='aid',how='left')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "data=pd.merge(data,user_feature,on='uid',how='left') "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>aid</th>\n",
       "      <th>uid</th>\n",
       "      <th>label</th>\n",
       "      <th>advertiserId</th>\n",
       "      <th>campaignId</th>\n",
       "      <th>creativeId</th>\n",
       "      <th>creativeSize</th>\n",
       "      <th>adCategoryId</th>\n",
       "      <th>productId</th>\n",
       "      <th>productType</th>\n",
       "      <th>LBS</th>\n",
       "      <th>age</th>\n",
       "      <th>appIdAction</th>\n",
       "      <th>appIdInstall</th>\n",
       "      <th>carrier</th>\n",
       "      <th>consumptionAbility</th>\n",
       "      <th>ct</th>\n",
       "      <th>education</th>\n",
       "      <th>gender</th>\n",
       "      <th>house</th>\n",
       "      <th>interest1</th>\n",
       "      <th>interest2</th>\n",
       "      <th>interest3</th>\n",
       "      <th>interest4</th>\n",
       "      <th>interest5</th>\n",
       "      <th>kw1</th>\n",
       "      <th>kw2</th>\n",
       "      <th>kw3</th>\n",
       "      <th>marriageStatus</th>\n",
       "      <th>os</th>\n",
       "      <th>topic1</th>\n",
       "      <th>topic2</th>\n",
       "      <th>topic3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>699</td>\n",
       "      <td>78508957</td>\n",
       "      <td>0</td>\n",
       "      <td>1082</td>\n",
       "      <td>295940</td>\n",
       "      <td>731679</td>\n",
       "      <td>59</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>576.0</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>70 100 48 49 23 36 11</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>124532 762727 771871 8561 610048</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>4908 9745 7070 4203 3292</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1991</td>\n",
       "      <td>3637295</td>\n",
       "      <td>0</td>\n",
       "      <td>702</td>\n",
       "      <td>42104</td>\n",
       "      <td>1441131</td>\n",
       "      <td>53</td>\n",
       "      <td>10</td>\n",
       "      <td>4669</td>\n",
       "      <td>11</td>\n",
       "      <td>353.0</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>93 109 75 29 59 49 6 5 73 46 22 36 11 3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>100 131 37 8 46</td>\n",
       "      <td>531499 542834 303322 503318 68211</td>\n",
       "      <td>89232 30101 117462 117548 85039</td>\n",
       "      <td>NaN</td>\n",
       "      <td>13</td>\n",
       "      <td>1</td>\n",
       "      <td>8445 6762 7608 523 7688</td>\n",
       "      <td>4518 862 3643 702 5247</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1119</td>\n",
       "      <td>19229018</td>\n",
       "      <td>0</td>\n",
       "      <td>3993</td>\n",
       "      <td>63752</td>\n",
       "      <td>798752</td>\n",
       "      <td>59</td>\n",
       "      <td>10</td>\n",
       "      <td>19256</td>\n",
       "      <td>11</td>\n",
       "      <td>94.0</td>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>93 37 47 75 29 46 17 9 56 36 11</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>255920 190937 576486 460368 135689</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>13 10</td>\n",
       "      <td>0</td>\n",
       "      <td>941 4674 9735 8953 5352</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2013</td>\n",
       "      <td>79277120</td>\n",
       "      <td>0</td>\n",
       "      <td>6937</td>\n",
       "      <td>186348</td>\n",
       "      <td>1427984</td>\n",
       "      <td>35</td>\n",
       "      <td>89</td>\n",
       "      <td>3791</td>\n",
       "      <td>9</td>\n",
       "      <td>48.0</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3 1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>49 18</td>\n",
       "      <td>42 9 1 46 70 14 29 10 35</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>92 116 128 42 5 121 64</td>\n",
       "      <td>101914 294138</td>\n",
       "      <td>57664 32928 40670 11395 79112</td>\n",
       "      <td>NaN</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>4538 6790 8850 3422 1940</td>\n",
       "      <td>2979 1877 358 666 4166</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>692</td>\n",
       "      <td>41528441</td>\n",
       "      <td>0</td>\n",
       "      <td>6946</td>\n",
       "      <td>296367</td>\n",
       "      <td>455396</td>\n",
       "      <td>59</td>\n",
       "      <td>24</td>\n",
       "      <td>3794</td>\n",
       "      <td>11</td>\n",
       "      <td>333.0</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1 3 4</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>70 109 12 75 76 106 29 59 49 41 111 85 17 56 1...</td>\n",
       "      <td>47 42 73 9 31 46 32 70 21 33 19 60 13 74 16 55...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>100 72 37 11 4 79 113 6 46 59</td>\n",
       "      <td>155585 562294 112872 273650 371405</td>\n",
       "      <td>67951 40801 56571 119723 16159</td>\n",
       "      <td>NaN</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>4535 5590 47 5570 2078</td>\n",
       "      <td>856 4587 1877 392 9103</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    aid       uid  label  advertiserId  campaignId  creativeId  creativeSize  \\\n",
       "0   699  78508957      0          1082      295940      731679            59   \n",
       "1  1991   3637295      0           702       42104     1441131            53   \n",
       "2  1119  19229018      0          3993       63752      798752            59   \n",
       "3  2013  79277120      0          6937      186348     1427984            35   \n",
       "4   692  41528441      0          6946      296367      455396            59   \n",
       "\n",
       "   adCategoryId  productId  productType    LBS  age appIdAction appIdInstall  \\\n",
       "0            13          0            6  576.0    1         NaN          NaN   \n",
       "1            10       4669           11  353.0    3         NaN          NaN   \n",
       "2            10      19256           11   94.0    4         NaN          NaN   \n",
       "3            89       3791            9   48.0    2         NaN          NaN   \n",
       "4            24       3794           11  333.0    1         NaN          NaN   \n",
       "\n",
       "   carrier  consumptionAbility     ct  education  gender  house  \\\n",
       "0        0                   1      0          1       1    NaN   \n",
       "1        1                   0      3          7       2    NaN   \n",
       "2        0                   0      0          1       1    NaN   \n",
       "3        1                   1    3 1          2       1    1.0   \n",
       "4        1                   1  1 3 4          7       1    NaN   \n",
       "\n",
       "                                           interest1  \\\n",
       "0                              70 100 48 49 23 36 11   \n",
       "1            93 109 75 29 59 49 6 5 73 46 22 36 11 3   \n",
       "2                    93 37 47 75 29 46 17 9 56 36 11   \n",
       "3                                              49 18   \n",
       "4  70 109 12 75 76 106 29 59 49 41 111 85 17 56 1...   \n",
       "\n",
       "                                           interest2 interest3 interest4  \\\n",
       "0                                                NaN       NaN       NaN   \n",
       "1                                                NaN       NaN       NaN   \n",
       "2                                                NaN       NaN       NaN   \n",
       "3                           42 9 1 46 70 14 29 10 35       NaN       NaN   \n",
       "4  47 42 73 9 31 46 32 70 21 33 19 60 13 74 16 55...       NaN       NaN   \n",
       "\n",
       "                       interest5                                 kw1  \\\n",
       "0                            NaN    124532 762727 771871 8561 610048   \n",
       "1                100 131 37 8 46   531499 542834 303322 503318 68211   \n",
       "2                            NaN  255920 190937 576486 460368 135689   \n",
       "3         92 116 128 42 5 121 64                       101914 294138   \n",
       "4  100 72 37 11 4 79 113 6 46 59  155585 562294 112872 273650 371405   \n",
       "\n",
       "                               kw2  kw3 marriageStatus os  \\\n",
       "0                              NaN  NaN             10  0   \n",
       "1  89232 30101 117462 117548 85039  NaN             13  1   \n",
       "2                              NaN  NaN          13 10  0   \n",
       "3    57664 32928 40670 11395 79112  NaN             10  1   \n",
       "4   67951 40801 56571 119723 16159  NaN             10  1   \n",
       "\n",
       "                     topic1                  topic2 topic3  \n",
       "0  4908 9745 7070 4203 3292                     NaN    NaN  \n",
       "1   8445 6762 7608 523 7688  4518 862 3643 702 5247    NaN  \n",
       "2   941 4674 9735 8953 5352                     NaN    NaN  \n",
       "3  4538 6790 8850 3422 1940  2979 1877 358 666 4166    NaN  \n",
       "4    4535 5590 47 5570 2078  856 4587 1877 392 9103    NaN  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(11064803, 33)"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "data=data.fillna('-1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>aid</th>\n",
       "      <th>uid</th>\n",
       "      <th>label</th>\n",
       "      <th>advertiserId</th>\n",
       "      <th>campaignId</th>\n",
       "      <th>creativeId</th>\n",
       "      <th>creativeSize</th>\n",
       "      <th>adCategoryId</th>\n",
       "      <th>productId</th>\n",
       "      <th>productType</th>\n",
       "      <th>LBS</th>\n",
       "      <th>age</th>\n",
       "      <th>appIdAction</th>\n",
       "      <th>appIdInstall</th>\n",
       "      <th>carrier</th>\n",
       "      <th>consumptionAbility</th>\n",
       "      <th>ct</th>\n",
       "      <th>education</th>\n",
       "      <th>gender</th>\n",
       "      <th>house</th>\n",
       "      <th>interest1</th>\n",
       "      <th>interest2</th>\n",
       "      <th>interest3</th>\n",
       "      <th>interest4</th>\n",
       "      <th>interest5</th>\n",
       "      <th>kw1</th>\n",
       "      <th>kw2</th>\n",
       "      <th>kw3</th>\n",
       "      <th>marriageStatus</th>\n",
       "      <th>os</th>\n",
       "      <th>topic1</th>\n",
       "      <th>topic2</th>\n",
       "      <th>topic3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>699</td>\n",
       "      <td>78508957</td>\n",
       "      <td>0</td>\n",
       "      <td>1082</td>\n",
       "      <td>295940</td>\n",
       "      <td>731679</td>\n",
       "      <td>59</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>576</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>70 100 48 49 23 36 11</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>124532 762727 771871 8561 610048</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>4908 9745 7070 4203 3292</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1991</td>\n",
       "      <td>3637295</td>\n",
       "      <td>0</td>\n",
       "      <td>702</td>\n",
       "      <td>42104</td>\n",
       "      <td>1441131</td>\n",
       "      <td>53</td>\n",
       "      <td>10</td>\n",
       "      <td>4669</td>\n",
       "      <td>11</td>\n",
       "      <td>353</td>\n",
       "      <td>3</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>2</td>\n",
       "      <td>-1</td>\n",
       "      <td>93 109 75 29 59 49 6 5 73 46 22 36 11 3</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>100 131 37 8 46</td>\n",
       "      <td>531499 542834 303322 503318 68211</td>\n",
       "      <td>89232 30101 117462 117548 85039</td>\n",
       "      <td>-1</td>\n",
       "      <td>13</td>\n",
       "      <td>1</td>\n",
       "      <td>8445 6762 7608 523 7688</td>\n",
       "      <td>4518 862 3643 702 5247</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1119</td>\n",
       "      <td>19229018</td>\n",
       "      <td>0</td>\n",
       "      <td>3993</td>\n",
       "      <td>63752</td>\n",
       "      <td>798752</td>\n",
       "      <td>59</td>\n",
       "      <td>10</td>\n",
       "      <td>19256</td>\n",
       "      <td>11</td>\n",
       "      <td>94</td>\n",
       "      <td>4</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>93 37 47 75 29 46 17 9 56 36 11</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>255920 190937 576486 460368 135689</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>13 10</td>\n",
       "      <td>0</td>\n",
       "      <td>941 4674 9735 8953 5352</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2013</td>\n",
       "      <td>79277120</td>\n",
       "      <td>0</td>\n",
       "      <td>6937</td>\n",
       "      <td>186348</td>\n",
       "      <td>1427984</td>\n",
       "      <td>35</td>\n",
       "      <td>89</td>\n",
       "      <td>3791</td>\n",
       "      <td>9</td>\n",
       "      <td>48</td>\n",
       "      <td>2</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3 1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>49 18</td>\n",
       "      <td>42 9 1 46 70 14 29 10 35</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>92 116 128 42 5 121 64</td>\n",
       "      <td>101914 294138</td>\n",
       "      <td>57664 32928 40670 11395 79112</td>\n",
       "      <td>-1</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>4538 6790 8850 3422 1940</td>\n",
       "      <td>2979 1877 358 666 4166</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>692</td>\n",
       "      <td>41528441</td>\n",
       "      <td>0</td>\n",
       "      <td>6946</td>\n",
       "      <td>296367</td>\n",
       "      <td>455396</td>\n",
       "      <td>59</td>\n",
       "      <td>24</td>\n",
       "      <td>3794</td>\n",
       "      <td>11</td>\n",
       "      <td>333</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1 3 4</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>70 109 12 75 76 106 29 59 49 41 111 85 17 56 1...</td>\n",
       "      <td>47 42 73 9 31 46 32 70 21 33 19 60 13 74 16 55...</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>100 72 37 11 4 79 113 6 46 59</td>\n",
       "      <td>155585 562294 112872 273650 371405</td>\n",
       "      <td>67951 40801 56571 119723 16159</td>\n",
       "      <td>-1</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>4535 5590 47 5570 2078</td>\n",
       "      <td>856 4587 1877 392 9103</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    aid       uid  label  advertiserId  campaignId  creativeId  creativeSize  \\\n",
       "0   699  78508957      0          1082      295940      731679            59   \n",
       "1  1991   3637295      0           702       42104     1441131            53   \n",
       "2  1119  19229018      0          3993       63752      798752            59   \n",
       "3  2013  79277120      0          6937      186348     1427984            35   \n",
       "4   692  41528441      0          6946      296367      455396            59   \n",
       "\n",
       "   adCategoryId  productId  productType  LBS  age appIdAction appIdInstall  \\\n",
       "0            13          0            6  576    1          -1           -1   \n",
       "1            10       4669           11  353    3          -1           -1   \n",
       "2            10      19256           11   94    4          -1           -1   \n",
       "3            89       3791            9   48    2          -1           -1   \n",
       "4            24       3794           11  333    1          -1           -1   \n",
       "\n",
       "   carrier  consumptionAbility     ct  education  gender house  \\\n",
       "0        0                   1      0          1       1    -1   \n",
       "1        1                   0      3          7       2    -1   \n",
       "2        0                   0      0          1       1    -1   \n",
       "3        1                   1    3 1          2       1     1   \n",
       "4        1                   1  1 3 4          7       1    -1   \n",
       "\n",
       "                                           interest1  \\\n",
       "0                              70 100 48 49 23 36 11   \n",
       "1            93 109 75 29 59 49 6 5 73 46 22 36 11 3   \n",
       "2                    93 37 47 75 29 46 17 9 56 36 11   \n",
       "3                                              49 18   \n",
       "4  70 109 12 75 76 106 29 59 49 41 111 85 17 56 1...   \n",
       "\n",
       "                                           interest2 interest3 interest4  \\\n",
       "0                                                 -1        -1        -1   \n",
       "1                                                 -1        -1        -1   \n",
       "2                                                 -1        -1        -1   \n",
       "3                           42 9 1 46 70 14 29 10 35        -1        -1   \n",
       "4  47 42 73 9 31 46 32 70 21 33 19 60 13 74 16 55...        -1        -1   \n",
       "\n",
       "                       interest5                                 kw1  \\\n",
       "0                             -1    124532 762727 771871 8561 610048   \n",
       "1                100 131 37 8 46   531499 542834 303322 503318 68211   \n",
       "2                             -1  255920 190937 576486 460368 135689   \n",
       "3         92 116 128 42 5 121 64                       101914 294138   \n",
       "4  100 72 37 11 4 79 113 6 46 59  155585 562294 112872 273650 371405   \n",
       "\n",
       "                               kw2 kw3 marriageStatus os  \\\n",
       "0                               -1  -1             10  0   \n",
       "1  89232 30101 117462 117548 85039  -1             13  1   \n",
       "2                               -1  -1          13 10  0   \n",
       "3    57664 32928 40670 11395 79112  -1             10  1   \n",
       "4   67951 40801 56571 119723 16159  -1             10  1   \n",
       "\n",
       "                     topic1                  topic2 topic3  \n",
       "0  4908 9745 7070 4203 3292                      -1     -1  \n",
       "1   8445 6762 7608 523 7688  4518 862 3643 702 5247     -1  \n",
       "2   941 4674 9735 8953 5352                      -1     -1  \n",
       "3  4538 6790 8850 3422 1940  2979 1877 358 666 4166     -1  \n",
       "4    4535 5590 47 5570 2078  856 4587 1877 392 9103     -1  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "one_hot_feature=['LBS','age','carrier','consumptionAbility','education','gender','house','os','ct','marriageStatus','advertiserId','campaignId', 'creativeId',\n",
    "       'adCategoryId', 'productId', 'productType']\n",
    "vector_feature=['appIdAction','appIdInstall','interest1','interest2','interest3','interest4','interest5','kw1','kw2','kw3','topic1','topic2','topic3']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "for feature in one_hot_feature:\n",
    "    try:\n",
    "        data[feature] = LabelEncoder().fit_transform(data[feature].apply(int))\n",
    "    except:\n",
    "        data[feature] = LabelEncoder().fit_transform(data[feature])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(11064803, 33)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>aid</th>\n",
       "      <th>uid</th>\n",
       "      <th>label</th>\n",
       "      <th>advertiserId</th>\n",
       "      <th>campaignId</th>\n",
       "      <th>creativeId</th>\n",
       "      <th>creativeSize</th>\n",
       "      <th>adCategoryId</th>\n",
       "      <th>productId</th>\n",
       "      <th>productType</th>\n",
       "      <th>LBS</th>\n",
       "      <th>age</th>\n",
       "      <th>appIdAction</th>\n",
       "      <th>appIdInstall</th>\n",
       "      <th>carrier</th>\n",
       "      <th>consumptionAbility</th>\n",
       "      <th>ct</th>\n",
       "      <th>education</th>\n",
       "      <th>gender</th>\n",
       "      <th>house</th>\n",
       "      <th>interest1</th>\n",
       "      <th>interest2</th>\n",
       "      <th>interest3</th>\n",
       "      <th>interest4</th>\n",
       "      <th>interest5</th>\n",
       "      <th>kw1</th>\n",
       "      <th>kw2</th>\n",
       "      <th>kw3</th>\n",
       "      <th>marriageStatus</th>\n",
       "      <th>os</th>\n",
       "      <th>topic1</th>\n",
       "      <th>topic2</th>\n",
       "      <th>topic3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>699</td>\n",
       "      <td>78508957</td>\n",
       "      <td>0</td>\n",
       "      <td>15</td>\n",
       "      <td>108</td>\n",
       "      <td>61</td>\n",
       "      <td>59</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>490</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>70 100 48 49 23 36 11</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>124532 762727 771871 8561 610048</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4908 9745 7070 4203 3292</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1991</td>\n",
       "      <td>3637295</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>38</td>\n",
       "      <td>135</td>\n",
       "      <td>53</td>\n",
       "      <td>4</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "      <td>299</td>\n",
       "      <td>3</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>33</td>\n",
       "      <td>7</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>93 109 75 29 59 49 6 5 73 46 22 36 11 3</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>100 131 37 8 46</td>\n",
       "      <td>531499 542834 303322 503318 68211</td>\n",
       "      <td>89232 30101 117462 117548 85039</td>\n",
       "      <td>-1</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>8445 6762 7608 523 7688</td>\n",
       "      <td>4518 862 3643 702 5247</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1119</td>\n",
       "      <td>19229018</td>\n",
       "      <td>0</td>\n",
       "      <td>20</td>\n",
       "      <td>53</td>\n",
       "      <td>69</td>\n",
       "      <td>59</td>\n",
       "      <td>4</td>\n",
       "      <td>28</td>\n",
       "      <td>3</td>\n",
       "      <td>82</td>\n",
       "      <td>4</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>93 37 47 75 29 46 17 9 56 36 11</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>255920 190937 576486 460368 135689</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>941 4674 9735 8953 5352</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2013</td>\n",
       "      <td>79277120</td>\n",
       "      <td>0</td>\n",
       "      <td>25</td>\n",
       "      <td>93</td>\n",
       "      <td>133</td>\n",
       "      <td>35</td>\n",
       "      <td>23</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>41</td>\n",
       "      <td>2</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>49 18</td>\n",
       "      <td>42 9 1 46 70 14 29 10 35</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>92 116 128 42 5 121 64</td>\n",
       "      <td>101914 294138</td>\n",
       "      <td>57664 32928 40670 11395 79112</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4538 6790 8850 3422 1940</td>\n",
       "      <td>2979 1877 358 666 4166</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>692</td>\n",
       "      <td>41528441</td>\n",
       "      <td>0</td>\n",
       "      <td>26</td>\n",
       "      <td>109</td>\n",
       "      <td>37</td>\n",
       "      <td>59</td>\n",
       "      <td>8</td>\n",
       "      <td>11</td>\n",
       "      <td>3</td>\n",
       "      <td>283</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>70 109 12 75 76 106 29 59 49 41 111 85 17 56 1...</td>\n",
       "      <td>47 42 73 9 31 46 32 70 21 33 19 60 13 74 16 55...</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>100 72 37 11 4 79 113 6 46 59</td>\n",
       "      <td>155585 562294 112872 273650 371405</td>\n",
       "      <td>67951 40801 56571 119723 16159</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4535 5590 47 5570 2078</td>\n",
       "      <td>856 4587 1877 392 9103</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    aid       uid  label  advertiserId  campaignId  creativeId  creativeSize  \\\n",
       "0   699  78508957      0            15         108          61            59   \n",
       "1  1991   3637295      0             9          38         135            53   \n",
       "2  1119  19229018      0            20          53          69            59   \n",
       "3  2013  79277120      0            25          93         133            35   \n",
       "4   692  41528441      0            26         109          37            59   \n",
       "\n",
       "   adCategoryId  productId  productType  LBS  age appIdAction appIdInstall  \\\n",
       "0             5          0            1  490    1          -1           -1   \n",
       "1             4         14            3  299    3          -1           -1   \n",
       "2             4         28            3   82    4          -1           -1   \n",
       "3            23         10            2   41    2          -1           -1   \n",
       "4             8         11            3  283    1          -1           -1   \n",
       "\n",
       "   carrier  consumptionAbility  ct  education  gender  house  \\\n",
       "0        0                   1   0          1       1      0   \n",
       "1        1                   0  33          7       2      0   \n",
       "2        0                   0   0          1       1      0   \n",
       "3        1                   1  34          2       1      1   \n",
       "4        1                   1  10          7       1      0   \n",
       "\n",
       "                                           interest1  \\\n",
       "0                              70 100 48 49 23 36 11   \n",
       "1            93 109 75 29 59 49 6 5 73 46 22 36 11 3   \n",
       "2                    93 37 47 75 29 46 17 9 56 36 11   \n",
       "3                                              49 18   \n",
       "4  70 109 12 75 76 106 29 59 49 41 111 85 17 56 1...   \n",
       "\n",
       "                                           interest2 interest3 interest4  \\\n",
       "0                                                 -1        -1        -1   \n",
       "1                                                 -1        -1        -1   \n",
       "2                                                 -1        -1        -1   \n",
       "3                           42 9 1 46 70 14 29 10 35        -1        -1   \n",
       "4  47 42 73 9 31 46 32 70 21 33 19 60 13 74 16 55...        -1        -1   \n",
       "\n",
       "                       interest5                                 kw1  \\\n",
       "0                             -1    124532 762727 771871 8561 610048   \n",
       "1                100 131 37 8 46   531499 542834 303322 503318 68211   \n",
       "2                             -1  255920 190937 576486 460368 135689   \n",
       "3         92 116 128 42 5 121 64                       101914 294138   \n",
       "4  100 72 37 11 4 79 113 6 46 59  155585 562294 112872 273650 371405   \n",
       "\n",
       "                               kw2 kw3  marriageStatus  os  \\\n",
       "0                               -1  -1               1   0   \n",
       "1  89232 30101 117462 117548 85039  -1               6   1   \n",
       "2                               -1  -1               7   0   \n",
       "3    57664 32928 40670 11395 79112  -1               1   1   \n",
       "4   67951 40801 56571 119723 16159  -1               1   1   \n",
       "\n",
       "                     topic1                  topic2 topic3  \n",
       "0  4908 9745 7070 4203 3292                      -1     -1  \n",
       "1   8445 6762 7608 523 7688  4518 862 3643 702 5247     -1  \n",
       "2   941 4674 9735 8953 5352                      -1     -1  \n",
       "3  4538 6790 8850 3422 1940  2979 1877 358 666 4166     -1  \n",
       "4    4535 5590 47 5570 2078  856 4587 1877 392 9103     -1  "
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "train=data[data.label!=-1]\n",
    "train_y=train.pop('label')\n",
    "# train, test, train_y, test_y = train_test_split(train,train_y,test_size=0.2, random_state=2018)\n",
    "test=data[data.label==-1]\n",
    "res=test[['aid','uid']]\n",
    "test=test.drop('label',axis=1)\n",
    "enc = OneHotEncoder()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_x=train[['creativeSize']] # 这是广告牌大小，当然不能编码啦，否则失去意义\n",
    "test_x=test[['creativeSize']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "one-hot prepared !\n"
     ]
    }
   ],
   "source": [
    "for feature in one_hot_feature:\n",
    "    enc.fit(data[feature].values.reshape(-1, 1))\n",
    "    train_a=enc.transform(train[feature].values.reshape(-1, 1)) \n",
    "    test_a = enc.transform(test[feature].values.reshape(-1, 1))\n",
    "    train_x= sparse.hstack((train_x, train_a))\n",
    "    test_x = sparse.hstack((test_x, test_a))\n",
    "print('one-hot prepared !')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cv=CountVectorizer()\n",
    "for feature in vector_feature:\n",
    "    cv.fit(data[feature])\n",
    "    train_a = cv.transform(train[feature])\n",
    "    test_a = cv.transform(test[feature])\n",
    "    train_x = sparse.hstack((train_x, train_a))\n",
    "    test_x = sparse.hstack((test_x, test_a))\n",
    "print('cv prepared !')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def LGB_test(train_x,train_y,test_x,test_y):\n",
    "    print(\"LGB test\")\n",
    "    clf = lgb.LGBMClassifier(\n",
    "        boosting_type='gbdt', num_leaves=31, reg_alpha=0.0, reg_lambda=1,\n",
    "        max_depth=-1, n_estimators=1000, objective='binary',\n",
    "        subsample=0.7, colsample_bytree=0.7, subsample_freq=1,\n",
    "        learning_rate=0.05, min_child_weight=50,random_state=2018,n_jobs=-1\n",
    "    )\n",
    "    clf.fit(train_x, train_y,eval_set=[(train_x, train_y),(test_x,test_y)],eval_metric='auc',early_stopping_rounds=100)\n",
    "    # print(clf.feature_importances_)\n",
    "    return clf,clf.best_score_[ 'valid_1']['auc']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def LGB_predict(train_x,train_y,test_x,res):\n",
    "    print(\"LGB test\")\n",
    "    clf = lgb.LGBMClassifier(\n",
    "        boosting_type='gbdt', num_leaves=31, reg_alpha=0.0, reg_lambda=1,\n",
    "        max_depth=-1, n_estimators=1500, objective='binary',\n",
    "        subsample=0.7, colsample_bytree=0.7, subsample_freq=1,\n",
    "        learning_rate=0.05, min_child_weight=50, random_state=2018, n_jobs=-1\n",
    "    )\n",
    "    clf.fit(train_x, train_y, eval_set=[(train_x, train_y)], eval_metric='auc',early_stopping_rounds=100)\n",
    "    res['score'] = clf.predict_proba(test_x)[:,1]\n",
    "    res['score'] = res['score'].apply(lambda x: float('%.6f' % x))\n",
    "    res.to_csv('./dataAfterProcess/submission.csv', index=False)\n",
    "    os.system('zip ./dataAfterProcess/baseline.zip ./dataAfterProcess/submission.csv')\n",
    "    return clf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model=LGB_predict(train_x,train_y,test_x,res)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}