content-based.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Load the Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd \n",
    "\n",
    "# train\n",
    "train_click = pd.read_csv(\"data/train_click_log.csv\")\n",
    "articles = pd.read_csv(\"data/articles.csv\")\n",
    "articles = articles.rename(columns={'article_id': 'click_article_id'})  #重命名，方便后续match\n",
    "articles_emb = pd.read_csv(\"data/articles_emb.csv\")\n",
    "\n",
    "# test\n",
    "test_click = pd.read_csv(\"data/testA_click_log.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import faiss\n",
    "from tqdm import tqdm\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "# only consider the content of test click\n",
    "train_click = test_click"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>article_id</th>\n",
       "      <th>emb_0</th>\n",
       "      <th>emb_1</th>\n",
       "      <th>emb_2</th>\n",
       "      <th>emb_3</th>\n",
       "      <th>emb_4</th>\n",
       "      <th>emb_5</th>\n",
       "      <th>emb_6</th>\n",
       "      <th>emb_7</th>\n",
       "      <th>emb_8</th>\n",
       "      <th>...</th>\n",
       "      <th>emb_240</th>\n",
       "      <th>emb_241</th>\n",
       "      <th>emb_242</th>\n",
       "      <th>emb_243</th>\n",
       "      <th>emb_244</th>\n",
       "      <th>emb_245</th>\n",
       "      <th>emb_246</th>\n",
       "      <th>emb_247</th>\n",
       "      <th>emb_248</th>\n",
       "      <th>emb_249</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>-0.161183</td>\n",
       "      <td>-0.957233</td>\n",
       "      <td>-0.137944</td>\n",
       "      <td>0.050855</td>\n",
       "      <td>0.830055</td>\n",
       "      <td>0.901365</td>\n",
       "      <td>-0.335148</td>\n",
       "      <td>-0.559561</td>\n",
       "      <td>-0.500603</td>\n",
       "      <td>...</td>\n",
       "      <td>0.321248</td>\n",
       "      <td>0.313999</td>\n",
       "      <td>0.636412</td>\n",
       "      <td>0.169179</td>\n",
       "      <td>0.540524</td>\n",
       "      <td>-0.813182</td>\n",
       "      <td>0.286870</td>\n",
       "      <td>-0.231686</td>\n",
       "      <td>0.597416</td>\n",
       "      <td>0.409623</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>-0.523216</td>\n",
       "      <td>-0.974058</td>\n",
       "      <td>0.738608</td>\n",
       "      <td>0.155234</td>\n",
       "      <td>0.626294</td>\n",
       "      <td>0.485297</td>\n",
       "      <td>-0.715657</td>\n",
       "      <td>-0.897996</td>\n",
       "      <td>-0.359747</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.487843</td>\n",
       "      <td>0.823124</td>\n",
       "      <td>0.412688</td>\n",
       "      <td>-0.338654</td>\n",
       "      <td>0.320786</td>\n",
       "      <td>0.588643</td>\n",
       "      <td>-0.594137</td>\n",
       "      <td>0.182828</td>\n",
       "      <td>0.397090</td>\n",
       "      <td>-0.834364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>-0.619619</td>\n",
       "      <td>-0.972960</td>\n",
       "      <td>-0.207360</td>\n",
       "      <td>-0.128861</td>\n",
       "      <td>0.044748</td>\n",
       "      <td>-0.387535</td>\n",
       "      <td>-0.730477</td>\n",
       "      <td>-0.066126</td>\n",
       "      <td>-0.754899</td>\n",
       "      <td>...</td>\n",
       "      <td>0.454756</td>\n",
       "      <td>0.473184</td>\n",
       "      <td>0.377866</td>\n",
       "      <td>-0.863887</td>\n",
       "      <td>-0.383365</td>\n",
       "      <td>0.137721</td>\n",
       "      <td>-0.810877</td>\n",
       "      <td>-0.447580</td>\n",
       "      <td>0.805932</td>\n",
       "      <td>-0.285284</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>-0.740843</td>\n",
       "      <td>-0.975749</td>\n",
       "      <td>0.391698</td>\n",
       "      <td>0.641738</td>\n",
       "      <td>-0.268645</td>\n",
       "      <td>0.191745</td>\n",
       "      <td>-0.825593</td>\n",
       "      <td>-0.710591</td>\n",
       "      <td>-0.040099</td>\n",
       "      <td>...</td>\n",
       "      <td>0.271535</td>\n",
       "      <td>0.036040</td>\n",
       "      <td>0.480029</td>\n",
       "      <td>-0.763173</td>\n",
       "      <td>0.022627</td>\n",
       "      <td>0.565165</td>\n",
       "      <td>-0.910286</td>\n",
       "      <td>-0.537838</td>\n",
       "      <td>0.243541</td>\n",
       "      <td>-0.885329</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>-0.279052</td>\n",
       "      <td>-0.972315</td>\n",
       "      <td>0.685374</td>\n",
       "      <td>0.113056</td>\n",
       "      <td>0.238315</td>\n",
       "      <td>0.271913</td>\n",
       "      <td>-0.568816</td>\n",
       "      <td>0.341194</td>\n",
       "      <td>-0.600554</td>\n",
       "      <td>...</td>\n",
       "      <td>0.238286</td>\n",
       "      <td>0.809268</td>\n",
       "      <td>0.427521</td>\n",
       "      <td>-0.615932</td>\n",
       "      <td>-0.503697</td>\n",
       "      <td>0.614450</td>\n",
       "      <td>-0.917760</td>\n",
       "      <td>-0.424061</td>\n",
       "      <td>0.185484</td>\n",
       "      <td>-0.580292</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 251 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   article_id     emb_0     emb_1     emb_2     emb_3     emb_4     emb_5  \\\n",
       "0           0 -0.161183 -0.957233 -0.137944  0.050855  0.830055  0.901365   \n",
       "1           1 -0.523216 -0.974058  0.738608  0.155234  0.626294  0.485297   \n",
       "2           2 -0.619619 -0.972960 -0.207360 -0.128861  0.044748 -0.387535   \n",
       "3           3 -0.740843 -0.975749  0.391698  0.641738 -0.268645  0.191745   \n",
       "4           4 -0.279052 -0.972315  0.685374  0.113056  0.238315  0.271913   \n",
       "\n",
       "      emb_6     emb_7     emb_8  ...   emb_240   emb_241   emb_242   emb_243  \\\n",
       "0 -0.335148 -0.559561 -0.500603  ...  0.321248  0.313999  0.636412  0.169179   \n",
       "1 -0.715657 -0.897996 -0.359747  ... -0.487843  0.823124  0.412688 -0.338654   \n",
       "2 -0.730477 -0.066126 -0.754899  ...  0.454756  0.473184  0.377866 -0.863887   \n",
       "3 -0.825593 -0.710591 -0.040099  ...  0.271535  0.036040  0.480029 -0.763173   \n",
       "4 -0.568816  0.341194 -0.600554  ...  0.238286  0.809268  0.427521 -0.615932   \n",
       "\n",
       "    emb_244   emb_245   emb_246   emb_247   emb_248   emb_249  \n",
       "0  0.540524 -0.813182  0.286870 -0.231686  0.597416  0.409623  \n",
       "1  0.320786  0.588643 -0.594137  0.182828  0.397090 -0.834364  \n",
       "2 -0.383365  0.137721 -0.810877 -0.447580  0.805932 -0.285284  \n",
       "3  0.022627  0.565165 -0.910286 -0.537838  0.243541 -0.885329  \n",
       "4 -0.503697  0.614450 -0.917760 -0.424061  0.185484 -0.580292  \n",
       "\n",
       "[5 rows x 251 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "articles_emb.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Get the Article Embedding Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(364047, 250)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "item_emb_cols = [x for x in articles_emb.columns if 'emb' in x]\n",
    "item_emb_np = np.ascontiguousarray(articles_emb[item_emb_cols].values, dtype=np.float32)\n",
    "item_emb_np = item_emb_np / np.linalg.norm(item_emb_np, axis=1, keepdims=True)  # normalize\n",
    "item_emb_np.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3. Get the User Profile Embedding Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_item_time_pair(df):\n",
    "    return list(zip(df['click_article_id'], df['click_timestamp']))\n",
    "\n",
    "train_click = train_click.sort_values('click_timestamp')\n",
    "\n",
    "user_item_time_df = train_click.groupby('user_id')['click_article_id', 'click_timestamp']\\\n",
    "                    .apply(lambda x: make_item_time_pair(x))\\\n",
    "                    .reset_index().rename(columns={0: 'item_time_list'})\n",
    "user_item_time_dict = dict(zip(user_item_time_df['user_id'], user_item_time_df['item_time_list']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(195839, 1507030363999), (191971, 1507030393999), (194300, 1507651461280)]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_item_time_dict[200000]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 50000/50000 [07:09<00:00, 116.45it/s]\n"
     ]
    }
   ],
   "source": [
    "user_profile = dict()\n",
    "\n",
    "for user_id, item_time_list in tqdm(user_item_time_dict.items()): \n",
    "    articles_emb_sum = 0\n",
    "    for i, i_click_time in item_time_list:  # i: article id \n",
    "        articles_emb_sum += articles_emb.iloc[i][1:]\n",
    "    user_profile[user_id] = articles_emb_sum / len(item_time_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>emb_0</th>\n",
       "      <th>emb_1</th>\n",
       "      <th>emb_2</th>\n",
       "      <th>emb_3</th>\n",
       "      <th>emb_4</th>\n",
       "      <th>emb_5</th>\n",
       "      <th>emb_6</th>\n",
       "      <th>emb_7</th>\n",
       "      <th>emb_8</th>\n",
       "      <th>emb_9</th>\n",
       "      <th>...</th>\n",
       "      <th>emb_240</th>\n",
       "      <th>emb_241</th>\n",
       "      <th>emb_242</th>\n",
       "      <th>emb_243</th>\n",
       "      <th>emb_244</th>\n",
       "      <th>emb_245</th>\n",
       "      <th>emb_246</th>\n",
       "      <th>emb_247</th>\n",
       "      <th>emb_248</th>\n",
       "      <th>emb_249</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>200000</th>\n",
       "      <td>-0.707095</td>\n",
       "      <td>-0.966142</td>\n",
       "      <td>0.215613</td>\n",
       "      <td>-0.540992</td>\n",
       "      <td>-0.097314</td>\n",
       "      <td>-0.711129</td>\n",
       "      <td>-0.388401</td>\n",
       "      <td>-0.214854</td>\n",
       "      <td>0.070674</td>\n",
       "      <td>-0.683168</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.247624</td>\n",
       "      <td>-0.250399</td>\n",
       "      <td>0.712416</td>\n",
       "      <td>-0.092248</td>\n",
       "      <td>0.391472</td>\n",
       "      <td>0.133109</td>\n",
       "      <td>0.304943</td>\n",
       "      <td>0.722092</td>\n",
       "      <td>0.112895</td>\n",
       "      <td>0.090947</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>200001</th>\n",
       "      <td>-0.826997</td>\n",
       "      <td>-0.962994</td>\n",
       "      <td>-0.415602</td>\n",
       "      <td>-0.804719</td>\n",
       "      <td>0.354680</td>\n",
       "      <td>-0.163473</td>\n",
       "      <td>-0.524052</td>\n",
       "      <td>-0.354726</td>\n",
       "      <td>0.079817</td>\n",
       "      <td>-0.082757</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.919703</td>\n",
       "      <td>-0.377261</td>\n",
       "      <td>-0.093302</td>\n",
       "      <td>0.559935</td>\n",
       "      <td>0.416835</td>\n",
       "      <td>-0.507327</td>\n",
       "      <td>0.212867</td>\n",
       "      <td>-0.435329</td>\n",
       "      <td>0.732461</td>\n",
       "      <td>0.171060</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>200002</th>\n",
       "      <td>-0.325285</td>\n",
       "      <td>-0.965357</td>\n",
       "      <td>-0.193873</td>\n",
       "      <td>-0.135098</td>\n",
       "      <td>-0.014686</td>\n",
       "      <td>0.087815</td>\n",
       "      <td>-0.581007</td>\n",
       "      <td>0.027498</td>\n",
       "      <td>-0.156805</td>\n",
       "      <td>0.038896</td>\n",
       "      <td>...</td>\n",
       "      <td>0.141499</td>\n",
       "      <td>-0.149350</td>\n",
       "      <td>0.100440</td>\n",
       "      <td>-0.060986</td>\n",
       "      <td>0.208628</td>\n",
       "      <td>-0.074240</td>\n",
       "      <td>-0.051672</td>\n",
       "      <td>0.074080</td>\n",
       "      <td>0.113742</td>\n",
       "      <td>0.192526</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>200003</th>\n",
       "      <td>-0.158326</td>\n",
       "      <td>-0.962052</td>\n",
       "      <td>0.077880</td>\n",
       "      <td>0.002835</td>\n",
       "      <td>-0.036167</td>\n",
       "      <td>0.184051</td>\n",
       "      <td>-0.406622</td>\n",
       "      <td>-0.042503</td>\n",
       "      <td>0.101619</td>\n",
       "      <td>-0.047257</td>\n",
       "      <td>...</td>\n",
       "      <td>0.040039</td>\n",
       "      <td>0.223023</td>\n",
       "      <td>-0.288795</td>\n",
       "      <td>-0.150053</td>\n",
       "      <td>0.205294</td>\n",
       "      <td>-0.067531</td>\n",
       "      <td>0.050926</td>\n",
       "      <td>0.128357</td>\n",
       "      <td>-0.062348</td>\n",
       "      <td>0.145443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>200004</th>\n",
       "      <td>-0.362147</td>\n",
       "      <td>-0.966331</td>\n",
       "      <td>-0.212863</td>\n",
       "      <td>-0.450467</td>\n",
       "      <td>-0.183054</td>\n",
       "      <td>-0.256189</td>\n",
       "      <td>-0.091871</td>\n",
       "      <td>-0.051382</td>\n",
       "      <td>0.013033</td>\n",
       "      <td>-0.139210</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.342469</td>\n",
       "      <td>-0.128845</td>\n",
       "      <td>-0.185921</td>\n",
       "      <td>-0.216158</td>\n",
       "      <td>0.152122</td>\n",
       "      <td>-0.001820</td>\n",
       "      <td>0.006968</td>\n",
       "      <td>0.133303</td>\n",
       "      <td>0.288489</td>\n",
       "      <td>0.272411</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 250 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           emb_0     emb_1     emb_2     emb_3     emb_4     emb_5     emb_6  \\\n",
       "200000 -0.707095 -0.966142  0.215613 -0.540992 -0.097314 -0.711129 -0.388401   \n",
       "200001 -0.826997 -0.962994 -0.415602 -0.804719  0.354680 -0.163473 -0.524052   \n",
       "200002 -0.325285 -0.965357 -0.193873 -0.135098 -0.014686  0.087815 -0.581007   \n",
       "200003 -0.158326 -0.962052  0.077880  0.002835 -0.036167  0.184051 -0.406622   \n",
       "200004 -0.362147 -0.966331 -0.212863 -0.450467 -0.183054 -0.256189 -0.091871   \n",
       "\n",
       "           emb_7     emb_8     emb_9  ...   emb_240   emb_241   emb_242  \\\n",
       "200000 -0.214854  0.070674 -0.683168  ... -0.247624 -0.250399  0.712416   \n",
       "200001 -0.354726  0.079817 -0.082757  ... -0.919703 -0.377261 -0.093302   \n",
       "200002  0.027498 -0.156805  0.038896  ...  0.141499 -0.149350  0.100440   \n",
       "200003 -0.042503  0.101619 -0.047257  ...  0.040039  0.223023 -0.288795   \n",
       "200004 -0.051382  0.013033 -0.139210  ... -0.342469 -0.128845 -0.185921   \n",
       "\n",
       "         emb_243   emb_244   emb_245   emb_246   emb_247   emb_248   emb_249  \n",
       "200000 -0.092248  0.391472  0.133109  0.304943  0.722092  0.112895  0.090947  \n",
       "200001  0.559935  0.416835 -0.507327  0.212867 -0.435329  0.732461  0.171060  \n",
       "200002 -0.060986  0.208628 -0.074240 -0.051672  0.074080  0.113742  0.192526  \n",
       "200003 -0.150053  0.205294 -0.067531  0.050926  0.128357 -0.062348  0.145443  \n",
       "200004 -0.216158  0.152122 -0.001820  0.006968  0.133303  0.288489  0.272411  \n",
       "\n",
       "[5 rows x 250 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "user_profile_emb = pd.DataFrame.from_dict(user_profile, orient='index')\n",
    " \n",
    "user_profile_emb.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50000, 250)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_profile_emb_np = np.ascontiguousarray(user_profile_emb[item_emb_cols].values, dtype=np.float32)\n",
    "user_profile_emb_np = user_profile_emb_np / np.linalg.norm(user_profile_emb_np, axis=1, keepdims=True)  # normalize\n",
    "user_profile_emb_np.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 4. Recommend"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50000, 10)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "topk = 10\n",
    "item_index = faiss.IndexFlatIP(item_emb_np.shape[1])  # Inner Product\n",
    "item_index.add(item_emb_np)\n",
    "\n",
    "# 相似度查询，给每个索引位置上的向量返回topk个item以及相似度\n",
    "sim, idx = item_index.search(user_profile_emb_np, topk) # 返回的是列表\n",
    "sim.shape  # 注意是排了顺序的"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.93727946, 0.9213254 , 0.92104375, 0.91865325, 0.9176518 ,\n",
       "       0.9175933 , 0.91751045, 0.9173319 , 0.9167841 , 0.91651976],\n",
       "      dtype=float32)"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sim[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>195839</td>\n",
       "      <td>195035</td>\n",
       "      <td>188845</td>\n",
       "      <td>188395</td>\n",
       "      <td>188018</td>\n",
       "      <td>188125</td>\n",
       "      <td>189021</td>\n",
       "      <td>194300</td>\n",
       "      <td>194920</td>\n",
       "      <td>189030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>175040</td>\n",
       "      <td>173466</td>\n",
       "      <td>173968</td>\n",
       "      <td>170466</td>\n",
       "      <td>172248</td>\n",
       "      <td>173533</td>\n",
       "      <td>172966</td>\n",
       "      <td>173021</td>\n",
       "      <td>171282</td>\n",
       "      <td>175486</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>299656</td>\n",
       "      <td>300704</td>\n",
       "      <td>205479</td>\n",
       "      <td>295398</td>\n",
       "      <td>299603</td>\n",
       "      <td>295905</td>\n",
       "      <td>300765</td>\n",
       "      <td>298673</td>\n",
       "      <td>301011</td>\n",
       "      <td>161269</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>161408</td>\n",
       "      <td>158039</td>\n",
       "      <td>157039</td>\n",
       "      <td>153006</td>\n",
       "      <td>157808</td>\n",
       "      <td>155685</td>\n",
       "      <td>162385</td>\n",
       "      <td>160231</td>\n",
       "      <td>159342</td>\n",
       "      <td>154567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>85630</td>\n",
       "      <td>285712</td>\n",
       "      <td>236065</td>\n",
       "      <td>233608</td>\n",
       "      <td>286064</td>\n",
       "      <td>47906</td>\n",
       "      <td>284110</td>\n",
       "      <td>235061</td>\n",
       "      <td>344756</td>\n",
       "      <td>285027</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49995</th>\n",
       "      <td>284110</td>\n",
       "      <td>284705</td>\n",
       "      <td>282761</td>\n",
       "      <td>279733</td>\n",
       "      <td>286064</td>\n",
       "      <td>284220</td>\n",
       "      <td>303288</td>\n",
       "      <td>284096</td>\n",
       "      <td>285372</td>\n",
       "      <td>47906</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49996</th>\n",
       "      <td>160974</td>\n",
       "      <td>156560</td>\n",
       "      <td>150616</td>\n",
       "      <td>161100</td>\n",
       "      <td>158828</td>\n",
       "      <td>159764</td>\n",
       "      <td>162765</td>\n",
       "      <td>159034</td>\n",
       "      <td>159299</td>\n",
       "      <td>162172</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49997</th>\n",
       "      <td>133022</td>\n",
       "      <td>132913</td>\n",
       "      <td>124638</td>\n",
       "      <td>133342</td>\n",
       "      <td>124065</td>\n",
       "      <td>123953</td>\n",
       "      <td>132119</td>\n",
       "      <td>123372</td>\n",
       "      <td>124090</td>\n",
       "      <td>124634</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49998</th>\n",
       "      <td>233492</td>\n",
       "      <td>208218</td>\n",
       "      <td>236121</td>\n",
       "      <td>246238</td>\n",
       "      <td>237264</td>\n",
       "      <td>157252</td>\n",
       "      <td>234681</td>\n",
       "      <td>233420</td>\n",
       "      <td>234356</td>\n",
       "      <td>236899</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49999</th>\n",
       "      <td>162504</td>\n",
       "      <td>214039</td>\n",
       "      <td>36011</td>\n",
       "      <td>162709</td>\n",
       "      <td>159581</td>\n",
       "      <td>159417</td>\n",
       "      <td>155317</td>\n",
       "      <td>158040</td>\n",
       "      <td>153023</td>\n",
       "      <td>157122</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>50000 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            0       1       2       3       4       5       6       7       8  \\\n",
       "0      195839  195035  188845  188395  188018  188125  189021  194300  194920   \n",
       "1      175040  173466  173968  170466  172248  173533  172966  173021  171282   \n",
       "2      299656  300704  205479  295398  299603  295905  300765  298673  301011   \n",
       "3      161408  158039  157039  153006  157808  155685  162385  160231  159342   \n",
       "4       85630  285712  236065  233608  286064   47906  284110  235061  344756   \n",
       "...       ...     ...     ...     ...     ...     ...     ...     ...     ...   \n",
       "49995  284110  284705  282761  279733  286064  284220  303288  284096  285372   \n",
       "49996  160974  156560  150616  161100  158828  159764  162765  159034  159299   \n",
       "49997  133022  132913  124638  133342  124065  123953  132119  123372  124090   \n",
       "49998  233492  208218  236121  246238  237264  157252  234681  233420  234356   \n",
       "49999  162504  214039   36011  162709  159581  159417  155317  158040  153023   \n",
       "\n",
       "            9  \n",
       "0      189030  \n",
       "1      175486  \n",
       "2      161269  \n",
       "3      154567  \n",
       "4      285027  \n",
       "...       ...  \n",
       "49995   47906  \n",
       "49996  162172  \n",
       "49997  124634  \n",
       "49998  236899  \n",
       "49999  157122  \n",
       "\n",
       "[50000 rows x 10 columns]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recall_df = pd.DataFrame(idx)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "submit = recall_df.iloc[:,:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>195839</td>\n",
       "      <td>195035</td>\n",
       "      <td>188845</td>\n",
       "      <td>188395</td>\n",
       "      <td>188018</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>175040</td>\n",
       "      <td>173466</td>\n",
       "      <td>173968</td>\n",
       "      <td>170466</td>\n",
       "      <td>172248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>299656</td>\n",
       "      <td>300704</td>\n",
       "      <td>205479</td>\n",
       "      <td>295398</td>\n",
       "      <td>299603</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>161408</td>\n",
       "      <td>158039</td>\n",
       "      <td>157039</td>\n",
       "      <td>153006</td>\n",
       "      <td>157808</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>85630</td>\n",
       "      <td>285712</td>\n",
       "      <td>236065</td>\n",
       "      <td>233608</td>\n",
       "      <td>286064</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        0       1       2       3       4\n",
       "0  195839  195035  188845  188395  188018\n",
       "1  175040  173466  173968  170466  172248\n",
       "2  299656  300704  205479  295398  299603\n",
       "3  161408  158039  157039  153006  157808\n",
       "4   85630  285712  236065  233608  286064"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "submit.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "submit[\"user_id\"] = submit.index + 200000\n",
    "submit = submit.rename(columns={0: 'article_1', 1: 'article_2', 2: 'article_3', 3: 'article_4', 4: 'article_5'})\n",
    "submit = submit[['user_id', 'article_1', 'article_2', 'article_3', 'article_4', 'article_5']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "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>user_id</th>\n",
       "      <th>article_1</th>\n",
       "      <th>article_2</th>\n",
       "      <th>article_3</th>\n",
       "      <th>article_4</th>\n",
       "      <th>article_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>200000</td>\n",
       "      <td>195839</td>\n",
       "      <td>195035</td>\n",
       "      <td>188845</td>\n",
       "      <td>188395</td>\n",
       "      <td>188018</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>200001</td>\n",
       "      <td>175040</td>\n",
       "      <td>173466</td>\n",
       "      <td>173968</td>\n",
       "      <td>170466</td>\n",
       "      <td>172248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>200002</td>\n",
       "      <td>299656</td>\n",
       "      <td>300704</td>\n",
       "      <td>205479</td>\n",
       "      <td>295398</td>\n",
       "      <td>299603</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>200003</td>\n",
       "      <td>161408</td>\n",
       "      <td>158039</td>\n",
       "      <td>157039</td>\n",
       "      <td>153006</td>\n",
       "      <td>157808</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>200004</td>\n",
       "      <td>85630</td>\n",
       "      <td>285712</td>\n",
       "      <td>236065</td>\n",
       "      <td>233608</td>\n",
       "      <td>286064</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   user_id  article_1  article_2  article_3  article_4  article_5\n",
       "0   200000     195839     195035     188845     188395     188018\n",
       "1   200001     175040     173466     173968     170466     172248\n",
       "2   200002     299656     300704     205479     295398     299603\n",
       "3   200003     161408     158039     157039     153006     157808\n",
       "4   200004      85630     285712     236065     233608     286064"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "submit.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "submit.to_csv(\"content_based_baseline.csv\", index=False, header=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python (Pytorch)",
   "language": "python",
   "name": "pytorch"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}