Untitled.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Environment Modules\n",
    "* Scikit-learn - 0.20.0\n",
    "* XGBoost - 0.90\n",
    "* numpy - 1.15.4\n",
    "* pandas - 0.24.2\n",
    "* seaborn - 0.9.0\n",
    "* matplotlib - 3.1.0\n",
    "* jupyter - 5.7.8\n  ",
    "## Used Training data set\n",
    "[WSDM](https://www.kaggle.com/c/kkbox-music-recommendation-challenge/data) - KKBox's Music Recommendation Challenge by isrc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_val_score, GridSearchCV, train_test_split\n",
    "from sklearn import metrics, ensemble\n",
    "import xgboost as xgb\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib as mpl\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "plt.style.use('ggplot')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('./data/train.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Sample of items(1%)\n",
    "* 0.01 > 0.1%\n",
    "* 0.1 > 1%"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.sample(frac=0.01)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load and Merge _songs_ data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "songs = pd.read_csv('./data/songs.csv')\n",
    "df = pd.merge(df, songs, on='song_id', how='left')\n",
    "del songs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load and Merge _members_ data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "members = pd.read_csv('./data/members.csv')\n",
    "df = pd.merge(df, members, on='msno', how='left')\n",
    "del members"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### DataFrame Information\n",
    "* Total Index : 73,774,170(100%) -> 1.0+@(__GB__)\n",
    "* Used Index : 737,741(1%) -> 106.9+@(__MB__)\n",
    "* predict Index : 7,374(0.1%) - > 10.7+@(__MB__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 73774 entries, 0 to 73773\n",
      "Data columns (total 18 columns):\n",
      "msno                      73774 non-null object\n",
      "song_id                   73774 non-null object\n",
      "source_system_tab         73536 non-null object\n",
      "source_screen_name        69764 non-null object\n",
      "source_type               73560 non-null object\n",
      "target                    73774 non-null int64\n",
      "song_length               73773 non-null float64\n",
      "genre_ids                 72574 non-null object\n",
      "artist_name               73773 non-null object\n",
      "composer                  57207 non-null object\n",
      "lyricist                  42168 non-null object\n",
      "language                  73773 non-null float64\n",
      "city                      73774 non-null int64\n",
      "bd                        73774 non-null int64\n",
      "gender                    44410 non-null object\n",
      "registered_via            73774 non-null int64\n",
      "registration_init_time    73774 non-null int64\n",
      "expiration_date           73774 non-null int64\n",
      "dtypes: float64(2), int64(6), object(10)\n",
      "memory usage: 10.7+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Null Data Count(%)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "msno                       0.000000\n",
       "song_id                    0.000000\n",
       "source_system_tab          0.322607\n",
       "source_screen_name         5.435519\n",
       "source_type                0.290075\n",
       "target                     0.000000\n",
       "song_length                0.001355\n",
       "genre_ids                  1.626589\n",
       "artist_name                0.001355\n",
       "composer                  22.456421\n",
       "lyricist                  42.841652\n",
       "language                   0.001355\n",
       "city                       0.000000\n",
       "bd                         0.000000\n",
       "gender                    39.802640\n",
       "registered_via             0.000000\n",
       "registration_init_time     0.000000\n",
       "expiration_date            0.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()/df.isnull().count()*100"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Replace(delete) NULL >> unknown"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in df.select_dtypes(include=['object']).columns:\n",
    "    df[i][df[i].isnull()] = 'unknown'\n",
    "df = df.fillna(value=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create Dates\n",
    "Date format :%Y%m%d  \n",
    "Used each date format"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "# registration_init_time(초기 등록 일자)\n",
    "df.registration_init_time = pd.to_datetime(df.registration_init_time, format='%Y%m%d', errors='ignore')\n",
    "df['registration_init_time_year'] = df['registration_init_time'].dt.year\n",
    "df['registration_init_time_month'] = df['registration_init_time'].dt.month\n",
    "df['registration_init_time_day'] = df['registration_init_time'].dt.day\n",
    "\n",
    "# expiration_date(마감 일자)\n",
    "df.expiration_date = pd.to_datetime(df.expiration_date,  format='%Y%m%d', errors='ignore')\n",
    "df['expiration_date_year'] = df['expiration_date'].dt.year\n",
    "df['expiration_date_month'] = df['expiration_date'].dt.month\n",
    "df['expiration_date_day'] = df['expiration_date'].dt.day"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "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>msno</th>\n",
       "      <th>song_id</th>\n",
       "      <th>source_system_tab</th>\n",
       "      <th>source_screen_name</th>\n",
       "      <th>source_type</th>\n",
       "      <th>target</th>\n",
       "      <th>song_length</th>\n",
       "      <th>genre_ids</th>\n",
       "      <th>artist_name</th>\n",
       "      <th>composer</th>\n",
       "      <th>...</th>\n",
       "      <th>gender</th>\n",
       "      <th>registered_via</th>\n",
       "      <th>registration_init_time</th>\n",
       "      <th>expiration_date</th>\n",
       "      <th>registration_init_time_year</th>\n",
       "      <th>registration_init_time_month</th>\n",
       "      <th>registration_init_time_day</th>\n",
       "      <th>expiration_date_year</th>\n",
       "      <th>expiration_date_month</th>\n",
       "      <th>expiration_date_day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ATnGnyeNcYFZKKRPtpQW6SkUpfSUl0Uz5Gg3Ru0Qwg8=</td>\n",
       "      <td>fdRAokHGxZlwa4pNIaed5LXuTq2i2SDNe/9IqKEMkqc=</td>\n",
       "      <td>my library</td>\n",
       "      <td>Local playlist more</td>\n",
       "      <td>local-playlist</td>\n",
       "      <td>1</td>\n",
       "      <td>229537.0</td>\n",
       "      <td>1259</td>\n",
       "      <td>Various Artists</td>\n",
       "      <td>unknown</td>\n",
       "      <td>...</td>\n",
       "      <td>unknown</td>\n",
       "      <td>7</td>\n",
       "      <td>2016-01-07</td>\n",
       "      <td>2017-10-07</td>\n",
       "      <td>2016</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>2017</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CA+85S3YBVtbz1jrRhPmuQFjMk+8a1CExDKLGJ77+Js=</td>\n",
       "      <td>vLkGTr3TJU9TGTJ1XBHCr4UoGCwaxqGPIHB05UeIscI=</td>\n",
       "      <td>search</td>\n",
       "      <td>Artist more</td>\n",
       "      <td>top-hits-for-artist</td>\n",
       "      <td>0</td>\n",
       "      <td>255059.0</td>\n",
       "      <td>465</td>\n",
       "      <td>李宇春 (Chris Lee)</td>\n",
       "      <td>陳奐仁</td>\n",
       "      <td>...</td>\n",
       "      <td>male</td>\n",
       "      <td>9</td>\n",
       "      <td>2011-02-17</td>\n",
       "      <td>2017-10-27</td>\n",
       "      <td>2011</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>2017</td>\n",
       "      <td>10</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>t9m6n01Sr6VIw278Vp3OYoZ9p9Q750+U8T+obxy2IEQ=</td>\n",
       "      <td>tbM8ByK21+iSDWZlT1yNgffPdQ5h9Wbk0hxS5RHQaCo=</td>\n",
       "      <td>my library</td>\n",
       "      <td>Local playlist more</td>\n",
       "      <td>local-library</td>\n",
       "      <td>0</td>\n",
       "      <td>249312.0</td>\n",
       "      <td>458</td>\n",
       "      <td>16個夏天 電視原聲帶</td>\n",
       "      <td>Peter Mansson| Vicky Karagiorgos</td>\n",
       "      <td>...</td>\n",
       "      <td>female</td>\n",
       "      <td>9</td>\n",
       "      <td>2011-10-11</td>\n",
       "      <td>2017-07-11</td>\n",
       "      <td>2011</td>\n",
       "      <td>10</td>\n",
       "      <td>11</td>\n",
       "      <td>2017</td>\n",
       "      <td>7</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>YN+zi4cXn8pNoA2qS0gtewM2bkFErdf8KDnqh47SBqQ=</td>\n",
       "      <td>PggZ9ma8Vpqehjl6iYVMLAhuKWDsMzKZnChkfA5bdvw=</td>\n",
       "      <td>discover</td>\n",
       "      <td>Search</td>\n",
       "      <td>online-playlist</td>\n",
       "      <td>0</td>\n",
       "      <td>219715.0</td>\n",
       "      <td>458</td>\n",
       "      <td>楊炅翰 &amp; 嚴藝丹 (Johan &amp; Ivyan)</td>\n",
       "      <td>Johan</td>\n",
       "      <td>...</td>\n",
       "      <td>female</td>\n",
       "      <td>3</td>\n",
       "      <td>2015-06-13</td>\n",
       "      <td>2018-01-04</td>\n",
       "      <td>2015</td>\n",
       "      <td>6</td>\n",
       "      <td>13</td>\n",
       "      <td>2018</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>uUsGH020Ypo4Oa4cAdGSkHpeWyjFDRd45flQKrPmhGg=</td>\n",
       "      <td>tBGEEPX42d7Vlb+XFKKSv2W7PIv+74GvkIEvV3KY+Ok=</td>\n",
       "      <td>discover</td>\n",
       "      <td>unknown</td>\n",
       "      <td>song</td>\n",
       "      <td>0</td>\n",
       "      <td>239177.0</td>\n",
       "      <td>465</td>\n",
       "      <td>郁可唯 (Yisa Yu)</td>\n",
       "      <td>潘協慶</td>\n",
       "      <td>...</td>\n",
       "      <td>female</td>\n",
       "      <td>7</td>\n",
       "      <td>2016-03-11</td>\n",
       "      <td>2017-10-02</td>\n",
       "      <td>2016</td>\n",
       "      <td>3</td>\n",
       "      <td>11</td>\n",
       "      <td>2017</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           msno  \\\n",
       "0  ATnGnyeNcYFZKKRPtpQW6SkUpfSUl0Uz5Gg3Ru0Qwg8=   \n",
       "1  CA+85S3YBVtbz1jrRhPmuQFjMk+8a1CExDKLGJ77+Js=   \n",
       "2  t9m6n01Sr6VIw278Vp3OYoZ9p9Q750+U8T+obxy2IEQ=   \n",
       "3  YN+zi4cXn8pNoA2qS0gtewM2bkFErdf8KDnqh47SBqQ=   \n",
       "4  uUsGH020Ypo4Oa4cAdGSkHpeWyjFDRd45flQKrPmhGg=   \n",
       "\n",
       "                                        song_id source_system_tab  \\\n",
       "0  fdRAokHGxZlwa4pNIaed5LXuTq2i2SDNe/9IqKEMkqc=        my library   \n",
       "1  vLkGTr3TJU9TGTJ1XBHCr4UoGCwaxqGPIHB05UeIscI=            search   \n",
       "2  tbM8ByK21+iSDWZlT1yNgffPdQ5h9Wbk0hxS5RHQaCo=        my library   \n",
       "3  PggZ9ma8Vpqehjl6iYVMLAhuKWDsMzKZnChkfA5bdvw=          discover   \n",
       "4  tBGEEPX42d7Vlb+XFKKSv2W7PIv+74GvkIEvV3KY+Ok=          discover   \n",
       "\n",
       "    source_screen_name          source_type  target  song_length genre_ids  \\\n",
       "0  Local playlist more       local-playlist       1     229537.0      1259   \n",
       "1          Artist more  top-hits-for-artist       0     255059.0       465   \n",
       "2  Local playlist more        local-library       0     249312.0       458   \n",
       "3               Search      online-playlist       0     219715.0       458   \n",
       "4              unknown                 song       0     239177.0       465   \n",
       "\n",
       "                 artist_name                          composer  ...   gender  \\\n",
       "0            Various Artists                           unknown  ...  unknown   \n",
       "1            李宇春 (Chris Lee)                               陳奐仁  ...     male   \n",
       "2                16個夏天 電視原聲帶  Peter Mansson| Vicky Karagiorgos  ...   female   \n",
       "3  楊炅翰 & 嚴藝丹 (Johan & Ivyan)                             Johan  ...   female   \n",
       "4              郁可唯 (Yisa Yu)                               潘協慶  ...   female   \n",
       "\n",
       "   registered_via  registration_init_time  expiration_date  \\\n",
       "0               7              2016-01-07       2017-10-07   \n",
       "1               9              2011-02-17       2017-10-27   \n",
       "2               9              2011-10-11       2017-07-11   \n",
       "3               3              2015-06-13       2018-01-04   \n",
       "4               7              2016-03-11       2017-10-02   \n",
       "\n",
       "  registration_init_time_year  registration_init_time_month  \\\n",
       "0                        2016                             1   \n",
       "1                        2011                             2   \n",
       "2                        2011                            10   \n",
       "3                        2015                             6   \n",
       "4                        2016                             3   \n",
       "\n",
       "  registration_init_time_day expiration_date_year  expiration_date_month  \\\n",
       "0                          7                 2017                     10   \n",
       "1                         17                 2017                     10   \n",
       "2                         11                 2017                      7   \n",
       "3                         13                 2018                      1   \n",
       "4                         11                 2017                     10   \n",
       "\n",
       "   expiration_date_day  \n",
       "0                    7  \n",
       "1                   27  \n",
       "2                   11  \n",
       "3                    4  \n",
       "4                    2  \n",
       "\n",
       "[5 rows x 24 columns]"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Type of category\n",
    "make date group for each date format"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Date to category\n",
    "df['registration_init_time'] = df['registration_init_time'].astype('category')\n",
    "df['expiration_date'] = df['expiration_date'].astype('category')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "# Object data to category\n",
    "for col in df.select_dtypes(include=['object']).columns:\n",
    "    df[col] = df[col].astype('category')\n",
    "    \n",
    "# Encoding categorical features\n",
    "for col in df.select_dtypes(include=['category']).columns:\n",
    "    df[col] = df[col].cat.codes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Сorrelation matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=[10,8])\n",
    "sns.heatmap(df.corr())\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Drop columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop(['expiration_date', 'lyricist'], 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train & Test split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "target = df.pop('target')\n",
    "train_data, test_data, train_labels, test_labels = train_test_split(df, target, test_size = 0.3)\n",
    "\n",
    "# Delete dataframe\n",
    "del df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create XGBoost model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "       colsample_bynode=1, colsample_bytree=1, gamma=0, learning_rate=0.1,\n",
       "       max_delta_step=0, max_depth=15, min_child_weight=5, missing=None,\n",
       "       n_estimators=250, n_jobs=1, nthread=None,\n",
       "       objective='binary:logistic', random_state=0, reg_alpha=0,\n",
       "       reg_lambda=1, scale_pos_weight=1, seed=None, silent=None,\n",
       "       subsample=1, verbosity=1)"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = xgb.XGBClassifier(learning_rate=0.1, max_depth=15, min_child_weight=5, n_estimators=250)\n",
    "model.fit(train_data, train_labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 1, ..., 1, 1, 0], dtype=int64)"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predict_labels = model.predict(test_data)\n",
    "predict_labels"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Accuracy\n",
    "* Total Index : 73,774,170(100%) -> (__Memory over__)\n",
    "* Used Index : 737,741(1%) -> (__68%__)\n",
    "* predict Index : 7,374(0.1%) - > (__63%__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.63      0.60      0.62     10921\n",
      "           1       0.63      0.65      0.64     11212\n",
      "\n",
      "   micro avg       0.63      0.63      0.63     22133\n",
      "   macro avg       0.63      0.63      0.63     22133\n",
      "weighted avg       0.63      0.63      0.63     22133\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(metrics.classification_report(test_labels, predict_labels))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}