ch9-predict-future-sales-실전문제_part.ipynb

{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.14","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":8587,"databundleVersionId":868304,"sourceType":"competition"}],"dockerImageVersionId":30786,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# Predict Future Sales 실전 문제\n\n#### !_contents_!\n- 훈련 데이터에서 테스트 데이터에 있는 상점ID만 추출\n- 추가적인 피처 엔지니어링을 통한 성능 개선 시도\n\n# 1. Loading data, libraries (+data downcasting)","metadata":{}},{"cell_type":"code","source":"import numpy as np\nimport pandas as pd\n\nfrom datetime import datetime\nimport matplotlib.pyplot as plt\nimport seaborn as sns\n\nimport lightgbm as lgb\n\nimport warnings\nwarnings.filterwarnings(action='ignore')\n\n# 데이터 경로 지정\ndata_path = '/kaggle/input/competitive-data-science-predict-future-sales/'\n\nsales_train = pd.read_csv(data_path + 'sales_train.csv')\nshops = pd.read_csv(data_path + 'shops.csv')\nitems = pd.read_csv(data_path + 'items.csv')\nitem_categories = pd.read_csv(data_path + 'item_categories.csv')\ntest = pd.read_csv(data_path + 'test.csv')\nsubmission = pd.read_csv(data_path + 'sample_submission.csv')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:00.746238Z","iopub.execute_input":"2024-11-30T13:14:00.746685Z","iopub.status.idle":"2024-11-30T13:14:07.815860Z","shell.execute_reply.started":"2024-11-30T13:14:00.746647Z","shell.execute_reply":"2024-11-30T13:14:07.814654Z"}},"outputs":[],"execution_count":1},{"cell_type":"code","source":"sales_train = sales_train.rename(columns={'date': '날짜', \n                                          'date_block_num': '월ID',\n                                          'shop_id': '상점ID',\n                                          'item_id': '상품ID',\n                                          'item_price': '판매가',\n                                          'item_cnt_day': '판매량'})\n\nshops = shops.rename(columns={'shop_name': '상점명',\n                              'shop_id': '상점ID'})\n\nitems = items.rename(columns={'item_name': '상품명',\n                              'item_id': '상품ID',\n                              'item_category_id': '상품분류ID'})\n\nitem_categories = item_categories.rename(columns=\n                                         {'item_category_name': '상품분류명',\n                                          'item_category_id': '상품분류ID'})\n\ntest = test.rename(columns={'shop_id': '상점ID',\n                            'item_id': '상품ID'})","metadata":{"papermill":{"duration":0.171343,"end_time":"2021-09-11T07:54:02.82152","exception":false,"start_time":"2021-09-11T07:54:02.650177","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:07.817558Z","iopub.execute_input":"2024-11-30T13:14:07.817911Z","iopub.status.idle":"2024-11-30T13:14:07.907974Z","shell.execute_reply.started":"2024-11-30T13:14:07.817878Z","shell.execute_reply":"2024-11-30T13:14:07.906816Z"}},"outputs":[],"execution_count":2},{"cell_type":"code","source":"def downcast(df, verbose=True):\n    start_mem = df.memory_usage().sum() / 1024**2\n    for col in df.columns:\n        dtype_name = df[col].dtype.name\n        if dtype_name == 'object':\n            pass\n        elif dtype_name == 'bool':\n            df[col] = df[col].astype('int8')\n        elif dtype_name.startswith('int') or (df[col].round() == df[col]).all():\n            df[col] = pd.to_numeric(df[col], downcast='integer')\n        else:\n            df[col] = pd.to_numeric(df[col], downcast='float')\n    end_mem = df.memory_usage().sum() / 1024**2\n    if verbose:\n        print('{:.1f}% 압축됨'.format(100 * (start_mem - end_mem) / start_mem))\n    \n    return df\n\nall_df = [sales_train, shops, items, item_categories, test]\nfor df in all_df:\n    df = downcast(df)","metadata":{"papermill":{"duration":0.422472,"end_time":"2021-09-11T07:54:03.287596","exception":false,"start_time":"2021-09-11T07:54:02.865124","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:07.908906Z","iopub.execute_input":"2024-11-30T13:14:07.909205Z","iopub.status.idle":"2024-11-30T13:14:08.258142Z","shell.execute_reply.started":"2024-11-30T13:14:07.909177Z","shell.execute_reply":"2024-11-30T13:14:08.257068Z"}},"outputs":[{"name":"stdout","text":"54.2% 압축됨\n38.6% 압축됨\n54.2% 압축됨\n39.9% 압축됨\n70.8% 압축됨\n","output_type":"stream"}],"execution_count":3},{"cell_type":"markdown","source":"# 2. preprocessing","metadata":{}},{"cell_type":"code","source":"# sales price 0~50000\nsales_train = sales_train[sales_train['판매가'] > 0]\nsales_train = sales_train[sales_train['판매가'] < 50000]\n# sales quantity 0~1000\nsales_train = sales_train[sales_train['판매량'] > 0]\nsales_train = sales_train[sales_train['판매량'] < 1000]","metadata":{"papermill":{"duration":0.525625,"end_time":"2021-09-11T07:54:03.858663","exception":false,"start_time":"2021-09-11T07:54:03.333038","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:08.260086Z","iopub.execute_input":"2024-11-30T13:14:08.260442Z","iopub.status.idle":"2024-11-30T13:14:08.858081Z","shell.execute_reply.started":"2024-11-30T13:14:08.260410Z","shell.execute_reply":"2024-11-30T13:14:08.857040Z"}},"outputs":[],"execution_count":4},{"cell_type":"code","source":"# considering some shops are same each other\nprint(shops['상점명'][0], '||', shops['상점명'][57])\nprint(shops['상점명'][1], '||', shops['상점명'][58])\nprint(shops['상점명'][10], '||', shops['상점명'][11])\nprint(shops['상점명'][39], '||', shops['상점명'][40])\n\n# in sales_train\nsales_train.loc[sales_train['상점ID'] == 0, '상점ID'] = 57\nsales_train.loc[sales_train['상점ID'] == 1, '상점ID'] = 58\nsales_train.loc[sales_train['상점ID'] == 10, '상점ID'] = 11\nsales_train.loc[sales_train['상점ID'] == 39, '상점ID'] = 40\n\n# in test\ntest.loc[test['상점ID'] == 0, '상점ID'] = 57\ntest.loc[test['상점ID'] == 1, '상점ID'] = 58\ntest.loc[test['상점ID'] == 10, '상점ID'] = 11\ntest.loc[test['상점ID'] == 39, '상점ID'] = 40","metadata":{"papermill":{"duration":0.058305,"end_time":"2021-09-11T07:54:03.962839","exception":false,"start_time":"2021-09-11T07:54:03.904534","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:08.859448Z","iopub.execute_input":"2024-11-30T13:14:08.859857Z","iopub.status.idle":"2024-11-30T13:14:08.892117Z","shell.execute_reply.started":"2024-11-30T13:14:08.859815Z","shell.execute_reply":"2024-11-30T13:14:08.891073Z"}},"outputs":[{"name":"stdout","text":"!Якутск Орджоникидзе, 56 фран || Якутск Орджоникидзе, 56\n!Якутск ТЦ \"Центральный\" фран || Якутск ТЦ \"Центральный\"\nЖуковский ул. Чкалова 39м? || Жуковский ул. Чкалова 39м²\nРостовНаДону ТРК \"Мегацентр Горизонт\" || РостовНаДону ТРК \"Мегацентр Горизонт\" Островной\n","output_type":"stream"}],"execution_count":5},{"cell_type":"markdown","source":"## 🔍 Selecting Store IDs Present in the Test Dataset(실전문제 실습)","metadata":{"papermill":{"duration":0.045234,"end_time":"2021-09-11T07:54:04.171455","exception":false,"start_time":"2021-09-11T07:54:04.126221","status":"completed"},"tags":[]}},{"cell_type":"code","source":"# 테스트 데이터에 있는 상점ID만 추출\nunique_test_shop_id = test['상점ID'].unique()\nsales_train = sales_train[sales_train['상점ID'].isin(unique_test_shop_id)]","metadata":{"papermill":{"duration":0.157325,"end_time":"2021-09-11T07:54:04.375122","exception":false,"start_time":"2021-09-11T07:54:04.217797","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:08.893513Z","iopub.execute_input":"2024-11-30T13:14:08.893938Z","iopub.status.idle":"2024-11-30T13:14:09.026814Z","shell.execute_reply.started":"2024-11-30T13:14:08.893891Z","shell.execute_reply":"2024-11-30T13:14:09.025803Z"}},"outputs":[],"execution_count":6},{"cell_type":"code","source":"shops['도시'] = shops['상점명'].apply(lambda x: x.split()[0])","metadata":{"papermill":{"duration":0.056743,"end_time":"2021-09-11T07:54:04.477338","exception":false,"start_time":"2021-09-11T07:54:04.420595","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:09.028225Z","iopub.execute_input":"2024-11-30T13:14:09.028658Z","iopub.status.idle":"2024-11-30T13:14:09.035720Z","shell.execute_reply.started":"2024-11-30T13:14:09.028613Z","shell.execute_reply":"2024-11-30T13:14:09.034637Z"}},"outputs":[],"execution_count":7},{"cell_type":"code","source":"shops['도시'].unique()","metadata":{"papermill":{"duration":0.08467,"end_time":"2021-09-11T07:54:04.60678","exception":false,"start_time":"2021-09-11T07:54:04.52211","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:09.037043Z","iopub.execute_input":"2024-11-30T13:14:09.037652Z","iopub.status.idle":"2024-11-30T13:14:09.056528Z","shell.execute_reply.started":"2024-11-30T13:14:09.037620Z","shell.execute_reply":"2024-11-30T13:14:09.055475Z"}},"outputs":[{"execution_count":8,"output_type":"execute_result","data":{"text/plain":"array(['!Якутск', 'Адыгея', 'Балашиха', 'Волжский', 'Вологда', 'Воронеж',\n       'Выездная', 'Жуковский', 'Интернет-магазин', 'Казань', 'Калуга',\n       'Коломна', 'Красноярск', 'Курск', 'Москва', 'Мытищи', 'Н.Новгород',\n       'Новосибирск', 'Омск', 'РостовНаДону', 'СПб', 'Самара', 'Сергиев',\n       'Сургут', 'Томск', 'Тюмень', 'Уфа', 'Химки', 'Цифровой', 'Чехов',\n       'Якутск', 'Ярославль'], dtype=object)"},"metadata":{}}],"execution_count":8},{"cell_type":"code","source":"shops.loc[shops['도시'] =='!Якутск', '도시'] = 'Якутск'","metadata":{"papermill":{"duration":0.062415,"end_time":"2021-09-11T07:54:04.725994","exception":false,"start_time":"2021-09-11T07:54:04.663579","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:09.057950Z","iopub.execute_input":"2024-11-30T13:14:09.058460Z","iopub.status.idle":"2024-11-30T13:14:09.068689Z","shell.execute_reply.started":"2024-11-30T13:14:09.058426Z","shell.execute_reply":"2024-11-30T13:14:09.067591Z"}},"outputs":[],"execution_count":9},{"cell_type":"code","source":"from sklearn.preprocessing import LabelEncoder\n\n# 레이블 인코더 생성\nlabel_encoder = LabelEncoder()\n# 도시 피처 레이블 인코딩\nshops['도시'] = label_encoder.fit_transform(shops['도시'])","metadata":{"papermill":{"duration":0.929735,"end_time":"2021-09-11T07:54:05.703749","exception":false,"start_time":"2021-09-11T07:54:04.774014","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:09.073485Z","iopub.execute_input":"2024-11-30T13:14:09.073825Z","iopub.status.idle":"2024-11-30T13:14:09.081463Z","shell.execute_reply.started":"2024-11-30T13:14:09.073796Z","shell.execute_reply":"2024-11-30T13:14:09.080430Z"}},"outputs":[],"execution_count":10},{"cell_type":"markdown","source":"## 🔍 Removing `_name` Features","metadata":{}},{"cell_type":"code","source":"shops = shops.drop(['상점명'], axis=1)\nshops.head()","metadata":{"papermill":{"duration":0.063621,"end_time":"2021-09-11T07:54:05.813109","exception":false,"start_time":"2021-09-11T07:54:05.749488","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:09.082483Z","iopub.execute_input":"2024-11-30T13:14:09.082791Z","iopub.status.idle":"2024-11-30T13:14:09.106923Z","shell.execute_reply.started":"2024-11-30T13:14:09.082754Z","shell.execute_reply":"2024-11-30T13:14:09.105582Z"}},"outputs":[{"execution_count":11,"output_type":"execute_result","data":{"text/plain":"   상점ID  도시\n0     0  29\n1     1  29\n2     2   0\n3     3   1\n4     4   2","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>상점ID</th>\n      <th>도시</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>29</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>1</td>\n      <td>29</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>2</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>3</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>4</td>\n      <td>2</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}],"execution_count":11},{"cell_type":"code","source":"items = items.drop(['상품명'], axis=1)\nitems.head()","metadata":{"papermill":{"duration":0.053707,"end_time":"2021-09-11T07:54:05.911886","exception":false,"start_time":"2021-09-11T07:54:05.858179","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:09.108388Z","iopub.execute_input":"2024-11-30T13:14:09.108725Z","iopub.status.idle":"2024-11-30T13:14:09.119319Z","shell.execute_reply.started":"2024-11-30T13:14:09.108694Z","shell.execute_reply":"2024-11-30T13:14:09.118331Z"}},"outputs":[{"execution_count":12,"output_type":"execute_result","data":{"text/plain":"   상품ID  상품분류ID\n0     0      40\n1     1      76\n2     2      40\n3     3      40\n4     4      40","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>상품ID</th>\n      <th>상품분류ID</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>40</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>1</td>\n      <td>76</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>2</td>\n      <td>40</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>3</td>\n      <td>40</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>4</td>\n      <td>40</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}],"execution_count":12},{"cell_type":"markdown","source":"# 3. Feature Engineering: textbook ver.\n\n## 🔍 Creating Feature: `첫 판매월`, `대분류` ","metadata":{}},{"cell_type":"code","source":"# 상품이 맨 처음 팔린 날을 피처로 추가\nitems['첫 판매월'] = sales_train.groupby('상품ID').agg({'월ID': 'min'})['월ID']\nitems.head()","metadata":{"papermill":{"duration":0.124783,"end_time":"2021-09-11T07:54:06.081871","exception":false,"start_time":"2021-09-11T07:54:05.957088","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:09.120739Z","iopub.execute_input":"2024-11-30T13:14:09.121175Z","iopub.status.idle":"2024-11-30T13:14:09.184705Z","shell.execute_reply.started":"2024-11-30T13:14:09.121123Z","shell.execute_reply":"2024-11-30T13:14:09.183613Z"}},"outputs":[{"execution_count":13,"output_type":"execute_result","data":{"text/plain":"   상품ID  상품분류ID  첫 판매월\n0     0      40    NaN\n1     1      76   15.0\n2     2      40    NaN\n3     3      40    NaN\n4     4      40    NaN","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>상품ID</th>\n      <th>상품분류ID</th>\n      <th>첫 판매월</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>40</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>1</td>\n      <td>76</td>\n      <td>15.0</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>2</td>\n      <td>40</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>3</td>\n      <td>40</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>4</td>\n      <td>40</td>\n      <td>NaN</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}],"execution_count":13},{"cell_type":"code","source":"items[items['첫 판매월'].isna()]","metadata":{"papermill":{"duration":0.064072,"end_time":"2021-09-11T07:54:06.19249","exception":false,"start_time":"2021-09-11T07:54:06.128418","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:09.185971Z","iopub.execute_input":"2024-11-30T13:14:09.186414Z","iopub.status.idle":"2024-11-30T13:14:09.200059Z","shell.execute_reply.started":"2024-11-30T13:14:09.186364Z","shell.execute_reply":"2024-11-30T13:14:09.199110Z"}},"outputs":[{"execution_count":14,"output_type":"execute_result","data":{"text/plain":"        상품ID  상품분류ID  첫 판매월\n0          0      40    NaN\n2          2      40    NaN\n3          3      40    NaN\n4          4      40    NaN\n5          5      40    NaN\n...      ...     ...    ...\n21975  21975      61    NaN\n22022  22022      40    NaN\n22028  22028      40    NaN\n22035  22035      40    NaN\n22137  22137      40    NaN\n\n[1090 rows x 3 columns]","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>상품ID</th>\n      <th>상품분류ID</th>\n      <th>첫 판매월</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>40</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>2</td>\n      <td>40</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>3</td>\n      <td>40</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>4</td>\n      <td>40</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>5</th>\n      <td>5</td>\n      <td>40</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>21975</th>\n      <td>21975</td>\n      <td>61</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>22022</th>\n      <td>22022</td>\n      <td>40</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>22028</th>\n      <td>22028</td>\n      <td>40</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>22035</th>\n      <td>22035</td>\n      <td>40</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>22137</th>\n      <td>22137</td>\n      <td>40</td>\n      <td>NaN</td>\n    </tr>\n  </tbody>\n</table>\n<p>1090 rows × 3 columns</p>\n</div>"},"metadata":{}}],"execution_count":14},{"cell_type":"code","source":"# 첫 판매월 피처의 결측값을 34로 대체\nitems['첫 판매월'] = items['첫 판매월'].fillna(34)","metadata":{"papermill":{"duration":0.055152,"end_time":"2021-09-11T07:54:06.294514","exception":false,"start_time":"2021-09-11T07:54:06.239362","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:09.201529Z","iopub.execute_input":"2024-11-30T13:14:09.201938Z","iopub.status.idle":"2024-11-30T13:14:09.212598Z","shell.execute_reply.started":"2024-11-30T13:14:09.201897Z","shell.execute_reply":"2024-11-30T13:14:09.211411Z"}},"outputs":[],"execution_count":15},{"cell_type":"code","source":"# 상품분류명의 첫 단어를 대분류로 추출\nitem_categories['대분류'] = item_categories['상품분류명'].apply(lambda x: x.split()[0])  ","metadata":{"papermill":{"duration":0.058487,"end_time":"2021-09-11T07:54:06.399546","exception":false,"start_time":"2021-09-11T07:54:06.341059","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:09.214467Z","iopub.execute_input":"2024-11-30T13:14:09.214819Z","iopub.status.idle":"2024-11-30T13:14:09.226123Z","shell.execute_reply.started":"2024-11-30T13:14:09.214785Z","shell.execute_reply":"2024-11-30T13:14:09.225000Z"}},"outputs":[],"execution_count":16},{"cell_type":"code","source":"item_categories['대분류'].value_counts()","metadata":{"papermill":{"duration":0.059877,"end_time":"2021-09-11T07:54:06.506786","exception":false,"start_time":"2021-09-11T07:54:06.446909","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:09.227255Z","iopub.execute_input":"2024-11-30T13:14:09.227620Z","iopub.status.idle":"2024-11-30T13:14:09.243499Z","shell.execute_reply.started":"2024-11-30T13:14:09.227589Z","shell.execute_reply":"2024-11-30T13:14:09.242352Z"}},"outputs":[{"execution_count":17,"output_type":"execute_result","data":{"text/plain":"대분류\nИгры          14\nКниги         13\nПодарки       12\nИгровые        8\nАксессуары     7\nМузыка         6\nПрограммы      6\nКарты          5\nКино           5\nСлужебные      2\nЧистые         2\nPC             1\nБилеты         1\nДоставка       1\nЭлементы       1\nName: count, dtype: int64"},"metadata":{}}],"execution_count":17},{"cell_type":"code","source":"def make_etc(x):\n    if len(item_categories[item_categories['대분류']==x]) >= 5:\n        return x\n    else:\n        return 'etc'\n\n# 5개 미만이면 'etc'로 통합\nitem_categories['대분류'] = item_categories['대분류'].apply(make_etc)\nitem_categories.head()","metadata":{"papermill":{"duration":0.087865,"end_time":"2021-09-11T07:54:06.641637","exception":false,"start_time":"2021-09-11T07:54:06.553772","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:09.244930Z","iopub.execute_input":"2024-11-30T13:14:09.245396Z","iopub.status.idle":"2024-11-30T13:14:09.295733Z","shell.execute_reply.started":"2024-11-30T13:14:09.245349Z","shell.execute_reply":"2024-11-30T13:14:09.294636Z"}},"outputs":[{"execution_count":18,"output_type":"execute_result","data":{"text/plain":"                     상품분류명  상품분류ID         대분류\n0  PC - Гарнитуры/Наушники       0         etc\n1         Аксессуары - PS2       1  Аксессуары\n2         Аксессуары - PS3       2  Аксессуары\n3         Аксессуары - PS4       3  Аксессуары\n4         Аксессуары - PSP       4  Аксессуары","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>상품분류명</th>\n      <th>상품분류ID</th>\n      <th>대분류</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>PC - Гарнитуры/Наушники</td>\n      <td>0</td>\n      <td>etc</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>Аксессуары - PS2</td>\n      <td>1</td>\n      <td>Аксессуары</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>Аксессуары - PS3</td>\n      <td>2</td>\n      <td>Аксессуары</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>Аксессуары - PS4</td>\n      <td>3</td>\n      <td>Аксессуары</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>Аксессуары - PSP</td>\n      <td>4</td>\n      <td>Аксессуары</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}],"execution_count":18},{"cell_type":"code","source":"label_encoder = LabelEncoder()\nitem_categories['대분류'] = label_encoder.fit_transform(item_categories['대분류'])\n\nitem_categories = item_categories.drop('상품분류명', axis=1)","metadata":{"papermill":{"duration":0.05765,"end_time":"2021-09-11T07:54:06.856297","exception":false,"start_time":"2021-09-11T07:54:06.798647","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:09.296968Z","iopub.execute_input":"2024-11-30T13:14:09.297387Z","iopub.status.idle":"2024-11-30T13:14:09.307298Z","shell.execute_reply.started":"2024-11-30T13:14:09.297353Z","shell.execute_reply":"2024-11-30T13:14:09.306219Z"}},"outputs":[],"execution_count":19},{"cell_type":"markdown","source":"## 🔍 Generating Features by Combining `월ID`, `상점ID`, `상품ID`","metadata":{}},{"cell_type":"code","source":"from itertools import product\n\ntrain = []\n# 월ID, 상점ID, 상품ID 조합 생성\nfor i in sales_train['월ID'].unique():\n    all_shop = sales_train.loc[sales_train['월ID']==i, '상점ID'].unique()\n    all_item = sales_train.loc[sales_train['월ID']==i, '상품ID'].unique()\n    train.append(np.array(list(product([i], all_shop, all_item))))\n\nidx_features = ['월ID', '상점ID', '상품ID'] # 기준 피처\ntrain = pd.DataFrame(np.vstack(train), columns=idx_features)","metadata":{"papermill":{"duration":17.704679,"end_time":"2021-09-11T07:54:24.610052","exception":false,"start_time":"2021-09-11T07:54:06.905373","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:09.308484Z","iopub.execute_input":"2024-11-30T13:14:09.309180Z","iopub.status.idle":"2024-11-30T13:14:14.414164Z","shell.execute_reply.started":"2024-11-30T13:14:09.309140Z","shell.execute_reply":"2024-11-30T13:14:14.413183Z"}},"outputs":[],"execution_count":20},{"cell_type":"code","source":"group = sales_train.groupby(idx_features).agg({'판매량': 'sum',\n                                               '판매가': 'mean'})\ngroup = group.reset_index()\ngroup = group.rename(columns={'판매량': '월간 판매량', '판매가': '평균 판매가'})\n\ntrain = train.merge(group, on=idx_features, how='left')\n\ntrain.head()","metadata":{"papermill":{"duration":5.084102,"end_time":"2021-09-11T07:54:29.741605","exception":false,"start_time":"2021-09-11T07:54:24.657503","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:14.415590Z","iopub.execute_input":"2024-11-30T13:14:14.416003Z","iopub.status.idle":"2024-11-30T13:14:18.919924Z","shell.execute_reply.started":"2024-11-30T13:14:14.415957Z","shell.execute_reply":"2024-11-30T13:14:18.918786Z"}},"outputs":[{"execution_count":21,"output_type":"execute_result","data":{"text/plain":"   월ID  상점ID   상품ID  월간 판매량  평균 판매가\n0    0    59  22154     1.0   999.0\n1    0    59   2552     NaN     NaN\n2    0    59   2554     NaN     NaN\n3    0    59   2555     NaN     NaN\n4    0    59   2564     NaN     NaN","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>월ID</th>\n      <th>상점ID</th>\n      <th>상품ID</th>\n      <th>월간 판매량</th>\n      <th>평균 판매가</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>59</td>\n      <td>22154</td>\n      <td>1.0</td>\n      <td>999.0</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0</td>\n      <td>59</td>\n      <td>2552</td>\n      <td>NaN</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0</td>\n      <td>59</td>\n      <td>2554</td>\n      <td>NaN</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0</td>\n      <td>59</td>\n      <td>2555</td>\n      <td>NaN</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0</td>\n      <td>59</td>\n      <td>2564</td>\n      <td>NaN</td>\n      <td>NaN</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}],"execution_count":21},{"cell_type":"code","source":"import gc\n\ndel group\ngc.collect();","metadata":{"papermill":{"duration":0.206373,"end_time":"2021-09-11T07:54:29.996939","exception":false,"start_time":"2021-09-11T07:54:29.790566","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:18.921205Z","iopub.execute_input":"2024-11-30T13:14:18.921652Z","iopub.status.idle":"2024-11-30T13:14:19.026204Z","shell.execute_reply.started":"2024-11-30T13:14:18.921607Z","shell.execute_reply":"2024-11-30T13:14:19.025022Z"}},"outputs":[],"execution_count":22},{"cell_type":"code","source":"# 상품 \"판매\"건수 피처 추가\ngroup = sales_train.groupby(idx_features).agg({'판매량': 'count'})\ngroup = group.reset_index()\ngroup = group.rename(columns={'판매량': '판매건수'})\n\ntrain = train.merge(group, on=idx_features, how='left')\n\ndel group, sales_train\ngc.collect()\n\ntrain.head()","metadata":{"papermill":{"duration":4.809101,"end_time":"2021-09-11T07:54:34.854265","exception":false,"start_time":"2021-09-11T07:54:30.045164","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:19.027865Z","iopub.execute_input":"2024-11-30T13:14:19.028308Z","iopub.status.idle":"2024-11-30T13:14:23.361858Z","shell.execute_reply.started":"2024-11-30T13:14:19.028236Z","shell.execute_reply":"2024-11-30T13:14:23.360706Z"}},"outputs":[{"execution_count":23,"output_type":"execute_result","data":{"text/plain":"   월ID  상점ID   상품ID  월간 판매량  평균 판매가  판매건수\n0    0    59  22154     1.0   999.0   1.0\n1    0    59   2552     NaN     NaN   NaN\n2    0    59   2554     NaN     NaN   NaN\n3    0    59   2555     NaN     NaN   NaN\n4    0    59   2564     NaN     NaN   NaN","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>월ID</th>\n      <th>상점ID</th>\n      <th>상품ID</th>\n      <th>월간 판매량</th>\n      <th>평균 판매가</th>\n      <th>판매건수</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>59</td>\n      <td>22154</td>\n      <td>1.0</td>\n      <td>999.0</td>\n      <td>1.0</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0</td>\n      <td>59</td>\n      <td>2552</td>\n      <td>NaN</td>\n      <td>NaN</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0</td>\n      <td>59</td>\n      <td>2554</td>\n      <td>NaN</td>\n      <td>NaN</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0</td>\n      <td>59</td>\n      <td>2555</td>\n      <td>NaN</td>\n      <td>NaN</td>\n      <td>NaN</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0</td>\n      <td>59</td>\n      <td>2564</td>\n      <td>NaN</td>\n      <td>NaN</td>\n      <td>NaN</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}],"execution_count":23},{"cell_type":"code","source":"# 테스트 데이터 월ID를 34로 설정\ntest['월ID'] = 34\n\n# train과 test 이어붙이기\nall_data = pd.concat([train, test.drop('ID', axis=1)],\n                     ignore_index=True,\n                     keys=idx_features)\n\n# 결측값을 0으로 대체\nall_data = all_data.fillna(0)\nall_data.head()","metadata":{"papermill":{"duration":0.489766,"end_time":"2021-09-11T07:54:35.39284","exception":false,"start_time":"2021-09-11T07:54:34.903074","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:23.362934Z","iopub.execute_input":"2024-11-30T13:14:23.363249Z","iopub.status.idle":"2024-11-30T13:14:23.717391Z","shell.execute_reply.started":"2024-11-30T13:14:23.363210Z","shell.execute_reply":"2024-11-30T13:14:23.716323Z"}},"outputs":[{"execution_count":24,"output_type":"execute_result","data":{"text/plain":"   월ID  상점ID   상품ID  월간 판매량  평균 판매가  판매건수\n0    0    59  22154     1.0   999.0   1.0\n1    0    59   2552     0.0     0.0   0.0\n2    0    59   2554     0.0     0.0   0.0\n3    0    59   2555     0.0     0.0   0.0\n4    0    59   2564     0.0     0.0   0.0","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>월ID</th>\n      <th>상점ID</th>\n      <th>상품ID</th>\n      <th>월간 판매량</th>\n      <th>평균 판매가</th>\n      <th>판매건수</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>59</td>\n      <td>22154</td>\n      <td>1.0</td>\n      <td>999.0</td>\n      <td>1.0</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0</td>\n      <td>59</td>\n      <td>2552</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0</td>\n      <td>59</td>\n      <td>2554</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0</td>\n      <td>59</td>\n      <td>2555</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0</td>\n      <td>59</td>\n      <td>2564</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}],"execution_count":24},{"cell_type":"code","source":"# 나머지 데이터 병합\nall_data = all_data.merge(shops, on='상점ID', how='left')\nall_data = all_data.merge(items, on='상품ID', how='left')\nall_data = all_data.merge(item_categories, on='상품분류ID', how='left')\n\n# 다운캐스팅\nall_data = downcast(all_data)\n\ndel shops, items, item_categories\ngc.collect();","metadata":{"papermill":{"duration":6.839293,"end_time":"2021-09-11T07:54:42.280558","exception":false,"start_time":"2021-09-11T07:54:35.441265","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:23.718775Z","iopub.execute_input":"2024-11-30T13:14:23.719112Z","iopub.status.idle":"2024-11-30T13:14:28.286747Z","shell.execute_reply.started":"2024-11-30T13:14:23.719081Z","shell.execute_reply":"2024-11-30T13:14:28.285586Z"}},"outputs":[{"name":"stdout","text":"68.9% 압축됨\n","output_type":"stream"}],"execution_count":25},{"cell_type":"code","source":"def add_mean_features(df, mean_features, idx_features):\n    # 기준 피처 확인\n    assert (idx_features[0] == '월ID') and \\\n           len(idx_features) in [2, 3]\n    \n    # 파생 피처명 설정 \n    if len(idx_features) == 2:\n        feature_name = idx_features[1] + '별 평균 판매량'\n    else:\n        feature_name = idx_features[1] + ' ' + idx_features[2] + '별 평균 판매량'\n    \n    # 기준 피처를 토대로 그룹화해 월간 평균 판매량 구하기\n    group = df.groupby(idx_features).agg({'월간 판매량': 'mean'})\n    group = group.reset_index()\n    group = group.rename(columns={'월간 판매량': feature_name})\n    \n    # df와 group 병합 \n    df = df.merge(group, on=idx_features, how='left')\n    # 다운캐스팅 \n    df = downcast(df, verbose=False)\n    # 새로 만든 feature_name 피처명을 mean_features 리스트에 추가 \n    mean_features.append(feature_name)\n\n    # garbage collection\n    del group\n    gc.collect()\n    \n    return df, mean_features","metadata":{"papermill":{"duration":0.061429,"end_time":"2021-09-11T07:54:42.391138","exception":false,"start_time":"2021-09-11T07:54:42.329709","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:28.288453Z","iopub.execute_input":"2024-11-30T13:14:28.288765Z","iopub.status.idle":"2024-11-30T13:14:28.295968Z","shell.execute_reply.started":"2024-11-30T13:14:28.288736Z","shell.execute_reply":"2024-11-30T13:14:28.294840Z"}},"outputs":[],"execution_count":26},{"cell_type":"markdown","source":"## 🔍 Creating lag features","metadata":{}},{"cell_type":"code","source":"# 그룹화 기준 피처 중 '상품ID'가 포함된 파생 피처명을 담을 리스트\nitem_mean_features = []\n\n# ['월ID', '상품ID']로 그룹화한 월간 평균 판매량 파생 피처 생성\nall_data, item_mean_features = add_mean_features(df=all_data,\n                                                 mean_features=item_mean_features,\n                                                 idx_features=['월ID', '상품ID'])\n\n# ['월ID', '상품ID', '도시']로 그룹화한 월간 평균 판매량 파생 피처 생성\nall_data, item_mean_features = add_mean_features(df=all_data,\n                                                 mean_features=item_mean_features,\n                                                 idx_features=['월ID', '상품ID', '도시'])","metadata":{"papermill":{"duration":14.281785,"end_time":"2021-09-11T07:54:56.721833","exception":false,"start_time":"2021-09-11T07:54:42.440048","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:28.297570Z","iopub.execute_input":"2024-11-30T13:14:28.298016Z","iopub.status.idle":"2024-11-30T13:14:38.275092Z","shell.execute_reply.started":"2024-11-30T13:14:28.297968Z","shell.execute_reply":"2024-11-30T13:14:38.274218Z"}},"outputs":[],"execution_count":27},{"cell_type":"code","source":"# 그룹화 기준 피처 중 '상점ID'가 포함된 파생 피처명을 담을 리스트\nshop_mean_features = []\n\n# ['월ID', '상점ID', '상품분류ID']로 그룹화한 월간 평균 판매량 파생 피처 생성\nall_data, shop_mean_features = add_mean_features(df=all_data, \n                                                 mean_features=shop_mean_features,\n                                                 idx_features=['월ID', '상점ID', '상품분류ID'])","metadata":{"papermill":{"duration":3.694415,"end_time":"2021-09-11T07:55:00.465365","exception":false,"start_time":"2021-09-11T07:54:56.77095","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:38.280938Z","iopub.execute_input":"2024-11-30T13:14:38.281293Z","iopub.status.idle":"2024-11-30T13:14:40.733617Z","shell.execute_reply.started":"2024-11-30T13:14:38.281241Z","shell.execute_reply":"2024-11-30T13:14:40.732730Z"}},"outputs":[],"execution_count":28},{"cell_type":"code","source":"def add_lag_features(df, lag_features_to_clip, idx_features, \n                     lag_feature, nlags=3, clip=False):\n    # 시차 피처 생성에 필요한 DataFrame 부분만 복사 \n    df_temp = df[idx_features + [lag_feature]].copy() \n\n    # 시차 피처 생성 \n    for i in range(1, nlags+1):\n        # 시차 피처명 \n        lag_feature_name = lag_feature +'_시차' + str(i)\n        # df_temp 열 이름 설정 \n        df_temp.columns = idx_features + [lag_feature_name]\n        # df_temp의 date_block_num 피처에 1 더하기 \n        df_temp['월ID'] += 1\n        # idx_feature를 기준으로 df와 df_temp 병합하기\n        df = df.merge(df_temp.drop_duplicates(), \n                      on=idx_features, \n                      how='left')\n        # 결측값 0으로 대체 \n        df[lag_feature_name] = df[lag_feature_name].fillna(0)\n        # 0 ~ 20 사이로 제한할 시차 피처명을 lag_features_to_clip에 추가\n        if clip: \n            lag_features_to_clip.append(lag_feature_name)\n    \n    # 데이터 다운캐스팅\n    df = downcast(df, False)\n    # 가비지 컬렉션\n    del df_temp\n    gc.collect()\n    \n    return df, lag_features_to_clip","metadata":{"papermill":{"duration":0.06144,"end_time":"2021-09-11T07:55:00.575477","exception":false,"start_time":"2021-09-11T07:55:00.514037","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:40.734830Z","iopub.execute_input":"2024-11-30T13:14:40.735147Z","iopub.status.idle":"2024-11-30T13:14:40.742565Z","shell.execute_reply.started":"2024-11-30T13:14:40.735116Z","shell.execute_reply":"2024-11-30T13:14:40.741447Z"}},"outputs":[],"execution_count":29},{"cell_type":"code","source":"lag_features_to_clip = [] # 0 ~ 20 사이로 제한할 시차 피처명을 담을 리스트\nidx_features = ['월ID', '상점ID', '상품ID'] # 기준 피처\n\n# idx_features를 기준으로 월간 판매량의 세 달치 시차 피처 생성\nall_data, lag_features_to_clip = add_lag_features(df=all_data, \n                                                  lag_features_to_clip=lag_features_to_clip,\n                                                  idx_features=idx_features,\n                                                  lag_feature='월간 판매량', \n                                                  nlags=3,\n                                                  clip=True)","metadata":{"papermill":{"duration":29.072942,"end_time":"2021-09-11T07:55:29.696778","exception":false,"start_time":"2021-09-11T07:55:00.623836","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:14:40.743875Z","iopub.execute_input":"2024-11-30T13:14:40.744171Z","iopub.status.idle":"2024-11-30T13:15:03.756711Z","shell.execute_reply.started":"2024-11-30T13:14:40.744141Z","shell.execute_reply":"2024-11-30T13:15:03.755651Z"}},"outputs":[],"execution_count":30},{"cell_type":"code","source":"all_data.head().T","metadata":{"papermill":{"duration":0.068651,"end_time":"2021-09-11T07:55:29.814402","exception":false,"start_time":"2021-09-11T07:55:29.745751","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:15:03.758064Z","iopub.execute_input":"2024-11-30T13:15:03.758505Z","iopub.status.idle":"2024-11-30T13:15:03.774557Z","shell.execute_reply.started":"2024-11-30T13:15:03.758460Z","shell.execute_reply":"2024-11-30T13:15:03.773109Z"}},"outputs":[{"execution_count":31,"output_type":"execute_result","data":{"text/plain":"                                0            1            2            3  \\\n월ID                      0.000000     0.000000     0.000000     0.000000   \n상점ID                    59.000000    59.000000    59.000000    59.000000   \n상품ID                 22154.000000  2552.000000  2554.000000  2555.000000   \n월간 판매량                   1.000000     0.000000     0.000000     0.000000   \n평균 판매가                 999.000000     0.000000     0.000000     0.000000   \n판매건수                     1.000000     0.000000     0.000000     0.000000   \n도시                      30.000000    30.000000    30.000000    30.000000   \n상품분류ID                  37.000000    58.000000    58.000000    56.000000   \n첫 판매월                    0.000000     0.000000     0.000000     0.000000   \n대분류                      5.000000     7.000000     7.000000     7.000000   \n상품ID별 평균 판매량             0.457143     0.028571     0.028571     0.028571   \n상품ID 도시별 평균 판매량          1.000000     0.000000     0.000000     0.000000   \n상점ID 상품분류ID별 평균 판매량      0.096774     0.000000     0.000000     0.011494   \n월간 판매량_시차1               0.000000     0.000000     0.000000     0.000000   \n월간 판매량_시차2               0.000000     0.000000     0.000000     0.000000   \n월간 판매량_시차3               0.000000     0.000000     0.000000     0.000000   \n\n                               4  \n월ID                     0.000000  \n상점ID                   59.000000  \n상품ID                 2564.000000  \n월간 판매량                  0.000000  \n평균 판매가                  0.000000  \n판매건수                    0.000000  \n도시                     30.000000  \n상품분류ID                 59.000000  \n첫 판매월                   0.000000  \n대분류                     7.000000  \n상품ID별 평균 판매량            0.142857  \n상품ID 도시별 평균 판매량         0.000000  \n상점ID 상품분류ID별 평균 판매량     0.013245  \n월간 판매량_시차1              0.000000  \n월간 판매량_시차2              0.000000  \n월간 판매량_시차3              0.000000  ","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>월ID</th>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n    </tr>\n    <tr>\n      <th>상점ID</th>\n      <td>59.000000</td>\n      <td>59.000000</td>\n      <td>59.000000</td>\n      <td>59.000000</td>\n      <td>59.000000</td>\n    </tr>\n    <tr>\n      <th>상품ID</th>\n      <td>22154.000000</td>\n      <td>2552.000000</td>\n      <td>2554.000000</td>\n      <td>2555.000000</td>\n      <td>2564.000000</td>\n    </tr>\n    <tr>\n      <th>월간 판매량</th>\n      <td>1.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n    </tr>\n    <tr>\n      <th>평균 판매가</th>\n      <td>999.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n    </tr>\n    <tr>\n      <th>판매건수</th>\n      <td>1.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n    </tr>\n    <tr>\n      <th>도시</th>\n      <td>30.000000</td>\n      <td>30.000000</td>\n      <td>30.000000</td>\n      <td>30.000000</td>\n      <td>30.000000</td>\n    </tr>\n    <tr>\n      <th>상품분류ID</th>\n      <td>37.000000</td>\n      <td>58.000000</td>\n      <td>58.000000</td>\n      <td>56.000000</td>\n      <td>59.000000</td>\n    </tr>\n    <tr>\n      <th>첫 판매월</th>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n    </tr>\n    <tr>\n      <th>대분류</th>\n      <td>5.000000</td>\n      <td>7.000000</td>\n      <td>7.000000</td>\n      <td>7.000000</td>\n      <td>7.000000</td>\n    </tr>\n    <tr>\n      <th>상품ID별 평균 판매량</th>\n      <td>0.457143</td>\n      <td>0.028571</td>\n      <td>0.028571</td>\n      <td>0.028571</td>\n      <td>0.142857</td>\n    </tr>\n    <tr>\n      <th>상품ID 도시별 평균 판매량</th>\n      <td>1.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n    </tr>\n    <tr>\n      <th>상점ID 상품분류ID별 평균 판매량</th>\n      <td>0.096774</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.011494</td>\n      <td>0.013245</td>\n    </tr>\n    <tr>\n      <th>월간 판매량_시차1</th>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n    </tr>\n    <tr>\n      <th>월간 판매량_시차2</th>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n    </tr>\n    <tr>\n      <th>월간 판매량_시차3</th>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}],"execution_count":31},{"cell_type":"code","source":"lag_features_to_clip","metadata":{"papermill":{"duration":0.060314,"end_time":"2021-09-11T07:55:29.923872","exception":false,"start_time":"2021-09-11T07:55:29.863558","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:15:03.776157Z","iopub.execute_input":"2024-11-30T13:15:03.776527Z","iopub.status.idle":"2024-11-30T13:15:03.789079Z","shell.execute_reply.started":"2024-11-30T13:15:03.776480Z","shell.execute_reply":"2024-11-30T13:15:03.787817Z"}},"outputs":[{"execution_count":32,"output_type":"execute_result","data":{"text/plain":"['월간 판매량_시차1', '월간 판매량_시차2', '월간 판매량_시차3']"},"metadata":{}}],"execution_count":32},{"cell_type":"code","source":"# idx_features를 기준으로 판매건수 피처의 세 달치 시차 피처 생성\nall_data, lag_features_to_clip = add_lag_features(df=all_data, \n                                                  lag_features_to_clip=lag_features_to_clip,\n                                                  idx_features=idx_features,\n                                                  lag_feature='판매건수', \n                                                  nlags=3)\n\n# idx_features를 기준으로 평균 판매가 피처의 세 달치 시차 피처 생성\nall_data, lag_features_to_clip = add_lag_features(df=all_data, \n                                                  lag_features_to_clip=lag_features_to_clip,\n                                                  idx_features=idx_features,\n                                                  lag_feature='평균 판매가', \n                                                  nlags=3)","metadata":{"papermill":{"duration":70.188378,"end_time":"2021-09-11T07:56:40.162392","exception":false,"start_time":"2021-09-11T07:55:29.974014","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:15:03.790483Z","iopub.execute_input":"2024-11-30T13:15:03.790837Z","iopub.status.idle":"2024-11-30T13:15:52.217665Z","shell.execute_reply.started":"2024-11-30T13:15:03.790805Z","shell.execute_reply":"2024-11-30T13:15:52.216658Z"}},"outputs":[],"execution_count":33},{"cell_type":"code","source":"# 예상해야하는 test 데이터 기준으로 34로 월 number 지정\nX_test_temp = all_data[all_data['월ID'] == 34]\nX_test_temp[item_mean_features].sum()","metadata":{"papermill":{"duration":0.381471,"end_time":"2021-09-11T07:56:40.59783","exception":false,"start_time":"2021-09-11T07:56:40.216359","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:15:52.219092Z","iopub.execute_input":"2024-11-30T13:15:52.219534Z","iopub.status.idle":"2024-11-30T13:15:52.259591Z","shell.execute_reply.started":"2024-11-30T13:15:52.219497Z","shell.execute_reply":"2024-11-30T13:15:52.258320Z"}},"outputs":[{"execution_count":34,"output_type":"execute_result","data":{"text/plain":"상품ID별 평균 판매량       0.0\n상품ID 도시별 평균 판매량    0.0\ndtype: float32"},"metadata":{}}],"execution_count":34},{"cell_type":"code","source":"# idx_features를 기준으로 item_mean_features 요소별 시차 피처 생성\nfor item_mean_feature in item_mean_features:\n    all_data, lag_features_to_clip = add_lag_features(df=all_data, \n                                                      lag_features_to_clip=lag_features_to_clip, \n                                                      idx_features=idx_features, \n                                                      lag_feature=item_mean_feature, \n                                                      nlags=3,\n                                                      clip=True)\n# item_mean_features 피처 제거\nall_data = all_data.drop(item_mean_features, axis=1)","metadata":{"papermill":{"duration":67.950839,"end_time":"2021-09-11T07:57:48.601381","exception":false,"start_time":"2021-09-11T07:56:40.650542","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:15:52.260732Z","iopub.execute_input":"2024-11-30T13:15:52.261061Z","iopub.status.idle":"2024-11-30T13:16:42.266179Z","shell.execute_reply.started":"2024-11-30T13:15:52.261030Z","shell.execute_reply":"2024-11-30T13:16:42.265172Z"}},"outputs":[],"execution_count":35},{"cell_type":"code","source":"shop_mean_features","metadata":{"papermill":{"duration":0.061003,"end_time":"2021-09-11T07:57:48.71523","exception":false,"start_time":"2021-09-11T07:57:48.654227","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:16:42.267251Z","iopub.execute_input":"2024-11-30T13:16:42.267610Z","iopub.status.idle":"2024-11-30T13:16:42.274716Z","shell.execute_reply.started":"2024-11-30T13:16:42.267571Z","shell.execute_reply":"2024-11-30T13:16:42.273659Z"}},"outputs":[{"execution_count":36,"output_type":"execute_result","data":{"text/plain":"['상점ID 상품분류ID별 평균 판매량']"},"metadata":{}}],"execution_count":36},{"cell_type":"code","source":"# ['월ID', '상점ID', '상품분류ID']를 기준으로 shop_mean_features 요소별 시차 피처 생성\nfor shop_mean_feature in shop_mean_features:\n    all_data, lag_features_to_clip = add_lag_features(df=all_data,\n                                                      lag_features_to_clip=lag_features_to_clip, \n                                                      idx_features=['월ID', '상점ID', '상품분류ID'], \n                                                      lag_feature=shop_mean_feature, \n                                                      nlags=3,\n                                                      clip=True)\n# shop_mean_features 피처 제거\nall_data = all_data.drop(shop_mean_features, axis=1)","metadata":{"papermill":{"duration":12.127324,"end_time":"2021-09-11T07:58:00.893865","exception":false,"start_time":"2021-09-11T07:57:48.766541","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:16:42.276142Z","iopub.execute_input":"2024-11-30T13:16:42.276671Z","iopub.status.idle":"2024-11-30T13:16:51.556858Z","shell.execute_reply.started":"2024-11-30T13:16:42.276613Z","shell.execute_reply":"2024-11-30T13:16:51.555758Z"}},"outputs":[],"execution_count":37},{"cell_type":"code","source":"# 월ID 3미만인 데이터 제거\nall_data = all_data.drop(all_data[all_data['월ID'] < 3].index)","metadata":{"papermill":{"duration":2.561672,"end_time":"2021-09-11T07:58:03.508126","exception":false,"start_time":"2021-09-11T07:58:00.946454","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:16:51.558211Z","iopub.execute_input":"2024-11-30T13:16:51.558644Z","iopub.status.idle":"2024-11-30T13:16:53.109689Z","shell.execute_reply.started":"2024-11-30T13:16:51.558607Z","shell.execute_reply":"2024-11-30T13:16:53.108569Z"}},"outputs":[],"execution_count":38},{"cell_type":"markdown","source":"### 🔍 Generating Derived Features\n","metadata":{}},{"cell_type":"code","source":"all_data['월간 판매량 시차평균'] = all_data[['월간 판매량_시차1',\n                                          '월간 판매량_시차2', \n                                          '월간 판매량_시차3']].mean(axis=1)","metadata":{"papermill":{"duration":0.230902,"end_time":"2021-09-11T07:58:03.789932","exception":false,"start_time":"2021-09-11T07:58:03.55903","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:16:53.111315Z","iopub.execute_input":"2024-11-30T13:16:53.111784Z","iopub.status.idle":"2024-11-30T13:16:54.217676Z","shell.execute_reply.started":"2024-11-30T13:16:53.111735Z","shell.execute_reply":"2024-11-30T13:16:54.216470Z"}},"outputs":[],"execution_count":39},{"cell_type":"code","source":"# 0 ~ 20 사이로 값 제한\nall_data[lag_features_to_clip + ['월간 판매량', '월간 판매량 시차평균']] = all_data[lag_features_to_clip +['월간 판매량', '월간 판매량 시차평균']].clip(0, 20)","metadata":{"papermill":{"duration":3.005294,"end_time":"2021-09-11T07:58:06.846142","exception":false,"start_time":"2021-09-11T07:58:03.840848","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:16:54.219155Z","iopub.execute_input":"2024-11-30T13:16:54.219626Z","iopub.status.idle":"2024-11-30T13:16:56.279918Z","shell.execute_reply.started":"2024-11-30T13:16:54.219580Z","shell.execute_reply":"2024-11-30T13:16:56.278329Z"}},"outputs":[],"execution_count":40},{"cell_type":"code","source":"all_data['시차변화량1'] = all_data['월간 판매량_시차1']/all_data['월간 판매량_시차2']\nall_data['시차변화량1'] = all_data['시차변화량1'].replace([np.inf, -np.inf], \n                                                        np.nan).fillna(0)\n\nall_data['시차변화량2'] = all_data['월간 판매량_시차2']/all_data['월간 판매량_시차3']\nall_data['시차변화량2'] = all_data['시차변화량2'].replace([np.inf, -np.inf], \n                                                        np.nan).fillna(0)","metadata":{"papermill":{"duration":0.613439,"end_time":"2021-09-11T07:58:07.510465","exception":false,"start_time":"2021-09-11T07:58:06.897026","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:16:56.281786Z","iopub.execute_input":"2024-11-30T13:16:56.282297Z","iopub.status.idle":"2024-11-30T13:16:56.756041Z","shell.execute_reply.started":"2024-11-30T13:16:56.282218Z","shell.execute_reply":"2024-11-30T13:16:56.754956Z"}},"outputs":[],"execution_count":41},{"cell_type":"code","source":"all_data['신상여부'] = all_data['첫 판매월'] == all_data['월ID']","metadata":{"papermill":{"duration":0.06782,"end_time":"2021-09-11T07:58:07.628495","exception":false,"start_time":"2021-09-11T07:58:07.560675","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:16:56.757394Z","iopub.execute_input":"2024-11-30T13:16:56.758353Z","iopub.status.idle":"2024-11-30T13:16:56.768161Z","shell.execute_reply.started":"2024-11-30T13:16:56.758211Z","shell.execute_reply":"2024-11-30T13:16:56.767048Z"}},"outputs":[],"execution_count":42},{"cell_type":"code","source":"all_data['첫 판매 후 기간'] = all_data['월ID'] - all_data['첫 판매월']","metadata":{"papermill":{"duration":0.068014,"end_time":"2021-09-11T07:58:07.748119","exception":false,"start_time":"2021-09-11T07:58:07.680105","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:16:56.769408Z","iopub.execute_input":"2024-11-30T13:16:56.769799Z","iopub.status.idle":"2024-11-30T13:16:56.786481Z","shell.execute_reply.started":"2024-11-30T13:16:56.769766Z","shell.execute_reply":"2024-11-30T13:16:56.785214Z"}},"outputs":[],"execution_count":43},{"cell_type":"code","source":"all_data['월'] = all_data['월ID'] % 12","metadata":{"papermill":{"duration":0.094063,"end_time":"2021-09-11T07:58:07.892834","exception":false,"start_time":"2021-09-11T07:58:07.798771","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:16:56.787992Z","iopub.execute_input":"2024-11-30T13:16:56.788587Z","iopub.status.idle":"2024-11-30T13:16:56.830054Z","shell.execute_reply.started":"2024-11-30T13:16:56.788524Z","shell.execute_reply":"2024-11-30T13:16:56.829014Z"}},"outputs":[],"execution_count":44},{"cell_type":"code","source":"# 첫 판매월, 평균 판매가, 판매건수 피처 제거\nall_data = all_data.drop(['첫 판매월', '평균 판매가', '판매건수'], axis=1)","metadata":{"papermill":{"duration":0.950936,"end_time":"2021-09-11T07:58:08.893811","exception":false,"start_time":"2021-09-11T07:58:07.942875","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:16:56.831594Z","iopub.execute_input":"2024-11-30T13:16:56.832009Z","iopub.status.idle":"2024-11-30T13:16:57.768739Z","shell.execute_reply.started":"2024-11-30T13:16:56.831967Z","shell.execute_reply":"2024-11-30T13:16:57.767222Z"}},"outputs":[],"execution_count":45},{"cell_type":"markdown","source":"# 4. Training a Model and Making Predictions : `Public Score: 0.87701`\n\nUsing `all_data` to `'submission-authur1.csv'`","metadata":{}},{"cell_type":"code","source":"all_data = downcast(all_data, False) # 데이터 다운캐스팅","metadata":{"papermill":{"duration":1.812834,"end_time":"2021-09-11T07:58:10.759095","exception":false,"start_time":"2021-09-11T07:58:08.946261","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:16:57.770218Z","iopub.execute_input":"2024-11-30T13:16:57.770608Z","iopub.status.idle":"2024-11-30T13:16:59.066093Z","shell.execute_reply.started":"2024-11-30T13:16:57.770573Z","shell.execute_reply":"2024-11-30T13:16:59.065256Z"}},"outputs":[],"execution_count":46},{"cell_type":"code","source":"all_data.info()","metadata":{"papermill":{"duration":0.070224,"end_time":"2021-09-11T07:58:10.87972","exception":false,"start_time":"2021-09-11T07:58:10.809496","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:16:59.067186Z","iopub.execute_input":"2024-11-30T13:16:59.067516Z","iopub.status.idle":"2024-11-30T13:16:59.078775Z","shell.execute_reply.started":"2024-11-30T13:16:59.067486Z","shell.execute_reply":"2024-11-30T13:16:59.077668Z"}},"outputs":[{"name":"stdout","text":"<class 'pandas.core.frame.DataFrame'>\nRangeIndex: 8029765 entries, 827196 to 8856960\nData columns (total 31 columns):\n #   Column                   Dtype  \n---  ------                   -----  \n 0   월ID                      int8   \n 1   상점ID                     int8   \n 2   상품ID                     int16  \n 3   월간 판매량                   int8   \n 4   도시                       int8   \n 5   상품분류ID                   int8   \n 6   대분류                      int8   \n 7   월간 판매량_시차1               int8   \n 8   월간 판매량_시차2               int8   \n 9   월간 판매량_시차3               int8   \n 10  판매건수_시차1                 int8   \n 11  판매건수_시차2                 int8   \n 12  판매건수_시차3                 int8   \n 13  평균 판매가_시차1               float64\n 14  평균 판매가_시차2               float64\n 15  평균 판매가_시차3               float64\n 16  상품ID별 평균 판매량_시차1         float32\n 17  상품ID별 평균 판매량_시차2         float32\n 18  상품ID별 평균 판매량_시차3         float32\n 19  상품ID 도시별 평균 판매량_시차1      float32\n 20  상품ID 도시별 평균 판매량_시차2      float32\n 21  상품ID 도시별 평균 판매량_시차3      float32\n 22  상점ID 상품분류ID별 평균 판매량_시차1  float32\n 23  상점ID 상품분류ID별 평균 판매량_시차2  float32\n 24  상점ID 상품분류ID별 평균 판매량_시차3  float32\n 25  월간 판매량 시차평균              float32\n 26  시차변화량1                   float32\n 27  시차변화량2                   float32\n 28  신상여부                     int8   \n 29  첫 판매 후 기간                int8   \n 30  월                        int8   \ndtypes: float32(12), float64(3), int16(1), int8(15)\nmemory usage: 681.5 MB\n","output_type":"stream"}],"execution_count":47},{"cell_type":"code","source":"# 훈련 데이터 (피처)\nX_train = all_data[all_data['월ID'] < 33]\nX_train = X_train.drop(['월간 판매량'], axis=1)\n# 검증 데이터 (피처)\nX_valid = all_data[all_data['월ID'] == 33]\nX_valid = X_valid.drop(['월간 판매량'], axis=1)\n# 테스트 데이터 (피처)\nX_test = all_data[all_data['월ID'] == 34]\nX_test = X_test.drop(['월간 판매량'], axis=1)\n\n# 훈련 데이터 (타깃값)\ny_train = all_data[all_data['월ID'] < 33]['월간 판매량']\n# 검증 데이터 (타깃값)\ny_valid = all_data[all_data['월ID'] == 33]['월간 판매량']\n\n# # 가비지 컬렉션 : 뒤에서 전처리를 지속적으로 이용할 것으로 수행 X\n# del all_data\n# gc.collect();","metadata":{"papermill":{"duration":2.713052,"end_time":"2021-09-11T07:58:13.643914","exception":false,"start_time":"2021-09-11T07:58:10.930862","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:16:59.080426Z","iopub.execute_input":"2024-11-30T13:16:59.080885Z","iopub.status.idle":"2024-11-30T13:17:02.663606Z","shell.execute_reply.started":"2024-11-30T13:16:59.080837Z","shell.execute_reply":"2024-11-30T13:17:02.662464Z"}},"outputs":[],"execution_count":48},{"cell_type":"code","source":"# import lightgbm as lgb\nfrom lightgbm import early_stopping, log_evaluation # 버전차이로 지원하지 않는 부분 설정\n\nstart_time = datetime.now()\n\n# LightGBM 하이퍼파라미터\nparams = {'metric': 'rmse',  # 평가지표: RMSE\n          'num_leaves': 255,\n          'learning_rate': 0.005,  # 이전 학습율 0.01\n          'feature_fraction': 0.75,  # 추가1\n          'bagging_fraction': 0.75,  # 추가2\n          'bagging_freq': 5,  # 추가3\n          'force_col_wise': True,\n          'random_state': 10}\n\n# categorical 피처 설정 : 도시, 대분류, 월 추가\ncat_features = ['상점ID', '도시', '상품분류ID', '대분류', '월']\n\n# LightGBM 훈련 및 검증 데이터셋\ndtrain = lgb.Dataset(X_train, y_train)\ndvalid = lgb.Dataset(X_valid, y_valid)\n\n\n# LightGBM 모델 훈련\n\n# 100회 반복마다 검증지표 출력(verbose_eval 파라미터 대체)\n# early_stopping_rounds=150 대체 추가\ncallbacks = [log_evaluation(period=100),  # verbose_eval 대체\n            early_stopping(stopping_rounds=150)]  # early_stopping_rounds 대체\n\nlgb_model = lgb.train(params=params,\n                      train_set=dtrain,\n                      num_boost_round=1500,\n                      valid_sets=(dtrain, dvalid),\n                      categorical_feature=cat_features,\n                      callbacks=callbacks) # verbose_eval, early_stopping_rounds 대체\n\nend_time = datetime.now()\nprint(f\"runtime: {end_time - start_time}\")","metadata":{"papermill":{"duration":809.626892,"end_time":"2021-09-11T08:11:43.322732","exception":false,"start_time":"2021-09-11T07:58:13.69584","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:17:02.665069Z","iopub.execute_input":"2024-11-30T13:17:02.665447Z","iopub.status.idle":"2024-11-30T13:30:16.851318Z","shell.execute_reply.started":"2024-11-30T13:17:02.665412Z","shell.execute_reply":"2024-11-30T13:30:16.849378Z"}},"outputs":[{"name":"stdout","text":"[LightGBM] [Warning] Found whitespace in feature_names, replace with underlines\n[LightGBM] [Info] Total Bins 3657\n[LightGBM] [Info] Number of data points in the train set: 7593847, number of used features: 30\n[LightGBM] [Warning] Found whitespace in feature_names, replace with underlines\n[LightGBM] [Info] Start training from score 0.307756\nTraining until validation scores don't improve for 150 rounds\n[100]\ttraining's rmse: 1.0178\tvalid_1's rmse: 0.913728\n[200]\ttraining's rmse: 0.912629\tvalid_1's rmse: 0.841626\n[300]\ttraining's rmse: 0.859658\tvalid_1's rmse: 0.812341\n[400]\ttraining's rmse: 0.830685\tvalid_1's rmse: 0.800648\n[500]\ttraining's rmse: 0.811176\tvalid_1's rmse: 0.796129\n[600]\ttraining's rmse: 0.797613\tvalid_1's rmse: 0.794879\n[700]\ttraining's rmse: 0.78732\tvalid_1's rmse: 0.794375\n[800]\ttraining's rmse: 0.77898\tvalid_1's rmse: 0.794011\n[900]\ttraining's rmse: 0.771955\tvalid_1's rmse: 0.794313\n[1000]\ttraining's rmse: 0.765998\tvalid_1's rmse: 0.794295\nEarly stopping, best iteration is:\n[869]\ttraining's rmse: 0.773969\tvalid_1's rmse: 0.79395\nruntime: 0:13:14.176592\n","output_type":"stream"}],"execution_count":49},{"cell_type":"code","source":"# 예측\npreds = lgb_model.predict(X_test).clip(0, 20)\n\n# 제출 파일 생성\nsubmission['item_cnt_month'] = preds\nsubmission.to_csv('submission-authur.csv', index=False)","metadata":{"papermill":{"duration":12.512026,"end_time":"2021-09-11T08:11:55.890917","exception":false,"start_time":"2021-09-11T08:11:43.378891","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:30:16.853650Z","iopub.execute_input":"2024-11-30T13:30:16.853999Z","iopub.status.idle":"2024-11-30T13:30:29.218243Z","shell.execute_reply.started":"2024-11-30T13:30:16.853958Z","shell.execute_reply":"2024-11-30T13:30:29.217334Z"}},"outputs":[],"execution_count":50},{"cell_type":"code","source":"del X_train, y_train, X_valid, y_valid, X_test, lgb_model, dtrain, dvalid\ngc.collect();","metadata":{"papermill":{"duration":0.2264,"end_time":"2021-09-11T08:11:56.175099","exception":false,"start_time":"2021-09-11T08:11:55.948699","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:30:29.219555Z","iopub.execute_input":"2024-11-30T13:30:29.219868Z","iopub.status.idle":"2024-11-30T13:30:29.371778Z","shell.execute_reply.started":"2024-11-30T13:30:29.219836Z","shell.execute_reply":"2024-11-30T13:30:29.370544Z"}},"outputs":[],"execution_count":51},{"cell_type":"markdown","source":"valid_1's rmse: 0.79395\n\nPublic Score: 0.87701\r\n","metadata":{}},{"cell_type":"markdown","source":"# 5. Using the Hyperparameters Derived from _Bayesian Optimization_ \n\n## 5-1. in previous notebook : `Public Score: 0.88809`\n\n\n최적화된 하이퍼파라미터: {'bagging_fraction': 0.7949387309032754, 'bagging_freq': 9, 'feature_fraction': 0.5, 'learning_rate': 0.01, 'num_leaves': 213}\n\n`'submission-private-5-1.csv'`","metadata":{}},{"cell_type":"code","source":"# 훈련 데이터 (피처)\nX_train = all_data[all_data['월ID'] < 33]\nX_train = X_train.drop(['월간 판매량'], axis=1)\n# 검증 데이터 (피처)\nX_valid = all_data[all_data['월ID'] == 33]\nX_valid = X_valid.drop(['월간 판매량'], axis=1)\n# 테스트 데이터 (피처)\nX_test = all_data[all_data['월ID'] == 34]\nX_test = X_test.drop(['월간 판매량'], axis=1)\n\n# 훈련 데이터 (타깃값)\ny_train = all_data[all_data['월ID'] < 33]['월간 판매량']\n# 검증 데이터 (타깃값)\ny_valid = all_data[all_data['월ID'] == 33]['월간 판매량']\n\n# 가비지 컬렉션\n# del all_data\n# gc.collect();","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:30:29.373176Z","iopub.execute_input":"2024-11-30T13:30:29.373559Z","iopub.status.idle":"2024-11-30T13:30:32.703781Z","shell.execute_reply.started":"2024-11-30T13:30:29.373526Z","shell.execute_reply":"2024-11-30T13:30:32.702700Z"}},"outputs":[],"execution_count":52},{"cell_type":"code","source":"# import lightgbm as lgb\n# from lightgbm import early_stopping, log_evaluation # 버전차이로 지원하지 않는 부분 설정\n\nstart_time = datetime.now()\n\n# 얻은 LightGBM 하이퍼파라미터\nparams = {'metric': 'rmse',  # 평가지표: RMSE\n          'bagging_fraction': 0.7949387309032754, \n          'bagging_freq': 9, \n          'feature_fraction': 0.5, \n          'learning_rate': 0.01, \n          'num_leaves': 213,\n          'force_col_wise': True,\n          'random_state': 10}\n\n# categorical 피처 설정 : 도시, 대분류, 월 추가\ncat_features = ['상점ID', '도시', '상품분류ID', '대분류', '월']\n\n# LightGBM 훈련 및 검증 데이터셋\ndtrain = lgb.Dataset(X_train, y_train)\ndvalid = lgb.Dataset(X_valid, y_valid)\n\n\n# LightGBM 모델 훈련\n\n# 100회 반복마다 검증지표 출력(verbose_eval 파라미터 대체)\n# early_stopping_rounds=150 대체 추가\ncallbacks = [log_evaluation(period=100),  # verbose_eval 대체\n            early_stopping(stopping_rounds=150)]  # early_stopping_rounds 대체\n\nlgb_model = lgb.train(params=params,\n                      train_set=dtrain,\n                      num_boost_round=1500,\n                      valid_sets=(dtrain, dvalid),\n                      categorical_feature=cat_features,\n                      callbacks=callbacks)\n\nend_time = datetime.now()\nprint(f\"runtime: {end_time - start_time}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:30:32.705147Z","iopub.execute_input":"2024-11-30T13:30:32.705486Z","iopub.status.idle":"2024-11-30T13:38:00.866183Z","shell.execute_reply.started":"2024-11-30T13:30:32.705453Z","shell.execute_reply":"2024-11-30T13:38:00.864721Z"}},"outputs":[{"name":"stdout","text":"[LightGBM] [Warning] Found whitespace in feature_names, replace with underlines\n[LightGBM] [Info] Total Bins 3657\n[LightGBM] [Info] Number of data points in the train set: 7593847, number of used features: 30\n[LightGBM] [Warning] Found whitespace in feature_names, replace with underlines\n[LightGBM] [Info] Start training from score 0.307756\nTraining until validation scores don't improve for 150 rounds\n[100]\ttraining's rmse: 0.929383\tvalid_1's rmse: 0.848691\n[200]\ttraining's rmse: 0.848294\tvalid_1's rmse: 0.802654\n[300]\ttraining's rmse: 0.817299\tvalid_1's rmse: 0.794773\n[400]\ttraining's rmse: 0.799351\tvalid_1's rmse: 0.792995\n[500]\ttraining's rmse: 0.786615\tvalid_1's rmse: 0.792804\n[600]\ttraining's rmse: 0.776371\tvalid_1's rmse: 0.793823\nEarly stopping, best iteration is:\n[479]\ttraining's rmse: 0.789316\tvalid_1's rmse: 0.792472\nruntime: 0:07:28.152741\n","output_type":"stream"}],"execution_count":53},{"cell_type":"code","source":"# 예측\npreds = lgb_model.predict(X_test).clip(0, 20)\n\n# 제출 파일 생성\nsubmission['item_cnt_month'] = preds\nsubmission.to_csv('submission-private-5-1.csv', index=False)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:38:00.867881Z","iopub.execute_input":"2024-11-30T13:38:00.868254Z","iopub.status.idle":"2024-11-30T13:38:06.590482Z","shell.execute_reply.started":"2024-11-30T13:38:00.868219Z","shell.execute_reply":"2024-11-30T13:38:06.589184Z"}},"outputs":[],"execution_count":54},{"cell_type":"markdown","source":"valid_1's rmse: 0.792472\n\npublic score: 0.88809\r\n","metadata":{}},{"cell_type":"markdown","source":"## 5-2. with [실전문제 preprocessing] : `Public Score: 0.89295`\n\n\n\n최적화된 파라미터: {'bagging_fraction': 0.50446175516352, 'bagging_freq': 8, 'feature_fraction': 0.50446175516352, 'learning_rate': 0.005538531127282626, 'num_leaves': 135}\r\n","metadata":{}},{"cell_type":"code","source":"# 훈련 데이터 (피처)\nX_train = all_data[all_data['월ID'] < 33]\nX_train = X_train.drop(['월간 판매량'], axis=1)\n# 검증 데이터 (피처)\nX_valid = all_data[all_data['월ID'] == 33]\nX_valid = X_valid.drop(['월간 판매량'], axis=1)\n# 테스트 데이터 (피처)\nX_test = all_data[all_data['월ID'] == 34]\nX_test = X_test.drop(['월간 판매량'], axis=1)\n\n# 훈련 데이터 (타깃값)\ny_train = all_data[all_data['월ID'] < 33]['월간 판매량']\n# 검증 데이터 (타깃값)\ny_valid = all_data[all_data['월ID'] == 33]['월간 판매량']\n\n# 가비지 컬렉션\n# del all_data\n# gc.collect();","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:38:06.592186Z","iopub.execute_input":"2024-11-30T13:38:06.592761Z","iopub.status.idle":"2024-11-30T13:38:09.914586Z","shell.execute_reply.started":"2024-11-30T13:38:06.592696Z","shell.execute_reply":"2024-11-30T13:38:09.913375Z"}},"outputs":[],"execution_count":55},{"cell_type":"code","source":"# import lightgbm as lgb\n# from lightgbm import early_stopping, log_evaluation # 버전차이로 지원하지 않는 부분 설정\n\nstart_time = datetime.now()\n\n# 얻은 LightGBM 하이퍼파라미터\nparams = {'metric': 'rmse',  # 평가지표: RMSE\n          'bagging_fraction': 0.50446175516352, \n          'bagging_freq': 8, \n          'feature_fraction': 0.50446175516352, \n          'learning_rate': 0.005538531127282626, \n          'num_leaves': 135,\n          'force_col_wise': True,\n          'random_state': 10}\n\n# categorical 피처 설정 : 도시, 대분류, 월 추가\ncat_features = ['상점ID', '도시', '상품분류ID', '대분류', '월']\n\n# LightGBM 훈련 및 검증 데이터셋\ndtrain = lgb.Dataset(X_train, y_train)\ndvalid = lgb.Dataset(X_valid, y_valid)\n\n\n# LightGBM 모델 훈련\n\ncallbacks = [log_evaluation(period=100),  # verbose_eval 대체\n            early_stopping(stopping_rounds=150)]  # early_stopping_rounds 대체\n\nlgb_model = lgb.train(params=params,\n                      train_set=dtrain,\n                      num_boost_round=1500,\n                      valid_sets=(dtrain, dvalid),\n                      categorical_feature=cat_features,\n                      callbacks=callbacks)\n\nend_time = datetime.now()\nprint(f\"runtime: {end_time - start_time}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:38:09.916064Z","iopub.execute_input":"2024-11-30T13:38:09.916461Z","iopub.status.idle":"2024-11-30T13:46:33.292388Z","shell.execute_reply.started":"2024-11-30T13:38:09.916425Z","shell.execute_reply":"2024-11-30T13:46:33.291110Z"}},"outputs":[{"name":"stdout","text":"[LightGBM] [Warning] Found whitespace in feature_names, replace with underlines\n[LightGBM] [Info] Total Bins 3657\n[LightGBM] [Info] Number of data points in the train set: 7593847, number of used features: 30\n[LightGBM] [Warning] Found whitespace in feature_names, replace with underlines\n[LightGBM] [Info] Start training from score 0.307756\nTraining until validation scores don't improve for 150 rounds\n[100]\ttraining's rmse: 1.02178\tvalid_1's rmse: 0.911698\n[200]\ttraining's rmse: 0.924857\tvalid_1's rmse: 0.839315\n[300]\ttraining's rmse: 0.879219\tvalid_1's rmse: 0.811362\n[400]\ttraining's rmse: 0.855026\tvalid_1's rmse: 0.799988\n[500]\ttraining's rmse: 0.839214\tvalid_1's rmse: 0.795199\n[600]\ttraining's rmse: 0.828281\tvalid_1's rmse: 0.792876\n[700]\ttraining's rmse: 0.819763\tvalid_1's rmse: 0.791665\n[800]\ttraining's rmse: 0.813004\tvalid_1's rmse: 0.791262\n[900]\ttraining's rmse: 0.807528\tvalid_1's rmse: 0.791318\nEarly stopping, best iteration is:\n[827]\ttraining's rmse: 0.811659\tvalid_1's rmse: 0.791056\nruntime: 0:08:23.367892\n","output_type":"stream"}],"execution_count":56},{"cell_type":"code","source":"# 예측\npreds = lgb_model.predict(X_test).clip(0, 20)\n\n# 제출 파일 생성\nsubmission['item_cnt_month'] = preds\nsubmission.to_csv('submission-private-5-2.csv', index=False)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:46:33.294180Z","iopub.execute_input":"2024-11-30T13:46:33.294559Z","iopub.status.idle":"2024-11-30T13:46:43.020570Z","shell.execute_reply.started":"2024-11-30T13:46:33.294522Z","shell.execute_reply":"2024-11-30T13:46:43.019487Z"}},"outputs":[],"execution_count":57},{"cell_type":"markdown","source":"valid_1's rmse: 0.791056\n\npublic score: 0.89295","metadata":{}},{"cell_type":"markdown","source":"# 6. Trying Additional Feature Engineering \n\n## 6-1. 근접한 달에 labeling - 1 : `Public Score: 0.87945`\n\n(EDA에서 확인한 바) 예측해야하는 11월은 대부분 9, 10월과 비슷한 수준의 판매량을 보였으므로, 해당하는 월들에 대하여 labeling 진행  ","metadata":{}},{"cell_type":"code","source":"Dec = [11, 23, 35]\nnearNov = [val for i in Dec for val in [i-1, i-2, i-3]]\nnearNov","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:46:43.021903Z","iopub.execute_input":"2024-11-30T13:46:43.022236Z","iopub.status.idle":"2024-11-30T13:46:43.029901Z","shell.execute_reply.started":"2024-11-30T13:46:43.022203Z","shell.execute_reply":"2024-11-30T13:46:43.028747Z"}},"outputs":[{"execution_count":58,"output_type":"execute_result","data":{"text/plain":"[10, 9, 8, 22, 21, 20, 34, 33, 32]"},"metadata":{}}],"execution_count":58},{"cell_type":"code","source":"all_data['nearNov'] = all_data['월ID'].isin(nearNov).astype(int)\nall_data.sample(5)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:46:43.031057Z","iopub.execute_input":"2024-11-30T13:46:43.031456Z","iopub.status.idle":"2024-11-30T13:46:43.469120Z","shell.execute_reply.started":"2024-11-30T13:46:43.031416Z","shell.execute_reply":"2024-11-30T13:46:43.468000Z"}},"outputs":[{"execution_count":59,"output_type":"execute_result","data":{"text/plain":"         월ID  상점ID   상품ID  월간 판매량  도시  상품분류ID  대분류  월간 판매량_시차1  월간 판매량_시차2  \\\n6438944   24    16   3360       0  10      30    3           0           0   \n7383920   28    16   4577       0  10      56    7           0           0   \n6562593   24    38  13549       0  17      69    8           0           0   \n2784833    9    15  14841      10   9      65    8           1           5   \n4520355   16     5  21836       0   3      37    5           0           0   \n\n         월간 판매량_시차3  ...  상점ID 상품분류ID별 평균 판매량_시차1  상점ID 상품분류ID별 평균 판매량_시차2  \\\n6438944           0  ...                 1.115790                 1.069149   \n7383920           0  ...                 0.093750                 0.088235   \n6562593           0  ...                 0.387097                 0.254237   \n2784833           1  ...                 0.340909                 1.022222   \n4520355           0  ...                 0.094508                 0.082310   \n\n         상점ID 상품분류ID별 평균 판매량_시차3  월간 판매량 시차평균  시차변화량1  시차변화량2  신상여부  \\\n6438944                 1.321429     0.000000     0.0     0.0     0   \n7383920                 0.032258     0.000000     0.0     0.0     0   \n6562593                 0.206897     0.000000     0.0     0.0     0   \n2784833                 0.351852     2.333333     0.2     5.0     0   \n4520355                 0.073110     0.000000     0.0     0.0     0   \n\n         첫 판매 후 기간  월  nearNov  \n6438944         24  0        0  \n7383920         28  4        0  \n6562593          8  0        0  \n2784833          5  9        1  \n4520355          1  4        0  \n\n[5 rows x 32 columns]","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>월ID</th>\n      <th>상점ID</th>\n      <th>상품ID</th>\n      <th>월간 판매량</th>\n      <th>도시</th>\n      <th>상품분류ID</th>\n      <th>대분류</th>\n      <th>월간 판매량_시차1</th>\n      <th>월간 판매량_시차2</th>\n      <th>월간 판매량_시차3</th>\n      <th>...</th>\n      <th>상점ID 상품분류ID별 평균 판매량_시차1</th>\n      <th>상점ID 상품분류ID별 평균 판매량_시차2</th>\n      <th>상점ID 상품분류ID별 평균 판매량_시차3</th>\n      <th>월간 판매량 시차평균</th>\n      <th>시차변화량1</th>\n      <th>시차변화량2</th>\n      <th>신상여부</th>\n      <th>첫 판매 후 기간</th>\n      <th>월</th>\n      <th>nearNov</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>6438944</th>\n      <td>24</td>\n      <td>16</td>\n      <td>3360</td>\n      <td>0</td>\n      <td>10</td>\n      <td>30</td>\n      <td>3</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>1.115790</td>\n      <td>1.069149</td>\n      <td>1.321429</td>\n      <td>0.000000</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0</td>\n      <td>24</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>7383920</th>\n      <td>28</td>\n      <td>16</td>\n      <td>4577</td>\n      <td>0</td>\n      <td>10</td>\n      <td>56</td>\n      <td>7</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0.093750</td>\n      <td>0.088235</td>\n      <td>0.032258</td>\n      <td>0.000000</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0</td>\n      <td>28</td>\n      <td>4</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>6562593</th>\n      <td>24</td>\n      <td>38</td>\n      <td>13549</td>\n      <td>0</td>\n      <td>17</td>\n      <td>69</td>\n      <td>8</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0.387097</td>\n      <td>0.254237</td>\n      <td>0.206897</td>\n      <td>0.000000</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0</td>\n      <td>8</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>2784833</th>\n      <td>9</td>\n      <td>15</td>\n      <td>14841</td>\n      <td>10</td>\n      <td>9</td>\n      <td>65</td>\n      <td>8</td>\n      <td>1</td>\n      <td>5</td>\n      <td>1</td>\n      <td>...</td>\n      <td>0.340909</td>\n      <td>1.022222</td>\n      <td>0.351852</td>\n      <td>2.333333</td>\n      <td>0.2</td>\n      <td>5.0</td>\n      <td>0</td>\n      <td>5</td>\n      <td>9</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>4520355</th>\n      <td>16</td>\n      <td>5</td>\n      <td>21836</td>\n      <td>0</td>\n      <td>3</td>\n      <td>37</td>\n      <td>5</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0.094508</td>\n      <td>0.082310</td>\n      <td>0.073110</td>\n      <td>0.000000</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>4</td>\n      <td>0</td>\n    </tr>\n  </tbody>\n</table>\n<p>5 rows × 32 columns</p>\n</div>"},"metadata":{}}],"execution_count":59},{"cell_type":"code","source":"# 훈련 데이터 (피처)\nX_train = all_data[all_data['월ID'] < 33]\nX_train = X_train.drop(['월간 판매량'], axis=1)\n# 검증 데이터 (피처)\nX_valid = all_data[all_data['월ID'] == 33]\nX_valid = X_valid.drop(['월간 판매량'], axis=1)\n# 테스트 데이터 (피처)\nX_test = all_data[all_data['월ID'] == 34]\nX_test = X_test.drop(['월간 판매량'], axis=1)\n\n# 훈련 데이터 (타깃값)\ny_train = all_data[all_data['월ID'] < 33]['월간 판매량']\n# 검증 데이터 (타깃값)\ny_valid = all_data[all_data['월ID'] == 33]['월간 판매량']\n\n# 가비지 컬렉션\n# del all_data\n# gc.collect();","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:46:43.470437Z","iopub.execute_input":"2024-11-30T13:46:43.470717Z","iopub.status.idle":"2024-11-30T13:46:46.946986Z","shell.execute_reply.started":"2024-11-30T13:46:43.470689Z","shell.execute_reply":"2024-11-30T13:46:46.945748Z"}},"outputs":[],"execution_count":60},{"cell_type":"code","source":"# import lightgbm as lgb\n# from lightgbm import early_stopping, log_evaluation # 버전차이로 지원하지 않는 부분 설정\n\nstart_time = datetime.now()\n\n# LightGBM 하이퍼파라미터\nparams = {'metric': 'rmse',  # 평가지표: RMSE\n          'num_leaves': 255,\n          'learning_rate': 0.005,  # 이전 학습율 0.01\n          'feature_fraction': 0.75,  # 추가1\n          'bagging_fraction': 0.75,  # 추가2\n          'bagging_freq': 5,  # 추가3\n          'force_col_wise': True,\n          'random_state': 10}\n\n# categorical 피처 설정 : 도시, 대분류, 월 추가\ncat_features = ['상점ID', '도시', '상품분류ID', '대분류', '월']\n\n# LightGBM 훈련 및 검증 데이터셋\ndtrain = lgb.Dataset(X_train, y_train)\ndvalid = lgb.Dataset(X_valid, y_valid)\n\n\n# LightGBM 모델 훈련\n\n# 100회 반복마다 검증지표 출력(verbose_eval 파라미터 대체)\n# early_stopping_rounds=150 대체 추가\ncallbacks = [log_evaluation(period=100),  # verbose_eval 대체\n            early_stopping(stopping_rounds=150)]  # early_stopping_rounds 대체\n\nlgb_model = lgb.train(params=params,\n                      train_set=dtrain,\n                      num_boost_round=1500,\n                      valid_sets=(dtrain, dvalid),\n                      categorical_feature=cat_features,\n                      callbacks=callbacks)\n\nend_time = datetime.now()\nprint(f\"runtime: {end_time - start_time}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:46:46.948335Z","iopub.execute_input":"2024-11-30T13:46:46.948761Z","iopub.status.idle":"2024-11-30T13:58:05.051154Z","shell.execute_reply.started":"2024-11-30T13:46:46.948713Z","shell.execute_reply":"2024-11-30T13:58:05.050067Z"}},"outputs":[{"name":"stdout","text":"[LightGBM] [Warning] Found whitespace in feature_names, replace with underlines\n[LightGBM] [Info] Total Bins 3659\n[LightGBM] [Info] Number of data points in the train set: 7593847, number of used features: 31\n[LightGBM] [Warning] Found whitespace in feature_names, replace with underlines\n[LightGBM] [Info] Start training from score 0.307756\nTraining until validation scores don't improve for 150 rounds\n[100]\ttraining's rmse: 1.01856\tvalid_1's rmse: 0.91328\n[200]\ttraining's rmse: 0.91327\tvalid_1's rmse: 0.842114\n[300]\ttraining's rmse: 0.86017\tvalid_1's rmse: 0.812148\n[400]\ttraining's rmse: 0.830187\tvalid_1's rmse: 0.801014\n[500]\ttraining's rmse: 0.811293\tvalid_1's rmse: 0.79731\n[600]\ttraining's rmse: 0.798036\tvalid_1's rmse: 0.795851\n[700]\ttraining's rmse: 0.787631\tvalid_1's rmse: 0.79527\n[800]\ttraining's rmse: 0.779159\tvalid_1's rmse: 0.795479\nEarly stopping, best iteration is:\n[699]\ttraining's rmse: 0.78773\tvalid_1's rmse: 0.795252\nruntime: 0:11:18.094885\n","output_type":"stream"}],"execution_count":61},{"cell_type":"code","source":"# 예측\npreds = lgb_model.predict(X_test).clip(0, 20)\n\n# 제출 파일 생성\nsubmission['item_cnt_month'] = preds\nsubmission.to_csv('submission-private-6-1.csv', index=False)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:58:05.052419Z","iopub.execute_input":"2024-11-30T13:58:05.052698Z","iopub.status.idle":"2024-11-30T13:58:15.481836Z","shell.execute_reply.started":"2024-11-30T13:58:05.052670Z","shell.execute_reply":"2024-11-30T13:58:15.480692Z"}},"outputs":[],"execution_count":62},{"cell_type":"markdown","source":"valid_1's rmse: 0.795252\n\nPublic Score: 0. 0.87945","metadata":{}},{"cell_type":"markdown","source":"### 2. 근접한 달에 labeling - 2 : `Public Score: 0.88068`\n\nnearNov에 0, 0.5, November에 1로 차등적으로 값 주는 방식으로 수정하여 시도","metadata":{}},{"cell_type":"code","source":"Dec = [11, 23, 35]\nNov = [val-1 for val in Dec]\nnearNov = [val for i in Dec for val in [i-2, i-3]]\n\nall_data['nearNov'] = np.where(all_data['월ID'].isin(nearNov), 0.5, 0)\nall_data['nearNov'] = np.where(all_data['월ID'].isin(Nov), 1, 0)\nall_data.sample(5)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:58:15.483173Z","iopub.execute_input":"2024-11-30T13:58:15.483649Z","iopub.status.idle":"2024-11-30T13:58:15.985126Z","shell.execute_reply.started":"2024-11-30T13:58:15.483600Z","shell.execute_reply":"2024-11-30T13:58:15.983970Z"}},"outputs":[{"execution_count":63,"output_type":"execute_result","data":{"text/plain":"         월ID  상점ID   상품ID  월간 판매량  도시  상품분류ID  대분류  월간 판매량_시차1  월간 판매량_시차2  \\\n1078737    3    44  21508       0  20      66    8           0           0   \n3198513   11    19   2415       0  12      30    3           1           2   \n8303768   32    52  10600       0  25      43    6           0           0   \n2085867    7    56   6599       1  28      25    3           1           0   \n1658820    5    50   8807       0  24      40    5           0           0   \n\n         월간 판매량_시차3  ...  상점ID 상품분류ID별 평균 판매량_시차1  상점ID 상품분류ID별 평균 판매량_시차2  \\\n1078737           0  ...                 0.000000                 0.000000   \n3198513           0  ...                 1.888476                 2.032389   \n8303768           0  ...                 0.057471                 0.052632   \n2085867           0  ...                 0.500000                 0.305556   \n1658820           1  ...                 0.291667                 0.301327   \n\n         상점ID 상품분류ID별 평균 판매량_시차3  월간 판매량 시차평균  시차변화량1  시차변화량2  신상여부  \\\n1078737                 0.000000     0.000000     0.0     0.0     1   \n3198513                 2.298851     1.000000     0.5     0.0     0   \n8303768                 0.074766     0.000000     0.0     0.0     0   \n2085867                 0.297297     0.333333     0.0     0.0     0   \n1658820                 0.339123     0.333333     0.0     0.0     0   \n\n         첫 판매 후 기간   월  nearNov  \n1078737          0   3        0  \n3198513         11  11        0  \n8303768          1   8        0  \n2085867          7   7        0  \n1658820          5   5        0  \n\n[5 rows x 32 columns]","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>월ID</th>\n      <th>상점ID</th>\n      <th>상품ID</th>\n      <th>월간 판매량</th>\n      <th>도시</th>\n      <th>상품분류ID</th>\n      <th>대분류</th>\n      <th>월간 판매량_시차1</th>\n      <th>월간 판매량_시차2</th>\n      <th>월간 판매량_시차3</th>\n      <th>...</th>\n      <th>상점ID 상품분류ID별 평균 판매량_시차1</th>\n      <th>상점ID 상품분류ID별 평균 판매량_시차2</th>\n      <th>상점ID 상품분류ID별 평균 판매량_시차3</th>\n      <th>월간 판매량 시차평균</th>\n      <th>시차변화량1</th>\n      <th>시차변화량2</th>\n      <th>신상여부</th>\n      <th>첫 판매 후 기간</th>\n      <th>월</th>\n      <th>nearNov</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>1078737</th>\n      <td>3</td>\n      <td>44</td>\n      <td>21508</td>\n      <td>0</td>\n      <td>20</td>\n      <td>66</td>\n      <td>8</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.000000</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>3</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>3198513</th>\n      <td>11</td>\n      <td>19</td>\n      <td>2415</td>\n      <td>0</td>\n      <td>12</td>\n      <td>30</td>\n      <td>3</td>\n      <td>1</td>\n      <td>2</td>\n      <td>0</td>\n      <td>...</td>\n      <td>1.888476</td>\n      <td>2.032389</td>\n      <td>2.298851</td>\n      <td>1.000000</td>\n      <td>0.5</td>\n      <td>0.0</td>\n      <td>0</td>\n      <td>11</td>\n      <td>11</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>8303768</th>\n      <td>32</td>\n      <td>52</td>\n      <td>10600</td>\n      <td>0</td>\n      <td>25</td>\n      <td>43</td>\n      <td>6</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0.057471</td>\n      <td>0.052632</td>\n      <td>0.074766</td>\n      <td>0.000000</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>8</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>2085867</th>\n      <td>7</td>\n      <td>56</td>\n      <td>6599</td>\n      <td>1</td>\n      <td>28</td>\n      <td>25</td>\n      <td>3</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0.500000</td>\n      <td>0.305556</td>\n      <td>0.297297</td>\n      <td>0.333333</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0</td>\n      <td>7</td>\n      <td>7</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>1658820</th>\n      <td>5</td>\n      <td>50</td>\n      <td>8807</td>\n      <td>0</td>\n      <td>24</td>\n      <td>40</td>\n      <td>5</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>...</td>\n      <td>0.291667</td>\n      <td>0.301327</td>\n      <td>0.339123</td>\n      <td>0.333333</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0</td>\n      <td>5</td>\n      <td>5</td>\n      <td>0</td>\n    </tr>\n  </tbody>\n</table>\n<p>5 rows × 32 columns</p>\n</div>"},"metadata":{}}],"execution_count":63},{"cell_type":"code","source":"# 훈련 데이터 (피처)\nX_train = all_data[all_data['월ID'] < 33]\nX_train = X_train.drop(['월간 판매량'], axis=1)\n# 검증 데이터 (피처)\nX_valid = all_data[all_data['월ID'] == 33]\nX_valid = X_valid.drop(['월간 판매량'], axis=1)\n# 테스트 데이터 (피처)\nX_test = all_data[all_data['월ID'] == 34]\nX_test = X_test.drop(['월간 판매량'], axis=1)\n\n# 훈련 데이터 (타깃값)\ny_train = all_data[all_data['월ID'] < 33]['월간 판매량']\n# 검증 데이터 (타깃값)\ny_valid = all_data[all_data['월ID'] == 33]['월간 판매량']\n\n# 가비지 컬렉션\n# del all_data\n# gc.collect();","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:58:15.987219Z","iopub.execute_input":"2024-11-30T13:58:15.987686Z","iopub.status.idle":"2024-11-30T13:58:19.534145Z","shell.execute_reply.started":"2024-11-30T13:58:15.987638Z","shell.execute_reply":"2024-11-30T13:58:19.533121Z"}},"outputs":[],"execution_count":64},{"cell_type":"code","source":"# import lightgbm as lgb\n# from lightgbm import early_stopping, log_evaluation # 버전차이로 지원하지 않는 부분 설정\n\nstart_time = datetime.now()\n\n# LightGBM 하이퍼파라미터\nparams = {'metric': 'rmse',  # 평가지표: RMSE\n          'num_leaves': 255,\n          'learning_rate': 0.005,  # 이전 학습율 0.01\n          'feature_fraction': 0.75,  # 추가1\n          'bagging_fraction': 0.75,  # 추가2\n          'bagging_freq': 5,  # 추가3\n          'force_col_wise': True,\n          'random_state': 10}\n\n# categorical 피처 설정 : 도시, 대분류, 월 추가\ncat_features = ['상점ID', '도시', '상품분류ID', '대분류', '월']\n\n# LightGBM 훈련 및 검증 데이터셋\ndtrain = lgb.Dataset(X_train, y_train)\ndvalid = lgb.Dataset(X_valid, y_valid)\n\n\n# LightGBM 모델 훈련\n\n# 100회 반복마다 검증지표 출력(verbose_eval 파라미터 대체)\n# early_stopping_rounds=150 대체 추가\ncallbacks = [log_evaluation(period=100),  # verbose_eval 대체\n            early_stopping(stopping_rounds=150)]  # early_stopping_rounds 대체\n\nlgb_model = lgb.train(params=params,\n                      train_set=dtrain,\n                      num_boost_round=1500,\n                      valid_sets=(dtrain, dvalid),\n                      categorical_feature=cat_features,\n                      callbacks=callbacks) # verbose_eval, early_stopping_rounds 대체\n\nend_time = datetime.now()\nprint(f\"runtime: {end_time - start_time}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T13:58:19.535912Z","iopub.execute_input":"2024-11-30T13:58:19.536475Z","iopub.status.idle":"2024-11-30T14:09:53.708894Z","shell.execute_reply.started":"2024-11-30T13:58:19.536424Z","shell.execute_reply":"2024-11-30T14:09:53.707674Z"}},"outputs":[{"name":"stdout","text":"[LightGBM] [Warning] Found whitespace in feature_names, replace with underlines\n[LightGBM] [Info] Total Bins 3659\n[LightGBM] [Info] Number of data points in the train set: 7593847, number of used features: 31\n[LightGBM] [Warning] Found whitespace in feature_names, replace with underlines\n[LightGBM] [Info] Start training from score 0.307756\nTraining until validation scores don't improve for 150 rounds\n[100]\ttraining's rmse: 1.01866\tvalid_1's rmse: 0.913874\n[200]\ttraining's rmse: 0.913423\tvalid_1's rmse: 0.84161\n[300]\ttraining's rmse: 0.860369\tvalid_1's rmse: 0.811409\n[400]\ttraining's rmse: 0.830438\tvalid_1's rmse: 0.799544\n[500]\ttraining's rmse: 0.81143\tvalid_1's rmse: 0.795141\n[600]\ttraining's rmse: 0.798102\tvalid_1's rmse: 0.793469\n[700]\ttraining's rmse: 0.787683\tvalid_1's rmse: 0.792922\n[800]\ttraining's rmse: 0.779295\tvalid_1's rmse: 0.793229\nEarly stopping, best iteration is:\n[716]\ttraining's rmse: 0.786275\tvalid_1's rmse: 0.792896\nruntime: 0:11:34.163772\n","output_type":"stream"}],"execution_count":65},{"cell_type":"code","source":"# 예측\npreds = lgb_model.predict(X_test).clip(0, 20)\n\n# 제출 파일 생성\nsubmission['item_cnt_month'] = preds\nsubmission.to_csv('submission-private-6-2.csv', index=False)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-30T14:09:53.710395Z","iopub.execute_input":"2024-11-30T14:09:53.710731Z","iopub.status.idle":"2024-11-30T14:10:04.929032Z","shell.execute_reply.started":"2024-11-30T14:09:53.710698Z","shell.execute_reply":"2024-11-30T14:10:04.927780Z"}},"outputs":[],"execution_count":66},{"cell_type":"markdown","source":"valid_1's rmse: 0.792896\n\npublic score : 0.88068\r\n\r\n","metadata":{}}]}