{"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.11.11"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":10445,"databundleVersionId":200747,"sourceType":"competition"}],"dockerImageVersionId":31040,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import os\nimport pandas as pd\nfrom sklearn.preprocessing import LabelEncoder\nfrom xgboost import XGBRegressor\nimport matplotlib.pyplot as plt\n\nrseed = 71","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-29T13:23:15.702328Z","iopub.execute_input":"2025-06-29T13:23:15.703223Z","iopub.status.idle":"2025-06-29T13:23:15.707653Z","shell.execute_reply.started":"2025-06-29T13:23:15.703183Z","shell.execute_reply":"2025-06-29T13:23:15.706765Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# ----------------------------------------------\n# データ読み込み\n# ----------------------------------------------\nDATA_DIR = \"/kaggle/input/elo-merchant-category-recommendation\"\ntrain = pd.read_csv(f\"{DATA_DIR}/train.csv\")\ntest  = pd.read_csv(f\"{DATA_DIR}/test.csv\")\n\n# 元テーブルからID列と目的変数をコピー\ntrain_card_id = train['card_id'].copy()\ntest_card_id  = test['card_id'].copy()\ntrain_y       = train['target'].copy()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-29T13:23:15.714827Z","iopub.execute_input":"2025-06-29T13:23:15.715590Z","iopub.status.idle":"2025-06-29T13:23:16.162422Z","shell.execute_reply.started":"2025-06-29T13:23:15.715563Z","shell.execute_reply":"2025-06-29T13:23:16.161456Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# ----------------------------------------------\n# メインテーブルの特徴量作成\n# ・年月文字列は欠損補完・Label Encoding\n# ・他はそのまま\n# ----------------------------------------------\n\n# 元テーブルから特徴量をコピー（目的変数は削除）\ntrain_x = train.copy().drop(columns=[\"target\"])\ntest_x  = test.copy()\n\n# 年月文字列に欠損があるので補完\ntrain_x[\"first_active_month\"] = train_x[\"first_active_month\"].fillna(\"Unknown\")\ntest_x[\"first_active_month\"]  = test_x[\"first_active_month\"].fillna(\"Unknown\")\n\n# 年月文字列をLabel Encoding\nle = LabelEncoder()\nall_months = pd.concat([train_x[\"first_active_month\"], test_x[\"first_active_month\"]])\nle.fit(all_months)\ntrain_x[\"fam_label\"] = le.transform(train_x[\"first_active_month\"])\ntest_x[\"fam_label\"]  = le.transform(test_x[\"first_active_month\"])\n\n\n# 変換前の年月文字列を削除\ntrain_x = train_x.drop(columns=[\"first_active_month\"])\ntest_x  = test_x.drop(columns=[\"first_active_month\"])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-29T13:23:16.163887Z","iopub.execute_input":"2025-06-29T13:23:16.164257Z","iopub.status.idle":"2025-06-29T13:23:16.291758Z","shell.execute_reply.started":"2025-06-29T13:23:16.164225Z","shell.execute_reply":"2025-06-29T13:23:16.290748Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# ----------------------------------------------\n# 追加テーブルの特徴量作成\n# ----------------------------------------------\n\n# historical_transactions.csv\n# merchants.csv\n\nnew_trans = pd.read_csv(f\"{DATA_DIR}/historical_transactions.csv\")\nprint(new_trans.describe())\n\n# 集約特徴の作成パターンを辞書化\n\"\"\"\nagg_funcs = {\n    \"purchase_amount\": [\"min\", \"mean\", \"max\", \"sem\"], # 正規化済みの購入金額（負値 = 顧客の支出）\n    \"installments\":    [\"min\", \"mean\", \"max\", \"sem\"], # 分割払いの分割回数\n    \"month_lag\":       [\"min\", \"mean\", \"max\", \"sem\"], # 基準日からの経過月数\n    \"purchase_date\":   [\"count\"],      # 購入日時（countで空でない行を数える ⇔ 購入回数）\n}\n\"\"\"\n\nagg_funcs = {\n    'purchase_amount' : ['max','min','mean'],\n    'installments' :    ['max','min','mean'],\n    'month_lag' :       ['max','min','mean'],\n    'purchase_date' :   ['count'],\n}\n\"\"\"\n    'purchase_amount' : ['sum','max','min','mean','var','skew'],\n    'installments' :    ['sum','max','mean','var','skew'],\n    'month_lag' :       ['max','min','mean','var','skew'],\n    'purchase_date' :   ['count'],\n\"\"\"\n\n# 全ての行を card_id の値でグループ化\n# ⇒　上の集約パターンに従って実際に特徴を集約\nagg_new = new_trans.groupby(\"card_id\").agg(agg_funcs)\n\n# 備考）上の集約特徴のテーブルは，列が MultiIndex の状態になっている\n# 　　　⇒　列名のリストを表示して確認してみよう\nprint(agg_new.columns.to_list())\n\n# 1列を指定するために「変数名と集約関数」をペアで指定する必要があることがわかる\n\n# 新特徴の名前（列名）を機械的に生成してリスト化\n# 　　命名規則：元変数名_集約方法（「purchase_amount_mean」など）\nflat_cols = []\nfor col_name, func_name in agg_new.columns.to_flat_index():\n    new_name = f\"new_{col_name}_{func_name}\"\n    flat_cols.append(new_name)\n\n# MultiIndex状態の列を通常の状態に変換（列方向に平坦化）\nagg_new.columns = flat_cols\nprint(agg_new.columns.to_list())\n\n# groupby によるグループ化を解除（card_idを通常の列に戻す）\nagg_new = agg_new.reset_index()\n\n# メインテーブルの特徴と結合\ntrain_x = train_x.merge(agg_new, on=\"card_id\", how=\"left\")\ntest_x  = test_x.merge(agg_new, on=\"card_id\", how=\"left\")\n\n\nprint(\"\\n!!!!!!!!!!\\n\")\n\nnew_trans = pd.read_csv(f\"{DATA_DIR}/merchants.csv\")\nprint(new_trans.describe())\n\n# 集約特徴の作成パターンを辞書化\n\"\"\"\nagg_funcs = {\n    \"avg_purchases_log12\": [\"mean\", lambda x: x.quantile(0.25), lambda x: x.quantile(0.75)], # 直近12ヵ月間の稼動月数\n    \"avg_sales_lag12\" :    [\"mean\", lambda x: x.quantile(0.25), lambda x: x.quantile(0.75)], # 直近12ヶ月間の月間平均売上 ÷ 最近1ヶ月間の月間平均売上\n    \"active_months_lag12\": [\"mean\", lambda x: x.quantile(0.25), lambda x: x.quantile(0.75)]  # 直近12ヵ月間の月平均取引高 ÷ 最近1ヵ月間の月間取引高\n}\n\"\"\"\n\nmode = lambda x: Counter(x).most_common(1)[0][0]\nagg_funcs = {\n    \"avg_sales_lag3\"     : \"mean\",\n    \"avg_purchases_lag3\" : \"mean\",\n    \"active_months_lag3\" : mode,\n    \"avg_sales_lag6\"     : \"mean\",\n    \"avg_purchases_lag6\" : \"mean\",\n    \"active_months_lag6\" : mode,\n    \"avg_sales_lag12\"    : \"mean\",\n    \"avg_purchases_lag12\": \"mean\",\n    \"active_months_lag12\": mode,\n}\n\"\"\"\n    \"avg_sales_lag3\"     : \"mean\",\n    \"avg_purchases_lag3\" : \"mean\",\n    \"active_months_lag3\" : mode,\n    \"avg_sales_lag6\"     : \"mean\",\n    \"avg_purchases_lag6\" : \"mean\",\n    \"active_months_lag6\" : mode,\n\"\"\"\n\n# 全ての行を card_id の値でグループ化\n# ⇒　上の集約パターンに従って実際に特徴を集約\n#agg_new = new_trans.groupby(\"merchant_id\").agg(agg_funcs)\n\n# 備考）上の集約特徴のテーブルは，列が MultiIndex の状態になっている\n# 　　　⇒　列名のリストを表示して確認してみよう\nprint(agg_new.columns.to_list())\n\n\n\n# groupby によるグループ化を解除（card_idを通常の列に戻す）\nagg_new = agg_new.reset_index()\n\n\ntrain_x = train_x.merge(agg_new, how=\"left\")\ntest_x  = test_x.merge(agg_new, how=\"left\")\n\nprint(\"\\n??????????\\n\")\n\n# 追加テーブルの読み込み\nnew_trans = pd.read_csv(f\"{DATA_DIR}/new_merchant_transactions.csv\")\nprint(new_trans.describe())\n\n# 集約特徴の作成パターンを辞書化\n\"\"\"\nagg_funcs = {\n    \"purchase_amount\": [\"min\", \"mean\", \"max\", \"sem\"], # 正規化済みの購入金額（負値 = 顧客の支出）\n    \"installments\":    [\"min\", \"mean\", \"max\", \"sem\"], # 分割払いの分割回数\n    \"month_lag\":       [\"min\", \"mean\", \"max\", \"sem\"], # 基準日からの経過月数\n    \"purchase_date\":   [\"count\"],      # 購入日時（countで空でない行を数える ⇔ 購入回数）\n}\n\"\"\"\n\n\nagg_funcs = {\n    'purchase_amount' : ['max','min','mean'],\n    'installments' :    ['max','min','mean'],\n    'month_lag' :       ['max','min','mean'],\n    'purchase_date' :   ['count'],\n}\n\"\"\"\n\n\"\"\"\n\n# 例）\"month_lag\": [\"min\", \"mean\", \"max\"],\n# 　　⇔　変数 \"month_lag\" について，同一IDの中で\n# 　　　　最小値，平均値，最大値を取って集約特徴に追加する\n\n# 全ての行を card_id の値でグループ化\n# ⇒　上の集約パターンに従って実際に特徴を集約\nagg_new = new_trans.groupby(\"card_id\").agg(agg_funcs)\n\n# 備考）上の集約特徴のテーブルは，列が MultiIndex の状態になっている\n# 　　　⇒　列名のリストを表示して確認してみよう\nprint(agg_new.columns.to_list())\n\n# 1列を指定するために「変数名と集約関数」をペアで指定する必要があることがわかる\n\n# 新特徴の名前（列名）を機械的に生成してリスト化\n# 　　命名規則：元変数名_集約方法（「purchase_amount_mean」など）\nflat_cols = []\nfor col_name, func_name in agg_new.columns.to_flat_index():\n    new_name = f\"new_{col_name}_{func_name}\"\n    flat_cols.append(new_name)\n\n# MultiIndex状態の列を通常の状態に変換（列方向に平坦化）\nagg_new.columns = flat_cols\nprint(agg_new.columns.to_list())\n\n# groupby によるグループ化を解除（card_idを通常の列に戻す）\nagg_new = agg_new.reset_index()\n\n# メインテーブルの特徴と結合\ntrain_x = train_x.merge(agg_new, on=\"card_id\", how=\"left\")\ntest_x  = test_x.merge(agg_new, on=\"card_id\", how=\"left\")\n\n# Fill missing values created by the left join\ntrain_x = train_x.fillna(-1)\ntest_x  = test_x.fillna(-1)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-29T13:23:16.292836Z","iopub.execute_input":"2025-06-29T13:23:16.293125Z","iopub.status.idle":"2025-06-29T13:25:13.553824Z","shell.execute_reply.started":"2025-06-29T13:23:16.293104Z","shell.execute_reply":"2025-06-29T13:25:13.552953Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# ID列を削除\ntrain_x = train_x.drop(columns=[\"card_id\"])\ntest_x  = test_x.drop(columns=[\"card_id\"])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-29T13:25:13.555917Z","iopub.execute_input":"2025-06-29T13:25:13.556199Z","iopub.status.idle":"2025-06-29T13:25:13.578818Z","shell.execute_reply.started":"2025-06-29T13:25:13.556150Z","shell.execute_reply":"2025-06-29T13:25:13.577947Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# ----------------------------------------------\n# 使用する特徴量の可視化\n# ----------------------------------------------\n\n# 各特徴の統計量\nprint(pd.concat([train_x, train_y], axis=1).describe())\n\nprint(pd.concat([test_x], axis=1).describe())\n\n# 各特徴のヒストグラム\ntrain_x.hist(bins=100, color=\"blue\", grid=True, label='pandas')\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-06-29T13:25:13.579712Z","iopub.execute_input":"2025-06-29T13:25:13.579966Z","iopub.status.idle":"2025-06-29T13:25:19.658581Z","shell.execute_reply.started":"2025-06-29T13:25:13.579947Z","shell.execute_reply":"2025-06-29T13:25:19.657661Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# ------------------------------\n# XGBoostの学習・推論・submit\n# ------------------------------\nmodel = XGBRegressor(random_state=rseed)\nmodel.fit(train_x, train_y)\npreds = model.predict(test_x)\n\nsubmission = pd.DataFrame({\n    \"card_id\": test_card_id,\n    \"target\":  preds\n})\n\nsubmission.to_csv(\"submission.csv\", index=False)","metadata":{"_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","trusted":true,"execution":{"iopub.status.busy":"2025-06-29T13:25:19.659692Z","iopub.execute_input":"2025-06-29T13:25:19.660190Z","iopub.status.idle":"2025-06-29T13:25:21.779470Z","shell.execute_reply.started":"2025-06-29T13:25:19.660139Z","shell.execute_reply":"2025-06-29T13:25:21.778528Z"}},"outputs":[],"execution_count":null}]}