{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"nvidiaTeslaT4","dataSources":[{"sourceId":50160,"databundleVersionId":7921029,"sourceType":"competition"}],"dockerImageVersionId":30665,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import sys\nfrom pathlib import Path\nimport subprocess\nimport os\nimport gc\nfrom glob import glob\n\nimport numpy as np\nimport pandas as pd\nimport polars as pl\nfrom datetime import datetime\nimport seaborn as sns\nimport matplotlib.pyplot as plt\n\nimport warnings\nwarnings.filterwarnings('ignore')\n\nROOT = '/kaggle/input/home-credit-credit-risk-model-stability'\n\nfrom sklearn.model_selection import TimeSeriesSplit, GroupKFold, StratifiedGroupKFold\nfrom sklearn.base import BaseEstimator, RegressorMixin\nfrom sklearn.metrics import roc_auc_score\nimport lightgbm as lgb\n\nfrom imblearn.over_sampling import SMOTE\nfrom sklearn.preprocessing import OrdinalEncoder\nfrom sklearn.impute import KNNImputer","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2024-05-27T05:46:16.117486Z","iopub.execute_input":"2024-05-27T05:46:16.117855Z","iopub.status.idle":"2024-05-27T05:46:16.124994Z","shell.execute_reply.started":"2024-05-27T05:46:16.117825Z","shell.execute_reply":"2024-05-27T05:46:16.123792Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class Pipeline:\n\n    def set_table_dtypes(df):\n        for col in df.columns:\n            if col in [\"case_id\", \"WEEK_NUM\", \"num_group1\", \"num_group2\"]:\n                df = df.with_columns(pl.col(col).cast(pl.Int64))\n            elif col in [\"date_decision\"]:\n                df = df.with_columns(pl.col(col).cast(pl.Date))\n            elif col[-1] in (\"P\", \"A\"):\n                df = df.with_columns(pl.col(col).cast(pl.Float64))\n            elif col[-1] in (\"M\",):\n                df = df.with_columns(pl.col(col).cast(pl.String))\n            elif col[-1] in (\"D\",):\n                df = df.with_columns(pl.col(col).cast(pl.Date))\n        return df\n\n    def handle_dates(df):\n        for col in df.columns:\n            if col[-1] in (\"D\",):\n                df = df.with_columns(pl.col(col) - pl.col(\"date_decision\"))  #!!?\n                df = df.with_columns(pl.col(col).dt.total_days()) # t - t-1\n        df = df.drop(\"date_decision\", \"MONTH\")\n        return df\n\n    def filter_cols(df):\n        for col in df.columns:\n            if col not in [\"target\", \"case_id\", \"WEEK_NUM\"]:\n                isnull = df[col].is_null().mean()\n                if isnull > 0.7:\n                    df = df.drop(col)\n        \n        for col in df.columns:\n            if (col not in [\"target\", \"case_id\", \"WEEK_NUM\"]) & (df[col].dtype == pl.String):\n                freq = df[col].n_unique()\n                if (freq == 1) | (freq > 200):\n                    df = df.drop(col)\n        \n        return df\n\n\nclass Aggregator:\n    @staticmethod\n    def num_expr(df):\n         # 末尾が 「P 」または 「A 」の数値列をフィルタリング\n        cols = [col for col in df.columns if col[-1] in (\"P\", \"A\")]\n\n        # 各数値列について、最大値を計算する集約式を構築\n        expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n\n        # 得られる結果はリストで、このリストの各要素はPolars式\n        # これらの式は, 元のデータフレームの特定の列に対して実行される操作(この場合は最大値の計算)を定義し\n        # その結果に新しい列名を割り当てる\n        # リスト内の各式は元データの列に対応\n        # 集計処理が実行された後, 各式は元の列の名前に接頭辞 「max_」 を付加した新しい列を生成\n\n        return expr_max\n\n    @staticmethod\n    def date_expr(df):\n        # 筛选出列名以\"D\"结尾的日期列\n        cols = [col for col in df.columns if col[-1] in (\"D\",)]\n\n        # 为每个日期列构建一个计算最大值的聚合表达式\n        expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n        return expr_max\n\n    @staticmethod\n    def str_expr(df):\n        # 筛选出列名以\"M\"结尾的字符串列\n        cols = [col for col in df.columns if col[-1] in (\"M\",)]\n\n        # 为每个字符串列构建一个计算最大值的聚合表达式\n        expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n        return expr_max\n\n    @staticmethod\n    def other_expr(df):\n        # 筛选出列名以\"T\"或\"L\"结尾的其他类型列\n        cols = [col for col in df.columns if col[-1] in (\"T\", \"L\")]\n\n        # 为这些其他类型列构建一个计算最大值的聚合表达式\n        expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n        return expr_max\n\n    @staticmethod\n    def count_expr(df):\n        # 筛选出列名中包含\"num_group\"的列，这些列可能涉及分组计数或索引\n        cols = [col for col in df.columns if \"num_group\" in col]\n        # 为这些分组计数或索引列构建一个计算最大值的聚合表达式\n        expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n        return expr_max\n\n    @staticmethod\n    def get_exprs(df):\n        # 组合上述所有类型的聚合表达式\n        exprs = Aggregator.num_expr(df) + \\\n                Aggregator.date_expr(df) + \\\n                Aggregator.str_expr(df) + \\\n                Aggregator.other_expr(df) + \\\n                Aggregator.count_expr(df)\n        # 返回包含所有聚合表达式的列表\n        return exprs\n\ndef read_file(path, depth=None):\n    df = pl.read_parquet(path)\n    df = df.pipe(Pipeline.set_table_dtypes)\n    if depth in [1,2]:\n        df = df.group_by(\"case_id\").agg(Aggregator.get_exprs(df)) \n    return df\n\ndef read_files(regex_path, depth=None):\n    chunks = []\n    \n    for path in glob(str(regex_path)):\n        df = pl.read_parquet(path)\n        df = df.pipe(Pipeline.set_table_dtypes)\n        if depth in [1, 2]:\n            df = df.group_by(\"case_id\").agg(Aggregator.get_exprs(df))\n        chunks.append(df)\n    \n    df = pl.concat(chunks, how=\"vertical_relaxed\")\n    df = df.unique(subset=[\"case_id\"])\n    return df\n\ndef feature_eng(df_base, depth_0, depth_1, depth_2):\n    df_base = (\n        df_base\n        .with_columns(\n            month_decision = pl.col(\"date_decision\").dt.month(),\n            weekday_decision = pl.col(\"date_decision\").dt.weekday(),\n        )\n    )\n    for i, df in enumerate(depth_0 + depth_1 + depth_2):\n        df_base = df_base.join(df, how=\"left\", on=\"case_id\", suffix=f\"_{i}\")\n    df_base = df_base.pipe(Pipeline.handle_dates)\n    return df_base\n\ndef to_pandas(df_data, cat_cols=None):\n    df_data = df_data.to_pandas()\n    if cat_cols is None:\n        cat_cols = list(df_data.select_dtypes(\"object\").columns)\n    df_data[cat_cols] = df_data[cat_cols].astype(\"category\")\n    return df_data, cat_cols\n\ndef reduce_mem_usage(df):\n    \"\"\" iterate through all the columns of a dataframe and modify the data type\n        to reduce memory usage.        \n    \"\"\"\n    start_mem = df.memory_usage().sum() / 1024**2\n    print('Memory usage of dataframe is {:.2f} MB'.format(start_mem))\n    \n    for col in df.columns:\n        col_type = df[col].dtype\n        if str(col_type)==\"category\":\n            continue\n        \n        if col_type != object:\n            c_min = df[col].min()\n            c_max = df[col].max()\n            if str(col_type)[:3] == 'int':\n                if c_min > np.iinfo(np.int8).min and c_max < np.iinfo(np.int8).max:\n                    df[col] = df[col].astype(np.int8)\n                elif c_min > np.iinfo(np.int16).min and c_max < np.iinfo(np.int16).max:\n                    df[col] = df[col].astype(np.int16)\n                elif c_min > np.iinfo(np.int32).min and c_max < np.iinfo(np.int32).max:\n                    df[col] = df[col].astype(np.int32)\n                elif c_min > np.iinfo(np.int64).min and c_max < np.iinfo(np.int64).max:\n                    df[col] = df[col].astype(np.int64)  \n            else:\n                if c_min > np.finfo(np.float16).min and c_max < np.finfo(np.float16).max:\n                    df[col] = df[col].astype(np.float16)\n                elif c_min > np.finfo(np.float32).min and c_max < np.finfo(np.float32).max:\n                    df[col] = df[col].astype(np.float32)\n                else:\n                    df[col] = df[col].astype(np.float64)\n        else:\n            continue\n    end_mem = df.memory_usage().sum() / 1024**2\n    print('Memory usage after optimization is: {:.2f} MB'.format(end_mem))\n    print('Decreased by {:.1f}%'.format(100 * (start_mem - end_mem) / start_mem))\n    \n    return df","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:46:16.133882Z","iopub.execute_input":"2024-05-27T05:46:16.134182Z","iopub.status.idle":"2024-05-27T05:46:16.170448Z","shell.execute_reply.started":"2024-05-27T05:46:16.134157Z","shell.execute_reply":"2024-05-27T05:46:16.169510Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"ROOT            = Path(\"/kaggle/input/home-credit-credit-risk-model-stability\")\n\nTRAIN_DIR       = ROOT / \"parquet_files\" / \"train\"\nTEST_DIR        = ROOT / \"parquet_files\" / \"test\"","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:46:16.171859Z","iopub.execute_input":"2024-05-27T05:46:16.172116Z","iopub.status.idle":"2024-05-27T05:46:16.184489Z","shell.execute_reply.started":"2024-05-27T05:46:16.172094Z","shell.execute_reply":"2024-05-27T05:46:16.183563Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%%time\ndata_store = {\n    \"df_base\": read_file(TRAIN_DIR / \"train_base.parquet\"),\n    \"depth_0\": [\n        read_file(TRAIN_DIR / \"train_static_cb_0.parquet\"),\n        read_files(TRAIN_DIR / \"train_static_0_*.parquet\"),\n    ],\n    \"depth_1\": [\n        read_files(TRAIN_DIR / \"train_applprev_1_*.parquet\", 1),\n        read_file(TRAIN_DIR / \"train_tax_registry_a_1.parquet\", 1),\n        read_file(TRAIN_DIR / \"train_tax_registry_b_1.parquet\", 1),\n        read_file(TRAIN_DIR / \"train_tax_registry_c_1.parquet\", 1),\n        read_files(TRAIN_DIR / \"train_credit_bureau_a_1_*.parquet\", 1),\n        read_file(TRAIN_DIR / \"train_credit_bureau_b_1.parquet\", 1),\n        read_file(TRAIN_DIR / \"train_other_1.parquet\", 1),\n        read_file(TRAIN_DIR / \"train_person_1.parquet\", 1),\n        read_file(TRAIN_DIR / \"train_deposit_1.parquet\", 1),\n        read_file(TRAIN_DIR / \"train_debitcard_1.parquet\", 1),\n    ],\n    \"depth_2\": [\n        read_file(TRAIN_DIR / \"train_credit_bureau_b_2.parquet\", 2),\n        read_files(TRAIN_DIR / \"train_credit_bureau_a_2_*.parquet\", 2),\n        read_file(TRAIN_DIR / \"train_applprev_2.parquet\", 2),\n        read_file(TRAIN_DIR / \"train_person_2.parquet\", 2)\n    ]\n}","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:46:16.275538Z","iopub.execute_input":"2024-05-27T05:46:16.275872Z","iopub.status.idle":"2024-05-27T05:48:23.254672Z","shell.execute_reply.started":"2024-05-27T05:46:16.275846Z","shell.execute_reply":"2024-05-27T05:48:23.253703Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%%time\ndf_train = feature_eng(**data_store)\nprint(\"train data shape:\\t\", df_train.shape)\ndel data_store\ngc.collect()\ndf_train = df_train.pipe(Pipeline.filter_cols)\nprint(type(df_train))\ndf_train, cat_cols = to_pandas(df_train)\ndf_train = reduce_mem_usage(df_train)\nprint(\"train data shape:\\t\", df_train.shape)\nnums=df_train.select_dtypes(exclude='category').columns\n\nfrom itertools import combinations, permutations\n#df_train=df_train[nums]\nnans_df = df_train[nums].isna()\nnans_groups={}\nfor col in nums:\n    cur_group = nans_df[col].sum()\n    try:\n        nans_groups[cur_group].append(col)\n    except:\n        nans_groups[cur_group]=[col]\ndel nans_df; x=gc.collect()\n\ndef reduce_group(grps):\n    use = []\n    for g in grps:\n        mx = 0; vx = g[0]\n        for gg in g:\n            n = df_train[gg].nunique()\n            if n>mx:\n                mx = n\n                vx = gg\n            #print(str(gg)+'-'+str(n),', ',end='')\n        use.append(vx)\n        #print()\n    print('Use these',use)\n    return use\n\ndef group_columns_by_correlation(matrix, threshold=0.8):\n    # 计算列之间的相关性\n    correlation_matrix = matrix.corr()\n\n    # 分组列\n    groups = []\n    remaining_cols = list(matrix.columns)\n    while remaining_cols:\n        col = remaining_cols.pop(0)\n        group = [col]\n        correlated_cols = [col]\n        for c in remaining_cols:\n            if correlation_matrix.loc[col, c] >= threshold:\n                group.append(c)\n                correlated_cols.append(c)\n        groups.append(group)\n        remaining_cols = [c for c in remaining_cols if c not in correlated_cols]\n    \n    return groups\n\nuses=[]\nfor k,v in nans_groups.items():\n    if len(v)>1:\n            Vs = nans_groups[k]\n            #cross_features=list(combinations(Vs, 2))\n            #make_corr(Vs)\n            grps= group_columns_by_correlation(df_train[Vs], threshold=0.8)\n            use=reduce_group(grps)\n            uses=uses+use\n            #make_corr(use)\n    else:\n        uses=uses+v\n    print('####### NAN count =',k)\nprint(uses)\nprint(len(uses))\nuses=uses+list(df_train.select_dtypes(include='category').columns)\nprint(len(uses))\ndf_train=df_train[uses]","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:48:23.257000Z","iopub.execute_input":"2024-05-27T05:48:23.257446Z","iopub.status.idle":"2024-05-27T05:49:28.698849Z","shell.execute_reply.started":"2024-05-27T05:48:23.257412Z","shell.execute_reply":"2024-05-27T05:49:28.697768Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sample = pd.read_csv(\"/kaggle/input/home-credit-credit-risk-model-stability/sample_submission.csv\")\ndevice='gpu'\n#n_samples=200000\nn_est=6000\nDRY_RUN = True if sample.shape[0] == 10 else False   \nif DRY_RUN:\n    device='cpu'\n    df_train = df_train.iloc[:60000]\n    #n_samples=10000\n    n_est=600\nprint(device)","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:49:28.700253Z","iopub.execute_input":"2024-05-27T05:49:28.700622Z","iopub.status.idle":"2024-05-27T05:49:28.711952Z","shell.execute_reply.started":"2024-05-27T05:49:28.700586Z","shell.execute_reply":"2024-05-27T05:49:28.710784Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data_store = {\n    \"df_base\": read_file(TEST_DIR / \"test_base.parquet\"),\n    \"depth_0\": [\n        read_file(TEST_DIR / \"test_static_cb_0.parquet\"),\n        read_files(TEST_DIR / \"test_static_0_*.parquet\"),\n    ],\n    \"depth_1\": [\n        read_files(TEST_DIR / \"test_applprev_1_*.parquet\", 1),\n        read_file(TEST_DIR / \"test_tax_registry_a_1.parquet\", 1),\n        read_file(TEST_DIR / \"test_tax_registry_b_1.parquet\", 1),\n        read_file(TEST_DIR / \"test_tax_registry_c_1.parquet\", 1),\n        read_files(TEST_DIR / \"test_credit_bureau_a_1_*.parquet\", 1),\n        read_file(TEST_DIR / \"test_credit_bureau_b_1.parquet\", 1),\n        read_file(TEST_DIR / \"test_other_1.parquet\", 1),\n        read_file(TEST_DIR / \"test_person_1.parquet\", 1),\n        read_file(TEST_DIR / \"test_deposit_1.parquet\", 1),\n        read_file(TEST_DIR / \"test_debitcard_1.parquet\", 1),\n    ],\n    \"depth_2\": [\n        read_file(TEST_DIR / \"test_credit_bureau_b_2.parquet\", 2),\n        read_files(TEST_DIR / \"test_credit_bureau_a_2_*.parquet\", 2),\n        read_file(TEST_DIR / \"test_applprev_2.parquet\", 2),\n        read_file(TEST_DIR / \"test_person_2.parquet\", 2)\n    ]\n}","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:49:28.713202Z","iopub.execute_input":"2024-05-27T05:49:28.713601Z","iopub.status.idle":"2024-05-27T05:49:28.904487Z","shell.execute_reply.started":"2024-05-27T05:49:28.713542Z","shell.execute_reply":"2024-05-27T05:49:28.903356Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_test = feature_eng(**data_store)\nprint(\"test data shape:\\t\", df_test.shape)\ndel data_store\ngc.collect()\ndf_test = df_test.select([col for col in df_train.columns if col != \"target\"])\nprint(\"train data shape:\\t\", df_train.shape)\nprint(\"test data shape:\\t\", df_test.shape)\n\ndf_test, cat_cols = to_pandas(df_test, cat_cols)\ndf_test = reduce_mem_usage(df_test)\n\ngc.collect()","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:49:28.907396Z","iopub.execute_input":"2024-05-27T05:49:28.907712Z","iopub.status.idle":"2024-05-27T05:49:29.850455Z","shell.execute_reply.started":"2024-05-27T05:49:28.907685Z","shell.execute_reply":"2024-05-27T05:49:29.849466Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Feature Selection","metadata":{}},{"cell_type":"code","source":"df_train_p = df_train\ndf_test_p = df_test","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:49:29.851510Z","iopub.execute_input":"2024-05-27T05:49:29.851794Z","iopub.status.idle":"2024-05-27T05:49:29.857560Z","shell.execute_reply.started":"2024-05-27T05:49:29.851769Z","shell.execute_reply":"2024-05-27T05:49:29.856446Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_train_p[\"target\"].value_counts()","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:49:29.858821Z","iopub.execute_input":"2024-05-27T05:49:29.859152Z","iopub.status.idle":"2024-05-27T05:49:29.871630Z","shell.execute_reply.started":"2024-05-27T05:49:29.859128Z","shell.execute_reply":"2024-05-27T05:49:29.870687Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"y = df_train[\"target\"]\nweeks = df_train[\"WEEK_NUM\"]\ndf_train= df_train.drop(columns=[\"target\", \"case_id\", \"WEEK_NUM\"])\ncv = StratifiedGroupKFold(n_splits=5, shuffle=False)\n","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:49:29.872916Z","iopub.execute_input":"2024-05-27T05:49:29.873338Z","iopub.status.idle":"2024-05-27T05:49:29.929193Z","shell.execute_reply.started":"2024-05-27T05:49:29.873297Z","shell.execute_reply":"2024-05-27T05:49:29.928457Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_train[cat_cols] = df_train[cat_cols].astype(str)\ndf_test[cat_cols] = df_test[cat_cols].astype(str)","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:49:29.930403Z","iopub.execute_input":"2024-05-27T05:49:29.931065Z","iopub.status.idle":"2024-05-27T05:49:30.187697Z","shell.execute_reply.started":"2024-05-27T05:49:29.931027Z","shell.execute_reply":"2024-05-27T05:49:30.186770Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_train","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:49:30.189072Z","iopub.execute_input":"2024-05-27T05:49:30.189460Z","iopub.status.idle":"2024-05-27T05:49:30.272023Z","shell.execute_reply.started":"2024-05-27T05:49:30.189424Z","shell.execute_reply":"2024-05-27T05:49:30.270968Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_test","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:49:30.273307Z","iopub.execute_input":"2024-05-27T05:49:30.273711Z","iopub.status.idle":"2024-05-27T05:49:30.308486Z","shell.execute_reply.started":"2024-05-27T05:49:30.273676Z","shell.execute_reply":"2024-05-27T05:49:30.307406Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_train[\"month_decision\"].value_counts()","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:49:30.309548Z","iopub.execute_input":"2024-05-27T05:49:30.309838Z","iopub.status.idle":"2024-05-27T05:49:30.317989Z","shell.execute_reply.started":"2024-05-27T05:49:30.309813Z","shell.execute_reply":"2024-05-27T05:49:30.316886Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"params = {\n    \"boosting_type\": \"gbdt\",\n    \"objective\": \"binary\",\n    \"metric\": \"auc\",\n    \"max_depth\": 10,  \n    \"learning_rate\": 0.05,\n    \"n_estimators\": 2000,  \n    \"colsample_bytree\": 0.8,\n    \"colsample_bynode\": 0.8,\n    \"verbose\": -1,\n    \"random_state\": 42,\n    \"reg_alpha\": 0.1,\n    \"reg_lambda\": 10,\n    \"extra_trees\":True,\n    'num_leaves':64,\n    \"device\": device, \n    \"verbose\": -1,\n}\n\nparams2 = {\n    \"booster\": \"gbtree\",\n    \"objective\": \"binary:logistic\",\n    \"eval_metric\": \"auc\",\n    \"max_depth\": 10,\n    \"learning_rate\": 0.05,\n    \"n_estimators\": 1000,\n    \"colsample_bytree\": 0.8,\n    \"colsample_bynode\": 0.8,\n    \"alpha\": 0.1,  \n    \"lambda\": 10,  \n    \"tree_method\": 'gpu_hist' if device == 'gpu' else 'auto',\n    \"random_state\": 42,\n    \"verbosity\": 0,\n    \"enable_categorical\":True,\n}\n","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:49:30.319342Z","iopub.execute_input":"2024-05-27T05:49:30.320180Z","iopub.status.idle":"2024-05-27T05:49:30.327941Z","shell.execute_reply.started":"2024-05-27T05:49:30.320153Z","shell.execute_reply":"2024-05-27T05:49:30.326968Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%time\nfrom catboost import CatBoostClassifier, Pool\nimport xgboost as xgb\n\nfitted_models_cat = []\nfitted_models_lgb = []\nfitted_models_xgb = []\n\ncv_scores_cat = []\ncv_scores_lgb = []\ncv_scores_xgb = []\n\n\nfor idx_train, idx_valid in cv.split(df_train, y, groups=weeks):#\n    X_train, y_train = df_train.iloc[idx_train], y.iloc[idx_train]# \n    X_valid, y_valid = df_train.iloc[idx_valid], y.iloc[idx_valid]\n    train_pool = Pool(X_train, y_train,cat_features=cat_cols)\n    val_pool = Pool(X_valid, y_valid,cat_features=cat_cols)\n    clf = CatBoostClassifier(\n    eval_metric='AUC',\n    task_type='GPU',\n    learning_rate=0.03,\n    iterations=n_est)\n    random_seed=3107\n    clf.fit(train_pool, eval_set=val_pool,verbose=300)\n    fitted_models_cat.append(clf)\n    y_pred_valid = clf.predict_proba(X_valid)[:,1]\n    auc_score = roc_auc_score(y_valid, y_pred_valid)\n    cv_scores_cat.append(auc_score)\n    \n    \n    X_train[cat_cols] = X_train[cat_cols].astype(\"category\")\n    X_valid[cat_cols] = X_valid[cat_cols].astype(\"category\")\n    \n    model = lgb.LGBMClassifier(**params)\n    model.fit(\n        X_train, y_train,\n        eval_set = [(X_valid, y_valid)],\n        callbacks = [lgb.log_evaluation(200), lgb.early_stopping(100)] )\n    \n    fitted_models_lgb.append(model)\n    y_pred_valid = model.predict_proba(X_valid)[:,1]\n    auc_score = roc_auc_score(y_valid, y_pred_valid)\n    cv_scores_lgb.append(auc_score)\n    \n    \n    model2 = xgb.XGBClassifier(**params2)\n    model2.fit(\n        X_train, y_train,\n        eval_set=[(X_valid, y_valid)],\n        early_stopping_rounds=100, verbose=False)\n    \n    fitted_models_xgb.append(model2)\n    \n    y_pred_valid = model2.predict_proba(X_valid)[:, 1]\n    auc_score = roc_auc_score(y_valid, y_pred_valid)\n    cv_scores_xgb.append(auc_score)\n    \n    del clf, model, model2\n    gc.collect()\n    \n    \nprint(\"CV AUC scores: \", cv_scores_cat)\nprint(\"Maximum CV AUC score: \", max(cv_scores_cat))\n\n\nprint(\"CV AUC scores: \", cv_scores_lgb)\nprint(\"Maximum CV AUC score: \", max(cv_scores_lgb))\n\nprint(\"CV AUC scores: \", cv_scores_xgb)\nprint(\"Maximum CV AUC score: \", max(cv_scores_xgb))","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:49:30.333125Z","iopub.execute_input":"2024-05-27T05:49:30.333439Z","iopub.status.idle":"2024-05-27T05:59:00.159249Z","shell.execute_reply.started":"2024-05-27T05:49:30.333415Z","shell.execute_reply":"2024-05-27T05:59:00.158144Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#del y \n#del weeks\n#del df_train\n\n#gc.collect()","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:59:00.160416Z","iopub.execute_input":"2024-05-27T05:59:00.160677Z","iopub.status.idle":"2024-05-27T05:59:00.164943Z","shell.execute_reply.started":"2024-05-27T05:59:00.160653Z","shell.execute_reply":"2024-05-27T05:59:00.163897Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class VotingModel(BaseEstimator, RegressorMixin):\n    def __init__(self, estimators):\n        super().__init__()\n        self.estimators = estimators\n        \n    def fit(self, X, y=None):\n        return self\n    \n    def predict(self, X):\n        y_preds = [estimator.predict(X) for estimator in self.estimators]\n        return np.mean(y_preds, axis=0)\n    \n    def predict_proba(self, X):\n        \n        y_preds = [estimator.predict_proba(X) for estimator in self.estimators[:5]]\n        \n        X[cat_cols] = X[cat_cols].astype(\"category\")\n        y_preds += [estimator.predict_proba(X) for estimator in self.estimators[5:10]]\n        y_preds+=y_preds #tang trong so\n        y_preds += [estimator.predict_proba(X) for estimator in self.estimators[10:]]\n        print(len(y_preds))\n        return np.mean(y_preds, axis=0)\n\nmodel = VotingModel(fitted_models_cat+fitted_models_lgb+fitted_models_xgb)","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:59:00.166510Z","iopub.execute_input":"2024-05-27T05:59:00.166833Z","iopub.status.idle":"2024-05-27T05:59:00.177187Z","shell.execute_reply.started":"2024-05-27T05:59:00.166802Z","shell.execute_reply":"2024-05-27T05:59:00.176414Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.model_selection import cross_val_predict\nfrom catboost import CatBoostClassifier, Pool\nfrom lightgbm import LGBMClassifier\nfrom sklearn.ensemble import RandomForestClassifier\nfrom xgboost import XGBClassifier\nfrom sklearn.metrics import roc_auc_score\nimport pandas as pd\n\n# 第一层模型\nmodels = [\n    ('CatBoost', CatBoostClassifier(eval_metric='AUC', task_type='GPU', learning_rate=0.03, iterations=n_est, random_seed=3107)),\n    ('LightGBM', LGBMClassifier(**params)),\n    ('XGBoost', XGBClassifier(**params2))\n]\n\n# 第二层模型\nfrom sklearn.ensemble import GradientBoostingClassifier\n\nparams = {\n    'n_estimators': 12,\n    'learning_rate': 0.1,\n    'max_depth': 3,\n    'min_samples_split': 3,\n    'min_samples_leaf': 1\n}\n\nmeta_model = GradientBoostingClassifier(**params)\n\n\n\n# 存储第一层模型和相应的AUC分数\nfitted_models_cb = []\nfitted_models_lgb = []\nfitted_models_xgb = []\ncv_scores_cb = []\ncv_scores_lgb = []\ncv_scores_xgb = []\n\n# 存储第一层模型的预测结果\nmeta_features = pd.DataFrame(index=df_train.index, columns=['CatBoost', 'LightGBM', 'XGBoost'])\n\nfor name, model in models:\n    for idx_train, idx_valid in cv.split(df_train, y, groups=weeks):\n        X_train, y_train = df_train.iloc[idx_train], y.iloc[idx_train]\n        X_valid, y_valid = df_train.iloc[idx_valid], y.iloc[idx_valid]\n\n        if name == 'CatBoost':\n            X_train[cat_cols] = X_train[cat_cols].astype(str)\n            X_valid[cat_cols] = X_valid[cat_cols].astype(str)\n            train_pool = Pool(X_train, y_train, cat_features=cat_cols)\n            val_pool = Pool(X_valid, y_valid, cat_features=cat_cols)\n            model.fit(train_pool, eval_set=val_pool, verbose=False)\n            y_pred_valid = model.predict_proba(val_pool)[:, 1]\n            fitted_models_cb.append(model)\n            auc_score = roc_auc_score(y_valid, y_pred_valid)\n            cv_scores_cb.append(auc_score)\n        elif name == 'LightGBM':\n            X_train[cat_cols] = X_train[cat_cols].astype('category')\n            X_valid[cat_cols] = X_valid[cat_cols].astype('category')\n            model.fit(X_train, y_train, eval_set=[(X_valid, y_valid)], callbacks=[lgb.log_evaluation(200), lgb.early_stopping(100)])\n            fitted_models_lgb.append(model)\n            y_pred_valid = model.predict_proba(X_valid)[:, 1]\n            auc_score = roc_auc_score(y_valid, y_pred_valid)\n            cv_scores_lgb.append(auc_score)\n        else:  # XGBoost\n            X_train[cat_cols] = X_train[cat_cols].astype('category')\n            X_valid[cat_cols] = X_valid[cat_cols].astype('category')\n            model.fit(X_train, y_train, eval_set=[(X_valid, y_valid)], early_stopping_rounds=100, verbose=False)\n            fitted_models_xgb.append(model)\n            y_pred_valid = model.predict_proba(X_valid)[:, 1]\n            auc_score = roc_auc_score(y_valid, y_pred_valid)\n            cv_scores_xgb.append(auc_score)\n\n        meta_features.loc[X_valid.index, name] = y_pred_valid","metadata":{"execution":{"iopub.status.busy":"2024-05-27T05:59:00.178841Z","iopub.execute_input":"2024-05-27T05:59:00.179479Z","iopub.status.idle":"2024-05-27T06:08:23.335561Z","shell.execute_reply.started":"2024-05-27T05:59:00.179446Z","shell.execute_reply":"2024-05-27T06:08:23.334609Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"meta_model.fit(meta_features, y)","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:08:23.336869Z","iopub.execute_input":"2024-05-27T06:08:23.337876Z","iopub.status.idle":"2024-05-27T06:08:24.353957Z","shell.execute_reply.started":"2024-05-27T06:08:23.337842Z","shell.execute_reply":"2024-05-27T06:08:24.352978Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 特徴量重要度から新たなtrain, testデータを作成","metadata":{}},{"cell_type":"code","source":"df_test_week = df_test[[\"WEEK_NUM\"]]\ndf_test_id = df_test[[\"case_id\"]]","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:08:24.355404Z","iopub.execute_input":"2024-05-27T06:08:24.355767Z","iopub.status.idle":"2024-05-27T06:08:24.361736Z","shell.execute_reply.started":"2024-05-27T06:08:24.355730Z","shell.execute_reply":"2024-05-27T06:08:24.360737Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_test = df_test.drop(columns=[\"WEEK_NUM\"])\ndf_test = df_test.drop(columns=[\"case_id\"])","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:08:24.363195Z","iopub.execute_input":"2024-05-27T06:08:24.363597Z","iopub.status.idle":"2024-05-27T06:08:24.400633Z","shell.execute_reply.started":"2024-05-27T06:08:24.363559Z","shell.execute_reply":"2024-05-27T06:08:24.399775Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#XGBoost\nfeature_importances_xgb = pd.DataFrame([model.feature_importances_ for model in fitted_models_xgb],columns = df_test.columns).mean()\nfeature_importances_xgb_df = pd.DataFrame(feature_importances_xgb)\ndf_sort = pd.DataFrame(feature_importances_xgb_df.sort_values(0, ascending=False))\ndf_sort[\"feature\"] = df_sort.index\n\ndf_sort.index = range(0, len(df_sort))\ndf_sort.columns = [\"importance\", \"feature\"]\ndf_xgb_sort_30 = df_sort[:30]\n\n#LightGBM\nfeature_importances_lgb = pd.DataFrame([model.feature_importances_ for model in fitted_models_lgb],columns = df_test.columns).mean()\nfeature_importances_lgb_df = pd.DataFrame(feature_importances_lgb)\ndf_sort = pd.DataFrame(feature_importances_lgb_df.sort_values(0, ascending=False))\ndf_sort[\"feature\"] = df_sort.index\n\ndf_sort.index = range(0, len(df_sort))\ndf_sort.columns = [\"importance\", \"feature\"]\ndf_lgb_sort_30 = df_sort[:30]\n\n#CatBoost\nfeature_importances_cat = pd.DataFrame([model.feature_importances_ for model in fitted_models_cat],columns = df_test.columns).mean()\nfeature_importances_cat_df = pd.DataFrame(feature_importances_cat)\ndf_sort = pd.DataFrame(feature_importances_cat_df.sort_values(0, ascending=False))\ndf_sort[\"feature\"] = df_sort.index\n\ndf_sort.index = range(0, len(df_sort))\ndf_sort.columns = [\"importance\", \"feature\"]\ndf_cat_sort_30 = df_sort[:30]\n\n\n#df_train, df_test 更新\ndf_select = ['month_decision', 'weekday_decision']\nfor i in range(0, len(df_cat_sort_30)):\n    df_select.append(df_cat_sort_30[\"feature\"][i])\n    df_select.append(df_xgb_sort_30[\"feature\"][i])\n    df_select.append(df_lgb_sort_30[\"feature\"][i])\n\n#df_selectのかぶりを除去\ndf_re_select = list(dict.fromkeys(df_select))\n\ndf_train = df_train[df_re_select]\ndf_test = df_test[df_re_select]\ndf_train","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:08:24.401873Z","iopub.execute_input":"2024-05-27T06:08:24.402173Z","iopub.status.idle":"2024-05-27T06:08:24.570893Z","shell.execute_reply.started":"2024-05-27T06:08:24.402148Z","shell.execute_reply":"2024-05-27T06:08:24.569784Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_test","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:08:24.572137Z","iopub.execute_input":"2024-05-27T06:08:24.572446Z","iopub.status.idle":"2024-05-27T06:08:24.617846Z","shell.execute_reply.started":"2024-05-27T06:08:24.572418Z","shell.execute_reply":"2024-05-27T06:08:24.616784Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_test_tmp = df_test.copy()","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:08:24.618931Z","iopub.execute_input":"2024-05-27T06:08:24.619212Z","iopub.status.idle":"2024-05-27T06:08:24.626390Z","shell.execute_reply.started":"2024-05-27T06:08:24.619187Z","shell.execute_reply":"2024-05-27T06:08:24.625251Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_test = pd.concat([df_test_id, df_test_week, df_test_tmp], axis=1)\n#df_id = pd.DataFrame(df_test.index)\ndf_test","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:08:24.627912Z","iopub.execute_input":"2024-05-27T06:08:24.628311Z","iopub.status.idle":"2024-05-27T06:08:24.669979Z","shell.execute_reply.started":"2024-05-27T06:08:24.628273Z","shell.execute_reply":"2024-05-27T06:08:24.668982Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#catcolls直す\ndf_train = pl.from_pandas(df_train)\ndf_train, cat_cols = to_pandas(df_train)","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:08:24.671204Z","iopub.execute_input":"2024-05-27T06:08:24.671556Z","iopub.status.idle":"2024-05-27T06:08:24.891998Z","shell.execute_reply.started":"2024-05-27T06:08:24.671521Z","shell.execute_reply":"2024-05-27T06:08:24.891171Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# モデル再実行","metadata":{}},{"cell_type":"code","source":"params = {\n    \"boosting_type\": \"gbdt\",\n    \"objective\": \"binary\",\n    \"metric\": \"auc\",\n    \"max_depth\": 10,  \n    \"learning_rate\": 0.05,\n    \"n_estimators\": 2000,  \n    \"colsample_bytree\": 0.8,\n    \"colsample_bynode\": 0.8,\n    \"verbose\": -1,\n    \"random_state\": 42,\n    \"reg_alpha\": 0.1,\n    \"reg_lambda\": 10,\n    \"extra_trees\":True,\n    'num_leaves':64,\n    \"device\": device, \n    \"verbose\": -1,\n}\n\nparams2 = {\n    \"booster\": \"gbtree\",\n    \"objective\": \"binary:logistic\",\n    \"eval_metric\": \"auc\",\n    \"max_depth\": 10,\n    \"learning_rate\": 0.05,\n    \"n_estimators\": 1000,\n    \"colsample_bytree\": 0.8,\n    \"colsample_bynode\": 0.8,\n    \"alpha\": 0.1,  \n    \"lambda\": 10,  \n    \"tree_method\": 'gpu_hist' if device == 'gpu' else 'auto',\n    \"random_state\": 42,\n    \"verbosity\": 0,\n    \"enable_categorical\":True,\n}","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:08:24.893125Z","iopub.execute_input":"2024-05-27T06:08:24.893396Z","iopub.status.idle":"2024-05-27T06:08:24.900939Z","shell.execute_reply.started":"2024-05-27T06:08:24.893373Z","shell.execute_reply":"2024-05-27T06:08:24.899977Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%time\nfrom catboost import CatBoostClassifier, Pool\nimport xgboost as xgb\n\nfitted_models_cat = []\nfitted_models_lgb = []\nfitted_models_xgb = []\n\ncv_scores_cat = []\ncv_scores_lgb = []\ncv_scores_xgb = []\n\n\nfor idx_train, idx_valid in cv.split(df_train, y, groups=weeks):#\n    X_train, y_train = df_train.iloc[idx_train], y.iloc[idx_train]# \n    X_valid, y_valid = df_train.iloc[idx_valid], y.iloc[idx_valid]\n    train_pool = Pool(X_train, y_train,cat_features=cat_cols)\n    val_pool = Pool(X_valid, y_valid,cat_features=cat_cols)\n    clf = CatBoostClassifier(\n    eval_metric='AUC',\n    task_type='GPU',\n    learning_rate=0.03,\n    iterations=n_est)\n    random_seed=3107\n    clf.fit(train_pool, eval_set=val_pool,verbose=300)\n    fitted_models_cat.append(clf)\n    y_pred_valid = clf.predict_proba(X_valid)[:,1]\n    auc_score = roc_auc_score(y_valid, y_pred_valid)\n    cv_scores_cat.append(auc_score)\n    \n    \n    X_train[cat_cols] = X_train[cat_cols].astype(\"category\")\n    X_valid[cat_cols] = X_valid[cat_cols].astype(\"category\")\n    \n    model = lgb.LGBMClassifier(**params)\n    model.fit(\n        X_train, y_train,\n        eval_set = [(X_valid, y_valid)],\n        callbacks = [lgb.log_evaluation(200), lgb.early_stopping(100)] )\n    \n    fitted_models_lgb.append(model)\n    y_pred_valid = model.predict_proba(X_valid)[:,1]\n    auc_score = roc_auc_score(y_valid, y_pred_valid)\n    cv_scores_lgb.append(auc_score)\n    \n    \n    model2 = xgb.XGBClassifier(**params2)\n    model2.fit(\n        X_train, y_train,\n        eval_set=[(X_valid, y_valid)],\n        early_stopping_rounds=100, verbose=False)\n    \n    fitted_models_xgb.append(model2)\n    \n    y_pred_valid = model2.predict_proba(X_valid)[:, 1]\n    auc_score = roc_auc_score(y_valid, y_pred_valid)\n    cv_scores_xgb.append(auc_score)\n    \n    del clf, model, model2\n    gc.collect()\n    \n    \nprint(\"CV AUC scores: \", cv_scores_cat)\nprint(\"Maximum CV AUC score: \", max(cv_scores_cat))\n\n\nprint(\"CV AUC scores: \", cv_scores_lgb)\nprint(\"Maximum CV AUC score: \", max(cv_scores_lgb))\n\nprint(\"CV AUC scores: \", cv_scores_xgb)\nprint(\"Maximum CV AUC score: \", max(cv_scores_xgb))","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:08:24.902359Z","iopub.execute_input":"2024-05-27T06:08:24.902618Z","iopub.status.idle":"2024-05-27T06:12:37.153635Z","shell.execute_reply.started":"2024-05-27T06:08:24.902596Z","shell.execute_reply":"2024-05-27T06:12:37.152612Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class VotingModel(BaseEstimator, RegressorMixin):\n    def __init__(self, estimators):\n        super().__init__()\n        self.estimators = estimators\n        \n    def fit(self, X, y=None):\n        return self\n    \n    def predict(self, X):\n        y_preds = [estimator.predict(X) for estimator in self.estimators]\n        return np.mean(y_preds, axis=0)\n    \n    def predict_proba(self, X):\n        \n        y_preds = [estimator.predict_proba(X) for estimator in self.estimators[:5]]\n        \n        X[cat_cols] = X[cat_cols].astype(\"category\")\n        y_preds += [estimator.predict_proba(X) for estimator in self.estimators[5:10]]\n        y_preds+=y_preds #tang trong so\n        y_preds += [estimator.predict_proba(X) for estimator in self.estimators[10:]]\n        print(len(y_preds))\n        return np.mean(y_preds, axis=0)\n\nmodel = VotingModel(fitted_models_cat+fitted_models_lgb+fitted_models_xgb)","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:12:37.154994Z","iopub.execute_input":"2024-05-27T06:12:37.155380Z","iopub.status.idle":"2024-05-27T06:12:37.164295Z","shell.execute_reply.started":"2024-05-27T06:12:37.155349Z","shell.execute_reply":"2024-05-27T06:12:37.163359Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.model_selection import cross_val_predict\nfrom catboost import CatBoostClassifier, Pool\nfrom lightgbm import LGBMClassifier\nfrom sklearn.ensemble import RandomForestClassifier\nfrom xgboost import XGBClassifier\nfrom sklearn.metrics import roc_auc_score\nimport pandas as pd\n\n# 第一层模型\nmodels = [\n    ('CatBoost', CatBoostClassifier(eval_metric='AUC', task_type='GPU', learning_rate=0.03, iterations=n_est, random_seed=3107)),\n    ('LightGBM', LGBMClassifier(**params)),\n    ('XGBoost', XGBClassifier(**params2))\n]\n\n# 第二层模型\nfrom sklearn.ensemble import GradientBoostingClassifier\n\nparams = {\n    'n_estimators': 12,\n    'learning_rate': 0.1,\n    'max_depth': 3,\n    'min_samples_split': 3,\n    'min_samples_leaf': 1\n}\n\nmeta_model = GradientBoostingClassifier(**params)\n\n\n\n# 存储第一层模型和相应的AUC分数\nfitted_models_cb = []\nfitted_models_lgb = []\nfitted_models_xgb = []\ncv_scores_cb = []\ncv_scores_lgb = []\ncv_scores_xgb = []\n\n# 存储第一层模型的预测结果\nmeta_features = pd.DataFrame(index=df_train.index, columns=['CatBoost', 'LightGBM', 'XGBoost'])\n\nfor name, model in models:\n    for idx_train, idx_valid in cv.split(df_train, y, groups=weeks):\n        X_train, y_train = df_train.iloc[idx_train], y.iloc[idx_train]\n        X_valid, y_valid = df_train.iloc[idx_valid], y.iloc[idx_valid]\n\n        if name == 'CatBoost':\n            X_train[cat_cols] = X_train[cat_cols].astype(str)\n            X_valid[cat_cols] = X_valid[cat_cols].astype(str)\n            train_pool = Pool(X_train, y_train, cat_features=cat_cols)\n            val_pool = Pool(X_valid, y_valid, cat_features=cat_cols)\n            model.fit(train_pool, eval_set=val_pool, verbose=False)\n            y_pred_valid = model.predict_proba(val_pool)[:, 1]\n            fitted_models_cb.append(model)\n            auc_score = roc_auc_score(y_valid, y_pred_valid)\n            cv_scores_cb.append(auc_score)\n        elif name == 'LightGBM':\n            X_train[cat_cols] = X_train[cat_cols].astype('category')\n            X_valid[cat_cols] = X_valid[cat_cols].astype('category')\n            model.fit(X_train, y_train, eval_set=[(X_valid, y_valid)], callbacks=[lgb.log_evaluation(200), lgb.early_stopping(100)])\n            fitted_models_lgb.append(model)\n            y_pred_valid = model.predict_proba(X_valid)[:, 1]\n            auc_score = roc_auc_score(y_valid, y_pred_valid)\n            cv_scores_lgb.append(auc_score)\n        else:  # XGBoost\n            X_train[cat_cols] = X_train[cat_cols].astype('category')\n            X_valid[cat_cols] = X_valid[cat_cols].astype('category')\n            model.fit(X_train, y_train, eval_set=[(X_valid, y_valid)], early_stopping_rounds=100, verbose=False)\n            fitted_models_xgb.append(model)\n            y_pred_valid = model.predict_proba(X_valid)[:, 1]\n            auc_score = roc_auc_score(y_valid, y_pred_valid)\n            cv_scores_xgb.append(auc_score)\n\n        meta_features.loc[X_valid.index, name] = y_pred_valid","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:12:37.165754Z","iopub.execute_input":"2024-05-27T06:12:37.166069Z","iopub.status.idle":"2024-05-27T06:16:49.024511Z","shell.execute_reply.started":"2024-05-27T06:12:37.166043Z","shell.execute_reply":"2024-05-27T06:16:49.023569Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"meta_model.fit(meta_features, y)","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:16:49.025875Z","iopub.execute_input":"2024-05-27T06:16:49.029140Z","iopub.status.idle":"2024-05-27T06:16:50.008355Z","shell.execute_reply.started":"2024-05-27T06:16:49.029102Z","shell.execute_reply":"2024-05-27T06:16:50.007382Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Submision","metadata":{}},{"cell_type":"code","source":"df_test = df_test.drop(columns=[\"WEEK_NUM\"])\ndf_test = df_test.set_index(\"case_id\")","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:16:50.009482Z","iopub.execute_input":"2024-05-27T06:16:50.009772Z","iopub.status.idle":"2024-05-27T06:16:50.016685Z","shell.execute_reply.started":"2024-05-27T06:16:50.009747Z","shell.execute_reply":"2024-05-27T06:16:50.015726Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test_meta_features = pd.DataFrame(index=df_test.index, columns=['CatBoost', 'LightGBM', 'XGBoost'])","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:16:50.017946Z","iopub.execute_input":"2024-05-27T06:16:50.018293Z","iopub.status.idle":"2024-05-27T06:16:50.027453Z","shell.execute_reply.started":"2024-05-27T06:16:50.018258Z","shell.execute_reply":"2024-05-27T06:16:50.026521Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# CatBoost\n\n\nfor model in fitted_models_cat:\n    df_test[cat_cols] = df_test[cat_cols].astype(str)\n    y_pred_test = model.predict_proba(df_test)[:, 1]\n    test_meta_features['CatBoost'] = test_meta_features['CatBoost'].add(y_pred_test, fill_value=0)\n\ntest_meta_features['CatBoost'] /= len(fitted_models_cat)\n\n# LightGBM\nfor model in fitted_models_lgb:\n    df_test[cat_cols] = df_test[cat_cols].astype(\"category\")\n    y_pred_test = model.predict_proba(df_test)[:, 1]\n    test_meta_features['LightGBM'] = test_meta_features['LightGBM'].add(y_pred_test, fill_value=0)\n\ntest_meta_features['LightGBM'] /= len(fitted_models_lgb)\n\n# XGBoost\nfor model in fitted_models_xgb:\n    df_test[cat_cols] = df_test[cat_cols].astype(\"category\")\n    y_pred_test = model.predict_proba(df_test)[:, 1]\n    test_meta_features['XGBoost'] = test_meta_features['XGBoost'].add(y_pred_test, fill_value=0)\n\ntest_meta_features['XGBoost'] /= len(fitted_models_xgb)","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:16:50.028981Z","iopub.execute_input":"2024-05-27T06:16:50.029695Z","iopub.status.idle":"2024-05-27T06:16:50.268187Z","shell.execute_reply.started":"2024-05-27T06:16:50.029657Z","shell.execute_reply":"2024-05-27T06:16:50.267229Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test_meta_features","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:16:50.269396Z","iopub.execute_input":"2024-05-27T06:16:50.270242Z","iopub.status.idle":"2024-05-27T06:16:50.280763Z","shell.execute_reply.started":"2024-05-27T06:16:50.270210Z","shell.execute_reply":"2024-05-27T06:16:50.279961Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"y_pred = pd.Series(meta_model.predict_proba(test_meta_features)[:, 1], index=df_test.index)","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:16:50.282855Z","iopub.execute_input":"2024-05-27T06:16:50.283548Z","iopub.status.idle":"2024-05-27T06:16:50.291830Z","shell.execute_reply.started":"2024-05-27T06:16:50.283513Z","shell.execute_reply":"2024-05-27T06:16:50.290932Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_subm = pd.read_csv(ROOT / \"sample_submission.csv\")\ndf_subm = df_subm.set_index(\"case_id\")\n\ndf_subm[\"score\"] = y_pred\ndf_subm.to_csv(\"submission.csv\")\ndf_subm","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:16:50.293224Z","iopub.execute_input":"2024-05-27T06:16:50.294093Z","iopub.status.idle":"2024-05-27T06:16:50.318419Z","shell.execute_reply.started":"2024-05-27T06:16:50.294058Z","shell.execute_reply":"2024-05-27T06:16:50.317696Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"+α (Importance)","metadata":{}},{"cell_type":"markdown","source":"**XGBoost重要度**","metadata":{}},{"cell_type":"code","source":"feature_importances_xgb = pd.DataFrame([model.feature_importances_ for model in fitted_models_xgb],columns = df_test.columns).mean()\nplt.figure(figsize=(12, 6))\nfeature_importances_xgb.plot(kind='bar', color='red')\nplt.title('Feature Importances for XGBoost')\nplt.xlabel('Features')\nplt.ylabel('Importance')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:16:50.319387Z","iopub.execute_input":"2024-05-27T06:16:50.319657Z","iopub.status.idle":"2024-05-27T06:16:51.177035Z","shell.execute_reply.started":"2024-05-27T06:16:50.319633Z","shell.execute_reply":"2024-05-27T06:16:51.176099Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_importances_xgb_df = pd.DataFrame(feature_importances_xgb)\ndf_sort = pd.DataFrame(feature_importances_xgb_df.sort_values(0, ascending=False))\ndf_sort[\"feature\"] = df_sort.index\n\ndf_sort.index = range(0, len(df_sort))\ndf_sort.columns = [\"importance\", \"feature\"]\ndf_xgb_sort_30 = df_sort[:30]\ndf_xgb_sort_30\n#pd.DataFrame(df_sort[:30]).to_csv(\"XGBoost_importance_30.csv\", index = 'false')","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:16:51.178204Z","iopub.execute_input":"2024-05-27T06:16:51.178508Z","iopub.status.idle":"2024-05-27T06:16:51.195196Z","shell.execute_reply.started":"2024-05-27T06:16:51.178482Z","shell.execute_reply":"2024-05-27T06:16:51.194233Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**LightGBM重要度**","metadata":{}},{"cell_type":"code","source":"feature_importances_lgb = pd.DataFrame([model.feature_importances_ for model in fitted_models_lgb],columns = df_test.columns).mean()\nplt.figure(figsize=(12, 6))\nfeature_importances_lgb.plot(kind='bar', color='red')\nplt.title('Feature Importances for LightGBM')\nplt.xlabel('Features')\nplt.ylabel('Importance')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:16:51.196345Z","iopub.execute_input":"2024-05-27T06:16:51.196621Z","iopub.status.idle":"2024-05-27T06:16:52.037663Z","shell.execute_reply.started":"2024-05-27T06:16:51.196597Z","shell.execute_reply":"2024-05-27T06:16:52.036662Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_importances_lgb_df = pd.DataFrame(feature_importances_lgb)\ndf_sort = pd.DataFrame(feature_importances_lgb_df.sort_values(0, ascending=False))\ndf_sort[\"feature\"] = df_sort.index\n\ndf_sort.index = range(0, len(df_sort))\ndf_sort.columns = [\"importance\", \"feature\"]\ndf_lgb_sort_30 = df_sort[:30]\ndf_lgb_sort_30\n#pd.DataFrame(df_sort[:30]).to_csv(\"LightGBM_importance_30.csv\", index = 'false')","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:16:52.039084Z","iopub.execute_input":"2024-05-27T06:16:52.039710Z","iopub.status.idle":"2024-05-27T06:16:52.059065Z","shell.execute_reply.started":"2024-05-27T06:16:52.039675Z","shell.execute_reply":"2024-05-27T06:16:52.058164Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**CatBoost重要度**","metadata":{}},{"cell_type":"code","source":"feature_importances_cat = pd.DataFrame([model.feature_importances_ for model in fitted_models_cat],columns = df_test.columns).mean()\nplt.figure(figsize=(12, 6))\nfeature_importances_cat.plot(kind='bar', color='red')\nplt.title('Feature Importances for CatBoost')\nplt.xlabel('Features')\nplt.ylabel('Importance')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:16:52.065395Z","iopub.execute_input":"2024-05-27T06:16:52.065694Z","iopub.status.idle":"2024-05-27T06:16:52.921670Z","shell.execute_reply.started":"2024-05-27T06:16:52.065669Z","shell.execute_reply":"2024-05-27T06:16:52.920774Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_importances_cat_df = pd.DataFrame(feature_importances_cat)\ndf_sort = pd.DataFrame(feature_importances_cat_df.sort_values(0, ascending=False))\ndf_sort[\"feature\"] = df_sort.index\n\ndf_sort.index = range(0, len(df_sort))\ndf_sort.columns = [\"importance\", \"feature\"]\ndf_cat_sort_30 = df_sort[:30]\ndf_cat_sort_30\n#pd.DataFrame(df_sort[:30]).to_csv(\"CatBoost_importance_30.csv\", index = 'false')","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:16:52.922931Z","iopub.execute_input":"2024-05-27T06:16:52.923254Z","iopub.status.idle":"2024-05-27T06:16:52.939115Z","shell.execute_reply.started":"2024-05-27T06:16:52.923225Z","shell.execute_reply":"2024-05-27T06:16:52.938197Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_cat_sort_30\ndf_lgb_sort_30\ndf_xgb_sort_30","metadata":{"execution":{"iopub.status.busy":"2024-05-27T06:16:52.940140Z","iopub.execute_input":"2024-05-27T06:16:52.940438Z","iopub.status.idle":"2024-05-27T06:16:52.956060Z","shell.execute_reply.started":"2024-05-27T06:16:52.940414Z","shell.execute_reply":"2024-05-27T06:16:52.955239Z"},"trusted":true},"execution_count":null,"outputs":[]}]}