{"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":"none","dataSources":[{"sourceId":50160,"databundleVersionId":7921029,"sourceType":"competition"},{"sourceId":7584174,"sourceType":"datasetVersion","datasetId":4414761},{"sourceId":7912587,"sourceType":"datasetVersion","datasetId":4648691}],"dockerImageVersionId":30646,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import os\nimport gc\nfrom glob import glob\nfrom pathlib import Path\nfrom datetime import datetime\n\nimport numpy as np\nimport pandas as pd\nimport polars as pl\n\nimport matplotlib.pyplot as plt\nimport seaborn as sns\n\nfrom sklearn.model_selection import TimeSeriesSplit, GroupKFold, StratifiedGroupKFold\nfrom sklearn.base import BaseEstimator, RegressorMixin\n\nimport joblib\n\nimport lightgbm as lgb\n\nimport warnings\nwarnings.simplefilter(action='ignore', category=FutureWarning)\n\n\n\n\n\n\nclass Pipeline:\n    @staticmethod\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\n        return df\n    \n    @staticmethod\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())\n                \n        df = df.drop(\"date_decision\", \"MONTH\")\n\n        return df\n    \n    @staticmethod\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\n                if isnull > 0.95:\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\n                if (freq == 1) | (freq > 200):\n                    df = df.drop(col)\n\n        return df\n    \n    \n    \nclass Aggregator:\n    @staticmethod\n    def num_expr(df):\n        cols = [col for col in df.columns if col[-1] in (\"P\", \"A\")]\n\n        expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n\n        return expr_max\n\n    @staticmethod\n    def date_expr(df):\n        cols = [col for col in df.columns if col[-1] in (\"D\",)]\n\n        expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n\n        return expr_max\n\n    @staticmethod\n    def str_expr(df):\n        cols = [col for col in df.columns if col[-1] in (\"M\",)]\n        \n        expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n\n        return expr_max\n\n    @staticmethod\n    def other_expr(df):\n        cols = [col for col in df.columns if col[-1] in (\"T\", \"L\")]\n        \n        expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n\n        return expr_max\n    \n    @staticmethod\n    def count_expr(df):\n        cols = [col for col in df.columns if \"num_group\" in col]\n\n        expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n\n        return expr_max\n\n    @staticmethod\n    def get_exprs(df):\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\n        return exprs\n\n    \n    \n    \ndef read_file(path, depth=None):\n    df = pl.read_parquet(path)\n    df = df.pipe(Pipeline.set_table_dtypes)\n    \n    if depth in [1, 2]:\n        df = df.group_by(\"case_id\").agg(Aggregator.get_exprs(df))\n    \n    return df\n\ndef read_files(regex_path, depth=None):\n    chunks = []\n    for path in glob(str(regex_path)):\n        chunks.append(pl.read_parquet(path).pipe(Pipeline.set_table_dtypes))\n        \n    df = pl.concat(chunks, how=\"vertical_relaxed\")\n    if depth in [1, 2]:\n        df = df.group_by(\"case_id\").agg(Aggregator.get_exprs(df))\n    \n    return df\n\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        \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        \n    df_base = df_base.pipe(Pipeline.handle_dates)\n    \n    return df_base\n\n\n\ndef to_pandas(df_data, cat_cols=None):\n    df_data = df_data.to_pandas()\n    \n    if cat_cols is None:\n        cat_cols = list(df_data.select_dtypes(\"object\").columns)\n    \n    df_data[cat_cols] = df_data[cat_cols].astype(\"category\")\n    \n    return df_data, cat_cols\n","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2024-04-14T05:46:55.847599Z","iopub.execute_input":"2024-04-14T05:46:55.848872Z","iopub.status.idle":"2024-04-14T05:47:01.103477Z","shell.execute_reply.started":"2024-04-14T05:46:55.84881Z","shell.execute_reply":"2024-04-14T05:47:01.102121Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Configuration","metadata":{}},{"cell_type":"code","source":"ROOT            = Path(\"/kaggle/input/home-credit-credit-risk-model-stability\")\nTRAIN_DIR       = ROOT / \"parquet_files\" / \"train\"\nTEST_DIR        = ROOT / \"parquet_files\" / \"test\"","metadata":{"execution":{"iopub.status.busy":"2024-04-14T05:47:46.267097Z","iopub.execute_input":"2024-04-14T05:47:46.267571Z","iopub.status.idle":"2024-04-14T05:47:46.274596Z","shell.execute_reply.started":"2024-04-14T05:47:46.267534Z","shell.execute_reply":"2024-04-14T05:47:46.273179Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Train Files Read & Feature Engineering","metadata":{}},{"cell_type":"code","source":"data_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_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    ]\n}","metadata":{"execution":{"iopub.status.busy":"2024-04-14T05:47:48.48203Z","iopub.execute_input":"2024-04-14T05:47:48.482509Z","iopub.status.idle":"2024-04-14T05:48:30.472919Z","shell.execute_reply.started":"2024-04-14T05:47:48.482474Z","shell.execute_reply":"2024-04-14T05:48:30.471995Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 研究 label 数据","metadata":{}},{"cell_type":"code","source":"data = read_file(TRAIN_DIR / \"train_base.parquet\").to_pandas()\nprint(data.shape)\ndata.head(10)      #date_decision:贷款批准做出决定的日期","metadata":{"execution":{"iopub.status.busy":"2024-04-14T05:51:12.86355Z","iopub.execute_input":"2024-04-14T05:51:12.864059Z","iopub.status.idle":"2024-04-14T05:51:13.465892Z","shell.execute_reply.started":"2024-04-14T05:51:12.864023Z","shell.execute_reply":"2024-04-14T05:51:13.464729Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# case_id 没有重复行\ndata[data.loc[:,['case_id']].duplicated()]","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 研究静态数据 static_cb_0","metadata":{}},{"cell_type":"code","source":"#--------------------------------------------------------\n#                      翻译指标字典\n#--------------------------------------------------------\nfanyi = pd.read_csv('/kaggle/input/home-credit-translate-kpi/feature_definitions_translated.csv')\nven = fanyi['Variable'].tolist()\nvzh = fanyi['翻译'].tolist()\nfanyi_dict = dict(zip(ven,vzh))\n#--------------------------------------------------------\n#                      读取train_static_cb_0\n#--------------------------------------------------------\ndata = read_file(TRAIN_DIR / \"train_static_cb_0.parquet\").to_pandas()\nprint(data.shape)\n# data = data.head(10)\n\n\n\n#--------------------------------------------------------\n#              翻译train_static_cb_0的指标,缺失率\n#--------------------------------------------------------\nkpi_en = data.columns\nkpi_zh = data.columns.map(fanyi_dict)\nkpi_null_percent = round(data.isnull().sum()/data.shape[0]*100,2)\ntrain_static_cb_0_fanyi = pd.DataFrame({\n    'kpi_en'           :kpi_en\n    ,'kpi_zh'          :kpi_zh\n    ,'kpi_null_percent':kpi_null_percent\n}).set_index('kpi_en')\ntrain_static_cb_0_fanyi","metadata":{"execution":{"iopub.status.busy":"2024-04-14T06:06:13.191021Z","iopub.execute_input":"2024-04-14T06:06:13.193004Z","iopub.status.idle":"2024-04-14T06:06:17.841102Z","shell.execute_reply.started":"2024-04-14T06:06:13.192938Z","shell.execute_reply":"2024-04-14T06:06:17.839963Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import missingno as msno\nmsno.matrix(data)","metadata":{"execution":{"iopub.status.busy":"2024-04-14T06:14:02.331892Z","iopub.execute_input":"2024-04-14T06:14:02.33243Z","iopub.status.idle":"2024-04-14T06:14:28.592366Z","shell.execute_reply.started":"2024-04-14T06:14:02.332391Z","shell.execute_reply":"2024-04-14T06:14:28.591052Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#--------------------------------------------------------\n#                     数据清洗：生成新的出生日期列\n#  'birthdate_574D', 'dateofbirth_337D', 'dateofbirth_342D' 三列合并\n#--------------------------------------------------------\nbirthdate = data[['birthdate_574D', 'dateofbirth_337D', 'dateofbirth_342D']]\n\n# 相同日期是否可以对应的上\n#count = data[['birthdate_574D', 'dateofbirth_337D', 'dateofbirth_342D']].isnull().sum(axis = 1)\n#birthdate[count <1]  出生日期如果都有，可以对应上\n\n# 填充，构造出生日期列\nbirthdate = birthdate['birthdate_574D'].fillna(birthdate['dateofbirth_337D']).fillna(birthdate['dateofbirth_342D'])\nbirthdate\n\n# print(round(birthdate.isnull().sum()/birthdate.shape[0]*100,2)) \n# 填补后缺失率为 2.7 ","metadata":{"execution":{"iopub.status.busy":"2024-04-09T04:29:04.408721Z","iopub.execute_input":"2024-04-09T04:29:04.409407Z","iopub.status.idle":"2024-04-09T04:29:04.425539Z","shell.execute_reply.started":"2024-04-09T04:29:04.409359Z","shell.execute_reply":"2024-04-09T04:29:04.424015Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 研究 train_static_0_0、train_static_0_1\n1. train_static_0_0 和 train_static_0_1 列名相同`data_train_static_0_1.columns == data_train_static_0_0.columns`\n","metadata":{}},{"cell_type":"code","source":"#--------------------------------------------------------\n#                      读取train_static_0_0\n#--------------------------------------------------------\ndata = read_file(TRAIN_DIR / \"train_static_0_0.parquet\").to_pandas()\nprint(data.shape)\n\n# data = data.head(10)\ndata.head(10)\n\n\n\n#--------------------------------------------------------\n#                   翻译指标,缺失率\n#--------------------------------------------------------\nkpi_en = data.columns\nkpi_zh = data.columns.map(fanyi_dict)\nkpi_null_percent = round(data.isnull().sum()/data.shape[0]*100,2)\ntrain_static_0_0fanyi = pd.DataFrame({\n    'kpi_en'           :kpi_en\n    ,'kpi_zh'          :kpi_zh\n    ,'kpi_null_percent':kpi_null_percent\n}).set_index('kpi_en')\ntrain_static_0_0fanyi\n\ndata_train_static_0_0 = data\n\n\n#--------------------------------------------------------\n#                      读取train_static_0_1\n#--------------------------------------------------------\ndata = read_file(TRAIN_DIR / \"train_static_0_1.parquet\").to_pandas()\nprint(data.shape)\n\n# data = data.head(10)\ndata.head(10)\n\n\n#--------------------------------------------------------\n#                   翻译指标,缺失率\n#--------------------------------------------------------\nkpi_en = data.columns\nkpi_zh = data.columns.map(fanyi_dict)\nkpi_null_percent = round(data.isnull().sum()/data.shape[0]*100,2)\ntrain_static_0_1fanyi = pd.DataFrame({\n    'kpi_en'           :kpi_en\n    ,'kpi_zh'          :kpi_zh\n    ,'kpi_null_percent':kpi_null_percent\n}).set_index('kpi_en')\ntrain_static_0_1fanyi\ndata_train_static_0_1 = data\n\nresult = pd.concat([data_train_static_0_0,data_train_static_0_1],join='inner') #此时index可能无效\nmsno.matrix(result)\n","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data = pd.concat([train_static_cb_0_fanyi\n           , train_static_0_0fanyi\n           , train_static_0_1fanyi],keys=[ 'train_static_cb_0','train_static_0_0','train_static_0_1' ])\ndata = data.reset_index()\ndata = data.rename(columns = {'level_0': '表名'})\ndata","metadata":{"execution":{"iopub.status.busy":"2024-04-10T01:53:05.007287Z","iopub.execute_input":"2024-04-10T01:53:05.007672Z","iopub.status.idle":"2024-04-10T01:53:05.014751Z","shell.execute_reply.started":"2024-04-10T01:53:05.007647Z","shell.execute_reply":"2024-04-10T01:53:05.013909Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data.columns","metadata":{"execution":{"iopub.status.busy":"2024-04-09T04:12:19.251271Z","iopub.execute_input":"2024-04-09T04:12:19.252582Z","iopub.status.idle":"2024-04-09T04:12:19.26223Z","shell.execute_reply.started":"2024-04-09T04:12:19.252529Z","shell.execute_reply":"2024-04-09T04:12:19.260861Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"birthdate['birthdate_574D'].fillna(birthdate['dateofbirth_337D'])","metadata":{"execution":{"iopub.status.busy":"2024-04-09T04:08:58.736757Z","iopub.execute_input":"2024-04-09T04:08:58.737748Z","iopub.status.idle":"2024-04-09T04:08:58.765148Z","shell.execute_reply.started":"2024-04-09T04:08:58.737699Z","shell.execute_reply":"2024-04-09T04:08:58.763529Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"pd.read_csv('/kaggle/input/home-credit-translate-kpi/feature_definitions_translated.csv')","metadata":{"execution":{"iopub.status.busy":"2024-03-22T09:02:44.810533Z","iopub.execute_input":"2024-03-22T09:02:44.810944Z","iopub.status.idle":"2024-03-22T09:02:44.84249Z","shell.execute_reply.started":"2024-03-22T09:02:44.810914Z","shell.execute_reply":"2024-03-22T09:02:44.841618Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data = read_file(TRAIN_DIR / \"train_static_0_0.parquet\").to_pandas()\n#print(data.shape)\n#print(data.head(10))\n\n\nlieming = data.columns\nfeature_definitions_data = pd.read_csv('/kaggle/input/home-credit-translate-kpi/feature_definitions_translated.csv')\nfeature_definitions = feature_definitions_data.iloc[:,1].tolist()\n\n\nlie = []\na = []\nb = []\nfor i in lieming:\n    lie.append(i)\n    a_ = i in feature_definitions\n    a.append(a_)\n    \n    if a_:\n        b.extend(feature_definitions_data.loc[feature_definitions_data['Variable'] == i,'翻译'].tolist())\n    else:\n        b.append('')\n\nziduanhanyi = pd.DataFrame({'lie':lie\n              ,'a':a\n              ,'b':b})\n    \nziduanhanyi.columns = ['字段名称', '是否提供解释', '解释']    \nziduanhanyi","metadata":{"execution":{"iopub.status.busy":"2024-03-22T09:05:35.033475Z","iopub.execute_input":"2024-03-22T09:05:35.033923Z","iopub.status.idle":"2024-03-22T09:05:38.223516Z","shell.execute_reply.started":"2024-03-22T09:05:35.033888Z","shell.execute_reply":"2024-03-22T09:05:38.222534Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"ziduanhanyi['解释']","metadata":{"execution":{"iopub.status.busy":"2024-03-22T09:04:29.017197Z","iopub.execute_input":"2024-03-22T09:04:29.017586Z","iopub.status.idle":"2024-03-22T09:04:29.025017Z","shell.execute_reply.started":"2024-03-22T09:04:29.017552Z","shell.execute_reply":"2024-03-22T09:04:29.024242Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data = read_file(TRAIN_DIR / \"train_static_0_1.parquet\").to_pandas()\nprint(data.shape)\nprint(data.head(10))\n\n\nlieming = data.columns\nfeature_definitions_data = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/feature_definitions.csv')\nfeature_definitions = feature_definitions_data.iloc[:,0].tolist()\n\n\nlie = []\na = []\nb = []\nfor i in lieming:\n    lie.append(i)\n    a_ = i in feature_definitions\n    a.append(a_)\n    \n    if a_:\n        b.extend(feature_definitions_data.loc[feature_definitions_data['Variable'] == i,'Description'].tolist())\n    else:\n        b.append('')\n\nziduanhanyi = pd.DataFrame({'lie':lie\n              ,'a':a\n              ,'b':b})\n    \nziduanhanyi.columns = ['字段名称', '是否提供解释', '解释']    \nziduanhanyi","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data = read_file(TRAIN_DIR / \"train_static_cb_0.parquet\").to_pandas()\nprint(data.shape)\nprint(data.head(10))\n\n\nlieming = data.columns\nfeature_definitions_data = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/feature_definitions.csv')\nfeature_definitions = feature_definitions_data.iloc[:,0].tolist()\n\n\nlie = []\na = []\nb = []\nfor i in lieming:\n    lie.append(i)\n    a_ = i in feature_definitions\n    a.append(a_)\n    \n    if a_:\n        b.extend(feature_definitions_data.loc[feature_definitions_data['Variable'] == i,'Description'].tolist())\n    else:\n        b.append('')\n\nziduanhanyi = pd.DataFrame({'lie':lie\n              ,'a':a\n              ,'b':b})\n    \nziduanhanyi.columns = ['字段名称', '是否提供解释', '解释']    \nziduanhanyi","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 研究depth1 数据","metadata":{}},{"cell_type":"code","source":"path_ = \"train_person_1.parquet\"\ndata = read_file(TRAIN_DIR / path_).to_pandas()\nprint(data.shape)\nprint(data.head(10))\n\n\nlieming = data.columns\nfeature_definitions_data = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/feature_definitions.csv')\nfeature_definitions = feature_definitions_data.iloc[:,0].tolist()\n\n\nlie = []\na = []\nb = []\nfor i in lieming:\n    lie.append(i)\n    a_ = i in feature_definitions\n    a.append(a_)\n    \n    if a_:\n        b.extend(feature_definitions_data.loc[feature_definitions_data['Variable'] == i,'Description'].tolist())\n    else:\n        b.append('')\n\nziduanhanyi = pd.DataFrame({'lie':lie\n              ,'a':a\n              ,'b':b})\n    \nziduanhanyi.columns = ['字段名称', '是否提供解释', '解释']    \nziduanhanyi.to_csv('/kaggle/working/'+ path_ +'.csv')\nziduanhanyi","metadata":{"execution":{"iopub.status.busy":"2024-03-18T07:43:27.87833Z","iopub.execute_input":"2024-03-18T07:43:27.878833Z","iopub.status.idle":"2024-03-18T07:43:32.253233Z","shell.execute_reply.started":"2024-03-18T07:43:27.878797Z","shell.execute_reply":"2024-03-18T07:43:32.252035Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data[data['case_id'] == 147982]","metadata":{"execution":{"iopub.status.busy":"2024-03-18T07:47:17.785899Z","iopub.execute_input":"2024-03-18T07:47:17.786344Z","iopub.status.idle":"2024-03-18T07:47:17.819531Z","shell.execute_reply.started":"2024-03-18T07:47:17.786314Z","shell.execute_reply":"2024-03-18T07:47:17.81832Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data.case_id.value_counts()","metadata":{"execution":{"iopub.status.busy":"2024-03-18T07:45:27.649329Z","iopub.execute_input":"2024-03-18T07:45:27.649772Z","iopub.status.idle":"2024-03-18T07:45:27.963687Z","shell.execute_reply.started":"2024-03-18T07:45:27.649737Z","shell.execute_reply":"2024-03-18T07:45:27.962461Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"path_ = \"train_applprev_1_0.parquet\"\ndata = read_file(TRAIN_DIR / path_).to_pandas()\nprint(data.shape)\nprint(data.head(10))\n\n\nlieming = data.columns\nfeature_definitions_data = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/feature_definitions.csv')\nfeature_definitions = feature_definitions_data.iloc[:,0].tolist()\n\n\nlie = []\na = []\nb = []\nfor i in lieming:\n    lie.append(i)\n    a_ = i in feature_definitions\n    a.append(a_)\n    \n    if a_:\n        b.extend(feature_definitions_data.loc[feature_definitions_data['Variable'] == i,'Description'].tolist())\n    else:\n        b.append('')\n\nziduanhanyi = pd.DataFrame({'lie':lie\n              ,'a':a\n              ,'b':b})\n    \nziduanhanyi.columns = ['字段名称', '是否提供解释', '解释']    \nziduanhanyi.to_csv('/kaggle/working/'+ path_ +'.csv')\nziduanhanyi","metadata":{"execution":{"iopub.status.busy":"2024-03-18T07:37:11.315912Z","iopub.execute_input":"2024-03-18T07:37:11.316373Z","iopub.status.idle":"2024-03-18T07:37:20.73135Z","shell.execute_reply.started":"2024-03-18T07:37:11.31633Z","shell.execute_reply":"2024-03-18T07:37:20.729909Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"path_ = \"train_applprev_1_1.parquet\"\ndata = read_file(TRAIN_DIR / path_).to_pandas()\nprint(data.shape)\nprint(data.head(10))\n\n\nlieming = data.columns\nfeature_definitions_data = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/feature_definitions.csv')\nfeature_definitions = feature_definitions_data.iloc[:,0].tolist()\n\n\nlie = []\na = []\nb = []\nfor i in lieming:\n    lie.append(i)\n    a_ = i in feature_definitions\n    a.append(a_)\n    \n    if a_:\n        b.extend(feature_definitions_data.loc[feature_definitions_data['Variable'] == i,'Description'].tolist())\n    else:\n        b.append('')\n\nziduanhanyi = pd.DataFrame({'lie':lie\n              ,'a':a\n              ,'b':b})\n    \nziduanhanyi.columns = ['字段名称', '是否提供解释', '解释']    \nziduanhanyi.to_csv('/kaggle/working/'+ path_ +'.csv')\nziduanhanyi","metadata":{"execution":{"iopub.status.busy":"2024-03-18T07:36:46.328304Z","iopub.execute_input":"2024-03-18T07:36:46.329544Z","iopub.status.idle":"2024-03-18T07:36:53.031335Z","shell.execute_reply.started":"2024-03-18T07:36:46.329501Z","shell.execute_reply":"2024-03-18T07:36:53.030037Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data.case_id.unique().tolist()[:10]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data[data['case_id'] == 2]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data[data['case_id'] == 3]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data[data['case_id'] == 4]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data[data['case_id'] == 5]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 研究depth2","metadata":{}},{"cell_type":"code","source":"path_ = \"train_applprev_2.parquet\"\ndata = read_file(TRAIN_DIR / path_).to_pandas()\nprint(data.shape)\nprint(data.head(10))\n\n\nlieming = data.columns\nfeature_definitions_data = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/feature_definitions.csv')\nfeature_definitions = feature_definitions_data.iloc[:,0].tolist()\n\n\nlie = []\na = []\nb = []\nfor i in lieming:\n    lie.append(i)\n    a_ = i in feature_definitions\n    a.append(a_)\n    \n    if a_:\n        b.extend(feature_definitions_data.loc[feature_definitions_data['Variable'] == i,'Description'].tolist())\n    else:\n        b.append('')\n\nziduanhanyi = pd.DataFrame({'lie':lie\n              ,'a':a\n              ,'b':b})\n    \nziduanhanyi.columns = ['字段名称', '是否提供解释', '解释']    \nziduanhanyi.to_csv('/kaggle/working/'+ path_ +'.csv')\nziduanhanyi","metadata":{"execution":{"iopub.status.busy":"2024-03-18T07:38:58.703718Z","iopub.execute_input":"2024-03-18T07:38:58.704319Z","iopub.status.idle":"2024-03-18T07:39:01.92767Z","shell.execute_reply.started":"2024-03-18T07:38:58.704279Z","shell.execute_reply":"2024-03-18T07:39:01.926462Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data[data['case_id'] == 2]","metadata":{"execution":{"iopub.status.busy":"2024-03-18T07:40:52.267059Z","iopub.execute_input":"2024-03-18T07:40:52.267553Z","iopub.status.idle":"2024-03-18T07:40:52.316314Z","shell.execute_reply.started":"2024-03-18T07:40:52.267518Z","shell.execute_reply":"2024-03-18T07:40:52.314902Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 我的找变量意义","metadata":{}},{"cell_type":"code","source":"def read_file(path, depth=None):\n    df = pl.read_parquet(path)\n    df = df.pipe(Pipeline.set_table_dtypes)\n    \n#     if depth in [1, 2]:\n#         df = df.group_by(\"case_id\").agg(Aggregator.get_exprs(df))\n    \n    return df\n\n\ntrain_credit_bureau_a_2_0 = read_file(TRAIN_DIR / \"train_credit_bureau_a_2_0.parquet\", 1)\nlieming = train_credit_bureau_a_2_0.columns\n\nfeature_definitions_data = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/feature_definitions.csv')\nfeature_definitions = feature_definitions_data.iloc[:,0].tolist()\n\n\nlie = []\na = []\nb = []\nfor i in lieming:\n    lie.append(i)\n    a_ = i in feature_definitions\n    a.append(a_)\n    \n    if a_:\n        b.extend(feature_definitions_data.loc[feature_definitions_data['Variable'] == i,'Description'].tolist())\n    else:\n        b.append('')\n\nziduanhanyi = pd.DataFrame({'lie':lie\n              ,'a':a\n              ,'b':b})\n    \nziduanhanyi.columns = ['字段名称', '是否提供解释', '解释']    \nziduanhanyi.to_csv('ziduanhanyi.csv')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 研究数据","metadata":{}},{"cell_type":"code","source":"def read_file(path, depth=None):\n    df = pl.read_parquet(path)\n    df = df.pipe(Pipeline.set_table_dtypes)\n    \n#     if depth in [1, 2]:\n#         df = df.group_by(\"case_id\").agg(Aggregator.get_exprs(df))\n    \n    return df\n\n\ntrain_credit_bureau_a_2_0 = read_file(TRAIN_DIR / \"train_credit_bureau_a_2_0.parquet\", 1)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_credit_bureau_a_2_0.head(10)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data = train_credit_bureau_a_2_0","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data.shape","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data.to_pandas().collater_typofvalofguarant_407M.value_counts()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data.to_pandas().collater_typofvalofguarant_298M.value_counts()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data = train_credit_bureau_a_2_0.to_pandas()\n","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data.case_id.unique()[1:5]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"case_id_405 = data.loc[data['case_id'] == 405,:]\ncolumns_dayu2 = [i for i in case_id_405.columns if len(case_id_405[i].unique())>1]\ndata = case_id_405[columns_dayu2]\n\n\ndata","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"columns = []\nfor i in case_id_405.columns:\n    print(case_id_405[i].unique())","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_credit_bureau_a_2_0['num_group1'].value_counts()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_credit_bureau_a_2_0['num_group2'].value_counts()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data = lieminga","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitions","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_credit_bureau_a_2_3","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_train = feature_eng(**data_store)\n\nprint(\"train data shape:\\t\", df_train.shape)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Test Files Read & Feature Engineering","metadata":{}},{"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_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    ]\n}","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_test = feature_eng(**data_store)\n\nprint(\"test data shape:\\t\", df_test.shape)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_train = df_train.pipe(Pipeline.filter_cols)\ndf_test = df_test.select([col for col in df_train.columns if col != \"target\"])\n\nprint(\"train data shape:\\t\", df_train.shape)\nprint(\"test data shape:\\t\", df_test.shape)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_train, cat_cols = to_pandas(df_train)\ndf_test, cat_cols = to_pandas(df_test, cat_cols)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"del data_store\n\ngc.collect()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### EDA","metadata":{}},{"cell_type":"code","source":"print(\"Train is duplicated:\\t\", df_train[\"case_id\"].duplicated().any())\nprint(\"Train Week Range:\\t\", (df_train[\"WEEK_NUM\"].min(), df_train[\"WEEK_NUM\"].max()))\n\nprint()\n\nprint(\"Test is duplicated:\\t\", df_test[\"case_id\"].duplicated().any())\nprint(\"Test Week Range:\\t\", (df_test[\"WEEK_NUM\"].min(), df_test[\"WEEK_NUM\"].max()))","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sns.lineplot(\n    data=df_train,\n    x=\"WEEK_NUM\",\n    y=\"target\",\n)\nplt.show()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Training","metadata":{}},{"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        y_preds = [estimator.predict_proba(X) for estimator in self.estimators]\n        return np.mean(y_preds, axis=0)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.metrics import roc_auc_score\n\nX = df_train.drop(columns=[\"target\", \"case_id\",\"WEEK_NUM\"])\ny = df_train[\"target\"]\nweeks = df_train[\"WEEK_NUM\"]\n\ncv = StratifiedGroupKFold(n_splits=5, shuffle=False)\n\nparams = {\n    \"boosting_type\": \"gbdt\",\n    \"objective\": \"binary\",\n    \"metric\": \"auc\",\n    \"max_depth\": 10,\n    \"learning_rate\": 0.05,\n    \"max_bin\": 255,\n    \"n_estimators\": 1200,\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\": \"gpu\",  # Uncomment if you want to use GPU for training\n}\n\nfitted_models = []\ncv_scores = []  \n\nfor idx_train, idx_valid in cv.split(X, y, groups=weeks):\n    X_train, y_train = X.iloc[idx_train], y.iloc[idx_train]\n    X_valid, y_valid = X.iloc[idx_valid], y.iloc[idx_valid]\n\n    print(\"Valid week range: \", (weeks.iloc[idx_valid].min(), weeks.iloc[idx_valid].max()))\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(50), lgb.early_stopping(50)]\n    )\n\n    fitted_models.append(model)\n\n    y_pred_valid = model.predict_proba(X_valid)[:, 1]\n    auc_score = roc_auc_score(y_valid, y_pred_valid)\n    cv_scores.append(auc_score)\n\nmodel = VotingModel(fitted_models)\nprint(\"CV AUC scores: \", cv_scores)\nprint(\"Average CV AUC score: \", sum(cv_scores) / len(cv_scores))","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_test = df_test.drop(columns=[\"WEEK_NUM\"])\nX_test = X_test.set_index(\"case_id\")\n\nlgb_pred = pd.Series(model.predict_proba(X_test)[:, 1], index=X_test.index)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Submission","metadata":{}},{"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\"] = lgb_pred","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(\"Check null: \", df_subm[\"score\"].isnull().any())","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_subm.head()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_subm.to_csv(\"submission.csv\")","metadata":{"trusted":true},"execution_count":null,"outputs":[]}]}