{"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":"gpu","dataSources":[{"sourceId":50160,"databundleVersionId":7921029,"sourceType":"competition"}],"dockerImageVersionId":30646,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"raw","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19"}},{"cell_type":"code","source":"# 라이브러리 import \n\nimport polars as pl\nimport numpy as np\nimport pandas as pd\nimport lightgbm as lgb\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.metrics import roc_auc_score \n\ndataPath = \"/kaggle/input/home-credit-credit-risk-model-stability/\"","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:15:38.550467Z","iopub.execute_input":"2024-03-20T01:15:38.551435Z","iopub.status.idle":"2024-03-20T01:15:38.558361Z","shell.execute_reply.started":"2024-03-20T01:15:38.551389Z","shell.execute_reply":"2024-03-20T01:15:38.557379Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 데이터 타입 변경","metadata":{}},{"cell_type":"code","source":"# 데이터 타입 변경\n# 1. 숫자형 변수\n\ndef set_table_dtypes(df: pl.DataFrame) -> pl.DataFrame:\n    # 모든 열의 컬럼을 가져와서 \"P\"나 \"A\"인 경우에만 실행\n    for col in df.columns:\n        # 해당 열의 데이터 타입을 Float64로 변경\n        # P - DPD 연체일수 \n        # A - 금액\n        if col[-1] in (\"P\", \"A\"):\n            df = df.with_columns(pl.col(col).cast(pl.Float64).alias(col))\n\n    return df","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:15:39.255384Z","iopub.execute_input":"2024-03-20T01:15:39.256467Z","iopub.status.idle":"2024-03-20T01:15:39.261797Z","shell.execute_reply.started":"2024-03-20T01:15:39.256432Z","shell.execute_reply":"2024-03-20T01:15:39.260778Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# 2. 문자형 변수\n# 문자열 타입이 object나 string일 경우에 변환\n# 문자열 - 범주형으로 변환하고 Unknown 범주를 추가하 문자열 데이터 처리\n\ndef convert_strings(df: pd.DataFrame) -> pd.DataFrame:\n    for col in df.columns:  \n        if df[col].dtype.name in ['object', 'string']:\n            df[col] = df[col].astype(\"string\").astype('category')\n            current_categories = df[col].cat.categories\n            new_categories = current_categories.to_list() + [\"Unknown\"]\n            new_dtype = pd.CategoricalDtype(categories=new_categories, ordered=True)\n            df[col] = df[col].astype(new_dtype)\n    return df","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:15:39.619269Z","iopub.execute_input":"2024-03-20T01:15:39.619660Z","iopub.status.idle":"2024-03-20T01:15:39.626457Z","shell.execute_reply.started":"2024-03-20T01:15:39.619630Z","shell.execute_reply":"2024-03-20T01:15:39.625372Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 데이터 import","metadata":{}},{"cell_type":"code","source":"# base 테이블 \ntrain_basetable = pl.read_csv(dataPath + \"csv_files/train/train_base.csv\")\n\n# 최소한의 데이터만 사용하여 import\n\ntrain_static = pl.concat(\n    [\n        pl.read_csv(dataPath + \"csv_files/train/train_static_0_0.csv\").pipe(set_table_dtypes),\n        pl.read_csv(dataPath + \"csv_files/train/train_static_0_1.csv\").pipe(set_table_dtypes),\n    ],\n    how=\"vertical_relaxed\",\n)\ntrain_static_cb = pl.read_csv(dataPath + \"csv_files/train/train_static_cb_0.csv\").pipe(set_table_dtypes)\ntrain_person_1 = pl.read_csv(dataPath + \"csv_files/train/train_person_1.csv\").pipe(set_table_dtypes) \ntrain_credit_bureau_b_2 = pl.read_csv(dataPath + \"csv_files/train/train_credit_bureau_b_2.csv\").pipe(set_table_dtypes)","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:15:40.373623Z","iopub.execute_input":"2024-03-20T01:15:40.374479Z","iopub.status.idle":"2024-03-20T01:15:53.100202Z","shell.execute_reply.started":"2024-03-20T01:15:40.374447Z","shell.execute_reply":"2024-03-20T01:15:53.099370Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#test데이터도 동일하게 처리\ntest_basetable = pl.read_csv(dataPath + \"csv_files/test/test_base.csv\")\ntest_static = pl.concat(\n    [\n        pl.read_csv(dataPath + \"csv_files/test/test_static_0_0.csv\").pipe(set_table_dtypes),\n        pl.read_csv(dataPath + \"csv_files/test/test_static_0_1.csv\").pipe(set_table_dtypes),\n        pl.read_csv(dataPath + \"csv_files/test/test_static_0_2.csv\").pipe(set_table_dtypes),\n    ],\n    how=\"vertical_relaxed\",\n)\ntest_static_cb = pl.read_csv(dataPath + \"csv_files/test/test_static_cb_0.csv\").pipe(set_table_dtypes)\ntest_person_1 = pl.read_csv(dataPath + \"csv_files/test/test_person_1.csv\").pipe(set_table_dtypes) \ntest_credit_bureau_b_2 = pl.read_csv(dataPath + \"csv_files/test/test_credit_bureau_b_2.csv\").pipe(set_table_dtypes)","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:15:53.101695Z","iopub.execute_input":"2024-03-20T01:15:53.101992Z","iopub.status.idle":"2024-03-20T01:15:53.173833Z","shell.execute_reply.started":"2024-03-20T01:15:53.101968Z","shell.execute_reply":"2024-03-20T01:15:53.173009Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#person1 데이터 정제\n\ntrain_person_1_feats_1 = train_person_1.group_by(\"case_id\").agg(\n    pl.col(\"mainoccupationinc_384A\").max().alias(\"mainoccupationinc_384A_max\"),\n    (pl.col(\"incometype_1044T\") == \"SELFEMPLOYED\").max().alias(\"mainoccupationinc_384A_any_selfemployed\")\n)\n\n\n# num_group1 = 0 은 대출 실행한 사람이므로 해당 값만 남김\n# num_group1 해당 값은 모델구축시 제거\n# housetype_905L을 person_housetype로 이름 변경!\n\ntrain_person_1_feats_2 = train_person_1.select([\"case_id\", \"num_group1\", \"housetype_905L\"]).filter(\n    pl.col(\"num_group1\") == 0\n).drop(\"num_group1\").rename({\"housetype_905L\": \"person_housetype\"})\n\n","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:15:53.174971Z","iopub.execute_input":"2024-03-20T01:15:53.175256Z","iopub.status.idle":"2024-03-20T01:15:53.981517Z","shell.execute_reply.started":"2024-03-20T01:15:53.175232Z","shell.execute_reply":"2024-03-20T01:15:53.980671Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# train_credit_bureau_b_2 데이터 정제 \n# pmts_pmtsoverdue_635A 열의 최댓값을 구하고 pmts_pmtsoverdue_635A_max로 별칭을 지정\n# pmts_dpdvalue_108P 열이 31보다 큰 경우의 최댓값을 구하고 pmts_dpdvalue_108P_over31로 별칭을 지정\n\ntrain_credit_bureau_b_2_feats = train_credit_bureau_b_2.group_by(\"case_id\").agg(\n    pl.col(\"pmts_pmtsoverdue_635A\").max().alias(\"pmts_pmtsoverdue_635A_max\"),\n    (pl.col(\"pmts_dpdvalue_108P\") > 31).max().alias(\"pmts_dpdvalue_108P_over31\")\n)","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:15:53.983740Z","iopub.execute_input":"2024-03-20T01:15:53.984400Z","iopub.status.idle":"2024-03-20T01:15:54.040082Z","shell.execute_reply.started":"2024-03-20T01:15:53.984364Z","shell.execute_reply":"2024-03-20T01:15:54.039293Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# A타입과 M타입만 선택 금액과 범주만\nselected_static_cols = []\nfor col in train_static.columns:\n    if col[-1] in (\"A\", \"M\"):\n        selected_static_cols.append(col)\nprint(selected_static_cols)","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:15:54.041223Z","iopub.execute_input":"2024-03-20T01:15:54.041614Z","iopub.status.idle":"2024-03-20T01:15:54.047777Z","shell.execute_reply.started":"2024-03-20T01:15:54.041569Z","shell.execute_reply":"2024-03-20T01:15:54.046880Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"selected_static_cb_cols = []\nfor col in train_static_cb.columns:\n    if col[-1] in (\"A\", \"M\"):\n        selected_static_cb_cols.append(col)\nprint(selected_static_cb_cols)","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:18:09.941019Z","iopub.execute_input":"2024-03-20T01:18:09.941398Z","iopub.status.idle":"2024-03-20T01:18:09.946911Z","shell.execute_reply.started":"2024-03-20T01:18:09.941371Z","shell.execute_reply":"2024-03-20T01:18:09.945944Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# 모든 데이터 Join\ndata = train_basetable.join(\n    train_static.select([\"case_id\"]+selected_static_cols), how=\"left\", on=\"case_id\"\n).join(\n    train_static_cb.select([\"case_id\"]+selected_static_cb_cols), how=\"left\", on=\"case_id\"\n).join(\n    train_person_1_feats_1, how=\"left\", on=\"case_id\"\n).join(\n    train_person_1_feats_2, how=\"left\", on=\"case_id\"\n).join(\n    train_credit_bureau_b_2_feats, how=\"left\", on=\"case_id\"\n)","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:18:12.380710Z","iopub.execute_input":"2024-03-20T01:18:12.381300Z","iopub.status.idle":"2024-03-20T01:18:13.525211Z","shell.execute_reply.started":"2024-03-20T01:18:12.381271Z","shell.execute_reply":"2024-03-20T01:18:13.524362Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test_person_1_feats_1 = train_person_1.group_by(\"case_id\").agg(\n    pl.col(\"mainoccupationinc_384A\").max().alias(\"mainoccupationinc_384A_max\"),\n    (pl.col(\"incometype_1044T\") == \"SELFEMPLOYED\").max().alias(\"mainoccupationinc_384A_any_selfemployed\")\n)\n\ntest_person_1_feats_2 = train_person_1.select([\"case_id\", \"num_group1\", \"housetype_905L\"]).filter(\n    pl.col(\"num_group1\") == 0\n).drop(\"num_group1\").rename({\"housetype_905L\": \"person_housetype\"})\n\ntest_credit_bureau_b_2_feats = train_credit_bureau_b_2.group_by(\"case_id\").agg(\n    pl.col(\"pmts_pmtsoverdue_635A\").max().alias(\"pmts_pmtsoverdue_635A_max\"),\n    (pl.col(\"pmts_dpdvalue_108P\") > 31).max().alias(\"pmts_dpdvalue_108P_over31\")\n)\n\ndata_submission = test_basetable.join(\n    test_static.select([\"case_id\"]+selected_static_cols), how=\"left\", on=\"case_id\"\n).join(\n    test_static_cb.select([\"case_id\"]+selected_static_cb_cols), how=\"left\", on=\"case_id\"\n).join(\n    test_person_1_feats_1, how=\"left\", on=\"case_id\"\n).join(\n    test_person_1_feats_2, how=\"left\", on=\"case_id\"\n).join(\n    test_credit_bureau_b_2_feats, how=\"left\", on=\"case_id\"\n)","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:18:17.618109Z","iopub.execute_input":"2024-03-20T01:18:17.618480Z","iopub.status.idle":"2024-03-20T01:18:18.616762Z","shell.execute_reply.started":"2024-03-20T01:18:17.618450Z","shell.execute_reply":"2024-03-20T01:18:18.615967Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# case id의 고유한 열에 대해서만 추출\n# shuffle을 통해 섞음\n# train_test_split을 활용하여 훈련세트 60 / 검증 20 / 테스트 20으로 나눔\n\ncase_ids = data[\"case_id\"].unique().shuffle(seed=1)\ncase_ids_train, case_ids_test = train_test_split(case_ids, train_size=0.6, random_state=1)\ncase_ids_valid, case_ids_test = train_test_split(case_ids_test, train_size=0.5, random_state=1)\n","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:19:48.730390Z","iopub.execute_input":"2024-03-20T01:19:48.730807Z","iopub.status.idle":"2024-03-20T01:19:48.891098Z","shell.execute_reply.started":"2024-03-20T01:19:48.730782Z","shell.execute_reply":"2024-03-20T01:19:48.890192Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# data 데이터프레임의 열 이름을 순회하면서, 열 이름의 마지막 문자가 대문자이고 마지막 문자를 제외한 나머지 부분이 모두 소문자인 경우에만 해당 열을 cols_pred 리스트에 추가합니다.\n\ncols_pred = []\nfor col in data.columns:\n    if col[-1].isupper() and col[:-1].islower():\n        cols_pred.append(col)\n\nprint(cols_pred)","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:21:21.849030Z","iopub.execute_input":"2024-03-20T01:21:21.849403Z","iopub.status.idle":"2024-03-20T01:21:21.856292Z","shell.execute_reply.started":"2024-03-20T01:21:21.849374Z","shell.execute_reply":"2024-03-20T01:21:21.855242Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# polars데이터 프레임에서 pandas로 재변환\n\ndef from_polars_to_pandas(case_ids: pl.DataFrame) -> pl.DataFrame:\n    return (\n        data.filter(pl.col(\"case_id\").is_in(case_ids))[[\"case_id\", \"WEEK_NUM\", \"target\"]].to_pandas(),\n        data.filter(pl.col(\"case_id\").is_in(case_ids))[cols_pred].to_pandas(),\n        data.filter(pl.col(\"case_id\").is_in(case_ids))[\"target\"].to_pandas()\n    )\n\nbase_train, X_train, y_train = from_polars_to_pandas(case_ids_train)\nbase_valid, X_valid, y_valid = from_polars_to_pandas(case_ids_valid)\nbase_test, X_test, y_test = from_polars_to_pandas(case_ids_test)\n\nfor df in [X_train, X_valid, X_test]:\n    df = convert_strings(df)","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:21:22.756108Z","iopub.execute_input":"2024-03-20T01:21:22.756480Z","iopub.status.idle":"2024-03-20T01:21:30.464712Z","shell.execute_reply.started":"2024-03-20T01:21:22.756448Z","shell.execute_reply":"2024-03-20T01:21:30.463824Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import lightgbm as lgb\n\nlgb_train = lgb.Dataset(X_train, label=y_train)\nlgb_valid = lgb.Dataset(X_valid, label=y_valid, reference=lgb_train)\n\nparams = {\n    \"boosting_type\": \"gbdt\",\n    \"objective\": \"binary\",\n    \"metric\": \"auc\",\n    \"max_depth\": 3,\n    \"num_leaves\": 31,\n    \"learning_rate\": 0.01,\n    \"feature_fraction\": 0.9,\n    \"bagging_fraction\": 0.8,\n    \"bagging_freq\": 5,\n    \"n_estimators\": 1500,\n    \"verbose\": -1,\n}\n\nlgb_model = lgb.train(\n    params,\n    lgb_train,\n    valid_sets=lgb_valid,  \n    callbacks=[lgb.log_evaluation(50), lgb.early_stopping(100)]\n)","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:21:52.058621Z","iopub.execute_input":"2024-03-20T01:21:52.059495Z","iopub.status.idle":"2024-03-20T01:24:18.611376Z","shell.execute_reply.started":"2024-03-20T01:21:52.059462Z","shell.execute_reply":"2024-03-20T01:24:18.610517Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import xgboost as xgb\n\nxgb_model = xgb.XGBClassifier(\n    #device=\"cuda\",\n    objective='binary:logistic',\n    tree_method=\"hist\",\n    enable_categorical=True,\n    eval_metric='auc',\n    #gamma=0.7,\n    #reg_alpha=0.7,\n    max_depth=10,\n    n_estimators=1200,\n    random_state=42,\n)\n\n# Training the model on the training data\nxgb_model.fit(\n    X_train, y_train,\n    eval_set=[(X_valid, y_valid)],\n    early_stopping_rounds=100,\n    verbose=True,\n)","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:24:18.615840Z","iopub.execute_input":"2024-03-20T01:24:18.617722Z","iopub.status.idle":"2024-03-20T01:35:07.685585Z","shell.execute_reply.started":"2024-03-20T01:24:18.617691Z","shell.execute_reply":"2024-03-20T01:35:07.684585Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import pandas as pd\nfrom catboost import CatBoostClassifier\ncat_features = [col for col in X_train.columns if X_train[col].dtype.name == 'category' or X_train[col].dtype.name == 'object']\n\nfor col in cat_features:\n    X_train[col] = X_train[col].cat.add_categories('Missing').fillna('Missing')\n    X_valid[col] = X_valid[col].cat.add_categories('Missing').fillna('Missing')\n\ncat_model = CatBoostClassifier(\n    iterations=1000,                 \n    depth=10,                        \n    learning_rate=0.1,               \n    eval_metric='AUC',               \n    random_seed=42,                  \n    bootstrap_type='Bayesian',       \n    bagging_temperature=1,           \n    od_type='Iter',                  \n    od_wait=50                      \n)\n\n\ncat_model.fit(\n    X_train, y_train,\n    eval_set=(X_valid, y_valid),\n    cat_features=cat_features,  \n    use_best_model=True,\n    verbose=True\n)","metadata":{"execution":{"iopub.status.busy":"2024-03-20T01:35:07.686906Z","iopub.execute_input":"2024-03-20T01:35:07.687265Z","iopub.status.idle":"2024-03-20T02:10:51.519007Z","shell.execute_reply.started":"2024-03-20T01:35:07.687233Z","shell.execute_reply":"2024-03-20T02:10:51.518022Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_submission = data_submission[cols_pred].to_pandas()\nX_submission = convert_strings(X_submission)\nX_submission_processed = X_submission.copy()\ncategorical_cols = X_train.select_dtypes(include=['category']).columns\n\nfor col in categorical_cols:\n    train_categories = set(X_train[col].cat.categories)\n    submission_categories = set(X_submission[col].cat.categories)\n    new_categories = submission_categories - train_categories\n    X_submission.loc[X_submission[col].isin(new_categories), col] = \"Unknown\"\n    new_dtype = pd.CategoricalDtype(categories=train_categories, ordered=True)\n    X_train[col] = X_train[col].astype(new_dtype)\n    X_submission[col] = X_submission[col].astype(new_dtype)\n\nfor col in cat_features:\n    X_submission_processed[col] = X_submission_processed[col].cat.add_categories('Missing').fillna('Missing')\n\nlgb_pred = lgb_model.predict(X_submission)\nxgb_pred = xgb_model.predict(X_submission)\ncat_pred = cat_model.predict(X_submission_processed)\n\ny_submission_pred = np.mean([lgb_pred, xgb_pred, cat_pred], axis=0)\nsubmission = pd.DataFrame({\n    \"case_id\": data_submission[\"case_id\"].to_numpy(),\n    \"score\": y_submission_pred\n}).set_index('case_id')\nsubmission.to_csv(\"./submission.csv\")\nsubmission","metadata":{"execution":{"iopub.status.busy":"2024-03-20T02:10:51.521375Z","iopub.execute_input":"2024-03-20T02:10:51.522016Z","iopub.status.idle":"2024-03-20T02:10:51.659599Z","shell.execute_reply.started":"2024-03-20T02:10:51.521981Z","shell.execute_reply":"2024-03-20T02:10:51.658674Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"y_submission_pred","metadata":{"execution":{"iopub.status.busy":"2024-03-20T02:10:51.660853Z","iopub.execute_input":"2024-03-20T02:10:51.661174Z","iopub.status.idle":"2024-03-20T02:10:51.669323Z","shell.execute_reply.started":"2024-03-20T02:10:51.661136Z","shell.execute_reply":"2024-03-20T02:10:51.668353Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}