{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.12","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":7602123,"sourceType":"competition"}],"dockerImageVersionId":30635,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"ФИО: Жухлистов Станислав Борисович\n\nОригинальный ноутбук: https://www.kaggle.com/code/jetakow/home-credit-2024-starter-notebook\n\nУлучшения: добавлены 3 новых признака и подбор гиперпараметров модели при помощи optuna\n\nВ целом результат получился хуже, чем у лучших публичных ноутбуков, поэтому кажется, что lightgbm в этой задаче не работает, надо будет пробовать другие модели. Ещё в этом ноутбуке используется мало признаков, можно попробовать использовать больше.","metadata":{}},{"cell_type":"markdown","source":"## Load the data","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19"}},{"cell_type":"code","source":"import 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":"2024-02-22T17:14:34.217397Z","iopub.execute_input":"2024-02-22T17:14:34.217700Z","iopub.status.idle":"2024-02-22T17:14:40.367664Z","shell.execute_reply.started":"2024-02-22T17:14:34.217674Z","shell.execute_reply":"2024-02-22T17:14:40.366805Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def set_table_dtypes(df: pl.DataFrame) -> pl.DataFrame:\n    # implement here all desired dtypes for tables\n    # the following is just an example\n    for col in df.columns:\n        # last letter of column name will help you determine the type\n        if col[-1] in (\"P\", \"A\"):\n            df = df.with_columns(pl.col(col).cast(pl.Float64).alias(col))\n\n    return df\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":"2024-02-22T17:14:40.369599Z","iopub.execute_input":"2024-02-22T17:14:40.370312Z","iopub.status.idle":"2024-02-22T17:14:40.378104Z","shell.execute_reply.started":"2024-02-22T17:14:40.370274Z","shell.execute_reply":"2024-02-22T17:14:40.377185Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_basetable = pl.read_csv(dataPath + \"csv_files/train/train_base.csv\")\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":"2024-02-22T17:14:40.379249Z","iopub.execute_input":"2024-02-22T17:14:40.379635Z","iopub.status.idle":"2024-02-22T17:14:55.319025Z","shell.execute_reply.started":"2024-02-22T17:14:40.379600Z","shell.execute_reply":"2024-02-22T17:14:55.317999Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_credit_bureau_b_2","metadata":{"execution":{"iopub.status.busy":"2024-02-22T17:14:55.321811Z","iopub.execute_input":"2024-02-22T17:14:55.322511Z","iopub.status.idle":"2024-02-22T17:14:55.341860Z","shell.execute_reply.started":"2024-02-22T17:14:55.322450Z","shell.execute_reply":"2024-02-22T17:14:55.340817Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test_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":"2024-02-22T17:14:55.342892Z","iopub.execute_input":"2024-02-22T17:14:55.343141Z","iopub.status.idle":"2024-02-22T17:14:55.423236Z","shell.execute_reply.started":"2024-02-22T17:14:55.343120Z","shell.execute_reply":"2024-02-22T17:14:55.422496Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Feature engineering\n\nIn this part, we can see a simple example of joining tables via `case_id`. Here the loading and joining is done with polars library. Polars library is blazingly fast and has much smaller memory footprint than pandas. ","metadata":{}},{"cell_type":"code","source":"# We need to use aggregation functions in tables with depth > 1, so tables that contain num_group1 column or \n# also num_group2 column.\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# Here num_group1=0 has special meaning, it is the person who applied for the loan.\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# Here we have num_goup1 and num_group2, so we need to aggregate again.\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)\n\n# We will process in this examples only A-type and M-type columns, so we need to select them.\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)\n\nselected_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)\n\n# Join all tables together.\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":"2024-02-22T17:14:55.424235Z","iopub.execute_input":"2024-02-22T17:14:55.424484Z","iopub.status.idle":"2024-02-22T17:14:56.829898Z","shell.execute_reply.started":"2024-02-22T17:14:55.424449Z","shell.execute_reply":"2024-02-22T17:14:56.828888Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test_person_1_feats_1 = test_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 = test_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 = test_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":"2024-02-22T17:14:56.831113Z","iopub.execute_input":"2024-02-22T17:14:56.831414Z","iopub.status.idle":"2024-02-22T17:14:56.842089Z","shell.execute_reply.started":"2024-02-22T17:14:56.831389Z","shell.execute_reply":"2024-02-22T17:14:56.841348Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"case_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\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)\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":"2024-02-22T17:14:56.843091Z","iopub.execute_input":"2024-02-22T17:14:56.843346Z","iopub.status.idle":"2024-02-22T17:15:04.275168Z","shell.execute_reply.started":"2024-02-22T17:14:56.843323Z","shell.execute_reply":"2024-02-22T17:15:04.274395Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Добавим новые признаки:\n\n* Если у человека annuitynextmonth_57A = 0, то это значит, что ему не надо в следующем месяце делать платежей, что по идее должно делать его более привлекательным заёмщиком\n* Если у человека lastotherlnsexpense_631A является nan'ом, то это значит, что у него до этого не было кредитов, что делает его менее привлекательным заёмщиком\n* Введём debt_to_income = currdebt_22A / maininc_215A. Чем ниже это значение, тем лучше.","metadata":{}},{"cell_type":"code","source":"X_train[\"zero_next_annuity\"] = (X_train[\"annuitynextmonth_57A\"] == 0).astype(int)\nX_train[\"nan_last_expense\"] = (X_train[\"lastotherlnsexpense_631A\"].isna()).astype(int)\nX_train[\"debt_to_income\"] = X_train[\"currdebt_22A\"] / X_train[\"maininc_215A\"]\nX_valid[\"zero_next_annuity\"] = (X_valid[\"annuitynextmonth_57A\"] == 0).astype(int)\nX_valid[\"nan_last_expense\"] = (X_valid[\"lastotherlnsexpense_631A\"].isna()).astype(int)\nX_valid[\"debt_to_income\"] = X_valid[\"currdebt_22A\"] / X_valid[\"maininc_215A\"]\nX_test[\"zero_next_annuity\"] = (X_test[\"annuitynextmonth_57A\"] == 0).astype(int)\nX_test[\"nan_last_expense\"] = (X_test[\"lastotherlnsexpense_631A\"].isna()).astype(int)\nX_test[\"debt_to_income\"] = X_test[\"currdebt_22A\"] / X_test[\"maininc_215A\"]","metadata":{"execution":{"iopub.status.busy":"2024-02-22T17:15:04.276390Z","iopub.execute_input":"2024-02-22T17:15:04.277068Z","iopub.status.idle":"2024-02-22T17:15:04.314499Z","shell.execute_reply.started":"2024-02-22T17:15:04.277034Z","shell.execute_reply":"2024-02-22T17:15:04.313482Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train","metadata":{"execution":{"iopub.status.busy":"2024-02-22T17:15:04.317750Z","iopub.execute_input":"2024-02-22T17:15:04.318081Z","iopub.status.idle":"2024-02-22T17:15:04.621922Z","shell.execute_reply.started":"2024-02-22T17:15:04.318038Z","shell.execute_reply":"2024-02-22T17:15:04.621006Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(f\"Train: {X_train.shape}\")\nprint(f\"Valid: {X_valid.shape}\")\nprint(f\"Test: {X_test.shape}\")","metadata":{"execution":{"iopub.status.busy":"2024-02-22T17:15:04.623244Z","iopub.execute_input":"2024-02-22T17:15:04.623648Z","iopub.status.idle":"2024-02-22T17:15:04.629423Z","shell.execute_reply.started":"2024-02-22T17:15:04.623611Z","shell.execute_reply":"2024-02-22T17:15:04.628485Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train","metadata":{"execution":{"iopub.status.busy":"2024-02-22T17:15:04.630775Z","iopub.execute_input":"2024-02-22T17:15:04.631438Z","iopub.status.idle":"2024-02-22T17:15:04.907189Z","shell.execute_reply.started":"2024-02-22T17:15:04.631407Z","shell.execute_reply":"2024-02-22T17:15:04.906170Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Training LightGBM\n\nMinimal example of LightGBM training is shown below.","metadata":{}},{"cell_type":"code","source":"def gini_stability(base, w_fallingrate=88.0, w_resstd=-0.5):\n    gini_in_time = base.loc[:, [\"WEEK_NUM\", \"target\", \"score\"]]\\\n        .sort_values(\"WEEK_NUM\")\\\n        .groupby(\"WEEK_NUM\")[[\"target\", \"score\"]]\\\n        .apply(lambda x: 2*roc_auc_score(x[\"target\"], x[\"score\"])-1).tolist()\n    \n    x = np.arange(len(gini_in_time))\n    y = gini_in_time\n    a, b = np.polyfit(x, y, 1)\n    y_hat = a*x + b\n    residuals = y - y_hat\n    res_std = np.std(residuals)\n    avg_gini = np.mean(gini_in_time)\n    return avg_gini + w_fallingrate * min(0, a) + w_resstd * res_std","metadata":{"execution":{"iopub.status.busy":"2024-02-22T17:15:04.908717Z","iopub.execute_input":"2024-02-22T17:15:04.909448Z","iopub.status.idle":"2024-02-22T17:15:04.917085Z","shell.execute_reply.started":"2024-02-22T17:15:04.909411Z","shell.execute_reply":"2024-02-22T17:15:04.916116Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Улучшение оригинального ноутбука: тюнинг параметров. Сам подбор делается очень долго, поэтому в финальном решении мы его не запускаем, а просто записываем лучшие параметры и вставляем как константы.","metadata":{}},{"cell_type":"code","source":"import optuna","metadata":{"execution":{"iopub.status.busy":"2024-02-22T17:15:04.918115Z","iopub.execute_input":"2024-02-22T17:15:04.918381Z","iopub.status.idle":"2024-02-22T17:15:04.933793Z","shell.execute_reply.started":"2024-02-22T17:15:04.918347Z","shell.execute_reply":"2024-02-22T17:15:04.932921Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"lgb_train = lgb.Dataset(X_train, label=y_train)\nlgb_valid = lgb.Dataset(X_valid, label=y_valid, reference=lgb_train)\n\ndef objective(trial):\n    params = {\n        \"boosting_type\": \"gbdt\",\n        \"objective\": \"binary\",\n        \"metric\": \"auc\",\n        \"max_depth\": 3,\n        'lambda_l1': trial.suggest_float('lambda_l1', 1e-8, 10.0, log=True),\n        'lambda_l2': trial.suggest_float('lambda_l2', 1e-8, 10.0, log=True),\n        \"num_leaves\": trial.suggest_int('num_leaves', 2, 256),\n        \"learning_rate\": trial.suggest_float('lr', 1e-3, 0.1, log=True),\n        'feature_fraction': trial.suggest_float('feature_fraction', 0.4, 1.0),\n        'bagging_fraction': trial.suggest_float('bagging_fraction', 0.4, 1.0),\n        \"bagging_freq\": trial.suggest_int('bagging_freq', 1, 7),\n        \"n_estimators\": 1000,\n        \"verbose\": -1,\n        \"device\": \"gpu\"\n    }\n    gbm = lgb.train(\n        params,\n        lgb_train,\n        valid_sets=lgb_valid,\n        callbacks=[lgb.log_evaluation(999999999), lgb.early_stopping(10)]\n    )\n    y_pred = gbm.predict(X_valid, num_iteration=gbm.best_iteration)\n    base_valid[\"score\"] = y_pred\n    return gini_stability(base_valid)","metadata":{"execution":{"iopub.status.busy":"2024-02-22T17:15:04.935078Z","iopub.execute_input":"2024-02-22T17:15:04.935611Z","iopub.status.idle":"2024-02-22T17:15:04.946215Z","shell.execute_reply.started":"2024-02-22T17:15:04.935586Z","shell.execute_reply":"2024-02-22T17:15:04.945305Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"\"\"\"\nstudy = optuna.create_study(direction='maximize')\nstudy.optimize(objective, n_trials=100)\nprint('Best trial:')\nbest_trial = study.best_trial\nprint(f'  Loss: {best_trial.value:.2f}')\nprint('  Params: ')\nfor key, value in best_trial.params.items():\n    print(f'    {key}: {value}')\n\"\"\"","metadata":{"execution":{"iopub.status.busy":"2024-02-22T17:15:04.947180Z","iopub.execute_input":"2024-02-22T17:15:04.947430Z","iopub.status.idle":"2024-02-22T19:01:57.321179Z","shell.execute_reply.started":"2024-02-22T17:15:04.947408Z","shell.execute_reply":"2024-02-22T19:01:57.320233Z"},"collapsed":true,"jupyter":{"outputs_hidden":true},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"params = {\n    \"boosting_type\": \"gbdt\",\n    \"objective\": \"binary\",\n    \"metric\": \"auc\",\n    \"max_depth\": 3,\n    'lambda_l1': 2.956713008012641,\n    'lambda_l2': 1.853932491997414e-08,\n    \"num_leaves\": 16,\n    \"learning_rate\": 0.09063010807409543,\n    'feature_fraction': 0.889592160374155,\n    'bagging_fraction': 0.9347271617654018,\n    \"bagging_freq\": 3,\n    \"n_estimators\": 1000,\n    \"verbose\": -1,\n    \"device\": \"gpu\"\n}\ngbm = lgb.train(\n    params,\n    lgb_train,\n    valid_sets=lgb_valid,\n    callbacks=[lgb.log_evaluation(999999999), lgb.early_stopping(10)]\n)\nfor base, X in [(base_train, X_train), (base_valid, X_valid), (base_test, X_test)]:\n    y_pred = gbm.predict(X, num_iteration=gbm.best_iteration)\n    base[\"score\"] = y_pred\nstability_score_train = gini_stability(base_train)\nstability_score_valid = gini_stability(base_valid)\nstability_score_test = gini_stability(base_test)\n\nprint(f'The stability score on the train set is: {stability_score_train}') \nprint(f'The stability score on the valid set is: {stability_score_valid}') \nprint(f'The stability score on the test set is: {stability_score_test}')","metadata":{"execution":{"iopub.status.busy":"2024-02-22T19:07:27.764369Z","iopub.execute_input":"2024-02-22T19:07:27.764797Z","iopub.status.idle":"2024-02-22T19:09:30.091141Z","shell.execute_reply.started":"2024-02-22T19:07:27.764765Z","shell.execute_reply":"2024-02-22T19:09:30.090157Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"В оригинальном ноутбуке было:\n\nThe stability score on the train set is: 0.4976648127691175\n\nThe stability score on the valid set is: 0.4726726686264489\n\nThe stability score on the test set is: 0.4583643686935092","metadata":{}},{"cell_type":"markdown","source":"Качество получается сильно хуже, чем на лучших публичных ноутбуках, поэтому есть ощущение, что бустинг в этой задаче плох. В будущих попытках надо попробовать другие модели.\n\nЕщё мы используем только 48 оригинальных признаков и более, чем 400. В будущем можно будет попробовать использовать больше","metadata":{}},{"cell_type":"markdown","source":"## Submission\n\nScoring the submission dataset is below, we need to take care of new categories. Then we save the score as a last step. ","metadata":{}},{"cell_type":"code","source":"X_submission = data_submission[cols_pred].to_pandas()\nX_submission = convert_strings(X_submission)\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)\nX_submission[\"zero_next_annuity\"] = (X_submission[\"annuitynextmonth_57A\"] == 0).astype(int)\nX_submission[\"nan_last_expense\"] = (X_submission[\"lastotherlnsexpense_631A\"].isna()).astype(int)\nX_submission[\"debt_to_income\"] = X_submission[\"currdebt_22A\"] / X_submission[\"maininc_215A\"]\ny_submission_pred = gbm.predict(X_submission, num_iteration=gbm.best_iteration)","metadata":{"execution":{"iopub.status.busy":"2024-02-22T19:09:30.092797Z","iopub.execute_input":"2024-02-22T19:09:30.093092Z","iopub.status.idle":"2024-02-22T19:09:30.170141Z","shell.execute_reply.started":"2024-02-22T19:09:30.093067Z","shell.execute_reply":"2024-02-22T19:09:30.169320Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"submission = 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\")","metadata":{"execution":{"iopub.status.busy":"2024-02-22T19:09:30.171345Z","iopub.execute_input":"2024-02-22T19:09:30.171872Z","iopub.status.idle":"2024-02-22T19:09:30.187578Z","shell.execute_reply.started":"2024-02-22T19:09:30.171843Z","shell.execute_reply":"2024-02-22T19:09:30.186716Z"},"trusted":true},"execution_count":null,"outputs":[]}]}