{"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":"none","dataSources":[{"sourceId":50160,"databundleVersionId":7921029,"sourceType":"competition"}],"dockerImageVersionId":30635,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# Example Notebook\n\nWelcome to the example notebook for the Home Credit Kaggle competition. The goal of this competition is to determine how likely a customer is going to default on an issued loan. The main difference between the [first](https://www.kaggle.com/c/home-credit-default-risk) and this competition is that now your submission will be scored with a custom metric that will take into account how well the model performs in future. A decline in performance will be penalized. The goal is to create a model that is stable and performs well in the future.\n\nIn this notebook you will see how to:\n* Load the data\n* Join tables with Polars - a DataFrame library implemented in Rust language, designed to be blazingy fast and memory efficient.  \n* Create simple aggregation features\n* Train a LightGBM model\n* Create a submission table\n\n## 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 \nimport optuna\n\ndataPath = \"/kaggle/input/home-credit-credit-risk-model-stability/\"","metadata":{"execution":{"iopub.status.busy":"2024-03-19T02:26:09.725196Z","iopub.execute_input":"2024-03-19T02:26:09.725990Z","iopub.status.idle":"2024-03-19T02:26:12.410424Z","shell.execute_reply.started":"2024-03-19T02:26:09.725884Z","shell.execute_reply":"2024-03-19T02:26:12.408625Z"},"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\",\"L\"):\n            try:\n                df = df.with_columns(pl.col(col).cast(pl.Float64).alias(col))\n            except:\n                pass\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":"2024-03-19T02:26:12.420097Z","iopub.execute_input":"2024-03-19T02:26:12.420734Z","iopub.status.idle":"2024-03-19T02:26:12.434422Z","shell.execute_reply.started":"2024-03-19T02:26:12.420671Z","shell.execute_reply":"2024-03-19T02:26:12.432741Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_basetable = pl.read_csv(dataPath + \"csv_files/train/train_base.csv\")  # Leer un archivo CSV\ntrain_static = pl.concat(  # Concatenar DataFrames\n    [  \n        pl.read_csv(dataPath + \"csv_files/train/train_static_0_0.csv\").pipe(set_table_dtypes),  # Leer un archivo CSV y aplicar una función\n        pl.read_csv(dataPath + \"csv_files/train/train_static_0_1.csv\").pipe(set_table_dtypes),  # Leer un archivo CSV y aplicar una función\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)  # Leer un archivo CSV y aplicar una función\ntrain_person_1 = pl.read_csv(dataPath + \"csv_files/train/train_person_1.csv\").pipe(set_table_dtypes)   # Leer un archivo CSV y aplicar una función\ntrain_credit_bureau_b_1 = pl.read_csv(dataPath + \"csv_files/train/train_credit_bureau_b_1.csv\").pipe(set_table_dtypes)  # Leer un archivo CSV y aplicar una función\ntrain_credit_bureau_b_2 = pl.read_csv(dataPath + \"csv_files/train/train_credit_bureau_b_2.csv\").pipe(set_table_dtypes)   # Leer un archivo CSV y aplicar una función\ntrain_credit_debit_card_1 = pl.read_csv(dataPath + \"csv_files/train/train_debitcard_1.csv\").pipe(set_table_dtypes)   # Leer un archivo CSV y aplicar una función\ntrain_credit_deposit_1 = pl.read_csv(dataPath + \"csv_files/train/train_deposit_1.csv\").pipe(set_table_dtypes)   # Leer un archivo CSV y aplicar una función\n","metadata":{"execution":{"iopub.status.busy":"2024-03-19T02:26:12.436832Z","iopub.execute_input":"2024-03-19T02:26:12.438669Z","iopub.status.idle":"2024-03-19T02:26:31.111640Z","shell.execute_reply.started":"2024-03-19T02:26:12.438598Z","shell.execute_reply":"2024-03-19T02:26:31.110263Z"},"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_1 = pl.read_csv(dataPath + \"csv_files/test/test_credit_bureau_b_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) \ntest_credit_debit_card_1 = pl.read_csv(dataPath + \"csv_files/test/test_debitcard_1.csv\").pipe(set_table_dtypes) \ntest_credit_deposit_1 = pl.read_csv(dataPath + \"csv_files/test/test_deposit_1.csv\").pipe(set_table_dtypes) ","metadata":{"execution":{"iopub.status.busy":"2024-03-19T02:26:31.115615Z","iopub.execute_input":"2024-03-19T02:26:31.116125Z","iopub.status.idle":"2024-03-19T02:26:31.231270Z","shell.execute_reply.started":"2024-03-19T02:26:31.116088Z","shell.execute_reply":"2024-03-19T02:26:31.229671Z"},"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":"person_important = [\"registaddr_district_1083M\",\n                    \"education_927M\",\n                    \"mainoccupationinc_384A\", \n                    #\"incometype_1044T\",\n                    \"empladdr_district_926M\"]\n\ntrain_person_fin = train_person_1.select([\"case_id\", \"num_group1\"] + person_important).filter(\n    pl.col(\"num_group1\") == 0\n).drop(\"num_group1\")\n\ntest_person_fin = test_person_1.select([\"case_id\", \"num_group1\"] + person_important).filter(\n    pl.col(\"num_group1\") == 0\n).drop(\"num_group1\")","metadata":{"execution":{"iopub.status.busy":"2024-03-19T02:26:31.233218Z","iopub.execute_input":"2024-03-19T02:26:31.233626Z","iopub.status.idle":"2024-03-19T02:26:31.365660Z","shell.execute_reply.started":"2024-03-19T02:26:31.233589Z","shell.execute_reply":"2024-03-19T02:26:31.364747Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"cb_important = [\"pmtssum_45A\",\n               \"education_1103M\",\n               \"pmtaverage_3A\",\n               \"pmtaverage_4527227A\",\n               \"days30_165L\",\n               \"days360_512L\",\n               \"days180_256L\"]\ntrain_cb_fin = train_static_cb.select([\"case_id\"]+cb_important)\ntest_cb_fin = test_static_cb.select([\"case_id\"]+cb_important)","metadata":{"execution":{"iopub.status.busy":"2024-03-19T02:26:31.367348Z","iopub.execute_input":"2024-03-19T02:26:31.367972Z","iopub.status.idle":"2024-03-19T02:26:31.373907Z","shell.execute_reply.started":"2024-03-19T02:26:31.367933Z","shell.execute_reply":"2024-03-19T02:26:31.372877Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"static_important = ['price_1097A',\n 'mobilephncnt_593L',\n 'pmtnum_254L',\n 'avgdpdtolclosure24_3658938P',\n 'numrejects9m_859L',\n 'cntpmts24_3658933L',\n 'numinstunpaidmax_3546851L',\n 'eir_270L',\n 'numinstlsallpaid_934L',\n 'maxdbddpdtollast12m_3658940P',\n 'numincomingpmts_3546848L',\n 'pctinstlsallpaidlate1d_3546856L',\n 'maxdpdlast3m_392P',\n 'monthsannuity_845L',\n 'lastrejectreason_759M']\ntrain_static_fin = train_static.select([\"case_id\"]+static_important)\ntest_static_fin = test_static.select([\"case_id\"]+static_important)","metadata":{"execution":{"iopub.status.busy":"2024-03-19T02:26:31.375519Z","iopub.execute_input":"2024-03-19T02:26:31.375988Z","iopub.status.idle":"2024-03-19T02:26:31.387393Z","shell.execute_reply.started":"2024-03-19T02:26:31.375946Z","shell.execute_reply":"2024-03-19T02:26:31.386104Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_bureau_fin = 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\ntest_bureau_fin = 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\"))","metadata":{"execution":{"iopub.status.busy":"2024-03-19T02:26:31.389056Z","iopub.execute_input":"2024-03-19T02:26:31.389715Z","iopub.status.idle":"2024-03-19T02:26:31.439001Z","shell.execute_reply.started":"2024-03-19T02:26:31.389677Z","shell.execute_reply":"2024-03-19T02:26:31.437665Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data = train_basetable.join(\n    train_static_fin, how=\"left\", on=\"case_id\"\n).join(\n    train_cb_fin, how=\"left\", on=\"case_id\"\n).join(\n    train_person_fin, how=\"left\", on=\"case_id\"\n).join(\n    train_bureau_fin, how=\"left\", on=\"case_id\"\n)\ndata_submission = test_basetable.join(\n    test_static_fin, how=\"left\", on=\"case_id\"\n).join(\n    test_cb_fin, how=\"left\", on=\"case_id\"\n).join(\n    test_person_fin, how=\"left\", on=\"case_id\"\n).join(\n    test_bureau_fin, how=\"left\", on=\"case_id\"\n)","metadata":{"execution":{"iopub.status.busy":"2024-03-19T02:26:31.440438Z","iopub.execute_input":"2024-03-19T02:26:31.441188Z","iopub.status.idle":"2024-03-19T02:26:32.605523Z","shell.execute_reply.started":"2024-03-19T02:26:31.441149Z","shell.execute_reply":"2024-03-19T02:26:32.604226Z"},"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)\ncbflag = False\n\nfor df in [X_train, X_valid, X_test]:\n    df = convert_strings(df)\n    \nprint(f\"Train: {X_train.shape}\")\nprint(f\"Valid: {X_valid.shape}\")\nprint(f\"Test: {X_test.shape}\")","metadata":{"execution":{"iopub.status.busy":"2024-03-19T02:26:32.607772Z","iopub.execute_input":"2024-03-19T02:26:32.609358Z","iopub.status.idle":"2024-03-19T02:26:37.575114Z","shell.execute_reply.started":"2024-03-19T02:26:32.609301Z","shell.execute_reply":"2024-03-19T02:26:37.572857Z"},"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":"lgb_train = lgb.Dataset(X_train, label=y_train)\nlgb_valid = lgb.Dataset(X_valid, label=y_valid, reference=lgb_train)","metadata":{"execution":{"iopub.status.busy":"2024-03-19T02:26:37.577747Z","iopub.execute_input":"2024-03-19T02:26:37.578257Z","iopub.status.idle":"2024-03-19T02:26:37.586749Z","shell.execute_reply.started":"2024-03-19T02:26:37.578222Z","shell.execute_reply":"2024-03-19T02:26:37.584797Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def objective(trial):\n    params = {\n        \"objective\": \"binary\",\n        \"metric\": \"auc\",\n        \"n_estimators\": 1000,\n        \"verbosity\": -1,\n        \"max_depth\": trial.suggest_int(\"max_depth\", 3, 6),\n        \"num_leaves\": trial.suggest_int(\"num_leaves\", 15, 50),\n        \"learning_rate\": trial.suggest_float(\"learning_rate\", 0.001, 0.1),\n        \"feature_fraction\": 0.9,\n        \"bagging_fraction\": 0.8,\n        \"bagging_freq\": 5,\n    }\n\n    gbm = lgb.train(\n        params,\n        lgb_train,\n        valid_sets=lgb_valid,\n        callbacks=[lgb.log_evaluation(50), lgb.early_stopping(10)]\n    )\n    \n    predictions = gbm.predict(X_valid, num_iteration=gbm.best_iteration)\n    return roc_auc_score(base_valid[\"target\"], predictions)","metadata":{"execution":{"iopub.status.busy":"2024-03-19T02:26:37.588194Z","iopub.execute_input":"2024-03-19T02:26:37.589330Z","iopub.status.idle":"2024-03-19T02:26:37.604696Z","shell.execute_reply.started":"2024-03-19T02:26:37.589261Z","shell.execute_reply":"2024-03-19T02:26:37.602105Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"params = {\n    \"boosting_type\": \"gbdt\",\n    \"objective\": \"binary\",\n    \"metric\": \"auc\",\n    \"max_depth\": 8,\n    \"num_leaves\": 30,\n    \"learning_rate\": 0.05,\n    \"feature_fraction\": 0.9,\n    \"bagging_fraction\": 0.8,\n    \"bagging_freq\": 5,\n    \"n_estimators\": 2000,\n    \"verbose\": -1,\n}\n\ngbm = lgb.train(\n    params,\n    lgb_train,\n    valid_sets=lgb_valid,\n    callbacks=[lgb.log_evaluation(50), lgb.early_stopping(10)]\n)","metadata":{"execution":{"iopub.status.busy":"2024-03-19T02:26:37.611631Z","iopub.execute_input":"2024-03-19T02:26:37.612199Z","iopub.status.idle":"2024-03-19T02:27:39.015435Z","shell.execute_reply.started":"2024-03-19T02:26:37.612144Z","shell.execute_reply":"2024-03-19T02:27:39.013689Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"for 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\n\nprint(f'The AUC score on the train set is: {roc_auc_score(base_train[\"target\"], base_train[\"score\"])}') \nprint(f'The AUC score on the valid set is: {roc_auc_score(base_valid[\"target\"], base_valid[\"score\"])}') \nprint(f'The AUC score on the test set is: {roc_auc_score(base_test[\"target\"], base_test[\"score\"])}')  ","metadata":{"execution":{"iopub.status.busy":"2024-03-19T02:27:39.017258Z","iopub.execute_input":"2024-03-19T02:27:39.017715Z","iopub.status.idle":"2024-03-19T02:27:55.515163Z","shell.execute_reply.started":"2024-03-19T02:27:39.017647Z","shell.execute_reply":"2024-03-19T02:27:55.513501Z"},"trusted":true},"execution_count":null,"outputs":[]},{"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\n\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":"2024-03-19T02:27:55.516611Z","iopub.execute_input":"2024-03-19T02:27:55.516969Z","iopub.status.idle":"2024-03-19T02:27:56.664809Z","shell.execute_reply.started":"2024-03-19T02:27:55.516940Z","shell.execute_reply":"2024-03-19T02:27:56.663896Z"},"trusted":true},"execution_count":null,"outputs":[]},{"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)\n\ny_submission_pred = gbm.predict(X_submission, num_iteration=gbm.best_iteration)","metadata":{"execution":{"iopub.status.busy":"2024-03-19T02:27:56.666218Z","iopub.execute_input":"2024-03-19T02:27:56.666824Z","iopub.status.idle":"2024-03-19T02:27:56.728935Z","shell.execute_reply.started":"2024-03-19T02:27:56.666787Z","shell.execute_reply":"2024-03-19T02:27:56.727101Z"},"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":"2024-03-19T02:27:56.730551Z","iopub.execute_input":"2024-03-19T02:27:56.730962Z","iopub.status.idle":"2024-03-19T02:27:56.746620Z","shell.execute_reply.started":"2024-03-19T02:27:56.730910Z","shell.execute_reply":"2024-03-19T02:27:56.745167Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Best of luck, and most importantly, enjoy the process of learning and discovery! \n\n<img src=\"https://i.imgur.com/obVWIBh.png\" alt=\"Image\" width=\"700\"/>","metadata":{}}]}