{"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":"# Home Credit Kaggle Notebook\n\nЭтот ноутбук построен на основе https://www.kaggle.com/code/jetakow/home-credit-2024-starter-notebook. Однако расширен датасет, добавлены новые признаки, а также произведен подбор параметров модели с помощью optuna.\n\nСтруктура ноутбука осталась прежней:\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 \n\ndataPath = \"/kaggle/input/home-credit-credit-risk-model-stability/\"","metadata":{"execution":{"iopub.status.busy":"2024-03-16T21:04:39.457056Z","iopub.execute_input":"2024-03-16T21:04:39.457596Z","iopub.status.idle":"2024-03-16T21:04:42.570240Z","shell.execute_reply.started":"2024-03-16T21:04:39.457547Z","shell.execute_reply":"2024-03-16T21:04:42.569347Z"},"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        if col[-1] in (\"D\"): # to process dates\n            df = df.with_columns(pl.col(col).cast(pl.Date).alias(col))\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":"2024-03-16T21:04:42.571727Z","iopub.execute_input":"2024-03-16T21:04:42.572015Z","iopub.status.idle":"2024-03-16T21:04:42.580139Z","shell.execute_reply.started":"2024-03-16T21:04:42.571988Z","shell.execute_reply":"2024-03-16T21:04:42.579061Z"},"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":"2024-03-16T21:04:42.581762Z","iopub.execute_input":"2024-03-16T21:04:42.582584Z","iopub.status.idle":"2024-03-16T21:05:01.388263Z","shell.execute_reply.started":"2024-03-16T21:04:42.582532Z","shell.execute_reply":"2024-03-16T21:05:01.387149Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"## NEW DATA TO CONSIDER\ntrain_credit_bureau_b_1 = pl.read_csv(dataPath + \"csv_files/train/train_credit_bureau_b_1.csv\").pipe(set_table_dtypes) \ntrain_debitcard_1 = pl.read_csv(dataPath + \"csv_files/train/train_debitcard_1.csv\").pipe(set_table_dtypes)\ntrain_deposit_1 = pl.read_csv(dataPath + \"csv_files/train/train_deposit_1.csv\").pipe(set_table_dtypes)","metadata":{"execution":{"iopub.status.busy":"2024-03-16T21:05:01.390383Z","iopub.execute_input":"2024-03-16T21:05:01.390697Z","iopub.status.idle":"2024-03-16T21:05:01.758850Z","shell.execute_reply.started":"2024-03-16T21:05:01.390668Z","shell.execute_reply":"2024-03-16T21:05:01.757788Z"},"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":"2024-03-16T21:05:01.760442Z","iopub.execute_input":"2024-03-16T21:05:01.760776Z","iopub.status.idle":"2024-03-16T21:05:01.852926Z","shell.execute_reply.started":"2024-03-16T21:05:01.760745Z","shell.execute_reply":"2024-03-16T21:05:01.852038Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"## NEW DATA TO CONSIDER\ntest_credit_bureau_b_1 = pl.read_csv(dataPath + \"csv_files/test/test_credit_bureau_b_1.csv\").pipe(set_table_dtypes) \ntest_debitcard_1 = pl.read_csv(dataPath + \"csv_files/test/test_debitcard_1.csv\").pipe(set_table_dtypes)\ntest_deposit_1 = pl.read_csv(dataPath + \"csv_files/test/test_deposit_1.csv\").pipe(set_table_dtypes)","metadata":{"execution":{"iopub.status.busy":"2024-03-16T21:05:01.853984Z","iopub.execute_input":"2024-03-16T21:05:01.854299Z","iopub.status.idle":"2024-03-16T21:05:01.873379Z","shell.execute_reply.started":"2024-03-16T21:05:01.854266Z","shell.execute_reply":"2024-03-16T21:05:01.872489Z"},"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":"feature_definitions = pd.read_csv(dataPath + \"feature_definitions.csv\") # to find feature descriptions\npd.set_option('max_colwidth', 600)\nfeature_definitions.iloc[0:15]","metadata":{"execution":{"iopub.status.busy":"2024-03-16T21:05:01.875929Z","iopub.execute_input":"2024-03-16T21:05:01.876776Z","iopub.status.idle":"2024-03-16T21:05:01.914160Z","shell.execute_reply.started":"2024-03-16T21:05:01.876743Z","shell.execute_reply":"2024-03-16T21:05:01.913318Z"},"trusted":true},"execution_count":null,"outputs":[]},{"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## NEW_FEATURES\ntrain_credit_bureau_b_1_feats = train_credit_bureau_b_1.group_by(\"case_id\").agg(\n    pl.col(\"credlmt_1052A\").max().alias(\"credlmt_1052A_max\"),\n    pl.col(\"debtpastduevalue_732A\").max().alias(\"debtpastduevalue_732A_max\"),\n    pl.col(\"numberofinstls_810L\").mean().alias(\"numberofinstls_810L_mean\"),\n    (pl.col(\"totalamount_503A\") < 40000).max().alias(\"totalamount_503A_under40\"),\n    pl.col(\"residualamount_3940956A\").max().alias(\"residualamount_3940956A_max\")\n)\n\ntrain_debitcard_1_feats = train_debitcard_1.group_by(\"case_id\").agg(\n    pl.col(\"last180dayaveragebalance_704A\").mean().alias(\"last180dayaveragebalance_704A_mean\"),\n    pl.col(\"last30dayturnover_651A\").max().alias(\"maxlast30dayturnover_651A\"),\n)\n\ntrain_deposit_1_feats = train_deposit_1.group_by(\"case_id\").agg(\n    (pl.col(\"amount_416A\") < 1000).max().alias(\"amount_416A_under1000\"),\n    (pl.col(\"amount_416A\") > 20).max().alias(\"amount_416A_above20\"),\n)\n\n# We will process in this notebook not only A-type and M-type columns, but L type also\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).join(\n    train_credit_bureau_b_1_feats, how=\"left\", on=\"case_id\" #NEW\n).join(\n    train_debitcard_1_feats, how=\"left\", on=\"case_id\" #NEW\n).join(\n    train_deposit_1_feats, how=\"left\", on=\"case_id\" #NEW\n)","metadata":{"execution":{"iopub.status.busy":"2024-03-16T21:05:01.915960Z","iopub.execute_input":"2024-03-16T21:05:01.916658Z","iopub.status.idle":"2024-03-16T21:05:03.630256Z","shell.execute_reply.started":"2024-03-16T21:05:01.916625Z","shell.execute_reply":"2024-03-16T21:05:03.629447Z"},"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\n## NEW_FEATURES\ntest_credit_bureau_b_1_feats = test_credit_bureau_b_1.group_by(\"case_id\").agg(\n    pl.col(\"credlmt_1052A\").max().alias(\"credlmt_1052A_max\"),\n    pl.col(\"debtpastduevalue_732A\").max().alias(\"debtpastduevalue_732A_max\"),\n    pl.col(\"numberofinstls_810L\").mean().alias(\"numberofinstls_810L_mean\"),\n    (pl.col(\"totalamount_503A\") < 40000).max().alias(\"totalamount_503A_under40\"),\n    pl.col(\"residualamount_3940956A\").max().alias(\"residualamount_3940956A_max\")\n)\n\ntest_debitcard_1_feats = test_debitcard_1.group_by(\"case_id\").agg(\n    pl.col(\"last180dayaveragebalance_704A\").mean().alias(\"last180dayaveragebalance_704A_mean\"),\n    pl.col(\"last30dayturnover_651A\").max().alias(\"maxlast30dayturnover_651A\"),\n)\n\ntest_deposit_1_feats = test_deposit_1.group_by(\"case_id\").agg(\n    (pl.col(\"amount_416A\") < 1000).max().alias(\"amount_416A_under1000\"),\n    (pl.col(\"amount_416A\") > 20).max().alias(\"amount_416A_above20\"),\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).join(\n    test_credit_bureau_b_1_feats, how=\"left\", on=\"case_id\" #NEW\n).join(\n    test_debitcard_1_feats, how=\"left\", on=\"case_id\" #NEW\n).join(\n    test_deposit_1_feats, how=\"left\", on=\"case_id\" #NEW\n)","metadata":{"execution":{"iopub.status.busy":"2024-03-16T21:05:03.631099Z","iopub.execute_input":"2024-03-16T21:05:03.631438Z","iopub.status.idle":"2024-03-16T21:05:03.652258Z","shell.execute_reply.started":"2024-03-16T21:05:03.631407Z","shell.execute_reply":"2024-03-16T21:05:03.651382Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Рассмотрим категориальные колонки. Нас могут заинтересовать сочетания категорий","metadata":{}},{"cell_type":"code","source":"categorical_columns = [column for column in data.columns if column.endswith(\"M\")]\ncategorical_columns","metadata":{"execution":{"iopub.status.busy":"2024-03-16T21:05:03.655159Z","iopub.execute_input":"2024-03-16T21:05:03.655760Z","iopub.status.idle":"2024-03-16T21:05:03.663155Z","shell.execute_reply.started":"2024-03-16T21:05:03.655728Z","shell.execute_reply":"2024-03-16T21:05:03.662415Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# some more features to add \n\ndef add_new_features(df: pl.DataFrame) -> pl.DataFrame:\n    df = df.with_columns((df[\"education_1103M\"] + df[\"maritalst_385M\"]).alias(\"education+maritalst\")) #to consider cases of being divorced and with different level of education on one depth\n    df = df.with_columns((df[\"currdebt_22A\"] / df[\"maxannuity_159A\"]).alias(\"currdebt_22A_per_maxannuity_159A\"))\n    df = df.with_columns((df[\"person_housetype\"] + df[\"education_88M\"]).alias(\"person_housetype+education_88M\")) #to find rare cases when housetype is not common for this type of education\n    df = df.with_columns((df[\"currdebt_22A\"] / df['maininc_215A']).alias(\"debttoincome\"))\n    return df","metadata":{"execution":{"iopub.status.busy":"2024-03-16T21:05:03.665088Z","iopub.execute_input":"2024-03-16T21:05:03.665404Z","iopub.status.idle":"2024-03-16T21:05:03.672794Z","shell.execute_reply.started":"2024-03-16T21:05:03.665376Z","shell.execute_reply":"2024-03-16T21:05:03.672069Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data = add_new_features(data)\ndata_submission = add_new_features(data_submission)","metadata":{"execution":{"iopub.status.busy":"2024-03-16T21:05:03.674123Z","iopub.execute_input":"2024-03-16T21:05:03.674443Z","iopub.status.idle":"2024-03-16T21:05:03.830934Z","shell.execute_reply.started":"2024-03-16T21:05:03.674416Z","shell.execute_reply":"2024-03-16T21:05:03.829796Z"},"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":"2024-03-16T21:05:03.832313Z","iopub.execute_input":"2024-03-16T21:05:03.832668Z","iopub.status.idle":"2024-03-16T21:05:11.401653Z","shell.execute_reply.started":"2024-03-16T21:05:03.832638Z","shell.execute_reply":"2024-03-16T21:05:11.400723Z"},"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":"2024-03-16T21:05:11.403049Z","iopub.execute_input":"2024-03-16T21:05:11.403524Z","iopub.status.idle":"2024-03-16T21:05:11.409125Z","shell.execute_reply.started":"2024-03-16T21:05:11.403483Z","shell.execute_reply":"2024-03-16T21:05:11.408200Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Training LightGBM\n\nMinimal example of LightGBM + Optuna training is shown below.","metadata":{}},{"cell_type":"code","source":"import optuna\n\nlgb_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\": trial.suggest_int(\"max_depth\", 3, 7),\n        \"num_leaves\": trial.suggest_int(\"num_leaves\", 10, 50),\n        \"learning_rate\": trial.suggest_float(\"learning_rate\", 1e-4, 1e-1, log=True),\n        \"feature_fraction\": trial.suggest_float(\"feature_fraction\", 0.1, 0.9),\n        \"bagging_fraction\": trial.suggest_float(\"bagging_fraction\", 0.1, 0.9),\n        \"bagging_freq\":trial.suggest_int(\"bagging_freq\", 1, 10),\n        \"n_estimators\": trial.suggest_int(\"n_estimators\", 500, 2000),\n        \"verbose\": -1,\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    score = gbm.predict(X_valid, num_iteration=gbm.best_iteration)\n\n\n    return roc_auc_score(base_valid[\"target\"], score)\n\n\nstudy = optuna.create_study(direction=\"maximize\")\nstudy.optimize(objective, n_trials=40) # here we can put even more trials, but it takes so long\n\nprint(\"Number of finished trials: {}\".format(len(study.trials)))\n\nprint(\"Best trial:\")\ntrial = study.best_trial\n\nprint(\"  Value: {}\".format(trial.value))\n\nprint(\"  Params: \")\nfor key, value in trial.params.items():\n    print(\"    {}: {}\".format(key, value))","metadata":{"execution":{"iopub.status.busy":"2024-03-16T21:05:11.410413Z","iopub.execute_input":"2024-03-16T21:05:11.411092Z","iopub.status.idle":"2024-03-16T21:07:16.310998Z","shell.execute_reply.started":"2024-03-16T21:05:11.411058Z","shell.execute_reply":"2024-03-16T21:07:16.309635Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"trial = study.best_trial\nbest_params = trial.params\n\nparams = {\n    \"boosting_type\": \"gbdt\",\n    \"objective\": \"binary\",\n    \"metric\": \"auc\",\n    \"verbose\": -1,\n    **best_params\n}\n\nlgb_train = lgb.Dataset(X_train, label=y_train)\nlgb_valid = lgb.Dataset(X_valid, label=y_valid, reference=lgb_train)\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":"2024-03-16T21:07:16.311942Z","iopub.status.idle":"2024-03-16T21:07:16.313089Z","shell.execute_reply.started":"2024-03-16T21:07:16.312876Z","shell.execute_reply":"2024-03-16T21:07:16.312900Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Evaluation with AUC and then comparison with the stability metric is shown below.","metadata":{}},{"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":"2024-03-16T21:07:16.313990Z","iopub.status.idle":"2024-03-16T21:07:16.314937Z","shell.execute_reply.started":"2024-03-16T21:07:16.314719Z","shell.execute_reply":"2024-03-16T21:07:16.314743Z"},"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":"2024-03-16T21:07:16.316413Z","iopub.status.idle":"2024-03-16T21:07:16.316952Z","shell.execute_reply.started":"2024-03-16T21:07:16.316771Z","shell.execute_reply":"2024-03-16T21:07:16.316790Z"},"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    new_dtype = pd.CategoricalDtype(categories=train_categories, ordered=True)\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":"2024-03-16T21:07:16.317836Z","iopub.status.idle":"2024-03-16T21:07:16.318637Z","shell.execute_reply.started":"2024-03-16T21:07:16.318436Z","shell.execute_reply":"2024-03-16T21:07:16.318458Z"},"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":"2024-03-16T21:07:16.319866Z","iopub.status.idle":"2024-03-16T21:07:16.320588Z","shell.execute_reply.started":"2024-03-16T21:07:16.320381Z","shell.execute_reply":"2024-03-16T21:07:16.320401Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<img src=\"https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcQGoEMYAc5RRCfCvwLCcN0BnWpMEO4Q3gxNZc7iJsFOuA&s\" alt=\"Image\" width=\"700\"/>","metadata":{}}]}