{"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":"# ДЗ по САД, Арсений Чеканов\n\n## Что изменилось после перезапуска соревнования?\n\nИз тестовой выборки убрали временные колонки, чтобы нельзя было специально ухудшить модель так, чтобы итоговая метрика выросла из-за понизившегося штрафа за нестабильность (i.e. ухудшение качества на данных с более поздним WEEK_NUM).\n\n## Проделанная мной работа\n\nСразу после форка ноутбука метрики датасетов были следующие:\n\n```\nThe stability score on the train set is: 0.3916579445516054\nThe stability score on the valid set is: 0.38715930628645623\nThe stability score on the test set is: 0.3692546564985046\n```\n\nВ качестве feature engineering я решил исследовать таблицы с налоговыми данными, так как они показались мне не достаточно изученными в других ноутбуках и EDA. Увы, после их добавления качество не изменилось.\n\nТакже я добавил автоматический поиск гиперпараметров с помощью библиотеки hyperopt.\n\nСтали такими:\n\n```\nThe stability score on the train set is: 0.4976648127691175\nThe stability score on the valid set is: 0.4726726686264489\nThe stability score on the test set is: 0.4583643686935092\n```","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":"2024-03-18T20:28:29.386089Z","iopub.execute_input":"2024-03-18T20:28:29.386594Z","iopub.status.idle":"2024-03-18T20:28:32.315883Z","shell.execute_reply.started":"2024-03-18T20:28:29.386537Z","shell.execute_reply":"2024-03-18T20:28:32.314629Z"},"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\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":"2024-03-18T20:28:32.318137Z","iopub.execute_input":"2024-03-18T20:28:32.318518Z","iopub.status.idle":"2024-03-18T20:28:32.329666Z","shell.execute_reply.started":"2024-03-18T20:28:32.318485Z","shell.execute_reply":"2024-03-18T20:28:32.328271Z"},"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":"2024-03-18T20:28:32.331102Z","iopub.execute_input":"2024-03-18T20:28:32.331531Z","iopub.status.idle":"2024-03-18T20:28:52.725521Z","shell.execute_reply.started":"2024-03-18T20:28:32.331498Z","shell.execute_reply":"2024-03-18T20:28:52.724615Z"},"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":"2024-03-18T20:28:52.727923Z","iopub.execute_input":"2024-03-18T20:28:52.728436Z","iopub.status.idle":"2024-03-18T20:28:52.823258Z","shell.execute_reply.started":"2024-03-18T20:28:52.728405Z","shell.execute_reply":"2024-03-18T20:28:52.822296Z"},"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. \n\n### Налоговые данные","metadata":{}},{"cell_type":"code","source":"def agg_tax(table, col):\n    return table.select([\"case_id\", col]).group_by(\"case_id\").agg(\n        pl.col(col).max().alias(col + '_max'),\n        pl.col(col).mean().alias(col + '_mean'),\n        pl.col(col).min().alias(col + '_min'),\n    )","metadata":{"execution":{"iopub.status.busy":"2024-03-18T20:28:52.825082Z","iopub.execute_input":"2024-03-18T20:28:52.825527Z","iopub.status.idle":"2024-03-18T20:28:52.832968Z","shell.execute_reply.started":"2024-03-18T20:28:52.825481Z","shell.execute_reply":"2024-03-18T20:28:52.831837Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_tax_a_1 = pl.read_csv(dataPath + \"csv_files/train/train_tax_registry_a_1.csv\").pipe(set_table_dtypes) \ntrain_tax_b_1 = pl.read_csv(dataPath + \"csv_files/train/train_tax_registry_b_1.csv\").pipe(set_table_dtypes) \ntest_tax_a_1 = pl.read_csv(dataPath + \"csv_files/test/test_tax_registry_a_1.csv\").pipe(set_table_dtypes) \ntest_tax_b_1 = pl.read_csv(dataPath + \"csv_files/test/test_tax_registry_b_1.csv\").pipe(set_table_dtypes)\n\ntrain_tax = agg_tax(train_tax_a_1, 'amount_4527230A')\\\n    .join(agg_tax(train_tax_b_1, 'amount_4917619A'), how='left', on='case_id')\\\n\ntest_tax = agg_tax(test_tax_a_1, 'amount_4527230A')\\\n    .join(agg_tax(test_tax_b_1, 'amount_4917619A'), how='left', on='case_id')","metadata":{"execution":{"iopub.status.busy":"2024-03-18T20:28:52.834209Z","iopub.execute_input":"2024-03-18T20:28:52.834531Z","iopub.status.idle":"2024-03-18T20:28:54.119002Z","shell.execute_reply.started":"2024-03-18T20:28:52.834504Z","shell.execute_reply":"2024-03-18T20:28:54.117453Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Остальные данные (не изменилось после форка)","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_credit_bureau_b_2_feats, how=\"left\", on=\"case_id\"\n).join(\n    train_tax, how=\"left\", on=\"case_id\"\n)","metadata":{"execution":{"iopub.status.busy":"2024-03-18T20:28:54.122113Z","iopub.execute_input":"2024-03-18T20:28:54.122497Z","iopub.status.idle":"2024-03-18T20:28:56.009970Z","shell.execute_reply.started":"2024-03-18T20:28:54.122463Z","shell.execute_reply":"2024-03-18T20:28:56.008922Z"},"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_credit_bureau_b_2_feats, how=\"left\", on=\"case_id\"\n).join(\n    test_tax, how=\"left\", on=\"case_id\"\n)","metadata":{"execution":{"iopub.status.busy":"2024-03-18T20:28:56.011031Z","iopub.execute_input":"2024-03-18T20:28:56.011369Z","iopub.status.idle":"2024-03-18T20:28:56.028898Z","shell.execute_reply.started":"2024-03-18T20:28:56.011338Z","shell.execute_reply":"2024-03-18T20:28:56.027539Z"},"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":"2024-03-18T20:28:56.031000Z","iopub.execute_input":"2024-03-18T20:28:56.031908Z","iopub.status.idle":"2024-03-18T20:29:06.014023Z","shell.execute_reply.started":"2024-03-18T20:28:56.031855Z","shell.execute_reply":"2024-03-18T20:29:06.012632Z"},"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":"2024-03-18T20:29:06.016991Z","iopub.execute_input":"2024-03-18T20:29:06.017451Z","iopub.status.idle":"2024-03-18T20:29:06.024509Z","shell.execute_reply.started":"2024-03-18T20:29:06.017409Z","shell.execute_reply":"2024-03-18T20:29:06.022972Z"},"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":"2024-03-18T20:29:30.369578Z","iopub.execute_input":"2024-03-18T20:29:30.370040Z","iopub.status.idle":"2024-03-18T20:29:30.375970Z","shell.execute_reply.started":"2024-03-18T20:29:30.370004Z","shell.execute_reply":"2024-03-18T20:29:30.374936Z"},"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    \"num_leaves\": 31,\n    \"learning_rate\": 0.05,\n    \"feature_fraction\": 0.9,\n    \"bagging_fraction\": 0.8,\n    \"bagging_freq\": 5,\n    \"n_estimators\": 1000,\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":"2024-03-18T20:35:40.697637Z","iopub.execute_input":"2024-03-18T20:35:40.698058Z","iopub.status.idle":"2024-03-18T20:37:01.417754Z","shell.execute_reply.started":"2024-03-18T20:35:40.698022Z","shell.execute_reply":"2024-03-18T20:37:01.416566Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import lightgbm as lgb\nimport numpy as np\nfrom sklearn.metrics import roc_auc_score\nfrom hyperopt import hp, tpe, Trials, fmin\n\n# Define search space for hyperparameters\nspace = {\n    \"max_depth\": hp.choice(\"max_depth\", np.arange(3, 11, dtype=int)),\n    \"num_leaves\": hp.choice(\"num_leaves\", np.arange(10, 51, dtype=int)),\n    \"learning_rate\": hp.uniform(\"learning_rate\", 0.01, 0.5),\n    \"feature_fraction\": hp.uniform(\"feature_fraction\", 0.6, 1.0),\n    \"bagging_fraction\": hp.uniform(\"bagging_fraction\", 0.6, 1.0),\n    \"bagging_freq\": hp.choice(\"bagging_freq\", np.arange(1, 11, dtype=int)),\n    \"n_estimators\": 1000,  # Fixed for early stopping\n}\n\ndef objective(params):\n    params[\"boosting_type\"] = \"gbdt\"\n    params[\"objective\"] = \"binary\"\n    params[\"metric\"] = \"auc\"\n    params[\"verbose\"] = -1\n\n    gbm = lgb.train(\n        params,\n        lgb_train,\n        valid_sets=[lgb_valid],\n        early_stopping_rounds=10,\n        verbose_eval=False\n    )\n\n    y_pred = gbm.predict(X_valid)\n    auc = roc_auc_score(y_valid, y_pred)\n    return -auc\n\nsubmit = True\nif not submit:\n    trials = Trials()\n    best = fmin(\n        fn=objective,\n        space=space,\n        algo=tpe.suggest,\n        max_evals=10,  # Adjust as needed\n        trials=trials\n    )\n    best_params = space_eval(space, best)\n    best_gbm = lgb.train(\n        best_params,\n        lgb_train,\n        valid_sets=[lgb_valid],\n        verbose_eval=False\n    )","metadata":{"execution":{"iopub.status.busy":"2024-03-18T20:29:32.837235Z","iopub.execute_input":"2024-03-18T20:29:32.837791Z","iopub.status.idle":"2024-03-18T20:34:51.641320Z","shell.execute_reply.started":"2024-03-18T20:29:32.837744Z","shell.execute_reply":"2024-03-18T20:34:51.639609Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import numpy as np\nfrom sklearn.pipeline import make_pipeline\nfrom sklearn.preprocessing import StandardScaler\nfrom sklearn.svm import SVC\nfrom sklearn.impute import KNNImputer\nfrom sklearn.experimental import enable_iterative_imputer\nfrom sklearn.impute import IterativeImputer, SimpleImputer\n\nif not submit:\n    imp = IterativeImputer(max_iter=3, random_state=0)\n    clf = make_pipeline(StandardScaler(), SimpleImputer(), SVC(gamma='auto'))\n    clf.fit(X_train._get_numeric_data(), y_train)","metadata":{"execution":{"iopub.status.busy":"2024-03-18T20:18:03.038376Z","iopub.execute_input":"2024-03-18T20:18:03.038851Z"},"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":"2024-03-18T20:37:39.085379Z","iopub.execute_input":"2024-03-18T20:37:39.085938Z","iopub.status.idle":"2024-03-18T20:38:03.292749Z","shell.execute_reply.started":"2024-03-18T20:37:39.085892Z","shell.execute_reply":"2024-03-18T20:38:03.291702Z"},"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":"2024-03-18T20:38:03.295026Z","iopub.execute_input":"2024-03-18T20:38:03.295384Z","iopub.status.idle":"2024-03-18T20:38:04.433851Z","shell.execute_reply.started":"2024-03-18T20:38:03.295343Z","shell.execute_reply":"2024-03-18T20:38:04.432806Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import matplotlib.pyplot as plt\nfig, ax = plt.subplots(figsize=(12,12), ncols=1, nrows=1)\nlgb.plot_importance(gbm, max_num_features=50, height=8, ax=ax)","metadata":{"execution":{"iopub.status.busy":"2024-03-18T20:38:04.435053Z","iopub.execute_input":"2024-03-18T20:38:04.435362Z","iopub.status.idle":"2024-03-18T20:38:05.670816Z","shell.execute_reply.started":"2024-03-18T20:38:04.435333Z","shell.execute_reply":"2024-03-18T20:38:05.669511Z"},"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":"2024-03-18T20:38:05.673042Z","iopub.execute_input":"2024-03-18T20:38:05.673428Z","iopub.status.idle":"2024-03-18T20:38:05.806535Z","shell.execute_reply.started":"2024-03-18T20:38:05.673393Z","shell.execute_reply":"2024-03-18T20:38:05.805231Z"},"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":"2024-03-18T20:38:17.083741Z","iopub.execute_input":"2024-03-18T20:38:17.084172Z","iopub.status.idle":"2024-03-18T20:38:17.098943Z","shell.execute_reply.started":"2024-03-18T20:38:17.084137Z","shell.execute_reply":"2024-03-18T20:38:17.097659Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}