{"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":30648,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"## ДЗ 2. Соревнование.\n\nВыполнил: Чайчук Михаил БПМИ201\n\n### Краткое описание процесса:\n\nЗа основу я взял кернел https://www.kaggle.com/code/greysky/home-credit-baseline?rvi=1 так как в нем очень удобно и понятно сделан процесс аггрегации таблиц.\n\nПайплайн предобработки таблиц из этого кернела я оставил в целом таким же.\n\nИзменения: \n\nЯ провел некоторую работу с фичами. Изменил фильтрацию, заполнил пропуски, удалил некоторые признаки. Также подобрал гиперпараметры, чего не было в оригинальной тетрадке + по-другому обучил модель","metadata":{}},{"cell_type":"code","source":"import os\nimport gc\nfrom glob import glob\nfrom pathlib import Path\nfrom datetime import datetime\n\nimport numpy as np\nimport pandas as pd\nimport polars as pl\n\nimport matplotlib.pyplot as plt\nimport seaborn as sns\n\nfrom sklearn.model_selection import cross_val_score, StratifiedGroupKFold\nfrom sklearn.base import BaseEstimator, ClassifierMixin\n\nimport lightgbm as lgb\n\nimport warnings\nwarnings.simplefilter(action='ignore', category=FutureWarning)","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Часть 1. собираем таблицы в одну","metadata":{}},{"cell_type":"markdown","source":"### Pipeline","metadata":{}},{"cell_type":"code","source":"class Pipeline:\n    @staticmethod\n    def set_table_dtypes(df):\n        for col in df.columns:\n            if col in [\"case_id\", \"WEEK_NUM\", \"num_group1\", \"num_group2\"]:\n                df = df.with_columns(pl.col(col).cast(pl.Int32))\n            elif col in [\"date_decision\"]:\n                df = df.with_columns(pl.col(col).cast(pl.Date))\n            elif col[-1] in (\"P\", \"A\"):\n                df = df.with_columns(pl.col(col).cast(pl.Float64))\n            elif col[-1] in (\"M\",):\n                df = df.with_columns(pl.col(col).cast(pl.String))\n            elif col[-1] in (\"D\",):\n                df = df.with_columns(pl.col(col).cast(pl.Date))            \n\n        return df\n    \n    @staticmethod\n    def handle_dates(df):\n        for col in df.columns:\n            if col[-1] in (\"D\",):\n                df = df.with_columns(pl.col(col) - pl.col(\"date_decision\"))\n                df = df.with_columns(pl.col(col).dt.total_days())\n                df = df.with_columns(pl.col(col).cast(pl.Float32))\n                \n        df = df.drop(\"date_decision\", \"MONTH\")\n\n        return df\n    \n    @staticmethod\n    def filter_cols(df): # Тут добавил возможность регулировать, столбцы с каким количеством нулевых значений выбрасывать\n        for col in df.columns:\n            if col not in [\"target\", \"case_id\", \"WEEK_NUM\"]:\n                isnull = df[col].is_null().mean()\n\n                if isnull > 0.8:\n                    df = df.drop(col)\n\n        for col in df.columns:\n            if (col not in [\"target\", \"case_id\", \"WEEK_NUM\"]) & (df[col].dtype == pl.String):\n                freq = df[col].n_unique()\n\n                if (freq == 1) | (freq > 50):\n                    df = df.drop(col)\n\n        return df","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Automatic Aggregation","metadata":{}},{"cell_type":"code","source":"class Aggregator:\n    @staticmethod\n    def num_expr(df):\n        cols = [col for col in df.columns if col[-1] in (\"P\", \"A\")]\n\n        expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n\n        return expr_max\n\n    @staticmethod\n    def date_expr(df):\n        cols = [col for col in df.columns if col[-1] in (\"D\",)]\n\n        expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n\n        return expr_max\n\n    @staticmethod\n    def str_expr(df):\n        cols = [col for col in df.columns if col[-1] in (\"M\",)]\n        \n        expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n\n        return expr_max\n\n    @staticmethod\n    def other_expr(df):\n        cols = [col for col in df.columns if col[-1] in (\"T\", \"L\")]\n        \n        expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n\n        return expr_max\n    \n    @staticmethod\n    def count_expr(df):\n        cols = [col for col in df.columns if \"num_group\" in col]\n\n        expr_max = [pl.max(col).alias(f\"max_{col}\") for col in cols]\n\n        return expr_max\n\n    @staticmethod\n    def get_exprs(df):\n        exprs = Aggregator.num_expr(df) + \\\n                Aggregator.date_expr(df) + \\\n                Aggregator.str_expr(df) + \\\n                Aggregator.other_expr(df) + \\\n                Aggregator.count_expr(df)\n\n        return exprs","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### File I/O","metadata":{}},{"cell_type":"code","source":"def read_file(path, depth=None):\n    df = pl.read_parquet(path)\n    df = df.pipe(Pipeline.set_table_dtypes)\n    \n    if depth in [1, 2]:\n        df = df.group_by(\"case_id\").agg(Aggregator.get_exprs(df))\n    \n    return df\n\ndef read_files(regex_path, depth=None):\n    chunks = []\n    for path in glob(str(regex_path)):\n        df = pl.read_parquet(path)\n        df = df.pipe(Pipeline.set_table_dtypes)\n        \n        if depth in [1, 2]:\n            df = df.group_by(\"case_id\").agg(Aggregator.get_exprs(df))\n        \n        chunks.append(df)\n        \n    df = pl.concat(chunks, how=\"vertical_relaxed\")\n    df = df.unique(subset=[\"case_id\"])\n    \n    return df","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Feature Engineering","metadata":{}},{"cell_type":"code","source":"def feature_eng(df_base, depth_0, depth_1, depth_2):\n    df_base = (\n        df_base\n        .with_columns(\n            month_decision = pl.col(\"date_decision\").dt.month(),\n            weekday_decision = pl.col(\"date_decision\").dt.weekday(),\n        )\n    )\n        \n    for i, df in enumerate(depth_0 + depth_1 + depth_2):\n        df_base = df_base.join(df, how=\"left\", on=\"case_id\", suffix=f\"_{i}\")\n        \n    df_base = df_base.pipe(Pipeline.handle_dates)\n    \n    return df_base","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def to_pandas(df_data, cat_cols=None):\n    df_data = df_data.to_pandas()\n    \n    if cat_cols is None:\n        cat_cols = list(df_data.select_dtypes(\"object\").columns)\n    \n    df_data[cat_cols] = df_data[cat_cols].astype(\"category\")\n    \n    return df_data, cat_cols","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Configuration","metadata":{}},{"cell_type":"code","source":"ROOT            = Path(\"/kaggle/input/home-credit-credit-risk-model-stability\")\nTRAIN_DIR       = ROOT / \"parquet_files\" / \"train\"\nTEST_DIR        = ROOT / \"parquet_files\" / \"test\"","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Train Files Read & Feature Engineering","metadata":{}},{"cell_type":"code","source":"data_store = {\n    \"df_base\": read_file(TRAIN_DIR / \"train_base.parquet\"),\n    \"depth_0\": [\n        read_file(TRAIN_DIR / \"train_static_cb_0.parquet\"),\n        read_files(TRAIN_DIR / \"train_static_0_*.parquet\"),\n    ],\n    \"depth_1\": [\n        read_files(TRAIN_DIR / \"train_applprev_1_*.parquet\", 1),\n        read_file(TRAIN_DIR / \"train_tax_registry_a_1.parquet\", 1),\n        read_file(TRAIN_DIR / \"train_tax_registry_b_1.parquet\", 1),\n        read_file(TRAIN_DIR / \"train_tax_registry_c_1.parquet\", 1),\n        read_files(TRAIN_DIR / \"train_credit_bureau_a_1_*.parquet\", 1),\n        read_file(TRAIN_DIR / \"train_credit_bureau_b_1.parquet\", 1),\n        read_file(TRAIN_DIR / \"train_other_1.parquet\", 1),\n        read_file(TRAIN_DIR / \"train_person_1.parquet\", 1),\n        read_file(TRAIN_DIR / \"train_deposit_1.parquet\", 1),\n        read_file(TRAIN_DIR / \"train_debitcard_1.parquet\", 1),\n    ],\n    \"depth_2\": [\n        read_file(TRAIN_DIR / \"train_credit_bureau_b_2.parquet\", 2),\n        read_files(TRAIN_DIR / \"train_credit_bureau_a_2_*.parquet\", 2),\n    ]\n}","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_train = feature_eng(**data_store)\n\nprint(\"train data shape:\\t\", df_train.shape)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Test Files Read & Feature Engineering","metadata":{}},{"cell_type":"code","source":"data_store = {\n    \"df_base\": read_file(TEST_DIR / \"test_base.parquet\"),\n    \"depth_0\": [\n        read_file(TEST_DIR / \"test_static_cb_0.parquet\"),\n        read_files(TEST_DIR / \"test_static_0_*.parquet\"),\n    ],\n    \"depth_1\": [\n        read_files(TEST_DIR / \"test_applprev_1_*.parquet\", 1),\n        read_file(TEST_DIR / \"test_tax_registry_a_1.parquet\", 1),\n        read_file(TEST_DIR / \"test_tax_registry_b_1.parquet\", 1),\n        read_file(TEST_DIR / \"test_tax_registry_c_1.parquet\", 1),\n        read_files(TEST_DIR / \"test_credit_bureau_a_1_*.parquet\", 1),\n        read_file(TEST_DIR / \"test_credit_bureau_b_1.parquet\", 1),\n        read_file(TEST_DIR / \"test_other_1.parquet\", 1),\n        read_file(TEST_DIR / \"test_person_1.parquet\", 1),\n        read_file(TEST_DIR / \"test_deposit_1.parquet\", 1),\n        read_file(TEST_DIR / \"test_debitcard_1.parquet\", 1),\n    ],\n    \"depth_2\": [\n        read_file(TEST_DIR / \"test_credit_bureau_b_2.parquet\", 2),\n        read_files(TEST_DIR / \"test_credit_bureau_a_2_*.parquet\", 2),\n    ]\n}","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_test = feature_eng(**data_store)\n\nprint(\"test data shape:\\t\", df_test.shape)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Часть 2. Работа с признаками","metadata":{}},{"cell_type":"markdown","source":"Тут я отфильтровал фичи, у которых более 80% пропусков + категориальные фичи, у которых более 80 категорий","metadata":{}},{"cell_type":"code","source":"df_train = df_train.pipe(Pipeline.filter_cols)\ndf_test = df_test.select([col for col in df_train.columns if col != \"target\"])\n\nprint(\"train data shape:\\t\", df_train.shape)\nprint(\"test data shape:\\t\", df_test.shape)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Pandas Conversion","metadata":{}},{"cell_type":"code","source":"df_train, cat_cols = to_pandas(df_train)\ndf_test, cat_cols = to_pandas(df_test, cat_cols)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_train","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"len([col for col in df_train.columns if 'num_group' in col])","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Уберем также все признаки num_group. Их мало и они не особо информативные","metadata":{}},{"cell_type":"code","source":"num_group_columns = [col for col in df_train.columns if 'num_group' in col]\ndf_train.drop(columns=num_group_columns, inplace=True)\ndf_test.drop(columns=num_group_columns, inplace=True)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Garbage Collection","metadata":{}},{"cell_type":"code","source":"del data_store\n\ngc.collect()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Для обучения я буду использовать CatBoost. Поэтому надо предварительно избавиться от NaN во всех фичах","metadata":{}},{"cell_type":"code","source":"import catboost\nfrom catboost.utils import eval_metric\nimport hyperopt\nfrom hyperopt import hp, fmin, tpe, Trials, STATUS_OK\nfrom functools import partial\nfrom sklearn.model_selection import train_test_split","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X = df_train.drop(columns=[\"target\", \"case_id\", \"WEEK_NUM\"])\ny = df_train[\"target\"]\nweeks = df_train[\"WEEK_NUM\"]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"num_features = X._get_numeric_data().columns\ncat_features = [col for col in X.columns if col not in num_features]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"for col in num_features:\n    X[col] = X[col].fillna(X[col].mean())\n    \nfor col in cat_features:\n    categories = list(pd.Categorical(X[col]).categories) + ['Missing']\n    X[col] = pd.Categorical(X[col], categories=categories)\n    X[col] = X[col].fillna('Missing')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_test = df_test.drop(columns=[\"WEEK_NUM\"])","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test_num_features = X_test._get_numeric_data().columns\ntest_cat_features = [col for col in X_test.columns if col not in num_features]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"for col in test_num_features:\n    X_test[col] = X_test[col].fillna(X_test[col].mean())\n    \nfor col in cat_features:\n    categories = list(pd.Categorical(X_test[col]).categories) + ['Missing']\n    X_test[col] = pd.Categorical(X_test[col], categories=categories)\n    X_test[col] = X_test[col].fillna('Missing')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_test = X_test.set_index(\"case_id\")","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"del df_train\ndel df_test\ngc.collect()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train, X_val, y_train, y_val = train_test_split(X, y, test_size=0.15, random_state=2204)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_pool = catboost.Pool(X_train, y_train, cat_features=cat_features)\nval_pool = catboost.Pool(X_val, y_val, cat_features=cat_features)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Обучим один простой катбуст и попробуем помотреть на важность признаков","metadata":{}},{"cell_type":"code","source":"model = catboost.CatBoostClassifier(iterations=1000, verbose=100, early_stopping_rounds=50, max_depth=8, random_seed=2204, task_type='GPU', eval_metric='AUC')\nmodel.fit(train_pool, eval_set=val_pool)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"importances = model.get_feature_importance(prettified=True)\nimportances","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sns.histplot(importances['Importances'])","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Как видим, у нас присутствует много фичей с очень низкой полезностью. Оставим только те, важность которых > 0.2","metadata":{}},{"cell_type":"code","source":"important_features = importances[importances['Importances'] >= 0.2]['Feature Id']","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train = X_train[important_features]\nX_val = X_val[important_features]\nX_test = X_test[important_features]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"##### importances[importances['Importances'] > 0.2]","metadata":{"execution":{"iopub.status.busy":"2024-03-18T08:03:07.877417Z","iopub.execute_input":"2024-03-18T08:03:07.878327Z","iopub.status.idle":"2024-03-18T08:03:07.890927Z","shell.execute_reply.started":"2024-03-18T08:03:07.878289Z","shell.execute_reply":"2024-03-18T08:03:07.889695Z"}}},{"cell_type":"code","source":"num_features = X_train._get_numeric_data().columns\ncat_features = [col for col in X_train.columns if col not in num_features]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Часть 3. Подбор гиперпараметров\n\nИспользуем библиотеку hyperopt","metadata":{}},{"cell_type":"code","source":"def objective(params, train_pool, val_pool):\n    estimator = catboost.CatBoostClassifier(**params, iterations=1000, verbose=100, early_stopping_rounds=50, eval_metric='AUC', random_seed=2204, task_type='GPU')\n    estimator.fit(train_pool, eval_set=val_pool)\n    y_pred = estimator.predict_proba(X_val)[:, 1]\n\n    return   {'loss': -eval_metric(y_val.tolist(), y_pred, 'AUC')[0], 'params': params, 'status': STATUS_OK}","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"search_space = {\n                'learning_rate': hp.choice('learning_rate', options=np.logspace(-4, -1, 10)),\n                'max_depth': hp.choice('max_depth', options=[4, 6, 8, 10, 12]),\n                'l2_leaf_reg': hp.choice('l2_leaf_reg', options=[1, 3, 5, 7, 9]),\n                }\n\ntrials = Trials()\n\ntrain_pool = catboost.Pool(X_train, y_train, cat_features=cat_features)\nval_pool = catboost.Pool(X_val, y_val, cat_features=cat_features)\n\nfmin(fn=partial(objective, train_pool=train_pool, val_pool=val_pool),\n     space=search_space,\n     algo=tpe.suggest,\n     max_evals=20,\n     trials=trials,\n     show_progressbar=True\n     )","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"best_params = trials.best_trial['result']['params']\nbest_params","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Часть 4. Обучение","metadata":{}},{"cell_type":"markdown","source":"### Voting Model. Класс для ансамблирования моделей","metadata":{}},{"cell_type":"code","source":"class VotingModel(BaseEstimator, ClassifierMixin):\n    def __init__(self, estimators):\n        super().__init__()\n        self.estimators = estimators\n        \n    def fit(self, X, y=None):\n        return self\n    \n    def predict(self, X):\n        y_preds = [estimator.predict(X) for estimator in self.estimators]\n        return np.mean(y_preds, axis=0)\n    \n    def predict_proba(self, X):\n        y_preds = [estimator.predict_proba(X) for estimator in self.estimators]\n        return np.mean(y_preds, axis=0)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from tqdm.notebook import tqdm","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X = X[X_train.columns]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"cv = StratifiedGroupKFold(n_splits=5, shuffle=False)\n\n\nfitted_models = []\n\nfor idx_train, idx_val in tqdm(cv.split(X, y, groups=weeks)):\n    X_train, y_train = X.iloc[idx_train], y.iloc[idx_train]\n    X_val, y_val = X.iloc[idx_val], y.iloc[idx_val]\n    \n    train_pool = catboost.Pool(X_train, y_train, cat_features=cat_features)\n    val_pool = catboost.Pool(X_val, y_val, cat_features=cat_features)\n    \n    model = catboost.CatBoostClassifier(**best_params, iterations=1000, verbose=100, early_stopping_rounds=50, random_seed=2204, task_type='GPU', eval_metric='AUC')\n    model.fit(train_pool, eval_set=val_pool)\n\n    fitted_models.append(model)\n\nmodel = VotingModel(fitted_models)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Часть 5. Предсказание","metadata":{}},{"cell_type":"code","source":"y_pred = pd.Series(model.predict_proba(X_test)[:, 1], index=X_test.index)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_subm = pd.read_csv(ROOT / \"sample_submission.csv\")\ndf_subm = df_subm.set_index(\"case_id\")\n\ndf_subm[\"score\"] = y_pred\ndf_subm.to_csv(\"submission.csv\")","metadata":{"trusted":true},"execution_count":null,"outputs":[]}]}