{"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":"none","dataSources":[{"sourceId":50160,"databundleVersionId":7921029,"sourceType":"competition"},{"sourceId":9163661,"sourceType":"datasetVersion","datasetId":5536584}],"dockerImageVersionId":30746,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\nimport os\nimport seaborn as sns\nimport matplotlib.pyplot as plt","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2024-08-20T05:03:03.496281Z","iopub.execute_input":"2024-08-20T05:03:03.497083Z","iopub.status.idle":"2024-08-20T05:03:03.502660Z","shell.execute_reply.started":"2024-08-20T05:03:03.497046Z","shell.execute_reply":"2024-08-20T05:03:03.501283Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"pd.set_option('display.max_columns', None)  # Отображать все столбцы","metadata":{"execution":{"iopub.status.busy":"2024-08-20T05:00:11.715643Z","iopub.status.idle":"2024-08-20T05:00:11.716204Z","shell.execute_reply.started":"2024-08-20T05:00:11.715940Z","shell.execute_reply":"2024-08-20T05:00:11.715962Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## train_applprev_1_0 depth=1","metadata":{}},{"cell_type":"code","source":"train_applprev_1_0 = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_applprev_1_0.csv')\nfeature_definitons = pd.read_excel('/kaggle/input/delete/feature_definitions.xlsx')","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:50:22.411225Z","iopub.execute_input":"2024-08-15T09:50:22.412150Z","iopub.status.idle":"2024-08-15T09:50:54.846521Z","shell.execute_reply.started":"2024-08-15T09:50:22.412115Z","shell.execute_reply":"2024-08-15T09:50:54.845489Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_applprev_1_0.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:50:54.849173Z","iopub.execute_input":"2024-08-15T09:50:54.849619Z","iopub.status.idle":"2024-08-15T09:50:54.900858Z","shell.execute_reply.started":"2024-08-15T09:50:54.849589Z","shell.execute_reply":"2024-08-15T09:50:54.899651Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"pd.set_option('display.max_colwidth', None)\nfeature_definitons[feature_definitons['Variable'].isin(train_applprev_1_0.columns)]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:50:54.902418Z","iopub.execute_input":"2024-08-15T09:50:54.903214Z","iopub.status.idle":"2024-08-15T09:50:54.925801Z","shell.execute_reply.started":"2024-08-15T09:50:54.903172Z","shell.execute_reply":"2024-08-15T09:50:54.924694Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"ИНФА ПО СТАРЫМ ЗАЯВКАМ НА КРЕДИТ","metadata":{}},{"cell_type":"markdown","source":"## train_applprev_2.csv","metadata":{}},{"cell_type":"code","source":"train_applprev_2 =  pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_applprev_2.csv')\ntrain_applprev_2.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:50:54.927306Z","iopub.execute_input":"2024-08-15T09:50:54.927644Z","iopub.status.idle":"2024-08-15T09:51:06.181422Z","shell.execute_reply.started":"2024-08-15T09:50:54.927615Z","shell.execute_reply":"2024-08-15T09:51:06.180123Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_applprev_2.columns)]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:51:06.182956Z","iopub.execute_input":"2024-08-15T09:51:06.183312Z","iopub.status.idle":"2024-08-15T09:51:06.196580Z","shell.execute_reply.started":"2024-08-15T09:51:06.183281Z","shell.execute_reply":"2024-08-15T09:51:06.195193Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"ИНФА ПО СТАРЫМ ЗАЯВКАМ НА КРЕДИТ","metadata":{}},{"cell_type":"markdown","source":"## train_base","metadata":{}},{"cell_type":"code","source":"train_base = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_base.csv')\ntrain_base.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:51:06.198190Z","iopub.execute_input":"2024-08-15T09:51:06.198538Z","iopub.status.idle":"2024-08-15T09:51:07.171748Z","shell.execute_reply.started":"2024-08-15T09:51:06.198508Z","shell.execute_reply":"2024-08-15T09:51:07.170420Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_base.columns)].head(40)","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:51:07.173307Z","iopub.execute_input":"2024-08-15T09:51:07.173727Z","iopub.status.idle":"2024-08-15T09:51:07.185085Z","shell.execute_reply.started":"2024-08-15T09:51:07.173687Z","shell.execute_reply":"2024-08-15T09:51:07.183791Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## train_credit_bureau_a_1_0","metadata":{}},{"cell_type":"code","source":"train_credit_bureau_a_1_0 = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_credit_bureau_a_1_0.csv')\ntrain_credit_bureau_a_1_0.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-20T05:00:23.304961Z","iopub.execute_input":"2024-08-20T05:00:23.305632Z","iopub.status.idle":"2024-08-20T05:01:08.469863Z","shell.execute_reply.started":"2024-08-20T05:00:23.305584Z","shell.execute_reply":"2024-08-20T05:01:08.468707Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sns.histplot(train_credit_bureau_a_1_0[train_credit_bureau_a_1_0['annualeffectiverate_199L'] > 0].annualeffectiverate_199L, bins=50, color='skyblue', edgecolor='black', log_scale=True)\nplt.xlabel('annualeffectiverate_199L')\nplt.ylabel('Frequency')\nplt.title('Гистограмма annualeffectiverate_199L')\n\n# Отобразить график\nplt.tight_layout()\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-08-20T05:03:35.591457Z","iopub.execute_input":"2024-08-20T05:03:35.591874Z","iopub.status.idle":"2024-08-20T05:03:36.533942Z","shell.execute_reply.started":"2024-08-20T05:03:35.591843Z","shell.execute_reply":"2024-08-20T05:03:36.532780Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sns.histplot(train_credit_bureau_a_1_0[train_credit_bureau_a_1_0['annualeffectiverate_63L'] > 0].annualeffectiverate_63L, bins=50, color='skyblue', edgecolor='black', log_scale=True)\nplt.xlabel('annualeffectiverate_63L')\nplt.ylabel('Frequency')\nplt.title('Гистограмма annualeffectiverate_63L')\n\n# Отобразить график\nplt.tight_layout()\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-08-20T05:04:06.111622Z","iopub.execute_input":"2024-08-20T05:04:06.112050Z","iopub.status.idle":"2024-08-20T05:04:07.241958Z","shell.execute_reply.started":"2024-08-20T05:04:06.111994Z","shell.execute_reply":"2024-08-20T05:04:07.240587Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_credit_bureau_a_1_0['annualeffectiverate_199L'].value_counts()","metadata":{"execution":{"iopub.status.busy":"2024-08-20T05:01:45.020112Z","iopub.execute_input":"2024-08-20T05:01:45.020500Z","iopub.status.idle":"2024-08-20T05:01:45.050471Z","shell.execute_reply.started":"2024-08-20T05:01:45.020468Z","shell.execute_reply":"2024-08-20T05:01:45.049142Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_credit_bureau_a_1_0.columns)].head(40)","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:51:47.326200Z","iopub.execute_input":"2024-08-15T09:51:47.326570Z","iopub.status.idle":"2024-08-15T09:51:47.344365Z","shell.execute_reply.started":"2024-08-15T09:51:47.326540Z","shell.execute_reply":"2024-08-15T09:51:47.342789Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_credit_bureau_a_1_0.columns)].tail(37)","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:51:47.346193Z","iopub.execute_input":"2024-08-15T09:51:47.346721Z","iopub.status.idle":"2024-08-15T09:51:47.369343Z","shell.execute_reply.started":"2024-08-15T09:51:47.346678Z","shell.execute_reply":"2024-08-15T09:51:47.367943Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## train_credit_bureau_a_2_0.csv","metadata":{}},{"cell_type":"code","source":"train_credit_bureau_a_2_0 = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_credit_bureau_a_2_0.csv')\ntrain_credit_bureau_a_2_0.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:51:47.370909Z","iopub.execute_input":"2024-08-15T09:51:47.371331Z","iopub.status.idle":"2024-08-15T09:52:01.237871Z","shell.execute_reply.started":"2024-08-15T09:51:47.371296Z","shell.execute_reply":"2024-08-15T09:52:01.236399Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_credit_bureau_a_2_0.columns)]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:01.239195Z","iopub.execute_input":"2024-08-15T09:52:01.239498Z","iopub.status.idle":"2024-08-15T09:52:01.253394Z","shell.execute_reply.started":"2024-08-15T09:52:01.239473Z","shell.execute_reply":"2024-08-15T09:52:01.252161Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## train_credit_bureau_b_1.csv","metadata":{}},{"cell_type":"code","source":"train_credit_bureau_b_1 = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_credit_bureau_b_1.csv')\ntrain_credit_bureau_b_1.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:01.255157Z","iopub.execute_input":"2024-08-15T09:52:01.255593Z","iopub.status.idle":"2024-08-15T09:52:01.937774Z","shell.execute_reply.started":"2024-08-15T09:52:01.255549Z","shell.execute_reply":"2024-08-15T09:52:01.936694Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_credit_bureau_b_1.columns)]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:01.939352Z","iopub.execute_input":"2024-08-15T09:52:01.939761Z","iopub.status.idle":"2024-08-15T09:52:01.956306Z","shell.execute_reply.started":"2024-08-15T09:52:01.939707Z","shell.execute_reply":"2024-08-15T09:52:01.955199Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## train_credit_bureau_b_2.csv","metadata":{}},{"cell_type":"code","source":"train_credit_bureau_b_2 = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_credit_bureau_b_2.csv')\ntrain_credit_bureau_b_2.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:01.957525Z","iopub.execute_input":"2024-08-15T09:52:01.957881Z","iopub.status.idle":"2024-08-15T09:52:02.836047Z","shell.execute_reply.started":"2024-08-15T09:52:01.957852Z","shell.execute_reply":"2024-08-15T09:52:02.834618Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_credit_bureau_b_2[train_credit_bureau_b_2['case_id'] == 1934]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:02.838091Z","iopub.execute_input":"2024-08-15T09:52:02.838409Z","iopub.status.idle":"2024-08-15T09:52:02.860533Z","shell.execute_reply.started":"2024-08-15T09:52:02.838383Z","shell.execute_reply":"2024-08-15T09:52:02.859350Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_credit_bureau_b_2.groupby('case_id').agg(\n\n    pmts_pmtsoverdue_635A_max=('pmts_pmtsoverdue_635A', 'max'))","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:02.861995Z","iopub.execute_input":"2024-08-15T09:52:02.862437Z","iopub.status.idle":"2024-08-15T09:52:02.924815Z","shell.execute_reply.started":"2024-08-15T09:52:02.862397Z","shell.execute_reply":"2024-08-15T09:52:02.923556Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_credit_bureau_b_2.columns)]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:02.926334Z","iopub.execute_input":"2024-08-15T09:52:02.926804Z","iopub.status.idle":"2024-08-15T09:52:02.940260Z","shell.execute_reply.started":"2024-08-15T09:52:02.926763Z","shell.execute_reply":"2024-08-15T09:52:02.938804Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## train_debitcard_1.csv","metadata":{}},{"cell_type":"code","source":"train_debitcard_1 = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_debitcard_1.csv')\ntrain_debitcard_1.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:02.941462Z","iopub.execute_input":"2024-08-15T09:52:02.941871Z","iopub.status.idle":"2024-08-15T09:52:03.063287Z","shell.execute_reply.started":"2024-08-15T09:52:02.941836Z","shell.execute_reply":"2024-08-15T09:52:03.062124Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_debitcard_1.columns)]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:03.064822Z","iopub.execute_input":"2024-08-15T09:52:03.065169Z","iopub.status.idle":"2024-08-15T09:52:03.078218Z","shell.execute_reply.started":"2024-08-15T09:52:03.065139Z","shell.execute_reply":"2024-08-15T09:52:03.076731Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## train_deposit_1.csv","metadata":{}},{"cell_type":"code","source":"train_deposit_1 = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_deposit_1.csv')\ntrain_deposit_1.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:03.079953Z","iopub.execute_input":"2024-08-15T09:52:03.080430Z","iopub.status.idle":"2024-08-15T09:52:03.229294Z","shell.execute_reply.started":"2024-08-15T09:52:03.080384Z","shell.execute_reply":"2024-08-15T09:52:03.227942Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_deposit_1.columns)]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:03.230637Z","iopub.execute_input":"2024-08-15T09:52:03.230969Z","iopub.status.idle":"2024-08-15T09:52:03.243970Z","shell.execute_reply.started":"2024-08-15T09:52:03.230941Z","shell.execute_reply":"2024-08-15T09:52:03.242374Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## train_other_1","metadata":{}},{"cell_type":"code","source":"train_other_1 = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_other_1.csv')\ntrain_other_1.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:03.245627Z","iopub.execute_input":"2024-08-15T09:52:03.246100Z","iopub.status.idle":"2024-08-15T09:52:03.322826Z","shell.execute_reply.started":"2024-08-15T09:52:03.246069Z","shell.execute_reply":"2024-08-15T09:52:03.321491Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_other_1.columns)]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:03.324650Z","iopub.execute_input":"2024-08-15T09:52:03.325113Z","iopub.status.idle":"2024-08-15T09:52:03.338968Z","shell.execute_reply.started":"2024-08-15T09:52:03.325073Z","shell.execute_reply":"2024-08-15T09:52:03.337722Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## train_person_1.csv","metadata":{}},{"cell_type":"code","source":"train_person_1 = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_person_1.csv')\ntrain_person_1.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:03.340253Z","iopub.execute_input":"2024-08-15T09:52:03.340579Z","iopub.status.idle":"2024-08-15T09:52:22.420116Z","shell.execute_reply.started":"2024-08-15T09:52:03.340551Z","shell.execute_reply":"2024-08-15T09:52:22.418832Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_person_1.columns)]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:22.426547Z","iopub.execute_input":"2024-08-15T09:52:22.427191Z","iopub.status.idle":"2024-08-15T09:52:22.444026Z","shell.execute_reply.started":"2024-08-15T09:52:22.427145Z","shell.execute_reply":"2024-08-15T09:52:22.442668Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## train_person_2.csv","metadata":{}},{"cell_type":"code","source":"train_person_2 = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_person_2.csv')\ntrain_person_2.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:22.445849Z","iopub.execute_input":"2024-08-15T09:52:22.446419Z","iopub.status.idle":"2024-08-15T09:52:24.995177Z","shell.execute_reply.started":"2024-08-15T09:52:22.446366Z","shell.execute_reply":"2024-08-15T09:52:24.993797Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_person_2.columns)]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:24.996633Z","iopub.execute_input":"2024-08-15T09:52:24.997094Z","iopub.status.idle":"2024-08-15T09:52:25.010229Z","shell.execute_reply.started":"2024-08-15T09:52:24.997054Z","shell.execute_reply":"2024-08-15T09:52:25.009082Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## train_static_0_0.csv","metadata":{}},{"cell_type":"code","source":"train_static_0_0 = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_static_0_0.csv')\ntrain_static_0_0.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:25.012074Z","iopub.execute_input":"2024-08-15T09:52:25.013219Z","iopub.status.idle":"2024-08-15T09:52:49.910194Z","shell.execute_reply.started":"2024-08-15T09:52:25.013183Z","shell.execute_reply":"2024-08-15T09:52:49.908807Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_static_0_0.columns)].iloc[:41]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:49.911836Z","iopub.execute_input":"2024-08-15T09:52:49.912181Z","iopub.status.idle":"2024-08-15T09:52:49.928490Z","shell.execute_reply.started":"2024-08-15T09:52:49.912151Z","shell.execute_reply":"2024-08-15T09:52:49.927369Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_static_0_0.columns)].iloc[40:81]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:49.930045Z","iopub.execute_input":"2024-08-15T09:52:49.930754Z","iopub.status.idle":"2024-08-15T09:52:49.951254Z","shell.execute_reply.started":"2024-08-15T09:52:49.930695Z","shell.execute_reply":"2024-08-15T09:52:49.950101Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_static_0_0.columns)].iloc[81:120]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:49.952679Z","iopub.execute_input":"2024-08-15T09:52:49.953099Z","iopub.status.idle":"2024-08-15T09:52:49.977259Z","shell.execute_reply.started":"2024-08-15T09:52:49.953062Z","shell.execute_reply":"2024-08-15T09:52:49.976016Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_static_0_0.columns)].iloc[119:]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:49.978560Z","iopub.execute_input":"2024-08-15T09:52:49.978948Z","iopub.status.idle":"2024-08-15T09:52:49.997627Z","shell.execute_reply.started":"2024-08-15T09:52:49.978908Z","shell.execute_reply":"2024-08-15T09:52:49.996443Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## train_static_cb_0.csv","metadata":{}},{"cell_type":"code","source":"train_static_cb_0 = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_static_cb_0.csv')\ntrain_static_cb_0.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:49.999273Z","iopub.execute_input":"2024-08-15T09:52:50.000218Z","iopub.status.idle":"2024-08-15T09:52:58.888104Z","shell.execute_reply.started":"2024-08-15T09:52:50.000181Z","shell.execute_reply":"2024-08-15T09:52:58.886991Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_static_cb_0['riskassesment_302T']","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:58.889290Z","iopub.execute_input":"2024-08-15T09:52:58.889594Z","iopub.status.idle":"2024-08-15T09:52:58.898165Z","shell.execute_reply.started":"2024-08-15T09:52:58.889568Z","shell.execute_reply":"2024-08-15T09:52:58.896900Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_static_cb_0.columns)]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:58.899689Z","iopub.execute_input":"2024-08-15T09:52:58.900139Z","iopub.status.idle":"2024-08-15T09:52:58.922799Z","shell.execute_reply.started":"2024-08-15T09:52:58.900103Z","shell.execute_reply":"2024-08-15T09:52:58.921391Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## train_tax_registry_a_1.csv","metadata":{}},{"cell_type":"code","source":"train_tax_registry_a_1 = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_tax_registry_a_1.csv')\ntrain_tax_registry_a_1.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:52:58.924698Z","iopub.execute_input":"2024-08-15T09:52:58.925103Z","iopub.status.idle":"2024-08-15T09:53:01.876300Z","shell.execute_reply.started":"2024-08-15T09:52:58.925067Z","shell.execute_reply":"2024-08-15T09:53:01.875030Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_tax_registry_a_1.columns)]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:53:01.878066Z","iopub.execute_input":"2024-08-15T09:53:01.879052Z","iopub.status.idle":"2024-08-15T09:53:01.896717Z","shell.execute_reply.started":"2024-08-15T09:53:01.879004Z","shell.execute_reply":"2024-08-15T09:53:01.895190Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## train_tax_registry_b_1.csv","metadata":{}},{"cell_type":"code","source":"train_tax_registry_b_1 = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_tax_registry_b_1.csv')\ntrain_tax_registry_b_1.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:53:01.898350Z","iopub.execute_input":"2024-08-15T09:53:01.898909Z","iopub.status.idle":"2024-08-15T09:53:02.958078Z","shell.execute_reply.started":"2024-08-15T09:53:01.898874Z","shell.execute_reply":"2024-08-15T09:53:02.956773Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_tax_registry_b_1.columns)]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:53:02.959758Z","iopub.execute_input":"2024-08-15T09:53:02.960095Z","iopub.status.idle":"2024-08-15T09:53:02.972523Z","shell.execute_reply.started":"2024-08-15T09:53:02.960067Z","shell.execute_reply":"2024-08-15T09:53:02.971270Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## train_tax_registry_c_1.csv","metadata":{}},{"cell_type":"code","source":"train_tax_registry_c_1 = pd.read_csv('/kaggle/input/home-credit-credit-risk-model-stability/csv_files/train/train_tax_registry_c_1.csv')\ntrain_tax_registry_c_1.head()","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:53:02.973864Z","iopub.execute_input":"2024-08-15T09:53:02.974194Z","iopub.status.idle":"2024-08-15T09:53:05.962464Z","shell.execute_reply.started":"2024-08-15T09:53:02.974167Z","shell.execute_reply":"2024-08-15T09:53:05.961308Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"\n13th place solution - pmts_year_1139T postprocess\n----\n**Era**\nдесятиелетие возраста\n```\ndf_base = df_base.with_columns(\n        ((pl.col(\"first_birth_259D\") / 10).floor() * 10).alias(\"era\").cast(pl.Int32),\n)\n```\n----\n\n**Age at Start of Employment**\n\n(дата начала работы - Дата рождения человека).дней\n```\ndf = df.with_columns(\n        ((pl.col(\"empl_employedfrom_271D\") - pl.col(\"birth_259D\")).dt.total_days() // 365).cast(pl.Int32).alias(\"agestartofemploymentA\"), \n)\n```\n----\n\n**Employment Period**\n```\ndf_base = df_base.with_columns(\n        (pl.col(\"first_birth_259D\") - pl.col(\"first_agestartofemploymentA\")).alias(\"durationofemploymentA\"),\n)\n```\n----\n**Date processing other than suffix D**\n```\ndef handle_dates(df):\n    for colin 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                elif \"year\" in col:\n            df = df.with_columns(pl.col(col) - pl.col(\"date_decision\").dt.year())\n            df = df.with_columns(pl.col(col).cast(pl.Int32))\n```\n\n**Merge tax_registry tables**\n\nFrom some case_id with multiple provider information, we inferred the correspondence of each table column and made it into one table.\n\n**Aggregation of String type (mode and n_unique)**\nWe used the process pl.col(col).drop_nans().drop_nulls().mode().sort().first() for reproducibility in polars.\n\n**Removal of features that fluctuate greatly during the training data period**\n\nWe manually checked and removed features that fluctuate greatly with each WEEK_NUM.","metadata":{}},{"cell_type":"code","source":"feature_definitons[feature_definitons['Variable'].isin(train_tax_registry_c_1.columns)]","metadata":{"execution":{"iopub.status.busy":"2024-08-15T09:53:05.963811Z","iopub.execute_input":"2024-08-15T09:53:05.964134Z","iopub.status.idle":"2024-08-15T09:53:05.976275Z","shell.execute_reply.started":"2024-08-15T09:53:05.964107Z","shell.execute_reply":"2024-08-15T09:53:05.975148Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Info\n\n**53th Place Solution (without metric hack)**\n\nWe've mostly used public codes for feature engineering and made some modifications to them. For example, we scaled the total day difference between the date features and the date_decision by dividing them by -365. This minor adjustment affected all date-based features and enhanced our cross-validation performance.\n\nдни свести к годам?\n```\ndef 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() / -365)\n    df = df.drop(\"date_decision\", \"MONTH\")\n    return df\n```\nWe examined the most important features of the previous Home Credit competition and tried to extract some features that had characteristics similar to them. Here are some features that improved our CV.\n```\nЦена кредита. / Ежемесячная аннуитетная сумма.\n(df_base['price_1097A'] / df_base['annuity_780A']).alias('credit_annuity_ratio')\n\nПроцентная ставка/ Цена кредита\n(df_base['eir_270L'] / df_base['price_1097A']).alias('interest_share')\n\nВыданная сумма кредита после консолидации./ Сумма кредита или кредитный лимит.\n\n(df_base['disbursedcredamount_1113A'] / df_base['credamount_770A']).alias('cred_disbursed_ratio')\n\nОбщая сумма долга. / (1 + Сумма кредита или кредитный лимит)\n(df_base['totaldebt_9A'] / (1 + df_base['credamount_770A'])).alias('totaldebt_credamount_ratio')\n```","metadata":{}},{"cell_type":"markdown","source":"**57th place, 0.528. Without hack.**\n\nWhat helped improve the models:\n* Sorting by numgroups before aggregations - at least it worked for \"first\" and \"last\" aggregators. For some models it was used with combination of aggregation by numgroup2 and then by numgroup1 (after that some tables can also be concatenated, for example applprev, credit_bureau_a, credit_bureau_b).\n* Models using datasets without date columns - the scores of these models weren't as big, but they complemented the ensemble well enough.\n* Using \"Coefficient of Variations\" aggregator - it helped a lot to improve the score of the CatBoost model.\n* Count encoding of categorical columns (from public notebooks)\n* Filtering of correlated columns (from public notebooks).\n* Collapse rare categories in one group.\n* Aggregators used: max, min, first, last, mean, var, sum, std plus coef_var (distributed differently among all models, column types, or used only for certain columns).\n\nWhat didn't work (at least in my experiments):\n\n* Feature selection based on adversarial validation (within train ds), permutation importance (including stability metric).\n* I tried to create some loss functions (based on functions simpler than initial metric - from topics about metrics; incorporating weights and so on) for GBDT models, but without improvements.\n* CV strategies based on splitting train ds on week numbers using distributions of some statistics.\n* Tuning GBDT models using some specific parameters like dart for LGBM, ordered boosting_type for CatBoost and so on.\n* Using IsolationForest with SMOTENC\n* Different scalers for 1DCNN model like PowerTransformer, QuantileTransformer. Finally I used Log1p transformation only for columns with max_min-mean ratio greater than 100.\n* Replace nums NaN with mean (or constant values like -99). BTW LAMA pipeline uses this approach, and I tried to use 0s, but LB didn't improve and I rejected this idea, but after deadline I saw that private score was improved for that submission.\n* Compress some categories (e.g. by template P0_111_121 to P0 or P0_111)\n* Some non-standard, custom aggregators such as entropy, mean-to-max, zero frequency, normalization within group, and so on. The goal was also to find aggragators with relative, not absolute, values.\n* Additional feature engineering like ratios between some columns didn't improve models. As I understand after first competition was created very good features and squeezing out from dataset additional gains was not so simple task.\n* By the results of previous competition I tried Regularized Greedy Forest (FastRGF version) but it has very poor performance and worsened final ensemble. Another idea from that competition was to train MLP only on some group of tables, but again without any results, so I discarded this idea as well. I also experimented with other models like DAE, TabTransformer, NN-cat_embeddings, packages like pytorch_tabular, but also I can't train them to good enough results.\n* Combining some categorical features (similar to how CatBoost works).\n* Balancing classes using parameters like scale_pos_weight.\n* Include 1DCNN in the final ensemble. I struggled a lot to overcome the overfitting of this model, but eventually it is worsening the final ensemble from 0.529 to 0.528. Some variations of this model (with another ds) has relatively good private score (0.485) but LB was not so good (only 0.555), so I didn't include the best and maybe was digging in the wrong direction. Besides, without metric hacking this model will not contribute much to the final ensemble/position.\n* Include some additional data such as FedFunds.\n* Filter some weeks.\n* Of course ideas from the community, discussions, public notebooks helped a lot.\n\nUpdate. 29.05.2024\nJust out of curiosity, I tried adding the CatBoost model with score 0.520 to the final ensemble and the total private score increased to 0.536.\n\nUpdate. 5.06.2024\nFeature importance by gain I used in selecting top-20 features from credit_bureau_a tables for lama-mlp model. This helped to avoid overfitting this model.","metadata":{}},{"cell_type":"markdown","source":"**59th Place Solution for the Home Credit - Credit Risk Model Stability Competition**\n\n\nFor each of the categorical features of appl_prev_1, we counted the number of occurrences of each categories, which increase the LB by 0.013. Khiem also perform count encoding on the categorical features.\n```\n# видимо что-то вроде Count encoding \n\nfor col in df.select([pl.col(pl.String), pl.col(pl.Boolean)]).columns:\n    values = df[col].unique().to_list()\n    if len(values) <= 10 and df[col].is_null().mean() < 0.9:\n        for value in values:\n            agg_cols += [pl.col(col).filter(pl.col(col) == value).count().alias(f\"{col}_{value}_C\")]\n```\n\nWe got the idea to use the riskassesment_302T from this notebook. This increased LB by 0.001.\n```\ndef transform_cols(df: pl.DataFrame) -> pl.DataFrame:\n    \"\"\"\n    Transforms columns in the DataFrame according to predefined rules.\n\n    Args:\n    - df (pl.DataFrame): Input DataFrame.\n\n    Returns:\n    - pl.DataFrame: DataFrame with transformed columns.\n    \"\"\"\n    if \"riskassesment_302T\" in df.columns:\n        if df[\"riskassesment_302T\"].dtype == pl.Null:\n            df = df.with_columns(\n                [\n                    pl.Series(\n                        \"riskassesment_302T_rng\", df[\"riskassesment_302T\"], pl.UInt8\n                    ),\n                    pl.Series(\n                        \"riskassesment_302T_mean\", df[\"riskassesment_302T\"], pl.UInt8\n                    ),\n                ]\n            )\n        else:\n            pct_low: pl.Series = (\n                df[\"riskassesment_302T\"]\n                .str.split(\" - \")\n                .apply(lambda x: x[0].replace(\"%\", \"\"))\n                .cast(pl.UInt8)\n            )\n            pct_high: pl.Series = (\n                df[\"riskassesment_302T\"]\n                .str.split(\" - \")\n                .apply(lambda x: x[1].replace(\"%\", \"\"))\n                .cast(pl.UInt8)\n            )\n\n            diff: pl.Series = pct_high - pct_low\n            avg: pl.Series = ((pct_low + pct_high) / 2).cast(pl.Float32)\n\n            del pct_high, pct_low\n            gc.collect()\n\n            df = df.with_columns(\n                [\n                    diff.alias(\"riskassesment_302T_rng\"),\n                    avg.alias(\"riskassesment_302T_mean\"),\n                ]\n            )\n\n        df.drop(\"riskassesment_302T\")\n\n    return df\n ```\n\nWe computed the number of months of each candidate on previous loan.\n```\ndf.group_by([\"case_id\", \"num_group1\"]).agg(\n        pl.count(\"pmts_month_158T\").alias(\"count_pmts_month_158T\"),\n        pl.count(\"pmts_month_706T\").alias(\"count_pmts_month_706T\"),\n    ).group_by(\"case_id\").agg(\n        pl.when(pl.col(\"count_pmts_month_158T\") > 0).then(pl.col(\"count_pmts_month_158T\")).otherwise(None).min().alias(\"min_count_pmts_month_158T\"),\n        pl.when(pl.col(\"count_pmts_month_158T\") > 0).then(pl.col(\"count_pmts_month_158T\")).otherwise(None).max().alias(\"max_count_pmts_month_158T\"),\n        pl.when(pl.col(\"count_pmts_month_158T\") > 0).then(pl.col(\"count_pmts_month_158T\")).otherwise(None).mean().alias(\"mean_count_pmts_month_158T\"),\n        pl.when(pl.col(\"count_pmts_month_158T\") > 0).then(pl.col(\"count_pmts_month_158T\")).otherwise(None).std().alias(\"std_count_pmts_month_158T\"),\n        pl.when(pl.col(\"count_pmts_month_706T\") > 0).then(pl.col(\"count_pmts_month_706T\")).otherwise(None).min().alias(\"min_count_pmts_month_706T\"),\n        pl.when(pl.col(\"count_pmts_month_706T\") > 0).then(pl.col(\"count_pmts_month_706T\")).otherwise(None).max().alias(\"max_count_pmts_month_706T\"),\n        pl.when(pl.col(\"count_pmts_month_706T\") > 0).then(pl.col(\"count_pmts_month_706T\")).otherwise(None).mean().alias(\"mean_count_pmts_month_706T\"),\n        pl.when(pl.col(\"count_pmts_month_706T\") > 0).then(pl.col(\"count_pmts_month_706T\")).otherwise(None).std().alias(\"std_count_pmts_month_706T\")\n)\n```","metadata":{}},{"cell_type":"markdown","source":"![image.png](attachment:846f5d93-42f5-4de5-8357-ecbcd452aaf3.png)","metadata":{},"attachments":{"846f5d93-42f5-4de5-8357-ecbcd452aaf3.png":{"image/png":"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"}}},{"cell_type":"markdown","source":"## Вопросы:\n1) Какие из файлов в соревновании, а также колонок внутри есть у нас, помимо кредитного бюро? как внутренние так и внешние\n\n2) Есть ли уже они в готовом виде, может какой глоссарий для сортировки их и убрать сразу лишние.\n\n3) Акутальность данных для построения, тут с начала 2019 трейн по ноябрь 2020, т.е. попадает ковид, сильно меняются данные там и после ковида видимо меняется сама процедура (цифровизация займов?)\n\n4) может их тогда подрезать, но подрезать до ковида, или от ковида, вопрос!\n\n5) денежные фичи (\"выраженные в деньгах\"), говорил не использовать, но можно ведь сделать фичи относительно денег, какие нибудь пропорции от условно среднего, максимального значения, \"аля нормализация\". Или в зависимости от стран / цен/ других факторов эта величина плавающая и нестабильная\n\n","metadata":{}},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}