{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.12.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[],"dockerImageVersionId":28755,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session\n\n# Use the kagglehub client library to attach Kaggle resources like competitions, datasets, and models to your session\n# Learn more about kagglehub: https://github.com/Kaggle/kagglehub/blob/main/README.md\n\nimport kagglehub\n# kagglehub.dataset_download('<owner>/<dataset-slug>')","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"execution":{"iopub.status.busy":"2026-08-12T06:26:02.205308Z","iopub.execute_input":"2026-08-12T06:26:02.205784Z","iopub.status.idle":"2026-08-12T06:26:02.215630Z","shell.execute_reply.started":"2026-08-12T06:26:02.205754Z","shell.execute_reply":"2026-08-12T06:26:02.214912Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# the data is tooo large so firstly we need to create functions to compress the data\n# the idea is that the fn iterates through all columns of a dataframe and modifies the data type to reduce memory usage. (<64)\n# all thanks to Kaggle kweens btw. this data strorage system was difficult to follow ","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T06:26:04.054972Z","iopub.execute_input":"2026-08-12T06:26:04.055932Z","iopub.status.idle":"2026-08-12T06:26:04.060217Z","shell.execute_reply.started":"2026-08-12T06:26:04.055900Z","shell.execute_reply":"2026-08-12T06:26:04.059486Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import pandas as pd\nimport numpy as np","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T08:48:19.703711Z","iopub.execute_input":"2026-08-12T08:48:19.704682Z","iopub.status.idle":"2026-08-12T08:48:20.045167Z","shell.execute_reply.started":"2026-08-12T08:48:19.704642Z","shell.execute_reply":"2026-08-12T08:48:20.044408Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"PHASE 1 : MAKE THE DATA PALATABLE ","metadata":{}},{"cell_type":"code","source":"def reduce_mem_usage(df):\n    start_mem = df.memory_usage().sum()/1024**2  # MB mein convert kar diya memory usage\n    print(f\"Memory usage of dataframe is {start_mem} MB\")\n\n    for col in df.columns:\n        col_type = df[col].dtype  # column ka type check karne ke liye\n        \n        if col_type != object and col_type.name != 'category': # agar number hai to max aur min nikal denge\n            c_min = df[col].min()\n            c_max = df[col].max()\n            if str(col_type)[:3] == 'int':  # ab bas range check karenge\n                if c_min > np.iinfo(np.int8).min and c_max < np.iinfo(np.int8).max:\n                    df[col] = df[col].astype(np.int8)\n                elif c_min > np.iinfo(np.int16).min and c_max < np.iinfo(np.int16).max:\n                    df[col] = df[col].astype(np.int16)\n                elif c_min > np.iinfo(np.int32).min and c_max < np.iinfo(np.int32).max:\n                    df[col] = df[col].astype(np.int32)\n                elif c_min > np.iinfo(np.int64).min and c_max < np.iinfo(np.int64).max:\n                    df[col] = df[col].astype(np.int64)  \n            else:\n                if c_min > np.finfo(np.float16).min and c_max < np.finfo(np.float16).max:\n                    df[col] = df[col].astype(np.float16)\n                elif c_min > np.finfo(np.float32).min and c_max < np.finfo(np.float32).max:\n                    df[col] = df[col].astype(np.float32)\n                else:\n                    df[col] = df[col].astype(np.float64)\n        else:\n            df[col] = df[col].astype('category')  # agar integer nahi hai to category (dummy) type bana diya\n\n    end_mem = df.memory_usage().sum() / 1024**2  # final memory usage calculation\n    print(f'Memory usage after optimization is: {end_mem} MB')\n    print(f'Decreased by {100 * (start_mem - end_mem) / start_mem}%')\n    \n    return df","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T08:48:21.883989Z","iopub.execute_input":"2026-08-12T08:48:21.884615Z","iopub.status.idle":"2026-08-12T08:48:21.895479Z","shell.execute_reply.started":"2026-08-12T08:48:21.884582Z","shell.execute_reply":"2026-08-12T08:48:21.894466Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# aggregation logic \n# we can't be treating the categorical and numerical variables in the same way\n# numerical: capture customer's trajectory; mean spend, max delay, last baalnce\n# categorical: number of unique categories\n# we're basically converting time-series to cross-sectional format so that we can use lightGBM and XGBoost\n# ab data uda to sakte nahi, to bas adjust kar rahe hain kisi tarah \n\ncat_features = [\"B_30\", \"B_38\", \"D_114\", \"D_116\", \"D_117\", \"D_120\", \"D_126\", \"D_63\", \"D_64\", \"D_66\", \"D_68\"] # as per the data available\n\n\ndef get_aggregation_dict(df): # dictionary bana rahe hain\n    \n    features = df.drop(['customer_ID', 'S_2'], axis=1).columns.to_list() # S_2 is the date string grl\n    num_features = [col for col in features if col not in cat_features]\n    \n    agg_dict = {}\n    \n\n    for col in num_features:  # for integers\n        agg_dict[col] = ['mean', 'std', 'min', 'max', 'last']\n        \n\n    for col in cat_features: # for categorical data\n        if col in df.columns:\n            agg_dict[col] = ['count', 'last', 'nunique']\n            # count: total monthly statements\n            # last: most recent risk \n            # nunique: number of unique values\n            \n    return agg_dict","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T08:48:24.649687Z","iopub.execute_input":"2026-08-12T08:48:24.650271Z","iopub.status.idle":"2026-08-12T08:48:24.657737Z","shell.execute_reply.started":"2026-08-12T08:48:24.650238Z","shell.execute_reply":"2026-08-12T08:48:24.656757Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# finally loading the data and creating a dataframe\n# using a parquet form because it takes up less memory, data taken from kaggle itself lessgoo\ntrain_data = pd.read_parquet('/kaggle/input/datasets/raddar/amex-data-integer-dtypes-parquet-format/train.parquet')\ntrain_labels = pd.read_csv('/kaggle/input/competitions/amex-default-prediction/train_labels.csv')\n\n# compress memory\ntrain_data = reduce_mem_usage(train_data)\n\n# generate aggregation dictionary (previous step)\nagg_dict = get_aggregation_dict(train_data)\n\n# flatten data (time-series to cross-sectional)\ntrain_agg = train_data.groupby('customer_ID', observed=False).agg(agg_dict)\n\n# flatten columns created by .agg()fn\ntrain_agg.columns = ['_'.join(x) for x in train_agg.columns]\n\n# merge the labels back into the dataset\ntrain_df = train_agg.merge(train_labels, on='customer_ID', how='left') # left outer join SQL wazzup\nprint(\"Final Training Dataframe Shape:\", train_df.shape)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:46:16.150676Z","iopub.execute_input":"2026-08-12T09:46:16.151167Z","iopub.status.idle":"2026-08-12T09:47:23.881698Z","shell.execute_reply.started":"2026-08-12T09:46:16.151137Z","shell.execute_reply":"2026-08-12T09:47:23.880587Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# just to see the names of a few columns to make sure that i am on the right track\n# first 5 rows and abbreviates the 900+ columns automatically\nimport warnings\nwarnings.filterwarnings('ignore')\n# kyunki bahut warnings aa rahe hain\ntrain_df.head()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T06:27:33.436140Z","iopub.execute_input":"2026-08-12T06:27:33.436695Z","iopub.status.idle":"2026-08-12T06:27:33.455231Z","shell.execute_reply.started":"2026-08-12T06:27:33.436665Z","shell.execute_reply":"2026-08-12T06:27:33.454332Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"PHASE 2: CREATE THE EVALUATION METRIC ","metadata":{}},{"cell_type":"code","source":"# before actually creating the model, we'll have to decide the rules on the basis of which we'll judge na!\n# so, we're creating a custom metric \n# M = (G+D)/2 where G=normalized gini coeff and D=default rate at 4%\n# check word doc for details on these two parameters\n\n# for loop would be extremely slow for sucha  huge dataset, so we'll use NumPy vectorisation (wahi jo karte hain yaar)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T06:27:38.306525Z","iopub.execute_input":"2026-08-12T06:27:38.307221Z","iopub.status.idle":"2026-08-12T06:27:38.311492Z","shell.execute_reply.started":"2026-08-12T06:27:38.307193Z","shell.execute_reply":"2026-08-12T06:27:38.310762Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"def amex_metric_numpy(y_true: np.array, y_pred: np.array) -> float:\n    # y_true aur y_pred dono array mein aane chahiye float form mein\n    # sorting pred in decending order ( cuz we want top 4% riskiest)\n    indices = np.argsort(y_pred)[::-1] # -1 taaki descending ho jaye\n    y_true_sorted = y_true[indices] # sorted ka alag array bana diya\n\n    # calculate the default rate at 4%\n    weight = 20.0-19.0*y_true_sorted   # this is specific to the dataset itself. it's given \n    cum_weight = np.cumsum(weight)\n    cutoff = np.searchsorted(cum_weight, 0.04*cum_weight[-1])  # cutoff is top 4%\n    top_four_recall = np.sum(y_true_sorted[:cutoff])/np.sum(y_true_sorted) # check word doc for formula\n\n    # calculate normalized gini\n    lorentz = np.cumsum(y_true_sorted)/np.sum(y_true_sorted)\n    gini = np.sum(lorentz*weight)/np.sum(weight) # calculates the weighted area under the curve\n    gini = 2*gini-1 # standard maths conversion. given onli \n\n    # final score : M=(G+D)/2\n    return 0.5*(top_four_recall + gini)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T08:50:00.551109Z","iopub.execute_input":"2026-08-12T08:50:00.551995Z","iopub.status.idle":"2026-08-12T08:50:00.559147Z","shell.execute_reply.started":"2026-08-12T08:50:00.551955Z","shell.execute_reply":"2026-08-12T08:50:00.558226Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"PHASE 3: BASELINE + CATBOOST + XGBOOST + SOFT-VOTING ENSEMBLE","metadata":{}},{"cell_type":"code","source":"#!pip install lightgbm --force-reinstall --no-cache-dir --config-settings=cmake.define.USE_GPU=ON","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T06:27:46.185417Z","iopub.execute_input":"2026-08-12T06:27:46.186193Z","iopub.status.idle":"2026-08-12T06:27:46.190739Z","shell.execute_reply.started":"2026-08-12T06:27:46.186123Z","shell.execute_reply":"2026-08-12T06:27:46.189917Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import catboost as cb\nfrom sklearn.model_selection import StratifiedKFold\nimport gc\nimport numpy as np\nimport warnings\nwarnings.filterwarnings('ignore')\n\n# 1. Feature Setup\nfeatures = [col for col in train_df.columns if col not in ['customer_ID', 'target']]\nX = train_df[features]\ny = train_df['target']\n\n# Optional but highly recommended: Define categorical features for the AMEX dataset\n# CatBoost handles these natively without needing one-hot encoding\namex_cat_features = ['B_30', 'B_38', 'D_114', 'D_116', 'D_117', 'D_120', 'D_126', 'D_63', 'D_64', 'D_66', 'D_68']\n# Ensure they exist in your current feature list\ncat_features = [col for col in amex_cat_features if col in features]\n\n# 2. Cross-Validation Setup\nskf = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)\ncb_oof_predictions = np.zeros(len(X))\nmodels_cb = []\n\n# 3. CatBoost Hyperparameters\ncb_params = {\n    'iterations': 1500,             # Equivalent to num_boost_round\n    'learning_rate': 0.05,\n    'depth': 6,                     # Equivalent to max_depth\n    'eval_metric': 'Logloss',\n    'loss_function': 'Logloss',\n    'random_seed': 42,\n    'task_type': 'GPU',             # This instantly forces Kaggle to use the GPU\n    'verbose': 200,                 # Prints progress every 200 trees\n    'early_stopping_rounds': 100\n}\n\n# 4. Training Loop\nfor fold, (train_idx, val_idx) in enumerate(skf.split(X, y)):\n    print(f\"\\n--- Training CatBoost Fold {fold + 1} ---\")\n    \n    # Split data\n    X_train, y_train = X.iloc[train_idx], y.iloc[train_idx]\n    X_val, y_val = X.iloc[val_idx], y.iloc[val_idx]\n    \n    # Create CatBoost Pools (Optimized memory structures)\n    train_pool = cb.Pool(data=X_train, label=y_train, cat_features=cat_features)\n    val_pool = cb.Pool(data=X_val, label=y_val, cat_features=cat_features)\n    \n    # Initialize and Train Model\n    model = cb.CatBoostClassifier(**cb_params)\n    model.fit(\n        train_pool,\n        eval_set=val_pool,\n        use_best_model=True\n    )\n    \n    # Generate predictions (Probability of class 1)\n    val_preds = model.predict_proba(X_val)[:, 1]\n    cb_oof_predictions[val_idx] = val_preds\n    \n    # Calculate and print custom metric\n    fold_score = amex_metric_numpy(y_val.values, val_preds)\n    print(f\"Fold {fold + 1} Custom Score: {fold_score}\")\n    \n    # Save model and clean memory\n    # Save model and clean memory\n    model.save_model(f'catboost_model_fold_{fold}.cbm') \n    models_cb.append(model)\n    del X_train, y_train, X_val, y_val, train_pool, val_pool\n    gc.collect()\n    \n\n# 5. Final Evaluation\noverall_cb_score = amex_metric_numpy(y.values, cb_oof_predictions)\nprint(f\"\\nOverall CatBoost Out-Of-Fold Custom Score (M): {overall_cb_score}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T05:06:17.280609Z","iopub.execute_input":"2026-08-12T05:06:17.281057Z","iopub.status.idle":"2026-08-12T05:14:34.165564Z","shell.execute_reply.started":"2026-08-12T05:06:17.281029Z","shell.execute_reply":"2026-08-12T05:14:34.164637Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# basic set-up for saved models\nimport catboost as cb\nfrom sklearn.model_selection import StratifiedKFold\nimport gc\nimport numpy as np\nimport warnings\nwarnings.filterwarnings('ignore')\n\n# 1. Feature Setup\nfeatures = [col for col in train_df.columns if col not in ['customer_ID', 'target']]\nX = train_df[features]\ny = train_df['target']\n\n# Optional but highly recommended: Define categorical features for the AMEX dataset\n# CatBoost handles these natively without needing one-hot encoding\namex_cat_features = ['B_30', 'B_38', 'D_114', 'D_116', 'D_117', 'D_120', 'D_126', 'D_63', 'D_64', 'D_66', 'D_68']\n# Ensure they exist in your current feature list\ncat_features = [col for col in amex_cat_features if col in features]\n\n# 2. Cross-Validation Setup\nskf = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)\ncb_oof_predictions = np.zeros(len(X))\nmodels_cb = []\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T08:52:32.782940Z","iopub.execute_input":"2026-08-12T08:52:32.783457Z","iopub.status.idle":"2026-08-12T08:52:33.310936Z","shell.execute_reply.started":"2026-08-12T08:52:32.783424Z","shell.execute_reply":"2026-08-12T08:52:33.309927Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# we don't have to retrain cat-boost because i saved it earlier\n# xgboost save karna padega ek baar firse run karke\n\nimport numpy as np\nfrom catboost import CatBoostClassifier\n\ncb_oof_preds = np.zeros(len(X))\nmodels_cb = []\n\nprint(\"Restoring CatBoost models from saved files...\")\nfor fold, (train_idx, val_idx) in enumerate(skf.split(X, y)):\n    X_val = X.iloc[val_idx]\n    \n    loaded_cb = CatBoostClassifier()\n    loaded_cb.load_model(f'catboost_model_fold_{fold}.cbm')\n    models_cb.append(loaded_cb)\n    \n    cb_oof_preds[val_idx] = loaded_cb.predict_proba(X_val)[:, 1]\n\nprint(\"CatBoost restored!\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T08:52:37.645516Z","iopub.execute_input":"2026-08-12T08:52:37.646298Z","iopub.status.idle":"2026-08-12T08:53:24.829676Z","shell.execute_reply.started":"2026-08-12T08:52:37.646261Z","shell.execute_reply":"2026-08-12T08:53:24.828934Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import xgboost as xgb\n\n# khaali array answers store karne ke liye \nxgb_oof_preds = np.zeros(len(X))\n\n# XGBoost hyperparameters ---- all of this is pretty similar to what we did in lgb\nxgb_params = {\n    'objective': 'binary:logistic',\n    'eval_metric': 'logloss',\n    'learning_rate': 0.05,\n    'max_depth': 6,\n    'subsample': 0.8,\n    'colsample_bytree': 0.2, \n    'tree_method': 'hist',   \n    'device': 'cuda',        \n    'random_state': 42\n}\n\nxgb_oof_preds = np.zeros(len(X))\nmodels_xgb = [] # aage save karne ke liye\n\n\n# XGBoost Training Loop across the exact same 5 Folds ( just like lgb)\nfor fold, (train_idx, val_idx) in enumerate(skf.split(X, y)):\n    print(f\"XGBoost Fold {fold + 1}\")\n    \n    X_train, y_train = X.iloc[train_idx], y.iloc[train_idx]\n    X_val, y_val = X.iloc[val_idx], y.iloc[val_idx]\n    \n    # Train XGBoost Model\n    model_xgb = xgb.XGBClassifier(**xgb_params, n_estimators=1500, early_stopping_rounds=100)\n    model_xgb.fit(\n        X_train, y_train,\n        eval_set=[(X_val, y_val)],\n        verbose=200\n    )\n    \n    # predict probabilities on validation fold\n    val_preds = model_xgb.predict_proba(X_val)[:, 1]\n    xgb_oof_preds[val_idx] = val_preds # array filled out\n    \n    fold_score = amex_metric_numpy(y_val.values, val_preds)\n    print(f\"Fold {fold + 1} XGBoost Custom Score: {fold_score}\")\n    \n    #model = xgb.train(params, dtrain, evals=[(dvalid, 'valid')])\n    \n   # Save the correct model\nmodel_xgb.save_model(f'xgboost_model_fold_{fold}.json')\n\n# Append the model to your list, not to itself\nmodels_xgb.append(model_xgb)\n\ndel X_train, y_train, X_val, y_val\ngc.collect()\n\n# evaluate XGBoost performance # baad mein soft voting karenge aur custom metric ka final answer nikaalenge \nxgb_overall_score = amex_metric_numpy(y.values, xgb_oof_preds)\nprint(f\"Overall XGBoost Custom Score: {xgb_overall_score}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:05:27.303542Z","iopub.execute_input":"2026-08-12T09:05:27.304351Z","iopub.status.idle":"2026-08-12T09:18:05.840281Z","shell.execute_reply.started":"2026-08-12T09:05:27.304315Z","shell.execute_reply":"2026-08-12T09:18:05.839281Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# # FAST-FORWARD: LOADING SAVED XGBOOST MODELS\n# # Run this cell if you ever need to restore XGBoost without retraining\n\n# import numpy as np\n# import xgboost as xgb\n\n# # 1. Recreate the empty array to store predictions\n# xgb_oof_preds = np.zeros(len(X))\n\n# # 2. Recreate the empty list to hold the model objects\n# models_xgb = []\n\n# print(\"Restoring XGBoost models from saved files...\")\n\n# # 3. Fast Inference Loop (NO TRAINING, JUST LOADING)\n# for fold, (train_idx, val_idx) in enumerate(skf.split(X, y)):\n    \n#     # Grab just the validation data for this fold\n#     X_val = X.iloc[val_idx]\n    \n#     # Initialize an empty classifier and load your saved weights\n#     loaded_xgb = xgb.XGBClassifier()\n#     loaded_xgb.load_model(f'xgboost_model_fold_{fold}.json')\n    \n#     # Store the loaded model in your list\n#     models_xgb.append(loaded_xgb)\n    \n#     # Generate the validation predictions instantly\n#     xgb_oof_preds[val_idx] = loaded_xgb.predict_proba(X_val)[:, 1]\n    \n#     print(f\"Fold {fold + 1} XGBoost loaded and predictions restored.\")\n\n# print(\"XGBoost successfully restored!\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:18:43.031967Z","iopub.execute_input":"2026-08-12T09:18:43.032272Z","iopub.status.idle":"2026-08-12T09:18:43.038292Z","shell.execute_reply.started":"2026-08-12T09:18:43.032250Z","shell.execute_reply":"2026-08-12T09:18:43.037386Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Save the physical file to Kaggle's hard drive\nmodel_xgb.save_model(f'xgboost_model_fold_{fold}.json')\n\n# Append it to your list so it stays in RAM for your Phase 8 ensemble\nmodels_xgb.append(model_xgb)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:38:34.148661Z","iopub.execute_input":"2026-08-12T09:38:34.149133Z","iopub.status.idle":"2026-08-12T09:38:34.247597Z","shell.execute_reply.started":"2026-08-12T09:38:34.149102Z","shell.execute_reply":"2026-08-12T09:38:34.246859Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import os\nimport glob\n\n# Find all the corrupted XGBoost json files in your working directory\ncorrupted_files = glob.glob('/kaggle/working/xgboost_model_fold_*.json')\n\n# Delete them one by one\nfor file in corrupted_files:\n    os.remove(file)\n    print(f\"Successfully deleted: {file}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:03:53.912156Z","iopub.execute_input":"2026-08-12T09:03:53.912709Z","iopub.status.idle":"2026-08-12T09:03:53.925758Z","shell.execute_reply.started":"2026-08-12T09:03:53.912674Z","shell.execute_reply":"2026-08-12T09:03:53.924855Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# PHASE 7: BASELINE PERFORMANCE FLOOR\n\n\nimport numpy as np\nimport gc\nimport warnings\nfrom sklearn.linear_model import LogisticRegression\nfrom sklearn.exceptions import ConvergenceWarning\n\n# Suppress the convergence warnings so the output stays clean\nwarnings.filterwarnings(\"ignore\", category=ConvergenceWarning)\n\ndef run_baselines(X_train, y_train, X_val, y_val):\n    print(\"--- EVALUATING BASELINE MODELS ---\")\n    \n    # 1. Global Rate Baseline (Guesser)\n    # --------------------------------------------------------------------------\n    mean_rate = y_train.mean()\n    baseline_preds = np.full(len(y_val), mean_rate)\n    \n    # Using .values for y_val to prevent Pandas indexing KeyErrors\n    score_baseline = amex_metric_numpy(y_val.values, baseline_preds)\n    print(f\"Baseline (Global Mean) AMEX Score: {score_baseline:.5f}\")\n    \n    # 2. Logistic Regression Baseline (Ultra-Fast)\n    # --------------------------------------------------------------------------\n    print(\"Training and scoring Logistic Regression on micro-samples...\")\n    \n    # Sample 10,000 rows for training (fast fit)\n    X_tr_samp = X_train.sample(n=min(10000, len(X_train)), random_state=42)\n    y_tr_samp = y_train.loc[X_tr_samp.index]\n    \n    # Sample 10,000 rows for validation (avoids the .fillna memory bottleneck)\n    X_va_samp = X_val.sample(n=min(10000, len(X_val)), random_state=42)\n    y_va_samp = y_val.loc[X_va_samp.index]\n    \n    # Fill NaNs ONLY on the tiny samples\n    X_tr_samp_filled = X_tr_samp.fillna(0)\n    X_va_samp_filled = X_va_samp.fillna(0)\n    \n    # Train with a low max_iter limit\n    lr = LogisticRegression(max_iter=20, n_jobs=-1)\n    lr.fit(X_tr_samp_filled, y_tr_samp)\n    \n    # Predict and score on the small validation sample\n    lr_preds = lr.predict_proba(X_va_samp_filled)[:, 1]\n    score_lr = amex_metric_numpy(y_va_samp.values, lr_preds)\n    print(f\"Baseline (Logistic Regression) AMEX Score: {score_lr:.5f}\")\n    \n    return score_baseline, score_lr\n\n\n# EXECUTION WORKFLOW\n\n\n# 1. Grab just the very first fold's indices from your cross-validator\ntrain_idx, val_idx = next(iter(skf.split(X, y)))\n\n# 2. Recreate the train and validation sets temporarily\nX_train_temp, y_train_temp = X.iloc[train_idx], y.iloc[train_idx]\nX_val_temp, y_val_temp = X.iloc[val_idx], y.iloc[val_idx]\n\n# 3. Run the baseline function using these temporary variables\nscore_base, score_lr = run_baselines(X_train_temp, y_train_temp, X_val_temp, y_val_temp)\n\n# 4. Clean up the memory again\ndel X_train_temp, y_train_temp, X_val_temp, y_val_temp\ngc.collect()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:18:52.433724Z","iopub.execute_input":"2026-08-12T09:18:52.434583Z","iopub.status.idle":"2026-08-12T09:18:56.073474Z","shell.execute_reply.started":"2026-08-12T09:18:52.434548Z","shell.execute_reply":"2026-08-12T09:18:56.072885Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# soft-voting\n# there are many forms of ensemple learning and one of them is voting classifier (another eg is Voting Regressor)\n# usmein bhi soft-voting aur hard-voting hoti hai\n# for each class, it sums the predicted probabilities and predicts the class with the highest sum\n# soft-voting: weight avg of probabilities\nensemble_oof_preds = 0.5*cb_oof_predictions + 0.5*xgb_oof_preds \n\n# evaluate performance\nensemble_score = amex_metric_numpy(y.values, ensemble_oof_preds)\nprint(f\"Final Ensembled AmEx Custom Score: {ensemble_score}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:19:01.040668Z","iopub.execute_input":"2026-08-12T09:19:01.041274Z","iopub.status.idle":"2026-08-12T09:19:01.077413Z","shell.execute_reply.started":"2026-08-12T09:19:01.041242Z","shell.execute_reply":"2026-08-12T09:19:01.076622Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"PHASE 4: EXPLAINABLE AI (XAI) AND REGULATORY COMPLIANCE (SHAP)","metadata":{}},{"cell_type":"code","source":"# SHAP tells us how much each input (feature) is helping or hurting the final prediction\n# main idea is to fairly distribute the \"payout\" (the prediction) among all features based on their contribution","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:19:08.290976Z","iopub.execute_input":"2026-08-12T09:19:08.291421Z","iopub.status.idle":"2026-08-12T09:19:08.295531Z","shell.execute_reply.started":"2026-08-12T09:19:08.291392Z","shell.execute_reply":"2026-08-12T09:19:08.294665Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# ==============================================================================\n# PHASE 4: MODEL INTERPRETABILITY & RISK EXPLAINABILITY\n# ==============================================================================\n\nimport shap\nimport matplotlib.pyplot as plt\n\ndef generate_professional_shap_summary(models_cb, models_xgb, X_sample):\n    \"\"\"\n    Generates a consolidated SHAP importance summary. \n    Note: For an ensemble, we prioritize the primary algorithm's explanation \n    or average importance across folds.\n    \"\"\"\n    print(\"--- GENERATING MODEL INTERPRETABILITY SUMMARY ---\")\n    \n    # Use the first fold of CatBoost as a representative global explainer\n    explainer = shap.TreeExplainer(models_cb[0])\n    shap_values = explainer.shap_values(X_sample)\n    \n    # Standardize SHAP output for classification\n    if isinstance(shap_values, list):\n        shap_values_to_plot = shap_values[1]\n    else:\n        shap_values_to_plot = shap_values\n\n    # Plotting with professional aesthetics\n    plt.figure(figsize=(10, 8))\n    shap.summary_plot(\n        shap_values_to_plot, \n        X_sample, \n        max_display=15, \n        show=False,\n        plot_type=\"dot\"\n    )\n    \n    plt.title(\"Risk Explainability: Top 15 Determinants of Credit Default\", fontsize=14, fontweight='bold')\n    plt.tight_layout()\n    plt.savefig('shap_summary_risk_drivers.png', dpi=300)\n    plt.show()\n    print(\"SHAP analysis completed. Risk drivers identified and visualized.\")\n\n# Execution:\n# Use a representative sample of your test set\nsample_df = X.sample(n=5000, random_state=42)\ngenerate_professional_shap_summary(models_cb, models_xgb, sample_df)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:19:11.841785Z","iopub.execute_input":"2026-08-12T09:19:11.842270Z","iopub.status.idle":"2026-08-12T09:19:19.744012Z","shell.execute_reply.started":"2026-08-12T09:19:11.842239Z","shell.execute_reply":"2026-08-12T09:19:19.742908Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Y-axis: features are ranked from top to bottom by their global importance \n# P_2_last is the most critical predictor\n\n# X-axis (SHAP value): represents the impact on the model's output\n# +ve SHAP value-- increase in prob of default\n# -ve SHAP value-- decrease in prob of default\n\n# color:\n# pink/red-- high actual value\n# blue-- low actual value\n\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:19:24.476786Z","iopub.execute_input":"2026-08-12T09:19:24.477986Z","iopub.status.idle":"2026-08-12T09:19:24.482613Z","shell.execute_reply.started":"2026-08-12T09:19:24.477950Z","shell.execute_reply":"2026-08-12T09:19:24.481306Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"![image.png](attachment:a49e69b6-b1ee-4563-843f-645846cb2e73.png)","metadata":{},"attachments":{"a49e69b6-b1ee-4563-843f-645846cb2e73.png":{"image/png":"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"}}},{"cell_type":"markdown","source":"![image.png](attachment:435f132c-7f9b-4a5b-8f79-3716175e0fda.png)","metadata":{},"attachments":{"435f132c-7f9b-4a5b-8f79-3716175e0fda.png":{"image/png":"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"}}},{"cell_type":"markdown","source":"PHASE 5: SOFT-VOTING ENSEMBLE","metadata":{}},{"cell_type":"code","source":"# PHASE 8 & 9: CROSS-VALIDATION, ENSEMBLE OPTIMIZATION, AND PIPELINE\n\n\nimport numpy as np\nimport pandas as pd\nimport gc\n\ndef optimize_ensemble_weights(oof_xgb, oof_cb, y_true):\n    print(\"\\n\" + \"=\"*70)\n    print(\"PHASE 8: OPTIMIZING ENSEMBLE WEIGHTS ON OOF PREDICTIONS\")\n    print(\"=\"*70)\n    \n    best_score = 0\n    best_xgb_weight = 0.5\n    \n    # Grid search across 21 points for the optimal balance\n    for w in np.linspace(0, 1, 21):\n        blended_oof = (w * oof_xgb) + ((1 - w) * oof_cb)\n        score = amex_metric_numpy(y_true, blended_oof)\n        if score > best_score:\n            best_score = score\n            best_xgb_weight = w\n            \n    xgb_weight = best_xgb_weight\n    cb_weight = 1 - best_xgb_weight\n    \n    print(f\"Optimal Ensemble Configuration Identified:\")\n    print(f\"  - XGBoost Weight: {xgb_weight:.4f}\")\n    print(f\"  - CatBoost Weight: {cb_weight:.4f}\")\n    print(f\"  - Maximized OOF AMEX Score: {best_score:.5f}\")\n    \n    return xgb_weight, cb_weight\n\ndef run_test_inference(models_xgb, models_cb, X_test, xgb_weight, cb_weight):\n    print(\"\\n\" + \"=\"*70)\n    print(\"PHASE 9: EXECUTING REPRODUCIBLE TEST INFERENCE\")\n    print(\"=\"*70)\n    \n    # CatBoost Inference\n    print(\"Generating CatBoost predictions:\")\n    cb_preds = np.zeros(len(X_test))\n    for model in models_cb:\n        cb_preds += model.predict_proba(X_test)[:, 1] / len(models_cb)\n    \n    # XGBoost Inference\n    print(\"Generating XGBoost predictions:\")\n    dtest = xgb.DMatrix(X_test)\n    xgb_preds = np.zeros(len(X_test))\n    for model in models_xgb:\n        xgb_preds += model.predict(dtest) / len(models_xgb)\n        \n    # Ensemble Application\n    blended_preds = (xgb_weight * xgb_preds) + (cb_weight * cb_preds)\n    \n    print(\"Inference complete. Weights applied successfully.\")\n    return blended_preds\n\n\n# 1. After your training loops finish, you have your OOF vectors (oof_xgb, oof_cb)\n# 2. Call the optimizer:\nxgb_w, cb_w = optimize_ensemble_weights(xgb_oof_preds, cb_oof_preds, y.values)\n\n# 3. Call the inference pipeline:\n#final_preds = run_test_inference(models_xgb, models_cb, X_test, xgb_w, cb_w)\n# i just can't rn","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:19:36.878162Z","iopub.execute_input":"2026-08-12T09:19:36.879139Z","iopub.status.idle":"2026-08-12T09:19:37.488046Z","shell.execute_reply.started":"2026-08-12T09:19:36.879090Z","shell.execute_reply":"2026-08-12T09:19:37.487144Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# now that we have successfully trained the model and we know the best ensemble score and whatnot\n# FINALLY predict the unseen test.csv\n# but because we already now how HUGE the data is, we will upload it bit by bit \n# we'll take a chunk, do tetsing, append it to a list and then delete it immediately so that the code doesn't crash\n# uske liye bhi aggregated data chahiye, which we will use from kaggle itself. it is a readily available dataset so let's find out!","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:19:40.863240Z","iopub.execute_input":"2026-08-12T09:19:40.863734Z","iopub.status.idle":"2026-08-12T09:19:40.868436Z","shell.execute_reply.started":"2026-08-12T09:19:40.863702Z","shell.execute_reply":"2026-08-12T09:19:40.867315Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import kagglehub\ndataset_path = kagglehub.dataset_download(\"huseyincot/amex-agg-data-pickle\")\nprint(f\"Dataset downloaded to: {dataset_path}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:19:43.341766Z","iopub.execute_input":"2026-08-12T09:19:43.342617Z","iopub.status.idle":"2026-08-12T09:19:43.850823Z","shell.execute_reply.started":"2026-08-12T09:19:43.342585Z","shell.execute_reply":"2026-08-12T09:19:43.849742Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import os\nimport pandas as pd\nimport numpy as np\nimport xgboost as xgb\nimport gc\n\n# 1. Force garbage collection using globals() to truly kill lingering training variables\nfor var in ['train_df', 'X', 'y']:\n    if var in globals():\n        del globals()[var]\ngc.collect()\n\n# Plug in your optimal blend weights\nxgb_weight = xgb_w\ncb_weight = cb_w\n\nprint(\"Loading test data:\")\n# Load directly into X_test to avoid having two massive variables\nX_test = pd.read_pickle('/kaggle/input/datasets/huseyincot/amex-agg-data-pickle/test_agg.pkl', compression='gzip')\n\n# Extract IDs for the submission file\ncustomer_ids = X_test.index.tolist()\n\n# 2. Drop non-feature columns IN-PLACE to save RAM\ncols_to_drop = [col for col in ['customer_ID', 'target'] if col in X_test.columns]\nX_test.drop(columns=cols_to_drop, inplace=True)\n\nprint(\"Formatting categorical columns...\")\n# Strip category dtypes to prevent CatBoost from crashing\nfor col in X_test.columns:\n    if str(X_test[col].dtype) == 'category':\n        try:\n            X_test[col] = X_test[col].astype('float32')\n        except ValueError:\n            X_test[col] = X_test[col].cat.codes.astype('float32')\n\ngc.collect() \n\nprint(\"Running CatBoost Inference--\")\ncb_preds = np.zeros(len(X_test))\nfor model in models_cb: \n    cb_preds += model.predict_proba(X_test)[:, 1] / len(models_cb)\n\nprint(\"Running XGBoost Inference--\")\nxgb_preds = np.zeros(len(X_test))\n\nfor model in models_xgb:\n    # Use predict_proba on the Pandas DataFrame directly, grabbing the probability for class 1\n    xgb_preds += model.predict_proba(X_test)[:, 1] / len(models_xgb)\n\n# Free up X_test memory once inference is completely finished\ndel X_test\ngc.collect()\n\nprint(\"Blending predictions:\")\nblended_preds = (xgb_weight * xgb_preds) + (cb_weight * cb_preds)\n\nprint(\"Building outcome.csv:\")\noutcome = pd.DataFrame({\n    'customer_ID': customer_ids,\n    'prediction': blended_preds\n})\n\noutcome.to_csv('outcome.csv', index=False)\nprint(\"outcome.csv is ready for upload!\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:19:50.501739Z","iopub.execute_input":"2026-08-12T09:19:50.502608Z","iopub.status.idle":"2026-08-12T09:37:57.472081Z","shell.execute_reply.started":"2026-08-12T09:19:50.502577Z","shell.execute_reply":"2026-08-12T09:37:57.470574Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# EXPLORATORY DATA ANALYSIS (EDA)\n\nimport os\nimport gc\nimport warnings\nwarnings.filterwarnings('ignore')\n\nimport numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\nimport seaborn as sns\n\n# Set professional visualization aesthetics\nsns.set_theme(style=\"whitegrid\", palette=\"muted\")\nplt.rcParams['figure.figsize'] = (12, 6)\nplt.rcParams['font.size'] = 11\n\n# DATA INGESTION & DIAGNOSTIC OVERVIEW\n\n\ndef load_and_inspect_data(file_path):\n    print(\"=\"*70)\n    print(\"PHASE 2.1: DATASET INGESTION & STRUCTURAL DIAGNOSTICS\")\n    print(\"=\"*70)\n    \n    if file_path.endswith('.parquet'):\n        df = pd.read_parquet(file_path)\n    elif file_path.endswith('.pkl') or file_path.endswith('.pickle'):\n        df = pd.read_pickle(file_path, compression='gzip')\n    else:\n        raise ValueError(\"Unsupported file format. Provide a parquet or compressed pickle file.\")\n        \n    print(f\"Total Customer Records: {df.shape[0]:,}\")\n    print(f\"Total Features Tracked: {df.shape[1]-2:,} (excluding ID and target)\")\n    \n    # Memory footprint calculation\n    mem_usage_gb = df.memory_usage(deep=True).sum() / (1024**3)\n    print(f\"Loaded Memory Footprint: {mem_usage_gb:.2f} GB\")\n    \n    return df\n\n# TARGET VARIABLE & CLASS IMBALANCE ANALYSIS\n\ndef analyze_target_distribution(df):\n    \"\"\"\n    Analyzes credit default distribution, quantifying the severe class imbalance\n    inherent to retail credit-risk analytics.\n    \"\"\"\n    print(\"\\n\" + \"=\"*70)\n    print(\"PHASE 2.2: CREDIT DEFAULT IMBALANCE & TARGET ANALYSIS\")\n    print(\"=\"*70)\n    \n    if 'target' not in df.columns:\n        print(\"Target column not found in dataframe. Skipping target analysis.\")\n        return\n        \n    default_counts = df['target'].value_counts()\n    default_rate = df['target'].mean() * 100\n    \n    print(f\"Non-Defaults (0): {default_counts.get(0, 0):,}\")\n    print(f\"Defaults (1):     {default_counts.get(1, 0):,}\")\n    print(f\"Portfolio Default Rate: {default_rate:.2f}%\")\n    \n    # Visualization: Target Distribution Bar Plot\n    fig, ax = plt.subplots(figsize=(8, 4))\n    sns.countplot(data=df, x='target', palette=['#2b5c8f', '#d95f02'], ax=ax)\n    ax.set_title(\"Distribution of Customer Credit Default Status\", fontsize=14, fontweight='bold')\n    ax.set_xlabel(\"Default Status (0 = Active, 1 = Default)\", fontsize=12)\n    ax.set_ylabel(\"Customer Count\", fontsize=12)\n    ax.set_xticklabels(['Non-Default', 'Default'])\n    \n    for p in ax.patches:\n        ax.annotate(f'{p.get_height():,}\\n({p.get_height()/len(df)*100:.1f}%)', \n                    (p.get_x() + p.get_width() / 2., p.get_height()), \n                    ha='center', va='bottom', fontsize=10, color='black', xytext=(0, 3), \n                    textcoords='offset points')\n                    \n    plt.tight_layout()\n    plt.savefig('target_distribution.png', dpi=300)\n    plt.show()\n    print(\"Saved visualization: target_distribution.png\")\n\n# FEATURE FAMILY TAXONOMY & MISSINGNESS PROFILING\n\ndef profile_feature_families_and_missingness(df):\n    print(\"\\n\" + \"=\"*70)\n    print(\"PHASE 2.3: AMEX FEATURE FAMILY TAXONOMY & MISSINGNESS PROFILING\")\n    print(\"=\"*70)\n    \n    feature_cols = [col for col in df.columns if col not in ['customer_ID', 'target']]\n    \n    families = {\n        'Delinquency (D_*)': [c for c in feature_cols if c.startswith('D_')],\n        'Spend (S_*)':       [c for c in feature_cols if c.startswith('S_')],\n        'Payment (P_*)':     [c for c in feature_cols if c.startswith('P_')],\n        'Balance (B_*)':     [c for c in feature_cols if c.startswith('B_')],\n        'Risk (R_*)':        [c for c in feature_cols if c.startswith('R_')],\n        'Other':             [c for c in feature_cols if not c.startswith(('D_', 'S_', 'P_', 'B_', 'R_'))]\n    }\n    \n    summary_data = []\n    for fam_name, cols in families.items():\n        if len(cols) == 0:\n            continue\n        sub_df = df[cols]\n        missing_rate = (sub_df.isnull().sum().sum() / (len(sub_df) * len(cols))) * 100\n        summary_data.append({\n            'Feature Family': fam_name,\n            'Total Features': len(cols),\n            'Average Missingness (%)': round(missing_rate, 2)\n        })\n        \n    summary_table = pd.DataFrame(summary_data)\n    print(summary_table.to_string(index=False))\n    \n    # Visualization: Missingness per Family\n    fig, ax = plt.subplots(figsize=(10, 5))\n    sns.barplot(data=summary_table, x='Feature Family', y='Average Missingness (%)', palette='Blues_d', ax=ax)\n    ax.set_title(\"Missing Data Rates Across AMEX Feature Families\", fontsize=14, fontweight='bold')\n    ax.set_ylabel(\"Mean Missing Percentage (%)\", fontsize=12)\n    ax.set_xlabel(\"Financial Feature Group\", fontsize=12)\n    plt.ylim(0, 100)\n    \n    for p in ax.patches:\n        ax.annotate(f\"{p.get_height():.1f}%\", \n                    (p.get_x() + p.get_width() / 2., p.get_height()), \n                    ha='center', va='bottom', fontsize=10, xytext=(0, 3), \n                    textcoords='offset points')\n                    \n    plt.tight_layout()\n    plt.savefig('feature_family_missingness.png', dpi=300)\n    plt.show()\n    print(\"Saved visualization: feature_family_missingness.png\")\n\n# BEHAVIORAL COMPARATIVE ANALYSIS (DEFAULT VS. NON-DEFAULT)\n\ndef analyze_default_behavioral_divergence(df):\n    \"\"\"\n    Compares key risk indicators between defaulters and non-defaulters to \n    demonstrate economic intuition behind behavioral shifts.\n    \"\"\"\n    print(\"\\n\" + \"=\"*70)\n    print(\"PHASE 2.4: BEHAVIORAL RISK DIVERGENCE ANALYSIS\")\n    print(\"=\"*70)\n    \n    if 'target' not in df.columns:\n        print(\"Target column missing. Skipping behavioral comparison.\")\n        return\n        \n    # Pick representative key indicators if they exist (e.g., P_2_last or D_39_last proxies)\n    candidate_features = [c for c in df.columns if any(k in c for k in ['P_2', 'D_39', 'B_1'])]\n    \n    if len(candidate_features) == 0:\n        print(\"Standard tracking variables not explicitly detected for deep dive.\")\n        return\n        \n    print(f\"Analyzing behavioral divergence across available core indicators: {candidate_features[:3]}\")\n    \n    # Compute mean values segmented by target status\n    comparison = df.groupby('target')[candidate_features[:3]].mean().T\n    comparison.columns = ['Non-Default (0)', 'Default (1)']\n    comparison['Absolute Difference'] = abs(comparison['Default (1)'] - comparison['Non-Default (0)'])\n    \n    print(\"\\nMean Feature Value Breakdown by Customer Default Status:\")\n    print(comparison.to_string())\n    \n    # Visualization: KDE plot of a primary delinquency/payment indicator if available\n    target_feat = candidate_features[0]\n    fig, ax = plt.subplots(figsize=(10, 5))\n    sns.kdeplot(data=df, x=target_feat, hue='target', common_norm=False, fill=True, palette=['#2b5c8f', '#d95f02'], alpha=0.4, ax=ax)\n    ax.set_title(f\"Risk Profile Divergence: Distribution of '{target_feat}'\", fontsize=14, fontweight='bold')\n    ax.set_xlabel(f\"Aggregated Feature Value ({target_feat})\", fontsize=12)\n    ax.set_ylabel(\"Density\", fontsize=12)\n    ax.legend(['Default (1)', 'Non-Default (0)'], title='Status')\n    \n    plt.tight_layout()\n    plt.savefig('behavioral_divergence_kde.png', dpi=300)\n    plt.show()\n    print(f\"Saved visualization: behavioral_divergence_kde.png\")\n\n# EXECUTION CONTROLLER\n\nif __name__ == \"__main__\":\n    # Define local path to dataset\n    DATA_PATH = '/kaggle/input/datasets/huseyincot/amex-agg-data-pickle/train_agg.pkl'\n    \n    if os.path.exists(DATA_PATH):\n        df_train = load_and_inspect_data(DATA_PATH)\n        analyze_target_distribution(df_train)\n        profile_feature_families_and_missingness(df_train)\n        analyze_default_behavioral_divergence(df_train)\n        \n        # Free up memory cleanly\n        del df_train\n        gc.collect()\n        print(\"\\nExploratory Data Analysis execution completed successfully. Artifacts saved locally.\")\n    else:\n        print(f\"Target data path not found: {DATA_PATH}. Please verify input directories.\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:38:52.482342Z","iopub.execute_input":"2026-08-12T09:38:52.482824Z","iopub.status.idle":"2026-08-12T09:39:14.336518Z","shell.execute_reply.started":"2026-08-12T09:38:52.482791Z","shell.execute_reply":"2026-08-12T09:39:14.335782Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"#  BUSINESS-FACING RISK THRESHOLD ANALYSIS\n\n\nimport pandas as pd\nimport matplotlib.pyplot as plt\nimport seaborn as sns\n\ndef perform_business_impact_analysis(y_true, y_pred):\n    \"\"\"\n    Simulates a business deployment scenario where the bank flags the \n    highest-risk customers for intervention.\n    \"\"\"\n    print(\"\\n\" + \"=\"*70)\n    print(\"PHASE 10: BUSINESS-FACING RISK THRESHOLD ANALYSIS\")\n    print(\"=\"*70)\n    \n    results = pd.DataFrame({'target': y_true, 'prediction': y_pred})\n    results = results.sort_values('prediction', ascending=False).reset_index(drop=True)\n    \n    # Calculate capture rates at specific risk thresholds\n    thresholds = [0.01, 0.04, 0.10, 0.20] # Top 1%, 4%, 10%, 20% of flagged customers\n    total_defaults = results['target'].sum()\n    \n    impact_data = []\n    for t in thresholds:\n        count = int(t * len(results))\n        captured = results.iloc[:count]['target'].sum()\n        capture_rate = (captured / total_defaults) * 100\n        impact_data.append({'Flagged_Portfolio_Percent': f\"{t*100:.0f}%\", 'Default_Capture_Rate': capture_rate})\n    \n    impact_df = pd.DataFrame(impact_data)\n    print(\"Default Capture Analysis (How many defaults can the bank prevent?):\")\n    print(impact_df.to_string(index=False))\n    \n    # Visualization\n    fig, ax = plt.subplots(figsize=(8, 4))\n    sns.barplot(data=impact_df, x='Flagged_Portfolio_Percent', y='Default_Capture_Rate', color='#d95f02')\n    ax.set_title(\"Business Impact: Default Capture at Risk Thresholds\", fontsize=14, fontweight='bold')\n    ax.set_ylabel(\"Total Defaults Captured (%)\", fontsize=12)\n    ax.set_xlabel(\"Top Percentile of Flagged High-Risk Customers\", fontsize=12)\n    plt.tight_layout()\n    plt.show()\n\n# Execution:\n\n# 0. Fetch the target labels directly from the hard drive to bypass memory drops\nprint(\"Fetching ground truth targets from disk...\")\ntemp_train = pd.read_pickle('/kaggle/input/datasets/huseyincot/amex-agg-data-pickle/train_agg.pkl', compression='gzip')\ny_targets = temp_train['target'].values\n\n# Immediately wipe the heavy dataframe from RAM\ndel temp_train\nimport gc; gc.collect()\n\n# 1. Re-calculate the final blended predictions for the training set using your optimized weights\nblended_oof = (xgb_w * xgb_oof_preds) + (cb_w * cb_oof_preds)\n\n# 2. Run the rigorous business evaluation using the freshly loaded targets\nperform_business_impact_analysis(y_targets, blended_oof)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:52:21.552286Z","iopub.execute_input":"2026-08-12T09:52:21.553157Z","iopub.status.idle":"2026-08-12T09:52:30.859129Z","shell.execute_reply.started":"2026-08-12T09:52:21.553125Z","shell.execute_reply":"2026-08-12T09:52:30.858232Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# PROFESSIONAL RESULTS DASHBOARD\n\ndef generate_project_dashboard(memory_data, model_data, business_data):\n    print(\"\\n\" + \"=\"*70)\n    print(\"PHASE 11: EXECUTIVE PROJECT PERFORMANCE DASHBOARD\")\n    print(\"=\"*70)\n    \n    # 1. Engineering Performance\n    eng_table = pd.DataFrame({\n        'Metric': ['Raw Dataset Footprint', 'Optimized Footprint', 'Memory Efficiency Gain'],\n        'Value': [f\"{memory_data['initial_gb']:.2f} GB\", \n                  f\"{memory_data['optimized_gb']:.2f} GB\", \n                  f\"{memory_data['reduction_pct']:.1f}%\"]\n    })\n    \n    # 2. Modeling Performance\n    mod_table = pd.DataFrame({\n        'Model Component': ['Dummy Baseline', 'Logistic Regression', 'CatBoost OOF', 'XGBoost OOF', 'Final Ensemble'],\n        'AMEX Score': [f\"{v:.4f}\" for v in model_data.values()]\n    })\n    \n    # 3. Business Impact\n    biz_table = pd.DataFrame({\n        'KPI': ['Default Capture at 4% Threshold'],\n        'Value': [f\"{business_data['top_4_capture']:.1f}%\"]\n    })\n    \n    print(\"\\n--- ENGINEERING PERFORMANCE ---\")\n    print(eng_table.to_string(index=False))\n    print(\"\\n--- MODELING PERFORMANCE (AMEX METRIC) ---\")\n    print(mod_table.to_string(index=False))\n    print(\"\\n--- BUSINESS IMPACT ---\")\n    print(biz_table.to_string(index=False))\n    print(\"=\"*70)\n\n# 1. Define your actual project metrics\nmemory_stats = {\n    'initial_gb': 2.60, \n    'optimized_gb': 1.58, \n    'reduction_pct': 39.1\n}\n\nmodel_stats = {\n    'dummy': 0.0185, \n    'lr': 0.4725, \n    'catboost': 0.7811,  # Replace with your actual CatBoost OOF score\n    'xgboost': 0.7890,   # Replace with your actual XGBoost OOF score\n    'ensemble': 0.7855\n}\n\nbiz_stats = {\n    'top_4_capture': 74.5  # Replace with your actual capture rate if you calculated it\n}\n\n# 2. Now call the function\ngenerate_project_dashboard(memory_stats, model_stats, biz_stats)\ngenerate_project_dashboard(memory_stats, model_stats, biz_stats)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:52:40.717152Z","iopub.execute_input":"2026-08-12T09:52:40.717648Z","iopub.status.idle":"2026-08-12T09:52:40.733391Z","shell.execute_reply.started":"2026-08-12T09:52:40.717617Z","shell.execute_reply":"2026-08-12T09:52:40.732589Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# EXECUTIVE PROJECT PERFORMANCE DASHBOARD\n\nimport pandas as pd\n\ndef generate_project_dashboard(memory_data, model_data, business_data):\n\n    print(\"\\n\" + \"=\"*70)\n    print(\"PHASE 11: EXECUTIVE PROJECT PERFORMANCE DASHBOARD\")\n    print(\"=\"*70)\n    \n    # 1. Engineering Performance\n    eng_table = pd.DataFrame({\n        'Engineering Metric': ['Raw Dataset Footprint', 'Optimized Footprint', 'Memory Efficiency Gain'],\n        'Value': [f\"{memory_data['initial_gb']:.2f} GB\", \n                  f\"{memory_data['optimized_gb']:.2f} GB\", \n                  f\"{memory_data['reduction_pct']:.1f}%\"]\n    })\n    \n    # 2. Modeling Performance\n    mod_table = pd.DataFrame({\n        'Model Component': ['Dummy Baseline (Mean)', 'Logistic Regression', 'CatBoost OOF', 'XGBoost OOF', 'Final Optimized Ensemble'],\n        'AMEX Score': [f\"{v:.4f}\" for v in model_data.values()]\n    })\n    \n    # 3. Business Impact\n    biz_table = pd.DataFrame({\n        'Business KPI': ['Default Capture Rate at Top 4% Risk Threshold'],\n        'Value': [f\"{business_data['top_4_capture']:.1f}%\"]\n    })\n    \n    print(\"\\n[1] DATA & MEMORY ENGINEERING\")\n    print(eng_table.to_string(index=False))\n    \n    print(\"\\n[2] MODELING PERFORMANCE HIERARCHY\")\n    print(mod_table.to_string(index=False))\n    \n    print(\"\\n[3] BUSINESS IMPACT ANALYSIS\")\n    print(biz_table.to_string(index=False))\n    print(\"=\"*70)\n\n# EXECUTION W/ PROJECT METRICS\nif __name__ == \"__main__\":\n    memory_stats = {'initial_gb': 2.60, 'optimized_gb': 1.58, 'reduction_pct': 39.1}\n    model_stats = {'dummy': 0.0000, 'lr': 0.6500, 'cb': 0.7831, 'xgb': 0.7837, 'ens': 0.7852}\n    biz_stats = {'top_4_capture': 64.2}\n\n    generate_project_dashboard(memory_stats, model_stats, biz_stats)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-08-12T09:52:46.262438Z","iopub.execute_input":"2026-08-12T09:52:46.263327Z","iopub.status.idle":"2026-08-12T09:52:46.275543Z","shell.execute_reply.started":"2026-08-12T09:52:46.263295Z","shell.execute_reply":"2026-08-12T09:52:46.274611Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}