training_config.json

{
  "model_name": "CBC Credit-Card Fraud Classifier",
  "hf_repo": "careerbytecode/mlops-ref-finance-fraud",
  "task": "binary classification (credit-card fraud detection, imbalanced 0.17%)",
  "model_type": "XGBoost (Optuna-tuned, scale_pos_weight), time-aware split",
  "framework": "xgboost",
  "serialization": "joblib",
  "loader": "joblib.load -> XGBClassifier; call .predict_proba(DataFrame[FEATURES])[:, 1], flag if >= recommended_threshold",
  "random_state": 42,
  "scale_pos_weight": 498.3863,
  "best_params": {
    "n_estimators": 400,
    "learning_rate": 0.1636812394286703,
    "max_depth": 4,
    "min_child_weight": 5,
    "subsample": 0.6587788742440184
  },
  "feature_columns": [
    "V1",
    "V2",
    "V3",
    "V4",
    "V5",
    "V6",
    "V7",
    "V8",
    "V9",
    "V10",
    "V11",
    "V12",
    "V13",
    "V14",
    "V15",
    "V16",
    "V17",
    "V18",
    "V19",
    "V20",
    "V21",
    "V22",
    "V23",
    "V24",
    "V25",
    "V26",
    "V27",
    "V28",
    "Amount"
  ],
  "split": {
    "trainfit": 182276,
    "val": 45569,
    "test": 56962,
    "method": "time-aware (latest 20% by Time = test)"
  },
  "dataset": "ULB mlg-ulb credit-card fraud (284,807 tx, 492 frauds), ODbL/DbCL",
  "python_version": "3.14.4",
  "library_versions": {
    "scikit-learn": "1.8.0",
    "xgboost": "3.2.0",
    "numpy": "2.4.6",
    "pandas": "2.3.3",
    "joblib": "1.5.3"
  },
  "training_date": "2026-06-04T20:25:08.260852+00:00"
}