data/metadata.json

{
    "random_seed": 14,
    "test_size": 0.2,
    "target_feature": "stroke",
    "original_features": [
        "gender",
        "age",
        "hypertension",
        "heart_disease",
        "ever_married",
        "work_type",
        "Residence_type",
        "avg_glucose_level",
        "bmi",
        "smoking_status"
    ],
    "binary_encoding": {
        "columns": [
            "ever_married"
        ],
        "mapping": {
            "No": 0,
            "Yes": 1
        }
    },
    "imputation": {
        "method": "SimpleImputer",
        "strategy": "median",
        "columns": [
            "bmi"
        ]
    },
    "binning": {
        "method": "pandas.cut",
        "columns": [
            "age",
            "avg_glucose_level",
            "bmi"
        ]
    },
    "encoding": {
        "method": "OrdinalEncoder",
        "columns": [
            "gender",
            "work_type",
            "residence_type",
            "smoking_status",
            "age",
            "bmi",
            "avg_glucose_level"
        ],
        "verbose_feature_names": false
    },
    "over_sampling": {
        "method": "SMOTE",
        "note": "Applied only to train data."
    },
    "processed_features": [
        "stroke",
        "ever_married",
        "heart_disease",
        "hypertension",
        "avg_glucose_level",
        "bmi",
        "age",
        "smoking_status",
        "residence_type",
        "work_type",
        "gender"
    ],
    "artifacts_dir": "data/artifacts",
    "processed_data_dir": "data/processed",
    "train_shape": [
        7778,
        11
    ],
    "test_shape": [
        1022,
        11
    ],
    "timestamp": "2026-05-19T02:47:52.834577"
}