def transform()

in src/mlmax/preprocessing.py [0:0]

• 13 lines of code
• 2 McCabe index (conditional complexity)

def transform(df, args, preprocess=None):
    if preprocess is None:
        model_directory = os.path.join(args.data_dir, "model")
        print(f"Reading model from {model_directory}")
        with tarfile.open(
            os.path.join(model_directory, "proc_model.tar.gz"), mode="r:gz"
        ) as archive:
            print(f"Exctracting tarfile to {model_directory}")
            archive.extractall(path=model_directory)
        preprocess = joblib.load(os.path.join(model_directory, "model.joblib"))
    features = preprocess.transform(df)
    print(f"Data shape after preprocessing: {features.shape}")
    return features