data_io.py 文件源码

def load_data(path, dense=False):
    """Load data from a CSV or libsvm format file.

    Args:
        path (str): A path to the CSV or libsvm format file containing data.
        dense (boolean): An optional variable indicating if the return matrix
                         should be dense.  By default, it is false.
    """

    with open(path, 'r') as f:
        line = f.readline().strip()

    if ':' in line:
        X, y = load_svmlight_file(path)
        X = X.astype(np.float32)
        if dense:
            X = X.todense()
    elif ',' in line:
        X = np.loadtxt(path, delimiter=',',
                       skiprows=0 if is_number(line.split(',')[0]) else 1)
        y = X[:, 0]
        X = X[:, 1:]
    else:
        raise NotImplementedError, "Neither CSV nor LibSVM formatted file."

    return X, y