为什么我得到 AttributeError: 'numpy.ndarray' object has no attribute 'encode' 错误

Question

这是我的代码，function get_labels_for_msg 是从另一个 py 文件中收集的。

def bayes_clas(mean,alpha,beta,t):
    l = []
    m = (mean+(alpha*beta))/2
    for i in range(t.shape[0]):
        if train[1][i] >= m:
            l.append(1)
        else:
            l.append(0)
    return np.array(l)

def _int2bytes(i):
    hex_string = '%x' % i
    n = len(hex_string)
    return binascii.unhexlify(hex_string.zfill(n + (n & 1)))

def _text_from_bits(bits, encoding='ascii', errors='surrogatepass'):
    n = int(bits, 2)
    return _int2bytes(n).decode(encoding, errors)

def get_msg_for_labels(l):
    """
    Convert a one-dimensional label array to a message-string.

    Args:
        l (1D array): Predicted (binary) labels which should be converted to a
                      message. The first eight elements are assumed to be the
                      8-bit ASCII representation of the first letter in the
                      message, the next eight elements are assumed to represent
                      the next letter, and so on. len(l) must be a multiple of 8
    Returns:
        msg (string): Decoded message.
    """
    st = "".join([str(i) for i in l])
    return _text_from_bits(st)

test = pd.read_csv("/Users/leanderkirkeland/MLF/A1/optdigits-1d-test.csv")
bt = bayes_clas(m,9,b,test)


print(get_labels_for_msg(bt))

我得到错误AttributeError: 'numpy.ndarray' object has no attribute 'encode' 。 这是我在此页面上的第一个问题，所以不知道该怎么做