shape = x.get_shape().as_list()# a list: [None, 9, 2]. You can proceed it easily with tf.reshape() without knowing the batch size.
確定！ 回上一頁
