I am training a ConvLstm Model with the sequence size of (20, 230, 230, 3), which means there are 20 images per sequence and each size 230 * 230 * 3, however, when I tried t