Created by: maoyun95

请问改怎样将图像矩阵转换成LOD_Tensor序列数据呢，我用的minist数据集。图像数据定义如下x = fluid.layers.data(name='x', shape=[1, 28, 28], dtype='float32', lod_level=1) 出现如下报错：Error: the provided recursive_sequence_lengths info is invalid [Hint: Expected CheckLoD(new_offset_lod, vectorize(self.dims()).front()) == true, but received CheckLoD(new_offset_lod, vectorize(self.dims()).front()):0 != true:1.] at (D:\1.8.3\paddle\paddle\fluid\pybind\pybind.cc:723)