The problem of disorganized data output by mnn model

I have used Ali's for the past two daysMNN：
https://github.com/alibaba/MNN

还挺好用的,Just do not know withopenclHow to use the backend enginePython API调用

---

I encountered a small hole,The output data is so disorganized：

而不是这样的：

反复debug,发现：

It turns out that the output of the model cannot be directly getData() 

output_tensor = interpreter.getSessionOutput(session)       # 获得模型的输出
tmp_output = MNN.Tensor((1, 2, 224, 224),                   # Temporary variable used for output
                        MNN.Halide_Type_Float, 
                        np.ones([1, 2, 224, 224]).astype(np.float32), 
                        MNN.Tensor_DimensionType_Caffe)
output_tensor.copyToHostTensor(tmp_output)                  # Give the output of the model to tmp_output 变量
x = tmp_output.getNumpyData()[0]                            # 获取 numpy 格式的数据

这段代码没啥问题,But put the last line：

x = tmp_output.getNumpyData()[0]                            # 获取 numpy 格式的数据

替换为：

x = output_tensor.getNumpyData()[0]                         # 获取 numpy 格式的数据

It becomes a mess of data,It may be that there is a problem with the data first and then the column??

他俩都是 MNN.Tensor 的数据类型

所以我感觉,MNN模型输出的Tensor,Convert to the corresponding format firstMNN.Tensor_DimensionType_Caffe,才能打印出来

In other words, this step is to convert the data format：

tmp_output = MNN.Tensor((1, 2, 224, 224),                   # Temporary variable used for output
                        MNN.Halide_Type_Float, 
                        np.ones([1, 2, 224, 224]).astype(np.float32), 
                        MNN.Tensor_DimensionType_Caffe)