JackieWu

Hi @vb123er951 , I am not familiar with it. You can ask the author @Heliang-Zheng.

@RogerChern Thanks for your support! Currently, MobulaOP doesn't support math helpers. I'm going to refactor MobulaOP with C++, and add the support of ONNX operators, which play the roles of...

@tiancity-bytedance Thank you for the report. I will check it.

@tiancity-bytedance `mobula.func.cumsum` is similar to [`np.cumsum`](https://docs.scipy.org/doc/numpy-1.15.1/reference/generated/numpy.cumsum.html)

@tiancity-bytedance `cumsum_kernel(const int N, const T* X, T* I, const int att_size)` The four parameters are `batch size`, `input`, `output`, `the number of elements in a batch` respectively.

Sorry, I have not tested it. I'm busy recently. I will train it.

Hi, @tiancity-bytedance . I have tested it and got the 86~87 accuracy on CUB-200-2011. Setting: Number of GPUs: 4 Batch Size: 48 MobulaOP/mobula/config.yaml : USING_ASYNC_EXEC: 0

``` INFO:root:Epoch[299] Train-att_net_accuracy=1.000000 INFO:root:Epoch[299] Train-part_net_accuracy=0.984375 INFO:root:Epoch[299] Train-master_net_accuracy=0.984375 INFO:root:Epoch[299] Train-part_net_aux_accuracy=1.000000 INFO:root:Epoch[299] Train-master_net_aux_accuracy=1.000000 INFO:root:Epoch[299] Train-distillation_loss=4.100089 INFO:root:Epoch[299] Time cost=20.978 INFO:root:Saved checkpoint to "./model/tasn-0300.params" INFO:root:Epoch[299] Validation-att_net_accuracy=0.804986 INFO:root:Epoch[299] Validation-part_net_accuracy=0.856728 INFO:root:Epoch[299] Validation-master_net_accuracy=0.860485 INFO:root:Epoch[299] Validation-part_net_aux_accuracy=0.864754 INFO:root:Epoch[299] Validation-master_net_aux_accuracy=0.869023...

Hi @vb123er951 , the function `mx.nd.cumsum` is supported in the eldder version of MXNet. Please use the latest version such as MXNet 1.6 : )

Hi @vb123er951 , I have updated the code, which supports the old version of MXNet without `cumsum`.