How to replace Conv2dStaticSamePadding and MaxPool2dStaticSamePadding in order to be able to transfer out dlc on the Qualcomm platform

datvuthanh / HybridNets

HybridNets: End-to-End Perception Network

MIT License

585 stars 120 forks source link

How to replace Conv2dStaticSamePadding and MaxPool2dStaticSamePadding in order to be able to transfer out dlc on the Qualcomm platform #48

Open aimuch opened 2 years ago

aimuch commented 2 years ago

As shown above, using Qualcomm toolchain to export dlc file, there is error:

Encountered Error: ERROR_ASYMMETRIC_PADS_VALUES: Asymmetric pads values is not supported

I try EfficientNet-PyTorch to exprot dlc successfully. I found the problem to arise Conv2dStaticSamePadding and MaxPool2dStaticSamePadding, the using F.pad to asymmetric padding. So, How to replace them?