about a err Assertion `(outDim - 1) * stride < inputDim + prePad' failed

I know that the setPoolingOutputDimensionsFormula() api is deprecated in tensorrt6.xxx, so i use the setpaddingmode() api to set the pooling layer’s pad.but after i set pool->setPaddingMode(nvinfer1::PaddingMode::kCAFFE_ROUND_UP);there is a err when i run the code,the err is “helpers.cpp:474: int nvinfer1::poolingAxisOutputDim(int, int, int, int, int, nvinfer1::PaddingMode): Assertion `(outDim - 1) * stride < inputDim + prePad’ failed”.I would like to ask if I set it wrong?
as i know, kCAFFE_ROUND_UP=EXPLICIT_ROUND_DOWN - ((EXPLICIT_ROUND_DOWN - 1) * S >= I + B);I don’t think this error should happen

Hi,

Are you using TRT 7.0 with dynamic mode?

Could you please provide details on the platforms you are using:
o Linux distro and version
o GPU type
o Nvidia driver version
o CUDA version
o CUDNN version
o Python version [if using python]
o Tensorflow and PyTorch version
o TensorRT version
If possible, please share the script & model file along with the error log to reproduce the issue.

Also, if you have padding, ensure that the last pooling starts strictly inside the image (instead of at the padding); otherwise clip the last.

Thanks