./detr -s ../detr.wts detr.engine error

[avBuffer]

trafficstars

Cuda11.4 TensorRT8.4.0 model: detr-r50-e632da11.pth

[06/14/2022-16:09:31] [E] [TRT] 4: (Unnamed Layer* 186) [Matrix Multiply]: last dimension of input0 = 850 and last dimension of input1 = 32 but must match. Your platform support int8: true Building engine, please wait for a while... [06/14/2022-16:09:31] [E] [TRT] 4: (Unnamed Layer* 186) [Matrix Multiply]: last dimension of input0 = 850 and last dimension of input1 = 32 but must match. [06/14/2022-16:09:31] [E] [TRT] 4: [network.cpp::validate::2922] Error Code 4: Internal Error (Could not compute dimensions for (Unnamed Layer* 186) [Matrix Multiply]_output, because the network is not valid.) Build engine successfully! detr: /home/cambricon/data/mgg/tensorrtx/detr/detr.cpp:577: void BuildDETRModel(unsigned int, nvinfer1::IHostMemory**, const string&, std::string): Assertion `engine != nullptr' failed. Aborted (core dumped)

[wang-xinyu]

Looks like your pytorch model doesn't match with the Trt model.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.