Depth Multiple & Width Mulitple Modification

[ysl19971124]

Hi, 我目前操作YOLOv4-p5時, 我有試著改過Depth, Width的Multiple數字, 但目前看起來在createEngine時, 會卡在這個Layer, bottleneck_CSP2 addBatchNorm2d裡面的addScale, 所以想在此詢問一下, 謝謝 ! 目前下面是出現的log

yolov4-p5: /home/<USERNAME>/scaled_yolov4_inference/Scaled-YOLOv4-TensorRT/yolov4-p5-tensorrt/common.hpp:211: nvinfer1::IScaleLayer* addBatchNorm2d(nvinfer1::INetworkDefinition*, std::map<std::__cxx11::basic_string, nvinfer1::Weights>&, nvinfer1::ITensor&, std::__cxx11::string, float): Assertion scale_1' failed. test_inference.sh: line 6: 12739 Aborted (core dumped) ./yolov4-p5 -s ${depth_multiple} ${width_multiple} [07/04/2022-05:27:46] [E] [TRT] Parameter check failed at: runtime.cpp::deserializeCudaEngine::30, condition: (blob) != nullptr yolov4-p5: /home/<USERNAME>/scaled_yolov4_inference/Scaled-YOLOv4-TensorRT/yolov4-p5-tensorrt/yolov4-p5_v2.cpp:282: int main(int, char**): Assertion engine != nullptr' failed.

[07/04/2022-05:27:44] [E] [TRT] Parameter check failed at: ../builder/Network.cpp::addScale::482, condition: shift.count > 0 ? (shift.values != nullptr) : (shift.values == nullptr)

謝謝 !