YOLOv5_PyTorch_cpp icon indicating copy to clipboard operation
YOLOv5_PyTorch_cpp copied to clipboard

A c++ implementation of https://github.com/ultralytics/yolov5 with PyTorch c++ API (LibTorch)

Results 5 YOLOv5_PyTorch_cpp issues
Sort by recently updated
recently updated
newest added

after successful build running this command yields error while loading shared libraries: libopencv_dnn.so.4.4: ./main --input-dir ../../input_image/ --model-file ../../model/yolov5s_torchscript_B1-C3-H640-W640_torch1-6-0_cuda10-2_cpu.pt I am not sure what am I doing wrong

Hello, my cuda10.2 libtorch1.7.1 model yolov5 was slow in the first two tests and the second one was the slowest and then normal. What is the reason? ![图片](https://user-images.githubusercontent.com/51087601/105516143-7f356e00-5d10-11eb-9985-aebc36a4fe8b.png) ![图片](https://user-images.githubusercontent.com/51087601/105516180-89f00300-5d10-11eb-9f8f-ca48851b7fdb.png)

请问在导出gpu版本的模型的时候,设置的model.model[-1].export = False吗?