GeneralJing
GeneralJing
use this command, i can build the project smoothly. but when i execute the build/demo/demo file, it occurs segmentation fault(core dumped). another question, how to use the local_pipeline_ubuntu.sh file? in...
it occurs like this: root@832ede5b5bd8:/home/xxxDesktop/jzy/dynamic-occupancy-grid-map/dogm/build# ./demo/demo Segmentation fault (core dumped) No more information output。 How can I verify whether it is caused by different CUDA versions？ drwxr-xr-x 2 root root...
最近看到这段代码的时候，也有点疑惑，有更好的资料解释这个吗？
> @SyGoing @lucky-xu-1994 Check this out it could be useful: [https://github.com/xingyizhou/CenterNet2/tree/73ff02f2967a87ab4877898e7d6207cde439a8c5/detectron2/export](url) > > I am also trying to export a CenterNet2 model to do inference with TensorRT blank page.
执行export_model.py的时候，报错： ModuleNotFoundError: No module named 'detectron2.projects.point_rend'
> 执行export_model.py的时候，报错： ModuleNotFoundError: No module named 'detectron2.projects.point_rend' 这个问题解决了，但是在导出onnx模型的时候遇到了其他问题： Traceback (most recent call last): File "tools/deploy/export_model.py", line 237, in exported_model = export_tracing(torch_model, sample_inputs) File "tools/deploy/export_model.py", line 142, in export_tracing torch.onnx.export(traceable_model, (image,),...
same question. someone knows?
reply so quickly,thank you
Do you know what caused this reason?I'm not familiar with swift and metal. Do you have some demos using oc and opengles?
I can understand that onnx can only be used as an intermediate expression for model conversion, and cannot be inferred？@ wenbingl