About torch::jit::trace

[Hugo699]

Hello,
Thank you for sharing your work, and I want to use libtorch to deploy this network in C++, but when using torch::jit::trace(), I get this error(executing test_sample.py can run successfully):


Because torch::jit::trace() cannot handle dictionary, I changed the output of depth_decoder to list, and there is a line "import hr_networks" in test_sample.py, but I did not find hr_networks, I don't know if this affectstorch::jit::trace().

Thank you very much！

[brandleyzhou]

• you can delete the 'import hr_network', but I don't think it is the reason.
• IF you deploy it when testing, you can let depth_decoder only output a full-scale depth map instead of multi-scale depth maps
• for this error, Sorry, I am not familiar with the torch.jit.trace. have you tried it on monodepth2? and met the same error?

[Hugo699]

Thank you for your time, I don't have the same problem on monodepth2, I think there might be something in HRNet that cannot be traced， but I don't know how to solve it.

[brandleyzhou]

I think so, I guess maybe you have to change the way of initializing a hrnet e.g. hrnet_config.py which uses a yacs lib