Use torch.jit.trace in unit-test

[nihui]

No description provided.

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you all sign our Contributor License Agreement before we can accept your contribution.
1 out of 2 committers have signed the CLA.

✅ zhiqwang
❌ nihui
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[codecov]

Codecov Report

Merging #205 (e97b14b) into main (fd53543) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##             main     #205   +/-   ##
=======================================
  Coverage   96.98%   96.98%           
=======================================
  Files          10       10           
  Lines         630      630           
=======================================
  Hits          611      611           
  Misses         19       19           

Flag	Coverage Δ
unittests	96.98% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update fd53543...e97b14b. Read the comment docs.

[zhiqwang]

Hi up @nihui , Thanks for your contributions here!

[zhiqwang]

Hi @nihui , I previously wrapped the torch.jit.trace in get_trace_module as follows. I changed your code here directly to this calling method. Also, I made some changes to the C++ part of the code to accommodate torch.jit.trace.

https://github.com/zhiqwang/yolov5-rt-stack/blob/fd53543e00b534ab11f5dfb689d4cdfb0160ad86/yolort/relaying/trace_wrapper.py#L37-L61