Update export.py

[nanmi]

🛠️ PR Summary

_{Made with ❤️ by Ultralytics Actions}

🌟 Summary

This PR appears to be either an empty or placeholder PR, with no code changes included.

📊 Key Changes

• There are no code changes to highlight as the PR diff is empty.

🎯 Purpose & Impact

• Without code changes, there's no direct impact on the software or users.
• It could be a PR for discussion, planning, or an error that needs to be closed without merging.

[github-actions]

👋 Hello @nanmi, thank you for submitting a 🚀 PR! To allow your work to be integrated as seamlessly as possible, we advise you to:

• ✅ Verify your PR is up-to-date with upstream/master. If your PR is behind upstream/master an automatic GitHub actions rebase may be attempted by including the /rebase command in a comment body, or by running the following code, replacing 'feature' with the name of your local branch:

git remote add upstream https://github.com/ultralytics/yolov5.git
git fetch upstream
git checkout feature  # <----- replace 'feature' with local branch name
git merge upstream/master
git push -u origin -f

• ✅ Verify all Continuous Integration (CI) checks are passing.
• ✅ Reduce changes to the absolute minimum required for your bug fix or feature addition. "It is not daily increase but daily decrease, hack away the unessential. The closer to the source, the less wastage there is." -Bruce Lee

[glenn-jocher]

@nanmi hi there. What is the bug you are trying to fix with this? For ONNX dynamic models the correct export command is:

python export.py --include onnx --dynamic

[nanmi]

@nanmi你好。您要修复的错误是什么？对于 ONNX 动态模型，正确的导出命令是：

python export.py --include onnx --dynamic

@glenn-jocher Hi, when I fine tune the model and save onnx directly, it is easy to generate output the processing after the detection head, if I export the pre training model onnx separately, there will be no error 😀

[glenn-jocher]

@nanmi I'm sorry I'm not understanding. The Detect layer onnx_dynamic attribute is already set correctly here:

yolov5/export.py

Lines 293 to 295 in ac2c49a

	elif isinstance(m, Detect):
	m.inplace = inplace
	m.onnx_dynamic = dynamic

If you are seeing an error in YOLOv5 that this PR is fixing, please raise a bug report first and then link this PR to it as a solution. As is I don't understand the problem this PR is fixing.

[glenn-jocher]

/rebase