-
-
Notifications
You must be signed in to change notification settings - Fork 16.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add --include torchscript onnx coreml
argument
#3137
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
👋 Hello @CristiFati, 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 origin/master. If your PR is behind origin/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 rebase 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
@CristiFati thanks for the PR! I think the assumption is that users would not install packages for export channels that were not of interest to them, like coremltools etc. One thing also is that coremltools export first requires a TorchScript export. I see you've made some additional improvements also, I'll take a look. |
renamed --skip-format to --exclude
@CristiFati I updated this a bit to bring the terminology in line with other 'include'/'exclude' functions like check_requirements(), but something doesn't seem right about going with the exclude pathway vs offering an include pathway instead. i.e. if I want to export onnx, I would rather opt-in to onnx export than opt-out of other exports (which I'm potentially unaware of). |
Thanks for the quick replies. Going one at a time:
Let me know what else I could do. |
@CristiFati yeah that all makes sense. I think what we want is to remove the parser.add_argument('--include', nargs='+', default=['torchscript', 'onnx', 'coreml'], help='included formats') |
Done, now with that CoreML -> TorchScript dependency: although I don't see one's input depending on the other's output, would it make sense that line 79 to be |
--include torchscript onnx coreml
argument
@CristiFati changes look good, I added a CoreML check for TorchScript. Merging PR now. Thank you for your contributions! |
Glad to be helpful. Looking forward to more! |
* Allow users to skip exporting in formats that they don't care about * Correct comments * Update export.py renamed --skip-format to --exclude * Switched format from exclude to include (as instructed by @glenn-jocher) * cleanup Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> (cherry picked from commit d9b4e6b)
* Allow users to skip exporting in formats that they don't care about * Correct comments * Update export.py renamed --skip-format to --exclude * Switched format from exclude to include (as instructed by @glenn-jocher) * cleanup Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
Some people might only want to export the models in certain formats (e.g. I only wanted ONNX). Although exporting in all the formats doesn't take much time (in my case, bulk exporting all 8 files took ~1min), this will become more handy when more formats will be available.
Initially I had a patch for line 12 as well (as yesterday I downloaded v5 and I tried running the script from a different folder, which obviously failed), then I cloned the repo and noticed it was already fixed in master, so the only thing left to do was to fix comments.
🛠️ PR Summary
Made with ❤️ by Ultralytics Actions
🌟 Summary
Enhanced model export capabilities for YOLOv5, now supporting CoreML format.
📊 Key Changes
PYTHONPATH
.--include
argument to specify which formats to export.--include
argument for format specification.🎯 Purpose & Impact
--include
option. The potential for simpler deployment processes and broader usage of models in different platforms is increased.