How to reproduce the mlmodels downloadable in v6.0 (preview not available with export.py)

[AtomScott]

Search before asking

• I have searched the YOLOv5 issues and discussions and found no similar questions.

Question

Recently, @glenn-jocher added support for mlmodel inference.
#6195 (comment)

This is great and demonstrated an easy way to check mlmodels. The same mlmodel can be opened in xcode with the preview option available. However, I can't reproduce the same mlmodel using export.py.

The torchhub mlmodel and the mlmodel downloadable from https://github.com/ultralytics/yolov5/releases/tag/v6.0 seem to be the same, but the mlmodel generate from

!python export.py --weights yolov5s.pt --include coreml  # CoreML export

is different.
Are there any other extra options necessary to reproduce the exact same thing?

It would be extremely helpful to be able to view yolov5 in xcode with the preview tab available.

Additional

Different description and output in when viewed in xcode

• (LEFT) mlmodel downloaded from v6 release page.
• (RIGHT) ml model exported with the following command

docker run ultralytics/yolov5 python export.py --weights yolov5n.pt --include coreml

I have also noticed that the output in preview tab is sometimes completely wrong depending on the scale and aspect ratio of the image. Should I open an extra issue for this? or is this behaviour expected?

Screen_Recording_2022-02-01_at_0.20.54.mov

[github-actions]

👋 Hello @AtomScott, thank you for your interest in YOLOv5 🚀! Please visit our ⭐️ Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.

If this is a 🐛 Bug Report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom training ❓ Question, please provide as much information as possible, including dataset images, training logs, screenshots, and a public link to online W&B logging if available.

For business inquiries or professional support requests please visit https://ultralytics.com or email [email protected].

Requirements

Python>=3.7.0 with all requirements.txt installed including PyTorch>=1.7. To get started:

git clone https://github.com/ultralytics/yolov5  # clone
cd yolov5
pip install -r requirements.txt  # install

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

• Google Colab and Kaggle notebooks with free GPU:
• Google Cloud Deep Learning VM. See GCP Quickstart Guide
• Amazon Deep Learning AMI. See AWS Quickstart Guide
• Docker Image. See Docker Quickstart Guide

Status

If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training (train.py), validation (val.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

[AtomScott]

@glenn-jocher Can I get input on this issue?

[nicolexirakia]

Hi @glenn-jocher , I am facing the exact same problem.

[github-actions]

👋 Hello, this issue has been automatically marked as stale because it has not had recent activity. Please note it will be closed if no further activity occurs.

Access additional YOLOv5 🚀 resources:

• Wiki – https://github.com/ultralytics/yolov5/wiki
• Tutorials – https://docs.ultralytics.com/yolov5
• Docs – https://docs.ultralytics.com

Access additional Ultralytics ⚡ resources:

• Ultralytics HUB – https://ultralytics.com/hub
• Vision API – https://ultralytics.com/yolov5
• About Us – https://ultralytics.com/about
• Join Our Team – https://ultralytics.com/work
• Contact Us – https://ultralytics.com/contact

Feel free to inform us of any other issues you discover or feature requests that come to mind in the future. Pull Requests (PRs) are also always welcomed!

Thank you for your contributions to YOLOv5 🚀 and Vision AI ⭐!