Core ML exports, python 3.12 support

[ambitious-octopus]

Search before asking

• [X] I have searched the YOLOv8 issues and found no similar feature requests.

Description

coreml export does not work with python 3.12.

Use case

No response

Additional

No response

Are you willing to submit a PR?

• [ ] Yes I'd like to help by submitting a PR!

[UltralyticsAssistant]

👋 Hello @ambitious-octopus, thank you for your interest in Ultralytics 🚀! We're here to help. This is an automated response, and an Ultralytics engineer will also assist you soon.

We recommend checking out our comprehensive Docs where new users can find many Python and CLI usage examples, which may already address some common questions.

As this appears to be a 🐛 Bug Report related to Core ML export compatibility with Python 3.12, please provide a minimum reproducible example if possible, to help us investigate further.

To ensure your environment is set up correctly, please upgrade to the latest ultralytics package and confirm you have all requirements installed in a Python>=3.8 environment, with PyTorch>=1.8:

pip install -U ultralytics

For collaboration and support from the Ultralytics community, you can join us on Discord 🎧, engage in detailed discussions on Discourse, or explore threads on our Subreddit.

Environments

YOLOv8 can be utilized in any up-to-date, verified environment:

• Notebooks with free GPU:
• Google Cloud Deep Learning VM
• Amazon Deep Learning AMI
• Docker Image:

Status

If the CI badge is green, all Ultralytics CI tests are passing. They verify the proper operation of all YOLOv8 Modes and Tasks across various platforms every 24 hours and on every commit.