onnx2tf export, python 3.12 support

ambitious-octopus commented 3 days ago

Search before asking

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

Description

onnx2tf 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 commented 3 days ago

👋 Hello @ambitious-octopus, thanks for reaching out and for your interest in Ultralytics 🚀! This is an automated response; an Ultralytics engineer will assist you soon.

For now, please ensure you're using the latest version of our repository. You can upgrade the ultralytics package and all requirements in a Python>=3.8 environment with PyTorch>=1.8 by running:

pip install -U ultralytics

🐛 For bug reports such as yours, we kindly request a minimum reproducible example to help us debug the issue effectively.

Join our community for real-time support and discussions:

Discord 🎧 for chat
Discourse for discussions
Subreddit for sharing and learning

Environments

You can run YOLOv8 in these environments with all dependencies preinstalled, including CUDA/CUDNN:

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

Status

Our continuous integration tests Ultralytics CI ensure correct functionality across various platforms. Check the status here:

Feel free to reach out with more details to help us assist you better! 😊

inisis commented 3 days ago

it seems that python 3.12 has many problems with tf, here is one previous test for yolo exports

glenn-jocher commented 3 days ago

@inisis thank you for sharing the information. Python 3.12 compatibility issues with TensorFlow are known, and we recommend using Python 3.8 to 3.11 for a smoother experience with YOLO exports.

ultralytics / ultralytics