YOLO Classfication on Rockchip

HalilHuseyinCaliskan commented 1 week ago

Search before asking

[X] I have searched the Ultralytics YOLO issues and discussions and found no similar questions.

Question

Hello, can I perform image classification with YOLOv8's classification method on Rockchip? If I can classify an image using YOLOv8 on the Rockchip 3588, how can I do that? I couldn't find any example code anywhere.Thank you.

Additional

No response

UltralyticsAssistant commented 1 week ago

👋 Hello @HalilHuseyinCaliskan, thank you for reaching out to Ultralytics 🚀! This is an automated response to assist you while an Ultralytics engineer reviews your question.

For guidance on using YOLOv8 for classification, I recommend checking out our Docs, which include useful resources and examples for Python and CLI.

Regarding your query about running YOLOv8 classification on the Rockchip 3588, you're welcome to share a more detailed account of what you've attempted, alongside any error messages you’ve encountered. This additional context can boost the efficiency of our debugging efforts. If you have not done so, a minimum reproducible example could greatly help us understand your issue better. 🐛

Join the Ultralytics community to connect with others and have your questions addressed. You can chat with us live on Discord 🎧, or for more in-depth discussions, check out Discourse. Don't miss out on our Subreddit for great community interaction as well.

Upgrade

Before proceeding, ensure you're on the latest ultralytics package version, with all requirements installed in a Python>=3.8 environment with PyTorch>=1.8:

pip install -U ultralytics

Environments

YOLO models can be run in the following verified environments. Make sure your setup is compatible:

Notebooks with free GPU access:
Google Cloud Deep Learning VM: GCP Quickstart Guide
Amazon Deep Learning AMI: AWS Quickstart Guide
Docker Image: Docker Quickstart Guide

Status

The status badge above reflects the current state of our integrations and testing workflows, ensuring all YOLO Modes and Tasks are functioning correctly.

Stay tuned for a personal response from a member of our team. Thank you for your patience and participation in the Ultralytics community! 🍀

glenn-jocher commented 1 week ago

@lakshanthad Rockchip question!

lakshanthad commented 2 hours ago

@HalilHuseyinCaliskan RKNN will initially only support detection models. Classification models will be supported later.

We are still working on it. You can follow the steps here for the time being.

Thanks for your patience.

ultralytics / ultralytics