Yolov8 CoreML

🚀 YOLOv8 CoreML

YOLOv8 models converted to CoreML format, enabling object detection on Apple devices with the support of the Neural Engine.

🚀 Quick Start

The YOLOv8 CoreML project provides YOLOv8 models converted into CoreML format. These models can be utilized for object - detection tasks on Apple devices and take advantage of the Apple Neural Engine for enhanced performance.

✨ Features

• Converted YOLOv8 models to CoreML format.
• Compatible with multiple Apple devices including M1, M1 Ultra, and M4 (iPad Pro 2024).
• Can leverage the Apple Neural Engine for efficient inference.

📦 Installation

The project uses the following dependencies:

• pytorch 2.2.0
• coremltools 7.2
• ultralytics 8.2.35

You can install these dependencies using the appropriate package managers. For example, using pip:

pip install torch==2.2.0 coremltools==7.2 ultralytics==8.2.35

📚 Documentation

Tested Devices

The models have been tested on the following Apple devices:

• M1
• M1 Ultra
• M4 (iPad Pro 2024)

Model Sources

• Original weights: https://huggingface.co/Ultralytics/YOLOv8/tree/main
• Model sources: https://github.com/ultralytics/ultralytics

📄 License

This project is licensed under the AGPL - 3.0 license.