ultralytics/yolov3
YOLOv3 real-time object detection model implemented in PyTorch with export to ONNX, CoreML, and TFLite formats.

Velocity · 7d
+3.7
★ / day
Trend
→steady
star history
This repository implements YOLOv3, a deep learning object detection algorithm, in PyTorch. It provides training and inference scripts for the YOLO (You Only Look Once) family of models. The repository also includes utilities for exporting trained models to common deployment formats including ONNX, CoreML, and TFLite for inference serving across different platforms.