cuDLA-samples icon indicating copy to clipboard operation
cuDLA-samples copied to clipboard

Published 20 hours ago • verified publisher icon NVIDIA-AI-IOT

Quantization of yolov5

Open zifeiyu-tan opened this issue 1 year ago • 1 comments

If i only want to quantize with int8 in yolov5, how to change this code。

zifeiyu-tan avatar Jan 16 '24 09:01 zifeiyu-tan

Hi @tangxiangyu0313 ,

The code in this link has already implemented quantization for YOLOv5. We provide two quantization modes for selection, and you can refer to the README for replication.

Thanks

liuanqi-libra7 avatar Feb 05 '24 09:02 liuanqi-libra7