I am looking for FP16-INT8 mixed precision yolo-v3 model. I could not find it from the list of models published by Nvidia. Any thoughts?
TensorRT can convert a model into FP16 or INT8 or mixed after training.
So you can try this with a standard YOLOv3 model directly.
Or you can check our NGC website for the model trained by our TAO toolkit.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.