lpkoh

Hi thank you for replying on this. I am confused. You said here and https://github.com/ceccocats/tkDNN/issues/186 and https://github.com/ceccocats/tkDNN/issues/173 that what demo prints on screen is the preprocessing + inference + postprocessing....

Ah gotcha. So I guess the difference between ~27 fps on yolo4-416x416 vs ~44 in your repo is probably down to MAXN? Could the Tensorrt version difference be an issue?...

I am facing same issue

@shaydeci Thank you so much for the response, they pretty much explain exactly how one would do it. Can I however, understand why QAT is applied after PTQ? It says...

Also, can I clarify, suppose I go through PTQ and QAT and obtain: [/content/sg_checkpoints_dir/yolo_nas_s_soccer_players/yolo_nas_s_soccer_players_32x3x640x640_ptq.onnx](https://colab.research.google.com/drive/1yHrHkUR1X2u2FjjvNMfUbSXTkUul6o1P?usp=sharing#) and [/content/sg_checkpoints_dir/yolo_nas_s_soccer_players/yolo_nas_s_soccer_players_32x3x640x640_qat.onnx](https://colab.research.google.com/drive/1yHrHkUR1X2u2FjjvNMfUbSXTkUul6o1P?usp=sharing#)". When I convert these models to tensorrt, should I still be converting using --fp16...