Huy Tran Van

> There's a drop on TensorRT compared to the PyTorch model. In some models, it's a relevant drop. In other models (like PPYOLOE and YOLO-NAS), it's a small. The test...

> The FP16 and FP32 mAP are equal. Yeah, I think so. In your opinion, what is the reason of fp32, fp16's mAP big drop compared with .pt models? I...

> In my opinion, TensorRT layers are performance focused, making some tweaks to precisions and parameters. So it's faster, but loses some of the accuracy. Thanks for sharing.