In jetson nano, inference time of fp16 and fp_32 is almost same?

[xxy90]

No description provided.

[jaybdub]

Hi xxy90,

Thanks for reaching out!

Do you mind sharing which model architecture you're referring to? The relative performance of FP32 vs. FP16 may depend on model architecture. I think the scaling also might not be linear with bit depth, because of various overhead when using reduced precision.

Best,
John

[zhangchenwei115]

Hi xxy90,

Thanks for reaching out!

Do you mind sharing which model architecture you're referring to? The relative performance of FP32 vs. FP16 may depend on model architecture. I think the scaling also might not be linear with bit depth, because of various overhead when using reduced precision.

Best,
John

Hi John,
I had the same problem which on Jetson Nano, weather in converting,fp16=True or False. the speed is the same. I used the model of lightweight openpose,here is the link, https://github.com/Daniil-Osokin/lightweight-human-pose-estimation.pytorch

[xxy90]

The model architecture is new anchor-free object-detection Nanodet, whose backbone is shuffleNetv2

[shadowuyl]

I also meet the same problem while using YOLOX-Nano(ref https://github.com/Megvii-BaseDetection/YOLOX) on jetson nano.