What about gpu inference time?

[mahdip72]

Are there any experiments regarding the speed of efficientformer on gpu devices?

[edwardyehuang]

Same. Could you please provide ablation stuides that only use GPU/CPU on iPhone? Depending on the iPhone runtime, when NPU (ANE) is enabled, some operations are not supported and may run outside NPU.

[liyy201912]

Thanks for your interest in our work.

We are going to add the following speed results in the final manuscript:

1. Speed on NVIDIA A100, deployed with TensorRT
2. Speed on Google 6, deployed with NNAPI
3. Speed on iPhone CPU, deployed with CoreML.

[alanspike]

Hi @edwardyehuang, with the latest iOS and Xcode, you could get the latency directly on CPU or CPU & GPU following this link.