What is the actual latency on ARM or CPU?
zeyu-liu opened this issue · 2 comments
zeyu-liu commented
Thanks for the great work! And I also find that ESPNetv2 has comparable performance with MobileNet or ShuffleNet with even less FLOPS, but is the actual inference speed (images / sec) on ARM or CPU faster than the other light weighted architectures?
sacmehta commented
I have not tested on ARM devices, but it delivers comparable speed to otherwhile being more power efficient on other edge devices.
P.S: I am not sure if PyTorch has optimized libraries for CPU or ARM
zeyu-liu commented
Got it, thx for the reply :)