CPU usage when train
laisimiao opened this issue · 2 comments
hello, thanks your clean code.
But i found a strange phenomenon that is my cpu usage is very high(almost 100% usage per kernel spied by linux system monitor) so that it run smoothly and stop for a while and continue run...
Have you also experienced this phenomenon when training?What may cause this phenomenon in code?
BTW, its GPU usage is not so tight as cpu usage
Different from other implementation on Siamese trackers. This demo directly crop trian image in original GOT10K dataset rather than crop the whole dataset before train. So during the train process, there will be a big burden on CPU. I have tried this demo on I7 9700k CPU RTX2080Ti GPU, it will taken 100% usage of CPU while token no more than 5% usage of GPU. If you don't have a powerful CPU and large RAM, I strongly recommend you to crop the training datasets for the first time.