tensorflow_multigpu_imagenet icon indicating copy to clipboard operation
tensorflow_multigpu_imagenet copied to clipboard

run slowly,my code run very slowly,gpu use rate is to low

Open encichou opened this issue 5 years ago • 2 comments

encichou avatar Jul 14 '19 08:07 encichou

The same problem. In a training epoch, the most time, gpu is not used. Moreover, as the training epoch increasing, the run time of a training epoch will be longer.

AILearnerLi avatar Jul 22 '19 07:07 AILearnerLi

Same problem. GPU utilization rate keep 0% almost all of the time.

xh-yuan avatar May 19 '20 13:05 xh-yuan