Hi, I have some questions about training speed.

[Okityou]

When I use the code you provide to train on my server. GPU Memory Usage is normal but Volatile GPU-Util is 0%. I have used tensorflow, but I have never used lua+torch, I guess it may be that the data is not queued. But since I am a beginner, I can't see if the data is trained to use the queued mode. If not, whether the above reasons are caused, or other reasons. Can you help me with it?
Thank you!