Training with your configurations cannot reach the performance reported in the paper

[lrx02]

I have trained the model on a single gpu with your latest configurations, but the performance is severely lower than that reported by your paper. I want to ask if there are some additional settings for training?

[plaiea]

what operating system you are using？win？linux？mac？

[lrx02]

what operating system you are using？win？linux？mac？

linux