I had the pleasure of reading your paper,but there was a problem that...

[ChuxiaYang]

Do you have any other tricks in your code? Why can't I get your secore with GPU 13060 and 1V100?

[chenhao2345]

According to my experience, the combination of gpu number, batchsize and learning rate can always affect results.

I used 4 GPUs to train. Thus, the batchsize and learning rate values in our paper are set based on 4 GPUs.

If you have enough GPUs, try to use 4 GPUs. Otherwise, you may need to search optimal batchsize and learning rate values to find best results with 2 GPUs. Wish this can help you.

[ChuxiaYang]

Thanks for your reply.Let me try again.