Closed ShenZheng2000 closed 4 months ago
Hi, thanks for your interest in our work !
I use 4 GPUs for training the model, in fact 2 GPUs is also enough, and the converge time is around >= 1 hour.
We set the steps in 62500 and 12500 is an epoch, I think 25000 (2 epochs) is also enough for converge. But the results maybe not as good as 62500 (5 epochs).
How many GPUs do you use for training, and how many hours does it take to converge the model?