effect of shared_decoder
lx709 opened this issue · 2 comments
lx709 commented
Hello @JonasSchult , Thanks for sharing the code of your nice work. I'm just wondering if you have checked the effect of using shared decoder transformer layers. Let's say will the performance decrease if we set shared_decoder=False.
JonasSchult commented
Hi!
Great question.
In my experiments, the effect was rather minimal while saving quite some memory.
Best,
Jonas
lx709 commented
Thanks for your quick response, much appreciate that.