Vincent Nguyen

not that much since at the beginning I use small batches of 32. EDIT: 4.6GB to be specific.

Hi @vikrant97 where do you stand on target features, are you still willing ot work on it ?

Thanks for the contribution, however you will need to change the logic. As a first PR you just need to add Bleu as an extra validation metric (look at how...

no, the class in its own file is fine. I am just saying that Bleu is another metric at the same level as PPL or ACC that's it.

Can you please post your command line for both preprocessing and training ? thanks.

@memray @francoishernandez Maybe this has been fixed with pytorch 1.2 Worth to try.

Sorry to hijack this post :) Marcin, paper and website are not in line for training speeds on transformer. one says about 42K w/s other says 60k w/s both for...

do you fit approx 4096 tokens per GPU for base and 2048 tokens for big ? anyway you're about 25% faster I would say, makes sense. I'll have to work...

@kpu for some reason there are issues with RTX 2080 ti on some i7 cpu. not getting full speeds.

oh yeah I know it's fp32, but with fp16 supposed to x3 according to FB, right ?