Junseong Kim

@crazyofapple Totally agree haha. Now I'm trying to train this model with korean corpus with 1080ti x2. But seriously, the model is too big for individual researcher.... we need some...

@MrRace Love to do it, if I have enough lot's of 2080ti. https://twitter.com/Tim_Dettmers/status/1050787783004942336 > Regarding compute for BERT: Uses 256 TPU-hours similar to the OpenAI model. Lots of TPUs parallelize...

@MrRace > On a standard, affordable GPU machine with 4 GPUs one can expect to train BERT for about 99 days using 16-bit or about 21 days using 8-bit. Haha...

> 준성님 이제는 논문까지 내시려고... > 멋지십니다 ! 저도 논문 써보고 싶어요...ㅎㅎ 과연 쓸 수 있을까요?

I'm facing the same issue and solved it with @williamih 's solution! This issue should be fixed in the future release. > I'm also receiving this error when using a...

The way that I found. But there is some issue. ```python get_custom_objects().update({'CRF':CRF, 'CRFLoss':CRF.loss_function}) ``` Under code shows the CRF layers loss_function() But as you can see, CRF loss function calculating...