Open Arjunsankarlal opened 5 years ago
Pretrained models will be released soon!
Looking forward for it. Also I am interested to know the inference time taken by the model, speaking of removing the pipeline architecture followed previously on other models, is the inference time is reduced? Also time taken for encoding text with BERT representations would take more time right?
I am asking it, since it requires huge computational power. Is there any plan to upload them anytime soon? Which would be really helpful for the fellow community.