I myself am not new to NLP, but for some reason I am unable to grasp purity of BERT. I have seen a ton of blogs, github repos, but none could clarify BERT usage to me.

It would be helpful if you could provide two things :

  1. A clear implementation of BERT, preferably in ipython notebook.
  2. Some papers on BERT excluding the original paper by google.
  • 1
    $\begingroup$ googles implementation is pretty good and its also in tfhub. Best rec for paper is the one on the architecture its about (the transformer-- arxiv.org/abs/1706.03762). The novelty of BERT isnt the model but the training procedure, how it claims and verifies that a bidirectional representation solving for word recovery and next sentence provides a strong contextual embedding $\endgroup$ – mshlis May 26 at 4:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.