Best course to learn how to classify text with NN and BERT models

What is the best course for:
a practical example of how to use a bert transformer model to retrain BERT on a large corpus of long texts (around 1000 words), in order to classify.
A practical example how to use LM to train a chatbox to answer any question based on a large document.

If your solution space is limited to, then that is the course. It addresses the topics you enumerate, but it is done using Google trax, which seems to have been abandoned by Google and not updated in 18 months (when the creator split to OpenAI)

Interweb search on your keywords returns other options