Question for paper "Adapting BERT for Continual Learning of a Sequence of Aspect Sentiment Classification Tasks"
leducthanguet opened this issue · 2 comments
leducthanguet commented
ZixuanKe commented
Thank you so much for your interest.
Are you asking where the output of the first layer of TSM goes to? We have two layers in TSM, so it simply goes to the next layer. You can see the code here:
https://github.com/ZixuanKe/PyContinual/blob/main/src/networks/base/adapters.py#L440
Feel free to let us know if we are misunderstanding your question.
leducthanguet commented
Thank you for your answer.