Transformer Encoder to get the new representation need to be train?

#45
by feyhong44 - opened

Do transformer to get only last hidden state for downstream task will transformer be train the model?
eg: I use graphormerModel from higgingface
Everytime, I run the same input I get different outcome. Do it normal?

Your need to confirm your account before you can post a new comment.

Sign up or log in to comment