#StackBounty: #nlp #transformer Why does vanilla transformer has fixed-length input?

Bounty: 50

I know that in the math on which the transformer is based there is no restriction on the length of input. But I still can’t understand why we should fix it in the frameworks (PyTorch). Because of this problem Transformer-XL have been created.

Can you explain to me where this problem is hiding, please?


Get this bounty!!!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.