hierarchical-attention-networks icon indicating copy to clipboard operation
hierarchical-attention-networks copied to clipboard

Published 20 hours ago • verified publisher icon ematvey

Embeddings for special tokens/padding?

Open Kreiswolke opened this issue 7 years ago • 0 comments

I was wondering where in the code you are initializing the embeddings for the special tokens in the vocabulary (like unknown and padding words) - shouldn't these be set to zero-embeddings and excluded from training? Or how are your dealing with these?

Kreiswolke avatar Aug 09 '18 08:08 Kreiswolke