class DecoderModel(ModelBase) using padding token for prediction

Question

class DecoderModel(ModelBase) using padding token for prediction

richardbaihe opened this issue 2 years ago · 1 comments

Hi There,

In the t0/model.py's DecoderModel class, I noticed that there are padding tokens between the input_ids and labels. It seems padding tokens are used when predicting the labels. Is this an issue?

Answer 1 · 2022-07-13T21:04:02.000Z

fixed with #32 thanks to @thomasw21