nlpodyssey / cybertron

Cybertron: the home planet of the Transformers in Go
BSD 2-Clause "Simplified" License
289 stars 26 forks source link

Cross Encoder Support #35

Open phillebaba opened 1 year ago

phillebaba commented 1 year ago

I am looking to do some tests with cross encoder models like MS Marco. After reading up on the python code used, I realized that all examples are based on being able to tokenize multiple sentences into a single input. As it stands now it seems that only a single input is possible for text classification models. Is this correct or have I missed something?