This toolkit was designed for the fast and efficient development of modern machine comprehension models, including both published models and original prototypes.
I think 'skip' answer type is actually answerable (but no span exist).
why is skip type skipped in training time? is there any performance gain?
I'm afraid that skipped conversation ruins context of conversation history.
btw, thanks for good code.
I think 'skip' answer type is actually answerable (but no span exist). why is skip type skipped in training time? is there any performance gain? I'm afraid that skipped conversation ruins context of conversation history. btw, thanks for good code.