Aoife Cahill

Results 5 comments of Aoife Cahill

OR alternatively, we could consider something like https://sklearn-crfsuite.readthedocs.io/en/latest/

Specifically, in block 10, the code ``` train_bar.n = 0 val_bar.n = 0 epoch_bar.update() ``` seems to be duplicated which leads to the epoch bar incorrectly updating twice for each...

This was presumably where there was no trimming? If there is still some overlap in range (though perhaps only partial), would you still compute kappa?

certainly would be nice to have, yes :)