f1 only 0.61 by training your model

CogComp / multirc

Reasoning over Multiple Sentences (Multi-RC)

http://cogcomp.org/multirc/

Other

32 stars 10 forks source link

Open Jasperty opened 5 years ago

Jasperty commented 5 years ago

hi,

i train your LR model without changing anything, but i only get f1 of 0.61, how to run to get 0.66?

Jasperty commented 5 years ago

i get 0.61 by predicting all as 1, if i set threshold as 0.5, then i will get f1 of 0.37; if i set threshold as 0.2, then i will get f1 of 0.51,

it is much lower than your result. and i run other models and get 0.5+