Interpretation of the output

#2
by Yanith - opened

Hello there,

I am currently trying to use this model to make inference on EU legislative acts. The output came out as {label, opportunity, score}. In this case, I just want to clarify on how I can interpret score. In this case, would it be the probability of the paragraph being labelled positively/negatively? Thank you! Also, I would appreciate it if you could recommend some data visualisation packages that could complement the outputs from the model (I am very beginner at this, sorry!)

Warm regards,
Yanith

ClimateBert org

Hi,

Yes exactly, score is the probability of the paragraph belonging to the predicted class (i.e. a confidence score). As visualization I would probably try seaborn which is build on top of matplotline.

Best, Mathias

Hello Mathias,

Thank you so much for your clarification!

Warm regards,
Yanith

Sign up or log in to comment