inconsistent metric names

ExpressAI / eaas_client

https://expressai.github.io/autoeval/

Apache License 2.0

0 stars 1 forks source link

Closed pfliu-nlp closed 2 years ago

pfliu-nlp commented 2 years ago

rouge2 (in class definition) <-> rouge_2 (in sample-level output json)

pfliu-nlp commented 2 years ago

similar case on bert_score

yyy-Apple commented 2 years ago

Does this mean we should have bert_score_p, bert_score_r, bert_score_f as separate metrics?

pfliu-nlp commented 2 years ago

Thanks for fixing these issues.