You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Straightforward alternative "single number" evaluation metric for ML algorithms--- should make comparisons to e.g. Rea et al (2019) and Churchill (2019) a bit easier.
More generally, we could output various Fγ scores, such as the F0.5, F1, F2 measures.
Straightforward alternative "single number" evaluation metric for ML algorithms--- should make comparisons to e.g. Rea et al (2019) and Churchill (2019) a bit easier.
More generally, we could output various Fγ scores, such as the F0.5, F1, F2 measures.
See eq. 2 in Rea et al (2019), van Rijsbergen (1979) pg 134, Sasaki (2007), or https://en.wikipedia.org/wiki/F1_score for example.
The text was updated successfully, but these errors were encountered: