Adding supervised and unsupervised scoring function types

inspirehep / beard

Bibliographic Entity Automatic Recognition and Disambiguation

Other

66 stars 36 forks source link

Adding supervised and unsupervised scoring function types #44

Closed natsheh closed 9 years ago

natsheh commented 9 years ago

Currently, if we pass a supervised scoring function like b^3_F_score to the clusterer, the code will raise an error for the case of no ground truth (all y's are -1). This PR allows a proper usage of the clustering passing a supervised or unsupervised scoring function or both. @MSusik @glouppe @etzemis

MSusik commented 9 years ago

:+1:

glouppe commented 9 years ago

Looks OK, merging