From 230698dc83b512a78df06dd9816e9fd63143a04d Mon Sep 17 00:00:00 2001 From: Paul O'Leary McCann Date: Thu, 12 Aug 2021 18:22:08 +0900 Subject: [PATCH] Fix bug in scorer Scoring code was just using one metric, not all three of interest. --- spacy/pipeline/coref.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/spacy/pipeline/coref.py b/spacy/pipeline/coref.py index a703c3a37..94677e2bf 100644 --- a/spacy/pipeline/coref.py +++ b/spacy/pipeline/coref.py @@ -357,7 +357,7 @@ class CoreferenceResolver(TrainablePipe): # we need to handle the average ourselves. scores = [] for metric in (b_cubed, muc, ceafe): - evaluator = Evaluator(b_cubed) + evaluator = Evaluator(metric) for ex in examples: p_clusters = doc2clusters(ex.predicted, self.span_cluster_prefix)