[chore] BERTScore averaging

Huffon · Jan 1, 2024 · 9ac7210 · 9ac7210
1 parent 665a20e
commit 9ac7210
Show file tree

Hide file tree

Showing 2 changed files with 5 additions and 10 deletions.
diff --git a/README.md b/README.md
@@ -96,16 +96,6 @@ Answers based on SUMMARY (Questions are generated from Summary)
 
 QAGS Score: 0.3333333333333333
 
-SOURCE Triples
-('Messi', 'is', 'Argentine')
-('Messi', 'is', 'professional')
-
-SUMMARY Triples
-('Messi', 'is', 'Spanish')
-('Messi', 'is', 'professional')
-
-Triple Score: 0.5
-
 Avg. ROUGE-1: 0.4415584415584415
 Avg. ROUGE-2: 0.3287671232876712
 Avg. ROUGE-L: 0.4415584415584415

diff --git a/factsumm/factsumm.py b/factsumm/factsumm.py
@@ -362,6 +362,11 @@ def calculate_bert_score(
             scores["recall"] += recall.item()
             scores["f1"] += f1.item()
 
+        if len(summary_lines) > 1:
+            scores["precision"] /= len(summary_lines)
+            scores["recall"] /= len(summary_lines)
+            scores["f1"] /= len(summary_lines)
+
         logging.info("<BERTScore Score>\nPrecision: %s\nRecall: %s\nF1: %s", scores["precision"], scores["recall"], scores["f1"])
 
         return scores