Update README.md

README.md

@@ -4,4 +4,4 @@ ToTTo is an open-domain English Table-to-Text dataset with over 120,000 training
 
 **Model**
 
-The pre-trained Text-to-Text "t5-base" model is fine-tuned with the Table-to-Text ToTTo dataset(downstream task) for the complete train dataset split of around 120,761 examples. During the fine-tuning process for this downstream task, it has been evaluated with bert-score metric for few examples from validation split.
+The pre-trained Text-to-Text "t5-base" model is fine-tuned with the Table-to-Text ToTTo dataset(downstream task) for the complete train dataset split of around 120,761 examples. During the fine-tuning process for this downstream task, BertScore metric was used as an evaluation metric instead of the standard BLEU metric.