Transformers-sklearn: a toolkit for medical language understanding with transformer-based models

Table 5 The experimental results of transformers-sklearn, transformers and UER in four medical NLP tasks (mode_type = “bert”)

Name	Score			Second			Lines of code			Pre-trained model
Name	Ours	Transformers	UER	Ours	Transformers	UER	Ours	Transformers	UER	Pre-trained model
TrialClassification	0.8225^a	0.8312^a	0.8213^a	1198	1227	764	38	246	412	bert-base-chinese
BC5CDR	0.8703^a	0.8635^a	-	471	499	-	41	309	-	bert-base-cased
DiabetesNER	0.6908^a	0.6962^a	0.7166^a	1254	1548	2805	63	309	372	bert-base-chinese
BIOSSES	0.8260^b	0.8200^b	-	19	15	-	41	246	-	bert-base-cased

^aThe value of Macro F1, where the bolded one indicates the best performance.
^bThe value of Person correlation, where the bolded one indicates the best performance.