Journal of Big Data

Table 6 The setup of the transfer learning models

From: Toward multi-label sentiment analysis: a transfer learning based approach

	BERT			XLNet
	Yelp	Wine	Movie	Yelp	Wine	Movie
Epochs	27	40	10	9	14	8
Batch size	32			16
Maximum length	128			128
Learning rate	\(2 \times 10^{-5}\)			\(2\times 10^{-5}\)

Back to article page