uds-lsv/bert-stable-fine-tuning
On the Stability of Fine-tuning BERT: Misconceptions, Explanations, and Strong Baselines
PythonApache-2.0
Issues
- 0
Adam Epsilon Choice
#5 opened by jacky18008 - 0
Loss surface axis
#4 opened by anyuzoey - 3
- 1
- 4
Did RoBERTa longer trained?
#1 opened by bayartsogt-ya