issues
search
NLPInBLCU
/
BiaffineDependencyParsing
BERT+Self-attention Encoder ; Biaffine Decoder ; Pytorch Implement
MIT License
73
stars
17
forks
source link
[TEXT-dev-LAS:0.8232] BERT+Transformer
#2
Closed
LiangsLi
closed
4 years ago
LiangsLi
commented
4 years ago
BERT之后加Transformer可以明显提升模型性能
Layer Attention效果不好(削弱了模型的性能)
模型似乎需要更强的正则化
更新了模型的输出保存
目前最优结果