我基于该项目实现了pytorch的版本
wireless911 opened this issue · comments
hello!
感谢您开源出来的代码和数据,让我们可以进行研究和试验,鉴于allennlp我不是很熟悉,我在基于这个存储库实现了lstm版本的span-aste,我想可能会有人跟我会碰到同样的问题,我将我的代码也开源了出来,希望可以帮到更多的人,期待您提出宝贵的建议 。https://github.com/wireless911/span-aste.git
谢谢
That's great, are you able to reproduce the result?
Thank you for your reply. After the bert version I implemented was trained in the 15res dataset, the f1 value of the validation set was 69.1 and the result(70.75) in the paper was still slightly different. I am still exploring.
太感谢了
已经让复杂的包搞得晕头转向了,总是包冲突
太好了,我也是各种包搞得头疼,Span-ASTE用的包都太过时了