bdbc-kg-nlp / covolution_over_dependency_tree_emnlp2019 Goto Github PK

Hi,

Thanks for your code. Could you please provide us with more hyper parameters? I ran the code on Laptop dataset and found the result is about 2% less than the result reported in the paper (both acc and f1). For Restaurant14 dataset, the gap is 1%.

Thanks for your outstanding work!

I've tried the code in this repo, but failed to reproduce the result reported in your paper though I follow the guidance in README and code. Could you please release the pretrained models that you use to produce your result?

Thank you!

Wonderful job, thanks! Could you please share the code with the building dependency tree?

能提供一下代码吗