search

xdqkid / S2S-AMR-Parser

Improving AMR parsing with Sequence-to-Sequence Pre-training
42 stars 4 forks source link

How to train this model from scratch #4

Closed Wangpeiyi9979 closed 2 years ago

Wangpeiyi9979 commented 3 years ago

Hi, Thanks for your nice work. However, I am struggled with training the model from scratch? Could you help me!!

xdqkid commented 3 years ago

Following https://github.com/RikVN/AMR and https://github.com/OpenNMT/OpenNMT-py, it is easy to reproduce baseline. Do not use fairseq with my parameter settings!!! You should pay more attention to the paper and the code details.