JZhao

same issue

> 请问已经生成STC_result.txt，想生成像Readme 评价指标表格中的 PPL | BLEU-2 | BLEU-4 | Dist-1 | Dist-2 | Greedy Matching | Embedding Average 这些值，请问你们是怎么计算的？ 请问你评测指标的代码的结果符合论文里的了吗，可以分享一下吗，我的一直些问题。

so ,could i konw how to solve the overfitting problem,thanks !!^_^

Firstly,thanks for your improvemen for the original version,it's very helpful!But i got a problem in this code :my train accuracy is very high at the begining(about 55% );and the validation...

> Do you solve this question? you can use the transforemr in fairseq by Facebook to reach the score.

Sorry, It's my fault. I wasn't clear. I mean how could I gain AMR from my own dataset(a group of sentences). ------------------ 原始邮件 ------------------ 发件人: "Linfeng Song"

Thanks for your reply^_^. ------------------ 原始邮件 ------------------ 发件人: "Linfeng Song"

Hi, Could I know how to calculate the "unbiased" accuracy in the paper? I can't got the results in the paper.