Cheng Yang
Cheng Yang
@ginsent it works,thanks!
> > I believe current normal fairseq does not provide such feature. Not in command line for sure. > > -- As for the implementation, If your data is not...
And one more thing, it looks like that you normalize the format of time to `XX:XX` in `flight-easy.json`: However in `Combined_Flights_2022.csv`, the format is as follows, which will cause mismatch...
> 在情感预测模型中，prefix是“很满意。”，候选答案是“很”和“不”，目前mark_idx是1，这样的话mask后的句子输入是“很[MASK]意”+text，这样看起来挖空的位置没什么实际含义。请问是我理解有什么不对吗，为什么不将mark_idx设为0呢？ 你好，根据我的理解，你没有考虑到BERT输入时，句子前面会有一个`[CLS]`哦。这个mark_idx是考虑了`CLS`之后的idx~即`[CLS]很满意。`
Have you finished this issue？
Ok, thank you~
Have you checked the quality of the English version？