Does Chinese data need additional change in the code?

[drxmy]

Or will it work just fine with Chinese? Really interesting work!

[hardyqr]

Thanks for your interest! In principle you shouldn't need to do much. You need to switch the English BERT models (incl. the initial bi-encoder and the base model) to Chinese ones. E.g., for bi-encoder you could use simcse-chinese-roberta-wwm-ext (I randomly retrieved this one from google); and the corresponding base model is hfl/chinese-roberta-wwm-ext.

[drxmy]

Thank you for replying so quickly. Yes, the model definitely need to change. I was mainly concerned about data processing. I will try it tomorrow.

[zpp13]

@drxmy 请问你做过尝试了吗，效果怎么样

[drxmy]

@drxmy 请问你做过尝试了吗，效果怎么样

我用的自己的数据集，现在的结果有一点奇怪，可能代码哪里没改对。bi-encoder验证指标过高，后面还会nan。不过最近一直没时间看是什么问题