fishfree

Is it enough to just put the `roberta_(e19_2025.02.28_09.36).pt` file under the data folder?

``` ubuntu@ip-172-31-18-132:~/www/wordflow$ npm run dev > [email protected] dev > vite --port 3000 file:///home/ubuntu/www/wordflow/node_modules/vite/bin/vite.js:7 await import('source-map-support').then((r) => r.default.install()) ^^^^^ SyntaxError: Unexpected reserved word at Loader.moduleStrategy (internal/modules/esm/translators.js:133:18) at async link (internal/modules/esm/module_job.js:42:21) ```...

Does it support to bind different top domain names / sub domain names / sub-paths to different subsites?

1

I didn't find anything about this in the doc.

论文里给的是英文的例子，我还看到有个[这个文件](https://github.com/liyucheng09/MetaphorFrame/blob/08b38aa5b5c6a16ed0062b01e529f71872304ca8/lyc/hfds_scripts/chinese_metaphor_dataset.py)似乎是处理中文语料。但不知道具体如何来训练中文的模型。 还麻烦给个示例代码，先谢过了！

这样可以更完善的维护。

请问下如何使用该repo计算一段古文的嵌入向量？

1

我看到REAME里有说用到了四库全书来训练嵌入向量的，但没看到使用示范代码。

希望能增加从古汉语翻译为现代汉语的功能。现在大语言模型使用提示词就能达到这个目的，不需要API的。

Trankit model was trained from UD 2.5. Now UD releases 2.15 which has larger copora. But classical-Chinese is not in the list of [the trainable languages list](https://trankit.readthedocs.io/en/latest/pkgnames.html#trainable-languages). So I have...

This repo seems to adpot udpipe model to process NLP tasks in spaCy. Can we do it reversely? Because UDPipe model is very inaccurate for Asian languages.

请问支持古汉语分词吗？

1

谢谢！