zhuzhongshu123
zhuzhongshu123
Could you provide additional debugging information, such as project configuration and trace logs?
The reason is that when extracting triples, the data returned by LLM is not in the correct JSON format. You can try customizing the triple_prompt based on your data. Please...
> > The reason is that when extracting triples, the data returned by LLM is not in the correct JSON format. You can try customizing the triple_prompt based on your...
Please refer to the quick start documentation： https://openspg.yuque.com/ndx6g9/docs_en/tx0gd5759hg4xi56 If you need further assistance or additional context, feel free to ask!
Additionally, you can also upgrade KAG version to 0.6, where the vectorizer has a default batch size set to 32.
> > > 提供 Schema-constraint 模式，知识库构建阶段，严格按照 Schema 的定义进行操作，从而实现更细粒度和更复杂的知识抽取。 > > > 这个模式在哪开启呢，是需要再schema定义中添加constrain关键字规则吗 > > > > > > You can refer to [KAG examples of Schema-constraint extractor](https://github.com/OpenSPG/KAG/blob/master/kag/examples/baike/kag_config.yaml) to get detail...
You can use the developer mode to see if it prints any error logs. Here is the documentation: https://openspg.yuque.com/ndx6g9/docs_en/rs7gr8g4s538b1n7#cikso
The log indicates that the LLM-based NER has failed. Could you check if the LLM service is available?
Thank you so much for your assistance! Would you be able to submit a PR to merge your changes into the master branch?