Luo Peng

Although using empty messages or a single message with empty content serves no practical purpose, I personally feel there is indeed a slight issue from the perspective of Python —...

From my observation, the built-in readers of `SimpleDirectoryReader` are mainly focus on parsing unstructured texts from various unstructured or structured files. As for a sample CSV as below: ```csv a,...

Hi @JulianOestreich90, in addition to the API you provided, llama.cpp also supports [OpenAI-compatible Chat Completions API](https://github.com/ggerganov/llama.cpp/tree/master/examples/server) (search `POST /v1/chat/completions`). If you use that API, I guess [OpenAI](https://docs.llamaindex.ai/en/stable/examples/llm/openai/) should suffice.

AFAIK, CondenseQuestionChatEngine is designed to rewrite the question, see [docs](https://docs.llamaindex.ai/en/stable/examples/chat_engine/chat_engine_condense_question/). > ... give me the same answer when I call query_engine. If this is the case, why is there a...

Could be off-topic, after reading through the implementation code: https://github.com/run-llama/llama_index/blob/631cb60b60af4569ffd67406704cfa199f171cff/llama-index-core/llama_index/core/chat_engine/condense_question.py#L114-L127 I'm considering whether it's better to condense the question only when `chat_history` is not empty, as it cloud reduce unnecessary...

I noticed that `CondensePlusContextChatEngine` has a special boolean property `skip_condense`, which controls whether to skip condensing: https://github.com/run-llama/llama_index/blob/f1ff1ebafccd6bfd3aec98ac325447c870072bf0/llama-index-core/llama_index/core/chat_engine/condense_plus_context.py#L138 As for `CondenseQuestionChatEngine`, this might be exactly what ~@logan-markewich~ (**UPDATE**: sorry, should be...

@Ninlawat-Puhu I'm afraid I don't fully understand your use case. To my understanding, a [RouterQueryEngine](https://docs.llamaindex.ai/en/stable/examples/query_engine/RouterQueryEngine/) typically uses a selector (backed by LLM), rather than a [ChatEngine](https://docs.llamaindex.ai/en/stable/module_guides/deploying/chat_engines/) (e.g. `CondenseQuestionChatEngine `), to...

See #41.

我也遇到了同样的问题，参考 @lycfight （感谢🙏）的思路对 PaddleOCR 进行了非官方修复： - 分支：https://github.com/RussellLuo/PaddleOCR/tree/hotfix-ppstructure-ocr - 修改：https://github.com/PaddlePaddle/PaddleOCR/compare/release/2.7...RussellLuo:PaddleOCR:hotfix-ppstructure-ocr 安装该修复版本： ```bash pip install git+https://github.com/RussellLuo/PaddleOCR.git@hotfix-ppstructure-ocr ``` 使用说明： ```bash # 官方命令：paddleocr --image_dir=ppstructure/docs/table/1.png --type=structure paddleocr --image_dir=ppstructure/docs/table/1.png --type=structurex ``` 如果大家也遇到了这个问题，希望这个修复版本能够起到缓解作用。最终，还是期待官方的解决方案！

@jzhang533 感谢认可，很荣幸可以给 PaddleOCR 社区作贡献！我尝试提交了一个 PR：#11916 🚀