无法创建数据库

[Rex-dby]

(chat) PS C:\Users\admin\Desktop\LLM_email_sys> chatchat kb -r --embed-model="C:\Users\admin\Desktop\LLM_email_sys\BAAI\bge-small-zh-v1.5"
recreating all vector stores
C:\Users\admin\anaconda3\envs\chat\lib\site-packages\langchain\_api\module_import.py:87: LangChainDeprecationWarning: Importing GuardrailsOutputParser from langchain.output_parsers is deprecated. Please replace the import with the following:
from langchain_community.output_parsers.rail_parser import GuardrailsOutputParser
  warnings.warn(
2025-05-12 15:33:02.182 | INFO     | chatchat.server.knowledge_base.kb_cache.faiss_cache:load_vector_store:109 - loading vector store in 'samples/vector_store/C_\Users\admin\Desktop\LLM_email_sys\BAAI\bge-small-zh-v1.5' from disk.
2025-05-12 15:33:02.204 | ERROR    | chatchat.server.utils:get_Embeddings:341 - failed to create Embeddings for model: C:\Users\admin\Desktop\LLM_email_sys\BAAI\bge-small-zh-v1.5.
2025-05-12 15:33:02.210 | ERROR    | chatchat.server.knowledge_base.kb_cache.faiss_cache:load_vector_store:140 - 'NoneType' object has no attribute 'embed_documents'
2025-05-12 15:33:02.211 | ERROR    | chatchat.init_database:worker:61 - 向量库 samples 加载失败。

使用本地embedding模型报错, 可以对话功能正常, 有大佬知道怎么做吗

[dosubot]

Hello @Rex-dby! I'm Dosu, a bot here to help you with any bugs, questions, or contributions you might have while you wait for a human maintainer. Let's tackle this issue together!

To resolve the error when using a local embedding model that prevents the creation of a database, you can try the following steps:

1. Reinitialize the Knowledge Base: Use the command to recreate all vector stores:

   python chatchat/cli.py kb --recreate-vs

2. Initialize with Specific Models: If you encounter a vector store loading failure, it might be due to the default embedding model not being found. Initialize Langchain-Chatchat with specific LLM and embedding models using:

   chatchat init -l qwen1.5-chat -e bge-m3 -r

   • -l specifies the LLM model (e.g., qwen1.5-chat)
   • -e specifies the embedding model (e.g., bge-m3)
   • -r is used for initialization

3. Check Embedding Model Accessibility: Ensure that the embedding model is accessible by using the check_embed_model function. This function verifies the accessibility of the specified embedding model and defaults to the standard model if none is provided. If the model is inaccessible, it logs an error message indicating the failure.

These steps should help in resolving the issue with the local embedding model and the database creation ^[1][2].

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Bug Report | Other}

[azhe1234]

(chat) PS C:\Users\admin\Desktop\LLM_email_sys> chatchat kb -r --embed-model="C:\Users\admin\Desktop\LLM_email_sys\BAAI\bge-small-zh-v1.5"
recreating all vector stores
C:\Users\admin\anaconda3\envs\chat\lib\site-packages\langchain\_api\module_import.py:87: LangChainDeprecationWarning: Importing GuardrailsOutputParser from langchain.output_parsers is deprecated. Please replace the import with the following:
from langchain_community.output_parsers.rail_parser import GuardrailsOutputParser
  warnings.warn(
2025-05-12 15:33:02.182 | INFO     | chatchat.server.knowledge_base.kb_cache.faiss_cache:load_vector_store:109 - loading vector store in 'samples/vector_store/C_\Users\admin\Desktop\LLM_email_sys\BAAI\bge-small-zh-v1.5' from disk.
2025-05-12 15:33:02.204 | ERROR    | chatchat.server.utils:get_Embeddings:341 - failed to create Embeddings for model: C:\Users\admin\Desktop\LLM_email_sys\BAAI\bge-small-zh-v1.5.
2025-05-12 15:33:02.210 | ERROR    | chatchat.server.knowledge_base.kb_cache.faiss_cache:load_vector_store:140 - 'NoneType' object has no attribute 'embed_documents'
2025-05-12 15:33:02.211 | ERROR    | chatchat.init_database:worker:61 - 向量库 samples 加载失败。

使用本地embedding模型报错, 可以对话功能正常, 有大佬知道怎么做吗

在model_settings指定默认的向量模型呢，不要直接使用参数行指定向量模型试试