LLM:Yi-34B on OpenAIAPI
EMBEDDING:text2vec-large-chinese on OpenAIAPI
What happened
1、Setting remote Milvus Service as Vector DB,when starting the project, display error: ERROR connect vector store failed: Can't instantiate abstract class MilvusStore with abstract method truncate
2024-09-10 14:12:25 NoteBook.local dbgpt.rag.summary.db_summary_client[4741] WARNING socal_base_data, mysql summary error!Can't instantiate abstract class MilvusStore with abstract method truncate, detail: Traceback (most recent call last):
File "/XX/LLM/dbgpt/DB-GPT/dbgpt/rag/summary/db_summary_client.py", line 75, in init_db_summary
self.db_summary_embedding(item["db_name"], item["db_type"])
File "/XX/LLM/dbgpt/DB-GPT/dbgpt/rag/summary/db_summary_client.py", line 45, in db_summary_embedding
self.init_db_profile(db_summary_client, dbname)
File "/XX/LLM/dbgpt/DB-GPT/dbgpt/rag/summary/db_summary_client.py", line 95, in init_db_profile
vector_connector = VectorStoreConnector.from_default(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/XX/LLM/dbgpt/DB-GPT/dbgpt/serve/rag/connector.py", line 153, in from_default
return cls(real_vector_store_type, vector_store_config)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/XX/Documents/LLM/dbgpt/DB-GPT/dbgpt/serve/rag/connector.py", line 128, in init
raise e
File "/Users/shaozhijing/Documents/LLM/dbgpt/DB-GPT/dbgpt/serve/rag/connector.py", line 124, in init
client = self.connector_class(config)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: Can't instantiate abstract class MilvusStore with abstract method truncate
Search before asking
Operating system information
Linux
Python version information
DB-GPT version
main
Related scenes
Installation Information
[X] Installation From Source
[ ] Docker Installation
[ ] Docker Compose Installation
[ ] Cluster Installation
[ ] AutoDL Image
[ ] Other
Device information
CPU ECS, DB-GPT running under Proxy model
Models information
LLM:Yi-34B on OpenAIAPI EMBEDDING:text2vec-large-chinese on OpenAIAPI
What happened
1、Setting remote Milvus Service as Vector DB,when starting the project, display error: ERROR connect vector store failed: Can't instantiate abstract class MilvusStore with abstract method truncate 2024-09-10 14:12:25 NoteBook.local dbgpt.rag.summary.db_summary_client[4741] WARNING socal_base_data, mysql summary error!Can't instantiate abstract class MilvusStore with abstract method truncate, detail: Traceback (most recent call last): File "/XX/LLM/dbgpt/DB-GPT/dbgpt/rag/summary/db_summary_client.py", line 75, in init_db_summary self.db_summary_embedding(item["db_name"], item["db_type"]) File "/XX/LLM/dbgpt/DB-GPT/dbgpt/rag/summary/db_summary_client.py", line 45, in db_summary_embedding self.init_db_profile(db_summary_client, dbname) File "/XX/LLM/dbgpt/DB-GPT/dbgpt/rag/summary/db_summary_client.py", line 95, in init_db_profile vector_connector = VectorStoreConnector.from_default( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/XX/LLM/dbgpt/DB-GPT/dbgpt/serve/rag/connector.py", line 153, in from_default return cls(real_vector_store_type, vector_store_config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/XX/Documents/LLM/dbgpt/DB-GPT/dbgpt/serve/rag/connector.py", line 128, in init raise e File "/Users/shaozhijing/Documents/LLM/dbgpt/DB-GPT/dbgpt/serve/rag/connector.py", line 124, in init client = self.connector_class(config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ TypeError: Can't instantiate abstract class MilvusStore with abstract method truncate
Milvus vector db config
VECTOR_STORE_TYPE=Milvus MILVUS_URL=127.0.0.1 MILVUS_PORT=19530 MILVUS_USERNAME= MILVUS_PASSWORD= MILVUS_SECURE=
2、on uploading docs to Knowledge database,errors displayed blew
document add error (pymysql.err.IntegrityError) (1048, "Column 'doc_token' cannot be null") [SQL: INSERT INTO knowledge_document (doc_name, doc_type, doc_token, space, chunk_size, status, last_sync, content, result, vector_ids, summary, gmt_created, gmt_modified, questions) VALUES (%(doc_name)s, %(doc_type)s, %(doc_token)s, %(space)s, %(chunk_size)s, %(status)s, %(last_sync)s, %(content)s, %(result)s, %(vector_ids)s, %(summary)s, %(gmt_created)s, %(gmt_modified)s, %(questions)s)] [parameters: {'doc_name': 'XXX.pdf', 'doc_type': 'DOCUMENT', 'doc_token': None, 'space': 'db1', 'chunk_size': 0.0, 'status': 'TODO', 'last_sync': datetime.datetime(2024, 9, 10, 14, 53, 57, 799748), 'content': '/LLM/dbgpt/DB-GPT/pilot/data/db1/XXX.pdf', 'result': '', 'vector_ids': None, 'summary': '', 'gmt_created': datetime.datetime(2024, 9, 10, 14, 53, 57, 799927), 'gmt_modified': datetime.datetime(2024, 9, 10, 14, 53, 57, 799931), 'questions': None}] (Background on this error at: https://sqlalche.me/e/20/gkpj)
What you expected to happen
1、EMBEDDING db schema correct 2、RAG with Milvus running correct
How to reproduce
1、setting Milvus as Vector in .env 2、setting MySQL DB as datasouce of chatdata 3、restart project
Additional context
No response
Are you willing to submit PR?