The main problem is that BOR remains quite unstable when adding or deleting files. Additionally, there are bugs affecting the file view generation. To resolve these issues, a deeper investigation into the /update_file, /add_file, and /delete_file endpoints in BOR's code is necessary. These endpoints are closely related to the embeddings mechanism of the MemgraphManager.py file.
One known issue causing these bugs is related to the query length. Queries must be broken down into smaller segments. For example:
INFO: 172.25.0.1:47514 - "PUT /knowledge_base/notes/add_file HTTP/1.1" 500 Internal Server Error
2024-03-06 09:00:42 ERROR: Exception in ASGI application
...
openai.BadRequestError: Error code: 400 - {'error': {'message': "This model's maximum context length is 8192 tokens. However, your messages resulted in 9063 tokens. Please reduce the length of the messages.", 'type': 'invalid_request_error', 'param': 'messages', 'code': 'context_length_exceeded'}}
This error suggests that the query length exceeds the permissible limit, leading to server errors. Contributions to resolve these or any other challenges with BOR endpoints are welcome.
Feel free to contribute, any input or assistance is highly appreciated!
The main problem is that BOR remains quite unstable when adding or deleting files. Additionally, there are bugs affecting the file view generation. To resolve these issues, a deeper investigation into the
/update_file,/add_file, and/delete_fileendpoints in BOR's code is necessary. These endpoints are closely related to the embeddings mechanism of theMemgraphManager.pyfile.One known issue causing these bugs is related to the query length. Queries must be broken down into smaller segments. For example:
This error suggests that the query length exceeds the permissible limit, leading to server errors. Contributions to resolve these or any other challenges with BOR endpoints are welcome.
Feel free to contribute, any input or assistance is highly appreciated!