eosphoros
Replies: 1 comment 1 reply
-
|
The current performance of my use of llama-2-13b is quite impressive. The main challenge lies in how to extract the desired data from the vector database. Currently, this aspect is not well achieved in db-gpt. |
Beta Was this translation helpful? Give feedback.
1 reply
-
|
Of course, varying individual needs contribute to this, but speaking for my own requirements, I need to make adjustments to the vector database. |
Beta Was this translation helpful? Give feedback.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
How to evaluate the performance of different models in DB-GPT KBQA？
Beta Was this translation helpful? Give feedback.
All reactions