Find answers from the community

Updated 2 months ago

Hello Guys,

Hello Guys,
I have an hybrid Qdrant database (i.e sparse and dense vectors).
When querying, I notice that the scores are very weak, even if the answer is correct. it rarely exceeds 0.5.

Do you have any idea how can I properly set up the query_engine parameters ? (similarity_top_k, alpha,...) ?
Attachment
image.png
W
1 comment
Are you using normal embedding model or or something related to your language.
If the model is not good with your language data it may not be predicting correctly.


Second case could be that the splitting of data is not suiting to your needs. You can try either making smaller nodes or larger nodes and see if the similarity score increases.
Add a reply
Sign up and join the conversation on Discord