How this method "similarity_search_with_score()" calculate the score. how you guys decided top_k = 5

[VikRathod]

Hello Team,
I just want to know how you guys calculating the scores. I am quite interesting for score. whether you are using cosine similarity or other techniques you are using.

[anakin87]

Hello @VikRathod!

This method/function is not part of the codebase.
You may be referring to another framework... 😃

[VikRathod]

@anakin87 Thanks,
Then how did you retrieve most similar documents based on query.
result = pipeline.run(query, params={"Retriever": {"top_k": 2}})

top_k :2 =?

[anakin87]

Please provide the rest of the code you are using...