i am trying to figure out how to use the query_by + query_by_weights and receiving a strange result. when i do not use query_by_weights, i get the correct result as the top result. however when i use query_by_weights even if i sent every single weight to the same value (50), i receive a different result. in fact, the top result i was looking for is not even included in the top 100.

Could you give me an example of what you mean? A few sample documents, search query (with all search params) would be helpful to figure out what’s going on

could i dm this to you?

I’d prefer to respond publicly in #C01P749MET0 for the benefit of future users who might be looking for info around this in the future. It doesn’t have to be your actual dataset. Just an example to reproduce the issue you’re talking about would be helpful: https://gist.github.com/jasonbosco/7c3432713216c378472f13e72246f46b

i think i figured it out actually- in your docs you mention that the fields listed earlier in the query_by parameter are weighted higher. i assume when i supply the query_by_weights parameter, it overrides this higher weighting provided to the fields listed earlier in query_by. is there a place where i can learn more about how typesense assigns these higher weights to the terms that come earlier in query_by? would love to know what the actual weights are so i can tweak them when i supply my query_by_weights param