Open shaliniml opened 1 week ago
It's difficult to say without knowing more. Why not just use get_matches
? Which version are you using? Also, what code are you currently using and could you provide an example?
Thanks for responding. We are matching current payment notes (non-semantic long text) with historical ones. Historical data will be large in lakhs. So, to make code more efficient planning to pass historical payment notes in fit and current payment notes in predict. Planning to retrain it monthly. I will also share code and sample data, but this is the brief
closing this issue and will create a new one from personal account
I was using get_matches() to get top 5 matches. Now, since moving to production thought of using Fit/Predict but seems it returns only top first matches for each item. Is there any other way to get top 5 matches in Fit/Predict