You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Submit "What is kurfessss's phone number?" as a question and hit send
Wait for a response
See non-normalized response (will respond with kurfessss's phone number is xxx)
Expected behavior
I expected the output response to be normalized / "correct"
Screenshots
If applicable, add screenshots to help explain your problem.
Additional context
Should be a pretty easy fix - famous last words from @cameron-toy (but in all actuality, it should be - just need to return the normalized Entity name in the output)
The text was updated successfully, but these errors were encountered:
Unfortunately, it won't be quite as easy as returning the normalized Entity name. The normalized Entity is just the Entity string without titles (Professor, Dr., etc). To actually get a normalized name, we'll have to modify the db_query and format_answer methods for each QA object.
We'd also have to find a way to describe how to identify each type of entity. For example, professor names could be normalized as "Professor {last_name}" or as "{first_name} {last_name}"
More and more I'm feeling like we should have some sort of single "identifying_name" column for each entity.
Describe the bug
Non-normalized responses from chatbot - (not capitalizing the first letter of the sentence, uses whatever was typed in).
Bug severity
Minor, non-blocker, purely aesthetic.
To Reproduce
Steps to reproduce the behavior:
Expected behavior
I expected the output response to be normalized / "correct"
Screenshots
If applicable, add screenshots to help explain your problem.
Additional context
Should be a pretty easy fix - famous last words from @cameron-toy (but in all actuality, it should be - just need to return the normalized Entity name in the output)
The text was updated successfully, but these errors were encountered: