After the summer students started using samromur I saw that a db query was taking up all the cpu from the database on amazon.
This branch optimizes it and only runs it when required.
The optimization is very simple, the count(*) was very heavy when all the clips where included in the search, filtering out the standard dataset 'samromur' from the query made it manageable.
I instead add the default dataset as an option in labels in the frontend.
After the summer students started using samromur I saw that a db query was taking up all the cpu from the database on amazon.
This branch optimizes it and only runs it when required.
The optimization is very simple, the count(*) was very heavy when all the clips where included in the search, filtering out the standard dataset 'samromur' from the query made it manageable.
I instead add the default dataset as an option in labels in the frontend.