Thank you for your answer! The links were really helpful too. Question - I did look at the sklearn implementation but it says "find their average". Is there a way I can extract just the sample-wise recall without calculating the average?

Have you tried BERT? It's a pre-trained transfer learning model that you can feed your context-specific data into.

Ah, that makes sense. I have a serious shortage of data...the dataset I have itself is too small. I'm looking for ways to really apply ML on the dataset I have, let alone improving accuracy.

Hi Carlos! Thanks for your answer. Will smote work if I eliminated the classes with only one example? Can I use smote for multilabel classification then?