For a imbalanced binary classification model for F1_score, should I use micro, macro, or weighted?

Question

My understanding:

Micro -> should use this if I want to get a score that does not favour any class in particular.

macro -> Weighs each class equally. So