(UAR or accuracy)[https://ogunlao.github.io/blog/2021/04/24/consider_uar_accuracy.html]
accuracy: most used metric to valuate classification tasks in ML: ratio of the number of correct predictions to the total number of examples.
it has limitations, which is why other metrics like precision, recall, F1-score etc. have been used to mitigate these problems.
what is UAR?
UAR stands for unweighted average recall - and this is a good metric to optimize when the sample class ratio is imbalanced and it is closely related to accuracy.