A lot has been said about the use of classification metrics for binary targets. But how to evaluate the performance of multiclass classification models effectively?
In this video, I discuss essential metrics like accuracy, precision, recall, and F1-score in the context of multi-class targets. I explain the use of macro and micro averaging of the metrics for balanced and imbalanced datasets and how to interpret these values, highlighting their advantages, and also their limitations.
Want to learn more? Check out our course: https://www.trainindata.com/p/machine-learning-with-imbalanced-data
