You have an Azure Machine Learning workspace. You plan to use the workspace to set up automated machine learning training for an image classification model. You need to choose the primary metric to…

r2_score is a metric typically used for regression problems, measuring how well future samples are likely to be predicted by the model.

mean_absolute_error is a metric used for regression problems, measuring the average magnitude of errors in a set of predictions.

Intersection Over Union (IoU) is a standard metric used for image classification, particularly when the task involves object detection or semantic segmentation, which are common within the broader category of "image classification models" in Azure Machine Learning's Automated ML for computer vision. Optimizing for IoU helps improve the overlap between predicted and actual bounding boxes or segmentation masks.

median_absolute_error is also a metric used for regression problems, representing the median of all absolute differences between the prediction and actual value.