In AutoML (Automated Machine Learning) frameworks, for classification problems, multiple evaluation metrics are automatically computed for each model run to assess performance comprehensively. These include:

• Accuracy: The proportion of true results (true positives and true negatives) among all cases.
• Area Under the ROC Curve (AUC-ROC): Indicates the model's ability to distinguish between classes, with 1 being perfect and 0.5 no better than random.
• Recall: The proportion of actual positives correctly identified (sensitivity).
• F1 Score: The harmonic mean of precision and recall, balancing the two.

AutoML platforms compute these metrics to offer a detailed view of each model's performance, aiding in selecting the most suitable model based on specific application needs.