How do I evaluate my model's prediction results?

To evaluate your model's prediction results, you can use the various Kumo graphs and metrics to evaluate prediction results under different problem settings, and analyze explanations regarding what they might mean in a particular scenario. Different metrics will apply depending on the problem or question you're trying to answer—whether you're dealing with a binary classification, multiclass classification, multilabel classification, regression, or link prediction problem.

Learn More:

• Evaluation
• Evaluation Metrics