Global Explainers

Global explanations help to understand the model’s general behavior.

There are multiple forms of global explanations. For example, global explanations:

• Can identify the important features that the model considers when making its predictions.

• Highlight the relationship between different feature values and the model’s predictions.

• Present the instances that are most influential towards the prediction of a given class and value.

Supported global explanation techniques:

• Feature Permutation Importance Explanations
  • Overview
  • Description
  • Interpretation
  • Examples
  • References
• Feature Dependence Explanations
  • Overview
  • Description
  • Interpretation
  • Examples
  • References
• Accumulated Local Effects
  • Overview
  • Description
  • Interpretation
  • Examples
  • Disadvantages
  • References