A flat PDP hides ICE curves with opposite slopes. What does this reveal?
Classic MLLevel: IntermediateAveraging hides interactions
A flat PDP hides ICE curves with opposite slopes. What does this reveal?
Hints
- An average can cancel opposite changes.
- Other features may modify the response.
Answer
The average cancels heterogeneous model effects across observations, often due to interactions. Inspect subgroup or individual responses rather than concluding the model ignores the feature.
Source and licence: scikit-learn 1.7.2 — partial dependence · BSD-3-Clause · answer: Quantoros editorial