You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Partial dependencies can be computed directly from the data distribution in the leaves of the trees in the booster, rather than by calling predict, see references below. This tends to be faster. It would be nice if LightGBM allowed for direct computation of the partial dependencies using this method.
Motivation
Partial dependency plots are widely used to interpret a model. Providing a fast method for computing them would be a big asset.
The original idea goes back to Friedman, Jerome H. 'Greedy function approximation: a gradient boosting machine.' Annals of statistics (2001): 1189-1232.
The text was updated successfully, but these errors were encountered:
Closed in favor of being in #2302. We decided to keep all feature requests in one place.
Welcome to contribute this feature! Please re-open this issue (or post a comment if you are not a topic starter) if you are actively working on implementing this feature.
Summary
Partial dependencies can be computed directly from the data distribution in the leaves of the trees in the booster, rather than by calling
predict
, see references below. This tends to be faster. It would be nice if LightGBM allowed for direct computation of the partial dependencies using this method.Motivation
Partial dependency plots are widely used to interpret a model. Providing a fast method for computing them would be a big asset.
References
The text was updated successfully, but these errors were encountered: