VARISTA
Machine Learning Knowledge Center
Interpretability
Individual Conditional Expectation Plot

Individual Conditional Expectation Plot

The Individual Conditional Expectation Plot can show the dependence of the target variable on any explanatory variable as well as the Partial Dependence Plot (PDP).

PDP has a problem that it does not display correctly if there are correlations between arbitrarily selected explanatory variables and other explanatory variables, but ICE Plot improves this point.

The ICE Plot visualizes the degree of influence on the prediction with a single line for each instance.
Unlike PDPs, ICE plots allow you to visualize and display only one explanatory variable of your choice. (Only one explanatory variable, because multiple explanatory variables will be plotted.


VARISTAは機械学習モデルの開発、管理をノーコードで効率的に行うことができる新しいプラットフォームです。
データをお持ちでしたらすぐに始められますので、是非ともお試しください。

or

Made with
by VARISTA Team.
© COLLESTA, Inc. 2021. All rights reserved.