New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Regression solutions which work with transformed input variables are not correctly combined to ensembles #3114

Open

HeuristicLab-Trac-Bot opened this issue Mar 11, 2021 · 0 comments

Assignees

Labels

bug

Milestone

HeuristicLab 3.3....

Collaborator

HeuristicLab-Trac-Bot commented Mar 11, 2021

Issue migrated from trac ticket # 3114

milestone: HeuristicLab 3.3.x Backlog | component: Problems.DataAnalysis | priority: medium

2021-03-11 16:46:54: @gkronber created the issue

The RegressionProblemData class allows to specify transformations for input variables. When using cross-validation the solutions of all folds are combined to an ensemble.

These two features in combination do not work correctly. The predictions produced by the ensemble are way off. Presumably, the transformations are not correctly applied for the ensemble solution.

HeuristicLab-Trac-Bot assigned gkronber

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment