Skip to content

Downscale one-hot-encoded categorical variable differences for correlation compuation #75

@tiadams

Description

@tiadams

Current implementation just one-hot-encodes categorical variables, giving them in principle more weight for the computation of the correlation difference.

Correlation differences from columns created by one-hot-encoding should be downsclaed based on the number of categories for that column.

Metadata

Metadata

Assignees

Labels

enhancementNew feature or request

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions