scikit learn - Using a transformer (estimator) to transform the target labels in sklearn.pipeline -
i understand 1 can chain several estimators implement transform method transform x (the feature set) in sklearn.pipeline. have use case transform target labels (like transform labels [1...k] instead of [0, k-1] , love component in pipeline. possible @ using sklearn.pipeline.?
no, pipelines pass y
through unchanged. transformation outside pipeline.
(this known design flaw in scikit-learn, it's never been pressing enough change or extend api.)
Comments
Post a Comment