Per in-person discussion, this is a feature that substantially increases the complexity of FeatureMatrixExtractor implementation and understanding to support a use case we don't think exists. If it does, it can be substituted by creating a stripped-down FeatureMatrixExtractor for a list of all studies with low efficiency loss.
Per in-person discussion, this is a feature that substantially increases the complexity of
FeatureMatrixExtractorimplementation and understanding to support a use case we don't think exists. If it does, it can be substituted by creating a stripped-downFeatureMatrixExtractorfor a list of all studies with low efficiency loss.