Given the refactor of outputs, one idea would be to make how much is written out configurable.
It may be useful to sometimes only output group level results, for example, if running a model on a very large dataset, as the results might be quite large in that case.
Another idea is to allow run level outputs to serve as inputs to a new group level analysis, to prevent re-running and writing on first level models.
Given the refactor of outputs, one idea would be to make how much is written out configurable.
It may be useful to sometimes only output group level results, for example, if running a model on a very large dataset, as the results might be quite large in that case.
Another idea is to allow run level outputs to serve as inputs to a new group level analysis, to prevent re-running and writing on first level models.