Easily combine predictions from multiple Bayesian models using techniques including (pseudo) Bayesian model averaging, hierarchical stacking, and more!
Currently, the .predict method returns averaging/stacking model weights (which may indeed be predictions, particularly in the case of the hierarchical stacking model). Then, blending is done through a separate .blend method.
However, it would be more convenient/user friendly to move all of .predict functionality into .weights, then move the current .blend functionality into .predict. This way, the stacking models perform the typical .fit -> .predict workflow common to many ML/stats packages.
Currently, the
.predict
method returns averaging/stacking model weights (which may indeed be predictions, particularly in the case of the hierarchical stacking model). Then, blending is done through a separate.blend
method.However, it would be more convenient/user friendly to move all of
.predict
functionality into.weights
, then move the current.blend
functionality into.predict
. This way, the stacking models perform the typical.fit -> .predict
workflow common to many ML/stats packages.