bias_variance_decomposition for multi-output regressor

[Baghapour]

Hi, I would appreciate to know how I can use the bias_variance_decompostion function in mlxtend for a multi-output regressor?
(For instance, using the function with MultiOutputRegressor(GradientBoostingRegressor()) model)
Thanks a lot

[rasbt]

Good question. I am not sure this is currently supported. At least, I haven't tried :P

[Baghapour]

Thanks Sebastian, I hope it will be considered sometime later in mlxtend since it would be great to see also the effect of targets interactions by such an error decompaction.