Open guyko81 opened 3 years ago
cc @stefanwebb
@guyko81 the .inv
method swaps the forward and inverse operations and returns a new transform, but it keeps the same parameters. Have you tried called transform._inverse(Y_train)
on the original transform?
@stefanwebb no, I wasn't aware of the function name, thanks for sharing! I'll check and get back if it doesn't work, thank you!
Hi,
I can't find the proper way to make a Spline transform inverse. What I would like to do is a 2 step approach: 1) train a Spline transform from a 1D Standard Normal distribution on a data
2) get the "more Normal" version of the target variable, And there is my problem, I don't know how to inverse the Spline transform and run through the Y_train on it to get back the "normalized" version of it. I would think something like that, but please help me - couldn't find such a thing in the documentation.
Apparently the parameters for transform.inv is not implemented. The following throws an error: AttributeError: '_InverseTransform' object has no attribute 'parameters'
While the normal version (without the .inv) of it prints out the parameters. What am I missing?