You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Add continuous and discrete variables as attributes to the Preprocessor
Problem:
Currently we have to pass continuous and discrete variable names to the fit and transform functions of the Preprocessor It would be better to pass those variable names only once to the fit function and then reuse this information in the transform function.
Reason:
As far as I am aware, the columns should never change for the Preprocessor so why should we pass them several times to the same object? Passing lists that are supposed to stay the same several times can cause errors and can also confuse the user.
Task Description
This issue will add continuous and discrete variable names as attributes to the Preprocessor object to be able to define those only once in the fit function or in the object creation.
The Preprocessor then should be refactored to use this attribute
Check if everything is still working as expected
Note:
The Preprocessor should still be able to preprocess a DataFrame that contains not all the variables (in case we want to use the same Preprocessor with data where a column is missing).
The text was updated successfully, but these errors were encountered:
Add continuous and discrete variables as attributes to the Preprocessor
Problem:
Currently we have to pass continuous and discrete variable names to the
fit
andtransform
functions of thePreprocessor
It would be better to pass those variable names only once to thefit
function and then reuse this information in thetransform
function.Reason:
As far as I am aware, the columns should never change for the
Preprocessor
so why should we pass them several times to the same object? Passing lists that are supposed to stay the same several times can cause errors and can also confuse the user.Task Description
Preprocessor
object to be able to define those only once in the fit function or in the object creation.Preprocessor
then should be refactored to use this attributeNote:
The Preprocessor should still be able to preprocess a
DataFrame
that contains not all the variables (in case we want to use the samePreprocessor
with data where a column is missing).The text was updated successfully, but these errors were encountered: