"1. If Training Dataset has 'n' cardinal values for any categorical column, it creates 'n' or 'n-1'(in case of dropfirst) dummy columns as per **ONE HOT ENCODING**.\n", "2. Due to the above, modified ...
Using sklearn.preprocessing.OneHotEncoder() with parameter handle_unknown='ignore' will get a wrong output, when I manually set categories_ && all objects are of string type && an unknown category is ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback