CODEWITHSUNDEEP
pythonscikit-learnlogistic-regression
Andrew Buchanan
Andrew Buchanan

Reputation: 81

Custom Scaling features in Logistic Regression using sklearn

I am doing a course on udemy and I am using a custom scaler function they've provided to scale the features that aren't dummies in the dataset but I am receiving an error. I am pretty certain that I've followed the code line for line, but I could have missed something

I've attached the code and a file to my github

Below are the functions:

#Custom Scaler to avoid scaling dummies

from sklearn.base import BaseEstimator, TransformerMixin
from sklearn.preprocessing import StandardScaler

class CustomScaler(BaseEstimator, TransformerMixin):
    
    def _init_(self,columns, copy=True, with_mean=True, with_std=True):
        self.scaler = StandardScaler(copy, with_mean, with_std) 
        self.columns = columns
        self.mean_ = None
        self.var_ = None

        
    def fit(self, X, y=None):
        self.scaler.fit(X[self.columns], y)
        self.mean_ = np.mean(X[self.columns]) 
        self.var_ = np.var(X[self.columns])
        return self
    
    def transform(self, X, y=None, copy=None):
        init_col_order = X.columns
        X_scaled = pd.DataFrame(self.scaler.transform(X[self.columns]), columns=self.columns)
        X_not_scaled = X.loc[:,~X.columns.isin(self.columns)]
        return pd.concat([X_not_scaled, X_scaled], axis=1)[init_col_order]

Then I pull the columns that I want to scale into a variable and call the function:

unscaled_inputs.columns.values

columns_to_scale=['Month Value','Day of the Week', 'Transportation Expense', 'Distance to Work',
       'Age', 'Daily Work Load Average', 'Body Mass Index', 'Children', 'Pets']

absenteeism_scaler = CustomScaler(columns_to_scale)

Then the following error is outputted:

TypeError                                 Traceback (most recent call last)
<ipython-input-26-c6390babc5b1> in <module>
----> 1 absenteeism_scaler = CustomScaler(columns_to_scale)

TypeError: CustomScaler() takes no arguments

This seems weird as I thought ordinarily this was avoided by including self in the parentheses of the functions.

If anyone is able to help I'd be most grateful!

Upvotes: 0

Views: 614

Answers (1)

afsharov
afsharov

Reputation: 5164

Your class definition has a typo. The __init__() function is a special method to instantiate customized objects. Observe that it has two leading and two trailing _ enclosing the word init, while yours has only one on each side.

The correct definition should be:

class CustomScaler(BaseEstimator, TransformerMixin):
    
    def __init__(self,columns, copy=True, with_mean=True, with_std=True):
        self.scaler = StandardScaler(copy, with_mean, with_std) 
        self.columns = columns
        self.mean_ = None
        self.var_ = None

    ...

Now you can pass columns_to_scale to the constructor of CustomScaler as you do in your code snippet:

absenteeism_scaler = CustomScaler(columns_to_scale)

Upvotes: 1

Related Questions