Sklearn. Pipeline. Several transformers. get_feature_names_out

Question

I'he realised custom transformer of sklearn, where I porcess a column of text data.

I create a pipeline, where I combine two transofrmers - NameTransformer, OneHotEncoder. But I have got an error.

from sklearn.base import BaseEstimator, TransformerMixin
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import OneHotEncoder
import re

import pandas as pd

def get_title_cat(row):
    name = str(row.Name)
    return name.split()[0]

class NameTransformer(BaseEstimator, TransformerMixin):

    def __init__(self) -> None:
        super().__init__()
        self.name_col = 'Name'

    def fit(self, X, y=None, **fit_params):
        return self

    def transform(self, X: pd.DataFrame, y=None, **fit_params):
        return pd.DataFrame(X.apply(get_title_cat, axis='columns'), columns=['Title_cat'])

    def get_feature_names_out(self):
        return ['Title_cat']


name_pipe = Pipeline([
    ('name_cat', NameTransformer()), 
    ('ohe_tr', OneHotEncoder(handle_unknown='infrequent_if_exist', sparse_output=False))
    ])


data = {'Name': ['Braund, Mr. Owen Harris', 
                 'Cumings, Mrs. John Bradley (Florence Briggs Thayer)', 
                 'Heikkinen, Miss. Laina', 
                 'Byles, Rev. Thomas Roussel Davids']}

df = pd.DataFrame(data)

name_pipe.fit_transform(df)
name_pipe.get_feature_names_out()

How to correctly realise get_feature_names_out to not get this error?

    750     if not hasattr(transform, "get_feature_names_out"):
    751         raise AttributeError(
    752             "Estimator {} does not provide get_feature_names_out. "
    753             "Did you mean to call pipeline[:-1].get_feature_names_out"
    754             "()?".format(name)
    755         )
--> 756     feature_names_out = transform.get_feature_names_out(feature_names_out)
    757 return feature_names_out

TypeError: get_feature_names_out() takes 1 positional argument but 2 were given

Answer 1

You defined the function with just one argument:

def get_feature_names_out(self):
    return ['Title_cat']

But you call it with 2 arguments. Either call it without argument

transform.get_feature_names_out()

or add a second argument to your function definition.

def get_feature_names_out(self, feature_names_out):
    return ['Title_cat']