2018_08_07_step_03.py

def create_manufacturer(DataFrame, N_cases = 3):
    """
    return the top_N + 1 manufacturers in one-hot encoding format
    """
    man = DataFrame['name'].apply(lambda x: x.split(' ')[0])
    cases = man.value_counts().index
    
    manufacturer_encoded = man.replace(cases[N_cases::], 
                                       len(cases[N_cases::]) * ['OTHER_MAN'])
    
    return pd.get_dummies( manufacturer_encoded )


class cat_cleaner(BaseEstimator, TransformerMixin):
    """
    Creates and encoded the manufacturer variable
    
    Inputs:
        N_cases = Numer of N + 1 manufacturers to be encoded
    """
        
    def __init__(self, n_cases=2):
        self.n_cases = n_cases
    
    def fit(self, X, y=None):
        return self
    
    def transform(self, X):
        assert isinstance(X, pd.DataFrame)
        return create_manufacturer(X, N_cases = self.n_cases)

# create a tupple with the name transformer
cat_data_cleaner = ('cat_cleaner', cat_cleaner(n_cases=5) )