This class is intended for faster and simpler categorical/ label encoding on large data, in which sklearn's LabelEncoder might be too slow.

CategoricalEncoder.py

'''
Author: Kung-hsiang, Huang (Steeve)
Date: 2019/Mar/15
'''

class CategoricalEncoder():
    '''
    This class is for those operating on large data, in which sklearn's LabelEncoder class may take too much time.
    This encoder is only suitable for 1-d array/ list. You may modify it to become n-d compatible.
    '''
    def __init__(self):
        self.f_dict = {}
        self.r_dict = {}

    def fit(self, array):
        '''

        :param array: list or np array
        :return: None
        '''

        unique_elements = set(array)
        self.n_elements = 0
        for e in unique_elements:
            self.f_dict[e] = self.n_elements
            self.r_dict[self.n_elements] = e
            self.n_elements += 1

    def reverse_transform(self, transformed_array, to_np=False):
        '''

        :param transformed_array: list or np array
        :return: array: np array with the same shape as input
        '''


        array = [self.r_dict[e] for e in transformed_array]
        if to_np:
            array = np.array(array)
        return array


    def transform(self, array, to_np=False):
        '''

        :param array: array list or np array
        :return: list or np array with the same shape as the input
        '''
        transformed_array = [self.f_dict[e] for e in array]
        if to_np:
            transformed_array = np.array(transformed_array)
        return transformed_array

    def fit_transform(self, array, to_np=False):
        '''

        :param array: array list or np array
        :return: list or np array with the same shape as the input
        '''
        self.fit(array)
        return self.transform(array, to_np)