If the input data has a different index, the reverse transformed data may be out of order

[amontanez24]

Environment Details

Please indicate the following details about the environment in which you found the bug:

• RDT version: Any
• Python version: Any
• Operating System: Any

Error Description

If the DataFrame input into theHyperTransformer has an index that isn't in incremental order, the reverse transformed data might come back in the wrong order. Some transformers reset the index and when the BaseTransformer assigns the new columns it is sometimes out of order. This happens in the following lines because the rows of columns_data will reorder to match the index of data.

RDT/rdt/transformers/base.py

Lines 126 to 130 in 10a996e

	def _set_columns_data(data, columns_data, columns):
	if len(columns_data.shape) == 1:
	data[columns[0]] = columns_data
	else:
	data[columns] = columns_data

Steps to reproduce

data = pd.DataFrame({
    'integer': [1, 2, 1, 3],
    'float': [0.1, 0.2, 0.1, 0.1],
    'categorical': ['a', 'a', 'b', 'a'],
    'bool': [False, False, True, False],
    'names': ['Jon', 'Arya', 'Jon', 'Jon'],
})
ht = HyperTransformer(data_type_transformers={'categorical': CategoricalTransformer})
ht.fit(data)
transformed = ht.transform(data)
reversed = ht.reverse_transform(data)