Allow to provide custom column name for transformed feature

[ragrawal]

I have a datetime values in a column, "submitted_at", and I am trying to construct two features out of it, namely: day of week and hour of day. However the output dataframe contains "submitted_at" twice. Once for day of week feature and another for hour of day feature.

[dukebody]

Hi @ragrawal . Can you post the code you are using? You can always change afterwards the name of the output columns to the ones you want.

[dukebody]

Sorry, didn't really read the title of the issue. What do you think about an API like:

mapper = DataFrameMapper([
    ('target', Transformer1(), 'transformed_target_1'),
    ('target', Transformer2(), 'transformed_target_2')
])

Meaning that the result of applying Transformer1 to the column target should be placed in a column with name transformed_target_1, and equivalent for the second one?

This is probably quite easy to do by passing the chosen names to the get_names method, if present.

[ragrawal]

submitted a pull request related to this: #91

[dukebody]

Fixed in #92.