category

Categorical transformation for data science

Installation

pip install works for this library.

pip install category

Single Category

>>> from category import Category
>>> book = Category(['a', 'b', 'c', 'Category_d', 'e', 'f', 'g', 'h', 'i', 'j'], pad_mst = False)
>>> book.i2c[2]
'c'

>>> book.c2i[['Category_d','f']]
array([3, 5])

You can set pad_mst to True to handle the missing token

>>> from category import Category
>>> book = Category(['a', 'b', 'c', 'Category_d', 'e', 'f', 'g', 'h', 'i', 'j'], pad_mst = True)
>>> book.i2c[2] # the 1st token is the missing token, not 'a' any more
'b'
>>> book.c2i[['Stranger','Category_d','Unknown','f']]
array([0, 4, 0, 6])