Reordering categories

Working with Categorical Data in Python

Kasey Jones

Research data scientist

Why would you reorder?

  1. Creating a ordinal variable
  2. To set the order that variables are displayed in analysis
  3. Memory savings
Working with Categorical Data in Python

Reordering example

dogs['coat'] = dogs["coat"].cat.reorder_categories(
  new_categories = ['short', 'medium', 'wirehaired', 'long'],

  ordered=True
)

Using inplace:

dogs["coat"].cat.reorder_categories(
  new_categories = ['short', 'medium', 'wirehaired', 'long'],
  ordered=True,

  inplace=True
)
Working with Categorical Data in Python

Grouping when ordered=True

dogs['coat'] = dogs["coat"].cat.reorder_categories(
  new_categories = ['short', 'medium', 'wirehaired', 'long'],
  ordered=True
)

dogs.groupby(by=['coat'])['age'].mean()

coat
short         8.364746
medium        9.027982
wirehaired    8.424136
long          9.552056
Working with Categorical Data in Python

Grouping when ordered=False

dogs['coat'] = dogs["coat"].cat.reorder_categories(
  new_categories = ['short', 'medium', 'long', 'wirehaired'],

  ordered=False
)

dogs.groupby(by=['coat'])['age'].mean()

coat
short         8.364746
medium        9.027982
long          9.552056
wirehaired    8.424136
Working with Categorical Data in Python

Reordering practice

Working with Categorical Data in Python

Preparing Video For Download...