ssb-klass-python / KLASS

A Python package built on top of Statistics Norway's code- and classification-system "KLASS".
The package aims to make Klass's API for retrieving data easier to use by re-representing Klass's internal hierarchy as python-classes. Containing methods for easier traversal down, search classes and widgets, reasonable defaults to parameters etc. Where data is possible to fit into pandas DataFrames, this will be preferred, but hiererachical data / objects containing metadata will be kept as json / dict structure.

Installing

The package is available on Pypi, and can be installed by for example poetry like this:

poetry add ssb-klass-python

Example usages

Getting started

from klass import search_classification
# Opens a ipywidget in notebooks for searching for classifications and copying code, to get started
search_classification(no_dupes=True)

Getting a classification directly

from klass import get_classification # Import the utility-function
nus = get_classification(36)

# Does the same as the code above, but does not shy away from using the class directly
from klass import KlassClassification # Import the class for KlassClassifications
nus = KlassClassification(36)  # Use ID for classification

When you have the classification stored in an object, you can "dig into" the API from there.

codes = nus.get_codes() # codes from current date
print(codes)
codes.data  # Pandas dataframe available under the .data attribute

From searching through "families", down to a specific codelist

from klass import KlassSearchFamilies
search = KlassSearchFamilies(360)
print(search)
>>> "Family ID: 20 - Utdanning - Number of classifications: 5"
utdanning = search.get_family(20)
print(utdanning)
>>> "The Klass Family "Utdanning" has id 20."
>>> "And contains the following classifications:"
>>>  "36: Standard for utdanningsgruppering (NUS)"
nus = utdanning.get_classification(36)
print(nus)
>>> "Classification 36: Standard for utdanningsgruppering (NUS)..."
nus_codes = nus.get_codes("2023-01-01")
print(nus_codes)
>>> "Codelist for classification: 36"
>>> " From date: 2023-01-01"
nus_codes.data  # A pandas dataframe