This repository serves as a demonstration of how to setup unit tests for functions that are used in a data analysis. This is one step away from making a Python package.
The demo data analysis is shown here:
notebooks/eda.ipynb
Those literate code documents call a function, count_classes,
which lives in src/count_classes.py
and who calculates the number observations of each class of a data set.
The tests for count_classes live in tests/test-count_classes.py,
and the helper data for the tests lives in tests/helper-count_classes.py.
The test suite can be run via:
pytest tests/*
Jupyter, Python and the following packages:
- pandas
- pytest