piso 0.2.0

Pandas Interval Set Operations: methods for set operations for pandas' Interval, IntervalArray and IntervalIndex

piso - pandas interval set operations

piso exists to bring set operations (union, intersection, difference + more) to pandas' interval classes, specifically

- pandas.Interval
- pandas.arrays.IntervalArray
- pandas.IntervalIndex

Currently, there is a lack of such functionality in pandas, although it has been earmarked for development. Until this eventuates, piso aims to fill the void. Many of the methods can be used via accessors, which can be registered to pandas.arrays.IntervalArray and pandas.IntervalIndex classes, for example:

>>> import pandas as pd
>>> import piso
>>> piso.register_accessors()

>>> arr = pd.arrays.IntervalArray.from_tuples(
...        [(1,5), (3,6), (2,4)]
...    )

>>> arr.piso.intersection()
<IntervalArray>
[(3, 4]]
Length: 1, closed: right, dtype: interval[int64]

The domain of the intervals can be either numerical, pandas.Timestamp or pandas.Timedelta.

A small case study using piso can be found in the user guide. Further examples, and a detailed explanation of functionality, are provided in the API reference.

Visit https://piso.readthedocs.io for the documentation.

Installation

piso can be installed from PyPI or Anaconda.

To install the latest version from PyPI::

python -m pip install piso

To install the latest version through conda-forge::

conda install -c conda-forge piso

Versioning

SemVer is used by piso for versioning releases. For versions available, see the tags on this repository.

License

This project is licensed under the MIT License

Acknowledgments

Currently, piso is a pure-python implentation which relies heavily on staircase and pandas. It is clearly designed to operate as part of the pandas ecosystem. The colours for the piso logo have been assimilated from pandas as a homage, and is not to intended to imply and affiliation with, or endorsement by, pandas.

Additionally, two classes have been borrowed, almost verbatim, from the pandas source code:

- `pandas.util._decorators.Appender`
- `pandas.core.accessor.CachedAccessor`