Welcome to skpro#
skpro
is a library for supervised probabilistic prediction and
tabular probability distributions in python.
Features#
skpro
provides unified, sklearn
and skbase
compatible interfaces to:
tabular supervised regressors for probabilistic prediction - interval, quantile and distribution predictions
metrics to evaluate probabilistic predictions, e.g., pinball loss, empirical coverage, CRPS
reductions to turn
sklearn
regressors into probabilisticskpro
regressors, such as bootstrap or conformalbuilding pipelines and composite models, including tuning via probabilistic performance metrics
symbolic probability distributions with value domain of pandas.DataFrame-s and pandas-like interface
Technical specification#
In-memory computation of a single machine, no distributed computing
Medium-sized data in pandas and NumPy based containers
Modular, principled and object-oriented API
Using interactive Python interpreter, no command-line interface or graphical user interface
Contents#
From here, you can navigate to:
Get started using skpro
quickly.
Find user documentation.
Understand skpro
’s API.
Find out how you can contribute.
See how the package has changed.
Learn more about skpro
.