This package is python wrapper for RuleKit library - a versatile tool for rule learning.
Based on a sequential covering induction algorithm, it is suitable for classification, regression, and survival problems.
NOTE: This package is a wrapper for Java library, and requires Java Development Kit version 8 or later to be installed on the computer. Both Open JDK and Oracle implementations are supported.
pip install rulekitIf you're running tests for the first time (or you want to update existing tests resources) you need to download tests resources from RuleKit repository. You can do it by running:
python tests/resources.py download
Runing tests:
In directory where setup.py file exists.
python -m unittest discover ./tests
from sklearn.datasets import load_iris
from rulekit.classification import RuleClassifier
X, y = load_iris(return_X_y=True)
clf = RuleClassifier()
clf.fit(X, y)
prediction = clf.predict(X)
print(prediction)Full documentation is available here
The software is publicly available under GNU AGPL-3.0 license. Any derivative work obtained under this license must be licensed under the AGPL if this derivative work is distributed to a third party. For commercial projects that require the ability to distribute RuleKit code as part of a program that cannot be distributed under the AGPL, it may be possible to obtain an appropriate license from the authors.