decision-tree-id3

decision-tree-id3 is a module created to derive decision trees using the ID3 algorithm. It is written to be compatible with Scikit-learn's API using the guidelines for Scikit-learn-contrib. It is licensed under the 3-clause BSD license.

Important Links

HTML Documentation - https://svaante.github.io/decision-tree-id3

Installation

Dependencies

• Python (>= 2.7 or >= 3.3)
• NumPy (>= 1.6.1)
• Scikit-learn (>= 0.17)

The package by itself comes with a single estimator Id3Estimator. To install the module:

pip install decision-tree-id3

or clone the project using:

git clone https://github.com/svaante/decision-tree-id3.git
cd decision-tree-id3
python setup.py install

Usage

If the installation is successful, you should be able to execute the following in Python:

>>> from sklearn.datasets import load_breast_cancer
>>> from id3 import Id3Estimator
>>> from id3 import export_graphviz

>>> bunch = load_breast_cancer()
>>> estimator = Id3Estimator()
>>> estimator = estimator.fit(bunch.data, bunch.target)
>>> tree = export_graphviz(estimator.tree_, 'tree.dot', bunch.feature_names)

And to generate a PDF of the decision tree using GraphViz:

dot -Tpdf tree.dot -o tree.pdf

There are a number of different default parameters to control the growth of the tree: - max_depth, the max depth of the tree. - min_samples_split, the minimum number of samples in a split to be considered. - prune, if the tree should be post-pruned to avoid overfitting and cut down on size. - gain_ratio, if the algorithm should use gain ratio when splitting the data. - min_entropy_decrease, the minimum decrease in entropy to consider a split. - is_repeating, repeat the use of features.

For more in depth information see the documentation https://svaante.github.io/decision-tree-id3