Canonical Correlation Forests (CCFs)

CCFs are a decision tree ensemble method for classification and regression. CCFs naturally
accommodate multiple outputs, provide a similar computational complexity to random forests,
and inherit their impressive robustness to the choice of input parameters.

This implementation is completely done using Numpy and SciPy, which are open-sourced
numerical computing libraries.

CCFs results on Spiral Dataset	CCFs results on Camel Dataset

Setup

pip3 install -e .

(This code base was developed on Python3.6)

Run Examples

For classification example run the following command:

cd tests
python3 classification_example.py

For regression example run the following command:

python3 regression_example.py

Contribution

Any improvements to the code base are welcomed, submit a merge request.

Original Code

https://github.com/twgr/ccfs

Paper citation

@article{rainforth2015canonical,
  title={Canonical correlation forests},
  author={Rainforth, Tom and Wood, Frank},
  journal={arXiv preprint arXiv:1507.05444},
  year={2015}
}