Python implementation of Hierarchical Probabilistic Principal Component Analysis proposed in [1].
HPPCA significantly improves dimesionality reduction performance by absorbing our prior knowledge about the group structure of the features and by decreasing the number of parameter from individual components.
originated from [1].
python setup.py install
or
pip install .
in the root of this repository.
- Python >3.5
- scikit-learn
- numpy
See examples
[1] Aiga Suzuki, Hayaru Shouno, "Generative Model of Textures Using Hierarchical Probabilistic Principal Component Analysis", Proc. of PDPTA’17, CSREA Press, pp.333-338, 2017.
Apache License v2.0, refer to LICENSE for more detail.
Aiga SUZUKI [email protected]