vbnigmm 2.2.0

Variational Bayes algorithm for normal inverse Gaussian mixture models

Variational Bayes algorithm for Normal Inverse Gaussian Mixture Models

Demonstration

The results of the sample simulation data can be checked by the following procedure:

poetry run jupyter lab
# open example.ipynb in jupyter environment

Installation

The package can be build using poetry and installed using pip:

poetry build
pip install dist/vbnigmm-2.0.0-py3-none-any.whl

Examples

If you want to apply vbnigmm to your data, you can run the following code:

from vbnigmm import NormalInverseGaussMixture as Model

# x is numpy.ndarray of 2D

model = Model()
model.fit(x)
label = model.predict(x)

Citation

If you use vbnigmm in a scientific paper, please consider citing the following paper:

Takashi Takekawa, Clustering of non-Gaussian data by variational Bayes for normal inverse Gaussian mixture models. arXiv preprint arXiv:2009.06002 (2020).