Variational Bayes algorithm for Normal Inverse Gaussian Mixture Models
The package can be build using poetry and installed using pip:
pip install vbnigmm
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)
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).