Metadata-Version: 2.1
Name: gmsdb
Version: 3.0
Summary: GMSDB Clusterer
Home-page: https://github.com/berng/GMSDB
Author: Oleg I.Berngardt
Author-email: berng@rambler.ru
License: UNKNOWN
Keywords: clusterization
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3.11
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
Requires-Dist: matplotlib
Requires-Dist: mlxtend
Requires-Dist: numpy
Requires-Dist: scikit-learn
Requires-Dist: scipy
Requires-Dist: statsmodels

# GMSDB Clusterer

# Use
pip install gmsdb

## Simple use (sklearn -like style), no speed improvement, recommended significance level alpha=0.1. 
Maximal number of gaussian components=50:

from gmsdb import GMSDB

clf=GMSDB(n_components=50)

clf.fit(X)

Y=clf.predict(X)

## Complex use (with speed improvement for stages 1 and 3):

clf=GMSDB(min_components=2,step_components=100,n_components=900,rand_search=1000,rand_level=0.5)

## Complex use (with custom significance level alpha=0.15):

clf=GMSDB(n_components=50,alpha_stage2=0.15,alpha_stage4=0.15)

## Verbose use (show debug information):

clf=GMSDB(n_components=50,verbose=True)


# Desciption:
https://arxiv.org/abs/2309.02623



