cgre-aachen / bayseg Goto Github PK

An unsupervised machine learning algorithm for the segmentation of spatial data sets.

License: GNU Lesser General Public License v3.0

Python 0.47% MATLAB 0.35% Jupyter Notebook 99.18%

segmentation python3 hidden-markov-models bayesian-methods gaussian-mixture-models gibbs-energy gibbs-sampling markov-chain machine-learning mixture-model

bayseg's Issues

information entropy for results

labels probability
1d
2d
a:b iter

[2D] MCR calculation

[2D] vectorized 8-stamp

test function for gibbs energy calculation

function for label matching

clean up/out jupyter notebooks

2D Gibbs energy calculation

Vectorized implementation

use numpy ndarrays for storage variables

requirements.txt file for dependecies

2d coloring for uneven extent

storage for acceptance ratios

MAP estimate of the segmentation

maximum a posterior

Take into account np.nan values for gibbs energy and labels draw

Add pip support

Steps to add pip support:
add pyproject.toml

[build-system]
requires = ["setuptools>=62"]
build-backend = "setuptools.build_meta"

and setup.cfg

[metadata]
name = bayseg
version = 0.0.1
author = Alexander Schaaf
author_email = [email protected]
description = An unsupervised machine learning algorithm for the segmentation of spatial data sets. 
long_description = file: README.md
long_description_content_type = text/markdown
url = https://github.com/cgre-aachen/bayseg
project_urls =
    Bug Tracker = https://github.com/cgre-aachen/bayseg/issues
classifiers =
    Programming Language :: Python :: 3
    License :: OSI Approved :: GNU Lesser General Public License v3 (LGPLv3)
    Operating System :: OS Independent

[options]
package_dir =
	= src
packages = find:
python_requires = >=3.6
install_requires = 
	pytest
	scipy
	numpy
	pandas
	matplotlib
	tqdm
	scikit_learn

[options.packages.find]
where = src

move the directory bayseg to the src directory