Metadata-Version: 2.3
Name: ksom
Version: 0.9.1
Summary: Self Organising Maps in pytorch
Project-URL: Homepage, https://github.com/mdaquin/ksom
Project-URL: Issues, https://github.com/mdaquin/ksom/issues
Author-email: Mathieu d'Aquin <mathieu.daquin.w+ksom@gmail.com>
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Requires-Python: >=3.8
Description-Content-Type: text/markdown

# KSOM - Simple, but relatively efficient, pytorch-based self organising map

This is a simple implementation of self-organising map training in python, using pytorch for efficiency. Check ``test-img.py`` for a simple use case creating a square color map of an image. To use, try
```
pip install -r requirements.txt
```
and then
```
python test/test_img.py chica.jpg 6
```
This shoud show an updating 6x6 map of the colours in the ``chica.jpg`` image. Change the image or the size of the map by changing the corresponding parameters.

Another example is included in the ``test_cheese.py`` creating a map of cheeses based on various binary attributes.