An implementation of Kohonen's self-organizing maps with a number of distinguishing features:
-
An easy to use interface for building and evaluating self-organizing maps.
-
Automatic centroid detection and visualization using starbursts.
-
Two models of the data: (a) a self organizing map model, (b) a centroid based clustering model.
-
A number of easily accessible quality metrics for the self organizing map and the centroid based cluster model.
Other documents:
-
For a worked Python example check this notebook on Kaggle.
-
For an example in R check this notebook on Kaggle.
-
The Python documentation can be found here.
-
For the R documentation as part of CRAN check here.