Mapper using a self-organizing map (SOM) for dimensionality reduction.
This mapper provides a simple, but pretty fast implementation of a self-organizing map using an unsupervised training algorithm. It performs a ND -> 2D mapping, which can for, example, be used for visualization of high-dimensional data.
This SOM implementation uses squared Euclidean distance to determine the best matching Kohonen unit and a Gaussian neighborhood influence kernel.
Notes
Available conditional attributes:
(Conditional attributes enabled by default suffixed with +)
Parameters : | kshape : (int, int)
niter : int
learning_rate : float
iradius : float or None
enable_ca : None or list of str
disable_ca : None or list of str
|
---|
Provide access to the Kohonen layer.
With some care.