Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Kohonen.py

Kohonen neural network implementation in Python.

Dependencies

  • python >= 2.6
  • numpy
  • matplotlib

Usage

python main.py mnist_path valid_count test_count train_count network_width
  • mnist_path: the cpickled mnist database
  • valid_count: the size of the validation dataset (<= 10000)
  • test_count: the size of the testing dataset (<= 10000)
  • train_count: the size of the training dataset (<= 60000)
  • network_width: the size of the neural network (between 4 and 30 is good)

For example:

python main.py data/mnist.pkl.gz 100 100 600 10

is a good starting point

More info on while1read.com.

About

Kohonen's Neural Network (with conscience) implementation in Python

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages