Moka is an application tool developed for the GraphESN-SOM machine learning model. It allows the training of the model over a dataset and the usage of the trained model, showing the resulting SOM (Self-organizing map) and related informations. This tool, together with the GraphESN-SOM, could be extreamly useful to visually inspetting a trained model in order to understand what it have learned.
This application was developed for the final exam at the University of Pisa, in the field of Machine Learning, in the 2013.
See more in documents contained within the docs folder.