Just leaving this here in case someone wants to play around with this on python 3.
While the SNN implemented by this is simple, it's one of the few practical uses of a basic SNN that is small enough to understand and has source code to look through. Therefore I've updated the code to conform to a more modular python approach (mostly by moving the modules from the training directory to a new parent module directory 'snn' and making the rest of the code depend on it)
I've also updated the code to be python 3 compatible, not rely on opencv for image reading and writing (since imageio is part of the standard scipy toolkit), and I've converted most of the tabs to spaces although this still requires more formatting.
I don't expect this to be pulled in and merged, at least not without the deleted Readme's being replaced with comments in the snn module and a proper re-format. But if anyone else looks here for a reference, this should save them some time.
Just leaving this here in case someone wants to play around with this on python 3.
While the SNN implemented by this is simple, it's one of the few practical uses of a basic SNN that is small enough to understand and has source code to look through. Therefore I've updated the code to conform to a more modular python approach (mostly by moving the modules from the training directory to a new parent module directory 'snn' and making the rest of the code depend on it)
I've also updated the code to be python 3 compatible, not rely on opencv for image reading and writing (since imageio is part of the standard scipy toolkit), and I've converted most of the tabs to spaces although this still requires more formatting.
I don't expect this to be pulled in and merged, at least not without the deleted Readme's being replaced with comments in the snn module and a proper re-format. But if anyone else looks here for a reference, this should save them some time.