First layer drawing information from bottom source layer.
First layer drawing information from bottom source layer.
7exy is a neural network framework written in python. It is designed to be very general. 7exy will observe an information stream and try derive structure from it by taking many distributed samples and applying statistical methods.
It is released under the BSD 2-clause license to Joseph Wright 2011.
A network in 7exy is built with nodes and layers. A network is made of stacked layers. Layers are a set of nodes. Nodes are analogous to neurons and do manage the information processing whereas the layer and the network structures are there for organization and plumbing.
Python was chosen for its ease of use in prototyping. NumPy handles the number crunching well, speed is not a goal yet.
I have been playing with and studying neural network related topics for years. After thinking about theory for a while a clear vision has formed for what I think a useful AI tool should look like.
Date: 2011-09-17 00:30:08 EDT
HTML generated by org-mode 6.21b in emacs 23