A home-made Deep Learning framework for my own sake of understanding Deep Learning architectures. Currently, it supports FeedForwardNet only and the performance is not good due to the Numpy-based implementation instead of Theano or Tensorflow which can ultilize GPUs.
Why the name Cerebro? People always try to find catchy names for their software. Me too. It's actually simple, people who invented Perceptron were inspired by Human brain (and you know Google Brain, right?) and in Spanish, Cerebro means 'Brain', :)