Python implementation of Neural Network class, fully vectorized and makes use on numpy's matrix multiplication for better performance.
Can be used to create a neural net with as many hidden layers as needed.
Usage:
# create a network with 10 input, 5 hidden, and 2 output nodes
n = neuralNet(10, 5, 2)
# training
n.learn(data_train)
#test
n.test(data_test)