Neural Binary Gates is a simple artificial neural network that represents a problem related with digital binary gates. The operation that predict this network is X = (A xor B) and (A or C).
For resolve this problem I has designed a network with the foundations of MLP (Multilayer Perceptron) and an architecture 3-3-1.
This network was built with Python 3.6 and Keras package. The network is represented with Dense objects layers and use the optimizer SGD.
Finally for easy the use of this representation and test it I has deployed the abstraction trained model on Google Cloud ML Engine, you can use the python file predict.py for check the results.