it contains a few Artificial Neural Networks models implemented in C++. The models it contains are: => Adaline => Single Perceptron => Multilayer Perceptron - MP
The Multilayer Perceptron follows the backpropagation algorithm in order to train a model. In addition, the SDG is implemented with the MSE error function.
SDG: Stocastic Gradient Decent MSE: Mean Squared Error
In order to use the MP, you have to store the data you want to train into data directory inside backpropagation directory. Once you have write 'make' down in the terminal you may modify the execute.sh script in order to generate the model you want. Finally, the results will be given in csv format at results directory.