ufpa-backprop

Implementation of backpropagation algorithm in Python, developed for evaluation in Artificial Neural Networks class (2019.4).

1. Implement backpropagation algorithm in any language (used Python).
2. Plot Mean Square Error evolution graph for all the interactions/epochs.
3. Plot Desired x Obtained Output graph.
4. It must be possible to change the number of neurons in the hidden layer (use only one hidden layer).
5. The values of sinaptic weights of the outputs must be printed at the end of training stage.
6. The network must be used in an aplication.