Hello! First I would like to apologize for the grammar errors. HAHA :D

This source is a simple example of neural network to learning play a pong similar game. You will need install the pygame library.

For this case have only 2 generations:

• Initial generation: Creates genomes randomly until value of constant MAX_RANDOM_GENOME.

• Post initial generation: Takes the best genome of initial generation and apply mutations it. Whenever the genome make the new record, we store the weights his on variables best_genome_input_weights and best_genome_hidden_weights. If he lose, the score will be reseted and a new genome will be created using the weights of the best genome, and applying mutations.