Hands-On Neuroevolution with Python

Initializing the first CPPN genome population

In the following code, at first, we initialize the random number generator seed with the current system time. After that, we create the appropriate substrate configuration for the discriminator ANN that is able to operate over the dimensionality of the experiment's visual field. Next, we create the CPPN genome using the created substrate configuration:

    # random seed    seed = int(time.time())    # Create substrate    substrate = create_substrate(num_dimensions)    # Create CPPN genome and population    g = NEAT.Genome(0,                    substrate.GetMinCPPNInputs(),                    0,                    substrate.GetMinCPPNOutputs(),                    False,                    NEAT.ActivationFunction.UNSIGNED_SIGMOID,                    NEAT.ActivationFunction.UNSIGNED_SIGMOID,                    0,                    params, 0) pop = NEAT.Population(g, ...