public DecisionTreeDNANode(int p_num_inputs, Range <float> p_threshold_range, int p_num_outputs, Range <float> p_outputs_range)
{
m_input_index = RandomCalc.Rand(new Range <int>(0, p_num_inputs - 1));
m_min_input = RandomCalc.Rand(p_threshold_range);
m_max_input = RandomCalc.Rand(new Range <float>(m_min_input, p_threshold_range.Max));
m_output_values = ArrayCalc.functionInitialize(p_num_outputs, () => { return(RandomCalc.Rand(p_outputs_range)); });
}
///<summary>Returns a random object with fitter objects more likely to be chosen.</summary>
public T getRandomObject()
{
float indexNum = Mathf.Pow(RandomCalc.Rand(new Range <float>(0f, 1f)), 2);
int index = (int)Mathf.Round(indexNum * (m_objects.Count - 1));
return(m_objects[index].Object);
}
void spawnResource()
{
GameObject obj = Instantiate(m_resource_prefab, Vector3Calc.randomDirection() * 8f, Quaternion.identity);
obj.GetComponent <Resource>().Initalize(RandomCalc.Rand(new Range <float>(10f, 100f)));
ObjectLogger.log(obj, "RESOURCE");
}
///<summary>Return a piece of DNA by doing crossover and mutation on random DNA in gene pool weighted by fitness</summary>
public T birth()
{
ADNA <T> dna = m_gene_pool.getRandomObject().DNAcrossover(m_gene_pool.getRandomObject().getSelf());
if (RandomCalc.ChanceRoll(m_mutation_chance_percentage * 100))
{
dna = dna.DNAmutate();
}
m_creatures_birthed++;
return(dna.getSelf());
}
public Gene mutate()
{
Gene clone = this.Clone();
for (int i = 0; i < m_mutation_iterations; i++)
{
int index = ArrayCalc.randomIndex(clone.m_genes);
clone.m_genes[index] += RandomCalc.Rand(m_mutation_range);
}
return(clone);
}
public Gene(int p_number, Range <float> p_range, int p_mutation_iterations, Range <float> p_mutation_range)
{
m_genes = new float[p_number];
for (int i = 0; i < m_genes.Length; i++)
{
m_genes[i] = RandomCalc.Rand(p_range);
}
m_mutation_iterations = p_mutation_iterations;
m_mutation_range = p_mutation_range;
}
public DecisionTreeDNANode mutate(int p_num_inputs, Range <float> p_mutation_mult)
{
int mut_input_index = RandomCalc.Rand(new Range <int>(0, p_num_inputs - 1));
float mut_min_input = m_min_input * RandomCalc.Rand(p_mutation_mult);
float mut_max_input = m_max_input * RandomCalc.Rand(p_mutation_mult);
if (mut_max_input < mut_min_input)
{
mut_max_input = mut_min_input;
}
float[] mut_output_values = ArrayCalc.shallowClone(m_output_values);
for (int i = 0; i < mut_output_values.Length; i++)
{
mut_output_values[i] *= RandomCalc.Rand(p_mutation_mult);
}
return(new DecisionTreeDNANode(mut_input_index, mut_min_input, mut_max_input, mut_output_values));
}
public ADNA <DecisionNetDNA <T> > randomInstance()
{
return(new DecisionNetDNA <T>(m_id, m_inputs, m_outputs, Matrix <float> .Build.Dense(m_inputs.Length, m_outputs.Length, (i, j) => { return RandomCalc.Rand(m_weight_range); }), m_mutation_multiplier));
}
public NeuralNetDNA(int p_id, DInputFactory <T>[] p_inputs, DOutputFactory <T>[] p_outputs, int p_hidden_number, int p_hidden_size,
Range <float> p_weight_range, Range <float> p_mutation_multiplier)
{
m_id = p_id;
m_inputs = ArrayCalc.shallowClone(p_inputs);
m_outputs = ArrayCalc.shallowClone(p_outputs);
m_weights = new Matrix <float> [p_hidden_number + 1];
m_weights[0] = Matrix <float> .Build.Dense(m_inputs.Length, p_hidden_size, (i, j) => { return(RandomCalc.Rand(p_weight_range)); });
for (int i = 1; i < m_weights.Length - 1; i++)
{
m_weights[i] = Matrix <float> .Build.Dense(p_hidden_size, p_hidden_size, (x, y) => { return(RandomCalc.Rand(p_weight_range)); });
}
m_weights[m_weights.Length - 1] = Matrix <float> .Build.Dense(p_hidden_size, m_outputs.Length, (i, j) => { return(RandomCalc.Rand(p_weight_range)); });
m_mutation_multiplier = p_mutation_multiplier;
}
