public void Mutate1(VRandom rng, double rate)
{
if (rng.RandBool(P_Insert * rate))
{
//selects a random character in the ASCII range
int c = rng.RandInt(32, 127);
mystr = mystr + (char)c;
}
else
{
//converts the string to a char array
char[] temp = mystr.ToCharArray();
//increments or decrements a random character
int index = rng.RandInt(0, temp.Length);
bool add = rng.RandBool();
int c = (int)temp[index] + (add ? 1 : -1);
//makes shure the resulting character is valid
if (c < 32)
{
c = 32;
}
if (c > 126)
{
c = 126;
}
temp[index] = (char)c;
//sets the new mutated string
mystr = new String(temp);
}
}
//This is the (Old) One. It assumes a rate between 0.0 and 1.0
public void MutateAlt(VRandom rng, double rate)
{
//clamps the rate to be between zero and one
rate = VMath.Clamp(rate);
//expands the size of the network at random
//if (rng.RandBool(P_EXPAND)) Expand(rng);
if (rng.RandBool(rate))
{
Expand(rng);
}
//used in itterating the structure
var ittr1 = nurons.ListItems();
var ittr2 = axons.ListItems();
//foreach (Nuron n in ittr1)
//{
// //skips over input nurons
// if (n.IsInput) continue;
// //mutates nodes based on the augmented mutation rate
// if (!rng.RandBool(P_NODE * rate)) continue;
// //updates the activation funciton
// n.Func = GetRandomActivation(rng);
//}
foreach (Axon ax in ittr2)
{
if (rng.RandBool(P_TOGGEL * rate))
{
//toggeles the enabled state
ax.Enabled = !ax.Enabled;
if (ax.Enabled)
{
ax.Weight = 0.0;
}
}
if (ax.Enabled)
{
//permutes the weight by a small amount
double delta = rng.RandGauss() * SD_NEW;
ax.Weight = ax.Weight + (delta * rate);
}
}
}
public void Mutate(VRandom rng, double rate)
{
if (rng.RandBool(P_Insert * rate))
{
int index = rng.RandInt(1, mystr.Length);
int len = mystr.Length;
//splits the string at the instertion point
string s1 = mystr.Substring(0, index);
string s2 = mystr.Substring(index, len - index);
//selects a random character in the ASCII range
int c = rng.RandInt(32, 127);
mystr = s1 + (char)c + s2;
}
//converts the string to a char array
char[] temp = mystr.ToCharArray();
if (rng.RandBool(P_Swap * rate))
{
//select two random indicies
int i = rng.RandInt(0, temp.Length);
int j = rng.RandInt(0, temp.Length);
//swaps the characters at the two incicies
char c = temp[i];
temp[i] = temp[j];
temp[j] = c;
}
for (int i = 0; i < temp.Length; i++)
{
//skips over characters with the inverse rate
if (!rng.RandBool(rate))
{
continue;
}
//selects a random character in the ASCII range
int c = rng.RandInt(32, 127);
temp[i] = (char)c;
}
//sets the new mutated string
mystr = new String(temp);
}
/// <summary>
/// Combines the genes of the curent nural net with the genes of
/// another network to create a brand new offspring. The idea is
/// that the child network will possess trates from both its
/// parents, similar to sexual reproduction.
/// </summary>
/// <param name="rng">Random number generator</param>
/// <param name="mate">Mate of the curent network</param>
/// <returns>The child of both networks</returns>
public NetworkCPP Combine(VRandom rng, NetworkCPP mate)
{
//makes a clone of the dominate parent
NetworkCPP child = new NetworkCPP(rng, this);
//determins weather or not to do liniar crossover
bool liniar = rng.RandBool(P_Linear);
double a = rng.NextDouble();
//lists all the axons in the mate
var axons = mate.ListAxons();
foreach (Axon ax in axons)
{
//obtains the matching child axon
Axon axc = child.FindMatch(ax);
if (axc == null)
{
continue;
}
if (liniar)
{
//chooses a value between the two weights
double weight = axc.Weight * (1.0 - a);
axc.Weight = weight + (ax.Weight * a);
}
else
{
//determins the new weight based on crossover
bool cross = rng.RandBool();
if (cross)
{
axc.Weight = ax.Weight;
}
}
//has a chance of enabling if either are disabled
bool en = ax.Enabled && axc.Enabled;
axc.Enabled = en || rng.RandBool(0.25);
}
return(child);
}
public void Mutate2(VRandom rng, double rate)
{
if (rng.RandBool(P_Insert * rate))
{
//selects a random character in the ASCII range
int c = rng.RandInt(32, 127);
mystr = mystr + (char)c;
}
else
{
//converts the string to a char array
char[] temp = mystr.ToCharArray();
for (int i = 0; i < temp.Length; i++)
{
//alters each character with a chance of 'rate'
if (!rng.RandBool(rate))
{
continue;
}
//increments or decrements the character
bool add = rng.RandBool();
int c = (int)temp[i] + (add ? 1 : -1);
//makes shure the resulting character is valid
if (c < 32)
{
c = 32;
}
if (c > 126)
{
c = 126;
}
temp[i] = (char)c;
}
//sets the new mutated string
mystr = new String(temp);
}
}
private void Procreate(Species <T> spec, Organism <T> rep, T scratch)
{
Organism <T> mom = spec.GetRandMember(rng);
Organism <T> dad = null;
if (crossover && rng.RandBool(CrossRate))
{
if (rng.RandBool(TransSpecies))
{
//obtains the best dad from a random species
int sindex = rng.RandInt(species.Count);
dad = species[sindex].Prototype;
Console.Write("T");
}
else
{
//obtains a random dad from the curent species
dad = spec.GetRandMember(rng);
}
if (mom.TrueFit > dad.TrueFit)
{
//swaps the parents if mom is better than dad
Organism <T> temp = dad;
dad = mom;
mom = dad;
}
//creates a new offspring from both parents
scratch.Overwrite(dad.Genome);
scratch.Crossover(rng, mom.Genome);
scratch.Mutate(rng, rate);
}
else
{
//creates a mutated offspring from the single parent
scratch.Overwrite(mom.Genome);
scratch.Mutate(rng, rate);
}
}
/// <summary>
/// Preterbs the current neural net by some random amount without
/// creating a clone. The rate of mutaiton determins how many of the
/// network connections are preturbed. For exampe, a mutation rate
/// of 0.5 indicates that half the weights will be perturbed.
/// </summary>
/// <param name="rate">Rate of mutation</param>
public void MutateSelf(double rate)
{
//clamps the rate to be between zero and one
rate = VMath.Clamp(rate);
if (rng.RandBool(P_Node))
{
//changes the activation funciton of a single node
NuronOld node = RandNuron();
node.Func = RandFunc();
return;
}
//lists all the axons in the network
var axons = ListAxons();
foreach (Axon ax in axons)
{
//mutates weights based on the rate of mutation
if (rng.RandBool(1.0 - rate))
{
continue;
}
if (ax.Enabled)
{
//permutes the weight by a small amount
double delta = rng.RandGauss() * SDS;
ax.Weight = ax.Weight + delta;
}
else
{
//resets the neuron to a small weight
ax.Weight = rng.RandGauss() * SDS;
ax.Enabled = true;
}
}
}
public void Mutate4(VRandom rng, double rate)
{
if (rng.RandBool(P_Insert * rate))
{
//selects a random character in the ASCII range
int c = rng.RandInt(32, 127);
mystr = mystr + (char)c;
}
else
{
//converts the string to a char array
char[] temp = mystr.ToCharArray();
//increments or decrements a random character
int index = rng.RandInt(0, temp.Length);
int c = rng.RandInt(32, 127);
temp[index] = (char)c;
//sets the new mutated string
mystr = new String(temp);
}
}
private void Expand(VRandom rng)
{
Nuron n1, n2;
//generates a pair of random nurons on diffrent levels
bool pass = GetRandomPair(rng, out n1, out n2);
if (!pass)
{
return;
}
//determins if this should be a recurent conneciton
bool testrec = recurent;
testrec = testrec && !n1.IsInput;
testrec = testrec && rng.RandBool(P_Recur);
if (testrec)
{
Nuron nx = n1;
n1 = n2;
n2 = nx;
}
//creates a temporary axon between the two nurons
double w = rng.RandGauss() * SD_NEW;
Axon temp = new Axon(n1.Index, n2.Index, w);
//atempts to gain the actual axon if it exists
Axon actual = axons.GetValue(temp.Index);
if (actual == null)
{
//adds the new axon into the network
axons.Add(temp.Index, temp);
}
else if (actual.Enabled == false)
{
//reneables the axon and updates it weight
actual.Enabled = true;
actual.Weight = temp.Weight;
}
else
{
//determins if we are able to insert a new node
int nid1 = GetRandomIndex(rng);
if (nid1 < 0)
{
return;
}
//sets the level of the new node in the middle
if (Math.Abs(n1.Level - n2.Level) < 3)
{
return;
}
int level = (n1.Level + n2.Level) / 2;
//adds a new node to the network
ActFunc act = GetRandomActivation(rng);
Nuron n3 = new Nuron(this, act, nid1, level);
nurons.Add(n3.Index, n3);
//inserts two new axons where the old one used to be
int nid0 = actual.Source;
int nid2 = actual.Target;
//Axon ax1 = new Axon(nid0, nid1, actual.Weight);
//Axon ax2 = new Axon(nid1, nid2, temp.Weight);
Axon ax1 = new Axon(nid0, nid1, actual.Weight);
Axon ax2 = new Axon(nid1, nid2, 1.0);
//we use the overwitre comand because there may be hash collisions
axons.Overwrite(ax1.Index, ax1);
axons.Overwrite(ax2.Index, ax2);
//deactivates the old axon
actual.Enabled = false;
}
}
public void Crossover(VRandom rng, NetworkAuto genome)
{
//throw new NotImplementedException();
//determins weather or not to do liniar crossover
bool liniar = rng.RandBool(P_Linear);
double a = rng.NextDouble();
//lists all the axons and nurons in the mate
var ittr1 = genome.nurons.ListItems();
var ittr2 = genome.axons.ListItems();
foreach (Nuron n in ittr1)
{
//obtains the matching child nuron
Nuron nc = nurons.GetValue(n.Index);
if (nc == null)
{
//adds the missing nuron
nc = new Nuron(this, n);
nurons.Add(nc.Index, nc);
}
//else
//{
// //determins the activation based on crossover
// bool cross = rng.RandBool();
// if (cross) nc.Func = n.Func;
//}
}
foreach (Axon ax in ittr2)
{
//obtains the matching child axon
Axon axc = axons.GetValue(ax.Index);
if (axc == null)
{
//adds the missing axon, but disabled
axc = new Axon(ax);
axc.Enabled = false;
axons.Add(axc.Index, axc);
}
else
{
if (liniar)
{
//chooses a value between the two weights
double weight = axc.Weight * (1.0 - a);
axc.Weight = weight + (ax.Weight * a);
}
else
{
//determins the new weight based on crossover
bool cross = rng.RandBool();
if (cross)
{
axc.Weight = ax.Weight;
}
}
//if the axon is present in both networks, it has a
//strong chance of becoming enabled
bool en = ax.Enabled && axc.Enabled;
axc.Enabled = en || rng.RandBool(0.25);
}
}
}