int main() {
std::cout << "[t-eoPartiallyMappedXover] => START" << std::endl;
Solution sol1, sol2;
sol1.resize(9);
sol2.resize(9);
for(int i = 0; i < sol1.size(); i++)
sol1[i] = i;
sol2[0] = 3;
sol2[1] = 4;
sol2[2] = 1;
sol2[3] = 0;
sol2[4] = 7;
sol2[5] = 6;
sol2[6] = 5;
sol2[7] = 8;
sol2[8] = 2;
std::cout << sol1 << std::endl;
std::cout << sol2 << std::endl;
eoPartiallyMappedXover<Solution> xover;
xover(sol1, sol2);
std::cout << "apres" << std::endl;
std::cout << sol1 << std::endl;
std::cout << sol2 << std::endl;
int verif[9];
for(int i = 0; i < sol1.size(); i++)
verif[i] = -1;
for(int i = 0; i < sol1.size(); i++)
verif[ sol1[i] ] = 1;
for(int i = 0; i < sol1.size(); i++)
assert(verif[i] != -1);
for(int i = 0; i < sol2.size(); i++)
verif[i] = -1;
for(int i = 0; i < sol2.size(); i++)
verif[ sol2[i] ] = 1;
for(int i = 0; i < sol2.size(); i++)
assert(verif[i] != -1);
std::cout << "[t-eoPartiallyMappedXover] => OK" << std::endl;
return EXIT_SUCCESS;
}
void main_function()
{
const unsigned SIZE = 8;
unsigned i, j;
eoBooleanGenerator gen;
Chrom chrom(SIZE), chrom2;
chrom.fitness(binary_value(chrom)); chrom2.fitness(binary_value(chrom2));
std::cout << "chrom: " << chrom << std::endl;
chrom[0] = chrom[SIZE - 1] = true; chrom.fitness(binary_value(chrom));
std::cout << "chrom: " << chrom << std::endl;
chrom[0] = chrom[SIZE - 1] = false; chrom.fitness(binary_value(chrom));
std::cout << "chrom: " << chrom << std::endl;
chrom[0] = chrom[SIZE - 1] = true; chrom.fitness(binary_value(chrom));
std::cout << "chrom.className() = " << chrom.className() << std::endl;
std::cout << "chrom: " << chrom << std::endl
<< "chrom2: " << chrom2 << std::endl;
std::ostringstream os;
os << chrom;
std::istringstream is(os.str());
is >> chrom2; chrom.fitness(binary_value(chrom2));
std::cout << "\nTesting reading, writing\n";
std::cout << "chrom: " << chrom << "\nchrom2: " << chrom2 << '\n';
std::fill(chrom.begin(), chrom.end(), false);
std::cout << "--------------------------------------------------"
<< std::endl << "eoMonOp's aplied to .......... " << chrom << std::endl;
eoInitFixedLength<Chrom>
random(chrom.size(), gen);
random(chrom); chrom.fitness(binary_value(chrom));
std::cout << "after eoBinRandom ............ " << chrom << std::endl;
eoOneBitFlip<Chrom> bitflip;
bitflip(chrom); chrom.fitness(binary_value(chrom));
std::cout << "after eoBitFlip .............. " << chrom << std::endl;
eoBitMutation<Chrom> mutation(0.5);
mutation(chrom); chrom.fitness(binary_value(chrom));
std::cout << "after eoBinMutation(0.5) ..... " << chrom << std::endl;
eoBitInversion<Chrom> inversion;
inversion(chrom); chrom.fitness(binary_value(chrom));
std::cout << "after eoBinInversion ......... " << chrom << std::endl;
eoBitNext<Chrom> next;
next(chrom); chrom.fitness(binary_value(chrom));
std::cout << "after eoBinNext .............. " << chrom << std::endl;
eoBitPrev<Chrom> prev;
prev(chrom); chrom.fitness(binary_value(chrom));
std::cout << "after eoBinPrev .............. " << chrom << std::endl;
std::fill(chrom.begin(), chrom.end(), false); chrom.fitness(binary_value(chrom));
std::fill(chrom2.begin(), chrom2.end(), true); chrom2.fitness(binary_value(chrom2));
std::cout << "--------------------------------------------------"
<< std::endl << "eoBinOp's aplied to ... "
<< chrom << " " << chrom2 << std::endl;
eo1PtBitXover<Chrom> xover;
std::fill(chrom.begin(), chrom.end(), false);
std::fill(chrom2.begin(), chrom2.end(), true);
xover(chrom, chrom2);
chrom.fitness(binary_value(chrom)); chrom2.fitness(binary_value(chrom2));
std::cout << "eoBinCrossover ........ " << chrom << " " << chrom2 << std::endl;
for (i = 1; i < SIZE; i++)
{
eoNPtsBitXover<Chrom> nxover(i);
std::fill(chrom.begin(), chrom.end(), false);
std::fill(chrom2.begin(), chrom2.end(), true);
nxover(chrom, chrom2);
chrom.fitness(binary_value(chrom)); chrom2.fitness(binary_value(chrom2));
std::cout << "eoBinNxOver(" << i << ") ........ "
<< chrom << " " << chrom2 << std::endl;
}
for (i = 1; i < SIZE / 2; i++)
for (j = 1; j < SIZE / 2; j++)
{
eoBitGxOver<Chrom> gxover(i, j);
std::fill(chrom.begin(), chrom.end(), false);
std::fill(chrom2.begin(), chrom2.end(), true);
gxover(chrom, chrom2);
chrom.fitness(binary_value(chrom)); chrom2.fitness(binary_value(chrom2));
std::cout << "eoBinGxOver(" << i << ", " << j << ") ..... "
<< chrom << " " << chrom2 << std::endl;
}
// test SGA algorithm
eoGenContinue<Chrom> continuator1(50);
eoFitContinue<Chrom> continuator2(65535.f);
eoCombinedContinue<Chrom> continuator(continuator1, continuator2);
eoCheckPoint<Chrom> checkpoint(continuator);
eoStdoutMonitor monitor;
checkpoint.add(monitor);
eoSecondMomentStats<Chrom> stats;
monitor.add(stats);
checkpoint.add(stats);
eoProportionalSelect<Chrom> select;
eoEvalFuncPtr<Chrom> eval(binary_value);
eoSGA<Chrom> sga(select, xover, 0.8f, bitflip, 0.1f, eval, checkpoint);
eoInitFixedLength<Chrom> init(16, gen);
eoPop<Chrom> pop(100, init);
apply<Chrom>(eval, pop);
sga(pop);
pop.sort();
std::cout << "Population " << pop << std::endl;
std::cout << "\nBest: " << pop[0].fitness() << '\n';
/*
Commented this out, waiting for a definite decision what to do with the mOp's
// Check multiOps
eoMultiMonOp<Chrom> mOp( &next );
mOp.adOp( &bitflip );
std::cout << "before multiMonOp............ " << chrom << std::endl;
mOp( chrom );
std::cout << "after multiMonOp .............. " << chrom << std::endl;
eoBinGxOver<Chrom> gxover(2, 4);
eoMultiBinOp<Chrom> mbOp( &gxover );
mOp.adOp( &bitflip );
std::cout << "before multiBinOp............ " << chrom << " " << chrom2 << std::endl;
mbOp( chrom, chrom2 );
std::cout << "after multiBinOp .............. " << chrom << " " << chrom2 <<std::endl;
*/
}
int
main(int argc, char** argv)
{
const unsigned int T_SIZE = 3; // size for tournament selection
const unsigned int VEC_SIZE = 3; // Number of object variables in genotypes
const unsigned int POP_SIZE = 10; // Size of population
const unsigned int MAX_GEN = 1000; // Maximum number of generation before STOP
const unsigned int MIN_GEN = 10; // Minimum number of generation before ...
const unsigned int STEADY_GEN = 10; // stop after STEADY_GEN gen. without improvement
const float P_CROSS = 0.8; // Crossover probability
const float P_MUT = 0.5; // mutation probability
const double EPSILON = 0.01; // range for real uniform mutation
double SIGMA = 0.3; // std dev. for normal mutation
// some parameters for chosing among different operators
const double hypercubeRate = 0.5; // relative weight for hypercube Xover
const double segmentRate = 0.5; // relative weight for segment Xover
const double uniformMutRate = 0.5; // relative weight for uniform mutation
const double detMutRate = 0.5; // relative weight for det-uniform mutation
const double normalMutRate = 0.5; // relative weight for normal mutation
const unsigned int SEED = 42; // seed for random number generator
po::options_description desc;
std::string matrixdir, model, nullmodel, trainingset, testingset, lastrun;
desc.add_options()
("matrixdir", po::value<std::string>(&matrixdir),
"The directory set up by SOFT2Matrix")
("model", po::value<std::string>(&model),
"The gene regulatory network model")
("nullmodel", po::value<std::string>(&nullmodel),
"The gene regulatory network null (scrambled) model")
("trainingset", po::value<std::string>(&trainingset),
"The list of arrays which have been selected for inclusion in the training set")
("testingset", po::value<std::string>(&testingset),
"The list of arrays which have been selected for inclusion in the testing set")
("lastrun", po::value<std::string>(&lastrun),
"The output file from the last run, to re-use scores from (optional)")
;
po::variables_map vm;
po::store(po::parse_command_line(argc, argv, desc), vm);
po::notify(vm);
std::string wrong;
if (!vm.count("help"))
{
if (!vm.count("matrixdir"))
wrong = "matrixdir";
else if (!vm.count("model"))
wrong = "model";
else if (!vm.count("trainingset"))
wrong = "trainingset";
else if (!vm.count("testingset"))
wrong = "testingset";
else if (!vm.count("nullmodel"))
wrong = "nullmodel";
}
if (wrong != "")
std::cerr << "Missing option: " << wrong << std::endl;
if (vm.count("help") || wrong != "")
{
std::cout << desc << std::endl;
return 1;
}
if (!fs::is_directory(matrixdir))
{
std::cout << "Matrix directory doesn't exist."
<< std::endl;
return 1;
}
if (!fs::is_regular(model))
{
std::cout << "Model file doesn't exist or not regular file."
<< std::endl;
return 1;
}
if (!fs::is_regular(nullmodel))
{
std::cout << "Null model file doesn't exist or not regular file."
<< std::endl;
return 1;
}
if (!fs::is_regular(trainingset))
{
std::cout << "Training set file doesn't exist or not regular file."
<< std::endl;
return 1;
}
if (!fs::is_regular(testingset))
{
std::cout << "Testing set file doesn't exist or not regular file."
<< std::endl;
return 1;
}
std::list<double> lastRun;
if (fs::is_regular(lastrun))
{
std::ifstream flastrun(lastrun.c_str());
static const boost::regex prev(".*SVM Result: .* Result \\(([^\\)]+)\\).*");
while (flastrun.good())
{
std::string l;
std::getline(flastrun, l);
boost::smatch m;
if (!boost::regex_match(l, m, prev))
{
continue;
}
lastRun.push_back(strtod(m[1].str().c_str(), NULL));
}
}
ExpressionMatrixProcessor emp(matrixdir);
GRNModel m(model, emp, 30);
GRNModel m2(nullmodel, emp, 30);
std::list<std::string> trainingArrays, testingArrays;
m.loadArraySet(trainingset, trainingArrays);
m.loadArraySet(testingset, testingArrays);
m.loadSVMTrainingData(trainingArrays);
m2.loadArraySet(trainingset, trainingArrays);
m2.loadArraySet(testingset, testingArrays);
m2.loadSVMTrainingData(trainingArrays);
// We seed it just so we can restart if need be.
rng.reseed(SEED);
EvaluateSVMFit eval(m, m2, trainingArrays, lastRun, 30, testingArrays.size());
std::vector<double> minVals, maxVals;
// log(gamma)
minVals.push_back(-15);
maxVals.push_back(15);
// log(C)
minVals.push_back(-15);
maxVals.push_back(2);
// nu
minVals.push_back(0);
maxVals.push_back(1);
eoRealVectorBounds rvb(minVals, maxVals);
eoRealInitBounded<Indi> random(rvb);
eoPop<Indi> pop(POP_SIZE, random);
apply<Indi>(eval, pop);
pop.sort();
std::cout << "Initial Population" << std::endl;
std::cout << pop;
eoDetTournamentSelect<Indi> selectOne(T_SIZE);
eoSelectPerc<Indi> select(selectOne);// by default rate==1
eoGenerationalReplacement<Indi> replace;
eoSegmentCrossover<Indi> xoverS;
eoHypercubeCrossover<Indi> xoverA;
eoPropCombinedQuadOp<Indi> xover(xoverS, segmentRate);
xover.add(xoverA, hypercubeRate, true);
eoUniformMutation<Indi> mutationU(EPSILON);
eoDetUniformMutation<Indi> mutationD(EPSILON);
eoNormalMutation<Indi> mutationN(SIGMA);
eoPropCombinedMonOp<Indi> mutation(mutationU, uniformMutRate);
mutation.add(mutationD, detMutRate);
mutation.add(mutationN, normalMutRate, true);
eoGenContinue<Indi> genCont(MAX_GEN);
eoSteadyFitContinue<Indi> steadyCont(MIN_GEN, STEADY_GEN);
eoCombinedContinue<Indi> continuator(genCont);
continuator.add(steadyCont);
eoSGATransform<Indi> transform(xover, P_CROSS, mutation, P_MUT);
eoEasyEA<Indi> gga(continuator, eval, select, transform, replace);
gga(pop);
pop.sort();
std::cout << "Final Population:"
<< std::endl << pop << std::endl;
return 0;
}
// GA function accepts data matrix, legal gene matrix, and number of total genes wanted in final matrix
// New stopping criterion: when PD does not increase for maxGenStalled generations
int optimizeDecisivnessGA (vector < vector <int> > & data,
vector < vector <int> > const& legal, int const& numAdd, bool const& referenceTaxonPresent,
int const& numProcs)
{
// initialize random number generator - already done in main
// srand((unsigned)time(0));
// Constants. Some should probably be user specified: tournamentSize, populationSize, mutationProbability, maxgen
// int numTaxa = data.size(); // not used
// int numLoci = data[0].size(); // not used
int numInitialLoci = countLociPresent(data);
int tournamentSize = 5; // size of tournament group.
const int populationSize = 500;
double mutationProbability = 0.9; // probability of mutation, otherwise xover
//double stepReplaceProbability = 0.01;
int maxgen = 100000; // maximum number of generations; ramp up to 100000 or whatever
int numTrees = 100; // ramp up to 1000
double bestSolution = 0.0;
vector < vector <int> > bestConfiguration;
int genStalled = 0; // number of generations without an increase in best fitness.
int maxGenStalled = 5000;
// Store fitness
double fitness[populationSize];
// cout << "initialize " << endl;
// Initialize population
vector < vector <int> > population[populationSize];
// int numAdd = maxGenes - numInitialLoci; // number of genes to add
int maxGenes = numInitialLoci + numAdd;
// cout << "Initial count = " << numInitialLoci << "; adding " << numAdd
// << " sampled characters for total of " << maxGenes << " occupied cells of a possible "
// << data.size() * data[0].size() << "." << endl;
// Function currently errors out when too many genes initially present.
// This should be changed to allow genes to be removed.
cout << endl << "SETTING UP INITIAL POPULATION" << endl << endl;
cout << "Population consists of " << populationSize << " individuals." << endl;
if (numAdd < 0) { // subtract genes; not sure when this will be used...
cout << numAdd << " populated cells will be removed." << endl;
for (int i = 0; i < populationSize; i++) {
population[i] = data;
for (int j = numAdd; j < 0; j++) {
mutate(population[i], legal, 0, 1);
}
}
} else { // add genes
cout << numAdd << " additional random taxon-character cells will be populated for each individual..." << endl;
for (int i = 0; i < populationSize; i++) {
population[i] = data;
for (int j = 0; j < numAdd; j++) { // Randomly add numAdd genes in legal positions
mutate(population[i], legal, 1, 0); // last 2 arguments are add, remove (boolean)
}
}
}
cout << "Done." << endl << endl;
//cout << "Best solution from initial population = " << bestSolution << "." << endl;
// Calculate initial fitness. For each matrix, crunch through 1000 trees
cout << "Calculating initial fitnesses for all individuals..." << endl;
for (int i = 0; i < populationSize; i++) {
fitness[i] = calcFit(referenceTaxonPresent, numTrees, population[i], numProcs);
// cout << "Fitness for individual " << i << " is: " << fitness[i] << "." << endl;
if (fitness[i] > bestSolution) {
bestSolution = fitness[i];
bestConfiguration = population[i];
// cout << "Best solution from initial population = " << bestSolution << "." << endl;
}
}
cout << "Done." << endl;
cout << "Best solution from initial population = " << bestSolution << "." << endl;
bool done = 0; // flag to end evolution loop
int gen = 0; // counter for number of generations
// Check for out of bounds parameter values
if (populationSize < 3) {
cout << "Population size must be greater than 2" << endl;
return(1);
}
if (tournamentSize < 2) {
cout << "Tournament group size must be greater than 1" << endl;
return(1);
}
// Evolution loop
cout << endl << "STARTING EVOLUTION" << endl << endl;
cout << "Running for a total of " << maxgen << " generations." << endl << endl;
// cout << "Progress:" << endl;
while (!done) {
gen = gen + 1;
cout << "Generation " << gen;
// Choose tournament group
// initialize best and worst with random
int random = (int)(rand() % (populationSize));
int best = random; // integer to hold position of best member of tournament group
int worst = random; // position of worst member of tournament group
double bestFit = fitness[random]; // best fitness
double worstFit = fitness[random]; // worst fitness
// selection step
//for (int i = 1; i < tournamentSize; i++)
for (int i = 0; i < tournamentSize; i++) {
random = (int)(rand() % populationSize);
if (fitness[random] > bestFit) {
bestFit = fitness[random];
best = random;
}
if (fitness[random] < worstFit) {
worstFit = fitness[random];
worst = random;
}
}
// make sure best and worst are not the same
// if they are the same, choose new best randomly from population
while (best == worst) {
best = (int)(rand() % (populationSize));
}
// Copy best individual to worst individual
population[worst] = population[best];
// enter stepwise function. too demanding.
// if (gen % 1000 == 0) {
// cout << ": Entering stepwise phase." << endl;
// stepReplace (population[worst], legal, maxGenes, referenceTaxonPresent, numProcs);
// }
// else {
// Mutate with some probability, otherwise xover
double randmu = (double) rand() / RAND_MAX;
if (randmu < mutationProbability) {
// mutate; change one cell.
mutate(population[worst], legal, 1, 1);
} else {
// xover with random individual in population
int randx = (int)(rand() % populationSize);
while (randx == worst) {
randx = (int)(rand() % populationSize);
}
xover(population[worst], population[randx], legal, maxGenes);
}
// }
// Calculate fitness of new individual
fitness[worst] = calcFit(referenceTaxonPresent, numTrees, population[worst], numProcs);
if (fitness[worst] > bestSolution) {
bestSolution = fitness[worst];
bestConfiguration = population[worst];
genStalled = 0;
} else {
genStalled += 1;
}
// Check to see if max number of generations has been reached
cout << ": best solution thus far = " << bestSolution << "." << endl;
if (gen == maxgen || genStalled == maxGenStalled) {
done = 1;
}
}
// End of evolution loop
// double bestFit = 0;
// int bestInd = 0;
// for (int i = 0; i < populationSize; i++)
// {
// if (fitness[i] > bestFit) {
// bestFit = fitness[i];
// bestInd = i;
// }
// }
// cout << endl << "Best result found: " << bestFit << endl;
// copy best evolved matrix to original data matrix
// data = population[bestInd];
//printData(data);
cout << endl << "Best result found: " << bestSolution << endl;
data = bestConfiguration;
cout << endl;
return(0);
}
