bool Trainee::dump(const char *traindatapath)
{
std::ofstream dat(traindatapath);
if(!dat.good()) return false;
dat << n_inputvec << ' ' << n_hid1vec << ' ' << n_hid2vec << ' ' << n_outputvec << '\n';
int i, j;
for(i=0;i<n_hid1vec;i++){
for(j=0;j<n_inputvec-1;j++) dat << weight1(i, j) << ' ';
dat << weight1(i, j) << '\n';
}
for(j=0;j<n_hid1vec-1;j++) dat << bias1(j) << ' ';
dat << bias1(j) << '\n';
for(i=0;i<n_hid2vec;i++){
for(j=0;j<n_hid1vec-1;j++) dat << weight2(i, j) << ' ';
dat << weight2(i, j) << '\n';
}
for(j=0;j<n_hid2vec-1;j++) dat << bias2(j) << ' ';
dat << bias2(j) << '\n';
for(i=0;i<n_outputvec;i++){
for(j=0;j<n_hid2vec-1;j++) dat << weight3(i, j) << ' ';
dat << weight3(i, j) << '\n';
}
for(j=0;j<n_outputvec-1;j++) dat << bias3(j) << ' ';
dat << bias3(j) << '\n';
if(!dat.good()) return false;
return true;
}
TEST(learnWithBPTest, XOR100)
{
Entry<double> a;
Entry<double> b;
Bias<double> bias1(1);
Bias<double> bias2(1);
Neuron<double, LinearActivationFunction<double >> n11(0.1);
Neuron<double, LinearActivationFunction<double >> n12(0.1);
Neuron<double, LinearActivationFunction<double >> n22(0.1);
Exit<double> exit;
std::vector < Link<double >> links(10);
//wejścia do 1 neuronu
a.setLinkOut(&links[0]);
b.setLinkOut(&links[1]);
bias1.setLinkOut(&links[2]);
n11.setLinkIn(&links[2]);
n11.setLinkIn(&links[0]);
n11.setLinkIn(&links[1]);
//wejścia do drugiego neuronu
a.setLinkOut(&links[3]);
b.setLinkOut(&links[4]);
bias1.setLinkOut(&links[5]);
n12.setLinkIn(&links[5]);
n12.setLinkIn(&links[3]);
n12.setLinkIn(&links[4]);
//połączenia międzyneuronowe
n11.setLinkOut(&links[6]);
n12.setLinkOut(&links[7]);
bias2.setLinkOut(&links[8]);
n22.setLinkIn(&links[8]);
n22.setLinkIn(&links[6]);
n22.setLinkIn(&links[7]);
//do wyjścia
n22.setLinkOut(&links[9]);
bias1.sendBiasToLinks();
bias2.sendBiasToLinks();
//n.printWages();
exit.setLinkIn(&links[9]);
//uczenie
for (int i = 0; i < 5000; i++)
{
a.setEntry(0);
b.setEntry(0);
n11.calculateOutput();
n12.calculateOutput();
n22.calculateOutput();
exit.learn(0);
n22.propagateAnswer();
n22.learnDelta();
n11.learnBP();
n12.learnBP();
// Qstd::cout<< "0,0: " << exit.getExit() << "\n";
a.setEntry(1);
b.setEntry(0);
n11.calculateOutput();
n12.calculateOutput();
n22.calculateOutput();
exit.learn(1);
n22.propagateAnswer();
n22.learnDelta();
n11.learnBP();
n12.learnBP();
// Qstd::cout<< " 1,0: " << exit.getExit() << "\n";
a.setEntry(0);
b.setEntry(1);
n11.calculateOutput();
n12.calculateOutput();
n22.calculateOutput();
exit.learn(1);
n22.propagateAnswer();
n22.learnDelta();
n11.learnBP();
n12.learnBP();
// Qstd::cout<< " 0,1: " << exit.getExit() << "\n";
a.setEntry(1);
b.setEntry(1);
n11.calculateOutput();
n12.calculateOutput();
n22.calculateOutput();
exit.learn(0);
n22.propagateAnswer();
n22.learnDelta();
n11.learnBP();
n12.learnBP();
// Qstd::cout<< " 1,1: " << exit.getExit() << "\n";
// for (auto l : links)
// {
// // Qstd::cout<< " " << l.getValue();
// }
// // Qstd::cout<< "\n";
//n11.printWages();
//n12.printWages();
//n22.printWages();
}
n11.printWages();
n12.printWages();
n22.printWages();
a.setEntry(0);
b.setEntry(0);
n11.calculateOutput();
n12.calculateOutput();
n22.calculateOutput();
ASSERT_GT(0.1, exit.getExit());
a.setEntry(1);
b.setEntry(0);
n11.calculateOutput();
n12.calculateOutput();
n22.calculateOutput();
ASSERT_LT(0.9, exit.getExit());
a.setEntry(0);
b.setEntry(1);
n11.calculateOutput();
n12.calculateOutput();
n22.calculateOutput();
ASSERT_LT(0.9, exit.getExit());
a.setEntry(1);
b.setEntry(1);
n11.calculateOutput();
n12.calculateOutput();
n22.calculateOutput();
ASSERT_GT(0.1, exit.getExit());
}
