int main(void) {
mocapy_seed((uint)5556575);
// Number of trainining sequences
int N = 500;
// Sequence lengths
int T = 100;
// Gibbs sampling parameters
int MCMC_BURN_IN = 10;
// HMM hidden and observed node sizes
uint H_SIZE=2;
bool init_random=false;
CPD th0_cpd;
th0_cpd.set_shape(2); th0_cpd.set_values(vec(0.1, 0.9));
CPD th1_cpd;
th1_cpd.set_shape(2,2); th1_cpd.set_values( vec(0.95, 0.05, 0.1, 0.9));
// The target DBN (This DBN generates the data)
Node* th0 = NodeFactory::new_discrete_node(H_SIZE, "th0", init_random, th0_cpd);
Node* th1 = NodeFactory::new_discrete_node(H_SIZE, "th1", init_random, th1_cpd);
Node* to0 = NodeFactory::new_kent_node("to0");
DBN tdbn;
tdbn.set_slices(vec(th0, to0), vec(th1, to0));
tdbn.add_intra("th0", "to0");
tdbn.add_inter("th0", "th1");
tdbn.construct();
// The model DBN (this DBN will be trained)
// For mh0, get the CPD from th0 and fix parameters
Node* mh0 = NodeFactory::new_discrete_node(H_SIZE, "mh0", init_random, CPD(), th0, true );
Node* mh1 = NodeFactory::new_discrete_node(H_SIZE, "mh1", init_random);
Node* mo0 = NodeFactory::new_kent_node("mo0");
DBN mdbn;
mdbn.set_slices(vec(mh0, mo0), vec(mh1, mo0));
mdbn.add_intra("mh0", "mo0");
mdbn.add_inter("mh0", "mh1");
mdbn.construct();
cout << "*** TARGET ***" << endl;
cout << *th0 << endl;
cout << *th1 << endl;
cout << *to0 << endl;
cout << "*** MODEL ***" << endl;
cout << *mh0 << endl;
cout << *mh1 << endl;
cout << *mo0 << endl;
vector<Sequence> seq_list;
vector< MDArray<eMISMASK> > mismask_list;
cout << "Generating data" << endl;
MDArray<eMISMASK> mismask;
mismask.repeat(T, vec(MOCAPY_HIDDEN, MOCAPY_OBSERVED));
// Generate the data
double sum_LL(0);
for (int i=0; i<N; i++) {
pair<Sequence, double> seq_ll = tdbn.sample_sequence(T);
sum_LL += seq_ll.second;
seq_list.push_back(seq_ll.first);
mismask_list.push_back(mismask);
}
cout << "Average LL: " << sum_LL/N << endl;
GibbsRandom mcmc = GibbsRandom(&mdbn);
EMEngine em = EMEngine(&mdbn, &mcmc, &seq_list, &mismask_list);
cout << "Starting EM loop" << endl;
double bestLL=-1000;
uint it_no_improvement(0);
uint i(0);
// Start EM loop
while (it_no_improvement<10) {
em.do_E_step(1, MCMC_BURN_IN, true);
double ll = em.get_loglik();
cout << "LL= " << ll;
if (ll > bestLL) {
cout << " * saving model *" << endl;
mdbn.save("kent_hmm.dbn");
bestLL = ll;
it_no_improvement=0;
}
else { it_no_improvement++; cout << endl; }
i++;
em.do_M_step();
}
cout << "DONE" << endl;
mdbn.load("kent_hmm.dbn");
cout << "*** TARGET ***" << endl;
cout << *th0 << endl;
cout << *th1 << endl;
cout << *to0 << endl;
cout << "*** MODEL ***" << endl;
cout << *mh0 << endl;
cout << *mh1 << endl;
cout << *mo0 << endl;
delete th0;
delete th1;
delete to0;
delete mh0;
delete mh1;
delete mo0;
return EXIT_SUCCESS;
}
void execute(DBN& dbn, task& task, const std::vector<std::string>& actions) {
print_title("Network");
dbn.display();
using dbn_t = std::decay_t<DBN>;
//Execute all the actions sequentially
for (auto& action : actions) {
if (action == "pretrain") {
print_title("Pretraining");
if (task.pretraining.samples.empty()) {
std::cout << "dllp: error: pretrain is not possible without a pretraining input" << std::endl;
return;
}
std::vector<Container> pt_samples;
//Try to read the samples
if (!read_samples<Three>(task.pretraining.samples, pt_samples)) {
std::cout << "dllp: error: failed to read the pretraining samples" << std::endl;
return;
}
if (task.pt_desc.denoising) {
std::vector<Container> clean_samples;
//Try to read the samples
if (!read_samples<Three>(task.pretraining_clean.samples, clean_samples)) {
std::cout << "dllp: error: failed to read the clean samples" << std::endl;
return;
}
//Pretrain the network
cpp::static_if<dbn_t::layers_t::is_denoising>([&](auto f) {
f(dbn).pretrain_denoising(pt_samples.begin(), pt_samples.end(), clean_samples.begin(), clean_samples.end(), task.pt_desc.epochs);
});
} else {
//Pretrain the network
dbn.pretrain(pt_samples.begin(), pt_samples.end(), task.pt_desc.epochs);
}
} else if (action == "train") {
print_title("Training");
if (task.training.samples.empty() || task.training.labels.empty()) {
std::cout << "dllp: error: train is not possible without samples and labels" << std::endl;
return;
}
std::vector<Container> ft_samples;
std::vector<std::size_t> ft_labels;
//Try to read the samples
if (!read_samples<Three>(task.training.samples, ft_samples)) {
std::cout << "dllp: error: failed to read the training samples" << std::endl;
return;
}
//Try to read the labels
if (!read_labels(task.training.labels, ft_labels)) {
std::cout << "dllp: error: failed to read the training labels" << std::endl;
return;
}
using last_layer = typename dbn_t::template layer_type<dbn_t::layers - 1>;
//Train the network
cpp::static_if<sgd_possible<last_layer>::value>([&](auto f) {
auto ft_error = f(dbn).fine_tune(ft_samples, ft_labels, task.ft_desc.epochs);
std::cout << "Train Classification Error:" << ft_error << std::endl;
});
} else if (action == "test") {
print_title("Testing");
if (task.testing.samples.empty() || task.testing.labels.empty()) {
std::cout << "dllp: error: test is not possible without samples and labels" << std::endl;
return;
}
std::vector<Container> test_samples;
std::vector<std::size_t> test_labels;
//Try to read the samples
if (!read_samples<Three>(task.testing.samples, test_samples)) {
std::cout << "dllp: error: failed to read the test samples" << std::endl;
return;
}
//Try to read the labels
if (!read_labels(task.testing.labels, test_labels)) {
std::cout << "dllp: error: failed to read the test labels" << std::endl;
return;
}
auto classes = dbn_t::output_size();
etl::dyn_matrix<std::size_t, 2> conf(classes, classes, 0.0);
std::size_t n = test_samples.size();
std::size_t tp = 0;
for (std::size_t i = 0; i < test_samples.size(); ++i) {
auto sample = test_samples[i];
auto label = test_labels[i];
auto predicted = dbn.predict(sample);
if (predicted == label) {
++tp;
}
++conf(label, predicted);
}
double test_error = (n - tp) / double(n);
std::cout << "Error rate: " << test_error << std::endl;
std::cout << "Accuracy: " << (1.0 - test_error) << std::endl
<< std::endl;
std::cout << "Results per class" << std::endl;
double overall = 0.0;
std::cout << " | Accuracy | Error rate |" << std::endl;
for (std::size_t l = 0; l < classes; ++l) {
std::size_t total = etl::sum(conf(l));
double acc = (total - conf(l, l)) / double(total);
std::cout << std::setw(3) << l;
std::cout << "|" << std::setw(10) << (1.0 - acc) << "|" << std::setw(12) << acc << "|" << std::endl;
overall += acc;
}
std::cout << std::endl;
std::cout << "Overall Error rate: " << overall / classes << std::endl;
std::cout << "Overall Accuracy: " << 1.0 - (overall / classes) << std::endl
<< std::endl;
std::cout << "Confusion Matrix (%)" << std::endl
<< std::endl;
std::cout << " ";
for (std::size_t l = 0; l < classes; ++l) {
std::cout << std::setw(5) << l << " ";
}
std::cout << std::endl;
for (std::size_t l = 0; l < classes; ++l) {
std::size_t total = etl::sum(conf(l));
std::cout << std::setw(3) << l << "|";
for (std::size_t p = 0; p < classes; ++p) {
std::cout << std::setw(5) << std::setprecision(2) << 100.0 * (conf(l, p) / double(total)) << "|";
}
std::cout << std::endl;
}
std::cout << std::endl;
} else if (action == "save") {
print_title("Save Weights");
dbn.store(task.w_desc.file);
std::cout << "Weights saved" << std::endl;
} else if (action == "load") {
print_title("Load Weights");
dbn.load(task.w_desc.file);
std::cout << "Weights loaded" << std::endl;
} else {
std::cout << "dllp: error: Invalid action: " << action << std::endl;
}
}
//TODO
}
int main (int argc, char * argv[])
{
std::streambuf *psbuf = NULL;
std::ofstream log_stream;
std::streambuf *pStreambuf = std::cout.rdbuf();
//--------------RNG INIT STUFF
srand((unsigned)time(0));
long seed;
r = gsl_rng_alloc (gsl_rng_rand48); // pick random number generator
seed = time (NULL) * getpid();
gsl_rng_set (r, seed); // set seed
// TESTING AREA
//---------------
DBN *dbn;
if (command_option_exists(argv, argv+argc, "-n")) {
dbn = return_network();
std::cout << dbn << std::endl;
std::cout << "Ready to learn" << std::endl;
std::cout << "If you are using visualization:" << std::endl;
std::cout << "'V': pauses visualization" << std::endl;
std::cout << "<SPACE>: pauses learning" << std::endl;
std::cout << "'+'/'-' increases/decreases learning rate" << std::endl;
std::cout << "'['/']' increases/decreases output threshold" << std::endl;
std::cout << "'L' stops learning for layer (skips to next if any)" << std::endl;
std::cout << "Current visualization is the features displayed on top with the plot of the reconstruction cost in the box (gl text not supported yet)" << std::endl;
std::cout << "Press <ENTER> to start learning: ";
std::cin.get();
}
else if (command_option_exists(argv, argv+argc, "-l")) {
std::string filename = get_command_line(argv, argv+argc, "-l");
MLP mlp = load_MLP(filename);
dbn = new DBN(mlp);
dbn->data_layers.clear();
dbn->view();
exit(EXIT_SUCCESS);
}
else if (command_option_exists(argv, argv+argc, "-f")) {
if (argc != 3) {
print_usage();
exit(EXIT_FAILURE);
}
//Initialization, time, logfile stuff, etc.
time_t t = time(0); // get time now
struct tm * the_time = localtime( & t );
mkdir(out_path.c_str(), S_IRWXU | S_IRWXG | S_IROTH | S_IXOTH);
out_path += convert_to_string(*the_time) + "/";
mkdir((out_path).c_str(), S_IRWXU | S_IRWXG | S_IROTH | S_IXOTH);
std::string filename = get_command_line(argv, argv+argc, "-f");
dbn = return_aod_network(filename, log_stream, psbuf);
}
else if (command_option_exists(argv, argv+argc, "-F")) {
std::string filename = get_command_line(argv, argv+argc, "-F");
MLP mlp = load_MLP(filename);
Autoencoder ae(mlp);
ae.name = (mlp).name + "fine_tuning";
Gradient_Descent gd(0);
gd.teachAE(ae);
MLP ae_mlp;
save(ae);
exit(EXIT_SUCCESS);
}
else if (command_option_exists(argv, argv+argc, "-stack")) {
std::string filename = get_command_line(argv, argv+argc, "-stack");
dbn = load_and_stack(filename, log_stream, psbuf);
}
else {
print_usage();
exit(EXIT_SUCCESS);
}
ContrastiveDivergence cd(1000);
dbn->learn(cd);
std::cout.rdbuf(pStreambuf);
log_stream.close();
exit(1);
return 0;
}
int main(void) {
mocapy_seed((uint)1);
// Number of training sequences
int N = 1000;
// Allocate storage in each node.
int max_num_dat_points = N;
// Sequence lengths
int T = 1;
// Gibbs sampling parameters
int MCMC_BURN_IN = 10;
// HMM hidden and observed node sizes
uint H_SIZE=1;
uint O_SIZE=20;
bool init_random=true;
// The target DBN (This DBN generates the data)
Node* th0 = NodeFactory::new_discrete_node(H_SIZE, "th0", init_random);
Node* th1 = NodeFactory::new_discrete_node(H_SIZE, "th1", init_random);
Node* to0 = NodeFactory::new_GDM_node("to0", O_SIZE , max_num_dat_points);
DBN tdbn;
tdbn.set_slices(vec(th0, to0), vec(th1, to0));
tdbn.add_intra("th0", "to0");
tdbn.add_inter("th0", "th1");
cout << "model has been defined" << endl;
tdbn.construct();
// The model DBN (this DBN will be trained)
// For mh0, get the CPD from th0 and fix parameters
Node* mh0 = NodeFactory::new_discrete_node(H_SIZE, "mh0", init_random, CPD(), th0, true );
Node* mh1 = NodeFactory::new_discrete_node(H_SIZE, "mh1", init_random);
Node* mo0 = NodeFactory::new_GDM_node("mo0",O_SIZE, max_num_dat_points);
DBN mdbn;
mdbn.set_slices(vec(mh0, mo0), vec(mh1, mo0));
mdbn.add_intra("mh0", "mo0");
mdbn.add_inter("mh0", "mh1");
mdbn.construct();
cout << "*** TARGET ***" << endl;
cout << *th0 << endl;
cout << *th1 << endl;
cout << *to0 << endl;
cout << "*** MODEL ***" << endl;
cout << *mh0 << endl;
cout << *mh1 << endl;
cout << *mo0 << endl;
vector<Sequence> seq_list;
vector< MDArray<eMISMASK> > mismask_list;
cout << "Generating data" << endl;
MDArray<eMISMASK> mismask;
mismask.repeat(T, vec(MOCAPY_HIDDEN, MOCAPY_OBSERVED));
// Setting parameters
CPD pars;
double s = 4;
pars.set_shape(H_SIZE,2*(O_SIZE-1));
vector<double> parvec;
for(uint i = 0;i < O_SIZE-1;i++){
parvec.push_back(s);
}
for(double i = s*(O_SIZE-1);i >= 1;i=i-s){
parvec.push_back(i);
}
/*
for(uint i = 0;i < O_SIZE-1;i++){
parvec.push_back(s);
}
for(double i = O_SIZE-1;i >= 1;i=i-s){
parvec.push_back(s*i);
}
*/
pars.set_values(parvec);
((GDMNode*)to0)->get_densities()->set_parameters(pars);
((GDMNode*)to0)->get_densities()->set_SumOfCounts(4000);
pair<Sequence, double> seq_ll;
// Generate the data
double sum_LL(0);
for (int i=0; i<N; i++) {
seq_ll = tdbn.sample_sequence(T);
sum_LL += seq_ll.second;
seq_list.push_back(seq_ll.first);
mismask_list.push_back(mismask);
}
cout << "Average LL: " << sum_LL/N << endl;
GibbsRandom mcmc = GibbsRandom(&mdbn);
EMEngine em = EMEngine(&mdbn, &mcmc, &seq_list, &mismask_list);
cout << "Starting EM loop" << endl;
double bestLL=-1000;
uint it_no_improvement(0);
uint i(0);
bool l;
// Start EM loop
while (it_no_improvement<100) {
em.do_E_step(1, MCMC_BURN_IN, true);
double ll = em.get_loglik();
cout << "LL= " << ll;
if (ll > bestLL) {
cout << " * saving model *" << endl;
cout << *mo0 << endl;
l = true;
//mdbn.save("discrete_hmm.dbn");
bestLL = ll;
it_no_improvement=0;
}
else {
it_no_improvement++;
cout << endl;
}
i++;
em.do_M_step();
if (l){
l=false;
}
}
cout << "DONE" << endl;
// mdbn.load("discrete_hmm.dbn");
cout << "*** TARGET ***" << endl;
cout << *th0 << endl;
cout << *th1 << endl;
cout << *to0 << endl;
cout << "*** MODEL ***" << endl;
cout << *mh0 << endl;
cout << *mh1 << endl;
cout << *mo0 << endl;
// em.do_E_step(1, MCMC_BURN_IN, true);/
// ((GDMNode*)mo0)->get_densities()->set_parameters(pars);
// double ll = em.get_loglik();
// cout << "LL= " << ll;
// cout << *mo0 << endl;
delete th0;
delete th1;
delete to0;
delete mh0;
delete mh1;
delete mo0;
return EXIT_SUCCESS;
}
DBN* KjaerulfsBNetModel()
{
DBN *net;
net = new DBN();
textcolor(WHITE);
net->AddNode(discrete^"node0-0 node1-0 node2-0 node3-0 node4-0 node5-0 node6-0 node7-0", "true false");
printf("\n net->AddNode(discrete^\"node0-0 node1-0 node2-0 \n\t node3-0 node4-0 node5-0 node6-0 node7-0\", \"true false\");");
textcolor(LIGHTGREEN);
printf("\t\t\tAdding of nodes is in process....");
_sleep(2000);
textcolor(WHITE);
net->AddNode(discrete^"node0-1 node1-1 node2-1 node3-1 node4-1 node5-1 node6-1 node7-1", "true false");
printf("\n net->AddNode(discrete^\"node0-1 node1-1 node2-1 \n\t node3-1 node4-1 node5-1 node6-1 node7-1\", \"true false\");");
_sleep(1000);
textcolor(LIGHTGREEN);
printf("\n ......All nodes are added....\n");
getch();
// arcs
textcolor(WHITE);
net->AddArc("node0-0", "node1-0 node2-0");
printf("\n net->AddArc(\"node0-0\", \"node1-0 node2-0\");");
textcolor(LIGHTGREEN);
printf("\t\t\t\t\t\tAdding of arcs is in process....");
_sleep(2000);
textcolor(WHITE);
net->AddArc("node1-0", "node2-0");
printf("\n net->AddArc(\"node1-0\", \"node2-0\");");
net->AddArc("node2-0", "node3-0");
printf("\n net->AddArc(\"node2-0\", \"node3-0\");");
net->AddArc("node3-0", "node4-0 node5-0");
printf("\n net->AddArc(\"node3-0\", \"node4-0 node5-0\");");
net->AddArc("node4-0 node5-0", "node6-0");
printf("\n net->AddArc(\"node4-0 node5-0\", \"node6-0\");");
net->AddArc("node6-0", "node7-0");
printf("\n net->AddArc(\"node6-0\", \"node7-0\");");
net->AddArc("node0-1", "node1-1 node2-1");
printf("\n net->AddArc(\"node0-1\", \"node1-1 node2-1\");");
net->AddArc("node1-1", "node2-1");
printf("\n net->AddArc(\"node1-1\", \"node2-1\");");
net->AddArc("node2-1", "node3-1");
printf("\n net->AddArc(\"node2-1\", \"node3-1\");");
net->AddArc("node3-1", "node4-1 node5-1");
printf("\n net->AddArc(\"node3-1\", \"node4-1 node5-1\");");
net->AddArc("node4-1 node5-1", "node6-1");
printf("\n net->AddArc(\"node4-1 node5-1\", \"node6-1\");");
net->AddArc("node6-1", "node7-1");
printf("\n net->AddArc(\"node6-1\", \"node7-1\");");
net->AddArc("node0-0", "node0-1");
printf("\n net->AddArc(\"node0-0\", \"node0-1\");");
net->AddArc("node3-0", "node3-1");
printf("\n net->AddArc(\"node3-0\", \"node3-1\");");
net->AddArc("node7-0", "node7-1");
printf("\n net->AddArc(\"node7-0\", \"node7-1\");");
textcolor(LIGHTGREEN);
printf("\n ......All arcs are added....\n");
getch();
// distributions
textcolor(WHITE);
net->SetPTabular("node0-0^false node0-0^true", "0.5 0.5");
printf("\n net->SetPTabular(\"node0-0^false node0-0^true\", \"0.5 0.5\");");
textcolor(LIGHTGREEN);
printf("\t\t\t\tAdding of distributions is in process....");
_sleep(2000);
textcolor(WHITE);
net->SetPTabular("node1-0^false node1-0^true", "0.98 0.02", "node0-0^false");
printf("\n net->SetPTabular(\"node1-0^false node1-0^true\", \"0.98 0.02\", \n\t \"node0-0^false\");");
net->SetPTabular("node1-0^false node1-0^true", "0.5 0.5", "node0-0^true");
printf("\n net->SetPTabular(\"node1-0^false node1-0^true\", \"0.5 0.5\", \n\t \"node0-0^true\");");
net->SetPTabular("node2-0^false node2-0^true", "1.0 0.0", "node0-0^false node1-0^false");
printf("\n net->SetPTabular(\"node2-0^false node2-0^true\", \"1.0 0.0\", \n\t \"node0-0^false node1-0^false\");");
net->SetPTabular("node2-0^false node2-0^true", "0.0 1.0", "node0-0^false node1-0^true");
printf("\n net->SetPTabular(\"node2-0^false node2-0^true\", \"0.0 1.0\", \n\t \"node0-0^false node1-0^true\");");
net->SetPTabular("node2-0^false node2-0^true", "0.0 1.0", "node0-0^true node1-0^false");
printf("\n net->SetPTabular(\"node2-0^false node2-0^true\", \"0.0 1.0\", \n\t \"node0-0^true node1-0^false\");");
net->SetPTabular("node2-0^false node2-0^true", "0.5 0.5", "node0-0^true node1-0^true");
printf("\n net->SetPTabular(\"node2-0^false node2-0^true\", \"0.5 0.5\", \n\t \"node0-0^true node1-0^true\");");
net->SetPTabular("node3-0^false node3-0^true", "0.99 0.01", "node2-0^false");
printf("\n net->SetPTabular(\"node3-0^false node3-0^true\", \"0.99 0.01\", \n\t \"node2-0^false\");");
net->SetPTabular("node3-0^false node3-0^true", "0.8 0.2", "node2-0^true");
printf("\n net->SetPTabular(\"node3-0^false node3-0^true\", \"0.8 0.2\", \n\t \"node2-0^true\");");
net->SetPTabular("node4-0^false node4-0^true", "0.99 0.01", "node3-0^false");
printf("\n net->SetPTabular(\"node4-0^false node4-0^true\", \"0.99 0.01\", \n\t \"node3-0^false\");");
net->SetPTabular("node4-0^false node4-0^true", "0.92 0.08", "node3-0^true");
printf("\n net->SetPTabular(\"node4-0^false node4-0^true\", \"0.92 0.08\", \n\t \"node3-0^true\");");
net->SetPTabular("node5-0^false node5-0^true", "0.79 0.21", "node3-0^false");
printf("\n net->SetPTabular(\"node5-0^false node5-0^true\", \"0.79 0.21\", \n\t \"node3-0^false\");");
net->SetPTabular("node5-0^false node5-0^true", "0.65 0.35", "node3-0^true");
printf("\n net->SetPTabular(\"node5-0^false node5-0^true\", \"0.65 0.35\", \n\t \"node3-0^true\");");
net->SetPTabular("node6-0^false node6-0^true", "0.9 0.1", "node4-0^false node5-0^false");
printf("\n net->SetPTabular(\"node6-0^false node6-0^true\", \"0.9 0.1\", \n\t \"node4-0^false node5-0^false\");");
net->SetPTabular("node6-0^false node6-0^true", "0.37 0.63", "node4-0^false node5-0^true");
printf("\n net->SetPTabular(\"node6-0^false node6-0^true\", \"0.37 0.63\", \n\t \"node4-0^false node5-0^true\");");
net->SetPTabular("node6-0^false node6-0^true", "0.21 0.79", "node4-0^true node5-0^false");
printf("\n net->SetPTabular(\"node6-0^false node6-0^true\", \"0.21 0.79\", \n\t \"node4-0^true node5-0^false\");");
net->SetPTabular("node6-0^false node6-0^true", "0.2 0.8", "node4-0^true node5-0^true");
printf("\n net->SetPTabular(\"node6-0^false node6-0^true\", \"0.2 0.8\", \n\t \"node4-0^true node5-0^true\");");
net->SetPTabular("node7-0^false node7-0^true", "0.91 0.09", "node6-0^false");
printf("\n net->SetPTabular(\"node7-0^false node7-0^true\", \"0.91 0.09\", \n\t \"node6-0^false\");");
net->SetPTabular("node7-0^false node7-0^true", "0.22 0.78", "node6-0^true");
printf("\n net->SetPTabular(\"node7-0^false node7-0^true\", \"0.22 0.78\", \n\t \"node6-0^true\");");
net->SetPTabular("node0-1^false node0-1^true", "0.45 0.55", "node0-0^false");
printf("\n net->SetPTabular(\"node0-1^false node0-1^true\", \"0.45 0.55\", \n\t \"node0-0^false\");");
net->SetPTabular("node0-1^false node0-1^true", "0.24 0.76", "node0-0^true");
printf("\n net->SetPTabular(\"node0-1^false node0-1^true\", \"0.24 0.76\", \n\t \"node0-0^true\");");
net->SetPTabular("node1-1^false node1-1^true", "0.51 0.49", "node0-1^false");
printf("\n net->SetPTabular(\"node1-1^false node1-1^true\", \"0.51 0.49\", \n\t \"node0-1^false\");");
net->SetPTabular("node1-1^false node1-1^true", "0.29 0.71", "node0-1^true");
printf("\n net->SetPTabular(\"node1-1^false node1-1^true\", \"0.29 0.71\", \n\t \"node0-1^true\");");
net->SetPTabular("node2-1^false node2-1^true", "0.98 0.02", "node0-1^false node1-1^false");
printf("\n net->SetPTabular(\"node2-1^false node2-1^true\", \"0.98 0.02\", \n\t \"node0-1^false node1-1^false\");");
net->SetPTabular("node2-1^false node2-1^true", "0.4 0.6", "node0-1^false node1-1^true");
printf("\n net->SetPTabular(\"node2-1^false node2-1^true\", \"0.4 0.6\", \n\t \"node0-1^false node1-1^true\");");
net->SetPTabular("node2-1^false node2-1^true", "0.2 0.8", "node0-1^true node1-1^false");
printf("\n net->SetPTabular(\"node2-1^false node2-1^true\", \"0.2 0.8\", \n\t \"node0-1^true node1-1^false\");");
net->SetPTabular("node2-1^false node2-1^true", "0.5 0.5", "node0-1^true node1-1^true");
printf("\n net->SetPTabular(\"node2-1^false node2-1^true\", \"0.5 0.5\", \n\t \"node0-1^true node1-1^true\");");
net->SetPTabular("node3-1^false node3-1^true", "0.36 0.64", "node3-0^false node2-1^false");
printf("\n net->SetPTabular(\"node3-1^false node3-1^true\", \"0.36 0.64\", \n\t \"node3-0^false node2-1^false\");");
net->SetPTabular("node3-1^false node3-1^true", "0.23 0.77", "node3-0^false node2-1^true");
printf("\n net->SetPTabular(\"node3-1^false node3-1^true\", \"0.23 0.77\", \n\t \"node3-0^false node2-1^true\");");
net->SetPTabular("node3-1^false node3-1^true", "0.78 0.22", "node3-0^true node2-1^false");
printf("\n net->SetPTabular(\"node3-1^false node3-1^true\", \"0.78 0.22\", \n\t \"node3-0^true node2-1^false\");");
net->SetPTabular("node3-1^false node3-1^true", "0.11 0.89", "node3-0^true node2-1^true");
printf("\n net->SetPTabular(\"node3-1^false node3-1^true\", \"0.11 0.89\", \n\t \"node3-0^true node2-1^true\");");
net->SetPTabular("node4-1^false node4-1^true", "0.955 0.045", "node3-1^false");
printf("\n net->SetPTabular(\"node4-1^false node4-1^true\", \"0.955 0.045\", \n\t \"node3-1^false\");");
net->SetPTabular("node4-1^false node4-1^true", "0.42 0.58", "node3-1^true");
printf("\n net->SetPTabular(\"node4-1^false node4-1^true\", \"0.42 0.58\", \n\t \"node3-1^true\");");
net->SetPTabular("node5-1^false node5-1^true", "0.57 0.43", "node3-1^false");
printf("\n net->SetPTabular(\"node5-1^false node5-1^true\", \"0.57 0.43\", \n\t \"node3-1^false\");");
net->SetPTabular("node5-1^false node5-1^true", "0.09 0.91", "node3-1^true");
printf("\n net->SetPTabular(\"node5-1^false node5-1^true\", \"0.09 0.91\", \n\t \"node3-1^true\");");
net->SetPTabular("node6-1^false node6-1^true", "0.91 0.09", "node4-1^false node5-1^false");
printf("\n net->SetPTabular(\"node6-1^false node6-1^true\", \"0.91 0.09\", \n\t \"node4-1^false node5-1^false\");");
net->SetPTabular("node6-1^false node6-1^true", "0.13 0.87", "node4-1^false node5-1^true");
printf("\n net->SetPTabular(\"node6-1^false node6-1^true\", \"0.13 0.87\", \n\t \"node4-1^false node5-1^true\");");
net->SetPTabular("node6-1^false node6-1^true", "0.12 0.88", "node4-1^true node5-1^false");
printf("\n net->SetPTabular(\"node6-1^false node6-1^true\", \"0.12 0.88\", \n\t \"node4-1^true node5-1^false\");");
net->SetPTabular("node6-1^false node6-1^true", "0.01 0.99", "node4-1^true node5-1^true");
printf("\n net->SetPTabular(\"node6-1^false node6-1^true\", \"0.01 0.99\", \n\t \"node4-1^true node5-1^true\");");
net->SetPTabular("node7-1^false node7-1^true", "0.39 0.61", "node6-1^false node7-0^false");
printf("\n net->SetPTabular(\"node7-1^false node7-1^true\", \"0.39 0.61\", \n\t \"node6-1^false node7-0^false\");");
net->SetPTabular("node7-1^false node7-1^true", "0.37 0.63", "node6-1^false node7-0^true");
printf("\n net->SetPTabular(\"node7-1^false node7-1^true\", \"0.37 0.63\", \n\t \"node6-1^false node7-0^true\");");
net->SetPTabular("node7-1^false node7-1^true", "0.255 0.745", "node6-1^true node7-0^false");
printf("\n net->SetPTabular(\"node7-1^false node7-1^true\", \"0.255 0.745\", \n\t \"node6-1^true node7-0^false\");");
net->SetPTabular("node7-1^false node7-1^true", "0.1 0.9", "node6-1^true node7-0^true");
printf("\n net->SetPTabular(\"node7-1^false node7-1^true\", \"0.1 0.9\", \n\t \"node6-1^true node7-0^true\");");
_sleep(1000);
textcolor(LIGHTGREEN);
printf("\n ......All distributions are added....\n");
textcolor(WHITE);
return net;
}
DBN* DBNModel()
{
DBN *net;
net = new DBN();
textcolor(WHITE);
net->AddNode(discrete^"Rain-0 Umbrella-0", "True False");
printf("\n net->AddNode(discrete^\"Rain-0 Umbrella-0\", \"True False\");");
textcolor(LIGHTGREEN);
printf("\t\t\t\t\tAdding of nodes is in process....");
_sleep(2000);
textcolor(WHITE);
net->AddNode(discrete^"Rain-1 Umbrella-1", "True False");
printf("\n net->AddNode(discrete^\"Rain-1 Umbrella-1\", \"True False\");");
textcolor(LIGHTGREEN);
printf("\n ......All nodes are added....\n");
getch();
// arcs
textcolor(WHITE);
net->AddArc("Rain-0", "Umbrella-0");
printf("\n net->AddArc(\"Rain-0\", \"Umbrella-0\");");
textcolor(LIGHTGREEN);
printf("\t\t\t\t\t\t\t\tAdding of arcs is in process....");
_sleep(2000);
textcolor(WHITE);
net->AddArc("Rain-0", "Rain-1");
printf("\n net->AddArc(\"Rain-0\", \"Rain-1\");");
_sleep(1000);
net->AddArc("Rain-1", "Umbrella-1");
printf("\n net->AddArc(\"Rain-1\", \"Umbrella-1\");");
textcolor(LIGHTGREEN);
printf("\n ......All arcs are added....\n");
getch();
// distributions
textcolor(WHITE);
net->SetPTabular("Rain-0^True Rain-0^False", "0.5 0.5");
printf("\n net->SetPTabular(\"Rain-0^True Rain-0^False\", \"0.5 0.5\");");
textcolor(LIGHTGREEN);
printf("\t\t\t\t\tAdding of distributions is in process....");
_sleep(2000);
textcolor(WHITE);
net->SetPTabular("Umbrella-0^True Umbrella-0^False", "0.9 0.1", "Rain-0^True");
printf("\n net->SetPTabular(\"Umbrella-0^True Umbrella-0^False\", \"0.9 0.1\", \"Rain-0^True\");");
_sleep(1000);
net->SetPTabular("Umbrella-0^True Umbrella-0^False", "0.2 0.8", "Rain-0^False");
printf("\n net->SetPTabular(\"Umbrella-0^True Umbrella-0^False\", \"0.2 0.8\", \"Rain-0^False\");");
_sleep(1000);
net->SetPTabular("Rain-1^True Rain-1^False", "0.7 0.3", "Rain-0^True");
printf("\n net->SetPTabular(\"Rain-1^True Rain-1^False\", \"0.7 0.3\", \"Rain-0^True\");");
_sleep(1000);
net->SetPTabular("Rain-1^True Rain-1^False", "0.3 0.7", "Rain-0^False");
printf("\n net->SetPTabular(\"Rain-1^True Rain-1^False\", \"0.3 0.7\", \"Rain-0^False\");");
_sleep(1000);
net->SetPTabular("Umbrella-1^True Umbrella-1^False", "0.9 0.1", "Rain-1^True");
printf("\n net->SetPTabular(\"Umbrella-1^True Umbrella-1^False\", \"0.9 0.1\", \"Rain-1^True\");");
_sleep(1000);
net->SetPTabular("Umbrella-1^True Umbrella-1^False", "0.2 0.8", "Rain-1^False");
printf("\n net->SetPTabular(\"Umbrella-1^True Umbrella-1^False\", \"0.2 0.8\", \"Rain-1^False\");");
textcolor(LIGHTGREEN);
printf("\n ......All distributions are added....\n");
getch();
textcolor(WHITE);
return net;
}