/**
* @brief Creates a trained proxy to a FFNet model
*
* @param data training data
*
* @return proxy to FFNet model
*/
shark::ArgMaxConverter<shark::FFNet<shark::LogisticNeuron, shark::LinearNeuron>>
createFFNetModel(shark::ClassificationDataset& data) {
using namespace shark;
using namespace std;
FFNet<LogisticNeuron, LinearNeuron> model;
vector<size_t> layers = {inputDimension(data), 10, numberOfClasses(data)};
model.setStructure(layers, FFNetStructures::Full, true);
initRandomUniform(model,-0.1,0.1);
CrossEntropy loss;
ErrorFunction<RealVector, unsigned int> errorFunction(data, &model, &loss);
IRpropPlus optimizer;
optimizer.init(errorFunction);
// TrainingError<> stoppingCriterion(10, 1.e-5);
// Lower precision for faster learning
TrainingError<> stoppingCriterion(10, 1.e-3);
OptimizationTrainer<FFNet<LogisticNeuron, LinearNeuron>, unsigned int> trainer(&loss, &optimizer, &stoppingCriterion);
trainer.train(model, data);
shark::ArgMaxConverter<FFNet<LogisticNeuron, LinearNeuron>> converter;
converter.decisionFunction() = model;
return std::move(converter);
}
int main(){
//get problem data
Problem problem(1.0);
LabeledData<RealVector,unsigned int> training = problem.generateDataset(1000);
LabeledData<RealVector,unsigned int> test = problem.generateDataset(100);
std::size_t inputs=inputDimension(training);
std::size_t outputs = numberOfClasses(training);
std::size_t hiddens = 10;
unsigned numberOfSteps = 1000;
//create network and initialize weights random uniform
FFNet<LogisticNeuron,LinearNeuron> network;
network.setStructure(inputs,hiddens,outputs);
initRandomUniform(network,-0.1,0.1);
//create error function
CrossEntropy loss;
ErrorFunction error(training,&network,&loss);
// loss for evaluation
// The zeroOneLoss for multiclass problems assigns the class to the highest output
ZeroOneLoss<unsigned int, RealVector> loss01;
// evaluate initial network
Data<RealVector> prediction = network(training.inputs());
cout << "classification error before learning:\t" << loss01.eval(training.labels(), prediction) << endl;
//initialize Rprop
IRpropPlus optimizer;
optimizer.init(error);
for(unsigned step = 0; step != numberOfSteps; ++step)
optimizer.step(error);
// evaluate solution found by training
network.setParameterVector(optimizer.solution().point); // set weights to weights found by learning
prediction = network(training.inputs());
cout << "classification error after learning:\t" << loss01(training.labels(), prediction) << endl;
}