DeepNetworkTraining.cpp File Reference

Go to the source code of this file.


typedef Autoencoder< RectifierNeuron, LinearNeuronAutoencoderModel
typedef FFNet< RectifierNeuron, LinearNeuronNetwork


LabeledData< RealVector, unsigned int > createProblem ()
template<class AutoencoderModel >
AutoencoderModel trainAutoencoderModel (UnlabeledData< RealVector > const &data, std::size_t numHidden, double regularisation, std::size_t iterations)
Network unsupervisedPreTraining (UnlabeledData< RealVector > const &data, std::size_t numHidden1, std::size_t numHidden2, std::size_t numOutputs, double regularisation, std::size_t iterations)
int main ()

Typedef Documentation

◆ AutoencoderModel

Definition at line 79 of file DeepNetworkTraining.cpp.

◆ Network

Definition at line 80 of file DeepNetworkTraining.cpp.

Function Documentation

◆ createProblem()

LabeledData<RealVector,unsigned int> createProblem ( )

Definition at line 19 of file DeepNetworkTraining.cpp.

References shark::random::coinToss(), and shark::random::globalRng.

Referenced by main().

◆ main()

◆ trainAutoencoderModel()

◆ unsupervisedPreTraining()

Network unsupervisedPreTraining ( UnlabeledData< RealVector > const &  data,
std::size_t  numHidden1,
std::size_t  numHidden2,
std::size_t  numOutputs,
double  regularisation,
std::size_t  iterations 

Definition at line 83 of file DeepNetworkTraining.cpp.

References shark::dataDimension(), and shark::initRandomNormal().

Referenced by main().