public class BackpropTrainer extends IterableTrainer
IterableTrainer.StoppingCondition
Trainer.DataNotInitializedException
Constructor and Description |
---|
BackpropTrainer(Trainable network,
java.util.List<java.util.List<Neuron>> layers)
Construct the backprop trainer.
|
Modifier and Type | Method and Description |
---|---|
void |
apply()
Apply the algorithm.
|
double |
getError()
Get the current MSE error.
|
double |
getLearningRate() |
double |
getMomentum() |
void |
randomize()
Randomize the network.
|
protected void |
randomize(java.util.List<Neuron> layer)
Randomize the specified layer.
|
void |
setLearningRate(double learningRate) |
void |
setMomentum(double momentum) |
protected void |
updateNetwork()
Update internally constructed network.
|
addErrorListener, fireErrorUpdated, getErrorListeners, getErrorThreshold, getIteration, getIterationsBeforeStopping, getMinimumNumRows, getStoppingCond, getStoppingCondition, getValidationError, incrementIteration, isUpdateCompleted, iterate, removeErrorListener, setErrorThreshold, setIteration, setIterationsBeforeStopping, setStoppingCond, setStoppingCondition, setUpdateCompleted
addListener, fireProgressUpdate, fireTrainingBegin, fireTrainingEnd, getListeners, getTrainable, getTrainableNetwork, removeListener, revalidateSynapseGroups
public void apply()
Trainer
public void randomize()
randomize
in class IterableTrainer
protected void randomize(java.util.List<Neuron> layer)
layer
- the layer to randomizepublic double getError()
getError
in class IterableTrainer
protected void updateNetwork()
public double getLearningRate()
public void setLearningRate(double learningRate)
learningRate
- the learningRate to setpublic double getMomentum()
public void setMomentum(double momentum)
momentum
- the momentum to set