Uses of Interface
jsat.classifiers.neuralnetwork.initializers.WeightInitializer

Packages that use WeightInitializer

Package	Description
jsat.classifiers.neuralnetwork
jsat.classifiers.neuralnetwork.initializers

Uses of WeightInitializer in jsat.classifiers.neuralnetwork

Methods in jsat.classifiers.neuralnetwork that return WeightInitializer

Modifier and Type	Method and Description
WeightInitializer	SGDNetworkTrainer.getWeightInit()

Methods in jsat.classifiers.neuralnetwork with parameters of type WeightInitializer

Modifier and Type	Method and Description
void	SGDNetworkTrainer.setWeightInit(WeightInitializer weightInit) Sets the method used to initialize matrix connection weights

Uses of WeightInitializer in jsat.classifiers.neuralnetwork.initializers

Classes in jsat.classifiers.neuralnetwork.initializers that implement WeightInitializer

Modifier and Type	Class and Description
class	GaussianNormalInit This object initializes the values of weights by sampling from the zero mean Gaussian
class	TanhInitializer This initializer samples the weights from an adjusted uniform distribution in order to provided better behavior of neuron activation and gradients See: Glorot, X., & Bengio, Y.

Methods in jsat.classifiers.neuralnetwork.initializers that return WeightInitializer

Modifier and Type	Method and Description
WeightInitializer	WeightInitializer.clone()