MNIST Namespace Reference

Classes
class	CatHinge
	Categorical hinge loss. Use if weights are restricted to be >0. More...
class	MLP
	The standard densely connected multilayer Perceptron. Template arguments provide the loss function, the activation function of neurons (experimental) and a possible constraint for the weights. More...
class	MLPBase
	Base class for Multi Layer Networks (–> currently Perceptron only). Allows us to use polymorphism with templated class. More...
class	MSE
	Root Mean Squared Error. More...
class	NoConstraint
	Constraint for weights in neural network: No constraint. More...
class	PositiveLimitedWeights
class	PositiveWeights
	Constraint for weights in neural network: Only weights >0. More...
class	ReLU
	ActivationFunction ReLU: Rectified Linear Unit. More...