Trait for functions and their derivatives for functional layers
Layer properties of affine transformations, that is y=A*x+b
Model of Affine layer
Model of Feed Forward Neural Network.
Model of Feed Forward Neural Network. Implements forward, gradient computation and can return weights in vector format.
Feed forward ANN
Functional layer properties, y = f(x)
Functional layer model.
Functional layer model. Holds no weights.
Trait that holds Layer properties, that are needed to instantiate it.
Trait that holds Layer properties, that are needed to instantiate it. Implements Layer instantiation.
Trait that holds Layer weights (or parameters).
Trait that holds Layer weights (or parameters). Implements functions needed for forward propagation, computing delta and gradient. Can return weights in Vector format.
Trait for loss function
Implements Sigmoid activation function
Trait for the artificial neural network (ANN) topology properties
Trait for ANN topology model
Fabric for Affine layer models
Implements in-place application of functions in the arrays
In-place DGEMM and DGEMV for Breeze
In-place DGEMM and DGEMV for Breeze
Fabric for feed forward ANN models
Factory for some of the frequently-used topologies
Most of the contents of this package were taken from Spark and adapter for use without training or a Spark Context