backprop

Type Members

1. class BackpropClassifier extends AnyRef

   

   A convenience shortcut for applying BackpropNet to classification problems.

   A convenience shortcut for applying BackpropNet to classification problems.

   Assumes consecutive class indices 0..N

2. class BackpropNet extends LazyLogging

   

   A neural network with one hidden layer using backpropagation.

   A neural network with one hidden layer using backpropagation.

   Use BackpropNet.randomNet() to create a new network with random weights.

   Ref.: Robert Gordon University, "The Back Propagation Algorithm" https://web.archive.org/web/20150317210621/https://www4.rgu.ac.uk/files/chapter3%20-%20bp.pdf https://en.wikipedia.org/wiki/Backpropagation

3. case class Node(weights: Array[Float], output: Float = 0, error: Float = 0) extends Product with Serializable