layered.network module

class Layer(size, activation)

Bases: object

apply(incoming)

Store the incoming activation, apply the activation function and store the result as outgoing activation.

delta(above)

The derivative of the activation function at the current state.

class Matrices(shapes, elements=None)

Bases: object

__getitem__(index)

__setitem__(index, data)

copy()

class Network(layers)

Bases: object

feed(weights, data)

Evaluate the network with alternative weights on the input data and return the output activation.

static forward(weight, activations)

static backward(weight, activations)