MachineLearning.Hopfield

Implementation of Hopfield Network training and asssociating

Documentation

data HopfieldNet

HopfieldNet maintains the state and weights of the Hopfield Network, and is the major datastructure used in this code.

Constructors

HopfieldNet

Fields

_state :: Vector Float

_weights :: Matrix Float

Instances

Show HopfieldNet

initializeWith :: Matrix Float -> HopfieldNet

Initializes the HopfieldNet with the given training patterns.

activity :: Float -> Float

Maps the activation of a neuron to the output.

train :: HopfieldNet -> Matrix Float -> HopfieldNet

Updates the weights of the Hopfield network with the given training patterns.

associate :: MonadRandom m => HopfieldNet -> Int -> Vector Float -> m (Vector Float)

Repeatedly adjusts the Hopfield network's state to minimize the energy of the current configuration.

energy :: HopfieldNet -> Float

The energy of the current configuration of the Hopfield network.