Neuron

Neurons

Useful Links

• "Simple" Description of Neural Networks
• Neural Network Basics
• The Role of Bias in Neural Networks

Usage

Getting Started

Creating a Neuron

let Neuron = require('@liquid-carrot/carrot').Neuron

let n = new Neuron()

Activating a Neuron

let Neuron = require('@liquid-carrot/carrot').Neuron

let n = new Neuron()

n.activate(Math.random(), function(error, results) {
  console.log(results) // 0.4254387327
})

Teaching a Neuron

let Neuron = require('@liquid-carrot/carrot').Neuron

let n = new Neuron()

n.propagate(0, function(error, fault) {
  console.log(results) // 0.124511366
})

Connecting two Neurons

let Neuron = require('@liquid-carrot/carrot').Neuron

let n0 = new Neuron()
let n1 = new Neuron()

n0.project(n1, function(error, connection) {
  console.log(connection)
})

Reference

new Neuron()

• new Neuron(): Creates a new neuron.
• new Neuron({ inputs: [n0, n1], outputs: [n2] }): Creates a new neuron with n0 and n1 as incoming connections, and n2 as an outgoing connection.
• new Neuron(n0): Creates a new neuron with the same connections as n0

.is.input([callback])

• neuron.is.input(): Returns true if neuron has no incoming connections; Invokes callback(error, isInput)

.is.output([callback])

• neuron.is.output(): Returns true if neuron has no outgoing connections; Invokes callback(error, isOutput)

.project(object[, callback])

• neuron.project(other_neuron): Connects neuron to other_neuron; Invokes callback(error, connections)
• neuron.project(layer): Connects neuron to every neuron in layer; Invokes callback(error, connections)
• neuron.project(group): Connects neuron to every neuron in group; Invokes callback(error, connections)

.activate(inputs[,callback])

• .activate([0, 1, 0, 1]): Activates neuron with the given inputs; inputs.length must equal connections.length; Invokes callback(error, results)

.propagate(feedback[,callback])

• .propagate([1, 0, 1, 0]): Calculates incoming connection errors; Invokes callback(error, results)

Appendix

Instance Functions

Key	Description
Neuron.prototype.is.input()	Tests whether neuron has no input connections
Neuron.prototype.is.output()	Tests whether neuron has no output connections
Neuron.prototype.project()	Connects to another neuron, layer, or group
Neuron.prototype.activate()	Activates neuron and forward propagates results
Neuron.prototype.learn()	Calculates error, updates weights, and backward propagates error

Class Constants

Key	Description
Neuron.activations	An object of typical activation/squash functions
Neuron.activations.SIGMOID	sigmoid Squash Function
Neuron.activations.RELU	ReLU Squash Function
Neuron.activations.TANH	tanh Squash Function
Neuron.activations.LINEAR	identity Squash Function

Properties

Key	Type	Default	Description
connections	Object	[]	All neuron connections
connections.incoming	[Connection]	[]	All incoming neuron connections
connections.outgoing	[Connection]	[]	All outgoing neuron connections
bias	Number	Math.random()	Check Out: The Role of Bias in Neural Networks
rate	Number	0.3	Check Out: What is a "Learning Rate" in a Neural Network? Learning Rates and Deep Learning Performance
activation	"relu"|"sigmoid"|"tanh"|"linear"|Function	"sigmoid"	Check Out: Fundamentals of Deep Learning: Activation Functions Understanding Activation Functions in Neural Networks Activation Functions: Neural Networks