Module: Neuronet

Defined in:

lib/neuronet.rb

Overview

Neuronet module

Defined Under Namespace

Modules: Tao, TaoYang, TaoYin, TaoYinYang, Yang, Yin, YinYang Classes: Connection, FeedForward, Gaussian, InputLayer, Layer, LogNormal, Neuron, Node, Scale, ScaledNetwork

Constant Summary

VERSION =

'6.0.0'

Class Method Summary

• .noise ⇒ Object

  By default, Neuronet builds a zeroed network.

• .squash(unsquashed) ⇒ Object

  The squash function for Neuronet is the sigmoid function.

• .unsquash(squashed) ⇒ Object

Class Method Details

.noise ⇒ Object

By default, Neuronet builds a zeroed network. Noise adds random fluctuations to create a search for minima.

# File 'lib/neuronet.rb', line 19

def self.noise
  rand + rand
end

.squash(unsquashed) ⇒ Object

The squash function for Neuronet is the sigmoid function. One should scale the problem with most data points between -1 and 1, extremes under 2s, and no outbounds above 3s. Standard deviations from the mean is probably a good way to figure the scale of the problem.

# File 'lib/neuronet.rb', line 9

def self.squash(unsquashed)
  1.0 / (1.0 + Math.exp(-unsquashed))
end

.unsquash(squashed) ⇒ Object

# File 'lib/neuronet.rb', line 13

def self.unsquash(squashed)
  Math.log(squashed / (1.0 - squashed))
end