Neuronet

DESCRIPTION

Library to create neural networks.

Features perceptron, MLP, and deep feed forward networks. Uses a logistic squash function.

INSTALL

$ gem install neuronet

Required Ruby version: >= 3.4

SYNOPSIS

The library is meant to be read, but here are some quick bits:

require 'neuronet'

# Perceptron
# It can mirror, equivalent to "copy":
np = Neuronet::Perceptron.new(3, 3)
np.output_layer.mirror
values = np * [-1, 0, 1] #=> [-1.0, 0.0, 1.0]
# It can anti-mirror, equivalent to "not":
np.output_layer.mirror(-1)
values = np * [-1, 0, 1] #=> [1.0, 0.0, -1.0]

# MPL: Multi-Layer(3) Perceptron
# It can "and".
# In this example, NoisyMiddleNeuron is needed to differentiate the neurons:
mlp = Neuronet::MLP.new(2, 4, 1,
                        middle_neuron: Neuronet::NoisyMiddleNeuron)
pairs = [
  [[1, 1], [1]],
  [[-1, 1], [-1]],
  [[1, -1], [-1]],
  [[-1, -1], [-1]],
]
while pairs.any? { |input, target| (mlp * input).map(&:round) != target }
  mlp.pairs(pairs) # Training...
end
(mlp * [1, 1]).map(&:round)   #=> [1]
(mlp * [-1, 1]).map(&:round)  #=> [-1]
(mlp * [1, -1]).map(&:round)  #=> [-1]
(mlp * [-1, -1]).map(&:round) #=> [-1]

# To export to a file:
#     mlp.export_to_file(filename)
# To import from a file:
#     mlp.import_from_file(filename)
# These will export/import the network's biases and weights.

HELP

When reading the library, this is order the order I would read it:

Once you understand these files, the rest should all make sense. For some math on neural networks, see the Wiki.

LICENSE

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.