https://github.com/viliusluneckas/brainz

Artificial Neural Network library written in Ruby
https://github.com/viliusluneckas/brainz

artificial-neural-networks neural-network ruby

Last synced: 12 months ago
JSON representation

Artificial Neural Network library written in Ruby

• Host: GitHub
• URL: https://github.com/viliusluneckas/brainz
• Owner: ViliusLuneckas
• License: mit
• Created: 2012-03-05T19:44:28.000Z (over 14 years ago)
• Default Branch: master
• Last Pushed: 2019-07-12T18:30:53.000Z (almost 7 years ago)
• Last Synced: 2024-09-18T16:19:46.187Z (almost 2 years ago)
• Topics: artificial-neural-networks, neural-network, ruby
• Language: Ruby
• Homepage:
• Size: 33.2 KB
• Stars: 10
• Watchers: 2
• Forks: 2
• Open Issues: 3
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Brainz

Artificial Neural Network Library written in Ruby.

## Supported training algorithms

* Backpropagation

## Installation

Add this line to your application's Gemfile:

    gem 'brainz'

And then execute:

    $ bundle

Or install it yourself as:

    $ gem install brainz

## Usage

### Simple example, logical AND operation

    require 'brainz'

    

    brainz = Brainz::Brainz.new

    

    brainz.teach do |iteration, error|

      that(1, 1).is(1)

      that(1, 0).is(0)

      that(0, 1).is(0)

      that(0, 0).is(0)

      p "error_rate = #{'%.3f' % error || 0 } after #{iteration} iterations" if iteration % 10 == 0

    end

    

    puts "0 and 0 = #{brainz.guess(a: 0, b: 0)}"

    puts "0 and 1 = #{brainz.guess(a: 0, b: 1)}"

    puts "1 and 1 = #{brainz.guess(a: 1, b: 1)}"

    puts "1 and 0 = #{brainz.guess(a: 1, b: 0)}"

### Advanced usage, ligical XOR operation

    require 'brainz'

    brainz = Brainz::Brainz.new

    

    # specify number of hidden layers

    # 3 hidden layers with 4, 7 and 3 neurons

    brainz.num_hidden = [4, 7, 3]

    

    # tune learning parameters: learning_rate, momentum, wanted_error (mse)

    brainz.teach(learning_rate: 0.2, momentum: 0.05, wanted_error: 0.01) do |iteration, error|

      that(1, 1).is(0)

      that(1, 0).is(1)

      that(0, 1).is(1)

      that(0, 0).is(0)

    end

    puts "Learning took #{brainz.last_iterations}, error: #{brainz.error}, time: #{brainz.learning_time} s."

## Contributing

1. Fork it

2. Create your feature branch (`git checkout -b my-new-feature`)

3. Commit your changes (`git commit -am 'Added some feature'`)

4. Push to the branch (`git push origin my-new-feature`)

5. Create new Pull Request

6.