Skip to content

Latest commit

 

History

History
128 lines (84 loc) · 3.95 KB

README.md

File metadata and controls

128 lines (84 loc) · 3.95 KB
Raw

Gem Version

rb-libsvm -- Ruby language bindings for LIBSVM

This package provides Ruby bindings to the LIBSVM library. SVM is a machine learning and classification algorithm, and LIBSVM is a popular free implementation of it, written by Chih-Chung Chang and Chih-Jen Lin, of National Taiwan University, Taipei. See the book "Programming Collective Intelligence," among others, for a usage example.

There is a JRuby implementation of this gem named jrb-libsvm by Andreas Eger.

Note: There exist some other Ruby bindings for LIBSVM. One is named Ruby SVM, written by Rudi Cilibrasi. The other, more actively developed one is libsvm-ruby-swig by Tom Zeng, which is built using SWIG.

LIBSVM includes a number of command line tools for preprocessing training data and finding parameters. These tools are not included in this gem. You should install the original package if you need them.

It is helpful to consult the README of the LIBSVM package for reference when configuring the training parameters.

Currently this package includes libsvm version 3.24.

Dependencies

None. LIBSVM is bundled with the project. Just install and go!

Installation

For building this gem from source on OS X (which is the default packaging) you will need to have Xcode installed, and from within Xcode you need to install the command line tools. Those contain the compiler which is necessary for the native code, and similar tools.

To install the gem run this command

gem install rb-libsvm

Usage

This is a short example of how to use the gem.

require 'libsvm'

# This library is namespaced.
problem = Libsvm::Problem.new
parameter = Libsvm::SvmParameter.new

parameter.cache_size = 1 # in megabytes

parameter.eps = 0.001
parameter.c = 10

examples = [ [1,0,1], [-1,0,-1] ].map {|ary| Libsvm::Node.features(ary) }
labels = [1, -1]

problem.set_examples(labels, examples)

model = Libsvm::Model.train(problem, parameter)

pred = model.predict(Libsvm::Node.features(1, 1, 1))
puts "Example [1, 1, 1] - Predicted #{pred}"

If you want to rely on Bundler for loading dependencies in a project, (i.e. use Bundler.require or use an environment that relies on it, like Rails), then you will need to specify rb-libsvm in the Gemfile like this:

gem 'rb-libsvm', require: 'libsvm'

This is because the loadable name (libsvm) is different from the gem's name (rb-libsvm).

Release

The process to make a release of the gem package to rubygems.org has a number of steps.

  • manually change the version in lib/libsvm/version.rb
  • clean, build, and run tests successfully
  • update code and documentation
  • push
  • sign into https://rubygems.org/
  • save API token from https://rubygems.org/profile/edit and store in .gem/credentials by running gem signin
  • perform actual release: bundle exec rake release

Author

Written by C. Florian Ebeling.

Contributors

License

This software can be freely used under the terms of the MIT license, see file MIT-LICENSE.

This package includes the source of LIBSVM, which is free to use under the license in the file LIBSVM-LICENSE.

Posts about using SVMs with Ruby

https://www.practicalai.io/implementing-classification-using-a-svm-in-ruby/

http://neovintage.blogspot.com/2011/11/text-classification-using-support.html

http://www.igvita.com/2008/01/07/support-vector-machines-svm-in-ruby/