Home / Ruby / Machine Learning

Ruby language bindings for LIBSVM

rb-libsvm -- Ruby language bindings for LIBSVM This package provides a 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
Information
Category: Ruby / Machine Learning
Watchers: 17
Star: 279
Fork: 35
Last update: Nov 27, 2023
Resource links
https://github.com/febeling/rb-libsvm

rb-libsvm -- Ruby language bindings for LIBSVM

Gem Version Build Status

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 3.23.

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).

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/

1 Open More issues
Closed

wrapper for "svm_predict_probability"

wrote a wrapper for the svm_predict_probability method

opened Jan 16, 2013 by andreaseger 10
Closed

Multi Class

It does not have NU_CVC and C_CVC enabled? Returns an error.

How to enable and use it?

opened Mar 8, 2017 by arjunmenon 7
Closed

The gem documentation is sparse

Information I assume should be explained to a beginner doesn't seem to be easily accessible. For example, in both examples, the code

parameter.cache_size = 1 # in megabytes
parameter.eps = 0.001
parameter.c   = 10

is not explained and only the first line is somehow self evident. Yet without these lines, the examples do not run (because they each default to 0). I assume .c corresponds to the formula at libsvm's homepage (PDF) ‘C’. I'm not sure about .eps (except that it's probably an epsilon).

Similarly, it's not obvious until you look deeper into the .public_methods that the kernel type should be set as a parameter passed to the model.

Another one is the presence of both an examples attribute and a set_examples method, with clear indications on how to use only the latter. It's not clear whether set_examples can accept data online, which I believe SVM should be able to handle.

The README should also probably mention the presence of more examples in the example folder.

Thanks for your work.

EDIT:

Reading the source code a bit more, I found this:

    double eps; /* stopping criteria */
    double C;   /* for C_SVC, EPSILON_SVR and NU_SVR */

so I assume eps is used to determine if the last iteration made enough changes to the weight vector to warrant a new one (?)

EDIT2: I am currently reading the libsvm README and linking to it might be a good enough minimum.

It also indicates that the basic executable svm-train uses sensible defaults for epsilon, c, and the cache size, see:


-c cost : set the parameter C of C-SVC, epsilon-SVR, and nu-SVR (default 1)
...
-m cachesize : set cache memory size in MB (default 100)
-e epsilon : set tolerance of termination criterion (default 0.001)
opened Jun 1, 2012 by ms 7
Closed

Adjust paths to modules

In Ruby community there is a strong directory structure pattern, which makes the modules correspond to directories.

I know that the library name might not be changed to libsvm, since this a different implementation of the libsvm wrapper, but I think rb-libsvm would be less surprising for developers, if the paths inside the lib directory were corresponding the the module Libsvm.

So instead of

  lib/rb-libsvm

we would have

  lib/libsvm

This also applies to the rb-libsvm.so which is generated during library installation.

opened Mar 16, 2012 by apohllo 7
Closed

predict_probability

I tried to get prediction probability, but even with the simple example provided in readme file, the return value of method is always 0.0 for each predicted class.

Example

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)

test = [0,0,0]

pred = model.predict_probability(Libsvm::Node.features(test))
puts "Example #{test} - Predicted #{pred}"

>>> Example [0, 0, 0] - Predicted [-1.0, [0.0, 0.0]]

I'm using OS X 10.8.4 but I tried also on latest ubuntu with the same result.

opened Aug 12, 2013 by mirobeka 6
Closed

File formatting tabs -> spaces

Pleas convert tabs in source files into two spaces. This is a very strong convention in the Ruby community.

opened Mar 2, 2012 by apohllo 6
Open

require 'libsvm' fails (MS windows)

I've installed ruby, devkit and rubygems and sucessfully installed the rb-libsvm gem. When I try to run the examples or go require 'libsvm' in irb i get this:

C:\dev>ruby
require 'libsvm'
^Z
c:/dev/ruby193/lib/ruby/site_ruby/1.9.1/rubygems/custom_require.rb:36:in `require': 193: %1 is not a valid Win32 application.   - c:/dev/ruby193/lib/ruby/gems/1.9.1/gems/rb-libsvm-1.0.8/lib/libsvm/libsvm_ext.so (LoadError)
        from c:/dev/ruby193/lib/ruby/site_ruby/1.9.1/rubygems/custom_require.rb:36:in `require'
        from c:/dev/ruby193/lib/ruby/gems/1.9.1/gems/rb-libsvm-1.0.8/lib/libsvm.rb:2:in `'
        from c:/dev/ruby193/lib/ruby/site_ruby/1.9.1/rubygems/custom_require.rb:60:in `require'
        from c:/dev/ruby193/lib/ruby/site_ruby/1.9.1/rubygems/custom_require.rb:60:in `rescue in require'
        from c:/dev/ruby193/lib/ruby/site_ruby/1.9.1/rubygems/custom_require.rb:35:in `require'
        from -:1:in `
'
feature request 
opened Sep 6, 2012 by jeppevb 12

v1.4.0(Sep 18, 2015)

Changed

  • Complete transition to expect specs.
  • Raise exception in Libsvm::Model.load if model can't be loaded (fixes #21).

Fixed

  • Fix a bug in Libsvm::Node#== when comparing with nil.

Contents

    titsuki A Raku bindings for libsvm
    datastream libsvm go version
    melisgl Common Lisp wrapper for the libsvm support vector machine library.
    ryanmelt An easy to install gem version of the Ruby bindings to Qt
    ralfbiedert nightly ffsvm - A libSVM compatible support vector machine, but up to 10x faster, for games or VR
    libgit2 ruby bindings to libgit2
    white-oak qml - Safe QML bindings for Rust
    tmattio High-quality JavaScript bindings for OCaml
    webview Tiny cross-platform webview library for C/C++/Golang. Uses WebKit (Gtk/Cocoa) and Edge (Windows)
    White-Oak qml-rust — QML bindings.
    artichoke 💎 Artichoke is a Ruby made with Rust
    lxc ruby bindings for liblxc
    sparklemotion Ruby bindings for the SQLite3 embedded database
    Telroshan Up-to-date Go bindings for SFML, the Simple and Fast Multimedia Library.
    nvzqz sys rosy - Ruby bindings for Rust
    hynek Low-level Python CFFI Bindings for Argon2
    chrisseaton Rhizome - a JIT for Ruby, implemented in pure Ruby
    Vurv78 luau bindings for the Rust programming language
    fullstaq-ruby A server-optimized Ruby distribution: less memory, faster, easy to install and security-patch via APT/YUM
    jruby JRuby, an implementation of Ruby on the JVM
    oracle A high performance implementation of the Ruby programming language. Built on the GraalVM by Oracle Labs.
    Share this repo

    Related Repos


    Machine Learning
    18
    Khiva - high-performance time series algorithms - for Ruby
    ankane Khiva - high-performance time series algorithms - for Ruby
     
    Machine Learning
    19
    Deploy machine learning models in Ruby (and Rails)
    ankane Deploy machine learning models in Ruby (and Rails)
     
    Machine Learning
    24
    🍅 tomoto - high performance topic modeling - for Ruby
    ankane 🍅 tomoto - high performance topic modeling - for Ruby
     
    Machine Learning
    21
    MITIE - named-entity recognition and binary relation detection - for Ruby
    ankane MITIE - named-entity recognition and binary relation detection - for Ruby
     
    Machine Learning
    595
    Deep learning for Ruby, powered by LibTorch
    ankane Torch-rb 🔥 Deep learning for Ruby, powered by LibTorch This gem is currently experimental. There may be breaking changes between each release. Installation First, install LibTorch. For Homebrew, use: bre
     
    Machine Learning
    486
    Collaborative filtering for Ruby
    ankane Disco 🔥 Collaborative filtering for Ruby Supports user-based and item-based recommendations Works with explicit and implicit feedback Uses high-performance matrix factorization Installation Add thi
     
    Machine Learning
    180
    Efficient text classification and representation learning for Ruby
    ankane fastText fastText - efficient text classification and representation learning - for Ruby Installation Add this line to your application’s Gemfile: gem 'fasttext' Getting Started fastText has two prim
     
    Machine Learning
    57
    Fast online machine learning for Ruby
    ankane Vowpal Wabbit Vowpal Wabbit - fast online machine learning - for Ruby 🔥 Uses the C API for blazing performance Installation First, install the Vowpal Wabbit C++ library. For Homebrew, use: brew install v
     
    Machine Learning
    53
    High performance factorization machines for Ruby
    ankane xLearn xLearn - the high performance machine learning library - for Ruby 🔥 Uses the C API for blazing performance Supports: Linear models Factorization machines Field-aware factorization machines
     
    Machine Learning
    360
    Deep learning for Ruby
    ankane TensorFlow TensorFlow - the end-to-end machine learning platform - for Ruby 🔥 Uses the C API under the hood Installation Install TensorFlow. For Homebrew, use: brew install tensorflow Add this line to
     
    Machine Learning
    713
    Rumale is a machine learning library in Ruby
    yoshoku Rumale Rumale (Ruby machine learning) is a machine learning library in Ruby. Rumale provides machine learning algorithms with interfaces similar to Scikit-Learn in Python. Rumale supports Support Vector Machine, Logistic
     
    Machine Learning
    631
    Machine learning for Ruby
    ankane Eps Linear regression for Ruby Build models quickly and easily Serve models built in Ruby, Python, R, and more Automatically handles categorical variables No external dependencies Works great with the SciRuby ecosyste
     
    Machine Learning
    509
    A ground-up and standalone reimplementation of TensorFlow for ruby. Comes with a pure ruby...
    jedld TensorStream TensorStream is an opensource framework for machine learning for ruby, its goal is to allow machine learning models to be easily built and run them in various hardware like GPUs and CPUs. It is heavily based on Te
     
    Machine Learning
    2k
    Curated list: Resources for machine learning in Ruby
    arbox Machine Learning with Ruby Useful resources for machine learning in Ruby This curated list comprises awesome resources, libraries, information sources about Machine Learning with the Ruby programming language. This list
     
    Machine Learning
    191
    PredictionIO Ruby SDK
    apache PredictionIO Ruby SDK The Ruby SDK provides a convenient wrapper for the PredictionIO API. It allows you to quickly record your users' behavior and retrieve personalized predictions for them. Documentation Full R