ml 0.1.0

data/Gemfile

@@ -0,0 +1,13 @@
+source :gemcutter
+
+# for plotting svg
+gem 'rubyvis'
+gem 'nokogiri'
+
+group :development do
+  gem "bacon", ">= 0"
+  gem "yard", "~> 0.6.0"
+  gem "bundler", "~> 1.0.0"
+  gem "jeweler", "~> 1.6.4"
+  gem "rcov", ">= 0"
+end

data/Gemfile.lock

@@ -0,0 +1,26 @@
+GEM
+  remote: http://rubygems.org/
+  specs:
+    bacon (1.1.0)
+    git (1.2.5)
+    jeweler (1.6.4)
+      bundler (~> 1.0)
+      git (>= 1.2.5)
+      rake
+    nokogiri (1.5.0)
+    rake (0.9.2)
+    rcov (0.9.10)
+    rubyvis (0.5.0)
+    yard (0.6.8)
+
+PLATFORMS
+  ruby
+
+DEPENDENCIES
+  bacon
+  bundler (~> 1.0.0)
+  jeweler (~> 1.6.4)
+  nokogiri
+  rcov
+  rubyvis
+  yard (~> 0.6.0)

data/LICENSE.txt

@@ -0,0 +1,20 @@
+Copyright (c) 2011 Andrew Liu
+
+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.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

data/README.md

@@ -0,0 +1,18 @@
+# ml 
+
+Machine learning library for ruby
+
+# Algorithm Implemented
+
+* Perceptron Learning Algorithm
+
+# Tools
+
+* Data generator
+* Data plotter
+
+# Copyright
+
+Copyright (c) 2011 Andrew Liu. See LICENSE.txt for
+further details.
+

data/Rakefile

@@ -0,0 +1,46 @@
+# encoding: utf-8
+
+require 'rubygems'
+require 'bundler'
+begin
+  Bundler.setup(:default, :development)
+rescue Bundler::BundlerError => e
+  $stderr.puts e.message
+  $stderr.puts "Run `bundle install` to install missing gems"
+  exit e.status_code
+end
+require 'rake'
+
+require 'jeweler'
+Jeweler::Tasks.new do |gem|
+  # gem is a Gem::Specification... see http://docs.rubygems.org/read/chapter/20 for more options
+  gem.name = "ml"
+  gem.homepage = "http://github.com/eggegg/ml"
+  gem.license = "MIT"
+  gem.summary = %Q{Machine learning library for Ruby}
+  gem.description = %Q{Machine learning library in Ruby}
+  gem.email = "andrewliu33@gmail.com"
+  gem.authors = ["Andrew Liu"]
+  # dependencies defined in Gemfile
+end
+Jeweler::RubygemsDotOrgTasks.new
+
+require 'rake/testtask'
+Rake::TestTask.new(:spec) do |spec|
+  spec.libs << 'lib' << 'spec'
+  spec.pattern = 'spec/**/*_spec.rb'
+  spec.verbose = true
+end
+
+require 'rcov/rcovtask'
+Rcov::RcovTask.new do |spec|
+  spec.libs << 'spec'
+  spec.pattern = 'spec/**/*_spec.rb'
+  spec.verbose = true
+  spec.rcov_opts << '--exclude "gems/*"'
+end
+
+task :default => :spec
+
+#require 'yard'
+#YARD::Rake::YardocTask.new

data/VERSION

@@ -0,0 +1 @@
+0.1.0

data/lib/data/generator.rb

@@ -0,0 +1,98 @@
+require 'matrix'
+
+module ML
+  # Generating sample points on 2D plane
+  class Generator2D
+    # Generate point from line
+    #
+    # @param [Array] [a,b,c] for ax+by+c=0
+    # @param [Number] x value
+    # @return [Array] 
+    def self.point_from_line coef, x
+      [x, (-coef[2]-(coef[0] * x))/coef[1]]
+    end
+
+    # Initialize a generator
+    #
+    # @param [Integer] x range
+    # @param [Integer] y range
+    def initialize x_range = 100, y_range = 100
+      @x_range = x_range
+      @y_range = y_range
+    end
+
+    # Generate two groups of points on 2d plain
+    #
+    # @param [Integer] the number of points of each set
+    # @param [Array] [a,b,c] for ax+by+c=0
+    # @return [Array] two sets of points
+    def points_2d points, coef = [-1.0, 1.0, 0.0]
+      result = []
+      # for each group
+      [1, -1].each do |grp|
+        result << []
+
+        points.times do
+          while true
+            point = generate_point
+            prod = Matrix.column_vector(point).transpose * Matrix.column_vector(coef)
+            if (prod[0,0] <=> 0) == grp
+              result[-1] << point
+              break
+            end
+          end
+        end
+      end
+      result
+    end
+
+  private
+    def generate_point
+      [@x_range * rand, @y_range * rand, 1.0]
+    end
+  end
+
+  # General generator for n-dimentional space
+  class Generator
+    # Initial generator
+    #
+    # @param [Integer] dimension
+    def initialize dim
+      @dim = dim
+    end
+
+    # Generate two groups of points
+    #
+    # @param [Integer] the number of points of each set
+    # @param [Array] array of the size of dimension to specify the hyper plane
+    # @return [Array] two sets of points
+    def points points, coef
+      result = []
+      # for each group
+      [1, -1].each do |grp|
+        result << []
+
+        points.times do
+          while true
+            point = Generator.generate_vector(@dim)
+            prod = Matrix.column_vector(point).transpose * Matrix.column_vector(coef)
+            if (prod[0,0] <=> 0) == grp
+              result[-1] << point
+              break
+            end
+          end
+        end
+      end
+      result
+    end
+
+    # Generating a random vector
+    #
+    # @param [Integer] the dimension of the vector
+    # @return [Array] random vector
+    def self.generate_vector dim
+      result = Array.new(dim) { rand - 0.5 } 
+      result << 1.0
+    end
+  end
+end

data/lib/data/parser.rb

@@ -0,0 +1,27 @@
+module ML
+  # Parser for traing/testing data
+  class Parser
+    # Parse the vector file with supervised result
+    #
+    # @return [Hash] map from data to supervised result
+    def parse_supervised filename
+      result = {}
+      lines = IO.readlines(filename)
+
+      lines.each do |line|
+        splitted = line.split.map(&:to_f)
+        result[splitted[1..-1] + [1.0]] = splitted[0]
+      end
+
+      result
+    end
+
+    # Parse the vector file
+    #
+    # @return [Array] array of vectors
+    def parse_unsupervised filename
+      lines = IO.readlines(filename)
+      lines.map {|line| line.split.map(&:to_f) }
+    end
+  end
+end

data/lib/data/plotter.rb

@@ -0,0 +1,77 @@
+require 'rubyvis'
+
+module ML
+  # Plotting the data to svg
+  class Plotter
+    # Initializer of plotter
+    #
+    # @param [Integer] x value range
+    # @param [Integer] y value range
+    # @param [Integer] x plot size
+    # @param [Integer] y plot size
+    def initialize x_range = 100, y_range = 100, x_size = 100, y_size = 100
+      @x_range = x_range
+      @y_range = y_range
+      @x_size = x_size
+      @y_size = y_size
+
+      @x = pv.Scale.linear(0, @x_range).range(0, @x_size)
+      @y = pv.Scale.linear(0, @y_range).range(0, @y_size)
+
+      @vis = pv.Panel.new.width(@x_size).height(@y_size)
+               .bottom(20).left(20).right(10).top(5)
+      
+      @vis.add(pv.Rule).data(@y.ticks()).bottom(@y)
+          .stroke_style(lambda {|d| d!=0 ? "#eee" : "#000"})
+          .anchor("left").add(pv.Label)
+          .visible(lambda {|d| d > 0 and d < x_range})
+          .text(@y.tick_format)
+ 
+      @vis.add(pv.Rule).data(@x.ticks()).left(@x)
+          .stroke_style(lambda {|d| d!=0 ? "#eee" : "#000"})
+          .anchor("bottom").add(pv.Label)
+          .visible(lambda {|d| d > 0 and d < y_range})
+          .text(@x.tick_format)
+
+      yield(self) if block_given?
+    end
+
+    # Plotting points with shape and color
+    #
+    # @param [Array] points to plot
+    # @param [String] shape of the points
+    # @param [String] color of the points
+    def dot points, shape = "circle", color = "#000"
+      # FIXME: dirty hack for instance_exec
+      x = @x
+      y = @y
+
+      @vis.add(pv.Dot).data(points)         
+          .left(lambda {|d| x.scale(d[0])})
+          .bottom(lambda {|d| y.scale(d[1])})
+          .shape(shape)
+          .stroke_style(color)
+    end
+
+    # Plotting line with color
+    #
+    # @param [Array] 2 points which represents line
+    def line points, color = "#000"
+      x = @x
+      y = @y
+
+      @vis.add(pv.Line).data(points)
+          .left(lambda {|d| x.scale(d[0])})
+          .bottom(lambda {|d| y.scale(d[1])})
+          .stroke_style(color)
+    end
+
+    # Convert to svg
+    #
+    # @return [String] svg plot
+    def to_svg
+      @vis.render
+      @vis.to_svg
+    end
+  end
+end

data/lib/method/perceptron.rb

@@ -0,0 +1,66 @@
+require 'matrix'
+
+module ML
+  # Implementation of Perceptron Learning Algorithm
+  class PerceptronLearner
+    # Initialize a perceptron learner
+    #
+    # @param [Integer] the number of dimension
+    def initialize dim
+      @dim = dim
+      @w = Matrix.column_vector(Array.new(dim + 1, 0))
+    end
+
+    # Train with supervised data
+    #
+    # @param [Hash] supervised input data (mapping from array to integer)
+    # @return [PerceptronLearner] self object 
+    def train! data
+      @update = 0
+      while true
+        misclassified = false
+
+        for dat, result in data
+          aug_data = Matrix.column_vector(dat)
+
+          if classify(aug_data) != result
+            misclassified = true
+
+            @w = @w + result * aug_data
+            @update += 1
+            break
+          end
+        end
+
+        break unless misclassified
+      end
+    end
+
+    # The final coefficient of the line
+    #
+    # @return [line] [a,b,c] for ax+by+c=0
+    def line
+      @w.column(0).to_a
+    end
+
+    # The number for updates
+    #
+    # @return [Integer] update count
+    def update_count
+      @update
+    end
+
+    # Predict certain data
+    #
+    # @param [Array] data in question
+    # @param [Integer] prediction
+    def predict data
+      classify Matrix.column_vector(data + [1.0])  
+    end
+
+  private
+    def classify data
+      (@w.transpose * data)[0,0] <=> 0
+    end
+  end
+end

data/lib/ml.rb

@@ -0,0 +1,13 @@
+require 'rubygems'
+require 'bundler/setup'
+
+require 'data/plotter'
+require 'data/generator'
+require 'data/parser'
+
+require 'method/perceptron'
+
+module ML
+end
+
+MachingLearning = ML

data/ml.gemspec

@@ -0,0 +1,71 @@
+# Generated by jeweler
+# DO NOT EDIT THIS FILE DIRECTLY
+# Instead, edit Jeweler::Tasks in Rakefile, and run 'rake gemspec'
+# -*- encoding: utf-8 -*-
+
+Gem::Specification.new do |s|
+  s.name = "ml"
+  s.version = "0.1.0"
+
+  s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
+  s.authors = ["Andrew Liu"]
+  s.date = "2011-10-03"
+  s.description = "Machine learning library in Ruby"
+  s.email = "andrewliu33@gmail.com"
+  s.extra_rdoc_files = [
+    "LICENSE.txt",
+    "README.md"
+  ]
+  s.files = [
+    "Gemfile",
+    "Gemfile.lock",
+    "LICENSE.txt",
+    "README.md",
+    "Rakefile",
+    "VERSION",
+    "lib/data/generator.rb",
+    "lib/data/parser.rb",
+    "lib/data/plotter.rb",
+    "lib/method/perceptron.rb",
+    "lib/ml.rb",
+    "ml.gemspec",
+    "spec/learning_spec.rb",
+    "spec/spec_helper.rb"
+  ]
+  s.homepage = "http://github.com/eggegg/ml"
+  s.licenses = ["MIT"]
+  s.require_paths = ["lib"]
+  s.rubygems_version = "1.8.10"
+  s.summary = "Machine learning library for Ruby"
+
+  if s.respond_to? :specification_version then
+    s.specification_version = 3
+
+    if Gem::Version.new(Gem::VERSION) >= Gem::Version.new('1.2.0') then
+      s.add_runtime_dependency(%q<rubyvis>, [">= 0"])
+      s.add_runtime_dependency(%q<nokogiri>, [">= 0"])
+      s.add_development_dependency(%q<bacon>, [">= 0"])
+      s.add_development_dependency(%q<yard>, ["~> 0.6.0"])
+      s.add_development_dependency(%q<bundler>, ["~> 1.0.0"])
+      s.add_development_dependency(%q<jeweler>, ["~> 1.6.4"])
+      s.add_development_dependency(%q<rcov>, [">= 0"])
+    else
+      s.add_dependency(%q<rubyvis>, [">= 0"])
+      s.add_dependency(%q<nokogiri>, [">= 0"])
+      s.add_dependency(%q<bacon>, [">= 0"])
+      s.add_dependency(%q<yard>, ["~> 0.6.0"])
+      s.add_dependency(%q<bundler>, ["~> 1.0.0"])
+      s.add_dependency(%q<jeweler>, ["~> 1.6.4"])
+      s.add_dependency(%q<rcov>, [">= 0"])
+    end
+  else
+    s.add_dependency(%q<rubyvis>, [">= 0"])
+    s.add_dependency(%q<nokogiri>, [">= 0"])
+    s.add_dependency(%q<bacon>, [">= 0"])
+    s.add_dependency(%q<yard>, ["~> 0.6.0"])
+    s.add_dependency(%q<bundler>, ["~> 1.0.0"])
+    s.add_dependency(%q<jeweler>, ["~> 1.6.4"])
+    s.add_dependency(%q<rcov>, [">= 0"])
+  end
+end
+

data/spec/learning_spec.rb

@@ -0,0 +1,7 @@
+require 'spec_helper'
+
+describe "Learning" do
+  it "fails" do
+    should.flunk "hey buddy, you should probably rename this file and start specing for real"
+  end
+end

data/spec/spec_helper.rb

@@ -0,0 +1,16 @@
+require 'rubygems'
+require 'bundler'
+begin
+  Bundler.setup(:default, :development)
+rescue Bundler::BundlerError => e
+  $stderr.puts e.message
+  $stderr.puts "Run `bundle install` to install missing gems"
+  exit e.status_code
+end
+require 'bacon'
+
+$LOAD_PATH.unshift(File.dirname(__FILE__))
+$LOAD_PATH.unshift(File.join(File.dirname(__FILE__), '..', 'lib'))
+require 'learning'
+
+Bacon.summary_on_exit

metadata

@@ -0,0 +1,141 @@
+--- !ruby/object:Gem::Specification
+name: ml
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+  prerelease: 
+platform: ruby
+authors:
+- Andrew Liu
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2011-10-03 00:00:00.000000000Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: rubyvis
+  requirement: &2154897000 !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: *2154897000
+- !ruby/object:Gem::Dependency
+  name: nokogiri
+  requirement: &2154896440 !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: *2154896440
+- !ruby/object:Gem::Dependency
+  name: bacon
+  requirement: &2154895820 !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: *2154895820
+- !ruby/object:Gem::Dependency
+  name: yard
+  requirement: &2154895300 !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 0.6.0
+  type: :development
+  prerelease: false
+  version_requirements: *2154895300
+- !ruby/object:Gem::Dependency
+  name: bundler
+  requirement: &2154894780 !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 1.0.0
+  type: :development
+  prerelease: false
+  version_requirements: *2154894780
+- !ruby/object:Gem::Dependency
+  name: jeweler
+  requirement: &2154894220 !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 1.6.4
+  type: :development
+  prerelease: false
+  version_requirements: *2154894220
+- !ruby/object:Gem::Dependency
+  name: rcov
+  requirement: &2154893720 !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: *2154893720
+description: Machine learning library in Ruby
+email: andrewliu33@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files:
+- LICENSE.txt
+- README.md
+files:
+- Gemfile
+- Gemfile.lock
+- LICENSE.txt
+- README.md
+- Rakefile
+- VERSION
+- lib/data/generator.rb
+- lib/data/parser.rb
+- lib/data/plotter.rb
+- lib/method/perceptron.rb
+- lib/ml.rb
+- ml.gemspec
+- spec/learning_spec.rb
+- spec/spec_helper.rb
+homepage: http://github.com/eggegg/ml
+licenses:
+- MIT
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  none: false
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+      segments:
+      - 0
+      hash: 3039846434950023552
+required_rubygems_version: !ruby/object:Gem::Requirement
+  none: false
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 1.8.10
+signing_key: 
+specification_version: 3
+summary: Machine learning library for Ruby
+test_files: []