reddavis-k_means 0.0.1

data/.document

@@ -0,0 +1,5 @@
+README.rdoc
+lib/**/*.rb
+bin/*
+features/**/*.feature
+LICENSE

data/.gitignore

@@ -0,0 +1,5 @@
+*.sw?
+.DS_Store
+coverage
+rdoc
+pkg

data/LICENSE

@@ -0,0 +1,20 @@
+Copyright (c) 2009 reddavis
+
+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.rdoc

@@ -0,0 +1,18 @@
+= k_means
+
+Attempting to build a memory efficient fast KMeans.
+
+== Note on Patches/Pull Requests
+ 
+* Fork the project.
+* Make your feature addition or bug fix.
+* Add tests for it. This is important so I don't break it in a
+  future version unintentionally.
+* Commit, do not mess with rakefile, version, or history.
+  (if you want to have your own version, that is fine but
+   bump version in a commit by itself I can ignore when I pull)
+* Send me a pull request. Bonus points for topic branches.
+
+== Copyright
+
+Copyright (c) 2009 reddavis. See LICENSE for details.

data/Rakefile

@@ -0,0 +1,57 @@
+require 'rubygems'
+require 'rake'
+
+begin
+  require 'jeweler'
+  Jeweler::Tasks.new do |gem|
+    gem.name = "k_means"
+    gem.summary = %Q{K Means algorithm}
+    gem.description = %Q{Attempting to create a fast memory efficient KMeans}
+    gem.email = "reddavis@gmail.com"
+    gem.homepage = "http://github.com/reddavis/k_means"
+    gem.authors = ["reddavis"]
+    # gem is a Gem::Specification... see http://www.rubygems.org/read/chapter/20 for additional settings
+  end
+
+rescue LoadError
+  puts "Jeweler (or a dependency) not available. Install it with: sudo gem install jeweler"
+end
+
+require 'rake/testtask'
+Rake::TestTask.new(:test) do |test|
+  test.libs << 'lib' << 'test'
+  test.pattern = 'test/**/*_test.rb'
+  test.verbose = true
+end
+
+begin
+  require 'rcov/rcovtask'
+  Rcov::RcovTask.new do |test|
+    test.libs << 'test'
+    test.pattern = 'test/**/*_test.rb'
+    test.verbose = true
+  end
+rescue LoadError
+  task :rcov do
+    abort "RCov is not available. In order to run rcov, you must: sudo gem install spicycode-rcov"
+  end
+end
+
+
+
+
+task :default => :test
+
+require 'rake/rdoctask'
+Rake::RDocTask.new do |rdoc|
+  if File.exist?('VERSION')
+    version = File.read('VERSION')
+  else
+    version = ""
+  end
+
+  rdoc.rdoc_dir = 'rdoc'
+  rdoc.title = "k_means #{version}"
+  rdoc.rdoc_files.include('README*')
+  rdoc.rdoc_files.include('lib/**/*.rb')
+end

data/VERSION

@@ -0,0 +1 @@
+0.0.1

data/benchmark/benchmark.rb

@@ -0,0 +1,25 @@
+require 'benchmark'
+require 'rubygems'
+require 'ai4r'
+require File.dirname(__FILE__) + '/../lib/k_means'
+
+data = Array.new(200) {Array.new(50) {rand(10)}}
+
+puts data.inspect
+
+ai4r_data = Ai4r::Data::DataSet.new(:data_items=> data)
+
+# Clustering can happen in magical ways
+# so lets do it over multiple times
+n = 2
+
+Benchmark.bm do |x|
+  x.report('Mine') do
+    a = KMeans.new(4)
+    n.times { a.clustify(data) }
+  end
+  x.report("Ai4R") do
+    b = Ai4r::Clusterers::KMeans.new
+    n.times { b.build(ai4r_data, 4) }
+  end
+end

data/k_means.gemspec

@@ -0,0 +1,56 @@
+# Generated by jeweler
+# DO NOT EDIT THIS FILE
+# Instead, edit Jeweler::Tasks in Rakefile, and run `rake gemspec`
+# -*- encoding: utf-8 -*-
+
+Gem::Specification.new do |s|
+  s.name = %q{k_means}
+  s.version = "0.0.1"
+
+  s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
+  s.authors = ["reddavis"]
+  s.date = %q{2009-08-15}
+  s.description = %q{Attempting to create a fast memory efficient KMeans}
+  s.email = %q{reddavis@gmail.com}
+  s.extra_rdoc_files = [
+    "LICENSE",
+     "README.rdoc"
+  ]
+  s.files = [
+    ".document",
+     ".gitignore",
+     "LICENSE",
+     "README.rdoc",
+     "Rakefile",
+     "VERSION",
+     "benchmark/benchmark.rb",
+     "k_means.gemspec",
+     "lib/basic_cache_store.rb",
+     "lib/ext/enumerable.rb",
+     "lib/k_means.rb",
+     "profiling/profile.rb",
+     "test/ext/test_enumerable.rb",
+     "test/helper.rb",
+     "test/test_k_means.rb"
+  ]
+  s.homepage = %q{http://github.com/reddavis/k_means}
+  s.rdoc_options = ["--charset=UTF-8"]
+  s.require_paths = ["lib"]
+  s.rubygems_version = %q{1.3.5}
+  s.summary = %q{K Means algorithm}
+  s.test_files = [
+    "test/ext/test_enumerable.rb",
+     "test/helper.rb",
+     "test/test_k_means.rb"
+  ]
+
+  if s.respond_to? :specification_version then
+    current_version = Gem::Specification::CURRENT_SPECIFICATION_VERSION
+    s.specification_version = 3
+
+    if Gem::Version.new(Gem::RubyGemsVersion) >= Gem::Version.new('1.2.0') then
+    else
+    end
+  else
+  end
+end

data/lib/basic_cache_store.rb

@@ -0,0 +1,15 @@
+class BasicCacheStore
+  
+  def initialize
+    @store = {}
+  end
+  
+  def set(key, data)
+    @store[key] = data
+  end
+  
+  def get(key)
+    @store[key]
+  end
+  
+end

data/lib/ext/enumerable.rb

@@ -0,0 +1,10 @@
+module Enumerable      
+  def euclidean_distance(other)
+    sum = 0.0
+    self.each_index do |i|
+      sum += (self[i] - other[i])**2
+    end
+    Math.sqrt(sum)
+  end
+end
+

data/lib/k_means.rb

@@ -0,0 +1,128 @@
+$: << File.dirname(__FILE__)
+require 'rubygems'
+require 'basic_cache_store'
+require 'ext/enumerable'
+
+class KMeans
+  
+  def initialize(k=4, options={})
+    @k = k
+    @verbose = options[:verbose] == true ? true : nil
+    @last_matches = nil
+  end
+  
+  def clustify(data)
+    @data = data
+    place_centroids
+    perform_cluster_process
+    @best_matches
+  end
+  
+  private
+  
+  def get_best_distance(data_index, centroid_index, data)
+    if cached_data = @cache.get("#{data_index}_#{centroid_index}")
+      cached_data
+    else
+      data.euclidean_distance(@centroids[centroid_index])
+    end
+  end
+  
+  def set_best_distance(data_index, centroid_index, data)
+    @cache.set("#{data_index}_#{centroid_index}", data.euclidean_distance(@centroids[centroid_index]))
+  end
+  
+  def perform_cluster_process
+    100.times do |t|
+      verbose_message("Iteration #{t}")
+      
+      # Prepare best matches array
+      @best_matches = create_best_matches_array
+      # A little bit of caching
+      @cache = BasicCacheStore.new
+      
+      # See which centroid is closest to which data
+      @data.each_with_index do |data, index|
+        best_match = 0
+        
+        @k.times do |i|
+          # Calculate the distance between the centroid and the data
+          distance = data.euclidean_distance(@centroids[i])
+          # Check to see if our new distance is better than what we had before
+          if distance < get_best_distance(index, best_match, data)#data.euclidean_distance(@centroids[best_match])
+            best_match = i
+            set_best_distance(index, best_match, data)
+          end #if distance...
+        end #@k.times
+        @best_matches[best_match] << index
+      end #@data.each_with...
+      
+      # Stop the loop if centroids have stopped moving
+      break if @last_matches == @best_matches
+      @last_matches = @best_matches
+      
+      reposition_centroids
+    end
+  end
+  
+  # Move the centroids to the average of their surrounding data
+  def reposition_centroids
+    @k.times do |i|
+      averages = [0.0] * @data[0].size # The average data
+      # Here we create an average of all the data in @best_matches[i]
+      # and then move the centroid (basically replacing the centroids own data with the average)
+      # i.e the data is the posistion of the element, if that makes sense?)
+      if @best_matches[i].size > 0 # Check the centroid has any matches
+        @best_matches[i].each do |data_index|
+          @data[data_index].each_with_index do |data, index| 
+            averages[index] += data
+          end
+        end 
+        
+        # Calculate last part of the average
+        averages.each do |average|
+          average /= @best_matches[i].size
+        end
+        @centroids[i] = averages
+      end #if @best_matches
+      
+    end
+  end
+  
+  def place_centroids
+    @centroids = []
+    ranges = create_ranges
+    
+    @k.times do |i|
+      line_size = @data.first.size
+            
+      ranges.each do |range|
+        group = []
+        line_size.times do |n|
+          group << rand * (range[1] - range[0]) + range[0]
+        end
+        @centroids << group
+      end
+    end
+  end
+  
+  # Calculate the ranges for each points
+  def create_ranges
+    ranges = []
+    @data.each do |line|
+      ranges << [line.max, line.min]
+    end
+    ranges
+  end
+  
+  def verbose_message(message)
+    puts message if @verbose
+  end
+  
+  def create_best_matches_array
+    array = []
+    @k.times { array << []}
+    array
+  end
+    
+end

data/profiling/profile.rb

@@ -0,0 +1,13 @@
+require File.dirname(__FILE__) + '/../lib/k_means'
+require 'rubygems'
+require 'ruby-prof'
+
+data = Array.new(100) {Array.new(2) {rand}}
+
+result = RubyProf.profile do
+  a = KMeans.new(4)
+  a.clustify(data)
+end
+
+printer = RubyProf::GraphPrinter.new(result)
+printer.print(STDOUT, 0)

data/test/ext/test_enumerable.rb

@@ -0,0 +1,11 @@
+require 'helper'
+
+class TestEnumerable < Test::Unit::TestCase
+  context "Euclidean Distance" do
+    
+    should "return 5" do
+      assert_equal 5, [10].euclidean_distance([5])
+    end
+    
+  end
+end

data/test/helper.rb

@@ -0,0 +1,10 @@
+require 'rubygems'
+require 'test/unit'
+require 'shoulda'
+
+$LOAD_PATH.unshift(File.join(File.dirname(__FILE__), '..', 'lib'))
+$LOAD_PATH.unshift(File.dirname(__FILE__))
+require 'k_means'
+
+class Test::Unit::TestCase
+end

data/test/test_k_means.rb

@@ -0,0 +1,28 @@
+require 'helper'
+
+class TestKMeans < Test::Unit::TestCase
+  context "A KMeans Instance" do
+    
+    setup do
+      @kmeans = KMeans.new(4)
+      @data = Array.new(10) {Array.new(2) {rand}}
+    end
+  
+    should "return an array" do
+      assert_kind_of Array, @kmeans.clustify(@data)
+    end
+    
+    should "have 4 centroids" do
+      centroids = @kmeans.clustify(@data).size
+      assert_equal(4, centroids)
+    end
+    
+    should "return same amount of data that went in" do
+      output_data_count = @kmeans.clustify(@data).inject(0) do |sum, n|
+        sum += n.size
+      end
+      assert_equal(@data.size, output_data_count)
+    end
+    
+  end
+end

metadata

@@ -0,0 +1,71 @@
+--- !ruby/object:Gem::Specification 
+name: reddavis-k_means
+version: !ruby/object:Gem::Version 
+  version: 0.0.1
+platform: ruby
+authors: 
+- reddavis
+autorequire: 
+bindir: bin
+cert_chain: []
+
+date: 2009-08-15 00:00:00 -07:00
+default_executable: 
+dependencies: []
+
+description: Attempting to create a fast memory efficient KMeans
+email: reddavis@gmail.com
+executables: []
+
+extensions: []
+
+extra_rdoc_files: 
+- LICENSE
+- README.rdoc
+files: 
+- .document
+- .gitignore
+- LICENSE
+- README.rdoc
+- Rakefile
+- VERSION
+- benchmark/benchmark.rb
+- k_means.gemspec
+- lib/basic_cache_store.rb
+- lib/ext/enumerable.rb
+- lib/k_means.rb
+- profiling/profile.rb
+- test/ext/test_enumerable.rb
+- test/helper.rb
+- test/test_k_means.rb
+has_rdoc: false
+homepage: http://github.com/reddavis/k_means
+licenses: 
+post_install_message: 
+rdoc_options: 
+- --charset=UTF-8
+require_paths: 
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+required_rubygems_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+requirements: []
+
+rubyforge_project: 
+rubygems_version: 1.3.5
+signing_key: 
+specification_version: 3
+summary: K Means algorithm
+test_files: 
+- test/ext/test_enumerable.rb
+- test/helper.rb
+- test/test_k_means.rb