k_means 0.0.3 → 0.0.4

data/README.rdoc

@@ -16,6 +16,34 @@ Attempting to build a fast, memory efficient K-Means program.
   kmeans.inspect  # Use kmeans.view to get hold of the un-inspected array
   => [[3, 4], [0, 1, 2]]
   
+== Distance Measurements
+
+KMeans uses the Distance Measures Gem (http://github.com/reddavis/Distance-Measures) so we get quite a range of distance measurements.
+
+The measurements currently available are:
+
+  euclidean_distance
+
+  cosine_similarity
+
+  jaccard_index
+
+  jaccard_distance
+
+  binary_jaccard_index
+
+  binary_jaccard_distance
+
+  tanimoto_coefficient
+  
+To specify a particular one to use in the KMeans algorithm, just provide it as an option:
+
+  KMeans.new(@data, :similarity_measure => :jaccard_index)
+  KMeans.new(@data, :similarity_measure => :cosine_similarity)
+  KMeans.new(@data, :similarity_measure => :tanimoto_coefficient)
+  
+You get the idea...
+  
 == Benchmarks
 
   # 1000 records with 50 dimensions

data/Rakefile

@@ -10,6 +10,7 @@ begin
     gem.email = "reddavis@gmail.com"
     gem.homepage = "http://github.com/reddavis/k_means"
     gem.authors = ["reddavis"]
+    gem.add_dependency('distance_measures', '>= 0.0.0')
     # gem is a Gem::Specification... see http://www.rubygems.org/read/chapter/20 for additional settings
   end
   Jeweler::GemcutterTasks.new

data/VERSION

@@ -1 +1 @@
-0.0.3
+0.0.4

data/benchmark/benchmark_ai4r.rb

@@ -3,13 +3,13 @@ require 'rubygems'
 require 'ai4r'
 require File.dirname(__FILE__) + '/../lib/k_means'
 
-data = Array.new(500) {Array.new(50) {rand(10)}}
+data = Array.new(1000) {Array.new(50) {rand(10)}} 
 
 ai4r_data = Ai4r::Data::DataSet.new(:data_items=> data)
 
 # Clustering can happen in magical ways
 # so lets do it over multiple times
-n = 20
+n = 5
 
 
 Benchmark.bm do |x|

data/k_means.gemspec

@@ -5,11 +5,11 @@
 
 Gem::Specification.new do |s|
   s.name = %q{k_means}
-  s.version = "0.0.3"
+  s.version = "0.0.4"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.authors = ["reddavis"]
-  s.date = %q{2009-11-25}
+  s.date = %q{2010-01-25}
   s.description = %q{Attempting to create a fast, memory efficient KMeans}
   s.email = %q{reddavis@gmail.com}
   s.extra_rdoc_files = [
@@ -25,18 +25,17 @@ Gem::Specification.new do |s|
      "VERSION",
      "benchmark/benchmark_ai4r.rb",
      "k_means.gemspec",
-     "lib/centroid.rb",
-     "lib/ext/enumerable.rb",
      "lib/ext/object.rb",
      "lib/k_means.rb",
-     "lib/node.rb",
+     "lib/k_means/centroid.rb",
+     "lib/k_means/k_means.rb",
+     "lib/k_means/node.rb",
      "profiling/profile.rb",
-     "test/ext/test_enumerable.rb",
      "test/ext/test_object.rb",
      "test/helper.rb",
-     "test/test_centroid.rb",
-     "test/test_k_means.rb",
-     "test/test_node.rb"
+     "test/k_means/test_centroid.rb",
+     "test/k_means/test_k_means.rb",
+     "test/k_means/test_node.rb"
   ]
   s.homepage = %q{http://github.com/reddavis/k_means}
   s.rdoc_options = ["--charset=UTF-8"]
@@ -44,12 +43,11 @@ Gem::Specification.new do |s|
   s.rubygems_version = %q{1.3.5}
   s.summary = %q{K Means algorithm}
   s.test_files = [
-    "test/ext/test_enumerable.rb",
-     "test/ext/test_object.rb",
+    "test/ext/test_object.rb",
      "test/helper.rb",
-     "test/test_centroid.rb",
-     "test/test_k_means.rb",
-     "test/test_node.rb"
+     "test/k_means/test_centroid.rb",
+     "test/k_means/test_k_means.rb",
+     "test/k_means/test_node.rb"
   ]
 
   if s.respond_to? :specification_version then
@@ -57,9 +55,12 @@ Gem::Specification.new do |s|
     s.specification_version = 3
 
     if Gem::Version.new(Gem::RubyGemsVersion) >= Gem::Version.new('1.2.0') then
+      s.add_runtime_dependency(%q<distance_measures>, [">= 0.0.0"])
     else
+      s.add_dependency(%q<distance_measures>, [">= 0.0.0"])
     end
   else
+    s.add_dependency(%q<distance_measures>, [">= 0.0.0"])
   end
 end
 

data/lib/k_means/k_means.rb

@@ -0,0 +1,75 @@
+require 'ext/object'
+
+class KMeans
+  
+  attr_reader :centroids, :nodes
+  
+  def initialize(data, options={})
+    k = options[:centroids] || 4
+    @verbose = options[:verbose]
+    
+    similarity_measure = options[:similarity_measure] || :euclidean_distance
+    @nodes = Node.create_nodes(data, similarity_measure)
+    @centroids = Centroid.create_centroids(k, @nodes)
+    
+    perform_cluster_process
+  end
+  
+  def inspect
+    @centroid_pockets.inspect
+  end
+  
+  def view
+    @centroid_pockets
+  end
+  
+  private
+  
+  def perform_cluster_process
+    iterations, updates = 0, 1
+    while updates > 0 && iterations < 100 
+      iterations += 1
+      verbose_message("Iteration #{iterations}")
+      updates = 0  
+      updates += update_nodes
+      reposition_centroids
+    end
+    place_nodes_into_pockets
+  end
+  
+  # This creates an array of arrays
+  # Each internal array represents a centroid
+  # and each in the array represents the nodes index
+  def place_nodes_into_pockets
+    centroid_pockets = Array.new(@centroids.size) {[]}
+    @centroids.each_with_index do |centroid, centroid_index|
+      @nodes.each_with_index do |node, node_index|
+        if node.closest_centroid == centroid
+          centroid_pockets[centroid_index] << node_index
+        end
+      end
+    end
+    @centroid_pockets = centroid_pockets
+  end
+    
+  def update_nodes
+    sum = 0
+    @nodes.each do |node|
+      sum += node.update_closest_centroid(@centroids)
+    end
+    sum
+  end
+  
+  def reposition_centroids
+    @centroids.each do |centroid|
+      nodes = [] 
+      @nodes.each {|n| nodes << n if n.closest_centroid == centroid}
+      centroid.reposition(nodes)
+    end
+  end
+        
+  def verbose_message(message)
+    puts message if @verbose
+  end
+      
+end

data/lib/{node.rb → k_means/node.rb}

@@ -1,10 +1,10 @@
 class Node
   
   class << self
-    def create_nodes(data)
+    def create_nodes(data, similarity_measure)
       nodes = []
       data.each do |position|
-        nodes << new(position)
+        nodes << new(position, similarity_measure)
       end
       nodes
     end
@@ -12,8 +12,9 @@ class Node
   
   attr_accessor :position, :best_distance, :closest_centroid
   
-  def initialize(position)
+  def initialize(position, similarity_measure)
     @position = position
+    @similarity_measure = similarity_measure
   end
   
   def update_closest_centroid(centroids)
@@ -38,7 +39,11 @@ class Node
   end
   
   def calculate_distance(centroid)
-    @position.euclidean_distance(centroid.position)
+    begin
+      @position.send(@similarity_measure, centroid.position)
+    rescue NoMethodError
+      raise "Hey, that's not a measurement. Read the REAdME for available measurements"
+    end
   end
   
 end

data/lib/k_means.rb

@@ -1,78 +1,8 @@
-$: << File.dirname(__FILE__)
-require 'centroid'
-require 'node'
-require 'ext/enumerable'
-require 'ext/object'
+$:.unshift(File.dirname(__FILE__) + '/../lib')
 
-class KMeans
-  
-  attr_reader :centroids, :nodes
-  
-  def initialize(data, options={})
-    k = options[:centroids] || 4
-    @verbose = options[:verbose] == true ? true : nil
-        
-    @nodes = Node.create_nodes(data)
-    @centroids = Centroid.create_centroids(k, @nodes)
-    
-    perform_cluster_process
-  end
-  
-  def inspect
-    @centroid_pockets.inspect
-  end
-  
-  def view
-    @centroid_pockets
-  end
-  
-  private
-  
-  def perform_cluster_process
-    iterations, updates = 0, 1
-    while updates > 0 && iterations < 100 
-      iterations += 1
-      verbose_message("Iteration #{iterations}")
-      updates = 0  
-      updates += update_nodes
-      reposition_centroids
-    end
-    place_nodes_into_pockets
-  end
-  
-  # This creates an array of arrays
-  # Each internal array represents a centroid
-  # and each in the array represents the nodes index
-  def place_nodes_into_pockets
-    centroid_pockets = Array.new(@centroids.size) {[]}
-    @centroids.each_with_index do |centroid, centroid_index|
-      @nodes.each_with_index do |node, node_index|
-        if node.closest_centroid == centroid
-          centroid_pockets[centroid_index] << node_index
-        end
-      end
-    end
-    @centroid_pockets = centroid_pockets
-  end
-    
-  def update_nodes
-    sum = 0
-    @nodes.each do |node|
-      sum += node.update_closest_centroid(@centroids)
-    end
-    sum
-  end
-  
-  def reposition_centroids
-    @centroids.each do |centroid|
-      nodes = [] 
-      @nodes.each {|n| nodes << n if n.closest_centroid == centroid}
-      centroid.reposition(nodes)
-    end
-  end
-        
-  def verbose_message(message)
-    puts message if @verbose
-  end
-      
-end
+require 'k_means/k_means'
+require 'k_means/centroid'
+require 'k_means/node'
+
+# Gems
+require 'distance_measures'

data/test/{test_centroid.rb → k_means/test_centroid.rb}

@@ -40,6 +40,6 @@ class TestCentroid < Test::Unit::TestCase
   private
   
   def create_nodes
-    Node.create_nodes([[1,2,3], [4,5,6]])
+    Node.create_nodes([[1,2,3], [4,5,6]], :euclidean)
   end
 end

data/test/{test_k_means.rb → k_means/test_k_means.rb}

@@ -5,7 +5,7 @@ class TestKMeans < Test::Unit::TestCase
     
     setup do
       @data = Array.new(200) {Array.new(2) {rand}}
-      @kmeans = KMeans.new(@data, :centroids => 2)
+      @kmeans = KMeans.new(@data, :centroids => 2, :similarity_measure => :cosine_similarity)
     end
   
     should "return an array" do

data/test/{test_node.rb → k_means/test_node.rb}

@@ -4,7 +4,7 @@ class TestNode < Test::Unit::TestCase
   context "A Data Instance" do
     
     setup do
-      @node = Node.new([4, 4])
+      @node = Node.new([4, 4], :euclidean_distance)
     end
   
     should "return an array" do
@@ -13,13 +13,13 @@ class TestNode < Test::Unit::TestCase
     
     should "create an array of nodes" do
       data = Array.new(10) {Array.new(2) {rand}}
-      nodes = Node.create_nodes(data)
+      nodes = Node.create_nodes(data, :euclidean_distance)
       assert_kind_of Array, nodes
     end
     
     should "create 10 nodes" do
       data = Array.new(10) {Array.new(2) {rand}}
-      nodes = Node.create_nodes(data)
+      nodes = Node.create_nodes(data, :euclidean_distance)
       assert_equal 10, nodes.size
     end
     
@@ -35,9 +35,16 @@ class TestNode < Test::Unit::TestCase
       @node.update_closest_centroid(centroids)
       a = @node.closest_centroid
       @node.update_closest_centroid([Centroid.new([4,4])])
+      
       assert_not_equal a, @node.closest_centroid
       assert_equal 0, @node.best_distance
     end
+    
+    should "raise error if a false measure is specified" do
+      assert_raise NoMethodError do
+        Node.new([9,9], :fakey).calculate_distance([1,1])
+      end
+    end
         
   end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: k_means
 version: !ruby/object:Gem::Version 
-  version: 0.0.3
+  version: 0.0.4
 platform: ruby
 authors: 
 - reddavis
@@ -9,10 +9,19 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2009-11-25 00:00:00 +00:00
+date: 2010-01-25 00:00:00 +00:00
 default_executable: 
-dependencies: []
-
+dependencies: 
+- !ruby/object:Gem::Dependency 
+  name: distance_measures
+  type: :runtime
+  version_requirement: 
+  version_requirements: !ruby/object:Gem::Requirement 
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        version: 0.0.0
+    version: 
 description: Attempting to create a fast, memory efficient KMeans
 email: reddavis@gmail.com
 executables: []
@@ -31,18 +40,17 @@ files:
 - VERSION
 - benchmark/benchmark_ai4r.rb
 - k_means.gemspec
-- lib/centroid.rb
-- lib/ext/enumerable.rb
 - lib/ext/object.rb
 - lib/k_means.rb
-- lib/node.rb
+- lib/k_means/centroid.rb
+- lib/k_means/k_means.rb
+- lib/k_means/node.rb
 - profiling/profile.rb
-- test/ext/test_enumerable.rb
 - test/ext/test_object.rb
 - test/helper.rb
-- test/test_centroid.rb
-- test/test_k_means.rb
-- test/test_node.rb
+- test/k_means/test_centroid.rb
+- test/k_means/test_k_means.rb
+- test/k_means/test_node.rb
 has_rdoc: true
 homepage: http://github.com/reddavis/k_means
 licenses: []
@@ -72,9 +80,8 @@ signing_key:
 specification_version: 3
 summary: K Means algorithm
 test_files: 
-- test/ext/test_enumerable.rb
 - test/ext/test_object.rb
 - test/helper.rb
-- test/test_centroid.rb
-- test/test_k_means.rb
-- test/test_node.rb
+- test/k_means/test_centroid.rb
+- test/k_means/test_k_means.rb
+- test/k_means/test_node.rb

data/lib/ext/enumerable.rb

@@ -1,10 +0,0 @@
-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/test/ext/test_enumerable.rb

@@ -1,11 +0,0 @@
-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