measurable 0.0.6 → 0.0.7

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 8e94547914fc6532f23f86d4d8118100aad64398
-  data.tar.gz: f5eae04dfb5c5b2e041d59527921be80ba1b27bc
+  metadata.gz: aaf11caa585477217fa9724f7f8c8a794490194f
+  data.tar.gz: 2e802d508407c4b656501c9a18af5130ae488979
 SHA512:
-  metadata.gz: 8992dfff181353a0feb8d893f90c4705572f2706e158e6ee51f6d398c31ddd0b39f8f1153c123252eb064213984c11ec6d98cf720814065973a3d5a59884ebe7
-  data.tar.gz: 8643ac26d9471a6e1a78ca9b25c684196ad7fd61d7f930b0d2242b9d68d7b857ff7dd5c49da667517fd7a9ebf4593528947ebbb812ef9045026d381b5b76f978
+  metadata.gz: 971e40facf06a090f515d1cd450006ce370d6ff460c8a9af65c9791ced5fddba58defdaff70c51cd9ce202ce712c7e0074b387507155de2c93250725cd3547f0
+  data.tar.gz: 7f2714aeca034f418567b555c0fd5f99bcfce34d2ab73d595a2d4aa5d6cb5abdd620e2502182ce8537d0d7c1dccdba4285dabe4f9877e175a4973ae1ff155126

data/.gitignore

@@ -1,4 +1,6 @@
 pkg
 tmp/*
 benchmarks/*
-lib/*.bundle
+lib/*.bundle
+html/
+Gemfile.lock

data/.travis.yml

@@ -0,0 +1,8 @@
+language: ruby
+rvm:
+  - 1.9.3
+  - 2.0
+  - 2.1
+  - rbx
+# uncomment this line if your project needs to run something other than `rake`:
+# script: bundle exec rspec spec

data/Gemfile

@@ -1,3 +1,4 @@
 # Gemfile
-source "http://rubygems.org"
+source "https://rubygems.org"
+
 gemspec

data/README.md

@@ -1,5 +1,8 @@
 # Measurable
 
+[![Build Status](https://travis-ci.org/agarie/measurable.svg?branch=master)](https://travis-ci.org/agarie/measurable)
+[![Code Climate](https://codeclimate.com/github/agarie/measurable.png)](https://codeclimate.com/github/agarie/measurable)
+
 A gem to test what metric is best for certain kinds of datasets in machine learning.
 
 Besides the `Array` class, I also want to support `NVector` (from [NMatrix](http://github.com/sciruby/nmatrix)).
@@ -30,6 +33,7 @@ The following are the similarity measures supported at the moment:
 - Minkowski (Cityblock or Manhattan) distance
 - Chebyshev distance
 - Hamming distance
+- [Levenshtein distance](http://en.wikipedia.org/wiki/Levenshtein_distance)
 
 These still need to be implemented:
 
@@ -49,8 +53,8 @@ The API I intend to support is something like this:
 ```ruby
 require "measurable"
 
-u = NVector.ones(2)
-v = NVector.zeros(2)
+u = NMatrix.ones([2, 1])
+v = NMatrix.zeros([2, 1])
 w = [1, 0]
 x = [2, 2]
 
@@ -68,11 +72,9 @@ Measurable.euclidean_squared([3, 4]) # => 25
 `RDoc` syntax is used to document the project. To build it locally, you'll need to install the [Fivefish generator](https://github.com/ged/rdoc-generator-fivefish) (`gem install rdoc-generator-fivefish`) and run the following command:
 
 ```bash
-rdoc -f fivefish -m README.md *.md LICENSE lib/
+rake rdoc
 ```
 
-I want to be able to use a Rake task to generate the documentation, thus allowing me to forget the specific command. However, there's a bug in `RDoc::Task` in which [custom generators (like Fivefish) can't be used](https://github.com/rdoc/rdoc/issues/246).
-
 If there's something wrong with an explanation or if there's information missing, please open an issue or send a pull request.
 
 ## License
@@ -82,4 +84,4 @@ See LICENSE for details.
 The original `distance_measures` gem is copyrighted by [@reddavis][reddavis].
 
 [maxmin]: http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=05156398
-[reddavis]: (https://github.com/reddavis)
+[reddavis]: (https://github.com/reddavis)

data/Rakefile

@@ -1,7 +1,7 @@
 require 'rake'
 require 'bundler/gem_tasks'
 require "rspec/core/rake_task"
-# require 'rdoc/task' # See below.
+require 'rdoc/task'
 
 # Setup the necessary gems, specified in the gemspec.
 require 'bundler'
@@ -13,20 +13,26 @@ rescue Bundler::BundlerError => e
   exit e.status_code
 end
 
+task :default => [:spec]
+
 # Run all the specs.
 RSpec::Core::RakeTask.new(:spec)
 
-# RDoc task isn't working with custom generators, as can be seen in:
-# https://github.com/rdoc/rdoc/issues/246
-#
-# Whenever this issue is fixed, I'll resume using this task.
-#
-# RDoc::Task.new do |rdoc|
-#   rdoc.main = "README.md"
-#   rdoc.rdoc_files.include("README.md", "LICENSE", "lib")
-#   rdoc.generator = "fivefish"
-#   rdoc.external = true
-# end
+RDoc::Task.new do |rdoc|
+  rdoc.main = "README.md"
+  rdoc.rdoc_files.include("README.md", "LICENSE", "lib")
+  rdoc.generator = "fivefish"
+  rdoc.external = true
+end
+
+desc "Open IRB with Measurable loaded."
+task :console do
+  require 'irb'
+  require 'irb/completion'
+  require 'measurable'
+  ARGV.clear
+  IRB.start
+end
 
 # Compile task.
 # Rake::ExtensionTask.new do |ext|

data/lib/measurable.rb

@@ -10,6 +10,7 @@ require 'measurable/chebyshev'
 require 'measurable/maxmin'
 require 'measurable/haversine'
 require 'measurable/hamming'
+require 'measurable/levenshtein'
 
 module Measurable
   # PI / 180 degrees.

data/lib/measurable/cosine.rb

@@ -1,9 +1,9 @@
 module Measurable
 
   # call-seq:
-  #     cosine(u, v) -> Float
+  #     cosine_similarity(u, v) -> Float
   #
-  # Calculate the similarity between the orientation of two vectors.
+  # Calculate the cosine similarity between the orientation of two vectors.
   #
   # See: http://en.wikipedia.org/wiki/Cosine_similarity
   #
@@ -16,7 +16,7 @@ module Measurable
   # * *Raises* :
   #   - +ArgumentError+ -> The sizes of +u+ and +v+ don't match.
   #
-  def cosine(u, v)
+  def cosine_similarity(u, v)
     # TODO: Change this to a more specific, custom-made exception.
     raise ArgumentError if u.size != v.size
 
@@ -24,4 +24,27 @@ module Measurable
 
     dot_product / (euclidean(u) * euclidean(v))
   end
-end
+
+  # call-seq:
+  #     cosine_distance(u, v) -> Float
+  #
+  # Calculate the cosine distance between the orientation of two vectors.
+  #
+  # See: http://en.wikipedia.org/wiki/Cosine_similarity
+  #
+  # * *Arguments* :
+  #   - +u+ -> An array of Numeric objects.
+  #   - +v+ -> An array of Numeric objects.
+  # * *Returns* :
+  #   - The normalized dot product of +u+ and +v+, that is, the angle between
+  #     them in the n-dimensional space.
+  # * *Raises* :
+  #   - +ArgumentError+ -> The sizes of +u+ and +v+ don't match.
+  #
+  def cosine_distance(u, v)
+    # TODO: Change this to a more specific, custom-made exception.
+    raise ArgumentError if u.size != v.size
+
+    1 - cosine_similarity(u, v)
+  end
+end

data/lib/measurable/hamming.rb

@@ -3,9 +3,11 @@ module Measurable
   # call-seq:
   #     hamming(s1, s2) -> Integer
   #
+  # Count the number of different characters between strings +s1+ and +s2+,
+  # that is, how many substitutions are necessary to change +s1+ into +s2+ and
+  # vice-versa.
   #
-  #
-  # See: http://en.wikipedia.org/wiki/Cosine_similarity
+  # See: http://en.wikipedia.org/wiki/Hamming_distance
   #
   # * *Arguments* :
   #   - +s1+ -> A String.

data/lib/measurable/jaccard.rb

@@ -10,15 +10,15 @@ module Measurable
   # cardinality of set x.
   #
   # For example:
-  #   jaccard_index([1, 0, 1], [1, 1, 1]) == 0.666...
+  #   jaccard_index([1, 0, 1], [1, 1, 1]) == 0.5
   #
-  # Because |intersection| = |(1, 0, 1)| = 2 and |union| = |(1, 1, 1)| = 3.
+  # Because |intersection| = |(1)| = 1 and |union| = |(0, 1)| = 2.
   #
   # See: http://en.wikipedia.org/wiki/Jaccard_coefficient
   #
   # * *Arguments* :
-  #   - +u+ -> Array of 1s and 0s.
-  #   - +v+ -> Array of 1s and 0s.
+  #   - +u+ -> Array.
+  #   - +v+ -> Array.
   # * *Returns* :
   #   - Float value representing the Jaccard similarity coefficient between
   #     +u+ and +v+.
@@ -29,17 +29,10 @@ module Measurable
     # TODO: Change this to a more specific, custom-made exception.
     raise ArgumentError if u.size != v.size
 
-    intersection = u.zip(v).reduce(0) do |acc, elem|
-      # Both u and v must have this element.
-      elem[0] + elem[1] == 2 ? (acc + 1) : acc
-    end
+    intersection = u & v
+    union = u | v
 
-    union = u.zip(v).reduce(0) do |acc, elem|
-      # One of u and v must have this element.
-      elem[0] + elem[1] >= 1 ? (acc + 1) : acc
-    end
-
-    intersection.to_f / union
+    intersection.length.to_f / union.length
   end
 
   # call-seq:
@@ -56,8 +49,8 @@ module Measurable
   #   - Triangular inequality: jaccard(u, v) <= jaccard(u, w) + jaccard(w, v)
   #
   # * *Arguments* :
-  #   - +u+ -> Array of 1s and 0s.
-  #   - +v+ -> Array of 1s and 0s.
+  #   - +u+ -> Array.
+  #   - +v+ -> Array.
   # * *Returns* :
   #   - Float value representing the dissimilarity between +u+ and +v+.
   # * *Raises* :
@@ -66,4 +59,4 @@ module Measurable
   def jaccard(u, v)
     1 - jaccard_index(u, v)
   end
-end
+end

data/lib/measurable/levenshtein.rb

@@ -0,0 +1,58 @@
+module Measurable
+  module Levenshtein
+
+    # call-seq:
+    #     levenshtein(u, v) -> Integer
+    #
+    # Give the edit distance between two binary sequences +u+ and +v+ where each
+    # edit (insertion, deletion, substitution) required to change on into the
+    # other increments the total distance.
+    #
+    # For example:
+    #   levenshtein('kitten', 'sitting') == 3
+    #
+    # Because
+    # 1. kitten -> sitten (substitution "s" for "k")
+    # 2. sitten -> sittin (substitution "i" for "e")
+    # 3. sittin -> sitting (insertion of "g" at the end)
+    #
+    # See: http://en.wikipedia.org/wiki/Levenshtein_distance
+    #
+    # * *Arguments* :
+    #   - +u+ -> Array or String.
+    #   - +v+ -> Array or String.
+    # * *Returns* :
+    #   - Integer value representing the Levenshtein distance between
+    #     +u+ and +v+.
+    #
+    def levenshtein(u, v)
+      return 0 if u == v
+      return u.size if v.size == 0
+      return v.size if u.size == 0
+
+      matrix = Array.new(u.size+1) { (0..v.size).to_a }
+
+      if v.size < u.size
+        u, v = v, u
+      end
+
+      (1..u.size).each do |i|
+        (1..v.size).each do |j|
+          if u[i] == v[j]
+            matrix[i][j] = matrix[i-1][j-1]
+          else
+            matrix[i][j] = [
+              matrix[i-1][j] + 1,   # deletion
+              matrix[i][j-1] + 1,   # insertion
+              matrix[i-1][j-1] + 1, # substitution
+            ].min
+          end
+        end
+      end
+
+      matrix[u.size][v.size]
+    end
+  end
+
+  extend Measurable::Levenshtein
+end

data/lib/measurable/version.rb

@@ -1,3 +1,3 @@
 module Measurable
-  VERSION = "0.0.6" # :nodoc:
-end
+  VERSION = "0.0.7" # :nodoc:
+end

data/measurable.gemspec

@@ -24,6 +24,8 @@ Gem::Specification.new do |gem|
   gem.required_ruby_version = '>= 1.9.3'
 
   gem.add_development_dependency 'bundler'
-  gem.add_development_dependency 'rake', '~> 0.9'
-  gem.add_development_dependency 'rspec', '~> 2.9.0'
+  gem.add_development_dependency 'rake', '~> 10.1'
+  gem.add_development_dependency 'rdoc', '~> 4.1'
+  gem.add_development_dependency 'rspec', '~> 2.1'
+  gem.add_development_dependency 'rdoc-generator-fivefish'
 end

data/spec/cosine_spec.rb

@@ -1,29 +1,61 @@
-describe "Cosine distance" do
+describe "Cosine" do
 
-  before :all do
-    @u = [1, 2]
-    @v = [2, 3]
-    @w = [4, 5]
-  end
+  context "Similarity" do
+    before :all do
+      @u = [1, 2]
+      @v = [2, 3]
+      @w = [4, 5]
+    end
 
-  it "accepts two arguments" do
-    expect { Measurable.cosine(@u, @v) }.to_not raise_error
-    expect { Measurable.cosine(@u, @v, @w) }.to raise_error(ArgumentError)
-  end
+    it "accepts two arguments" do
+      expect { Measurable.cosine_similarity(@u, @v) }.to_not raise_error
+      expect { Measurable.cosine_similarity(@u, @v, @w) }.to raise_error(ArgumentError)
+    end
 
-  it "should be symmetric" do
-    x = Measurable.cosine(@u, @v)
-    y = Measurable.cosine(@v, @u)
+    it "should be symmetric" do
+      x = Measurable.cosine_similarity(@u, @v)
+      y = Measurable.cosine_similarity(@v, @u)
 
-    x.should be_within(TOLERANCE).of(y)
-  end
+      x.should be_within(TOLERANCE).of(y)
+    end
 
-  it "should return the correct value" do
-    x = Measurable.cosine(@u, @v)
-    x.should be_within(TOLERANCE).of(0.992277877)
+    it "should return the correct value" do
+      x = Measurable.cosine_similarity(@u, @v)
+      x.should be_within(TOLERANCE).of(0.992277877)
+    end
+
+    it "shouldn't work with vectors of different length" do
+      expect { Measurable.cosine_similarity(@u, [1, 3, 5, 7]) }.to raise_error(ArgumentError)
+    end
   end
 
-  it "shouldn't work with vectors of different length" do
-    expect { Measurable.cosine(@u, [1, 3, 5, 7]) }.to raise_error(ArgumentError)
+  context "Distance" do
+    before :all do
+      @u = [1, 2]
+      @v = [2, 3]
+      @w = [4, 5]
+    end
+
+    it "accepts two arguments" do
+      expect { Measurable.cosine_distance(@u, @v) }.to_not raise_error
+      expect { Measurable.cosine_distance(@u, @v, @w) }.to raise_error(ArgumentError)
+    end
+
+    it "should be symmetric" do
+      x = Measurable.cosine_distance(@u, @v)
+      y = Measurable.cosine_distance(@v, @u)
+
+      x.should be_within(TOLERANCE).of(y)
+    end
+
+    it "should return the correct value" do
+      x = Measurable.cosine_distance(@u, @v)
+      # TODO: Use a real example.
+      x.should be_within(TOLERANCE).of(1.0 - 0.992277877)
+    end
+
+    it "shouldn't work with vectors of different length" do
+      expect { Measurable.cosine_distance(@u, [1, 3, 5, 7]) }.to raise_error(ArgumentError)
+    end
   end
-end
+end

data/spec/jaccard_spec.rb

@@ -22,7 +22,7 @@ describe "Jaccard" do
     it "should return the correct value" do
       x = Measurable.jaccard_index(@u, @v)
 
-      x.should be_within(TOLERANCE).of(2.0 / 3.0)
+      x.should be_within(TOLERANCE).of(1.0 / 2.0)
     end
 
     it "shouldn't work with vectors of different length" do
@@ -52,11 +52,11 @@ describe "Jaccard" do
     it "should return the correct value" do
       x = Measurable.jaccard(@u, @v)
 
-      x.should be_within(TOLERANCE).of(1.0 / 3.0)
+      x.should be_within(TOLERANCE).of(1.0 / 2.0)
     end
 
     it "shouldn't work with vectors of different length" do
       expect { Measurable.jaccard(@u, [1, 2, 3, 4]) }.to raise_error(ArgumentError)
     end
   end
-end
+end

data/spec/levenshtein_spec.rb

@@ -0,0 +1,55 @@
+describe Measurable::Levenshtein do
+
+  context "strings" do
+
+    it "handles empty" do
+      Measurable.levenshtein("", "").should == 0
+      Measurable.levenshtein("", "abcd").should == 4
+      Measurable.levenshtein("abcd", "").should == 4
+    end
+
+    it "should not count equality" do
+      Measurable.levenshtein("aa", "aa").should == 0
+    end
+
+    it "should count deletion" do
+      Measurable.levenshtein("ab", "a").should == 1
+    end
+
+    it "should count insertion" do
+      Measurable.levenshtein("ab", "abc").should == 1
+    end
+
+    it "should count substitution" do
+      Measurable.levenshtein("aa", "ab").should == 1
+    end
+
+  end
+
+  context "arrays" do
+
+    it "handles empty" do
+      Measurable.levenshtein([], []).should == 0
+      Measurable.levenshtein([], %w[ a b c d ]).should == 4
+      Measurable.levenshtein(%w[ a b c d ], []).should == 4
+    end
+
+    it "should not count equality" do
+      Measurable.levenshtein(%w[ a ], %w[ a ]).should == 0
+    end
+
+    it "should count deletion" do
+      Measurable.levenshtein(%w[ a b ], %w[ a ]).should == 1
+    end
+
+    it "should count insertion" do
+      Measurable.levenshtein(%w[ a b ], %w[ a b c ]).should == 1
+    end
+
+    it "should count substitution" do
+      Measurable.levenshtein(%w[ a a ], %w[ a b ]).should == 1
+    end
+
+  end
+
+end

data/spec/tanimoto_spec.rb

@@ -21,10 +21,10 @@ describe "Tanimoto distance" do
   it "should return the correct value" do
     x = Measurable.tanimoto(@u, @v)
 
-    x.should be_within(TOLERANCE).of(-Math.log2(2.0 / 3.0))
+    x.should be_within(TOLERANCE).of(-Math.log2(1.0 / 2.0))
   end
 
   it "shouldn't work with vectors of different length" do
     expect { Measurable.tanimoto(@u, [1, 3, 5, 7]) }.to raise_error(ArgumentError)
   end
-end
+end

metadata

@@ -1,67 +1,95 @@
 --- !ruby/object:Gem::Specification
 name: measurable
 version: !ruby/object:Gem::Version
-  version: 0.0.6
+  version: 0.0.7
 platform: ruby
 authors:
 - Carlos Agarie
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-08-04 00:00:00.000000000 Z
+date: 2014-04-14 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ~>
+    - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.9'
+        version: '10.1'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ~>
+    - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.9'
+        version: '10.1'
+- !ruby/object:Gem::Dependency
+  name: rdoc
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '4.1'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '4.1'
 - !ruby/object:Gem::Dependency
   name: rspec
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ~>
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.1'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.1'
+- !ruby/object:Gem::Dependency
+  name: rdoc-generator-fivefish
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
       - !ruby/object:Gem::Version
-        version: 2.9.0
+        version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ~>
+    - - ">="
       - !ruby/object:Gem::Version
-        version: 2.9.0
+        version: '0'
 description: A Ruby gem with a lot of distance measures for your projects.
 email: carlos.agarie@gmail.com
 executables: []
 extensions: []
 extra_rdoc_files: []
 files:
-- .gitignore
-- .rspec
+- ".gitignore"
+- ".rspec"
+- ".travis.yml"
 - Gemfile
-- Gemfile.lock
 - LICENSE
 - README.md
 - Rakefile
@@ -72,6 +100,7 @@ files:
 - lib/measurable/hamming.rb
 - lib/measurable/haversine.rb
 - lib/measurable/jaccard.rb
+- lib/measurable/levenshtein.rb
 - lib/measurable/maxmin.rb
 - lib/measurable/minkowski.rb
 - lib/measurable/tanimoto.rb
@@ -83,6 +112,7 @@ files:
 - spec/hamming_spec.rb
 - spec/haversine_spec.rb
 - spec/jaccard_spec.rb
+- spec/levenshtein_spec.rb
 - spec/maxmin_spec.rb
 - spec/minkowski_spec.rb
 - spec/spec_helper.rb
@@ -97,17 +127,17 @@ require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - ">="
     - !ruby/object:Gem::Version
       version: 1.9.3
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.0.6
+rubygems_version: 2.2.2
 signing_key: 
 specification_version: 4
 summary: A Ruby gem with a lot of distance measures for your projects.
@@ -118,6 +148,7 @@ test_files:
 - spec/hamming_spec.rb
 - spec/haversine_spec.rb
 - spec/jaccard_spec.rb
+- spec/levenshtein_spec.rb
 - spec/maxmin_spec.rb
 - spec/minkowski_spec.rb
 - spec/spec_helper.rb

data/Gemfile.lock

@@ -1,27 +0,0 @@
-PATH
-  remote: .
-  specs:
-    measurable (0.0.6)
-
-GEM
-  remote: http://rubygems.org/
-  specs:
-    diff-lcs (1.1.3)
-    rake (0.9.6)
-    rspec (2.9.0)
-      rspec-core (~> 2.9.0)
-      rspec-expectations (~> 2.9.0)
-      rspec-mocks (~> 2.9.0)
-    rspec-core (2.9.0)
-    rspec-expectations (2.9.1)
-      diff-lcs (~> 1.1.3)
-    rspec-mocks (2.9.0)
-
-PLATFORMS
-  ruby
-
-DEPENDENCIES
-  bundler
-  measurable!
-  rake (~> 0.9)
-  rspec (~> 2.9.0)