machine_learner 0.0.1

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: ae7126878df4103a2ee33bd433c109970ec39c3e
+  data.tar.gz: 8f33eb69cd5b5f86bebe1659c3e529c789180217
+SHA512:
+  metadata.gz: 73bd4e5130971a2de7766f6c044d1e77792dfcb48e32eb7d16314a63aebfa22b3bf6e69c85c1d646e252e913e45a7637e692e838691ed29ce0719544bbe91616
+  data.tar.gz: 22439d8d45b57399efe48fb4addd0b9411b5b53be0d0c84c46b347719d666328c8be528c86acabe17d23ce116eb14e49bc4063f10f6cf101c002514067d49f5c

data/.gitignore

@@ -0,0 +1,15 @@
+/.bundle/
+/.yardoc
+/Gemfile.lock
+/_yardoc/
+/coverage/
+/doc/
+/pkg/
+/spec/reports/
+/tmp/
+*.bundle
+*.so
+*.o
+*.a
+mkmf.log
+machine_learner-*.gem

data/.rspec

@@ -0,0 +1,2 @@
+--format documentation
+--color

data/.travis.yml

@@ -0,0 +1,3 @@
+language: ruby
+rvm:
+  - 2.1.1

data/Gemfile

@@ -0,0 +1,4 @@
+source 'https://rubygems.org'
+
+# Specify your gem's dependencies in machine_learner.gemspec
+gemspec

data/LICENSE.txt

@@ -0,0 +1,22 @@
+Copyright (c) 2014 YutaTanaka
+
+MIT License
+
+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 +1,31 @@
+# MachineLearner
+
+TODO: Write a gem description
+
+## Installation
+
+Add this line to your application's Gemfile:
+
+```ruby
+gem 'machine_learner'
+```
+
+And then execute:
+
+    $ bundle
+
+Or install it yourself as:
+
+    $ gem install machine_learner
+
+## Usage
+
+TODO: Write usage instructions here
+
+## Contributing
+
+1. Fork it ( https://github.com/[my-github-username]/machine_learner/fork )
+2. Create your feature branch (`git checkout -b my-new-feature`)
+3. Commit your changes (`git commit -am 'Add some feature'`)
+4. Push to the branch (`git push origin my-new-feature`)
+5. Create a new Pull Request

data/Rakefile

@@ -0,0 +1,7 @@
+require "bundler/gem_tasks"
+require "rspec/core/rake_task"
+
+RSpec::Core::RakeTask.new(:spec)
+
+task :default => :spec
+

data/lib/machine_learner/adaboost.rb

@@ -0,0 +1,115 @@
+#coding utf-8
+
+require_relative 'dataset.rb'
+require_relative 'learner.rb'
+
+module MachineLearner
+  # 複数の識別器から最良のものを自動選択する学習器
+  class SimpleLearner < Learner
+    
+    # データを元に学習を行う
+    # classifiersの各戻り値に対し識別を行い、結果のもっともよいものを採用する
+    # @param datas [Array<DataSet>] トレーニングデータの配列
+    # @param ds [Array<Float>] 各トレーナーデータの重みの配列
+    # @return [Array<Boolean>] 識別結果の配列
+    def learn(datas, ds)
+      @log = []
+      enum_classifier(datas, ds) do |cl|
+        results = datas.map{|data| cl.test(data) }
+        epsilon = results.zip(ds).map{|r, w| r ? 0 : w }.inject(:+)
+        @log << {epsilon: epsilon, classifier: cl, results: results}
+      end
+      
+      @best = @log.min {|h1, h2| h1[:epsilon] <=> h2[:epsilon] }
+      @best[:results]
+    end
+    
+    # 識別器を列挙する
+    # 識別器はyieldでblockに渡す
+    # @param datas [Array<DataSet>] トレーニングデータの配列
+    # @param ds [Array<Float>] 各トレーナーデータの重みの配列
+    def classifiers(datas, ds)
+      raise NotImplementedError.new
+    end
+    
+    # 識別器を文字列で表現する
+    # @return [String] 識別器を表す文字列
+    def to_s
+      cl.to_s
+    end
+    
+    # 識別を行う
+    # @param x 特徴空間
+    # @return [Fixnum] 識別結果
+    def classify(x)
+      cl.classify(x)
+    end
+    
+    private
+    
+    # 性能の最も良い識別器を取得する
+    # @return [Classifier] 性能の最も良い識別器を取得する
+    def cl
+      @best[:classifier]
+    end
+    
+    # データを識別器に掛けた結果
+    # ハッシュの配列であり、各ハッシュはキー[:classifier, :result, :epsiron]を持つ
+    # @return log [Array<Hash>] 識別結果の配列
+    def log
+      @log
+    end
+    
+  end
+  
+  # AdaBoost クラス
+  class AdaBoost < Learner
+    
+    # コンストラクタ
+    # @param learners [Array<Learner>] 学習器のリスト
+    def initialize(learners)
+      @learners = learners
+      @alphas = []
+    end
+    
+    # データを元に学習を行う
+    # @param datas [Array<DataSet>] トレーニングデータの配列
+    # @return [Array<Boolean>] 識別結果の配列
+    def learn(datas, ds = nil)
+      ds ||= [1.0 / datas.size] * datas.size
+      
+      @learners.each do |learner|
+        # 学習記に順番に学習させる
+        # results : トレーニングデータの正解、不正解の配列
+        results = learner.learn(datas, ds)
+        # epsilon : 学習器のエラー率(失敗した学習データの重みの総和)
+        epsilon = results.zip(ds).map{|r, w| r ? 0 : w }.inject(:+)
+        epsilon = ds.min * 0.1 if(epsilon == 0)
+        # alpha : 学習器の重み(エラー率epsilonが低いほど高い値を取る)
+        alpha = Math.log((1 - epsilon) / epsilon) / 2
+        @alphas << alpha
+        # 重みの更新
+        ds = ds.map.with_index{|w, i| w * Math.exp(alpha * (results[i] ? -1 : 1)) }
+        z = ds.inject(&:+)
+        ds.map!{|w| w / z}
+      end
+    end
+    
+    # 識別を行う
+    # @param x 特徴空間
+    # @return [Fixnum] 識別結果
+    def classify(x)
+      score = [@learners, @alphas].transpose.reduce(0) { |score, (l, a)|
+        score += l.classify(x) * a 
+      }
+      score > 0 ? 1 : -1
+    end
+    
+    # Learnerを表すログ
+    def to_s
+      [@learners, @alphas].transpose.map {|l, a| "(#{a.round(3)} * #{l})" }.join(" + ");
+    end
+  end
+end
+
+# vim: set et ts=2 sts=2 sw=2:

data/lib/machine_learner/bayes.rb

@@ -0,0 +1,72 @@
+#coding utf-8
+
+require_relative 'dataset.rb'
+require_relative 'learner.rb'
+
+module MachineLearner
+  
+  # ナイブベイズ学習器
+  class BayesLearner < Learner
+    
+    # データを元に学習を行う
+    # @param datas [Array<DataSet>] トレーニングデータの配列
+    # @return [Array<Boolean>] 識別結果の配列
+    def learn(datas, ds = nil)
+      @training = datas
+      @ys = @training.map{|data| data.y }
+      @candidate_y = @ys.uniq.compact
+    end
+    
+    # 識別を行う
+    # @param x 特徴空間
+    # @return [Fixnum] 識別結果
+    def classify(xs)
+      # 最も尤度の高い候補 y を探す
+      max_y, max_likelihood = 0, -Float::INFINITY
+      @candidate_y.each do |y|
+        likelihood = p_y_x(y, xs)
+        max_y, max_likelihood = y, likelihood if max_likelihood < likelihood
+      end
+      return max_y
+    end
+    
+    private
+  
+    # P(Y | X) を計算
+    def p_y_x(y, xs)
+      p = Math.log(p_y(y))
+      xs.each_with_index do |xi, i|
+        next if xi.nil?
+        p += Math.log(p_xi_y(xi, i, y))
+      end
+      return p
+    end
+  
+    # P(Xi | Y) を計算
+    def p_xi_y(x, i, y)
+      total = count = 0
+      @ys.each_with_index do |ysj, j|
+        next if ysj != y
+        total += 1
+        count += 1 if @training[j].x[i] == x
+      end
+      return (count + 1).to_f / (total + candidate_x(i).size + 1).to_f
+    end
+  
+    # P(Y) を計算
+    def p_y(y)
+      count = @ys.count(y)
+      return (count + 1).to_f / (@ys.size + @candidate_y.size + 1).to_f
+    end
+  
+    # xの候補を列挙
+    def candidate_x(field)
+      @candidate_x ||= []
+      @candidate_x[field] ||= @training.map{|data| data.x[field]}.uniq.compact
+    end
+  end
+  
+end
+
+# vim: set et ts=2 sts=2 sw=2:
+

data/lib/machine_learner/dataset.rb

@@ -0,0 +1,24 @@
+#coding utf-8
+
+module MachineLearner
+  # 学習器にかけるデータ
+  # 特徴空間xと識別結果yを持つ
+  class DataSet
+    attr_reader :x, :y
+    
+    # コンストラクタ
+    # @param x 特徴空間
+    # @param y 識別結果
+    def initialize(x, y)
+      @x, @y = x, y
+    end
+    
+    # @return データセットを表現する文字列
+    def to_s
+      "(#{@x}=>#{@y})"
+    end
+    
+  end
+end
+
+# vim: set et ts=2 sts=2 sw=2:

data/lib/machine_learner/learner.rb

@@ -0,0 +1,44 @@
+#coding utf-8
+
+module MachineLearner
+  
+  # 識別器クラス
+  class Classifier
+    
+    # 識別を行う
+    # @param x 特徴空間
+    # @return [Fixnum] 識別結果
+    def classify(x)
+      raise NotImplementedError.new
+    end
+    
+    # 性能の測定を行う
+    # @param data[DataSet] データ
+    # @return [Boolean] 識別できたらtrue, 外れたらfalse
+    def test(data)
+      classify(data.x) == data.y
+    end
+    
+    # 性能の測定を行う
+    # @param datas[Array<DataSet>] テストデータセット
+    # @return [Float] 正解率(0以上1以下)
+    def evaluate(datas)
+      datas.count{|data| test(data)} / datas.size.to_f
+    end
+  end
+  
+  # 学習器クラス
+  class Learner < Classifier
+    
+    # トレーニングデータを元に学習を行う
+    # @param datas [Array<DataSet>] トレーニングデータセット
+    # @param ds [Array<Float>] 各データセットの重み
+    # @return [Array<Boolean>] 学習後の識別結果
+    def learn(datas, ds)
+      raise NotImplementedError.new
+    end
+  end
+  
+end
+
+# vim: set et ts=2 sts=2 sw=2:

data/lib/machine_learner/version.rb

@@ -0,0 +1,3 @@
+module MachineLearner
+  VERSION = "0.0.1"
+end

data/lib/machine_learner.rb

@@ -0,0 +1,3 @@
+require "machine_learner/version"
+require "machine_learner/adaboost"
+require "machine_learner/bayes"

data/machine_learner.gemspec

@@ -0,0 +1,27 @@
+# coding: utf-8
+lib = File.expand_path('../lib', __FILE__)
+$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
+require 'machine_learner/version'
+
+Gem::Specification.new do |spec|
+  spec.name          = "machine_learner"
+  spec.version       = MachineLearner::VERSION
+  spec.authors       = ["YutaTanaka"]
+  spec.email         = ["yuta84q.ihcarok@gmail.com"]
+  spec.summary       = %q{Library for machine learning.}
+  spec.description   = "This is a library for machine learning.\n" + 
+                       "You can use AdaBoost and Naive Bayes easily.\n"
+  spec.homepage      = "https://github.com/84q/machine_learning"
+  spec.license       = "MIT"
+
+  spec.files         = `git ls-files -z`.split("\x0")
+  spec.executables   = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
+  spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
+  spec.require_paths = ["lib"]
+
+  spec.add_development_dependency "bundler", "~> 1.6"
+  spec.add_development_dependency "rake", "~> 10.0"
+  spec.add_development_dependency 'rspec', '~> 0'
+end
+
+# vim: set et ts=2 sts=2 sw=2:

data/spec/machine_learner_spec.rb

@@ -0,0 +1,11 @@
+require 'spec_helper'
+
+describe MachineLearner do
+  it 'has a version number' do
+    expect(MachineLearner::VERSION).not_to be nil
+  end
+
+  it 'does something useful' do
+    expect(false).to eq(true)
+  end
+end

data/spec/spec_helper.rb

@@ -0,0 +1,2 @@
+$LOAD_PATH.unshift File.expand_path('../../lib', __FILE__)
+require 'machine_learner'

metadata

@@ -0,0 +1,107 @@
+--- !ruby/object:Gem::Specification
+name: machine_learner
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+platform: ruby
+authors:
+- YutaTanaka
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2014-08-10 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: bundler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.6'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.6'
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '10.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '10.0'
+- !ruby/object:Gem::Dependency
+  name: rspec
+  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'
+description: |
+  This is a library for machine learning.
+  You can use AdaBoost and Naive Bayes easily.
+email:
+- yuta84q.ihcarok@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- ".gitignore"
+- ".rspec"
+- ".travis.yml"
+- Gemfile
+- LICENSE.txt
+- README.md
+- Rakefile
+- lib/machine_learner.rb
+- lib/machine_learner/adaboost.rb
+- lib/machine_learner/bayes.rb
+- lib/machine_learner/dataset.rb
+- lib/machine_learner/learner.rb
+- lib/machine_learner/version.rb
+- machine_learner.gemspec
+- spec/machine_learner_spec.rb
+- spec/spec_helper.rb
+homepage: https://github.com/84q/machine_learning
+licenses:
+- MIT
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.2.2
+signing_key: 
+specification_version: 4
+summary: Library for machine learning.
+test_files:
+- spec/machine_learner_spec.rb
+- spec/spec_helper.rb
+has_rdoc: