machine_learner 0.0.1
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/.gitignore +15 -0
- data/.rspec +2 -0
- data/.travis.yml +3 -0
- data/Gemfile +4 -0
- data/LICENSE.txt +22 -0
- data/README.md +31 -0
- data/Rakefile +7 -0
- data/lib/machine_learner/adaboost.rb +115 -0
- data/lib/machine_learner/bayes.rb +72 -0
- data/lib/machine_learner/dataset.rb +24 -0
- data/lib/machine_learner/learner.rb +44 -0
- data/lib/machine_learner/version.rb +3 -0
- data/lib/machine_learner.rb +3 -0
- data/machine_learner.gemspec +27 -0
- data/spec/machine_learner_spec.rb +11 -0
- data/spec/spec_helper.rb +2 -0
- metadata +107 -0
checksums.yaml
ADDED
@@ -0,0 +1,7 @@
|
|
1
|
+
---
|
2
|
+
SHA1:
|
3
|
+
metadata.gz: ae7126878df4103a2ee33bd433c109970ec39c3e
|
4
|
+
data.tar.gz: 8f33eb69cd5b5f86bebe1659c3e529c789180217
|
5
|
+
SHA512:
|
6
|
+
metadata.gz: 73bd4e5130971a2de7766f6c044d1e77792dfcb48e32eb7d16314a63aebfa22b3bf6e69c85c1d646e252e913e45a7637e692e838691ed29ce0719544bbe91616
|
7
|
+
data.tar.gz: 22439d8d45b57399efe48fb4addd0b9411b5b53be0d0c84c46b347719d666328c8be528c86acabe17d23ce116eb14e49bc4063f10f6cf101c002514067d49f5c
|
data/.gitignore
ADDED
data/.rspec
ADDED
data/.travis.yml
ADDED
data/Gemfile
ADDED
data/LICENSE.txt
ADDED
@@ -0,0 +1,22 @@
|
|
1
|
+
Copyright (c) 2014 YutaTanaka
|
2
|
+
|
3
|
+
MIT License
|
4
|
+
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining
|
6
|
+
a copy of this software and associated documentation files (the
|
7
|
+
"Software"), to deal in the Software without restriction, including
|
8
|
+
without limitation the rights to use, copy, modify, merge, publish,
|
9
|
+
distribute, sublicense, and/or sell copies of the Software, and to
|
10
|
+
permit persons to whom the Software is furnished to do so, subject to
|
11
|
+
the following conditions:
|
12
|
+
|
13
|
+
The above copyright notice and this permission notice shall be
|
14
|
+
included in all copies or substantial portions of the Software.
|
15
|
+
|
16
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
17
|
+
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
18
|
+
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
|
19
|
+
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
|
20
|
+
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
|
21
|
+
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
|
22
|
+
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
data/README.md
ADDED
@@ -0,0 +1,31 @@
|
|
1
|
+
# MachineLearner
|
2
|
+
|
3
|
+
TODO: Write a gem description
|
4
|
+
|
5
|
+
## Installation
|
6
|
+
|
7
|
+
Add this line to your application's Gemfile:
|
8
|
+
|
9
|
+
```ruby
|
10
|
+
gem 'machine_learner'
|
11
|
+
```
|
12
|
+
|
13
|
+
And then execute:
|
14
|
+
|
15
|
+
$ bundle
|
16
|
+
|
17
|
+
Or install it yourself as:
|
18
|
+
|
19
|
+
$ gem install machine_learner
|
20
|
+
|
21
|
+
## Usage
|
22
|
+
|
23
|
+
TODO: Write usage instructions here
|
24
|
+
|
25
|
+
## Contributing
|
26
|
+
|
27
|
+
1. Fork it ( https://github.com/[my-github-username]/machine_learner/fork )
|
28
|
+
2. Create your feature branch (`git checkout -b my-new-feature`)
|
29
|
+
3. Commit your changes (`git commit -am 'Add some feature'`)
|
30
|
+
4. Push to the branch (`git push origin my-new-feature`)
|
31
|
+
5. Create a new Pull Request
|
data/Rakefile
ADDED
@@ -0,0 +1,115 @@
|
|
1
|
+
#coding utf-8
|
2
|
+
|
3
|
+
require_relative 'dataset.rb'
|
4
|
+
require_relative 'learner.rb'
|
5
|
+
|
6
|
+
module MachineLearner
|
7
|
+
# 複数の識別器から最良のものを自動選択する学習器
|
8
|
+
class SimpleLearner < Learner
|
9
|
+
|
10
|
+
# データを元に学習を行う
|
11
|
+
# 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
|
12
|
+
# @param datas [Array<DataSet>] トレーニングデータの配列
|
13
|
+
# @param ds [Array<Float>] 各トレーナーデータの重みの配列
|
14
|
+
# @return [Array<Boolean>] 識別結果の配列
|
15
|
+
def learn(datas, ds)
|
16
|
+
@log = []
|
17
|
+
enum_classifier(datas, ds) do |cl|
|
18
|
+
results = datas.map{|data| cl.test(data) }
|
19
|
+
epsilon = results.zip(ds).map{|r, w| r ? 0 : w }.inject(:+)
|
20
|
+
@log << {epsilon: epsilon, classifier: cl, results: results}
|
21
|
+
end
|
22
|
+
|
23
|
+
@best = @log.min {|h1, h2| h1[:epsilon] <=> h2[:epsilon] }
|
24
|
+
@best[:results]
|
25
|
+
end
|
26
|
+
|
27
|
+
# 識別器を列挙する
|
28
|
+
# 識別器はyieldでblockに渡す
|
29
|
+
# @param datas [Array<DataSet>] トレーニングデータの配列
|
30
|
+
# @param ds [Array<Float>] 各トレーナーデータの重みの配列
|
31
|
+
def classifiers(datas, ds)
|
32
|
+
raise NotImplementedError.new
|
33
|
+
end
|
34
|
+
|
35
|
+
# 識別器を文字列で表現する
|
36
|
+
# @return [String] 識別器を表す文字列
|
37
|
+
def to_s
|
38
|
+
cl.to_s
|
39
|
+
end
|
40
|
+
|
41
|
+
# 識別を行う
|
42
|
+
# @param x 特徴空間
|
43
|
+
# @return [Fixnum] 識別結果
|
44
|
+
def classify(x)
|
45
|
+
cl.classify(x)
|
46
|
+
end
|
47
|
+
|
48
|
+
private
|
49
|
+
|
50
|
+
# 性能の最も良い識別器を取得する
|
51
|
+
# @return [Classifier] 性能の最も良い識別器を取得する
|
52
|
+
def cl
|
53
|
+
@best[:classifier]
|
54
|
+
end
|
55
|
+
|
56
|
+
# データを識別器に掛けた結果
|
57
|
+
# ハッシュの配列であり、各ハッシュはキー[:classifier, :result, :epsiron]を持つ
|
58
|
+
# @return log [Array<Hash>] 識別結果の配列
|
59
|
+
def log
|
60
|
+
@log
|
61
|
+
end
|
62
|
+
|
63
|
+
end
|
64
|
+
|
65
|
+
# AdaBoost クラス
|
66
|
+
class AdaBoost < Learner
|
67
|
+
|
68
|
+
# コンストラクタ
|
69
|
+
# @param learners [Array<Learner>] 学習器のリスト
|
70
|
+
def initialize(learners)
|
71
|
+
@learners = learners
|
72
|
+
@alphas = []
|
73
|
+
end
|
74
|
+
|
75
|
+
# データを元に学習を行う
|
76
|
+
# @param datas [Array<DataSet>] トレーニングデータの配列
|
77
|
+
# @return [Array<Boolean>] 識別結果の配列
|
78
|
+
def learn(datas, ds = nil)
|
79
|
+
ds ||= [1.0 / datas.size] * datas.size
|
80
|
+
|
81
|
+
@learners.each do |learner|
|
82
|
+
# 学習記に順番に学習させる
|
83
|
+
# results : トレーニングデータの正解、不正解の配列
|
84
|
+
results = learner.learn(datas, ds)
|
85
|
+
# epsilon : 学習器のエラー率(失敗した学習データの重みの総和)
|
86
|
+
epsilon = results.zip(ds).map{|r, w| r ? 0 : w }.inject(:+)
|
87
|
+
epsilon = ds.min * 0.1 if(epsilon == 0)
|
88
|
+
# alpha : 学習器の重み(エラー率epsilonが低いほど高い値を取る)
|
89
|
+
alpha = Math.log((1 - epsilon) / epsilon) / 2
|
90
|
+
@alphas << alpha
|
91
|
+
# 重みの更新
|
92
|
+
ds = ds.map.with_index{|w, i| w * Math.exp(alpha * (results[i] ? -1 : 1)) }
|
93
|
+
z = ds.inject(&:+)
|
94
|
+
ds.map!{|w| w / z}
|
95
|
+
end
|
96
|
+
end
|
97
|
+
|
98
|
+
# 識別を行う
|
99
|
+
# @param x 特徴空間
|
100
|
+
# @return [Fixnum] 識別結果
|
101
|
+
def classify(x)
|
102
|
+
score = [@learners, @alphas].transpose.reduce(0) { |score, (l, a)|
|
103
|
+
score += l.classify(x) * a
|
104
|
+
}
|
105
|
+
score > 0 ? 1 : -1
|
106
|
+
end
|
107
|
+
|
108
|
+
# Learnerを表すログ
|
109
|
+
def to_s
|
110
|
+
[@learners, @alphas].transpose.map {|l, a| "(#{a.round(3)} * #{l})" }.join(" + ");
|
111
|
+
end
|
112
|
+
end
|
113
|
+
end
|
114
|
+
|
115
|
+
# vim: set et ts=2 sts=2 sw=2:
|
@@ -0,0 +1,72 @@
|
|
1
|
+
#coding utf-8
|
2
|
+
|
3
|
+
require_relative 'dataset.rb'
|
4
|
+
require_relative 'learner.rb'
|
5
|
+
|
6
|
+
module MachineLearner
|
7
|
+
|
8
|
+
# ナイブベイズ学習器
|
9
|
+
class BayesLearner < Learner
|
10
|
+
|
11
|
+
# データを元に学習を行う
|
12
|
+
# @param datas [Array<DataSet>] トレーニングデータの配列
|
13
|
+
# @return [Array<Boolean>] 識別結果の配列
|
14
|
+
def learn(datas, ds = nil)
|
15
|
+
@training = datas
|
16
|
+
@ys = @training.map{|data| data.y }
|
17
|
+
@candidate_y = @ys.uniq.compact
|
18
|
+
end
|
19
|
+
|
20
|
+
# 識別を行う
|
21
|
+
# @param x 特徴空間
|
22
|
+
# @return [Fixnum] 識別結果
|
23
|
+
def classify(xs)
|
24
|
+
# 最も尤度の高い候補 y を探す
|
25
|
+
max_y, max_likelihood = 0, -Float::INFINITY
|
26
|
+
@candidate_y.each do |y|
|
27
|
+
likelihood = p_y_x(y, xs)
|
28
|
+
max_y, max_likelihood = y, likelihood if max_likelihood < likelihood
|
29
|
+
end
|
30
|
+
return max_y
|
31
|
+
end
|
32
|
+
|
33
|
+
private
|
34
|
+
|
35
|
+
# P(Y | X) を計算
|
36
|
+
def p_y_x(y, xs)
|
37
|
+
p = Math.log(p_y(y))
|
38
|
+
xs.each_with_index do |xi, i|
|
39
|
+
next if xi.nil?
|
40
|
+
p += Math.log(p_xi_y(xi, i, y))
|
41
|
+
end
|
42
|
+
return p
|
43
|
+
end
|
44
|
+
|
45
|
+
# P(Xi | Y) を計算
|
46
|
+
def p_xi_y(x, i, y)
|
47
|
+
total = count = 0
|
48
|
+
@ys.each_with_index do |ysj, j|
|
49
|
+
next if ysj != y
|
50
|
+
total += 1
|
51
|
+
count += 1 if @training[j].x[i] == x
|
52
|
+
end
|
53
|
+
return (count + 1).to_f / (total + candidate_x(i).size + 1).to_f
|
54
|
+
end
|
55
|
+
|
56
|
+
# P(Y) を計算
|
57
|
+
def p_y(y)
|
58
|
+
count = @ys.count(y)
|
59
|
+
return (count + 1).to_f / (@ys.size + @candidate_y.size + 1).to_f
|
60
|
+
end
|
61
|
+
|
62
|
+
# xの候補を列挙
|
63
|
+
def candidate_x(field)
|
64
|
+
@candidate_x ||= []
|
65
|
+
@candidate_x[field] ||= @training.map{|data| data.x[field]}.uniq.compact
|
66
|
+
end
|
67
|
+
end
|
68
|
+
|
69
|
+
end
|
70
|
+
|
71
|
+
# vim: set et ts=2 sts=2 sw=2:
|
72
|
+
|
@@ -0,0 +1,24 @@
|
|
1
|
+
#coding utf-8
|
2
|
+
|
3
|
+
module MachineLearner
|
4
|
+
# 学習器にかけるデータ
|
5
|
+
# 特徴空間xと識別結果yを持つ
|
6
|
+
class DataSet
|
7
|
+
attr_reader :x, :y
|
8
|
+
|
9
|
+
# コンストラクタ
|
10
|
+
# @param x 特徴空間
|
11
|
+
# @param y 識別結果
|
12
|
+
def initialize(x, y)
|
13
|
+
@x, @y = x, y
|
14
|
+
end
|
15
|
+
|
16
|
+
# @return データセットを表現する文字列
|
17
|
+
def to_s
|
18
|
+
"(#{@x}=>#{@y})"
|
19
|
+
end
|
20
|
+
|
21
|
+
end
|
22
|
+
end
|
23
|
+
|
24
|
+
# vim: set et ts=2 sts=2 sw=2:
|
@@ -0,0 +1,44 @@
|
|
1
|
+
#coding utf-8
|
2
|
+
|
3
|
+
module MachineLearner
|
4
|
+
|
5
|
+
# 識別器クラス
|
6
|
+
class Classifier
|
7
|
+
|
8
|
+
# 識別を行う
|
9
|
+
# @param x 特徴空間
|
10
|
+
# @return [Fixnum] 識別結果
|
11
|
+
def classify(x)
|
12
|
+
raise NotImplementedError.new
|
13
|
+
end
|
14
|
+
|
15
|
+
# 性能の測定を行う
|
16
|
+
# @param data[DataSet] データ
|
17
|
+
# @return [Boolean] 識別できたらtrue, 外れたらfalse
|
18
|
+
def test(data)
|
19
|
+
classify(data.x) == data.y
|
20
|
+
end
|
21
|
+
|
22
|
+
# 性能の測定を行う
|
23
|
+
# @param datas[Array<DataSet>] テストデータセット
|
24
|
+
# @return [Float] 正解率(0以上1以下)
|
25
|
+
def evaluate(datas)
|
26
|
+
datas.count{|data| test(data)} / datas.size.to_f
|
27
|
+
end
|
28
|
+
end
|
29
|
+
|
30
|
+
# 学習器クラス
|
31
|
+
class Learner < Classifier
|
32
|
+
|
33
|
+
# トレーニングデータを元に学習を行う
|
34
|
+
# @param datas [Array<DataSet>] トレーニングデータセット
|
35
|
+
# @param ds [Array<Float>] 各データセットの重み
|
36
|
+
# @return [Array<Boolean>] 学習後の識別結果
|
37
|
+
def learn(datas, ds)
|
38
|
+
raise NotImplementedError.new
|
39
|
+
end
|
40
|
+
end
|
41
|
+
|
42
|
+
end
|
43
|
+
|
44
|
+
# vim: set et ts=2 sts=2 sw=2:
|
@@ -0,0 +1,27 @@
|
|
1
|
+
# coding: utf-8
|
2
|
+
lib = File.expand_path('../lib', __FILE__)
|
3
|
+
$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
|
4
|
+
require 'machine_learner/version'
|
5
|
+
|
6
|
+
Gem::Specification.new do |spec|
|
7
|
+
spec.name = "machine_learner"
|
8
|
+
spec.version = MachineLearner::VERSION
|
9
|
+
spec.authors = ["YutaTanaka"]
|
10
|
+
spec.email = ["yuta84q.ihcarok@gmail.com"]
|
11
|
+
spec.summary = %q{Library for machine learning.}
|
12
|
+
spec.description = "This is a library for machine learning.\n" +
|
13
|
+
"You can use AdaBoost and Naive Bayes easily.\n"
|
14
|
+
spec.homepage = "https://github.com/84q/machine_learning"
|
15
|
+
spec.license = "MIT"
|
16
|
+
|
17
|
+
spec.files = `git ls-files -z`.split("\x0")
|
18
|
+
spec.executables = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
|
19
|
+
spec.test_files = spec.files.grep(%r{^(test|spec|features)/})
|
20
|
+
spec.require_paths = ["lib"]
|
21
|
+
|
22
|
+
spec.add_development_dependency "bundler", "~> 1.6"
|
23
|
+
spec.add_development_dependency "rake", "~> 10.0"
|
24
|
+
spec.add_development_dependency 'rspec', '~> 0'
|
25
|
+
end
|
26
|
+
|
27
|
+
# vim: set et ts=2 sts=2 sw=2:
|
data/spec/spec_helper.rb
ADDED
metadata
ADDED
@@ -0,0 +1,107 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: machine_learner
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.0.1
|
5
|
+
platform: ruby
|
6
|
+
authors:
|
7
|
+
- YutaTanaka
|
8
|
+
autorequire:
|
9
|
+
bindir: bin
|
10
|
+
cert_chain: []
|
11
|
+
date: 2014-08-10 00:00:00.000000000 Z
|
12
|
+
dependencies:
|
13
|
+
- !ruby/object:Gem::Dependency
|
14
|
+
name: bundler
|
15
|
+
requirement: !ruby/object:Gem::Requirement
|
16
|
+
requirements:
|
17
|
+
- - "~>"
|
18
|
+
- !ruby/object:Gem::Version
|
19
|
+
version: '1.6'
|
20
|
+
type: :development
|
21
|
+
prerelease: false
|
22
|
+
version_requirements: !ruby/object:Gem::Requirement
|
23
|
+
requirements:
|
24
|
+
- - "~>"
|
25
|
+
- !ruby/object:Gem::Version
|
26
|
+
version: '1.6'
|
27
|
+
- !ruby/object:Gem::Dependency
|
28
|
+
name: rake
|
29
|
+
requirement: !ruby/object:Gem::Requirement
|
30
|
+
requirements:
|
31
|
+
- - "~>"
|
32
|
+
- !ruby/object:Gem::Version
|
33
|
+
version: '10.0'
|
34
|
+
type: :development
|
35
|
+
prerelease: false
|
36
|
+
version_requirements: !ruby/object:Gem::Requirement
|
37
|
+
requirements:
|
38
|
+
- - "~>"
|
39
|
+
- !ruby/object:Gem::Version
|
40
|
+
version: '10.0'
|
41
|
+
- !ruby/object:Gem::Dependency
|
42
|
+
name: rspec
|
43
|
+
requirement: !ruby/object:Gem::Requirement
|
44
|
+
requirements:
|
45
|
+
- - "~>"
|
46
|
+
- !ruby/object:Gem::Version
|
47
|
+
version: '0'
|
48
|
+
type: :development
|
49
|
+
prerelease: false
|
50
|
+
version_requirements: !ruby/object:Gem::Requirement
|
51
|
+
requirements:
|
52
|
+
- - "~>"
|
53
|
+
- !ruby/object:Gem::Version
|
54
|
+
version: '0'
|
55
|
+
description: |
|
56
|
+
This is a library for machine learning.
|
57
|
+
You can use AdaBoost and Naive Bayes easily.
|
58
|
+
email:
|
59
|
+
- yuta84q.ihcarok@gmail.com
|
60
|
+
executables: []
|
61
|
+
extensions: []
|
62
|
+
extra_rdoc_files: []
|
63
|
+
files:
|
64
|
+
- ".gitignore"
|
65
|
+
- ".rspec"
|
66
|
+
- ".travis.yml"
|
67
|
+
- Gemfile
|
68
|
+
- LICENSE.txt
|
69
|
+
- README.md
|
70
|
+
- Rakefile
|
71
|
+
- lib/machine_learner.rb
|
72
|
+
- lib/machine_learner/adaboost.rb
|
73
|
+
- lib/machine_learner/bayes.rb
|
74
|
+
- lib/machine_learner/dataset.rb
|
75
|
+
- lib/machine_learner/learner.rb
|
76
|
+
- lib/machine_learner/version.rb
|
77
|
+
- machine_learner.gemspec
|
78
|
+
- spec/machine_learner_spec.rb
|
79
|
+
- spec/spec_helper.rb
|
80
|
+
homepage: https://github.com/84q/machine_learning
|
81
|
+
licenses:
|
82
|
+
- MIT
|
83
|
+
metadata: {}
|
84
|
+
post_install_message:
|
85
|
+
rdoc_options: []
|
86
|
+
require_paths:
|
87
|
+
- lib
|
88
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
89
|
+
requirements:
|
90
|
+
- - ">="
|
91
|
+
- !ruby/object:Gem::Version
|
92
|
+
version: '0'
|
93
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
94
|
+
requirements:
|
95
|
+
- - ">="
|
96
|
+
- !ruby/object:Gem::Version
|
97
|
+
version: '0'
|
98
|
+
requirements: []
|
99
|
+
rubyforge_project:
|
100
|
+
rubygems_version: 2.2.2
|
101
|
+
signing_key:
|
102
|
+
specification_version: 4
|
103
|
+
summary: Library for machine learning.
|
104
|
+
test_files:
|
105
|
+
- spec/machine_learner_spec.rb
|
106
|
+
- spec/spec_helper.rb
|
107
|
+
has_rdoc:
|