see5 0.1.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/.gitignore +6 -0
- data/.rubocop.yml +11 -0
- data/Gemfile +8 -0
- data/README.md +33 -0
- data/Rakefile +12 -0
- data/lib/see5.rb +7 -0
- data/lib/see5/input_file_writer.rb +53 -0
- data/lib/see5/model.rb +22 -0
- data/lib/see5/rule.rb +30 -0
- data/lib/see5/rules_file_parser.rb +101 -0
- data/lib/see5/schema.rb +26 -0
- data/lib/see5/version.rb +5 -0
- data/see5.gemspec +38 -0
- metadata +100 -0
checksums.yaml
ADDED
@@ -0,0 +1,7 @@
|
|
1
|
+
---
|
2
|
+
SHA256:
|
3
|
+
metadata.gz: 6d2b54a9d4d3c9f3271505bf233a46772cacfd1702312a68dab174b197265f59
|
4
|
+
data.tar.gz: cb5a5d275b08cb8ce7ca00be58c81fd015902578d31c1d6e8a7d1b5d2936225a
|
5
|
+
SHA512:
|
6
|
+
metadata.gz: 903fb8b64822a99a15c14383fc57932af364bb1c355046b4b022b9588c2d7359c5bdbcf5221f4282ba9b1f9a4200f746d390d962367bcad8a39ecd61e2671839
|
7
|
+
data.tar.gz: 98594a2c2be18ede5633ee7ba66894d2ce2b2eddead9d10eb4531728c349b8b9db95e439f7ca1407d9c92a6abed363549127a7b731a8f5c53af984096d0fe1d5
|
data/.gitignore
ADDED
data/.rubocop.yml
ADDED
data/Gemfile
ADDED
data/README.md
ADDED
@@ -0,0 +1,33 @@
|
|
1
|
+
# Ruby::See5
|
2
|
+
|
3
|
+
A Ruby frontend for the See5/C5.0 family of classifiers and modellers. Builds models from Ruby objects.
|
4
|
+
|
5
|
+
## Installation
|
6
|
+
|
7
|
+
Add this line to your application's Gemfile:
|
8
|
+
|
9
|
+
```ruby
|
10
|
+
gem 'ruby-see5'
|
11
|
+
```
|
12
|
+
|
13
|
+
And then execute:
|
14
|
+
|
15
|
+
$ bundle
|
16
|
+
|
17
|
+
Or install it yourself as:
|
18
|
+
|
19
|
+
$ gem install ruby-see5
|
20
|
+
|
21
|
+
## Usage
|
22
|
+
|
23
|
+
TODO: Write usage instructions here
|
24
|
+
|
25
|
+
## Development
|
26
|
+
|
27
|
+
After checking out the repo, run `bin/setup` to install dependencies. Then, run `rake test` to run the tests. You can also run `bin/console` for an interactive prompt that will allow you to experiment.
|
28
|
+
|
29
|
+
To install this gem onto your local machine, run `bundle exec rake install`. To release a new version, update the version number in `version.rb`, and then run `bundle exec rake release`, which will create a git tag for the version, push git commits and tags, and push the `.gem` file to [rubygems.org](https://rubygems.org).
|
30
|
+
|
31
|
+
## Contributing
|
32
|
+
|
33
|
+
Bug reports and pull requests are welcome on GitHub at https://github.com/elebow/ruby-see5.
|
data/Rakefile
ADDED
data/lib/see5.rb
ADDED
@@ -0,0 +1,53 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
require_relative "schema"
|
4
|
+
|
5
|
+
module See5
|
6
|
+
# Writes names and data files suitable for See5, Cubist, or GritBot.
|
7
|
+
class InputFileWriter
|
8
|
+
def self.write_files(data:, schema: nil, names_io: nil, data_io: nil)
|
9
|
+
new(data: data, schema: schema, names_io: names_io, data_io: data_io)
|
10
|
+
.write_files
|
11
|
+
end
|
12
|
+
|
13
|
+
def initialize(data:, schema: nil, names_io: nil, data_io: nil)
|
14
|
+
@data = data
|
15
|
+
@schema = schema # TODO: automatic schema from data objects' attributes
|
16
|
+
@names_io = names_io
|
17
|
+
@data_io = data_io
|
18
|
+
end
|
19
|
+
|
20
|
+
def write_files
|
21
|
+
write_names_file
|
22
|
+
write_data_file
|
23
|
+
end
|
24
|
+
|
25
|
+
def write_names_file
|
26
|
+
names_io.write(@schema.names_file_contents)
|
27
|
+
end
|
28
|
+
|
29
|
+
def write_data_file
|
30
|
+
# TODO: missing or N/A
|
31
|
+
@data.each do |record|
|
32
|
+
data_io.write(row(record))
|
33
|
+
data_io.write("\n")
|
34
|
+
end
|
35
|
+
end
|
36
|
+
|
37
|
+
private
|
38
|
+
|
39
|
+
def row(record)
|
40
|
+
@schema.attributes.map do |attr, _vals|
|
41
|
+
record.send(attr)
|
42
|
+
end.join(",")
|
43
|
+
end
|
44
|
+
|
45
|
+
def names_io
|
46
|
+
@names_io ||= File.open("/tmp/ruby-see5.names", "w")
|
47
|
+
end
|
48
|
+
|
49
|
+
def data_io
|
50
|
+
@data_io ||= File.open("/tmp/ruby-see5.data", "w")
|
51
|
+
end
|
52
|
+
end
|
53
|
+
end
|
data/lib/see5/model.rb
ADDED
@@ -0,0 +1,22 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
module See5
|
4
|
+
class Model
|
5
|
+
attr_reader :rules
|
6
|
+
|
7
|
+
def initialize(default_classification:, rules:)
|
8
|
+
@default_classification = default_classification
|
9
|
+
@rules = rules
|
10
|
+
end
|
11
|
+
|
12
|
+
def classify(data)
|
13
|
+
# See5 orders rules by confidence within each class (TODO verify),
|
14
|
+
# so the first matching rule is the one with the highest confidence.
|
15
|
+
first_matching_rule = rules.find { |rule| rule.match?(data) }
|
16
|
+
|
17
|
+
return first_matching_rule.classification unless first_matching_rule.nil?
|
18
|
+
|
19
|
+
@default_classification
|
20
|
+
end
|
21
|
+
end
|
22
|
+
end
|
data/lib/see5/rule.rb
ADDED
@@ -0,0 +1,30 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
module See5
|
4
|
+
class Rule
|
5
|
+
attr_reader :classification, :confidence, :rule_info, :conditions
|
6
|
+
|
7
|
+
def initialize(rule_info, conditions, class_info)
|
8
|
+
@rule_info = rule_info
|
9
|
+
@conditions = conditions
|
10
|
+
@classification = class_info[:classification]
|
11
|
+
@confidence = class_info[:confidence]
|
12
|
+
end
|
13
|
+
|
14
|
+
def match?(data)
|
15
|
+
conditions
|
16
|
+
.map { |attr, val| data[attr] == val }
|
17
|
+
.all? { |matched| matched == true }
|
18
|
+
end
|
19
|
+
|
20
|
+
def to_s
|
21
|
+
[
|
22
|
+
"See5::Rule",
|
23
|
+
"@classification=#{@classification}",
|
24
|
+
"@conditions=#{@conditions}"
|
25
|
+
]
|
26
|
+
.join(", ")
|
27
|
+
.yield_self { |s| "#<#{s}>" }
|
28
|
+
end
|
29
|
+
end
|
30
|
+
end
|
@@ -0,0 +1,101 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
require_relative "rule"
|
4
|
+
|
5
|
+
module See5
|
6
|
+
# Read See5 rules output and return an array of hashes representing the rules
|
7
|
+
# Note that this is the output normally sent to stdout, NOT the .rules file!
|
8
|
+
# The .rules file lacks some important information like confidence.
|
9
|
+
class RulesFileParser
|
10
|
+
def self.parse_file(fname)
|
11
|
+
new(fname).model
|
12
|
+
end
|
13
|
+
|
14
|
+
def initialize(fname)
|
15
|
+
@file = File.open(fname)
|
16
|
+
@rules = []
|
17
|
+
|
18
|
+
parse_file
|
19
|
+
end
|
20
|
+
|
21
|
+
def model
|
22
|
+
{
|
23
|
+
default_classification: @default_classification,
|
24
|
+
rules: @rules
|
25
|
+
}
|
26
|
+
end
|
27
|
+
|
28
|
+
def parse_file
|
29
|
+
discard_header
|
30
|
+
|
31
|
+
while (line = lines.next)
|
32
|
+
if line.start_with?("Rule ")
|
33
|
+
@rules << parse_rule(line)
|
34
|
+
elsif line.start_with?("Default class:")
|
35
|
+
@default_classification = line.split(":").last.strip
|
36
|
+
|
37
|
+
break
|
38
|
+
end
|
39
|
+
end
|
40
|
+
end
|
41
|
+
|
42
|
+
private
|
43
|
+
|
44
|
+
def lines
|
45
|
+
# TODO: lazy unnecessary given that rules are small?
|
46
|
+
@file.each_line.lazy
|
47
|
+
end
|
48
|
+
|
49
|
+
# Discard the file header and advance to the rules section
|
50
|
+
# TODO: save the data from the header, in case user wants it?
|
51
|
+
def discard_header
|
52
|
+
while (line = lines.next)
|
53
|
+
break if line == "Rules:\n"
|
54
|
+
end
|
55
|
+
# discard the final blank line
|
56
|
+
lines.next
|
57
|
+
end
|
58
|
+
|
59
|
+
def parse_rule(line)
|
60
|
+
rule_info = parse_rule_info_line(line)
|
61
|
+
conditions = []
|
62
|
+
|
63
|
+
while (line = lines.next.strip)
|
64
|
+
if line.start_with?("->")
|
65
|
+
class_info = parse_class_line(line)
|
66
|
+
|
67
|
+
break
|
68
|
+
end
|
69
|
+
|
70
|
+
conditions << parse_condition_line(line)
|
71
|
+
end
|
72
|
+
|
73
|
+
Rule.new(rule_info, conditions.to_h, class_info)
|
74
|
+
end
|
75
|
+
|
76
|
+
def parse_class_line(line)
|
77
|
+
matches = line.match(/class ([\w]+) \[(.+)\]/)
|
78
|
+
|
79
|
+
{
|
80
|
+
classification: matches[1],
|
81
|
+
confidence: matches[2].to_f
|
82
|
+
}
|
83
|
+
end
|
84
|
+
|
85
|
+
def parse_rule_info_line(line)
|
86
|
+
matches = line.match(%r{Rule \d+: \((\d+)(?:/)?([^,]*), lift (.+)\)})
|
87
|
+
|
88
|
+
{
|
89
|
+
cases_covered: matches[1].to_i,
|
90
|
+
cases_not_covered: matches[2]&.to_i,
|
91
|
+
lift: matches[3].to_f
|
92
|
+
}
|
93
|
+
end
|
94
|
+
|
95
|
+
def parse_condition_line(line)
|
96
|
+
(attr, val) = line.split("=").map(&:strip)
|
97
|
+
|
98
|
+
[attr.to_sym, val]
|
99
|
+
end
|
100
|
+
end
|
101
|
+
end
|
data/lib/see5/schema.rb
ADDED
@@ -0,0 +1,26 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
module See5
|
4
|
+
class Schema
|
5
|
+
attr_reader :class_attribute, :attributes
|
6
|
+
|
7
|
+
def initialize(classes:, attributes:, class_attribute: nil)
|
8
|
+
@classes = classes
|
9
|
+
@attributes = attributes
|
10
|
+
@class_attribute = class_attribute&.to_sym || :class_attribute
|
11
|
+
|
12
|
+
# if the class attribute doesn't exist in the attributes, add it
|
13
|
+
unless @attributes.key?(@class_attribute)
|
14
|
+
@attributes[@class_attribute] = @classes
|
15
|
+
end
|
16
|
+
end
|
17
|
+
|
18
|
+
def names_file_contents
|
19
|
+
class_attribute.to_s + # TODO: continuous class attribute
|
20
|
+
"\n" +
|
21
|
+
attributes.map do |attr, vals|
|
22
|
+
"#{attr}: #{vals.join(',')}"
|
23
|
+
end.join("\n")
|
24
|
+
end
|
25
|
+
end
|
26
|
+
end
|
data/lib/see5/version.rb
ADDED
data/see5.gemspec
ADDED
@@ -0,0 +1,38 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
lib = File.expand_path("lib", __dir__)
|
4
|
+
$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
|
5
|
+
require "see5/version"
|
6
|
+
|
7
|
+
Gem::Specification.new do |spec|
|
8
|
+
spec.name = "see5"
|
9
|
+
spec.version = See5::VERSION
|
10
|
+
spec.authors = ["Eddie Lebow"]
|
11
|
+
spec.email = ["elebow@users.noreply.github.com"]
|
12
|
+
|
13
|
+
spec.summary = "A Ruby frontend for the See5/C5.0 family of classifiers and modellers."
|
14
|
+
#spec.description = "TODO: Write a longer description or delete this line."
|
15
|
+
spec.homepage = "https://github.com/elebow/ruby-see5"
|
16
|
+
|
17
|
+
if spec.respond_to?(:metadata)
|
18
|
+
spec.metadata["homepage_uri"] = spec.homepage
|
19
|
+
spec.metadata["source_code_uri"] = "https://github.com/elebow/ruby-see5"
|
20
|
+
spec.metadata["changelog_uri"] = "https://github.com/elebow/ruby-see5/blob/master/CHANGELOG.md"
|
21
|
+
else
|
22
|
+
raise "RubyGems 2.0 or newer is required to protect against " \
|
23
|
+
"public gem pushes."
|
24
|
+
end
|
25
|
+
|
26
|
+
# Specify which files should be added to the gem when it is released.
|
27
|
+
# The `git ls-files -z` loads the files in the RubyGem that have been added into git.
|
28
|
+
spec.files = Dir.chdir(File.expand_path(__dir__)) do
|
29
|
+
`git ls-files -z`.split("\x0").reject { |f| f.match(%r{^(test|spec|features)/}) }
|
30
|
+
end
|
31
|
+
spec.bindir = "exe"
|
32
|
+
spec.executables = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
|
33
|
+
spec.require_paths = ["lib"]
|
34
|
+
|
35
|
+
spec.add_development_dependency "bundler", "~> 1.17"
|
36
|
+
spec.add_development_dependency "minitest", "~> 5.0"
|
37
|
+
spec.add_development_dependency "rake", "~> 10.0"
|
38
|
+
end
|
metadata
ADDED
@@ -0,0 +1,100 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: see5
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.1.0
|
5
|
+
platform: ruby
|
6
|
+
authors:
|
7
|
+
- Eddie Lebow
|
8
|
+
autorequire:
|
9
|
+
bindir: exe
|
10
|
+
cert_chain: []
|
11
|
+
date: 2020-03-03 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: '1.17'
|
20
|
+
type: :development
|
21
|
+
prerelease: false
|
22
|
+
version_requirements: !ruby/object:Gem::Requirement
|
23
|
+
requirements:
|
24
|
+
- - "~>"
|
25
|
+
- !ruby/object:Gem::Version
|
26
|
+
version: '1.17'
|
27
|
+
- !ruby/object:Gem::Dependency
|
28
|
+
name: minitest
|
29
|
+
requirement: !ruby/object:Gem::Requirement
|
30
|
+
requirements:
|
31
|
+
- - "~>"
|
32
|
+
- !ruby/object:Gem::Version
|
33
|
+
version: '5.0'
|
34
|
+
type: :development
|
35
|
+
prerelease: false
|
36
|
+
version_requirements: !ruby/object:Gem::Requirement
|
37
|
+
requirements:
|
38
|
+
- - "~>"
|
39
|
+
- !ruby/object:Gem::Version
|
40
|
+
version: '5.0'
|
41
|
+
- !ruby/object:Gem::Dependency
|
42
|
+
name: rake
|
43
|
+
requirement: !ruby/object:Gem::Requirement
|
44
|
+
requirements:
|
45
|
+
- - "~>"
|
46
|
+
- !ruby/object:Gem::Version
|
47
|
+
version: '10.0'
|
48
|
+
type: :development
|
49
|
+
prerelease: false
|
50
|
+
version_requirements: !ruby/object:Gem::Requirement
|
51
|
+
requirements:
|
52
|
+
- - "~>"
|
53
|
+
- !ruby/object:Gem::Version
|
54
|
+
version: '10.0'
|
55
|
+
description:
|
56
|
+
email:
|
57
|
+
- elebow@users.noreply.github.com
|
58
|
+
executables: []
|
59
|
+
extensions: []
|
60
|
+
extra_rdoc_files: []
|
61
|
+
files:
|
62
|
+
- ".gitignore"
|
63
|
+
- ".rubocop.yml"
|
64
|
+
- Gemfile
|
65
|
+
- README.md
|
66
|
+
- Rakefile
|
67
|
+
- lib/see5.rb
|
68
|
+
- lib/see5/input_file_writer.rb
|
69
|
+
- lib/see5/model.rb
|
70
|
+
- lib/see5/rule.rb
|
71
|
+
- lib/see5/rules_file_parser.rb
|
72
|
+
- lib/see5/schema.rb
|
73
|
+
- lib/see5/version.rb
|
74
|
+
- see5.gemspec
|
75
|
+
homepage: https://github.com/elebow/ruby-see5
|
76
|
+
licenses: []
|
77
|
+
metadata:
|
78
|
+
homepage_uri: https://github.com/elebow/ruby-see5
|
79
|
+
source_code_uri: https://github.com/elebow/ruby-see5
|
80
|
+
changelog_uri: https://github.com/elebow/ruby-see5/blob/master/CHANGELOG.md
|
81
|
+
post_install_message:
|
82
|
+
rdoc_options: []
|
83
|
+
require_paths:
|
84
|
+
- lib
|
85
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
86
|
+
requirements:
|
87
|
+
- - ">="
|
88
|
+
- !ruby/object:Gem::Version
|
89
|
+
version: '0'
|
90
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
91
|
+
requirements:
|
92
|
+
- - ">="
|
93
|
+
- !ruby/object:Gem::Version
|
94
|
+
version: '0'
|
95
|
+
requirements: []
|
96
|
+
rubygems_version: 3.0.3
|
97
|
+
signing_key:
|
98
|
+
specification_version: 4
|
99
|
+
summary: A Ruby frontend for the See5/C5.0 family of classifiers and modellers.
|
100
|
+
test_files: []
|