natto2classifier 0.2.0 → 0.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 51ad6e9876016e7a4202e0c5e97e2c1e67d208c4
-  data.tar.gz: 9684166a82e326f30b5b21a63658cd708540baad
+  metadata.gz: f8ef6d3851edf130a6a2b7ca3a6ba7c305cf39be
+  data.tar.gz: a73d1ab45544793c3a27a704233b29a8a1898b04
 SHA512:
-  metadata.gz: b2811c6898c36d2e5d37267023a2f6121e38fa953e3c3d7191225b8d704c0c1d83c651f20353a09530739c46a8d0239f1d638153ffc221cd21509f6785323e44
-  data.tar.gz: a9e57167f00b04feac3653cbe2a8362ac0cfc0946dad2e34912764173a03ca35c8db77404feb285cde009e8380a26033b8e501ab53b9a6e873d7cfc63be52e77
+  metadata.gz: b1b3dc2ca4d798aaffbcd1be95bd03d62a822550520b029b734a13265161582ca82c25d1720e6d05ebcab6c1c126e62e98b64ac345403e52c4b544914a0edc4a
+  data.tar.gz: 89fcf93bc0d47633659bfffa12b375d34f25cad5a8a774a7746b3c9e7dcbc99cba8eb6a72009b29d6660c41deb029eea2609c5c79fe8aae224c19b5ab1c0a2f4

data/.circleci/config.yml

@@ -12,13 +12,3 @@ workflows:
   test:
     jobs:
       - test
-  scheduled-workflow:
-    triggers:
-      - schedule:
-          cron: "0 * * * *"
-          filters:
-            branches:
-              only:
-                - master
-    jobs:
-      - test

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    natto2classifier (0.2.0)
+    natto2classifier (0.3.0)
       classifier-reborn
       natto
       rb-gsl

data/README.md

@@ -39,6 +39,17 @@ lsi.classify '納豆はいつも朝食べている' #=> '朝食'
 lsi.find_related '納豆はいつも朝食べている' #=> ['今日 キョウ の ノ 朝食 チョウショク は ハ 納豆 ナットウ だ ダ', '今日 キョウ の ノ 夕食 ユウショク は ハ 湯豆腐 ユドウフ だ ダ']
 ```
 
+### validate methods
+
+```
+sample_data = CSV.read('./data/train.csv')
+bayes = Natto2classifier::Bayes.new '朝食', '夕食'
+cross_validate(bayes, sample_data) #=> report...
+
+test_data, training_data = sample_data.partition.with_index { |_, i| (i % 2).zero? }
+validate(bayes, training_data, test_data) #=> {"夕食"=>{"夕食"=>3, "朝食"=>0}, "朝食"=>{"夕食"=>...}}
+```
+
 ## Development
 
 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.

data/lib/natto2classifier.rb

@@ -3,3 +3,4 @@ require 'natto2classifier/natto'
 require 'classifier-reborn'
 require 'natto2classifier/bayes'
 require 'natto2classifier/lsi'
+require 'natto2classifier/validator'

data/lib/natto2classifier/bayes.rb

@@ -0,0 +1,17 @@
+
+module Natto2classifier
+  # It is a library that classifies Japanese language.
+  class Bayes < ClassifierReborn::Bayes
+    alias_method :__train__, :train
+    alias_method :__classify__, :classify
+    private :__train__, :__classify__
+
+    def train(category, word)
+      __train__ category, Natto2classifier::Natto.parse(word).join(' ')
+    end
+
+    def classify(word)
+      __classify__ Natto2classifier::Natto.parse(word).join(' ')
+    end
+  end
+end

data/lib/natto2classifier/lsi.rb

@@ -0,0 +1,22 @@
+
+module Natto2classifier
+  # It is a library that classifies Japanese language.
+  class LSI < ClassifierReborn::LSI
+    alias_method :__add_item__, :add_item
+    alias_method :__classify__, :classify
+    alias_method :__find_related__, :find_related
+    private :__add_item__, :__classify__, :__find_related__
+
+    def add_item(word, category)
+      __add_item__ Natto2classifier::Natto.parse(word).join(' '), category
+    end
+
+    def classify(word)
+      __classify__ Natto2classifier::Natto.parse(word).join(' ')
+    end
+
+    def find_related(word)
+      __find_related__ Natto2classifier::Natto.parse(word).join(' ')
+    end
+  end
+end

data/lib/natto2classifier/version.rb

@@ -1,3 +1,3 @@
 module Natto2classifier
-  VERSION = "0.2.0"
+  VERSION = "0.3.0"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: natto2classifier
 version: !ruby/object:Gem::Version
-  version: 0.2.0
+  version: 0.3.0
 platform: ruby
 authors:
 - kanayannet
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2018-04-27 00:00:00.000000000 Z
+date: 2018-05-01 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -129,6 +129,8 @@ files:
 - bin/setup
 - data/train.csv
 - lib/natto2classifier.rb
+- lib/natto2classifier/bayes.rb
+- lib/natto2classifier/lsi.rb
 - lib/natto2classifier/natto.rb
 - lib/natto2classifier/version.rb
 - natto2classifier.gemspec