classifier-reborn 2.0.1 → 2.0.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 95e6a82c745772e5180c3765d2ae24995f02e2db
-  data.tar.gz: e1c6fb9a5f8a4651dac1d3dc5fb327cb8e586e44
+  metadata.gz: 1219c6a6bbe3f8c2a0820d953da2ef4c85bb5e52
+  data.tar.gz: 66575a30d03763da10f35ee669942bc111838e6e
 SHA512:
-  metadata.gz: 8d276020491d285ef656d85de17013c3fbcbb3161116ed139798602107dcd2229b85da121d0e0e3d00d6a5be095ae110a24e864d99f8efc2ab62fb4aa26b6b07
-  data.tar.gz: 735ebae19d3d920efad2e58dadb8580ae04d512528f319a63d1a746f539d5166405d2a8de5efc487f6c4832700652fbc222d71d672b31826ddaeaefd2687d365
+  metadata.gz: 7d8012f760b929a521a03aa43914def24bd2ddcf688c6816c3699d549d584cccf91e5b8499f1c988b8eceb8d4cf767ccad2585cb2cacabcd9092ce040d64b02e
+  data.tar.gz: 6e6d7e9a50f9ab05d8548b7032c208729c852fa32452a2c411f7ebd50b9ccf53750f0155907a4cfefd697ec12a0dd24fc25e8fd6d9131e7318fbf04350f6b93e

data/README.markdown

@@ -29,7 +29,7 @@ This should install automatically with RubyGems.
 If you would like to speed up LSI classification by at least 10x, please install the following libraries:
 
 * [GNU GSL](http://www.gnu.org/software/gsl)
-* [rb-gsl](http://rb-gsl.rubyforge.org)
+* [rb-gsl](https://rubygems.org/gems/rb-gsl)
 
 Notice that LSI will work without these libraries, but as soon as they are installed, Classifier will make use of them. No configuration changes are needed, we like to keep things ridiculously easy for you.
 
@@ -40,7 +40,7 @@ A Bayesian classifier by Lucas Carlson. Bayesian Classifiers are accurate, fast,
 ### Usage
 
 ```ruby
-require 'classifier'
+require 'classifier-reborn'
 b = ClassifierReborn::Bayes.new 'Interesting', 'Uninteresting'
 b.train_interesting "here are some good words. I hope you love them"
 b.train_uninteresting "here are some bad words, I hate you"
@@ -74,7 +74,7 @@ theoretically simulates human learning.
 ### Usage
 
 ```ruby
-require 'classifier'
+require 'classifier-reborn'
 lsi = ClassifierReborn::LSI.new
 strings = [ ["This text deals with dogs. Dogs.", :dog],
             ["This text involves dogs too. Dogs! ", :dog],

data/lib/classifier-reborn.rb

@@ -25,6 +25,6 @@
 # License::   LGPL
 
 require 'rubygems'
-require_relative 'classifier-reborn/extensions/string'
+require_relative 'classifier-reborn/category_namer'
 require_relative 'classifier-reborn/bayes'
 require_relative 'classifier-reborn/lsi'

data/lib/classifier-reborn/bayes.rb

@@ -2,7 +2,7 @@
 # Copyright:: Copyright (c) 2005 Lucas Carlson
 # License::   LGPL
 
-require_relative 'extensions/string'
+require_relative 'category_namer'
 
 module ClassifierReborn
   class Bayes
@@ -11,9 +11,10 @@ module ClassifierReborn
     #      b = ClassifierReborn::Bayes.new 'Interesting', 'Uninteresting', 'Spam'
     def initialize(*categories)
       @categories = Hash.new
-      categories.each { |category| @categories[category.prepare_category_name] = Hash.new }
+      categories.each { |category| @categories[CategoryNamer.prepare_name(category)] = Hash.new }
       @total_words = 0
       @category_counts = Hash.new(0)
+      @category_word_count = Hash.new
     end
 
     # Provides a general training method for all categories specified in Bayes#new
@@ -23,11 +24,13 @@ module ClassifierReborn
     #     b.train "that", "That text"
     #     b.train "The other", "The other text"
     def train(category, text)
-      category = category.prepare_category_name
-                  @category_counts[category] += 1
+      category = CategoryNamer.prepare_name(category)
+      @category_word_count[category] ||= 0
+      @category_counts[category] += 1
       Hasher.word_hash(text).each do |word, count|
         @categories[category][word]     ||=     0
         @categories[category][word]      +=     count
+        @category_word_count[category]   += count
         @total_words += count
       end
     end
@@ -40,17 +43,22 @@ module ClassifierReborn
     #     b.train :this, "This text"
     #     b.untrain :this, "This text"
     def untrain(category, text)
-      category = category.prepare_category_name
+      category = CategoryNamer.prepare_name(category)
+      @category_word_count[category] ||= 0
       @category_counts[category] -= 1
       Hasher.word_hash(text).each do |word, count|
         if @total_words >= 0
           orig = @categories[category][word] || 0
-          @categories[category][word] ||= 0
-          @categories[category][word] -= count
+          @categories[category][word]     ||=     0
+          @categories[category][word]      -=     count
           if @categories[category][word] <= 0
             @categories[category].delete(word)
             count = orig
           end
+
+          if @category_word_count[category] >= count
+            @category_word_count[category] -= count
+          end
           @total_words -= count
         end
       end
@@ -62,13 +70,14 @@ module ClassifierReborn
     # The largest of these scores (the one closest to 0) is the one picked out by #classify
     def classifications(text)
       score = Hash.new
-      training_count = @category_counts.values.inject { |x,y| x+y }.to_f
+      word_hash = Hasher.word_hash(text)
+      training_count = @category_counts.values.reduce(:+).to_f
       @categories.each do |category, category_words|
         score[category.to_s] = 0
-        total = category_words.values.inject(0) {|sum, element| sum+element}
-        Hasher.word_hash(text).each do |word, count|
+        total = (@category_word_count[category] || 1).to_f
+        word_hash.each do |word, count|
           s = category_words.has_key?(word) ? category_words[word] : 0.1
-          score[category.to_s] += Math.log(s/total.to_f)
+          score[category.to_s] += Math.log(s/total)
         end
         # now add prior probability for the category
         s = @category_counts.has_key?(category) ? @category_counts[category] : 0.1
@@ -93,7 +102,8 @@ module ClassifierReborn
     #     b.untrain_that "That text"
     #     b.train_the_other "The other text"
     def method_missing(name, *args)
-      category = name.to_s.gsub(/(un)?train_([\w]+)/, '\2').prepare_category_name
+      cleaned_name = name.to_s.gsub(/(un)?train_([\w]+)/, '\2')
+      category = CategoryNamer.prepare_name(cleaned_name)
       if @categories.has_key? category
         args.each { |text| eval("#{$1}train(category, text)") }
       elsif name.to_s =~ /(un)?train_([\w]+)/
@@ -120,7 +130,7 @@ module ClassifierReborn
     # more criteria than the trained selective categories. In short,
     # try to initialize your categories at initialization.
     def add_category(category)
-      @categories[category.prepare_category_name] = Hash.new
+      @categories[CategoryNamer.prepare_name(category)] = Hash.new
     end
 
     alias append_category add_category

data/lib/classifier-reborn/{extensions/string.rb → category_namer.rb}

@@ -5,6 +5,11 @@
 require 'fast_stemmer'
 require 'classifier-reborn/extensions/hasher'
 
-class Object
-  def prepare_category_name; to_s.gsub("_"," ").capitalize.intern end
+module ClassifierReborn
+  module CategoryNamer
+    extend self
+    def prepare_name(name) 
+      name.to_s.gsub("_"," ").capitalize.intern 
+    end
+  end
 end

data/lib/classifier-reborn/extensions/vector.rb

@@ -5,18 +5,6 @@
 
 require 'matrix'
 
-class Array
-  def sum(identity = 0, &block)
-    return identity unless size > 0
-
-    if block_given?
-      map(&block).sum
-    else
-      reduce(:+) || 0
-    end
-  end
-end
-
 class Vector
   def magnitude
     sumsqs = 0.0

data/lib/classifier-reborn/lsi/content_node.rb

@@ -44,7 +44,7 @@ module ClassifierReborn
       end
 
       # Perform the scaling transform and force floating point arithmetic
-      total_words = vec.sum.to_f
+      total_words = vec.reduce(0, :+).to_f
       total_unique_words = 0
 
       if $GSL

data/lib/classifier-reborn/version.rb

@@ -1,3 +1,3 @@
 module ClassifierReborn
-  VERSION = '2.0.1'
+  VERSION = '2.0.2'
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: classifier-reborn
 version: !ruby/object:Gem::Version
-  version: 2.0.1
+  version: 2.0.2
 platform: ruby
 authors:
 - Lucas Carlson
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-08-14 00:00:00.000000000 Z
+date: 2014-11-09 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: fast-stemmer
@@ -71,8 +71,8 @@ files:
 - bin/summarize.rb
 - lib/classifier-reborn.rb
 - lib/classifier-reborn/bayes.rb
+- lib/classifier-reborn/category_namer.rb
 - lib/classifier-reborn/extensions/hasher.rb
-- lib/classifier-reborn/extensions/string.rb
 - lib/classifier-reborn/extensions/vector.rb
 - lib/classifier-reborn/extensions/vector_serialize.rb
 - lib/classifier-reborn/lsi.rb