luisparravicini-classifier 1.3.8 → 1.3.9

data/{README → README.rdoc}

@@ -9,8 +9,9 @@ rb-gsl:: http://rb-gsl.rubyforge.org
 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.
 
 == Changes in this branch
-I made this branch to fix a TypeError on untrain (classifier-1.3.1), then merge KirinDave branch[http://github.com/KirinDave/classifier/] to fix an incompatibility problem with ActiveSupport (with Array#sum), then francois[http://github.com/francois/classifier/] branch for jeweler and all the changes yuri[http://github.com/yury/classifier/] made on his branch (specially the use of ruby-stemmer, which I needed).
+I made this branch to fix a TypeError on untrain (classifier-1.3.1), then francois[http://github.com/francois/classifier/] branch for jeweler and all the changes yuri[http://github.com/yury/classifier/] made on his branch (specially the use of ruby-stemmer, and the incompatibility fix on Array#sum, which I needed).
 After that I added support for loading the stopwords of certain language from a file (before the list was embedded on the source code) and a stopword list for Spanish.
+This branch only works with Ruby 1.9
 
 == Bayes
 A Bayesian classifier by Lucas Carlson. Bayesian Classifiers are accurate, fast, and have modest memory requirements.
@@ -33,8 +34,17 @@ A Bayesian classifier by Lucas Carlson. Bayesian Classifiers are accurate, fast,
 
 Using Madeleine, your application can persist the learned data over time.
 
-You can also specify language and encoding for internal stemmer
-	  b = Classifier::Bayes.new :categories => ['Interesting', 'Uninteresting'], :language => 'ro', :encoding => 'ISO_8859_2'
+=== Stemmer configuration
+
+You can specify language and encoding for internal stemmer:
+
+	  b = Classifier::Bayes.new :categories => ['Interesting', 'Uninteresting'],
+      :language => 'ro', :encoding => 'ISO_8859_2'
+
+The default values are 'en' for language and 'UTF-8' for the encoding.
+
+Each language uses a word list to exclude certain words (stopwords). classifier comes with three included stopword lists, for English, Russian and Spanish.
+The English list is the list that comes with the original gem (don't know where it was taken from) and the Russian and Spanish are from snowball[http://snowball.tartarus.org/algorithms/].
 
 === Bayesian Classification
 
@@ -71,7 +81,8 @@ theoretically simulates human learning.
 Please see the Classifier::LSI documentation for more information. It is possible to index, search and classify
 with more than just simple strings. 
 
-You can also specify language and encoding for internal stemmer
+The configuration for the stemmer is the same used for Bayes:
+
 	  lsi = Classifier::LSI.new :language => 'ro', :encoding => 'ISO_8859_2'
 
 

data/Rakefile

@@ -8,8 +8,8 @@ begin
     s.summary     = "A general classifier module to allow Bayesian and other types of classifications."
     s.description = "Bayesian classifier and others."
     s.homepage    = "http://github.com/luisparravicini/classifier"
-    s.author      = "Yury Korolev"
-    s.email       = "yury.korolev@gmail.com"
+    s.author      = "Luis Parravicini"
+    s.email       = "lparravi@gmail.com"
 
     s.add_dependency "activesupport", ">= 2.2.2"
     s.add_dependency "ruby-stemmer", ">= 0.5.1"

data/VERSION.yml

@@ -1,5 +1,5 @@
 --- 
 :major: 1
 :minor: 3
-:patch: 8
+:patch: 9
 :build: 

data/lib/classifier.rb

@@ -24,7 +24,6 @@
 # Copyright:: Copyright (c) 2005 Lucas Carlson
 # License::   LGPL
 
-require 'rubygems'
 require 'activesupport'
 require 'lingua/stemmer'
 require 'classifier/base'

data/lib/classifier/bayes.rb

@@ -131,6 +131,15 @@ class Bayes < Classifier::Base
 	end
 	
 	alias append_category add_category
+
+  def marshal_dump
+    [@categories, @total_words, @options ]
+  end
+
+  def marshal_load(data)
+    @categories, @total_words, @options = data
+  end
+
 end
 
 end

data/lib/classifier/extensions/vector.rb

@@ -6,15 +6,6 @@
 require 'matrix'
 require 'mathn'
 
-# Conflicts with ActiveSupport
-unless Array.new.respond_to?(:sum)
-  class Array
-    def sum
-      inject(0) { |sum,term| sum += term  }.to_f
-    end
-  end
-end
-
 class Vector
   def magnitude
     sumsqs = 0.0

data/lib/classifier/lsi.rb

@@ -295,6 +295,17 @@ module Classifier
       return top_n.collect { |x| @word_list.word_for_index(arr.index(x))}
     end
 
+    def marshal_dump
+      [ @auto_rebuild, @word_list, @items, @version, @built_at_version,
+        @options,
+      ]
+    end
+
+    def marshal_load(data)
+      @auto_rebuild, @word_list, @items, @version, @built_at_version,
+        @options = data
+    end
+
     private
     def build_reduced_matrix( matrix, cutoff=0.75 )
       # TODO: Check that M>=N on these dimensions! Transpose helps assure this

data/luisparravicini-classifier.gemspec

@@ -5,21 +5,21 @@
 
 Gem::Specification.new do |s|
   s.name = %q{luisparravicini-classifier}
-  s.version = "1.3.8"
+  s.version = "1.3.9"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
-  s.authors = ["Yury Korolev"]
-  s.date = %q{2009-12-24}
+  s.authors = ["Luis Parravicini"]
+  s.date = %q{2009-12-26}
   s.description = %q{Bayesian classifier and others.}
-  s.email = %q{yury.korolev@gmail.com}
+  s.email = %q{lparravi@gmail.com}
   s.extra_rdoc_files = [
     "LICENSE",
-     "README"
+     "README.rdoc"
   ]
   s.files = [
     "LICENSE",
      "Manifest",
-     "README",
+     "README.rdoc",
      "Rakefile",
      "VERSION.yml",
      "lib/classifier.rb",

data/test/bayes/bayesian_test.rb

@@ -55,4 +55,14 @@ class BayesianTest < Test::Unit::TestCase
 	  assert_equal c.classifications("ХОРОШО"), c.classifications("хорошо")
 	  assert_equal c.classifications("плОХО"), c.classifications("плохо")
   end
+
+  def test_serialize
+    txt = "this can be serialized"
+    b = Classifier::Bayes.new(:categories => ['Interesting', 'Uninteresting'])
+    b.train_interesting(txt)
+    b.train_uninteresting("really uninteresting")
+
+    b2 = Marshal::load(Marshal::dump(b))
+    assert_equal b.classify(txt), b2.classify(txt)
+  end
 end

data/test/stopwords_test.rb

@@ -1,5 +1,6 @@
 # coding:utf-8
 require File.dirname(__FILE__) + '/test_helper'
+
 class SkipWordsTest < Test::Unit::TestCase
   def test_en
     assert_equal 80, Classifier::SkipWords.for('en').size

metadata

@@ -1,15 +1,15 @@
 --- !ruby/object:Gem::Specification 
 name: luisparravicini-classifier
 version: !ruby/object:Gem::Version 
-  version: 1.3.8
+  version: 1.3.9
 platform: ruby
 authors: 
-- Yury Korolev
+- Luis Parravicini
 autorequire: 
 bindir: bin
 cert_chain: []
 
-date: 2009-12-24 00:00:00 -02:00
+date: 2009-12-26 00:00:00 -02:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -33,18 +33,18 @@ dependencies:
         version: 0.5.1
     version: 
 description: Bayesian classifier and others.
-email: yury.korolev@gmail.com
+email: lparravi@gmail.com
 executables: []
 
 extensions: []
 
 extra_rdoc_files: 
 - LICENSE
-- README
+- README.rdoc
 files: 
 - LICENSE
 - Manifest
-- README
+- README.rdoc
 - Rakefile
 - VERSION.yml
 - lib/classifier.rb