te_rex 0.0.11 → 0.0.12

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 929e90be95279fcf5093f8d865457509bcc5e692
-  data.tar.gz: 98e1e7ad046d7e3f423934187b0df0546a6500a0
+  metadata.gz: eba4b0e904897695a6ffe59f676cd41129d9f9d3
+  data.tar.gz: b9714ec30ca138fed335894c95503c113c8e605d
 SHA512:
-  metadata.gz: ad648071a76d79757ecdbc793a31753bdfdf94f6c0e8b462b9f315c8d514e9fdc07c942d2adae49dc20c7d35911554ff48c7f41c6804cf62c1727c16542d897a
-  data.tar.gz: a6397c1ea3b21735813c8f2cd739b889720d08746a5ea1903db1976890ea2c9eaac2e316ea886b2d6305c1e8ef3e02053c560b62dacaa3cfc5f0390094989893
+  metadata.gz: 2f0e6f419f099c3f485c1b829d1e5421935092af35fc2184fdd9a70de0f7127cbf95bcfc100e6e8f0d0efc7eeda15f4b8cc5f696211c852a92933220ae2894ff
+  data.tar.gz: 3f017ed552f0fdcb0ba89a0ffcc188f76c13780571b67ca4edd254980605ff5166af536610da2d3b2570b8bb5c39a3b1e0dc57d1873c1de89b370447b2426859

data/lib/format/basic_file.rb

@@ -2,10 +2,11 @@ module TeRex
   module Format
     class BasicFile
 
-      attr_reader :sentences, :path
+      attr_reader :sentences, :path, :category
 
-      def initialize(file_path)
+      def initialize(file_path, klass)
         @path = file_path
+        @category = klass
       end
 
       # Each line of file with Array object,

data/lib/format/brown_file.rb

@@ -2,10 +2,11 @@ module TeRex
   module Format
     class BrownFile
 
-      attr_accessor :sentences
+      attr_reader :sentences, :path, :category
 
-      def initialize(file_path)
+      def initialize(file_path, klass)
         @path = file_path
+        @category = klass
       end
 
       # Each line of file with Array object,

data/lib/format/error_file.rb

@@ -3,12 +3,13 @@ module TeRex
     require 'csv'
     class ErrorFile
 
-      attr_reader :sentences, :path
+      attr_reader :sentences, :path, :category
 
       @@csv_conf = {:headers => true}
 
-      def initialize(file_path)
+      def initialize(file_path, klass)
         @path = file_path
+        @category = klass
       end
 
       # Each row of csv as Array object, strip it and return 

data/lib/format/format.rb

@@ -1,7 +1,7 @@
 module TeRex
   module Format
     def self.category_term(t)
-      t.capitalize.intern
+      t.intern
     end
   end
 end

data/lib/te_rex/bayes.rb

@@ -5,8 +5,7 @@ module TeRex
   module Classifier
     class Bayes
 
-      attr_accessor :category_counts, :total_words
-      attr_reader :messages
+      attr_reader :category_counts, :total_words, :messages
 
       # categories = [{:tag => "Thing1", :msg => "Thing1 message"}, {:tag => "Thing2", :msg => "Thing2 message"}]
       # initialize({:tag => "Refund", :msg => "You'll get a refund"}, {:tag => "Nonrefund", :msg => "You won't get a refund"})

data/lib/te_rex/corpus.rb

@@ -30,18 +30,24 @@ module TeRex
       end
 
       def sentence_partition
-        #super_set = build_superset
         corpus_set = partition_files_for_sentences
-        #@sample_size = (superset.count.to_f * 0.75).round
-        @sample_size = 0.0
         @training = partition_training_by_sentence(corpus_set)
         @testing = partition_test_by_sentence(corpus_set)
-        count_all
+        c = count_all
+        @sample_size = (c.to_f * 0.75)
+        c
+      end
+
+      def build_superset
+        @set.reduce([]) do |memo,formatter|
+          memo << formatter.sentences
+        end.flatten
       end
 
+      private
       def define_set
         @set ||= Dir[@glob].map do |file|
-          @format_klass.new(file)
+          @format_klass.new(file, @category_klass)
         end
         @set
       end
@@ -72,12 +78,6 @@ module TeRex
         c_set.sample(c_set.count * 0.25)
       end
 
-      def build_superset
-        @set.reduce([]) do |memo,formatter|
-          memo << formatter.sentences
-        end.flatten
-      end
-
       def count_all
         counter = 0
         @set.map{|f| counter += f.sentences.count}

data/lib/te_rex/version.rb

@@ -1,3 +1,3 @@
 module TeRex
-  VERSION = "0.0.11"
+  VERSION = "0.0.12"
 end

data/test/corpus_test.rb

@@ -48,6 +48,11 @@ class CorpusTest < MicroTest::Test
     assert (22...27).map{|i| i}.include?((ratio * 100).to_i)
   end
 
+  test "sample size equals size of training set" do
+    sample = @@sent_corpus.total_sentences.to_f * 0.75
+    assert @@sent_corpus.sample_size == sample
+  end
+
   test "sentence counts are correct" do
     assert @@sent_corpus.set.count == 3
     assert @@sent_corpus.training.count == 9

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: te_rex
 version: !ruby/object:Gem::Version
-  version: 0.0.11
+  version: 0.0.12
 platform: ruby
 authors:
 - Joshua Bowles
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-11-13 00:00:00.000000000 Z
+date: 2014-11-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: fast-stemmer