treat 1.1.0 → 1.1.1

data/lib/treat/workers/formatters/serializers/mongo.rb

@@ -6,55 +6,61 @@ class Treat::Workers::Formatters::Serializers::Mongo
 
   DefaultOptions = {
     :recursive => true,
-    :stop_at => :token
+    :stop_at => nil
   }
-  
+
   def self.serialize(entity, options = {})
-    
+
     options = DefaultOptions.merge(options)
-    stop_at = options[:stop_at] ? 
+    options[:stop_at] = options[:stop_at] ?
     Treat::Entities.const_get(
-    options[:stop_at].to_s.capitalize) : 
-    Treat::Entities::Token
-    
+    options[:stop_at].to_s.capitalize) : nil
+
     if !Treat.databases.mongo.db && !options[:db]
       raise Treat::Exception,
       'Must supply the database name in config. ' +
       '(Treat.databases.mongo.db = ...) or pass ' +
       'it as a parameter to #serialize.'
     end
-    
+
     @@database ||= Mongo::Connection.
     new(Treat.databases.mongo.host).
     db(Treat.databases.mongo.db || options[:db])
-    
-    type = cl(entity.class.superclass).downcase
-    type = entity.type.to_s if type == 'entity'
-    types = type + 's'
 
-    coll = @@database.collection(types)
-    
+    supertype =  cl(Treat::Entities.const_get(
+    entity.type.to_s.capitalize.intern).superclass).downcase
+    supertype = entity.type.to_s if supertype == 'entity'
+    supertypes = supertype + 's'
+      
+    coll = @@database.collection(supertypes)
+    entity_token = self.do_serialize(entity, options)
+    coll.update({id: entity.id}, entity_token, {upsert: true})
+  end
+
+  def self.do_serialize(entity, options)
+
+    children = []
+
+    if options[:recursive] && entity.has_children?
+      entity.each do |child|
+        next if options[:stop_at] && child.class.
+        compare_with(options[:stop_at]) < 0
+        children << self.do_serialize(child, options)
+      end
+    end
+
     entity_token = {
       :id => entity.id,
       :value => entity.value,
       :string => entity.to_s,
-      :type => entity.type,
-      :children => entity.children.map { |c| [c.id, c.type] },
+      :type => entity.type.to_s,
+      :children => children,
       :parent => (entity.has_parent? ? entity.parent.id : nil),
       :features => entity.features
     }
-    
-    coll.insert(entity_token)
 
-    if options[:recursive] && entity.has_children?
-      entity.each do |child|
-        next if child.class.compare_with(stop_at) < 0
-        self.serialize(child, options)
-      end
-    end
-     
+    entity_token
+
   end
 
 end
-
-

data/lib/treat/workers/formatters/unserializers/mongo.rb

@@ -1,20 +1,9 @@
 module Treat::Workers::Formatters::Unserializers::Mongo
-
-  DefaultOptions = { 
-    :recursive => true,
-    :stop_at => nil
-  }
   
   require 'mongo'
   
   def self.unserialize(entity, options={})
 
-    options = DefaultOptions.merge(options)
-    options[:stop_at] = options[:stop_at] ? 
-    Treat::Entities.const_get(
-    options[:stop_at].to_s.capitalize) : 
-    Treat::Entities::Token
-    
     if !Treat.databases.mongo.db && !options[:db]
       raise Treat::Exception,
       'Must supply the database name in config. ' +
@@ -25,12 +14,6 @@ module Treat::Workers::Formatters::Unserializers::Mongo
     @@database ||= Mongo::Connection.
     new(Treat.databases.mongo.host).
     db(Treat.databases.mongo.db || options[:db])
-
-    self.do_unserialize(entity, options)
-    
-  end
-
-  def self.do_unserialize(entity, options)
     
     supertype =  cl(Treat::Entities.const_get(
     entity.type.to_s.capitalize.intern).superclass).downcase
@@ -39,38 +22,33 @@ module Treat::Workers::Formatters::Unserializers::Mongo
     
     coll = @@database.collection(supertypes)
     record = coll.find_one(:id => entity.id)
-
+    
     unless record
       raise Treat::Exception,
       "Couldn't find record ID #{entity.id}."
     end
+
+    self.do_unserialize(record, options)
+    
+  end
+
+  def self.do_unserialize(record, options)
+    
+    entity = Treat::Entities.
+    const_get(record['type'].
+    capitalize.intern).new(
+    record['value'], record['id'])
     
-    # Convert feature keys to symbols.
     features = record['features']
     new_feat = {}
     features.each do |feature, value|
       new_feat[feature.intern] = value
     end
-    entity.features = new_feat
     
-    # Set the entity's value.
-    entity.value = record['value']
-    
-    if entity.class.compare_with(
-      options[:stop_at]) == 0
-      entity.value = record['string']
-    end
+    entity.features = new_feat
     
-    return entity unless options[:recursive]
-
     record['children'].each do |c|
-      cid, ctype = *c
-      cklass = Treat::Entities.const_get(
-      ctype.capitalize.intern)
-      next if cklass.compare_with(
-      options[:stop_at]) < 0
-      entity << self.do_unserialize(
-      cklass.new('', cid), options)
+      entity << self.do_unserialize(c, options)
     end
 
     entity

data/lib/treat/workers/learners/classifiers/id3.rb

@@ -7,22 +7,21 @@ class Treat::Workers::Learners::Classifiers::ID3
   def self.classify(entity, options = {})
     
     set = options[:training]
-    cl = set.classification
+    cl = set.problem
     
     if !@@classifiers[cl]
       dec_tree = DecisionTree::ID3Tree.new(
-      set.labels.map { |l| l.to_s }, set.items, 
-      cl.default, cl.mode)
+      cl.labels.map { |l| l.to_s }, set.items, cl.question.default, 
+      cl.question.type)
       dec_tree.train
       @@classifiers[cl] = dec_tree
     else
       dec_tree = @@classifiers[cl]
+      dec_tree.graph('testingbitch')
     end
-    
     dec_tree.predict(
       cl.export_item(entity, false)
     )
-    
   end
   
 end

data/lib/treat/workers/learners/classifiers/mlp.rb

@@ -8,7 +8,7 @@ class Treat::Workers::Learners::Classifiers::MLP
   def self.classify(entity, options = {})
     
     set = options[:training]
-    cl = set.classification
+    cl = set.problem
       
     if !@@mlps[cl]
       net = Ai4r::NeuralNetwork::

data/lib/treat/workers.rb

@@ -34,7 +34,7 @@ module Treat::Workers
   def self.create_category(name, conf)
     category = self.const_set(name, Module.new)
     conf.each_pair do |group, worker|
-      name = group.to_s.capitalize.intern
+      name = cc(group.to_s).intern
       category.module_eval do
         @@methods = []; def methods; 
         @@methods; end; def groups; 

data/spec/entity.rb

@@ -113,12 +113,14 @@ describe Treat::Entities::Entity do
   describe "Exportable" do
 
     context "when supplied with a classification to export" do
-      classification = Treat::Core::Classification.new(:word, :tag, :is_keyword)
+      feature = Treat::Core::Feature.new(:tag)
+      question = Treat::Core::Question.new(:is_keyword, :word, :discrete, false)
+      problem = Treat::Core::Problem.new(question, feature)
       it "returns a data set with the exported features" do
-        ds = @sentence.export(classification)
-        ds.classification.should eql classification
-        ds.labels.should eql [:tag]
-        ds.ids.should eql @sentence.words.map { |w| w.id }
+        ds = @sentence.export(problem)
+        ds.problem.should eql problem
+        ds.problem.labels.should eql [:tag]
+        ds.entities.should eql @sentence.words.map { |w| w.id }
         ds.items.should eql [
           ["DT", false], ["JJ", false],
           ["NN", false], ["VBZ", false],

data/spec/phrase.rb

@@ -42,7 +42,7 @@ describe Treat::Entities::Phrase do
     
     describe "#time" do
       it "returns a DateTime object representing the time in the phrase" do
-        Treat.languages.english[:workers][:extractors][:time].each do |e|
+        Treat.languages.english.workers.extractors.time.each do |e|
           t = 'october 2006'.time(e)
           t.month.should eql 10
         end
@@ -55,7 +55,7 @@ describe Treat::Entities::Phrase do
     describe "#tokenize" do
 
       it "splits a phrase/sentence into tokens and adds them as children of the phrase" do
-        Treat.languages.english[:workers][:processors][:tokenizers].each do |t|
+        Treat.languages.english.workers.processors.tokenizers.each do |t|
           @phrase = Treat::Entities::Phrase.new('a phrase to tokenize')
           @phrase.tokenize(t)
           @phrase.children.should eql @phrase.tokens

data/spec/zone.rb

@@ -5,13 +5,12 @@ describe Treat::Entities::Zone do
   describe "Processable" do
 
     before do
-      @processors = Treat.languages.
-      english[:workers][:processors]
+      @processors = Treat.languages.english.workers.processors
     end
     describe "#segment" do
 
       it "splits a zone into phrases/sentences and adds them as children of the zone" do
-        @processors[:segmenters].each do |s|
+        @processors.segmenters.each do |s|
           paragraph = Treat::Entities::Paragraph.new(
           "This is a first sentence inside the first paragraph. " +
           "This is the second sentence that is inside the paragraph.")

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: treat
 version: !ruby/object:Gem::Version
-  version: 1.1.0
+  version: 1.1.1
   prerelease: 
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-07-11 00:00:00.000000000 Z
+date: 2012-07-19 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: schiphol
@@ -27,6 +27,38 @@ dependencies:
     - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
+- !ruby/object:Gem::Dependency
+  name: birch
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: sourcify
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: rspec
   requirement: !ruby/object:Gem::Requirement
@@ -67,11 +99,6 @@ extensions: []
 extra_rdoc_files: []
 files:
 - bin/MANIFEST
-- bin/stanford/bridge.jar
-- bin/stanford/joda-time.jar
-- bin/stanford/stanford-corenlp.jar
-- bin/stanford/stanford-parser.jar
-- bin/stanford/xom.jar
 - lib/treat/config/core/acronyms.rb
 - lib/treat/config/core/encodings.rb
 - lib/treat/config/core/entities.rb
@@ -110,10 +137,11 @@ files:
 - lib/treat/config/workers/processors.rb
 - lib/treat/config/workers/retrievers.rb
 - lib/treat/config.rb
-- lib/treat/core/classification.rb
 - lib/treat/core/data_set.rb
+- lib/treat/core/feature.rb
 - lib/treat/core/node.rb
-- lib/treat/core/server.rb
+- lib/treat/core/problem.rb
+- lib/treat/core/question.rb
 - lib/treat/core.rb
 - lib/treat/entities/abilities/buildable.rb
 - lib/treat/entities/abilities/checkable.rb
@@ -224,18 +252,12 @@ files:
 - spec/samples/mathematicians/gauss.pdf
 - spec/samples/mathematicians/leibniz.txt
 - spec/samples/mathematicians/newton.doc
-- spec/sandbox.rb
 - spec/token.rb
 - spec/treat.rb
 - spec/word.rb
 - spec/zone.rb
-- tmp/english.yaml
 - tmp/MANIFEST
-- files/21552208.html
-- files/3_2_release_notes.html
 - files/MANIFEST
-- files/nethttp-cheat-sheet-2940.html
-- files/weather-central-canada-heat-wave.html
 - README.md
 - LICENSE
 homepage: https://github.com/louismullie/treat