thera 0.0.2 → 0.0.3

data/ext/quarry/quarry_toolkit.cpp

@@ -60,7 +60,7 @@ extern "C" {
     rb_mText.define_module_function("standard_pipeline", &Preprocessing::Text::StandardPipeline);
     Data_Type<Preprocessing::Text::TextPipeline> rb_cTextPipeline = define_class_under<Preprocessing::Text::TextPipeline>(rb_mQuarry, "ImplTextPipeline")
       .define_constructor(Constructor<Preprocessing::Text::TextPipeline>());
-//      .define_method("process_text", &Preprocessing::Text::TextPipeline::process_text);
+      //.define_method("process_text", &Preprocessing::Text::TextPipeline::process_text);
     
     // storage
     Data_Type<Storage::Storage> rb_cStorage = define_class_under<Storage::Storage>(rb_mQuarry, "ImplStorage");
@@ -90,6 +90,8 @@ extern "C" {
       .define_method("train_text", &Model::Model::train_text)
       .define_method("classify", &Model::Model::classify)
       .define_method("classify_text", &Model::Model::classify_text)
+      .define_method("process_text", &Model::Model::process_text)
+      .define_method("add_text_example", &Model::Model::add_text_example)
       .define_method("set_data_set", &Model::Model::set_data_set)
       .define_method("get_data_set", &Model::Model::get_data_set)
       .define_method("set_classifier", &Model::Model::set_classifier)
@@ -110,6 +112,7 @@ extern "C" {
     
     Data_Type<DataSet::Example> rb_cDataSetExample = define_class_under<DataSet::Example>(rb_mDataSet, "ImplExample")
       .define_method("category_index", &DataSet::Example::category_index)
+      .define_method("set_category_index", &DataSet::Example::set_category_index)
       .define_method("get_value", &DataSet::Example::get_value)
       .define_method("set_value", &DataSet::Example::set_value)
       .define_constructor(Constructor<DataSet::Example, int>());

data/lib/quarry/src/classifier/naive_bayes/naive_bayes_classifier.cpp

@@ -21,6 +21,7 @@ void Classifier::NaiveBayesClassifier::prepare() {
     feature_caches[i].resize(features_size);
     
     for(int j = 0; j < features_size; j++) {
+      // FIXME: need to wipe numeric_features[j] here in case category j was, now isn't numeric on a second call to prepare
       if(!numeric_features[j])
         continue;
       feature = (DataSet::NumericFeature *) data_set->features[j];
@@ -62,12 +63,12 @@ double Classifier::NaiveBayesClassifier::score(int category, DataSet::Example *e
 }
 
 void Classifier::NaiveBayesClassifier::write_binary(Storage::Binary *file) {
-  int category_count = feature_caches.size();
+  int category_count = feature_caches.size() - 1;
   file->write_int(category_count);
   file->write_vector<double>(&category_probabilities);
   
   for(int i = 1; i <= category_count; i++)
-    file->write_vector<NumericFeatureCache>(&feature_caches[i]);
+    file->write_vector<NumericFeatureCache>(&(feature_caches[i]));
 }
 
 void Classifier::NaiveBayesClassifier::read_binary(Storage::Binary *file) {

data/lib/quarry/src/model/model.cpp

@@ -6,6 +6,15 @@ void Model::Model::train(DataSet::Example *example) {
 void Model::Model::train_text(string text) {
 }
 
+DataSet::Example *Model::Model::process_text(string text, bool create_features) {
+  return text_pipeline->process_text((DataSet::SparseDataSet *)data_set, (char *)text.c_str(), create_features);
+}
+
+void Model::Model::add_text_example(string text, string category) {
+  DataSet::Example *example = text_pipeline->process_text((DataSet::SparseDataSet *)data_set, (char *)text.c_str(), true);
+  example->set_category_index(data_set, data_set->category_feature()->value_index(category));
+}
+
 int Model::Model::classify(DataSet::Example *example) {
   return classifier->classify(example);
 }

data/lib/quarry/src/model/model.h

@@ -16,6 +16,8 @@ namespace Model {
     
     void train(DataSet::Example *example);
     void train_text(string text);
+    DataSet::Example *process_text(string text, bool create_features);
+    void add_text_example(string text, string category);
     int classify(DataSet::Example *example);
     int classify_text(string text);
     vector<Classifier::Score> *rank(DataSet::Example *example);

data/lib/quarry/src/storage/binary.cpp

@@ -448,7 +448,6 @@ Model::Model *Storage::Binary::read_model() {
 void Storage::Binary::write_model(Model::Model *model) {
   open_for_writing();
   
-  // write the 3 model components
   write_data_set(model->data_set);
   write_classifier(model->classifier);
   write_text_pipeline(model->text_pipeline);

data/lib/quarry_rb/data_set/data_set.rb

@@ -15,15 +15,15 @@ module Quarry
       end
       
       def examples
-        @examples ||= EnumerableHelper.new(self, @data_set, Example, :examples_size, :get_example_by_index)
+        @examples ||= EnumerableHelper.new(self, @data_set, ::Quarry::DataSet::Example, :examples_size, :get_example_by_index)
       end
       
       def categories
-        @categories ||= EnumerableHelper.new(self, @data_set, Category, :categories_size, :get_category_by_index)
+        @categories ||= EnumerableHelper.new(self, @data_set, ::Quarry::DataSet::Category, :categories_size, :get_category_by_index)
       end
       
       def features
-        @features ||= EnumerableHelper.new(self, @data_set, Feature, :features_size, :get_feature_by_index)
+        @features ||= EnumerableHelper.new(self, @data_set, ::Quarry::DataSet::Feature, :features_size, :get_feature_by_index)
       end
       
       def stratify(classifier, folds, skip_fold)

data/lib/quarry_rb/data_set/example.rb

@@ -25,8 +25,7 @@ module Quarry
       end
       
       def category=(new_category)
-        raise "new_category must be an instance of Quarry::DataSet::Category" unless new_category.is_a?(Category)
-        @example.set_category(new_category.category)
+        @example.set_category_index(@data_set.data_set, new_category)
       end
     end
   end

data/lib/quarry_rb/model/model.rb

@@ -6,11 +6,12 @@ module Quarry
     end
     
     def data_set
-      Quarry::DataSet::DataSet.new(@model.get_data_set)
+      @data_set ||= Quarry::DataSet::DataSet.new(@model.get_data_set)
     end
     
     def data_set=(ds)
       @model.set_data_set(ds.data_set)
+      @data_set = ds
     end
     
     # def classifier
@@ -22,11 +23,12 @@ module Quarry
     end
     
     def text_pipeline
-      TextPipeline.new(@model.get_text_pipeline)
+      @text_pipeline ||= TextPipeline.new(@model.get_text_pipeline)
     end
     
     def text_pipeline=(t)
       @model.set_text_pipeline(t.text_pipeline)
+      @text_pipeline = t
     end
     
     def train(example)
@@ -45,6 +47,14 @@ module Quarry
       @model.classify_text(text)
     end
     
+    def process_text(text, create_features = true)
+      DataSet::Example.new(@model.process_text(text, create_features), data_set)
+    end
+    
+    def add_text_example(text, category_name)
+      @model.add_text_example(text, category_name)
+    end
+    
     def rank(example)
       @model.rank(example)
     end

data/lib/quarry_rb/text_pipeline.rb

@@ -6,7 +6,7 @@ module Quarry
     end
     
     def process_text(data_set, text, create_features = false)
-      Example.new(@text_pipeline.process_text(data_set, text, create_features))
+      ::Quarry::DataSet::Example.new(@text_pipeline.process_text(data_set.data_set, text, create_features))
     end
     
     def self.standard_pipeline

data/thera.gemspec

@@ -9,7 +9,7 @@ Gem::Specification.new do |s|
   s.description = "C++ Data Mining Library for Ruby"
   s.email = "me@willcannings.com"
   s.authors = ["Will Cannings"]
-  s.version = '0.0.2'
+  s.version = '0.0.3'
   s.extensions = ["ext/extconf.rb"]
 
   s.files         = `git ls-files`.split("\n")

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: thera
 version: !ruby/object:Gem::Version
-  version: 0.0.2
+  version: 0.0.3
   prerelease: 
 platform: ruby
 authors:
@@ -13,7 +13,7 @@ date: 2011-12-07 00:00:00.000000000Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rice
-  requirement: &70171050802120 !ruby/object:Gem::Requirement
+  requirement: &70129254365400 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -21,7 +21,7 @@ dependencies:
         version: '0'
   type: :runtime
   prerelease: false
-  version_requirements: *70171050802120
+  version_requirements: *70129254365400
 description: C++ Data Mining Library for Ruby
 email: me@willcannings.com
 executables: []