feature_selection 0.0.1 → 0.0.2

data/README.rdoc

@@ -32,6 +32,18 @@ Example:
 
   a.rank_features
     #=> {:spam => {term => score, term => score}, :ham => {term => score}}
+    
+== Logging
+
+There are two ways to log the activity of algorithms:
+
+  # Provide a path to somewhere to log to
+  log = File.expand_path(File.dirname(__FILE__) + '/log.txt')
+  FeatureSelection::MutualInformation.new(data, :log_to => log)
+  
+  # Provide an existing Logger object
+  log = Logger.new('log.txt')
+  FeatureSelection::MutualInformation.new(data, :log_to => log)
           
 == Copyright
 

data/VERSION

@@ -1 +1 @@
-0.0.1
+0.0.2

data/feature_selection.gemspec

@@ -5,11 +5,11 @@
 
 Gem::Specification.new do |s|
   s.name = %q{feature_selection}
-  s.version = "0.0.1"
+  s.version = "0.0.2"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.authors = ["reddavis"]
-  s.date = %q{2010-01-07}
+  s.date = %q{2010-01-11}
   s.description = %q{A library of feature selection algorithms}
   s.email = %q{reddavis@gmail.com}
   s.extra_rdoc_files = [
@@ -26,10 +26,11 @@ Gem::Specification.new do |s|
      "benchmark/benchmark.rb",
      "feature_selection.gemspec",
      "lib/feature_selection.rb",
+     "lib/feature_selection/algorithms/chi_squared.rb",
+     "lib/feature_selection/algorithms/frequency_based.rb",
+     "lib/feature_selection/algorithms/mutual_information.rb",
      "lib/feature_selection/base.rb",
-     "lib/feature_selection/chi_squared.rb",
-     "lib/feature_selection/frequency_based.rb",
-     "lib/feature_selection/mutual_information.rb",
+     "lib/feature_selection/log_helpers.rb",
      "spec/feature_selection/base_spec.rb",
      "spec/feature_selection/chi_squared_spec.rb",
      "spec/feature_selection/frequency_based_spec.rb",

data/lib/feature_selection/{chi_squared.rb → algorithms/chi_squared.rb}

@@ -10,10 +10,15 @@ module FeatureSelection
       #=> {:class => {'term' => score, 'term' => score}}
       @results = {}
       
+      n = 1
+      
       classes.each do |klass|
         @results[klass] = {}
         
         uniq_terms.each do |term|
+          log_calculations_complete(n)
+          n += 1
+          
           answer = calculate_contribution(term, klass)
           @results[klass][term] = answer
         end #terms.each

data/lib/feature_selection/{frequency_based.rb → algorithms/frequency_based.rb}

@@ -2,14 +2,22 @@ module FeatureSelection
   class FrequencyBased < Base
    
     def rank_features
+      write_to_log("Starting to rank features...")
       # Returns:
       #=> {:class => {'term' => count, 'term' => count}}
       @results = {}
       
+      # For logger
+      total_calculations = classes.size * terms.size
+      n = 1
+      
       classes.each do |klass|
         @results[klass] = {}
         
         terms.each do |term|
+          log_calculations_complete(n)
+          n += 1
+          
           if @results[klass].key?(term)
             @results[klass][term] += 1
           else
@@ -19,6 +27,13 @@ module FeatureSelection
       end #classes.each
       @results
     end
+    
+    private
+    
+    # Overwrite Base#total_calculations
+    def total_calculations
+      classes.size * terms.size
+    end
         
   end
 end

data/lib/feature_selection/{mutual_information.rb → algorithms/mutual_information.rb}

@@ -21,10 +21,15 @@ module FeatureSelection
       #=> {:class => {'term' => score, 'term' => score}}
       @results = {}
       
+      n = 1
+            
       classes.each do |klass|
         @results[klass] = {}
         
         uniq_terms.each do |term|
+          log_calculations_complete(n)
+          n += 1
+          
           answer = calculate_contribution(term, klass)
           @results[klass][term] = answer
         end #terms.each

data/lib/feature_selection/base.rb

@@ -1,8 +1,10 @@
 module FeatureSelection
   class Base
+    include LogHelpers
     
-    def initialize(data)
+    def initialize(data, options={})
       @data = data
+      create_log(options[:log_to]) if options[:log_to]
     end
     
     def classes
@@ -160,5 +162,9 @@ module FeatureSelection
       @terms ||= @data.map {|x| x[1]}.flatten
     end
     
+    def total_calculations
+      @total_calculations ||= uniq_terms.size * classes.size 
+    end
+    
   end
 end

data/lib/feature_selection/log_helpers.rb

@@ -0,0 +1,25 @@
+require 'logger'
+
+module LogHelpers
+  # 2 outcomes
+  # - Filepath given: create a log to that file
+  # - Logger object given: write to that object
+  def create_log(log_to)
+    if log_to.is_a?(Logger)
+      @log = log_to
+    else
+      @log = Logger.new(log_to)
+    end
+  end
+  
+  def write_to_log(message)
+    if @log
+      @log.info(message)
+    end
+  end
+  
+  # Writes the number of calculations completed to the log
+  def log_calculations_complete(n)
+    write_to_log("#{n}/#{total_calculations} calculations complete.")
+  end
+end

data/lib/feature_selection.rb

@@ -1,4 +1,5 @@
+require File.expand_path(File.dirname(__FILE__) + '/feature_selection/log_helpers')
 require File.expand_path(File.dirname(__FILE__) + '/feature_selection/base')
-require File.expand_path(File.dirname(__FILE__) + '/feature_selection/mutual_information')
-require File.expand_path(File.dirname(__FILE__) + '/feature_selection/chi_squared')
-require File.expand_path(File.dirname(__FILE__) + '/feature_selection/frequency_based')
+require File.expand_path(File.dirname(__FILE__) + '/feature_selection/algorithms/mutual_information')
+require File.expand_path(File.dirname(__FILE__) + '/feature_selection/algorithms/chi_squared')
+require File.expand_path(File.dirname(__FILE__) + '/feature_selection/algorithms/frequency_based')

data/spec/feature_selection/base_spec.rb

@@ -10,5 +10,75 @@ describe "Base" do
     @a.classes.should include(:spam)
     @a.classes.should include(:ham)
   end
+  
+  describe "Logger" do
+    before do
+      # Remove test files
+      Dir[File.expand_path(File.dirname(__FILE__) + "/../logger/*")].each do |file|
+        FileUtils.rm(file)
+      end
+    end
+    
+    describe "New Logger" do
+      before do
+        @log_file = log_path
+      end
+      
+      describe "Base" do
+        it "should work create a log file called test_1.txt" do
+          FeatureSelection::Base.new(data, :log_to => @log_file)
+          File.exist?(@log_file).should be_true
+        end
+      end
+      
+      describe "MutualInformation" do
+        it "should work and create a log file called test_1.txt" do
+          FeatureSelection::MutualInformation.new(data, :log_to => @log_file).rank_features
+          File.exist?(@log_file).should be_true
+        end
+      end
+      
+      describe "ChiSquared" do
+        it "should work and create a log file called test_1.txt" do
+          FeatureSelection::ChiSquared.new(data, :log_to => @log_file).rank_features
+          File.exist?(@log_file).should be_true
+        end
+      end
+      
+      describe "FrequencyBased" do
+        it "should work and create a log file called test_1.txt" do
+          FeatureSelection::FrequencyBased.new(data, :log_to => @log_file).rank_features
+          File.exist?(@log_file).should be_true
+        end
+      end
+    end
+    
+    describe "Existing Log" do
+      describe "FrequencyBased" do
+        it "should work, therefore return a hash" do
+          a = FeatureSelection::FrequencyBased.new(data, :log_to => @log_file)
+          a.rank_features.should be_a(Hash)
+        end
+      end
+      
+      describe "MutualInformation" do
+        it "should work, therefore return a hash" do
+          a = FeatureSelection::MutualInformation.new(data, :log_to => @log_file)
+          a.rank_features.should be_a(Hash)
+        end
+      end
+      
+      describe "ChiSquared" do
+        it "should work, therefore return a hash" do
+          a = FeatureSelection::ChiSquared.new(data, :log_to => @log_file)
+          a.rank_features.should be_a(Hash)
+        end
+      end
+    end
+    
+    def log_path
+      File.expand_path(File.dirname(__FILE__) + "/../logger/test_1.txt")
+    end
+  end
       
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: feature_selection
 version: !ruby/object:Gem::Version 
-  version: 0.0.1
+  version: 0.0.2
 platform: ruby
 authors: 
 - reddavis
@@ -9,7 +9,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-01-07 00:00:00 +00:00
+date: 2010-01-11 00:00:00 +00:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -41,10 +41,11 @@ files:
 - benchmark/benchmark.rb
 - feature_selection.gemspec
 - lib/feature_selection.rb
+- lib/feature_selection/algorithms/chi_squared.rb
+- lib/feature_selection/algorithms/frequency_based.rb
+- lib/feature_selection/algorithms/mutual_information.rb
 - lib/feature_selection/base.rb
-- lib/feature_selection/chi_squared.rb
-- lib/feature_selection/frequency_based.rb
-- lib/feature_selection/mutual_information.rb
+- lib/feature_selection/log_helpers.rb
 - spec/feature_selection/base_spec.rb
 - spec/feature_selection/chi_squared_spec.rb
 - spec/feature_selection/frequency_based_spec.rb