bio-band 0.1.3 → 0.1.4

data/lib/bio-band/weka/classifiers/rules/rules_utils.rb

@@ -20,14 +20,21 @@ module Rules_utils
     build_classifier(@dataset)
   end
 
+  #Set data for instance classifier
+  #ARGV
+  # data -> an Instances object
   def set_data(data)
     @dataset = data 
   end
-
+  
+  #Set a class index for the input dataset
   def set_class_index(class_index)
     @class_index = class_index 
   end
 
+  #Set options for the selected classifier
+  #ARGS:
+  #options -> a String, i.e. "-K 3"
   def set_options(options)
     options_inst = Utils.splitOptions(options)
     setOptions(options_inst)
@@ -41,10 +48,19 @@ module Rules_utils
     puts globalInfo
   end
 
+  # perform crossvalidation on a trained classifier
+  #ARGV:
+  #fold -> 'int' value
   def cross_validate(fold)
-    eval = Weka::Classifier::Evaluation.new self.class.data
-    eval.crossValidateModel(self.class.ancestors[2].new, self.class.data, fold.to_java(:int), Random.new(1))
-    eval.summary
+    if self.class.data
+      eval = Weka::Classifier::Evaluation.new self.class.data
+      eval.crossValidateModel(self.class.ancestors[2].new, self.class.data, fold.to_java(:int), Random.new(1))
+      eval.summary 
+    else
+      eval = Weka::Classifier::Evaluation.new @dataset
+      eval.crossValidateModel(self.class.ancestors[1].new, @dataset, fold.to_java(:int), Random.new(1))
+      eval.summary
+    end
   end
 
   #Class methods module

data/lib/bio-band/weka/classifiers/trees/trees.rb

@@ -3,6 +3,7 @@ require 'trees_utils'
 
 module Weka
 	module Classifier
+    #This module contains classifiers from the 'weka.classifiers.trees' package
 		module Trees
       java_import 'weka.classifiers.trees.J48'
       java_import 'weka.classifiers.trees.FT'

data/lib/bio-band/weka/classifiers/trees/trees_utils.rb

@@ -20,14 +20,21 @@ module Trees_utils
     build_classifier(@dataset)
   end
 
+  #Set input data for the selected classifier
+  #ARGV:
+  #data -> an Instances class object
   def set_data(data)
     @dataset = data 
   end
 
+  #Set the class index for the input dataset
   def set_class_index(class_index)
     @class_index = class_index 
   end
 
+  #Set options for the instance classifier
+  #ARGS:
+  #options -> A String object, i.e. "-K 3"
   def set_options(options)
     options_inst = Utils.splitOptions(options)
     setOptions(options_inst)
@@ -37,14 +44,24 @@ module Trees_utils
     listOptions.each {|key| puts "#{key.synopsis} #{key.description}"}
   end
 
+  #Return a short description for the selected classifier
   def description
     puts globalInfo
   end
 
+  # perform crossvalidation on a trained classifier
+  #ARGV:
+  #fold -> 'int' value
   def cross_validate(fold)
-    eval = Weka::Classifier::Evaluation.new self.class.data
-    eval.crossValidateModel(self.class.ancestors[2].new, self.class.data, fold.to_java(:int), Random.new(1))
-    eval.summary
+    if self.class.data
+      eval = Weka::Classifier::Evaluation.new self.class.data
+      eval.crossValidateModel(self.class.ancestors[2].new, self.class.data, fold.to_java(:int), Random.new(1))
+      eval.summary 
+    else
+      eval = Weka::Classifier::Evaluation.new @dataset
+      eval.crossValidateModel(self.class.ancestors[1].new, @dataset, fold.to_java(:int), Random.new(1))
+      eval.summary
+    end
   end
 
   #Class methods module

data/lib/bio-band/weka/clusterers/clusterers.rb

@@ -2,6 +2,7 @@ $:.unshift File.dirname(__FILE__)
 require 'clusterers_utils'
 
 module Weka
+  #This module contains the clusterers from the 'weka.clusterers' package
 	module Clusterer
     java_import 'weka.clusterers.SimpleKMeans'
     java_import 'weka.clusterers.FarthestFirst'
@@ -13,9 +14,13 @@ module Weka
     class Cobweb
       include Clusterer_utils
       class Base < Cobweb
-        def initialize
+        def initialize(&block)
           super
-          init_clusterer
+          if block_given?
+            init_instance_clusterer(&block)
+          else
+            init_clusterer  
+          end
         end
       end
     end
@@ -23,9 +28,13 @@ module Weka
     class EM
       include Clusterer_utils
       class Base < EM
-        def initialize
+        def initialize(&block)
           super
-          init_clusterer
+          if block_given?
+            init_instance_clusterer(&block)
+          else
+            init_clusterer  
+          end
         end
       end
     end
@@ -33,20 +42,27 @@ module Weka
     class HierarchicalClusterer
       include Clusterer_utils
       class Base < HierarchicalClusterer
-        def initialize
+        def initialize(&block)
           super
-          init_clusterer
+          if block_given?
+            init_instance_clusterer(&block)
+          else
+            init_clusterer  
+          end
         end
       end
     end
 
-
     class SimpleKMeans
       include Clusterer_utils
       class Base < SimpleKMeans
-        def initialize
+        def initialize(&block)
           super
-          init_clusterer
+          if block_given?
+            init_instance_clusterer(&block)
+          else
+            init_clusterer  
+          end
         end
       end
     end
@@ -54,9 +70,13 @@ module Weka
     class FarthestFirst
       include Clusterer_utils
       class Base < FarthestFirst
-        def initialize
+        def initialize(&block)
           super
-          init_clusterer
+          if block_given?
+            init_instance_clusterer(&block)
+          else
+            init_clusterer  
+          end
         end
       end
     end
@@ -64,9 +84,13 @@ module Weka
     class XMeans
       include Clusterer_utils
       class Base < XMeans
-        def initialize
+        def initialize(&block)
           super
-          init_clusterer
+          if block_given?
+            init_instance_clusterer(&block)
+          else
+            init_clusterer  
+          end
         end
       end
     end

data/lib/bio-band/weka/clusterers/clusterers_utils.rb

@@ -1,61 +1,86 @@
 #This module is used by the classes from the Clusterer module 
 #to inherit the following methods (instance and class methods)
 module Clusterer_utils
-	java_import "weka.core.Utils"
-	java_import "weka.clusterers.ClusterEvaluation"
+  java_import "weka.core.Utils"
+  java_import "weka.clusterers.ClusterEvaluation"
 
-	def init_clusterer
+  def init_clusterer
     set_options(self.class.options) if self.class.options 
     buildClusterer(self.class.data)
-	end
+  end
+
+  def init_instance_clusterer(&block)
+    self.instance_eval(&block)
+    #@dataset.setClassIndex(@class_index)
+    buildClusterer(@dataset)
+  end
 
-	#Instance methods list
+  #Instance methods list
   def self.included(base)
     base.extend(ClassMethods)
   end
 
+  #set instance data for the clusterer
+  def set_data(data)
+    @dataset = data 
+  end
+
+  #set options for the clusterer
   def set_options(options)
-  	options_inst = Utils.splitOptions(options)
-		setOptions(options_inst)
+    options_inst = Utils.splitOptions(options)
+    setOptions(options_inst)
   end
 
-	def list_options
-		listOptions.map {|key| "#{key.synopsis} #{key.description}"}.join("\n")
-	end
+  def list_options
+    listOptions.map {|key| "#{key.synopsis} #{key.description}"}.join("\n")
+  end
 
-	def description
+  #the description provided by the Weka Documentation
+  def description
     globalInfo
-	end
+  end
 
+  #list cluster centroids with coordinates
   def get_centroids 
     getClusterCentroids
   end
 
+  #list cluster's capabilities with attributes (i.e Numeric, Nominal...)
   def list_capabilities
     get_capabilities.to_s
   end
 
-  # 'data' is an Instances class object
- 	def evaluate
- 		eval = ClusterEvaluation.new
- 		eval.setClusterer(self)
- 		eval.evaluateClusterer(self.class.data)
- 		eval.clusterResultsToString
- 	end 
-
-	#Class methods module
-	module ClassMethods
-		
-		def self.classifier_attr_accessor(*args)
-	    args.each do |arg|
-	      #Here's the getter
-	      self.class_eval("def #{arg};@#{arg};end")
-	      #Here's the setter
-	      self.class_eval("def set_#{arg}(val);@#{arg}=val;end")
-	  	end
-  	end
-
-		classifier_attr_accessor :options,:data
-
-	end
+  # Validate clusterer. If the evaluation needs to be performed on a different dataset this function accepts 
+  # an optional parameter (an Instances class object)
+  def evaluate(*args)
+    eval = ClusterEvaluation.new
+    eval.setClusterer(self)
+    if not args[0]
+      if self.class.data
+        eval.evaluateClusterer(self.class.data) 
+      else
+        eval.evaluateClusterer(@dataset)
+      end
+    else
+      eval.evaluateClusterer(args[0])
+    end
+    puts 'performing evaluation'
+    eval.clusterResultsToString
+  end 
+
+  #Class methods module
+  module ClassMethods
+    
+    def self.classifier_attr_accessor(*args)
+      args.each do |arg|
+        #Here's the getter
+        self.class_eval("def #{arg};@#{arg};end")
+        #Here's the setter
+        self.class_eval("def set_#{arg}(val);@#{arg}=val;end")
+      end
+    end
+
+    classifier_attr_accessor :options,:data
+
+  end
 end

data/lib/bio-band/weka/filters/unsupervised/attribute/attribute.rb

@@ -27,6 +27,10 @@ module Weka
 
         class AddCluster
           include Unsupervised_Util
+          alias_method :clusterer, :set_clusterer
+          def set_clusterer(index)
+            set_clusterer(index)
+          end
         end
 
   			class Center
@@ -52,8 +56,12 @@ module Weka
   			class PrincipalComponents
           include Unsupervised_Util
   			end
-
+      
   			class Remove
+          alias_method :attribute_indices, :setAttributeIndices
+          def setAttributeIndices(index)
+            setAttributeIndices(index)
+          end
           include Unsupervised_Util
   			end
 

data/test/helper.rb

@@ -0,0 +1,18 @@
+require 'rubygems'
+require 'bundler'
+begin
+  Bundler.setup(:default, :development)
+rescue Bundler::BundlerError => e
+  $stderr.puts e.message
+  $stderr.puts "Run `bundle install` to install missing gems"
+  exit e.status_code
+end
+require 'test/unit'
+require 'shoulda'
+
+$LOAD_PATH.unshift(File.join(File.dirname(__FILE__), '..', 'lib'))
+$LOAD_PATH.unshift(File.dirname(__FILE__))
+require 'bio-band'
+
+class Test::Unit::TestCase
+end

data/test/test_apacheCorrelation.rb

@@ -0,0 +1,22 @@
+require 'helper'
+
+class TestApacheCorrelation < Test::Unit::TestCase
+
+  context "Apache correlation methods" do
+    should "calculate covariance having as input two Ruby arrays" do
+      result = Apache::Stat::Correlation.covariance([1,2,3,4],[6,5,2,0])
+      assert_equal -3.5, result
+    end
+
+    should "calculate Pearson correlation having as input two Ruby arrays" do
+      result = Apache::Stat::Correlation.pearson_correlation([1,2,3,4],[1,2,3,4])
+      assert_equal 1, result
+    end
+
+    should "calculate Spearman correlation having as input two Ruby arrays" do
+      result = Apache::Stat::Correlation.spearman_correlation([1,2,3,4],[4,3,2,1])
+      assert_equal -1, result
+    end
+
+  end
+end

data/test/test_apacheInference.rb

@@ -0,0 +1,46 @@
+require 'helper'
+
+class TestApacheInference < Test::Unit::TestCase
+
+  context "Apache inference module" do
+    should "Perform Wilcoxon signed rank test" do
+      val,p_val = Apache::Stat::Inference.wilcoxon_test([1,2,3,4],[6,5,2,0])
+      assert_equal 6,val
+      assert_equal 0.875,p_val
+    end
+
+    should "Computes the Chi-Square statistic comparing observed and expected frequency counts" do
+      val,p_val = Apache::Stat::Inference.chi_square([1,2,3,4,5],[1,2,3,4,5])
+      assert_equal 0,val
+      assert_equal 1,p_val
+      val,p_val = Apache::Stat::Inference.chi_square [[1,2,3,4,5],[1,2,3,4,5]]
+      assert_equal 0,val
+      assert_equal 1,p_val
+    end
+
+    should "Perform the Mann-Whitney U test on two input datasets" do
+      val,p_val = Apache::Stat::Inference.mann_whitney_u([1,2,3,4,5],[1,2,3,4,5])
+      assert_equal 12.5,val
+      assert_equal 1,p_val
+    end
+
+    should "Perform a homoscedastic T test on two input datasets" do
+      val,p_val = Apache::Stat::Inference.t_test([1,2,3,4,5],[10,11,12,13,14],homoscedastic=true)
+      assert_equal -9,val
+      assert_equal 1.853118429643006e-05,p_val
+    end
+
+    should "Perform a paired T test on two input datasets" do
+      val,p_val = Apache::Stat::Inference.t_test([1,2,3,4,5],[10,11,12,13,14])
+      assert_equal -9,val
+      assert_equal 1.853118429643006e-05,p_val
+    end
+
+    should "Calculate one-way ANOVA (analysis of variance) statistics on input data" do
+      val,p_val = Apache::Stat::Inference.one_way_anova [[1,2,3,4,5],[10,11,12,13,14]]
+      assert_equal 81,val
+      assert_equal 1.8531184296399772e-05,p_val
+    end
+
+  end
+end

data/test/test_bio-band.rb

@@ -0,0 +1,9 @@
+require 'helper'
+
+class TestBioBand < Test::Unit::TestCase
+#  should "probably rename this file and start testing for real" do
+#    flunk "hey buddy, you should probably rename this file and start testing for real"
+#  end
+end
+
+