fselector 0.9.0 → 1.0.0

data/lib/fselector/ensemble.rb

@@ -4,9 +4,11 @@
 module FSelector
   # select feature by an ensemble of ranking algorithms
   class Ensemble < Base
-    # new()
     #
-    # @param [Array] rankers multiple feature ranking algorithms
+    # initialize from multiple algorithms
+    #
+    # @param [Array] algos multiple feature selection algorithms
+    #
     def initialize(*algos)
       super(nil)
       

data/lib/fselector/entropy.rb

@@ -1,20 +1,22 @@
 #
 # entropy-related functions for discrete data
 #
+# ref: [Wikipedia](http://en.wikipedia.org/wiki/Mutual_information)
+#
 module Entropy
   #
-  # get the marginal entropy of array (X)
+  # get the marginal entropy of vector (X)
   #
   #     H(X) = -1 * sigma_i (P(x_i) log2 P(x_i))
   #
-  # @param [Array] arrX array of interest
+  # @param [Array] vecX vector of interest
   # @return [Float] H(X)
-   def get_marginal_entropy(arrX)
+   def get_marginal_entropy(vecX)
     h = 0.0
-    n = arrX.size.to_f
+    n = vecX.size.to_f
 
-    arrX.uniq.each do |x_i|
-      p = arrX.count(x_i)/n
+    vecX.uniq.each do |x_i|
+      p = vecX.count(x_i)/n
       h += -1.0 * (p * Math.log2(p))
     end
 
@@ -23,28 +25,28 @@ module Entropy
   
   
   #
-  # get the conditional entropy of array (X) given another array (Y)
+  # get the conditional entropy of vector (X) given another vector (Y)
   #
-  #     H(X|Y) = sigma_j (P(y_j) * H(C|y_j))
+  #     H(X|Y) = sigma_j (P(y_j) * H(X|y_j))
   #     
   #     where H(X|y_j) = -1 * sigma_i (P(x_i|y_j) log2 P(x_i|y_j))
   #
-  # @param [Array] arrX the first array
-  # @param [Array] arrY the second array
+  # @param [Array] vecX the first vector
+  # @param [Array] vecY the second vector
   # @return [Float] H(X|Y)
-  # @note arrX and arrY must be of same length
-   def get_conditional_entropy(arrX, arrY)
+  # @note vecX and vecY must be of same length
+   def get_conditional_entropy(vecX, vecY)
     abort "[#{__FILE__}@#{__LINE__}]: "+
-          "array must be of same length" if not arrX.size == arrY.size
+          "vector must be of same length" if not vecX.size == vecY.size
 
     hxy = 0.0
-    n = arrX.size.to_f
+    n = vecX.size.to_f
 
-    arrY.uniq.each do |y_j|
-      p1 = arrY.count(y_j)/n
+    vecY.uniq.each do |y_j|
+      p1 = vecY.count(y_j)/n
 
-      indices = (0...n).to_a.select { |k| arrY[k] == y_j }
-      xvs = arrX.values_at(*indices)
+      indices = (0...n).to_a.select { |k| vecY[k] == y_j }
+      xvs = vecX.values_at(*indices)
       m = xvs.size.to_f
 
       xvs.uniq.each do |x_i|
@@ -59,97 +61,65 @@ module Entropy
   
   
   #
-  # get the joint entropy of array (X) and array (Y)
+  # get the joint entropy of vector (X) and vector (Y)
   # 
   #     H(X,Y) = H(Y) + H(X|Y)
   #            = H(X) + H(Y|X)
   #     
   #     i.e. H(X,Y) == H(Y,X)
   #
-  # @param [Array] arrX the first array
-  # @param [Array] arrY the second array
+  # @param [Array] vecX the first vector
+  # @param [Array] vecY the second vector
   # @return [Float] H(X,Y)
-  # @note arrX and arrY must be of same length
-   def get_joint_entropy(arrX, arrY)
-    abort "[#{__FILE__}@#{__LINE__}]: "+
-        "array must be of same length" if not arrX.size == arrY.size
-
-    get_marginal_entropy(arrY) + get_conditional_entropy(arrX, arrY)
+  # @note vecX and vecY must be of same length
+  #
+   def get_joint_entropy(vecX, vecY)
+    get_marginal_entropy(vecY) + get_conditional_entropy(vecX, vecY)
   end # get_joint_entropy
   
   
   #
-  # get the symmetrical uncertainty of array (X) and array (Y)
+  # get the information gain of vector (X) given another vector (Y)
+  # 
+  #     IG(X;Y) = H(X) - H(X|Y)
+  #             = H(Y) - H(Y|X) = IG(Y;X)
   #
-  # @param [Array] arrX the first array
-  # @param [Array] arrY the second array
-  # @return [Float] SU(X,Y)
+  # @param [Array] vecX the first vector
+  # @param [Array] vecY the second vector
+  # @return [Float] IG(X;Y)
+  # @note vecX and vecY must be of same length
   #
-  def get_symmetrical_uncertainty(arrX, arrY)
-    abort "[#{__FILE__}@#{__LINE__}]: "+
-        "array must be of same length" if not arrX.size == arrY.size
-    
-    hx = get_marginal_entropy(arrX)
-    hxy = get_conditional_entropy(arrX, arrY)
-    hy = get_marginal_entropy(arrY)
+   def get_information_gain(vecX, vecY)
+    get_marginal_entropy(vecX) - get_conditional_entropy(vecX, vecY)
+  end # get_joint_entropy
+  
+  
+  #
+  # get the symmetrical uncertainty of vector (X) and vector (Y)
+  #
+  #                      IG(X;Y)
+  #     SU(X;Y) = 2 * -------------
+  #                     H(X) + H(Y)
+  #
+  #                    H(X) - H(X|Y)         H(Y) - H(Y|X)
+  #             = 2 * --------------- = 2 * --------------- = SU(Y;X)
+  #                     H(X) + H(Y)           H(X) + H(Y)
+  #
+  # @param [Array] vecX the first vector
+  # @param [Array] vecY the second vector
+  # @return [Float] SU(X;Y)
+  # @note vecX and vecY must be of same length
+  #
+  def get_symmetrical_uncertainty(vecX, vecY)  
+    hx = get_marginal_entropy(vecX)
+    hxy = get_conditional_entropy(vecX, vecY)
+    hy = get_marginal_entropy(vecY)
     
     su = 0.0
     su = 2*(hx-hxy)/(hx+hy) if not (hx+hy).zero?
+    
+    su
   end
   
   
 end # module
-
-
-=begin
-
-class Test
-  include Entropy
-end
-
-labels = ['A', 'B', 'C']
-arrX, arrY = [], []
-#40.times { arrX << labels[rand(labels.size)] }
-#40.times { arrY << labels[rand(labels.size)] }
-
-data = {
-  :c1 => [
-    {:f1 => 1},{:f1 => 1},{:f1 => 1},{:f1 => 1},{:f1 => 1},
-    {:f1 => 0}
-  ],
-  :c2 => [
-    {:f1 => 1},
-    {:f1 => 1},
-    {:f1 => 1},
-    {:f1 => 0},{:f1 => 0},{:f1 => 0},{:f1 => 0},{:f1 => 0},{:f1 => 0},{:f1 => 0},
-    {:f1 => 0},{:f1 => 0},{:f1 => 0},{:f1 => 0},{:f1 => 0},{:f1 => 0},{:f1 => 0},
-    {:f1 => 0},{:f1 => 0},{:f1 => 0},{:f1 => 0},{:f1 => 0},{:f1 => 0},{:f1 => 0},
-    {:f1 => 0},{:f1 => 0},{:f1 => 0},{:f1 => 0},{:f1 => 0},{:f1 => 0}
-  ]
-}
-
-data.each do |c, ss|
-  ss.each do |s|
-    arrX << c
-  arrY << s[:f1]
-  end
-end
-
-puts arrX.join(',')
-puts arrY.join(',')
-
-t = Test.new
-hx = t.get_marginal_entropy(arrX)
-hy = t.get_marginal_entropy(arrY)
-hxy = t.get_conditional_entropy(arrX, arrY)
-hyx = t.get_conditional_entropy(arrY, arrX)
-ig1 = hx-hxy
-ig2 = hy-hyx
-hx_y = t.get_joint_entropy(arrX, arrY)
-hy_x = t.get_joint_entropy(arrY, arrX)
-
-puts
-puts [hx, hxy, hy, hyx, ig1, ig2, ig1-ig2 ].join(',')
-puts [hx_y, hy_x, hx_y-hy_x].join(',')
-
-=end

data/lib/fselector/fileio.rb

@@ -21,7 +21,7 @@
 #
 module FileIO
   #
-  # read from random data (for test)
+  # read from random data (read only, for test purpose)
   #
   # @param [Integer] nsample number of total samples
   # @param [Integer] nclass number of classes
@@ -203,7 +203,7 @@ module FileIO
       else # data rows
         label, *fvs = ln.chomp.split(/,/)
         label = label.to_sym
-        data[label] = [] if not data.has_key? label
+        data[label] ||= []
         
         fs = {}
         fvs.each_with_index do |v, i|

data/lib/fselector/normalizer.rb

@@ -2,65 +2,74 @@
 # normalize continuous feature
 #
 module Normalizer
-   # log transformation, requires positive feature values 
-   def normalize_by_log!(base=10)
-     each_sample do |k, s|
-       s.keys.each do |f|
-         if s[f] > 0.0
-           s[f] = Math.log(s[f], base) 
-         else
-           abort "[#{__FILE__}@#{__LINE__}]: "+
-              "feature value must be positive"
-         end
-       end
-     end
-   end
-   
-   
-   # scale to [min, max], max > min
-   def normalize_by_min_max!(min=0.0, max=1.0)
-     # first determine min and max for each feature
-     f2min_max = {}
-          
-     each_feature do |f|
-       fvs = get_feature_values(f)
-       f2min_max[f] = [fvs.min, fvs.max]
-     end
-     
-     # then normalize
-     each_sample do |k, s|
-       s.keys.each do |f|
-         min_v, max_v = f2min_max[f]
-         s[f] = min + (s[f]-min_v) * (max-min) / (max_v-min_v)
-       end
-     end
-   end
-   
-   
-   # by z-score
-   #
-   # ref: [Wikipedia](http://en.wikipedia.org/wiki/Zscore)
-   def normalize_by_zscore!
-     # first determine mean and sd for each feature
-     f2mean_sd = {}
-     
-     each_feature do |f|
-       fvs = get_feature_values(f)
-       f2mean_sd[f] = fvs.mean, fvs.sd
-     end
-     
-     # then normalize
-     each_sample do |k, s|
-       s.keys.each do |f|
-         mean, sd = f2mean_sd[f]
-         if sd.zero?
-           s[f] = 0.0
-         else
-           s[f] = (s[f]-mean)/sd
-         end
-       end
-     end
-   end
+  #
+  # log transformation, requires positive feature values
+  #
+  # @param [Integer] base base for log
+  #
+  def normalize_by_log!(base=10)
+    each_sample do |k, s|
+      s.keys.each do |f|
+        if s[f] > 0.0
+          s[f] = Math.log(s[f], base) 
+        else
+          abort "[#{__FILE__}@#{__LINE__}]: "+
+             "feature value must be positive"
+        end
+      end
+    end
+  end # normalize_by_log!
+  
+  
+  #
+  # scale to [min, max], max > min
+  #
+  # @param [Float] min lower bound
+  # @param [Float] max upper bound
+  #
+  def normalize_by_min_max!(min=0.0, max=1.0)
+    # first determine min and max for each feature
+    f2min_max = {}
+         
+    each_feature do |f|
+      fvs = get_feature_values(f)
+      f2min_max[f] = [fvs.min, fvs.max]
+    end
+    
+    # then normalize
+    each_sample do |k, s|
+      s.keys.each do |f|
+        min_v, max_v = f2min_max[f]
+        s[f] = min + (s[f]-min_v) * (max-min) / (max_v-min_v)
+      end
+    end
+  end # normalize_by_min_max!
+  
+  
+  # convert to z-score
+  #
+  # ref: [Wikipedia](http://en.wikipedia.org/wiki/Zscore)
+  def normalize_by_zscore!
+    # first determine mean and sd for each feature
+    f2mean_sd = {}
+    
+    each_feature do |f|
+      fvs = get_feature_values(f)
+      f2mean_sd[f] = fvs.mean, fvs.sd
+    end
+    
+    # then normalize
+    each_sample do |k, s|
+      s.keys.each do |f|
+        mean, sd = f2mean_sd[f]
+        if sd.zero?
+          s[f] = 0.0
+        else
+          s[f] = (s[f]-mean)/sd
+        end
+      end
+    end
+  end # normalize_by_zscore!
 
   
 end # module

data/lib/fselector/replace_missing_values.rb

@@ -4,7 +4,7 @@
 module ReplaceMissingValues
   #
   # replace missing feature value by a fixed value, 
-  # applicable for both discrete and continuous feature
+  # applicable to both discrete and continuous feature
   #
   # @note data structure will be altered
   #

data/lib/fselector/util.rb

@@ -71,7 +71,7 @@ class Array
   end
   
   
-  # to symbol
+  # convert to symbol
   # @return [Array<Symbol>] converted symbols
   def to_sym
     self.collect { |x| x.to_sym }
@@ -81,7 +81,7 @@ class Array
   # pearson's correlation coefficient, 
   # two vectors must be of the same length
   #
-  # @param [Array] v the second array
+  # @param [Array] v the second vector
   # @return [Float] pearson's r
   def pearson_r(v)
     sm, vm = self.ave, v.ave
@@ -130,7 +130,7 @@ class String
   # e.g. 'a,"b, c",d'.split_me(/,/, '"') => [a, 'b, c', d]
   #
   # @param [Regex] delim_regex regular expression for split
-  # @param [String] quote quote char such as ' and "
+  # @param [String] quote_char quote char such as ' and "
   # @return [Array<String>]
   #
   def split_me(delim_regex, quote_char="'")

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: fselector
 version: !ruby/object:Gem::Version
-  version: 0.9.0
+  version: 1.0.0
   prerelease: 
 platform: ruby
 authors:
@@ -9,11 +9,11 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-04-25 00:00:00.000000000 Z
+date: 2012-05-04 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rinruby
-  requirement: &25980288 !ruby/object:Gem::Requirement
+  requirement: &25438824 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -21,7 +21,7 @@ dependencies:
         version: 2.0.2
   type: :runtime
   prerelease: false
-  version_requirements: *25980288
+  version_requirements: *25438824
 description: FSelector is a Ruby gem that aims to integrate various feature selection/ranking
   algorithms and related functions into one single package. Welcome to contact me
   (need47@gmail.com) if you'd like to contribute your own algorithms or report a bug.
@@ -70,6 +70,7 @@ files:
 - lib/fselector/algo_discrete/GMean.rb
 - lib/fselector/algo_discrete/GSSCoefficient.rb
 - lib/fselector/algo_discrete/InformationGain.rb
+- lib/fselector/algo_discrete/INTERACT.rb
 - lib/fselector/algo_discrete/LasVegasFilter.rb
 - lib/fselector/algo_discrete/LasVegasIncremental.rb
 - lib/fselector/algo_discrete/MatthewsCorrelationCoefficient.rb
@@ -86,6 +87,7 @@ files:
 - lib/fselector/algo_discrete/Sensitivity.rb
 - lib/fselector/algo_discrete/Specificity.rb
 - lib/fselector/algo_discrete/SymmetricalUncertainty.rb
+- lib/fselector/consistency.rb
 - lib/fselector/discretizer.rb
 - lib/fselector/ensemble.rb
 - lib/fselector/entropy.rb