diverge 1.4 → 1.5

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: b3c6dfef4a14c085ab041478e61c5105726ad5ef
-  data.tar.gz: cf3a2fb9db9b13a332809bca504ec5b66064902f
+  metadata.gz: ba2b6cc8caffede3c5d2f8ab9638623c066d454e
+  data.tar.gz: c5cbf22b8fdbb2b4db61b5fc971c01b7e3df70e8
 SHA512:
-  metadata.gz: 19f099057566ce796d8b90553efad207402ea8c4ec6cd88abed6b87970108de5c33697d09926c1958afcc515cf9fb292f178db67a88b5ad7b79d385b8addf552
-  data.tar.gz: 653f177767f13a8bdd28b1663d6be09fdc36fdb8c4cf3d89d2dd720ad3ed967798bf0de2e4cd26ad9b6868dbb2cf8e6e23c87ccb17f4e5d1bf5fc652aee0115c
+  metadata.gz: 906006fe28fc85baa620c23f191b9e1da02118d249a00afa5ed938addb5488b86d213cea215c98b3036b6ed68863aedfe2363ce8000af8611d4b0857d39be6c0
+  data.tar.gz: 9230cde1bd5d72ac864c31a7a09e30deb50b8a28acff4e98febad6d8d7ffca9bd03969b2671a3c084a8acb74503c412c8755e170b5b73d82a3126577fcc3d684

data/lib/diverge.rb

@@ -1,5 +1,3 @@
-require "gsl"
-
 class Diverge
   class << self
     attr_accessor :debug, :method_names
@@ -62,17 +60,19 @@ class Diverge
   alias :tvd :total_variation_distance
   
   def correlation
-    GSL::Stats::correlation(GSL::Vector.alloc(p), GSL::Vector.alloc(q))
+    square = ->(x) { x ** 2 }
+    top    = size * p.zip(q).map { |x, y| x * y }.inject(&:+) - p.inject(&:+) * q.inject(&:+)
+    bottom = Math.sqrt((size * p.map(&square).inject(&:+) - p.inject(&:+) ** 2) * (size * q.map(&square).inject(&:+) - q.inject(&:+) ** 2))
+
+    bottom.zero? ? 0 : top / bottom
   end
   
   alias :corr :correlation
 
   def spearman_correlation
-    p_ranks   = list_ranks(p)
-    q_ranks   = list_ranks(q)
-    d_squared = p.zip(q).map { |x, y| (p_ranks[x] - q_ranks[y]) ** 2 }
+    sum_d_squared = list_ranks(p).zip(list_ranks(q)).inject(0.0) { |sum, (p_data, q_data)| sum + (p_data.rank_order - q_data.rank_order) ** 2 }
 
-    1 - ((6 * d_squared.inject(&:+)) / (size * (size ** 2 - 1)))
+    1 - ((6 * sum_d_squared) / (size * (size ** 2 - 1)))
   end
 
   alias :s_corr :spearman_correlation
@@ -94,16 +94,16 @@ class Diverge
   private
 
   def list_ranks(list)
-    uniq_list = list.sort.uniq
-    Hash[*
-      list.
-        sort.
-        each_with_index.
-        group_by { |x, i| x }.
-        values.
-        map { |a| a.map(&:first).zip([a.map(&:last).avg + 1] * a.length).uniq }.
-        flatten
-    ]
+    element = Struct.new(:value, :order, :rank_order)
+    list.
+      each_with_index.
+      map { |x, i| element.new(x, i, nil) }.
+      sort_by(&:value).
+      each_with_index.
+      group_by { |e, i| e.value }.
+      values.
+      inject([]) { |memo, a| memo + a.map(&:first).map { |e| e.tap { e.rank_order = a.map(&:last).avg + 1 } } }.
+      sort_by(&:order)
   end
 
   def size

metadata

@@ -1,29 +1,15 @@
 --- !ruby/object:Gem::Specification
 name: diverge
 version: !ruby/object:Gem::Version
-  version: '1.4'
+  version: '1.5'
 platform: ruby
 authors:
 - Evan Senter
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-09-04 00:00:00.000000000 Z
-dependencies:
-- !ruby/object:Gem::Dependency
-  name: gsl
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - '>='
-      - !ruby/object:Gem::Version
-        version: 1.14.0
-  type: :runtime
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - '>='
-      - !ruby/object:Gem::Version
-        version: 1.14.0
+date: 2013-10-11 00:00:00.000000000 Z
+dependencies: []
 description: A simple collection of functions for determining the divergence between
   two distributions.
 email: evansenter@gmail.com
@@ -43,17 +29,17 @@ require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.0.6
+rubygems_version: 2.1.5
 signing_key: 
 specification_version: 4
 summary: Distribution divergences.