aliastable 3.0.2 → 3.0.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 95ea83614738c6b2c92b8db5abc1afd806f05d39
-  data.tar.gz: 780af8b387b6caf1cdf6597b6a551cb2aca25f5f
+  metadata.gz: 64997be896934a0e70ba0c4776e26f08a68fae50
+  data.tar.gz: 7e7d8a38a21f8355be9c0e83559fb3e5689ea556
 SHA512:
-  metadata.gz: 336a3bec2883cd91589968dd0b5a9c79274b3278d074ecd7301c9f4b8fd2edae0daf9c55d8244a625cd0a2b0092d79fc2a4a7b448fe9ff0c07e90215c2ac1099
-  data.tar.gz: 3232ecaf98cbda4cc48d45989da5b62d08c68836b774bfb8438f4a95a90065a86ab0a57d60c50a7128255bf609c47d24f07ff8adfa23ef7d2650d2ff9897de29
+  metadata.gz: e49595d9a6da320b2e22db5672f5fc4546a4919edc3915dbd6e9b1cec69beff947b16006f76a55727c147946dd08cec4f2aeaad4be31df6e9a606de27190ed6d
+  data.tar.gz: f696a2b2f0b5de5eb7b9932f4a121f0b1e94e988284eb0123e79c45bef8e59dee1c3b089ab5e24ddd448ab17e22acfe367cbb52e8b3130e7745307c3bd4ae638

data/aliastable.gemspec

@@ -1,17 +1,17 @@
 # -*- ruby -*-
-_VERSION = "3.0.2"
+_VERSION = "3.0.3"
 
 Gem::Specification.new do |s|
   s.name = "aliastable"
   s.version = _VERSION
-  s.date = "2015-04-19"
+  s.date = "2015-09-10"
   s.summary = "Efficiently generate random outcomes from an arbitrary categorical distribution."
   s.email = "pjs@alum.mit.edu"
   s.description = "If a categorical distribution has k distinct values, traditional approaches will require O(k) work to pick an outcome with the correct probabilities.  This algorithm uses conditional probability to construct a table which will yield outcomes with the correct probabilities, but in O(1) time."
   s.author = "Paul J Sanchez"
   s.files = %w[
     aliastable.gemspec
-    lib/alias.rb
+    lib/aliastable.rb
     Rakefile
     test/infile.bad.1
     test/infile.bad.2

data/lib/{alias.rb → aliastable.rb}

@@ -12,55 +12,47 @@ class AliasTable
   # Construct an alias table from a set of values and their associated
   # probabilities.  Values and their probabilities must be synchronized,
   # i.e., they must be arrays of the same length.  Values can be
-  # anything, but the probabilities must be positive numbers that
-  # sum to one.
+  # anything, but the probabilities must be positive Rational numbers
+  # that sum to one.
   #
   # *Arguments*::
-  #   - +x_set+ -> the set of values to generate from.
+  #   - +x_set+ -> the set of values from which to generate.
   #   - +p_value+ -> the synchronized set of probabilities associated
   #     with the value set. These values should be Rationals to avoid
   #     rounding errors.
   # *Raises*::
   #   - RuntimeError if +x_set+ and +p_value+s are different lengths.
-  #   - RuntimeError if any +p_value+ are negative.
-  #   - RuntimeError if +p_value+ don't sum to one. Use Rationals to avoid this.
+  #   - RuntimeError if any +p_value+ is negative.
+  #   - RuntimeError if +p_value+s don't sum to one. Rationals will avoid this.
   #
-  def initialize(x_values, p_values)
-    if x_values.length != p_values.length
-      raise "Args to AliasTable must be vectors of the same length."
+  def initialize(x_set, p_value)
+    if x_set.length != p_value.length
+      fail 'Args to AliasTable must be vectors of the same length.'
     end
-    p_val = p_values.map do |current_p|
-      tmp = current_p.to_r
-      raise "p_values must be positive" if tmp <= 0.0
-      tmp
-    end
-    unless p_val.reduce(:+) == Rational(1)
-      raise "p_values must sum to 1.0"
-    end
-    @x = x_values.clone.freeze
+    fail 'p_values must be positive' unless p_value.all? { |value| value > 0 }
+    @p_primary = p_value.map(&:to_r)
+    fail 'p_values must sum to 1' unless @p_primary.reduce(:+) == Rational(1)
+    @x = x_set.clone.freeze
     @alias = Array.new(@x.length)
-    @p_primary = Array.new(@x.length).map{Rational(1)}
-    equiprob = Rational(1, @x.length)
-    deficit_set = []
-    surplus_set = []
-    @x.each_index do |i|
-      unless p_val[i] == equiprob
-        (p_val[i] < equiprob ? deficit_set : surplus_set) << i
-      end
-    end
-    until deficit_set.empty? do
+    parity = Rational(1, @x.length)
+    group = @p_primary.each_index.group_by { |i| @p_primary[i] <=> parity }
+    deficit_set = group[-1]
+    surplus_set = group[1]
+    until deficit_set.empty?
       deficit = deficit_set.pop
       surplus = surplus_set.pop
-      @p_primary[deficit] = p_val[deficit] / equiprob
+      @p_primary[surplus] -= parity - @p_primary[deficit]
+      @p_primary[deficit] /= parity
       @alias[deficit] = @x[surplus]
-      p_val[surplus] -= equiprob - p_val[deficit]
-      unless p_val[surplus] == equiprob
-        (p_val[surplus] < equiprob ? deficit_set : surplus_set) << surplus
+      if @p_primary[surplus] == parity
+        @p_primary[surplus] = Rational(1)
+      else
+        (@p_primary[surplus] < parity ? deficit_set : surplus_set) << surplus
       end
     end
   end
 
-  # Returns a random outcome from this object's distribution.
+  # Return a random outcome from this object's distribution.
   # The generate method is O(1) time, but is not an inversion
   # since two uniforms are used for each value that gets generated.
   #
@@ -68,5 +60,4 @@ class AliasTable
     column = rand(@x.length)
     rand <= @p_primary[column] ? @x[column] : @alias[column]
   end
-
 end

data/test/test_alias.rb

@@ -1,42 +1,42 @@
 #!/usr/bin/env ruby -w
 
-require 'alias'
+require_relative '../lib/aliastable.rb'
 
 nvars = 1_000_000
 begin
-  at = AliasTable.new(["yes", "no"], [0.3, 0.3, 0.4])
-  nvars.times {print at.generate, "\n"}
-rescue Exception => e
-  puts e.message
+  at = AliasTable.new(%w(yes no), [0.3, 0.3, 0.4])
+  nvars.times { print at.generate, "\n" }
+rescue RuntimeError => e
+  p e
   puts
 end
-Dir["test/infile.*"].each do |f_name|
+Dir['test/infile.*'].each do |f_name|
   x = []
   probs = []
-  f = File.open(f_name, "r")
+  f = File.open(f_name, 'r')
   counts = {}
   expected_counts = {}
-  while line = f.gets do
+  while line = f.gets
     inputs = line.strip.split(/[\s,;:]+/)
     x << inputs[0]
     counts[inputs[0]] = 0
     probs << inputs[1].to_r
     n_hat = probs[-1] * nvars
-    half_width = 2.5 * Math::sqrt(n_hat * (1.0 - probs[-1])) if n_hat > 0
+    half_width = 2.5 * Math.sqrt(n_hat * (1.0 - probs[-1])) if n_hat > 0
     expected_counts[inputs[0]] = [n_hat, half_width]
   end
   f.close
   begin
     at = AliasTable.new(x, probs)
-    nvars.times {counts[at.generate] += 1}
-    puts "All values should be in range almost always:"
+    nvars.times { counts[at.generate] += 1 }
+    puts 'All values should be in range almost always:'
     counts.each_key do |k|
       printf "%s: Allowable Range = %d, Expected - Observed = %d\n",
-        k, expected_counts[k][1], expected_counts[k][0] - counts[k]
+             k, expected_counts[k][1], expected_counts[k][0] - counts[k]
     end
     puts
-  rescue Exception => e
-    puts e.message
+  rescue RuntimeError => e
+    p e
     puts
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: aliastable
 version: !ruby/object:Gem::Version
-  version: 3.0.2
+  version: 3.0.3
 platform: ruby
 authors:
 - Paul J Sanchez
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-04-19 00:00:00.000000000 Z
+date: 2015-09-10 00:00:00.000000000 Z
 dependencies: []
 description: If a categorical distribution has k distinct values, traditional approaches
   will require O(k) work to pick an outcome with the correct probabilities.  This
@@ -21,7 +21,7 @@ extra_rdoc_files: []
 files:
 - Rakefile
 - aliastable.gemspec
-- lib/alias.rb
+- lib/aliastable.rb
 - test/infile.bad.1
 - test/infile.bad.2
 - test/infile.bad.3
@@ -49,7 +49,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.4.5
+rubygems_version: 2.4.5.1
 signing_key: 
 specification_version: 4
 summary: Efficiently generate random outcomes from an arbitrary categorical distribution.