krippendorff_alpha 0.0.2

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: ac1085af7a90a35a6c158b0b40a5785b4c801b24
+  data.tar.gz: db5562b2b1731fddb1a773b35bb3d66c9bce319f
+SHA512:
+  metadata.gz: 2a3c4ee44f8b3873f048d828ffdfbed80172693de54813c4544c2d5562320b7e6b2be71f1f62a9b5d4da882daa6a678f07de116e0308d1c3ea2b2e3b5450b345
+  data.tar.gz: 7532060d8915c324de3251a960cd34be828c2b16ca14673beffec0f4ebc161c26f6ec9c630a0a24dc1b5f59dbc858e3a37de87634df107a987be505c84da273e

data/ext/Rakefile

@@ -0,0 +1,7 @@
+require 'ffi-compiler/compile_task'
+
+FFI::Compiler::CompileTask.new('krippendorff_alpha') do |c|
+  %w(stddef stdlib stdbool math assert).each do |fname|
+    c.have_header?("#{fname}.h", '/usr/include')
+  end
+end

data/ext/krippendorff_alpha.c

@@ -0,0 +1,109 @@
+#include <stddef.h>
+#include <stdlib.h>
+#include <stdbool.h>
+#include <math.h>
+#include <assert.h>
+
+bool _skip_flag(int index, int *skip_indexes, size_t skip_indexes_size)
+{
+  for (size_t s_i = 0; s_i < skip_indexes_size; s_i++) {
+    if (skip_indexes[s_i] == index) {
+      return true;
+    }
+  }
+
+  return false;
+}
+
+double _metric(double c, double k)
+{
+  return pow(c - k, 2);
+}
+
+double _krippendorff_alpha(double *in_array, size_t n_rows, size_t n_columns, int *skip_indexes, size_t skip_indexes_size, double *unique_values, size_t unique_values_size)
+{
+  // Construct value-by-unit matrix
+  unsigned int *vu = calloc(n_columns * unique_values_size, sizeof(unsigned int));
+  assert(vu != NULL);
+
+  unsigned int *unit_totals = calloc(n_columns, sizeof(unsigned int));
+  assert(unit_totals != NULL);
+
+  unsigned int *unique_value_totals = calloc(unique_values_size, sizeof(unsigned int));
+  assert(unique_value_totals != NULL);
+
+  // Count the number of times the value occurs for each unit, accumulating overall totals as well
+  for (size_t v_i = 0; v_i < unique_values_size; v_i++) {
+    for (size_t u_i = 0; u_i < n_columns; u_i++) {
+      for (size_t r_i = 0; r_i < n_rows; r_i++) {
+        size_t flat_i = (u_i * n_rows) + r_i;
+
+        // Check if we match the unique value we're currently comparing against and tabulate,
+        // and skip over any values marked to be skipped
+        if (!_skip_flag(flat_i, skip_indexes, skip_indexes_size) &&
+            in_array[flat_i] == unique_values[v_i]) {
+          vu[(v_i * n_columns) + u_i] += 1;
+          unit_totals[u_i] += 1;
+        }
+      }
+    }
+  }
+
+  unsigned int total_pairable_values = 0;
+
+  // Calculate the marginal sums, now that we know if there are any lone values
+  for (size_t v_i = 0; v_i < unique_values_size; v_i++) {
+    for (size_t u_i = 0; u_i < n_columns; u_i++) {
+      // Skip any units with lone values
+      if (unit_totals[u_i] > 1) {
+        unique_value_totals[v_i] += vu[(v_i * n_columns) + u_i];
+      }
+    }
+
+    total_pairable_values += unique_value_totals[v_i];
+  }
+
+  // Calculate the observed disagreement
+  double d_o = 0.0;
+  for (size_t u_i = 0; u_i < n_columns; u_i++) {
+
+    double denominator = unit_totals[u_i] - 1;
+    if (denominator > 0) {
+
+      double sum_diff = 0.0;
+      for (size_t c_i = 0; c_i < unique_values_size; c_i++) {
+        for (size_t k_i = c_i + 1; k_i < unique_values_size; k_i++) {
+          double c = unique_values[c_i];
+          double k = unique_values[k_i];
+          double n_uc = vu[(c_i * n_columns) + u_i];
+          double n_uk = vu[(k_i * n_columns) + u_i];
+
+          sum_diff += n_uc * n_uk * _metric(c, k);
+        }
+      }
+
+      d_o += sum_diff / denominator;
+    }
+  }
+
+  // Calculate the expected disagreement
+  double d_e = 0.0;
+  for (size_t c_i = 0; c_i < unique_values_size; c_i++) {
+    for (size_t k_i = c_i + 1; k_i < unique_values_size; k_i++) {
+      double c = unique_values[c_i];
+      double k = unique_values[k_i];
+      double dot_c = unique_value_totals[c_i];
+      double dot_k = unique_value_totals[k_i];
+
+      d_e += dot_c * dot_k * _metric(c, k);
+    }
+  }
+
+  double a = 1 - (total_pairable_values - 1) * (d_o / d_e);
+
+  free(vu);
+  free(unit_totals);
+  free(unique_value_totals);
+
+  return a;
+}

data/ext/mkrf_conf.rb

@@ -0,0 +1 @@
+task :default { `make` }

data/krippendorff_alpha.gemspec

@@ -0,0 +1,20 @@
+Gem::Specification.new do |s|
+  s.name        = 'krippendorff_alpha'
+  s.version     = '0.0.2'
+  s.date        = '2017-08-30'
+  s.summary     = "Krippendorff Alpha implementation"
+  s.description = "A Krippendorff Alpha implementation for std-ruby Matrix library"
+  s.authors     = ["Brad Folkens"]
+  s.email       = 'bfolkens@gmail.com'
+  s.files       = %w(krippendorff_alpha.gemspec) + Dir.glob("{lib,spec,ext}/**/*")
+  s.license     = 'MIT'
+	s.homepage    = 'http://github.com/bfolkens/ruby-krippendorff-alpha'
+
+  s.extensions << 'ext/Rakefile'
+  s.extensions << 'ext/mkrf_conf.rb'
+
+  s.add_development_dependency 'minitest', '~> 0'
+	s.add_development_dependency 'rake', '~> 0'
+
+  s.add_dependency 'ffi-compiler', '~> 1.0'
+end

data/lib/krippendorff_alpha.rb

@@ -0,0 +1,65 @@
+require 'matrix'
+require 'ffi'
+require 'ffi-compiler/loader'
+
+module KrippendorffAlphaLib
+  extend FFI::Library
+
+  ffi_lib FFI::Library::LIBC
+
+  # memory allocators
+  attach_function :malloc, [:size_t], :pointer
+  attach_function :calloc, [:size_t], :pointer
+  attach_function :valloc, [:size_t], :pointer
+  attach_function :realloc, [:pointer, :size_t], :pointer
+  attach_function :free, [:pointer], :void
+
+  # memory movers
+  attach_function :memcpy, [:pointer, :pointer, :size_t], :pointer
+  attach_function :bcopy, [:pointer, :pointer, :size_t], :void
+
+  ffi_lib FFI::Compiler::Loader.find('krippendorff_alpha')
+  attach_function :_krippendorff_alpha, [:pointer, :size_t, :size_t, :pointer, :size_t, :pointer, :size_t], :double
+end
+
+module KrippendorffAlpha
+  include KrippendorffAlphaLib
+
+  # Calculate Krippendorff's alpha (inter-rater reliability)
+  # http://repository.upenn.edu/cgi/viewcontent.cgi?article=1043&context=asc_papers
+  #
+  # Assumed input (Matrix)
+  # [
+  #   [ nil, nil, nil, nil, nil, 3, 4, 1, 2, 1, 1, 3, 3, nil, 3 ],    # coder 1
+  #   [ 1, nil, 2, 1, 3, 3, 4, 3, nil, nil, nil, nil, nil, nil, nil], # coder 2
+  #   [ nil, nil, 2, 1, 3, 4, 4, nil, 2, 1, 1, 3, 3, nil, 4 ]         # coder 3
+  # ]
+  def krippendorff_alpha
+    in_array = self.to_a
+    in_array_flattened = in_array.transpose.flatten
+    unique_values = in_array.flatten.compact.uniq
+
+    # We need to keep track of the skip indexes separately since we can't send nils via C array of double
+    skip_indexes = []
+    in_array_flattened.each_with_index do |element, i|
+      skip_indexes << i if element.nil?
+    end
+
+    # Reformat the in_array to not have nil
+    skip_indexes.each {|i| in_array_flattened[i] = 0 }
+
+    FFI::MemoryPointer.new(:double, in_array_flattened.size) do |in_array_ptr|
+      FFI::MemoryPointer.new(:double, unique_values.size) do |unique_values_ptr|
+        FFI::MemoryPointer.new(:int, skip_indexes.size) do |skip_indexes_ptr|
+          in_array_ptr.write_array_of_double(in_array_flattened)
+          unique_values_ptr.write_array_of_double(unique_values)
+          skip_indexes_ptr.write_array_of_int(skip_indexes)
+
+          return _krippendorff_alpha(in_array_ptr, row_count, column_count, skip_indexes_ptr, skip_indexes.size, unique_values_ptr, unique_values.size)
+        end
+      end
+    end
+  end
+end
+
+Matrix.send(:include, KrippendorffAlpha)

metadata

@@ -0,0 +1,94 @@
+--- !ruby/object:Gem::Specification
+name: krippendorff_alpha
+version: !ruby/object:Gem::Version
+  version: 0.0.2
+platform: ruby
+authors:
+- Brad Folkens
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2017-08-30 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: minitest
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: ffi-compiler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
+description: A Krippendorff Alpha implementation for std-ruby Matrix library
+email: bfolkens@gmail.com
+executables: []
+extensions:
+- ext/Rakefile
+- ext/mkrf_conf.rb
+extra_rdoc_files: []
+files:
+- ext/Rakefile
+- ext/krippendorff_alpha.c
+- ext/mkrf_conf.rb
+- ext/x86_64-darwin/krippendorff_alpha.o
+- ext/x86_64-darwin/libkrippendorff_alpha.bundle
+- krippendorff_alpha.gemspec
+- lib/krippendorff_alpha.rb
+homepage: http://github.com/bfolkens/ruby-krippendorff-alpha
+licenses:
+- MIT
+metadata: {}
+post_install_message: 
+rdoc_options: []
+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: 2.4.8
+signing_key: 
+specification_version: 4
+summary: Krippendorff Alpha implementation
+test_files: []