idhja22 0.14.2 → 0.14.3

data/.yardopts

@@ -0,0 +1 @@
+--no-private  lib/**/*.rb - LICENSE.txt

data/idhja22.gemspec

@@ -8,7 +8,7 @@ Gem::Specification.new do |gem|
   gem.version       = Idhja22::VERSION
   gem.authors       = ["Henry Addison"]
   gem.description   = %q{Decision Trees}
-  gem.summary       = %q{A different take on decision trees}
+  gem.summary       = %q{A gem for creating decision trees}
   gem.homepage      = "https://github.com/henryaddison/idhja22"
 
   gem.files         = `git ls-files`.split($/)
@@ -20,4 +20,6 @@ Gem::Specification.new do |gem|
   gem.add_development_dependency "rake"
   gem.add_development_dependency 'debugger'
   gem.add_development_dependency 'simplecov'
+  gem.add_development_dependency 'yard'
+  gem.add_development_dependency 'redcarpet'
 end

data/lib/idhja22/node.rb

@@ -1,5 +1,57 @@
 module Idhja22
   class Node
+    class << self
+      def build_node(dataset, attributes_available, depth, parent_probability = nil)
+        if(dataset.size < Idhja22::MIN_DATASET_SIZE)
+          return Idhja22::LeafNode.new(probability_guess(parent_probability, depth), dataset.category_label)
+        end
+
+        #if successful termination - create and return a leaf node
+        if(dataset.terminating? && depth > 0) # don't terminate without splitting the data at least once
+          return Idhja22::LeafNode.new(dataset.probability, dataset.category_label)
+        end
+
+        if(depth >= 3) # don't let trees get too long
+          return Idhja22::LeafNode.new(dataset.probability, dataset.category_label)
+        end
+
+        #if we have no more attributes left to split the dataset on, then return a leafnode
+        if(attributes_available.empty?)
+          return Idhja22::LeafNode.new(dataset.probability, dataset.category_label)
+        end
+
+        data_split, best_attribute = best_attribute(dataset, attributes_available)
+
+        node = Idhja22::DecisionNode.new(data_split, best_attribute, attributes_available-[best_attribute], depth, dataset.probability)
+
+        return node
+      end
+
+      private
+      def best_attribute(dataset, attributes_available)
+        data_split = best_attribute = nil
+        igain = - Float::INFINITY
+
+        attributes_available.each do |attr_label|
+          possible_split = dataset.partition(attr_label)
+          possible_igain = dataset.entropy
+          possible_split.each do |value, ds|
+            possible_igain -= (ds.size.to_f/dataset.size.to_f)*ds.entropy
+          end
+          if(possible_igain > igain)
+            igain = possible_igain
+            data_split = possible_split
+            best_attribute = attr_label
+          end
+        end
+        return data_split, best_attribute
+      end
+
+      def probability_guess(parent_probability, depth)
+        return (parent_probability + (Idhja22::DEFAULT_PROBABILITY-parent_probability)/2**depth)
+      end
+    end
+
     def ==(other)
       return self.class == other.class
     end
@@ -11,7 +63,7 @@ module Idhja22
       @decision_attribute = decision_attribute
       @branches = {}
       data_split.each do |value, dataset|
-        node = Tree.build_node(dataset, attributes_available, depth+1, parent_probability)
+        node = Node.build_node(dataset, attributes_available, depth+1, parent_probability)
         if(node.is_a?(DecisionNode) && node.branches.values.all? { |n| n.is_a?(LeafNode) })
           probs = node.branches.values.collect(&:probability)
           if(probs.max - probs.min < 0.01)

data/lib/idhja22/tree.rb

@@ -1,11 +1,16 @@
 module Idhja22
+  # The main entry class for a training, viewing and evaluating a decision tree.
   class Tree
     attr_accessor :root
     class << self
+      # Trains a Tree using the provided Dataset.
       def train(dataset)
         new(dataset, dataset.attribute_labels)
       end
 
+      # Takes a dataset and splits it randomly into training and validation data. 
+      # Uses the training data to train a tree whose perfomance then measured using the validation data.
+      # @param [Float] Proportion of dataset to use for training. The rest will be used to validate the resulting tree.
       def train_and_validate(dataset, training_proportion=0.5)
         training_set, validation_set = dataset.split(training_proportion)
         tree = self.train(training_set)
@@ -13,70 +18,24 @@ module Idhja22
         return tree, validation_value
       end
 
+      # see #train
+      # @note Takes a CSV filename rather than a Dataset
       def train_from_csv(filename)
         ds = Dataset.from_csv(filename)
         train(ds)
       end
 
+      # see #train_and_validate
+      # @note Takes a CSV filename rather than a Dataset
       def train_and_validate_from_csv(filename, training_proportion=0.5)
         ds = Dataset.from_csv(filename)
         train_and_validate(ds, training_proportion)
       end
-
-      def build_node(dataset, attributes_available, depth, parent_probability = nil)
-        if(dataset.size < Idhja22::MIN_DATASET_SIZE)
-          return Idhja22::LeafNode.new(probability_guess(parent_probability, depth), dataset.category_label)
-        end
-
-        #if successful termination - create and return a leaf node
-        if(dataset.terminating? && depth > 0) # don't terminate without splitting the data at least once
-          return Idhja22::LeafNode.new(dataset.probability, dataset.category_label)
-        end
-
-        if(depth >= 3) # don't let trees get too long
-          return Idhja22::LeafNode.new(dataset.probability, dataset.category_label)
-        end
-
-        #if we have no more attributes left to split the dataset on, then return a leafnode
-        if(attributes_available.empty?)
-          return Idhja22::LeafNode.new(dataset.probability, dataset.category_label)
-        end
-
-        data_split , best_attribute = best_attribute(dataset, attributes_available)
-
-        node = Idhja22::DecisionNode.new(data_split, best_attribute, attributes_available-[best_attribute], depth, dataset.probability)
-
-        return node
-      end
-
-      private
-      def best_attribute(dataset, attributes_available)
-        data_split = best_attribute = nil
-        igain = - Float::INFINITY
-
-        attributes_available.each do |attr_label|
-          possible_split = dataset.partition(attr_label)
-          possible_igain = dataset.entropy
-          possible_split.each do |value, ds|
-            possible_igain -= (ds.size.to_f/dataset.size.to_f)*ds.entropy
-          end
-          if(possible_igain > igain)
-            igain = possible_igain
-            data_split = possible_split
-            best_attribute = attr_label
-          end
-        end
-        return data_split, best_attribute
-      end
-
-      def probability_guess(parent_probability, depth)
-        return (parent_probability + (Idhja22::DEFAULT_PROBABILITY-parent_probability)/2**depth)
-      end
     end
 
     def initialize(dataset, attributes_available)
       raise Idhja22::Dataset::InsufficientData, "require at least #{Idhja22::MIN_DATASET_SIZE} data points, only have #{dataset.size} in data set provided" if(dataset.size < Idhja22::MIN_DATASET_SIZE)
-      @root = self.class.build_node(dataset, attributes_available, 0)
+      @root = Node.build_node(dataset, attributes_available, 0)
     end
 
     def get_rules

data/lib/idhja22/version.rb

@@ -1,3 +1,3 @@
 module Idhja22
-  VERSION = "0.14.2"
+  VERSION = "0.14.3"
 end

data/spec/version_spec.rb

@@ -3,7 +3,7 @@ require 'spec_helper'
 describe Idhja22 do
   describe 'VERSION' do
     it 'should be current version' do
-      Idhja22::VERSION.should == '0.14.2'
+      Idhja22::VERSION.should == '0.14.3'
     end
   end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: idhja22
 version: !ruby/object:Gem::Version
-  version: 0.14.2
+  version: 0.14.3
   prerelease: 
 platform: ruby
 authors:
@@ -75,6 +75,38 @@ dependencies:
     - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
+- !ruby/object:Gem::Dependency
+  name: yard
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: redcarpet
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
 description: Decision Trees
 email: 
 executables:
@@ -84,6 +116,7 @@ extra_rdoc_files: []
 files:
 - .gitignore
 - .travis.yml
+- .yardopts
 - Gemfile
 - LICENSE.txt
 - README.md
@@ -121,7 +154,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: -4104544286961851710
+      hash: 2323453043414878291
 required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
   requirements:
@@ -130,13 +163,13 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
       segments:
       - 0
-      hash: -4104544286961851710
+      hash: 2323453043414878291
 requirements: []
 rubyforge_project: 
 rubygems_version: 1.8.24
 signing_key: 
 specification_version: 3
-summary: A different take on decision trees
+summary: A gem for creating decision trees
 test_files:
 - spec/another_large_spec_data.csv
 - spec/dataset/example_spec.rb
@@ -147,3 +180,4 @@ test_files:
 - spec/spec_helper.rb
 - spec/tree_spec.rb
 - spec/version_spec.rb
+has_rdoc: