decision-tree 0.0.1

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 37f1ad53408f84cac2ee0dfcc71cfe43708b08ca
+  data.tar.gz: 5b39986b348b91f4c4d52c10b4c293ce27404461
+SHA512:
+  metadata.gz: 171879112eb7730b21238e5ba14ebdcba0be35466caba611279a7022788f2a9f801b39bd0fe1827a64793f9ab70bc8bb281ed552d227f832daf39a4036000c70
+  data.tar.gz: 373ab5d902c1640606f4f660a4a675f51dcae142ac46061364580b53d7c31be4c711052ea854ab8ca9bdff99bed9beac403f3475cd9cdcf9a640e7b967a0051b

data/lib/decision-tree.rb

@@ -0,0 +1,137 @@
+require 'set'
+require 'json'
+
+module Enumerable
+    def entropy
+        dataset = Hash.new(0)
+        self.each{|x| dataset[x] += 1 }
+
+        entropy = 0.0
+        dataset.each do |k,v|
+            p = v.to_f / self.size
+            entropy += (-p)*Math.log2(p)
+        end
+
+        return entropy
+    end
+
+    def concitional_entropy_with(label)
+        dataset = Hash.new{|h,k| h[k] = Array.new }
+        self.each_with_index{|v,i| dataset[v] << label[i] }
+
+        new_entropy = 0.0
+        dataset.each{|k,v| new_entropy += (v.size.to_f / self.size)*v.entropy }
+        return new_entropy
+    end
+end
+
+
+class DecisionTree
+	def initialize(entries, columns=nil, algorithm='c45', dimension=nil, parent_node=nil, threshold=nil, path=nil)
+		@parent_node = parent_node
+		@path = if path.nil?
+			Array.new
+		else
+			path
+		end
+
+        @threshold = threshold
+
+        @algorithm = if algorithm=='c45' or algorithm=='id3'
+            algorithm
+        else
+            raise "Unknown algorithm"
+        end
+
+		@dimension = if dimension.nil?
+			entries[0][:features].size
+		else
+			dimension
+		end
+
+		@columns = if columns.nil?
+			@dimension.times.map{|i| "feature_#{i}"}
+		elsif columns.size != @dimension
+			raise "The number of columns is incorrect"
+		else
+			columns
+		end
+
+
+		@labels = entries.map{|x| x[:label]}
+		@entropy = @labels.entropy
+		@child_nodes = Hash.new
+
+		return if @path.size == @dimension
+		return if @entropy==0.0
+
+		@path << choose_best_feature(entries)
+
+		build_child_nodes(entries)
+	end
+
+
+	def feature_index
+		@path[-1]
+	end
+
+
+	def feature_name
+		@columns[ @path[-1] ]
+	end
+
+    def to_pseudo_code(buff=nil,indent="")
+        buff = Array.new if buff.nil?
+
+        if @child_nodes.size==0
+            result = @labels.to_set.to_a
+            if result.size==1
+                buff << "#{indent}return #{result[0]}"
+            else
+                buff << "#{indent}return #{@labels}"
+            end
+        end
+
+        @child_nodes.each do |feature_value,child_node|
+            buff << "#{indent}if(#{feature_name} == #{feature_value}){"
+            child_node.to_pseudo_code(buff, indent+"  " )
+            buff << "#{indent}}"
+        end
+        return buff
+    end
+
+    private
+	def choose_best_feature(entries)
+
+		labels = entries.map{|x| x[:label]}
+
+		max_ig = {index: -1, ig: -1}
+		@dimension.times do |i|
+			next if @path.include?(i)
+			child_entropy = entries.map{|x| x[:features][i]}.concitional_entropy_with(labels)
+
+            ig = if @algorithm=='id3'
+                @entropy - child_entropy
+            else# c45
+                (@entropy - child_entropy) / entries.map{|x| x[:features][i]}.entropy
+            end
+
+            max_ig = {index: i, ig: ig} if ig > max_ig[:ig]
+		end
+		return max_ig[:index]
+	end
+
+
+	def build_child_nodes(entries)
+
+		buff = Hash.new{|h,feature_value| h[feature_value] = Array.new}
+		entries.each do |e|
+			feature_value = e[:features][feature_index]
+			buff[feature_value] << e
+		end
+
+		buff.each do |feature_value,child_entries|
+			@child_nodes[feature_value] = DecisionTree.new(child_entries, @columns, @algorithm, @dimension, self, feature_value, @path.dup)
+		end
+	end
+end

metadata

@@ -0,0 +1,45 @@
+--- !ruby/object:Gem::Specification
+name: decision-tree
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+platform: ruby
+authors:
+- ireullin
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2017-04-17 00:00:00.000000000 Z
+dependencies: []
+description: 'A decision tree library which implemented ID3 & C4.5 of algorithms '
+email:
+- ireullin@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/decision-tree.rb
+homepage: https://github.com/ireullin/decision-tree
+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: 2.2.2
+signing_key: 
+specification_version: 4
+summary: A decision tree library which implemented ID3 & C4.5 of algorithms
+test_files: []