kmeans-crystal 0.1.2

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: d58aa8e75aff1be703afe7ad5c3d89ff09d17512
+  data.tar.gz: 51c94d22cee9f8c7f46b5a29d60b324b656bd5ae
+SHA512:
+  metadata.gz: 6aa29a139ed0aff295ea7f17d78f22b40b1130137415cbfa6963bd20f14777df4eec21b960565988bb606b06c29a9695afc341411a4e8019e7ffd4c531ac7271
+  data.tar.gz: b1be6bc5b49282eb210d7d39c6ccc8f66f8dd113f644fe46ebbe2dc98a3d0b07506635fa0cc99c55cd2e5b4b0ab5ded7c2a19c62d1d07dadfc94067a8e3beb11

data/lib/kmeans-crystal.rb

@@ -0,0 +1,101 @@
+module KMeansCrystal
+class Cluster
+    attr_reader :centroid
+    attr_reader :entries
+    attr_reader :name
+
+    def initialize(name, centroid, vector_name)
+        @name = name
+        @centroid = centroid
+        @entries = Array.new
+        @vector_name = vector_name
+    end
+
+    def output
+        return { name: @name, centroid: @centroid, entries: @entries }
+    end
+
+    def distance(entry)
+        sum = 0.0
+        @centroid.size.times{|i| sum += (@centroid[i]-entry[@vector_name][i])**2}
+        return Math.sqrt(sum)
+    end
+
+    def update_centroid
+        new_centroid = Array.new(entries[0][@vector_name].size, 0.0)
+        entries.each do |entry|
+            entry[@vector_name].size.times do |i|
+                new_centroid[i] += entry[@vector_name][i]
+            end
+        end
+        new_centroid.map!{|e| e/entries.size }
+        return new_centroid
+    end
+end
+
+
+class Model
+    def initialize(cluster_num, entries, vector_name = :features)
+        raise 'too less cluster_num to evaluate k-means' if entries.size < cluster_num
+        @cluster_num = cluster_num
+        @entries = entries
+        @vector_name = vector_name
+    end
+
+    def train
+        init_centroids = @entries.sample(@cluster_num).map{|x| x[@vector_name]}
+        @clusters = new_clusters(init_centroids)
+        i = 0
+        while true do
+            @entries.each do |entry|
+                min_cluster = get_min(@clusters, entry)
+                min_cluster.entries << entry
+            end
+
+            yield(i+=1, @clusters.map{|x| x.output} )
+
+            new_centroids = get_new_centroids(@clusters)
+            @clusters = new_clusters(new_centroids)
+        end
+    end
+
+    def result
+        return @clusters.map{|x| x.output}
+    end
+
+    def predict(entry)
+        raise 'has not been trained' if @clusters.nil?
+        predicted_cluster = get_min(@clusters, entry)
+        return predicted_cluster.name
+    end
+
+    private
+    def get_new_centroids(clusters)
+        centroids = Array.new
+        clusters.each do |cluster|
+            centroids << cluster.update_centroid
+        end
+        return centroids
+    end
+
+    def get_min(centroids, entry)
+        min_cluster = centroids[0]
+        last_distance = centroids[0].distance(entry)
+        1.upto(centroids.size-1) do |i|
+            current_distance = centroids[i].distance(entry)
+            next if current_distance > last_distance
+            min_cluster = centroids[i]
+            last_distance = current_distance
+        end
+        return min_cluster
+    end
+
+    def new_clusters(centroids)
+        clusters = Array.new
+        centroids.each_with_index do |centroid, i|
+            clusters << Cluster.new("cluster#{i}", centroid, @vector_name)
+        end
+        return clusters
+    end
+end
+end

metadata

@@ -0,0 +1,52 @@
+--- !ruby/object:Gem::Specification
+name: kmeans-crystal
+version: !ruby/object:Gem::Version
+  version: 0.1.2
+platform: ruby
+authors:
+- ireullin
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2016-06-29 00:00:00.000000000 Z
+dependencies: []
+description: |-
+  The library for data clustering is implemented by k-means algorithm.With the library, you can monitor the model’s training processand end the training if the result is converged.
+
+  這是一個分群用的library。他實作了k-means演算法。透過這個library你可以監看整個model訓練的過程，並且在結果收斂的時候結束訓練。
+
+  https://github.com/ireullin/kmeans-crystal
+email:
+- ireullin@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/kmeans-crystal.rb
+homepage: https://github.com/ireullin/kmeans-crystal
+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: 2.4.6
+signing_key: 
+specification_version: 4
+summary: With this library, you can monitor the model’s training process and end the
+  training if the result is converged.  https://github.com/ireullin/kmeans-crystal
+test_files: []
+has_rdoc: