bayesic_matching 0.1.0 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 5cea0e170af73887df8e76272b3cd884b83ddf06
-  data.tar.gz: a3448c71f0ea4614fea06983678f988bcde33c27
+  metadata.gz: df98c7d569ac9e2e548452b9a3b40656b3626a91
+  data.tar.gz: 8da8d971e361eb88303054ca11ab83deeab10a6c
 SHA512:
-  metadata.gz: 159b74bf8b9224c7cba78e483b88f2f69c6ac943d5bcabee6bebc2a65077755d2402512139e94adde5b3b0e3be4c074d37d8fcd32262c3d6ee8b822dc6925294
-  data.tar.gz: 3e451a929ab45a35cd7462f72fd23ad962b844d7a3b854cf77479cb4347e34d647fab4d4f801c94d29e2f67015461bb5e87d2e44b6500a1fd723db0815dcdd1c
+  metadata.gz: 3b0bffa3d5381e085f7376e76053ed47b3593b9cc51a64590ea15f889a763b8fd71a40a8a5f0820444221d31990fed2936ee17d06e23fdab8cd752e2ed55e3f7
+  data.tar.gz: 64d046aff06b505337e8f3593f9cd4a89e365fdcad65cd482d2d4c3782ed6f5e33b696f0a372c93ec7257eff3f39d5681410d7ba0787cee5ebaf0f2ab72c4687

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    bayesic_matching (0.1.0)
+    bayesic_matching (0.2.0)
 
 GEM
   remote: https://rubygems.org/

data/lib/bayesic_matching/version.rb

@@ -1,3 +1,3 @@
 class BayesicMatching
-  VERSION = "0.1.0"
+  VERSION = "0.2.0"
 end

data/lib/bayesic_matching.rb

@@ -1,15 +1,15 @@
 require "bayesic_matching/version"
+require "set"
 
 class BayesicMatching
   def initialize
-    @classifications = []
+    @classifications = ::Set.new
     @classifications_by_token = {}
     @tokens_by_classification = {}
-    @total_token_entries = 0.0
   end
 
   def classify(tokens)
-    tokens = tokens.reject{|t| @classifications_by_token[t].nil? }
+    tokens = tokens.reject{|t| @classifications_by_token[t].nil? }.uniq
     tokens.each_with_object({}) do |token, hash|
       @classifications_by_token[token].each do |c|
         p_klass = hash[c] || (1.0 / @classifications.size)
@@ -23,12 +23,11 @@ class BayesicMatching
 
   def train(tokens, classification)
     @classifications << classification
-    @tokens_by_classification[classification] ||= []
-    @tokens_by_classification[classification].concat(tokens)
+    @tokens_by_classification[classification] ||= ::Set.new
     tokens.each do |token|
-      @classifications_by_token[token] ||= []
+      @tokens_by_classification[classification] << token
+      @classifications_by_token[token] ||= ::Set.new
       @classifications_by_token[token] << classification
-      @total_token_entries += 1.0
     end
   end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: bayesic_matching
 version: !ruby/object:Gem::Version
-  version: 0.1.0
+  version: 0.2.0
 platform: ruby
 authors:
 - Michael Ries