judgee 0.1.0 → 0.1.1

checksums.yaml

@@ -0,0 +1,15 @@
+---
+!binary "U0hBMQ==":
+  metadata.gz: !binary |-
+    Yjg5MGQ0YjhlNTcwYmM4ZTMwNjgxMzBkMGEyNjNmOWI4NWNhZDJmNg==
+  data.tar.gz: !binary |-
+    ZTkwY2E4ZjZiZjUxNzljOWQwNTcwOWRjZTFiNmVmYmQyODVkNGFjZQ==
+!binary "U0hBNTEy":
+  metadata.gz: !binary |-
+    MjJlYTFhMmQ4MzVmZDEyMjk1ZjdlYjk0MjdlOWU3YzI0NzI3ODE4YWIyY2U0
+    NWJjOTE2ZmQ2ZjMwMjg5NTJjODUxZmU3Yzg4ODRiZDdkZTM5ZGQyY2Q3ZTFj
+    NTJiMmUzOWE3YTMzZWU1NGRlOTljM2MzZGI1NmEyMWRlNTQzOTA=
+  data.tar.gz: !binary |-
+    ZWQwZmJlNDkxOTgxNmJlNWFjYTM0MWE2Mjc1NzYyMWNlYzZlMTZkNTQxM2Nl
+    ODUxYTkzZjU1OWRmNWIzNGQ5OWJhOThmYzMzOTg5NTU3NDdiYTQ2ZTNlNDMx
+    MGMyMWE0MWNhMTQyZmNlYzgwODFmYmVlYTMxODhhYTkyMGY5ZjU=

data/README.md

@@ -32,7 +32,6 @@ It supports strings, hashes, lists, sets, sorted sets and offers an incredible p
 	judgee = Judgee::Classifier.new(:path => "/tmp/redis.sock")
 
 
-
 	# Now you can train the classifier
 	judgee.train(:spam, ["bad", "worse", "stupid", "idiotic"])
 	judgee.train(:ham, ["good", "better", "best", "lovely"])
@@ -51,4 +50,17 @@ If you read the source code, you might stumble upon the confusing method names.
 There are two methods for training (train, train_fast), two methods for untraining (untrain, untrain_fast) and two methods for classification (classify, classify_fast).
 The difference is quite simple. As the name suggests, all methods with the suffix '_fast' are (really) faster (3x to 10x) in processing the data, but virtually unreadable.
 
-So use the '_fast' methods if you need performance, e.g. in production and the methods without the suffix for learning purposes.
+So use the '_fast' methods if you need performance, e.g. in production and the methods without the suffix for learning purposes.
+
+## For higher performance, use the _fast methods
+
+	# Now you can train the classifier
+	judgee.train_fast(:spam, ["bad", "worse", "stupid", "idiotic"])
+	judgee.train_fast(:ham, ["good", "better", "best", "lovely"])
+
+	# After training, classify your text sample
+	judgee.classify_fast(["good", "better", "best", "worse"]) # => :ham
+
+
+	# Want to untrain some words?
+	judgee.untrain_fast(:spam, ["bad", "worse"])

data/lib/judgee/classifier.rb

@@ -5,9 +5,7 @@ require "redis"
 module Judgee
   class Classifier
 
-    ###
-    # Constants
-    ###
+    ### Constants ###
     CATEGORIES_KEY  = "judgee:categories"
     CATEGORY_KEY    = "judgee:category"
     ALPHA           = 1.0
@@ -84,8 +82,8 @@ module Judgee
       occurances  = count_occurance(data)
 
       categories.each do |category|
-        numerator   = Hash[occurances.keys.zip(redis.hmget(redis_category_key(category), occurances.keys))].inject({}) { |hash, (key, value)| hash[key] = value.to_f + ALPHA; hash }
-        denominator = categories.map { |category| Hash[occurances.keys.zip(redis.hmget(redis_category_key(category), occurances.keys))] }.inject(Hash.new(0)) { |main_hash, sub_hash| main_hash.merge(sub_hash) { |key, value_first, value_second| value_first.to_f + value_second.to_f} }.inject(Hash.new(0)) { |hash, (key, value)| hash[key] = value.to_f + (ALPHA * data.length); hash }
+        numerator         = Hash[occurances.keys.zip(redis.hmget(redis_category_key(category), occurances.keys))].inject({}) { |hash, (key, value)| hash[key] = value.to_f + ALPHA; hash }
+        denominator       = categories.map { |category| Hash[occurances.keys.zip(redis.hmget(redis_category_key(category), occurances.keys))] }.inject(Hash.new(0)) { |main_hash, sub_hash| main_hash.merge(sub_hash) { |key, value_first, value_second| value_first.to_f + value_second.to_f} }.inject(Hash.new(0)) { |hash, (key, value)| hash[key] = value.to_f + (ALPHA * data.length); hash }
         result[category] += numerator.merge(denominator) { |key, value_numerator, value_denominator| (occurances[key] * Math.log(value_numerator / value_denominator)).abs }.values.inject(0, :+)
       end
 
@@ -112,7 +110,7 @@ module Judgee
     def count_occurance(data='')
       bag_of_words = Hash.new(0)
 
-      data = [data].flatten.map! do |word|
+      data = [data].flatten.map do |word|
         word.to_s.strip
       end.delete_if(&:empty?)
 
@@ -121,7 +119,7 @@ module Judgee
       end
       bag_of_words
     rescue
-      raise ArgumentError, 'input must be a single String or an Array of Strings'
+      raise ArgumentError, 'Input must be a single String or an Array of Strings'
     end
 
 

data/lib/judgee/version.rb

@@ -1,5 +1,5 @@
 # encoding: UTF-8
 
 module Judgee
-  VERSION = "0.1.0"
+  VERSION = "0.1.1"
 end

metadata

@@ -1,8 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: judgee
 version: !ruby/object:Gem::Version
-  version: 0.1.0
-  prerelease: 
+  version: 0.1.1
 platform: ruby
 authors:
 - Railsmechanic
@@ -14,7 +13,6 @@ dependencies:
 - !ruby/object:Gem::Dependency
   name: redis
   requirement: !ruby/object:Gem::Requirement
-    none: false
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
@@ -22,7 +20,6 @@ dependencies:
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
-    none: false
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
@@ -30,7 +27,6 @@ dependencies:
 - !ruby/object:Gem::Dependency
   name: rspec
   requirement: !ruby/object:Gem::Requirement
-    none: false
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
@@ -38,7 +34,6 @@ dependencies:
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
-    none: false
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
@@ -62,27 +57,26 @@ files:
 - spec/judgee_spec.rb
 homepage: https://github.com/railsmechanic/judgee
 licenses: []
+metadata: {}
 post_install_message: 
 rdoc_options: []
 require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
-  none: false
   requirements:
   - - ! '>='
     - !ruby/object:Gem::Version
       version: '0'
 required_rubygems_version: !ruby/object:Gem::Requirement
-  none: false
   requirements:
   - - ! '>='
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 1.8.24
+rubygems_version: 2.0.3
 signing_key: 
-specification_version: 3
+specification_version: 4
 summary: Judgee is a simple Bayesian Classifier with additive smoothing, which uses
   Redis for persistance.
 test_files: