anomaly 0.0.1 → 0.0.2

data/README.md

@@ -1,6 +1,6 @@
 # Anomaly
 
-Anomaly detection using a normal distribution.
+Easy-to-use anomaly detection
 
 ## Installation
 
@@ -16,6 +16,14 @@ And then execute:
 bundle install
 ```
 
+For max performance (about 3x faster), also install the NArray gem:
+
+```ruby
+gem "narray"
+```
+
+Anomaly will automatically detect it and use it.
+
 ## How to Use
 
 Train the detector with **only non-anomalies**. Each row is a sample.
@@ -54,11 +62,18 @@ Here's sample to code to help you find the best ε for your application.
 # TODO
 ```
 
-You can easily persist the detector in a file or database.
+You can easily persist the detector to a file or database - it's very tiny.
 
 ```ruby
-# TODO Finish example
-Marshal.dump(ad)
+serialized_ad = Marshal.dump(ad)
+
+# Save to a file
+File.open("anomaly_detector.dump", "w") {|f| f.write(serialized_ad) }
+
+# ...
+
+# Read it later
+ad2 = Marshal.load(File.open("anomaly_detector.dump", "r").read)
 ```
 
 ## Contributing

data/anomaly.gemspec

@@ -4,8 +4,8 @@ require File.expand_path('../lib/anomaly/version', __FILE__)
 Gem::Specification.new do |gem|
   gem.authors       = ["Andrew Kane"]
   gem.email         = ["andrew@getformidable.com"]
-  gem.description   = %q{Anomaly detection using a normal distribution.}
-  gem.summary       = %q{Anomaly detection using a normal distribution.}
+  gem.description   = %q{Easy-to-use anomaly detection}
+  gem.summary       = %q{Easy-to-use anomaly detection}
   gem.homepage      = "https://github.com/ankane/anomaly"
 
   gem.executables   = `git ls-files -- bin/*`.split("\n").map{ |f| File.basename(f) }

data/lib/anomaly/detector.rb

@@ -1,25 +1,38 @@
 module Anomaly
   class Detector
 
-    def initialize(data)
-      # Use NMatrix if possible
-      if defined?(NMatrix) and (!defined?(Matrix) or !data.is_a?(Matrix))
-        d = data.is_a?(NMatrix) ? data : NMatrix.to_na(data)
+    def initialize(data = nil)
+      @trained = false
+      train(data) if data
+    end
 
+    def train(data)
+      if defined?(NMatrix)
+        d = NMatrix.to_na(data)
         # Convert these to an array for Marshal.dump
         @mean = d.mean(1).to_a
         @std = d.stddev(1).to_a
       else
-        d = data.is_a?(Matrix) ? data : Matrix.rows(data)
-        cols = d.column_size.times.map{|i| d.column(i)}
+        # Default to Array, since built-in Matrix does not give us a big performance advantage.
+        d = data.to_a
+        cols = d.first.size.times.map{|i| d.map{|r| r[i]}}
         @mean = cols.map{|c| mean(c)}
         @std = cols.each_with_index.map{|c,i| std(c, @mean[i])}
       end
 
-      raise "Standard deviation cannot be zero" if @std.find_index{|i| i == 0 or i.nan?}
+      @std.map!{|std| (std == 0 or std.nan?) ? Float::MIN : std}
+
+      # raise "Standard deviation cannot be zero" if @std.find_index{|i| i == 0 or i.nan?}
+
+      @trained = true
+    end
+
+    def trained?
+      @trained
     end
 
     def probability(x)
+      raise "Train me first" unless trained?
       raise ArgumentError, "x must have #{@mean.size} elements" if x.size != @mean.size
       x.each_with_index.map{|a,i| normal_pdf(a, @mean[i], @std[i]) }.reduce(1, :*)
     end
@@ -32,8 +45,10 @@ module Anomaly
 
     SQRT2PI = Math.sqrt(2*Math::PI)
 
+    # Return 1 (exclude feature) if std ~ 0
     def normal_pdf(x, mean = 0, std = 1)
-      1/(SQRT2PI*std)*Math.exp(-((x - mean)**2/(2.0*(std**2))))
+      p = 1.0/(SQRT2PI*std)*Math.exp(-((x - mean)**2/(2.0*(std**2))))
+      p.nan? ? 1 : p
     end
 
     # Not used for NArray

data/lib/anomaly/version.rb

@@ -1,3 +1,3 @@
 module Anomaly
-  VERSION = "0.0.1"
+  VERSION = "0.0.2"
 end

data/spec/anomaly/detector_spec.rb

@@ -14,18 +14,22 @@ describe Anomaly::Detector do
   end
 
   context "when standard deviation is 0" do
-    let(:data) { [[1],[1]] }
+    let(:data) { [[0],[0]] }
 
-    it "raises error" do
-      expect{ ad }.to raise_error RuntimeError, "Standard deviation cannot be zero"
+    it "returns infinity for mean" do
+      ad.probability([0]).should == 1
+    end
+
+    it "returns 0 for not mean" do
+      ad.probability([1]).should == 0
     end
   end
 
   context "when one training example" do
-    let(:data) { [[1]] }
+    let(:data) { [[0]] }
 
-    it "raises error" do
-      expect{ ad }.to raise_error RuntimeError, "Standard deviation cannot be zero"
+    it "returns infinity" do
+      ad.probability([0]).should == 1
     end
   end
 
@@ -34,7 +38,9 @@ describe Anomaly::Detector do
     let(:sample) { [rand, rand] }
 
     it "returns the same probability as an NMatrix" do
-      ad.probability(sample).should == Anomaly::Detector.new(Matrix.rows(data)).probability(sample)
+      prob = ad.probability(sample)
+      Object.send(:remove_const, :NMatrix)
+      prob.should == Anomaly::Detector.new(data).probability(sample)
     end
   end
 end

data/spec/spec_helper.rb

@@ -2,7 +2,7 @@ require "rubygems"
 require "bundler/setup"
 
 require "anomaly"
-require "matrix"
+require "narray"
 
 RSpec.configure do |config|
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: anomaly
 version: !ruby/object:Gem::Version
-  version: 0.0.1
+  version: 0.0.2
   prerelease: 
 platform: ruby
 authors:
@@ -9,11 +9,11 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2011-12-11 00:00:00.000000000Z
+date: 2011-12-12 00:00:00.000000000Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
-  requirement: &2156676440 !ruby/object:Gem::Requirement
+  requirement: &2160640240 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -21,10 +21,10 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *2156676440
+  version_requirements: *2160640240
 - !ruby/object:Gem::Dependency
   name: rspec
-  requirement: &2156675180 !ruby/object:Gem::Requirement
+  requirement: &2160639580 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -32,8 +32,8 @@ dependencies:
         version: 2.0.0
   type: :development
   prerelease: false
-  version_requirements: *2156675180
-description: Anomaly detection using a normal distribution.
+  version_requirements: *2160639580
+description: Easy-to-use anomaly detection
 email:
 - andrew@getformidable.com
 executables: []
@@ -75,7 +75,7 @@ rubyforge_project:
 rubygems_version: 1.8.10
 signing_key: 
 specification_version: 3
-summary: Anomaly detection using a normal distribution.
+summary: Easy-to-use anomaly detection
 test_files:
 - spec/anomaly/detector_spec.rb
 - spec/spec_helper.rb