statsample-glm 0.0.1 → 0.1.0

data/spec/normal_spec.rb

@@ -0,0 +1,15 @@
+describe Statsample::GLM::Normal do
+  context "MLE algorithm", focus: true do
+    before do
+      # Below data set taken from http://dl.dropbox.com/u/10246536/Web/RTutorialSeries/dataset_multipleRegression.csv
+      @ds = Statsample::CSV.read "spec/data/normal.csv"
+    end
+
+    it "reports correct values as an array" do
+      @glm = Statsample::GLM.compute @ds, 'roll', :normal, {algorithm: :mle}
+      
+      expect_similar_vector @glm.coefficients, [450.12450365911894, 
+        0.4064837278023981, 4.27485769721736,-9153.254462671905]
+    end
+  end
+end

data/spec/poisson_spec.rb

@@ -0,0 +1,32 @@
+describe Statsample::GLM::Poisson do 
+  context "IRLS algorithm" do
+    before :each do
+      x1      = Statsample::Vector.new([0.537322309644812,-0.717124209978434,-0.519166718891331,0.434970973986765,-0.761822002215759,1.51170030921189,0.883854199811195,-0.908689798854196,1.70331977539793,-0.246971150634099,-1.59077593922623,-0.721548040910253,0.467025703920194,-0.510132788447137,0.430106510266798,-0.144353683251536,-1.54943800728303,0.849307651309298,-0.640304240933579,1.31462478279425,-0.399783455165345,0.0453055645017902,-2.58212161987746,-1.16484414309359,-1.08829266466281,-0.243893919684792,-1.96655661929441,0.301335373291024,-0.665832694463588,-0.0120650855753837,1.5116066367604,0.557300353673344,1.12829931872045,0.234443748015922,-2.03486690662651,0.275544751380246,-0.231465849558696,-0.356880153225012,-0.57746647541923,1.35758352580655,1.23971669378224,-0.662466275100489,0.313263561921793,-1.08783223256362,1.41964722846899,1.29325100940785,0.72153880625103,0.440580131022748,0.0351917814720056, -0.142353224879252],:scale)
+      x2      = Statsample::Vector.new([-0.866655707911859,-0.367820249977585,0.361486610435,0.857332626245179,0.133438466268095,0.716104533073575,1.77206093023382,-0.10136697295802,-0.777086491435508,-0.204573554913706,0.963353531412233,-1.10103024900542,-0.404372761837392,-0.230226345183469,0.0363730246866971,-0.838265540390497,1.12543549657924,-0.57929175648001,-0.747060244805248,0.58946979365152,-0.531952663697324,1.53338594419818,0.521992029051441,1.41631763288724,0.611402316795129,-0.518355638373296,-0.515192557101107,-0.672697937866108,1.84347042325327,-0.21195540664804,-0.269869371631611,0.296155694010096,-2.18097898069634,-1.21314663927206,1.49193669881581,1.38969280369493,-0.400680808117106,-1.87282814976479,1.82394870451051,0.637864732838274,-0.141155946382493,0.0699950644281617,1.32568550595165,-0.412599258349398,0.14436832227506,-1.16507785388489,-2.16782049922428,0.24318371493798,0.258954871320764,-0.151966534521183],:scale)
+      @y_pois = Statsample::Vector.new([1,2,1,3,3,1,10,1,1,2,15,0,0,2,1,2,18,2,1,1,1,8,18,13,7,1,1,0,26,0,2,2,0,0,25,7,0,0,21,0,0,1,5,0,3,0,0,1,0,0],:scale)
+
+      @df     = Statsample::Dataset.new({"x1"=>x1,"x2"=>x2, "y" => @y_pois})
+
+    end
+
+    it "reports return values as arrays for IRLS" do
+      @glm    = Statsample::GLM.compute @df,:y,:poisson, 
+                  {algorithm: :irls, constant: 1}
+
+      expect_similar_vector(@glm.coefficients,[-0.586359358356708,
+        1.28511323439258,0.32993246633711])
+    end
+
+    it "reports return values as a hash for IRLS" do
+      @glm    = Statsample::GLM.compute @df,:y,:poisson, 
+                  {algorithm: :irls, constant: 1}
+
+      expect_similar_hash(@glm.coefficients(:hash), {:constant => 0.32993246633711,
+        :x1 => -0.586359358356708, :x2 => 1.28511323439258})
+    end
+  end
+
+  context "MLE algorithm" do
+    # TODO: Implement MLE for poisson
+  end
+end

data/spec/probit_spec.rb

@@ -0,0 +1,19 @@
+describe Statsample::GLM::Probit do
+  context "IRLS algorithm" do
+    # TODO : Implement this!
+  end
+
+  context "MLE algorithm" do
+    before do
+      @data_set = Statsample::CSV.read 'spec/data/logistic_mle.csv'
+      @glm      = Statsample::GLM.compute @data_set, :y, :probit, 
+                    {algorithm: :mle, constant: 1}
+    end
+
+    it "reports correct values as an array" do
+      expect_similar_vector(@glm.coefficients,[0.1763,0.4483,-0.2240,-3.0670],0.001)
+
+      expect(@glm.log_likelihood).to be_within(0.0001).of(-38.31559)
+    end
+  end
+end

data/spec/spec_helper.rb

@@ -0,0 +1,50 @@
+require 'rubygems'
+require 'bundler'
+begin
+  Bundler.setup(:default, :development)
+rescue Bundler::BundlerError => e
+  $stderr.puts e.message
+  $stderr.puts "Run `bundle install` to install missing gems"
+  exit e.status_code
+end
+
+require 'rspec'
+
+$LOAD_PATH.unshift(File.join(File.dirname(__FILE__), '..', 'lib'))
+$LOAD_PATH.unshift(File.dirname(__FILE__))
+require 'statsample-glm'
+
+def expect_similar_vector(exp, obs, delta=1e-10,msg=nil)
+  expect(exp.size).to eq(obs.size)
+
+  exp.data_with_nils.each_with_index do |v,i|
+    expect(v).to be_within(delta).of(obs[i])
+  end
+end
+
+def expect_similar_hash(exp, obs, delta=1e-10,msg=nil)
+  expect(exp.size).to eq(obs.size)
+
+  exp.each_key do |k|
+    expect(exp[k]).to be_within(delta).of(obs[k])
+  end
+end
+
+def expect_equal_vector(exp,obs,delta=1e-10,msg=nil)
+  expect(exp.size).to eq(obs.size)
+
+  exp.size.times do |i|
+    expect(exp[i]).to be_within(delta).of(obs[i])
+  end
+end
+
+def expect_equal_matrix(exp,obs,delta=1e-10,msg=nil)
+  expect(exp.row_size).to eq(obs.row_size)
+  expect(exp.column_size).to eq(obs.column_size)
+
+  exp.row_size.times do |i|
+    exp.column_size.times do |j|
+      expect(exp[i,j]).to be_within(delta).of(obs[i,j])    
+    end
+  end
+end

data/statsample-glm.gemspec

@@ -0,0 +1,35 @@
+# coding: utf-8
+$:.unshift File.expand_path("../lib", __FILE__)
+
+require 'statsample-glm/version'
+
+DESCRIPTION = <<MSG
+Statsample-GLM is an extension to Statsample, an advance statistics suite in
+Ruby. This gem includes modules for Regression techniques such as Poisson 
+and Logistic Regression using the IRLS algorithm and Logistic,  
+Probit and Normal Regression using the Newton Raphson algorithm.
+MSG
+
+Gem::Specification.new do |spec|
+  spec.name          = 'statsample-glm'
+  spec.version       = Statsample::GLM::VERSION
+  spec.authors       = ['Ankur Goel', 'Sameer Deshmukh']
+  spec.email         = ['sameer.deshmukh93@gmail.com']
+  spec.summary       = %q{Generalized Linear Models for Statsample}
+  spec.description   = DESCRIPTION
+  spec.homepage      = "http://github.com/sciruby/statsample-glm"
+  spec.license       = 'BSD-2'
+
+  spec.files         = `git ls-files -z`.split("\x0")
+  spec.executables   = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
+  spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
+  spec.require_paths = ["lib"]
+
+  spec.add_runtime_dependency 'statsample', '~> 1.2'
+
+  spec.add_development_dependency 'bundler'
+  spec.add_development_dependency 'rake'
+  spec.add_development_dependency 'rspec'
+  spec.add_development_dependency 'awesome_print'
+  spec.add_development_dependency 'rdoc', '~> 3.12'
+end

metadata

@@ -1,164 +1,89 @@
 --- !ruby/object:Gem::Specification
 name: statsample-glm
 version: !ruby/object:Gem::Version
-  version: 0.0.1
-  prerelease: 
+  version: 0.1.0
 platform: ruby
 authors:
 - Ankur Goel
+- Sameer Deshmukh
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-09-25 00:00:00.000000000 Z
+date: 2014-10-01 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: statsample
   requirement: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - ! '>='
-      - !ruby/object:Gem::Version
-        version: 1.2.0
-  type: :runtime
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ! '>='
-      - !ruby/object:Gem::Version
-        version: 1.2.0
-- !ruby/object:Gem::Dependency
-  name: activesupport
-  requirement: !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - '='
+    - - ~>
       - !ruby/object:Gem::Version
-        version: 3.2.10
+        version: '1.2'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - '='
+    - - ~>
       - !ruby/object:Gem::Version
-        version: 3.2.10
+        version: '1.2'
 - !ruby/object:Gem::Dependency
-  name: shoulda
+  name: bundler
   requirement: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - ! '>='
+    - - '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - ! '>='
+    - - '>='
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
-  name: rdoc
-  requirement: !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ~>
-      - !ruby/object:Gem::Version
-        version: '3.12'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ~>
-      - !ruby/object:Gem::Version
-        version: '3.12'
-- !ruby/object:Gem::Dependency
-  name: minitest
-  requirement: !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ~>
-      - !ruby/object:Gem::Version
-        version: 4.7.5
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ~>
-      - !ruby/object:Gem::Version
-        version: 4.7.5
-- !ruby/object:Gem::Dependency
-  name: cucumber
+  name: rake
   requirement: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - ! '>='
+    - - '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - ! '>='
+    - - '>='
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
-  name: bundler
+  name: rspec
   requirement: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - ~>
+    - - '>='
       - !ruby/object:Gem::Version
-        version: 1.3.5
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ~>
-      - !ruby/object:Gem::Version
-        version: 1.3.5
-- !ruby/object:Gem::Dependency
-  name: jeweler
-  requirement: !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ~>
-      - !ruby/object:Gem::Version
-        version: 1.8.4
+        version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - ~>
+    - - '>='
       - !ruby/object:Gem::Version
-        version: 1.8.4
+        version: '0'
 - !ruby/object:Gem::Dependency
-  name: bio
+  name: awesome_print
   requirement: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - ! '>='
+    - - '>='
       - !ruby/object:Gem::Version
-        version: 1.4.2
+        version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - ! '>='
+    - - '>='
       - !ruby/object:Gem::Version
-        version: 1.4.2
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: rdoc
   requirement: !ruby/object:Gem::Requirement
-    none: false
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
@@ -166,81 +91,82 @@ dependencies:
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
-    none: false
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
         version: '3.12'
-- !ruby/object:Gem::Dependency
-  name: mocha
-  requirement: !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ~>
-      - !ruby/object:Gem::Version
-        version: 0.14.0
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ~>
-      - !ruby/object:Gem::Version
-        version: 0.14.0
-description: Statsample-GLM is an extension to Statsample, an advance statistics suite
-  in Ruby. This gem includes modules for Regression techniques such as Poisson Regression,
-  Logistic Regression and Exponential Regression
-email: ankurgel@gmail.com
+description: "Statsample-GLM is an extension to Statsample, an advance statistics
+  suite in\nRuby. This gem includes modules for Regression techniques such as Poisson
+  \nand Logistic Regression using the IRLS algorithm and Logistic,  \nProbit and Normal
+  Regression using the Newton Raphson algorithm.\n"
+email:
+- sameer.deshmukh93@gmail.com
 executables: []
 extensions: []
-extra_rdoc_files:
-- LICENSE.txt
-- README.rdoc
+extra_rdoc_files: []
 files:
 - .document
+- .gitignore
+- .rspec
 - .travis.yml
 - Gemfile
 - LICENSE.txt
 - README.rdoc
 - Rakefile
-- VERSION
-- features/bio-statsample-glm.feature
-- features/step_definitions/bio-statsample-glm_steps.rb
-- features/support/env.rb
 - lib/statsample-glm.rb
-- lib/statsample-glm/regression.rb
-- lib/statsample-glm/regression/logistic.rb
-- lib/statsample-glm/regression/poisson.rb
-- test/helper.rb
-- test/test_glm.rb
-- test/test_glm_logistic.rb
-- test/test_glm_poisson.rb
-homepage: http://github.com/AnkurGel/statsample-glm
+- lib/statsample-glm/glm.rb
+- lib/statsample-glm/glm/base.rb
+- lib/statsample-glm/glm/irls/base.rb
+- lib/statsample-glm/glm/irls/logistic.rb
+- lib/statsample-glm/glm/irls/poisson.rb
+- lib/statsample-glm/glm/logistic.rb
+- lib/statsample-glm/glm/mle/base.rb
+- lib/statsample-glm/glm/mle/logistic.rb
+- lib/statsample-glm/glm/mle/normal.rb
+- lib/statsample-glm/glm/mle/probit.rb
+- lib/statsample-glm/glm/normal.rb
+- lib/statsample-glm/glm/poisson.rb
+- lib/statsample-glm/glm/probit.rb
+- lib/statsample-glm/version.rb
+- spec/data/logistic.csv
+- spec/data/logistic_mle.csv
+- spec/data/normal.csv
+- spec/logistic_spec.rb
+- spec/normal_spec.rb
+- spec/poisson_spec.rb
+- spec/probit_spec.rb
+- spec/spec_helper.rb
+- statsample-glm.gemspec
+homepage: http://github.com/sciruby/statsample-glm
 licenses:
-- MIT
+- BSD-2
+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'
-      segments:
-      - 0
-      hash: 891920669
 required_rubygems_version: !ruby/object:Gem::Requirement
-  none: false
   requirements:
-  - - ! '>='
+  - - '>='
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 1.8.25
+rubygems_version: 2.2.2
 signing_key: 
-specification_version: 3
+specification_version: 4
 summary: Generalized Linear Models for Statsample
-test_files: []
+test_files:
+- spec/data/logistic.csv
+- spec/data/logistic_mle.csv
+- spec/data/normal.csv
+- spec/logistic_spec.rb
+- spec/normal_spec.rb
+- spec/poisson_spec.rb
+- spec/probit_spec.rb
+- spec/spec_helper.rb