bayesnet 0.0.1 → 0.0.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 9501e7b169bda69caf717e26a63ba141fbfcb413607f8888cd62a66eaf3ab78c
-  data.tar.gz: 4ca8a547e3f71d4b5690a73ae4430700a7115f97bbee974974da094701e192b3
+  metadata.gz: 9f17178cdfa472aea14e7769f1831f1903ceee2613b6e7b71e1135990dc974ab
+  data.tar.gz: c55cb95134d4dbb479e89839be7a4634b6bc6a8be52c5c066f1cc1ad21c584a7
 SHA512:
-  metadata.gz: 11c61de5da370ee1a5b75014c1d74644d8b38cb45a5daa81e2292efac233beaf35ee85059074b09296d7d3ce842e67353921fbcf54f92109c5ace3b09459159a
-  data.tar.gz: 9abb2ee0bcd8661546b6ca3e27f9a3d3f96a2679c5b6b60e06ea7e26b759542f043f39a8fa7dc1de8270032b9485fc531b23a0895ca2001119a357de13797216
+  metadata.gz: fda426d9dd319b2ad8300a8ba59acae511740cc271729f40b36ce7a7765d8019a901d9d83a2e955ccd964901ee604a3ad96fba092ce41394da5980a77d1bb198
+  data.tar.gz: 03d4beb66b11ba684007dd49f67ef2ae085ecd5c6e5ca9b8b63327eee109ada3b1f389731543fa6d970bff110230231bc1533022640367a906f57dec1c41bc6d

data/.ruby-version

@@ -1 +1 @@
-3.0.3
+3.1.0

data/CHANGELOG.md

@@ -1,5 +1,10 @@
 ## [Unreleased]
 
-## [0.1.0] - 2021-12-18
+## [0.0.2] - 2021-12-28
+
+- README, CI/CD for Ruby 2.6, 2.7, 3.1 added
+
+## [0.0.1] - 2021-12-28
 
 - Initial release
+

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    bayesnet (0.0.1)
+    bayesnet (0.0.2)
 
 GEM
   remote: https://rubygems.org/

data/README.md

@@ -1,8 +1,70 @@
 # Bayesnet
 
-Welcome to your new gem! In this directory, you'll find the files you need to be able to package up your Ruby library into a gem. Put your Ruby code in the file `lib/bayesnet`. To experiment with that code, run `bin/console` for an interactive prompt.
+This gem provides an API for building a Bayesian network and executing some queries against it.
 
-TODO: Delete this and the text above, and describe your gem
+### Example:
+
+Someone has decided to study how some sleep hours followed by a cup of coffee are affecting a morning mood. 
+Here is the graphical model:
+
+![model](./doc/morning-mood-model.png "Morning Mood Model")
+
+`Sleep hours` could have values `:six, :seven, :eight`. `Mood` could be `:bad, :moderate, :good` and variable 
+`Coffee` could take values `:yes, :no`. Here is how accumulated statistics could be compiled in a Bayesian network and then defined in the code:
+
+```
+net = Bayesnet.define do
+      node :coffee do
+        values yes: 0.7, no: 0.3
+      end
+
+      node :sleep_hours do
+        values six: 0.1, seven: 0.3, eight: 0.6
+      end
+
+      node :mood, parents: [:coffee, :sleep_hours] do
+        values [:good, :moderate, :bad] do
+          distributions do
+            as [0.8, 0.1, 0.1], given: [:yes, :eight]
+            as [0.6, 0.2, 0.2], given: [:yes, :seven]
+            as [0.4, 0.4, 0.2], given: [:yes, :six]
+            as [0.7, 0.2, 0.1], given: [:no, :eight]
+            as [0.5, 0.3, 0.2], given: [:no, :seven]
+            as [0.3, 0.4, 0.3], given: [:no, :six]
+          end
+        end
+      end
+    end
+```
+
+Above, must be read as:
+someone could be observed having morning coffee in 70% of all cases and when that person has a morning coffee after sleeping 8 hours, chances to find her in
+  - `:good` mood - 80%
+  - `:moderate` mood - 10%
+  - `:bad` mood - 10%
+  
+i.e. `[0.8, 0.1, 0.1]` is a conditional distribution.
+
+Given the model above the following queries could be executed:
+
+#### Given sleep time is six hours and mood is good, what chances cup of coffee has been consumed?
+```
+net.chances({coffee: :yes}, evidence: {mood: :good, sleep_hours: :six}) # 0.757
+```
+
+#### Given sleep time is six hours and mood is good, what is the most likely value for a `:coffee` variable?
+```
+net.most_likely_value(:coffee, evidence: {mood: :good, sleep_hours: :six}) # :yes
+```
+
+#### A broader question than the two above: Given sleep time is six hours and mood is good, what is the distribution for a `:coffee` variable?
+```
+net.distribution(over: [:coffee], evidence: {mood: :good, sleep_hours: :six}) # [:yes, 0.757], [:no, 0.243]
+```
+
+The inference is based on summing over joint distribution, i.e. it is the simplest and
+most expensive way to calculate it. No optimization is implemented in this version; the code
+is more a proof of API.
 
 ## Installation
 
@@ -20,9 +82,6 @@ Or install it yourself as:
 
     $ gem install bayesnet
 
-## Usage
-
-TODO: Write usage instructions here
 
 ## Development
 

data/bayesnet.gemspec

@@ -15,7 +15,7 @@ Gem::Specification.new do |spec|
 
   spec.metadata["homepage_uri"] = spec.homepage
   spec.metadata["source_code_uri"] = spec.homepage
-  spec.metadata["changelog_uri"] = "#{spec.homepage}/blob/main/CHANGELOG.md"
+  spec.metadata["changelog_uri"] = "#{spec.homepage}/blob/master/CHANGELOG.md"
 
   # Specify which files should be added to the gem when it is released.
   # The `git ls-files -z` loads the files in the RubyGem that have been added into git.

data/lib/bayesnet/factor.rb

@@ -15,7 +15,7 @@ module Bayesnet
     # Specifies function values for args. Latest args is an function value, all previous are argument values
     def val(*args)
       args = args[0] if args.size == 1 && args[0].is_a?(Array)
-      @vals[args[..-2]] = args[-1]
+      @vals[args[0..-2]] = args[-1]
     end
 
     def var_names
@@ -52,7 +52,9 @@ module Bayesnet
     end
 
     def limit_by(evidence)
-      vars = @vars.except(*evidence.keys)
+      # todo: use Hash#except when Ruby 2.6 support no longer needed
+      evidence_keys_set = evidence.keys.to_set
+      vars = @vars.reject { |k, _| evidence_keys_set.include?(k) }
 
       evidence_vals = evidence.values
       indices = evidence.keys.map { |k| index_by_var_name[k] }

data/lib/bayesnet/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Bayesnet
-  VERSION = "0.0.1"
+  VERSION = "0.0.2"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: bayesnet
 version: !ruby/object:Gem::Version
-  version: 0.0.1
+  version: 0.0.2
 platform: ruby
 authors:
 - Aleksandr Furmanov
@@ -85,6 +85,7 @@ files:
 - bayesnet.gemspec
 - bin/console
 - bin/setup
+- doc/morning-mood-model.png
 - lib/bayesnet.rb
 - lib/bayesnet/dsl.rb
 - lib/bayesnet/error.rb
@@ -98,7 +99,7 @@ licenses:
 metadata:
   homepage_uri: https://github.com/afurmanov/bayesnet
   source_code_uri: https://github.com/afurmanov/bayesnet
-  changelog_uri: https://github.com/afurmanov/bayesnet/blob/main/CHANGELOG.md
+  changelog_uri: https://github.com/afurmanov/bayesnet/blob/master/CHANGELOG.md
 post_install_message:
 rdoc_options: []
 require_paths:
@@ -114,7 +115,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.2.32
+rubygems_version: 3.3.3
 signing_key:
 specification_version: 4
 summary: Byesian network DSL + inference API