yannitor 0.5.0 → 0.6.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
-SHA1:
-  metadata.gz: 86f4b0ff2fca0aaa5270a85df9fcf5a802154eed
-  data.tar.gz: 55d3f801ad79c800177ca453e08be692eedaf249
+SHA256:
+  metadata.gz: 3aa7e3ec52c46173c2b250128ae91232b56f5f77128b96076e03a4bb754d7ebc
+  data.tar.gz: 96ba16f09a7d85186d97228758f4a2ff91f996f45204f05f82b6fb1c1b741c4f
 SHA512:
-  metadata.gz: b636f749e8a5bae71eb3cbf953925522279ea8272ee6a1b1b63620617743d878d7202f5182d5486debe33034055f6c544fdbb15e47bbf471be8455eac4ccf998
-  data.tar.gz: 7ac6c13650a2463901647d809773060206e4c6c684f6a731eece17f95ce34acb34d52bfeaecb15e73a81052aee685bf4cc81c11eca8e42e519acab2e5e87b23f
+  metadata.gz: f786eaffca03c775ab53f0931ce45ca1c66896132e9a3d052547d4a71711d32ef044378370b077b4bdfcbeea79f939ee002ff18f8fe821367113de0aa694f7d8
+  data.tar.gz: f87a486ac25649dab0f2e6ffb10e3030061d6e6d96c12e938fa4e2efb0e9f43b2b12720dcb29ccf5befc26cb78d28b5bbe135e60a430eeb84dac970e7f552208

data/lib/yannitor/cleaner.rb

@@ -1,59 +1,64 @@
-require 'active_record'
-# require 'active_record/version'
-# require 'active_support/core_ext/module'
+# frozen_string_literal: true
 
-# begin
-#   require 'rails/engine'
-# end
+require 'active_record'
 
 module Yannitor
   module Broom
-    attr_accessor :features
+    attr_accessor :yannitor_features
 
     def yannitor_is_cleaning(feats = {})
-      self.features = feats
+      self.yannitor_features = feats
     end
 
-    def to_one_hot target_column, type = 'text'
-      sorted_value_array = self.pluck("distinct(#{target_column})")
-
-      _table_name   = self.table_name
-      values_select = %Q(SELECT value FROM (values ('#{ sorted_value_array.join("'), ('") }')) s(value))
-
-      self.select(%Q(
-        #{_table_name}.id,
-        ARRAY_AGG(CASE WHEN sorted_value_table.value::#{type} = #{_table_name}.#{target_column}::#{type} THEN 1 ELSE 0 END) AS o#{target_column}
-      )).joins(%Q(
+    def to_one_hot(target_column, type = 'text')
+      sorted_value_array = pluck("distinct(#{target_column})").join("'), ('")
+
+      table_name = self.table_name
+      values_select = %(
+        SELECT value FROM (values ('#{sorted_value_array}')) s(value)
+      )
+
+      self.select(%(
+        #{table_name}.id,
+        ARRAY_AGG(CASE
+          WHEN sorted_value_table.value::#{type} = #{table_name}.#{target_column}::#{type}
+          THEN 1
+          ELSE 0
+          END
+        ) AS o#{target_column}
+      )).joins(%(
         LEFT JOIN (#{values_select}) AS sorted_value_table ON 1=1
-      )).group("#{_table_name}.id")
+      )).group("#{table_name}.id")
     end
 
     def vectorize
-      _table_name   = self.table_name
-
-      select('*, ' + features[:linear].map do |feature|
-        min = all.minimum(feature)
-        max = all.maximum(feature)
-        "CAST((#{_table_name}.#{feature}::float - #{min}::float) / (#{max}::float - #{min}::float) AS float) as n#{feature}"
-      end.join(', ')).all.map do |obj|
+      select('*, ' + linear_feature_select).build_linear_features
+    end
 
-        obj.class.features[:linear].map do |feature|
+    def build_linear_features
+      all.map do |obj|
+        obj.class.yannitor_features[:linear].map do |feature|
           obj.send("n#{feature}").to_f
         end
       end
     end
 
-    def nelect(feature)
-      _table_name   = self.table_name
+    def linear_feature_select
+      yannitor_features[:linear].map do |feature|
+        min = all.minimum(feature)
+        max = all.maximum(feature)
+        "CAST((#{table_name}.#{feature}::float - #{min}::float) / (#{max}::float - #{min}::float) AS float) as n#{feature}"
+      end.join(', ')
+    end
 
+    def nelect(feature)
       min = all.minimum(feature)
       max = all.maximum(feature)
-      
-      self.select("*, (#{_table_name}.#{feature}::float - #{min}::float) / (#{max}::float - #{min}::float)::float as n#{feature}")
+
+      select("*, (#{table_name}.#{feature}::float - #{min}::float) / (#{max}::float - #{min}::float)::float as n#{feature}")
     end
 
     def normalize(feature)
-      print "Normalizing #{feature}"
       min = all.minimum(feature)
       max = all.maximum(feature)
       data = all.nelect(feature).map do |e|
@@ -63,14 +68,13 @@ module Yannitor
       [data, min, max]
     end
 
-    def to_file
-      CSV.open("data.csv", "wb", {col_sep: ' '}) do |csv|
+    def to_file(file_name = 'data.csv', separator = ' ')
+      CSV.open(file_name, 'wb', col_sep: separator) do |csv|
         all.vectorize.each { |v| csv << v }
       end
 
       nil
     end
-
   end
 end
 

data/lib/yannitor/version.rb

@@ -1,3 +1,3 @@
 module Yannitor
-  VERSION = "0.5.0"
+  VERSION = "0.6.0"
 end

data/lib/yannitor.rb

@@ -1,6 +1,7 @@
-require "yannitor/version"
-require "yannitor/cleaner"
+# frozen_string_literal: true
+
+require 'yannitor/version'
+require 'yannitor/cleaner'
 
 module Yannitor
-  # Your code goes here...
 end

data/yannitor.gemspec

@@ -7,21 +7,12 @@ Gem::Specification.new do |spec|
   spec.name          = "yannitor"
   spec.version       = Yannitor::VERSION
   spec.authors       = ["Danielius Visockas"]
-  spec.email         = ["danielius@wisemonks.com"]
+  spec.email         = ["danieliusvisockas@gmail.com"]
 
-  spec.summary       = %q{Gem for preprocessing Active record collections}
+  spec.summary       = %q{Helps you build one-hot or min-max encoded vectors from ActiveRecord collections}
   spec.description   = %q{I'll clean your data}
   spec.homepage      = "https://github.com"
 
-  # Prevent pushing this gem to RubyGems.org. To allow pushes either set the 'allowed_push_host'
-  # to allow pushing to a single host or delete this section to allow pushing to any host.
-  # if spec.respond_to?(:metadata)
-  #   spec.metadata['allowed_push_host'] = "'http://mygemserver.com'"
-  # else
-  #   raise "RubyGems 2.0 or newer is required to protect against " \
-  #     "public gem pushes."
-  # end
-
   spec.files         = `git ls-files -z`.split("\x0").reject do |f|
     f.match(%r{^(test|spec|features)/})
   end
@@ -29,6 +20,7 @@ Gem::Specification.new do |spec|
   spec.executables   = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
   spec.require_paths = ["lib"]
 
+  spec.add_runtime_dependency "active_record", ["> 3.2.0"]
   spec.add_development_dependency "bundler", "~> 1.14"
   spec.add_development_dependency "rake", "~> 10.0"
   spec.add_development_dependency "rspec", "~> 3.0"

metadata

@@ -1,15 +1,29 @@
 --- !ruby/object:Gem::Specification
 name: yannitor
 version: !ruby/object:Gem::Version
-  version: 0.5.0
+  version: 0.6.0
 platform: ruby
 authors:
 - Danielius Visockas
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2017-11-11 00:00:00.000000000 Z
+date: 2019-04-05 00:00:00.000000000 Z
 dependencies:
+- !ruby/object:Gem::Dependency
+  name: active_record
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">"
+      - !ruby/object:Gem::Version
+        version: 3.2.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">"
+      - !ruby/object:Gem::Version
+        version: 3.2.0
 - !ruby/object:Gem::Dependency
   name: bundler
   requirement: !ruby/object:Gem::Requirement
@@ -54,7 +68,7 @@ dependencies:
         version: '3.0'
 description: I'll clean your data
 email:
-- danielius@wisemonks.com
+- danieliusvisockas@gmail.com
 executables: []
 extensions: []
 extra_rdoc_files: []
@@ -91,8 +105,8 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.6.14
+rubygems_version: 2.7.8
 signing_key: 
 specification_version: 4
-summary: Gem for preprocessing Active record collections
+summary: Helps you build one-hot or min-max encoded vectors from ActiveRecord collections
 test_files: []