nimbus 2.0.1 → 2.1.0

data/README.md

@@ -45,6 +45,12 @@ Nimbus can be used both with regression and classification problems.
 * The split of nodes uses the Gini index as loss function. 
 * Labeling of nodes is made finding the majority fenotype class of the individuals in the node.
 
+## Variable importances
+
+By default Nimbus will estimate SNP importances everytime a training file is run to create a forest.
+
+You can disable this behaviour (and speed up the training process) by setting the parameter `var_importances: No` in the configuration file.
+
 ## Install
 
 You need to have Ruby (1.9.2 or higher) and Rubygems installed in your computer. Then install nimbus with:
@@ -98,6 +104,7 @@ Under the forest chapter:
  * `SNP_sample_size_mtry`: size of the random sample of SNPs to be used in every tree node.
  * `SNP_total_count`: total count of SNPs in the training and/or testing files
  * `node_min_size`: minimum amount of individuals in a tree node to make a split.
+ * `var_importances`: **optional**. If set to `No` Nimbus will not calculate SNP importances.
 
 
 ## Input files
@@ -126,7 +133,7 @@ After training:
  * `random_forest.yml`: A file defining the structure of the computed Random Forest. It can be used as input forest file.
  * `generalization_errors.txt`: A file with the generalization error for every tree in the forest.
  * `training_file_predictions.txt`: A file with predictions for every individual from the training file.
- * `snp_importances.txt`: A file with the computed importance for every SNP.
+ * `snp_importances.txt`: A file with the computed importance for every SNP. _(unless `var_importances` set to `No` in config file)_
 
 After testing:
 

data/lib/nimbus/application.rb

@@ -34,7 +34,7 @@ module Nimbus
           output_random_forest_file(@forest)
           output_tree_errors_file(@forest)
           output_training_file_predictions(@forest)
-          output_snp_importances_file(@forest)
+          output_snp_importances_file(@forest) if @config.do_importances
         end
 
         if @config.do_testing

data/lib/nimbus/configuration.rb

@@ -25,6 +25,7 @@ module Nimbus
       :loss_function_continuous,
       :do_training,
       :do_testing,
+      :do_importances,
       :training_set,
       :output_forest_file,
       :output_training_file,
@@ -63,6 +64,7 @@ module Nimbus
     def initialize
       @do_training = false
       @do_testing  = false
+      @do_importances = true
 
       @forest_size              = DEFAULTS[:forest_size]
       @tree_SNP_sample_size     = DEFAULTS[:tree_SNP_sample_size]
@@ -137,6 +139,8 @@ module Nimbus
         @tree_SNP_total_count = user_config_params['forest']['SNP_total_count'].to_i if user_config_params['forest']['SNP_total_count']
         @tree_SNP_sample_size = user_config_params['forest']['SNP_sample_size_mtry'].to_i  if user_config_params['forest']['SNP_sample_size_mtry']
         @tree_node_min_size   = user_config_params['forest']['node_min_size'].to_i if user_config_params['forest']['node_min_size']
+        @do_importances       = user_config_params['forest']['var_importances'].to_s.strip.downcase
+        @do_importances       = (@do_importances != 'no' && @do_importances != 'false')
       end
 
       check_configuration
@@ -170,7 +174,7 @@ module Nimbus
           next if line.strip == ''
           data_id, *snp_list = line.strip.split
           raise Nimbus::InputFileError, "There are individuals with no ID, please check data in Testing file." unless (!data_id.nil? && data_id.strip != '')
-          raise Nimbus::InputFileError, "Individual ##{data_id} from testing set has no value for all #{@tree_SNP_total_count} SNPs, please check structure of the testing file." unless snp_list.size == @tree_SNP_total_count
+          raise Nimbus::InputFileError, "Individual ##{data_id} from testing set has no value for all #{@tree_SNP_total_count} SNPs." unless snp_list.size == @tree_SNP_total_count
           individual_test = Nimbus::Individual.new(data_id.to_i, nil, snp_list.map{|snp| snp.to_i})
           yield individual_test
         end

data/lib/nimbus/forest.rb

@@ -42,11 +42,11 @@ module Nimbus
         tree = tree_class.new @options.tree
         @trees << tree.seed(@options.training_set.individuals, tree_individuals_bag, @options.training_set.ids_fenotypes)
         @tree_errors << tree.generalization_error_from_oob(tree_out_of_bag)
-        @tree_snp_importances << tree.estimate_importances(tree_out_of_bag)
+        @tree_snp_importances << tree.estimate_importances(tree_out_of_bag) if @options.do_importances
         acumulate_predictions tree.predictions
         Nimbus.clear_line!
       end
-      average_snp_importances
+      average_snp_importances if @options.do_importances
       totalize_predictions
     end
 

data/lib/nimbus/version.rb

@@ -1,3 +1,3 @@
 module Nimbus
-  VERSION = "2.0.1"
+  VERSION = "2.1.0"
 end

data/spec/configuration_spec.rb

@@ -11,6 +11,7 @@ describe Nimbus::Configuration do
     config.testing_file.should == fixture_file('regression_testing.data')
     config.forest_file.should == fixture_file('regression_random_forest.yml')
     config.classes.should be_nil
+    config.do_importances.should be
 
     config.forest_size.should == 3
     config.tree_SNP_sample_size.should == 60
@@ -24,6 +25,7 @@ describe Nimbus::Configuration do
     config.testing_file.should == fixture_file('classification_testing.data')
     config.forest_file.should == fixture_file('classification_random_forest.yml')
     config.classes.should == ['0','1']
+    config.do_importances.should_not be
 
     config.forest_size.should == 3
     config.tree_SNP_sample_size.should == 33

data/spec/fixtures/classification_config.yml

@@ -10,4 +10,5 @@ forest:
   forest_size: 3 #how many trees
   SNP_sample_size_mtry: 33 #mtry
   SNP_total_count: 100
-  node_min_size: 5
+  node_min_size: 5
+  var_importances: NO

data/spec/forest_spec.rb

@@ -34,6 +34,13 @@ describe Nimbus::Forest do
       @forest.snp_importances.values.each{|v| v.should be_kind_of Numeric}
     end
 
+    it 'does not compute SNP importances if config set to false' do
+      @forest.snp_importances.should == {}
+      @forest.options.do_importances = false
+      @forest.grow
+      @forest.snp_importances.should == {}
+    end
+
     it 'traverses a set of testing individuals through every tree in the forest and returns predictions' do
       @forest = @config.load_forest
       @forest.predictions.should == {}
@@ -85,11 +92,19 @@ describe Nimbus::Forest do
 
     it 'computes averaged SNP importances for every SNP' do
       @forest.snp_importances.should == {}
+      @forest.options.do_importances = true
       @forest.grow
       @forest.snp_importances.keys.sort.should == (1..100).to_a # 100 snps in the training file
       @forest.snp_importances.values.each{|v| v.should be_kind_of Numeric}
     end
 
+    it 'does not compute SNP importances if config set to false' do
+      @forest.snp_importances.should == {}
+      @forest.options.do_importances = false
+      @forest.grow
+      @forest.snp_importances.should == {}
+    end
+
     it 'traverses a set of testing individuals through every tree in the forest and returns predictions' do
       @forest = @config.load_forest
       @forest.predictions.should == {}

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: nimbus
 version: !ruby/object:Gem::Version
-  version: 2.0.1
+  version: 2.1.0
   prerelease: 
 platform: ruby
 authors:
@@ -10,11 +10,11 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-07-27 00:00:00.000000000 Z
+date: 2012-07-28 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rspec
-  requirement: &2160657580 !ruby/object:Gem::Requirement
+  requirement: &2152757020 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -22,7 +22,7 @@ dependencies:
         version: 2.11.0
   type: :development
   prerelease: false
-  version_requirements: *2160657580
+  version_requirements: *2152757020
 description: Nimbus is a Ruby gem to implement Random Forest in a genomic selection
   context.
 email: