svmkit 0.6.2 → 0.6.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 9655d7d990f37468c79de9713e55f74c6134b00c1cda9471832097c678cb6ded
-  data.tar.gz: fb294e6256d16272e80c2ade2f5b223dd792b14c7fd39caf1de8a9500a3ab55e
+  metadata.gz: 8f6c3a9b6df704497579a44ed1649baeff28364d8992cc781babb03ee1fe836a
+  data.tar.gz: 99502858d1cecbe65efe5ee342cc6719278f227b1305c75d5305d3362d00e295
 SHA512:
-  metadata.gz: f7386071fe57df51bd8223d4945dc67069769464892bd64b972b5c1ba26cd206b7b67d50e600f34d79a3bff9f19803c0fdae06dd92fdf8f6ef87f1d5e982cf2d
-  data.tar.gz: 473a1233e0109672b80b8bf17933366276b0a81fab3c75699fb1a07f92923d66cc3f064e0a2df36e493bf1554aa4c34d567d60607d1802e10ab949eafe1187d3
+  metadata.gz: 86bf9a6ce8de82c7b51c9f7ff3451c9834ca73a84abb48394769e65bf80ebf8b00c565f27b69882719ad81c5b33a27af0156cc1b3cc2949f530eeacf11560718
+  data.tar.gz: 14a4e5a9becd0b41f43ba1a4cdee545f12f8d18f339feae789593c527025a9c6c0bb50d7471c5ffefffbe1ba96dc46e409aaa25e781ef52830adac476531ce58

data/HISTORY.md

@@ -1,3 +1,6 @@
+# 0.6.3
+- Fix bug on setting random seed and max_features parameter of Random Forest estimators.
+
 # 0.6.2
 - Refactor decision tree classes for improving performance.
 

data/lib/svmkit/ensemble/random_forest_classifier.rb

@@ -85,22 +85,22 @@ module SVMKit
         SVMKit::Validation.check_sample_label_size(x, y)
         # Initialize some variables.
         n_samples, n_features = x.shape
-        @params[:max_features] = n_features unless @params[:max_features].is_a?(Integer)
-        @params[:max_features] = [[1, @params[:max_features]].max, Math.sqrt(n_features).to_i].min
+        @params[:max_features] = Math.sqrt(n_features).to_i unless @params[:max_features].is_a?(Integer)
+        @params[:max_features] = [[1, @params[:max_features]].max, n_features].min
         @classes = Numo::Int32.asarray(y.to_a.uniq.sort)
+        @feature_importances = Numo::DFloat.zeros(n_features)
         # Construct forest.
-        @estimators = Array.new(@params[:n_estimators]) do |_n|
+        @estimators = Array.new(@params[:n_estimators]) do
           tree = Tree::DecisionTreeClassifier.new(
             criterion: @params[:criterion], max_depth: @params[:max_depth],
             max_leaf_nodes: @params[:max_leaf_nodes], min_samples_leaf: @params[:min_samples_leaf],
-            max_features: @params[:max_features], random_seed: @params[:random_seed]
+            max_features: @params[:max_features], random_seed: @rng.rand(int_max)
           )
           bootstrap_ids = Array.new(n_samples) { @rng.rand(0...n_samples) }
           tree.fit(x[bootstrap_ids, true], y[bootstrap_ids])
+          @feature_importances += tree.feature_importances
+          tree
         end
-        # Calculate feature importances.
-        @feature_importances = Numo::DFloat.zeros(n_features)
-        @estimators.each { |tree| @feature_importances += tree.feature_importances }
         @feature_importances /= @feature_importances.sum
         self
       end
@@ -157,8 +157,11 @@ module SVMKit
       # Dump marshal data.
       # @return [Hash] The marshal data about RandomForestClassifier.
       def marshal_dump
-        { params: @params, estimators: @estimators, classes: @classes,
-          feature_importances: @feature_importances, rng: @rng }
+        { params: @params,
+          estimators: @estimators,
+          classes: @classes,
+          feature_importances: @feature_importances,
+          rng: @rng }
       end
 
       # Load marshal data.
@@ -171,6 +174,12 @@ module SVMKit
         @rng = obj[:rng]
         nil
       end
+
+      private
+
+      def int_max
+        @int_max ||= 2**([42].pack('i').size * 16 - 2) - 1
+      end
     end
   end
 end

data/lib/svmkit/ensemble/random_forest_regressor.rb

@@ -80,21 +80,22 @@ module SVMKit
         check_sample_tvalue_size(x, y)
         # Initialize some variables.
         n_samples, n_features = x.shape
-        @params[:max_features] ||= n_features
-        @params[:max_features] = [[1, @params[:max_features]].max, Math.sqrt(n_features).to_i].min
+        @params[:max_features] = Math.sqrt(n_features).to_i unless @params[:max_features].is_a?(Integer)
+        @params[:max_features] = [[1, @params[:max_features]].max, n_features].min
+        @feature_importances = Numo::DFloat.zeros(n_features)
         single_target = y.shape[1].nil?
         # Construct forest.
-        @estimators = Array.new(@params[:n_estimators]) do |_n|
+        @estimators = Array.new(@params[:n_estimators]) do
           tree = Tree::DecisionTreeRegressor.new(
             criterion: @params[:criterion], max_depth: @params[:max_depth],
             max_leaf_nodes: @params[:max_leaf_nodes], min_samples_leaf: @params[:min_samples_leaf],
-            max_features: @params[:max_features], random_seed: @params[:random_seed]
+            max_features: @params[:max_features], random_seed: @rng.rand(int_max)
           )
           bootstrap_ids = Array.new(n_samples) { @rng.rand(0...n_samples) }
           tree.fit(x[bootstrap_ids, true], single_target ? y[bootstrap_ids] : y[bootstrap_ids, true])
+          @feature_importances += tree.feature_importances
+          tree
         end
-        # Calculate feature importances.
-        @feature_importances = @estimators.map(&:feature_importances).reduce(&:+)
         @feature_importances /= @feature_importances.sum
         self
       end
@@ -135,6 +136,12 @@ module SVMKit
         @rng = obj[:rng]
         nil
       end
+
+      private
+
+      def int_max
+        @int_max ||= 2**([42].pack('i').size * 16 - 2) - 1
+      end
     end
   end
 end

data/lib/svmkit/version.rb

@@ -3,5 +3,5 @@
 # SVMKit is a machine learning library in Ruby.
 module SVMKit
   # @!visibility private
-  VERSION = '0.6.2'.freeze
+  VERSION = '0.6.3'.freeze
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: svmkit
 version: !ruby/object:Gem::Version
-  version: 0.6.2
+  version: 0.6.3
 platform: ruby
 authors:
 - yoshoku
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2018-09-17 00:00:00.000000000 Z
+date: 2018-11-25 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: numo-narray