svmkit 0.7.0 → 0.7.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
-SHA256:
-  metadata.gz: 1e3564c00ae91e7c4cc9db2d19b19df0158803571a18ba9b1446bea22c36b93e
-  data.tar.gz: ebd9a2d55e3935533144bc25498bc9db24d95cfa74fa9b2e8d6299a8b7d54409
+SHA1:
+  metadata.gz: 1f708500e8ba450849736cd02b152b30e302c1bb
+  data.tar.gz: 210256fcd91375e96b4d68015fae2bf2a5b0d4be
 SHA512:
-  metadata.gz: 3fc3e77783fa89bb73b68bd39c21251eae553f8f70259dab31189569b1950f760b72c41b5f839e4d133d156f952995cd2201f2538076435828d19393ea8df007
-  data.tar.gz: 953a07537d01e28b4c00714aebcea67789e654896af7ed5bb18900fcbf3d3fdd0be6d60bf8bfe43932e9e6b0e17fc1ade13aa7c9a9aeb564cbcfd79626979bcc
+  metadata.gz: 4a03201b32dc7a5c0db43bfdfe742cc5b369449c2ce0dc63f4905e3e0b9186276f5b382c1b6bff45acdfb91f61712befc58538c1dfa74874722127bec5e2bf03
+  data.tar.gz: deea7c688685935ebe2574448973903115e727cc479b12ea42f382ea697f7218b0de27cccf978c7ea0f4aa0f0fa0bfd5a97ccbbb88e2ea0b87cd1a5ca00cc5f3

data/.travis.yml

@@ -7,5 +7,6 @@ rvm:
   - 2.3
   - 2.4
   - 2.5
+  - 2.6
 before_install:
-  - gem install --no-document bundler -v '~> 1.16'
+  - gem install --no-document bundler -v '>= 1.17'

data/HISTORY.md

@@ -1,3 +1,7 @@
+# 0.7.1
+- Fix to use CSV class in parsing libsvm format file.
+- Refactor ensemble estimators.
+
 # 0.7.0
 - Add class for AdaBoost classifier.
 - Add class for AdaBoost regressor.

data/lib/svmkit.rb

@@ -4,6 +4,8 @@ require 'numo/narray'
 
 require 'svmkit/version'
 require 'svmkit/validation'
+require 'svmkit/values'
+require 'svmkit/utils'
 require 'svmkit/pairwise_metric'
 require 'svmkit/dataset'
 require 'svmkit/probabilistic_output'

data/lib/svmkit/dataset.rb

@@ -1,5 +1,7 @@
 # frozen_string_literal: true
 
+require 'csv'
+
 module SVMKit
   # Module for loading and saving a dataset file.
   module Dataset
@@ -16,11 +18,11 @@ module SVMKit
         ftvecs = []
         labels = []
         n_features = 0
-        File.read(filename).split("\n").each do |line|
+        CSV.foreach(filename, col_sep: "\s", headers: false) do |line|
           label, ftvec, max_idx = parse_libsvm_line(line, zero_based)
           labels.push(label)
           ftvecs.push(ftvec)
-          n_features = [n_features, max_idx].max
+          n_features = max_idx if n_features < max_idx
         end
         [convert_to_matrix(ftvecs, n_features), Numo::NArray.asarray(labels)]
       end
@@ -48,16 +50,17 @@ module SVMKit
       private
 
       def parse_libsvm_line(line, zero_based)
-        tokens = line.split
-        label = parse_label(tokens.shift)
-        ftvec = tokens.map do |el|
+        label = parse_label(line.shift)
+        adj_idx = zero_based == false ? 1 : 0
+        max_idx = -1
+        ftvec = []
+        while (el = line.shift)
           idx, val = el.split(':')
-          idx = idx.to_i - (zero_based == false ? 1 : 0)
+          idx = idx.to_i - adj_idx
           val = val.to_i.to_s == val ? val.to_i : val.to_f
-          [idx, val]
+          max_idx = idx if max_idx < idx
+          ftvec.push([idx, val])
         end
-        max_idx = ftvec.map { |el| el[0] }.max
-        max_idx ||= 0
         [label, ftvec, max_idx]
       end
 

data/lib/svmkit/ensemble/ada_boost_classifier.rb

@@ -1,6 +1,8 @@
 # frozen_string_literal: true
 
 require 'svmkit/validation'
+require 'svmkit/values'
+require 'svmkit/utils'
 require 'svmkit/base/base_estimator'
 require 'svmkit/base/classifier'
 require 'svmkit/tree/decision_tree_classifier'
@@ -22,6 +24,7 @@ module SVMKit
     class AdaBoostClassifier
       include Base::BaseEstimator
       include Base::Classifier
+      include Validation
 
       # Return the set of estimators.
       # @return [Array<DecisionTreeClassifier>]
@@ -52,15 +55,16 @@ module SVMKit
       #   If nil is given, split process considers all features.
       # @param random_seed [Integer] The seed value using to initialize the random generator.
       #   It is used to randomly determine the order of features when deciding spliting point.
-      def initialize(n_estimators: 50, criterion: 'gini', max_depth: nil, max_leaf_nodes: nil, min_samples_leaf: 1,
+      def initialize(n_estimators: 50,
+                     criterion: 'gini', max_depth: nil, max_leaf_nodes: nil, min_samples_leaf: 1,
                      max_features: nil, random_seed: nil)
-        SVMKit::Validation.check_params_type_or_nil(Integer, max_depth: max_depth, max_leaf_nodes: max_leaf_nodes,
-                                                             max_features: max_features, random_seed: random_seed)
-        SVMKit::Validation.check_params_integer(n_estimators: n_estimators, min_samples_leaf: min_samples_leaf)
-        SVMKit::Validation.check_params_string(criterion: criterion)
-        SVMKit::Validation.check_params_positive(n_estimators: n_estimators, max_depth: max_depth,
-                                                 max_leaf_nodes: max_leaf_nodes, min_samples_leaf: min_samples_leaf,
-                                                 max_features: max_features)
+        check_params_type_or_nil(Integer, max_depth: max_depth, max_leaf_nodes: max_leaf_nodes,
+                                          max_features: max_features, random_seed: random_seed)
+        check_params_integer(n_estimators: n_estimators, min_samples_leaf: min_samples_leaf)
+        check_params_string(criterion: criterion)
+        check_params_positive(n_estimators: n_estimators, max_depth: max_depth,
+                              max_leaf_nodes: max_leaf_nodes, min_samples_leaf: min_samples_leaf,
+                              max_features: max_features)
         @params = {}
         @params[:n_estimators] = n_estimators
         @params[:criterion] = criterion
@@ -82,9 +86,9 @@ module SVMKit
       # @param y [Numo::Int32] (shape: [n_samples]) The labels to be used for fitting the model.
       # @return [AdaBoostClassifier] The learned classifier itself.
       def fit(x, y) # rubocop:disable Metrics/AbcSize
-        SVMKit::Validation.check_sample_array(x)
-        SVMKit::Validation.check_label_array(y)
-        SVMKit::Validation.check_sample_label_size(x, y)
+        check_sample_array(x)
+        check_label_array(y)
+        check_sample_label_size(x, y)
         ## Initialize some variables.
         n_samples, n_features = x.shape
         @estimators = []
@@ -100,12 +104,12 @@ module SVMKit
         observation_weights = Numo::DFloat.zeros(n_samples) + 1.fdiv(n_samples)
         @params[:n_estimators].times do |_t|
           # Fit classfier.
-          ids = weighted_sampling(observation_weights)
+          ids = SVMKit::Utils.choice_ids(n_samples, observation_weights, @rng)
           break if y[ids].to_a.uniq.size != n_classes
           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: @rng.rand(int_max)
+            max_features: @params[:max_features], random_seed: @rng.rand(SVMKit::Values::int_max)
           )
           tree.fit(x[ids, true], y[ids])
           # Calculate estimator error.
@@ -134,7 +138,7 @@ module SVMKit
       # @param x [Numo::DFloat] (shape: [n_samples, n_features]) The samples to compute the scores.
       # @return [Numo::DFloat] (shape: [n_samples, n_classes]) Confidence score per sample.
       def decision_function(x)
-        SVMKit::Validation.check_sample_array(x)
+        check_sample_array(x)
         n_samples, = x.shape
         n_classes = @classes.size
         sum_probs = Numo::DFloat.zeros(n_samples, n_classes)
@@ -150,7 +154,7 @@ module SVMKit
       # @param x [Numo::DFloat] (shape: [n_samples, n_features]) The samples to predict the labels.
       # @return [Numo::Int32] (shape: [n_samples]) Predicted class label per sample.
       def predict(x)
-        SVMKit::Validation.check_sample_array(x)
+        check_sample_array(x)
         n_samples, = x.shape
         probs = decision_function(x)
         Numo::Int32.asarray(Array.new(n_samples) { |n| @classes[probs[n, true].max_index] })
@@ -161,7 +165,7 @@ module SVMKit
       # @param x [Numo::DFloat] (shape: [n_samples, n_features]) The samples to predict the probailities.
       # @return [Numo::DFloat] (shape: [n_samples, n_classes]) Predicted probability of each class per sample.
       def predict_proba(x)
-        SVMKit::Validation.check_sample_array(x)
+        check_sample_array(x)
         n_classes = @classes.size
         probs = Numo::NMath.exp(1.fdiv(n_classes - 1) * decision_function(x))
         sum_probs = probs.sum(1)
@@ -172,8 +176,11 @@ module SVMKit
       # Dump marshal data.
       # @return [Hash] The marshal data about AdaBoostClassifier.
       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.
@@ -186,27 +193,6 @@ module SVMKit
         @rng = obj[:rng]
         nil
       end
-
-      private
-
-      def weighted_sampling(weights)
-        Array.new(weights.size) do
-          target = @rng.rand
-          chosen = 0
-          weights.each_with_index do |w, idx|
-            if target <= w
-              chosen = idx
-              break
-            end
-            target -= w
-          end
-          chosen
-        end
-      end
-
-      def int_max
-        @int_max ||= 2**([42].pack('i').size * 16 - 2) - 1
-      end
     end
   end
 end

data/lib/svmkit/ensemble/ada_boost_regressor.rb

@@ -1,6 +1,7 @@
 # frozen_string_literal: true
 
 require 'svmkit/validation'
+require 'svmkit/values'
 require 'svmkit/base/base_estimator'
 require 'svmkit/base/regressor'
 require 'svmkit/tree/decision_tree_regressor'
@@ -106,11 +107,11 @@ module SVMKit
         # Construct forest.
         @params[:n_estimators].times do |_t|
           # Fit weak learner.
-          ids = weighted_sampling(observation_weights)
+          ids = SVMKit::Utils.choice_ids(n_samples, observation_weights, @rng)
           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: @rng.rand(int_max)
+            max_features: @params[:max_features], random_seed: @rng.rand(SVMKit::Values::int_max)
           )
           tree.fit(x[ids, true], y[ids])
           p = tree.predict(x)
@@ -174,27 +175,6 @@ module SVMKit
         @rng = obj[:rng]
         nil
       end
-
-      private
-
-      def weighted_sampling(weights)
-        Array.new(weights.size) do
-          target = @rng.rand
-          chosen = 0
-          weights.each_with_index do |w, idx|
-            if target <= w
-              chosen = idx
-              break
-            end
-            target -= w
-          end
-          chosen
-        end
-      end
-
-      def int_max
-        @int_max ||= 2**([42].pack('i').size * 16 - 2) - 1
-      end
     end
   end
 end

data/lib/svmkit/ensemble/random_forest_classifier.rb

@@ -1,6 +1,7 @@
 # frozen_string_literal: true
 
 require 'svmkit/validation'
+require 'svmkit/values'
 require 'svmkit/base/base_estimator'
 require 'svmkit/base/classifier'
 require 'svmkit/tree/decision_tree_classifier'
@@ -20,6 +21,7 @@ module SVMKit
     class RandomForestClassifier
       include Base::BaseEstimator
       include Base::Classifier
+      include Validation
 
       # Return the set of estimators.
       # @return [Array<DecisionTreeClassifier>]
@@ -50,15 +52,16 @@ module SVMKit
       #   If nil is given, split process considers all features.
       # @param random_seed [Integer] The seed value using to initialize the random generator.
       #   It is used to randomly determine the order of features when deciding spliting point.
-      def initialize(n_estimators: 10, criterion: 'gini', max_depth: nil, max_leaf_nodes: nil, min_samples_leaf: 1,
+      def initialize(n_estimators: 10,
+                     criterion: 'gini', max_depth: nil, max_leaf_nodes: nil, min_samples_leaf: 1,
                      max_features: nil, random_seed: nil)
-        SVMKit::Validation.check_params_type_or_nil(Integer, max_depth: max_depth, max_leaf_nodes: max_leaf_nodes,
-                                                             max_features: max_features, random_seed: random_seed)
-        SVMKit::Validation.check_params_integer(n_estimators: n_estimators, min_samples_leaf: min_samples_leaf)
-        SVMKit::Validation.check_params_string(criterion: criterion)
-        SVMKit::Validation.check_params_positive(n_estimators: n_estimators, max_depth: max_depth,
-                                                 max_leaf_nodes: max_leaf_nodes, min_samples_leaf: min_samples_leaf,
-                                                 max_features: max_features)
+        check_params_type_or_nil(Integer, max_depth: max_depth, max_leaf_nodes: max_leaf_nodes,
+                                          max_features: max_features, random_seed: random_seed)
+        check_params_integer(n_estimators: n_estimators, min_samples_leaf: min_samples_leaf)
+        check_params_string(criterion: criterion)
+        check_params_positive(n_estimators: n_estimators, max_depth: max_depth,
+                              max_leaf_nodes: max_leaf_nodes, min_samples_leaf: min_samples_leaf,
+                              max_features: max_features)
         @params = {}
         @params[:n_estimators] = n_estimators
         @params[:criterion] = criterion
@@ -80,9 +83,9 @@ module SVMKit
       # @param y [Numo::Int32] (shape: [n_samples]) The labels to be used for fitting the model.
       # @return [RandomForestClassifier] The learned classifier itself.
       def fit(x, y)
-        SVMKit::Validation.check_sample_array(x)
-        SVMKit::Validation.check_label_array(y)
-        SVMKit::Validation.check_sample_label_size(x, y)
+        check_sample_array(x)
+        check_label_array(y)
+        check_sample_label_size(x, y)
         # Initialize some variables.
         n_samples, n_features = x.shape
         @params[:max_features] = Math.sqrt(n_features).to_i unless @params[:max_features].is_a?(Integer)
@@ -94,7 +97,7 @@ module SVMKit
           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: @rng.rand(int_max)
+            max_features: @params[:max_features], random_seed: @rng.rand(SVMKit::Values::int_max)
           )
           bootstrap_ids = Array.new(n_samples) { @rng.rand(0...n_samples) }
           tree.fit(x[bootstrap_ids, true], y[bootstrap_ids])
@@ -110,7 +113,7 @@ module SVMKit
       # @param x [Numo::DFloat] (shape: [n_samples, n_features]) The samples to predict the labels.
       # @return [Numo::Int32] (shape: [n_samples]) Predicted class label per sample.
       def predict(x)
-        SVMKit::Validation.check_sample_array(x)
+        check_sample_array(x)
         n_samples, = x.shape
         n_classes = @classes.size
         classes_arr = @classes.to_a
@@ -130,7 +133,7 @@ module SVMKit
       # @param x [Numo::DFloat] (shape: [n_samples, n_features]) The samples to predict the probailities.
       # @return [Numo::DFloat] (shape: [n_samples, n_classes]) Predicted probability of each class per sample.
       def predict_proba(x)
-        SVMKit::Validation.check_sample_array(x)
+        check_sample_array(x)
         n_samples, = x.shape
         n_classes = @classes.size
         classes_arr = @classes.to_a
@@ -150,7 +153,7 @@ module SVMKit
       # @param x [Numo::DFloat] (shape: [n_samples, n_features]) The samples to predict the labels.
       # @return [Numo::Int32] (shape: [n_samples, n_estimators]) Leaf index for sample.
       def apply(x)
-        SVMKit::Validation.check_sample_array(x)
+        check_sample_array(x)
         Numo::Int32[*Array.new(@params[:n_estimators]) { |n| @estimators[n].apply(x) }].transpose
       end
 
@@ -174,12 +177,6 @@ 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

@@ -1,6 +1,7 @@
 # frozen_string_literal: true
 
 require 'svmkit/validation'
+require 'svmkit/values'
 require 'svmkit/base/base_estimator'
 require 'svmkit/base/regressor'
 require 'svmkit/tree/decision_tree_regressor'
@@ -46,7 +47,8 @@ module SVMKit
       #   If nil is given, split process considers all features.
       # @param random_seed [Integer] The seed value using to initialize the random generator.
       #   It is used to randomly determine the order of features when deciding spliting point.
-      def initialize(n_estimators: 10, criterion: 'mse', max_depth: nil, max_leaf_nodes: nil, min_samples_leaf: 1,
+      def initialize(n_estimators: 10,
+                     criterion: 'mse', max_depth: nil, max_leaf_nodes: nil, min_samples_leaf: 1,
                      max_features: nil, random_seed: nil)
         check_params_type_or_nil(Integer, max_depth: max_depth, max_leaf_nodes: max_leaf_nodes,
                                           max_features: max_features, random_seed: random_seed)
@@ -89,7 +91,7 @@ module SVMKit
           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: @rng.rand(int_max)
+            max_features: @params[:max_features], random_seed: @rng.rand(SVMKit::Values::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])
@@ -136,12 +138,6 @@ 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/utils.rb

@@ -0,0 +1,22 @@
+# frozen_string_literal: true
+
+module SVMKit
+  # @!visibility private
+  module Utils
+    module_function
+
+    # @!visibility private
+    def choice_ids(size, probs, rng=nil)
+      rng ||= Random.new
+      Array.new(size) do
+        target = rng.rand
+        chosen = 0
+        probs.each_with_index do |p, idx|
+          break (chosen = idx) if target <= p
+          target -= p
+        end
+        chosen
+      end
+    end
+  end
+end

data/lib/svmkit/values.rb

@@ -0,0 +1,13 @@
+# frozen_string_literal: true
+
+module SVMKit
+  # @!visibility private
+  module Values
+    module_function
+
+    # @!visibility private
+    def int_max
+      @int_max ||= 2**([42].pack('i').size * 16 - 2) - 1
+    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 = '0.7.0'.freeze
+  VERSION = '0.7.1'.freeze
 end

data/svmkit.gemspec

@@ -35,7 +35,7 @@ MSG
 
   spec.add_runtime_dependency 'numo-narray', '>= 0.9.1'
 
-  spec.add_development_dependency 'bundler', '~> 1.16'
+  spec.add_development_dependency 'bundler', '>= 1.16'
   spec.add_development_dependency 'coveralls', '~> 0.8'
   spec.add_development_dependency 'rake', '~> 12.0'
   spec.add_development_dependency 'rspec', '~> 3.0'

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: svmkit
 version: !ruby/object:Gem::Version
-  version: 0.7.0
+  version: 0.7.1
 platform: ruby
 authors:
 - yoshoku
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2018-12-02 00:00:00.000000000 Z
+date: 2018-12-17 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: numo-narray
@@ -28,14 +28,14 @@ dependencies:
   name: bundler
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
         version: '1.16'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
         version: '1.16'
 - !ruby/object:Gem::Dependency
@@ -168,7 +168,9 @@ files:
 - lib/svmkit/tree/decision_tree_classifier.rb
 - lib/svmkit/tree/decision_tree_regressor.rb
 - lib/svmkit/tree/node.rb
+- lib/svmkit/utils.rb
 - lib/svmkit/validation.rb
+- lib/svmkit/values.rb
 - lib/svmkit/version.rb
 - svmkit.gemspec
 homepage: https://github.com/yoshoku/svmkit
@@ -191,7 +193,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.7.6
+rubygems_version: 2.5.2.3
 signing_key: 
 specification_version: 4
 summary: SVMKit is a machine learninig library in Ruby. SVMKit provides machine learning