svmkit 0.2.2 → 0.2.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: c7c326db290b1847234f890914fe5d670a4b1d36
-  data.tar.gz: ee5a624c92bf6b35edcccf4df469f9f552c2174d
+  metadata.gz: 6271a50754a13199f7c3c12c6f1b9e2a0a2075d5
+  data.tar.gz: ecdc84a2987f22d49ad1b8435397862771f96f37
 SHA512:
-  metadata.gz: 9256fc3d36e6247fae44ac1a14672eaf9b3ba414176b48c592be5aa8631d232dbaddba9f3884198e0ba751616a3017ad461da2fb7ec40ef26b9ab2b2417aadf5
-  data.tar.gz: e198dcbe0c7e782162a31e7131b961f619e76bf509b43b596299412ba5bb5ea16ea02e21bb2a79909d70bb230c81484083df94ec65db6e770e4e5adc712da174
+  metadata.gz: 9974eb62cd19ebca32ca92cafbcbf2e34a978d41c0aa9bb0765c001d56963189aa43f71806fa5a2492a4b3e50bcd939a018aa784198f5aef08e2159682682dd2
+  data.tar.gz: eee0f089449a71f79576165aa083c41fe1d9812fcc251c31c2a5c7cf06e6dfd4293dddd6685fa6285c8d89a3744bb3a72bd9db6f36b7cbc586b7a71800c1fb78

data/.travis.yml

@@ -3,8 +3,9 @@ os: linux
 dist: trusty
 language: ruby
 rvm:
-  - 2.2.9
-  - 2.3.6
-  - 2.4.3
+  - 2.2
+  - 2.3
+  - 2.4
+  - 2.5
 before_install:
   - gem install --no-document bundler -v '~> 1.16'

data/HISTORY.md

@@ -1,5 +1,10 @@
+# 0.2.3
+- Added class for cross validation.
+- Added specs for base modules.
+- Fixed validation of the number of splits when a negative label is given.
+
 # 0.2.2
-- Added classes for K-fold cross validation.
+- Added data splitter classes for K-fold cross validation.
 
 # 0.2.1
 - Added class for K-nearest neighbors classifier.

data/README.md

@@ -66,6 +66,27 @@ transformed = transformer.transform(normalized)
 puts(sprintf("Accuracy: %.1f%%", 100.0 * classifier.score(transformed, labels)))
 ```
 
+5-fold cross-validation:
+
+```ruby
+require 'svmkit'
+
+samples, labels = SVMKit::Dataset.load_libsvm_file('pendigits')
+
+kernel_svc =
+  SVMKit::KernelMachine::KernelSVC.new(reg_param: 1.0, max_iter: 1000, random_seed: 1)
+ovr_kernel_svc = SVMKit::Multiclass::OneVsRestClassifier.new(estimator: kernel_svc)
+
+kf = SVMKit::ModelSelection::StratifiedKFold.new(n_splits: 5, shuffle: true, random_seed: 1)
+cv = SVMKit::ModelSelection::CrossValidation.new(estimator: ovr_kernel_svc, splitter: kf)
+
+kernel_mat = SVMKit::PairwiseMetric::rbf_kernel(samples, nil, 0.005)
+report = cv.perform(kernel_mat, labels)
+
+mean_accuracy = report[:test_score].inject(:+) / kf.n_splits
+puts(sprintf("Mean Accuracy: %.1f%%", 100.0 * mean_accuracy))
+```
+
 ## Development
 
 After checking out the repo, run `bin/setup` to install dependencies. Then, run `rake spec` to run the tests. You can also run `bin/console` for an interactive prompt that will allow you to experiment.

data/lib/svmkit.rb

@@ -19,3 +19,4 @@ require 'svmkit/preprocessing/min_max_scaler'
 require 'svmkit/preprocessing/standard_scaler'
 require 'svmkit/model_selection/k_fold'
 require 'svmkit/model_selection/stratified_k_fold'
+require 'svmkit/model_selection/cross_validation'

data/lib/svmkit/base/splitter.rb

@@ -9,7 +9,7 @@ module SVMKit
 
       # An abstract method for splitting dataset.
       def split
-        raise NoImplementedError, "#{__method__} has to be implemented in #{self.class}."
+        raise NotImplementedError, "#{__method__} has to be implemented in #{self.class}."
       end
     end
   end

data/lib/svmkit/kernel_machine/kernel_svc.rb

@@ -68,7 +68,7 @@ module SVMKit
           weight_vec[target_id] += 1.0 if func < 1.0
         end
         # Store the learned model.
-        @weight_vec = weight_vec * Numo::DFloat.asarray(bin_y)
+        @weight_vec = weight_vec * Numo::DFloat[*bin_y]
         self
       end
 
@@ -78,7 +78,7 @@ module SVMKit
       #     The kernel matrix between testing samples and training samples to compute the scores.
       # @return [Numo::DFloat] (shape: [n_testing_samples]) Confidence score per sample.
       def decision_function(x)
-        @weight_vec.dot(x.transpose)
+        x.dot(@weight_vec)
       end
 
       # Predict class labels for samples.

data/lib/svmkit/linear_model/logistic_regression.rb

@@ -74,7 +74,7 @@ module SVMKit
         end
         # Initialize some variables.
         n_samples, n_features = samples.shape
-        rand_ids = [*0..n_samples - 1].shuffle(random: @rng)
+        rand_ids = [*0...n_samples].shuffle(random: @rng)
         weight_vec = Numo::DFloat.zeros(n_features)
         # Start optimization.
         @params[:max_iter].times do |t|

data/lib/svmkit/linear_model/svc.rb

@@ -70,7 +70,7 @@ module SVMKit
         end
         # Initialize some variables.
         n_samples, n_features = samples.shape
-        rand_ids = [*0..n_samples - 1].shuffle(random: @rng)
+        rand_ids = [*0...n_samples].shuffle(random: @rng)
         weight_vec = Numo::DFloat.zeros(n_features)
         # Start optimization.
         @params[:max_iter].times do |t|

data/lib/svmkit/model_selection/cross_validation.rb

@@ -0,0 +1,82 @@
+require 'svmkit/base/splitter'
+
+module SVMKit
+  # This module consists of the classes for model validation techniques.
+  module ModelSelection
+    # CrossValidation is a class that evaluates a given classifier with cross-validation method.
+    #
+    # @example
+    #   svc = SVMKit::LinearModel::SVC.new
+    #   kf = SVMKit::ModelSelection::StratifiedKFold.new(n_splits: 5)
+    #   cv = SVMKit::ModelSelection::CrossValidation.new(estimator: svc, splitter: kf)
+    #   report = cv.perform(samples, lables)
+    #   mean_test_score = report[:test_score].inject(:+) / kf.n_splits
+    #
+    class CrossValidation
+      # Return the classifier of which performance is evaluated.
+      # @return [Classifier]
+      attr_reader :estimator
+
+      # Return the splitter that divides dataset.
+      # @return [Splitter]
+      attr_reader :splitter
+
+      # Return the flag indicating whether to caculate the score of training dataset.
+      # @return [Boolean]
+      attr_reader :return_train_score
+
+      # Create a new evaluator with cross-validation method.
+      #
+      # @param estimator [Classifier] The classifier of which performance is evaluated.
+      # @param splitter [Splitter] The splitter that divides dataset to training and testing dataset.
+      # @param return_train_score [Boolean] The flag indicating whether to calculate the score of training dataset.
+      def initialize(estimator: nil, splitter: nil, return_train_score: false)
+        @estimator = estimator
+        @splitter = splitter
+        @return_train_score = return_train_score
+      end
+
+      # Perform the evalution of given classifier with cross-validation method.
+      #
+      # @param x [Numo::DFloat] (shape: [n_samples, n_features])
+      #   The dataset to be used to evaluate the classifier.
+      # @param y [Numo::Int32] (shape: [n_samples])
+      #   The labels to be used to evaluate the classifier.
+      # @return [Hash] The report summarizing the results of cross-validation.
+      #   * :fit_time (Array<Float>) The calculation times of fitting the estimator for each split.
+      #   * :test_score (Array<Float>) The scores of testing dataset for each split.
+      #   * :train_score (Array<Float>) The scores of training dataset for each split. This option is nil if
+      #     the return_train_score is false.
+      def perform(x, y)
+        # Initialize the report of cross validation.
+        report = {test_score: [], train_score: nil, fit_time: []}
+        report[:train_score] = [] if @return_train_score
+        # Evaluate the estimator on each split.
+        @splitter.split(x, y).each do |train_ids, test_ids|
+          # Split dataset into training and testing dataset.
+          feature_ids = !kernel_machine? || train_ids
+          train_x = x[train_ids, feature_ids]
+          train_y = y[train_ids]
+          test_x = x[test_ids, feature_ids]
+          test_y = y[test_ids]
+          # Fit the estimator.
+          start_time = Time.now.to_i
+          @estimator.fit(train_x, train_y)
+          # Calculate scores and prepare the report.
+          report[:fit_time].push(Time.now.to_i - start_time)
+          report[:test_score].push(@estimator.score(test_x, test_y))
+          report[:train_score].push(@estimator.score(train_x, train_y)) if @return_train_score
+        end
+        report
+      end
+
+      private
+
+      def kernel_machine?
+        class_name = @estimator.class.to_s
+        class_name = @estimator.params[:estimator].class.to_s if class_name.include?('Multiclass')
+        class_name.include?('KernelMachine')
+      end
+    end
+  end
+end

data/lib/svmkit/model_selection/k_fold.rb

@@ -16,7 +16,7 @@ module SVMKit
     class KFold
       include Base::Splitter
 
-      # Return the proportion of the test set to the dataset.
+      # Return the flag indicating whether to shuffle the dataset.
       # @return [Boolean]
       attr_reader :shuffle
 

data/lib/svmkit/model_selection/stratified_k_fold.rb

@@ -16,7 +16,7 @@ module SVMKit
     class StratifiedKFold
       include Base::Splitter
 
-      # Return the proportion of the test set to the dataset.
+      # Return the flag indicating whether to shuffle the dataset.
       # @return [Boolean]
       attr_reader :shuffle
 
@@ -47,7 +47,7 @@ module SVMKit
       # @return [Array] The set of data indices for constructing the training and testing dataset in each fold.
       def split(x, y) # rubocop:disable Lint/UnusedMethodArgument
         # Check the number of samples in each class.
-        unless y.bincount.to_a.all? { |n_samples| @n_splits.between?(2, n_samples) }
+        unless valid_n_splits?(y)
           raise ArgumentError,
                 'The value of n_splits must be not less than 2 and not more than the number of samples in each class.'
         end
@@ -59,6 +59,10 @@ module SVMKit
 
       private
 
+      def valid_n_splits?(y)
+        y.to_a.uniq.map { |label| y.eq(label).where.size }.all? { |n_samples| @n_splits.between?(2, n_samples) }
+      end
+
       def fold_sets(y, label)
         sample_ids = y.eq(label).where.to_a
         sample_ids.shuffle!(random: @rng) if @shuffle

data/lib/svmkit/version.rb

@@ -1,5 +1,5 @@
 # SVMKit is an experimental library of machine learning in Ruby.
 module SVMKit
   # @!visibility private
-  VERSION = '0.2.2'.freeze
+  VERSION = '0.2.3'.freeze
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: svmkit
 version: !ruby/object:Gem::Version
-  version: 0.2.2
+  version: 0.2.3
 platform: ruby
 authors:
 - yoshoku
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2018-01-13 00:00:00.000000000 Z
+date: 2018-01-15 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: numo-narray
@@ -112,6 +112,7 @@ files:
 - lib/svmkit/kernel_machine/kernel_svc.rb
 - lib/svmkit/linear_model/logistic_regression.rb
 - lib/svmkit/linear_model/svc.rb
+- lib/svmkit/model_selection/cross_validation.rb
 - lib/svmkit/model_selection/k_fold.rb
 - lib/svmkit/model_selection/stratified_k_fold.rb
 - lib/svmkit/multiclass/one_vs_rest_classifier.rb