rumale 0.19.2 → 0.19.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 1bff2e1e6182aa954be00ed107ed1bd81220298f89514b4b31304f8890ff27c4
-  data.tar.gz: '09b185f468baf9dbec6280fa6c06984c95919308f1d2247277bf30348ed392bc'
+  metadata.gz: dc3413c05ad7c365117adc4abbc304ff1851fa9a3ff69fef3c69e730d9a2b834
+  data.tar.gz: 8895cce8b350c4e245aabb5e3e4c4036655fa8d24a72f6481e0d2f8c9869fa54
 SHA512:
-  metadata.gz: 6d8f1fcaffcd6714c6156fc615d87e6b6950e82ab40fc7434cfc5a014d6c08eb0170ee7c45d8fed978c2a52f839b1ce647fd6e088cbab2ea45e517b34c88407a
-  data.tar.gz: b255ae4c24cdc91ebad59f79ee5a58c5d2a5ffa79bda0ac221e3a33bd824d2fd94e5cd83f3a06e54a2dc537a074276cea5a71651deeee2a304d23e963ff92c9d
+  metadata.gz: fe10c975f286a4c9ac155d29310d61d1f180cbcc909ec7bdba3925973b6b9857635befc9bf4938cf28a6ef50c8011894b7b15768735f6200c27ce912907e5fb1
+  data.tar.gz: 327cce25145c1ca3f5623f84b4163560bdbee8245009c3d8e1c3318f61dec94b58a7395ebc3538e7b04a9702af08b077e7f426c54f7c5d4fd3b0fcf11c4744cf

data/.rubocop.yml

@@ -24,6 +24,15 @@ Style/HashTransformKeys:
 Style/HashTransformValues:
   Enabled: true
 
+Lint/DeprecatedOpenSSLConstant:
+  Enabled: true
+
+Lint/DuplicateElsifCondition:
+  Enabled: true
+
+Lint/MixedRegexpCaptureTypes:
+  Enabled: true
+
 Lint/RaiseException:
   Enabled: true
 
@@ -34,7 +43,6 @@ Layout/LineLength:
   Max: 145
   IgnoredPatterns: ['(\A|\s)#']
 
-
 Metrics/ModuleLength:
   Max: 200
 
@@ -70,15 +78,48 @@ Naming/MethodParameterName:
 Naming/ConstantName:
   Enabled: false
 
+Style/AccessorGrouping:
+  Enabled: true
+
+Style/ArrayCoercion:
+  Enabled: true
+
+Style/BisectedAttrAccessor:
+  Enabled: true
+
+Style/CaseLikeIf:
+  Enabled: true
+
 Style/ExponentialNotation:
   Enabled: true
 
 Style/FormatStringToken:
   Enabled: false
 
+Style/HashAsLastArrayItem:
+  Enabled: true
+
+Style/HashLikeCase:
+  Enabled: true
+
 Style/NumericLiterals:
   Enabled: false
 
+Style/RedundantAssignment:
+  Enabled: true
+
+Style/RedundantFetchBlock:
+  Enabled: true
+
+Style/RedundantFileExtensionInRequire:
+  Enabled: true
+
+Style/RedundantRegexpCharacterClass:
+  Enabled: true
+
+Style/RedundantRegexpEscape:
+  Enabled: true
+
 Style/SlicingWithRange:
   Enabled: true
 
@@ -91,6 +132,30 @@ Layout/EmptyLinesAroundAttributeAccessor:
 Layout/SpaceAroundMethodCallOperator:
   Enabled: true
 
+Performance/AncestorsInclude:
+  Enabled: true
+
+Performance/BigDecimalWithNumericArgument:
+  Enabled: true
+
+Performance/RedundantSortBlock:
+  Enabled: true
+
+Performance/RedundantStringChars:
+  Enabled: true
+
+Performance/ReverseFirst:
+  Enabled: true
+
+Performance/SortReverse:
+  Enabled: true
+
+Performance/Squeeze:
+  Enabled: true
+
+Performance/StringInclude:
+  Enabled: true
+
 RSpec/MultipleExpectations:
   Enabled: false
 

data/CHANGELOG.md

@@ -1,3 +1,8 @@
+# 0.19.3
+- Add preprocessing class for [Binarizer](https://yoshoku.github.io/rumale/doc/Rumale/Preprocessing/Binarizer.html)
+- Add preprocessing class for [MaxNormalizer](https://yoshoku.github.io/rumale/doc/Rumale/Preprocessing/MaxNormalizer.html)
+- Refactor some codes with Rubocop.
+
 # 0.19.2
 - Fix L2Normalizer to avoid zero divide.
 - Add preprocssing class for [L1Normalizer](https://yoshoku.github.io/rumale/doc/Rumale/Preprocessing/L1Normalizer.html).

data/lib/rumale.rb

@@ -96,6 +96,7 @@ require 'rumale/feature_extraction/feature_hasher'
 require 'rumale/feature_extraction/tfidf_transformer'
 require 'rumale/preprocessing/l2_normalizer'
 require 'rumale/preprocessing/l1_normalizer'
+require 'rumale/preprocessing/max_normalizer'
 require 'rumale/preprocessing/min_max_scaler'
 require 'rumale/preprocessing/max_abs_scaler'
 require 'rumale/preprocessing/standard_scaler'
@@ -104,6 +105,7 @@ require 'rumale/preprocessing/label_binarizer'
 require 'rumale/preprocessing/label_encoder'
 require 'rumale/preprocessing/one_hot_encoder'
 require 'rumale/preprocessing/ordinal_encoder'
+require 'rumale/preprocessing/binarizer'
 require 'rumale/preprocessing/polynomial_features'
 require 'rumale/model_selection/k_fold'
 require 'rumale/model_selection/stratified_k_fold'

data/lib/rumale/clustering/hdbscan.rb

@@ -232,7 +232,7 @@ module Rumale
       end
 
       def flatten(tree, stabilities)
-        node_ids = stabilities.keys.sort { |a, b| b <=> a }.slice(0, stabilities.size - 1)
+        node_ids = stabilities.keys.sort.reverse.slice(0, stabilities.size - 1)
 
         cluster_tree = tree.select { |edge| edge.n_elements > 1 }
         is_cluster = node_ids.each_with_object({}) { |n_id, h| h[n_id] = true }

data/lib/rumale/clustering/k_means.rb

@@ -103,7 +103,7 @@ module Rumale
         # random initialize
         n_samples = x.shape[0]
         sub_rng = @rng.dup
-        rand_id = [*0...n_samples].sample(@params[:n_clusters], random: sub_rng)
+        rand_id = Array(0...n_samples).sample(@params[:n_clusters], random: sub_rng)
         @cluster_centers = x[rand_id, true].dup
         return unless @params[:init] == 'k-means++'
 

data/lib/rumale/clustering/k_medoids.rb

@@ -124,7 +124,7 @@ module Rumale
         # random initialize
         n_samples = distance_mat.shape[0]
         sub_rng = @rng.dup
-        @medoid_ids = Numo::Int32.asarray([*0...n_samples].sample(@params[:n_clusters], random: sub_rng))
+        @medoid_ids = Numo::Int32.asarray(Array(0...n_samples).sample(@params[:n_clusters], random: sub_rng))
         return unless @params[:init] == 'k-means++'
 
         # k-means++ initialize

data/lib/rumale/clustering/mini_batch_k_means.rb

@@ -67,7 +67,7 @@ module Rumale
         init_cluster_centers(x, sub_rng)
         # optimization with mini-batch sgd.
         @params[:max_iter].times do |_t|
-          sample_ids = [*0...n_samples].shuffle(random: sub_rng)
+          sample_ids = Array(0...n_samples).shuffle(random: sub_rng)
           old_centers = @cluster_centers.dup
           until (subset_ids = sample_ids.shift(@params[:batch_size])).empty?
             # sub sampling
@@ -120,7 +120,7 @@ module Rumale
       def init_cluster_centers(x, sub_rng)
         # random initialize
         n_samples = x.shape[0]
-        rand_id = [*0...n_samples].sample(@params[:n_clusters], random: sub_rng)
+        rand_id = Array(0...n_samples).sample(@params[:n_clusters], random: sub_rng)
         @cluster_centers = x[rand_id, true].dup
         return unless @params[:init] == 'k-means++'
 

data/lib/rumale/dataset.rb

@@ -81,7 +81,7 @@ module Rumale
         y = Numo::Int32.hstack([Numo::Int32.zeros(n_samples_out), Numo::Int32.ones(n_samples_in)])
         # shuffle data indices.
         if shuffle
-          rand_ids = [*0...n_samples].shuffle(random: rng.dup)
+          rand_ids = Array(0...n_samples).shuffle(random: rng.dup)
           x = x[rand_ids, true].dup
           y = y[rand_ids].dup
         end
@@ -118,7 +118,7 @@ module Rumale
         y = Numo::Int32.hstack([Numo::Int32.zeros(n_samples_out), Numo::Int32.ones(n_samples_in)])
         # shuffle data indices.
         if shuffle
-          rand_ids = [*0...n_samples].shuffle(random: rng.dup)
+          rand_ids = Array(0...n_samples).shuffle(random: rng.dup)
           x = x[rand_ids, true].dup
           y = y[rand_ids].dup
         end
@@ -173,7 +173,7 @@ module Rumale
         end
         # shuffle data.
         if shuffle
-          rand_ids = [*0...n_samples].shuffle(random: rng.dup)
+          rand_ids = Array(0...n_samples).shuffle(random: rng.dup)
           x = x[rand_ids, true].dup
           y = y[rand_ids].dup
         end

data/lib/rumale/kernel_approximation/nystroem.rb

@@ -69,7 +69,7 @@ module Rumale
         n_components = [1, [@params[:n_components], n_samples].min].max
 
         # random sampling.
-        @component_indices = Numo::Int32.cast([*0...n_samples].shuffle(random: sub_rng)[0...n_components])
+        @component_indices = Numo::Int32.cast(Array(0...n_samples).shuffle(random: sub_rng)[0...n_components])
         @components = x[@component_indices, true]
 
         # calculate normalizing factor.

data/lib/rumale/kernel_machine/kernel_svc.rb

@@ -172,7 +172,7 @@ module Rumale
         # Start optimization.
         @params[:max_iter].times do |t|
           # random sampling
-          rand_ids = [*0...n_training_samples].shuffle(random: sub_rng) if rand_ids.empty?
+          rand_ids = Array(0...n_training_samples).shuffle(random: sub_rng) if rand_ids.empty?
           target_id = rand_ids.shift
           # update the weight vector
           func = (weight_vec * bin_y).dot(x[target_id, true].transpose).to_f

data/lib/rumale/linear_model/base_linear_model.rb

@@ -56,7 +56,7 @@ module Rumale
         samples = @params[:fit_bias] ? expand_feature(x) : x
         # Initialize some variables.
         n_samples, n_features = samples.shape
-        rand_ids = [*0...n_samples].shuffle(random: @rng.dup)
+        rand_ids = Array(0...n_samples).shuffle(random: @rng.dup)
         weight = Numo::DFloat.zeros(n_features)
         optimizer = @params[:optimizer].dup
         # Optimization.

data/lib/rumale/linear_model/base_sgd.rb

@@ -209,7 +209,7 @@ module Rumale
         l1_penalty = LinearModel::Penalty::L1Penalty.new(reg_param: l1_reg_param) if apply_l1_penalty?
         # Optimization.
         @params[:max_iter].times do |t|
-          sample_ids = [*0...n_samples]
+          sample_ids = Array(0...n_samples)
           sample_ids.shuffle!(random: sub_rng)
           until (subset_ids = sample_ids.shift(@params[:batch_size])).empty?
             # sampling

data/lib/rumale/model_selection/cross_validation.rb

@@ -69,10 +69,11 @@ module Rumale
       #     the return_train_score is false.
       def perform(x, y)
         x = check_convert_sample_array(x)
-        if @estimator.is_a?(Rumale::Base::Classifier)
+        case @estimator
+        when Rumale::Base::Classifier
           y = check_convert_label_array(y)
           check_sample_label_size(x, y)
-        elsif @estimator.is_a?(Rumale::Base::Regressor)
+        when Rumale::Base::Regressor
           y = check_convert_tvalue_array(y)
           check_sample_tvalue_size(x, y)
         else

data/lib/rumale/model_selection/k_fold.rb

@@ -62,7 +62,7 @@ module Rumale
         end
         sub_rng = @rng.dup
         # Splits dataset ids to each fold.
-        dataset_ids = [*0...n_samples]
+        dataset_ids = Array(0...n_samples)
         dataset_ids.shuffle!(random: sub_rng) if @shuffle
         fold_sets = Array.new(@n_splits) do |n|
           n_fold_samples = n_samples / @n_splits

data/lib/rumale/model_selection/shuffle_split.rb

@@ -74,7 +74,7 @@ module Rumale
         end
         sub_rng = @rng.dup
         # Returns array consisting of the training and testing ids for each fold.
-        dataset_ids = [*0...n_samples]
+        dataset_ids = Array(0...n_samples)
         Array.new(@n_splits) do
           test_ids = dataset_ids.sample(n_test_samples, random: sub_rng)
           train_ids = if @train_size.nil?

data/lib/rumale/multiclass/one_vs_rest_classifier.rb

@@ -1,7 +1,7 @@
 # frozen_string_literal: true
 
-require 'rumale/base/base_estimator.rb'
-require 'rumale/base/classifier.rb'
+require 'rumale/base/base_estimator'
+require 'rumale/base/classifier'
 
 module Rumale
   # This module consists of the classes that implement multi-class classification strategy.

data/lib/rumale/nearest_neighbors/vp_tree.rb

@@ -30,7 +30,7 @@ module Rumale
         @params = {}
         @params[:min_samples_leaf] = min_samples_leaf
         @data = x
-        @tree = build_tree(Numo::Int32.cast([*0...@data.shape[0]]))
+        @tree = build_tree(Numo::Int32.cast(Array(0...@data.shape[0])))
       end
 
       # Search k-nearest neighbors of given query point.

data/lib/rumale/neural_network/base_mlp.rb

@@ -222,7 +222,7 @@ module Rumale
         n_samples = x.shape[0]
 
         @params[:max_iter].times do |t|
-          sample_ids = [*0...n_samples]
+          sample_ids = Array(0...n_samples)
           sample_ids.shuffle!(random: srng)
           until (subset_ids = sample_ids.shift(@params[:batch_size])).empty?
             # random sampling

data/lib/rumale/polynomial_model/base_factorization_machine.rb

@@ -69,7 +69,7 @@ module Rumale
         factor_optimizers = Array.new(@params[:n_factors]) { @params[:optimizer].dup }
         # Start optimization.
         @params[:max_iter].times do |t|
-          sample_ids = [*0...n_samples]
+          sample_ids = Array(0...n_samples)
           sample_ids.shuffle!(random: sub_rng)
           until (subset_ids = sample_ids.shift(@params[:batch_size])).empty?
             # Sampling.

data/lib/rumale/preprocessing/binarizer.rb

@@ -0,0 +1,60 @@
+# frozen_string_literal: true
+
+require 'rumale/base/base_estimator'
+require 'rumale/base/transformer'
+
+module Rumale
+  module Preprocessing
+    # Binarize samples according to a threshold
+    #
+    # @example
+    #   binarizer = Rumale::Preprocessing::Binarizer.new
+    #   x = Numo::DFloat[[-1.2, 3.2], [2.4, -0.5], [4.5, 0.8]]
+    #   b = binarizer.transform(x)
+    #   p b
+    #
+    #   # Numo::DFloat#shape=[3, 2]
+    #   # [[0, 1],
+    #   #  [1, 0],
+    #   #  [1, 1]]
+    class Binarizer
+      include Base::BaseEstimator
+      include Base::Transformer
+
+      # Create a new transformer for binarization.
+      # @param threshold [Float] The threshold value for binarization.
+      def initialize(threshold: 0.0)
+        check_params_numeric(threshold: threshold)
+        @params = { threshold: threshold }
+      end
+
+      # This method does nothing and returns the object itself.
+      # For compatibility with other transformer, this method exists.
+      #
+      # @overload fit() -> Binarizer
+      #
+      # @return [Binarizer]
+      def fit(_x = nil, _y = nil)
+        self
+      end
+
+      # Binarize each sample.
+      #
+      # @param x [Numo::DFloat] (shape: [n_samples, n_features]) The samples to be binarized.
+      # @return [Numo::DFloat] The binarized samples.
+      def transform(x)
+        x = check_convert_sample_array(x)
+        x.class.cast(x.gt(@params[:threshold]))
+      end
+
+      # The output of this method is the same as that of the transform method.
+      # For compatibility with other transformer, this method exists.
+      #
+      # @param x [Numo::DFloat] (shape: [n_samples, n_features]) The samples to be binarized.
+      # @return [Numo::DFloat] The binarized samples.
+      def fit_transform(x, _y = nil)
+        fit(x).transform(x)
+      end
+    end
+  end
+end

data/lib/rumale/preprocessing/max_normalizer.rb

@@ -0,0 +1,62 @@
+# frozen_string_literal: true
+
+require 'rumale/base/base_estimator'
+require 'rumale/base/transformer'
+
+module Rumale
+  module Preprocessing
+    # Normalize samples with the maximum of the absolute values.
+    #
+    # @example
+    #   normalizer = Rumale::Preprocessing::MaxNormalizer.new
+    #   new_samples = normalizer.fit_transform(samples)
+    class MaxNormalizer
+      include Base::BaseEstimator
+      include Base::Transformer
+
+      # Return the vector consists of the maximum norm for each sample.
+      # @return [Numo::DFloat] (shape: [n_samples])
+      attr_reader :norm_vec # :nodoc:
+
+      # Create a new normalizer for normaliing to max-norm.
+      def initialize
+        @params = {}
+        @norm_vec = nil
+      end
+
+      # Calculate the maximum norms of each sample.
+      #
+      # @overload fit(x) -> MaxNormalizer
+      #
+      # @param x [Numo::DFloat] (shape: [n_samples, n_features]) The samples to calculate the maximum norms.
+      # @return [MaxNormalizer]
+      def fit(x, _y = nil)
+        x = check_convert_sample_array(x)
+        @norm_vec = x.abs.max(1)
+        @norm_vec[@norm_vec.eq(0)] = 1
+        self
+      end
+
+      # Calculate the maximums norm of each sample, and then normalize samples with the norms.
+      #
+      # @overload fit_transform(x) -> Numo::DFloat
+      #
+      # @param x [Numo::DFloat] (shape: [n_samples, n_features]) The samples to calculate maximum norms.
+      # @return [Numo::DFloat] The normalized samples.
+      def fit_transform(x, _y = nil)
+        x = check_convert_sample_array(x)
+        fit(x)
+        x / @norm_vec.expand_dims(1)
+      end
+
+      # Calculate the maximum norms of each sample, and then normalize samples with the norms.
+      # This method calls the fit_transform method. This method exists for the Pipeline class.
+      #
+      # @param x [Numo::DFloat] (shape: [n_samples, n_features]) The samples to calculate maximum norms.
+      # @return [Numo::DFloat] The normalized samples.
+      def transform(x)
+        fit_transform(x)
+      end
+    end
+  end
+end

data/lib/rumale/version.rb

@@ -3,5 +3,5 @@
 # Rumale is a machine learning library in Ruby.
 module Rumale
   # The version of Rumale you are using.
-  VERSION = '0.19.2'
+  VERSION = '0.19.3'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: rumale
 version: !ruby/object:Gem::Version
-  version: 0.19.2
+  version: 0.19.3
 platform: ruby
 authors:
 - yoshoku
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2020-06-20 00:00:00.000000000 Z
+date: 2020-07-23 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: numo-narray
@@ -167,11 +167,13 @@ files:
 - lib/rumale/polynomial_model/factorization_machine_classifier.rb
 - lib/rumale/polynomial_model/factorization_machine_regressor.rb
 - lib/rumale/preprocessing/bin_discretizer.rb
+- lib/rumale/preprocessing/binarizer.rb
 - lib/rumale/preprocessing/l1_normalizer.rb
 - lib/rumale/preprocessing/l2_normalizer.rb
 - lib/rumale/preprocessing/label_binarizer.rb
 - lib/rumale/preprocessing/label_encoder.rb
 - lib/rumale/preprocessing/max_abs_scaler.rb
+- lib/rumale/preprocessing/max_normalizer.rb
 - lib/rumale/preprocessing/min_max_scaler.rb
 - lib/rumale/preprocessing/one_hot_encoder.rb
 - lib/rumale/preprocessing/ordinal_encoder.rb