rumale-manifold 0.28.1 → 1.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 146643ddd999165173dcb3388d01240400288e9093d2d599aeff4348486eb8ce
-  data.tar.gz: bc2c0eca2e8acfa5c07a28dd7b13283ba505c69c066ae94139ac0dc4033933e5
+  metadata.gz: 90e4ed5028c5ec00bc02d2cbc982d524881d64d3a83b9e874aaa25d693a42b58
+  data.tar.gz: 5dd03085a372ba20432c842dfea839c1490ace268631bf76040ac45de36f5a4d
 SHA512:
-  metadata.gz: 308de32a6c4870c308f8da145a5cd4f6e3558d74457ea0aeb8dbc21c06babbdbb52a97e3dbe6a53b1d01cb622aebad79c69b612737023fbef24ef2a83be422f7
-  data.tar.gz: 6abd9ef1208b3eee3265bc23b4d785e5e70bf34bd88cd05cd1df457dfc02dc7757de63407ee338cb77d818ca2e7d238b50d14b97d9fb76a687511366ebcf7988
+  metadata.gz: 8145595c87833e14579640edfaa30eff46b5cdd34a5883d751bee9548fbd92dc070c5f69a6984ba68465912a9a246f47367ae4dff917b97ec70fdd54a0dd5610
+  data.tar.gz: 47ba4f7e544adbec19f864e519886357a9a9757eb44d352ab94a2f100d93f43aee671876f920ab7903682cfc3a39a2b412888b607cdd9ae3f25dd482d517ea15

data/LICENSE.txt

@@ -1,4 +1,4 @@
-Copyright (c) 2022-2023 Atsushi Tatsuma
+Copyright (c) 2022-2024 Atsushi Tatsuma
 All rights reserved.
 
 Redistribution and use in source and binary forms, with or without

data/lib/rumale/manifold/hessian_eigenmaps.rb

@@ -41,9 +41,9 @@ module Rumale
 
       # Fit the model with given training data.
       #
-      # @overload fit(x) -> LocallyLinearEmbedding
+      # @overload fit(x) -> HessianEigenmaps
       #   @param x [Numo::DFloat] (shape: [n_samples, n_features]) The training data to be used for fitting the model.
-      #   @return [LocallyLinearEmbedding] The learned transformer itself.
+      #   @return [HessianEigenmaps] The learned transformer itself.
       def fit(x, _y = nil) # rubocop:disable Metrics/AbcSize
         raise 'HessianEigenmaps#fit requires Numo::Linalg but that is not loaded' unless enable_linalg?(warning: false)
 

data/lib/rumale/manifold/local_tangent_space_alignment.rb

@@ -0,0 +1,142 @@
+# frozen_string_literal: true
+
+require 'rumale/base/estimator'
+require 'rumale/base/transformer'
+require 'rumale/pairwise_metric'
+require 'rumale/validation'
+
+module Rumale
+  module Manifold
+    # LocalTangentSpaceAlignment is a class that implements Local Tangent Space Alignment.
+    #
+    # @example
+    #   require 'numo/linalg/autoloader'
+    #   require 'rumale/manifold/local_tangent_space_alignment'
+    #
+    #   lem = Rumale::Manifold::LocalTangentSpaceAlignment.new(n_components: 2, n_neighbors: 15)
+    #   z = lem.fit_transform(x)
+    #
+    # *Reference*
+    # - Zhang, A., and Zha, H., "Principal Manifolds and Nonlinear Diemnsion Reduction via Local Tangent Space Alignment," SIAM Journal on Scientific Computing, vol. 26, iss. 1, pp. 313-338, 2004.
+    class LocalTangentSpaceAlignment < Rumale::Base::Estimator
+      include Rumale::Base::Transformer
+
+      # Return the data in representation space.
+      # @return [Numo::DFloat] (shape: [n_samples, n_components])
+      attr_reader :embedding
+
+      # Create a new transformer with Local Tangent Space Alignment.
+      #
+      # @param n_components [Integer] The number of dimensions on representation space.
+      # @param n_neighbors [Integer] The number of nearest neighbors for finding k-nearest neighbors
+      # @param reg_param [Float] The reguralization parameter for local gram matrix in transform method.
+      def initialize(n_components: 2, n_neighbors: 10, reg_param: 1e-3)
+        super()
+        @params = {
+          n_components: n_components,
+          n_neighbors: [1, n_neighbors].max,
+          reg_param: reg_param
+        }
+      end
+
+      # Fit the model with given training data.
+      #
+      # @overload fit(x) -> LocalTangentSpaceAlignment
+      #   @param x [Numo::DFloat] (shape: [n_samples, n_features]) The training data to be used for fitting the model.
+      #   @return [LocalTangentSpaceAlignment] The learned transformer itself.
+      def fit(x, _y = nil)
+        unless enable_linalg?(warning: false)
+          raise 'LocalTangentSpaceAlignment#fit requires Numo::Linalg but that is not loaded'
+        end
+
+        x = Rumale::Validation.check_convert_sample_array(x)
+
+        n_samples = x.shape[0]
+        distance_mat = Rumale::PairwiseMetric.squared_error(x)
+        neighbor_ids = neighbor_ids(distance_mat, @params[:n_neighbors], true)
+
+        affinity_mat = Numo::DFloat.zeros(n_samples, n_samples)
+        x_tangent = Numo::DFloat.zeros(@params[:n_neighbors], @params[:n_components] + 1)
+        x_tangent[true, 0] = 1.fdiv(Math.sqrt(@params[:n_neighbors]))
+
+        n_samples.times do |n|
+          x_local = x[neighbor_ids[n, true], true]
+          x_tangent[true, 1...] = right_singular_vectors(x_local, @params[:n_components])
+          weight_mat = x_tangent.dot(x_tangent.transpose)
+          neighbor_ids[n, true].each_with_index do |m, i|
+            affinity_mat[m, neighbor_ids[n, true]] -= weight_mat[i, true]
+            affinity_mat[m, m] += 1
+          end
+        end
+
+        kernel_mat = 0.5 * (affinity_mat.transpose + affinity_mat)
+        _, eig_vecs = Numo::Linalg.eigh(kernel_mat, vals_range: 1...(1 + @params[:n_components]))
+
+        @embedding = @params[:n_components] == 1 ? eig_vecs[true, 0].dup : eig_vecs.dup
+        @x_train = x.dup
+
+        self
+      end
+
+      # Fit the model with training data, and then transform them with the learned model.
+      #
+      # @overload fit_transform(x) -> Numo::DFloat
+      #   @param x [Numo::DFloat] (shape: [n_samples, n_features]) The training data to be used for fitting the model.
+      #   @return [Numo::DFloat] (shape: [n_samples, n_components]) The transformed data
+      def fit_transform(x, _y = nil)
+        unless enable_linalg?(warning: false)
+          raise 'LocalTangentSpaceAlignment#fit_transform requires Numo::Linalg but that is not loaded'
+        end
+
+        fit(x).transform(x)
+      end
+
+      # Transform the given data with the learned model.
+      # For out-of-sample data embedding, the same method as Locally Linear Embedding is used.
+      #
+      # @param x [Numo::DFloat] (shape: [n_samples, n_features]) The data to be transformed with the learned model.
+      # @return [Numo::DFloat] (shape: [n_samples, n_components]) The transformed data.
+      def transform(x)
+        x = Rumale::Validation.check_convert_sample_array(x)
+
+        n_samples = x.shape[0]
+        tol = @params[:reg_param].fdiv(@params[:n_neighbors])
+        distance_mat = Rumale::PairwiseMetric.squared_error(x, @x_train)
+        neighbor_ids = neighbor_ids(distance_mat, @params[:n_neighbors], false)
+        weight_mat = Numo::DFloat.zeros(n_samples, @x_train.shape[0])
+
+        n_samples.times do |n|
+          x_local = @x_train[neighbor_ids[n, true], true] - x[n, true]
+          gram_mat = x_local.dot(x_local.transpose)
+          gram_mat += tol * weight_mat.trace * Numo::DFloat.eye(@params[:n_neighbors])
+          weights = Numo::Linalg.solve(gram_mat, Numo::DFloat.ones(@params[:n_neighbors]))
+          weights /= weights.sum + 1e-8
+          weight_mat[n, neighbor_ids[n, true]] = weights
+        end
+
+        weight_mat.dot(@embedding)
+      end
+
+      private
+
+      def neighbor_ids(distance_mat, n_neighbors, contain_self)
+        n_samples = distance_mat.shape[0]
+        neighbor_ids = Numo::Int32.zeros(n_samples, n_neighbors)
+        if contain_self
+          n_samples.times { |n| neighbor_ids[n, true] = (distance_mat[n, true].sort_index.to_a - [n])[0...n_neighbors] }
+        else
+          n_samples.times { |n| neighbor_ids[n, true] = distance_mat[n, true].sort_index.to_a[0...n_neighbors] }
+        end
+        neighbor_ids
+      end
+
+      def right_singular_vectors(x_local, n_singulars)
+        n_samples = x_local.shape[0]
+        x_local -= x_local.mean(0)
+        gram_mat = x_local.dot(x_local.transpose)
+        _, evecs = Numo::Linalg.eigh(gram_mat, vals_range: (n_samples - n_singulars)...n_samples)
+        evecs.reverse(1).dup
+      end
+    end
+  end
+end

data/lib/rumale/manifold/locally_linear_embedding.rb

@@ -7,7 +7,7 @@ require 'rumale/validation'
 
 module Rumale
   module Manifold
-    # LocallyLinearEmbedding is a class that implements Loccaly Linear Embedding.
+    # LocallyLinearEmbedding is a class that implements Locally Linear Embedding.
     #
     # @example
     #   require 'numo/linalg/autoloader'

data/lib/rumale/manifold/version.rb

@@ -5,6 +5,6 @@ module Rumale
   # Module for data embedding algorithms.
   module Manifold
     # @!visibility private
-    VERSION = '0.28.1'
+    VERSION = '1.0.0'
   end
 end

data/lib/rumale/manifold.rb

@@ -5,6 +5,7 @@ require 'numo/narray'
 require_relative 'manifold/laplacian_eigenmaps'
 require_relative 'manifold/locally_linear_embedding'
 require_relative 'manifold/hessian_eigenmaps'
+require_relative 'manifold/local_tangent_space_alignment'
 require_relative 'manifold/mds'
 require_relative 'manifold/tsne'
 require_relative 'manifold/version'

metadata

@@ -1,14 +1,13 @@
 --- !ruby/object:Gem::Specification
 name: rumale-manifold
 version: !ruby/object:Gem::Version
-  version: 0.28.1
+  version: 1.0.0
 platform: ruby
 authors:
 - yoshoku
-autorequire:
 bindir: exe
 cert_chain: []
-date: 2023-12-24 00:00:00.000000000 Z
+date: 2025-01-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: numo-narray
@@ -30,28 +29,28 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.28.1
+        version: 1.0.0
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.28.1
+        version: 1.0.0
 - !ruby/object:Gem::Dependency
   name: rumale-decomposition
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.28.1
+        version: 1.0.0
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.28.1
+        version: 1.0.0
 description: |
   Rumale::Manifold provides data embedding algorithms,
   such as Multi-dimensional Scaling, Locally Linear Embedding, Laplacian Eigenmaps, Hessian Eigenmaps,
@@ -68,6 +67,7 @@ files:
 - lib/rumale/manifold.rb
 - lib/rumale/manifold/hessian_eigenmaps.rb
 - lib/rumale/manifold/laplacian_eigenmaps.rb
+- lib/rumale/manifold/local_tangent_space_alignment.rb
 - lib/rumale/manifold/locally_linear_embedding.rb
 - lib/rumale/manifold/mds.rb
 - lib/rumale/manifold/tsne.rb
@@ -80,7 +80,6 @@ metadata:
   source_code_uri: https://github.com/yoshoku/rumale/-/tree/main/rumale-manifold
   changelog_uri: https://github.com/yoshoku/rumale/-/blob/main/CHANGELOG.md
   rubygems_mfa_required: 'true'
-post_install_message:
 rdoc_options: []
 require_paths:
 - lib
@@ -95,8 +94,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.4.22
-signing_key:
+rubygems_version: 3.6.2
 specification_version: 4
 summary: Rumale::Manifold provides data embedding algorithms with Rumale interface.
 test_files: []