Class: Rumale::KernelApproximation::RBF

Inherits:
Object
  • Object
show all
Includes:
Base::BaseEstimator, Base::Transformer
Defined in:
lib/rumale/kernel_approximation/rbf.rb

Overview

Class for RBF kernel feature mapping.

Refernce:

    1. Rahimi and B. Recht, “Random Features for Large-Scale Kernel Machines,” Proc. NIPS’07, pp.1177–1184, 2007.

Examples:

transformer = Rumale::KernelApproximation::RBF.new(gamma: 1.0, n_coponents: 128, random_seed: 1)
new_training_samples = transformer.fit_transform(training_samples)
new_testing_samples = transformer.transform(testing_samples)

Instance Attribute Summary collapse

Attributes included from Base::BaseEstimator

#params

Instance Method Summary collapse

Constructor Details

#initialize(gamma: 1.0, n_components: 128, random_seed: nil) ⇒ RBF

Create a new transformer for mapping to RBF kernel feature space.

Parameters:

  • gamma (Float) (defaults to: 1.0) —

    The parameter of RBF kernel: exp(-gamma * x^2).

  • n_components (Integer) (defaults to: 128) —

    The number of dimensions of the RBF kernel feature space.

  • random_seed (Integer) (defaults to: nil) —

    The seed value using to initialize the random generator.



40
41
42
43
44
45
46
47
48
49
50
51
52
53
 
# File 'lib/rumale/kernel_approximation/rbf.rb', line 40

def initialize(gamma: 1.0, n_components: 128, random_seed: nil)
  check_params_float(gamma: gamma)
  check_params_integer(n_components: n_components)
  check_params_type_or_nil(Integer, random_seed: random_seed)
  check_params_positive(gamma: gamma, n_components: n_components)
  @params = {}
  @params[:gamma] = gamma
  @params[:n_components] = n_components
  @params[:random_seed] = random_seed
  @params[:random_seed] ||= srand
  @random_mat = nil
  @random_vec = nil
  @rng = Random.new(@params[:random_seed])
end

Instance Attribute Details

#random_mat ⇒ Numo::DFloat (readonly)

Return the random matrix for transformation.

Returns:

  • (Numo::DFloat) —

    (shape: [n_features, n_components])



25
26
27
 
# File 'lib/rumale/kernel_approximation/rbf.rb', line 25

def random_mat
  @random_mat
end

#random_vec ⇒ Numo::DFloat (readonly)

Return the random vector for transformation.

Returns:

  • (Numo::DFloat) —

    (shape: [n_components])



29
30
31
 
# File 'lib/rumale/kernel_approximation/rbf.rb', line 29

def random_vec
  @random_vec
end

#rng ⇒ Random (readonly)

Return the random generator for transformation.

Returns:

  • (Random) 


33
34
35
 
# File 'lib/rumale/kernel_approximation/rbf.rb', line 33

def rng
  @rng
end

Instance Method Details

#fit(x) ⇒ RBF

Fit the model with given training data.

Parameters:

  • x (Numo::NArray) —

    (shape: [n_samples, n_features]) The training data to be used for fitting the model. This method uses only the number of features of the data.

Returns:

  • (RBF) —

    The learned transformer itself.



62
63
64
65
66
67
68
69
70
71
72
73
 
# File 'lib/rumale/kernel_approximation/rbf.rb', line 62

def fit(x, _y = nil)
  check_sample_array(x)

  n_features = x.shape[1]
  @params[:n_components] = 2 * n_features if @params[:n_components] <= 0
  @random_mat = Rumale::Utils.rand_normal([n_features, @params[:n_components]], @rng) * (2.0 * @params[:gamma])**0.5
  n_half_components = @params[:n_components] / 2
  @random_vec = Numo::DFloat.zeros(@params[:n_components] - n_half_components).concatenate(
    Numo::DFloat.ones(n_half_components) * (0.5 * Math::PI)
  )
  self
end

#fit_transform(x) ⇒ Numo::DFloat

Fit the model with training data, and then transform them with the learned model.

Parameters:

  • x (Numo::DFloat) —

    (shape: [n_samples, n_features]) The training data to be used for fitting the model.

Returns:

  • (Numo::DFloat) —

    (shape: [n_samples, n_components]) The transformed data



81
82
83
84
85
 
# File 'lib/rumale/kernel_approximation/rbf.rb', line 81

def fit_transform(x, _y = nil)
  check_sample_array(x)

  fit(x).transform(x)
end

#marshal_dump ⇒ Hash

Dump marshal data.

Returns:

  • (Hash) —

    The marshal data about RBF.



103
104
105
106
107
108
 
# File 'lib/rumale/kernel_approximation/rbf.rb', line 103

def marshal_dump
  { params: @params,
    random_mat: @random_mat,
    random_vec: @random_vec,
    rng: @rng }
end

#marshal_load(obj) ⇒ nil

Load marshal data.

Returns:

  • (nil) 


112
113
114
115
116
117
118
 
# File 'lib/rumale/kernel_approximation/rbf.rb', line 112

def marshal_load(obj)
  @params = obj[:params]
  @random_mat = obj[:random_mat]
  @random_vec = obj[:random_vec]
  @rng = obj[:rng]
  nil
end

#transform(x) ⇒ Numo::DFloat

Transform the given data with the learned model.

Parameters:

  • x (Numo::DFloat) —

    (shape: [n_samples, n_features]) The data to be transformed with the learned model.

Returns:

  • (Numo::DFloat) —

    (shape: [n_samples, n_components]) The transformed data.



93
94
95
96
97
98
99
 
# File 'lib/rumale/kernel_approximation/rbf.rb', line 93

def transform(x)
  check_sample_array(x)

  n_samples, = x.shape
  projection = x.dot(@random_mat) + @random_vec.tile(n_samples, 1)
  Numo::NMath.sin(projection) * ((2.0 / @params[:n_components])**0.5)
end