Class: Rumale::KernelApproximation::RBF
- Inherits:
-
Base::Estimator
- Object
- Base::Estimator
- Rumale::KernelApproximation::RBF
- Includes:
- Base::Transformer
- Defined in:
- rumale-kernel_approximation/lib/rumale/kernel_approximation/rbf.rb
Overview
Class for RBF kernel feature mapping.
Refernce:
-
Rahimi, A., and Recht, B., “Random Features for Large-Scale Kernel Machines,” Proc. NIPS’07, pp.1177–1184, 2007.
Instance Attribute Summary collapse
-
#random_mat ⇒ Numo::DFloat
readonly
Return the random matrix for transformation.
-
#random_vec ⇒ Numo::DFloat
readonly
Return the random vector for transformation.
-
#rng ⇒ Random
readonly
Return the random generator for transformation.
Attributes inherited from Base::Estimator
Instance Method Summary collapse
-
#fit(x) ⇒ RBF
Fit the model with given training data.
-
#fit_transform(x) ⇒ Numo::DFloat
Fit the model with training data, and then transform them with the learned model.
-
#initialize(gamma: 1.0, n_components: 128, random_seed: nil) ⇒ RBF
constructor
Create a new transformer for mapping to RBF kernel feature space.
-
#transform(x) ⇒ Numo::DFloat
Transform the given data with the learned model.
Constructor Details
#initialize(gamma: 1.0, n_components: 128, random_seed: nil) ⇒ RBF
Create a new transformer for mapping to RBF kernel feature space.
41 42 43 44 45 46 47 48 49 |
# File 'rumale-kernel_approximation/lib/rumale/kernel_approximation/rbf.rb', line 41 def initialize(gamma: 1.0, n_components: 128, random_seed: nil) super() @params = { gamma: gamma, n_components: n_components, random_seed: random_seed || srand } @rng = Random.new(@params[:random_seed]) end |
Instance Attribute Details
#random_mat ⇒ Numo::DFloat (readonly)
Return the random matrix for transformation.
26 27 28 |
# File 'rumale-kernel_approximation/lib/rumale/kernel_approximation/rbf.rb', line 26 def random_mat @random_mat end |
#random_vec ⇒ Numo::DFloat (readonly)
Return the random vector for transformation.
30 31 32 |
# File 'rumale-kernel_approximation/lib/rumale/kernel_approximation/rbf.rb', line 30 def random_vec @random_vec end |
#rng ⇒ Random (readonly)
Return the random generator for transformation.
34 35 36 |
# File 'rumale-kernel_approximation/lib/rumale/kernel_approximation/rbf.rb', line 34 def rng @rng end |
Instance Method Details
#fit(x) ⇒ RBF
Fit the model with given training data.
57 58 59 60 61 62 63 64 65 66 67 68 69 |
# File 'rumale-kernel_approximation/lib/rumale/kernel_approximation/rbf.rb', line 57 def fit(x, _y = nil) x = ::Rumale::Validation.check_convert_sample_array(x) n_features = x.shape[1] sub_rng = @rng.dup @params[:n_components] = 2 * n_features if @params[:n_components] <= 0 @random_mat = ::Rumale::Utils.rand_normal([n_features, @params[:n_components]], sub_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.
76 77 78 79 80 |
# File 'rumale-kernel_approximation/lib/rumale/kernel_approximation/rbf.rb', line 76 def fit_transform(x, _y = nil) x = ::Rumale::Validation.check_convert_sample_array(x) fit(x).transform(x) end |
#transform(x) ⇒ Numo::DFloat
Transform the given data with the learned model.
87 88 89 90 91 92 93 |
# File 'rumale-kernel_approximation/lib/rumale/kernel_approximation/rbf.rb', line 87 def transform(x) x = ::Rumale::Validation.check_convert_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 |