Class: Rumale::KernelMachine::KernelSVC

Inherits:
Base::Estimator show all
Includes:
Base::Classifier
Defined in:
rumale-kernel_machine/lib/rumale/kernel_machine/kernel_svc.rb

Overview

Note:

Rumale::SVM provides kernel support vector classifier based on LIBSVM. If you prefer execution speed, you should use Rumale::SVM::SVC. github.com/yoshoku/rumale-svm

KernelSVC is a class that implements (Nonlinear) Kernel Support Vector Classifier with stochastic gradient descent (SGD) optimization. For multiclass classification problem, it uses one-vs-the-rest strategy.

Reference

  • Shalev-Shwartz, S., Singer, Y., Srebro, N., and Cotter, A., “Pegasos: Primal Estimated sub-GrAdient SOlver for SVM,” Mathematical Programming, vol. 127 (1), pp. 3–30, 2011.

Examples:

require 'rumale/pairwise_metric'
require 'rumale/kernel_machine/kernel_svc'

training_kernel_matrix = Rumale::PairwiseMetric::rbf_kernel(training_samples)
estimator =
  Rumale::KernelMachine::KernelSVC.new(reg_param: 1.0, max_iter: 1000, random_seed: 1)
estimator.fit(training_kernel_matrix, traininig_labels)
testing_kernel_matrix = Rumale::PairwiseMetric::rbf_kernel(testing_samples, training_samples)
results = estimator.predict(testing_kernel_matrix)

Instance Attribute Summary collapse

Attributes inherited from Base::Estimator

#params

Instance Method Summary collapse

Methods included from Base::Classifier

#score

Constructor Details

#initialize(reg_param: 1.0, max_iter: 1000, probability: false, n_jobs: nil, random_seed: nil) ⇒ KernelSVC

Create a new classifier with Kernel Support Vector Machine by the SGD optimization.

Parameters:

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

    The regularization parameter.

  • max_iter (Integer) (defaults to: 1000) —

    The maximum number of iterations.

  • probability (Boolean) (defaults to: false) —

    The flag indicating whether to perform probability estimation.

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

    The number of jobs for running the fit and predict methods in parallel. If nil is given, the methods do not execute in parallel. If zero or less is given, it becomes equal to the number of processors. This parameter is ignored if the Parallel gem is not loaded.

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

    The seed value using to initialize the random generator.



57
58
59
60
61
62
63
64
65
66
67
 
# File 'rumale-kernel_machine/lib/rumale/kernel_machine/kernel_svc.rb', line 57

def initialize(reg_param: 1.0, max_iter: 1000, probability: false, n_jobs: nil, random_seed: nil)
  super()
  @params = {
    reg_param: reg_param,
    max_iter: max_iter,
    probability: probability,
    n_jobs: n_jobs,
    random_seed: random_seed || srand
  }
  @rng = Random.new(@params[:random_seed])
end

Instance Attribute Details

#classes ⇒ Numo::Int32 (readonly)

Return the class labels.

Returns:

  • (Numo::Int32) —

    (shape: [n_classes])



41
42
43
 
# File 'rumale-kernel_machine/lib/rumale/kernel_machine/kernel_svc.rb', line 41

def classes
  @classes
end

#rng ⇒ Random (readonly)

Return the random generator for performing random sampling.

Returns:

  • (Random) 


45
46
47
 
# File 'rumale-kernel_machine/lib/rumale/kernel_machine/kernel_svc.rb', line 45

def rng
  @rng
end

#weight_vec ⇒ Numo::DFloat (readonly)

Return the weight vector for Kernel SVC.

Returns:

  • (Numo::DFloat) —

    (shape: [n_classes, n_trainig_sample])



37
38
39
 
# File 'rumale-kernel_machine/lib/rumale/kernel_machine/kernel_svc.rb', line 37

def weight_vec
  @weight_vec
end

Instance Method Details

#decision_function(x) ⇒ Numo::DFloat

Calculate confidence scores for samples.

Parameters:

  • x (Numo::DFloat) —

    (shape: [n_testing_samples, n_training_samples]) The kernel matrix between testing samples and training samples to compute the scores.

Returns:

  • (Numo::DFloat) —

    (shape: [n_testing_samples, n_classes]) Confidence score per sample.



113
114
115
116
117
 
# File 'rumale-kernel_machine/lib/rumale/kernel_machine/kernel_svc.rb', line 113

def decision_function(x)
  x = ::Rumale::Validation.check_convert_sample_array(x)

  x.dot(@weight_vec.transpose)
end

#fit(x, y) ⇒ KernelSVC

Fit the model with given training data.

Parameters:

  • x (Numo::DFloat) —

    (shape: [n_training_samples, n_training_samples]) The kernel matrix of the training data to be used for fitting the model.

  • y (Numo::Int32) —

    (shape: [n_training_samples]) The labels to be used for fitting the model.

Returns:

  • (KernelSVC) —

    The learned classifier itself.



75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
 
# File 'rumale-kernel_machine/lib/rumale/kernel_machine/kernel_svc.rb', line 75

def fit(x, y)
  x = ::Rumale::Validation.check_convert_sample_array(x)
  y = ::Rumale::Validation.check_convert_label_array(y)
  ::Rumale::Validation.check_sample_size(x, y)

  @classes = Numo::Int32[*y.to_a.uniq.sort]
  n_classes = @classes.size
  n_features = x.shape[1]

  if n_classes > 2
    @weight_vec = Numo::DFloat.zeros(n_classes, n_features)
    @prob_param = Numo::DFloat.zeros(n_classes, 2)
    models = if enable_parallel?
               parallel_map(n_classes) do |n|
                 bin_y = Numo::Int32.cast(y.eq(@classes[n])) * 2 - 1
                 partial_fit(x, bin_y)
               end
             else
               Array.new(n_classes) do |n|
                 bin_y = Numo::Int32.cast(y.eq(@classes[n])) * 2 - 1
                 partial_fit(x, bin_y)
               end
             end
    models.each_with_index { |model, n| @weight_vec[n, true], @prob_param[n, true] = model }
  else
    negative_label = y.to_a.uniq.min
    bin_y = Numo::Int32.cast(y.ne(negative_label)) * 2 - 1
    @weight_vec, @prob_param = partial_fit(x, bin_y)
  end

  self
end

#predict(x) ⇒ Numo::Int32

Predict class labels for samples.

Parameters:

  • x (Numo::DFloat) —

    (shape: [n_testing_samples, n_training_samples]) The kernel matrix between testing samples and training samples to predict the labels.

Returns:

  • (Numo::Int32) —

    (shape: [n_testing_samples]) Predicted class label per sample.



124
125
126
127
128
129
130
131
132
133
134
135
136
137
 
# File 'rumale-kernel_machine/lib/rumale/kernel_machine/kernel_svc.rb', line 124

def predict(x)
  x = ::Rumale::Validation.check_convert_sample_array(x)

  return Numo::Int32.cast(decision_function(x).ge(0.0)) * 2 - 1 if @classes.size <= 2

  n_samples, = x.shape
  decision_values = decision_function(x)
  predicted = if enable_parallel?
                parallel_map(n_samples) { |n| @classes[decision_values[n, true].max_index] }
              else
                Array.new(n_samples) { |n| @classes[decision_values[n, true].max_index] }
              end
  Numo::Int32.asarray(predicted)
end

#predict_proba(x) ⇒ Numo::DFloat

Predict probability for samples.

Parameters:

  • x (Numo::DFloat) —

    (shape: [n_testing_samples, n_training_samples]) The kernel matrix between testing samples and training samples to predict the labels.

Returns:

  • (Numo::DFloat) —

    (shape: [n_samples, n_classes]) Predicted probability of each class per sample.



144
145
146
147
148
149
150
151
152
153
154
155
156
157
 
# File 'rumale-kernel_machine/lib/rumale/kernel_machine/kernel_svc.rb', line 144

def predict_proba(x)
  x = ::Rumale::Validation.check_convert_sample_array(x)

  if @classes.size > 2
    probs = 1.0 / (Numo::NMath.exp(@prob_param[true, 0] * decision_function(x) + @prob_param[true, 1]) + 1.0)
    return (probs.transpose / probs.sum(axis: 1)).transpose.dup
  end

  n_samples, = x.shape
  probs = Numo::DFloat.zeros(n_samples, 2)
  probs[true, 1] = 1.0 / (Numo::NMath.exp(@prob_param[0] * decision_function(x) + @prob_param[1]) + 1.0)
  probs[true, 0] = 1.0 - probs[true, 1]
  probs
end