Class: SVMKit::LinearModel::SVC

Inherits:
Object
  • Object
show all
Includes:
Base::BaseEstimator, Base::Classifier
Defined in:
lib/svmkit/linear_model/svc.rb

Overview

SVC is a class that implements Support Vector Classifier with mini-batch stochastic gradient descent optimization. For multiclass classification problem, it uses one-vs-the-rest strategy.

Reference

    1. Shalev-Shwartz and Y. Singer, “Pegasos: Primal Estimated sub-GrAdient SOlver for SVM,” Proc. ICML’07, pp. 807–814, 2007.

Examples:

estimator =
  SVMKit::LinearModel::SVC.new(reg_param: 1.0, max_iter: 1000, batch_size: 20, random_seed: 1)
estimator.fit(training_samples, traininig_labels)
results = estimator.predict(testing_samples)

Instance Attribute Summary collapse

Attributes included from Base::BaseEstimator

#params

Instance Method Summary collapse

Methods included from Base::Classifier

#score

Constructor Details

#initialize(reg_param: 1.0, fit_bias: false, bias_scale: 1.0, max_iter: 1000, batch_size: 20, probability: false, optimizer: nil, random_seed: nil) ⇒ SVC

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

Parameters:

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

    The regularization parameter.

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

    The flag indicating whether to fit the bias term.

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

    The scale of the bias term.

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

    The maximum number of iterations.

  • batch_size (Integer) (defaults to: 20) —

    The size of the mini batches.

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

    The flag indicating whether to perform probability estimation.

  • optimizer (Optimizer) (defaults to: nil) —

    The optimizer to calculate adaptive learning rate. If nil is given, Nadam is used.

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

    The seed value using to initialize the random generator.



56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
 
# File 'lib/svmkit/linear_model/svc.rb', line 56

def initialize(reg_param: 1.0, fit_bias: false, bias_scale: 1.0,
               max_iter: 1000, batch_size: 20, probability: false, optimizer: nil, random_seed: nil)
  check_params_float(reg_param: reg_param, bias_scale: bias_scale)
  check_params_integer(max_iter: max_iter, batch_size: batch_size)
  check_params_boolean(fit_bias: fit_bias, probability: probability)
  check_params_type_or_nil(Integer, random_seed: random_seed)
  check_params_positive(reg_param: reg_param, bias_scale: bias_scale, max_iter: max_iter, batch_size: batch_size)
  @params = {}
  @params[:reg_param] = reg_param
  @params[:fit_bias] = fit_bias
  @params[:bias_scale] = bias_scale
  @params[:max_iter] = max_iter
  @params[:batch_size] = batch_size
  @params[:probability] = probability
  @params[:optimizer] = optimizer
  @params[:optimizer] ||= Optimizer::Nadam.new
  @params[:random_seed] = random_seed
  @params[:random_seed] ||= srand
  @weight_vec = nil
  @bias_term = nil
  @prob_param = nil
  @classes = nil
  @rng = Random.new(@params[:random_seed])
end

Instance Attribute Details

#bias_term ⇒ Numo::DFloat (readonly)

Return the bias term (a.k.a. intercept) for SVC.

Returns:

  • (Numo::DFloat) —

    (shape: [n_classes])



35
36
37
 
# File 'lib/svmkit/linear_model/svc.rb', line 35

def bias_term
  @bias_term
end

#classes ⇒ Numo::Int32 (readonly)

Return the class labels.

Returns:

  • (Numo::Int32) —

    (shape: [n_classes])



39
40
41
 
# File 'lib/svmkit/linear_model/svc.rb', line 39

def classes
  @classes
end

#rng ⇒ Random (readonly)

Return the random generator for performing random sampling.

Returns:

  • (Random) 


43
44
45
 
# File 'lib/svmkit/linear_model/svc.rb', line 43

def rng
  @rng
end

#weight_vec ⇒ Numo::DFloat (readonly)

Return the weight vector for SVC.

Returns:

  • (Numo::DFloat) —

    (shape: [n_classes, n_features])



31
32
33
 
# File 'lib/svmkit/linear_model/svc.rb', line 31

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_samples, n_features]) The samples to compute the scores.

Returns:

  • (Numo::DFloat) —

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



128
129
130
131
 
# File 'lib/svmkit/linear_model/svc.rb', line 128

def decision_function(x)
  check_sample_array(x)
  x.dot(@weight_vec.transpose) + @bias_term
end

#fit(x, y) ⇒ SVC

Fit the model with given training data.

Parameters:

  • x (Numo::DFloat) —

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

  • y (Numo::Int32) —

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

Returns:

  • (SVC) —

    The learned classifier itself.



86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
 
# File 'lib/svmkit/linear_model/svc.rb', line 86

def fit(x, y)
  check_sample_array(x)
  check_label_array(y)
  check_sample_label_size(x, y)

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

  if n_classes > 2
    @weight_vec = Numo::DFloat.zeros(n_classes, n_features)
    @bias_term = Numo::DFloat.zeros(n_classes)
    @prob_param = Numo::DFloat.zeros(n_classes, 2)
    n_classes.times do |n|
      bin_y = Numo::Int32.cast(y.eq(@classes[n])) * 2 - 1
      weight, bias = binary_fit(x, bin_y)
      @weight_vec[n, true] = weight
      @bias_term[n] = bias
      @prob_param[n, true] = if @params[:probability]
                               SVMKit::ProbabilisticOutput.fit_sigmoid(x.dot(weight.transpose) + bias, bin_y)
                             else
                               Numo::DFloat[1, 0]
                             end
    end
  else
    negative_label = y.to_a.uniq.min
    bin_y = Numo::Int32.cast(y.ne(negative_label)) * 2 - 1
    @weight_vec, @bias_term = binary_fit(x, bin_y)
    @prob_param = if @params[:probability]
                    SVMKit::ProbabilisticOutput.fit_sigmoid(x.dot(@weight_vec.transpose) + @bias_term, bin_y)
                  else
                    Numo::DFloat[1, 0]
                  end
  end

  self
end

#marshal_dump ⇒ Hash

Dump marshal data.

Returns:

  • (Hash) —

    The marshal data about SVC.



168
169
170
171
172
173
174
175
 
# File 'lib/svmkit/linear_model/svc.rb', line 168

def marshal_dump
  { params: @params,
    weight_vec: @weight_vec,
    bias_term: @bias_term,
    prob_param: @prob_param,
    classes: @classes,
    rng: @rng }
end

#marshal_load(obj) ⇒ nil

Load marshal data.

Returns:

  • (nil) 


179
180
181
182
183
184
185
186
187
 
# File 'lib/svmkit/linear_model/svc.rb', line 179

def marshal_load(obj)
  @params = obj[:params]
  @weight_vec = obj[:weight_vec]
  @bias_term = obj[:bias_term]
  @prob_param = obj[:prob_param]
  @classes = obj[:classes]
  @rng = obj[:rng]
  nil
end

#predict(x) ⇒ Numo::Int32

Predict class labels for samples.

Parameters:

  • x (Numo::DFloat) —

    (shape: [n_samples, n_features]) The samples to predict the labels.

Returns:

  • (Numo::Int32) —

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



137
138
139
140
141
142
143
144
145
 
# File 'lib/svmkit/linear_model/svc.rb', line 137

def predict(x)
  check_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)
  Numo::Int32.asarray(Array.new(n_samples) { |n| @classes[decision_values[n, true].max_index] })
end

#predict_proba(x) ⇒ Numo::DFloat

Predict probability for samples.

Parameters:

  • x (Numo::DFloat) —

    (shape: [n_samples, n_features]) The samples to predict the probailities.

Returns:

  • (Numo::DFloat) —

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



151
152
153
154
155
156
157
158
159
160
161
162
163
164
 
# File 'lib/svmkit/linear_model/svc.rb', line 151

def predict_proba(x)
  check_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
  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