Module: SVMKit::PairwiseMetric

Defined in:
lib/svmkit/pairwise_metric.rb

Overview

Module for calculating pairwise distances, similarities, and kernels.

Class Method Summary collapse

Class Method Details

.euclidean_distance(x, y = nil) ⇒ NMatrix

Calculate the pairwise euclidean distances between x and y.

Parameters:

  • x (NMatrix) —

    (shape: [n_samples_x, n_features])

  • y (NMatrix) (defaults to: nil) —

    (shape: [n_samples_y, n_features])

Returns:

  • (NMatrix) —

    (shape: [n_samples_x, n_samples_x] or [n_samples_x, n_samples_y] if y is given)



10
11
12
13
14
15
16
17
18
19
 
# File 'lib/svmkit/pairwise_metric.rb', line 10

def euclidean_distance(x, y = nil)
  y = x if y.nil?
  sum_x_vec = (x**2).sum(1)
  sum_y_vec = (y**2).sum(1)
  dot_xy_mat = x.dot(y.transpose)
  distance_matrix = dot_xy_mat * -2.0 +
                    sum_x_vec.repeat(y.shape[0], 1) +
                    sum_y_vec.transpose.repeat(x.shape[0], 0)
  distance_matrix.abs.sqrt
end

.linear_kernel(x, y = nil) ⇒ NMatrix

Calculate the linear kernel between x and y.

Parameters:

  • x (NMatrix) —

    (shape: [n_samples_x, n_features])

  • y (NMatrix) (defaults to: nil) —

    (shape: [n_samples_y, n_features])

Returns:

  • (NMatrix) —

    (shape: [n_samples_x, n_samples_x] or [n_samples_x, n_samples_y] if y is given)



39
40
41
42
 
# File 'lib/svmkit/pairwise_metric.rb', line 39

def linear_kernel(x, y = nil)
  y = x if y.nil?
  x.dot(y.transpose)
end

.polynomial_kernel(x, y = nil, degree = 3, gamma = nil, coef = 1) ⇒ NMatrix

Calculate the polynomial kernel between x and y.

Parameters:

  • x (NMatrix) —

    (shape: [n_samples_x, n_features])

  • y (NMatrix) (defaults to: nil) —

    (shape: [n_samples_y, n_features])

  • degree (Integer) (defaults to: 3) —

    The parameter of polynomial kernel.

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

    The parameter of polynomial kernel, if nil it is 1 / n_features.

  • coef (Integer) (defaults to: 1) —

    The parameter of polynomial kernel.

Returns:

  • (NMatrix) —

    (shape: [n_samples_x, n_samples_x] or [n_samples_x, n_samples_y] if y is given)



52
53
54
55
56
 
# File 'lib/svmkit/pairwise_metric.rb', line 52

def polynomial_kernel(x, y = nil, degree = 3, gamma = nil, coef = 1)
  y = x if y.nil?
  gamma ||= 1.0 / x.shape[1]
  (x.dot(y.transpose) * gamma + coef)**degree
end

.rbf_kernel(x, y = nil, gamma = nil) ⇒ NMatrix

Calculate the rbf kernel between x and y.

Parameters:

  • x (NMatrix) —

    (shape: [n_samples_x, n_features])

  • y (NMatrix) (defaults to: nil) —

    (shape: [n_samples_y, n_features])

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

    The parameter of rbf kernel, if nil it is 1 / n_features.

Returns:

  • (NMatrix) —

    (shape: [n_samples_x, n_samples_x] or [n_samples_x, n_samples_y] if y is given)



27
28
29
30
31
32
 
# File 'lib/svmkit/pairwise_metric.rb', line 27

def rbf_kernel(x, y = nil, gamma = nil)
  y = x if y.nil?
  gamma ||= 1.0 / x.shape[1]
  distance_matrix = euclidean_distance(x, y)
  ((distance_matrix**2) * -gamma).exp
end

.sigmoid_kernel(x, y = nil, gamma = nil, coef = 1) ⇒ NMatrix

Calculate the sigmoid kernel between x and y.

Parameters:

  • x (NMatrix) —

    (shape: [n_samples_x, n_features])

  • y (NMatrix) (defaults to: nil) —

    (shape: [n_samples_y, n_features])

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

    The parameter of polynomial kernel, if nil it is 1 / n_features.

  • coef (Integer) (defaults to: 1) —

    The parameter of polynomial kernel.

Returns:

  • (NMatrix) —

    (shape: [n_samples_x, n_samples_x] or [n_samples_x, n_samples_y] if y is given)



65
66
67
68
69
 
# File 'lib/svmkit/pairwise_metric.rb', line 65

def sigmoid_kernel(x, y = nil, gamma = nil, coef = 1)
  y = x if y.nil?
  gamma ||= 1.0 / x.shape[1]
  (x.dot(y.transpose) * gamma + coef).tanh
end