Module: Tools::ClassifierMethods

Included in:
LinearRegression, Perceptron
Defined in:
lib/rubyml/tools.rb

Overview

Methods to test classifier accuracy via K-fold cross validation.

Instance Method Summary collapse

Instance Method Details

#correct_count(ypred, ytest, c, t, n) ⇒ Object 



83
84
85
86
87
88
89
90
 
# File 'lib/rubyml/tools.rb', line 83

def correct_count(ypred, ytest, c, t, n)
  count = 0.0
  ypred.row_count.times do |r|
    count += handle_epsilon(ypred, ytest, r)
  end
  p "Fold #{n} Accuracy: #{(count / ypred.row_count * 100.0).round(3)}%"
  [c + count, t + ypred.row_count]
end

#generate_folds(x, y, num, folds) ⇒ Object 



55
56
57
58
59
60
61
 
# File 'lib/rubyml/tools.rb', line 55

def generate_folds(x, y, num, folds)
  sin = String(num * (x.row_count / folds))
  ein = String([(num + 1) * (x.row_count / folds), x.row_count].min)
  train = generate_train_set(x, y, sin, ein)
  test = generate_test_set(x, y, sin, ein)
  train + test
end

#generate_test_set(x, y, sin, ein) ⇒ Object 



69
70
71
72
73
 
# File 'lib/rubyml/tools.rb', line 69

def generate_test_set(x, y, sin, ein)
  xtest = x[sin + ':' + ein, ':']
  ytest = y[sin + ':' + ein, ':']
  [xtest, ytest]
end

#generate_train_set(x, y, sin, ein) ⇒ Object 



63
64
65
66
67
 
# File 'lib/rubyml/tools.rb', line 63

def generate_train_set(x, y, sin, ein)
  xtrain = x[':' + sin, ':'].vstack(x[ein + ':', ':'])
  ytrain = y[':' + sin, ':'].vstack(y[ein + ':', ':'])
  [xtrain, ytrain]
end

#handle_epsilon(ypred, ytest, r) ⇒ Object 



75
76
77
78
79
80
81
 
# File 'lib/rubyml/tools.rb', line 75

def handle_epsilon(ypred, ytest, r)
  if @epsilon
    ((ypred[r, 0] - ytest[r, 0]).abs < @epsilon ? 1.0 : 0.0)
  else
    (ypred[r, 0] == ytest[r, 0] ? 1.0 : 0.0)
  end
end

#training_accuracy(x, y) ⇒ Object 



92
93
94
95
96
97
98
99
100
101
102
 
# File 'lib/rubyml/tools.rb', line 92

def training_accuracy(x, y)
  correct = 0.0
  total = 0.0
  @folds.times do |n|
    xtrain, ytrain, xtest, ytest = generate_folds(x, y, n, @folds)
    fit(xtrain, ytrain)
    ypred = predict(xtest)
    correct, total = correct_count(ypred, ytest, correct, total, n)
  end
  (correct / total).round(5)
end