Class: Eps::LabelEncoder

Inherits:
Object
  • Object
show all
Defined in:
lib/eps/label_encoder.rb

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize ⇒ LabelEncoder

Returns a new instance of LabelEncoder.



5
6
7
 
# File 'lib/eps/label_encoder.rb', line 5

def initialize
  @labels = {}
end

Instance Attribute Details

#labels ⇒ Object (readonly)

Returns the value of attribute labels.



3
4
5
 
# File 'lib/eps/label_encoder.rb', line 3

def labels
  @labels
end

Instance Method Details

#fit(y) ⇒ Object 



9
10
11
12
13
14
15
 
# File 'lib/eps/label_encoder.rb', line 9

def fit(y)
  labels = {}
  y.compact.map(&:to_s).uniq.sort.each_with_index do |label, i|
    labels[label] = i
  end
  @labels = labels
end

#fit_transform(y) ⇒ Object 



17
18
19
20
 
# File 'lib/eps/label_encoder.rb', line 17

def fit_transform(y)
  fit(y)
  transform(y)
end

#inverse_transform(y) ⇒ Object 



38
39
40
41
42
43
 
# File 'lib/eps/label_encoder.rb', line 38

def inverse_transform(y)
  inverse = @labels.map(&:reverse).to_h
  y.map do |yi|
    inverse[yi.to_i]
  end
end

#transform(y) ⇒ Object 



22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
 
# File 'lib/eps/label_encoder.rb', line 22

def transform(y)
  y.map do |yi|
    if yi.nil?
      nil
    else
      # use an additional label for unseen values
      # this is only used during training for the LightGBM eval_set
      # LightGBM ignores them (only uses seen categories for predictions)
      # https://github.com/microsoft/LightGBM/issues/1936
      # the evaluator also ignores them (to be consistent with LightGBM)
      # but doesn't use this code
      @labels[yi.to_s] || @labels.size
    end
  end
end