Class: Lda::DataDocument

Inherits:
Document show all
Defined in:
lib/lda-ruby/document/data_document.rb

Instance Attribute Summary

Attributes inherited from Document

#corpus, #counts, #length, #tokens, #total, #words

Instance Method Summary collapse

Methods inherited from Document

#handle, #has_text?, #recompute, #tokenize

Constructor Details

#initialize(corpus, data) ⇒ DataDocument

Returns a new instance of DataDocument.



16
17
18
19
20
21
22
23
24
25
26
27
28
# File 'lib/lda-ruby/document/data_document.rb', line 16

def initialize(corpus, data)
  super(corpus)

  items = data.split(/\s+/)
  pairs = items[1..items.size].map { |item| item.split(':') }

  pairs.each do |feature_identifier, feature_weight|
    @words << feature_identifier.to_i
    @counts << feature_weight.to_i
  end

  recompute
end