ppr.py

def PPR(ref, hyp):

  def find_words_before_commas(tokens):
    words_before_commas = set()
    for i, token in enumerate(tokens):
      if token == "," and i > 0:
        word = tokens[i-1]
        words_before_commas.add(word)
    return words_before_commas

  hyp_words_before_commas = find_words_before_commas(hyp.split())
  ref_words_before_commas = find_words_before_commas(ref.split())
  diff = hyp_words_before_commas.symmetric_difference(ref_words_before_commas)

  return len(diff)