implemented hapax legomena index

src/TRUNAJOD/ttr.py

@@ -200,3 +200,30 @@ def d_estimate(
  y = ttrs ** 2
  d = np.linalg.lstsq(A, y, rcond=None)[0]
  return d[0]
+
+
+def hapax_legomena_index(doc: Doc) -> int:
+ """Hapax Legomena Index from a text.
+
+ Hapax Legomena Index is the number of words occuring once in a text.
+
+ :param doc: Processed spaCy Doc
+ :type doc: Doc
+ :return: Texts' Hapex Legomena Index
+ :rtype: int
+ """
+ word_counter = 0
+ word_dupe_counter = 0
+ words = {}
+ for token in doc:
+ if is_word(token.pos_):
+ word_counter += 1
+ if str(token.pos_) not in words:
+ words[str(token.pos_)] = 1
+ else:
+ words[str(token.pos_)] += 1
+
+ for key, value in words.items():
+ if int(value) > 1:
+ word_dupe_counter += int(value)
+ return word_counter - word_dupe_counter

tests/ttr_test.py

@@ -90,3 +90,19 @@ def test_d_estimate():
 
  np.random.seed(0)
  assert ttr.d_estimate(doc) == 119.4468681409897
+
+
+def test_hapax_legomena_index():
+ """Test hapax_legomena_index."""
+ Token = namedtuple("Token", "lemma_ pos_")
+ doc = [
+ Token("hola", "hola"),
+ Token("hola", "hola"),
+ Token("chao", "chao"),
+ Token("hola", "hola"),
+ Token("perro", "perro"),
+ Token("hola", "hola"),
+ ]
+
+ answer = 2
+ assert ttr.hapax_legomena_index(doc) == answer