def _load_docs(cls): test_dir = os.path.dirname(__file__) reuters_ldac_fn = os.path.join(test_dir, 'data', 'reuters.ldac') with open(reuters_ldac_fn, 'r') as f: cls.docs = utils.docs_from_ldac(f) cls.V = utils.num_terms(cls.docs) cls.N = len(cls.docs)
def test_num_terms(): docs = [[0, 1, 2], [1, 2, 3]] assert utils.num_terms(docs) == 4