def test_optimal_spearman_is_borda(self):
""" Check that Borda count and Spearman optimal rank aggregation returns the same output on the template matrix.
"""
m_template = pd.read_csv('data/matrix.csv')
m_template.index = m_template['index']
m_template = m_template.drop('index', axis=1).rename_axis(None, axis=0)
borda_rank = rk.rank(rk.borda(m_template), reverse=True)
optimal_spearman_rank = rk.rank(
rk.center(m_template, method='spearman'))
print(borda_rank)
print(optimal_spearman_rank)
np.testing.assert_array_equal(borda_rank, optimal_spearman_rank)
def test_generator():
print('Testing generator...')
G = rk.Generator()
R = [5, 4, 3, 2, 1]
G.fit(R)
print('Judgement matrix')
m = G.sample(n=9)
print(m)
print('Score ranking')
r = rk.score(m)
print(r)
distance = rk.dist(rk.rank(R), rk.rank(r))
correlation = rk.corr(R, r)
print('hamming distance: {}'.format(distance))
print('kendall tau correlation: {}'.format(correlation))
print('Uninominal ranking')
r = rk.uninominal(m)
print(r)
distance = rk.dist(rk.rank(R), rk.rank(r))
correlation = rk.corr(R, r)
print('hamming distance: {}'.format(distance))
print('kendall tau correlation: {}'.format(correlation))
def test_rank(self):
rank_M = np.array([[2., 3., 3.], [1., 1., 1.], [4., 2., 2.],
[3., 4., 4.], [5., 5., 5.]])
np.testing.assert_array_equal(rk.rank(self.__class__.M), rank_M)