def test_entropy(use_handle):
handle, stream = get_handle(use_handle)
# The outcome of a fair coin is the most uncertain:
# in base 2 the result is 1 (One bit of entropy).
cluster = np.array([0, 1], dtype=np.int32)
assert_almost_equal(entropy(cluster, base=2., handle=handle), 1.)
# The outcome of a biased coin is less uncertain:
cluster = np.array(([0] * 9) + [1], dtype=np.int32)
assert_almost_equal(entropy(cluster, base=2., handle=handle), 0.468995593)
# base e
assert_almost_equal(entropy(cluster, handle=handle), 0.32508297339144826)
def test_entropy_random(n_samples, base, use_handle):
handle, stream = get_handle(use_handle)
clustering, _ = \
generate_random_labels(lambda rng: rng.randint(0, 1000, n_samples))
# generate unormalized probabilities from clustering
pk = np.bincount(clustering)
# scipy's entropy uses probabilities
sp_S = sp_entropy(pk, base=base)
# we use a clustering
S = entropy(np.array(clustering, dtype=np.int32), base, handle=handle)
assert_almost_equal(S, sp_S, decimal=2)
def test_entropy_random(n_samples, base, use_handle):
if has_scipy():
from scipy.stats import entropy as sp_entropy
else:
pytest.skip('Skipping test_entropy_random because Scipy is missing')
handle, stream = get_handle(use_handle)
clustering, _, _, _ = \
generate_random_labels(lambda rng: rng.randint(0, 1000, n_samples))
# generate unormalized probabilities from clustering
pk = np.bincount(clustering)
# scipy's entropy uses probabilities
sp_S = sp_entropy(pk, base=base)
# we use a clustering
S = entropy(np.array(clustering, dtype=np.int32), base, handle=handle)
assert_almost_equal(S, sp_S, decimal=2)
