def test_counts(counts):
estimator = ndd.estimators.AutoEstimator()
assert ndd.entropy(counts.nk) == approx(
counts.entropy(estimator=estimator))
def test_unnormalized_pmf():
counts = numpy.random.random(size=100) # pylint: disable=no-member
pk = counts / counts.sum()
assert ndd.entropy(counts) == approx(Pmf().entropy_from_pmf(pk))
def test_pmf(pmf):
ref = pmf.entropy
assert ndd.entropy(pmf.pk) == approx(ref)
def test_pmf_with_zeros(pmf_with_zeros):
ref = pmf_with_zeros.entropy
print(pmf_with_zeros.pk.sum())
assert ndd.entropy(pmf_with_zeros.pk) == approx(ref)
def test_nsb(counts_1d, multi_1d):
estimate_from_counts = fnsb.nsb(counts_1d, K)[0]
estimate_from_multiplicities = fnsb.nsb_from_multiplicities(
multi_1d[0], multi_1d[1], K)[0]
assert estimate_from_multiplicities == approx(estimate_from_counts)
def test_PmfPlugin(pmf):
"""Test estimator from PMF."""
estimator = ndd.estimators.PmfPlugin()
ref = pmf.entropy
assert estimator(pmf.pk) == approx(ref)
def test_millermadow_estimator(counts_1d, multi_1d):
est = MillerMadow()
estimate_from_counts = est(counts_1d, k=K)
nk, zk = multi_1d
estimate_from_multiplicities = est(nk, zk=zk, k=K)
assert estimate_from_multiplicities == approx(estimate_from_counts)
def test_nsb_from_multiplicities(data_1d):
frequencies = compute_frequencies(data_1d)
hn, hz = compute_multiplicities(data_1d)
estimate_from_counts = fnsb.nsb(frequencies, K)[0]
estimate_from_multiplicities = fnsb.nsb_from_multiplicities(hn, hz, K)[0]
assert estimate_from_multiplicities == approx(estimate_from_counts)
def test_plugin_estimator(counts_1d, multi_1d):
est = Plugin()
estimate_from_counts = est(counts_1d, k=K)
nk, zk = multi_1d
estimate_from_multiplicities = est(nk, zk=zk, k=K)
assert estimate_from_multiplicities == approx(estimate_from_counts)
def test_grassberger_estimator(counts_1d, multi_1d):
est = Grassberger()
estimate_from_counts = est(counts_1d, k=K)
nk, zk = multi_1d
estimate_from_multiplicities = est(nk, zk=zk, k=K)
assert estimate_from_multiplicities == approx(estimate_from_counts)
def test_ww_estimator(counts_1d, multi_1d):
alpha = 0.1
estimate_from_counts = Nsb(alpha=alpha)(counts_1d, k=K)
nk, zk = multi_1d
estimate_from_multiplicities = Nsb(alpha=alpha)(nk, zk=zk, k=K)
assert estimate_from_multiplicities == approx(estimate_from_counts)
def test_ww(counts_1d, multi_1d):
alpha = 0.1
estimate_from_counts = fnsb.ww(counts_1d, K, alpha)[0]
estimate_from_multiplicities = fnsb.ww_from_multiplicities(
multi_1d[0], multi_1d[1], K, alpha)[0]
assert estimate_from_multiplicities == approx(estimate_from_counts)
def test_nsb_estimator(counts_1d, multi_1d):
estimate_from_counts = Nsb()(counts_1d, k=K)
nk, zk = multi_1d
estimate_from_multiplicities = Nsb()(nk, zk=zk, k=K)
assert estimate_from_multiplicities == approx(estimate_from_counts)