def test_small_input_b(self):
py_similarity = similarities.dice_coefficient_python(
(self.filters1, self.filters2[:10]), self.default_threshold,
self.default_k)
c_similarity = similarities.dice_coefficient_accelerated(
(self.filters1, self.filters2[:10]), self.default_threshold,
self.default_k)
self.assert_similarity_matrices_equal(py_similarity, c_similarity)
def test_same_score_k_none(self):
cffi_cands = similarities.dice_coefficient_accelerated(
self.filters, self.default_threshold, None)
cffi_scores, _ = cffi_cands
python_cands = similarities.dice_coefficient_python(
self.filters, self.default_threshold, None)
python_scores, _ = python_cands
assert cffi_scores == python_scores
def test_same_score(self):
c_cands = similarities.dice_coefficient_accelerated(
self.filters, self.default_threshold, self.default_k)
c_scores, _ = c_cands
python_cands = similarities.dice_coefficient_python(
self.filters, self.default_threshold, self.default_k)
python_scores, _ = python_cands
assert c_scores == python_scores
def test_not_multiple_of_64(self, k, threshold, bytes_n):
datasets = [[bitarray('01001011') * bytes_n],
[bitarray('01001011') * bytes_n]]
py_similarity = similarities.dice_coefficient_python(
datasets, self.default_threshold, k)
c_similarity = similarities.dice_coefficient_accelerated(datasets,
threshold,
k=k)
self.assert_similarity_matrices_equal(py_similarity, c_similarity)
def test_memory_use(self):
n = 10
f1 = self.filters1[:n]
f2 = self.filters2[:n]
# If memory is not handled correctly, then this would allocate
# several terabytes of RAM.
big_k = 1 << 50
py_similarity = similarities.dice_coefficient_python(
(f1, f2), self.default_threshold, big_k)
c_similarity = similarities.dice_coefficient_accelerated(
(f1, f2), self.default_threshold, big_k)
self.assert_similarity_matrices_equal(py_similarity, c_similarity)
def test_cffi_manual(self):
nl = randomnames.NameList(30)
s1, s2 = nl.generate_subsets(5, 1.0)
keys = generate_key_lists(('test1', 'test2'), len(nl.schema_types))
f1 = tuple(
f[0]
for f in bloomfilter.stream_bloom_filters(s1, keys, nl.SCHEMA))
f2 = tuple(
f[0]
for f in bloomfilter.stream_bloom_filters(s2, keys, nl.SCHEMA))
py_similarity = similarities.dice_coefficient_python(
(f1, f2), self.default_threshold, self.default_k)
c_similarity = similarities.dice_coefficient_accelerated(
(f1, f2), self.default_threshold, self.default_k)
self.assert_similarity_matrices_equal(py_similarity, c_similarity)
def test_cffi(self):
similarity = similarities.dice_coefficient_accelerated(
self.filters, self.default_threshold, self.default_k)
self._check_proportion(similarity)
def test_not_multiple_of_64(self, k, threshold, bytes_n):
datasets = [[bitarray('01001011') * bytes_n],
[bitarray('01001011') * bytes_n]]
with pytest.raises(NotImplementedError):
similarities.dice_coefficient_accelerated(datasets, threshold, k=k)
def test_not_multiple_of_8_raises(self, ):
datasets = [[bitarray('010')], [bitarray('010')]]
with pytest.raises(NotImplementedError):
similarities.dice_coefficient_accelerated(
datasets, threshold=self.default_threshold)
