def test_remove_non_binary(self):
b1, b2, b3 = (DiscreteVariable(c, values=tuple("12")) for c in "abc")
d1, d2, d3 = (DiscreteVariable(c, values=tuple("123")) for c in "def")
c1, c2 = (ContinuousVariable(c) for c in "xy")
t = StringVariable("t")
domain = Domain([d1, b1, b2, c1], d2, [c2, d3, t, b3])
data = Table.from_domain(domain, 5)
reduced = distance.remove_nonbinary_features(data)
self.assertEqual(reduced.domain.attributes, (b1, b2))
self.assertEqual(reduced.domain.class_var, d2)
self.assertEqual(reduced.domain.metas, (c2, d3, t, b3))
reduced = distance.remove_nonbinary_features(data, to_metas=True)
self.assertEqual(reduced.domain.attributes, (b1, b2))
self.assertEqual(reduced.domain.class_var, d2)
self.assertEqual(reduced.domain.metas, (c2, d3, t, b3, d1, c1))
def _fix_nonbinary():
nonlocal data
if metric is distance.Jaccard and not issparse(data.X):
nbinary = sum(a.is_discrete and len(a.values) == 2
for a in data.domain.attributes)
if not nbinary:
self.Error.no_binary_features()
return False
elif nbinary < len(data.domain.attributes):
self.Warning.ignoring_nonbinary()
data = distance.remove_nonbinary_features(data)
return True
def _fix_nonbinary():
nonlocal data
if metric is distance.Jaccard and not issparse(data.X):
nbinary = sum(a.is_discrete and len(a.values) == 2
for a in data.domain.attributes)
if not nbinary:
self.Error.no_binary_features()
return False
elif nbinary < len(data.domain.attributes):
self.Warning.ignoring_nonbinary()
data = distance.remove_nonbinary_features(data)
return True
