def test_that_it_raises_error_if_wrong_shape(self) -> None:
with pytest.raises(ValueError, match="two-dimensional"):
_ = ProbabilityMatrix(distributions=[])
with pytest.raises(ValueError, match="two-dimensional") as info:
# noinspection PyTypeChecker
_ = ProbabilityMatrix(distributions=[[[]]]) # type: ignore
assert info.match("Expected a two-dimensional")
assert info.match("but received array of shape")
assert info.match("(1, 1, 0)")
def test_that_it_raises_error_if_not_normalized(self) -> None:
with pytest.raises(AssertionError, match="1.0"):
_ = ProbabilityMatrix(distributions=[[0.9, 0.11, 0.0]])
with pytest.raises(AssertionError, match="1.0"):
_ = ProbabilityMatrix(distributions=[[0.9, 0.0, 0.0]])
with pytest.raises(AssertionError, match="1.0"):
_ = ProbabilityMatrix(distributions=[[]])
def test_that_it_raises_if_n_classes_different(self) -> None:
with pytest.raises(ValueError, match="classes") as info:
_ = ClassificationData(
target=ProbabilityMatrix(distributions=[[0.8, 0.1, 0.1, 0.0],
[0.8, 0.1, 0.1, 0.0]]),
pred=ProbabilityMatrix(
distributions=[[0.8, 0.1, 0.1], [0.0, 0.9, 0.1]]),
)
assert info.match("4 classes")
assert info.match("3 classes")
def test_that_it_does_not_smoke(self) -> None:
cd = ClassificationData(
target=ProbabilityMatrix(distributions=[
[0.8, 0.1, 0.1],
[0.2, 0.2, 0.6],
[0.0, 0.9, 0.1],
[0.0, 0.9, 0.1],
]),
pred=ProbabilityMatrix(distributions=[
[0.8, 0.1, 0.1],
[0.2, 0.2, 0.6],
[0.0, 0.9, 0.1],
[0.0, 0.9, 0.1],
]),
)
assert cd.n_samples == 4
assert cd.n_classes == 3
def test_properties(self) -> None:
pm = ProbabilityMatrix(distributions=[
[0.8, 0.1, 0.1],
[0.2, 0.2, 0.6],
[0.0, 0.9, 0.1],
[0.0, 0.9, 0.1],
])
assert pm.n_classes == 3
assert pm.n_samples == 4
npt.assert_equal(pm.argmaxes, desired=[0, 2, 1, 1])
npt.assert_equal(
pm.argmaxes_one_hot,
np.array([[1.0, 0.0, 0.0], [0.0, 0.0, 1.0], [0.0, 1.0, 0.0],
[0.0, 1.0, 0.0]]),
)
def test_that_it_can_be_initialized_from_array(self) -> None:
_ = ProbabilityMatrix(distributions=np.eye(5))
def test_that_it_can_be_initialized_from_nested_list(self) -> None:
_ = ProbabilityMatrix(
distributions=[[0.8, 0.1, 0.1], [0.0, 0.9, 0.1], [0.2, 0.2, 0.6]])