def test_urand_seed_reproduction(sut, shape_1d):
# Arrange
sut_data, data = generate_data(sut, shape_1d, float)
seed = 0
# Act
sut_seed_ndarray = []
seed_ndarray = []
for i in range(2):
sut.urand(sut_data, seed)
backend.urand(data, seed)
sut_seed_ndarray.append(sut.to_ndarray(sut_data))
seed_ndarray.append(backend.to_ndarray(data))
# Assert
np.testing.assert_array_equal(seed_ndarray[0], seed_ndarray[1])
np.testing.assert_array_equal(sut_seed_ndarray[0], sut_seed_ndarray[1])
def test_urand(sut, seed, shape_1d):
# Arrange
sut_data, data = generate_data(sut, shape_1d, float)
sut_idx, idx = generate_idx(sut, shape_1d, 'asc')
length = shape_1d[0]
# Act
sut.urand(sut_data, seed)
backend.urand(data, seed)
# Assert
assert sut_data.shape == data.shape
assert sut_data.dtype == data.dtype
assert backend.amin(data, idx, length) >= 0
assert backend.amax(data, idx, length) <= 1
assert sut.amin(sut_data, sut_idx, length) >= 0
assert sut.amax(sut_data, sut_idx, length) <= 1
def test_urand_seed_response(sut, shape_1d):
# Arrange
sut_data, data = generate_data(sut, shape_1d, float)
# Act
sut_seed_ndarray = []
seed_ndarray = []
for i in range(2):
sut.urand(sut_data, seed=i)
backend.urand(data, seed=i)
sut_seed_ndarray.append(sut.to_ndarray(sut_data))
seed_ndarray.append(backend.to_ndarray(data))
# Assert
TestMathsMethods.assert_array_significantly_different(
seed_ndarray[0], seed_ndarray[1])
TestMathsMethods.assert_array_significantly_different(
sut_seed_ndarray[0], sut_seed_ndarray[1])