def test_add_exp():
np.random.seed(3)
raw_data = np.random.randint(10, size=(4, 3))
noisy_data = noise_aug(
"additive_exponential", data, 11, 2, {
"low_bound_gauss": 0.01,
"up_bound_gauss": 0.15,
"scale": 4,
"snr_thres": 10,
"rate_thres": 1
})
assert not (raw_data == noisy_data).all()
def test_mult_ray():
np.random.seed(3)
raw_data = np.random.randint(10, size=(4, 3))
noisy_data = noise_aug(
"multiplicative_rayleigh", data, 11, 2, {
"low_bound_gauss": 0.01,
"up_bound_gauss": 0.15,
"scale": 4,
"snr_thres": 10,
"rate_thres": 1
})
assert not (raw_data == noisy_data).all()
def test_dimension():
""" Checks whether the dimension of the original data is equal to the augmented data. """
raw_data_shape = data.shape
noisy_data_shape = noise_aug(
"multiplicative_exponential", data, 11, 2, {
"low_bound_gauss": 0.01,
"up_bound_gauss": 0.15,
"scale": 4,
"snr_thres": 10,
"rate_thres": 1
}).shape
assert raw_data_shape == noisy_data_shape
def test_random():
""" Checks whether the functions in noise_aug return the same noisy_data """
np.random.seed(3)
augmented1 = noise_aug(
"multiplicative_exponential", data, 11, 2, {
"low_bound_gauss": 0.01,
"up_bound_gauss": 0.15,
"scale": 4,
"snr_thres": 10,
"rate_thres": 1
})
np.random.seed(3)
augmented2 = noise_aug(
"multiplicative_exponential", data, 11, 2, {
"low_bound_gauss": 0.01,
"up_bound_gauss": 0.15,
"scale": 4,
"snr_thres": 10,
"rate_thres": 1
})
assert (augmented1 == augmented2).all()