def test_multi_scale_loss_kernel():
"""
Test multi-scale loss kernel returns the appropriate
loss tensor for same inputs and jaccard cal.
"""
loss_values = np.asarray([1, 2, 3])
array_eye = np.identity((3))
tensor_pred = np.zeros((3, 3, 3, 3))
tensor_eye = np.zeros((3, 3, 3, 3))
tensor_eye[:, :, 0:3, 0:3] = array_eye
tensor_pred[:, :, 0, 0] = array_eye
tensor_eye = tf.convert_to_tensor(tensor_eye, dtype=tf.double)
tensor_pred = tf.convert_to_tensor(tensor_pred, dtype=tf.double)
list_losses = np.array([
label.single_scale_loss(
y_true=label.separable_filter3d(tensor_eye,
label.gauss_kernel1d(s)),
y_pred=label.separable_filter3d(tensor_pred,
label.gauss_kernel1d(s)),
loss_type="jaccard",
) for s in loss_values
])
expect = np.mean(list_losses, axis=0)
get = label.multi_scale_loss(tensor_eye, tensor_pred, "jaccard",
loss_values)
assert assertTensorsEqual(get, expect)
def test_gauss_kernel1d_0():
"""
Testing case where sigma = 0, expect 0 return
"""
sigma = tf.constant(0, dtype=tf.float32)
expect = tf.constant(0, dtype=tf.float32)
get = label.gauss_kernel1d(sigma)
assert get == expect
def test_gauss_kernel1d_else():
"""
Testing case where sigma is not 0,
expect a tensor returned.
"""
sigma = 3
get = tf.cast(label.gauss_kernel1d(sigma), dtype=tf.float64)
list_vals = range(-sigma * 3, sigma * 3 + 1)
exp = [np.exp(-0.5 * x**2 / sigma**2) for x in list_vals]
expect = tf.convert_to_tensor(exp, dtype=tf.float64)
expect = expect / tf.reduce_sum(expect)
assert assertTensorsEqual(get, expect)
def test_gauss_kernel1d_else():
"""
Testing case where sigma is not 0,
expect a tensor returned.
"""
sigma = 3
get = tf.cast(label.gauss_kernel1d(sigma), dtype=tf.float32)
expect = [
np.exp(-0.5 * x**2 / sigma**2)
for x in range(-sigma * 3, sigma * 3 + 1)
]
expect = tf.convert_to_tensor(expect, dtype=tf.float32)
expect = expect / tf.reduce_sum(expect)
assert is_equal_tf(get, expect)