def test_line_model_residuals():
model = LineModel()
model._params = (0, 0)
assert_equal(abs(model.residuals(np.array([[0, 0]]))), 0)
assert_equal(abs(model.residuals(np.array([[0, 10]]))), 0)
assert_equal(abs(model.residuals(np.array([[10, 0]]))), 10)
model._params = (5, np.pi / 4)
assert_equal(abs(model.residuals(np.array([[0, 0]]))), 5)
assert_equal(abs(model.residuals(np.array([[np.sqrt(50), 0]]))), 5)
def test_line_model_residuals():
model = LineModel()
model.params = (0, 0)
assert_equal(abs(model.residuals(np.array([[0, 0]]))), 0)
assert_equal(abs(model.residuals(np.array([[0, 10]]))), 0)
assert_equal(abs(model.residuals(np.array([[10, 0]]))), 10)
model.params = (5, np.pi / 4)
assert_equal(abs(model.residuals(np.array([[0, 0]]))), 5)
assert_almost_equal(abs(model.residuals(np.array([[np.sqrt(50), 0]]))), 0)
def test_line_model_residuals():
model = LineModel()
model.params = (0, 0)
assert_equal(model.residuals(np.array([[0, 0]])), 0)
assert_equal(model.residuals(np.array([[0, 10]])), 10)
assert_equal(model.residuals(np.array([[10, 0]])), 0)
model.params = (1, 0)
assert_equal(model.residuals(np.array([[0, 0]])), 0)
assert_almost_equal(model.residuals(np.array([[1, 0]])), np.sqrt(2) / 2.)
def test_line_model_estimate():
# generate original data without noise
model0 = LineModel()
model0.params = (0, 1)
x0 = np.arange(-100, 100)
y0 = model0.predict_y(x0)
data0 = np.column_stack([x0, y0])
data = data0 + (np.random.random(data0.shape) - 0.5)
# estimate parameters of noisy data
model_est = LineModel()
model_est.estimate(data)
assert_almost_equal(model_est.residuals(data0), np.zeros(len(data)), 1)
# test whether estimated parameters almost equal original parameters
random_state = np.random.RandomState(1234)
x = random_state.rand(100, 2)
assert_almost_equal(model0.predict_y(x), model_est.predict_y(x), 1)
def keep_border_points(cloud_pts, convex_hull_threshold):
''' Return the points that are localized onto the convex hull of the cloud
of input points.
Input and output are (N,2) numpy array '''
convexHulls = convex_hull(cloud_pts.tolist())
convexHulls[:+1][:] = convexHulls[0][:]
nsegments = len(convexHulls)
to_keep = np.zeros(len(cloud_pts), dtype=bool)
for P1, P2 in zip(convexHulls[:-1][:], convexHulls[1:][:]):
data = np.asarray([P1, P2])
model = LineModel()
model.estimate(data)
to_keep = np.logical_or(
to_keep,
abs(model.residuals(cloud_pts)) < convex_hull_threshold)
return cloud_pts[to_keep, :]
