Cubic interpolation #107
Cubic interpolation #107
Conversation
| @@ -88,4 +88,77 @@ def test_linearity_linear(): | |||
| linear_points = x_and_y @ mapping | |||
|
|
|||
| for t, e in zip(new_x.in_si(), linear_points.in_si()): | |||
There was a problem hiding this comment.
Not really a change made in this branch but this file should be in the test directory (at the root of the repository). This might be something we move later in the refactor branch but it needs to be moved at some point because iirc the CI test runner is not running this test.
There was a problem hiding this comment.
If you want, you can resolve this straight away but lets make sure we make this change once this has been merged.
| test_points = NamedQuantity("x_test", np.linspace(-5000, 5000, 11), units.millimeters) | ||
|
|
||
| mapping, _ = calculate_interpolation_matrix_1d(original_points, test_points, order=InterpolationOptions.CUBIC) |
There was a problem hiding this comment.
Could this logic be abstracted into another function? Both test_cubic_interpolate_matrix_inside , and test_cubic_interpolate_different_units seem to have a lot of repeated logic.
| y_values_expected = y_expected.in_units_of(test_units) | ||
| # | ||
| # print(y_values_test) |
There was a problem hiding this comment.
Does this comment still need to be here? Or if it does, is there perhaps a way to create an if statement so it only makes the plots under a certain condition?
This pr extends
calculate_interpolation_matrix_1dto support cubic interpolation. I've also duplicated the linear interpolation tests to also test the cubic interpolation. Those tests were a significant helpful at catching a boundary condition bug that popped up when I vectorized the code (solved by theifstatement on line 184).