Interpolate on axis

[pietsjoh]

Description of the change

The idea for this feature came from the wish to create the functionality to remove the background from a signal (remove_background_signal). For this problem the signal dimension is usually 1 and the background signal is loaded from a seperate file. The axis of the background signal b doesn't necessarily match the axis of the measured signal s. Thus for the subtraction of the background it would be nice to simply interpolate the b onto the axis of s and then subtract s-b.

This PR aims to solve the interpolation part, which may be more generally useful. The function interpolate_on_axis() takes an axis (UniformDataAxis, FunctionalDataAxis or DataAxis) as input and replaces an axis (specified by an index). The data is interpolated using scipy.interpolate.interp1d in the process. This feature works both on signal and navigation axes and replaces one axes for a signal with arbitrary dimensions. Moreover, it can be used to switch between different axis types.

Progress of the PR

• Change implemented
• handle extrapolation (custom error msg or use scipy bounds_error?)
• which index to use (from set_axis or from axes_manager?)
• update docstring (if appropriate),
• update user guide (if appropriate),
• add an changelog entry in the upcoming_changes folder (see upcoming_changes/README.rst),
• Check formatting changelog entry in the readthedocs doc build of this PR (link in github checks)
• add tests,
• ready for review.

Minimal example of the bug fix or the new feature

import hyperspy.api as hs
import numpy as np

x = {"offset": 0, "scale": 1, "size": 10, "name": "X", "units": "µm", "navigate": True}
s = {"offset": 0, "scale": 1, "size": 50, "name": "S", "units": "eV", "navigate": False}

x_new = UniformDataAxis(offset=1.5, scale=0.8, size=7, navigate=True, name="X", units="nm")
s_new = DataAxis(axis=np.arange(23)**2, navigate=False, name="S", units="eV")

d = np.random.random((10, 50))
s = hs.signals.Signal1D(d, axes=[x, s])

s2 = s.interpolate_on_axis(x_new, 0, inplace=False)
s2.interpolate_on_axis(s_new, 1, inplace=True)

[codecov]

Codecov Report

Patch coverage: 100.00% and project coverage change: +0.12% 🎉

Comparison is base (c1a0de8) 81.25% compared to head (aebbd25) 81.38%.
Report is 140 commits behind head on RELEASE_next_major.

Additional details and impacted files

@@                  Coverage Diff                   @@
##           RELEASE_next_major    #3214      +/-   ##
======================================================
+ Coverage               81.25%   81.38%   +0.12%     
======================================================
  Files                     173      173              
  Lines                   24192    24352     +160     
  Branches                 5621     5677      +56     
======================================================
+ Hits                    19657    19818     +161     
+ Misses                   3236     3222      -14     
- Partials                 1299     1312      +13     

Files Changed	Coverage Δ
hyperspy/signal.py	77.73% <100.00%> (+0.09%)	⬆️

... and 19 files with indirect coverage changes

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.