fix issue with plotting negative-sum images

[ncullen93]

If an image has a sum less than zero, then ants.plot rejects it. This is due to code that incorrectly identifies whether an image is all-zero.

This should work:

import ants
img = ants.image_read(ants.get_ants_data('r16'))
img2 = (img - img.mean()) / img.std()
img2.plot()

[stnava]

thanks for working on this - it has been on my todo list for several years

[cookpa]

This causes CI tests to fail, try

np.all(np.equal(image.numpy(), 0.0))

[cookpa]

Running viz tests
E
======================================================================
ERROR: test_plot_example (__main__.TestModule_plot)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/Users/runner/work/ANTsPy/ANTsPy/tests/test_viz.py", line 33, in test_plot_example
    ants.plot(img, filename=filename)
  File "/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages/ants/viz/plot.py", line 231, in plot
    if sum(sum(image != 0)) == 0:
ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

[stnava]

also - I dont this this should raise an error but instead fail "gracefully"

[cookpa]

@stnava you changed it to output a warning instead, which I think is the right call. #562 is just improving the test. It just needs to reduce to a scalar true / false to avoid ambiguity

[stnava]

ok

[cookpa]

@ncullen93 I can fix this real quick, now it looks like all the other runners are done

[cookpa]

#566

[ncullen93]

Yes, thanks. Can you consider reducing the number of python versions tested in the CI to 1 or 2 instead of 5? It takes so long otherwise and doesn't add much. I will try to wait for it to finish going forward though.

[cookpa]

Yeah, it can get painful when there's a lot of activity. I will have a think about how we can do better.

One option might be to put branch protection on master, and insist that all changes go through PR, and then build a couple of wheels in the PR. That would avoid duplication where PR builds pass and then the same wheels get built again on merge