Switch to np array not needed when reading in data

[sstevens2]

How could the content be improved?

maize_roots = np.array(maize_roots) isn't needed. The data reads in as a np array.

Which part of the content does your suggestion apply to?

https://datacarpentry.github.io/image-processing/03-skimage-images.html

[K-Meech]

Thanks for opening these issues @sstevens2 - really appreciate the suggestions! We'll discuss these issues at our next maintainer meeting and make some updates 😄

[sstevens2]

Let me know if you'd like me to put in a PR for anything. I'm certainly willing to do so for smaller suggestions. Not sure I have the time for any bigger PRs at the moment though

[uschille]

Thanks @sstevens2! I recall that, at some point, there was an issue with the array being read only/immutable. The lines img = np.array(img) were added to make the array mutable. We should check what the current module versions do and perhaps add a callout.

[mkcor]

Right, details are provided in the callout note: https://github.com/datacarpentry/image-processing/blob/main/episodes/03-skimage-images.md#loading-images-with-imageio-read-only-arrays

The note mentions "in certain situations" and I think it's safer to cover all situations, since we are not super strict setup-wise (i.e., file environment.yml lists imageio as a dependency, without specifying a pinned version). Therefore, I would advise against removing the img = np.array(img) lines.