It does not produce identical results. Compare the second plot where the "innermost" part of the top-right contour is no longer yellow.

With that said, I do not know if it should be there, etc, or if one should add an example to show the difference with using the origin keyword.

@oscargus which versions are you comparing? There seems to have been a bug in recent Matplotlib versions (but not main or the release candidate) that caused that yellow not to appear. I already confused myself with that a couple of days ago: https://discourse.matplotlib.org/t/cmap-set-over-and-set-under-not-working-in-newer-python-environments/23999

I also wondered if it worth adding an example to show use of origin, and I see @gautamsagar99 asked the same on the issue (#26600). I know there is also a tension with “how much is too much to put in an example”.

Ahh, sorry I mixed up the versions (checked on my phone)...

I also wondered if it worth adding an example to show use of origin

I think here it'd be useful to clarify what sounds like a point of confusion. I'd also make it a standalone "origin" section of an existing example?

Hey,

To clarify I am using matplotlib - 3.7.1 version and I used colab for testing all the examples.

Here is the link for colab:

https://colab.research.google.com/drive/1XnsYLa8Tr0zuLe4hRDRu4KBJcJKlEnXX?usp=sharing

If you want me to add a new section with the below example I can add it.

import matplotlib.pyplot as plt
import numpy as np

x = np.arange(1, 10)
y = x.reshape(-1, 1)
h = x * y

cs = plt.contourf(h, levels=[10, 30, 50],
colors=['#808080', '#A0A0A0', '#C0C0C0'], extend='both', origin="upper")
cs.cmap.set_over('red')
cs.cmap.set_under('blue')
cs.changed()

building on your example, I think a side by side might be clearest:

x = np.arange(1, 10)
y = x.reshape(-1, 1)
h = x * y

fog, (ax1, ax2) = plt.subplots(ncols=2)
for (ax, origin) in [(ax1, 'upper'), (ax2, 'lower')]:
  ax.set_title(f"origin={origin}")
  cs = ax.contourf(h, levels=[10, 30, 50],
      colors=['#808080', '#A0A0A0', '#C0C0C0'], extend='both', origin=origin)
  cs.cmap.set_over('red')
  cs.cmap.set_under('blue')
  cs.changed()

and I'm a big fan here of using sphinx headings for each of the examples 'cause they all demonstrate a different feature?

@rcomer Let me add a new section below which clearly populates the example using origin keyword.

@story645 Yes we can add side by side which will be clearly differentiating for lower and upper origin value.

Let me Just add it and push it.

Thanks

Finally I want to add The below section to the documentation:

"**Customizing Contour Plots with the origin Keyword: **

This section demonstrates how to use the origin parameter in Matplotlib's contourf function to adjust the positioning of grid data. The origin parameter specifies whether data coordinates originate from the upper left ('upper') or lower left ('lower') corner of the grid.

import matplotlib.pyplot as plt
import numpy as np

x = np.arange(1, 10)
y = x.reshape(-1, 1)
h = x * y

fig, (ax1, ax2) = plt.subplots(ncols=2)

for (ax, origin) in [(ax1, 'upper'), (ax2, 'lower')]:
    ax.set_title(f"origin={origin}")
    cs = ax.contourf(h, levels=[10, 30, 50],
                     colors=['#808080', '#A0A0A0', '#C0C0C0'],
                     extend='both', origin=origin)
    cs.cmap.set_over('red')
    cs.cmap.set_under('blue')
    cs.changed()

plt.show()

This code creates two side-by-side contour plots with different origin settings ('upper' and 'lower'). The contourf function utilizes the origin parameter to control how grid data coordinates are interpreted for plotting. The resulting visualizations highlight the effect of changing the origin on the contour plot's appearance."

@gautamsagar99 seems good to me. Since you're modifying this example, can you also put in headings id'ing the topics of the other examples? I also understand if not because it's out of scope here.

@story645 I will keep this as heading "Customizing Contour Plots with the origin Keyword:", under it I will provide the example.

@story645 I'm Confused regarding the images. Do we need to upload the image for the output(for the above example) or It is going to auto generate?

Auto generate - see https://matplotlib.org/devdocs/devel/document.html#write-examples-and-tutorials

@story645 Thank you for your assistance! I've made the necessary changes and have also generated the documentation. Please review the updates and let me know if any further adjustments are required.

@story645 I have done all the changes I hope everything looks good. Let me know if there is anything if you might want me to modify.

Huge thanks to @story645 and @timhoffm

I wanted to express my sincere appreciation for your tremendous support during my first open source contribution. Your guidance in understanding automatic documentations with proper formatting and your help in fixing my mistakes were incredibly valuable. Thanks for making this journey so enriching!

Thanks! You need to make the change @QuLogic suggested, and also much credit to @rcomer for opening the issue & her and @oscargus for clarifying.

Sure, Let me make that small change and push it.

congrats on your first merged pr 🎊 hope to see you back for more!

Thank you so much for your guidance throughout the process! Contributing to this open-source project has been an amazing learning experience, and I'm really grateful for the opportunity. I completely understand that reviewing and merging PRs is no small task, and your support in guiding me through it means a lot. Looking forward to continuing this journey and contributing more in the future! 🙌😊