The coolest thing I ever discovered: Git has an alias "g" 😱

I've started working on an R package that makes it easy to write Typst markup dynamically! #rstats

It's still in its early stages, but it implements:
- ✅ all Typst functions
- ✅ set rules
- ✅ conversion from R to Typst (`NULL` becomes `none`, etc)
- ✅ no dependencies

github.com/y-sunflower/r2typ

I'm starting my company!!

Very happy to announce Yellow Sunflower, an open-source & consulting company that focuses on data science, especially data visualization and storytelling!

Learn more in the announcement blog post: ysunflower.com/blog/introducing-yellow-sunflower

The real difference between R and Python that no one talks about: emotional support

I can't get enough of testthat messages 🥲

Error Lens is a big one for me! It prints errors/warnings directly in the editor

New feature available in dayplot! 📅

You can now easily add captions (and customize them) to the calendar heatmaps.

Now, I think it's time for me to spend more time finding better examples.

Learn more about dayplot: github.com/y-sunflower/...

New release of morethemes 😁🎨 with 2 new themes for matplotlib!

These themes (try to) be more color-blind friendly, as requested by some users.

If you have any ideas for themes, please don't hesitate to reach out!

Browse all themes: y-sunflower.github.io/morethemes/

@matplotlib.org is getting more and more interactivity!

"plotjs" converts any existing matplotlib charts to an interactive, web-based version of them, with quite a lot of customization options.

Try it out here: github.com/y-sunflower/...

More examples: y-sunflower.github.io/plotjs/

I've started working on a minimalist PoC that converts @matplotlib.org plots into interactive graphics.

There are a bazillion things to do before it's something robust and stable, but I'm confident it will go somewhere.

Online demo: y-sunflower.github.io/plotjs/

New #tidytuesday chart! It's all about gas prices

• Highlight the overall trend rather than precise values
• Use a scale-free format
• Keep the design minimalist

Code: github.com/JosephBARBIERDARNAL/tidytuesday/blob/main/src/2025/2025-07-01/main.py

New (small) release of pypalettes! 😁

- A create_cmap() function to easily create a colormap object
- A show_cmap() function to preview the colors
- Improved documentation
- Minor bug fixes and internal code cleanup

More info on the project repository: github.com/JosephBARBIERDARNAL/pypalettes

Is it enough to be educated to have opportunities? That's not for sure.

I literally spent days with @yan-holtz.bsky.social to end up with a simple static image, but that's how dataviz is sometimes.

If you want to learn how to make this type of visualisation, take a look at matplotlib-journey.com

Matplotlib is enormous—but just how big is it, really?

Even other well-known tools like Altair and Plotly haven’t reached Matplotlib’s level of adoption—though they do have their own unique features that Matplotlib doesn’t offer.

Code: github.com/JosephBARBIERDARNAL/anim-manim/blob/main/scene.py

My confidence in my data visualization skills over the past few years has been... interesting.

By the way, I'm really curious to hear about YOUR dataviz journey!

made with the @matplotlib.bsky.social plt.xkcd() tool, which is quite amusing!

@matplotlib.bsky.social is much more than just a data visualization library.

This is ~100 lines of code, using numpy+matplotlib

If you want to learn how to create anything with Matplotlib, the first step is to check out our new project with @yan-holtz.bsky.social: matplotlib-journey.com

Sometimes, I enjoy reinventing the wheel to suit my needs.

So I ended up creating gifing—a minimalist yet customizable GIF maker for Python.

Right now, it’s a simple, foundational version, but I have some cool feature ideas!

repo: github.com/JosephBARBIERDARNAL/gifing