#python

More of a general question about community. I want to draw a pie plot, in a package/rendering engine that is not #matplotlib . But I know that matplotlib does do the math I need.

Theoretically, the "correct" approach would be to isolate that math, make a new package and hook it in so that both matplotlib and my new package can now use the same math, same package. I can reuse the math I need without their rendering assumptions.

But I don't think they would do this. (1/2) ...

Here’s Google search interest for “remove president” from 2020 to 2025.

Not sure if I will use it for anything, but it was fun to make!

I plotted a graph with some very small numbers. The scale of the vertical axis is listed as "1e-5+5.802e-1". What does this mean? I understand scientific notation, but not a sum of two numbers in this context. Where do the invisible parentheses go?

#30DayChartChallenge Day1 Fractions - animal rescue incidents attended by the London fire brigade.

5 of 10 animals rescued are cats, but there are some interesting differences between boroughs - birds in Westminster, horse in Bexley, foxes in Hammersmith.

Made in #python #matplotlib using custom svg markers + #pyfonts to read in Google font.

Full code https://github.com/Lisa-Ho/small-data-projects?tab=readme-ov-file#042025-animal-rescue-incidents

Seeking recommendations for a #WebMapping tutorial / course?

Slightly at sea on where to start.

- My current JS skill level is _extreme novice_.
- I don't have access to ArcGIS.
- Comfortable with #QGIS [*] and the #python #geospatial ecosystem (#geopandas #xarray #rasterio and plotting with #matplotlib)

Suggestions welcome. TIA.

* I have looked at the qgis2web plugin, but having some issues associated with my aged laptop (2012 mbp running Ubuntu) and a 'Wayland session'.

Weekend viz. Marvel money tree showing box office sales by film and series.

Avengers are the most succesful series in in terms of box office sales with Avengers: End Game on top ($2,797m). Captain Marvel is the most succesful standalone film ($1,129m).

Data from Information is Beautiful (up until Jun 2023). Visual made 100% in #python #matplotlib.

Full code here (not pretty though ) https://github.com/Lisa-Ho/small-data-projects/blob/main/README.md#032025-marvel-money-tree

"Plotando estatísticas básicas com #Polars e #Matplotlib - #NLP 04 " @dunossauro #Python

https://www.youtube.com/watch?v=4HpSFIekqDw

Hoje eu aprendi uma ideia ótima do Dunossauro que é imaginar que o ax do fig, ax do matplotlib (axis/eixo) é como uma haste onde penduramos as coisas! Como é fundamental o trabalho dele pra nossa comunidade.

Update: inicialmente achei que era uma tradução corrente mas ele me explicou que não.

Only today found out that there’s a built-in function for labeling bars in Matplotlib.

It's been there since version 3.4. of Matplotlib, out in 2021
https://matplotlib.org/stable/gallery/lines_bars_and_markers/bar_label_demo.html#sphx-glr-gallery-lines-bars-and-markers-bar-label-demo-py
#Matplotlib

Did anyone here have trouble with #uv using a #Python build that breaks interactive #matplotlib and/or #tkinter? I think it might be breaking #FreeSimpleGUI too :(

https://github.com/astral-sh/uv/issues/6893

Update: Also https://github.com/astral-sh/python-build-standalone/issues/146

If you have been using #py5 for a while, this page about #matplotlib integration is a documentation gem It opens up the possibility of making "live", real time and interactive, maplotlib charts, but even if you are not into #dataviz, it shows the beautiful #profiling tools integrated with py5 and how to use #threading to improve performance. In the end you also learn about named colors and the clever "Colormap Color Mode" feature.

https://www.py5coding.org/integrations/matplotlib.html #python #processing4 #profilers #colormapps

On a mapping run - Bauhaus inspired grid map of europe. Each country is coloured by the first letter of their ISO name.

Maybe a little puzzle to figure out the grid I used

Initial map made in #python using #matplotlib then refined in Figma. Code: https://github.com/Lisa-Ho/small-data-projects/tree/main?tab=readme-ov-file#012025-grid-map-of-europe

#gnuplot is great. I've been feeding the results of #sqlite queries into it via org-babel, and it works almost perfectly; the only exception being that I can't use column names in the gnuplot dataset.

Maybe I'll write a blog post about that... In some moderately distant future.

It feels much less accessible compared to #matplotlib, but not more so than #emacs, I guess. And it's great not to carry any dependencies except the gnuplot library, particularly for the Org Mode use case.

The charts sometimes look like a hello from the 90s, but to me it's a plus that they don't give the "matplotlib on defaults" vibe which is omnipresent in modern science :D

Mastodon.social alt text analysis report!

Me and my friend Cristal just published a report on image description usage on mastodon.social, as a group project for the Introduction to Data Science course of the Artificial Intelligence and Sustainable Societies Erasmus Mundus Joint Master program.

Thoughts and feedback are welcome

A huge thanks to @stefan for publishing the dataset on which we based our analysis!

NOTE: We are absolutely aware that the report has very little actual relevance, as the dataset contains a super limited amount of posts from one instance only. It was mainly an experimentation to test our data analysis skills.