Leonardo Hansa
LightNews LightNews
leonardohansa.com
Leonardo Hansa
@leonardohansa.com
1K followers 550 following 110 posts
I love chocolate and opera, but it's too expensive so I only watch it on YouTube.
#stats #RStats
Posts Replies Media Videos
leonardohansa.com
Every time I open PowerPoint I think:

"So we're living in the era of Artificial Intelligence"
November 11, 2025 at 9:01 AM
leonardohansa.com
The belief that numbers offer a single, objective description of reality overlooks a crucial truth: data does not speak for itself. Every dataset results from choices, which inevitably reflect implicit, and sometimes ideological, assumptions.

arxiv.org/abs/2507.03628

@freakonometrics.bsky.social
When Numbers Mislead Us
The belief that numbers offer a single, objective description of reality overlooks a crucial truth: data does not speak for itself. Every dataset results from choices-what to measure, how, when, and w...
arxiv.org
October 10, 2025 at 8:28 AM
leonardohansa.com
Lately the sentence I say the most is "I don't remember"
August 31, 2025 at 12:22 PM
leonardohansa.com
What alternative to Google Calendar do you like?

Notion, Proton...?

#android
#app
#calendar
#productivity
August 22, 2025 at 2:36 PM
Reposted by Leonardo Hansa
pydataparis.bsky.social
📢 Talk Announcement

"From Jupyter Notebook to Publish-Ready Report: Effortless Sharing with Quarto", by @cderv.bsky.social from @posit.co.

📜 Talk info: pretalx.com/pydata-paris-2025/talk/K7CAAE
📅 Schedule: pydata.org/paris2025/schedule
🎟 Tickets: pydata.org/paris2025/tickets
August 15, 2025 at 7:01 AM
leonardohansa.com
Have you read this by @anthropic.com ?

Any explanation for dummies?

Are they saying that not-so-random number sequences might hide info only readable by AI?

alignment.anthropic.com/2025/sublimi... #llm
Subliminal Learning: Language Models Transmit Behavioral Traits via Hidden Signals in Data
alignment.anthropic.com
August 1, 2025 at 3:15 PM
leonardohansa.com
Have you ever loved saving money so much that you've WALKED from an airport to the city center?
July 29, 2025 at 1:52 PM
leonardohansa.com
On a ML Explainability course, the author compares one concept with coeffs on a linear regression.

Then says: “if you don't know about linear regression forget about this comparison”.

My question: how on earth someone who doesn't know about linear regression is studying ML?

#ML #stats
July 12, 2025 at 8:08 AM
Reposted by Leonardo Hansa
curiouspolymath.bsky.social
#statistics #mathsky
June 19, 2025 at 11:47 AM
Reposted by Leonardo Hansa
tellarpao.bsky.social
Happy early summer weekend, #stats!
June 20, 2025 at 9:09 AM
leonardohansa.com
«América se tragaba a Europa.

»—Nos comen. Somos pobres, muy pobres. Unos miserables que solo entendemos de tomar el sol.»

La Regenta, Leopoldo Alas, Clarín (1884)
June 5, 2025 at 12:44 PM
leonardohansa.com
We should first define 'dead' for answering the question below.

As examples:
- Is Pascal dead?
- Is Fortran dead?
- Is SPSS dead? (alas this one isn't)
leonardohansa.com Leonardo Hansa @leonardohansa.com · Jun 3
I'm reading lately lots of posts by R users claiming R is not dead.

Does it mean that R is almost dead?

#rstats
June 3, 2025 at 2:13 PM
leonardohansa.com
I'm reading lately lots of posts by R users claiming R is not dead.

Does it mean that R is almost dead?

#rstats
June 3, 2025 at 2:11 PM
leonardohansa.com
On bubble plots, what should the bubble size variable be associated to: the radius or the area?

#dataviz
May 31, 2025 at 5:57 AM
Reposted by Leonardo Hansa
withenoughcoffee.com
Help internet where can I buy this hoodie
May 31, 2025 at 4:57 AM
leonardohansa.com
Estimates are distributions, points are decisions (Richard McElreath)
May 30, 2025 at 9:23 AM
leonardohansa.com
20 #pandas one-liners. No idea about some of them, and seem pretty elegant

www.nb-data.com/p/20-pandas-...

Via #TLDRData #python
20 Pandas One-Liners That Can Save You Hours of Work
Simple code that you should not miss at all
www.nb-data.com
May 29, 2025 at 11:01 AM
leonardohansa.com
While attending online meetings, today I'm listening to Marin Marais by Brandon Acker and Craig Trompeter in the background.

It's the only thing worth it in today's meetings.
May 19, 2025 at 10:48 AM
leonardohansa.com
Some people say AI will take out my job...

Looking forward to it! 🎉

The poor thing it's going to get so bored that it'll beg me to give it back.
May 8, 2025 at 6:07 PM
leonardohansa.com
I'm starting to love Information Theory.

I tried some weeks ago these lectures from Oxford Uni but were too much for me, as a beginner.

I've read some texts about the subject and understand much better the intuition. Am I now ready for the lectures?

Let's try

www.youtube.com/watch?v=ScX2...
Information Theory, Lecture 1: Defining Entropy and Information - Oxford Mathematics 3rd Yr Lecture
YouTube video by Oxford Mathematics
www.youtube.com
May 7, 2025 at 10:15 AM
Reposted by Leonardo Hansa
simonsmart.bsky.social
Being able to cancel and restart slow-running jobs in shiny with {mirai} is a bit of a game-changer #rstats Full app in alt text / gist.github.com/simon-smart8... with a method for updating progress too
library(shiny) library(bslib) library(mirai) daemons(1) onStop(function() daemons(0)) ui <- page_fluid( p("The time is ", textOutput("current_time", inline = TRUE)), hr(), numericInput("n", "Sample size (n)", 100), numericInput("delay", "Seconds to take for plot", 5), actionButton("btn", "Go"), textOutput("progress_log"), plotOutput("plot"), ) my_func <- function(x, y, file){ writeLines(paste0("Starting..."), file) for (i in 1:y){ Sys.sleep(1) writeLines(paste0("Step ", i, " complete"), file) } writeLines(paste0("Finished"), file) runif(x) } server <- function(input, output, session) { progress_file <- tempfile() output$current_time <- renderText({ invalidateLater(1000) format(Sys.time(), "%H:%M:%S %p") }) task_var <- NULL task <- ExtendedTask$new( function(...) task_var <<- mirai(run(...), environment(), .args = list(run = my_func)) ) observeEvent(input$btn, { if (task$status() == "running"){ mirai::stop_mirai(task_var) } task$invoke(input$n, input$delay, progress_file)} ) output$plot <- renderPlot({ if (task$status() == "success"){ hist(task$result()) } }) progress_reader <- reactiveFileReader(100, session, progress_file, readFunc = function(file) { if (file.exists(file)) { readLines(file) } } ) output$progress_log <- renderPrint({ cat(paste(progress_reader(), collapse = "\n")) }) } shinyApp(ui = ui, server = server)
April 29, 2025 at 3:13 PM
leonardohansa.com
"As programmers use more AI-written code, debugging may end up being the only remaining programming skill. But for some reason, the vast majority of programming content is about writing code well, not about debugging. What’s different about debugging?"

www.seangoedecke.com/debugging/ #coding
Debugging, emotional resilience, and mental models
Being good at debugging is more useful than being good at writing code - you only write a piece of code once, but you may end up debugging it hundreds of times…
www.seangoedecke.com
April 27, 2025 at 12:40 PM
Reposted by Leonardo Hansa
jaquent.bsky.social
For #rstats people working on computing environments without GUI, you should check out the package `txtplot`.

E.g. this density plot is pretty useful to get an idea what my data looks like:
April 21, 2025 at 4:47 AM
leonardohansa.com
Computer questions:

🧐 Do you have a laptop AND a desktop?

🤔 If so, what do you use your desktop for?
April 20, 2025 at 3:50 PM