Replying to
Oh, my fucking god. The "Hallucination detector" is just another LLM. This is not serious.
Remote
Elio Campitelli
@eliocamp@mastodon.social
mastodon 4.8.0-nightly.2026-10-06For entertainment purposes only. Elio can make mistakes, check important information carefully.
#Rstats
1245 Followers
596 Following
50 Posts
Joined October 29, 2022
Website:
Pronouns:
they/elle
GitHub:
Location:
Australia 🇦🇺 (previously Agentina 🇦🇷)
Replying to
@diffrentcolours@tech.lgbt Have you read a bunch of weird puns, each more nonsensical than the previous one? You might be entitled to consternation.
@Pionir@masto.bike @patrislav@chaos.social @HighlandLawyer@mastodon.social @seabass@social.seabass.systems @dpnash@c.im @loren@flipping.rocks
Replying to
@jessta@aus.social My brother in law moved to Perth from Argentina like 4 years ago. Last year we went to visit and he had no idea how to use PT. Our trip to the city was his first train trip Australia.
Replying to
Racists policies against immigration not only are cruel, they are also expensive as hell. In 2023 this concentration camp, now empty, was costing #Australia AUD486 million.
https://www.bbc.com/news/world-australia-66027565
Replying to
This is also why the most important part of "open source" is not the source that is open but the people who write it and understand it.
Replying to
@henrikbengtsson@mastodon.social @trevorld@fosstodon.org I think you need to manually download the ISO. At least I had to a few weeks ago when I tried.
Replying to
@henrikbengtsson@mastodon.social @datastory@mstdn.ca Ah, yeah. The horizonal range is about 200 m. There's not going to be such a big difference in temperature. The dry adiabatic lapse rate is about 1°C/100m, so you wouldn't expect more than 2°C difference between the lowest and highest location.
Replying to
@trevorld@fosstodon.org You can download a windows 11 image from microsoft and install it in a virtual machine for free. I assume there's nothing illegal about it.
Is someone else having issues with #Positron using remote SSH? My #RStats sessions take FOR EVER to startup and the console has enormous latency and sometimes it just completely locks up and I need to close the whole session, sometimes even restart Positron completely.
Meanwhile, using R from the terminal via SSH works perfectly well. And VSCode using the normal R terminal also is working well.
Replying to
@jimgar@hachyderm.io They're not hard to understand if you translate them to lapply(). But the backward notation always trips me up.
I've just found this 22 year-old bash script to add history to the cd command. And it works perfectly! 🤯
Journal is asking for a persistent identifier (DOI or ISBN) for #RStast itself. `citation()` doesn't return a DOI and I can't find that there is. Any tips for that?
EDIT: answer in the first reply, thanks!
Why did they named it "Young Sheldon" instead of "The Small Bang Theory"?
Github is such a bloated POS. It got stuck following a link on a wiki. It was faster to reload the repo in another tab and click the link in the readme.
Replying to
@jimbob CDO will gladly produce 1x1 files to represent timeseries (e.g. from cdo fldmean), so I'm sceptical that it can't handle them afterwards.
btw, cdo distgrid will chop up your files into several independent regions that you can process and then collect with cdo collgrid.
Google calendar's timezone selector has always been bad. Now they changed it to have search, which is good, but it only understands cities, so you can't select UTC. How can they fuck this up so royally?
Replying to
@drgroftehauge@sigmoid.social R doesn't use dot as accessor. List.files is just a function.
How you do things to the files depends on what you want to do to them 🤷
Replying to
@jimgar@hachyderm.io @bmaxv@noc.social I am trying to design a class on analysing climate model experiments for early university students who might not be super comfortable with python or any programming language at all. ANY extra cognitive load is a massive loss because it's time I need to waste not teaching them about climate model output and more friction between them and getting cool and motivating results.
Replying to
@jimgar@hachyderm.io agreed. I think you need to understand for loops before understanding comprehension/lapply. But then, they are a better abstraction vs all the boilerplate of pre-assignment + for loop.
Replying to
I found a one-liner: list(Path(path).iterdir()). It's still rather convoluted and impossible to discover, but at least doesn´t require explicit loops.
Replying to
@bmaxv@noc.social That's the type of code I came across when googling the problem. That or for loops. 🤷
The journal typesetter modified every instance of "days" (as in "the forecast reaches its maximum error in about 30 days") to "d" (as in "the forecast reaches its maximum error in about 30 d"). I imagine the argument is that "days" is a unit of measurement and units of measurement are always abbreviated (like "km" vs "kilometers"), but this strikes me as making the text harder to read for the sake of consistency.
A few hours back in #Australia. I already miss bidets.
Preparing our haul to bring back to Australia.
Replying to
@jonny I couldn't make it work and then I realised I had written "upolad". Correcting the typo fixed it. So it seems it's pretty sensitive to the file name pattern.
Replying to
@utrenkner I try to do a proper deep dive once I'm not on vacation and have a proper computer with screens and keyboard instead of a phone. I can't do research on my phone.
Replying to
@pvanheus@mstdn.science You don't need to remember it. You can press tab and get a list of all arguments, or press F1 to open the documentation of the function.
Replying to
Replying to
Replying to
@PhilSalkie@mindly.social @Danpiker@mathstodon.xyz Huh. Does this means that we shouldn't be writing small text or rendering small details in blue?
Replying to
@gspeed0689@mapstodon.space That's much better. I wonder why no one proposed that answer in the various SO questions and articles I found when googling the problem.
Replying to
@jimgar@hachyderm.io @bmaxv@noc.social And even if this was my own project, having to write a whole function, having to think about for loops, preallocation and string formatting it's PITA to do for every single project. And this is a general issue that shows it's head everywhere in the language. There's A LOT of things in python that require the user to worry about these sort of low-level stuff.
Replying to
@PabloBzz@mastodon.la Igual me parece que la captura es falsa, no? Recién fui al perfil de este engendro y encontré un post parecido sobre el SMN nomaś.
Replying to
@jonny@neuromatch.social It's my favourite of the violin family.
Do you know about Zoe Keating?
Replying to
@whitequark@social.treehouse.systems Minute physics has been on a roll recently. The previous one about dissonance is fantastic.
Replying to
@pvanheus@mstdn.science the list.files() is the same as os.listdir() function in python. Except that can return full paths with an argument, obviating the need for using for loops or convoluted list comprehension expressions to join back the basedir.
Experimenting with supersampling with splines to smooth contours when the resolution is kinda coarse. My logic is that contour methods assume linear interpolation, but that's just one assumption. Why not assume a smoother function? So, I first interpolate to a finer grid using splines and then use regular contours. This is the result. First image is default contours, second image is with the interpolation step.