Sign in

Anatolii Tsyplenkov

@anatolii.nz
294 followers 281 following 91 posts

Holding PhD in hydrology, interested in sediment transport issues and all things #geospatial. Love mountains, mystery stories and #rstats 🤎 🌐 anatolii.nz | 🐙 github.com/atsyplenkov

PostsRepliesMedia
Reposted by Anatolii Tsyplenkov
Jakub Nowosad @jakubnowosad.com · 16/09/2026
A new R package for redistributing counts to flexible spatial grids 🗺️ Christopher D. Higgins’ `{pycnogrid}` implements pycnophylactic interpolation for H3, A5, S2, ISEA, raster, and local hexagonal grids, while preserving the original totals. higgicd.github.io/pycnogrid/ar... #rstats #RSpatial
higgicd.github.io
{pycnogrid}: Flexible pycnophylactic interpolation to discrete global and local grid systems
Christopher D. Higgins
1227
Reposted by Anatolii Tsyplenkov
Tomasz Kalinowski @t-kalinowski.bsky.social · 23/07/2026
I'm happy to share `ir`, a new command-line tool for running portable R scripts and Quarto documents. It's inspired by my two favorite parts of `uv`: self-describing scripts and tools you can run without installing first. opensource.posit.co/blog/2026-07...
opensource.posit.co
ir 0.1.0: self-describing R scripts and Quarto documents
`ir` is a new command-line tool for running portable R scripts and rendering Quarto documents whose package requirements, and optional R selection, live inside the file itself.
47521
Anatolii Tsyplenkov @anatolii.nz · 11/06/2026
I moved to @fastmail.com a couple of years ago, and it seems like I took the wrong path. I decided to start all my new conversations from a new email address and gradually reduce the Gmail ones. It is working, but very, very slowly.
130
Reposted by Anatolii Tsyplenkov
Pedro Batista @pedro-batista.bsky.social · 27/09/2025
🌍 We’re hiring 2 PhD students in #Soil Science & #Erosion Research! 🌱 Join the DFG Emmy Noether project “Trajectories of erosion-induced soil change on tropical arable land (TASC)” at the University of Augsburg 🇩🇪🇧🇷 👉 Job offer: tinyurl.com/yuut64m5 👉 Project info: tinyurl.com/repjumje
tinyurl.com
TASC
011
Anatolii Tsyplenkov @anatolii.nz · 20/09/2025
Reminded me of a Hogwarts logo
media.tenor.com
harry potter is sitting at a table reading a newspaper and drinking from a cup .
ALT: harry potter is sitting at a table reading a newspaper and drinking from a cup .
120
Reposted by Anatolii Tsyplenkov
Yutani @yutannihilation-en.bsky.social · 20/09/2025
100 stars🎉 Savvy is a niche framework for those who want to use Rust in an R package yet extendr is not their fit, so I didn't expect to see so much stars! github.com/yutannihilat...
github.com
GitHub - yutannihilation/savvy: A simple R extension interface using Rust
A simple R extension interface using Rust. Contribute to yutannihilation/savvy development by creating an account on GitHub.
053
Anatolii Tsyplenkov @anatolii.nz · 18/09/2025
Bloody AI is everywhere
Photo of a moving company truck with following slogan "Let's Get Moving with AI furniture"
010
Anatolii Tsyplenkov @anatolii.nz · 12/09/2025
Great words
010
Reposted by Anatolii Tsyplenkov
Emil Hvitfeldt @emilhvitfeldt.bsky.social · 20/08/2025
Happy to announce ✨quarto-revealjs-editable✨ This fully supersedes the imagemover extension, as I back then didn't realize the potential. You can now also move, resize, change font size and alignment for text in your slides github.com/EmilHvitfeld... #quarto #slidecrafting
1011636
Reposted by Anatolii Tsyplenkov
Jordi Rosell @jrosell.bsky.social · 29/08/2025
Scripts to avoid recomputation compared: dagster in #python vs {targets} in #rstats. gist.github.com/jrosell/e283...
gist.github.com
Example of a machine learning pipeline in Python using dagster to skip the recomputation of assets if last modification dates are not updated from since previous computations. Also an example in R usi...
Example of a machine learning pipeline in Python using dagster to skip the recomputation of assets if last modification dates are not updated from since previous computations. Also an example in R ...
131
Reposted by Anatolii Tsyplenkov
Nick Tierney @njtierney.bsky.social · 29/08/2025
📢 R Dev Day @ Australia 🏢 Monash University, Melbourne 🌐 Remote, Australia 📅 Fri 21 Nov It's happening! Apply to attend this free event, to collaborate on contributions to #RStats! You don't need to be an R guru! Deadline Sunday 16 Nov: pretix.eu/r-contributo... #RDevDay #RSEng
pretix.eu
R Dev Day @ Australia 2025
Fri, Nov. 21st, 2025
01014
Reposted by Anatolii Tsyplenkov
Konsta Happonen @konsta.happonen.eu · 27/08/2025
Has anybody come up with a solution to this problem: you have a perfectly reproducible analysis pipeline that produces beautiful dynamic reports. But once you bring on even one collaborator things immediately devolve into email-attached word documents named manuscript_finalFinal_2.docx #rstats
4212
Anatolii Tsyplenkov @anatolii.nz · 26/03/2025
4 for sure!
010
Anatolii Tsyplenkov @anatolii.nz · 22/03/2025
Oh, that’s a good idea. I am just afraid, that it will be biased towards R as it prevails it academia
110
Anatolii Tsyplenkov @anatolii.nz · 22/03/2025
expanding the idea suggested on mastodon, I analysed the number of questions posted on gis.stackexchange.com over the past decade, focusing on the most popular geospatial libraries in #rstats and #python. The cumsum of daily questions suggests a draw. gist.github.com/atsyplenkov/...
Cumulative graph of Python- and R-tagged questions on the gis.stackexchange.com
110
Anatolii Tsyplenkov @anatolii.nz · 21/03/2025
No, of course not
100
Anatolii Tsyplenkov @anatolii.nz · 21/03/2025
I am currently writing a grant proposal and as a researcher by nature got really curious. My impression that it's R — but I am biased and work mostly within "R information bubble"
010
Anatolii Tsyplenkov @anatolii.nz · 21/03/2025
Which language serves as the geospatial lingua franca — #pythonor #rstats? A quick look at Google Trends suggests R, but that’s a rather simplistic comparison. Has anyone come across a more in-depth study on the most popular languages for spatial analysis?
A plot showing google hits for ` Python spatial `, ` R spatial `, ` Python geospatial `, ` R geospatial ` keywords since 2015 to present
571
Anatolii Tsyplenkov @anatolii.nz · 10/03/2025
You have literally listed all my favs. Especially Inter is <3
000
Anatolii Tsyplenkov @anatolii.nz · 07/03/2025
I know I'm biased, but maaaan, Formalist and Air work so well together in the latest Positron update. Dear #rstats, you should give Positron a try. 💨 Air — posit-dev.github.io/air/ 🎩 Formalist — github.com/atsyplenkov/...
073
Anatolii Tsyplenkov @anatolii.nz · 02/03/2025
FYI, this is last year announcement
000
Anatolii Tsyplenkov @anatolii.nz · 15/02/2025
Oh, that’s super cool. It can be used not only for click baits, but for other links to. Like you hover, or click on the link, and a short summary of the page is popping up
110
Reposted by Anatolii Tsyplenkov
Rense Corten 🟥 @rensec.bsky.social · 14/02/2025
It still bugs me that nrow() in #RStats is called nrow() and not nrows(). I type it wrong every time.
021
Anatolii Tsyplenkov @anatolii.nz · 13/02/2025
Haven’t seen that in Positron. I doubt that it exists, tbh
000
Anatolii Tsyplenkov @anatolii.nz · 13/02/2025
Unfortunately there is too much of them. A good and up-to-date collection of them are here: rsources-en.netlify.app/courses
rsources-en.netlify.app
Courses – RSources
110
Anatolii Tsyplenkov @anatolii.nz · 13/02/2025
Noice! Do you have any suggestions for naming columns in a dataframe? I tried using PascalCase for column names to differentiate them from snake_case objects, functions, and variables. However, I keep forgetting to follow my convention 🙈
120
Anatolii Tsyplenkov @anatolii.nz · 12/02/2025
Take a look on {janitor}. Its adorn_percentages() can do the trick in two lines of code sfirke.github.io/janitor/refe...
sfirke.github.io
Convert a data.frame of counts to percentages. — adorn_percentages
This function defaults to excluding the first column of the input data.frame, assuming that it contains a descriptive variable, but this can be overridden by specifying the columns to adorn in the ......
110
Anatolii Tsyplenkov @anatolii.nz · 12/02/2025
I haven’t used it yet, but saw this repo. Maybe it will help you github.com/ocsmit/zed-r
github.com
GitHub - ocsmit/zed-r: R language support for Zed editor
R language support for Zed editor. Contribute to ocsmit/zed-r development by creating an account on GitHub.
000
Anatolii Tsyplenkov @anatolii.nz · 11/02/2025
For sure!
020
Anatolii Tsyplenkov @anatolii.nz · 11/02/2025
I was working with big data recently and found that Polars is faster than a DuckDB-based approach. Thanks to {tidypolars} you can use {tidyverse} syntax as well. However, if you are working with a medium-sized CSV like this (only 20K rows), {data.table} may still be faster but it loads csv in memory
141
Anatolii Tsyplenkov @anatolii.nz · 11/02/2025
Finally got to my laptop and realized I made a mistake. The arrow command is actually open_csv_dataset(). Here is the reprex. Both examples do not load the entire CSV into R, but if your CSV is enormous, it may be beneficial to create an on-disk DuckDB.
Open csv in R using duckdb and arrow using dataset from https://josiahparry.com/posts/2024-05-24-duckdb-and-r
251
Anatolii Tsyplenkov @anatolii.nz · 11/02/2025
Or one can use {arrow}. For example this will convert the Dataset object into a lazy DuckDB table: open_dataset("file.csv") |> to_duckdb()
120
Anatolii Tsyplenkov @anatolii.nz · 08/02/2025
Especially videos. I found v useful youtu.be/HWzSOhDVDk8?...
youtu.be
Building UwU: Rust-based UUIDs for R
YouTube video by Josiah Parry
020
Anatolii Tsyplenkov @anatolii.nz · 03/02/2025
There is one more way to tackle this — using `.x` for data masking. I think this is the recommended approach: rlang.r-lib.org/reference/to...
010
Anatolii Tsyplenkov @anatolii.nz · 30/01/2025
Thanks! Just keep in mind that this is faaaaaar away from being a stable version. If you encounter any problems, please report them by submitting an issue on gh
100
Anatolii Tsyplenkov @anatolii.nz · 30/01/2025
Ha! It’s literally me a week ago. However, I found it better for my projects to keep the renv library out of my projects by setting RENV_PATHS_LIBRARY to something like “/renlib.”
000
Anatolii Tsyplenkov @anatolii.nz · 30/01/2025
You can simply use ‘reframe()’. It drops groups by default dplyr.tidyverse.org/reference/re...
dplyr.tidyverse.org
Transform each group to an arbitrary number of rows — reframe
While summarise() requires that each argument returns a single value, and mutate() requires that each argument returns the same number of rows as the input, reframe() is a more general workhorse with ...
110
Anatolii Tsyplenkov @anatolii.nz · 30/01/2025
my favourite theme in RStudio was {KandleTheme} (github.com/TheMrKingent...) with {darkstudio} enabled (github.com/rileytwo/dar...). They are perfectly work together. Just look at this beauty
https://github.com/TheMrKingent/KandleTheme/blob/main/img/mainview_ed.png
121
Anatolii Tsyplenkov @anatolii.nz · 30/01/2025
To all new {wbw} users, please note that the package is still under active development, with only about 7% of the planned features currently implemented 🐢💨
163
Anatolii Tsyplenkov @anatolii.nz · 28/01/2025
Consider this: to slow down #AI development, the community could deliberately mislabel correct answers as incorrect. By doing so, the false positive rate would rise, confusing the #LLM and disrupting its learning process. 🤖🤔
Response quality control in ChatGPT web interface
020
Anatolii Tsyplenkov @anatolii.nz · 27/01/2025
{terra} and {sf} are great for sure, but if we are talking about impressing, then it is {geos} by @paleolimbot.bsky.social This is where the real speed is
152
Anatolii Tsyplenkov @anatolii.nz · 23/01/2025
I have some projects bookmarked: Gatsby — github.com/taniarascia/... Node — github.com/d3ward/d3war... 11ty — eleventy-excellent.netlify.app
github.com
110
Anatolii Tsyplenkov @anatolii.nz · 16/01/2025
On behalf of the people, I declare that we are interested 🤩
130
Reposted by Anatolii Tsyplenkov
Ben Zipperer @benzipperer.org · 08/01/2025
a lazy organizational tip for data analysts and researchers: create a single directory that holds all the exploratory or one-off scripts that don't yet have a proper home mine is called misc_R once a year (like today!) I satisfyingly shove the prior year's files into a subdirectory
screenshot of a directory browser in a text editor showing the directory called misc_R containing several subdirectories, one for each year
4142
Anatolii Tsyplenkov @anatolii.nz · 16/01/2025
I do like the vertical stack plots. Looks gorgeous
Vertical stack plots from https://ggstackplot.kopflab.org/
041
Anatolii Tsyplenkov @anatolii.nz · 14/01/2025
Fellow #rstats users working on WSL, has anyone successfully run Positron from a nix-shell? I’m struggling to configure a custom path to the R binaries when loading from the nix-shell. A more detailed question can be found here github.com/ropensci/rix... @brodriguesco.bsky.social ,any thoughts?
github.com
Configuring Positron and nix on Windows WSL · ropensci rix · Discussion #387
Hey guys, looking for some help and advice. I cannot connect to a custom R installation on Positron v2025.01.0-39 when connecting to a remote machine via SSH (Ubuntu 24 on WSL2). I am trying to con...
021
Anatolii Tsyplenkov @anatolii.nz · 14/01/2025
I just realized there’s no emoji to represent data frames, databases, or even spreadsheets! What do you typically use #rstats #databs ? Here are my favorites so far: • 🗄️ this cabinet reminds me of data storage • 🗓️ if you squint, it might resemble a spreadsheet with all its cells • 📊⚾️ data base?
150
Anatolii Tsyplenkov @anatolii.nz · 14/01/2025
Cool-cool-cool 🗄️💨 can’t wait to see the benchmarks! You mentioned that the speed might be slower because column-by-column predictions are the most common. Does this mean Arrow might perform better on large data frames that still fit in memory (vs tibble)? orbital.tidymodels.org/articles/pro...
orbital.tidymodels.org
Pros and Cons
110