🤖 GitHub Actions: Auto-generate CI workflows with Cachix and NAR caching out of the box
📈 Execution Reports: `pipeline_report` for informative Markdown reports
Check out the full changelog: https://tstats-project.org/changelog.html
🚀 Announcing T v0.53.0 "L'Initiation"! 🚀
This release takes Nix-native data science pipelines to the next level:
📊 Dynamic Branching: Map, cross, slice, head/tail, and sample pipeline nodes automatically
🖥️ Atelier TUI IDE: Tmux-based interactive environment with live REPL variable & plot sync
🛡️ Static Conditionals: `node_when` & `node_fork` for flexible DAG configurations
1/2
the people that built "basic" unix tools like tmux/screen, vim, ssh, etc, were true visionaries. These tools are more useful as ever in the current age of agentic ai
even if you believe LLM are crap at coding, I'd recommend you get that side project which you've never had time to build done using an LLM, you might suprised. Also, even if the codebase wouldn't be up to scratch, keep in mind that in 5 years, another LLM might successfully refactor it to match your vision
The first data science book that has a chapter on monads https://reproducible-data-science.dev/
Learn how to build robust #DataScience pipelines with #RStats, #Python , #Julia and #Nix !
Reproducible Polyglot Data Science
still a WIP, but I think there's already quite a lot of useful content. Will stay free forever: https://reproducible-data-science.dev/
buy me an espresso to show your support: https://buymeacoffee.com/brodriguesco
what the recent Tailwind drama showed is that the primary audience for tech books/docs is going to be LLMs now, no one's going to read anymore, not that people read anything before. Just upload a book's PDF and ask for a TLDR. Future tech docs will be written with this in mind
"Nix for Polyglot, Reproducible Data Science Workflows" read it here https://b-rodrigues.github.io/rix_paper/
Korppi , a local-only, open-source Markdown editor for asynchronous collaboration.
With Korppi, each document carries its full history, so you can import multiple versions of the same file and apply changes hunk by hunk, calmly and explicitly.
It's just so easy.
if someone wants to give feedback on a paper i’ll be submitting soon titled "Nix for Polyglot, Reproducible Data Science Workflows" that’d be greatly appreciated!
https://github.com/b-rodrigues/rix_paper/blob/master/paper.qmd
This is an example of how Nix can be useful for data analysis: a project that uses #Julia #Python and #RStats in which #Nix is used to set up the dev env in a reproducible way, and is also used as build automation tool to orchestrate the codebase https://b-rodrigues.github.io/rixpress_demos/rbc/index.html
Friends, I’ll be giving a lecture titled "Best practices for doing science with data" at the Luxembourg Institute of Health next week (online as well, don’t hesitate to register): https://www.lih.lu/en/event/%f0%9f%87%ac%f0%9f%87%a7-best-practices-for-doing-science-with-data-3/
rixpress is now an @rOpenSci package!
Link: docs.ropensci.org/rixpress/
@julialanguage and #Nix enjoyers, an important PR got merged today that makes using Julia on Nix a breeze: https://github.com/NixOS/nixpkgs/pull/433332
I am planning a new release of my R package rix with full support for generating Nix expressions that include Julia!
rixpress is currently in ropensci review, and I think that I've tackled most of teh reviewers suggestions. I would be very grateful if some people could give it a go, especially if you're already users of rix!
New blog post #Python needs its CRAN
I'll be releasing a new version of my #RStats package chronicler which makes it easy to log functions, now it's able to log ggplots as well!
