Uriel Menalled

@urielmenalled.bsky.social

Postdoctoral researcher at Virginia Tech studying the #agroecology of apple orchard systems 🍎. Soon to be Assistant Professor of Agroecology at Montana State University 🏔️! #Rstats Lab website: https://urielmenalled.github.io/lab-website

I recommend academics look into university-hosted AI (🧵 1/2): I'm lucky enough to access Gemma 4-31B via my university's HPC using an API key to code in #rstats. For now, I've been able to avoid any (increasingly expensive) subscriptions with that and #GitHub Copilot's "free" educator AI tools.

15% of fossil fuels extracted each year are used to make fertiliser, run farms, process foods and transport them. In a very real sense our food is made from oil and gas. We are about to feel a very painful food shock. 1/

Bild

New Article: "Creating a new oilseed crop, pennycress, by combining key domestication traits using CRISPR genome editing" rdcu.be/e0NAR De novo domestication by identifying and stacking CRISPR-induced mutations to create a new intermediate off-season oilseed crop.

Bild

ICYMI with a quick screenshot as sales pitch. Simply feed your estimation a bunch of .parquet files on disk and run it as normal. This example takes 180m rows of NYC taxi data and computes the model in < 1 second. On a laptop.

dbreg example showing how to estimate a model against a large dataset, comprising parquet files on disk. The model takes < 1 to run, including the time needed to read the data from disk.

dbreg(
   tip_amount ~ fare_amount + passenger_count | month + vendor_name,
   path     = "read_parquet('nyc-taxi/**/*.parquet')",
   vcov     = ~month,    # clustered SEs
   strategy = "compress" # skip auto strategy overhead
)
Grant McDermott@gmcd.bsky.social · 6mo ago

Things are grim. But in more frivolous news... @jamesbrandecon.bsky.social and I have been chipping away at `dbreg`, a 📦 for running big regression models on database backends. For the right kinds of problems, the speed-ups are near magical. Website: grantmcdermott.com/dbreg/ #rstats [1/2]

Are you stuck on the fundamentals of Bayesian analysis? For a long time, I was. Sure, I was familiar with the equation P(A|B) = [P(B|A)*P(A)] / P(B), but it was hard to internalize. To this end, I've made an #RStats workshop on Bayesian analysis. Check it out on GitHub: github.com/UrielMenalle...

GitHub - UrielMenalled/workshop-BayesianR: Learn about Bayesian analysis theory and its application in R

Learn about Bayesian analysis theory and its application in R - UrielMenalled/workshop-BayesianR

github.com

A while ago, I wrote about the future of Kernza and perennial grain research. A concern I voiced was that, as grain yields increase, the plant would move from investing resources in roots and biomass in favor of seed yield, denuding the plants ability to curtail erosion and improve soil health. (🧵)

A field of Kernza with the sun peaking behind it