Matt Moores

Helsinki today

Jira checklist showing the steps involved in preparing an R package for release to CRAN

also thanks to @gaborcsardi @maelle for rhub v2, which makes it easy to configure GitHub Actions for crossplatform builds on Linux, macOS and Windows

https://CRAN.R-project.org/package=rhub

#rstats #statsodon #github #jira

R package release checklist in Jira

list of bug fixes in Jira

cover of the 2nd edition of the book "R Packages"

I mostly followed the 2nd edition of "R Packages" by @hadleywickham @jennybryan with a few minor tweaks to account for the differences between GitHub and Bitbucket/Jira

https://r-pkgs.org/release.html

I highly recommend this book for anyone who is writing or maintaining R code! #rstats #statsodon

announcement by CRANberries of the new version of the R package bayesImageS

logo of the R package bayesImageS

version 0.7-0 of my R package `bayesImageS' is now available on CRAN for Linux and macOS
(Windows binaries are still being built and should be available soon)

The main change is a reduction in the console output for the exchange algorithm. There were also […]

[Original post on fediscience.org]

1 to 2 Starlink satellites are falling back to Earth each day

"There are currently one to two Starlink satellites falling back to Earth every day ... Soon, McDowell told us, there will be up to 5 satellite reentries per day." 😳

Brisbane Mastodon November Social A meetup for Brisbane mastodon folks, from bne.social and beyond!

Hello #Brisbane friends (and future friends). We're doing a (long-awaited) meetup in November! RSVP here. And spread the word! https://events.humanitix.com/november-social-ship-inn

marginal trace plots and densities for the 2 ERGM parameters

7529 is the minimum ESS for a 2 parameter model recommended by the R package mcmcse, for stable estimation with a relative tolerance of .05

The first 1000 iterations were discarded as burn-in.

social network graph of a karate club from Zachary (1977)

joint posterior distribution for the two ERGM parameters

This is the third in my series of examples using the exchange algorithm for Bayesian inference with intractable likelihoods. Here I fit an exponential random graph model (ERGM) to the karate club social network from Zachary (1977). Using the R package Bergm […]

[Original post on fediscience.org]

"In recent months, several publishers have announced that they are licensing their scholarly content for use as training data for LLMs. (...) To understand the dynamics around this fast-developing market, Ithaka S+R has launched a tracker of these licensing deals. In it, we catalog (when the […]

YouTube video by Post-Bayes seminar Pierre Alquier (ESSEC) - PAC Bayes: introduction and overview

The 3rd chapter of the "post-Bayes" seminar, focused on PAC-Bayes bounds, started yesterday. I gave a very introdutory talk, which is already on Youtube.

www.youtube.com/watch?v=hT-d...

There will be 5 more talks in this chapter, see the full schedule there: postbayes.github.io/seminar/

Blog post on The Carpentries web site "Announcing Withdrawal of NSF Open-Source Ecosystems Proposal" June 9, 2025

🧵@carpentries.carpentries.org was set to get an US National Science Foundation (NSF) grant to make an open source (computer coding) ecosystem from a proposal submitted last fall.

The Trump administration told them they had to cease all DEI activity to receive the funds.

They withdrew the grant.

MCMC traceplot showing burn-in for the inverse temperature, beta

MCMC traceplot showing burn-in for the second sufficient statistic, s2(z)

MCMC traceplot showing burn-in for the first sufficient statistic, s1(z)

the first 201 iterations were discarded as burn-in

An example using the approximate exchange algorithm (AEA) in the R package GiRaF, for a satellite image of polar ice floes https://www.kaggle.com/code/matttm/ice-floes-with-the-exchange-algorithm

ESS is only 2118, not the minimum 7529 recommended by mcmcse […]

[Original post on fediscience.org]

Springer now demand that your pre-acceptance version of a paper on the arXiv comes with a disclaimer:

“This preprint has not undergone peer review (when applicable) or any post-submission improvements or corrections. The Version of Record of this article is published in insert journal title] […]

It's wild reading about Bluesky as "the" left-of-center social media network.

2D plot of marching cubes polygonisation of a sphere.

#RStats Dev Diary: Polygonisation of a sphere using marching cubes algorithm

The meme humorously suggests using Vim, a text editor, as an alternative to various Google services. Each Google app is paired with the Vim icon, implying that users can replace these services with Vim for a more privacy-focused approach, even for tasks typically handled by apps.

Finally solved the #Degoogling debate. Anyone who finds fault with this list = Skill issue.

Stacked bar charts showing land cover classification for 10 cities around the world

Thanks to Eloise Skinner, Behzad Kiani, Benn Sartorius, Angela Cadavid Restrepo & all of the team at the University of Queensland Centre for Clinical Research ODeSI M3 for a fantastic first day with GIS and Google Earth Engine at the Spatial Epidemiology Workshop #GIS #EarthEngine #geospatial

Corrigendum to “Dimension-free bounds for sums of dependent matrices and operators with heavy-tailed distributions” We correct Theorem 4 of Nakakita et al. (2024) by introducing a log-Sobolev inequality in place of the boundedness condition. We show that the examples discussed in Nakakita et al. (2024) can be recovered under new conditions.

1) A paper we published last year in EJS contained a technical mistake. We just published a corrigendum today. In case you need dimension-free bounds on the estimation error of covariance matrices for time series, use the corrigendum rather than the original paper!

projecteuclid.org/journals/ele...

It's just so bleak to read about how many new age-blocking things have hit in the past week. The whole internet feels so precarious right now. Not doomed, not lost, but on the edge of becoming something far worse than it already is.

histogram of MCMC samples for the inverse temperature parameter of the Potts model

traceplot of the first 800 iterations of the approximate exchange algorithm

An example using the approximate exchange algorithm (AEA) in my R package bayesImageS, for a satellite image of Lake Menteith, in Scotland
https://www.kaggle.com/code/matttm/lake-menteith-with-the-exchange-algorithm

The axing of Clear Linux by Intel is a clear reminder to never choose a corporate backed distro over a community based one

Reactions to this morning's flurry of fairly pithy posts from Ben Williamson. Anger is an energy. Ed-tech is a mess, uncritical adoption dating back a long time led to the current situation.

#academia #academicchatter #education #ai #genai #Datasociety

The righteous ire of Ben Williamson | […]

Illustration of the process of pathogen removal by a membrane bioreactor

New publication "A critical review of the mechanisms of virus removal by membrane bioreactors–Influencing factors and correlation with operating parameters" in the Journal of Environmental Management

by my PhD student Sharmin Zaman, joint with her […]

[Original post on fediscience.org]

On top, a spectrum consisting of 3 Lorentzian line shapes in blue along with an example Gaussian process fit. The solid black line and the shaded area are the predictive mean and 95% confidence intervals of the Gaussian process. On the bottom, realizations drawn from the Gaussian process in blue.

Posterior mean (solid line) and 95% confidence intervals (shaded blue region) for the mean Lorentzian line width

New publication "Statistical estimation of mean Lorentzian line width in spectra by Gaussian processes" in Chemometrics & Intelligent Laboratory Systems

joint work with Eki Kuitunen (LUT) & Teemu Härkönen (Aalto)

https://doi.org/10.1016/j.chemolab.2024.105307