kfold woes 

🤦‍♂️ If the parameters in the log-likelihood function in the test dataset aren't calculated from the training dataset then your log likelihood calculation won't work.

I think this means that for a hierarchical model with grouped folds, this means, for example, that if you calculate SiteA parameter with the training data and then test with SiteB data, you gotta use the superpopulation Site param, not the specific Site param.

I'm feeling grumpy today, but I'm going to sit down and see if I can finish my k-fold cross-validation before lunch. And then I'm going to spend a couple hours in my garden.

Reading, biology, math, HIV 

Started reading Otto & Day’s Biologist’s guide to Mathematical Modeling today in preparation for a reading group I’ve joined. I hope I can keep up!

They discuss models for HIV in the first chapter and wow have things changed and not changed since 2007.

Good joint response to latest Canadian government copyright consultation from CARL-ABRC & CFLA-FCAB. Much to like in its recommendations & that it sheds light on more than just the short-sighted term extension imposed by CUSMA (NAFTA) carl-abrc.ca/news/joint-respon

Programming persistence 

Alrighty. Gonna try again to understand this very frustrating error in my code this morning. Wish me luck!

This is a very very important paper about heritability and I need every journalist who writes about genetics to read and understand it.


I didn't sleep well last night, but made a great graph today and worked in my garden.

Tomorrow I'm going to figure out how to implement k-fold cross validation since the answer to my question about reloo on the stan forums was basically "re-implement these portions of the rstanarm package to work with your model" at which point the extra time cost of learning and doing k-fold is worth it.

Reading about model comparison and validation this morning and delighted by LOO as an acronym.

This is a very soothing and beautiful video about working in forestry in BC bilibili.com/video/BV1j5411E7e

Vik Sohonie: "Too often we forget that the most defining event of the 20th century was not World War II or the Cold War, but the liberation of billions in Asia and Africa between the 1950s and 1980s as citizens of almost 100 new-born countries."

A cool article from the always-fascinating Africa is a Country site: africasacountry.com/2020/10/wh

deadlines in science 

Finding problems with my model/work is frustrating because of arbitrary deadlines. Without those deadlines, I'm curious and excited about what those problems reveal.

bookstore surprises, tiny lewd 

As a young and horny STEM undergrad, I picked up The History of Sexuality in the university bookstore. It didn't take me long to realize this book wasn't going to be what I thought it was, but too late! had already sucked me in.

Now I'm helping teach ecology students Griffith's Queer Theory for Lichens currents.journals.yorku.ca/ind (thanks again @lichen for putting this in my tl.

I like the direction my life is going. Let's read widely curiously, joyously!

scheduling challenges 

Scheduling a meeting that a bunch of busy people in British Columbia, Texas, Nigeria, New Zealand, and Australia can all make every week is ... maybe impossible? I wish we had more normalized asynchronous interactions in academia.

The Victorian era was whiplash for western political theory. Before the Victorian era when political theory took a hard turn away from emotion you had the Scottish Enlightenment talking about how moral sentiment was essential to politics. Adam Smith wrote about that!
Locke said happiness is property, which shows how much emotions were a part of all discussions of politics.
The 18th century revolved and focused on emotions and how to express them Well - a deep belief that we naturally care 1/


Good morning everyone! I like to plan my week on Sunday mornings and plan my month at the beginning of each month. It's very pleasing that the two coincide today, though it does mean I'll need to plan a little longer than usual.

Hope you all have a good week planned!

@lichen Just wanted to let you know that the "We're all queer lichens" article is now officially being taught in an upper level ecology class an as an example of feedback loops between science/humanities/society scholar.social/@scisus/1056115

I just helped develop a really cool ecology course and I feel good about that.

Programming frustration 

There is a bug in my code and I can’t tell if my recent changes created it or uncovered it and I’ve been at this for 3 hours now and even my good git habits aren’t saving me.

Gonna go cry-work for another hour and see what happens.

boring, technical, notes 

I've been using to take notes and I love it.

Need to do a bit of maintenance today though - I've been syncing across machines with and the syncing is broken somehow.

Show older
Scholar Social

Scholar Social is a microblogging platform for researchers, grad students, librarians, archivists, undergrads, academically inclined high schoolers, educators of all levels, journal editors, research assistants, professors, administrators—anyone involved in academia who is willing to engage with others respectfully.