Data Science. TileDB. Open Source. Quant Research. R. C++. Debian. Linux. Adjunct Clinical Professor, University of Illinois. Lots of coffee. And some running.
Me too 😀 It is:
- emacs27 with ligature support (requires build with proper fonts, plus a ligature elisp file)
- font is JetBrains mono, see my GH repos for font pkgs
- colours from @nordtheme used "everywhere": gnome terminal, emacs, rstudio (used sparingly these days)
_If_ you trial numbers are indeed unique, make it a factor and retrieve the factor levels.
Or use @Rdatatable which does that for you via grouping and access to the group index.
#rstats
How It Started How It's Going
Less Is More: #rstats testing edition.
Yesterday's post at dirk.eddelbuettel.com/blog/ has more on "less is more", and a vignette engine example.
The Economist built a model that estimates the excess mortality of the pandemic, corrected for underreporting.
They come to the conclusion that 18 million people have died globally.
We now show this data in our Explorer, along with uncertainty intervals: ourworldindata.org/explorers…
It's git. There are no rules. Only trial and error til we get there. And battle scars that remind us how we got there.
Kidding. I love git, and yes it surely beats rsyncing between machines and messing up.
Yep because you can rebase, or squash, or simply cherry pick.
It all varies. Some of my projects have more and smaller commits (and you may find 'temp' or 'snapshot' in the commit message -- per your need here). Others (i.e. work) squash.
"Mark Zuckerberg has only neutral feelings toward Peppa Pig, who he understands is a fictional character, and he blames the coronavirus pandemic on other factors."
xkcd.com/2551/
Yes. Even better to just do
suppressMessages({
library(splines)
library(ggplot2)
library(dplyt)
...
})
as you then also catch non-conformant packages NOT using startup messages. This really silences *evreything* which is a nice feature.