Data Science. TileDB. Open Source. Quant Research. R. C++. Debian. Linux. Adjunct Clinical Professor, University of Illinois. Lots of coffee. And some running.
Just had a great time at #useR2020 learning about using C++ in R using Rcpp with @eddelbuettel.
Now for an @opendatamcr workshop on how to access open data through APIs using R with @r_solymosi !
So much R!
Today is a monumental day for TileDB! We celebrate closing a $15M Series A led by Two Bear Capital, joined by Uncorrelated Ventures, @NexusVP, @intelcapital & Big Pi Ventures. Check out the news, see why complex data needs a universal engine. bwnews.pr/2WklPyf
New `-use-LTO` flag in R-devel to enable link-time optimization (i.e. "whole program optimization") for packages.
This will likely add a few percent in performance at the cost of slower compilation.
#rstats
drat 0.1.7 on CRAN: Moar features!
Create and manage R repositories with ease
dirk.eddelbuettel.com/blog/2…#rstats
With special thanks to (twitterless?) Felix Ernst for all the work on this release.
The #Rcpp interface is one of the most seminal changes to happen to #rstats. As someone who has used the native API, there is no question that Rcpp is by far the simplest way to include compiled C++ code in R packages, affording its power to many more users. Thanks, @eddelbuettel
The @CRANberries service is currently 'behind' because something broke at CRAN:
- Packages being added to src/contrib/
- but the PACKAGES.* files are not updating
- per-package webpages for new packages missing.
I notified the CRAN team. But weekend being weekend...
#rstats
One of the things I learned as a @KITKarlsruhe undergrad is that (Alsacian) Flammkuchen is absolutely awesome.
This is one of three pans I just made for Anna, Lisa and myself US-style with Trader Joe pancetta, vidalia onions, and, of course, creme fraiche.
A *really big moment* for #Rcpp:
- 2000 reverse dependecies
- One in eight packages on CRAN
A *big and heartfelt Thank You!* to everbody who has contributed on that journey with code, suggestions, bug reports, documentation and tips!
#rstats
One? Possible topics:
- general intro/buffers
- one consistent editor for R,sh,sql,C(++),Py, ...
- emacs --daemon; emacsclient: ideal for byobu!
- ESS for R and md
- magit (reason alone to switch)
- tramp for remote editing over ssh
- dired
- auctex for latex
- elpy for Python