Data Science. TileDB. Open Source. Quant Research. R. C++. Debian. Linux. Adjunct Clinical Professor, University of Illinois. Lots of coffee. And some running.
Chicago, IL, USA
Joined March 2007
- Tweets 5,361
- Following 350
- Followers 10,149
- Likes 7,646
Sure. Standard CRAN magic. Points that help
- no prior NOTE, WARNING or ERROR -- but OK everywhere
- no rev. deps so no check for 'change to worse' needed
- no compiled code hence no rchk/(ub)san issues
and drake hits all points -- but it is less easy for widely-used packages
In his #chirunconf wrap-up presentation, @wmlandau submitted an update of the #rstats package drake to CRAN.
Live.
And it’s been accepted.
1
9
Just in from a nice Madeleine Peyroux @mpeyrouxmusic show at @NSCPAS with Paula Cole @paulacolemusic opening.
1
RInside 0.2.15 on CRAN: Build robustifications
The easiest way to embed R in your C++ applications
dirk.eddelbuettel.com/blog/2…
#rcpp #rstats
1
7
Thanks for the kind note -- but I should state that I am merely part-time at the fabulous UofI as an adjunct. But really looking forward to the upcoming @IcorsLacsc2019 conference! See you all in May!!
/cc @IllinoisStat @Illinois_Alma
BIO. Dirk @Eddelbuettel (USA) Ph.D in Mathematical Economics. Professor in the Statistics Dept. @UniversityofIl. Over 20yrs of finance industry experience focused in quantitative research, development & trading. Author of many R packages on CRAN like Rcpp. bit.ly/2NIzc5K
1
4
10
New RPy2 release 3.0.0 (and bug fix release 3.0.1 is next) just uploaded to @Debian (and will get to @Ubuntu) thanks to a helping hand from @yarikoptic
RPy2 by Laurent Gautier allows Python users to access the goodness of R without leaving their surroundings
#rstats
3
11
Both points are valid:
- when teaching, I show 'bare-bones' approaches so Rcpp tutorials include bits about manual compiling and linking; and C language understanding is indeed useful
- yet often we prefer not to have to worry about low-level details hence the Rcpp popularity
1
3
Wow: Text as data with (I presume) a loess fit in @TheEconomist
Anybody have details has whether it is loess or gam or ... ?
/cc gelliottmorris
economist.com/business/2019/…
3
4
15
Yup: use #!/usr/bin/env r, chmod to 0755 and you're good. No need to wrap w/ bash. Give the scripts unique-ish names and tab-completion wins.
Many examples (build, install, checl, Rmd->pdf, ...) at github.com/eddelbuettel/litt…
6
Glad you like it -- drat is mostly "just" careful access to already existing Base R functionality without adding dependencies. Or what we may call "standard fare" :)
dirk.eddelbuettel.com/code/d…
cran.r-project.org/package=d…
github.com/eddelbuettel/drat
This tweet is unavailable
1
1
10
RcppStreamulus 0.1.3: Keeping CRAN happy
C++ stream event processing for R
dirk.eddelbuettel.com/blog/2…
#rcpp #rstats
3
12
RVowpalWabbit 0.0.13: Keeping CRAN happy
R Interface to the 'Vowpal Wabbit' fast out-of-core learner
dirk.eddelbuettel.com/blog/2…
#rstats #rcpp
1
1
4
Rblpapi 0.3.9 on CRAN: Requested Staged Install update
Direct and efficient R interface to Bloomberg Terminal
dirk.eddelbuettel.com/blog/2…
#rcpp #rstats
1
9
And @Debian packages have been uploaded -- or get @Ubuntu 18.10, 18.04 and 16.04 packages via my PPA at launchpad.net/~edd/+archive/…
And to boot RQuantLib (version 0.4.7.1 aka repo HEAD) builds without changes.
#rstats #rcpp
#QuantLib 1.15 was released today. Thanks to all contributors!
Get it at <quantlib.org/download.shtml>. List of changes at <quantlib.org/reference/histo…>.
7
10
Anna's team rocks, and they would love to get to Kansas to play a spring (== off-season) tournament. Please help with a buck or two.
gofundme.com/get-washu-women…
1
2
Fair enough! So I just added `motivate()` to the same dang package.
Still at github.com/eddelbuettel/dang -- commit going up in a moment.
#rstats #andIdontactuallyeverusedstatamyself
1
3
Came across 'demotivate', an interesting Stata command .
So I just "had to" port it to R on the commute home yesterday. Find it in package dang on @GitHub.
#rstats #andIdontactuallyeverusedstatamyself
/cc @FelipeOrjuelaR @KevinDenny
2
9
2
39
You describe what @Debian has been doing for 20+ years:
- New packages go into "unstable" and are available
- After a period (was ten days, now five) they migrate to "testing" unless bugs or regressions seen
(- Plus rules for mass migration, bulk dependent library updates etc)
"Update all packages to latest versions" is scary, because what if one of those releases is an hour old and breaks something? I'd much rather do "update all packages to the latest version that is at least three days old." Does any package tool support that kind of update?
3
23
Seconded :)
Despite calling for help at the repo (#187) I still have an open CSS issue I could not fix, so I moved on. But @GoHugo.io is still lovely, and works perfectly well with #rstats whether or not you use blogdown.
Help with github.com/matcornic/hugo-th… still welcome.
1
1
1