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
Replying to @zpneal
Excellent proposal. Maybe it should be a copy of the last released version of S-PLUS though?
1
Replying to @SamanthaLWilson
It's been a while since I looked at it but I am fairly certain that `caret` and/or `mlr3` already cover it for #rstats. It is a not uncommon task. Here is a tweet from just yesterday doing it for #rspatial data too:
New version of #rstats #rspatial package CAST allows visualizing whether training data for #MachineLearning have representative coverage of the prediction area and whether CV folds are appropriately chosen. Tutorial: hannameyer.github.io/CAST/ar… @MLdwig @edzerpebesma @carles_milagarc
3
1
2
Hey, look, round number at CRAN! A big, big thank you to the CRAN maintainers who are volunteers putting together an unparallel repository with unmatched quality guarantees, year in and out. Very much appreciated! (Even if #RStats package authors like myself grumble at times.)
11
3
49
My name is Dirk, and I endorse this package. As I did a year ago in a talk at the @Bioconductor developer forum; video at youtube.com/watch?v=nmY1jIhY… and slides at dirk.eddelbuettel.com/papers… #rstats
#rstats package tinytest now used to unit-test 200+ packages on CRAN! Thanks to all users for your trust in the package, and for all your valuable suggestions!
3
11
The second (and more advanced, covering 'live' (financial) market monitoring via pub/sub) of the two new vignettes in package `RcppRedis` is now also on arXiv at, respectively, arxiv.org/abs/2203.08323, and doi.org/10.48550/arXiv.2203.…. #rcpp #rstats
2
6
M-x R M-x rename-buffer *R:projA* and repeat for several buffers to give multiple (long-running) R sessions within Emacs. Which of course runs in daemon mode so that you can access it at the workstation or remotely ssh'ed in _accessing the same R sessions_. #rstats #emacs
1
5
Rcpp 1.0.8.3 on CRAN: Hot-Fix to a Hot-Fix Seamless R and C++ Integration dirk.eddelbuettel.com/blog/2… #rcpp #rstats
2
7
Dirk Eddelbuettel retweeted
If you're getting sliced and diced by the bleeding edge, #rocker to the rescue with a container for #rstats with gcc-12. Thanks to @eddelbuettel. New linter in gcc-12 identified a real problem in my code (first picked up by CRAN, verified with r-edge). github.com/rocker-org/rocker…
4
8
The first (and more introductory) of the two new vignettes in package `RcppRedis` is now on arXiv at, respectively, arxiv.org/abs/2203.06559 and doi.org/10.48550/arXiv.2203.…. #rcpp #rstats
3
9
Replying to @cristianquirozd
It's a NOTE. Not a WARNING or ERROR. There a lots of packages with larger installed footprints. So you can proceed. But you can consider it as hint to maybe reduce same sample data or documentation.
1
4
With daughter #1 in town for her first big grad school conference (hello to the APS meeting at McCormick Place) I fired up a family favourite recipe @bittman's HTCE: crispy pork with orange and black beans.
1
10
Replying to @ryancbriggs
An underappreciated #rstats simulation option is simple mixtures: eg 80% from N(0,1) and 20% from N(0,5) giving fatter tails "cheaply": > set.seed(42) > X <- c(rnorm(10000,0,1), rnorm(2500,0,5)) > e1071::skewness(X) [1] -0.00247981 > e1071::kurtosis(X) [1] 7.92564 >
2
12
Replying to @superboreen
The quote below is from the @duckdb documentation, but holds in general: don't use `insert` for bulk operations. Rather look at the _specific_ documentation for _your_ SQL backend and see what it recommends for bulk. Or else just be very patient.
1
2
RcppGSL 0.3.11 on CRAN: Small Maintenance Easier GNU GSL use from R dirk.eddelbuettel.com/blog/2… #rcpp #rstats
2
1
Rcpp 1.0.8.2 on CRAN: Hot-fix to silence compiler nags Seamless R and C++ Integration dirk.eddelbuettel.com/blog/2… #rcpp #rstats
3
8
"There are two types of dichotomy: False dichotomies, true dichotomies, and surprise trichotomies."
1
2
6
dtts 0.1.0 on CRAN: New Package data.table time series ops at nanosecond resolution dirk.eddelbuettel.com/blog/2… #rstats #rcpp Think `xts` but at nanosecond resolution. All the immense power of `data.table` at the hightest time resolution.
9
28
Dirk Eddelbuettel retweeted
When you get an error in your R code that will take the whole day to fix. #RStats
3
1
15
The @Debian package for #RStats 4.1.3 released today was updated a few hours ago, used to update the #RockerProject container, and PRed as usual into @Docker itself to update the official r-base container. @marutterstat will follow-up as usual with @Ubuntu binaries.
#rstats 4.1.3 "One Push-Up" (source version) has been released. Binaries will follow in due course.
2
4