are you streaming data science-related content on Twitch? drop your channel + a brief description so we can start sharing your work and join up our communities ♥️ all languages//tools welcome! #rstats #python #julia #javascript



Jesse Mostipak is prepping for #SLICED (@kierisi; 37181): are you streaming data science-related content on Twitch? drop your channel + a brief description so we can start sharing your work and join up our communities ♥️ > all languages//tools welcome! #rstats #python #julia #javascript

Will Landau (@wmlandau; 29658): Introducing Target Markdown, a game-changer for heavy-duty #rstats #rmarkdown projects: Now in development #rstats {targets}: Try out the Target Markdown template in RStudio:

#RStats Question A Day (@data_question; 24339): When you find out, not just statistics, you need to know linear algebra too > #DataScience #RStats

R posts you might have missed! (@icymi_r; 23858): ✍️📊 A Scientist’s Guide to R: Step 3.1 - data visualization with ggplot2+ > 👤 Craig Hutton @huttoncp
> 🔗 #rstats #datascience


R posts you might have missed! (@icymi_r; 16835): ✍️🖥 Building website with Blogdown > 👤 Daidai @daidaidai2014; R-Ladies Melbourne Inc @rladiesmelb
> 🔗 #rstats #datascience

R posts you might have missed! (@icymi_r; 8814): ✍️🖥 Building your website using R blogdown | A concise step-by-step guide > 👤 Shilaan Alzahawi @shilaan01
> 🔗 #rstats #datascience

Silvia Canelón (@spcanelon; 252): Congrats! 🥳 A big win for #RMarkdown users! > You can find a great example of @apreshill supporting a better user experience in the new #blogdown check functions > Feature request: Improving error messages:

Alison Presmanes Hill (@apreshill; 161): Thanks to an @RLadiesTunis workshop this week, the Hugo Apéro docs site gained a new “Get Started” section 🌸 > #rmarkdown #blogdown #GoHugo

Yihui Xie (@xieyihui; 152): @alfredogmarquez @MeghanMHall @thomas_mock After you run blogdown::check_site() and follow its instructions, you will no longer need to maintain Hugo.

Yihui Xie (@xieyihui; 60): @gavinprm Extremely stable, as long as you follow blogdown::check_site()’s instructions. The Hugo purgatory has been left long behind us.

Jesse Mostipak is prepping for #SLICED (@kierisi; 60): so it turns out this is an easy fix if you’re using the latest version of your theme 🙃 > blogdown::install_theme(theme = “hugo-apero/hugo-apero”, update_config = FALSE, force = TRUE) > (swap out “hugo-apero/hugo-apero” for the repo for your own theme to update!

Daidai (@Daidaidai2014; 52): @JenRichmondPhD @icymi_r @RLadiesMelb Hi @JenRichmondPhD. Yes there are xaringan slides. You can find the code from: I was inspired by @apreshill with all her amazing slides. I mainly used iPad to do all the noting and use it as a background image. 📝💻

Ⓜⓐⓙⓘⓓ Ⓚⓗⓐⓣⓣⓐⓚ (@MajidRusiya; 46): Building blogs with #julialanguage Franklin. #julialang #DataScience #MachineLearning #RStats #Maths #rmarkdown #blogdown

Meghan Hall (@MeghanMHall; 30): @alfredogmarquez @thomas_mock I have a site in Hugo/blogdown, as well (which I also really like!), so I can confirm distill was easier to set up! I’d highly recommend it for a personal blog or other simple site (I’m using it for a course website). Tom’s guide covers everything you need :)

Gavin Masterson, PhD 🐍🐸📊🇿🇦🇸🇪 (@gavinprm; 22): Hi #rstats. What’s the stability of Hugo like now? I’m thinking of using blogdown to showcase a project but my last blogdown website is stuck in Hugo purgatory. I don’t want to be left with endless rendering headaches.

Houston H. 💉💉🎉 Haynes (@h3techdev; 21): @isaac_abraham @LyalinDotCom @asp_net @qmatteoq FWIW I saw @ardalis use this feature with demoing TDD workflow with one of his own code katas and it was tight. I’m a fan of the concept - in fact the thing that sucked me into the #rstats world was @rstudio’s REPL and hot reload in blogdown - so I’m here for it. 👍

Brad Weiner (@brad_weiner; 10): @EpiEllie RStudio + Blogdown + Hugo + GitHub + Netlify.

Yihui Xie (@xieyihui; 10): @SorenWilke If you know which old Hugo version you used before, you could install it with blogdown::install_hugo(version) instead of downgrading blogdown. Then follow the instructions in the console. If anything breaks, usually blogdown::check_site() can tell you what to do to fix it.

Yihui Xie (@xieyihui; 10): @SorenWilke May I learn more about “repairing killed Hugo”? What does it mean? Thanks! (And to answer D4: blogdown::check_site() is almost surely the truth).

jebyrnes (@jebyrnes; 10): @juliasilge @FloatingForests Oh no! And to clean up that blog post. Still getting used to making blogdown do what I want it to with images….

Sören Wilke (@SorenWilke; 10): @xieyihui Yes, blogdown::install_hugo(“whatever netlify said”) saved the day (No.3). Thanks for the tip with check_site.


Nick HK (@nickchk; 20956): My causal inference textbook The Effect has done an editing round and is now available in Bookdown form! It’s also at its own easy-to-remember URL at > Check it out! The Bookdown version will remain up even after the published version is out.

R posts you might have missed! (@icymi_r; 8532): 📚 Meta-analysis of Ecological Data in R > 👤 Rob Crystal-Ornelas @rob_c_ornelas
> 🔗 #rstats #datascience

Anna Sophie Kümpel (is at #ica21) (@kuempelanna; 599): Did a lot of GLM and multilevel model stuff in R in the last couple of weeks and I can only recommend the great (and entirely free) resource: “Beyond Multiple Linear Regression” by Paul Roback and Julie Legler #RStats

Denis Mongin (@denis_mongin; 3713): How do you reveal tax paradise countries from the #Forbes list of 2021? Divide the number of #Billionaires by the population, and voilà: #Monaco, Bermuda Islands, #Bahamas, but also #Switzerland and #Singapore. > #Rstats, inspired by @R_Graph_Gallery and

Dr Simonis (they/them) knows what chem weapons are (@JuniperLSimonis; 2312): Just published > The Thermal Fogger: An Imperial Tetherball > on #bookdown and the @chemweaponsci site! > It’s a vastly expanded version of the book (no longer booklet) > updates: … 🧵

Paul C. Bauer (@p_c_bauer; 215): Workshop materials from the last iteration of Applied Data Visualization (for anyone interested):

Nick HK (@nickchk; 100): Finishing up the bookdown plus a full edit :) may as well run the diagnostics > i am highly skeptical of those reading grade-level estimates but if they’re right i’ll consider that a surprising success

Pablo Casas (@pabloc_ds; 94): 📗 Las herramientas y lib open-source de @rstudio me permitieron publicar en el 2018 la 1er ed. del Libro Vivo de Ciencia de Datos… > Y ganar el 1st Bookdown contest (RStudio)! > Este lunes hablaremos de open-source e IA con Edgar de RStudio 😱


Fred Oswald (@FredOswald; 94): @lisa_kath This guide is good:

Nick HK (@nickchk; 81): go patronize the guy who wrote the javascript to solve my bookdown problem, and pay him more than his list price, i did. worth it.

Nick HK (@nickchk; 70): Also, if anyone is enough of a Bookdown wizard to figure out how to get the search bar from regular Bookdown into the tufte version that would be pretty awesome. I may be able to pay you for it. > (I’ll make a repo soon with my own msmbstyle improvements)

Vikram_Singh Rawat (@Guru_GyanKhoji; 64): @klts0 @rstatstweet If you want to know best ways to run a for loop > #RStats

numericalguy (@numericalguy; 52): Our beta-version of a #OER book on Introduction to Matrix Algebra made using #Rmarkdown and #Bookdown #MatrixAlgebra Thanks to @xieyihui and @grrrck What do you think? @science360 @nsf @IuseProgram @ASEE_DC @EngineeringUSF @rstudio

Will Landau (@wmlandau; 50): By the way: I designed this system to work not only inside a single R Markdown report, but also with pipelines defined across multiple reports ({bookdown}, {workflowr}, etc., as long as the chunk labels are all unique).

Dr Simonis (they/them) knows what chem weapons are (@JuniperLSimonis; 50): References > References are now managed through BibTex which allows a more adaptable and updatable presentation. > Currently the references are rendered using out-of-the-box #bookdown formats/templates, which I havent explored at all.

Dr Simonis (they/them) knows what chem weapons are (@JuniperLSimonis; 50): Structure > - Using bookdown in place of the internal Wowchemy book page type > - Side menu and top menu are book-specific (the site is not visualized within the main chemweaponsresearch site) > - Top menu provides links to information, pdf download, font size/type options, search

Adam Fleischhacker (@preposterior; 50): Thank you @JennyBryan for this fabulous material! I do not reference it often, but when I do it is superbly helpful.
> Happy Git and GitHub for the useR #rmarkdown #bookdown

Nick HK (@nickchk; 40): i couldn’t figure out how to make bookdown do what i wanted, and stackexchange wasn’t answering, so i paid a dude to fix it and he did and it was excellent

Norimitsu Nishida (@NorimitsuNishi1; 40): @OkadaYohei YAMLにテンプレートを指定すればいけると思います。 >

blehbleh (@wiki_early_life; 30): @joe_r_Odonnell It’s a thing!

Peter Higgins (@ibddoctor; 30): @90sKeed @rstatstweet

Kaija Gahm (@kaija_bean; 20): @BroVic @Shel_Kariuki @xieyihui Have you checked out this documentation? Not sure if it’s totally comprehensive, though.

Andrew Brown (@humus_rocks; 12): I will have to get to trying {targets} Markdown with {bookdown} ASAP! #rstats

tidyverse tweets (@tidyversetweets; 12): LaTeX compilation error in pdf bookdown package, but HTML output works perfectly #tidyverse #rstats

jonny saunders (@json_dirs; 11): @hipsterelectron love how R is somehow always simultaneously 5 years ahead and behind in like immediately adjacent language features – PDFs only on CRAN, but leading the way with test running integrated in the package repo/typesetting w bookdown/etc.

Vikram_Singh Rawat (@Guru_GyanKhoji; 11): @haematobot @_ColinFay Yeah, but like there is so much more to code review than just styling… Its one of the part but your code can get rejected due to multiple reasons… I am trying to cover them up here… > Let me know if you like it > Best coding practices for #RStats >

@CarlosVelandiaJ (@carlosvelandiaj; 11): Después de la guerra | Narrativas de excombatientes #rmarkdown #bookdown

Sunglassbro (@guille_jmnz; 10): Realizado me siento después del pedazo de bookdown que me he marcado yessir

Will Landau (@wmlandau; 10): @comeseethepony Narrating, plus an interactive mode to prototype the targets in the notebook interface: Target Markdown is powered by a knitr language engine, and flexdashboard is an output format, so you might be able to use both together.

Steven Lancaster (@SLLancaster; 10): Phase 1: Lakens + GLM. Complete @lakens Inferences and Questions courses. Meanwhile work through @psforscher GLM 2 semester sequence, complete Roback and Legler Beyond MLR bookdown, and watch Eubanks’ “Its all the same” video.

Christophe Dervieux (@chrisderv; 10): @Hao_and_Y @bearloga @rstudio For bookdown, it is also possible to deploy to using GitHub action. We need to make that more documented and maybe with a helper function but examples are currently in our books repo like

Ashok Khosla (@amkhosla; 10): @apreshill @rstudio Oh that’s absolutely lovely. Congrats to you, and the R Team both. Can I ask for better bookdown support for tufte - like real example code that works and we can download?

PS_ratio (@celeryfamous; 10): @BonusLevelFund use rstudio and bookdown package…

IG: @pacmannai (@pacmannai; 10): Referensi: :

phoenix2life (@phoenix2life; 10): @icymi_r @ClausWilke Wow… > ‘The entire book is written in R Markdown, using #RStudio as my text editor and the #bookdown package to turn a collection of #markdown documents into a coherent whole. ’ 👌🏽👍🏽💯➕

Carlos A López Morales (@calm77; 10): @diego_lopezt muchas felicidades Diego!!! > usas la misma interfaz en RStudio que yo (público conocedor jajaja!) > Luego peloteamos experiencias (por ejemplo, yo uso bookdown para documentos largos basados en capítulos), y te da el pdf, el html y el e-book… hasta el word si se lo pides jaja

Sidhant Seth (@DrSidSeth; 10): @marinadpol @baveena_heer This is a good intro course to R

Thomas Sandmann (@thomas_sandmann; 10): @wmlandau @EmilyRiederer @ijlyttle @jdblischak Tardown? In the spirit of bookdown and pkgdown?

Dr. Dana Wanzer (@danawanzer; 0/2): Hey #rstats folks, is there a way to auto hide the sidebar in #bookdown? I am trying to integrate my book (hosted on Github) into @perusall which doesn’t allow toggling. I’m hoping for something I could append to the link to indicate I want the sidebar hidden JUST for Perusall.

tidyverse tweets (@tidyversetweets; 0/1): Bookdown child documents #tidyverse #rstats


Indrajeet Patil (@patilindrajeets; 409): Sometimes you may prefer a certain graphical device to render plots (e.g. PNG), but the journal or stakeholders might request different formats. 📸 > 📽️ shows how easy it is to produce plots for multiple devices at once using #rmarkdown docs! 🙌 > > #rstats

BayesAna 😷 (@AnaBayes; 181): Conversación matutina: > - Que tienes hoy? - Como con Pilar - Pues mira, le das a Knitr…

Will Landau (@wmlandau; 150): Made with input from @chrisderv, @EmilyRiederer, @ijlyttle, @thomas_sandmann, @gshotwell, @jaredlander, @hlageek, @rundel. Special thanks to @chrisderv, who patiently answered my endless questions about the advanced capabilities of {knitr} and circled me back to a solid strategy.

Will Landau (@wmlandau; 101): The key is to use a special {targets} knitr language engine to declare targets and global objects. >

tar_target(analysis, analyze_data(data)

Just updated to make this clearer.

Will Landau (@wmlandau; 80): Decided to pivot to a different interface: Slightly less Markdown/script-oriented, but far more powerful and with less risk of conflicting with the interfaces of {knitr} or {targets}.

tj mahr 🍍🍕 (@tjmahr; 50): @apreshill “Expoxy gives knitr a new engine” drops everything to read more

Rプログラマたん (@cppfriendsbot; 40): 統計学のレポートは、 R Markdown + knitr で HTML なり PDF なりを作るとよいでしょう。図をスライドにぺたぺたコピペするのは辛すぎます。データやスクリプトを変えたらコピペし直しなので、おっとコピペし忘れた、内容の一貫性が崩れた😭、となりがちです。

Doc (@DocK; 40): Btw… Gibt es eigentlich schon ein Tinder für Statistiker? > - ggplot2 oder lattice? - Sweave oder knitr? - xtable::xtable oder Hmisc::latex? - …

Royal Statistical Society (@RoyalStatSoc; 30): You can learn how to build automated reports using R with our course! #R #Rmarkdown #Knitr

Sa-kiera (you can call me Kiera) TJ Hudson 🏳️‍🌈 (@Sakiera_Hudson; 30): @PaoloAPalma @JohnSakaluk 😂😂😂 yep! Now tidyverse has a win over base r in reshaping data from long to wide format and back again. That’s it. > Oh yeah, Knitr can stay away from the verse too along with ggplot. They are redeemed.

Dilsher Singh Dhillon (@dhillon_stats; 30): @Ben_D_Douglas @rstats4ds You can write warning = FALSE, message = FALSE in whatever chunk you don’t want the warnings and messages to appear. If you wanted to do this for the whole document, you can do something like knitr$opts(warning = FALSE, message = FALSE) in your first chunk.

Garrick Aden-Buie (@grrrck; 30): @yoniceedee @chrisderv @wmlandau @jaredlander @gshotwell @iflores_siaca @PeeltothePithy I like the pattern in this {details} example a lot for registering knitr engines: using a helper function to register the engine and calling that function in .onLoad()

Christophe Dervieux (@chrisderv; 30): @wmlandau @gshotwell @grrrck @iflores_siaca @PeeltothePithy @jaredlander 😅 this part is still missing better docs. We did not include it in this version of the R Markdown Cookbook. Best way is to look how custom formats are implemented (like distill). Output format is way to control rmarkdown & knitr behavior, Including pre/post processing.

yoni sidi (@yoniceedee; 20): @grrrck @chrisderv @wmlandau @jaredlander @gshotwell @iflores_siaca @PeeltothePithy after i had two packages that i used regularly in workflows that i had chunk engines (in addition to knitr) for i had to generalize and make them identifiable and replicable (and simpler to debug/maintain) for loading multiple packages

Will Landau (@wmlandau; 11): Does anyone know how to supply language objects to chunk options in #knitr/#rmarkdown? Seems to work with fig.cap and fig.alt but not others, even when I set them in knitr::opts_knit$set(eval.after = …)

入院患者 (@takuto_patient; 10): knitrさせられてるMacBook Pro「ブオーーーーンwwwww」

Dale Maschette 🐟🧗‍♂️ (@Dale_Masch; 10): @AmyStringer11 So - I have discovered you can pull out pieces of code using knitr::read_chunk() but it relies on you putting a label tag ## —- in the chunk you cant use the rmd chunk label as far as I can tell.

Will Landau (@wmlandau; 10): @jaredlander @gshotwell @chrisderv @grrrck @iflores_siaca @PeeltothePithy @chrisderv, is it normal / good practice for a package to assign {knitr} engines in .onLoad()? I think it would be super convenient for users if you could just call library(targets) in a report and have the chunk engines immediately available.

Umut Oguzoglu (@HopeEcon; 10): @Ben_D_Douglas you can suppress messages/warnings/code with : knitr::opts_chunk$set( echo = FALSE, messages = FALSE, warnings = FALSE)

Marcelo (@m_r_avila; 10): @SallumMig isso diminui a probabilidade de ser encoding. se possivel, recrie em utf-8 (dá pra colocar como padrao pra todos os documentos); o autor do knitr diz que suportar multiplos encoding foi a pior decisão deste pacote… (

Gabe The Engineer (@gdbassett; 0/2): Does #rstats knitr purposefully not knit to the specified size if the fig.width/fig.height include a variable, but does if they are numbers? (when rendering individual chunks in line while testing)

tidyverse tweets (@tidyversetweets; 0/1): Suppress evaluation of knitr chunk options in custom language engine #tidyverse #rstats

tidyverse tweets (@tidyversetweets; 0/1): Suppress evaluation of knitr chunk options in custom language engine #tidyverse #rstats


CRAN Package Updates (@CRANberriesFeed; 22): CRAN updates: autoharp coneproj tfarima tinytex #rstats


Benjamin Wolfe (he/him) (@BenjaminWolfe; 188): Really cool stuff happening with parametrized reporting in #rstats. > I’m probably just not in the right circles, but I haven’t seen anything nearly as accessible in the #python/#jupyter world. > Really neat tabbed #xaringan slides by @apreshill showing sample data, code, + output.

Indrajeet Patil (@patilindrajeets; 187): The {xaringan} 📦 ( is the best for preparing reproducible presentations in #rstats! 🙌 > And it also offers an increasing number of beautiful themes via CSS (e.g. metropolis - my current favorite theme). > Here is how to list the available themes! 👇

Alastair Rushworth (@rushworth_a; 83): Here’s a new thing: 🎓 Papers you might have missed! The most shared papers on twitter in the last week (mostly stats & CS).
> Also with cute abstract summaries made using #rstats 📦 {xaringan} + {xaringanthemer} so you can quickly see what it’s about.

Emi Tanaka 🌾 (@statsgen; 50): @BHarrap You can have speakers notes with xaringan Rmd. Time to switch? Just saying!

David Keyes (@dgkeyes; 41): I’m using the {xaringan} package to make my slides. > I’ve gotten tons of help from @spcanelon and @grrrck have provided with regard to adding alt text on all of my images. >

Silvia Canelón (@spcanelon; 32): @smtology @FrankElavsky @liatrisbian @demartsc @DogGeneticsLLC True, but the flexibility of #rstats means we often create HTML outputs of one kind or another. People often say they gain web dev skills by accident as the result of wanting to customize #rmarkdown reports, slide decks (#xaringan), #shiny apps, websites, just to name a few

David Keyes (@dgkeyes; 30): @spcanelon I’m making some xaringan slides. Do you have any advice for adding alt text on background images?

Khia A. Johnson (@khia_johnson; 21): where do #rstats people stand on using R for poster-making?? like i want the #xaringan experience but for a poster 💁🏻‍♀️📰✨

Daniel Falbel (@dfalbel; 10): @TheZachMueller RMarkdown with xaringan is really great!

Dale Maschette 🐟🧗‍♂️ (@Dale_Masch; 10): @SineAndie Equations were the cause for me to jump ship to xaringan slides in R

tidyverse tweets (@tidyversetweets; 0/1): Vertically scrollable code with RStudio and xaringan #tidyverse #rstats

tidyverse tweets (@tidyversetweets; 0/1): Code chunk and output side by side in xaringan slide #tidyverse #rstats

tidyverse tweets (@tidyversetweets; 0/1): How can I disable highlightSpans for one slide in a xaringan presentation #tidyverse #rstats