I've been working on a total rewrite of dtplyr, which lets you take advantage of (almost all of) data.table's speed while using dplyr syntax. The package is still in development, but I'd love for you to try it out and let me know how it goes: https://t.co/MXogoZ2lvu #rstats

2019/07/24

#rstats

Hadley Wickham (@hadleywickham; 1536⁄361): I’ve been working on a total rewrite of dtplyr, which lets you take advantage of (almost all of) data.table’s speed while using dplyr syntax. The package is still in development, but I’d love for you to try it out and let me know how it goes: https://t.co/MXogoZ2lvu #rstats ↪

Emily Robinson (@robinson_es; 250⁄62): Want your mind blown today? Check out the data viz made in #rstats with ggplot2 by @CedScherer for #TidyTuesday > https://t.co/rm7Zt19q6Q https://t.co/TmqJJnfFdt ↪

Mitchell O’Hara-Wild (@mitchoharawild; 243⁄56): Time to update your résumé? 👷‍♂️👷‍♀️ Now’s the time as vitae v0.2.0 is now on CRAN! 🥳📦 > Release post: https://t.co/ffZKq6Obec > This #rstats release adds: - 5 new styles for your CV/résumé - a new vignette on using data from around the web with vitae - revamped bibliography support https://t.co/kjDa99jkZ8 ↪

Mara Averick (@dataandme; 187⁄27): ⚓️ @hrbrmstr’s a master of chart-recreation w/ #rstats… “Quick Hit: A Different (Diminutive) Look At Distributions w/ {ggeconodist}” https://t.co/O7qjBLCoXP #dataviz 📦 page: https://t.co/nrHdfCa0bc https://t.co/ePVOJv1e7g ↪

Mark Dingemanse (@DingemanseMark; 173⁄30): Science Twitter! A reviewer feels “there may be too much unnecessary information provided re software” and asks us remove references to #rstats packages. I strongly disagree: software is a critical part of research > 1⁄2 ↪

blogdown

Alison Hill (@apreshill; 60⁄7): So excited to see so many instructors I really admire here- a great resource for teams or individuals looking to learn #rstats #tidyverse or #shiny from people who know how to code AND how to teach. > Also, site built with #blogdown #madewithacademic 💪 > https://t.co/G3mHrFK5Ww https://t.co/r9dFIpzMmx ↪

Dr Victoria Brunsdon (@veabrunsdon; 39⁄4): I created my personal website using blogdown in R 😀👩‍💻 > 🖥️ https://t.co/Xlopk0X8y2 > 👇I followed this excellent thread! https://t.co/aKx9jaPToB ↪

Tony ElHabr (@TonyElHabr; 17⁄3): An #rstats “how to” for making a gallery of visualizations for your #blogdown site: https://t.co/jI1eU1dMUa. Inspiration from @R_by_Ryo. https://t.co/t7g1DZwF11 ↪

Guy Prochilo 🏳️‍🌈 (@GuyProchilo; 5⁄3): #Rstats: Hugo-academic blogdown theme is a beautiful template recommended for academics who want a professional-looking website. > Instructions on GitHub: https://t.co/mI8ICuTGz4… > Check out my website with the theme: https://t.co/eTVWEtFOaX > #phdchat https://t.co/2pEcWEjPuU ↪

Dr Andrew Laughland (@andrewlaughland; 5⁄1): A good reason to work with Blogdown… https://t.co/PHGm1x88Pe ↪

Seth Dobson (@sethdobson; 4⁄5): Hey everybody, I took the plunge and designed my own #rstats blog using the #blogdown package in R, HT to @NicholasStrayer for creating the lovely Tuftesque Hugo theme, the first post is up with more to come (hopefully not too infrequently) > https://t.co/E6muKQTlKq ↪

Ioannis Kosmidis (@IKosmidis_; 3⁄1): See > ⏹️ https://t.co/JC6Mg5L3LX for slides from my {cranly} #useR2019 🗣 > ⏹️ https://t.co/tYo1c9Wgdh for code to place interactive directives networks of #rstats 📦 on webpages: as simple as adding a few commands in Rmd & using {blogdown} > Stay tuned! More’s coming… > 5️⃣-5️⃣ https://t.co/EMHaJ6WTf1 ↪

Mathew ‘Night Science’ Ling (@lingtax; 3⁄1): @COBYolland @HDR_HeSSA Now @drjbeaudry should run a workshop on building an academic website in #blogdown and #rstats 😜 ↪

Kyle Tretina (@AllThingsApx; 3⁄0): My new website is now online (https://t.co/enHXfKmXVa). Constructive feedback appreciated. Thanks @dsquintana for the blogdown tutorial! ↪

Maëlle Salmon 🐟 (@ma_salmon; 2⁄0): @whatsgoodio I don’t know why emojis are bad in RStudio and have not reported the issue yet but in the meantime you can remove emojis by embedding the tweets with shortcodes instead of html https://t.co/1CtwP17TIT (Hugo shortcodes are great in general) ↪

Seth Dobson (@sethdobson; 1⁄2): ICYMI: my first post on my new #rstats blog made with #blogdown, a bit of R musing, some personal stuff, and dunking on SPSS via Google Trends > https://t.co/hcz6sboFdY ↪

John Blischak (@jdblischak; 1⁄1): @malaykbasu Subdirs are a fundamental issue for rmarkdown, and thus any pkg that builds on it. This is b/c it gets ambiguous how to link the external files (figs, CSS, JS). The only Rmd option with subdirs is blogdown: > https://t.co/cGcikfWmYK ↪

Chelsea Clifford (@DitchOntologist; 1⁄0): @trishuphigh I missed the PhD news- congrats & GOOD LUCK! Didn’t get website until defending/job search; still feel silly about it, but is nice to compile stuff scattered across internet. Lost good recent article about this, but it suggested #scicomm & code, beyond usual. See R blogdown? ↪

Sina Rüeger (@sinarueeger; 1⁄0): @R_by_Ryo @i_steves @ma_salmon @UseR2019_Conf I also had problems with the {{ showing up with blogdown/Rmd. How did you make it work? ↪

⚽️Ryo Nakagawara📊 (@R_by_Ryo; 1⁄0): @TonyElHabr nice and thanks for the shout-out! and I really need to upgrade to using {blogdown} soon … 😐 ↪

atusy (@Atsushi776; 1⁄0): @ymn753 レベルたかい……．blogdownも分類モデルも使える人って誰だろ……？ ↪

bookdown

R for the Rest of Us (@rfortherest; 110⁄28): This bookdown project from @nj_tierney, RMarkdown for Scientists, has some great tips on improving your workflow. https://t.co/2Wa3ikwVuB #rstats ↪

Edwin Thoen (@edwin_thoen; 24⁄1): Using #rstats bookdown for the first time. No matter how many typos you make, the layout alone makes it look so professional. ↪

Kristijan Bakaric (@kbakaric1; 16⁄6): YaRrr! The Pirate’s Guide to R Very fun and educational. > #rstats > https://t.co/mGjBnHUbzT ↪

Jesse Mostipak @ #NCES STATS DC (@kierisi; 16⁄5): I’m still very much in the learning stages of bookdown in #rstats, and I was curious: is there a way to deploy a bookdown project on https://t.co/BaNE2FOTP6? I realize this is a weird and possibly “why would you want to do that?!” question, and the answer is reasons. ↪

atusy (@Atsushi776; 14⁄7): これで knitr → rmarkdown → bookdown → pagedown と一連のパッケージの contributor になったぜ☆ ↪

Edwin Thoen (@edwin_thoen; 14⁄6): Agile Machine Learning with #rstats book in the making now uses continues integration using @rOpenSci’s tic.bookdown. Builds after every commit, so always up-to-date https://t.co/okmnBBU18n ↪

David Reinstein (@GivingTools; 12⁄7): The future of “textbooks”: https://t.co/EEfaV1S7TS. A great example of what one can do with R and #bookdown. “Open review” allows continuous feedback from readers! #econometrics text adapting Stock Watson #rstats. @xieyihui ↪

Philipp Bayer (@PhilippBayer; 10⁄0): @JaydenGreenwell I’d refer you to @RachaelLappan who wrote her whole PhD thesis in markdown (bookdown to be precise), but she’s not in Perth any more (booo ;) ) > I’d also play around with workflowr for whole reproducible projects in markdown https://t.co/KLbflT8bXA ↪

Alessio Passalacqua (@alessiopassah2o; 7⁄2): 📕 #Rbook: “Agile Machine Learning with R” by @edwin_thoen 💪 #getstrongeR #Rstats #machinelearning https://t.co/eEagrcEg7W ↪

Francois Dion (@f_dion; 6⁄2): @kierisi You could make a small shiny app that is just a fluid page with includeHTML and call bookdown::render_book(‘index.Rmd’, “bookdown::gitbook”) > #rstats ↪

Bram Zandbelt (@bbzandbelt; 4⁄1): @AkhmerovAnton I’m writing papers in RMarkdown, knitr, bookdown, and pandoc. Integrating text, code, and figures, and rendering to PDF is pretty straightforward. ↪

The Real PhDeal (@RPhdeal; 4⁄1): @you2nice2me @AcademicChatter Ooo actually for a basic introduction to R if you’ve never programmes before, this is a good place to start - https://t.co/ZFTo2M9A7l It’s also all pirate themed! ↪

Ben Marwick (@benmarwick; 4⁄1): @ishiijunpei I agree it’s confusing! There’s a very good explanation here https://t.co/Wt30ZA0Sm3 Briefly, seems like rmarkdown is for formatting & pandoc, while knitr is for code chunk execution & output ↪

Giovanni Pavolini (@gpavolini; 4⁄0): @RikaGorn @rstatstweet I use #Rmarkdown and choose powerpoint_presentation as the output in the YAML Metadata. Tables often need some heavy editing, but code is usually OK. > Check https://t.co/oNA8Hi7hzC ↪

Martin Frigaard (@mjfrigaard; 4⁄0): @kierisi @jrosenberg6432 @RyanEs @ebovee09 @ivelasq3 Congrats 😁 It actually is my jam! Will you be writing with bookdown? ↪

Kelly Bodwin (@KellyBodwin; 3⁄0): @JuanM_Ca @rstats4ds @RikaGorn I highly recommend {xaringan} as well! > https://t.co/bdSkyWvzmC ↪

Paula Andrea (@orchid00; 2⁄3): Sharing is caring #rstats #bookdown #bookdown #xaringan by @xieyihui https://t.co/eyIslLHRgd you might like to try it out! @RLadiesBrisbane @GrantChalmers ↪

Julian Gerez (@JulianEGerez; 2⁄1): I’m having issues rendering theorems and proofs in #rstats bookdown. For those who have used it in the past, I posted an issue to @StackOverflow here: https://t.co/bJK88pCL6y > Any suggestions would be much appreciated! ↪

Steven Senior (@stevensenior; 2⁄0): @NHSrCommunity I’d love a digital R library. I’ve been trying to think of a good way to keep a list of all those fantastic free R and other stats books that are out there. Stuff like https://t.co/pGekZ5k6vJ, https://t.co/dvBqlILPwK, https://t.co/Lffz1kJL0b, https://t.co/gKjUIRdEUO and so on. ↪

David Smith (@revodavid; 2⁄0): @kartar @kierisi Hmm @dataandme would know about https://t.co/6ghofeaNtZ, but I think there is a way to publish bookdown to https://t.co/zVpCs2QiSh if that helps. ↪

Solomon Kurz (@SolomonKurz; 2⁄0): @Heinonmatti @matherion @LauraMKoenig @JkayFlake @paulbuerkner @vuorre Also, the code needs a bit of an update, but you might check out the mediation chapters in my #brms translation of Andrew Hayes’ Introduction to Mediation, Moderation, and Conditional Process Analysis. You’ll get a lot of practice with multivariate models. https://t.co/SSZViepkrd ↪

matti heino (@Heinonmatti; 2⁄0): @LauraMKoenig @AlxEtz @JkayFlake @matherion @rlmcelreath’s Statistical Rethinking is wonderful, and the YouTube lectures (https://t.co/psJK6lFHyh) are pretty funny. @SolomonKurz translated the book/course into modern R (https://t.co/1jb4Ffsfep), so you can learn that too as you go :) ↪

prasanna (@thegymnosophist; 1⁄1): To this goodness, I’d also add bookdown-tools https://t.co/GzGYKdlc3O > which makes .Rmd editing in #vscode fairly painfree. #rstats ↪

Edwin Thoen (@edwin_thoen; 1⁄0): @ma_salmon But did not know there was something specifically aimed at bookdown! ↪

Danielle Johnson (@DJohnsonTrials; 1⁄0): Theres a few basic #R things I missed when I first learned, and this is proving to be such a good resource - https://t.co/XlRvKXzeb5 ↪

Francois Dion (@f_dion; 1⁄0): @kierisi I do something similar, but 1- run my own shiny server, 2- serve some of the static from an nginx server fronting everything as a proxy (plus 2FA), 3- only works on company network, either direct physical cable connection or vpn. 4- straight rmarkdown instead of bookdown ↪

Joshua Rosenberg (@jrosenberg6432; 1⁄0): @oscar_b123 @kierisi @ebovee09 @RyanEs @ivelasq3 @routledgebooks We were fortunate in that @xieyihui’s bookdown text, which was published with @CRCPress (which is a part of the same publisher as Routledge), had a web version, and so we were able to point to that when we asked ↪

Mara Averick (@dataandme; 1⁄0): @revodavid @kartar @kierisi I don’t see why you wouldn’t be able to, but usually I see almost the reverse embedded (like a shiny app embedded in a bookdown book) ↪

👨🏻‍💻Scott 👨🏻‍🔬📈📊📉 (@fusionet24; 1⁄0): @kierisi I believe its possible since shiny server can host static rmds. @xieyihui has a section in the rmarkdown book for something similar - https://t.co/EPmxjxViO3 ↪

Bram Zandbelt (@bbzandbelt; 1⁄0): @AkhmerovAnton If needed, you can still use LaTeX in your R Markdown document, e.g. see here https://t.co/QADFL7sdPR ↪

CientíficasEcuador🇪🇨👩🏽‍🔬 (@CientificasEC; 1⁄0): @GabySandovalM ¡Gracias @GabySandovalM por aportar al hilo! Excelente sugerencia la de #RMarkdown. Para quienes usan #python el equivalente es #JupyterNotebooks. Y si vamos un pasito más, que tal #bookdown. Podemos consolidar como un #libroDigital nuestro #coding👩🏽‍💻 ↪

Todd R. Johnson (@johnsontoddr4; 1⁄0): @stephenfloor In terms of format, https://t.co/ySbJVoLNcU is innovative, but clearly takes a lot of effort. https://t.co/1mSbjgCqu2 is easier. Bret Victor has good advice on redesigning scientific communication: https://t.co/0W3QKDmyd0 ↪

knitr

Gabby “ocelot & side project” Palomo (@gabbspalomo; 15⁄2): If you are using camtrapR I have a couple of tips for you today: 1. Save your detection history in an .rds file so you don’t have to run the whole code again. 2. Use knitr::kable to visualize the detection history (or any other table) -> looks nicer! https://t.co/qGpKzWqvvi ↪

atusy (@Atsushi776; 8⁄2): Pandoc’s markdown → knitr → rmarkdown → rmarkdownを拡張するパッケージって感じで本をわけていくか． knitr 本は kohske さんの「ドキュメント・プレゼンテーション作成」があるけど，Rmd特化で解説した本は多分ないよね？ ↪

Benjamin Wolfe (@BenjaminWolfe; 6⁄1): I love using roxygen-style #rstats comments— > #’ > —in part because @rstudio automatically continues the comment when I hit Enter. (I also love it for knitr::spin—thx @JennyBryan for that tip!!) > Is there an easy way, when you switch back to code, to get a new line without the #‘? https://t.co/Mntgq8jP9M ↪

Simson Garfinkel (@xchatty; 3⁄0): I was not aware of knitr, which apparently is supported on #overleaf! https://t.co/bnYELme4Q3 https://t.co/fdTXNo4ZfY ↪

boB 🇷udis (@hrbrmstr; 2⁄0): In more technical news, adding the pngquant binary to all the hrbrinfrastructure enabled adding: > knitr::knit_hooks$set(pngquant = knitr::hook_pngquant) > to all the places I stick images into R packages and enable hrbrthemes & ggalt to slip well under the CRAN MB limit radar. ↪

Dr. Kirby Conrod (@kirbyconrod; 2⁄0): reasons knitr doesn’t work currently: - overleaf gets too mad about the dollar signs - my data is in a weird shape and I do not know why so my old code isn’t even replicating graphs right now ↪

annakrystalli (@annakrystalli; 2⁄0): @kirstie_j @Jade_Pickering @EikoFried @pophealth3 @KMKing_Psych @mybinderteam @ajstewart_lang @rOpenSci Function write_bib might also be useful for writing package citations to a bibTex database: https://t.co/ZWgPWifV9j ↪

💧Morpheus #Progressive (@MorpheusBeing; 1⁄0): @_erikaroper You do use RStudio for your stats? And you do know about knitr that will all pdf output direct? ↪

Shabby Hats Only™ (@shabbychef; 1⁄0): @xchatty @healthyalgo or use knitr with python blocks. Side effect: you can run bash, R, SAS, whatever, in chunks. ↪

Ben Marwick (@benmarwick; 1⁄0): @ishiijunpei Right, if I understand correctly, the knitr pkg will use other pkgs to do the conversion between formats (we don’t need to load them by ourselves) ↪

石井淳平 (@ishiijunpei; 1⁄0): @benmarwick Thank you!! I could not understand the relationship with Rmarkdown.Because I could output HTML even with knitr. Document type conversion is not a knitr, but a feature of Rmarkdown? ↪

石井淳平 (@ishiijunpei; 1⁄0): rmarkdownとknitrの関係がうまく理解できていない。それぞれがどんな処理を担当してるんだかよくわからん。 #rmarkdown #knitr ↪

Jo Etzel (@JosetAEtzel; 0/1): PSA: If you can’t compile #knitr documents after updating #rstats and/or RStudio, try @xieyihui’s TinyTeX https://t.co/Q4vqqo5xnJ. ↪

pagedown

James Goldie (@rensa_co; 4⁄2): Been giving pagedown a real workout in #rstats this week. Very happy with it! So ready to leave LaTeX behind 😊 Thanks @xieyihui! > https://t.co/KrNgBaRPjr ↪

tinytex

Nicholas Tierney (@nj_tierney; 15⁄3): I got very frustrated trying to find + install packages in a LaTeX document. > So I wrote an #rstats 📦 https://t.co/FVmTe87k4g with Claire Miller > I have a sneaky feeling ninja hacker @xieyihui has already solved this in TinyTex, but for the moment, here’s our solution, enjoy! https://t.co/HANwfcsj8E ↪

Miles McBain (@MilesMcBain; 2⁄0): @nj_tierney @xieyihui Slightly different approach on offer in tinytex see parse_packages(). 👍 ↪

Nicholas Tierney (@nj_tierney; 1⁄0): @MilesMcBain @xieyihui Oh right! Yeah I’ll write an issue on tinytex and see what Yihui says. ↪

Miles McBain (@MilesMcBain; 1⁄0): @nj_tierney @xieyihui Yup. There’s a wrapper for this that will try to build and install missing packages repeatedly: latexmk() your approach is faster. Maybe you should PR it to tinytex? ↪

xaringan

Emi Tanaka 🌾 (@statsgen; 204⁄57): Collapsible code output for #xaringan #rstats > 👩‍💻🔗 https://t.co/Z5JwN63JDl https://t.co/bSIbh8Lh78 ↪

Yihui Xie (@xieyihui; 5⁄0): @er13_r Sorry, but I wasn’t there. Glad to know the popularity of xaringan, though. I guess people are still waiting for the day when yolo: true is no longer an inside joke :) ↪

Rizqy Amelia Zein (@ameliazein; 3⁄2): Slide bisa diunduh di https://t.co/3jbREvUbk1 Kode (xaringan) dapat diakses dimari https://t.co/6i0WByICwz. Lisensi CC0 (public domain) > Jangan lupa kunjungi https://t.co/GWqKQJUn9r untuk kegiatan2 yang lainnya :) @sainsterbuka https://t.co/VYhnMUWq8e ↪

Todd R. Johnson (@johnsontoddr4; 2⁄0): @nickchk I can recommend https://t.co/4HTejFjq1h ↪

Yihui Xie (@xieyihui; 2⁄0): @BeaMilz @juliasilge @jakub_nowosad Absolutely! The Github repo for the book is emitanaka/xaringan-book (I don’t want to publicize the actual link here since I’m far from being ready to start writing anything for the book yet). Any kind of contribution (idea, suggestion, writing, …) will be appreciated! ↪

Ingo Rohlfing (@ingorohlfing; 1⁄0): @JustinEsarey To send a really strong signal, you should use #Xaringan ↪

Dom $alva (@AKAperdido; 1⁄0): @TavaNaBad Xaringan ↪

yihui.name

Ben Marwick (@benmarwick; 7⁄3): @acerbialberto @CTennie Here are a few: https://t.co/SccGpOxnfd https://t.co/LEjOtd4rkX https://t.co/lrbOWoCet4 for articles, Rmd is ☑️ because it’s simpler (journal publisher will apply their own stylings), & can include #rstats code, & LaTeX for equations. Many templates here: https://t.co/w4wpZfB2dW ↪

Muhammad Aswan Syahputra (@aswansyahputra_; 5⁄0): @LizEisenhauer @rstatstweet @rstudio You may check {xfun} 📦https://t.co/MhhPDWrtrI ↪

R Weekly Live (@rweekly_live; 2⁄3): Romain Lesur: The Most Wonderful Collaborator I Have Ever Worked With @xieyihui #rstats #datascience https://t.co/lhL1aZgNtK ↪

Henning (@henningsway; 2⁄1): @LucyStats @jtleek https://t.co/pfAqdcqzpF ↪

Yohan J. Rodríguez (@hasdid; 1⁄1): #R #Automated | Romain Lesur: The Most Wonderful Collaborator I Have Ever Worked With https://t.co/1yevBsU8yO ↪