#rstats
R Function A Day (@rfunctionaday; 367/70): ROC curves provide a convenient way to compare responses and predictions of a binomial model.
The {performance_roc} function from {performance} 📦 computes AUC metric and visualizes ROC curves for a collection of models 🥇🥈🥉
https://t.co/ZKty2kA5Br
#rstats #DataScience https://t.co/LYJgfvZHIF ↪
R Function A Day (@rfunctionaday; 244/40): Sometimes you may wish to insert a textual table in a {ggplot2} plot the same way you insert text labels at a specified position.
The {geom_table} function from {ggpp} 📦 offers just the geometric layer! 🙌
https://t.co/bQdaxiMffR
#rstats #DataScience https://t.co/6raba4JIDV ↪
Oliver Stringham (@OliverStringham; 195/40): Setting up PostgreSQL and PostGIS to Run Spatial Queries in R
In this tutorial I cover installation of PostgresSQL, creating schemas, saving data to tables, connecting from R, and running spatial queries.
https://t.co/97KH8NcEa9
#datascience #rstats #tutorial #gis #postgis ↪
Catherine Adenle (@CatherineAdenle; 133/87): Infographic: Cheatsheet – Python & R codes for common Machine Learning Algorithms
#Data #DataScience #AI #cheatsheet #ML #Rcode #MachineLearning #Algorithms #datasets #DataScientists #Analytics #BigData #Rstats #Python #Tech #javascript #TensorFlow #Coding #100DaysOfCode https://t.co/gJACnsB1au ↪
blogdown
Beatriz Milz (@BeaMilz; 66/18): Pessoal, amanhã eu vou participar de uma live no canal @GeoCastBrasil !
Vai ser um live-coding mostrando como usar o novo tema #hugoapero com o pacote #blogdown para criar blogs com R, e fazendo o deploy com @Netlify !
#RStats
https://t.co/VXoiist5GQ https://t.co/5VKx8mefLQ ↪
Beatriz Milz (@BeaMilz; 23/1): Gente, a live foi super legal! Obrigada pelo convite @GeoCastBrasil !
O material está neste link: https://t.co/3NXHp5kch5E lembrando que tudo é muito baseado no importante e incrível trabalho que a @apreshill e a @rstudio tem feito. Eu apenas apresentei o conteúdo em Português! https://t.co/XkvJ8O3ae9 ↪
Norimitsu Nishida (@NorimitsuNishi1; 9/0): Rの優位性は、Rmarkdown blogdown bookdown shinyあたりでしょうか？（もちろん、これらPythonで同じ用なことが出来ます） ↪
Beatriz Milz (@BeaMilz; 8/0): eu to usando pra desenvolver esse site: https://t.co/wNiWJgrLuw
é do evento @sicam_procamusp
é feito com #blogdown :) ↪
Lucas Stefano (@_lucasstxs; 5/4): Day 03: I started customizing my academic site using #blogdown and @GoHugoIO . The idea now is to customize SCSS with my favorite colors. 😊
#100DaysOfCode https://t.co/R5LgnfQo2b ↪
Michael Dorman (@MichaelDorman84; 4/0): @Noa_Liberator https://t.co/ZMqIzmncAH https://t.co/rbhgzXxxEc ↪
Derek Dixon (@ddixonAI; 2/5): Here is my first attempt at a blog post using Blogdown and Netlify. I discuss some of the initial concepts underpinning machine learning. Feedback is appreciated and I hope you enjoy.
https://t.co/udzIQ8AEjp
#RStats #MachineLearning #Statistics ↪
Jonathan Kitt (@KittJonathan; 2/0): 3⃣Building your website with blogdown📦
https://t.co/6hgofChpIf
@shilaan01 ↪
Beatriz Milz (@BeaMilz; 1/0): @apreshill @GeoCastBrasil @rstudio thank you for all the materials and tools that you make! if it wasn’t for your blog, I think I might have given up on blogdown in 2018/2019 🤣 I remember that most times that I got an error, when I searched for the answer, it was in your blog. So thanks so much! ↪
Liz Hare PhD 😷 (@DogGeneticsLLC; 1/0): @mchua @WordPress @Ghost If you are into R there’s rmarkdown/ blogdown. I don’t know a lot about blogging, but I’m looking forward to reading yours! Thanks for your interest in screen reader accessibility ↪
Enrico Manlapig (@EnricoManlapig; 0/1): #rstats #blogdown peeps: Can anyone point me to a friendly tutorial for how to secure a hugo academic site with netlify identity? i’m new to this and the documentation is all a bit overwhelming 😵💫 ↪
bookdown
Keith McNulty (@dr_keithmcnulty; 105/21): I’ve updated my book to the dev version of #bookdown. Nice improvements to bs4_book copy-to-clipboard and the math expressions in the footnotes have been fixed. Lovely! #rstats #rmarkdown #datascience #python
https://t.co/vx49KF1Omd ↪
Keith McNulty (@dr_keithmcnulty; 103/23): My first textbook is officially released today. Many thanks to the #bookdown team who have made book writing in #rstats and #python exciting and fun, and to @crcgrubbsd and @CRCPress for a smooth production process. #datascience https://t.co/Q6ZsDB1gOg ↪
Indrajeet Patil (@patilindrajeets; 77/13): If you write a manuscript in #rmarkdown, the references are included by default at the end of the article.
But what if you also wish to include appendices? 🤔This demo 📽️ shows how you can force a position for a bibliography. 💡
For more, see:
https://t.co/4kqTvNeOSh#rstats https://t.co/IbDTHYcSBr ↪
心理的に柔軟なクジラ (@matsuchiy; 68/13): Rの無名関数について調べてたら，テキサス大学のSPHの先生がAuthorの『R for Epidemiology』というbookdownサイトを見つけた。purrrの使い方について，身近な関数での使い方で解説してくれていて学習がはかどりそう。疫学分野のR解説，親切で素敵 https://t.co/foJEABiCrv ↪
JP Gannon (@jp_gannon; 20/9): I’m super excited to present about my #rstats #datascience #hydrology bookdown (https://t.co/SgDZzsYqpk) in this education session at AGU, and to see YOUR awesome submissions too! https://t.co/Ly660mGedw ↪
Maëlle Salmon (@ma_salmon; 12/1): @noamross - Just noting you can set up a template repo on GitHub. https://t.co/y70hwHyxj8
- You can also have RStudio project templates e.g. https://t.co/Jw29neaq69 ↪
One R Package a Day (@RLangPackage; 11/4): bookdownplus - A collection and selector of R ‘bookdown’ templates. ‘bookdownplus’ helps you write academic journal articles, guitar books, chemical equations, mails, calendars, and diaries. R ‘bookdownplus’ extends the features of ‘bookdown’, a… #rstats https://t.co/m1SGi8ivGX ↪
memovocab (@memovocab; 11/0): [Book] Data science with R for psychologists. Open access https://t.co/JfbhHXb9tH ↪
Joris Meys (@JorisMeys; 6/1): Been experimenting with maps in #RStats , and I have to admit it’s more fiddly than I imagined. But here it is, my first “I can live with that” map :-)
Some sources I found interesting:
- “Geocomputation with R” by @robinlovelace :
https://t.co/5q52omFwTH
chef’s kiss https://t.co/A5hJ4Ccj9g ↪
Sébastien Rochette (@StatnMap; 6/0): @noamross At @thinkR_fr we have a package that starts everything needed on our GitLab using {gitlabr} (dev version)
- Creates Gitlab repo, with skeleton
- Set up CI with Pages depending on pkg, golem or bookdown
- Parametrised Wiki to inform client
- 1st issue to explain DevOps to new ones ↪
Shoei (@shoei05; 5/3): “R MarkdownではJavaScritpチャンクも使用可能。詳しくは公式ドキュメント(https://t.co/AyAuwfeIqj )を参照。jQueryを標準で読み込んでいるので、そのまま使える” https://t.co/dAjUUbe2Hm ↪
Sébastien Rochette (@StatnMap; 3/1): @noamross @ma_salmon @beeonaposy There is {bookprep} from @MarionLouveaux . She proposed such a replacement for templated #bookdown projects. She uses multiple pre-formated templates depending on context.
See the vignette here: https://t.co/BHBjmMPlXO https://t.co/sxzQ9G9cOa ↪
Colin Gillespie (@csgillespie; 2/0): @noamross we currently auto generate a google drive, asana project, gitlab repo, bookdown type handbook and slack channel. The various bits have templates. They are all linked together via client and project ids ↪
Dr. Elin Waring (@ElinWaring; 2/0): @ma_salmon @noamross 💚💙 I’ve been using Github templates for my courses but now that you said this, I’m definitely going to create one for general projects and bookdown sites. ↪
Chris Henson (@chenson_; 2/0): @JonTheGeek @kierisi @kaija_bean @NicholasStrayer Interesting! Something where people can share core examples is definitely doable.
Imagine a repo of shiny reprex apps and a bookdown doc that anyone can make a PR to. Do you think there would be much interest in this? If so I’d be willing to take a stab at setting it up ↪
Chris Henson (@chenson_; 2/0): @kierisi @kaija_bean @NicholasStrayer Thanks! Also another cool thing is that if you have these example apps deployed somewhere, you can embed them in documentation.
I make the app, then in bookdown Rmd, I put <iframe src=“link to app”> so that the shiny app appears inline with the discussion of the feature. ↪
tidyverse tweets (@tidyversetweets; 1/2): How can I get my R bookdown (gitbook output format) table of contents to start out initially not expanded? #tidyverse #rstats https://t.co/gD4Cvdzi7x ↪
tidyverse tweets (@tidyversetweets; 1/1): Bookdown error when building to word doc #tidyverse #rstats https://t.co/0NFdBSZrFM ↪
Dr. Robert M Flight, PhD (@rmflight; 1/1): Spent afternoon writing this little #rstats 📦 basically b/c I didn’t know about {bookdown::word_document2}.
Might be useful to someone else.
https://t.co/XiRdzMzTTp ↪
Matt DeCarlo (@profmattdecarlo; 1/0): @amhgruber Looks like it might fit between our intro MSW research open textbook (https://t.co/BPbp5vN9xm) and this open textbook on very advanced stats in social work (https://t.co/bzglADjR1j).
Your authors may want to consider teaching @JASPStats instead of SPSS. Better & #FOSS ↪
Alex Kyllo (@alexkyllo; 1/0): @aiexplorations Yes, I use Kaplan-Meier, CoxPH and the lifelines package a lot, they’re just limited in terms of ability to model longitudinal features. Joint models are a good approach, lifelines doesn’t include them, so you can’t do this in Python: https://t.co/cKch4ekbTW ↪
Daniel Sjoberg (@statistishdan; 1/0): @kylefbutts The results=asis was key! I ended up using this templating option I was not aware of
https://t.co/ZA0tieAZwz ↪
Jean-Paul R. Soucy (@JPSoucy; 1/0): @d_a_keldsen Some of the examples here are pretty close.
https://t.co/o2JOckRTrC ↪
Lyndsie Wszola (@Lyndsie_Wszola; 1/0): @KECoblentz Revising/commenting is the eternal dilemma. If it was just me, I would just write the whole thing in bookdown. ↪
Lisa DeBruine 🏳️🌈 (@LisaDeBruine; 1/0): @ibddoctor It’s just a link in the menu bar of a bookdown that takes you to the relevant GitHub page so you can make quick edits or others can suggest edits via fork and pull request (all online). ↪
knitr
Sophie Hill (@sophie_e_hill; 16/2): 🔥 #rstats take: if you’re making an html doc and you don’t need to show raw code, you should write in R not Rmd!
- faster to compile (no knitr)
- objects are loaded into the environment (easier to debug)
- can still edit css/js in separate files in RStudio ↪
Emily Nordmann (@emilynordmann; 3/0): @bmwiernik @LisaDeBruine @JenRichmondPhD @lingtax @cedricbatailler @page_gould Hmmm, probably knitr::kable() or patchwork (does that count as tidyverse-friendly or tidyverse?) ↪
atusy (@Atsushi776; 3/0): @kilometer00 Option templates?
https://t.co/07v3AANkmJ ↪
Matthew Martin (@hyperplanes; 3/0): My biggest obstacle to using r markdown and knitr is I still haven’t found a good approach for tables, especially regression tables. What do yall use? ↪
StackOverflow Bot (@soiqbot; 1/1): How to produce markdown document for each row of dataframe in R
https://t.co/xvTGz7tpOY
#r #knitr ↪
oh long johnson (@mendes_ph; 1/0): @RPColistete Muito bom! Eu tenho alguns templates que funcionam com rmarkdown e os outros pacotes do Rstudio (knitr, tinytex), são bem fáceis de usar
https://t.co/vYNqrp7hir ↪
Uriel Cohen Priva 🟣 (@urielcohenpriva; 1/0): I had to look at the knitr source to figure it out, so here’s my gift to you all: qtree(tikz) in rmarkdown:
```{tikz, echo=FALSE, engine.opts=list(extra.preamble="\usepackage{tikz-qtree}")}
\begin{tikzpicture}
\Tree [.VP [.VT tread ] [.NP [.N water ] ] ]
\end{tikzpicture}
``` ↪
Matthew Martin (@hyperplanes; 1/0): I like the idea of using markdown but 1)journal formatting requirements aren’t really compatible with any flavor of markdown, and 2)knitr has some major blind spots that honestly, if I were doing code review, are bad enough that it would not pass ↪
tinytex
Tinytex (@Tinytex_Aberoth; 2/0): #snapchatdown when you are supposed to “hang tight!” And it lasts 4 hours ↪
xaringan
Garrick Aden-Buie (@grrrck; 86/14): Hey #xaringan #RStats friends! Have you used xaringan for courses or workshops and included presenter notes? Please share! I’m looking for some good examples of real-world usage 🧑🏫🧑🎓 ↪
alakin4 (@kinenealan; 5/0): @grrrck Hi @grrrck, I used xaringan for my halftime PhD thesis presentation: https://t.co/GI0jtkg4J6. I hope these work for you. ↪
Jonathan Kitt (@KittJonathan; 4/0): 1⃣ Making great #slides with xaringan📦
https://t.co/qOSKvBWAMt
@grrrck, @spcanelon, @patriloto, @PipingHotData ↪
Gul Inan (@GulInan5; 3/1): @grrrck I prepared all my 14-week long lecture notes in xaringan this semester for this class: https://t.co/VK3VWHypj9. I do not disclose the lectures notes for a while, but here is one in .pdf:https://t.co/zDFCllWTyu. I used scribble a lot for writing/drawing on lecture notes. ↪
Emi Tanaka 🌾 (@statsgen; 3/0): @grrrck @visnut It means a lot coming from you, xaringan master! ↪
Lukas Burk / JeMScu (@Jemus42; 3/0): @hadleywickham @sharlagelfand @Emil_Hvitfeldt @theRcast Not for xaringan though as far as I remember, which is arguably the most important usecase for emoji 😁 ↪
Emil Hvitfeldt (@Emil_Hvitfeldt; 3/0): @apreshill @revealjs this is more related to how it is different than the default xaringan highlighter. The default barely highlights anything and isn’t super easy to configure. shiki does more highlighter, which is closer to what RStudio does ↪
Sergio Garcia Mora (@sergiogarciamor; 2/0): @grrrck I’ve been using #Xaringan for my training courses and also in presentations in People Analytics conferences like #PAWorld21 https://t.co/pnPEXN9c6T ↪
Maëlle Salmon (@ma_salmon; 2/0): @apreshill @revealjs I feel it’s like a punishment for not using xaringan 😅
–
With which things would be easier.
— ↪
Joseph Casillas (@jvcasill; 1/0): @grrrck Im pretty sure all if the xaringan slides here have presenter notes
https://t.co/tCjU1zuwvW ↪
yihui.name
Michael Sumner (@mdsumner; 1/0): @wycats best suggestion I ever saw was regular clearing of reputation, that would actually work and would work even now:
https://t.co/0FKpm0cRe2 https://t.co/KOw9nblhC4 ↪
Huihui Li (@lihuihui; 0/1): 随机漫步十八年的傻瓜 https://t.co/n1KgA2YkkI ↪