View profile

Quarto, Github and Twitter Networks | Next - Issue #38

Harshvardhan
Harshvardhan
Hi there,
Over the last week, I’ve been discussing Hans Rosling’s talk with a reader. They rightly pointed out how these fantastic visualisations hide many inequalities within groups, that logarithmic scales underestimate the actual differences and that purchasing power parity measures are faulty.
I responded with my understanding of these measures and why Rosling’s visualisations could be hiding information but are not misleading. Thoughtful responses are the most authentic motivation to continue writing these letters weekly. The exchange would be an exciting read for curious readers like you, and I’d love to know your opinion on it.
Also, I’m taking a month-long break. I look forward to writing to you all and exchanging more opinions soon. Next will hit your inbox again on August 2, 2022.
Let’s dive in.

Five Stories
Committing mistakes in public is embarrassing. But beyond that uneasiness is the grand opportunity to learn in public. Jesse Mostipak occasionally live-streamed her video games, shifting to #TidyTuesday live streams and soon competing in SLICED. And there are many benefits, as she notes.
As I was working through a tutorial there’d be an occasional message from someone in chat pointing out an additional resource, code snippet, or solution to a bug I was grappling with, and suddenly it seemed like a silent stream was simply another way for me to learn in isolation.
Check out her post!
In April 2022, Martin Henze launched a Kaggle competition to design engaging data science notebooks with great visuals, narration, and insights. In this blog post, Martin shows how he built the competition’s leaderboard using the gtExtras package in R. Here’s how the final table looks. ✨ (Yes, it was made using R.)
Open source projects are likely more prone to abandonment than proprietary projects. Not because they’re open-source; but because they’re usually thankless jobs. Here’s xkcd to explain.
How to manage your expectations when reporting a bug or making a feature request on a Github project? Maëlle has some advice.
  1. Look if the repository is archived,
  2. Read contribution files and readme,
  3. Explore the latest code activity,
  4. Check pending PRs and issues, and more!
Quarto is the new method to make reproducible data science notebooks. Being a markdown document at heart, it can produce presentations, reports, blogs, websites, books and many more. It works with your favourite language: R, Python or Julia. See Alison Hill’s post for a gentle introduction to Quarto.
In this video tutorial, Albert (aka RappA) guides how to create custom-style Quarto blogs. He doesn’t assume much background knowledge of HTML and CSS and thus is excellent for beginners.
If you have used Twitter with R, you are likely aware of rtweet package. In this blog post, Keith built a network of British Members of Parliament based on how much they interacted with each other. He uses igraph and ggraph packages in the process.
Also check out his free book Handbook of Graphs and Networks in People Analytics: With Examples in R and Python to learn more about people analytics.
Four Packages
gtExtras provides several helper functions to create better tables with the gt package in R. See the vignette here.
huggingfaceR being state-of-the-art natural language processing models to R. See the vignette here.
igraph is a network analysis package built for performance and ease of use. See the vignette here.
ggraph introduces grammar of graphics to network visualisation and is built around ggplot2. See the vignette here.
Three Jargons
Networks are a set of objects sharing a relationship with each other; a graph where nodes have a name and can be identified is a network. Learn more.
Recurrent Neural Networks are a particular class of neural networks with temporal connections between the inputs and response. This is super helpful in handwriting recognition, self-driving cars, etc., where temporal memory significantly enhances performance. Learn more.
LaMDA, Language Model for Dialogue Applications, is Google’s conversational language model. It is built transformers with deep learning on extensive texts of documents and dialogues consisting of 1.56 trillion words and further refined via manually annotated responses for sensibleness, interestingness, and safety. Learn more.
Two Tweets
Quarto
Today we begin a new series: “A Quarto tip a day keeps the docs away” -- a playful attempt to share Quarto tips daily for the next month. Follow along:

🐦 #quartotip
🔗 https://t.co/Zd67RsMYau

And, of course, don't keep the docs away, they're incredibly helpful and thorough!
One Meme
Bonus
Did you know students and educators have free access to Github Pro? That also includes access to Github Copilot (which means AI can save you some keystrokes). It’s not available with RStudio yet but can be used in Visual Studio with R, Python and more.
That's a wrap!
I hope you enjoyed today’s letter. As a reminder, I’m taking a break for next month, and the next Next will reach you on August 2, 2022. Have a fantastic summer!
Harsh
Did you enjoy this issue? Yes No
Harshvardhan
Harshvardhan @harshbutjust

A short and sweet curated collection of R-related works. Five stories. Four packages. Three jargons. Two tweets. One Meme.

List of all packages covered in past issues: https://www.harsh17.in/nextpackages/.

In order to unsubscribe, click here.
If you were forwarded this newsletter and you like it, you can subscribe here.
Created with Revue by Twitter.
Knoxville, Tennessee.