posit::glimpse() Newsletter

Welcome to our newsletter, posit::glimpse(). If you’re currently reading this on our blog, consider subscribing to Stories on our subscription page to receive this newsletter directly in your inbox.

Sharing is caring in data science, and this month there are even more ways to share the results of your data science analysis. Quarto 1.3 has multi-format publishing and Confluence publishing, Shiny is on Hugging Face, {pins} offers enhanced functionality with Connect, and tables with {gt} can make your tables look even better. Plus, if you’ve been wanting to try out LLMs for yourself, and share some chat applications with friends, we have a how-to with Shiny for Python. We’re excited to see what and how you share!

Roundup

• AI chat apps with Shiny. If you’ve wanted to make a web application for interacting with AI, you can do it with Shiny for Python. Winston Chang shows how you can easily create your own chat application with just a few lines of code. 
• Shiny + Hugging Face. We’ve partnered with Hugging Face to provide templated Spaces for both the R and Python Shiny packages. These Spaces allow you to deploy an R or Python Shiny app on Hugging Face with a click of a button.
• Quarto 1.3 released! Just some of the new features include code annotation, multi-format publishing, Jupyter cell embedding and Confluence publishing. Learn how to get started with Quarto or see a potpourri of Quarto features useful for academics in these videos from Mine Çetinkaya-Rundel. 
• Tidymodels Q1 roundup. New releases and features in tidymodels include tuning parameters, speed improvements and 24 releases to CRAN. 
• New features for better tables. Rich Iannone shares updates in the {gt} 0.9.0 release, and how to use data cell colorization and make good summary rows.
• LLaMA in R with Keras and TensorFlow. Tomasz Kalinowski walks through an implementation of LLaMA, a Large Language Model, in R, with TensorFlow and Keras.
• Join us at posit::conf(2023) in Chicago or online September 17-21. Workshop instructors share a few words in these videos about what they’ll be teaching, so you can select the right ones for you!

Learn. Teach. Share.

• Why Shiny for Python? Learn how Shiny for Python’s design philosophy sets it apart from Streamlit and Dash.
• Driving impact for data science communities. Rachael Dempsey shares guidance on building a successful data science community with lessons learned from pharma leaders doing this today.
• Deploying a Shiny for Python app with Posit Connect. Ryan Johnson shows how to deploy a Shiny for Python app using Posit Workbench and Posit Connect.
• Getting started with RStudio on Amazon SageMaker. RStudio is now available on Amazon SageMaker, which enables R users to start coding in AWS with just a few clicks without needing to configure a server environment. 
• Group equivalent neural networks escnn. Sigrid Keydana introduces the mathematical concepts and essential library use of escnn, an extension of PyTorch, that provides a high-level interface to designing and training group-equivariant neural networks. 
• Python made easier with Posit Team: Tom Mock discusses our commitment to Python from open-source projects to enterprise support, and how Posit Team helps you share and maintain your R & Python data products.

Selected Releases

• {pins} 1.2.0 – The updated pins version offers enhanced functionality, including reading and writing pins using Parquet, reading from Connect vanity URLs, and preventing duplicate pins.
• {desirability2} – A new tidymodels R package for desirability functions with an interface conducive to being used in-line via dplyr pipelines. 
• Posit Package Manager now allows you to block packages by open-source license and create curated repositories of Python packages from PyPI.
• Integrate Posit Workbench with your Altair Grid Engine. Altair has partnered with Posit PBC to integrate Workbench with Altair Grid Engine (AGE), enabling the execution of Workbench sessions on AGE-based HPC clusters.
• Posit Workbench on Google Cloud Workstations: You can now easily and securely access your favorite editors with Posit Workbench on Google Cloud Workstations.

Wrapping Up

Don’t miss out on future issues – subscribe to Stories on our subscription page to receive this newsletter directly in your inbox. If you prefer reading online, you can still view this newsletter on our website.

We value your feedback and suggestions for our newsletter! Share your thoughts with us by leaving a comment in the Posit Community. You can also stay updated by following us on our Twitter and Mastodon accounts, @posit_glimpse. Thank you for being a part of our community!

And finally:

What do you get when you cross a mosquito with a mountain climber? 

…

Nothing. You can’t cross a vector and a scalar.