#GPT4 is way better at generating code than previous versions! See a #rstats #shiny app fully generated by GPT-4. It displays a ggplot of mtcars, uses {shiny.semantic} for UI and has some nice styling in CSS 🚀 This level was impossible to achieve in GPT3. Here's the process🧵

Had a chance to talk to @zyla_kamil, who just released Rhino 1.9. I asked him what are the biggest updates in the recent releases. Rhino revolutionizes building R Shiny apps for enterprise. Best programming practices, automates them in code, adds modularity, easier maintenance.

Another huge release on the long-context LLMs front: Introducing Giraffe 70B, an open-source LLM fine-tuned from base Llama 2 with context length extended to 32K. The @abacusai team conducted an evaluation of the model assessing performance over long contexts. The 70B model improves significantly at the longest context windows (32k) for the document QA task vs. the 13B model, scoring 61% accuracy vs. the 18% accuracy of 13B on the AltQA dataset. I got excited with the earlier release of Giraffe because of the potential of these long context models to unlock unique applications. Even more exciting to see that the 70B Giraffe model shows the best performance of all the open-source models in key categories like extraction, coding, and math. Note that there is still a gap in performance to the best closed source models like GPT-4 but it's coming closer. I also wonder how these models could help with key capabilities like information extraction, large-form doc analysis, and reasoning. There is a lot to explore with these long-context models. Don't sleep on these. Great work by the Abacus team. Blog: blog.abacus.ai/blog/2023/09/… Model: huggingface.co/abacusai/Gira…

This is huge! @jcheng announced Shinylive for R at #positconf2023 🔥 Now you can: 1. Convert Shiny to static HTML/CSS/JS/wasm 2. Write Shiny apps directly in a browser 3. Embed Shiny as Quarto code chunks in other webapps #rshiny #rstats #positconf

💡Here is a list of really cool Shiny apps for Pharma and Biotech: go.appsilon.com/pharma-apps We've recently launched these two: FDA-Pilot-App: go.appsilon.com/fda-clinical… Drug Interactions App: go.appsilon.com/f 📩Contact us to collaborate on 🤖Machine Learning🤖 projects!

Who says @posit_pbc Connect is just for #RStats 📊 and #Python 🐍? Our team has paved the way for #Rust 🦀 #Actix APIs. Learn more: appsilon.com/deploy-rust-api… 🚀 #TechInnovation

🚀 Guess what? 🚀 {shiny.emptystate} has just landed on CRAN! 🎊 Meet a new member of #Rhinoverse and learn how to easily manage situations when there is nothing (yet) to show in your #RShiny application. Check the documentation to learn more: appsilon.github.io/shiny.emp… #RStats

Optimizing #RShiny applications with webR is a game-changer 🤯 By leveraging browser-based computation, we witness a significant reduction in server load and enhanced responsiveness. Dive deeper into the technical insights: appsilon.com/bringing-webr-i… #webR #RStats

🧬 We've successfully adapted the R Submissions Pilot 2 Shiny App submitted to the FDA to our latest framework - #Rhino 📊 Our open-source app (FDA pilot app) provides a groundbreaking example of an R/Shiny app that can be submitted to the FDA. 📖 appsilon.com/fda-clinical-tr… #FDA

Looking for tools to move your #RShiny application to the next level? Check out the #Rhinoverse - a family of #RStats packages built around {rhino}: rhinoverse.dev/

{data.validator} package lets you automate validation of the data your app relies on... and creates great-looking reports (in just a few lines of code!) We've just released a new version with a number of updates. #Rstats #Rshiny

🎉 Big news! Our #Data4Good project, Mbaza, made it to the Top 10 Outstanding AI Projects by @ircai_unesco! It helps preserve animals in national parks in Africa. Catch our Head of AI, Jędrzej's live presentation in just a few mins: youtube.com/watch?v=N8p4MEvD… #SDG #Mbaza @appsilon

Uses all the popular accoutrements such as LLaMA, LoRA, Langchain, etc.

Stop reading the text !!! Have a graphic dialogue with OpenAI’s #GPT4 with #Graphologue. It turns GPT4-generated text into interactive node-link diagrams in real-time. New Human-Computer Interaction #HCI research makes #AI easier to use. More at creativity.ucsd.edu/ai

What if AI agents could write their own tools/plugins? Introducing 🧰 Toolkit: the easiest way to create & discover AI Plugins for 🦜📎LangChain and ChatGPT toolkit.club Just describe a plugin in natural language, and get working LangChain Code. ChatGPT coming soon!

🤖 AutoGPTs will be a game changer, and here's why: 1. They overcome GPT models' small context length by leveraging external memory 🧠 2. They simplify coding by removing the need to copy stuff to/from the console 💻 3. They could potentially pave the way for AGI 🤔

We're launching a 🔥 series of #rstats events to keep @ShinyConf vibes alive all year! 🎉 Thrilled to kick off with @sparktuga from @appsilon, sharing insights on boosting Shiny apps with web tech & new standards. Fellow #Shiny folks, pls retweet to help spread the word! 📢✨

What is In-Context Learning (ICL)? It is a mysterious emergent behaviour in LLMs where a downstream task is well performed just by conditioning on input-output examples, without optimising any parameters. 👇 #machinelearning #deeplearning #prompting

Open-source ML is at it again! Introducing ColossalChat - an open-source solution for cloning ChatGPT with a complete RLHF pipeline. Here is what you need to know:

Exactly! GPT4 in apps development lets you spend more time on "what?" and "why?" and less on "how?"

Happening now: @elonmusk, Yoshua Bengio, Stuart Russell, Yuval Noah Harari, @Grady_Booch, @AndrewYang, @GaryMarcus, and >1000 others (myself included) have called for a temporary pause on training AI models bigger than GPT-4. futureoflife.org/open-letter…