RPIrates: Fun with OpenAI, GPTStudio and R!

Seemingly everyone has been talking about the impact of OpenAI's ChatGPT on, well, everything​, including writing code.  In this very special RPIrates we'll talk specifically about generating great, and sometimes not-so-great, R code based on OpenAI's "Codex" models, which are easily accessed via the OpenAI API with the help of RStudio extensions (addins) provided by the GPTStudio package. After a quick icebreaker in which we'll demo GPTStudio in action, we'll briefly review transformer neural networks; we'll talk about how code completion tools like Microsoft's IntelliCode Compose apply TNN frameworks to write scarily excellent code; we'll tour OpenAI's Codex documentation; and then we'll get back to more hands-on with GPTstudio. We'll also talk about why code completion frameworks like Codex are so great at Python and Javascript but spotty with R, and of course, the ethics; oh, the ethics!​ Many links for further learning and research will be provided. 

Selected links from the talk:

• GPTStudio: OpenAI extensions for RStudio: https://github.com/MichelNivard/gptstudio

• "Transformer Neural Networks: A Step-by-Step Breakdown": https://bit.ly/3lv78YB

• "Attention Is All You Need" (2017): https://arxiv.org/pdf/1706.03762.pdf

• "IntelliCode Compose: Code Generation using Transformer" (2021): https://bit.ly/3Xtwa7W

• "Evaluating Large Language Models Trained on Code" (2021): https://bit.ly/3E7VLfK

• "Competition-Level Code Generation with AlphaCode" (2022): https://bit.ly/3lwR9cC

• GitHub Copilot documentation: https://docs.github.com/en/copilot