/NLP-with-LLMs

Natural Language Processing with Large Language Models

Primary LanguageJupyter NotebookMIT LicenseMIT

Natural Language Processing with Large Language Models

Purpose

Jon Krohn created this repo to accompany his half-day training on NLP with GPT-4 and other LLMs: From Training to Deployment with Hugging Face and PyTorch Lightning, which was first offered at the Open Data Science Conference (ODSC) East in Boston on May 10th, 2023.

Code

Code can be found in the aptly named code directory:

  • Jupyter Notebooks are directly supported for execution in Google Colab
  • .py files are for running at the command line (see instructions)

N.B.: Code is intended to be accompanied by live instructions and so it will not necessarily be self-explanatory.

Repo Art

The "repo art" above was generated by prompting Midjourney v5 with this artistic take on LLMs (that was output by GPT-4):

Painting of a harmonious blend of alpacas and vicuñas in rich shades of caramel and soft gray, amidst a lush, futuristic landscape. The animals are surrounded by a web of glowing, pulsating neural network connections in hues of electric blue and neon green, symbolizing the cutting-edge and cost-effective AI training techniques. In the background, a dynamic matrix of binary code cascades down, further emphasizing the technological prowess of the scene.