To set the stage: what’s an NLP task? It could be anything from text classification (figuring out if a piece of text is positive or negative) to machine translation (translating one language into another). The course covers all sorts of classic NLP problems, so you can choose the ones that interest you most.

Now how this course works. Each chapter is designed to be completed in 1 week with around 6-8 hours of work per week (but feel free to take as much time as you need). The first four chapters cover the main concepts behind Transformer models, which are a type of neural network that have become really popular for NLP tasks. By the end of this part of the course, you’ll be able to use a model from the Hugging Face Hub (which is like a library of pre-trained models), fine-tune it on your own dataset, and share your results with others!

Chapters 5 through 8 teach you how to work with datasets and tokenizers (which are tools for breaking down text into smaller pieces) before diving into classic NLP tasks. By the end of this part, you’ll be able to tackle all sorts of problems by yourself!

If you have any questions during the course, just click on that “Ask a question” banner at the top of the page and it will automatically redirect you to the right section of the Hugging Face forums. And if you want some project ideas once you’ve completed the course, there’s a list available on the forums as well!

As for getting the code for this course, feel free to do so in any reasonable manner (just don’t suggest that we endorse you or your use). If you want to cite the course, here’s some BibTeX:

@misc{huggingfacecourse,
author = {Hugging Face},
title = {The Hugging Face Course, 2022},
howpublished = “\url{https://huggingface.co/course}”,
year = {2022},
note = “[Online; accessed ]”
}

And if you’re curious about the StackLLaMA model (which combines approaches from InstructGPT and LLaMA), check out this demo below!