Learn Hugging Face 🤗
Welcome to the 2nd best place on the internet to learn the Hugging Face
ecosystem (the first is the official
docs) and a bunch of AI and ML concepts
along the way.
This website dedicated to teaching the Hugging Face ecosystem with
practical examples.
Each example will include an end-to-end approach of starting with a
dataset (custom or existing), building and evaluating a model and
creating a demo to share.
Teaching style: a machine learning cooking show! 👨🍳
We focus on starting with raw ingredients such as a dataset or base
model and then we customize them to our own liking before uploading the
fine-tuned model and custom demo to Hugging Face so other people can try
it out.
This will be our (rough) workflow:
A general Hugging Face workflow from idea to shared model and demo using
tools from the Hugging Face ecosystem. These kind of workflows are not
set in stone and are more of guide than specific directions. See
information on each of the tools in the
Hugging Face documentation.
Contents
[!TIP]
Video Course Content
If you would like step-by-step guidance through each of the projects,
they are all available as video courses on the Zero to Mastery
platform.
You can sign up and go through them at your own liesure.
| Project | Description | Dataset | Model | Demo | Video Course |
|---|
| 0 - Text classification | Build project “Food Not Food”, a text classification model to classify image captions into “food” if they’re about food or “not_food” if they’re not about food. This is the ideal place to get started if you’ve never used the Hugging Face ecosystem. | Dataset | Model | Demo | Video Course |
| 1 - Object Detection | Build Trashify 🚮, an object detection model to detect “trash”, “hand”, “bin” to incentivize people to clean up their local area. Start with a dataset, customize an open-source object detection model and turn it into a demo application that others can use and try out on their own images. | Dataset | Model | Demo | Video Course |
| 2 - LLM Full Fine-tuning | Fully fine-tune Google’s Gemma 3 270M model to perform structured data extraction on any kind of text. | Dataset | Model | Demo | Video Course |
| 3 - VLM Fine-tuning | Fine-tune a small VLM model, SmolVLM2-500M for structured data extraction from images. | Dataset | Model | Demo | YouTube |
| 4 - Multimodal RAG (Retrieval Augmented Generation) | Level up your text-based RAG pipelines and learn how to embed text and images of documents into a shared embedding space. This allows you to query over a dataset of combined images and text. | Dataset | Model (embed) / Model (rerank) | Demo | YouTube |
| More to come soon! | Let me know if you’d like to see anything specific by leaving an issue. | | | | |
Extensions
| Extension | Description |
|---|
| Perform batched inference with an LLM and Hugging Face Transformers | Take an existing LLM and learn how to speed up prediction times by batching together samples. This is helpful if you need to perform inference over a large dataset. Uses a small model to run on Google Colab or locally. |
Updates
- 10 June 2026 - All videos for the LLM fine-tuning course are live
on the ZTM website, sign up to
learn how to fine-tune a Small Language Model (SLM) on your own custom
data.
- 16 Apr 2026 - Add batched inference with Hugging Face
Transformers
Notebook,
this extends from the LLM fine-tuning notebook and helps speedup
inference significantly.
- 1 Apr 2026 - Fully finished LLM fine-tuning
notebook,
recorded all video course videos for the notebook, they will be
available soon on ZTM.
- 26 Feb 2026 - Update links to LLM fine-tuning, VLM fine-tuning,
Multimodal RAG notebooks (these all work, however, I’m in the process
of tidying them up). Add dark mode to notebooks.
- 08 Jan 2026 - Add first iteration of full fine-tuning notebook
for Gemma 3
270M
(fully fine-tune a Small Language Model for structured data
extraction)
- 07 Nov 2025 - Videos for the object detection project are now
available on ZTM in the Hugging Face
Bootcamp!
- 18 June 2025 - All code has been completed for the object
detection
project,
train a custom object detection model and make a demo with it for
others to try! (video course to come soon)
- 1 Oct 2024 - Video course version of text classification is live
on ZTM! Inside,
we’ll walkthrough every line of code building the text classification
project
with Hugging Face Datasets, Transformers and Spaces.
Who is it for?
Ideal for:
- Beginners who love things explained in detail and would like to
fine-tune their own AI models.
- Someone who wants to create more of their own end-to-end AI/machine
learning projects.
Not ideal for:
- People with 2-3+ years of machine learning projects & experience^.
^Note: This being said, you may actually find some things helpful along
the way. Best to explore and see!
The course is very hands-on. We teach concepts interweaved with code and
always push towards deploying publishing real-world applications.
Our project style will be: data, model, demo.
- Data – Create a new/reuse an existing dataset.
- Model – Train/evaluate a model (Hugging Face hosts many 1000s of
models we can download and use).
- Demo – Build a demo to share (a demo is one of the quickest ways
to let other people try your work).
As practitioners, we’ve got several mottos:
- If in doubt, run the code. – Machine learning is very
experimental. So it’s good to get in the habit of continually trying
things (even if you think they won’t work).
- Visualize, visualize, visualize! - If you’re not sure of some
dataset or some operation or some predictions, visualize it/them.
- Experiment, experiment, experiment! - Again, machine learning is
very experimental. So keep trying different things!
- Data, model, demo! - Create/get a dataset, build/train/evaluate
a model, create a demo to share.
Prerequisites
- 3-6 months Python experience.
- 1x beginner machine learning or deep learning course (see my
begineer-friendly ML course to
learn Python + important ML concepts in one).
- PyTorch experience is a bonus (see my Learn PyTorch in a Day
video or
learnpytorch.io)
What is Hugging Face?
Hugging Face is a platform that offers access to many different kinds of
open-source machine learning models and datasets.
They’re also the creators of the popular transformers
library (and many
more helpful libraries) which is a Python-based library for working with
pre-trained models as well as custom models.
If you’re getting into the world of AI and machine learning, you’re
going to come across Hugging Face.
A handful of pieces from the Hugging Face ecosystem. There are many more
available in Hugging Face
documentation.
Why Hugging Face?
Many of the biggest companies in the world use Hugging Face for their
open-source machine learning projects including
Apple,
Google,
Facebook (Meta),
Microsoft,
OpenAI,
ByteDance and more.
Not only does Hugging Face make it so you can use state-of-the-art
machine learning models such as Stable
Diffusion (for
image generation) and
Whipser (for audio
transcription) easily, it also makes it so you can share your own
models, datasets and resources.
Aside from your own website, consider Hugging Face the homepage of your
AI/machine learning profile.
TODO
FAQ
Is this an official Hugging Face website?
No, it’s a personal project by myself (Daniel
Bourke) to learn and help others learn the
Hugging Face ecosystem.
How is this website made?
This is a Quarto website.
To learn more about Quarto websites visit
https://quarto.org/docs/websites.
