The Hugging Face Ecosystem

The Hugging Face Ecosystem

Duration: 2 Days

Description

Hugging Face has grown into the default hub for open-weight models, datasets, and tooling, and knowing your way around it is now a practical requirement for building with open AI. This hands-on course teaches developers to navigate the Hugging Face ecosystem end to end: searching and evaluating models and datasets, working with the transformers and datasets libraries, running local and hosted inference, and fine-tuning models for custom tasks. Coverage also includes deploying demos with Spaces and understanding licensing considerations when adopting open models in production. Participants leave able to source, adapt, and ship open models confidently in their own projects.
Class time is divided between instructor-led explanation (35 percent) and hands-on development (65 percent).

Audience

Developers and ML practitioners who want hands-on fluency with open models and the Hugging Face libraries, especially those looking to reduce reliance on closed, API-only models. Assumes basic Python and familiarity with ML or AI concepts.

Objectives

  • Navigate the Hugging Face Hub for models and datasets
  • Use the core libraries to load and run models
  • Run inference for common tasks across modalities
  • Work with datasets for training and evaluation
  • Fine-tune a model for a custom task
  • Apply best practices for using open models responsibly

Prerequisites

Participants should be comfortable with Python. The following will help:

  • Working knowledge of Python (see Python Foundations)
  • Familiarity with ML or AI concepts (see Machine Learning Foundations)

Related AI Courses

See the full AI training roadmap and course directory for how this fits into a broader learning path.

Pair this course with NLP Fundamentals to strengthen your language-model foundations, Machine Learning Foundations for the underlying theory, and MLOps & LLMOps to learn how to deploy and operate the models you fine-tune here.

Course Outline

  • Module 1 – Course Introduction
    • Welcome and objectives
    • The Hugging Face ecosystem overview
    • Open models and the community
  • Module 2 – The Hub
    • Finding and evaluating models
    • Datasets on the Hub
    • Model cards, licenses, and considerations
    • Hands-on exercise: explore and load a model
  • Module 3 – Running Models
    • The core libraries and pipelines
    • Inference for text, vision, and audio tasks
    • Tokenizers and preprocessing
    • Hands-on exercise: run inference tasks
  • Module 4 – Working with Datasets
    • Loading and processing datasets
    • Preparing data for training and evaluation
    • Metrics and evaluation
    • Hands-on exercise: prepare a dataset
  • Module 5 – Fine-Tuning
    • When and why to fine-tune
    • Fine-tuning workflow
    • Efficient fine-tuning approaches
    • Hands-on exercise: fine-tune a model
  • Module 6 – Course Wrap-Up
    • Key takeaways
    • Sharing and deploying models
    • Resources for continued learning