AI Engineering Foundations: sections and modules
Hugging Face Models & Gradio Demos
Log in to the Hub and load datasets and models.
- Lessons
- 5 lessons
- By
- Sourav Banerjee
- Prerequisites
- None
Lessons
Start moduleNotebook-only lessons
These course lessons have no page here yet. Work through their notebooks on GitHub.
Images, diffusion and video generation
- Understanding image data(optional)
- Understanding diffusion models(optional)
- Autopipelines diffusers(optional)
- Stable video diffusion(optional)
- Image2videogen xl(optional)
Audio: data, classification, transcription, generation
- Audio data(optional)
- Audio classification(optional)
- Audio transcription(optional)
- Audio generation(optional)
Practise and Check
- Deliverable
- Model demo: wrap an open embedding or text model from the Hub in a Gradio app that shows its output for a user-entered query.
- Assessment
- Answer the five Interview questions without notes, then run the lab.
- Interview probe
- Larkspur indexes each handbook page as one chunk with `all-MiniLM-L6-v2`. Support says "How fast does support answer a ticket?" never finds the FAQ page, though the answer is on it. What do you check?
- Common mistakes
- Loading weights from the
mainbranch instead of a pinned commit hash - Embedding a page longer than
max_seq_length, so its tail is cut - Building the pipeline inside the Gradio event function
- Loading weights from the
Your Progress
Saved in this browser when you leave the box.