#2525th of 41 in Learning and cookbooks
transformers-tutorials
Colab notebooks for running and fine-tuning Hugging Face Transformers models
- Stars
- 11.8k
- License
- MIT
- Last commit
- Sep 2026
- Language
- Jupyter Notebook
11,458 on 31 Mar 202611,767 on 20 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
A collection of Jupyter notebooks by Niels Rogge showing inference and fine-tuning with Hugging Face Transformers in PyTorch. Coverage is mostly vision and document models (DETR, LayoutLM v1-v3, Donut, MaskFormer, Perceiver), plus some BERT, audio and GPT-J notebooks. Each demo links to a Colab notebook and the source paper.
Who it is for: Engineers learning Transformers fine-tuning for vision and document tasks
Strengths
- Every demo has an Open in Colab badge, so no local setup is needed
- Covers document AI in depth: LayoutLM v1-v3, LayoutXLM, Donut, DiT, MarkupLM
- Includes both inference and fine-tuning notebooks, some using HF Trainer or PyTorch Lightning
- MIT licensed, with a recent commit on 2026-09-17
Weaknesses
- PyTorch only; no TensorFlow or JAX notebooks
- Many notebooks use older APIs (e.g. DetrFeatureExtractor) and may need version pinning
- Mostly older architectures; no modern LLM fine-tuning notebooks in the visible README
- No releases and no packaged library; notebooks only
What it needs
- GPU optional
- Needs Hugging Face Transformers, PyTorch, Google Colab
- Models: DETR, LayoutLM, LayoutLMv2, LayoutLMv3, Donut
Also in Learning and cookbooks
See all 41| Rank | Project | Score |
|---|---|---|
| 1 | ai-engineering-from-scratchHands-on AI engineering curriculum with 542 lessons across 20 phases | 89 out of 100 |
| 2 | generative-ai-for-beginners21-lesson course on building generative AI apps in Python and TypeScript | 81 out of 100 |
| 3 | ai-agents-for-beginners18-lesson course on building AI agents with Microsoft Agent Framework | 78 out of 100 |
| #4 | llms-from-scratchCode for the book on building a GPT-style LLM in PyTorch | 77 out of 100 |
| #5 | ml-for-beginners12-week, 26-lesson classic machine learning course using Scikit-learn | 76 out of 100 |