Skip to content

About

Tutorial on Prior-data Fitted Networks for the Neural Network Architectures course at AGH University

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Repository files navigation

Prior-data Fitted Networks (PFN) Tutorial

Open in Colab uv PyTorch License: MIT

An intuitive, hands-on tutorial and PyTorch implementation of Prior-data Fitted Networks (PFNs) — the breakthrough meta-learning framework powering foundational tabular models like TabPFN.

Getting Started

Ensure you have uv installed.

1. Clone & Sync Environment

git clone https://github.com/barhanc/pfn-tutorial.git
cd pfn-tutorial

# Synchronize the virtual environment with uv.lock in seconds
uv sync

2. Launch the Tutorial

Run JupyterLab or Jupyter Notebook directly via uv:

uv run jupyter lab
# or
uv run jupyter notebook

Alternatively, open the notebook in one click using Google Colab.

References

  • Prior-data Fitted Networks: Müller et al., "Transformers Can Do Bayesian Inference", ICLR 2022. [arXiv:2112.10510]
  • TabPFN: Hollmann et al., "TabPFN: A Transformer That Solves Small Tabular Classification Problems in a Second", Nature 2025 / ICLR 2023. [arXiv:2207.01848]

About

Tutorial on Prior-data Fitted Networks for the Neural Network Architectures course at AGH University

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages