Skip to content

README.md

Latest commit

 

History

History
93 lines (68 loc) · 3.62 KB

README.md

File metadata and controls

93 lines (68 loc) · 3.62 KB
Raw

Teaching materials

Start with the interactive tutorial. It runs directly in a modern browser with no installation, GPU, or remote inference service. The TabICLv2 checkpoint loads into the browser and inference remains local.

This package also contains a self-guided notebook and the source materials for Table as Prompt: An Interactive Guide to Tabular Foundation Models.

Optional local copy

Local serving is only needed to inspect or modify the downloaded website. It requires Python 3:

cd website
./serve.sh 8000

Then open http://localhost:8000/.

Self-guided notebook

Google Colab

Open the primer in Colab and run all cells. A GPU runtime is recommended but not required. The first cell installs the pinned tutorial packages while retaining Colab's managed Python and PyTorch environment. TabICL automatically selects CUDA in Colab and CUDA, XPU, or Apple MPS on supported local hardware, with CPU as the fallback.

Local or CI

On macOS, XGBoost also needs the OpenMP runtime:

brew install libomp

Use Python and OpenMP for the same processor architecture. On Apple Silicon, use native arm64 Python and Homebrew.

Use Python 3.11 in a fresh virtual environment:

python3.11 -m venv .venv
source .venv/bin/activate
python -m pip install --require-hashes -r requirements-lock.txt
jupyter nbconvert notebooks/01_tabicl_primer.ipynb \
  --to html --execute --ExecutePreprocessor.timeout=2400

requirements.txt lists the direct dependencies; requirements-lock.txt fully pins transitive packages and distribution hashes for Python 3.11. This is the exact reproducibility path; Colab is the convenience path. The notebook verifies its immutable Hugging Face revision and checkpoint checksum before model loading. An internet connection is needed for package, dataset, and checkpoint downloads.

Open notebooks/01_tabicl_primer.html to inspect the executed notebook. The locked environment includes nbconvert and the Python kernel. To edit the notebook in JupyterLab, install the optional editor separately:

python -m pip install jupyterlab
jupyter lab notebooks/01_tabicl_primer.ipynb

JupyterLab is not part of the locked execution environment.

Source and attribution

Licensing

Unless otherwise noted, original tutorial materials in this standalone package, including prose, diagrams, notebook narrative, and original software/code, are available under Creative Commons Attribution 4.0 International (CC BY 4.0). Original software/code is also available under the Apache License, Version 2.0; users may choose either license for that original software/code.

Third-party software, model artifacts, datasets, and adapted materials retain their separate terms; this dual-license grant does not relicense them. Affirm names and logos are excluded from these license grants. No trademark rights are granted.

Copyright (c) 2026, Affirm, Inc. See NOTICE and THIRD_PARTY_NOTICES for project and third-party terms, including those for the browser runtime, model artifact, adapted explorer, and datasets.