TabAbax
Copyright 2026 TabAbax

This file contains attribution and license notices for third-party
components included in this distribution. It must be retained with
redistributions where required by the applicable licenses.

TabAbax-original packaging is Apache-2.0. See LICENSES/Apache-2.0.txt.
The checkpoint weights are not in this source distribution. They are
CC-BY-NC-4.0 and are not licensed for commercial use.

TabICL-derived portions
-----------------------

The inference implementation is derived from TabICL v2:

https://github.com/soda-inria/tabicl

Copyright (c) 2025, Soda team @ Inria

Those portions remain under the BSD 3-Clause License.
See LICENSES/BSD-3-Clause.txt.

The comparison tag is v2.2.0, commit 0dbff3ec8fc68c123c87af77b0ea8b25cd2d23f3.
That commit is the version this distribution was compared against. It is not a
record of the commit that was originally copied.
See PROVENANCE.md for which files match that version and which are modified.
`_model/inference.py`, `_model/embedding.py`, `_model/interaction.py`,
`_model/model.py`, and the listed `_sklearn` modules are modified derivatives,
not unchanged copies.

TabICL-derived files:

- src/tababax/_model/
- src/tababax/_sklearn/base.py
- src/tababax/_sklearn/classifier.py
- src/tababax/_sklearn/regressor.py
- src/tababax/_sklearn/preprocessing.py
- src/tababax/_sklearn/sklearn_utils.py
- src/tababax/_torch_devices.py

preprocessing.py and sklearn_utils.py also contain the additional
third-party portions listed below. Those portions are not TabICL's
own code.

Additional third-party portions
--------------------------------

Rotary positional embedding code in
src/tababax/_model/rope.py is based on:

https://github.com/lucidrains/rotary-embedding-torch

Copyright (c) 2021 Phil Wang

Licensed under the MIT License.
See LICENSES/MIT.txt.

src/tababax/_sklearn/sklearn_utils.py contains code adapted from
scikit-learn:

https://github.com/scikit-learn/scikit-learn/blob/1eb422d6c5/sklearn/utils/validation.py

Copyright (c) 2007-2024 The scikit-learn developers.
All rights reserved.

The adapted portions are distributed under the BSD 3-Clause License.
The complete scikit-learn license, copyright line included, is reproduced in
LICENSES/scikit-learn-BSD-3-Clause.txt and ships with the package.

Parts of src/tababax/_sklearn/preprocessing.py are adapted from:

TabR:
https://github.com/yandex-research/tabular-dl-tabr/blob/75105013189c76bc4f247633c2fb856bc948e579/lib/data.py

Copyright (c) 2023 Authors of
"TabR: Unlocking the Power of Retrieval-Augmented Tabular Deep Learning"

Licensed under the MIT License.

PyTabKit:
https://github.com/dholzmueller/pytabkit/blob/949bf81e3964f65a33dd2c252c3713c239c17b2d/pytabkit/models/utils.py

Licensed under the Apache License, Version 2.0.
See LICENSES/Apache-2.0.txt.

TabICL already carried these secondary attributions. They are repeated
here because this distribution still contains the code.
