Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -1,3 +1,30 @@
|
|
| 1 |
-
|
| 2 |
-
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# TabDPT: Scaling Tabular Foundation Models
|
| 2 |
+
|
| 3 |
+
## Installation
|
| 4 |
+
```
|
| 5 |
+
git clone [email protected]:layer6ai-labs/TabDPT.git
|
| 6 |
+
cd TabDPT
|
| 7 |
+
pip install -e .
|
| 8 |
+
```
|
| 9 |
+
|
| 10 |
+
## Example Usage
|
| 11 |
+
Please refer to the GitHub repo.
|
| 12 |
+
For better performance, please increase `context_size` or increase `n_ensembles` to trade off speed and accuracy
|
| 13 |
+
|
| 14 |
+
## Updates
|
| 15 |
+
|
| 16 |
+
### Update April 2025: New Model
|
| 17 |
+
**Version 1.1 is now available.** We have improved the prediction performance of TabDPT through increased training stability.
|
| 18 |
+
|
| 19 |
+
### Update December 2024: Faster Inference
|
| 20 |
+
Added support for flash attention (with bf16 precision) and compile flag. Both are enabled to True by default and should lead to a significant speed-up.
|
| 21 |
+
|
| 22 |
+
## Citation
|
| 23 |
+
```
|
| 24 |
+
@article{ma2024tabdpt,
|
| 25 |
+
title={TabDPT: Scaling Tabular Foundation Models},
|
| 26 |
+
author={Ma, Junwei and Thomas, Valentin and Hosseinzadeh, Rasa and Kamkari, Hamidreza and Labach, Alex and Cresswell, Jesse C and Golestan, Keyvan and Yu, Guangwei and Volkovs, Maksims and Caterini, Anthony L},
|
| 27 |
+
journal={arXiv preprint arXiv:2410.18164},
|
| 28 |
+
year={2024}
|
| 29 |
+
}
|
| 30 |
+
```
|