Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: pytorch
|
| 3 |
+
tags:
|
| 4 |
+
- vision
|
| 5 |
+
- image-classification
|
| 6 |
+
- pytorch
|
| 7 |
+
license: mit
|
| 8 |
+
base_model: mwalmsley/baseline-encoder-regression-maxvit_tiny
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
# BTSbot
|
| 12 |
+
|
| 13 |
+
This is a maxvit fine-tuned for classifying images from the
|
| 14 |
+
Zwicky Transient Facility (ZTF) observatory.
|
| 15 |
+
[Rehemtulla et al. 2024](https://arxiv.org/abs/2401.15167) originally introduced
|
| 16 |
+
`BTSbot` and its classification task, and
|
| 17 |
+
[Rehemtulla et al. 2025](https://arxiv.org/abs/2512.11957) performed
|
| 18 |
+
architecture and pre-training benchmarking on this `BTSbot` image classification task.
|
| 19 |
+
|
| 20 |
+
**Base Model**:
|
| 21 |
+
[mwalmsley/baseline-encoder-regression-maxvit_tiny](https://huggingface.co/mwalmsley/baseline-encoder-regression-maxvit_tiny)
|
| 22 |
+
|
| 23 |
+
## Usage
|
| 24 |
+
|
| 25 |
+
Easily install the btsbot package and load this model with:
|
| 26 |
+
|
| 27 |
+
```python
|
| 28 |
+
git clone https://github.com/nabeelre/BTSbot.git
|
| 29 |
+
cd BTSbot
|
| 30 |
+
pip install -e .
|
| 31 |
+
|
| 32 |
+
import btsbot
|
| 33 |
+
model = btsbot.load_HF_model(
|
| 34 |
+
architecture="maxvit", multi_modal=False, pretrain="galaxyzoo"
|
| 35 |
+
)
|
| 36 |
+
```
|
| 37 |
+
|
| 38 |
+
Also see [`BTSbot/btsbot/inference_example.py`](https://github.com/nabeelre/BTSbot).
|
| 39 |
+
|
| 40 |
+
## Citation
|
| 41 |
+
|
| 42 |
+
If you use this model, please cite:
|
| 43 |
+
|
| 44 |
+
```bibtex
|
| 45 |
+
@ARTICLE{Rehemtulla+2025,
|
| 46 |
+
title={Pre-training vision models for the classification of alerts from wide-field time-domain surveys},
|
| 47 |
+
author={Nabeel Rehemtulla and Adam A. Miller and Mike Walmsley and Ved G. Shah and Theophile Jegou du Laz and Michael W. Coughlin and Argyro Sasli and Joshua Bloom and Christoffer Fremling and Matthew J. Graham and Steven L. Groom and David Hale and Ashish A. Mahabal and Daniel A. Perley and Josiah Purdum and Ben Rusholme and Jesper Sollerman and Mansi M. Kasliwal},
|
| 48 |
+
year={2025},
|
| 49 |
+
eprint={2512.11957},
|
| 50 |
+
archivePrefix={arXiv},
|
| 51 |
+
primaryClass={astro-ph.IM},
|
| 52 |
+
url={https://arxiv.org/abs/2512.11957},
|
| 53 |
+
}
|
| 54 |
+
```
|
| 55 |
+
|
| 56 |
+
## License
|
| 57 |
+
|
| 58 |
+
This model is released under the MIT License.
|
| 59 |
+
|
| 60 |
+
## Repository
|
| 61 |
+
|
| 62 |
+
For more information, see the [BTSbot GitHub repository](https://github.com/nabeelre/BTSbot).
|