# AstroPT Euclid VIS Model

Pre-trained AstroPT model for single-band analysis using Euclid VIS imaging.

## Overview

This is a pre-trained checkpoint for the **AstroPT** framework, trained on Euclid VIS band imaging from the Euclid Q1 dataset. 

**Citation**: Euclid Collaboration: Siudek, M et al. 2025 ([arXiv:2503.15312](https://ui.adsabs.harvard.edu/abs/2025arXiv250315312E/abstract))

## Quick Start

### Load Model

```python
import torch
from pathlib import Path

# Load model checkpoint
model_path = "astropt/090M/ckpt.pt"
device = "cuda" if torch.cuda.is_available() else "cpu"

# Load state dict
checkpoint = torch.load(model_path, map_location=device)
# Initialize your model architecture here
# model.load_state_dict(checkpoint)
```

### Inference

```python
from datasets import load_dataset
import torch

# Load dataset
dataset = load_dataset(
    "msiudek/astroPT_euclid_dataset",
    split="train_batch_1",
    streaming=True
)

# Run inference
model.eval()
with torch.no_grad():
    for sample in dataset:
        vis_image = sample['VIS_image']  # 224×224
        vis_image = torch.tensor(vis_image, dtype=torch.float32)
        vis_image = vis_image.unsqueeze(0).unsqueeze(0)  # [1, 1, 224, 224]
        
        # Get embeddings
        embeddings = model(vis_image)
```

## Training Data

- **Dataset**: [AstroPT Euclid Dataset](https://huggingface.co/datasets/msiudek/astroPT_euclid_dataset)
- **VIS Band**: Euclid VIS (0.55–0.90 μm)

## Related Models

- [AstroPT VIS+NISP Model](https://huggingface.co/msiudek/astroPT_euclid_VIS_NISP_model): Multi-band (VIS + NISP)
- [AstroPT VIS+NISP+SED Model](https://huggingface.co/msiudek/astroPT_euclid_VIS_NISP_SED_model): Multi-modal (imaging + photometry)

## Datasets

- **Imaging**: [astroPT_euclid_dataset](https://huggingface.co/datasets/msiudek/astroPT_euclid_dataset)
- **Metadata**: [astroPT_euclid_metadata](https://huggingface.co/datasets/msiudek/astroPT_euclid_metadata)

## Code & Documentation

For inference code, training scripts, and tutorials, visit the **[AstroPT GitHub Repository](https://github.com/Smith42/astroPT)**.

## Citation

```bibtex
@article{Siudek2025,
  title={AstroPT: Astronomical Physics Transformers for Multi-modal Learning},
  author={Siudek, M and others},
  journal={Euclid Collaboration},
  eprint={2503.15312},
  archivePrefix={arXiv},
  year={2025},
  url={https://ui.adsabs.harvard.edu/abs/2025arXiv250315312E/abstract}
}
```

## License

CC-BY-4.0

---

**Last Updated**: December 2025  
**Model Version**: 1.0