Applying multimodal learning to Classify transient Detections Early

AppleCiDEr (arXiv) is a multimodal transient classifier that uses photometry, metadata, images, and spectra. Name inspired by University of Minnesota's famous apple program.

AppleCiDEr in ZTF production

---

Repository layout

applecider/
├── src/applecider/
│   ├── datasets/              # dataset classes and sampling helpers
│   ├── models/                # AppleCiDEr and baseline model implementations
│   └── preprocessing_utils/   # multimodal preprocessing pipeline utilities
├── scripts/
│   └── fusion_preprocessing.py
├── tests/
│   └── applecider/
├── docs/
│   ├── conf.py
│   ├── index.rst
│   ├── notebooks/
│   ├── pre_executed/
│   └── static/
├── _archive/                  # legacy code/notebooks kept for reference
├── pyproject.toml
└── .pre-commit-config.yaml

Developer quickstart

Create and activate a conda environment:

conda create --name applecider python=3.12
conda activate applecider

Clone and enter the repository:

git clone https://github.com/skyportal/applecider.git
cd applecider

Install optional compiled dependencies used by parts of the preprocessing stack:

conda install -c conda-forge llvmlite numba

Install in editable mode with development dependencies:

python -m pip install -e '.[dev]'

Set up pre-commit hooks:

pre-commit install

Run all checks locally (recommended before push):

pre-commit run --all-files

Run tests directly:

python -m pytest --cov=./src --cov-report=html

Build docs directly:

sphinx-build -T -E -b html -d ./docs/_build/doctrees ./docs ./_readthedocs

Citation

@article{junell2025AppleCiDEr,
      title={Applying multimodal learning to Classify transient Detections Early (AppleCiDEr) I: Data set, methods, and infrastructure}, 
      author={Alexandra Junell and Argyro Sasli and Felipe Fontinele Nunes and Maojie Xu and Benny Border and Nabeel Rehemtulla and Mariia Rizhko and Yu-Jing Qin and Theophile Jegou Du Laz and Antoine Le Calloch and Sushant Sharma Chaudhary and Shaowei Wu and Jesper Sollerman and Niharika Sravan and Steven L. Groom and David Hale and Mansi M. Kasliwal and Josiah Purdum and Avery Wold and Matthew J. Graham and Michael W. Coughlin},
      year={2025},
      eprint={2507.16088},
      archivePrefix={arXiv},
      primaryClass={astro-ph.IM},
      url={https://arxiv.org/abs/2507.16088}, 
}