Spectuner is a tool for automated spectral line analysis of instellar molecules.
The code requires Python>=3.10. If you do not have Python installed, we recommend installing Anaconda. Then, we can install the code from the repository.
pip install spectuner
If you want to use the AI module, you need to install PyTorch.
pip install torch
Also, please download the neural network weights file from Hugging Face.
In addition, the code requires the Cologne Database for Molecular Spectroscopy (CDMS) as input. You may download the database using the following command.
wget https://cdms.astro.uni-koeln.de/static/cdms/download/official/cdms_sqlite__official-version__2024-01-01.db.gz
Read the docs at this link.
If you find this code useful in your research, please cite our work in the acknowledgment section using the following BibTeX entry:
@ARTICLE{2025ApJS..277...21Q,
author = {{Qiu}, Yisheng and {Zhang}, Tianwei and {M{\"o}ller}, Thomas and {Jiang}, Xue-Jian and {Song}, Zihao and {Chen}, Huaxi and {Quan}, Donghui},
title = "{Spectuner: A Framework for Automated Line Identification of Interstellar Molecules}",
journal = {\apjs},
keywords = {Spectral line identification, Interstellar medium, 2073, 847, Astrophysics - Astrophysics of Galaxies, Astrophysics - Instrumentation and Methods for Astrophysics, Astrophysics - Solar and Stellar Astrophysics},
year = 2025,
month = mar,
volume = {277},
number = {1},
eid = {21},
pages = {21},
doi = {10.3847/1538-4365/adaeba},
archivePrefix = {arXiv},
eprint = {2408.06004},
primaryClass = {astro-ph.GA},
adsurl = {https://ui.adsabs.harvard.edu/abs/2025ApJS..277...21Q},
adsnote = {Provided by the SAO/NASA Astrophysics Data System}
}