LUCI is a general-purpose emission-line fitting pipeline for SITELLE IFU data cubes and one-dimensional spectra.
You can find the RNAAS article here: Arxiv Paper: 2108.12428
LUCI is a general purpose fitting pipeline built specifically with SITELLE IFU
data cubes in mind; however, if you need to fit any emission line spectra, LUCI
will be able to help! Although other codes exist, we built LUCI specifically with the user
in mind. Thanks to the clear coding practices used to create LUCI and her detailed documentation,
users can easily modify the code by changing numerical solvers. Additionally, we implemented
a Bayesian Inference algorithm using MCMC (using emcee) to derive uncertainty estimates.
LUCI requires Python 3.10 or newer. Install the published package with pip:
python -m pip install luci-sitelleThe distribution is named luci-sitelle; the Python package you import is
luci. The legacy machine-learning training and component-counting tools are
available as an optional extra on Python 3.10:
python -m pip install "luci-sitelle[ml-legacy]"To contribute or run the examples from a checkout, use uv and the committed lockfile:
git clone https://github.com/crhea93/LUCI.git
cd LUCI
uv sync
uv run pytest -m "not slow"LUCI is now a proper installed package, so no sys.path juggling is required. Load it with:
from luci import SitelleCubeThe older spellings all still work — from LuciBase import Luci, from LUCI.LuciFit import Fit,
and so on — so existing notebooks and scripts run unchanged. See the
migration guide for the full old-to-new mapping, and for the list of
bug fixes that changed numerical results.
(Older documentation and notebooks may still show sys.path.insert(0, '/the/path/to/LUCI/') before
the import. That line is no longer needed and can be deleted.)
You can quickly test that everything is working for your system by entering the LUCI directory and running pytest. You may receive some warnings, but everything should pass! If not, please let me know :)
LUCI is - of course - first and foremost, a fitting algorithm. In order to ease the pain of fitting IFU data cubes, we have built several wrappers for the fitting functions (these can be found in Luci/LuciFit.py). These functionalities include, but are not limited to: reading in a data cube in the HDF5 format (LUCI()), fitting the entire data cube (LUCI.fit_entire_cube), fitting a region or masked region of the cube (LUCI.fit_region()), extracting and fitting an integrated region (LUCI.fit_spectrum_region), creating a deep image (LUCI.create_deep_image), and building a Signal-to-Noise ratio map of the cube (LUCI.create_snr_map).
You can also access the fitting directly by circumventing the LUCI wrappers and directly accessing the fitting algorithms (Luci.LuciFitting). With this, you can fit any spectrum using the Luci.LuciFitting.fit and Luci.LuciFitting.bayes_fit algorithms. We also provide basic plotting functionality which can be found in Luci.LuciPlotting.
Examples are paramount to the success of any open source code. Therefore, we have tried to make our examples as complete as possible. That said, we surely have forgotten something! If you wish to see an example that does not exist (or for an example to be better explained), please shoot us an email or open up an issue!
All examples can be found in two locations. Read-through examples can be found in our online documentation while jupyter notebooks can be found in the Examples folder. I suggest starting with https://crhea93.github.io/LUCI/example_basic_lite.html.
Documentation can also be found on the LUCI documentation site. In addition to documentation on each function in LUCI, you can also find a description of what LUCI calculates and how she does what she does!
If you wish to contribute, that's awesome! Please shoot me an email at carter.rhea@umontreal.ca. The easiest way to get involved is to make an issue or fork the repo, make your changes, and submit a well-documented pull request.
If you have any questions about how to install, use, or modify LUCI, please send an email to Carter Rhea.
Copyright (c) 2021-2026 Carter Rhea and LUCI contributors.
LUCI is distributed under the MIT License.
If you cite LUCI, please use both the following citations.
Software Citation: Carter Lee Rhea, Laurie Rousseau-Nepton, Jessie Covington, Leo Alcorn, Benjamin Vigneron, Julie Hlavacek-Larrondo, & Louis-Simon Guité. (2021). crhea93/LUCI: Luci Updates (v1.1). Zenodo. https://doi.org/10.5281/zenodo.5730149
Paper Citation: Carter Rhea et al 2021 Res. Notes AAS 5 208
If you use the mixture density network (MDN) implementation, please include the following citation: Carter Rhea et al 2021 Res. Notes AAS 12 276