CIANNA is a general-purpose deep learning framework primarily developed and used for astronomical data analysis. Functionalities and optimizations are added based on relevance for astrophysical problem-solving. CIANNA can be used to build and train large neural network models for various tasks and is provided with a high-level Python interface (similar to keras, pytorch, etc.). One of the specificities of CIANNA is its custom implementation of a YOLO-inspired object detector used in the context of galaxy detection in 2D or 3D radio-astronomical data products. The framework is fully GPU-accelerated through low-level CUDA programming.
Development team
David Cornu - creator and lead dev, post-doc researcher, AI Fellow PR[AI]RIE, FR - LUX / Observatoire de Paris, PSL
Gregory Sainton - dev, AI Research engineer, FR - LUX / Observatoire de Paris
Aristide Doussot - dev, HPC Research engineer, FR - LUX / Observatoire de Paris
Preferred contact point: david.cornu@observatoiredeparis.psl.eu
See Copyright © and License terms at the end.
Quick access:
CIANNA status upgrade (06/2025)
The first release (V-1.0), published in July 2024 (see the release page), marked a major milestone in CIANNA development. Since then, there have been mostly minor updates to this repo, but new developments are still ongoing on a private experimental version. We have chosen to maintain this repository as a stable reference version for the moment, as several publications based on V-1.0 are currently in preparation. We also wanted to give the community time to learn about CIANNA and explore the capabilities of this version. Coming back to the experimental version, in the past year, we noticeably added basic support for skip/residual connection layers, slightly reworked the normalization layers, and explored more advanced gradient optimizers (Adam and RMSprop). These changes enabled us to train much deeper architectures and to build Denoising Diffusion Probabilistic Models (see an example of galaxy image generation below), which we plan to utilize in future work. Our focus is now on enabling the construction of less linear architectures, allowing for multi-modality and the development of more complex structural pathways. These latest changes necessitate a rework of certain aspects of the framework and require extensive testing due to the increasing number of possible structures, which means that the next major update will take some time to arrive. Still, early access to these functionalities for research projects can be discussed on a case-by-case basis (see contact above).
Python scripts and Google-Colab-compatible notebooks are available under the examples directory for most of the following examples.
| Description - Dataset | Visualization | Animation or real time |
|---|---|---|
| *** | Classical computer vision examples |
*** |
| Image classification MNIST Top-1 accuracy ~99.3% Net. ~LeNet-5 630000 ips @28p*  |
||
| Image classification Imagenet - 1000 classes Top-1 acc ~74.7% Top-5 acc ~91.7% Net. ~Darknet19 740 ips @448p* |
||
| Object detection COCO - 1000 classes mAP@50 ~40.1% COCO-mAP ~21.9% Net. ~Darknet19 690 ips @416p* |
Real-time on a laptop GPU |
|
| *** | Astronomical dataset examples |
*** |
| Source detection SKA SDC1 2D continuum 560MHz - 1000h score 479372 pts Net. 17 conv. layers 500 ips @512p*  |
Not real-time here, only animated |
|
| Profile regression 3D Galactic extinction mapping Net. [C5x5.12-P2-{D3072}x2-D2048-D128] 120000 ips @64p*  |
Face-on view of the galactic plane in a 45° "cone" toward the Carina arm (derived from the 3D map) |
Per LOS prediction examples Integrated extinction skyview |
| Fake galaxy generation Based on galaxy zoo 2 Cascaded DDPM Ensemble of U-Nets ~40M param. A few ips @192p* Made with the dev-exp branch of CIANNA, unavailable ATM |
Cascading pipeline with 3 DDPM models |
Generated examples |
*Images (or Inputs) per second (ips) are given for an RTX 4090 GPU in inference using FP16C_FP32A mixed-precision at the specified resolution and with maximum batch size to saturate performances.
Please take a look at the system requirements and the installation instructions wiki pages.
=> A complete step-by-step installation guide of CIANNA and its dependencies from a fresh Ubuntu 20.04 is accessible here.
Please read the How to use Wiki page for a minimalistic tour of CIANNA capabilities on a simple example script and dataset.
A full description of all the Python interface functions is available as an API documentation page on the Wiki.
Please also consider consulting the Step-by-step installation guide to verify everything was installed correctly.
Several Python scripts and notebooks are provided as examples for different datasets and applications.
List of known publications that make use or directly refer to the CIANNA framework.
When referring to a specific functionality or application, feel free to cite the relevant publication. In all cases, if your work makes use of any version of CIANNA, please cite the non-version-specific DOI from Zenodo 10.5281/zenodo.12806324.
###########################################################################
These files are Copyright © 2024 David Cornu, but released under the Apache2 License.
While you are free to duplicate and modify this repository under the Apache2 License above, by being allowed to submit a contribution to this repository, you agree to the following terms:
-
You grant to the present CIANNA framework (and its Author) your copyright license to reproduce and distribute your contributions and such derivative works.
-
To the fullest extent permitted, you agree not to assert all of your "moral rights" in or relating to your contributions to the benefit of the present CIANNA framework.
-
Your contribution was created in whole or in part by you and you have the right to submit it under the open source license indicated in the LICENSE file; or the contribution is based upon previous work that, to the best of your knowledge, is covered under an appropriate open source license and you have the right to submit that work with modifications.