NOTE: this repository is outdated and deprecated. Please refer to mwalmsley/agnfinder for the original version, or maximerobeyns/spitorch for an updated version.
AGNFinder
Detect AGN from photometry in XXL data, as Euclid prep.
This fork explores the use of conditional generative modelling to to estimate galaxy parameters, using PyTorch.
Please see the documentation for more information about this project; in particular, the description of the inference procedure using autoregressive models and CVAEs may give a good overview of this project's goals.
Installation
The installation guide contains detailed information on how to install the project, but for users looking to get started quickly, the following steps should be sufficient.
AGNFinder only supports Python 3.9 onwards, since it makes use of recent language features (as of August, 2021).
To install, run
git clone https://github.com/MaximeRobeyns/agnfinder
cd agnfinder
make install
Please note that:
- the inital repo clone is about 100Mb large. This is slightly regrettable, however it avoids the need to store and maintain binary blobs on some third party server.
- the installation procedure (
make install) further downloads approximately 1.5Gb of additional data.
The installation process only modifies files within the /agnfinder directory, and can be run without root privileges.
To make sure your installation was successful, run
make test
For further help and available commands, run
make help
or consult the documentation.