ML-Poppyns is a new population synthesis framework for Galactic neutron stars that employs deep learning and simulation based inference (SBI) to investigate the birth, evolution and emission properties of the neutron star population in our Galaxy. It starts from some initial conditions on the birth positions in the Galaxy, kick velocities, birth spin-periods and magnetic field distributions. It then performs dynamical and magneto-rotational evolution to generate an evolved population of neutron star. Finally it models the radio emission and detection with radio surveys. To perform parameter inference it employs the power of neural networks, in particular a neural density estimator is trained on simulated data to learn a mapping between model parameters and simulated observables and reconstruct an approximation of the posterior distribution of the model parameters. Once trained the neural density estimator can be applied to the true observed neutron star population to infer its physical properties.
Download
https://github.com/ice-csic-astroexotic/ML-Poppyns-Open.git
Documentation
A guide for the installation can be found in the README file together with the instructions to compile the full documentation including a quick-start guide.Contact
Should you encounter any issues or have questions, please feel free to email us at ml-poppyns@ice.csis.es.If you use ML-Poppyns in your research, we kindly ask you to:
-
Cite the following publications in the text:
- Analyzing the Galactic Pulsar Distribution with Machine Learning, Ronchi et al. 2021
- Isolated Pulsar Population Synthesis with Simulation-based Inference, Graber et al. 2024
- Radio pulsar population synthesis with consistent flux measurements using simulation-based inference, Pardo-Araujo et al. 2025
-
Add the following sentence in the acknowledgment section of your publication:
"ML-Poppyns has been funded by the European Research Council via the ERC Consolidator grant 'MAGNESIA' (No. 817661; PI: N. Rea)."