This is a project which is currently making use of HPC facilities at Newcastle University. It is active.
For further information about this project, please contact:
We will train machine learning models to ingest dark matter properties of halos in hydrodynamic simulations to make predictions on the baryonic properties of the galaxies which sit in those halos. We will apply these models to dark matter only simulations and verify that the results of these agree with the hydrodynamic simulations and with observations.
We use the Extremely Randomized Trees algorithm in Python using the scikit-learn package. The input and output data for our models is processed using the Astropy and pandas Python packages. We also do hyperparameter optimization using a Gaussian Processes Emulator from the SciPy Python package. Furthermore, we compute cosmological two-point correlation functions using the Landy-Szalay estimator implemented in the treecorr Python package