This repository contains the associated code for 'Machine learning the deuteron' and focuses on solving the ground-state wavefunction of the deuteron with a neural-network quantum state (NQS) in momentum space.
You can reproduce the results of the paper by cloning the repository with,
git clone https://github.com/jwtkeeble/machine-learning-the-deuteron
and running the run.py script as shown below in the Usage section.
The requirements in order to run this script can be found in requirements.txt and can be installed via pip or conda.
The arguments for the run.py script are as follows:
| Argument | Type | Default | Description |
|---|---|---|---|
-H/--hidden_nodes |
int |
10 | Number of hidden neurons in the model |
--preepochs |
int |
10000 | Number of pre-epochs for the pretraining phase |
--epochs |
int |
250000 | Number of epochs for the energy minimisation phase |
The license of this repositry is Apache License 2.0.