- Notifications
You must be signed in to change notification settings - Fork0
Implementation of neural coherent states, arXiv:2105.15193
wrzadkow/ncs
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
Implementation of neural coherent states, a type of neural-network quantumstates introduced in the following preprint:
Artificial neural network states for non-additive systems
Wojciech Rzadkowski, Mikhail Lemeshko, Johan H. Mentink
arXiv:2105.15193
This code needsJax andFlax.Python 3.9 is recommended.
Runningpython main.py will perform learning procedure for a small systemwith two bosonic modes. Energies at each optimization step will be written tooutput.txt. Forsimplicity, adjusting both the physics and algorithm parameters is donedirectlyin the main file by editingphysics_pars andarg_pars variables.
The code runs on GPU without change. Consultthis material for running on TPU.
The tests can be run withpython tests.py. No errors indicate tests passing,whileAssertionErrors appearing correspond to their failure.