LISO
Author: Jun Zhu, zhujun981661@gmail.com
LISO (LInac Simulation and Optimization) is a library which provides a unified interface for numerical simulations, experiments and data management in the era of big data and artificial intelligence. LISO was initially created only for simulation and optimization back in 2017. After a more than two years interruption, it is live again but aims at:
- providing a high-level API to run a large numbers of beam dynamics simulations using a combination of different codes;
- providing a unified IO for the simulated and experimental data;
- providing an interface for deep learning and deep reinforcement learning studies on accelerator physics.
Documentation
The full documentation can be found at
Getting started
Installation
$ pip install lisoUse LISO in your experiments
Parameter scan
from liso import EuXFELInterface, MachineScan
from liso import doocs_channels as dc
m = EuXFELInterface()
m.add_control_channel('XFEL.RF/LLRF.CONTROLLER/VS.GUN.I1/PHASE.SAMPLE', dc.FLOAT,
write_address='XFEL.RF/LLRF.CONTROLLER/CTRL.GUN.I1/SP.PHASE')
m.add_control_channel('XFEL.RF/LLRF.CONTROLLER/VS.A1.I1/PHASE.SAMPLE', dc.FLOAT,
write_address='XFEL.RF/LLRF.CONTROLLER/CTRL.A1.I1/SP.PHASE')
m.add_diagnostic_channel('XFEL.DIAG/CAMERA/OTRC.64.I1D/IMAGE_EXT_ZMQ', dc.ARRAY,
shape=(1750, 2330), dtype='uint16')
sc = MachineScan(m)
sc.add_param('XFEL.RF/LLRF.CONTROLLER/VS.GUN.I1/PHASE.SAMPLE', lb=-3, ub=3)
sc.add_param('XFEL.RF/LLRF.CONTROLLER/VS.A1.I1/PHASE.SAMPLE', lb=-3, ub=3)
sc.scan(4000, n_tasks=8)Reading experimental data from files
from liso import open_run
run = open_run('r0001')
# Get an overview of the run
run.info()
# Loop over a run
for pid, data in run:
pass
# Access a single macropulse
# - by id
pid, data = run.from_id(123456789)
# - by index
pid, data = run.from_index(0)
# Access a channel data
ch_data = run.channel('XFEL.RF/LLRF.CONTROLLER/VS.GUN.I1/PHASE.SAMPLE')
# and convert it to a numpy array
ch_data_array = ch_data.numpy()Use LISO to run simulations
Building a linac
from liso import Linac
linac = Linac(2000)
linac.add_beamline('astra',
name='injector',
swd='../astra_files',
fin='injector.in',
template='injector.in.000',
pout='injector.0450.001')Running a parameter scan or a jitter study
from liso import LinacScan
sc = LinacScan(linac)
sc.add_param('gun_gradient', start=120, stop=140, num=10, sigma=-0.001)
sc.add_param('gun_phase', start=-10, stop=10, num=20, sigma=0.1)
sc.scan()Reading simulated data from files
from liso import open_sim
sim = open_sim('./')
# Get an overview of the simulation
sim.info()
# Loop over the simulated data
for sid, data in sim:
pass
# Access a single simulation
# - by id
sid, data = sim.from_id(1)
# - by index
sid, data = sim.from_index(0)
# Access a phasespace data
ch_data = sim.channel('injector/out', ['x', 'px'])
# and convert it to a numpy array
ch_data_array = ch_data.numpy()Cite LISO
A BibTeX entry that you can use to cite it in a publication:
@misc{liso,
Author = {J. Zhu},
Title = {LISO},
Year = {2020},
publisher = {GitHub},
journal = {https://github.com/zhujun98/liso},
version = {...}
}
