Python package for benchmarking quantum tomography (QT) methods. The theory behind the software could be found in [1].
To install package download the source files and run
python setup.py install
The following code shows a basic example of running data collection for a 2-qubit tomography method on random pure states.
from pyqtb import Dimension
from pyqtb.analyze import collect
from pyqtb.tests import rps_test
from pyqtb.methods.proto_fmub import proto_fmub
from pyqtb.methods.est_ppi import est_ppi
dim = Dimension([2, 2])
result = collect(dim, proto_fmub(dim), est_ppi(), rps_test(dim))
To get and print a basic report over benchmarks one runs the following commands.
from pyqtb.analyze import report, as_table
r = report(result)
print(as_table(report))
One can also compare different methods.
from pyqtb.analyze import compare, as_table
r = report([result1, result2])
print(as_table(report))
Module pyqtb.analyze
also contains plot functions.
All code found in this repository is licensed under GPL v3
[1] Bantysh B.I. et al. Quantum tomography benchmarking; arXiv:2012.15656