usnistgov/state_meas_tomo

Qubit state tomography with unknown measurement operators

state_meas_tomo

Joint Quantum State and Measurement Tomography

This software package performs joint quantum state and measurement tomography as described in arXiv:1803.08245. Included are three example scripts that simulate data for one or two trapped ion systems with either symmetric or asymmetric measurements:

• analysis_scripts/paper_simulations.py: produces all data and histograms shown in arXiv:1803.08245 with seed = 0. Also provides an example of symmetric measurements.

• analysis_scripts/asym_simulations.py: produces simulated data from asymmetric measruements by similar methods as in previous script.

• analysis_scripts/load_tutorial/load_simulations.py: gives an example of loading data with asymmetric measurements.

Features

• Produces point estimates of unknown quantum states and POVMs
• Bounds expectation values of states even when POVMs are not informationally complete
• Performs bootstrap resampling to establish confidence intervals
• Produces histogram plots and data shown in arXiv:1803.08245

Installation

You may obtain the software from https://github.com/usnistgov/state_meas_tomo.

Once the software is downloaded, update the following lines in each script so that installdir1 is the directory where the software is saved:

sys.path.append('installdir1/state_meas_tomo/')
sys.path.append('installdir1/state_meas_tomo/analysis_scripts')
sys.path.append('installdir1/state_meas_tomo/partial_tomography')

The software was written for Python 3.4. To create a Python 3.4 environment, we recommend using Anaconda. Once installed, an environment can be created by running the following in the terminal:

For Mac/Linux:

$ conda create —-name py34 python=3.4
$ source activate

For Windows:

> conda create --name py34 python=3.4 matplotlib numpy scipy".
> activate py34

For more information and instructions see Conda's User Guide

All three scripts mentioned above can be run from the terminal:
python xxxxx_simulations.py
from a Python Shell, or a developement environment like Spyder (included with Anaconda).

Without expectation value bounding

No further installation is required. (In the current version of each script, the expecation value bounding is commented out.)

With expectation value bounding

In order to use the expectation-value bounding features (described in Sec. V of arXiv:1803.08245) you additionally must have MATLAB installed and the included python engine extracted, see Mathworks's instructions.

To extract the engine, from within the py34 environment, run:

$ cd "matlabroot\extern\engines\python"
$ python setup.py install —prefix="installdir2"

where "matlabroot" is the location of the MATLAB installation (use function matlabroot in MATLAB), and "installdir2" is the directory you've specified to to install the engine.

Depending on your operating system permissions, you may need to build or install the MATLAB python engine in non-default locations. For instructions see these instructions.

You must also add installdir2 to the python path in each script by updating the following lines:

sys.path.append('installdir2/lib/python3.4/site-packages/')
sys.path.append('installdir2/lib/python3.4/site-packages/matlab')

The software also requires YALMIP and SeDuMi, which must be installed in MATLAB. Download and follow installation instructions from YALMIP and SeDuMi.

To run all three sample scripts with the expectation value bounding, first uncomment line 27 in analysis.py:
# import matlab.engine # uncomment for expectation-value bounding
and then uncomment the bottom lines of the chosen script.

Outputs

Each PartialMaxLike instance (contained in analysis.py module) that uses the .tomography or .bootstrapAnalysis methods creates a "autosave_name.hist" file, where name is the attribuite given for the instance (defaults to the date), which contains the PartialMaxLike instance with all the histograms, analysis parameters, estimites, and bootstrap log (if created).

The expectationValues function in analysis_tools.py creates a text file called "results-name.txt" output, which contains the expectation value lower and upper bounds for each specified observable as well as the confidence intervals and medians from the bootstrap resamples. It also creates .out file with the list of lower and upper bounds for each bootstrap resample.

The remaining functions in analysis_tools.py create figure 2 and 5 in arXiv:1803.08245 and other figures that may be useful.

Support

If you are having issues, please contact Scott Glancy.

If you use this software, please cite our paper:
Adam C. Keith, Charles H. Baldwin, Scott Glancy, E. Knill "Joint Quantum State and Measurement Tomography with Incomplete Measurements" arXiv:1803.08245.

Last updated 2018-Mar-24.