if-quantum / pairwise-tomography

Efficient pairwise tomography for qiskit circuits

pairwise_tomography

Efficient pairwise tomography for qiskit circuits.

Guillermo García-Pérez, Matteo A. C. Rossi, Boris Sokolov, Elsi-Mari Borrelli, Sabrina Maniscalco, Phys. Rev. Research 2, 023393 (2020)

Installation

Clone the repository

git clone https://github.com/if-quantum/pairwise-tomography.git

Install with

pip install ./pairwise-tomography

Usage

Check the tutorial notebook pairwise-tomography-tutorial.ipynb

Citation

If you find this package useful, please cite the article

Guillermo García-Pérez, Matteo A. C. Rossi, Boris Sokolov, Elsi-Mari Borrelli, Sabrina Maniscalco, Phys. Rev. Research 2, 023393 (2020)

Bibtex entry:

@article{garcia-perez2020pairwise,
  title = {Pairwise tomography networks for many-body quantum systems},
  author = {Garc\'{\i}a-P\'erez, Guillermo and Rossi, Matteo A. C. and Sokolov, Boris and Borrelli, Elsi-Mari and Maniscalco, Sabrina},
  journal = {Phys. Rev. Research},
  volume = {2},
  issue = {2},
  pages = {023393},
  numpages = {9},
  year = {2020},
  month = {Jun},
  publisher = {American Physical Society},
  doi = {10.1103/PhysRevResearch.2.023393},
  url = {https://link.aps.org/doi/10.1103/PhysRevResearch.2.023393}
}

pairwise-tomography's Issues

Tutorial notebook leads to error

Running the tutorial notebook yields

QiskitError: 'register size error'

'pairs_list' kwarg doesn't work

The fit method of the fitter is documented as having a kwarg pairs_list

pairs_list (list): A list of tuples containing the indices of the qubit pairs for which to perform tomography

However this doesn't seem to work correctly. For example, running

qc = QuantumCircuit(3)

tomo_circs = pairwise_state_tomography_circuits(qc, qc.qregs[0])
tomo_results = execute(tomo_circs,Aer.get_backend('qasm_simulator')).result()
fitter = PairwiseStateTomographyFitter(tomo_results, tomo_circs, qc.qregs[0])

rho = fitter.fit(pairs_list=[(0,1)])

gives the error

TypeError: lstsq_fit() got an unexpected keyword argument 'pairs_list'

Idea for alternative output

Would it be easy to give output in terms of expectation values instead of density matrices. So, for example, if you wanted the two-qubit Pauli expectation values for qubits 0 and 1 you could use something like

print(fit_result[(0,1)],output='expectation')

and then get a result that looks something like

{'XI':0.0,'YI':0.0,'ZI':0.0,'XX':1.0,'YY':-1.0,'ZZ':1.0,...