Qiskit / qiskit-addon-utils / 29438851097

97%

push

github

Add support for postselection using bit-flip checks (#174)

* necessary changes to post_selection passes to handle circuits with preselection circuitry

* add constants, stuff to init

* add pre_selection module to noise management: mirrors post_selection in many ways, made separate to avoid breaking changes to post_selection module as much as possible

* removed commented out code block

* added documentation + tests; all tests pass except test_if_else, which is currently skipped. Demo file still runs successfully

* docs

* pre-selection in expectation_values function

* handle if successor has no op field

* Code style

* remove pre_selection_mask from executor_expectation_values

* style

* Put pre/post selection in same post_selection sub-package (#175)

* Put pre/post selection in same post_selection sub-package

* Update module docstring for clarity

* Correct capitalization in module docstring

* Restore expval changes

* Support pre-selection in PostSelSummary module

* Refine docstring for PostSelectionSummary class

Removed redundant explanation about post-selection techniques.

* coverage

* Combine pre/post selectors into only the PostSelector

* docs fix

* bugfixing

* everything working except pre-->post with both edge-based

* test scripts

* bugfix

* Update tests to focus more on interplay between init checks and post-measurement checks

* lint

* Fixes

* Clean up selector and summary modules

* Pair spec msmts w terminal measurements using barrier

* reno

* Ensure backward compatibility

* Bugfix

* Clean up dead code, handle mismatches in circuit size and coupling map size

* Rename classes. Restore legacy post_selection path. Move new implementation to bitflip_checks sub-package.

* naming conventions

* bitflip to bit_flip throughout

* style

* presel spectators can be added before or after

* varname

* Add guide

* Remove transpiler sub-pkg, clean up guide, suppress docs warning from qiskit

* Ignore bitflip check guide

* ... (continued)

767 of 769 new or added lines in 12 files covered. (99.74%)

1583 of 1640 relevant lines covered (96.52%)

0.97 hits per line