Fixed incorrect edge width assignment in qubit selection plot.

Fixed a bug in the bootstrapping functionality of the GST benchmark and updated the respective Jupyter tutorial.

Updated CLOPS value and plot reporting, making explicit offline values, such as time spent in transpilation and in parameter assigning.

Fixed header levels in example_graphstate notebook for correct pages rendering.

Added graph state (bipartite entanglement negativity) benchmark.

Qiskit on IQM dependency updated to > 17.0.

Changed benchmark observation names and identifiers to be more consistent with guidelines.

Added optional configuration parameter (max_circuits_per_batch) to specify the maximum amount of circuits per batch.

Added rustworkx dependency range to fix wrong edge thickness assignment in qubit selection plot.

Added dynamical decoupling parameter option to configurations of all benchmarks.

Added visual aid plot for qubit selection (see, e.g., GHZ example notebook).

Included option to run GST in parallel if the specified qubits don’t overlap

Small runtime improvements in the GST benchmark.

Changed tensor order in GST outputs from Qiskit (bottom to top) to standard (top to bottom) order.

Fix for QScore errors when custom_qubits_array is specified.

Function to bootstrap counts added to utils file.

Standardizes observations for CLOPS and Mirror RB.

All functional tests extended to the fake Deneb backend.

Added backend transpilation to REM calibration circuits to fix errors with REM on fake Deneb.

Added notebook to benchmark IQM Star QPUs and bug fixes done for Qscore.

Update installation command for development mode. #41

Added readout error mitigation for Qscore benchmark.

Added optimal GHZ circuit generation and corresponding example notebook for all-to-all connected QPU topology.

Added devices folder in docs with notebook to benchmark IQM Spark.

Move all example notebooks to docs. #30

Added compatibility with IQM-Deneb by adapting the transpilation behavior in several benchmarks.

Report average native single-qubit gate fidelity estimates in observations of 1Q Clifford RB and 1Q IRB, and display in plots of 1Q Clifford RB.

Fix docs publishing by CI.

Add optional security-scanned lockfile.

Fixed a bug where optional dependencies related to gst were imported with other benchmarks, leading to a ModuleNotFoundError.

Fixed bugs in Qscore and enabled benchmark execution for pyrite.

Fixed bugs including wrong GHZ plot x-Axis labels and incorrect transpiled and untranspiled circuit storage for mGST.

Added note about optional dependency “mgst”.

Improved display and calculation method for Hamiltonian parameter output of rank 1 compressive GST.

Changed simulation method for MRB to ‘stabilizer’ and simulation circuits are compiled in circuit generation stage.

Changed Qscore to operate under the new base class.

Reverted QV simulation circuits to untranspiled ones (fixes bug giving all HOPs equal to zero).

Added Clifford RB example notebook to docs. #20

Fixed bug in RB plots for individual decays.

Adds Circuits, BenchmarkCircuit and CircuitGroup as a way to easily store and interact with multiple quantum circuits.

BenchmarkRunResult now takes a circuits argument, expecting an instance of Circuits. QuantumCircuit instances can now exist there instead of inside xarray Datasets. All analysis methods should also expect to use an instance of BenchmarkRunResult.

Ported all of the benchmarks subclassing from Benchmark to use the new containers.

Updates the usage of qiskit.QuantumCircuit to iqm.qiskit_iqm.IQMCircuit in many places.

Miscellaneous small bugs fixed.

Relaxes dependencies to allow for ranges.

Added API docs building and publishing.

Fixed bug (overwriting observations) in Quantum Volume.

Fixed small bug in CLOPS when calling plots in simulator execution.

Changed compressive GST to operate under the new base class and added multiple qubit layouts.

Added plot to GHZ benchmark and applied small fixes.

Added tutorial notebook for the GHZ benchmark.

Remove explicit dependency on qiskit, instead taking it from qiskit-on-iqm.

Minor change in dependencies for compatibility.

fit results are no longer BenchmarkObservation, and instead are moved into the datasets.

Renames:

• AnalysisResult -> BenchmarkAnalysisResult

• RunResult -> BenchmarkRunResult

Adds BenchmarkObservation class, and modifies BenchmarkAnalysisResult so observations now accepts a list[BenchmarkObservation].

Adds BenchmarkObservationIdentifier class.

Rebases RandomizedBenchmarking benchmarks, QuantumVolume, GHZ and CLOPS to use the new Observation class.

Fixes serialization of some circuits.

Adds AVAILABLE_BENCHMARKS to map a benchmark name to its class in __init__.

Adds benchmarks and configurations to __init__ for public import.

Other fixes.

Further improvements to type hints, docstrings, and error messages.

Minor improvements to type hints, docstrings, and error messages.

Fixed bug preventing execution on a generic IQM Backend.

Randomized Benchmarking (Clifford, Interleaved and Mirror), Quantum Volume, CLOPS and GHZ state fidelity all functioning exclusively under new Benchmark base class.

Updated separate example Jupyter notebooks.

Published Randomized Benchmarking (Clifford, Interleaved and Mirror), Quantum Volume, CLOPS and GHZ state fidelity all functioning exclusively under new Benchmark base class.

Updated separate example Jupyter notebooks.