Changed

• Updated QCS SDK to 0.31.1. Update QCS SDK for benchmarking campaign. Required for single-use
  refresh tokens.

Changed

• Updated QCS SDK to 0.30.0. Update QCS SDK for benchmarking campaign. Required for single-use
  refresh tokens.

Changed

• A Channel whose noise generator acts on more levels than its gate promotes the gate to the
  generator's dims, as the identity on the levels above its own: Channel.from_lindbladian(RX(theta, q), qx.lindbladians.leakage(rate)) is a qutrit channel without a hand-promoted gate in custom_gates.
  Channel.from_mixture does the same for constituents on more levels. Noise that cannot hold the gate
  raises a ValueError naming both dims.
• MeasurementChannel.from_confusion_and_transition takes a (num_outcomes, d) confusion matrix, so a
  two-state discriminator on a qutrit is a (2, 3) matrix, and its transition matrix defaults to the
  identity (a quantum non-demolition measurement).

Deprecated

APIs that pyQuil v5 removes now emit a PyQuilDeprecationWarning when used, and a deprecated
module warns when imported. It is a FutureWarning, so Python shows it by default; filter on
pyquil._deprecation.PyQuilDeprecationWarning to silence it.

• Replaced in pyQuil v5 by the JAX-based simulators and noise model. pyQuil v4 already includes
  them as private, experimental modules (pyquil.simulation._simulator and
  pyquil.noise._noise_model; see the
  simulation architecture documentation),
  so we recommend continuing to use the deprecated APIs until you upgrade to pyQuil v5. Note also,
  the qcs-sdk maintains support for executing Quil
  programs on the QVM as of its 0.27.0 release (see documentation for
  qcs_sdk.qvm):
  • The pyquil.pyqvm and pyquil.wavefunction modules.
  • pyquil.api.QVM, pyquil.api.QVMCompiler and pyquil.api.WavefunctionSimulator, and getting
    a QVM-backed quantum computer from get_qc (a -qvm or -pyqvm name, or as_qvm=True).
  • From pyquil.simulation: NumpyWavefunctionSimulator, ReferenceWavefunctionSimulator,
    ReferenceDensitySimulator, get_measure_probabilities, targeted_einsum,
    targeted_tensordot and zero_state_matrix.
  • The Kraus-map noise model in pyquil.noise: NoiseModel, KrausModel, NoisyGateUndefined,
    add_decoherence_noise, append_kraus_to_gate, apply_noise_model,
    bitstring_probs_to_z_moments, combine_kraus_maps, correct_bitstring_probs,
    corrupt_bitstring_probs, damping_after_dephasing, damping_kraus_map,
    decoherence_noise_with_asymmetric_ro, dephasing_kraus_map, estimate_assignment_probs,
    estimate_bitstring_probs, get_noisy_gate, pauli_kraus_map and tensor_kraus_maps. The
    constants ANGLE_TOLERANCE, INFINITY and NO_NOISE are deprecated too, but do not warn.
  • Program.define_noisy_gate, Program.define_noisy_readout and
    pyquil.quil.merge_with_pauli_noise, whose noise pragmas only the QVM understands.
• Replaced in pyQuil v5 by rigetti-qpu-hybrid-benchmark:
  • The pyquil.experiment module, with QuantumComputer.run_experiment and
    QuantumComputer.calibrate.
  • The pyquil.operator_estimation module, by its Estimator interface.
• Removed in pyQuil v5 without replacement: the pyquil.latex module.
• pyquil.api.local_forest_runtime warns that starting a QVM server is deprecated: in pyQuil v5 it
  starts only quilc.

Deprecated

APIs that pyQuil v5 removes now emit a PyQuilDeprecationWarning when used, and a deprecated
module warns when imported. It is a FutureWarning, so Python shows it by default; filter on
pyquil._deprecation.PyQuilDeprecationWarning to silence it.

• Replaced in pyQuil v5 by the JAX-based simulators and noise model. pyQuil v4 already includes
  them as private, experimental modules (pyquil.simulation._simulator and
  pyquil.noise._noise_model; see the
  simulation architecture documentation),
  so we recommend continuing to use the deprecated APIs until you upgrade to pyQuil v5. Note also,
  the qcs-sdk maintains support for executing Quil
  programs on the QVM as of its 0.27.0 release (see documentation for
  qcs_sdk.qvm):
  • The pyquil.pyqvm and pyquil.wavefunction modules.
  • pyquil.api.QVM, pyquil.api.QVMCompiler and pyquil.api.WavefunctionSimulator, and getting
    a QVM-backed quantum computer from get_qc (a -qvm or -pyqvm name, or as_qvm=True).
  • From pyquil.simulation: NumpyWavefunctionSimulator, ReferenceWavefunctionSimulator,
    ReferenceDensitySimulator, get_measure_probabilities, targeted_einsum,
    targeted_tensordot and zero_state_matrix.
  • The Kraus-map noise model in pyquil.noise: NoiseModel, KrausModel, NoisyGateUndefined,
    add_decoherence_noise, append_kraus_to_gate, apply_noise_model,
    bitstring_probs_to_z_moments, combine_kraus_maps, correct_bitstring_probs,
    corrupt_bitstring_probs, damping_after_dephasing, damping_kraus_map,
    decoherence_noise_with_asymmetric_ro, dephasing_kraus_map, estimate_assignment_probs,
    estimate_bitstring_probs, get_noisy_gate, pauli_kraus_map and tensor_kraus_maps. The
    constants ANGLE_TOLERANCE, INFINITY and NO_NOISE are deprecated too, but do not warn.
  • Program.define_noisy_gate, Program.define_noisy_readout and
    pyquil.quil.merge_with_pauli_noise, whose noise pragmas only the QVM understands.
• Replaced in pyQuil v5 by rigetti-qpu-hybrid-benchmark:
  • The pyquil.experiment module, with QuantumComputer.run_experiment and
    QuantumComputer.calibrate.
  • The pyquil.operator_estimation module, by its Estimator interface.
• Removed in pyQuil v5 without replacement: the pyquil.latex module.
• pyquil.api.local_forest_runtime warns that starting a QVM server is deprecated: in pyQuil v5 it
  starts only quilc.

Features

• TrajectorySimulator, a Monte-Carlo trajectory simulator for programs with mid-circuit
  measurements and resets. It samples pure-state trajectories under JAX, returning the
  measurement outcomes alongside the final state; sample streams trajectories in batches,
  data-parallel across the available devices, so the shot count is not bounded by device
  memory. compute is jax.jit-traceable (though not differentiable).
  (#1861)
• A gate angle may be any Quil arithmetic expression over declared memory, so the parametric
  programs quilc emits -- RX(theta[0]/2 + pi) 0 -- simulate without rewriting.
  (#1869). Restored from
  #1860, where it was errantly dropped during a
  rebase (prior to the 4.19.0 release).
• The simulators accept an initial_state to evolve from, so a circuit can be split in two and
  a prefix shared across many runs evolved once instead of once per run.

Fixed

• Tolerant Circuit equality (and its unhashability, which follows), the split of
  Circuit.__post_init__ into named coercion and validation steps, and a mypy fix for JAX's
  dynamically defined config flags. Restored from
  #1860, where it was errantly dropped during a
  rebase (prior to the 4.19.0 release).
• Circuit.to_kraus_maps no longer truncates the Kraus set.

Features

• Helper for adding leading delays to programs with leading classical computation, including
  cases with control flow. See pyquil._qpu.add_leading_delay_to_first_pulse_block.
  (#1851)
• Support for per-shot randomized compiling on the QPU control system. The top-level
  pyquil._qpu.RandomizedCompilingConfiguration provides methods for building the necessary
  classical instructions (build_quil_program), seed values (generate_seed_values), and
  memory map values (build_memory_map) to enable the method. It additionally provides methods
  for verifying final memory read off of the QPU for verification (verify_final_memory),
  along with tracking the randomizations shot by shot (track_pauli_frames).
  (#1851)
• Support Quil shift operators. (#1851)

Features

• Experimental exact simulators built on rigetti-quax: PureStateVectorSimulator and
  DensityMatrixSimulator in pyquil.simulation._simulator. They evolve a program exactly
  under JAX, so they are jit-compatible and differentiable with respect to gate
  parameters, and they accept a NoiseModel. The API is not yet stable; see
  the architecture guide.
  (#1855)
• A gate angle may be any Quil arithmetic expression over declared memory, so the parametric
  programs quilc emits -- RX(theta[0]/2 + pi) 0 -- simulate without rewriting.
  (#1869)

Features

drop Python 3.9 support, add rigetti-quax (1/5) (#1858)

qudit noise model (2/5) (#1854) (#1859)

Fixes

format complex params with nonzero real and unit imaginary part (#1850)

Features

drop Python 3.9 support, add rigetti-quax (1/5) (#1858)

qudit noise model (2/5) (#1854) (#1859)

Fixes

format complex params with nonzero real and unit imaginary part (#1850)

Features

drop Python 3.9 support, add rigetti-quax (1/5) (#1858)

Fixes

format complex params with nonzero real and unit imaginary part (#1850)