Skip to content

Latest commit

 

History

History
983 lines (617 loc) · 31 KB

CHANGELOG.md

File metadata and controls

983 lines (617 loc) · 31 KB

Release 0.30.0-dev

New features since last release

  • Add MCMC sampler. [(#384)] (#384)

Breaking changes

  • Lightning now works with the new return types specification that is now default in PennyLane. See the PennyLane qml.enable_return documentation for more information on this change. (#427)

Instead of creating potentially ragged numpy array, devices and QNode's now return an object of the same type as that returned by the quantum function.

>>> dev = qml.device('lightning.qubit', wires=1)
>>> @qml.qnode(dev, diff_method="adjoint")
... def circuit(x):
...     qml.RX(x, wires=0)
...     return qml.expval(qml.PauliY(0)), qml.expval(qml.PauliZ(0))
>>> x = qml.numpy.array(0.5)
>>> circuit(qml.numpy.array(0.5))
(array(-0.47942554), array(0.87758256))

Interfaces like Jax or Torch handle tuple outputs without issues:

>>> jax.jacobian(circuit)(jax.numpy.array(0.5))
(Array(-0.87758255, dtype=float32, weak_type=True),
Array(-0.47942555, dtype=float32, weak_type=True))

Autograd cannot differentiate an output tuple, so results must be converted to an array before use with qml.jacobian:

>>> qml.jacobian(lambda y: qml.numpy.array(circuit(y)))(x)
array([-0.87758256, -0.47942554])

Alternatively, the quantum function itself can return a numpy array of measurements:

>>> dev = qml.device('lightning.qubit', wires=1)
>>> @qml.qnode(dev, diff_method="adjoint")
>>> def circuit2(x):
...     qml.RX(x, wires=0)
...     return np.array([qml.expval(qml.PauliY(0)), qml.expval(qml.PauliZ(0))])
>>> qml.jacobian(circuit2)(np.array(0.5))
array([-0.87758256, -0.47942554])

Improvements

  • Lightning wheels are now checked with twine check post-creation for PyPI compatibility. (#430)

  • Lightning has been made compatible with the change in return types specification. (#427)

Documentation

Bug fixes

Contributors

This release contains contributions from (in alphabetical order):

Christina Lee

Release 0.29.0

Improvements

  • Remove runtime dependency on ninja build system. (#414)

  • Allow better integration and installation support with CMake targeted binary builds. (#403)

  • Remove explicit Numpy and Scipy requirements. (#412)

  • Get llvm installation root from the environment variable LLVM_ROOT_DIR (or fallback to brew). (#413)

  • Update AVX2/512 kernel infrastructure for additional gate/generator operations. (#404)

  • Remove unnecessary lines for resolving CodeCov issue. (#415)

  • Add more AVX2/512 gate operations. (#393)

Documentation

Bug fixes

  • Ensure error raised when asking for out of order marginal probabilities. Prevents the return of incorrect results. (#416)

  • Fix Github shields in README. (#402)

Contributors

Amintor Dusko, Vincent Michaud-Rioux, Lee James O'Riordan, Chae-Yeun Park

Release 0.28.2

Bug fixes

  • Fix Python module versioning for Linux wheels. (#408)

Contributors

This release contains contributions from (in alphabetical order):

Amintor Dusko, Shuli Shu, Trevor Vincent

Release 0.28.1

Bug fixes

  • Fix Pybind11 module versioning and locations for Windows wheels. (#400)

Contributors

This release contains contributions from (in alphabetical order):

Lee J. O'Riordan

Release 0.28.0

Breaking changes

  • Deprecate support for Python 3.7. (#391)

Improvements

  • Improve Lightning package structure for external use as a C++ library. (#369)

  • Improve the stopping condition method. (#386)

Bug fixes

  • Pin CMake to 3.24.x in wheel-builder to avoid Python not found error in CMake 3.25, when building wheels for PennyLane-Lightning-GPU. (#387)

Contributors

This release contains contributions from (in alphabetical order):

Amintor Dusko, Lee J. O'Riordan

Release 0.27.0

New features since last release

  • Enable building of python 3.11 wheels and upgrade python on CI/CD workflows to 3.8. (#381)

Breaking changes

Improvements

  • Update clang-tools version in Github workflows. (#351)

  • Improve tests and checks CI/CD pipelines. (#353)

  • Implement 3 Qubits gates (CSWAP & Toffoli) & 4 Qubits gates (DoubleExcitation, DoubleExcitationMinus, DoubleExcitationPlus) in LM manner. (#362)

  • Upgrade Kokkos and Kokkos Kernels to 3.7.00, and improve sparse matrix-vector multiplication performance and memory usage. (#361)

  • Update Linux (ubuntu-latest) architecture x86_64 wheel-builder from GCC 10.x to GCC 11.x. (#373)

  • Update gcc and g++ 10.x to 11.x in CI tests. This update brings improved support for newer C++ features. (#370)

  • Change Lightning to inherit from QubitDevice instead of DefaultQubit. (#365)

Documentation

Bug fixes

  • Use mutex when accessing cache in KernelMap. (#382)

Contributors

This release contains contributions from (in alphabetical order):

Amintor Dusko, Chae-Yeun Park, Monit Sharma, Shuli Shu

Release 0.26.1

Bug fixes

  • Fixes the transposition method used in the probability calculation. (#377)

Contributor

Amintor Dusko

Release 0.26.0

Improvements

  • Introduces requirements-dev.txt and improves dockerfile. (#330)

  • Support expval for a Hamiltonian. (#333)

  • Implements caching for Kokkos installation. (#316)

  • Supports measurements of operator arithmetic classes such as Sum, Prod, and SProd by deferring handling of them to DefaultQubit. (#349)

@qml.qnode(qml.device('lightning.qubit', wires=2))
def circuit():
    obs = qml.s_prod(2.1, qml.PauliZ(0)) + qml.op_sum(qml.PauliX(0), qml.PauliZ(1))
    return qml.expval(obs)

Bug fixes

  • Test updates to reflect new measurement error messages. (#334)

  • Updates to the release tagger to fix incompatibilities with RTD. (#344)

  • Update cancel-workflow-action and bot credentials. (#345)

Contributors

This release contains contributions from (in alphabetical order):

Amintor Dusko, Christina Lee, Lee J. O'Riordan, Chae-Yeun Park

Release 0.25.0

New features since last release

Breaking changes

  • We explicitly disable support for PennyLane's parameter broadcasting. #317

  • We explicitly remove support for PennyLane's Sum, SProd and Prod as observables. (#326)

Improvements

  • CI builders use a reduced set of resources and redundant tests for PRs. (#319)

  • Parallelize wheel-builds where applicable. (#314)

  • AVX2/512 kernels are now available on Linux/MacOS with x86-64 architecture. (#313)

Documentation

  • Updated ReadTheDocs runner version from Ubuntu 20.04 to 22.04 (#327)

Bug fixes

  • Test updates to reflect new additions to PennyLane. (#318)

Contributors

This release contains contributions from (in alphabetical order):

Amintor Dusko, Christina Lee, Rashid N H M, Lee J. O'Riordan, Chae-Yeun Park

Release 0.24.0

New features since last release

  • Add SingleExcitation and DoubleExcitation qchem gates and generators. (#289)

  • Add a new dispatch mechanism for future kernels. (#291)

  • Add IsingXY gate operation. (#303)

  • Support qml.state() in vjp and Hamiltonian in adjoint jacobian. (#294)

Breaking changes

  • Codebase is now moving to C++20. The default compiler for Linux is now GCC10. (#295)

  • Minimum macOS version is changed to 10.15 (Catalina). (#295)

Improvements

  • Split matrix operations, refactor dispatch mechanisms, and add a benchmark suits. (#274)

  • Add native support for the calculation of sparse Hamiltonians' expectation values. Sparse operations are offloaded to Kokkos and Kokkos-Kernels. (#283)

  • Device lightning.qubit now accepts a datatype for a statevector. (#290)

dev1 = qml.device('lightning.qubit', wires=4, c_dtype=np.complex64) # for single precision
dev2 = qml.device('lightning.qubit', wires=4, c_dtype=np.complex128) # for double precision

Documentation

Bug fixes

  • Fix the issue with using available clang-format version in format. (#288)

  • Fix a bug in the generator of DoubleExcitationPlus. (#298)

Contributors

This release contains contributions from (in alphabetical order):

Mikhail Andrenkov, Ali Asadi, Amintor Dusko, Lee James O'Riordan, Chae-Yeun Park, and Shuli Shu

Release 0.23.0

New features since last release

  • Add generate_samples() to lightning. (#247)

  • Add Lightning GBenchmark Suite. (#249)

  • Support runtime and compile information. (#253)

Improvements

  • Add ENABLE_BLAS build to CI checks. (#249)

  • Add more clang-tidy checks and kernel tests. (#253)

  • Add C++ code coverage to CI. (#265)

  • Skip over identity operations in "lightning.qubit". (#268)

Bug fixes

  • Update tests to remove JacobianTape. (#260)

  • Fix tests for MSVC. (#264)

  • Fix #include <cpuid.h> for PPC and AArch64 in Linux. (#266)

  • Remove deprecated tape execution methods. (#270)

  • Update qml.probs in test_measures.py. (#280)

Contributors

This release contains contributions from (in alphabetical order):

Ali Asadi, Chae-Yeun Park, Lee James O'Riordan, and Trevor Vincent

Release 0.22.1

Bug fixes

  • Ensure Identity kernel is registered to C++ dispatcher. (#275)

Release 0.22.0

New features since last release

Improvements

  • Update quantum tapes serialization and Python tests. (#239)

  • Clang-tidy is now enabled for both tests and examples builds under Github Actions. (#237)

  • The return type of StateVectorBase data is now derived-class defined. (#237)

  • Update adjointJacobian and VJP methods. (#222)

  • Set GitHub workflow to upload wheels to Test PyPI. (#220)

  • Finalize the new kernel implementation. (#212)

Documentation

  • Use of batching with OpenMP threads is documented. (#221)

Bug fixes

  • Fix for OOM errors when using adjoint with large numbers of observables. (#221)

  • Add virtual destructor to C++ state-vector classes. (#200)

  • Fix a bug in Python tests with operations' matrix calls. (#238)

  • Refactor utility header and fix a bug in linear algebra function with CBLAS. (#228)

Contributors

This release contains contributions from (in alphabetical order):

Ali Asadi, Chae-Yeun Park, Lee James O'Riordan

Release 0.21.0

New features since last release

  • Add C++ only benchmark for a given list of gates. (#199)

  • Wheel-build support for Python 3.10. (#186)

  • C++ support for probability, expectation value and variance calculations. (#185)

  • Add bindings to C++ expval, var, probs. (#214)

Improvements

  • setup.py adds debug only when --debug is given (#208)

  • Add new highly-performant C++ kernels for quantum gates. (#202)

The new kernels significantly improve the runtime performance of PennyLane-Lightning for both differentiable and non-differentiable workflows. Here is an example workflow using the adjoint differentiation method with a circuit of 5 strongly entangling layers:

import pennylane as qml
from pennylane import numpy as np
from pennylane.templates.layers import StronglyEntanglingLayers
from numpy.random import random
np.random.seed(42)
n_layers = 5
n_wires = 6
dev = qml.device("lightning.qubit", wires=n_wires)

@qml.qnode(dev, diff_method="adjoint")
def circuit(weights):
    StronglyEntanglingLayers(weights, wires=list(range(n_wires)))
    return [qml.expval(qml.PauliZ(i)) for i in range(n_wires)]

init_weights = np.random.random(StronglyEntanglingLayers.shape(n_layers=n_layers, n_wires=n_wires))
params = np.array(init_weights,requires_grad=True)
jac = qml.jacobian(circuit)(params)

The latest release shows improved performance on both single and multi-threaded evaluations!

  • Ensure debug info is built into dynamic libraries. (#201)

Documentation

  • New guidelines on adding and benchmarking C++ kernels. (#202)

Bug fixes

  • Update clang-format version (#219)

  • Fix failed tests on Windows. (#218)

  • Update clang-format version (#219)

  • Add virtual destructor to C++ state-vector classes. (#200)

  • Fix failed tests for the non-binary wheel. (#213)

  • Add virtual destructor to C++ state-vector classes. (#200)

Contributors

This release contains contributions from (in alphabetical order):

Ali Asadi, Amintor Dusko, Chae-Yeun Park, Lee James O'Riordan

Release 0.20.1

Bug fixes

  • Fix missing header-files causing build errors in algorithms module. (#193)

  • Fix failed tests for the non-binary wheel. (#191)

Release 0.20.2

Bug fixes

  • Introduce CY kernel to Lightning to avoid issues with decomposition. (#203)

Contributors

This release contains contributions from (in alphabetical order):

Lee J. O'Riordan

Release 0.20.1

Bug fixes

  • Fix missing header-files causing build errors in algorithms module. (#193)

  • Fix failed tests for the non-binary wheel. (#191)

Release 0.20.0

New features since last release

  • Add wheel-builder support for Python 3.10. (#186)

  • Add VJP support to PL-Lightning. (#181)

  • Add complex64 support in PL-Lightning. (#177)

  • Added examples folder containing aggregate gate performance test. (#165)

Breaking changes

Improvements

  • Update PL-Lightning to support new features in PL. (#179)

Documentation

  • Lightning setup.py build process uses CMake. (#176)

Contributors

This release contains contributions from (in alphabetical order):

Ali Asadi, Chae-Yeun Park, Isidor Schoch, Lee James O'Riordan

Release 0.19.0

  • Add Cache-Friendly DOTC, GEMV, GEMM along with BLAS Support. (#155)

Improvements

  • The performance of parametric gates has been improved. (#157)

  • AVX support is enabled for Linux users on Intel/AMD platforms. (#157)

  • PennyLane-Lightning has been updated to conform with clang-tidy recommendations for modernization, offering performance improvements across all use-cases. (#153)

Breaking changes

  • Linux users on x86_64 must have a CPU supporting AVX. (#157)

Bug fixes

  • OpenMP built with Intel MacOS CI runners causes failures on M1 Macs. OpenMP is currently disabled in the built wheels until this can be resolved with Github Actions runners. (#166)

Contributors

This release contains contributions from (in alphabetical order):

Ali Asadi, Lee James O'Riordan

Release 0.18.0

New features since last release

  • PennyLane-Lightning now provides a high-performance adjoint Jacobian method for differentiating quantum circuits. (#136)

    The adjoint method operates after a forward pass by iteratively applying inverse gates to scan backwards through the circuit. The method is already available in PennyLane's default.qubit device, but the version provided by lightning.qubit integrates with the C++ backend and is more performant, as shown in the plot below:

    The plot compares the average runtime of lightning.qubit and default.qubit for calculating the Jacobian of a circuit using the adjoint method for a range of qubit numbers. The circuit consists of ten BasicEntanglerLayers with a PauliZ expectation value calculated on each wire, repeated over ten runs. We see that lightning.qubit provides a speedup of around two to eight times, depending on the number of qubits.

    The adjoint method can be accessed using the standard interface. Consider the following circuit:

    import pennylane as qml

wires = 3
layers = 2
dev = qml.device("lightning.qubit", wires=wires)

@qml.qnode(dev, diff_method="adjoint")
def circuit(weights):
    qml.templates.StronglyEntanglingLayers(weights, wires=range(wires))
    return qml.expval(qml.PauliZ(0))

weights = qml.init.strong_ent_layers_normal(layers, wires, seed=1967)

    The circuit can be executed and its gradient calculated using:

    >>> print(f"Circuit evaluated: {circuit(weights)}")
Circuit evaluated: 0.9801286266677633
>>> print(f"Circuit gradient:\n{qml.grad(circuit)(weights)}")
Circuit gradient:
[[[-1.11022302e-16 -1.63051504e-01 -4.14810501e-04]
[ 1.11022302e-16 -1.50136528e-04 -1.77922957e-04]
[ 0.00000000e+00 -3.92874550e-02  8.14523075e-05]]

[[-1.14472273e-04  3.85963953e-02  0.00000000e+00]
[-5.76791765e-05 -9.78478343e-02  0.00000000e+00]
[-5.55111512e-17  0.00000000e+00 -1.11022302e-16]]]

  • PennyLane-Lightning now supports all of the operations and observables of default.qubit. (#124)

Improvements

  • A new state-vector class StateVectorManaged was added, enabling memory use to be bound to statevector lifetime. (#136)

  • The repository now has a well-defined component hierarchy, allowing each indepedent unit to be compiled and linked separately. (#136)

  • PennyLane-Lightning can now be installed without compiling its C++ binaries and will fall back to using the default.qubit implementation. Skipping compilation is achieved by setting the SKIP_COMPILATION environment variable, e.g., Linux/MacOS: export SKIP_COMPILATION=True, Windows: set SKIP_COMPILATION=True. This feature is intended for building a pure-Python wheel of PennyLane-Lightning as a backup for platforms without a dedicated wheel. (#129)

  • The C++-backed Python bound methods can now be directly called with wires and supplied parameters. (#125)

  • Lightning supports arbitrary unitary and non-unitary gate-calls from Python to C++ layer. (#121)

Documentation

  • Added preliminary architecture diagram for package. (#131)

  • C++ API built as part of docs generation. (#131)

Breaking changes

  • Wheels for MacOS <= 10.13 will no longer be provided due to XCode SDK C++17 support requirements. (#149)

Bug fixes

  • An indexing error in the CRY gate is fixed. (#136)

  • Column-major data in numpy is now correctly converted to row-major upon pass to the C++ layer. (#126)

Contributors

This release contains contributions from (in alphabetical order):

Thomas Bromley, Lee James O'Riordan

Release 0.17.0

New features

  • C++ layer now supports float (32-bit) and double (64-bit) templated complex data. (#113)

Improvements

  • The PennyLane device test suite is now included in coverage reports. (#123)

  • Static versions of jQuery and Bootstrap are no longer included in the CSS theme. (#118)

  • C++ tests have been ported to use Catch2 framework. (#115)

  • Testing now exists for both float and double precision methods in C++ layer. (#113) (#115)

  • Compile-time utility methods with constexpr have been added. (#113)

  • Wheel-build support for ARM64 (Linux and MacOS) and PowerPC (Linux) added. (#110)

  • Add support for Controlled Phase Gate (ControlledPhaseShift). (#114)

  • Move changelog to .github and add a changelog reminder. (#111)

  • Adds CMake build system support. (#104)

Breaking changes

  • Removes support for Python 3.6 and adds support for Python 3.9. (#127) (#128)

  • Compilers with C++17 support are now required to build C++ module. (#113)

  • Gate classes have been removed with functionality added to StateVector class. (#113)

  • We are no longer building wheels for Python 3.6. (#106)

Bug fixes

  • PowerPC wheel-builder now successfully compiles modules. (#120)

Documentation

  • Added community guidelines. (#109)

Contributors

This release contains contributions from (in alphabetical order):

Ali Asadi, Christina Lee, Thomas Bromley, Lee James O'Riordan

Release 0.15.1

Bug fixes

  • The PennyLane-Lightning binaries are now built with NumPy 1.19.5, to avoid ABI compatibility issues with the latest NumPy 1.20 release. See the NumPy release notes for more details. (#97)

Contributors

This release contains contributions from (in alphabetical order):

Josh Izaac, Antal Száva

Release 0.15.0

Improvements

  • For compatibility with PennyLane v0.15, the analytic keyword argument has been removed. Statistics can still be computed analytically by setting shots=None. (#93)

  • Inverse gates are now supported. (#89)

  • Add new lightweight backend with performance improvements. (#57)

  • Remove the previous Eigen-based backend. (#67)

Bug fixes

  • Re-add dispatch table after fixing static initialisation order issue. (#68)

Contributors

This release contains contributions from (in alphabetical order):

Thomas Bromley, Theodor Isacsson, Christina Lee, Thomas Loke, Antal Száva.

Release 0.14.1

Bug fixes

  • Fixes a bug where the QNode would swap LightningQubit to DefaultQubitAutograd on device execution due to the inherited passthru_devices entry of the capabilities dictionary. (#61)

Contributors

This release contains contributions from (in alphabetical order):

Antal Száva

Release 0.14.0

Improvements

  • Extends support from 16 qubits to 50 qubits. (#52)

Bug fixes

  • Updates applying basis state preparations to correspond to the changes in DefaultQubit. (#55)

Contributors

This release contains contributions from (in alphabetical order):

Thomas Loke, Tom Bromley, Josh Izaac, Antal Száva

Release 0.12.0

Bug fixes

  • Updates capabilities dictionary to be compatible with core PennyLane (#45)

  • Fix install of Eigen for CI wheel building (#44)

Contributors

This release contains contributions from (in alphabetical order):

Tom Bromley, Josh Izaac, Antal Száva

Release 0.11.0

Initial release.

This release contains contributions from (in alphabetical order):

Tom Bromley, Josh Izaac, Nathan Killoran, Antal Száva