NumPy & SciPy for GPU
This is the release note of v11.6.0. See here for the complete list of solved issues and merged PRs.
This is the last planned release for CuPy v11 series. Please start testing your workload with the v12 release candidate to get ready for the final v12 release. To install:pip install -U --pre cupy-cuda11x -f https://pip.cupy.dev/pre. See the Upgrade Guide for the list of possible breaking changes in v12.
We are running a Gitter chat for general discussions and quick questions. Feel free to join the channel to talk with developers and users!
This release fixes a critical performance regression in CUDA 12.0 that the on-disk kernel cache is ineffective, causing kernels to be recompiled for each python process. Users with CUDA 12.0 are strongly suggested to upgrade to this release.
runtime.getDeviceProperties (#7353)cupy.cuda.profiler.initialize deprecated as it is removed in CUDA 12 (#7379)arange() to raise TypeError in boolean case (#7407)cupyx.scipy.sparse.eigsh (#7361)TestRoundHalfway (#7362)nanargmin/max tests (#7381)fillvalue overflow in cupyx.scipy.signal test (#7401)The CuPy Team would like to thank all those who contributed to this release!
@asi1024 @emcastillo @kmaehashi @leofang @RisaKirisu
This is the release note of v12.0.0rc1. See here for the complete list of solved issues and merged PRs.
This is a release candidate of the CuPy v12 series. Please start testing your workload with this release to prepare for the final v12 release. To install: pip install -U --pre cupy-cuda11x -f https://pip.cupy.dev/pre. See the Upgrade Guide for the list of possible breaking changes in v12.
We are running a Gitter chat for general discussions and quick questions. Feel free to join the channel to talk with developers and users!
cupyx.scipy.interpolateThe following interpolators have been implemented: BPoly, Akima1DInterpolator, PchipInterpolator.
Acknowledgements: This work was done by Edgar Andrés Margffoy Tuay (@andfoy) and Evgeni Burovski (@ev-br) under the support of the Chan Zuckerberg Initiative's Essential Open Source Software for Science program.
CuPy is now compatible with DLPack v0.8 to allow importing/exporting bool arrays.
This release fixes a critical performance regression in CUDA 12.0 that the on-disk kernel cache is ineffective, causing kernels to be recompiled for each python process. Users with CUDA 12.0 are strongly suggested to upgrade to this release.
cupy.cuda.Device Behavior (#7427)The CUDA current device (set via cupy.cuda.Device.use() or underlying CUDA API cudaSetDevice()) will now be reactivated when exiting a cupy.cuda.Device context manager. This reverts the change introduced in CuPy v10, making the behavior identical to the one in CuPy v9 or earlier. Please refer to the Upgrade Guide for the background of this decision.
As per NEP 29, CuPy v12 drops support for Python 3.7 and NumPy 1.20. Support for SciPy 1.6 has been dropped as well.
Texture reference features (RawModule.get_texref() and TextureReference), which were marked deprecated in CUDA 10.1 and removed in CUDA 12.0, have been removed from CuPy.
cupyx.distributed._array implementation (#7040)PchipInterpolator to cupyx.scipy.interpolate (#7255)Akima1DInterpolator to cupyx.scipy.interpolate (#7260)cached_code to ElementwiseKernel and ReductionKernel (#7265)RegularGridInterpolator (#7334)BPoly to cupyx.scipy.interpolate module (#7343)runtime.getDeviceProperties (#7302)cupy.cuda.profiler.initialize deprecated as it is removed in CUDA 12 (#7377)cupy.cuda.Device behavior to v9 (#7427)ndarray.fill to raise ComplexWarning (#7393)arange() to raise TypeError in boolean case (#7394)cupyx.scipy.sparse.eigsh (#7356)CUPY_INCLUDE_PATH and CUPY_LIBRARY_PATH env vars (#7305)TestRoundHalfway (#7338)nanargmin/max tests (#7341)fillvalue overflow in cupyx.scipy.signal test (#7397)The CuPy Team would like to thank all those who contributed to this release!
Contributors: @andfoy @asi1024 @emcastillo @ev-br @kmaehashi @leofang @Nordicus @Raghav323 @RisaKirisu @seberg @wstolp
This is the release note of v11.5.0. See here for the complete list of solved issues and merged PRs.
We are running a Gitter chat for general discussions and quick questions. Feel free to join the channel to talk with developers and users!
CuPy now supports CUDA 12.0 and NVIDIA's latest H100 GPU. Binary packages are available for Linux (x86_64/aarch64) and Windows.
$ pip install cupy-cuda12x
For aarch64:
$ pip install cupy-cuda12x -f https://pip.cupy.dev/aarch64
Note that cuDNN support is unavailable at this time as cuDNN for CUDA 12 has not yet been released.
cupy-cuda12x to cupy-wheel (#7327)array_api (#7321)broadcast_to (#7291)The CuPy Team would like to thank all those who contributed to this release!
@anaruse @hubertlu-tw @kmaehashi @leofang @takagi
This is the release note of v12.0.0b3. See here for the complete list of solved issues and merged PRs.
We are running a Gitter chat for general discussions and quick questions. Feel free to join the channel to talk with developers and users!
CuPy now supports CUDA 12.0 and NVIDIA's latest H100 GPU. Binary packages are available for Linux (x86_64/aarch64) and Windows.
$ pip install cupy-cuda12x --pre -f https://pip.cupy.dev/pre
Note that cuDNN support is unavailable at this time as cuDNN for CUDA 12 has not yet been released.
NVTX support in CuPy is now backed by NVTX3 instead of the legacy NVTX1.
cupyx.scipy.interpolate.make_interp_spline (#7195)RegularGridInterpolator and interpn from scipy.interpolate (#7197)PPoly to cupyx.scipy.interpolate (#7204)uniform() to random generator (#7205)make_interp_spline(..., bc_type="periodic") (#7206)CubicHermiteSpline to cupyx.scipy.interpolate (#7242)cupyx.scipy.sparse.linalg.spsolve : allow two-dimensional right-hand sides in A @ X = B (#7219)cupyx.scipy.sparse.linalg.eigsh (#7269)RegularGridInterpolator (#7275)cupy-cuda12x to cupy-wheel (#7300)array_api (#7313)broadcast_to (#7271)cupyx.scipy.interpolate (#7262)The CuPy Team would like to thank all those who contributed to this release!
@anaruse @andfoy @asi1024 @ev-br @hubertlu-tw @ideasrule @kmaehashi @leofang @mandal-saswata @oishigyunyu @takagi
This is the release note of v11.4.0. See here for the complete list of solved issues and merged PRs.
We are running a Gitter chat for general discussions and quick questions. Feel free to join the channel to talk with developers and users!
lexsort (#7191)cupyx.scipy.ndimage.zoom for outputs of size 1 when mode is 'opencv' (#7202)list(kwargs) instead of list(kwargs.keys) (#7213)The CuPy Team would like to thank all those who contributed to this release!
@emcastillo @hadipash @jjmortensen @kmaehashi @takagi
This is the release note of v12.0.0b2. See here for the complete list of solved issues and merged PRs.
We are running a Gitter chat for general discussions and quick questions. Feel free to join the channel to talk with developers and users!
cupyx.scipy.interpolate APIs (#7086, #7190 and #7215)Increased coverage of cupyx.scipy.interpolate APIs, which now includes BSpline, RBFInterpolator, splantider and splder.
Acknowledgements: This work was done by Edgar Andrés Margffoy Tuay (@andfoy) and Evgeni Burovski (@ev-br) under the support of the Chan Zuckerberg Initiative's Essential Open Source Software for Science program.
cupyx.jit (#7145)Now it is possible to use the CUB reduction classes, cub::WarpReduce and cub::BlockReduce, in kernels written using CuPy JIT.
import cupy, cupyx
from cupy.cuda import runtime
from cupyx import jit
@jit.rawkernel()
def warp_reduce_sum(x, y):
WarpReduce = jit.cub.WarpReduce[cupy.int32]
temp_storage = jit.shared_memory(
dtype=WarpReduce.TempStorage, size=1)
i, j = jit.blockIdx.x, jit.threadIdx.x
value = x[i, j]
aggregator = WarpReduce(temp_storage[0])
aggregate = aggregator.Reduce(value, jit.cub.Sum())
if j == 0:
y[i] = aggregate
warp_size = 64 if runtime.is_hip else 32
h, w = (32, warp_size)
x = cupy.arange(h * w, dtype=cupy.int32).reshape(h, w)
cupy.random.shuffle(x)
y = cupy.zeros(h, dtype=cupy.int32)
warp_reduce_sum[h, w](x, y)
Acknowledgements: This work was done by Tsutsui Masayoshi (@TsutsuiMasayoshi) as a part of the internship program at Preferred Networks.
BSpline to interpolate module (#7086)cub::WarpReduce and cub::BlockReduce (#7145)cupyx.scipy.interpolate.RBFInterpolator (#7190)splder and splantider (#7215)_PerfCaseResult.to_str format (#7152)lexsort (#7178)cupyx.scipy.ndimage.zoom for outputs of size 1 when mode is 'opencv' (#7192)warnings.warn() (#7194)list(kwargs) instead of list(kwargs.keys) (#7203)comb() for Python 3.7 (#7221)The CuPy Team would like to thank all those who contributed to this release!
@anaruse @andfoy @asi1024 @emcastillo @ev-br @hadipash @jjmortensen @kmaehashi @takagi @TsutsuiMasayoshi
This is the release note of v11.3.0. See here for the complete list of solved issues and merged PRs.
We are running a Gitter chat for general discussions and quick questions. Feel free to join the channel to talk with developers and users!
This release adds support for CUDA 11.8 and the latest NVIDIA H100 GPUs. Note that CUDA 11.8 support is included in the cupy-cuda11x wheel.
Wheels are now available for Python 3.11.
cutensorPermutation (#7083)inf/nan in cupy.fuse (#7128)op.routine including in0_type (#7096)__del__ in TCPStore (#7111)cupy.nansum in fusing (#7114)TypeError in cupy._core.fusion._call_ufunc() (#7130)/test jenkins to trigger Jenkins only (#7129)The CuPy Team would like to thank all those who contributed to this release!
@anaruse @andfoy @asi1024 @emcastillo @kmaehashi @leofang @takagi
This is the release note of v12.0.0b1. See here for the complete list of solved issues and merged PRs.
We are running a Gitter chat for general discussions and quick questions. Feel free to join the channel to talk with developers and users!
This release adds support for CUDA 11.8 and the latest NVIDIA H100 GPUs. Note that CUDA 11.8 support is included in the cupy-cuda11x wheel.
Wheels are now available for Python 3.11.
ufunc MethodsThis release adds ufunc.reduce, ufunc.accumulate, ufunc.reduceat, and ufunc.at methods. See the documentation for more details.
cupyx.jit (#7054, #7139)Now it is possible to use the Thrust library device functions in kernels written using CuPy JIT.
import cupy, cupyx
@cupyx.jit.rawkernel()
def sort_by_key(x, y):
i = cupyx.jit.threadIdx.x
x_array = x[i]
y_array = y[i]
cupyx.jit.thrust.sort_by_key(
cupyx.jit.thrust.device,
x_array.begin(),
x_array.end(),
y_array.begin(),
)
h, w = (256, 256)
x = cupy.arange(h * w, dtype=cupy.int32)
cupy.random.shuffle(x)
x = x.reshape(h, w)
y = cupy.arange(h * w, dtype=cupy.int32)
cupy.random.shuffle(y)
y = y.reshape(h, w)
sort_by_key[1, 256](x, y)
Currently supported Thrust functions are count, copy, find, mismatch, sort, sort_by_key.
Acknowledgements: This work was done by Tsutsui Masayoshi (@TsutsuiMasayoshi) as a part of the internship program at Preferred Networks.
ndarray.scatter_{add,max,min} (#7097)cupy.ndarray.scatter_{add,max,min} methods are marked as deprecated. Use the corresponding ufunc methods (cupy.{add,maximum,minimum}.at) instead.
cupyx (#7013)Previously, CuPy has been providing high-level wrappers for CUDA libraries as cupy.cudnn, cupy.cusolver, cupy.cusparse, and cupy.cutensor. These modules are now moved to cupyx as a part of the cupy namespace cleanup. The old modules are still available but marked as deprecated. Note that these modules are still undocumented and may be subject to change.
axis to cupy.logspace (#6797)thrust::count, device in CuPy JIT (#7054)cupy.ndarray.searchsorted (#7059)add.at, maximum.at, minimum.at (#7077)subtract.at, bitwise_and.at, bitwise_or.at, bitwise_xor.at (#7099)ufunc.reduce and ufunc.accumulate (#7105)cupy.add.reduceat (#7115)cupy.min_scalar_type (#7136)cupy.cudnn cupy.cusolver cupy.cutensor cupy.cusparse to cupyx (#7013)ndarray.scatter_{add, max, min} (#7097)inf/nan in cupy.fuse (#7122)TypeError instead of ValueError in cupy.from_dlpack when CPU tensor is passed (#7133)__del__ in TCPStore (#6989)op.routine including in0_type (#7076)cupy.nansum in fusing (#7102)TypeError in cupy._core.fusion._call_ufunc() (#7113)_ufunc_method directory (#7116)/test jenkins to trigger Jenkins only (#7126)thrust::sort test (#7162)The CuPy Team would like to thank all those who contributed to this release!
@anaruse @andfoy @asi1024 @Diwakar-Gupta @emcastillo @IncubatorShokuhou @kmaehashi @MarcoGorelli @takagi @TsutsuiMasayoshi
This is the release note of v12.0.0a2. See here for the complete list of solved issues and merged PRs.
We are running a Gitter chat for general discussions and quick questions. Feel free to join the channel to talk with developers and users!
cupyx.scipy APIs (#6773, #6990, #7014, #7015, #7036)The coverage of SciPy interpolate & special APIs has increased. (Thanks @khushi-411 & @1MrEnot!)
ufunc methods (#7049)Starting from v12, CuPy will support the corresponding NumPy ufunc methods.
This release adds compatibility with ufunc.outer. Check the tracking issue (#7082) for detailed information.
cupyx.scipy.special.logsumexp (#6773)cupyx.scipy.interpolate.KroghInterpolator (#6990)scipy.special.expi and scipy.special.exp1 (#7014)cupy.byte_bounds (#7015)cupyx.scipy.special.k0, cupyx.scipy.special.k1, cupyx.scipy.special.k0e, cupyx.scipy.special.k1e (#7036)ufunc.outer (#7049)cupy.apply_along_axis for tuple retval (#7068)cutensorPermutation (#7070)csrsm2 memory leak (#7039)_routine_indexing.pyx (#7053)The CuPy Team would like to thank all those who contributed to this release!
@1MrEnot @andfoy @asi1024 @betatim @khushi-411 @kmaehashi @leofang @maronuu @takagi @wyli
This is the release note of v11.2.0. See here for the complete list of solved issues and merged PRs.
We are running a Gitter chat for general discussions and quick questions. Feel free to join the channel to talk with developers and users!
csrsm2 memory leak (#7041)_routine_indexing.pyx (#7056)distutils.utils (#7009)The CuPy Team would like to thank all those who contributed to this release!
@andfoy @asi1024 @betatim @kmaehashi @leofang @takagi @wyli