aboutsummaryrefslogtreecommitdiffhomepage
path: root/unsupported/Eigen/CXX11
Commit message (Collapse)AuthorAge
* Add block evaluation support to TensorOpsGravatar Eugene Zhulenev2018-07-31
|
* Rename Index to StorageIndex + use Eigen::Array and Eigen::Map when possibleGravatar Eugene Zhulenev2018-07-27
|
* Add tiled evaluation support to TensorExecutorGravatar Eugene Zhulenev2018-07-25
|
* TensorBlockIOGravatar Eugene Zhulenev2018-07-23
|
* Initial support of TensorBlockGravatar Eugene Zhulenev2018-07-20
|
* PR430: Convert count to the reducer type in MeanReducerGravatar Eugene Zhulenev2018-07-19
| | | | | | | | | | | | | Without explicit conversion Tensorflow fails to compile, pset1 template deduction fails. cannot convert '((const Eigen::internal::MeanReducer<Eigen::half>*)this) ->Eigen::internal::MeanReducer<Eigen::half>::packetCount_' (type 'const DenseIndex {aka const long int}') to type 'const type& {aka const Eigen::half&}' return pdiv(vaccum, pset1<Packet>(packetCount_)); Honestly I’m not sure why it works in Eigen tests, because Eigen::half constructor is explicit, and why it stopped working in TF, I didn’t find any relevant changes since previous Eigen upgrade. static_cast<T>(packetCount_) - breaks cxx11_tensor_reductions test for Eigen::half, also quite surprising.
* Reduce number of allocations in TensorContractionThreadPool.Gravatar Eugene Zhulenev2018-07-16
|
* bug #1569: fix Tensor<half>::mean() on AVX with respective unit test.Gravatar Gael Guennebaud2018-07-19
|
* Assert that no output kernel is defined for GPU contractionGravatar Eugene Zhulenev2018-07-18
|
* Specify default output kernel for TensorContractionOpGravatar Eugene Zhulenev2018-07-18
|
* Added a move constructor and move assignment operator to Tensor and wrote ↵Gravatar Viktor Csomor2018-02-07
| | | | some tests.
* Fix TensorContractionOp evaluators for GPU and SYCLGravatar Eugene Zhulenev2018-07-17
|
* applying EIGEN_DECLARE_TEST to *gpu* testsGravatar Deven Desai2018-07-17
| | | | | | | | | | | | | Also, a few minor fixes for GPU tests running in HIP mode. 1. Adding an include for hip/hip_runtime.h in the Macros.h file For HIP __host__ and __device__ are macros which are defined in hip headers. Their definitions need to be included before their use in the file. 2. Fixing the compile failure in TensorContractionGpu introduced by the commit to "Fuse computations into the Tensor contractions using output kernel" 3. Fixing a HIP/clang specific compile error by making the struct-member assignment explicit
* Call OutputKernel in evalGemvGravatar Eugene Zhulenev2018-07-12
|
* Remove SimpleThreadPool and always use {NonBlocking}ThreadPoolGravatar Eugene Zhulenev2018-07-16
|
* Fuse computations into the Tensor contractions using output kernelGravatar Eugene Zhulenev2018-07-10
|
* Get rid of alias for m_broadcast.Gravatar Rasmus Munk Larsen2018-07-13
|
* Optimize the case where broadcasting is a no-op.Gravatar Rasmus Munk Larsen2018-07-13
|
* Introduce gpu_assert for assertion in device-code, and disable them with ↵Gravatar Gael Guennebaud2018-07-13
| | | | clang-cuda.
* Add deprecated header files for TensorFlowGravatar Gael Guennebaud2018-07-12
|
* Updates corresponding to the latest round of PR feedbackGravatar Deven Desai2018-07-11
| | | | | | | | | | | | | | The major changes are 1. Moving CUDA/PacketMath.h to GPU/PacketMath.h 2. Moving CUDA/MathFunctions.h to GPU/MathFunction.h 3. Moving CUDA/CudaSpecialFunctions.h to GPU/GpuSpecialFunctions.h The above three changes effectively enable the Eigen "Packet" layer for the HIP platform 4. Merging the "hip_basic" and "cuda_basic" unit tests into one ("gpu_basic") 5. Updating the "EIGEN_DEVICE_FUNC" marking in some places The change has been tested on the HIP and CUDA platforms.
* merging updates from upstreamGravatar Deven Desai2018-07-11
|\
| * bug #1567: add optimized path for tensor broadcasting and 'Channel First' shapeGravatar Gael Guennebaud2018-07-09
| |
* | merging the CUDA and HIP implementation for the Tensor directory and the ↵Gravatar Deven Desai2018-06-20
| | | | | | | | unit tests
* | removing the *Hip files from the unsupported/Eigen/CXX11/src/Tensor and ↵Gravatar Deven Desai2018-06-20
| | | | | | | | unsupported/test directories
* | renaming *Cuda files to *Gpu in the unsupported/Eigen/CXX11/src/Tensor and ↵Gravatar Deven Desai2018-06-20
| | | | | | | | unsupported/test directories
* | updates based on PR feedbackGravatar Deven Desai2018-06-14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There are two major changes (and a few minor ones which are not listed here...see PR discussion for details) 1. Eigen::half implementations for HIP and CUDA have been merged. This means that - `CUDA/Half.h` and `HIP/hcc/Half.h` got merged to a new file `GPU/Half.h` - `CUDA/PacketMathHalf.h` and `HIP/hcc/PacketMathHalf.h` got merged to a new file `GPU/PacketMathHalf.h` - `CUDA/TypeCasting.h` and `HIP/hcc/TypeCasting.h` got merged to a new file `GPU/TypeCasting.h` After this change the `HIP/hcc` directory only contains one file `math_constants.h`. That will go away too once that file becomes a part of the HIP install. 2. new macros EIGEN_GPUCC, EIGEN_GPU_COMPILE_PHASE and EIGEN_HAS_GPU_FP16 have been added and the code has been updated to use them where appropriate. - `EIGEN_GPUCC` is the same as `(EIGEN_CUDACC || EIGEN_HIPCC)` - `EIGEN_GPU_DEVICE_COMPILE` is the same as `(EIGEN_CUDA_ARCH || EIGEN_HIP_DEVICE_COMPILE)` - `EIGEN_HAS_GPU_FP16` is the same as `(EIGEN_HAS_CUDA_FP16 or EIGEN_HAS_HIP_FP16)`
* | syncing this fork with upstreamGravatar Deven Desai2018-06-13
|\ \
| | * Fix oversharding bug in parallelFor.Gravatar Rasmus Munk Larsen2018-06-20
| |/
| * Merge from eigen/eigenGravatar Michael Figurnov2018-06-07
| |\
| | * Fix typos found using codespellGravatar Gael Guennebaud2018-06-07
| | |
| * | Derivative of the incomplete Gamma function and the sample of a Gamma random ↵Gravatar Michael Figurnov2018-06-06
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | variable. In addition to igamma(a, x), this code implements: * igamma_der_a(a, x) = d igamma(a, x) / da -- derivative of igamma with respect to the parameter * gamma_sample_der_alpha(alpha, sample) -- reparameterization derivative of a Gamma(alpha, 1) random variable sample with respect to the alpha parameter The derivatives are computed by forward mode differentiation of the igamma(a, x) code. Although gamma_sample_der_alpha can be implemented via igamma_der_a, a separate function is more accurate and efficient due to analytical cancellation of some terms. All three functions are implemented by a method parameterized with "mode" that always computes the derivatives, but does not return them unless required by the mode. The compiler is expected to (and, based on benchmarks, does) skip the unnecessary computations depending on the mode.
* | | Adding support for using Eigen in HIP kernels.Gravatar Deven Desai2018-06-06
|/ / | | | | | | | | | | | | | | | | This commit enables the use of Eigen on HIP kernels / AMD GPUs. Support has been added along the same lines as what already exists for using Eigen in CUDA kernels / NVidia GPUs. Application code needs to explicitly define EIGEN_USE_HIP when using Eigen in HIP kernels. This is because some of the CUDA headers get picked up by default during Eigen compile (irrespective of whether or not the underlying compiler is CUDACC/NVCC, for e.g. Eigen/src/Core/arch/CUDA/Half.h). In order to maintain this behavior, the EIGEN_USE_HIP macro is used to switch to using the HIP version of those header files (see Eigen/Core and unsupported/Eigen/CXX11/Tensor) Use the "-DEIGEN_TEST_HIP" cmake option to enable the HIP specific unit tests.
| * Performance improvements to tensor broadcast operationGravatar Vamsi Sripathi2018-05-23
|/ | | | | | 1. Added new packet functions using SIMD for NByOne, OneByN cases 2. Modified existing packet functions to reduce index calculations when input stride is non-SIMD 3. Added 4 test cases to cover the new packet functions
* Merged in mfigurnov/eigen (pull request PR-400)Gravatar Benoit Steiner2018-06-05
|\ | | | | | | | | | | Exponentially scaled modified Bessel functions of order zero and one. Approved-by: Benoit Steiner <benoit.steiner.goog@gmail.com>
* | Add a ThreadPoolInterface* getter for ThreadPoolDevice.Gravatar Penporn Koanantakool2018-06-02
| |
| * Exponentially scaled modified Bessel functions of order zero and one.Gravatar Michael Figurnov2018-05-31
|/ | | | | | The functions are conventionally called i0e and i1e. The exponentially scaled version is more numerically stable. The standard Bessel functions can be obtained as i0(x) = exp(|x|) i0e(x) The code is ported from Cephes and tested against SciPy.
* Hyperlink DOIs against preferred resolverGravatar Katrin Leinweber2018-05-24
|
* Merged in rmlarsen/eigen2 (pull request PR-393)Gravatar Benoit Steiner2018-05-16
|\ | | | | | | Rename scalar_clip_op to scalar_clamp_op to prevent collision with existing functor in TensorFlow.
| * Rename clip2 to clamp.Gravatar Rasmus Munk Larsen2018-05-16
| |
| * Rename scalar_clip_op to scalar_clip2_op to prevent collision with existing ↵Gravatar Rasmus Munk Larsen2018-05-16
| | | | | | | | functor in TensorFlow.
* | Merged in didierjansen/eigen (pull request PR-360)Gravatar Benoit Steiner2018-05-16
|\ \ | |/ |/| | | Fix bugs and typos in the contraction example of the tensor README
* | Add vectorized clip functor for Eigen Tensors.Gravatar Rasmus Munk Larsen2018-05-14
| |
* | Enable RawAccess to tensor slices whenever possinle.Gravatar Benoit Steiner2018-04-30
| | | | | | | | Avoid 32-bit integer overflow in TensorSlicingOp
* | Avoid using memcpy for non-POD elementsGravatar Weiming Zhao2018-04-11
| |
| * Fix typos in the contraction example of tensor READMEGravatar Lee.Deokjae2018-01-06
| |
* | Update the padding computation for PADDING_SAME to be consistent with ↵Gravatar Benoit Steiner2018-01-30
|\ \ | | | | | | | | | TensorFlow.
* | | Disable use of recurrence for computing twiddle factors. Fixes FFT precision ↵Gravatar RJ Ryan2017-12-31
| |/ |/| | | | | issues for large FFTs. https://github.com/tensorflow/tensorflow/issues/10749#issuecomment-354557689
* | Workaround nvcc 9.0 issue. See PR 351.Gravatar Gael Guennebaud2017-12-15
| | | | | | | | https://bitbucket.org/eigen/eigen/pull-requests/351
| * Update the padding computation for PADDING_SAME to be consistent with ↵Gravatar Yangzihao Wang2017-12-12
|/ | | | TensorFlow.
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-28 16:28:14 +0000