dlprimitives
ParallelReductionsBenchmark
dlprimitives | ParallelReductionsBenchmark | |
---|---|---|
7 | 2 | |
156 | 59 | |
- | - | |
3.8 | 4.6 | |
5 months ago | 5 months ago | |
C++ | C++ | |
MIT License | - |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
dlprimitives
- Dlprimitives: Deep Learning Primitives and Mini-Framework for OpenCL
-
[P] OpenCL backend for PyTorch - progress works with mainstream pytorch
I'm working on PyTorch OpenCL backend based on dlprimitives core library. It exists for a while but until now it required building custom pytorch version.
-
[P] DLPrimitives - wondering about best development direction
BTW Performance numbers: https://github.com/artyom-beilis/dlprimitives/blob/master/docs/benchmarks/benchmarks-gtx1080.md (I just added below TF2 that is missing in docs)
-
[P] DLPrimitives - an OpenCL miro-framework and inference library
Full benchmarks can be found there: https://github.com/artyom-beilis/dlprimitives/blob/master/docs/summary.md
- [P] OpenCL Deep Learning Primitives Library
ParallelReductionsBenchmark
-
Failing to Reach 204 GB/S DDR4 Bandwidth
For the single threaded version, they have a data hazard on the sums that could be smoothed out with a little loop unrolling and separate variables.
But in the [threaded version](https://github.com/unum-cloud/ParallelReductions/blob/fd16d9...) they have separate slots for an accumulator but it's still in a shared vector, which most likely has the issue I described.
What are some alternatives?
tensorflow-opencl - OpenCL support for TensorFlow
MatX - An efficient C++17 GPU numerical computing library with Python-like syntax
plaidml - PlaidML is a framework for making deep learning work everywhere.
ispc - IntelĀ® Implicit SPMD Program Compiler
AdaptiveCpp - Implementation of SYCL and C++ standard parallelism for CPUs and GPUs from all vendors: The independent, community-driven compiler for C++-based heterogeneous programming models. Lets applications adapt themselves to all the hardware in the system - even at runtime!
gpuowl - GPU Mersenne primality test.
pytorch-coriander - OpenCL build of pytorch - (in-progress, not useable)
alpaka - Abstraction Library for Parallel Kernel Acceleration :llama:
oneDNN - oneAPI Deep Neural Network Library (oneDNN)
cuda_memtest - Fork of CUDA GPU memtest :eyeglasses:
pytorch_dlprim - DLPrimitives/OpenCL out of tree backend for pytorch
eaminer - Heterogeneous Ethereum Miner with support for AMD, Intel and Nvidia GPUs using SYCL, OpenCL and CUDA backends