ksopyla / CudaDotProd
Different implementation of sparse matrix multiplication. All matrices are in CSR format. The code contains different CUDA kernels for multiply sparse matrix vs dense vector and sparse matrix vs another sparse matrix. It contains several cuda kernel for sparse matrix dense vector product and sparse matrix sparse matrix product.
☆16Updated 14 years ago
Alternatives and similar repositories for CudaDotProd:
Users that are interested in CudaDotProd are comparing it to the libraries listed below
- a heterogeneous multiGPU level-3 BLAS library☆45Updated 5 years ago
- CUDA Matrix Factorization Library with Stochastic Gradient Descent (SGD)☆71Updated 7 years ago
- CUDA Sparse-Matrix Vector Multiplication, using Sliced Coordinate format☆21Updated 6 years ago
- Kernel Fusion and Runtime Compilation Based on NNVM☆70Updated 8 years ago
- High Efficiency Convolution Kernel for Maxwell GPU Architecture☆134Updated 7 years ago
- sparse matrix pre-processing library☆81Updated 9 months ago
- A simple memory manager for CUDA designed to help Deep Learning frameworks manage memory☆296Updated 6 years ago
- Efficient LDA solution on GPUs.☆24Updated 6 years ago
- Medusa: Building GPU-based Parallel Sparse Graph Applications with Sequential C/C++ Code☆61Updated 4 years ago
- The Surprisingly ParalleL spArse Tensor Toolkit.☆70Updated 2 years ago
- This repository contains the cuStinger data structure used for dynamic graph representation.☆19Updated 6 years ago
- Full-speed Array of Structures access☆164Updated last year
- GPU-specialized parameter server for GPU machine learning.☆100Updated 6 years ago
- Sparse matrix computation library for GPU☆54Updated 4 years ago
- Greentea LibDNN - a universal convolution implementation supporting CUDA and OpenCL☆135Updated 7 years ago
- Simple MXNet sequence-to-sequence model (neural machine translation)☆24Updated 7 years ago
- Matrix Shadow:Lightweight CPU/GPU Matrix and Tensor Template Library in C++/CUDA for (Deep) Machine Learning☆33Updated 8 years ago
- CSR5-based SpMV on CPUs, GPUs and Xeon Phi☆102Updated 8 months ago
- ☆93Updated 8 years ago
- CuSha is a CUDA-based vertex-centric graph processing framework that uses G-Shards and CW representations.☆52Updated 9 years ago
- CSR-based SpMV on Heterogeneous Processors (Intel Broadwell, AMD Kaveri and nVidia Tegra K1)☆26Updated 9 years ago
- Benchmarking State-of-the-Art Deep Learning Software Tools☆170Updated 7 years ago
- CLTune: An automatic OpenCL & CUDA kernel tuner☆173Updated 2 years ago
- Library for fast image convolution in neural networks on Intel Architecture☆29Updated 7 years ago
- Machine Learning Toolkit for Extreme Scale (MaTEx)☆111Updated 6 years ago
- The Jacobi-type (hyperbolic) SVD for CUDA.☆10Updated 5 months ago
- Benchmarks for CNTK and other toolkits.☆44Updated 9 years ago
- flexible-gemm conv of deepcore☆17Updated 5 years ago
- Optimized half precision gemm assembly kernels (deprecated due to ROCm)☆47Updated 7 years ago
- profiling gemm on android☆10Updated 8 years ago