openai / blocksparse
Efficient GPU kernels for block-sparse matrix multiplication and convolution
☆1,036Updated last year
Alternatives and similar repositories for blocksparse:
Users that are interested in blocksparse are comparing it to the libraries listed below
- FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/☆1,264Updated this week
- Collective communications library with various primitives for multi-machine training.☆1,266Updated this week
- A domain specific language to express machine learning workloads.☆1,757Updated last year
- ATen: A TENsor library for C++11☆694Updated 5 years ago
- Fast Block Sparse Matrices for Pytorch☆546Updated 4 years ago
- TVM integration into PyTorch☆452Updated 5 years ago
- The Tensor Algebra SuperOptimizer for Deep Learning☆696Updated 2 years ago
- Benchmarking Deep Learning operations on different hardware☆1,081Updated 3 years ago
- A GPU performance profiling tool for PyTorch models☆501Updated 3 years ago
- Tutorial for building a custom CUDA function for Pytorch☆513Updated 6 years ago
- A GPipe implementation in PyTorch☆832Updated 7 months ago
- PyTorch elastic training☆729Updated 2 years ago
- Reference implementations of MLPerf™ training benchmarks☆1,647Updated 3 weeks ago
- Mesh TensorFlow: Model Parallelism Made Easier☆1,603Updated last year
- Open single and half precision gemm implementations☆376Updated last year
- ☆577Updated 6 years ago
- common in-memory tensor structure☆952Updated last week
- Examples of using sparse attention, as in "Generating Long Sequences with Sparse Transformers"☆1,552Updated 4 years ago
- Quantized Neural Network PACKage - mobile-optimized implementation of quantized neural network operators☆1,536Updated 5 years ago
- Library for faster pinned CPU <-> GPU transfer in Pytorch☆685Updated 5 years ago
- PyTorch implementation of the Quasi-Recurrent Neural Network - up to 16 times faster than NVIDIA's cuDNN LSTM☆1,259Updated 3 years ago
- Implements an efficient softmax approximation as described in the paper "Efficient softmax approximation for GPUs" (http://arxiv.org/abs/…☆394Updated 5 years ago
- Block-sparse primitives for PyTorch☆153Updated 3 years ago
- A library of GPU kernels for sparse matrix operations.☆259Updated 4 years ago
- Low-precision matrix multiplication☆1,792Updated last year
- Experimental ground for optimizing memory of pytorch models☆366Updated 6 years ago
- Running BERT without Padding☆470Updated 2 years ago
- ☆373Updated 7 years ago
- Training RNNs as Fast as CNNs (https://arxiv.org/abs/1709.02755)☆2,100Updated 3 years ago
- nGraph has moved to OpenVINO☆1,350Updated 4 years ago