• 3
• 0
• 0

AITemplate

AITemplate is a Python framework which renders neural network into high performance CUDA/HIP C++ code. Specialized for FP16 TensorCore (NVIDIA GPU)...

• 3
• 0
• 0

libfabric

Open Fabric Interfaces

• 3
• 0
• 0

pytorch_scatter

PyTorch Extension Library of Optimized Scatter Operations

• 3
• 0
• 0

MITuna

• 3
• 0
• 0

TransferBench

TransferBench is a utility capable of benchmarking simultaneous copies between user-specified devices (CPUs/GPUs)

• 3
• 0
• 0

pytorch-lightning

The lightweight PyTorch wrapper for high-performance AI research. Scale your models, not the boilerplate.

• 3
• 0
• 0

recommenders-addons

Additional utils and helpers to extend TensorFlow when build recommendation systems, contributed and maintained by SIG Recommenders.

• 3
• 0
• 0

rtg_tracer

• 3
• 0
• 0

kineto

A CPU+GPU Profiling library that provides access to timeline traces and hardware performance counters.

• 3
• 0
• 0

rocWMMA

rocWMMA

• 3
• 0
• 0

sukha-tools

Profiling tools intended for Nvidia or AMD GPUs

• 3
• 0
• 0

triton

Development repository for the Triton language and compiler

• 3
• 0
• 0

hipify_torch

• 3
• 0
• 0

composable_kernel_remove

Composable Kernel: Performance Portable Programming Model for Machine Learning Tensor Operators

• 3
• 0
• 0

gpufort

GPUFORT: S2S translation tool for CUDA Fortran and Fortran+X in the spirit of hipify

• 2
• 0
• 0

tensorflow-build

Build-related tools for TensorFlow

• 2
• 0
• 0

aws-ofi-rccl

• 2
• 0
• 0

rocGemmDriver

rocGemmDriver

• 2
• 0
• 0

audio

Data manipulation and transformation for audio signal processing, powered by PyTorch

• 2
• 0
• 0

hipOMB

OSU MPI benchmarks with ROCm support