LLM training in simple, raw C/CUDA

[ARCHIVED] Cooperative primitives for CUDA C++. See https://github.com/NVIDIA/cccl

how to optimize some algorithm in cuda.

[MICRO'23, MLSys'22] TorchSparse: Efficient Training and Inference Framework for Sparse Convolution on GPUs.

Efficient GPU kernels for block-sparse matrix multiplication and convolution

CUDA Kernel Benchmarking Library

Distributed multigrid linear solver library on GPU

Fast CUDA matrix multiplication from scratch

Chamfer Distance in Pytorch with f-score

Step-by-step optimization of CUDA SGEMM

CUDA implementation of the fundamental sum reduce operation. Aims to be as optimized as reasonable.

CUDA Program to perform Logistic regression on any CSV data files with Training output in 2nd Coloumn.

Using CUDA to calculate normals given a n-n matrix. The lenguage of the comments is Spanish.