LLM training in simple, raw C/CUDA

Fast parallel CTC.

Squeeze-and-Excitation Networks

This package contains the original 2012 AlexNet code.

how to optimize some algorithm in cuda.

Sample codes for my CUDA programming book

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

MatConvNet: CNNs for MATLAB

RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing …

Examples demonstrating available options to program multiple GPUs in a single node or a cluster

Automatically exported from code.google.com/p/cuda-convnet2

UNet diffusion model in pure CUDA

Causal depthwise conv1d in CUDA, with a PyTorch interface

A GPU implementation of Convolutional Neural Nets in C++

CUDA Learning guide

Unsupervised Learning of Video Representations using LSTMs

llama3.cuda is a pure C/CUDA implementation for Llama 3 model.

Alex Krizhevsky's original code from Google Code

CUDA Matrix Factorization Library with Alternating Least Square (ALS)

Some CUDA example code with READMEs.

Wang Yi's GPT solution

LLM training in simple, raw C/CUDA

Deep Object Co-Segmentation

CUDA extension for the SPORCO project

This project optimizes multi-GPU parallelism for machine learning training by accelerating multi-GPU using fused gradient buffers, NCCL AllReduce, and CUDA C kernel-level optimizations including me…