Search Results for "cuda"

Numbast is an automated toolchain that bridges CUDA C++ and Python by generating Numba-compatible bindings directly from CUDA header files. Its primary goal is to eliminate the manual effort required to expose CUDA libraries to Python, enabling developers to use GPU-accelerated functionality in Python environments more easily. The system parses CUDA C++ declarations and converts them into Python bindings that can be used within Numba, allowing seamless integration with Python-based GPU workflows. ...

The official SuiteSparse library: a suite of sparse matrix algorithms authored or co-authored by Tim Davis, Texas A&M University.

...With GPU acceleration via CUDA and OpenCV integration, it achieves high performance in image recognition tasks. Its simplicity, combined with powerful capabilities, has made Darknet one of the most influential projects in the computer vision community.

...Unlike the standard PyTorch-based Real-ESRGAN code, this variant is written in C/C++ and designed to run efficiently on many platforms (including Windows, Linux, and possibly Android) without requiring heavy frameworks like CUDA or Python. It provides command-line tools for upscaling images with selected models, allowing users to specify input/output paths, scaling factors, tile sizes, and model names from a compressed model set, which is particularly helpful for larger images or automated workflows. The Vulkan backend enables fast execution on GPUs from different vendors (Intel/AMD/Nvidia) with broad support, making it suitable for non-Python environments, production systems, or performance-constrained setups.

...The objective of this project (PMCGPU) is to bring together all the researchers working on the development of parallel simulators for P systems, specially those using the GPU (e.g. CUDA, OpenCL, etc). Other parallel platforms are also welcome (multicore and manycore, FPGAs, etc). This project has been initiated by the Research Group on Natural Computing (Department of Computer Science and Artificial Intelligence, University of Seville). PMCGPU was born inside the P-Lingua project, of the same research group. ...

HIPAcc development has moved to github: https://github.com/hipacc HIPAcc allows to design image processing kernels and algorithms in a domain-specific language (DSL). From this high-level description, low-level target code for GPU accelerators is generated using source-to-source translation. As back ends, the framework supports CUDA, OpenCL, and Renderscript. HIPAcc allows programmers to develop imaging applications while providing high productivity, flexibility and portability as well as competitive performance: the same algorithm description serves as basis for targeting different GPU accelerators and low-level languages.

cuLsoda is an implementation of the Livermore Solver for Ordinary Differential Equations with Automatic switching for stiff and non-stiff problems (LSODA), written in NVIDIA's C for CUDA GPGPU API. NOTE: This repository is now deprecated and replaced by a new one at GitHub: https://github.com/Celemourn/cuLsoda

hiCUDA is a directive-based abstraction that simplifies CUDA programming. This project aims to develop a source-to-source compiler, based on Open64, that translates a sequential program with hiCUDA directives into an equivalent CUDA program.

The Open64 Compiler and Tools site is dedicated to the continued development of the former SGI Pro64(TM) compiler for the IA64, x86, CUDA and MIPS architecture.

Fossa is a runtime CUDA memory extender for Linux. With fossa, users can run CUDA applications that require more than the available amount of GPU memory. Fossa is targeted toward command line driven applications.

Allows CUDA developers to deploy kernels requiring more global GPU memory than is available by transparently mapping pinned CPU memory into the GPU memory map. Works behind the scenes and is easily added to existing projects with minimal effort.