搜索资源列表
NCUDDA_BP_neuv
- 一种基于nvidia CUDA架构的BP神经网络程序源码,在G80,G992 GPU上能完成BP神经网络训练。速度较双核CPU提高十倍 可直接使用。 -Based on the completion of the BP neural network training on the G80, G992 GPU nvidia CUDA architecture BP neural network program source code. Faster than dual-core CPU to
cuda3.0Programingguid
- cuda3.0编程指南,由风尘大牛翻译编写,适合初学CUDA者使用-CUDA3.0 programing guid
CUDA4.0-Function-library
- GPU并行开发,要先安装cuda工具包,再在visual studio2005上可以运行。里面有CUBLAS_Library,CUBLAS_Library函数库,还有一些编程书籍,优化。-GPU parallel development, first install cuda toolkit can run visual studio2005. CUBLAS_Library, CUBLAS_Library library, there are some programming books, an
ECUDAzipa
- 深入浅出谈CUDA——一种新新的GPU编程工具可直接使用。 -Layman' s language to talk about CUDA- A new GPU programming tools can be used directly.
gpu11
- use CUDA to develop a GPU computing program. The program solves a laplacian equation using finite-difference method.
cuda-latex-sandcastle-fff0ec6
- A LaTeX Build Component for the SandCastle Help File Builder. Uses mimeTeX 1.7.4 to parse LaTeX code.
queuing_model
- queuing model on CUDA
programa18p
- implementation of a genetic algorithm in CUDA/C++ (parallelized)
programa18
- implementation of a genetic algorithm in CUDA/C++ (not parallelized!)
programa17p
- generic genetic algorithm in CUDA C++
example
- 使用C++和CUDA编写的双边滤波器。在平滑图像的同时可以保持边界-Using bilateral filter the C++ and CUDA written. Boundary can be kept in the smooth image while
sort
- 基于cuda的并行高效排序算法,并且计算了是否使用shared内存的不同效率-Cuda parallel efficient sorting algorithms, and calculate whether to use shared memory efficiency
Matrix-multiplication
- 采用共享内存的矩阵乘法的CUDA实现,是一种并行实现方式-Matrix multiplication
3
- 纹理存储器的使用例子,实现1D和2D的cuda使用-texture application
vincentfpgarcia-kNN-CUDA-70dc929
- KNN algorithm source code which allows you to find the k nearest neighbour.
simpleCUFFT
- 利用CUDA实现FFT,速度达到标准,完整版-With CUDA implementation the FFT, speed up to standard, full version
laplace2d
- Laplace 定律用CUDA實做 拉普拉斯變換是工程數學中常用的一種積分變換,又名拉氏轉換。拉氏變換是一個線性變換,可將一個有引數實數t(t ≥ 0)的函數轉換為一個引數為複數s的函數。- Laplace implement in CUDA The Laplace transform is a widely used integral transform with many applications in physics and engineering.
matrixMul_texture
- CUDA的Texture Memory實作 並有詳細住解-Texture Memory of CUDA implementation and live solution
multiStream_v1
- CUDA的mutiStream的運用及詳細住解-The CUDA mutiStream use and detailed live solution
merge
- CUDA的Merge Sort(Old and even)實作及詳細住解-CUDA of MERGE the Sort (Old and even) the solution implementation and detailed live