keras安装步骤的ppt 1、ANACONDA 安装 2、Cuda及cuDNN安装 3、Tensorflow-gpu版本安装 4、Keras安装 5、Anaconda的使用 6、Keras分类示例
2024-03-09 14:14:36 4.3MB tensorflow tensorflow anaconda keras
1
英伟达开发者社区免费课程NVIDIA GPU
2024-02-21 14:22:58 73.7MB
1
(1)嵌入式系统-linux (2)使用tvm的opencl后端调用mali-gpu (3)rk3588的mali-gpu安装包:G610
2024-02-20 15:37:00 12.04MB 人工智能 深度学习
1
本文介绍了使用pytorch2.0进行图像分类的实战案例,包括数据集的准备,卷积神经网络的搭建,训练和测试的过程,以及模型的保存和加载。本案例使用了CIFAR-10数据集,包含10个类别的彩色图像,每个类别有6000张图像,其中5000张用于训练,1000张用于测试。本案例使用了一个简单的卷积神经网络,包含两个卷积层和两个全连接层,使用ReLU激活函数和交叉熵损失函数,使用随机梯度下降优化器。本案例可以在GPU和CPU上运行,根据设备的不同自动切换。本案例适合入门pytorch深度学习和练手,也可以用到项目当中。代码精炼,容易修改进行二次完善和开发。
2024-01-16 14:08:43 325.06MB pytorch 数据集 计算机视觉
1
本资源为Vitis-AI3.0版本docker镜像的.tar文件的下载链接,适用于NVIDIA显卡硬件平台,内置pytorch量化编译器镜像以及pytorch优化器镜像 使用方法: 使用docker load指令将镜像文件导入后(导入后可以使用docker tag指令改名),再按照官方手册中的使用方法即可 docker镜像生成过程: 按照官方github提供的3.0.0.001版本源代码中的dockerfile进行docker创建 以NVIDIA提供的nvidia/cuda:11.3.1-cudnn8-runtime-ubuntu20.04镜像为基础 仅修改apt、python、conda为国内下载源,其他未作变动 dockerfile修改内容: 参考文章https://blog.csdn.net/qq_36745999/article/details/129920225
2024-01-15 17:47:55 78B pytorch docker Vitis-AI Vitis
1
本研究尝试通过使用最新的多核多CPU系统在陡峭的三维隔离山上进行大涡模拟(LES)。 结果,发现1)使用大约5000万个网格点进行湍流模拟是可行的; 2)使用该系统导致了很高的计算速度,超过了单个CPU所达到的并行计算速度在最新的超级计算机之一上。 此外,LES是通过使用多GPU系统进行的。 这些仿真的结果揭示了以下发现:1)使用NVDIA:registered:Tesla M2090或M2075的多GPU环境可以在多达约5000万个网格点的模型中模拟湍流。 2)多GPU环境实现的计算速度超过了并行计算的速度,并行计算使用的是最新超级计算机之一的4至6个CPU。
2024-01-11 12:00:10 3.41MB 多核多CPU计算 多GPU计算
1
Delve into the Broadcom VideoCore GPU used on the Raspberry Pi and master topics such as OpenGL ES and OpenMAX. Along the way, you’ll also learn some Dispmanx, OpenVG, and GPGPU programming. The author, Jan Newmarch bumped into a need to do this kind of programming while trying to turn the RPi into a karaoke machine: with the CPU busting its gut rendering MIDI files, there was nothing left for showing images such as karaoke lyrics except for the GPU, and nothing really to tell him how to do it. Raspberry Pi GPU Audio Video Programming scratches his itch and since he had to learn a lot about RPi GPU programming, he might as well share it with you. What started as a side issue turned into a full-blown project of its own; and this stuff is hard. What You'll Learn Use Dispmanx and EGL on Raspberry Pi Work with OpenMAX and its components, state, IL Client Library, * * Buffers, and more on RPi Process images and video on RPi Handle audio on RPi Render OpenMAX to OpenGL on the RPi Play multimedia files on the RPi Use OpenVG for text processing and more Master overlays Who This Book Is For You should be comfortable with C programming and at least some concurrency and thread programming using it. This book is for experienced programmers who are new or learning about Raspberry Pi. Table of Contents Chapter 1: Introduction to the Raspberry Pi Chapter 2: Khronos Group Chapter 3: Compiling Programs for the Raspberry Pi Chapter 4: Dispmanx on the Raspberry Pi Chapter 5: EGL on the Raspberry Pi Chapter 6: OpenGL ES on the Raspberry Pi Chapter 7: OpenMAX on the Raspberry Pi Concepts Chapter 8: OpenMAX Components on the Raspberry Pi Chapter 9: OpenMAX on the Raspberry Pi State Chapter 10: OpenMAX IL Client Library on the Raspberry Pi Chapter 11: OpenMAX Buffers on the Raspberry Pi Chapter 12: Image Processing on the Raspberry Pi Chapter 13: OpenMAX Video Processing on the Raspberry Pi Chapter 14: OpenMAX Audio on the Raspberry Pi Chapter 15: Rendering OpenMAX to OpenGL on the Raspberry Pi Chapter 16: Playing Multimedia Files on the Raspberry Pi Chapter 17: Basic OpenVG on the Raspberry Pi Chapter 18: Text Processing in OpenVG on the Raspberry Pi Chapter 19: Overlays on the Raspberry Pi
2023-12-21 12:01:36 3.71MB Raspberry GPU Programming
1
SIFT-GPU A CUDA implementation of SIFT: 配置 (待完成)见SIFT-GPU配置教程 测试 1.Release 模式 cd bin ./SimpleSIFT.exe Release模式,输出结果: [GPU VENDOR]: NVIDIA Corporation 1717MB TEXTURE: 16384 Image size : 800x600 Image loaded : ../data/800-1.jpg Features: 3358 Features MO: 3923 Image size : 640x480 Image loaded : ../data/640-1.jpg Features: 2383 Features MO: 2791 2279 sift matches were
2023-12-15 09:39:08 43.14MB gpu cuda sift
1
安卓下使用OpenCL进行GPU编程,测试平台为Nokia N1平板,GPU为PowerVR
2023-12-14 11:34:48 1.8MB OpenCL
1
openmm, OpenMM是一种使用高性能GPU代码进行分子模拟的工具包 : 高性能分子动态库简介OpenMM 是一个用于分子模拟的工具包。 它可以以用作运行模拟的独立应用程序,也可以以作为你自己的代码调用的库。 它提供了极端灵活性( 通过自定义力和积分器) 。开放性和高性能( 尤其是最近的gpu
2023-10-31 17:18:11 16.13MB
1