The last article talked about how to use Ubuntu 18 04 install nvidia graphics card driver, then install cuda and cudnn together in the environment where the graphics card driver is installed. The workload is also a lot. I unloaded it several times. The complete collection of deep learning environment and pit avoidance is coming!!!!
Added by nedpwolf on Mon, 07 Mar 2022 22:30:13 +0200
First, you need to install two necessary packages tensorrt and torch2trt. If tensorrt, you need to download the tar compressed package on the official website. It is recommended to download the tar package for installation, Official website , I downloaded version 7.2.3. torch2trt can clone projects on GitHub. My environment (tensorrt seems to w ...
Added by Sk8Er_GuY on Fri, 18 Feb 2022 18:51:22 +0200
Standard two body model implemented with CUDA Thrust
The space target orbit parallel computing technology based on CUDA has four sections, of which the contents of the first, second and third sections are as follows
1. Orbit calculation requirements and mission analysis of space target based on CUDA
2 Calculation of spatial coordinate syste ...
Added by mclamais on Thu, 06 Jan 2022 02:22:34 +0200
Chapter 5 thread execution efficiency and SIMD
Single instruction multiple data execution (SIMD) when a warp thread executes; threads in warp execute the same command at any timeControl divergence: occurs when threads in warp adopt different control paths through different control decisions. Threads adopting different control paths will eventu ...
Added by Threehearts on Wed, 29 Dec 2021 23:22:59 +0200
I want to be here c file using cuda's function, that is Content of cu
Installing nvcc is not the content here, but make sure that nvcc can be used, which is to ensure that it can be compiled On the premise of cu, view the version of nvcc, and the command is as follows
The output is ...
Added by lhcpr on Tue, 21 Dec 2021 15:43:03 +0200
Before installation, please explain a few points. Because you are not familiar with the installation method of linux, you can see that there are many installation methods, including source code compilation, dpkg, apt, etc. we all know the source code method here, compile through Makefile or cmake, and then install. The process is complex and er ...
Added by Fantast on Tue, 07 Dec 2021 20:56:40 +0200
In order to better illustrate how to realize the mixed programming of C + + and CUDA in Linux, I will next realize the modular operation of each element of a matrix.
1. Header file and file form
To write CUDA code in C + +, you need to import header files:
#include "device_launch_parameters.h" ...
Added by Shawnaize on Sat, 09 Oct 2021 12:45:06 +0300