零、平臺(tái)介紹

????0、軟件平臺(tái)

????????操作系統(tǒng)：Ubuntu20.04?

????????Linux Kernel：5.8.4-050804-generic

????? ? NVCC（ROCm）：rocm-dev3.7.0/Ubuntu 16.04 3.7.0-20 amd64

????????Docker：docker-ce/focal,now 5:19.03.12~3-0~ubuntu-focal amd64

????1、硬件平臺(tái)

????????CPU：Ryzen 3700X

????????GPU：Radeon VII

????????主板：ROG STRIX?B450-I

????????內(nèi)存：3200MHz 2*16GB

????2、ROCm介紹

????????ROCm是AMD顯卡的開(kāi)放計(jì)算平臺(tái)，提供了接近Cuda的API和轉(zhuǎn)換工具。目前支持Pytorch和TensorFlow。

一、平臺(tái)搭建

????0、安裝ROCm

????????(1)更新到最新版本內(nèi)核

????????前往?https://kernel.ubuntu.com/~kernel-ppa/mainline/v5.8.4/?下載內(nèi)核文件 ：

linux-headers-5.8.4-050804-generic_5.8.4-050804.202008260637_amd64.deb
inux-headers-5.8.4-050804_5.8.4-050804.202008260637_all.deb
linux-image-unsigned-5.8.4-050804-generic_5.8.4-050804.202008260637_amd64.deb
linux-modules-5.8.4-050804-generic_5.8.4-050804.202008260637_amd64.deb
????????切換到文件目錄，執(zhí)行sudo apt install ./linux-*-5.8.4*.deb。

????????（2）安裝ROCm

????????添加軟件源

wget -q -O - http://repo.radeon.com/rocm/rocm.gpg.key | sudo apt-key add -

echo 'deb [arch=amd64] http://repo.radeon.com/rocm/apt/debian/ xenial main' | sudo tee /etc/apt/sources.list.d/rocm.list

????????安裝軟件

sudo apt install rocm-dev

echo 'SUBSYSTEM=="kfd", KERNEL=="kfd", TAG+="uaccess", GROUP="video"' | sudo tee /etc/udev/rules.d/70-kfd.rules

????????配置環(huán)境和權(quán)限

sudo usermod -a -G video $LOGNAME?

sudo usermod -a -G render $LOGNAME

echo 'ADD_EXTRA_GROUPS=1' | sudo tee -a /etc/adduser.conf

echo 'EXTRA_GROUPS=video' | sudo tee -a /etc/adduser.conf

echo 'EXTRA_GROUPS=render' | sudo tee -a /etc/adduser.conf

echo 'export PATH=$PATH:/opt/rocm/bin:/opt/rocm/profiler/bin:/opt/rocm/opencl/bin' | sudo tee -a /etc/profile.d/rocm.sh

????????（3）安裝docker

????????為了加速安裝，教程使用阿里鏡像站

curl -fsSL https://mirrors.aliyun.com/docker-ce/linux/ubuntu/gpg | sudo apt-key add -

sudo add-apt-repository "deb [arch=amd64] https://mirrors.aliyun.com/docker-ce/linux/ubuntu $(lsb_release -cs) stable"

sudo apt?update

sudo apt-get?install docker-ce

????????安裝Pytorch和TersonFlow鏡像

sudo?docker pull rocm/pytorch:latest

sudo?docker pull rocm/tensorflow:latest

二、調(diào)用Pytorch和TensorFlow

sudo docker run -it -v $HOME:/data --privileged --rm --device=/dev/kfd --device=/dev/dri --group-add video rocm/pytorch:latest

sudo docker run -it -v $HOME:/data --privileged --rm --device=/dev/kfd --device=/dev/dri --group-add video rocm/tensorflow:latest