昇腾训练营应用营第一课笔记
第一步:购买ECS弹性云服务器
计费方式选择按需计费,CPU架构选择:鲲鹏计算,资源选择:ai1s.large.4,操作系统选择Ubuntu 18,硬盘40GB应该是足够了。选择流量的位置,建议选择按流量计费,这样带宽无论选择多少,价格都差不多。
第二步:使用mobax登录到ECS远程服务器
第三步:环境准备
1.修改HwHiAiUser密码
passwd HwHiAiUser
2.将python3.7.5设置为默认python环境
rm -rf /usr/bin/python
ln -sf /usr/bin/python3.7.5 /usr/bin/python
说明:需要把原来的连接删除才可以连接新的,接下去我们在使用python命令出来的就是python3.7.5的环境了
3.修改HwHiAiUser默认环境为/bin/bash并且切换
vim /etc/passwd
将文件最后一行HwHiAiUser最后的/bin/sh修改为/bin/bash
4.下载环境包
一共有四个,Ascend驱动,CANN_NNRT.CANN_TOOLKIT,MindStudio
退出ROOT用户,使用HwHiAiUser登录进行操作
打开 https://ascend.huawei.com/#/software/cann/community 找到相关的驱动包下载地址。
mkdir download
cd download
wget https://repo.huaweicloud.com/ascend/autoarchive/CANN/3.3.0.alpha001/Ascend-cann-toolkit_3.3.0.alpha001_linux-x86_64.run --no-check-certificate
wget https://repo.huaweicloud.com/ascend/autoarchive/CANN/3.3.0.alpha001/Ascend-cann-nnrt_3.3.0.alpha001_linux-x86_64.run --no-check-certificate
wget https://obs-9be7.obs.cn-east-2.myhuaweicloud.com/turing/resourcecenter/Software/20.2.0/A300-3010%201.0.9/A300-3010-npu-driver_20.2.0_ubuntu18.04-x86_64.run --no-check-certificate
wget https://obs-9be7.obs.cn-east-2.myhuaweicloud.com/turing/resource/mindstudio/2.0.0-beta3/MindStudio_2.0.0-beta3_linux.tar.gz --no-check-certificate
切换到ROOT用户修改UBUNTU APT源为华为云镜像站
sed -i "s@http://.*archive.ubuntu.com@http://repo.huaweicloud.com@g" /etc/apt/sources.list
sed -i "s@http://.*security.ubuntu.com@http://repo.huaweicloud.com@g" /etc/apt/sources.list
apt-get update
安装依赖库
apt-get install -y gcc g++ make cmake zlib1g zlib1g-dev libbz2-dev openssl libsqlite3-dev libssl-dev libxslt1-dev libffi-dev unzip pciutils net-tools libblas-dev gfortran libblas3 libopenblas-dev libncursesw5-dev
如果遇到error报错的话重新安装一下python3.7就好
apt-get install python3.7
切换回HwHiAiUser用户安装python库
pip3.7.5 install attrs numpy decorator sympy==1.4 cffi==1.12.3 pyyaml pathlib2 psutil protobuf scipy requests xlrd==1.2.0 --user -i https://mirrors.huaweicloud.com/repository/pypi/simple
切换到root用户将所有的安装包增加可执行权限
cd /home/HwHiAiUser/download
chmod a+x ./A*
使用root用户安装NPU驱动
./A300-3010-npu-driver_20.2.0_ubuntu18.04-x86_64.run --full
安装完成后重启,使用HwHiAiUser设置环境变量
vim ~/.bashrc
文件内容添加
#用于设置python3.7.5库文件路径
export LD_LIBRARY_PATH=/usr/local/python3.7.5/lib:$LD_LIBRARY_PATH
#如果用户环境存在多个python3版本,则指定使用python3.7.5版本
export PATH=/usr/local/python3.7.5/bin:$PATH
然后继续安装njnrt
./Ascend-cann-nnrt_3.3.0.alpha001_linux-x86_64.run --install
将这一段加入到bashrc文件
export LD_LIBRARY_PATH=/home/HwHiAiUser/Ascend/nnrt/latest/acllib/lib64:$LD_LIBRARY_PATH
export PYTHONPATH=/home/HwHiAiUser/Ascend/nnrt/latest/pyACL/python/site-packages/acl:$PYTHONPATH
export ASCEND_AICPU_PATH=/home/HwHiAiUser/Ascend/nnrt/latest
安装Toolkit
./Ascend-cann-toolkit_3.3.0.alpha001_linux-x86_64.run --install
离线、在线、模型转换三种场景对应的环境变量:
Offline infer develop (cannot coexist with other scenes):
PATH includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/toolkit/bin:
LD_LIBRARY_PATH includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/acllib/lib64:
PYTHONPATH includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/pyACL/python/site-packages/acl:
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/toolkit/python/site-packages:
ASCEND_AICPU_PATH includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest:
ASCEND_OPP_PATH includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/opp:
TOOLCHAIN_HOME includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/toolkit:
Third party AI framework develop (cannot coexist with other scenes):
PATH includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/fwkacllib/ccec_compiler/bin:
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/fwkacllib/bin:
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/toolkit/bin:
LD_LIBRARY_PATH includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/fwkacllib/lib64:
PYTHONPATH includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/fwkacllib/python/site-packages:
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/toolkit/python/site-packages:
ASCEND_OPP_PATH includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/opp:
ASCEND_AICPU_PATH includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest:
TOOLCHAIN_HOME includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/toolkit:
Model transformation and operator develop (cannot coexist with other scenes):
PATH includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/atc/ccec_compiler/bin:
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/atc/bin:
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/toolkit/bin:
LD_LIBRARY_PATH includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/atc/lib64:
PYTHONPATH includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/atc/python/site-packages:
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/toolkit/python/site-packages:
ASCEND_OPP_PATH includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/opp:
TOOLCHAIN_HOME includes :
/home/HwHiAiUser/Ascend/ascend-toolkit/latest/toolkit:
安装MindStudio
# 解压mindstudio
tar -xvf MindStudio_2.0.0-beta3_linux.tar.gz
# 运行
MindStudio/bin/MindStudio.sh
这一步可能会一段报错,根据报错使用root重新安装了一下库就可以解决了
apt-get -y install xterm firefox xdg-utils fonts-droid-fallback fonts-wqy-zenhei fonts-wqy-microhei fonts-arphic-ukai fonts-arphic-uming gnome-keyring
apt-get -y install openjdk-8-jdk #非必须看是否会提示缺这个库
使用HwHiAiUser用户安装pip库
pip3 install --user coverage gnureadline pylint matplotlib pandas absl-py -i https://mirrors.huaweicloud.com/repository/pypi/simple
安装完成后再次启动MindStudio
这一步可以直接OK
看到这个界面就说明成功了
进行到这边的时候在PATH位置选择一下toolkit的目录,然后我们新建一个项目
然后项目名字随便,直接下一步到如下界面
这里选择【Asecend App】然后下一步,跟着图选Resnet50,然后完成
根据https://gitee.com/HuaweiAscend/models/tree/master/computer_vision/classification/resnet50 下载原始模型文件
打开MindStudio的终端输入一下命令下载模型文件
mkdir model
cd model
wget https://gitee.com/HuaweiAscend/models/blob/master/computer_vision/classification/resnet50/resnet50.prototxt
wget https://obs-model-ascend.obs.cn-east-2.myhuaweicloud.com/resnet50/resnet50.caffemodel
如果prototxt文件下载有困难的,直接自己VIM新增一个文件,内容如下
下载完成后,回到MindStudio,选择菜单栏【Ascend】-【Model Conver】进行模型转换,设置如下图
后面直接一路下一步即可
出现这一步就说明成功了,回到界面,选择项目右键【Add Model】
模型文件选择前面模型转换时候输出的文件,基本在modelzoo目录下。
打开MindStudio终端,下载推理所需的图片并且执行图片处理程序:
wget https://c7xcode.obs.cn-north-4.myhuaweicloud.com/models/aclsample/dog1_1024_683.jpg
wget https://c7xcode.obs.cn-north-4.myhuaweicloud.com/models/aclsample/dog2_1024_683.jpg
python ../script/transferPic.py
运行成功后,执行菜单栏里的【build】-【edit config....】编辑配置
然后直接点【build】开始编译
看到sucessful就说明编译成功了,然后在菜单栏找到【RUN】,然后【edit configure】,然后按如下图设置
然后在菜单栏【run】,点击【run】就可以完成本次作业了
出现图上的LOG信息就说明推理成功了。作业也就完成了
- 点赞
- 收藏
- 关注作者
评论(0)