深度学习环境配置——Linux下的torch=1.2.0环境配置

**

下载Python-3.6.2.tgz软件包**

[root@localhost ~]# mkdir -p /server/tools/
[root@localhost tools]# cd /server/tools/
[root@localhost ~]# wget https://registry.npmmirror.com/-/binary/python/3.6.2/Python-3.6.2.tgz

[root@localhost tools]# tar -xf Python-3.6.2.tgz
[root@ tools]# cd Python-3.6.2
[root@localhost Python-3.6.2]# ./configure
[root@localhost Python-3.6.2]# make
[root@localhost Python-3.6.2]# make install

删除旧的python的软连接，创建新的python2和python3的软连接

[root@localhost Python-3.6.2]# cd /usr/bin/
[root@localhost bin]# rm -f python2
[root@localhost bin]# mv python python2.6.ori
[root@localhost bin]# ln -s python2.6 python2
[root@localhost bin]# ln -s /usr/local/bin/python3 /usr/bin/python

[root@localhost bin]# python -V
Python 3.6.2
[root@localhost bin]# python2 -V
Python 2.6.6
[root@localhost bin]# python3 -V
Python 3.6.2

[root@bogon ~]# wget --no-check-certificate https://pypi.python.org/packages/source/s/setuptools/setuptools-19.6.tar.gz#md5=c607dd118eae682c44ed146367a17e26
[root@localhost ~]# tar zxvf setuptools-19.6.tar.gz
[root@localhost ~]# cd setuptools-19.6
[root@localhost ~]# python3 setup.py build
[root@localhost ~]# python3 setup.py install

[root@localhost ~]# wget --no-check-certificate https://pypi.python.org/packages/source/p/pip/pip-8.0.2.tar.gz#md5=3a73c4188f8dbad6a1e6f6d44d117eeb
[root@localhost ~]# tar zvxf pip-8.0.2.tar.gz
[root@localhost ~]# cd pip-8.0.2
[root@localhost pip-8.0.2]# python3 setup.py build
[root@localhost pip-8.0.2]# python3 setup.py install

打开pytorch的官方安装方法：
https://pytorch.org/get-started/previous-versions/
官网推荐的安装代码如下，我使用的是Cuda10的版本，不太懂为什么要写3个=才能正确定位，两个=会定位到cuda92的whl：

# CUDA 10.0
pip install torch===1.2.0 torchvision===0.4.0 -f https://download.pytorch.org/whl/torch_stable.html

scipy==1.2.1
numpy==1.17.0
matplotlib==3.1.2
opencv_python==4.1.2.30
torch==1.2.0
torchvision==0.4.0
tqdm==4.60.0
Pillow==8.2.0
h5py==2.10.0

创建一个requirements.txt文件，复制上述内容到txt文件中。

pip install -r C:UsersDELLDesktoprequirements.txt

接下来就可以自己跑项目啦！！！！