PyTorch 安装相当简便。先去官网上配置环境，生成安装命令，然后执行即可。
1. 去官网配置环境，生成安装命令
打开官网网址，配置环境，生成安装命令，如下图所示：
2. 执行安装命令
执行安装命令：
Kevin@QIKEGU C:\Users\Kevin
> pip3 install torch==1.2.0+cpu torchvision==0.4.0+cpu -f https://download.pytorch.org/whl/torch_stable.html
Looking in links: https://download.pytorch.org/whl/torch_stable.html
Collecting torch==1.2.0+cpu
Downloading https://download.pytorch.org/whl/cpu/torch-1.2.0%2Bcpu-cp37-cp37m-win_amd64.whl (89.4MB)
|████████████████████████████████| 89.4MB 16kB/s
Collecting torchvision==0.4.0+cpu
Downloading https://download.pytorch.org/whl/cpu/torchvision-0.4.0%2Bcpu-cp37-cp37m-win_amd64.whl (684kB)
|████████████████████████████████| 686kB 13kB/s
Requirement already satisfied: numpy in d:\program files\python\lib\site-packages (from torch==1.2.0+cpu) (1.16.4)
Requirement already satisfied: six in c:\users\kevin\appdata\roaming\python\python37\site-packages (from torchvision==0.4.0+cpu) (1.12.0)
Collecting pillow>=4.1.1 (from torchvision==0.4.0+cpu)
Using cached https://files.pythonhosted.org/packages/ae/96/6f83deebfcd20a5d4ad35e4e989814a16559d8715741457e670aae1a5a09/Pillow-6.1.0-cp37-cp37m-win_amd64.whl
Installing collected packages: torch, pillow, torchvision
Successfully installed pillow-6.1.0 torch-1.2.0+cpu torchvision-0.4.0+cpu
3. 验证安装
进入Python:
> python
Python 3.7.3 (v3.7.3:ef4ec6ed12, Mar 25 2019, 22:22:05) [MSC v.1916 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>>
执行如下语句：
from __future__ import print_function
import torch
x = torch.rand(5, 3)
print(x)
如果安装正确，将输出以下内容：
tensor([[0.0799, 0.8004, 0.3068],
[0.9890, 0.6794, 0.4926],
[0.2001, 0.2437, 0.3993],
[0.9377, 0.5443, 0.6676],
[0.7629, 0.2589, 0.6175]])