Python 入门与实战教程系列
1. Python入门与实战教程（发布中）
https://edu.csdn.net/course/detail/24468
2. Python数据可视化库Seaborn（待上传视频）
https://edu.csdn.net/course/detail/24790
3. Python数据可视化教程：基于Plotly的动态可视化绘图（发布中）
https://edu.csdn.net/course/detail/24935
---------------------
作者：走在前方
mac pro pytorch 安装，花费了我近一个小时的时间，终于大功告成，接下来就可以在mac 上编写pytorch程序，并在CPU上运行，成功后，可以考虑迁移到公司的GPU上运行。
1. install pytorch
conda install pytorch torchvision -c pytorch
2. install torchvision
pip3 install https://download.pytorch.org/whl/cpu/torch-1.0.0-cp37-none-macosx_10_7_x86_64.whl
pip3 install torchvision
直接pip3 install https://xxxxx, 失败后手动下载文件，然后进行安装
下载 https://download.pytorch.org/whl/cpu/torch-1.0.0-cp37-none-macosx_10_7_x86_64.whl 后，然后安装
$ pip3 install torch-1.0.0-cp37-none-macosx_10_7_x86_64.whl
Looking in indexes: http://mirrors.aliyun.com/pypi/simple/
Processing ./torch-1.0.0-cp37-none-macosx_10_7_x86_64.whl
Installing collected packages: torch
Found existing installation: torch 1.0.1.post2
Uninstalling torch-1.0.1.post2:
Successfully uninstalled torch-1.0.1.post2
Successfully installed torch-1.0.0
3. verfication pytorch program
ipython enter function pages：
===>
In [6]: from __future__ import print_function
In [7]: import torch
In [8]: x = torch.rand(5,3)
In [9]: print(x)
tensor([[0.9826, 0.0903, 0.1729],
[0.8138, 0.7652, 0.4320],
[0.5192, 0.7816, 0.6091],
[0.9613, 0.1501, 0.9096],
[0.6060, 0.6852, 0.4438]])
reference
- pytorch 安装步骤 https://pytorch.org/get-started/locally/
- CPU 模式下安装pytorch https://pytorch.org/get-started/previous-versions/
