anaconda+pytorch安装说明

第一步：anconda的安装，大家可以参考以下博文，写的比较简约，看起来比较清爽

https://blog.csdn.net/ITLearnHall/article/details/81708148

Anacond的介绍
Anaconda指的是一个开源的Python发行版本，其包含了conda、Python等180多个科学包及其依赖项。 因为包含了大量的科学包，Anaconda 的下载文件比较大（约 531 MB），如果只需要某些包，或者需要节省带宽或存储空间，也可以使用Miniconda这个较小的发行版（仅包含conda和 Python）。

Conda是一个开源的包、环境管理器，可以用于在同一个机器上安装不同版本的软件包及其依赖，并能够在不同的环境之间切换

Anaconda包括Conda、Python以及一大堆安装好的工具包，比如：numpy、pandas等

Miniconda包括Conda、Python
 

第二步，使用conda来安装pytorch

登录pytorch的官网，然后选择和自己电脑配置以下的信息，会自动生成的对应conda命令

https://pytorch.org/get-started/locally/

其中因为我没有Gpu所以选择的是none cuda，py build最好是选择的稳定版，防止运行时候有问题

conda install pytorch-cpu torchvision-cpu -c pytorch

安装命令后，就会出来对应的说明

 

在proceed后选择y进行安装 ,第一遍  pytorch-cpu-1.0.1  下载一直卡着，于是重新run了一下命令，可以正常下载，但是提示用户没有写目录的权限。这个以管理员身份再启动一下prompt即可解决

ARNING: The conda.compat module is deprecated and will be removed in a future release.
Collecting package metadata: done
Solving environment: done

## Package Plan ##

  environment location: C:\ProgramData\Anaconda3

  added / updated specs:
    - pytorch-cpu
    - torchvision-cpu

The following packages will be downloaded:

    package                    |            build
    ---------------------------|-----------------
    conda-4.6.12               |           py37_1         2.1 MB
    ninja-1.9.0                |   py37h74a9793_0         263 KB
    pytorch-cpu-1.0.1          |      py3.7_cpu_1        59.0 MB  pytorch
    torchvision-cpu-0.2.2      |             py_3          44 KB  pytorch
    ------------------------------------------------------------
                                           Total:        61.4 MB

The following NEW packages will be INSTALLED:

  ninja              pkgs/main/win-64::ninja-1.9.0-py37h74a9793_0
  pytorch-cpu        pytorch/win-64::pytorch-cpu-1.0.1-py3.7_cpu_1
  torchvision-cpu    pytorch/noarch::torchvision-cpu-0.2.2-py_3

The following packages will be UPDATED:

  conda                                       4.6.11-py37_0 --> 4.6.12-py37_1

Proceed ([y]/n)? y

Downloading and Extracting Packages
torchvision-cpu-0.2. | 44 KB     | ############################################################################ | 100%
ninja-1.9.0          | 263 KB    | ############################################################################ | 100%
pytorch-cpu-1.0.1    | 59.0 MB   | 4                                                                            |   1%
conda-4.6.12         | 2.1 MB    | ############################################################################ | 100%

CondaHTTPError: HTTP 000 CONNECTION FAILED for url
Elapsed: -

An HTTP error occurred when trying to retrieve this URL.
HTTP errors are often intermittent, and a simple retry will get you on your way.

(base) C:\Users\shdxlzh>conda install pytorch-cpu torchvision-cpu -c pytorch
WARNING: The conda.compat module is deprecated and will be removed in a future release.
Collecting package metadata: done
Solving environment: done

## Package Plan ##

  environment location: C:\ProgramData\Anaconda3

  added / updated specs:
    - pytorch-cpu
    - torchvision-cpu

The following packages will be downloaded:

    package                    |            build
    ---------------------------|-----------------
    pytorch-cpu-1.0.1          |      py3.7_cpu_1        59.0 MB  pytorch
    ------------------------------------------------------------
                                           Total:        59.0 MB

The following NEW packages will be INSTALLED:

  ninja              pkgs/main/win-64::ninja-1.9.0-py37h74a9793_0
  pytorch-cpu        pytorch/win-64::pytorch-cpu-1.0.1-py3.7_cpu_1
  torchvision-cpu    pytorch/noarch::torchvision-cpu-0.2.2-py_3

The following packages will be UPDATED:

  conda                                       4.6.11-py37_0 --> 4.6.12-py37_1

Proceed ([y]/n)? y

Downloading and Extracting Packages
pytorch-cpu-1.0.1    | 59.0 MB   | ############################################################################ | 100%
Preparing transaction: done
Verifying transaction: failed

EnvironmentNotWritableError: The current user does not have write permissions to the target environment.
  environment location: C:\ProgramData\Anaconda3

(base) C:\Users\shdxlzh>
 

最后总算安装成功了，安装成功的状态如下