第七周周三 配置pytorch环境，运行YOLO v4

昨天安装好了vscode，今天配置它的环境来运行yolo。
首先在anaconda里创建一个新的环境——pytorch，并激活环境：

conda create -n pytorch python==3.6  #环境名:pytorch,py版本3.6
activate pytorch

之后配置下载源

# 配置国内源，方便安装Numpy,Matplotlib等
conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/
conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/
# 配置国内源，安装PyTorch用
conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/
# 显示源地址
conda config --set show_channel_urls yes

但是执行显示源地址后，并没有出现源地址。接下来安装pytorch1.2，显示：

Collecting package metadata (current_repodata.json): done
Solving environment: done

# All requested packages already installed.

但是之后安装模块时并不成功，执行

conda clean --packages --tarballs  
#再重新执行pytorch安装命令
conda install pytorch==1.2.0 torchvision==0.4.0 cudatoolkit=10.0

清除包这一步费时有点长，但是，重新执行pytorch安装命令，仍然是同样的结果，我先去运行了train.py，提示：

ModuleNotFoundError: No module named 'cv2'

搜索后发现要安装opencv-python，希望可以成功

anaconda search -t conda opencv-python
conda install -c https://api.anaconda.org//wheeler-microfluidics opencv-python

但是……，失败了：

(pytorch) C:\Users\ACER>conda install opencv-python
Collecting package metadata (current_repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Collecting package metadata (repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.

PackagesNotFoundError: The following packages are not available from current channels:

  - opencv-python

Current channels:

  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/win-64
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/noarch
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/win-64
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/noarch
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/win-64
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/noarch
  - https://repo.anaconda.com/pkgs/main/win-64
  - https://repo.anaconda.com/pkgs/main/noarch
  - https://repo.anaconda.com/pkgs/r/win-64
  - https://repo.anaconda.com/pkgs/r/noarch
  - https://repo.anaconda.com/pkgs/msys2/win-64
  - https://repo.anaconda.com/pkgs/msys2/noarch

To search for alternate channels that may provide the conda package you're
looking for, navigate to

    https://anaconda.org

and use the search bar at the top of the page.

网上搜索之后，执行如下命令：

anaconda search -t conda opencv-python

结果：

(pytorch) C:\Users\ACER>anaconda search -t conda opencv-python
Using Anaconda API: https://api.anaconda.org
Packages:
     Name                      |  Version | Package Types   | Platforms       | Builds
     ------------------------- |   ------ | --------------- | --------------- | ----------
     fastai/opencv-python-headless | 4.5.1.48 | conda           | linux-64, win-64, osx-64 | py37_0, py36_0, py38_0
                                          : Wrapper package for OpenCV python bindings.
     fastchan/opencv-python-headless | 4.5.1.48 | conda           | linux-64, win-64, osx-64 | py37_0, py36_0, py38_0
                                          : Wrapper package for OpenCV python bindings.
     khamutov/opencv-python-headless | 4.3.0.36 | conda           | linux-ppc64le   | py37_3
                                          : Power9 build for OpenCV
     ohsu-comp-bio/opencv-python-headless | 4.5.1.48 | conda           | linux-64, osx-64 | py37hb2a4cfb_1, py37hbeab489_0, py37hef7ab6a_1
                                          : opencv python headless
     rsignell/opencv-python    |    2.4.9 | conda           | win-32          | np18py27_0
     wheeler-microfluidics/opencv-python |  3.1.0.3 | conda           | win-32          | 0
Found 6 packages

Run 'anaconda show <USER/PACKAGE>' to get installation details

选择 wheeler-microfluidics/opencv-python，斜杠/换成空格：

conda install -c https://api.anaconda.org//wheeler-microfluidics opencv-python

又报错了：

 (pytorch) C:\Users\ACER>conda install -c https://api.anaconda.org//wheeler-microfluidics opencv-python
Collecting package metadata (current_repodata.json): failed

UnavailableInvalidChannel: The channel is not accessible or is invalid.
  channel name: wheeler-microfluidics
  channel url: https://api.anaconda.org/wheeler-microfluidics
  error code: 404

You will need to adjust your conda configuration to proceed.
Use `conda config --show channels` to view your configuration's current state,
and use `conda config --show-sources` to view config file locations.

检查设置后，发现只有之前配置的几个通道：

 (pytorch) C:\Users\ACER>conda config --show channels
channels:
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/
  - defaults

(pytorch) C:\Users\ACER>conda config --show-source
==> C:\Users\ACER\.condarc <==
ssl_verify: True
channels:
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/
  - defaults
show_channel_urls: True

尝试了其他方法下载cv2，明天再写。