Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

用 conda 管理 Python 开发环境 #26

Open
rainyear opened this issue Mar 13, 2016 · 0 comments
Open

用 conda 管理 Python 开发环境 #26

rainyear opened this issue Mar 13, 2016 · 0 comments
Assignees
@rainyear
Labels
Jupyter Note Python

Comments

@rainyear
Copy link
Owner

之前所有 Python 新项目都是用 python3 -m venv venv3 创建独立的虚拟环境，这次建一个基于 Jupiter 的科学计算环境，所以用到了 Anaconda：

THE FASTEST GROWING OPEN DATA SCIENCE PLATFORM

Anaconda 提供一个管理工具 conda，可以把 conda 看作是 pip + virtualenv + PVM (Python Version Manager) + 一些必要的底层库，也就是一个更完整也更大的集成管理工具。

1. 下载、安装 Anaconda

如果只想用 conda ，可以下载简化版的 Miniconda，需要 Anaconda 可以安装完全版，一共大概 400 M，用阿里云服务器从官方网站下载大概要 6 个小时…不过还好清华大学 TUNA 镜像源提供了镜像：

# 下载时要看清楚 Anaconda3 还是 2，对应的是 Python 3.5 和 2.7
wget https://mirrors.tuna.tsinghua.edu.cn/anaconda/Anaconda3-2.5.0-Linux-x86_64.sh
bash Anaconda3-2.5.0-Linux-x86_64.sh

设定安装目录 /home/rainy/.anaconda3 并一路确认，会在 ~/.bashrc 添加环境变量，因为我用的是 zsh，所以需要在 ~/.zshrc 添加：

# added by Anaconda3 2.5.0 installer
export PATH="/home/rainy/.anaconda3/bin:$PATH"

conda --version
# conda 3.19.1

2. 创建（clone）新的环境

conda env list
# conda environments:
#
root                  *  /home/rainy/.anaconda3

conda create --name nb --clone root

conda env list
# conda environments:
#
nb                       /home/rainy/.anaconda3/envs/nb
root                  *  /home/rainy/.anaconda3

切换环境：

source activate nb
# discarding /home/rainy/.anaconda3/bin from PATH
# prepending /home/rainy/.anaconda3/envs/nb/bin to PATH

此时变成 (nb) $ ，和 virtualenv 一样，只是在退出时不太一样：

which python
/home/rainy/.anaconda3/envs/nb/bin/python
source deactivate
discarding /home/rainy/.anaconda3/envs/nb/bin from PATH

需要重新打开新的窗口才能再切换。现在查看已安装的 package 列表：

source active nb
conda list
# packages in environment at /home/rainy/.anaconda3/envs/nb:
#
abstract-rendering        0.5.1               np110py35_0
alabaster                 0.7.7                    py35_0
anaconda                  2.5.0               np110py35_0
anaconda-client           1.2.2                    py35_0
argcomplete               1.0.0                    py35_1
astropy                   1.1.1               np110py35_0
...

3. 配置新的 Jupyter

Anaconda 已经集成了 Jupyter，可以直接使用。Jupyter 默认的配置文件在 ~/.jupyter/jupyter_notebook_config.py，新的 Jupyter 也会从这里读取配置文件，官方文档里面写的 jupyter {application} --generate-config 并不是想象中的用法：

jupyter app --generate-config
jupyter: 'app' is not a Jupiter command

根据 Google 的结果，应该是：

JUPYTER_CONFIG_DIR=./jupyter_config jupyter --generate-config

编辑配置：

# vim ./jupyter_config/jupyter_notebook_config.py
c.NotebookApp.password = u'sha1:****'
c.NotebookApp.ip = 'my domain.com'
c.NotebookApp.port = 8888

启动：

JUPYTER_CONFIG_DIR=./jupyter_config jupyter notebook
@rainyear rainyear changed the title 用 conda 管理 Python 开发环境 用 conda 管理 Python 开发环境 Mar 13, 2016
@rainyear rainyear self-assigned this Sep 8, 2017
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@rainyear rainyear

Labels
Jupyter Note Python
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@rainyear