« Back to home

Notes on Anaconda

Conda is an open source package management system and environment management system for installing multiple versions of software packages and their dependencies and switching easily between them.

Environments

Check installation details

conda info

List existing conda environments

conda info --envs
conda info -e

Create environment and install dependencies

conda create --name snowflakes biopython
conda create --n snowflakes biopython

> Fetching package metadata: ....
> Solving package specifications: .
> Package plan for installation 
> in environment C:\Users\USER\Anaconda\envs\snowflakes:>	
> 
> The following packages will be downloaded:
> 
>     package          |            build
>     -----------------|-----------------
>     biopython-1.65   |       np19py27_0    2.0 MB
> 
> The following NEW packages will be INSTALLED:
> 
>     biopython:  1.65-np19py27_0
>     numpy:      1.9.2-py27_0
>     pip:        7.1.0-py27_0
>     python:     2.7.10-0
>     setuptools: 18.0.1-py27_0
  • Dependency conflicts (between packages) are avoided by installing in the env at the same time

  • Environment default location C:\Users\USER\Anaconda\envs
    override with --path c:\env

Start using environment (UNIX systems require source)

(source) activate snowflakes

Exit with...

deactivate

Creating a new env with Python 3.4

conda create -n bunnies python=3.4 astroid

Clone an existing environment

conda create -n kittens --clone bunnies

Python

Check available python versions

conda search --full-name python
conda search -f python

Install Python3.4 in new env and activate

conda create -n snakes python=3.4
activate snakes

.condarc configuration

Get all keys from config file

conda config --get

Get channel keys from config file

conda config --get channels

Add new channel pandas. Looks for packages there

conda config --add channels pandas

Packages

See all packages in environment

conda list

Look for a package (if available)

conda search beautiful-soup	

Install package to environment (or current active). All packages available via conda install

conda install -n bunnies beatiful-soup

Update package in current environment

conda update beautiful-soup

Search in a specific location, e.g., pandas channel

conda search --override-channels -c pandas bottleneck

Install package from a given channel

conda install -c pandas bottleneck

Look for a package in Anaconda's repo

conda search --override-channels -c defaults beautiful-soup

Activate and pip-install a package

activate bunnies
pip install see

Install commercial packages

conda install iopro accelerate

Removing packages and environments

Remove 1 package from a named env, active environment

conda remove --name bunnies beautiful-soup
conda remove beautiful-soup

Remove multiple

conda remove --name bunnies beautiful-soup astroid

Remove environment

conda remove --name snakes --all

Using with Jupyter

Source activate your environment before running jupyter notebook

conda install nb_conda provides display and switch support on running notebooks

Sources

  1. http://conda.pydata.org/docs
  2. Conda cheat sheet

Comments

comments powered by Disqus