Anaconda 2016

From HPC users
Jump to navigationJump to search

Introduction

Conda is an open source package management system and environment management which can quickly install, run and update packages and their dependencies (see also https://conda.io/docs/). Conda is maintained by Continuum Analytics. Conda lets you find and install packages.

Installed Version

Anaconda3 / 4.3.0 (dependency)

conda 4.3.8

Using Anaconda

Creating Environments and Installing Packages

To keep your programs isolated, you should create separate environments for each application. It is not advisable to put programs into your base environment, which is the default one named 'base'.

To create a new environment and install a package, type in

conda create --name <environment_name> <package_name>

Subsequent, you must activate the new environment by typing:

source activate <environment_name>

From now on, every conda command will only affect the activated environment until it gets deactivated again.

source deactivate <environment_name>

To see a list of all your environments, type:

conda info --envs

The active environment is marked with an asterisk (*).

Different Versions

When you create a new environment, conda installs the same Python version which had been installed when Anaconda was installed. If you want to use a different version, you can create a new environment and specify the version of Python that you want. E.g.:

conda create --name <environment_name> python=3.5

In this example, Python should show 3.5 as the installed version:

python --version

Finding and Installing Software

If you want to find out whether you have a specific package installed, or if so, which version is installed, you can do this using the search function within the individual environments. Just type in

conda search <package_name>

and you will get a list of all the applying packages available from the Anaconda repository. If there is an asterisk in front of one of the displayed version numbers, this indicates that this version is already installed. To install one of the shown packages, just type in

conda install <package_name>

If you just want to get a short list of every single installed programs in the environment, type in

conda list