installing-software-conda.md - University of Oslo

Anaconda, Miniconda & Conda

You can install many python and non-python packages yourself using conda or especially for bioinformatics software bioconda.

Conda enables you to easily install complex packages and software. Creating multiple enviroments enables you to have installations of the same software in different versions or incompatible software collections at once. You can easily share a list of the installed packages with collaborators or colleagues, so they can setup the same eniviroment in a matter of minutes.

Setup

First you load the miniconda module which is like a python and r package manager. Conda makes it easy to have multiple environments for example one python2 and one python3 based parallel to each other without interfering.

Load conda module

Start by removing all preloaded modules which can complicate things. We then display all installed version and load the newest Miniconda one (4.6.14):

$ module purge
$ module avail conda
$ module Miniconda3/4.6.14

Setup conda activate command

To use conda activate interactively you have to initialise your shell once with:

$ conda init bash

Add channels

To install packages we first have to add the package repository to conda (we only have to do this once). This is the place conda will download the packages from.

$ conda config --add channels defaults
$ conda config --add channels conda-forge

If you want install bioinformatics packages you should also add the bioconda channel:

$ conda config --add channels bioconda

Supress unneccessary warnings

To suppress the warning that a newer version of conda exists which is usually not important for most users and will be fixed by us by installing a new module:

$ conda config --set notify_outdated_conda false

Create new environment

New environments are initialised with the conda create. During the creation you should list all the packages and software that should be installed in this environment instead of creating an empty one and installing them one by one. This makes the installation much faster and there is less chance for conda to get stuck in a dependency loop.

$ conda create --name ENVIRONMENT python=3 SOMESOFTWARE MORESOFTWARE

If you are planning on adding many libraries to your environment, you should consider placing it in a directory other than your $HOME, due to the {ref}storage restrictions <clusters-homedirectory> on that folder. One alternative could be to use the {ref}Project area <project-area>, please check out {ref}Storage areas on HPC clusters <clusters-overview> for other alternatives. To install conda in an alternative location, use the --prefix PATH or -p PATH option when creating a new environment.

conda create -p PATH SOMEPACKAGES

This enables multiple users of a project to share the conda environment by installing it into their project folder instead of the user's home.

Daily usage

Interactively

To load this environment you have to use the following commands either on the command line or in your job script:

$ module purge
$ module Miniconda3/4.6.14 # Replace with the version available on the system
$ conda activate ENVIRONMENT

Then you can use all software as usual.

To deactivate the current environment:

$ conda deactivate

If you need to install additional software or packages, we can search for it with:

$ conda search SOMESOFTWARE

and install it with:

$ conda install -n ENVIRONMENT SOMESOFTWARE

If the python package you are looking for is not available in conda you can use pip like usually from within a conda environment to install additional python packages:

$ pip install SOMEPACKAGE

To update a single package with conda:

$ conda update -n ENVIRONMENT SOMESOFTWARE

or to update all packages:

$ conda update -n ENVIRONMENT --all

In batch/job scripts

To be able to use this environment in a batch script (job script), you will need to include the following in your batch script, before calling the python program:

# load the Anaconda3
module load Anaconda3/2019.03

# Set the ${PS1} (needed in the source of the Anaconda environment)
export PS1=\$

# Source the conda environment setup
# The variable ${EBROOTANACONDA3} or ${EBROOTMINICONDA3}
# So use one of the following lines
# comes with the module load command
# source ${EBROOTANACONDA3}/etc/profile.d/conda.sh
source ${EBROOTMINICONDA3}/etc/profile.d/conda.sh

# Deactivate any spill-over environment from the login node
conda deactivate &>/dev/null

# Activate the environment by using the full path (not name)
# to the environment. The full path is listed if you do
# conda info --envs at the command prompt.
conda activate PATH_TO_ENVIRONMENT

# Execute the python program
python pdexample.py

Share your environment

Share with project members on the same machine

By creating conda environments in your project folder (conda create -p /cluster/projects/nnXXXXk/conda/ENVIROMENT) all your colleagues that are also member of that project have access to the environment and can load it with:

$ conda activate /cluster/projects/nnXXXXk/conda/ENVIROMENT

Export your package list

To export a list of all packages/programs installed with conda in a certain environment (in this case "ENVIRONMENT"):

$ conda list --explicit --name ENVIRONMENT > package-list.txt

To setup a new environment (let's call it "newpython") from an exported package list:

$ conda create --name newpython --file package-list.txt

Additional Conda information

Cheatsheet and built-in help

See this cheatsheet for an overview over the most important conda commands.

In case you get confused by the conda commands and command line options you can get help by adding --help to any conda command or have a look at the conda documentation.

Miniconda vs. Anaconda

Both Miniconda and Anaconda are distributions of the conda repository management system. But while Miniconda brings just the management system (the conda command), Anaconda comes with a lot of built-in packages.

Both are installed on Stallo but we advise the use of Miniconda. By explicitly installing packages into your own environment the chance for unwanted effects and errors due to wrong or incompatible versions is reduced. Also you can be sure that everything that happens with your setup is controlled by yourself.

^{CC Attribution: This page is maintained by the University of Oslo IT FFU-BT group.
It has either been modified from, or is a derivative of, "Installing Python packages"
by NRIS under CC-BY-4.0.}