👋 Hey there! This feature is under active development! This guide will walk you through how to use Renku’s environment customization functionality, and show you workarounds for the spots where we haven’t finished building the features yet.

</aside>

Choose your language:

Python

Renku can build an environment based on a Python dependencies file for you (no knowledge of Docker required)!

To get started, see How to create a custom environment from a code repository.

R

Add packages to your project dependencies

To add packages to your environment, add the packages to the dependencies file: install.R.

To specify your project dependencies in install.R, write the R package install commands.

Note that the base R install.packages function can only install the latest version of a package on CRAN. To install a specific version, use devtools::install_version. If the package is not on CRAN and exists as a github repo, you can use install_github instead. These two scenarios are shown in the example below.

devtools::install_version("ggplot2",
            version = "3.3.6",
            repos = "<http://cran.us.r-project.org>",
            upgrade="always")
devtools::install_github("thomasp85/patchwork")

Install the packages

Every time you start or resume the session, install the packages you listed in the dependency file by running the following command in the terminal:

**$** R -f install.R

Note: You need to install the dependencies every time you launch or resume your session.

</aside>

conda

If your image uses conda as the python package management system, follow the instructions.

Add packages to your project dependencies

To add packages to your environment, add the packages to the dependencies file: environment.yml.

To add your project’s dependencies to environment.yml, specify an environment name and then a list of dependencies.