Skip to content

Latest commit

 

History

History
115 lines (68 loc) · 3.74 KB

install-packages.rst

File metadata and controls

115 lines (68 loc) · 3.74 KB
Raw

Add Packages to your Renku Project Environment

The process for adding packages to your Renku project is a little different depending on which language you are using- for example Python or R. Alternatively, for both R and Python, you may use conda's environment management system. All three of these options work with Renku.

Select the tab below for the documentation relevant to you.

.. tabbed:: Python

    **Add packages to your project dependencies**

    To add packages to your environment, add the packages to the dependencies file.
    When you're working in Python, this is the ``requirements.txt`` file.

    To add your project's dependencies to ``requirements.txt``, simply list the package names.
    Optionally, you may specify specific package versions, as shown in the example below.

    .. code-block:: console

        pandas
        numpy==1.22.0

    **Install the packages**

    Install the packages you listed in the dependency file by running the following command in the terminal:

    .. code-block:: console

        $ pip install -r requirements.txt


.. tabbed:: R

    **Add packages to your project dependencies**

    To add packages to your environment, add the packages to the dependencies file.
    When you're working in R, it's ``install.R``.

    To specify your project dependencies in ``install.R``, write the R package install commands.
    This ``install.R`` script will be sourced and run to create your requested environment.

    .. code-block:: console

        install.packages(c("dplyr", "ggplot2"))

    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.

    .. code-block:: console

        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**

    Install the packages you listed in the dependency file by running the following command in the terminal:

    .. code-block:: console

        $ R -f install.R


.. tabbed:: conda

    **Add packages to your project dependencies**

    To add packages to your environment, add the packages to the dependencies file.
    When you're using conda, it's ``environment.yml``.

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

    .. code-block:: console

        name: stats
        dependencies:
          - numpy
          - pandas

    For more details, see `conda's documentation <https://docs.conda.io/projects/conda/en/latest/user-guide/tasks/manage-environments.html#create-env-file-manually>`_.

    **Install the packages**

    Install the packages you listed in the dependency file by running the following command in the terminal:

    .. code-block:: console

        $ conda env update --file environment.yml  --prune

Save your changes

After you've modified your project's dependencies, make sure to save your changes so that the packages will automatically be installed the next time you start a Renku session.

$ renku save -m "updated dependencies"

The next time you start a renku session, the packages will already be installed for you.

Looking for more options?

If you'd like to further customize your Renku project environment, take a look at :ref:`customizing` and :ref:`docker`.