This is the source tree or distribution for the rpy2 package.



pip should work out of the box:

pip install rpy2

The package is known to compile on Linux, MacOSX, and Windows (provided that developper tools are installed, and you are ready figure out how by yourself).

Alternatively, there is a Docker image available to try rpy2 out without concerns about the installation process.

To run the ipython console:

docker run -it --rm -p 8888:8888 rpy2/jupyter:2.9.x ipython

To run jupypter notebook on port 8888:

docker run --rm -p 8888:8888 rpy2/jupyter:2.9.x

More information about Docker images can be found in the documentation.

In case you find yourself with this source without any idea of what it takes to compile anything on your platform, try first

python setup.py install

If this fails, consider looking for pre-compiled binaries (they are available on Linux Red Hat, CentOS, Debian, Ubuntu, etc...) or using the matching Docker container.

Note that python setup.py develop will appear to work, but will result in an installation from the rpy directory here. The namespaces will be incorrect, so don't do that!


Documentation is available either in the source tree (to be built), or online (on readthedocs).


The testing machinery uses the new unittest functionality, requiring python 2.7+ (or potentially the backported unittest2 library for older python, but this is not supported). The test suite can be run (once rpy2 is installed) as follows:

pytest --pyargs rpy2

Individual tests can be run as follows:

pytest --pyargs rpy2.robjects.tests.testVector

The former recommended way to run tests (python -m rpy2.tests) may still work.


RPy2 can be used under the terms of the GNU General Public License Version 2 or later (see the file gpl-2.0.txt). This is the very same license R itself is released under.