

The optional Platform tag specifies the platform where the image is intended for: If unspecified, the tag points to an image using Python 3.7. All Python versions supported by Ray are available, e.g. The optional Python version tag specifies the Python version in the image. The most recent Ray development build (a recent commit from Github master)Ī specific development build (uses a SHA from the Github master, e.g. Ray version tag can be one of the following: The rayproject/base-deps and rayproject/ray-deps images are for the Linux and Python dependencies respectively. The rayproject/ray-ml images include the above as well as many additional ML libraries. It comes with anaconda and various versions of Python. The rayproject/ray images include Ray and all required dependencies. Most users should pull a Docker image from the Ray Docker Hub. However, should you need to build from source, follow these instructions for building Ray. Installing from pip should be sufficient for most Ray users. Pip install ray in the newly created environment. Using a conda environment, it is recommended to install Ray from PyPi using Ray conda packages are maintained by the community, not the Ray team.

You can manually install the package by following the instructions on theĪrch Wiki or use an AUR helper like yay (recommended for ease of install) Ray is available on Arch Linux via the Arch User Repository ( AUR) as Note: Installing Ray on Arch Linux is not tested by the Project Ray developers. Windows does not have a copy-on-write forking model, so spinning up new Is considerably slower than on other operating systems. Performance on Windows is known to be slower since opening files on Windows Ray-specific, the rest are true anywhere Windows is used):įilenames are tricky on Windows and there still may be a few places where RayĪssumes UNIX filenames rather than Windows ones. Ray supports running on Windows with the following caveats (only the first is To get started with local Ray development:Įnsure you’re using the miniforge environment (you should see (base) in your terminal). Ray supports machines running Apple Silicon (such as M1 macs). If you build Ray from source, remove the build option build -cxxopt="-D_GLIBCXX_USE_CXX11_ABI=0" from the file cpp/example/.bazelrc before running your application.
