CuArrays provides a fully-functional GPU array, which can give significant speedups over normal arrays without code changes. CuArrays are implemented fully in Julia, making the implementation elegant and extremely generic.
The package can be installed with the Julia package manager.
From the Julia REPL, type
] to enter the Pkg REPL mode and run:
pkg> add CuArrays
Or, equivalently, via the
julia> import Pkg; Pkg.add("CuArrays")
For usage instructions and other information, check-out the CUDA.jl documentation.
The package is tested against, and being developed for, Julia
1.0 and above. Main
development and testing happens on Linux, but the package is expected to work on macOS and
Windows as well.
Contributions are very welcome, as are feature requests and suggestions. Please open an issue if you encounter any problems.
4 days ago