FFT library for the embedded array language Accelerate. For details on Accelerate, refer to the main repository.
The following build flags control whether optimised implementations are used. Note that enabling these (which is the default) will require the corresponding Accelerate backend as a dependency:
llvm-ptx: For NVIDIA GPUsllvm-cpu: For multicore CPUs
Contributions and bug reports are welcome!
Please feel free to contact me through GitHub or gitter.im.