ExecutionCUDA

class nvmath.fft.ExecutionCUDA(device_id: int | None = None)

A data class for providing GPU execution options to the FFT object and the family of wrapper functions fft(), ifft(), rfft(), and irfft().

device_id

CUDA device ordinal (used if the operand resides on the CPU). Device 0

Type:

int | None

will be used if not specified.

See also

ExecutionCPU, FFT, fft(), ifft(), rfft(), and irfft().