Add complex128 support to FFT, FFT2D, FFT3D, IFFT, IFFT2D, and IFFT3D.
NumPy automatically upcasts to complex128 when computing FFTs, leading to issues like: #10749 This change allows users to choose between 32-bit and 64-bit precision FFTs on CPU and GPU. PiperOrigin-RevId: 195183206
Loading
Please sign in to comment