pytorch_fft
pytorch_fft copied to clipboard
PyTorch wrapper for FFTs
Looking forward for normalization argument, which is corresponding to numpy.fft, for example, np.fft.fft(x, norm='ortho') This is done by dividing the result by sqrt(N)
The following is the code: ``` import torch from torch.autograd import Variable import pytorch_fft.fft as fft f = fft.autograd.Fft2d() invf = fft.autograd.Ifft2d() x = Variable(torch.arange(0, 16).view(1, 1, 4, 4), requires_grad=True).cuda()...
In numpy's fft functions it is possible to perform ffts with respect to a particular axis: https://docs.scipy.org/doc/numpy-1.13.0/reference/routines.fft.html Is it possibly to add this to the code? Or is it already...
Either I am missing something or it is not present in this package: Is there a way to specify target output arrays for the fft? If not, is this planned?...