In each batch there are 8000 sample points, and I apply discrete Fourier transform on them. The original samples are real valued, so only the half of the result is needed. The end result is 4000 complex numbers.
I would like to build an autoencoder (with convolution and deconvolution), which can find the common features of these batches
Currently my loss function calculates the average distance between the 4000 input and output complex numbers on the complex plane. Unfortunately it produces a really low loss, because a really big fraction of numbers are close to zero
How should I define the loss function, if the input of the autoencoder is a large complex vector?
