Training a Variational Autoencoder on sine waves

2 Ansichten (letzte 30 Tage)
Uerm
Uerm am 21 Nov. 2019
Hi,
I am trying to train a variational autoencoder on sine waves using the code from https://se.mathworks.com/help/deeplearning/examples/train-a-variational-autoencoder-vae-to-generate-images.html
The code seems to be running. However, the ELBO loss is identical through all the epochs. Therefore, it seems that something is wrong with the training part. When I run the code with the data given on the webpage, I get the same results as on the webpage (the training part works, ELBO loss is reduced through the epochs). I attach two m-files. ConvAE.m is the main code and Sinus.m is the sine wave data set. I have not attached the other helper functions as they are identical to the ones on the webpage. Some parameters in ConvAE.m have been adjusted to match the data set that I use.
Can anyone spot the problem?

Antworten (0)

Kategorien

Mehr zu MATLAB finden Sie in Help Center und File Exchange

Produkte


Version

R2019b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by