I aim to use variational autoencoders (VAE) to find interpretable latent spaces for genetic data. So, I need to understand how they work, what activation function to use, etc. But all tutorials and courses that explain VAE use image data as their examples. Are there any good resources (courses, youtube videos, tutorials) that use non-image data? A proficiency level does not matter.
$\begingroup$
$\endgroup$
2
-
$\begingroup$ how is a table different from an image? $\endgroup$– AlbertoCommented Aug 14 at 19:11
-
$\begingroup$ in genetic data I have features and observations, so I can get latent spaces with weighted genes in them. I understand that an image also is a matrix of pixels, but I think models using images often have a purpose to reproduce/denoise/classify. I am interested in explainable latent spaces $\endgroup$– Yulia KentievaCommented Aug 14 at 20:45
Add a comment
|