Model compression for generative models (GAN or VAE)

[gulaotou (Sam Gu)]

AIMET is designed for CNN compression, but how about generative models (GAN or VAE)? Can it be used to compress the Generator in GAN, or Decoder in VAE?

[akhobare (Abhi Khobare)]

@gulaotou

IMET is designed for CNN compression, but how about generative models (GAN or VAE)? Can it be used to compress the Generator in GAN, or Decoder in VAE?

We have not attempted this yet. So not sure if this will work out of the box. Generally it seems to me that quantization is a better bang-for-the-buck than compression for generative models.