I want to use my own images in the test phase.

[yoyu66]

What do I need to realize that? acting in pytoｒｃｈ.

[LoreGoetschalckx]

Hi @yoyu66,

Thank you for your interest in our work. As we note in the paper, it is currently only possible to
modify seed images that are GAN-images themselves, not
user-supplied, real images. To make that possible, you would need a GAN inversion technique to translate your own images to the GAN's latent space.