Why do you train `refiner_model` with `synthetic_image_batch` for both x, y?

mjdietzx / SimGAN

Implementation of Apple's Learning from Simulated and Unsupervised Images through Adversarial Training

MIT License

411 stars 101 forks source link

Why do you train `refiner_model` with `synthetic_image_batch` for both x, y? #21

Closed chojuahn closed 4 years ago

chojuahn commented 4 years ago

I don't understand why there is synthetic_image_batch fed to train refiner_model for x and y. Shouldn't this be x as refiner_model and y as real_image_batch respectively?

chojuahn commented 4 years ago

self_regularization_loss is designed for ||R(x) - x|| where x is synthetic_image, which makes sense having synthetic_image_batch for both x and y.