Would you run into class balance issues if you do not have a growing pool of ini... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		anankaie on Nov 20, 2022 \| parent \| context \| favorite \| on: Deepfake detector can spot a real or fake video ba... Would you run into class balance issues if you do not have a growing pool of initial videos?

sigmoid10 on Nov 27, 2022 [–]

Not if you have a balanced dataset. If not, you're already getting into trouble with GANs, which is why they were superseded by diffusers on image generation tasks.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact