<![CDATA[Make Data Speak]]>https://amneumarkt.leima.isRSS for NodeFri, 01 May 2026 10:15:18 GMT60<![CDATA[Training GAN can be baffling]]>https://amneumarkt.leima.is/training-gan-can-be-bafflinghttps://amneumarkt.leima.is/training-gan-can-be-bafflingThu, 19 Aug 2021 08:30:55 GMTTraining GAN can be baffling. For example, the generator and the discriminator just don't "learn" at the same scale sometimes. Would you try to balance the generator loss and discriminator loss by hand? Soumith Chintala ( @ FAIR ) put together this list of tips for training GAN. "Don't balance loss via statistics" is one of the 17 tips by Chintala. The list is quite inspiring.

https://github.com/soumith/ganhacks

]]>