GAN Lecture 2
Conditional Generation by GAN
Algorithm
In each traing iteration:
- Sample m positive examples
from database
- Sample m noise samples
from a distribution
- Obtaining generated data
,
- Sample m objects
from database
- Update discriminator parameters
to maximize
Learning D
- Sample m noise samples
from a distribution
- Sample m conditions
from a database
- Update generator parameters
to maximize
-
,
-
Learning G
倾向推荐第二种网络架构
参考文献：StackGAN
参考文献：Patch GAN
参考例子：Github
