improved-gan
improved-gan copied to clipboard
Inception score for conditional GANs
IMO,there are tow ways to calculate Inception score of conditional GANs.
- For every "condition", calculate a IS and Average them.
- Sample "condition" from "real condition distribution" and generate fake data to calculate IS. Which way is reasonable. thx.
I also has this problem, I don't undersdand the detail of the calculation of Inception score perfectly.