You really didn't think I'd let you get out of this chapter without some homework, did you? I've got a few basic problems to get you ready for the following lessons:
- Produce a CNN based on ImageNet to classify all of the MSCOCO classes.
Hint: There are plenty of available models out there. Simply get one running in inference mode and see how it works.
- Draw the basic GAN architecture. Now, draw an improvement to that architecture based on the topics you learned in this chapter. Think of the generator and discriminator as building blocks.