I wrote a GAN implementation and I trained that to produce some sample images after training on a dataset. The images looked visually fine.
Now I want to test my implementation on the CI and make sure that as more changes are introduced in later PR's it does not produce any unwanted change that breaks the GAN. But it is not feasible to train a GAN during the time limit of the CI pipeline.
Would I have to locally train the GAN every time I change the implementation to see if it works correctly? Is there any simpler alternative that can be done within the time limit of say 5 minutes to test if the GAN works correctly?