I got a dataset splits into $2$ folder 'test' with $240$ images and 'train'. with $960$ images total. Each folder will have $4$ subfolder 'Action', 'Comedy', 'Horror', 'Romance', which includes many poster films images for that type. I used CNN to train the model and then using EfficientNet$\_$B$0$ as the base model, perform transfer learning to retrain the final layer using this new dataset. Now, I want to Use a generative AI tool for images, one of the ones recommended earlier or a choice of your own, to

make one or more of your own movie posters. You can also further edit the poster if you$$re feeling

artistic. Try to make a poster that you think will be classified correctly. Decide which genre you intend

your movie$/$s to be and see if the EfficientNet model identifies them correctly.

Explain your rationale for why you designed the poster the way you did, what about this design made

you think it would be classified correctly? Why or why not was classified correctly?

Now try to create a movie poster that you think will be classified incorrectly. Again, what was your

rationale, and why or why not was the intended result achieved?