I was working on the diabetic retinopathy problem on kaggle and was wondering what was a good approach to change images into a constant shape. Would it be better to use size = int where the image would be cropped and some information would be lost (sometimes over half the retina) or should i use size = (int,int) and resize the image to a size but change the aspect ratio hence warping the image?
Also my understanding is limited and correct me if i’m wrong about these functions or if theres a better alternative. Thank you.