I have a dataset of Apple, Pineapple and Strawberry images. In this dataset, there are mature/immature/partially mature ripe images of these fruits. The data was previously separated as train/test/validation. How can I split this data for CNN using sklearn/train_test_split? I am sharing the folder structure with you. Dataset Folder Structure
In previous studies, classes were available under the train folder. I was doing my project without any problems. However, in this dataset, there is a class within a class in the train folder. So I don't know what to do. I tried to do it using Python but it didn't work.