Those 3 notebook files represents 3 different types of classification problems.
To classify hand-written digit classification dataset is imported from tensorflow package directly.
To classify FMNIST dataset download csv files from internet. Link: https://www.kaggle.com/datasets/zalando-research/fashionmnist
To classify Cats & Dogs download images from internet. Link: https://www.kaggle.com/datasets/tongpython/cat-and-dog
For Classifying Cats & Dogs. The folder structure will be like:
Create folder "dataset". In that folder create subfolders "training_set", "testing_set", single_prediction.
In folders "training_set" & "testing_set" there should be two sub-folders "cats" & "dogs" containing images.
In folder "single_prediction" there will be images