What can you use this tool for?

[github-userx]

Apologies, big noob here

[NSchrading]

If you have a set of images of two classes, for example cats and dogs, you can use it to easily label those images with the ground truth labels (cat or dog). That's the extent of the functionality of this tool.

Why would you want to do this? These labels can then be used in a supervised machine learning algorithm which will take your labeled images to train its model. Then, you can pass that machine learning model an unlabeled, new image of a cat or dog, and if the training worked well it should be able to tell you accurately whether it is an image of a cat or an image of a dog.

Obviously, it doesn't have to be cats and dogs. It will work just as well on any two-class dataset you can think of (cars and motorcycles, hotdogs and not hotdogs; whatever you can think of)