So for example, let’s just say there is some unknown function of x, y=f(x). Now I can train the model with x and y and it would predict y when given x. But i don’t want exactly y, instead i want will y be greater than x or not? So the output should be 1 if it greater 0 if equal or lesser. How can I do this? So my data has 3 parameters: x, y, e(1 or 0). I need to train my model by giving out x and y and results as e. But then when I want predict i have to give x and y. Right? I want to give only x and get output 0 or 1. I could just train it with x as input and y as results and predict y from x and check them by myself but that is useless since I don’t want to know y it is just going to be waste of computational power.