How to create a weak classifier for the Viola-Jones algorithm?

Question

Mathematics

I'm writing a machine learning project for the purpose of self-education. So far I've made all possible Haar patterns (on the 32*32 image), i.e. white background and black rectangle (all possible sizes) in all possible positions on the image. The next step is to create a weak classifier, but I don’t understand how to do it at all, I read a lot of material, there is almost no information about this anywhere. Maybe I'm, of course, bad looking or something I do not understand.
Please tell me how to do it?
I do face recognition.