MP Neurons & Perceptron using Python: Binarization

I am not able to understand how changing the labels lead to network getting converged.

Also, I can’t figure out how the professor decided that class label 0 is benign and class label 1 is malignant. Since when you check breast_cancer.target_names it shows as [malignant, benign] which means label 0 is malignant and 1 is benign.
I think the professor did a mistake when he said class 0 is benign and 1 is malignant.

I would be grateful if somebody can shed some information on how to decide the labels. I think it would be difficult to figure out labels just from the data when it becomes complex(multi-labels).

Hi @Mohd_Zaid,

  1. Binarizarion of inputs leads to a simplification of data, which becomes easy for simple models like MP Neuron and Perceptron to generalize.
  2. Labelling can be different, but it merely makes a difference, as we are labelling it for our convenience, but inside a computer it’s still 0 or 1. It doesn’t even know the definition of benign and malignant.