# Understanding backprop for softmax

I'm looking on a given solution of the first assignment of cs231n course.

Down below a snippet from the loss function. I don't really understand lines 140-143. Can you explain why dscores (the derivative of scores) is calculated like that?

• What is y? and N in conjunction to lim_scores? Dec 22 '18 at 21:04
• I was looking for the answer to this as well. I think the answer is in section Computing the Analytic Gradient with Backpropagation of this link.
– CaTx
Sep 3 at 13:47

Be aware that posting code in images very annoying to copy/paste and it's bad for web reference ment.

This is due to the derivative of the softmax, but to me it's seems fishy.

If $$S$$ is the softmax vector, then the Jacobian $$DS$$ consists of $$S_j(\delta_{ij}-S_i)$$. This could explain the -=1 part, but not the /=N, and not the shape either.