This might be a dumb question. I don’t quite get the concept behind the equation of dice similarity coefficient. For the numerator, why is A intersect B translate to sum of y_pred*y_true. Does it mean the intersection equal to the sum of two matrix multiplication? Thank you!
Hi @Qichao_Wu,
the dice similarity coefficient equals the size of the intersection between A and B divided by the the size of the union of A and B. That is why we sum up the multiplication of y_pred * y_true in the numerator. So, yes your line of thought is correct.
Hope that helps!
Thank you for the explanation, Samuel! It’s helpful 