I am dealing with Yolo Loss Function (the following). $$\begin{align} &\lambda_{coord} \sum_{i=0}^{S^2}\sum_{j=0}^B \mathbb{1}_{ij}^{obj}[(x_i-\hat{x}_i)^2 + (y_i-\hat{y}_i)^2 ] \\&+ \lambda_{coord} \sum_{i=0}^{S^2}\sum_{j=0}^B \mathbb{1}_{ij}^{obj}[(\sqrt{w_i}-\sqrt{\hat{w}_i})^2 +(\sqrt{h_i}-\sqrt{\hat{h}_i})^2 ]\\ &+ \sum_{i=0}^{S^2}\sum_{j=0}^B \mathbb{1}_{ij}^{obj}(C_i - \hat{C}_i)^2 + \lambda_{noobj}\sum_{i=0}^{S^2}\sum_{j=0}^B \mathbb{1}_{ij}^{noobj}(C_i - \hat{C}_i)^2 \\ &+ \sum_{i=0}^{S^2} \mathbb{1}_{i}^{obj}\sum_{c \in classes}(p_i(c) - \hat{p}_i(c))^2 \\ \end{align}$$

I cannot understand how to use it. That is, the "hat" variables are the right ones (related to the training set), while the standard variables are the predicted ones. However, it has some sense if there is a one-to-one corrispondence between the two kind of variables.

But what should happens if my model predicts an airplane while there are not?

I know the predicted center ($x_i,y_i$) and the predicted whidth and height ($w_i,h_i$) but since there are no airplanes what sould be the related hat variables?

Also, what is the meaning of $\hat{p}_i(c)$ since I have the certainess that that object belong to that class? It should be 1?

Thanks in advance :)

  • $\begingroup$ Usually, the "hat" variables are the predicted ones and the "standard" ones are the right ones. Are you sure it is different in your case? $\endgroup$ – nope Jan 7 '19 at 12:22
  • 1
    $\begingroup$ No, but it doesn't matter. They're interchangable. $\endgroup$ – aleio1 Jan 7 '19 at 13:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.