You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
page 6 from your paper: Precisely, Focal-R loss based on L1 distance can be written as 1/n∑n i=1 σ(|βei|)γ ei, where ei is the L1 error for i-th sample, σ(·)
This is a good catch. Since there's already a hyper-parameter beta to control the "scaling" effect, we did not use the squared error -- but one could always try that (and tune the value of beta correspondingly).
QUESTION 2 : why is there 2*torch.abs(...)-1
This is because sigmoid(x) is in [0.5, 1] for all x>=0, so we simple scale it back to [0, 1]. If you choose activation as 'tanh', you do not need to rescale it. Since it's an engineering issue, for cleanness, we did not expand the detail in paper.
Hi authors,
page 6 from your paper:
Precisely, Focal-R loss based on L1 distance can be written as 1/n∑n i=1 σ(|βei|)γ ei, where ei is the L1 error for i-th sample, σ(·)
imbalanced-regression/agedb-dir/loss.py
Line 24 in 055a7b3
should be torch.abs((inputs - targets)**2) and not only torch.abs(inputs - targets), am I correct?
2*torch.abs(...)-1
? you do not have and -1 or 2* in the function in your paper?The text was updated successfully, but these errors were encountered: