RuntimeError: element 0 of variables does not require grad and does not have a grad_fn

[srishti-git1110 (Srishti Gureja)]

The following code is the source of your error -

@Dariush_Rnzdh

pred=(pred>.5).float()
loss=torch.where(yb==1,1-pred,pred).mean()#.requires_grad_()
   

You are essentially re-creating pred here which causes it to lose its computation graph.
It looks like a classification problem; try exploring loss functions like this one rather than defining your own like you’ve done here which also naturally has its requires_grad=False.