conf_loss = cross_entropy_loss(conf_preds.view(-1,num_classes),conf_targets.view(-1))
The shape of x and y are
X: torch.Size([69856, 40]) , Y: torch.Size([69856])
respectively.
Author has mentioned the size as x:[N,D] and y:[N,]
but my y size is [N] I need to calculate the difference but I am getting out of memory . Can anyone help with the dimension? I should get the final shape after taking the difference as [N,]. I need to calculate this print(log_sum_exp - x.gather(1, y.view(-1,1)))
def cross_entropy_loss(x, y):
'''Cross entropy loss w/o averaging across all samples.
Args:
x: (tensor) sized [N,D].
y: (tensor) sized [N,].
Return:
(tensor) cross entroy loss, sized [N,].
'''
print("X:",x.shape)
print("Y:",y.shape)
xmax = x.data.max()
log_sum_exp = torch.log(torch.sum(torch.exp(x-xmax), 1)) + xmax
print(log_sum_exp.shape) # torch.Size([69856])
print(x.gather(1,y.view(-1,1)).shape) #torch.Size([69856, 1])
#print(log_sum_exp - x.gather(1, y.view(-1,1)))
#return log_sum_exp - x.gather(1, y.view(-1,1))
