unable to explicitly pass reduce and size_average in loss function, what am I doing wrong?
Which version of Pytorch are you using? The reduce argument was introduced in v0.3 as far as I know.
You have to define the
size_average param when creating the loss:
loss = nn.MSELoss(size_average=True, reduce=False)
Sorry, haven’t noticed this issue in your first post.
oh you’re right, sorry!