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
{{ message }}
This repository has been archived by the owner on Nov 1, 2021. It is now read-only.
I am confused by the meaning of util.makeContiguous and its gradient being defined as zeroGradient(). There is also torch.contiguous() doing IMHO the same job, but having gradient function(g,ans,x) return g end. I have a strong feeling this is wrong. I have noticed this when trying to figure out why there is no backpropagation through torch.view(). I suggest util.makeContiguous get removed completely.
The text was updated successfully, but these errors were encountered:
I am confused by the meaning of
util.makeContiguous
and its gradient being defined aszeroGradient()
. There is alsotorch.contiguous()
doing IMHO the same job, but having gradientfunction(g,ans,x) return g end
. I have a strong feeling this is wrong. I have noticed this when trying to figure out why there is no backpropagation throughtorch.view()
. I suggestutil.makeContiguous
get removed completely.The text was updated successfully, but these errors were encountered: