Is Gradient Clippping in the code as it is on the paper?

[sangmain]

Is Gradient clipping gr = gq1|r|≤1 still used in the code?
The only part I see clipping is p.org.copy_(p.data.clamp_(-1,1)) in def train():

optimizer.zero_grad()
loss.backward()
for p in list(model.parameters()):
if hasattr(p,'org'):
p.data.copy_(p.org)
optimizer.step()
for p in list(model.parameters()):
if hasattr(p,'org'):
p.org.copy_(p.data.clamp_(-1,1))
If it is a gradient clipping, shouldn't that be used before optimizer.step() ?
And I also don't get the meaning of p.org.copy_(p.data.clamp_(-1,1)) since p.org is Binarized later afterall (Same result if p.data is not clamped).
Thank you

[whubaichuan]

good question. there is no gradient clip i think.