问题
What is the correct way to perform gradient clipping in pytorch?
I have an exploding gradients problem, and I need to program my way around it.
回答1:
clip_grad_norm
(which is actually deprecated in favor of clip_grad_norm_
following the more consistent syntax of a trailing _
when in-place modification is performed) clips the norm of the overall gradient by concatenating all parameters passed to the function, as can be seen from the documentation:
The norm is computed over all gradients together, as if they were concatenated into a single vector. Gradients are modified in-place.
From your example it looks like that you want clip_grad_value_ instead which has a similar syntax and also modifies the gradients in-place:
clip_grad_value_(model.parameters(), clip_value)
Another option is to register a backward hook. This takes the current gradient as an input and may return a tensor which will be used in-place of the previous gradient, i.e. modifying it. This hook is called each time after a gradient has been computed, i.e. there's no need for manually clipping once the hook has been registered:
for p in model.parameters():
p.register_hook(lambda grad: torch.clamp(grad, -clip_value, clip_value))
回答2:
A more complete example
optimizer.zero_grad()
loss, hidden = model(data, hidden, targets)
loss.backward()
torch.nn.utils.clip_grad_norm_(model.parameters(), args.clip)
optimizer.step()
Source: https://github.com/pytorch/pytorch/issues/309
回答3:
Reading through the forum discussion gave this:
clipping_value = 1 # arbitrary value of your choosing
torch.nn.utils.clip_grad_norm(model.parameters(), clipping_value)
I'm sure there is more depth to it than only this code snippet.
来源:https://stackoverflow.com/questions/54716377/how-to-do-gradient-clipping-in-pytorch