Calculating gradient of input except the gradient of model parameter

1

Hello,
I’m trying to get the gradient of input but without calculating the gradient of model parameters.
More specifically, there is an input A and this goes into a model M and then output P comes out.
With this output P and the input A, I will make another input B (for example, elementwise multiplication bla bla) and this goes into the model M and then I get final output Q.
Now, the loss function is gonna be formulated with this output Q and given label.

To make things clear, I will do
input A -> model M -> output P (process #1)
C(a result of doing something with an output P and input A) -> model M -> output Q (process #2)
and then I will do

loss = criterion(output Q, label)
loss.backward()

but the point is

  1. the gradient of loss wrt the model M’s parameter in process #2
    should not be saved in model.something.weight.grad and model.something.bias.grad
  2. but the gradient of loss wrt C in process #2 should be calculated and saved in C.grad.

Backpropagation flows through C and goes to process #1.
Here in process #1, on the other hand,

  1. the gradient of loss wrt the model M’s parameter should be calculated and saved.

As far as I know, .detach() makes all the superior parameters(parameters that are closer to the input) not calculate their gradient.
Is there any way to do this?

2

Hi,

The problem is that M is the same during process 1 and 2. You won’t be able to get 2 different behaviors here.

If would suggest doing two steps (add the corresponding optimizer zero grad and step of course):

# Process 1 as usual:
P = M(A)
C = some_func(P)

# Now break the graph and get a new C
C_2 = C.detach().requires_grad_()

# Process 2
Q = M(C_2)
loss = criterion(Q, label)

# Compute grads for C_2 (but don't change .grads in M)
C_2_grad = autograd.grad(loss, C_2)[0]

# Not do a regular backward for process 1 to populate the .grad in M
C.backward(C_2_grad)
3

Thanks a lot. It works!!