[solved] Getting batches of gradients

[tor]

@Giuseppe_Castiglione: could you please elaborate your answer more?
How did you use torch.autograd.grad() to get the gradient of each individual input vector from a batch?
Some code snippet will be appreciated.

One relevant discussion is

Efficient computation of per-sample examples

Additional discussion of [feature request] Simple and Efficient way to get gradients of each element of a sum @tom I hope I interpreted your last post correctly by posting here. The code to reproduce the numbers I see is available here. The main.py file runs different ways of computing individual gradients and checks/times them. Your recommendation seems to be along the line of Goodfellow’s derivation, implemented as goodfellow, timing output under goodf. I don’t think it is a bug per-se, …