Calculating the divergence

[pascal_notsawo (Tikquuss)]

Maybe this will help you :

How to compute Jacobian matrix in PyTorch?

Here are some functions that can help you. import torch def gradient(y, x, grad_outputs=None): """Compute dy/dx @ grad_outputs""" if grad_outputs is None: grad_outputs = torch.ones_like(y) grad = torch.autograd.grad(y, [x], grad_outputs = grad_outputs, create_graph=True)[0] return grad def jacobian(y, x): """Compute dy/dx = dy/dx @ grad_outputs; for grad_outputs in [1, 0, ..., 0], [0, 1, 0, ..., 0], ...., [0, ..., 0, 1]""" jac = torch.zeros(y.shape[0], x.s…