djl
djl copied to clipboard
[W TensorBody.h:491] Warning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed.
[W TensorBody.h:491] Warning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (function grad)
I found the same error in https://github.com/deepjavalibrary/djl/issues/2133, but there's no specific solution for it. My code does not have any errors; the only solution now is to suppress the output of these warnings, but this does not fundamentally solve the problem, it merely conceals the warnings.