torch_tensor_backward Interface

public interface torch_tensor_backward

Calls

interface~~torch_tensor_backward~~CallsGraph interface~torch_tensor_backward torch_tensor_backward proc~torch_tensor_backward_with_external_gradient torch_tensor_backward_with_external_gradient interface~torch_tensor_backward->proc~torch_tensor_backward_with_external_gradient proc~torch_tensor_backward_without_external_gradient torch_tensor_backward_without_external_gradient interface~torch_tensor_backward->proc~torch_tensor_backward_without_external_gradient proc~torch_tensor_get_rank torch_tensor%torch_tensor_get_rank proc~torch_tensor_backward_without_external_gradient->proc~torch_tensor_get_rank proc~torch_tensor_get_shape torch_tensor%torch_tensor_get_shape proc~torch_tensor_backward_without_external_gradient->proc~torch_tensor_get_shape proc~torch_tensor_get_shape->proc~torch_tensor_get_rank

Module Procedures

public subroutine torch_tensor_backward_with_external_gradient(tensor, external_gradient, retain_graph)

Performs back-propagation on a Torch Tensor, given some external gradient.

Arguments

Type IntentOptional Attributes Name
type(torch_tensor), intent(in) :: tensor

Tensor to compute gradients of
type(torch_tensor), intent(in) :: external_gradient

External tensor used as an initial scaling of the gradient calculation
logical, intent(in), optional :: retain_graph

Should the computational graph be retained?

public subroutine torch_tensor_backward_without_external_gradient(tensor, retain_graph)

Performs back-propagation on a Torch Tensor, with an assumed external_gradient of ones.

Arguments

Type IntentOptional Attributes Name
type(torch_tensor), intent(in) :: tensor

Tensor to compute gradients of
logical, intent(in), optional :: retain_graph

Should the computational graph be retained?