Function autograd::ops::stop_gradients [] [src]

pub fn stop_gradients(x: &Tensor) -> Tensor

Stops gradients.

Guarantees that the gradient is not propagated to the tensors behind this during gradient computation.