CPU gradient maps

The reverse sweeps themselves live with the tape in NN.Runtime.Autograd.Engine.Tape. These helpers turn a CPU sweep's result into the sparse map that training code reads.

def Runtime.Autograd.Tape.denseToHashMap {α : Type} [TorchLean.Storage α] (grads : Array (Option (Spec.SomeTensor α))) : Std.HashMap ℕ (Spec.SomeTensor α)

Convert the optional dense gradient array returned by backwardDense into a sparse HashMap.

Only entries that are present (some (some g)) are kept. The result records exactly the nodes reached by reverse-mode propagation.

def Runtime.Autograd.Tape.backward {α : Type} [TorchLean.Storage α] [Add α] (t : Tape (Spec.SomeTensor α)) (outId : ℕ) (seed : Spec.SomeTensor α) : Result (Std.HashMap ℕ (Spec.SomeTensor α))

Reverse-mode backpropagation returning a HashMap of only the nodes that received gradients.

This is the sparse public form of backwardDense: it computes dense gradients first, then drops nodes that did not receive a gradient.

def Runtime.Autograd.Tape.backwardScalar {α : Type} [TorchLean.Storage α] [Add α] [One α] (t : Tape (Spec.SomeTensor α)) (outId : ℕ) : Result (Std.HashMap ℕ (Spec.SomeTensor α))

Backpropagate from a scalar output with seed gradient 1.

PyTorch analogy: loss.backward() when loss is a scalar.