PyTorch autograd (loss.backward)

Tool / ProductMentioned in 1 video

The built-in automatic differentiation mechanism in PyTorch; the lecturer argues for replacing calls to loss.backward with a manual tensor-level backward pass for pedagogical and debugging reasons.

Videos Mentioning PyTorch autograd (loss.backward)

Building makemore Part 4: Becoming a Backprop Ninja

Building makemore Part 4: Becoming a Backprop Ninja

Andrej Karpathy

The built-in automatic differentiation mechanism in PyTorch; the lecturer argues for replacing calls to loss.backward with a manual tensor-level backward pass for pedagogical and debugging reasons.