Building makemore Part 4: Becoming a Backprop Ninja (opens in new tab)

<iframe id="ytplayer" type="text/html" width="640" height="360" src="https://www.youtube-nocookie.com/embed/q8SA3rM6ckI" frameborder="0" allowfullscreen="" referrerpolicy="strict-origin-when-cross-origin"></iframe><br><span style="white-space: pre-wrap;">We take the 2-layer MLP (with BatchNorm) from the previous video and backpropagate through it manually without using PyTorch autograd's loss.backward(): through the cross entropy loss, 2nd...</span>