Building makemore Part 3: Activations & Gradients, BatchNorm

569,207 views

About this video

Hand-picked by a BrighterTube curator · found via Developer-Y/cs-video-courses · 94% confidence

Rare, practical depth on why deep networks train or fail, with diagnostics viewers can reuse.

Makemore part 3: activations, gradients and batch normalisation — diagnosing a network's health from its statistics.

Video by Andrej Karpathy, played through YouTube's embedded player.

Part of Neural Networks: Zero to Hero