Autograd, Loss, Backpropagation, and Training Loops

demystify how neural networks learn.

Units

1. Unit 03.00: 03.00 Forward pass and loss
2. Unit 03.01: 03.01 Gradients and `requires_grad`
3. Unit 03.02: 03.02 Backpropagation intuition without heavy derivation
4. Unit 03.03: 03.03 Optimizer step and zeroing gradients
5. Unit 03.04: 03.04 Building a minimal training loop
6. Unit 03.05: 03.05 Saving metrics and checkpoints
7. Unit 03.06: 03.06 Project step: train a tiny model end to end

Module work

• Activity
• Knowledge check
• Summary