Current: Course overview

0%

Course workspace

AI Made Simple · From the basics to Mini GPT

Learn from data, parameters, and gradients through to a working Mini GPT. Work through the lessons in order, following the calculations and trying the exercises yourself.

0%Course progress · 0 of 321 sections
Start Week 1

12-week learning path

  1. Week 1Week 1 — Essential mathematical intuition: learn to read AI formulas
  2. Week 2Week 2 — Linear regression: from prediction to gradient descent
  3. Week 3Week 3 — Neural networks: from one neuron to multiple layers
  4. Week 4Week 4 — Backpropagation: learning across all the network's parameters
  5. Week 5Week 5 — Tensors and PyTorch: let the computer do the calculations
  6. Week 6Week 6 — From tokens to a small language model: understanding prediction through one example
  7. Week 7Week 7 — Attention: read the relevant preceding context
  8. Week 8Week 8 — Transformers: build a trainable network around attention
  9. Week 9Week 9 — Tokenizers: from raw text to training batches
  10. Week 10Week 10 — GPT architecture: assemble a trainable Mini GPT
  11. Week 11Week 11 — Training and inference: learn, evaluate, save, and generate with the same Mini GPT
  12. Week 12Week 12 — End-to-end Mini GPT: from text and training to generation after loading