MLP の forward → activation → loss → optimizer → 評価指標 までの一周。基本的な MNIST / 分類パイプラインの構成部品を一つずつ自前実装する週。
8 problems. Solve each template in place, then run the last cell check("...") to grade. Stuck? Open the linked solution in the rightmost column.
| Order | # | Problem | Difficulty | Solution | Colab |
|---|---|---|---|---|---|
| 1 | 40 | 40_linear_regression.ipynb — Linear Regression |
Medium | ↗ |  |
| 2 | 03 | 03_linear.ipynb — Simple Linear Layer |
Medium | ↗ | |
| 3 | 01 | 01_relu.ipynb — Implement ReLU |
Easy | ↗ | |
| 4 | 02 | 02_softmax.ipynb — Implement Softmax |
Easy | ↗ | |
| 5 | 16 | 16_cross_entropy.ipynb — Cross-Entropy Loss |
Easy | ↗ | |
| 6 | 53 | 53_nll_loss.ipynb — Negative Log-Likelihood Loss |
Easy | ↗ | |
| 7 | 54 | 54_sgd_momentum.ipynb — SGD with Momentum |
Medium | ↗ | |
| 8 | 52 | 52_top_k_accuracy.ipynb — Top-k Accuracy |
Easy | ↗ | |
# from the repo root
make run # or `docker compose up`
# then in JupyterLab, navigate to practice/W2/ and open a notebookIf you accidentally break a template, delete it and rerun python scripts/build_weeks.py to restore — it only copies missing files, so your other in-progress notebooks are preserved.