CIFAR-10 CNN を構成する全部品。Conv → Pool → GAP の forward + 基本 data augmentation (Normalize / HFlip / RandomCrop) のセット。
7 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 | 22 | 22_conv2d.ipynb — 2D Convolution |
Medium | ↗ |  |
| 2 | 41 | 41_maxpool2d.ipynb — 2D Max Pooling |
Medium | ↗ | |
| 3 | 49 | 49_avg_pool2d.ipynb — 2D Average Pooling |
Easy | ↗ | |
| 4 | 50 | 50_global_avg_pool.ipynb — Global Average Pooling |
Easy | ↗ | |
| 5 | 42 | 42_normalize.ipynb — Per-Channel Normalize |
Easy | ↗ | |
| 6 | 43 | 43_random_hflip.ipynb — Random Horizontal Flip |
Easy | ↗ | |
| 7 | 44 | 44_random_crop.ipynb — Random Crop with Padding |
Easy | ↗ | |
# from the repo root
make run # or `docker compose up`
# then in JupyterLab, navigate to practice/W4/ 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.