| Age | Commit message (Expand) | Author |
|---|---|---|
| 2026-03-27 | fix(cnn_v3): L1 loss + depth-grad tanh normalization to reduce flat convergence | skal |
| 2026-03-27 | fix(cnn_v3): remove dec0 ReLU, load FiLM MLP at runtime | skal |
| 2026-03-26 | fix(cnn_v3/training): fix defaults and help strings across py tools | skal |
| 2026-03-26 | feat(cnn_v3): upgrade architecture to enc_channels=[8,16] | skal |
| 2026-03-25 | fix(cnn_v3/training): rebuild optimizer before loading state on resume past F... | skal |
| 2026-03-25 | feat(cnn_v3): 3×3 dilated bottleneck + Sobel loss + FiLM warmup + architectu... | skal |
| 2026-03-25 | feat(cnn_v3/training): add --single-sample option + doc fixes | skal |
| 2026-03-22 | feat(cnn_v3): patch alignment search, resume, Ctrl-C save | skal |
| 2026-03-22 | docs(cnn_v3): add uv inline deps to train_cnn_v3.py + HOW_TO_CNN note | skal |
| 2026-03-22 | fix(cnn_v3): correct weight budget in docstring (3.9→5.4 KB f16) | skal |
| 2026-03-21 | feat(cnn_v3): Phase 6 — training script (train_cnn_v3.py + cnn_v3_utils.py) | skal |
 |
index : demo.git | |
| Vide-coded 64k demo system |
| summaryrefslogtreecommitdiff |