rogermt
/

nsgf-plusplus

Model card Files Files and versions

129 kB

Ctrl+K

Ctrl+K

1 contributor

History: 25 commits

rogermt's picture

Add LEARNING.md — all mistakes, war stories, examples, and principles from NSGF++ reproduction

c95af57 verified 13 days ago

.gitattributes

1.52 kB
initial commit 15 days ago
LEARNING.md

10.8 kB
Add LEARNING.md — all mistakes, war stories, examples, and principles from NSGF++ reproduction 13 days ago
README.md

1.79 kB
Upload README.md 15 days ago
SKILL.md

6.72 kB
Split SKILL.md into SKILL.md (rules) + LEARNING.md (stories/mistakes) + TODO.md (next steps)" 13 days ago
config.yaml

5.52 kB
config.yaml: CIFAR sinkhorn batch 128→32 for T4, pool batches 2500→10000 to compensate 13 days ago
dataset_loader.py

6.47 kB
Fix DataLoader batch size mismatch across training phases + --train-iters now overrides all phases 13 days ago
evaluation.py

9.18 kB
Upload evaluation.py 15 days ago
inference.py

4.82 kB
Upload inference.py 15 days ago
main.py

11.9 kB
main.py: --resume-phase, --checkpoint-dir, --sinkhorn-batch flags 13 days ago
model.py

12.8 kB
Upload model.py 15 days ago
requirements.txt

167 Bytes
Upload requirements.txt 15 days ago
sinkhorn_flow.py

8.93 kB
Fix geomloss tensor shape bug for images + optimize pool sampling 13 days ago
skill-creator.md

33.2 kB
Create skill-creator.md 13 days ago
trainer.py

15.7 kB
Add checkpointing, resume, CIFAR OOM fix, --sinkhorn-batch flag 13 days ago