AbstractPhil
/

vit-beans-v3

Image Classification

geometric-deep-learning

vision-transformer

geometric-coalescence

Model card Files Files and versions

Metrics Training metrics Community

AbstractPhil commited on Nov 24, 2025

Commit

dc551fe

·

verified ·

1 Parent(s): 43747e4

Upload README.md with huggingface_hub

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -34,9 +34,9 @@ This run uses **restart_lr_mult = 1.15x** with **GeometricCoalescenceLoss**:
 ## Current Run
-**Latest**: `cifar100_learned_ADAMW_WarmRestart_boost1.15x_coal0.5_20251124_011239`
 - **Dataset**: CIFAR100
-- **Fusion Mode**: learned
 - **Coalescence**: λ=0.5 ✓
 - **LR Boost**: 1.15x 🚀

 ## Current Run
+**Latest**: `cifar100_weighted_ADAMW_WarmRestart_boost1.15x_coal0.5_20251124_152227`
 - **Dataset**: CIFAR100
+- **Fusion Mode**: weighted
 - **Coalescence**: λ=0.5 ✓
 - **LR Boost**: 1.15x 🚀