| tags: | |
| - image-classification | |
| - cantor-fusion | |
| - geometric-deep-learning | |
| - safetensors | |
| - vision-transformer | |
| - warm-restarts | |
| - geometric-coalescence | |
| library_name: pytorch | |
| datasets: | |
| - cifar10 | |
| - cifar100 | |
| metrics: | |
| - accuracy | |
| # vit-beans-v3 | |
| **Geometric Deep Learning with Cantor Multihead Fusion + Shatter-Reconstruct Training** | |
| This repository contains training runs using Cantor fusion architecture with: | |
| - Pentachoron (5-simplex) structures for geometric routing | |
| - CosineAnnealingWarmRestarts for exploration cycles | |
| - GeometricCoalescenceLoss for shatter-reconstruct training | |
| ### π LR Boost + Geometric Coalescence | |
| This run uses **restart_lr_mult = 1.15x** with **GeometricCoalescenceLoss**: | |
| - LR boosts create aggressive exploration cycles | |
| - Coalescence loss provides geometric scaffolding during weight thrashing | |
| - Adaptive weighting: 0.1 β 0.8 during LR spikes | |
| - Model reconstructs from geometric first principles when patterns shatter | |
| ## Current Run | |
| **Latest**: `cifar100_weighted_ADAMW_WarmRestart_boost1.15x_coal0.5_20251124_152227` | |
| - **Dataset**: CIFAR100 | |
| - **Fusion Mode**: weighted | |
| - **Coalescence**: Ξ»=0.5 β | |
| - **LR Boost**: 1.15x π | |
| --- | |
| **Repository maintained by**: [@AbstractPhil](https://huggingface.co/AbstractPhil) | |