README.md

---
license: mit
tags:
- image-classification
- vit
- bittensor-sn72
- roadwork
base_model: natix-network-org/roadwork
---

# Roadwork classifier (fine-tuned for Bittensor SN72)

Fine-tuned from `natix-network-org/roadwork` against the validator augmentation pipeline
in `natix/utils/image_transforms.py` (rotation/flips/JPEG/blur/noise/color distortions).

## Training
- 4 epochs, AdamW lr=2e-5 cosine schedule, class weights for the 4:1 label imbalance
- Filter rule from validator's image cache applied to keep only consistent (description, label) pairs

## Eval (augmented test set, 2,263 examples)

| Metric | Baseline | This model | Cubic reward gain |
|---|---|---|---|
| Accuracy | 0.9633 | 0.9806 | |
| MCC | — | 0.8888 | |
| Reward proxy | 0.8564 | 0.9347 | +30.0% |