README.md

---
tags:
- traffic-forecasting
- time-series
- graph-neural-network
- stgformer_final
datasets:
- metr-la
---

# Spatial-Temporal Graph Transformer (Final) - METR-LA

Spatial-Temporal Graph Transformer (Final) (STGFORMER_FINAL) trained on METR-LA dataset for traffic speed forecasting.

## Model Description

STGFormer Chebyshev+TCN with Xavier initialization, DOW embeddings, exclude_missing_from_norm, and sparsity_k=16 [FINAL - 100 epochs]



## Dataset

**METR-LA**: Traffic speed data from highway sensors.

## Usage

```python
from utils.stgformer import load_from_hub

# Load model from Hub
model, scaler = load_from_hub("METR-LA", hf_repo_prefix="STGFORMER_FINAL")

# Get predictions
from utils.stgformer import get_predictions
predictions = get_predictions(model, scaler, test_dataset)
```

## Training

Model was trained using the STGFORMER_FINAL implementation with default hyperparameters.

## Citation

If you use this model, please cite the original STGFORMER_FINAL paper:

```bibtex
@inproceedings{lan2022stgformer,
  title={STGformer: Spatial-Temporal Graph Transformer for Traffic Forecasting},
  author={Lan, Shengnan and Ma, Yong and Huang, Weijia and Wang, Wanwei and Yang, Hui and Li, Peng},
  booktitle={IEEE Transactions on Neural Networks and Learning Systems},
  year={2022}
}
```

## License

This model checkpoint is released under the same license as the training code.