🌊 SEA-ViT: Sea Surface Currents Forecasting with Vision Transformers and GRUs

"SEA-ViT: Forecasting Sea Surface Currents Using a Vision Transformer and GRU-Based Spatio-Temporal Covariance Model"
πŸ“Œ Accepted at IEEE KST 2025 | πŸ”¬ Developed by Teerapong Panboonyuen (Kao)

πŸš€ Highlights


πŸ—‚οΈ Project Structure

sea-vit/
β”œβ”€β”€ src/
β”‚   β”œβ”€β”€ model.py         # SEA-ViT Model: ViT + BiGRU
β”‚   β”œβ”€β”€ train.py         # Training Pipeline
β”‚   β”œβ”€β”€ inference.py     # Run Inference on Radar Sequences
β”‚   β”œβ”€β”€ metrics.py       # Evaluation (RMSE, MAE, Corr)
β”‚   └── utils.py         # Dataset I/O, Preprocessing, Logging
β”œβ”€β”€ Dockerfile           # Reproducible Environment
β”œβ”€β”€ requirements.txt     # Python Dependencies
β”œβ”€β”€ README.md            # Project Overview (this file)
└── data/                # Example .npz radar data (optional)

πŸ“₯ Installation

Install dependencies via pip:

git clone https://github.com/kaopanboonyuen/GISTDA-SeaViT.git
cd GISTDA-SeaViT
pip install -r requirements.txt

Or build the full environment via Docker:

docker build -t sea-vit .

πŸ’» How to Run the Code

πŸ”§ 1. Train SEA-ViT on your dataset

Assuming your .npz dataset includes radar sequences + targets:

python src/train.py \
  --data-path ./data/train.npz \
  --batch-size 8 \
  --epochs 50 \
  --lr 1e-4

Input shape: [N, T, 2, H, W] for sequences
Target shape: [N, forecast_steps, 2] for (U, V)

🧠 2. Inference on a New Sequence

python src/inference.py \
  --model-path ./checkpoints/sea-vit.pth \
  --input-path ./data/sample_input.npz \
  --output-path ./output/forecast.npy

πŸ“ 3. Evaluate Metrics (RMSE, MAE, Correlation)

python src/metrics.py \
  --predictions ./output/forecast.npy \
  --ground-truth ./data/sample_ground_truth.npy

πŸ“ˆ Results Summary

Metric Value (Example)
RMSE 0.087
MAE 0.065
Corr 0.91

SEA-ViT demonstrates high accuracy across diverse oceanic conditions, outperforming CNN-GRU and ConvLSTM baselines.


πŸ“š Learn More


πŸ”¬ Citation

@inproceedings{panboonyuen2025sea,
  title={SEA-ViT: Forecasting Sea Surface Currents Using a Vision Transformer and GRU-Based Spatio-Temporal Covariance Model},
  author={Panboonyuen, Teerapong},
  booktitle={2025 17th International Conference on Knowledge and Smart Technology (KST)},
  pages={1--6},
  year={2025},
  organization={IEEE}
}

πŸ“« Contact

Teerapong Panboonyuen (Kao)
Postdoctoral Researcher, Chulalongkorn University
Senior Research Scientist, MARSAIL
πŸ“§ teerapong.panboonyuen@gmail.com


🀝 Contributing

Pull requests welcome! For feature ideas or bug fixes, feel free to open an issue.


β€œSEA-ViT doesn’t just see the ocean β€” it foresees it.” πŸŒŠπŸ“‘