This repo implements both traditional statistical interpolation methods and modern graph neural network(including train and inference) approaches to address the challenge of two-dimensional geospatial plane interpolation.
The methodologies used in this project are detailed below:
- ADW (Angular Distance Weighting Interpolation) https://dl.acm.org/doi/10.1145/800186.810616
- IDW (Inverse Distance Weighting Interpolation)
- Kriging
- GAT (Graph Attention Networks)
- GCN (Graph Convolutional Networks)
- KCN (Kriging Convolutional Networks) https://ojs.aaai.org/index.php/AAAI/article/view/5716
- Add some usage examples
- visualize the interpolated graph