Investigate using the real dataset as input for GNN. It doesn't need to run with all data, and it doesn't have to use the "real" parameters.
In other words, we can set max_distance, min_neighbours, and max_edge_distance to result in a small number of small graphs so that we can run locally.
The objective of this issue is to build the model, while #60 will handle dealing with large amount of data.
Investigate using the real dataset as input for GNN. It doesn't need to run with all data, and it doesn't have to use the "real" parameters. In other words, we can set
max_distance
,min_neighbours
, andmax_edge_distance
to result in a small number of small graphs so that we can run locally.The objective of this issue is to build the model, while #60 will handle dealing with large amount of data.