Working on Graph data

[JasonStraka]

Hi, thank you for your nice work and for sharing the code.

The datasets used in expernments are based on grid structure with Shape: $(N \times T \times H \times W)$ , but in README (which says UniST supports Data Format of Grid/Graph). How to Work on Graph data like METR-LA (Metro Traffic Los Angeles) with shape [N, T, C] (Number of nodes, time length, feature). There is no grid structure in raw dataset METR-LA.

[YuanYuan98]

To adapt UniST to graph-based data, you can adjust the spatial patch size to 1. This will allow you to apply the same spatio-temporal patching technique that is used for grid-based data.

Specifically, you would reshape the data to have a shape of $N\times T \times 1 \times W$, where $W$ is the number of nodes.

This will enable UniST to process the graph data by treating each node as a single spatial unit over time.

[HuangYH0921]

Hello, may I ask how to reflect the relationship between the spatial positions of each point after the data is reshaped into the shape of N × T × 1 × W? For example, how do I input the adjacency matrix?

[ZhangXiangyu23]

Hello, may I ask how to reflect the relationship between the spatial positions of each point after the data is reshaped into the shape of N × T × 1 × W? For example, how do I input the adjacency matrix?

Hi, I have the same question. According to the modifications provided by the authors, this four-dimensional tensor can represent time and features, but it cannot represent the connectivity between topologies (i.e. the adjacency matrix). Friend, have you solved this problem yet? Is it convenient to communicate?

[ZhangXiangyu23]

To adapt UniST to graph-based data, you can adjust the spatial patch size to 1. This will allow you to apply the same spatio-temporal patching technique that is used for grid-based data.

Specifically, you would reshape the data to have a shape of N × T × 1 × W , where W is the number of nodes.

This will enable UniST to process the graph data by treating each node as a single spatial unit over time.

Hello, author. The modification you mentioned, my understanding is: W-dimension data is the characteristic value corresponding to each graph node, that is, W= the number of topology graph nodes. H=1, N is the number of samples, and T is the length of time. This represents only the features in the topology, but how do you represent the adjacencies in the topology? Thank you！