WalmartTrips

class dhg.data.WalmartTrips(data_root=None)[source]

Bases: dhg.data.base.BaseData

The Walmart Trips dataset is a user-product network dataset for vertex classification task. In Walmart, nodes represent products being purchased at Walmart, and hyperedges equal sets of products purchased together; the node labels are the product categories. More details see this and the YOU ARE ALLSET: A MULTISET LEARNING FRAMEWORK FOR HYPERGRAPH NEURAL NETWORKS paper.

The content of the Walmart Trips dataset includes the following:

  • num_classes: The number of classes: \(12\).

  • num_vertices: The number of vertices: \(88,860\).

  • num_edges: The number of edges: \(69,906\).

  • edge_list: The edge list. List with length \(69,906\).

  • labels: The label list. torch.LongTensor with size \((88,860, )\).

Parameters

data_root (str, optional) – The data_root has stored the data. If set to None, this function will auto-download from server and save into the default direction ~/.dhg/datasets/. Defaults to None.