Flickr

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

Bases: dhg.data.base.BaseData

The Flickr dataset is a social network dataset for vertex classification task. It is a social network where nodes represent users and edges correspond to friendships among users. The labels represent the interest groups of the users.

Note

The L1-normalization for the feature is not recommended for this dataset.

The content of the Flickr dataset includes the following:

  • num_classes: The number of classes: \(9\).

  • num_vertices: The number of vertices: \(7,575\).

  • num_edges: The number of edges: \(479,476\).

  • dim_features: The dimension of features: \(12,047\).

  • features: The vertex feature matrix. torch.Tensor with size \((7,575 \times 12,047)\).

  • edge_list: The edge list. List with length \((479,476 \times 2)\).

  • labels: The label list. torch.LongTensor with size \((7,575, )\).

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.