BlogCatalog

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

Bases: dhg.data.base.BaseData

The BlogCatalog dataset is a social network dataset for vertex classification task. This is a network of social relationships of bloggers from the BlogCatalog website, where nodes’ attributes are constructed by keywords, which are generated by users as a short description of their blogs. The labels represent the topic categories provided by the authors.

Note

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

The content of the BlogCatalog dataset includes the following:

  • num_classes: The number of classes: \(6\).

  • num_vertices: The number of vertices: \(5,196\).

  • num_edges: The number of edges: \(343,486\).

  • dim_features: The dimension of features: \(8,189\).

  • features: The vertex feature matrix. torch.Tensor with size \((5,196 \times 8,189)\).

  • edge_list: The edge list. List with length \((343,486 \times 2)\).

  • labels: The label list. torch.LongTensor with size \((5,196, )\).

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.