QuestionsDataset

class dgl.data.QuestionsDataset(raw_dir=None, force_reload=False, verbose=True, transform=None)[source]

Bases: HeterophilousGraphDataset

Questions dataset from the β€˜A Critical Look at the Evaluation of GNNs under Heterophily: Are We Really Making Progress? <https://arxiv.org/abs/2302.11640>’__ paper.

This dataset is based on data from the question-answering website Yandex Q. Nodes are users, and an edge connects two nodes if one user answered the other user’s question. The task is to predict which users remained active on the website (were not deleted or blocked). Node features are the mean of word embeddings for words in the user description. Users that do not have description are indicated by a separate binary feature.

Statistics:

  • Nodes: 48921

  • Edges: 307080

  • Classes: 2

  • Node features: 301

  • 10 train/val/test splits

Parameters:

  • raw_dir (str, optional) – Raw file directory to store the processed data. Default: ~/.dgl/

  • force_reload (bool, optional) – Whether to re-download the data source. Default: False

  • verbose (bool, optional) – Whether to print progress information. Default: True

  • transform (callable, optional) – A transform that takes in a DGLGraph object and returns a transformed version. The DGLGraph object will be transformed before every access. Default: None

num_classes

Number of node classes

Type:

int

Examples

>>> from dgl.data import QuestionsDataset
>>> dataset = QuestionsDataset()
>>> g = dataset[0]
>>> num_classes = dataset.num_classes

>>> # get node features
>>> feat = g.ndata["feat"]

>>> # get the first data split
>>> train_mask = g.ndata["train_mask"][:, 0]
>>> val_mask = g.ndata["val_mask"][:, 0]
>>> test_mask = g.ndata["test_mask"][:, 0]

>>> # get labels
>>> label = g.ndata['label']
__getitem__(idx)

Gets the data object at index.

__len__()

The number of examples in the dataset.