14 datasets found

The dataset used in the paper is a real-world dataset for future link prediction task on dynamic graphs.

The TUDataset is a collection of benchmark datasets for learning with graphs. It contains 14 datasets, each with a different type of graph.

Graph pooling is an essential component of the architecture for many graph-level tasks, such as graph classification and graph generation.

COX2 is a molecular dataset where every graph is classified into one of two classes.

ogbg-molhiv dataset is a graph classification dataset containing 41k molecules.

Open Graph Benchmark (OGB) dataset contains many large-scale benchmark datasets.

Benchmarking graph neural networks.

NELL is a large-scale, open knowledge graph that contains a vast amount of structured data.

Revisiting link prediction: a data perspective.

A synthetic dataset generated for benchmarking GRAPHSHAP, consisting of graphs with controlled motifs injection.

The dataset is used for testing the proposed TopNets architecture on graph classification tasks.

The ENZYMES dataset contains 600 proteins from each of the 6 Enzyme Commission top-level enzyme classes (EC classes) and the goal was to correctly predict enzyme class...

A dataset for learning distributed representations of graphs using discrete substructure patterns and neural language models.

The dataset used in the paper is a synthetic dataset for graph classification tasks.