==========
图神经网络
==========

Dataset Class for How Powerful Are Graph Neural Networks?.
This is adapted from https://github.com/weihua916/powerful-gnns/blob/master/dataset.zip.
The class provides an interface for nine datasets used in the paper along with the paper-specific settings. The datasets are 'MUTAG', 'COLLAB', 'IMDBBINARY', 'IMDBMULTI', 'NCI1', 'PROTEINS', 'PTC', 'REDDITBINARY', 'REDDITMULTI5K'.

QM7b （化学分子数据）
QM7b dataset for graph property prediction (regression)
This dataset consists of 7,211 molecules with 14 regression targets. Nodes means atoms and edges means bonds. Edge data ‘h’ means the entry of Coulomb matrix.

QM9 dataset for graph property prediction (regression)
This dataset consists of 130,831 molecules with 12 regression targets. Nodes correspond to atoms and edges correspond to close atom pairs.

QM9Edge dataset for graph property prediction (regression)
This dataset consists of 130,831 molecules with 19 regression targets. Nodes correspond to atoms and edges correspond to bonds.

MUTAG
MUTAG dataset for node classification task

KarateClub
Zachary’s karate club is a social network of a university karate club, described in the paper “An Information Flow Model for Conflict and Fission in Small Groups” by Wayne W. Zachary. The network became a popular example of community structure in networks after its use by Michelle Girvan and Mark Newman in 2002. Official website: http://konect.cc/networks/ucidata-zachary/
Karate Club dataset statistics:

• Nodes: 34, Edges: 156, Number of Classes: 2

MiniGCDataset
The synthetic graph classification dataset class.

The datset contains 8 different types of graphs.

PubmedGraph
Pubmed citation network dataset.

CoauthorCS
Computer Science (CS)’ part of the Coauthor dataset for node classification task.

CoauthorPhysics
Physics’ part of the Coauthor dataset for node classification task.

CoraGraph
Cora citation network dataset.

AmazonCoBuyComputer
‘Computer’ part of the AmazonCoBuy dataset for node classification task.

CoraFull
CORA-Full dataset for node classification task.

TuDataset
TUDataset contains lots of graph kernel datasets for graph classification.

Sst'
Stanford Sentiment Treebank dataset.

PPIDataset （蛋白质相互作用数据）

Protein-Protein Interaction dataset for inductive node classification
A toy Protein-Protein Interaction network dataset. The dataset contains 24 graphs. The average number of nodes per graph is 2372. Each node has 50 features and 121 labels. 20 graphs for training, 2 for validation and 2 for testing.
Reference: http://snap.stanford.edu/graphsage/
Statistics: Train examples: 20, Valid examples: 2, Test examples: 2

LegacyTU
LegacyTUDataset contains lots of graph kernel datasets for graph classification.