- GraphData
import os.path as osp
import numpy as np
def load_npz(filepath):
filepath = osp.abspath(osp.expanduser(filepath))
if not filepath.endswith('.npz'):
filepath = filepath + '.npz'
if osp.isfile(filepath):
with np.load(filepath, allow_pickle=True) as loader:
loader = dict(loader)
for k, v in loader.items():
if v.dtype.kind in {'O', 'U'}:
loader[k] = v.tolist()
return loader
else:
raise ValueError(f"{filepath} doesn't exist.")e.g., run load_npz('cora') and it returns a dict instance loader, it might have the following keys:
adj_matrix: scipy.sparse.csr_matrix, adjacency matrix. NOTE: the adjacency matrix might not be symmetric.node_attr: scipy.sparse.csr_matrix or numpy.ndarray, node attribute matrixnode_label: scipy.sparse.csr_matrix or numpy.ndarray, node labelsmetadata: dict, additional metadata such as text.
| name | num_nodes | num_edges | num_attrs | density | is_directed |
|---|---|---|---|---|---|
| karate_club | 34 | 78 | 0 | 6.7474% | 0 |
| polblogs | 1,490 | 19,025 | 0 | 0.8569% | 1 |
| cora | 2,708 | 5,429 | 1,433 | 0.0740% | 1 |
| cora_ml | 2,995 | 8,416 | 2,879 | 0.0938% | 1 |
| acm | 3,025 | 13,128 | 1,870 | 0.1435% | 0 |
| uai | 3,067 | 28,314 | 4,973 | 0.3010% | 0 |
| citeseer | 3,312 | 4,715 | 3,703 | 0.0430% | 1 |
| citeseer_full | 4,230 | 5,358 | 602 | 0.0299% | 1 |
| blogcatalog | 5,196 | 171,743 | 8,189 | 0.6361% | 0 |
| flickr | 7,575 | 239,738 | 12,047 | 0.4178% | 0 |
| amazon_photo | 7,650 | 143,663 | 745 | 0.2455% | 1 |
| amazon_cs | 13,752 | 287,209 | 767 | 0.1519% | 1 |
| dblp | 17,716 | 52,867 | 1,639 | 0.0168% | 0 |
| coauthor_cs | 18,333 | 81,894 | 6,805 | 0.0244% | 0 |
| pubmed | 19,717 | 44,324 | 500 | 0.0114% | 0 |
| cora_full | 19,793 | 65,311 | 8,710 | 0.0167% | 1 |
| coauthor_phy | 34,493 | 247,962 | 8,415 | 0.0208% | 0 |
citation network and knowledge graph(NELL) datasets in https://github.com/kimiyoung/planetoid
nodes are documents and edges are citation links. Label rate denotes the number of labeled nodes that are used for training divided by the total number of nodes in each dataset.
@article{SenNBGGE08,
author = {Prithviraj Sen and
Galileo Namata and
Mustafa Bilgic and
Lise Getoor and
Brian Gallagher and
Tina Eliassi{-}Rad},
title = {Collective Classification in Network Data},
journal = {{AI} Mag.},
volume = {29},
number = {3},
pages = {93--106},
year = {2008}
}Amazon Computers and Amazon Photo are segments of the Amazon co-purchase graph,where nodes represent goods, edges indicate that two goods are frequently bought together, node features are bag-of-words encoded product reviews, and class labels are given by the product category.
@inproceedings{McAuleyTSH15,
author = {Julian J. McAuley and
Christopher Targett and
Qinfeng Shi and
Anton van den Hengel},
editor = {Ricardo Baeza{-}Yates and
Mounia Lalmas and
Alistair Moffat and
Berthier A. Ribeiro{-}Neto},
title = {Image-Based Recommendations on Styles and Substitutes},
booktitle = {Proceedings of the 38th International {ACM} {SIGIR} Conference on
Research and Development in Information Retrieval, Santiago, Chile,
August 9-13, 2015},
pages = {43--52},
publisher = {{ACM}},
year = {2015}
}Coauthor CS and Coauthor Physics are co-authorship graphs based on the Microsoft Academic Graph from the KDD Cup 2016 challenge. Here, nodes are authors, that are connected by an edge if they co-authored a paper; node features represent paper keywords for each author’s papers, and class labels indicate most active fields of study for each author.
https://kddcup2016.azurewebsites.net/
The above datasets are collected from
https://github.com/shchur/gnn-benchmark
@article{shchur2018pitfalls,
title={Pitfalls of Graph Neural Network Evaluation},
author={Shchur, Oleksandr and Mumme, Maximilian and Bojchevski, Aleksandar and G{\"u}nnemann, Stephan},
journal={Relational Representation Learning Workshop, NeurIPS 2018},
year={2018}
}@inproceedings{PanWZZW16,
author = {Shirui Pan and
Jia Wu and
Xingquan Zhu and
Chengqi Zhang and
Yang Wang},
editor = {Subbarao Kambhampati},
title = {Tri-Party Deep Network Representation},
booktitle = {Proceedings of the Twenty-Fifth International Joint Conference on
Artificial Intelligence, {IJCAI} 2016, New York, NY, USA, 9-15 July
2016},
pages = {1895--1901},
publisher = {{IJCAI/AAAI} Press},
year = {2016}
}@inproceedings{GilesBL98,
author = {C. Lee Giles and
Kurt D. Bollacker and
Steve Lawrence},
title = {CiteSeer: An Automatic Citation Indexing System},
booktitle = {Proceedings of the 3rd {ACM} International Conference on Digital Libraries,
June 23-26, 1998, Pittsburgh, PA, {USA}},
pages = {89--98},
publisher = {{ACM}},
year = {1998}
}CORA_Full, citation network dataset, an extended version of CORA
CORA-ML, extracted from the original data the entire network of CORA
@article{McCallumNRS00,
author = {Andrew McCallum and
Kamal Nigam and
Jason Rennie and
Kristie Seymore},
title = {Automating the Construction of Internet Portals with Machine Learning},
journal = {Inf. Retr.},
volume = {3},
number = {2},
pages = {127--163},
year = {2000}
}The above datasets are collected from
https://github.com/abojchevski/graph2gauss
@inproceedings{bojchevski2018deep,
title={Deep Gaussian Embedding of Graphs: Unsupervised Inductive Learning via Ranking},
author={Aleksandar Bojchevski and Stephan Günnemann},
booktitle={International Conference on Learning Representations},
year={2018},
url={https://openreview.net/forum?id=r1ZdKJ-0W},
}233K nodes, 11.6M edges, 602 node features
Link: http://snap.stanford.edu/graphsage/
@inproceedings{hamilton2017inductive,
author = {Hamilton, William L. and Ying, Rex and Leskovec, Jure},
title = {Inductive Representation Learning on Large Graphs},
booktitle = {NIPS},
year = {2017}
}Link: https://data.dgl.ai/dataset/reddit.zip
Link: https://ndownloader.figshare.com/files/23742119
https://github.com/kimiyoung/planetoid
@inproceedings{CarlsonBKSHM10,
author = {Andrew Carlson and
Justin Betteridge and
Bryan Kisiel and
Burr Settles and
Estevam R. Hruschka Jr. and
Tom M. Mitchell},
editor = {Maria Fox and
David Poole},
title = {Toward an Architecture for Never-Ending Language Learning},
booktitle = {Proceedings of the Twenty-Fourth {AAAI} Conference on Artificial Intelligence,
{AAAI} 2010, Atlanta, Georgia, USA, July 11-15, 2010},
publisher = {{AAAI} Press},
year = {2010}@inproceedings{DBLP:conf/icml/YangCS16,
author = {Zhilin Yang and
William W. Cohen and
Ruslan Salakhutdinov},
editor = {Maria{-}Florina Balcan and
Kilian Q. Weinberger},
title = {Revisiting Semi-Supervised Learning with Graph Embeddings},
booktitle = {Proceedings of the 33nd International Conference on Machine Learning,
{ICML} 2016, New York City, NY, USA, June 19-24, 2016},
series = {{JMLR} Workshop and Conference Proceedings},
volume = {48},
pages = {40--48},
publisher = {JMLR.org},
year = {2016}
}BlogCatalog : It is a dataset of a blog community social network, which contains 5,196 users as nodes, 171,743 edges indicating the user interactions, and 8,189 attribute categories denoting the keywords of their blogs. Users could register their blogs into six different predefined classes, which are set as labels.
Flickr: It is a benchmark attributed social network dataset containing 7,575 nodes. Each node is a Flickr user and each attribute category is a tag related to the photos shared by users. There are 239,738 undirected edges in this network, which indicate the following relationships among users. The nine groups that users have joined are considered as target labels.
@inproceedings{LiHTL15,
author = {Jundong Li and
Xia Hu and
Jiliang Tang and
Huan Liu},
editor = {James Bailey and
Alistair Moffat and
Charu C. Aggarwal and
Maarten de Rijke and
Ravi Kumar and
Vanessa Murdock and
Timos K. Sellis and
Jeffrey Xu Yu},
title = {Unsupervised Streaming Feature Selection in Social Media},
booktitle = {Proceedings of the 24th {ACM} International Conference on Information
and Knowledge Management, {CIKM} 2015, Melbourne, VIC, Australia,
October 19 - 23, 2015},
pages = {1041--1050},
publisher = {{ACM}},
year = {2015}
}both datasets are collected form
https://github.com/xhuang31/LANE
@inproceedings{HuangLH17,
author = {Xiao Huang and
Jundong Li and
Xia Hu},
editor = {Maarten de Rijke and
Milad Shokouhi and
Andrew Tomkins and
Min Zhang},
title = {Label Informed Attributed Network Embedding},
booktitle = {Proceedings of the Tenth {ACM} International Conference on Web Search
and Data Mining, {WSDM} 2017, Cambridge, United Kingdom, February
6-10, 2017},
pages = {731--739},
publisher = {{ACM}},
year = {2017}
}-
KDDS1 https://www.biendata.xyz/competition/kddcup_2020/data/
-
KDDS2 https://www.biendata.xyz/competition/kddcup_2020_formal/data/ the label of the last 50,000 test nodes are released at https://github.com/THUDM/GIAAD
Link: https://github.com/zhumeiqiBUPT/AM-GCN
@inproceedings{wang2018unified,
title={A Unified Weakly Supervised Framework for Community Detection and Semantic Matching},
author={Wang, Wenjun and Liu, Xiao and Jiao, Pengfei and Chen, Xue and Jin, Di},
booktitle={Pacific-Asia Conference on Knowledge Discovery and Data Mining},
pages={218--230},
year={2018},
organization={Springer}
}This network is extracted from ACM dataset where nodes represent papers and there is an edge between two papers if they have the same author. All the papers are divided into 3 classes (Database, Wireless Communication, DataMining). The features are the bag-of-words representa- tions of paper keywords.
Link1: https://github.com/Jhy1993/HAN Cite:
@article{han2019,
title={Heterogeneous Graph Attention Network},
author={Xiao, Wang and Houye, Ji and Chuan, Shi and Bai, Wang and Peng, Cui and P. , Yu and Yanfang, Ye},
journal={WWW},
year={2019}
}Link2: https://github.com/zhumeiqiBUPT/AM-GCN Cite:
@inproceedings{DBLP:conf/kdd/0017ZB0SP20,
author = {Xiao Wang and
Meiqi Zhu and
Deyu Bo and
Peng Cui and
Chuan Shi and
Jian Pei},
title = {{AM-GCN:} Adaptive Multi-channel Graph Convolutional Networks},
booktitle = {{KDD} '20: The 26th {ACM} {SIGKDD} Conference on Knowledge Discovery
and Data Mining, Virtual Event, CA, USA, August 23-27, 2020},
pages = {1243--1253},
publisher = {{ACM}},
year = {2020},
}a new benchmark dataset based on the Microsoft Academic Graph (MAG). Nodes represent papers, edges denote citations, and node features correspond to a bag-of-words representation of paper abstracts. The graph is augmented with "groundtruth" node labels corresponding to the papers’ field of study.
Link: https://figshare.com/articles/dataset/mag_scholar/12696653 Cite:
@inproceedings{bojchevski2020pprgo,
title={Scaling Graph Neural Networks with Approximate PageRank},
author={Bojchevski, Aleksandar and Klicpera, Johannes and Perozzi, Bryan and Kapoor, Amol and Blais, Martin and R{\'o}zemberczki, Benedek and Lukasik, Michal and G{\"u}nnemann, Stephan},
booktitle = {Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining},
year={2020},
publisher = {ACM},
address = {New York, NY, USA},
}Datasets from Karateclub
- deezer
- github
- lastfm
- twitch
- wikipedia
- reddit10k
Cite:
@inproceedings{karateclub,
title = {{Karate Club: An API Oriented Open-source Python Framework for Unsupervised Learning on Graphs}},
author = {Benedek Rozemberczki and Oliver Kiss and Rik Sarkar},
year = {2020},
pages = {3125–3132},
booktitle = {Proceedings of the 29th ACM International Conference on Information and Knowledge Management (CIKM '20)},
organization = {ACM},
}FB15K_URL = "https://dl.fbaipublicfiles.com/starspace/fb15k.tgz" FILENAMES = [ "FB15k/freebase_mtr100_mte100-train.txt", "FB15k/freebase_mtr100_mte100-valid.txt", "FB15k/freebase_mtr100_mte100-test.txt", ]
Datasets from MUSAE ╒═══════════╤═══════════════════════════════════════════════════════════════════╕ │ Names │ Description │ ╞═══════════╪═══════════════════════════════════════════════════════════════════╡ │ chameleon │ Wiki-chameleon dataset (classification) │ ├───────────┼───────────────────────────────────────────────────────────────────┤ │ crocodile │ Wiki-crocodile dataset (classification) │ ├───────────┼───────────────────────────────────────────────────────────────────┤ │ squirrel │ Wiki-squirrel dataset (node classification) │ ├───────────┼───────────────────────────────────────────────────────────────────┤ │ facebook │ facebook dataset (classification) │ ├───────────┼───────────────────────────────────────────────────────────────────┤ │ github │ github dataset, the same as KarateClub('github') (classification) │ ├───────────┼───────────────────────────────────────────────────────────────────┤ │ DE │ Twitch-DE dataset (classification, regression) │ ├───────────┼───────────────────────────────────────────────────────────────────┤ │ ENGB │ Twitch-ENGB dataset (classification, regression) │ ├───────────┼───────────────────────────────────────────────────────────────────┤ │ ES │ Twitch-ES dataset (classification, regression) │ ├───────────┼───────────────────────────────────────────────────────────────────┤ │ FR │ Twitch-FR dataset (classification, regression) │ ├───────────┼───────────────────────────────────────────────────────────────────┤ │ PTBR │ Twitch-PTBR dataset (classification, regression) │ ├───────────┼───────────────────────────────────────────────────────────────────┤ │ RU │ Twitch-RU dataset (classification, regression) │ ├───────────┼───────────────────────────────────────────────────────────────────┤ │ ZHTW │ Twitch-ZHTW dataset (classification, regression) │ ╘═══════════╧═══════════════════════════════════════════════════════════════════╛ Cite
@misc{rozemberczki2019multiscale,
title = {{Multi-scale Attributed Node Embedding}},
author = {Benedek Rozemberczki and Carl Allen and Rik Sarkar},
year = {2019},
eprint = {1909.13021},
archivePrefix = {arXiv},
primaryClass = {cs.LG}
}Synthetic node classification dataset from PDN: https://github.com/benedekrozemberczki/PDN Cite
@inproceedings{rozemberczki2021pdn,
title={{Pathfinder Discovery Networks for Neural Message Passing}},
author={Benedek Rozemberczki and Peter Englert and Amol Kapoor and Martin Blais and Bryan Perozzi},
booktitle = {Proceedings of The Web Conference 2021},
year={2021},
organization={ACM}
}