clefourrier HF staff commited on
Commit
7f79023
1 Parent(s): 3e167cc

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +123 -0
README.md CHANGED
@@ -1,3 +1,126 @@
1
  ---
2
  license: cc0-1.0
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: cc0-1.0
3
  ---
4
+
5
+ # Dataset Card for ogbg-ppa
6
+
7
+ ## Table of Contents
8
+ - [Table of Contents](#table-of-contents)
9
+ - [Dataset Description](#dataset-description)
10
+ - [Dataset Summary](#dataset-summary)
11
+ - [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
12
+ - [External Use](#external-use)
13
+ - [PyGeometric](#pygeometric)
14
+ - [Dataset Structure](#dataset-structure)
15
+ - [Data Properties](#data-properties)
16
+ - [Data Fields](#data-fields)
17
+ - [Data Splits](#data-splits)
18
+ - [Additional Information](#additional-information)
19
+ - [Licensing Information](#licensing-information)
20
+ - [Citation Information](#citation-information)
21
+ - [Contributions](#contributions)
22
+
23
+ ## Dataset Description
24
+
25
+ - **[Homepage](https://ogb.stanford.edu/docs/graphprop/#ogbg-ppa)**
26
+ - **[Repository](https://github.com/snap-stanford/ogb):**:
27
+ - **Paper:**: Open Graph Benchmark: Datasets for Machine Learning on Graphs (see citation)
28
+ - **Leaderboard:**: [OGB leaderboard](https://ogb.stanford.edu/docs/leader_graphprop/#ogbg-ppa) and [Papers with code leaderboard](https://paperswithcode.com/sota/graph-property-prediction-on-ogbg-ppa)
29
+
30
+ ### Dataset Summary
31
+
32
+ The `ogbg-ppa` dataset is "a set of undirected protein association neighborhoods extracted from the protein-protein association networks of 1,581 species", over 37 taxonomic groups, by teams at Stanford, to be a part of the Open Graph Benchmark. See their website for dataset postprocessing.
33
+
34
+ ### Supported Tasks and Leaderboards
35
+
36
+ `ogbg-ppa` should be used for taxonomic group prediction, a 37-way multi-class classification task. The score used is Average Precision on the test set.
37
+
38
+ ## External Use
39
+ ### PyGeometric
40
+ To load in PyGeometric, do the following:
41
+
42
+ ```python
43
+ from datasets import load_dataset
44
+
45
+ from torch_geometric.data import Data
46
+ from torch_geometric.loader import DataLoader
47
+
48
+ graphs_dataset = load_dataset("graphs-datasets/ogbg-ppa")
49
+ # For the train set (replace by valid or test as needed)
50
+ graphs_list = [Data(graph) for graph in graphs_dataset["train"]]
51
+ graphs_pygeometric = DataLoader(graph_list)
52
+
53
+ ```
54
+
55
+
56
+ ## Dataset Structure
57
+
58
+ ### Data Properties
59
+
60
+ | property | value |
61
+ |---|---|
62
+ | scale | small |
63
+ | #graphs | 158,100 |
64
+ | average #nodes | 243.4 |
65
+ | average #edges | 2,266.1 |
66
+ | average node degree | 18.3 |
67
+ | average cluster coefficient | 0.513 |
68
+ | MaxSCC ratio | 1.000 |
69
+ | graph diameter | 4.8 |
70
+
71
+ ### Data Fields
72
+
73
+ Each row of a given file is a graph, with:
74
+ - `edge_index` (list: 2 x #edges): pairs of nodes constituting edges
75
+ - `edge_attr` (list: #edges x #edge-features): for the aforementioned edges, contains their features
76
+ - `y` (list: 1 x #labels): contains the number of labels available to predict (here 1, equal to zero or one)
77
+ - `num_nodes` (int): number of nodes of the graph
78
+
79
+ The nodes don't have specific features and are implicit from the lists of edges
80
+
81
+ ### Data Splits
82
+
83
+ This data comes from the PyGeometric version of the dataset provided by OGB, and follows the provided data splits.
84
+ This information can be found back using
85
+ ```python
86
+ from ogb.graphproppred import PygGraphPropPredDataset
87
+
88
+ dataset = PygGraphPropPredDataset(name = 'ogbg-ppa')
89
+
90
+ split_idx = dataset.get_idx_split()
91
+ train = dataset[split_idx['train']] # valid, test
92
+ ```
93
+
94
+ ## Additional Information
95
+
96
+ ### Licensing Information
97
+ The dataset has been released under CC-0 license.
98
+
99
+ ### Citation Information
100
+ ```
101
+ @inproceedings{hu-etal-2020-open,
102
+ author = {Weihua Hu and
103
+ Matthias Fey and
104
+ Marinka Zitnik and
105
+ Yuxiao Dong and
106
+ Hongyu Ren and
107
+ Bowen Liu and
108
+ Michele Catasta and
109
+ Jure Leskovec},
110
+ editor = {Hugo Larochelle and
111
+ Marc Aurelio Ranzato and
112
+ Raia Hadsell and
113
+ Maria{-}Florina Balcan and
114
+ Hsuan{-}Tien Lin},
115
+ title = {Open Graph Benchmark: Datasets for Machine Learning on Graphs},
116
+ booktitle = {Advances in Neural Information Processing Systems 33: Annual Conference
117
+ on Neural Information Processing Systems 2020, NeurIPS 2020, December
118
+ 6-12, 2020, virtual},
119
+ year = {2020},
120
+ url = {https://proceedings.neurips.cc/paper/2020/hash/fb60d411a5c5b72b2e7d3527cfc84fd0-Abstract.html},
121
+ }
122
+ ```
123
+
124
+ ### Contributions
125
+
126
+ Thanks to [@clefourrier](https://github.com/clefourrier) for adding this dataset.