clefourrier HF staff commited on
Commit
65cb6d6
1 Parent(s): db03e64

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +119 -0
README.md CHANGED
@@ -1,3 +1,122 @@
1
  ---
2
  license: unknown
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: unknown
3
  ---
4
+
5
+ # Dataset Card for ZINC
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://zinc15.docking.org/)**
26
+ - **[Repository](https://www.dropbox.com/s/feo9qle74kg48gy/molecules.zip?dl=1):**:
27
+ - **Paper:**: ZINC 15 – Ligand Discovery for Everyone (see citation)
28
+ - **Leaderboard:**: [Papers with code leaderboard](https://paperswithcode.com/sota/)
29
+
30
+ ### Dataset Summary
31
+
32
+ The `ZINC` dataset is a "curated collection of commercially available chemical compounds prepared especially for virtual screening" (Wikipedia).
33
+
34
+ ### Supported Tasks and Leaderboards
35
+
36
+ `ZINC` should be used for molecular property prediction (aiming to predict the constrained solubility of the molecules), a graph regression task. The score used is the MAE.
37
+
38
+ The associated leaderboard is here: [Papers with code leaderboard](https://paperswithcode.com/sota/graph-regression-on-zinc).
39
+
40
+ ## External Use
41
+ ### PyGeometric
42
+ To load in PyGeometric, do the following:
43
+
44
+ ```python
45
+ from datasets import load_dataset
46
+
47
+ from torch_geometric.data import Data
48
+ from torch_geometric.loader import DataLoader
49
+
50
+ dataset_hf = load_dataset("graphs-datasets/<mydataset>")
51
+ # For the train set (replace by valid or test as needed)
52
+ dataset_pg_list = [Data(graph) for graph in dataset_hf["train"]]
53
+ dataset_pg = DataLoader(dataset_pg_list)
54
+
55
+ ```
56
+
57
+
58
+ ## Dataset Structure
59
+
60
+ ### Data Properties
61
+
62
+ | property | value |
63
+ |---|---|
64
+ | scale | big |
65
+ | #graphs | 220011 |
66
+ | average #nodes | 23.15 |
67
+ | average #edges | 49.81 |
68
+
69
+ ### Data Fields
70
+
71
+ Each row of a given file is a graph, with:
72
+ - `node_feat` (list: #nodes x #node-features): nodes
73
+ - `edge_index` (list: 2 x #edges): pairs of nodes constituting edges
74
+ - `edge_attr` (list: #edges x #edge-features): for the aforementioned edges, contains their features
75
+ - `y` (list: 1 x #labels): contains the number of labels available to predict (here 1, equal to zero or one)
76
+ - `num_nodes` (int): number of nodes of the graph
77
+
78
+ ### Data Splits
79
+
80
+ This data comes from the PyGeometric version of the dataset, and follows the provided data splits.
81
+ This information can be found back using
82
+ ```python
83
+ from torch_geometric.datasets import ZINC
84
+
85
+ dataset = ZINC(root = '', split='train') # valid, test
86
+ ```
87
+
88
+ ## Additional Information
89
+
90
+ ### Licensing Information
91
+ The dataset has been released under unknown license. Please open an issue if you know what is the license of this dataset.
92
+
93
+ ### Citation Information
94
+ ```bibtex
95
+ @article{doi:10.1021/acs.jcim.5b00559,
96
+ author = {Sterling, Teague and Irwin, John J.},
97
+ title = {ZINC 15 – Ligand Discovery for Everyone},
98
+ journal = {Journal of Chemical Information and Modeling},
99
+ volume = {55},
100
+ number = {11},
101
+ pages = {2324-2337},
102
+ year = {2015},
103
+ doi = {10.1021/acs.jcim.5b00559},
104
+ note ={PMID: 26479676},
105
+
106
+ URL = {
107
+ https://doi.org/10.1021/acs.jcim.5b00559
108
+
109
+ },
110
+ eprint = {
111
+ https://doi.org/10.1021/acs.jcim.5b00559
112
+
113
+ }
114
+
115
+ }
116
+
117
+
118
+ ```
119
+
120
+ ### Contributions
121
+
122
+ Thanks to [@clefourrier](https://github.com/clefourrier) for adding this dataset.