clefourrier HF staff commited on
Commit
21dd15a
1 Parent(s): 09bac48

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +120 -0
README.md ADDED
@@ -0,0 +1,120 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ licence: mit
3
+ ---
4
+
5
+ # Dataset Card for alchemy
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
+ - **[Homepage](https://alchemy.tencent.com/)**
25
+ - **Paper:**: (see citation)
26
+ - **Leaderboard:**: [Leaderboard](https://alchemy.tencent.com/)
27
+
28
+ ### Dataset Summary
29
+ The `alchemy` dataset is a molecular dataset, called Alchemy, which lists 12 quantum mechanical properties of 130,000+ organic molecules comprising up to 12 heavy atoms (C, N, O, S, F and Cl), sampled from the GDBMedChem database.
30
+
31
+ ### Supported Tasks and Leaderboards
32
+ `alchemy` should be used for organic quantum molecular property prediction, a regression task on 12 properties. The score used is MAE.
33
+
34
+
35
+ ## External Use
36
+ ### PyGeometric
37
+ To load in PyGeometric, do the following:
38
+
39
+ ```python
40
+ from datasets import load_dataset
41
+
42
+ from torch_geometric.data import Data
43
+ from torch_geometric.loader import DataLoader
44
+
45
+ dataset_hf = load_dataset("graphs-datasets/<mydataset>")
46
+ # For the train set (replace by valid or test as needed)
47
+ dataset_pg_list = [Data(graph) for graph in dataset_hf["train"]]
48
+ dataset_pg = DataLoader(dataset_pg_list)
49
+ ```
50
+
51
+ ## Dataset Structure
52
+
53
+ ### Data Properties
54
+ | property | value |
55
+ |---|---|
56
+ | scale | big |
57
+ | #graphs | 202578 |
58
+ | average #nodes | 10.101387606810183 |
59
+ | average #edges | 20.877326870011206 |
60
+
61
+ ### Data Fields
62
+
63
+ Each row of a given file is a graph, with:
64
+ - `node_feat` (list: #nodes x #node-features): nodes
65
+ - `edge_index` (list: 2 x #edges): pairs of nodes constituting edges
66
+ - `edge_attr` (list: #edges x #edge-features): for the aforementioned edges, contains their features
67
+ - `y` (list: 1 x #labels): contains the number of labels available to predict (here 1, equal to zero or one)
68
+ - `num_nodes` (int): number of nodes of the graph
69
+
70
+ ### Data Splits
71
+
72
+ This data is not split, and should be used with cross validation. It comes from the PyGeometric version of the dataset.
73
+
74
+ ## Additional Information
75
+
76
+ ### Licensing Information
77
+ The dataset has been released under license mit.
78
+
79
+ ### Citation Information
80
+ ```
81
+ @inproceedings{Morris+2020,
82
+ title={TUDataset: A collection of benchmark datasets for learning with graphs},
83
+ author={Christopher Morris and Nils M. Kriege and Franka Bause and Kristian Kersting and Petra Mutzel and Marion Neumann},
84
+ booktitle={ICML 2020 Workshop on Graph Representation Learning and Beyond (GRL+ 2020)},
85
+ archivePrefix={arXiv},
86
+ eprint={2007.08663},
87
+ url={www.graphlearning.io},
88
+ year={2020}
89
+ }
90
+ ```
91
+
92
+ ```
93
+
94
+ @article{DBLP:journals/corr/abs-1906-09427,
95
+ author = {Guangyong Chen and
96
+ Pengfei Chen and
97
+ Chang{-}Yu Hsieh and
98
+ Chee{-}Kong Lee and
99
+ Benben Liao and
100
+ Renjie Liao and
101
+ Weiwen Liu and
102
+ Jiezhong Qiu and
103
+ Qiming Sun and
104
+ Jie Tang and
105
+ Richard S. Zemel and
106
+ Shengyu Zhang},
107
+ title = {Alchemy: {A} Quantum Chemistry Dataset for Benchmarking {AI} Models},
108
+ journal = {CoRR},
109
+ volume = {abs/1906.09427},
110
+ year = {2019},
111
+ url = {http://arxiv.org/abs/1906.09427},
112
+ eprinttype = {arXiv},
113
+ eprint = {1906.09427},
114
+ timestamp = {Mon, 11 Nov 2019 12:55:11 +0100},
115
+ biburl = {https://dblp.org/rec/journals/corr/abs-1906-09427.bib},
116
+ bibsource = {dblp computer science bibliography, https://dblp.org}
117
+ }
118
+
119
+
120
+ ```