lucas-morin commited on
Commit
9d0d4f1
1 Parent(s): 611a45e

Update README.md

Browse files

Add dataset description

Files changed (1) hide show
  1. README.md +14 -0
README.md CHANGED
@@ -22,4 +22,18 @@ dataset_info:
22
  ---
23
  # Dataset Card for "USPTO-30K"
24
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
25
  [More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
 
22
  ---
23
  # Dataset Card for "USPTO-30K"
24
 
25
+ Existing benchmarks for Optical Chemical Structure Recognition have some limitations.
26
+ Being created using only a few documents, they contain batches of very similar molecules. For example in a patent, a molecule could typically be displayed together with all the substituent of one particular substructure, resulting in large batches of almost identical molecules.
27
+ Additionally, the existing sets contain molecules of different kinds, including superatom groups and various markush features, which should be evaluated independently.
28
+ In practice, it is important to delimit on which types of molecules models can be applied.
29
+
30
+ We introduce USPTO-30K, a large-scale benchmark dataset of annotated molecule images, which overcomes these limitations.
31
+ It is created using the pairs of images and MolFiles by the United States Patent and Trademark Office.
32
+ Each molecule was independently selected among all the available documents from 2001 to 2020.
33
+ The set consists of three subsets to decouple the study of clean molecules, molecules with abbreviations and large molecules.
34
+
35
+ - USPTO-10K contains 10,000 clean molecules, i.e. without any abbreviated groups.
36
+ - USPTO-10K-abb contains 10,000 molecules with superatom groups.
37
+ - USPTO-10K-L contains 10,000 clean molecules with more than 70 atoms.
38
+
39
  [More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)