rwightman HF staff commited on
Commit
70a6d94
1 Parent(s): 97985a3

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +71 -0
README.md CHANGED
@@ -3,4 +3,75 @@ license: other
3
  license_name: conceptual-captions
4
  license_link: >-
5
  https://github.com/google-research-datasets/conceptual-captions/blob/master/LICENSE
 
 
 
 
6
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  license_name: conceptual-captions
4
  license_link: >-
5
  https://github.com/google-research-datasets/conceptual-captions/blob/master/LICENSE
6
+ task_categories:
7
+ - image-to-text
8
+ size_categories:
9
+ - 1M<n<10M
10
  ---
11
+
12
+ # Dataset Card for Conceptual Captions (CC3M)
13
+
14
+ ## Table of Contents
15
+ - [Dataset Description](#dataset-description)
16
+ - [Dataset Summary](#dataset-summary)
17
+ - [Additional Information](#additional-information)
18
+ - [Dataset Curators](#dataset-curators)
19
+ - [Licensing Information](#licensing-information)
20
+ - [Citation Information](#citation-information)
21
+
22
+ ## Dataset Description
23
+
24
+ - **Homepage:** [Conceptual Captions homepage](https://ai.google.com/research/ConceptualCaptions/)
25
+ - **Repository:** [Conceptual Captions repository](https://github.com/google-research-datasets/conceptual-captions)
26
+ - **Paper:** [Conceptual Captions: A Cleaned, Hypernymed, Image Alt-text Dataset For Automatic Image Captioning](https://www.aclweb.org/anthology/P18-1238/)
27
+ - **Leaderboard:** [Conceptual Captions leaderboard](https://ai.google.com/research/ConceptualCaptions/competition?active_tab=leaderboard)https://ai.google.com/research/ConceptualCaptions/leaderboard?active_tab=leaderboard
28
+ - **Point of Contact:** [Conceptual Captions e-mail](mailto:[email protected])
29
+
30
+ ### Dataset Summary
31
+
32
+ Conceptual Captions is a dataset consisting of ~3.3M images annotated with captions. In contrast with the curated style of other image caption annotations, Conceptual Caption images and their raw descriptions are harvested from the web, and therefore represent a wider variety of styles. More precisely, the raw descriptions are harvested from the Alt-text HTML attribute associated with web images. To arrive at the current version of the captions, we have developed an automatic pipeline that extracts, filters, and transforms candidate image/caption pairs, with the goal of achieving a balance of cleanliness, informativeness, fluency, and learnability of the resulting captions.
33
+
34
+ ### Usage
35
+
36
+ This instance of Conceptual Captions is in [webdataset](https://github.com/webdataset/webdataset/commits/main) .tar format. It can be used with webdataset library or upcoming releases of Hugging Face `datasets`.
37
+
38
+ ...More Detail TBD
39
+
40
+ ### Data Splits
41
+
42
+ This dataset was downloaded using img2dataset.
43
+
44
+ #### Train
45
+ * `cc3m-train-*.tar`
46
+ * Downloaded on 2021/12/22
47
+ * 576 shards, 2905954 samples
48
+
49
+ #### Validation
50
+ * `cc3m-validation-*.tar`
51
+ * Downloaded on 2023/12/13 (original validation set download in 2021 was corrupted)
52
+ * 16 shards, 15840 samples
53
+
54
+ ## Additional Information
55
+
56
+ ### Dataset Curators
57
+
58
+ Piyush Sharma, Nan Ding, Sebastian Goodman and Radu Soricut.
59
+
60
+ ### Licensing Information
61
+
62
+ The dataset may be freely used for any purpose, although acknowledgement of
63
+ Google LLC ("Google") as the data source would be appreciated. The dataset is
64
+ provided "AS IS" without any warranty, express or implied. Google disclaims all
65
+ liability for any damages, direct or indirect, resulting from the use of the
66
+ dataset.
67
+
68
+ ### Citation Information
69
+
70
+ ```bibtex
71
+ @inproceedings{sharma2018conceptual,
72
+ title = {Conceptual Captions: A Cleaned, Hypernymed, Image Alt-text Dataset For Automatic Image Captioning},
73
+ author = {Sharma, Piyush and Ding, Nan and Goodman, Sebastian and Soricut, Radu},
74
+ booktitle = {Proceedings of ACL},
75
+ year = {2018},
76
+ }
77
+ ```