Sebastian Gehrmann commited on
Commit
1e5fcfa
1 Parent(s): 357067d
Files changed (1) hide show
  1. dataset_infos.json +126 -1
dataset_infos.json CHANGED
@@ -1 +1,126 @@
1
- {"abstract": {"description": "The OrangeSum dataset was inspired by the XSum dataset. It was created by scraping the \"Orange Actu\" website: https://actu.orange.fr/. Orange S.A. is a large French multinational telecommunications corporation, with 266M customers worldwide. Scraped pages cover almost a decade from Feb 2011 to Sep 2020. They belong to five main categories: France, world, politics, automotive, and society. The society category is itself divided into 8 subcategories: health, environment, people, culture, media, high-tech, unsual (\"insolite\" in French), and miscellaneous.\n\nEach article featured a single-sentence title as well as a very brief abstract, both professionally written by the author of the article. These two fields were extracted from each page, thus creating two summarization tasks: OrangeSum Title and OrangeSum Abstract.\n", "citation": "@article{eddine2020barthez,\n title={BARThez: a Skilled Pretrained French Sequence-to-Sequence Model},\n author={Eddine, Moussa Kamal and Tixier, Antoine J-P and Vazirgiannis, Michalis},\n journal={arXiv preprint arXiv:2010.12321},\n year={2020}\n}\n", "homepage": "https://github.com/Tixierae/OrangeSum/", "license": "", "features": {"text": {"dtype": "string", "id": null, "_type": "Value"}, "summary": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "text", "output": "summary"}, "builder_name": "OrangeSum", "config_name": "abstract", "version": {"version_str": "1.1.0", "description": null, "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 53531651, "num_examples": 21401, "dataset_name": "OrangeSum"}, "test": {"name": "test", "num_bytes": 3785207, "num_examples": 1500, "dataset_name": "OrangeSum"}, "validation": {"name": "validation", "num_bytes": 3698650, "num_examples": 1500, "dataset_name": "OrangeSum"}}, "download_checksums": {"https://raw.githubusercontent.com/Tixierae/OrangeSum/main/data/docs/splits/abstract.tgz": {"num_bytes": 23058350, "checksum": "eaa4321b70bcf41c758d02fb5a94e50d73509a2be32adb1f9aacdcfd5796434b"}}, "download_size": 23058350, "post_processing_size": null, "dataset_size": 61015508, "size_in_bytes": 84073858}, "title": {"description": "The OrangeSum dataset was inspired by the XSum dataset. It was created by scraping the \"Orange Actu\" website: https://actu.orange.fr/. Orange S.A. is a large French multinational telecommunications corporation, with 266M customers worldwide. Scraped pages cover almost a decade from Feb 2011 to Sep 2020. They belong to five main categories: France, world, politics, automotive, and society. The society category is itself divided into 8 subcategories: health, environment, people, culture, media, high-tech, unsual (\"insolite\" in French), and miscellaneous.\n\nEach article featured a single-sentence title as well as a very brief abstract, both professionally written by the author of the article. These two fields were extracted from each page, thus creating two summarization tasks: OrangeSum Title and OrangeSum Abstract.\n", "citation": "@article{eddine2020barthez,\n title={BARThez: a Skilled Pretrained French Sequence-to-Sequence Model},\n author={Eddine, Moussa Kamal and Tixier, Antoine J-P and Vazirgiannis, Michalis},\n journal={arXiv preprint arXiv:2010.12321},\n year={2020}\n}\n", "homepage": "https://github.com/Tixierae/OrangeSum/", "license": "", "features": {"text": {"dtype": "string", "id": null, "_type": "Value"}, "summary": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": {"input": "text", "output": "summary"}, "builder_name": "OrangeSum", "config_name": "title", "version": {"version_str": "1.1.0", "description": null, "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 65225136, "num_examples": 30659, "dataset_name": "OrangeSum"}, "test": {"name": "test", "num_bytes": 3176690, "num_examples": 1500, "dataset_name": "OrangeSum"}, "validation": {"name": "validation", "num_bytes": 3276713, "num_examples": 1500, "dataset_name": "OrangeSum"}}, "download_checksums": {"https://raw.githubusercontent.com/Tixierae/OrangeSum/main/data/docs/splits/title.tgz": {"num_bytes": 27321627, "checksum": "5d15823f7e1158f16f5428fdfc8fa26509f98325c0793d6a8880a33af9822301"}}, "download_size": 27321627, "post_processing_size": null, "dataset_size": 71678539, "size_in_bytes": 99000166}}
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "abstract": {
3
+ "description": "The OrangeSum dataset was inspired by the XSum dataset. It was created by scraping the \"Orange Actu\" website: https://actu.orange.fr/. Orange S.A. is a large French multinational telecommunications corporation, with 266M customers worldwide. Scraped pages cover almost a decade from Feb 2011 to Sep 2020. They belong to five main categories: France, world, politics, automotive, and society. The society category is itself divided into 8 subcategories: health, environment, people, culture, media, high-tech, unsual (\"insolite\" in French), and miscellaneous.\n\nEach article featured a single-sentence title as well as a very brief abstract, both professionally written by the author of the article. These two fields were extracted from each page, thus creating two summarization tasks: OrangeSum Title and OrangeSum Abstract.\n",
4
+ "citation": "@article{eddine2020barthez,\n title={BARThez: a Skilled Pretrained French Sequence-to-Sequence Model},\n author={Eddine, Moussa Kamal and Tixier, Antoine J-P and Vazirgiannis, Michalis},\n journal={arXiv preprint arXiv:2010.12321},\n year={2020}\n}\n",
5
+ "homepage": "https://github.com/Tixierae/OrangeSum/",
6
+ "license": "",
7
+ "features": {
8
+ "text": {
9
+ "dtype": "string",
10
+ "id": null,
11
+ "_type": "Value"
12
+ },
13
+ "summary": {
14
+ "dtype": "string",
15
+ "id": null,
16
+ "_type": "Value"
17
+ }
18
+ },
19
+ "post_processed": null,
20
+ "supervised_keys": {
21
+ "input": "text",
22
+ "output": "summary"
23
+ },
24
+ "builder_name": "OrangeSum",
25
+ "config_name": "abstract",
26
+ "version": {
27
+ "version_str": "1.1.0",
28
+ "description": null,
29
+ "major": 1,
30
+ "minor": 1,
31
+ "patch": 0
32
+ },
33
+ "splits": {
34
+ "train": {
35
+ "name": "train",
36
+ "num_bytes": 53531651,
37
+ "num_examples": 21401,
38
+ "dataset_name": "OrangeSum"
39
+ },
40
+ "test": {
41
+ "name": "test",
42
+ "num_bytes": 3785207,
43
+ "num_examples": 1500,
44
+ "dataset_name": "OrangeSum"
45
+ },
46
+ "validation": {
47
+ "name": "validation",
48
+ "num_bytes": 3698650,
49
+ "num_examples": 1500,
50
+ "dataset_name": "OrangeSum"
51
+ }
52
+ },
53
+ "download_checksums": {
54
+ "https://raw.githubusercontent.com/Tixierae/OrangeSum/main/data/docs/splits/abstract.tgz": {
55
+ "num_bytes": 23058350,
56
+ "checksum": "eaa4321b70bcf41c758d02fb5a94e50d73509a2be32adb1f9aacdcfd5796434b"
57
+ }
58
+ },
59
+ "download_size": 23058350,
60
+ "post_processing_size": null,
61
+ "dataset_size": 61015508,
62
+ "size_in_bytes": 84073858
63
+ },
64
+ "title": {
65
+ "description": "The OrangeSum dataset was inspired by the XSum dataset. It was created by scraping the \"Orange Actu\" website: https://actu.orange.fr/. Orange S.A. is a large French multinational telecommunications corporation, with 266M customers worldwide. Scraped pages cover almost a decade from Feb 2011 to Sep 2020. They belong to five main categories: France, world, politics, automotive, and society. The society category is itself divided into 8 subcategories: health, environment, people, culture, media, high-tech, unsual (\"insolite\" in French), and miscellaneous.\n\nEach article featured a single-sentence title as well as a very brief abstract, both professionally written by the author of the article. These two fields were extracted from each page, thus creating two summarization tasks: OrangeSum Title and OrangeSum Abstract.\n",
66
+ "citation": "@article{eddine2020barthez,\n title={BARThez: a Skilled Pretrained French Sequence-to-Sequence Model},\n author={Eddine, Moussa Kamal and Tixier, Antoine J-P and Vazirgiannis, Michalis},\n journal={arXiv preprint arXiv:2010.12321},\n year={2020}\n}\n",
67
+ "homepage": "https://github.com/Tixierae/OrangeSum/",
68
+ "license": "",
69
+ "features": {
70
+ "text": {
71
+ "dtype": "string",
72
+ "id": null,
73
+ "_type": "Value"
74
+ },
75
+ "summary": {
76
+ "dtype": "string",
77
+ "id": null,
78
+ "_type": "Value"
79
+ }
80
+ },
81
+ "post_processed": null,
82
+ "supervised_keys": {
83
+ "input": "text",
84
+ "output": "summary"
85
+ },
86
+ "builder_name": "OrangeSum",
87
+ "config_name": "title",
88
+ "version": {
89
+ "version_str": "1.1.0",
90
+ "description": null,
91
+ "major": 1,
92
+ "minor": 1,
93
+ "patch": 0
94
+ },
95
+ "splits": {
96
+ "train": {
97
+ "name": "train",
98
+ "num_bytes": 65225136,
99
+ "num_examples": 30659,
100
+ "dataset_name": "OrangeSum"
101
+ },
102
+ "test": {
103
+ "name": "test",
104
+ "num_bytes": 3176690,
105
+ "num_examples": 1500,
106
+ "dataset_name": "OrangeSum"
107
+ },
108
+ "validation": {
109
+ "name": "validation",
110
+ "num_bytes": 3276713,
111
+ "num_examples": 1500,
112
+ "dataset_name": "OrangeSum"
113
+ }
114
+ },
115
+ "download_checksums": {
116
+ "https://raw.githubusercontent.com/Tixierae/OrangeSum/main/data/docs/splits/title.tgz": {
117
+ "num_bytes": 27321627,
118
+ "checksum": "5d15823f7e1158f16f5428fdfc8fa26509f98325c0793d6a8880a33af9822301"
119
+ }
120
+ },
121
+ "download_size": 27321627,
122
+ "post_processing_size": null,
123
+ "dataset_size": 71678539,
124
+ "size_in_bytes": 99000166
125
+ }
126
+ }