Gaëtan Caillaut commited on
Commit
8e8eb22
1 Parent(s): 46a70d6

update dataset to v0.2.1

Browse files
README.md CHANGED
@@ -51,6 +51,19 @@ The dataset `entities` contains description for each Wikipedia pages.
51
  "wikipedia_id": "Identifier of the Wikipedia page",
52
  "wikipedia_url": "URL to the Wikipedia page",
53
  "wikidata_url": "URL to the Wikidata page. Can be null.",
 
 
 
 
 
 
 
 
 
 
 
 
 
54
  "words": ["words", "in", "the", "sentence"],
55
  "ner": ["ner", "labels", "of", "each", "words"],
56
  "el": ["el", "labels", "of", "each", "words"]
 
51
  "wikipedia_id": "Identifier of the Wikipedia page",
52
  "wikipedia_url": "URL to the Wikipedia page",
53
  "wikidata_url": "URL to the Wikidata page. Can be null.",
54
+ "sentences" : [
55
+ {
56
+ "text": "text of the current sentence",
57
+ "ner": ["list", "of", "ner", "labels"],
58
+ "mention_mappings": [
59
+ (start_of_first_mention, end_of_first_mention),
60
+ (start_of_second_mention, end_of_second_mention)
61
+ ],
62
+ "el_wikidata_id": ["wikidata id of first mention", "wikidata id of second mention"],
63
+ "el_wikipedia_id": [wikipedia id of first mention, wikipedia id of second mention],
64
+ "el_wikipedia_title": ["wikipedia title of first mention", "wikipedia title of second mention"]
65
+ }
66
+ ]
67
  "words": ["words", "in", "the", "sentence"],
68
  "ner": ["ner", "labels", "of", "each", "words"],
69
  "el": ["el", "labels", "of", "each", "words"]
data/frwiki-20220901/corpus.jsonl.gz CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:960f5dcf94cb73061c21de8f2f4b3979c388f3924419764a40da5db4d757855c
3
- size 3193932378
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c354336bc135f7b9da4ab95f0ba727c45233001e2bd0f7c457f7b9dcc1c7d1af
3
+ size 3195634980
data/frwiki-20220901/corpus_abstracts.jsonl.gz CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9b1954986631da26d31c9a7b8a4ad30e108981f7201728c57555a052ac77f210
3
- size 465701861
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3210691cc62dd2c17d41c5940cdecd19611431dd75a8f70f6e044597204383af
3
+ size 461618096
data/frwiki-20220901/corpus_mini.jsonl.gz CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ec52d912d96a771cb6374b2358024b47df7f3bde3b2ae03743fda567370ae827
3
- size 132297
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91414a72404a749999d1979ef089a4ab42a77d609220878e44c1a2f57e427c8d
3
+ size 138746
data/frwiki-20220901/entities.jsonl.gz CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3711542c9bee2dfbf5c93eb8bd71efda996b6161e808f0debe1a4df0af643585
3
- size 185405217
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b0c6295842f38647d9e4330c150fc291d14e11987f779b3bb324d33802bf274
3
+ size 184935939
frwiki_el.py CHANGED
@@ -45,7 +45,7 @@ class FrwikiElDataset(datasets.GeneratorBasedBuilder):
45
  """
46
  """
47
 
48
- VERSION = datasets.Version("0.2.0")
49
 
50
  # This is an example of a dataset with multiple configurations.
51
  # If you don't want/need to define several sub-sets in your dataset,
@@ -77,7 +77,7 @@ class FrwikiElDataset(datasets.GeneratorBasedBuilder):
77
  features = datasets.Features({
78
  "name": datasets.Value("string"),
79
  "wikidata_id": datasets.Value("string"),
80
- "wikipedia_id": datasets.Value("string"),
81
  "wikipedia_url": datasets.Value("string"),
82
  "wikidata_url": datasets.Value("string"),
83
  "sentences": [{
@@ -85,7 +85,7 @@ class FrwikiElDataset(datasets.GeneratorBasedBuilder):
85
  "ner": [datasets.Value("string")],
86
  "mention_mappings": [[datasets.Value("int16")]],
87
  "el_wikidata_id": [datasets.Value("string")],
88
- "el_wikipedia_id": [datasets.Value("string")],
89
  "el_wikipedia_title": [datasets.Value("string")],
90
  }]
91
  })
@@ -93,7 +93,7 @@ class FrwikiElDataset(datasets.GeneratorBasedBuilder):
93
  features = datasets.Features({
94
  "name": datasets.Value("string"),
95
  "wikidata_id": datasets.Value("string"),
96
- "wikipedia_id": datasets.Value("string"),
97
  "wikipedia_url": datasets.Value("string"),
98
  "wikidata_url": datasets.Value("string"),
99
  "description": datasets.Value("string"),
@@ -139,9 +139,6 @@ class FrwikiElDataset(datasets.GeneratorBasedBuilder):
139
  # This method handles input defined in _split_generators to yield (key, example) tuples from the dataset.
140
  # The `key` is here for legacy reason (tfds) and is not important in itself.
141
 
142
- def _identiy(x):
143
- return x
144
-
145
  # We need to use open before gzip.open in case the dataset is streamed
146
  # https://github.com/huggingface/datasets/issues/2607#issuecomment-883219727
147
  with gzip.open(open(path, 'rb'), "rt", encoding="UTF-8") as datafile:
 
45
  """
46
  """
47
 
48
+ VERSION = datasets.Version("0.2.1")
49
 
50
  # This is an example of a dataset with multiple configurations.
51
  # If you don't want/need to define several sub-sets in your dataset,
 
77
  features = datasets.Features({
78
  "name": datasets.Value("string"),
79
  "wikidata_id": datasets.Value("string"),
80
+ "wikipedia_id": datasets.Value("int32"),
81
  "wikipedia_url": datasets.Value("string"),
82
  "wikidata_url": datasets.Value("string"),
83
  "sentences": [{
 
85
  "ner": [datasets.Value("string")],
86
  "mention_mappings": [[datasets.Value("int16")]],
87
  "el_wikidata_id": [datasets.Value("string")],
88
+ "el_wikipedia_id": [datasets.Value("int32")],
89
  "el_wikipedia_title": [datasets.Value("string")],
90
  }]
91
  })
 
93
  features = datasets.Features({
94
  "name": datasets.Value("string"),
95
  "wikidata_id": datasets.Value("string"),
96
+ "wikipedia_id": datasets.Value("int32"),
97
  "wikipedia_url": datasets.Value("string"),
98
  "wikidata_url": datasets.Value("string"),
99
  "description": datasets.Value("string"),
 
139
  # This method handles input defined in _split_generators to yield (key, example) tuples from the dataset.
140
  # The `key` is here for legacy reason (tfds) and is not important in itself.
141
 
 
 
 
142
  # We need to use open before gzip.open in case the dataset is streamed
143
  # https://github.com/huggingface/datasets/issues/2607#issuecomment-883219727
144
  with gzip.open(open(path, 'rb'), "rt", encoding="UTF-8") as datafile: