Datasets:
Tasks:
Token Classification
Sub-tasks:
named-entity-recognition
Languages:
German
Size:
100K<n<1M
License:
Update files from the datasets library (from 1.1.3)
Browse filesRelease notes: https://github.com/huggingface/datasets/releases/tag/1.1.3
- dataset_infos.json +1 -1
- germeval_14.py +76 -14
dataset_infos.json
CHANGED
@@ -1 +1 @@
|
|
1 |
-
{"germeval_14": {"description": "The GermEval 2014 NER Shared Task builds on a new dataset with German Named Entity annotation with the following properties: - The data was sampled from German Wikipedia and News Corpora as a collection of citations. - The dataset covers over 31,000 sentences corresponding to over 590,000 tokens. - The NER annotation uses the NoSta-D guidelines, which extend the T\u00fcbingen Treebank guidelines, using four main NER categories with sub-structure, and annotating embeddings among NEs such as [ORG FC Kickers [LOC Darmstadt]].\n", "citation": "@inproceedings{benikova-etal-2014-nosta,\n title = {NoSta-D Named Entity Annotation for German: Guidelines and Dataset},\n author = {Benikova, Darina and\n Biemann, Chris and\n Reznicek, Marc},\n booktitle = {Proceedings of the Ninth International Conference on Language Resources and Evaluation ({LREC}'14)},\n month = {may},\n year = {2014},\n address = {Reykjavik, Iceland},\n publisher = {European Language Resources Association (ELRA)},\n url = {http://www.lrec-conf.org/proceedings/lrec2014/pdf/276_Paper.pdf},\n pages = {2524--2531},\n}\n", "homepage": "https://sites.google.com/site/germeval2014ner/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "source": {"dtype": "string", "id": null, "_type": "Value"}, "tokens": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "
|
|
|
1 |
+
{"germeval_14": {"description": "The GermEval 2014 NER Shared Task builds on a new dataset with German Named Entity annotation with the following properties: - The data was sampled from German Wikipedia and News Corpora as a collection of citations. - The dataset covers over 31,000 sentences corresponding to over 590,000 tokens. - The NER annotation uses the NoSta-D guidelines, which extend the T\u00fcbingen Treebank guidelines, using four main NER categories with sub-structure, and annotating embeddings among NEs such as [ORG FC Kickers [LOC Darmstadt]].\n", "citation": "@inproceedings{benikova-etal-2014-nosta,\n title = {NoSta-D Named Entity Annotation for German: Guidelines and Dataset},\n author = {Benikova, Darina and\n Biemann, Chris and\n Reznicek, Marc},\n booktitle = {Proceedings of the Ninth International Conference on Language Resources and Evaluation ({LREC}'14)},\n month = {may},\n year = {2014},\n address = {Reykjavik, Iceland},\n publisher = {European Language Resources Association (ELRA)},\n url = {http://www.lrec-conf.org/proceedings/lrec2014/pdf/276_Paper.pdf},\n pages = {2524--2531},\n}\n", "homepage": "https://sites.google.com/site/germeval2014ner/", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "source": {"dtype": "string", "id": null, "_type": "Value"}, "tokens": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "ner_tags": {"feature": {"num_classes": 25, "names": ["O", "B-LOC", "I-LOC", "B-LOCderiv", "I-LOCderiv", "B-LOCpart", "I-LOCpart", "B-ORG", "I-ORG", "B-ORGderiv", "I-ORGderiv", "B-ORGpart", "I-ORGpart", "B-OTH", "I-OTH", "B-OTHderiv", "I-OTHderiv", "B-OTHpart", "I-OTHpart", "B-PER", "I-PER", "B-PERderiv", "I-PERderiv", "B-PERpart", "I-PERpart"], "names_file": null, "id": null, "_type": "ClassLabel"}, "length": -1, "id": null, "_type": "Sequence"}, "nested_ner_tags": {"feature": {"num_classes": 25, "names": ["O", "B-LOC", "I-LOC", "B-LOCderiv", "I-LOCderiv", "B-LOCpart", "I-LOCpart", "B-ORG", "I-ORG", "B-ORGderiv", "I-ORGderiv", "B-ORGpart", "I-ORGpart", "B-OTH", "I-OTH", "B-OTHderiv", "I-OTHderiv", "B-OTHpart", "I-OTHpart", "B-PER", "I-PER", "B-PERderiv", "I-PERderiv", "B-PERpart", "I-PERpart"], "names_file": null, "id": null, "_type": "ClassLabel"}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": null, "supervised_keys": null, "builder_name": "germ_eval14", "config_name": "germeval_14", "version": {"version_str": "2.0.0", "description": null, "major": 2, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 13816714, "num_examples": 24000, "dataset_name": "germ_eval14"}, "validation": {"name": "validation", "num_bytes": 1266974, "num_examples": 2200, "dataset_name": "germ_eval14"}, "test": {"name": "test", "num_bytes": 2943201, "num_examples": 5100, "dataset_name": "germ_eval14"}}, "download_checksums": {"https://drive.google.com/uc?export=download&id=1Jjhbal535VVz2ap4v4r_rN1UEHTdLK5P": {"num_bytes": 7882358, "checksum": "1e5a803d81f5fe6ade54700a7e8e9107a45edba80469d42e41a360550d1758e7"}, "https://drive.google.com/uc?export=download&id=1ZfRcQThdtAR5PPRjIDtrVP7BtXSCUBbm": {"num_bytes": 723876, "checksum": "d69d1347847e3ac0d1bfd14d7e5c0713dcb82899624301ced6df807dbb070056"}, "https://drive.google.com/uc?export=download&id=1u9mb7kNJHWQCWyweMDRMuTFoOHOfeBTH": {"num_bytes": 1682738, "checksum": "9405e49532379f3aee048851d116b35823d31c04e9521b87a9c4e6572c269097"}}, "download_size": 10288972, "post_processing_size": null, "dataset_size": 18026889, "size_in_bytes": 28315861}}
|
germeval_14.py
CHANGED
@@ -85,8 +85,68 @@ class GermEval14(datasets.GeneratorBasedBuilder):
|
|
85 |
"id": datasets.Value("string"),
|
86 |
"source": datasets.Value("string"),
|
87 |
"tokens": datasets.Sequence(datasets.Value("string")),
|
88 |
-
"
|
89 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
90 |
}
|
91 |
),
|
92 |
supervised_keys=None,
|
@@ -110,8 +170,8 @@ class GermEval14(datasets.GeneratorBasedBuilder):
|
|
110 |
data = csv.reader(f, delimiter="\t", quoting=csv.QUOTE_NONE)
|
111 |
current_source = ""
|
112 |
current_tokens = []
|
113 |
-
|
114 |
-
|
115 |
sentence_counter = 0
|
116 |
for row in data:
|
117 |
if row:
|
@@ -120,37 +180,39 @@ class GermEval14(datasets.GeneratorBasedBuilder):
|
|
120 |
continue
|
121 |
id_, token, label, nested_label = row[:4]
|
122 |
current_tokens.append(token)
|
123 |
-
|
124 |
-
|
125 |
else:
|
126 |
# New sentence
|
127 |
if not current_tokens:
|
128 |
# Consecutive empty lines will cause empty sentences
|
129 |
continue
|
130 |
-
assert len(current_tokens) == len(
|
131 |
-
assert len(
|
|
|
|
|
132 |
assert current_source, "๐ฅ Source for new sentence was not set"
|
133 |
sentence = (
|
134 |
sentence_counter,
|
135 |
{
|
136 |
"id": str(sentence_counter),
|
137 |
"tokens": current_tokens,
|
138 |
-
"
|
139 |
-
"
|
140 |
"source": current_source,
|
141 |
},
|
142 |
)
|
143 |
sentence_counter += 1
|
144 |
current_tokens = []
|
145 |
-
|
146 |
-
|
147 |
current_source = ""
|
148 |
yield sentence
|
149 |
# Don't forget last sentence in dataset ๐ง
|
150 |
yield sentence_counter, {
|
151 |
"id": str(sentence_counter),
|
152 |
"tokens": current_tokens,
|
153 |
-
"
|
154 |
-
"
|
155 |
"source": current_source,
|
156 |
}
|
|
|
85 |
"id": datasets.Value("string"),
|
86 |
"source": datasets.Value("string"),
|
87 |
"tokens": datasets.Sequence(datasets.Value("string")),
|
88 |
+
"ner_tags": datasets.Sequence(
|
89 |
+
datasets.features.ClassLabel(
|
90 |
+
names=[
|
91 |
+
"O",
|
92 |
+
"B-LOC",
|
93 |
+
"I-LOC",
|
94 |
+
"B-LOCderiv",
|
95 |
+
"I-LOCderiv",
|
96 |
+
"B-LOCpart",
|
97 |
+
"I-LOCpart",
|
98 |
+
"B-ORG",
|
99 |
+
"I-ORG",
|
100 |
+
"B-ORGderiv",
|
101 |
+
"I-ORGderiv",
|
102 |
+
"B-ORGpart",
|
103 |
+
"I-ORGpart",
|
104 |
+
"B-OTH",
|
105 |
+
"I-OTH",
|
106 |
+
"B-OTHderiv",
|
107 |
+
"I-OTHderiv",
|
108 |
+
"B-OTHpart",
|
109 |
+
"I-OTHpart",
|
110 |
+
"B-PER",
|
111 |
+
"I-PER",
|
112 |
+
"B-PERderiv",
|
113 |
+
"I-PERderiv",
|
114 |
+
"B-PERpart",
|
115 |
+
"I-PERpart",
|
116 |
+
]
|
117 |
+
)
|
118 |
+
),
|
119 |
+
"nested_ner_tags": datasets.Sequence(
|
120 |
+
datasets.features.ClassLabel(
|
121 |
+
names=[
|
122 |
+
"O",
|
123 |
+
"B-LOC",
|
124 |
+
"I-LOC",
|
125 |
+
"B-LOCderiv",
|
126 |
+
"I-LOCderiv",
|
127 |
+
"B-LOCpart",
|
128 |
+
"I-LOCpart",
|
129 |
+
"B-ORG",
|
130 |
+
"I-ORG",
|
131 |
+
"B-ORGderiv",
|
132 |
+
"I-ORGderiv",
|
133 |
+
"B-ORGpart",
|
134 |
+
"I-ORGpart",
|
135 |
+
"B-OTH",
|
136 |
+
"I-OTH",
|
137 |
+
"B-OTHderiv",
|
138 |
+
"I-OTHderiv",
|
139 |
+
"B-OTHpart",
|
140 |
+
"I-OTHpart",
|
141 |
+
"B-PER",
|
142 |
+
"I-PER",
|
143 |
+
"B-PERderiv",
|
144 |
+
"I-PERderiv",
|
145 |
+
"B-PERpart",
|
146 |
+
"I-PERpart",
|
147 |
+
]
|
148 |
+
)
|
149 |
+
),
|
150 |
}
|
151 |
),
|
152 |
supervised_keys=None,
|
|
|
170 |
data = csv.reader(f, delimiter="\t", quoting=csv.QUOTE_NONE)
|
171 |
current_source = ""
|
172 |
current_tokens = []
|
173 |
+
current_ner_tags = []
|
174 |
+
current_nested_ner_tags = []
|
175 |
sentence_counter = 0
|
176 |
for row in data:
|
177 |
if row:
|
|
|
180 |
continue
|
181 |
id_, token, label, nested_label = row[:4]
|
182 |
current_tokens.append(token)
|
183 |
+
current_ner_tags.append(label)
|
184 |
+
current_nested_ner_tags.append(nested_label)
|
185 |
else:
|
186 |
# New sentence
|
187 |
if not current_tokens:
|
188 |
# Consecutive empty lines will cause empty sentences
|
189 |
continue
|
190 |
+
assert len(current_tokens) == len(current_ner_tags), "๐ between len of tokens & labels"
|
191 |
+
assert len(current_ner_tags) == len(
|
192 |
+
current_nested_ner_tags
|
193 |
+
), "๐ between len of labels & nested labels"
|
194 |
assert current_source, "๐ฅ Source for new sentence was not set"
|
195 |
sentence = (
|
196 |
sentence_counter,
|
197 |
{
|
198 |
"id": str(sentence_counter),
|
199 |
"tokens": current_tokens,
|
200 |
+
"ner_tags": current_ner_tags,
|
201 |
+
"nested_ner_tags": current_nested_ner_tags,
|
202 |
"source": current_source,
|
203 |
},
|
204 |
)
|
205 |
sentence_counter += 1
|
206 |
current_tokens = []
|
207 |
+
current_ner_tags = []
|
208 |
+
current_nested_ner_tags = []
|
209 |
current_source = ""
|
210 |
yield sentence
|
211 |
# Don't forget last sentence in dataset ๐ง
|
212 |
yield sentence_counter, {
|
213 |
"id": str(sentence_counter),
|
214 |
"tokens": current_tokens,
|
215 |
+
"ner_tags": current_ner_tags,
|
216 |
+
"nested_ner_tags": current_nested_ner_tags,
|
217 |
"source": current_source,
|
218 |
}
|