leondz commited on
Commit
f3fa5fb
1 Parent(s): cacdb08

initial commit

Browse files
Files changed (3) hide show
  1. README.md +233 -0
  2. dataset_infos.json +1 -0
  3. ipm_nel.py +168 -0
README.md ADDED
@@ -0,0 +1,233 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ annotations_creators:
3
+ - crowdsourced
4
+ language_creators:
5
+ - found
6
+ languages:
7
+ - en
8
+ licenses:
9
+ - other
10
+ multilinguality:
11
+ - monolingual
12
+ size_categories:
13
+ - 10K<n<100K
14
+ source_datasets:
15
+ - extended|other-reuters-corpus
16
+ task_categories:
17
+ - token-classification
18
+ task_ids:
19
+ - named-entity-recognition
20
+ - part-of-speech-tagging
21
+ paperswithcode_id: conll-2003
22
+ pretty_name: CoNLL-2003
23
+ ---
24
+
25
+ # Dataset Card for "conll2003"
26
+
27
+ ## Table of Contents
28
+ - [Dataset Description](#dataset-description)
29
+ - [Dataset Summary](#dataset-summary)
30
+ - [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
31
+ - [Languages](#languages)
32
+ - [Dataset Structure](#dataset-structure)
33
+ - [Data Instances](#data-instances)
34
+ - [Data Fields](#data-fields)
35
+ - [Data Splits](#data-splits)
36
+ - [Dataset Creation](#dataset-creation)
37
+ - [Curation Rationale](#curation-rationale)
38
+ - [Source Data](#source-data)
39
+ - [Annotations](#annotations)
40
+ - [Personal and Sensitive Information](#personal-and-sensitive-information)
41
+ - [Considerations for Using the Data](#considerations-for-using-the-data)
42
+ - [Social Impact of Dataset](#social-impact-of-dataset)
43
+ - [Discussion of Biases](#discussion-of-biases)
44
+ - [Other Known Limitations](#other-known-limitations)
45
+ - [Additional Information](#additional-information)
46
+ - [Dataset Curators](#dataset-curators)
47
+ - [Licensing Information](#licensing-information)
48
+ - [Citation Information](#citation-information)
49
+ - [Contributions](#contributions)
50
+
51
+ ## Dataset Description
52
+
53
+ - **Homepage:** [https://www.aclweb.org/anthology/W03-0419/](https://www.aclweb.org/anthology/W03-0419/)
54
+ - **Repository:** [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
55
+ - **Paper:** [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
56
+ - **Point of Contact:** [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
57
+ - **Size of downloaded dataset files:** 4.63 MB
58
+ - **Size of the generated dataset:** 9.78 MB
59
+ - **Total amount of disk used:** 14.41 MB
60
+
61
+ ### Dataset Summary
62
+
63
+ The shared task of CoNLL-2003 concerns language-independent named entity recognition. We will concentrate on
64
+ four types of named entities: persons, locations, organizations and names of miscellaneous entities that do
65
+ not belong to the previous three groups.
66
+
67
+ The CoNLL-2003 shared task data files contain four columns separated by a single space. Each word has been put on
68
+ a separate line and there is an empty line after each sentence. The first item on each line is a word, the second
69
+ a part-of-speech (POS) tag, the third a syntactic chunk tag and the fourth the named entity tag. The chunk tags
70
+ and the named entity tags have the format I-TYPE which means that the word is inside a phrase of type TYPE. Only
71
+ if two phrases of the same type immediately follow each other, the first word of the second phrase will have tag
72
+ B-TYPE to show that it starts a new phrase. A word with tag O is not part of a phrase. Note the dataset uses IOB2
73
+ tagging scheme, whereas the original dataset uses IOB1.
74
+
75
+ For more details see https://www.clips.uantwerpen.be/conll2003/ner/ and https://www.aclweb.org/anthology/W03-0419
76
+
77
+ ### Supported Tasks and Leaderboards
78
+
79
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
80
+
81
+ ### Languages
82
+
83
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
84
+
85
+ ## Dataset Structure
86
+
87
+ ### Data Instances
88
+
89
+ #### conll2003
90
+
91
+ - **Size of downloaded dataset files:** 4.63 MB
92
+ - **Size of the generated dataset:** 9.78 MB
93
+ - **Total amount of disk used:** 14.41 MB
94
+
95
+ An example of 'train' looks as follows.
96
+
97
+ ```
98
+ {
99
+ "chunk_tags": [11, 12, 12, 21, 13, 11, 11, 21, 13, 11, 12, 13, 11, 21, 22, 11, 12, 17, 11, 21, 17, 11, 12, 12, 21, 22, 22, 13, 11, 0],
100
+ "id": "0",
101
+ "ner_tags": [0, 3, 4, 0, 0, 0, 0, 0, 0, 7, 0, 0, 0, 0, 0, 7, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
102
+ "pos_tags": [12, 22, 22, 38, 15, 22, 28, 38, 15, 16, 21, 35, 24, 35, 37, 16, 21, 15, 24, 41, 15, 16, 21, 21, 20, 37, 40, 35, 21, 7],
103
+ "tokens": ["The", "European", "Commission", "said", "on", "Thursday", "it", "disagreed", "with", "German", "advice", "to", "consumers", "to", "shun", "British", "lamb", "until", "scientists", "determine", "whether", "mad", "cow", "disease", "can", "be", "transmitted", "to", "sheep", "."]
104
+ }
105
+ ```
106
+
107
+ The original data files have `-DOCSTART-` lines used to separate documents, but these lines are removed here.
108
+ Indeed `-DOCSTART-` is a special line that acts as a boundary between two different documents, and it is filtered out in this implementation.
109
+
110
+ ### Data Fields
111
+
112
+ The data fields are the same among all splits.
113
+
114
+ #### conll2003
115
+ - `id`: a `string` feature.
116
+ - `tokens`: a `list` of `string` features.
117
+ - `pos_tags`: a `list` of classification labels (`int`). Full tagset with indices:
118
+
119
+ ```python
120
+ {'"': 0, "''": 1, '#': 2, '$': 3, '(': 4, ')': 5, ',': 6, '.': 7, ':': 8, '``': 9, 'CC': 10, 'CD': 11, 'DT': 12,
121
+ 'EX': 13, 'FW': 14, 'IN': 15, 'JJ': 16, 'JJR': 17, 'JJS': 18, 'LS': 19, 'MD': 20, 'NN': 21, 'NNP': 22, 'NNPS': 23,
122
+ 'NNS': 24, 'NN|SYM': 25, 'PDT': 26, 'POS': 27, 'PRP': 28, 'PRP$': 29, 'RB': 30, 'RBR': 31, 'RBS': 32, 'RP': 33,
123
+ 'SYM': 34, 'TO': 35, 'UH': 36, 'VB': 37, 'VBD': 38, 'VBG': 39, 'VBN': 40, 'VBP': 41, 'VBZ': 42, 'WDT': 43,
124
+ 'WP': 44, 'WP$': 45, 'WRB': 46}
125
+ ```
126
+
127
+ - `chunk_tags`: a `list` of classification labels (`int`). Full tagset with indices:
128
+
129
+ ```python
130
+ {'O': 0, 'B-ADJP': 1, 'I-ADJP': 2, 'B-ADVP': 3, 'I-ADVP': 4, 'B-CONJP': 5, 'I-CONJP': 6, 'B-INTJ': 7, 'I-INTJ': 8,
131
+ 'B-LST': 9, 'I-LST': 10, 'B-NP': 11, 'I-NP': 12, 'B-PP': 13, 'I-PP': 14, 'B-PRT': 15, 'I-PRT': 16, 'B-SBAR': 17,
132
+ 'I-SBAR': 18, 'B-UCP': 19, 'I-UCP': 20, 'B-VP': 21, 'I-VP': 22}
133
+ ```
134
+
135
+ - `ner_tags`: a `list` of classification labels (`int`). Full tagset with indices:
136
+
137
+ ```python
138
+ {'O': 0, 'B-PER': 1, 'I-PER': 2, 'B-ORG': 3, 'I-ORG': 4, 'B-LOC': 5, 'I-LOC': 6, 'B-MISC': 7, 'I-MISC': 8}
139
+ ```
140
+
141
+ ### Data Splits
142
+
143
+ | name |train|validation|test|
144
+ |---------|----:|---------:|---:|
145
+ |conll2003|14041| 3250|3453|
146
+
147
+ ## Dataset Creation
148
+
149
+ ### Curation Rationale
150
+
151
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
152
+
153
+ ### Source Data
154
+
155
+ #### Initial Data Collection and Normalization
156
+
157
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
158
+
159
+ #### Who are the source language producers?
160
+
161
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
162
+
163
+ ### Annotations
164
+
165
+ #### Annotation process
166
+
167
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
168
+
169
+ #### Who are the annotators?
170
+
171
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
172
+
173
+ ### Personal and Sensitive Information
174
+
175
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
176
+
177
+ ## Considerations for Using the Data
178
+
179
+ ### Social Impact of Dataset
180
+
181
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
182
+
183
+ ### Discussion of Biases
184
+
185
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
186
+
187
+ ### Other Known Limitations
188
+
189
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
190
+
191
+ ## Additional Information
192
+
193
+ ### Dataset Curators
194
+
195
+ [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
196
+
197
+ ### Licensing Information
198
+
199
+ From the [CoNLL2003 shared task](https://www.clips.uantwerpen.be/conll2003/ner/) page:
200
+
201
+ > The English data is a collection of news wire articles from the Reuters Corpus. The annotation has been done by people of the University of Antwerp. Because of copyright reasons we only make available the annotations. In order to build the complete data sets you will need access to the Reuters Corpus. It can be obtained for research purposes without any charge from NIST.
202
+
203
+ The copyrights are defined below, from the [Reuters Corpus page](https://trec.nist.gov/data/reuters/reuters.html):
204
+
205
+ > The stories in the Reuters Corpus are under the copyright of Reuters Ltd and/or Thomson Reuters, and their use is governed by the following agreements:
206
+ >
207
+ > [Organizational agreement](https://trec.nist.gov/data/reuters/org_appl_reuters_v4.html)
208
+ >
209
+ > This agreement must be signed by the person responsible for the data at your organization, and sent to NIST.
210
+ >
211
+ > [Individual agreement](https://trec.nist.gov/data/reuters/ind_appl_reuters_v4.html)
212
+ >
213
+ > This agreement must be signed by all researchers using the Reuters Corpus at your organization, and kept on file at your organization.
214
+
215
+ ### Citation Information
216
+
217
+ ```
218
+ @inproceedings{tjong-kim-sang-de-meulder-2003-introduction,
219
+ title = "Introduction to the {C}o{NLL}-2003 Shared Task: Language-Independent Named Entity Recognition",
220
+ author = "Tjong Kim Sang, Erik F. and
221
+ De Meulder, Fien",
222
+ booktitle = "Proceedings of the Seventh Conference on Natural Language Learning at {HLT}-{NAACL} 2003",
223
+ year = "2003",
224
+ url = "https://www.aclweb.org/anthology/W03-0419",
225
+ pages = "142--147",
226
+ }
227
+
228
+ ```
229
+
230
+
231
+ ### Contributions
232
+
233
+ Thanks to [@jplu](https://github.com/jplu), [@vblagoje](https://github.com/vblagoje), [@lhoestq](https://github.com/lhoestq) for adding this dataset.
dataset_infos.json ADDED
@@ -0,0 +1 @@
 
1
+ {"ipm_nel": {"description": "This data is for the task of named entity recognition and linking/disambiguation over tweets. It comprises\nthe addition of an entity URI layer on top of an NER-annotated tweet dataset. The task is to detect entities\nand then provide a correct link to them in DBpedia, thus disambiguating otherwise ambiguous entity surface\nforms; for example, this means linking \"Paris\" to the correct instance of a city named that (e.g. Paris, \nFrance vs. Paris, Texas).\n\nThe data concentrates on ten types of named entities: company, facility, geographic location, movie, musical\nartist, person, product, sports team, TV show, and other.\n\nThe file is tab separated, in CoNLL format, with line breaks between tweets.\nData preserves the tokenisation used in the Ritter datasets.\nPoS labels are not present for all tweets, but where they could be found in the Ritter\ndata, they're given. In cases where a URI could not be agreed, or was not present in\nDBpedia, there is a NIL. See the paper for a full description of the methodology.\n\nFor more details see http://www.derczynski.com/papers/ner_single.pdf or https://www.sciencedirect.com/science/article/abs/pii/S0306457314001034\n", "citation": "@article{derczynski2015analysis,\n title={Analysis of named entity recognition and linking for tweets},\n author={Derczynski, Leon and Maynard, Diana and Rizzo, Giuseppe and Van Erp, Marieke and Gorrell, Genevieve and Troncy, Rapha{\"e}l and Petrak, Johann and Bontcheva, Kalina},\n journal={Information Processing \\& Management},\n volume={51},\n number={2},\n pages={32--49},\n year={2015},\n publisher={Elsevier}\n}\n", "homepage": "https://www.sciencedirect.com/science/article/pii/S0306457314001034", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "tokens": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "uris": {"dtype": "string", "id": null, "_type": "Value"}, "ner_tags": {"feature": {"num_classes": 21, "names": ["O", "B-company", "B-facility", "B-geo-loc", "B-movie", "B-musicartist", "B-other", "B-person", "B-product", "B-sportsteam", "B-tvshow", "I-company", "I-facility", "I-geo-loc", "I-movie", "I-musicartist", "I-other", "I-person", "I-product", "I-sportsteam", "I-tvshow"], "id": null, "_type": "ClassLabel"}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "ipm_nel2003", "config_name": "ipm_nel", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 96989, "num_examples": 183, "dataset_name": "ipm_nel2003"}}, "download_checksums": {"http://www.derczynski.com/resources/ipm_nel.tar.gz": {"num_bytes": 2409032, "checksum": "c5a2fb618f19b591e6091d1538906db60ae16d2dbe7280533e4c2f8f8dabda9c"}}, "download_size": 2409032, "post_processing_size": null, "dataset_size": 96989, "size_in_bytes": 2506021}}
ipm_nel.py ADDED
@@ -0,0 +1,168 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # coding=utf-8
2
+ # Copyright 2020 HuggingFace Datasets Authors.
3
+ #
4
+ # Licensed under the Apache License, Version 2.0 (the "License");
5
+ # you may not use this file except in compliance with the License.
6
+ # You may obtain a copy of the License at
7
+ #
8
+ # http://www.apache.org/licenses/LICENSE-2.0
9
+ #
10
+ # Unless required by applicable law or agreed to in writing, software
11
+ # distributed under the License is distributed on an "AS IS" BASIS,
12
+ # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13
+ # See the License for the specific language governing permissions and
14
+ # limitations under the License.
15
+
16
+ # Lint as: python3
17
+ """Introduction to the CoNLL-2003 Shared Task: Language-Independent Named Entity Recognition"""
18
+
19
+ import os
20
+
21
+ import datasets
22
+
23
+
24
+ logger = datasets.logging.get_logger(__name__)
25
+
26
+
27
+ _CITATION = """\
28
+ @article{derczynski2015analysis,
29
+ title={Analysis of named entity recognition and linking for tweets},
30
+ author={Derczynski, Leon and Maynard, Diana and Rizzo, Giuseppe and Van Erp, Marieke and Gorrell, Genevieve and Troncy, Rapha{\"e}l and Petrak, Johann and Bontcheva, Kalina},
31
+ journal={Information Processing \& Management},
32
+ volume={51},
33
+ number={2},
34
+ pages={32--49},
35
+ year={2015},
36
+ publisher={Elsevier}
37
+ }
38
+ """
39
+
40
+ _DESCRIPTION = """\
41
+ This data is for the task of named entity recognition and linking/disambiguation over tweets. It comprises
42
+ the addition of an entity URI layer on top of an NER-annotated tweet dataset. The task is to detect entities
43
+ and then provide a correct link to them in DBpedia, thus disambiguating otherwise ambiguous entity surface
44
+ forms; for example, this means linking "Paris" to the correct instance of a city named that (e.g. Paris,
45
+ France vs. Paris, Texas).
46
+
47
+ The data concentrates on ten types of named entities: company, facility, geographic location, movie, musical
48
+ artist, person, product, sports team, TV show, and other.
49
+
50
+ The file is tab separated, in CoNLL format, with line breaks between tweets.
51
+ Data preserves the tokenisation used in the Ritter datasets.
52
+ PoS labels are not present for all tweets, but where they could be found in the Ritter
53
+ data, they're given. In cases where a URI could not be agreed, or was not present in
54
+ DBpedia, there is a NIL. See the paper for a full description of the methodology.
55
+
56
+ For more details see http://www.derczynski.com/papers/ner_single.pdf or https://www.sciencedirect.com/science/article/abs/pii/S0306457314001034
57
+ """
58
+
59
+ _URL = "http://www.derczynski.com/resources/ipm_nel.tar.gz"
60
+ _TRAINING_FILE = "ipm_nel_corpus/ipm_nel.conll"
61
+
62
+
63
+ class IpmNelConfig(datasets.BuilderConfig):
64
+ """BuilderConfig for IPM NEL"""
65
+
66
+ def __init__(self, **kwargs):
67
+ """BuilderConfig for IPM NEL.
68
+
69
+ Args:
70
+ **kwargs: keyword arguments forwarded to super.
71
+ """
72
+ super(IpmNelConfig, self).__init__(**kwargs)
73
+
74
+
75
+ class IpmNel2003(datasets.GeneratorBasedBuilder):
76
+ """IpmNel2003 dataset."""
77
+
78
+ BUILDER_CONFIGS = [
79
+ IpmNelConfig(name="ipm_nel", version=datasets.Version("1.0.0"), description="IPM NEL dataset"),
80
+ ]
81
+
82
+ def _info(self):
83
+ return datasets.DatasetInfo(
84
+ description=_DESCRIPTION,
85
+ features=datasets.Features(
86
+ {
87
+ "id": datasets.Value("string"),
88
+ "tokens": datasets.Sequence(datasets.Value("string")),
89
+ "uris": datasets.Value("string"),
90
+ "ner_tags": datasets.Sequence(
91
+ datasets.features.ClassLabel(
92
+ names=[
93
+ "O",
94
+ "B-company",
95
+ "B-facility",
96
+ "B-geo-loc",
97
+ "B-movie",
98
+ "B-musicartist",
99
+ "B-other",
100
+ "B-person",
101
+ "B-product",
102
+ "B-sportsteam",
103
+ "B-tvshow",
104
+ "I-company",
105
+ "I-facility",
106
+ "I-geo-loc",
107
+ "I-movie",
108
+ "I-musicartist",
109
+ "I-other",
110
+ "I-person",
111
+ "I-product",
112
+ "I-sportsteam",
113
+ "I-tvshow",
114
+ ]
115
+ )
116
+ ),
117
+
118
+ }
119
+ ),
120
+ supervised_keys=None,
121
+ homepage="https://www.sciencedirect.com/science/article/pii/S0306457314001034",
122
+ citation=_CITATION,
123
+ )
124
+
125
+ def _split_generators(self, dl_manager):
126
+ """Returns SplitGenerators."""
127
+ downloaded_file = dl_manager.download_and_extract(_URL)
128
+ data_files = {
129
+ "train": os.path.join(downloaded_file, _TRAINING_FILE),
130
+ }
131
+
132
+ return [
133
+ datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": data_files["train"]}),
134
+ ]
135
+
136
+ def _generate_examples(self, filepath):
137
+ logger.info("⏳ Generating examples from = %s", filepath)
138
+ with open(filepath, encoding="utf-8") as f:
139
+ guid = 0
140
+ tokens = []
141
+ ner_tags = []
142
+ uris = []
143
+ for line in f:
144
+ if line.startswith("-DOCSTART-") or line.strip() == "":
145
+ if tokens:
146
+ yield guid, {
147
+ "id": str(guid),
148
+ "tokens": tokens,
149
+ "ner_tags": ner_tags,
150
+ "uris": uris,
151
+ }
152
+ guid += 1
153
+ tokens = []
154
+ uris = []
155
+ ner_tags = []
156
+ else:
157
+ # ipm_nel items are tab separated
158
+ splits = line.split("\t")
159
+ tokens.append(splits[0])
160
+ uris.append(splits[1])
161
+ ner_tags.append(splits[2].rstrip())
162
+ # last example
163
+ yield guid, {
164
+ "id": str(guid),
165
+ "tokens": tokens,
166
+ "ner_tags": ner_tags,
167
+ "uris": uris,
168
+ }