qanastek commited on
Commit
d3d64f1
1 Parent(s): 47fdd7e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +381 -362
README.md CHANGED
@@ -1,362 +1,381 @@
1
- # ANTILLES : An Open French Linguistically Enriched Part-of-Speech Corpus
2
-
3
- ## Table of Contents
4
- - [Dataset Card for [Needs More Information]](#dataset-card-for-needs-more-information)
5
- - [Table of Contents](#table-of-contents)
6
- - [Dataset Description](#dataset-description)
7
- - [Dataset Summary](#dataset-summary)
8
- - [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
9
- - [Languages](#languages)
10
- - [Dataset Structure](#dataset-structure)
11
- - [Data Instances](#data-instances)
12
- - [sent_id = fr-ud-dev_00005](#sent_id--fr-ud-dev_00005)
13
- - [text = Travail de trés grande qualité exécuté par un imprimeur artisan passionné.](#text--travail-de-trs-grande-qualit-excut-par-un-imprimeur-artisan-passionn)
14
- - [Data Fields](#data-fields)
15
- - [Data Splits](#data-splits)
16
- - [Dataset Creation](#dataset-creation)
17
- - [Curation Rationale](#curation-rationale)
18
- - [Source Data](#source-data)
19
- - [Initial Data Collection and Normalization](#initial-data-collection-and-normalization)
20
- - [Who are the source language producers?](#who-are-the-source-language-producers)
21
- - [Annotations](#annotations)
22
- - [Annotation process](#annotation-process)
23
- - [Who are the annotators?](#who-are-the-annotators)
24
- - [Personal and Sensitive Information](#personal-and-sensitive-information)
25
- - [Considerations for Using the Data](#considerations-for-using-the-data)
26
- - [Social Impact of Dataset](#social-impact-of-dataset)
27
- - [Discussion of Biases](#discussion-of-biases)
28
- - [Other Known Limitations](#other-known-limitations)
29
- - [Additional Information](#additional-information)
30
- - [Dataset Curators](#dataset-curators)
31
- - [Licensing Information](#licensing-information)
32
- - [Citation Information](#citation-information)
33
-
34
- ## Dataset Description
35
-
36
- - **Homepage:** https://qanastek.github.io/ANTILLES/
37
- - **Repository:** https://github.com/qanastek/ANTILLES
38
- - **Paper:** [Needs More Information]
39
- - **Leaderboard:** [Needs More Information]
40
- - **Point of Contact:** [email protected]
41
-
42
- ### Dataset Summary
43
-
44
- `ANTILLES` is a part-of-speech tagging corpora based on [UD_French-GSD](https://universaldependencies.org/treebanks/fr_gsd/index.html) which was originally created in 2015 and is based on the [universal dependency treebank v2.0](https://github.com/ryanmcd/uni-dep-tb).
45
-
46
- Originally, the corpora consists of 400,399 words (16,341 sentences) and had 17 different classes. Now, after applying our tags augmentation script `transform.py`, we obtain 60 different classes which add semantic information such as: the gender, number, mood, person, tense or verb form given in the different CoNLL-U fields from the original corpora.
47
-
48
- We based our tags on the level of details given by the [LIA_TAGG](http://pageperso.lif.univ-mrs.fr/frederic.bechet/download.html) statistical POS tagger written by [Frédéric Béchet](http://pageperso.lif.univ-mrs.fr/frederic.bechet/index-english.html) in 2001.
49
-
50
- <a rel="license" href="http://creativecommons.org/licenses/by-sa/4.0/"><img alt="Creative Commons License" style="border-width:0" src="https://i.creativecommons.org/l/by-sa/4.0/88x31.png" /></a><br />This work is licensed under a <a rel="license" href="http://creativecommons.org/licenses/by-sa/4.0/">Creative Commons Attribution-ShareAlike 4.0 International License</a>.
51
-
52
- ### Supported Tasks and Leaderboards
53
-
54
- `part-of-speech-tagging`: The dataset can be used to train a model for part-of-speech-tagging. The performance is measured by how high its F1 score is. A Flair Sequence-To-Sequence model trained to tag tokens from Wikipedia passages achieves a F1 score (micro) of 0.952.
55
-
56
- ### Languages
57
-
58
- The text in the dataset is in French, as spoken by [Wikipedia](https://en.wikipedia.org/wiki/Main_Page) users. The associated [BCP-47](https://tools.ietf.org/html/bcp47) code is `fr`.
59
-
60
- ## Load the dataset
61
-
62
- ### HuggingFace
63
-
64
- ```python
65
- from datasets import load_dataset
66
- dataset = load_dataset("qanastek/ANTILLES")
67
- print(dataset)
68
- ```
69
-
70
- ### FlairNLP
71
-
72
- ```python
73
- from flair.datasets import UniversalDependenciesCorpus
74
- corpus: Corpus = UniversalDependenciesCorpus(
75
- data_folder='ANTILLES',
76
- train_file="train.conllu",
77
- test_file="test.conllu",
78
- dev_file="dev.conllu"
79
- )
80
- ```
81
-
82
- ## Load the model
83
-
84
- ### Flair ([model](https://huggingface.co/qanastek/pos-french))
85
-
86
- ```python
87
- from flair.models import SequenceTagger
88
- tagger = SequenceTagger.load("qanastek/pos-french")
89
- ```
90
-
91
- ## HuggingFace Spaces
92
-
93
- <table style="width: fit-content;">
94
- <thead>
95
- <tr>
96
- <td>
97
- <a href="https://huggingface.co/spaces/qanastek/French-Part-Of-Speech-Tagging">
98
- <img src="https://huggingface.co/datasets/qanastek/ANTILLES/raw/main/imgs/en.png" width="160">
99
- </a>
100
- </td>
101
- <td>
102
- <a href="https://huggingface.co/spaces/qanastek/Etiqueteur-Morphosyntaxique-Etendu">
103
- <img src="https://huggingface.co/datasets/qanastek/ANTILLES/raw/main/imgs/fr.png" width="160">
104
- </a>
105
- </td>
106
- </tr>
107
- </thead>
108
- </table>
109
-
110
- ## Dataset Structure
111
-
112
- ### Data Instances
113
-
114
- ```plain
115
- # sent_id = fr-ud-dev_00005
116
- # text = Travail de trés grande qualité exécuté par un imprimeur artisan passionné.
117
- 1 Travail travail NMS _ Gender=Masc|Number=Sing 0 root _ wordform=travail
118
- 2 de de PREP _ _ 5 case _ _
119
- 3 trés trés ADV _ _ 4 advmod _ _
120
- 4 grande grand ADJFS _ Gender=Fem|Number=Sing 5 amod _ _
121
- 5 qualité qualité NFS _ Gender=Fem|Number=Sing 1 nmod _ _
122
- 6 exécuté exécuter VPPMS _ Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part 1 acl _ _
123
- 7 par par PREP _ _ 9 case _ _
124
- 8 un un DINTMS _ Definite=Ind|Gender=Masc|Number=Sing|PronType=Art 9 det _ _
125
- 9 imprimeur imprimeur NMS _ Gender=Masc|Number=Sing 6 obl:agent _ _
126
- 10 artisan artisan NMS _ Gender=Masc|Number=Sing 9 nmod _ _
127
- 11 passionné passionné ADJMS _ Gender=Masc|Number=Sing 9 amod _ SpaceAfter=No
128
- 12 . . YPFOR _ _ 1 punct _ _
129
- ```
130
-
131
- ### Data Fields
132
-
133
- | Abbreviation | Description | Examples | # tokens |
134
- |:--------:|:--------:|:--------:|:--------:|
135
- | PREP | Preposition | de | 63 738 |
136
- | AUX | Auxiliary Verb | est | 12 886 |
137
- | ADV | Adverb | toujours | 14 969 |
138
- | COSUB | Subordinating conjunction | que | 3 007 |
139
- | COCO | Coordinating Conjunction | et | 10 102 |
140
- | PART | Demonstrative particle | -t | 93 |
141
- | PRON | Pronoun | qui ce quoi | 667 |
142
- | PDEMMS | Singular Masculine Demonstrative Pronoun | ce | 1 950 |
143
- | PDEMMP | Plurial Masculine Demonstrative Pronoun | ceux | 108 |
144
- | PDEMFS | Singular Feminine Demonstrative Pronoun | cette | 1 004 |
145
- | PDEMFP | Plurial Feminine Demonstrative Pronoun | celles | 53 |
146
- | PINDMS | Singular Masculine Indefinite Pronoun | tout | 961 |
147
- | PINDMP | Plurial Masculine Indefinite Pronoun | autres | 89 |
148
- | PINDFS | Singular Feminine Indefinite Pronoun | chacune | 136 |
149
- | PINDFP | Plurial Feminine Indefinite Pronoun | certaines | 31 |
150
- | PROPN | Proper noun | houston | 22 135 |
151
- | XFAMIL | Last name | levy | 6 449 |
152
- | NUM | Numerical Adjectives | trentaine vingtaine | 67 |
153
- | DINTMS | Masculine Numerical Adjectives | un | 4 254 |
154
- | DINTFS | Feminine Numerical Adjectives | une | 3 543 |
155
- | PPOBJMS | Singular Masculine Pronoun complements of objects | le lui | 1 425 |
156
- | PPOBJMP | Plurial Masculine Pronoun complements of objects | eux y | 212 |
157
- | PPOBJFS | Singular Feminine Pronoun complements of objects | moi la | 358 |
158
- | PPOBJFP | Plurial Feminine Pronoun complements of objects | en y | 70 |
159
- | PPER1S | Personal Pronoun First Person Singular | je | 571 |
160
- | PPER2S | Personal Pronoun Second Person Singular | tu | 19 |
161
- | PPER3MS | Personal Pronoun Third Person Masculine Singular | il | 3 938 |
162
- | PPER3MP | Personal Pronoun Third Person Masculine Plurial | ils | 513 |
163
- | PPER3FS | Personal Pronoun Third Person Feminine Singular | elle | 992 |
164
- | PPER3FP | Personal Pronoun Third Person Feminine Plurial | elles | 121 |
165
- | PREFS | Reflexive Pronouns First Person of Singular | me m' | 120 |
166
- | PREF | Reflexive Pronouns Third Person of Singular | se s' | 2 337 |
167
- | PREFP | Reflexive Pronouns First / Second Person of Plurial | nous vous | 686 |
168
- | VERB | Verb | obtient | 21 131 |
169
- | VPPMS | Singular Masculine Participle Past Verb | formulé | 6 275 |
170
- | VPPMP | Plurial Masculine Participle Past Verb | classés | 1 352 |
171
- | VPPFS | Singular Feminine Participle Past Verb | appelée | 2 434 |
172
- | VPPFP | Plurial Feminine Participle Past Verb | sanctionnées | 813 |
173
- | VPPRE | Present participle | étant | 2 |
174
- | DET | Determinant | les l' | 25 206 |
175
- | DETMS | Singular Masculine Determinant | les | 15 444 |
176
- | DETFS | Singular Feminine Determinant | la | 10 978 |
177
- | ADJ | Adjective | capable sérieux | 1 075 |
178
- | ADJMS | Singular Masculine Adjective | grand important | 8 338 |
179
- | ADJMP | Plurial Masculine Adjective | grands petits | 3 274 |
180
- | ADJFS | Singular Feminine Adjective | franéaise petite | 8 004 |
181
- | ADJFP | Plurial Feminine Adjective | légéres petites | 3 041 |
182
- | NOUN | Noun | temps | 1 389 |
183
- | NMS | Singular Masculine Noun | drapeau | 29 698 |
184
- | NMP | Plurial Masculine Noun | journalistes | 10 882 |
185
- | NFS | Singular Feminine Noun | téte | 25 414 |
186
- | NFP | Plurial Feminine Noun | ondes | 7 448 |
187
- | PREL | Relative Pronoun | qui dont | 2 976 |
188
- | PRELMS | Singular Masculine Relative Pronoun | lequel | 94 |
189
- | PRELMP | Plurial Masculine Relative Pronoun | lesquels | 29 |
190
- | PRELFS | Singular Feminine Relative Pronoun | laquelle | 70 |
191
- | PRELFP | Plurial Feminine Relative Pronoun | lesquelles | 25 |
192
- | PINTFS | Singular Feminine Interrogative Pronoun | laquelle | 3 |
193
- | INTJ | Interjection | merci bref | 75 |
194
- | CHIF | Numbers | 1979 10 | 10 417 |
195
- | SYM | Symbol | é % | 705 |
196
- | YPFOR | Endpoint | . | 15 088 |
197
- | PUNCT | Ponctuation | : , | 28 918 |
198
- | MOTINC | Unknown words | Technology Lady | 2 022 |
199
- | X | Typos & others | sfeir 3D statu | 175 |
200
-
201
- ### Data Splits
202
-
203
- | | Train | Dev | Test |
204
- |:------------------:|:------:|:------:|:-----:|
205
- | # Docs | 14 449 | 1 476 | 416 |
206
- | Avg # Tokens / Doc | 24.54 | 24.19 | 24.08 |
207
-
208
- ## Dataset Creation
209
-
210
- ### Curation Rationale
211
-
212
- [Needs More Information]
213
-
214
- ### Source Data
215
-
216
- #### Initial Data Collection and Normalization
217
-
218
- [Needs More Information]
219
-
220
- #### Who are the source language producers?
221
-
222
- [Needs More Information]
223
-
224
- ### Annotations
225
-
226
- #### Annotation process
227
-
228
- [Needs More Information]
229
-
230
- #### Who are the annotators?
231
-
232
- [Needs More Information]
233
-
234
- ### Personal and Sensitive Information
235
-
236
- The corpora is free of personal or sensitive information since it has been based on `Wikipedia` articles content.
237
-
238
- ## Considerations for Using the Data
239
-
240
- ### Social Impact of Dataset
241
-
242
- [Needs More Information]
243
-
244
- ### Discussion of Biases
245
-
246
- The nature of the corpora introduce various biases such as the names of the streets which are temporaly based and can therefore introduce named entity like author or event names. For example, street names such as `Rue Victor-Hugo` or `Rue Pasteur` doesn't exist before the 20's century in France.
247
-
248
- ### Other Known Limitations
249
-
250
- [Needs More Information]
251
-
252
- ## Additional Information
253
-
254
- ### Dataset Curators
255
-
256
- __ANTILLES__: Labrak Yanis, Dufour Richard
257
-
258
- __UD_FRENCH-GSD__: de Marneffe Marie-Catherine, Guillaume Bruno, McDonald Ryan, Suhr Alane, Nivre Joakim, Grioni Matias, Dickerson Carly, Perrier Guy
259
-
260
- __Universal Dependency__: Ryan McDonald, Joakim Nivre, Yvonne Quirmbach-Brundage, Yoav Goldberg, Dipanjan Das, Kuzman Ganchev, Keith Hall, Slav Petrov, Hao Zhang, Oscar Tackstrom, Claudia Bedini, Nuria Bertomeu Castello and Jungmee Lee
261
-
262
- ### Licensing Information
263
-
264
- ```plain
265
- For the following languages
266
-
267
- German, Spanish, French, Indonesian, Italian, Japanese, Korean and Brazilian
268
- Portuguese
269
-
270
- we will distinguish between two portions of the data.
271
-
272
- 1. The underlying text for sentences that were annotated. This data Google
273
- asserts no ownership over and no copyright over. Some or all of these
274
- sentences may be copyrighted in some jurisdictions. Where copyrighted,
275
- Google collected these sentences under exceptions to copyright or implied
276
- license rights. GOOGLE MAKES THEM AVAILABLE TO YOU 'AS IS', WITHOUT ANY
277
- WARRANTY OF ANY KIND, WHETHER EXPRESS OR IMPLIED.
278
-
279
- 2. The annotations -- part-of-speech tags and dependency annotations. These are
280
- made available under a CC BY-SA 4.0. GOOGLE MAKES
281
- THEM AVAILABLE TO YOU 'AS IS', WITHOUT ANY WARRANTY OF ANY KIND, WHETHER
282
- EXPRESS OR IMPLIED. See attached LICENSE file for the text of CC BY-NC-SA.
283
-
284
- Portions of the German data were sampled from the CoNLL 2006 Tiger Treebank
285
- data. Hans Uszkoreit graciously gave permission to use the underlying
286
- sentences in this data as part of this release.
287
-
288
- Any use of the data should reference the above plus:
289
-
290
- Universal Dependency Annotation for Multilingual Parsing
291
- Ryan McDonald, Joakim Nivre, Yvonne Quirmbach-Brundage, Yoav Goldberg,
292
- Dipanjan Das, Kuzman Ganchev, Keith Hall, Slav Petrov, Hao Zhang,
293
- Oscar Tackstrom, Claudia Bedini, Nuria Bertomeu Castello and Jungmee Lee
294
- Proceedings of ACL 2013
295
- ```
296
-
297
- ### Citation Information
298
-
299
- Please cite the following paper when using this model.
300
-
301
- ANTILLES extended corpus:
302
-
303
- ```latex
304
- @software{Labrak_ANTILLES_An_2021,
305
- author = {Labrak, Yanis and Dufour, Richard},
306
- month = {11},
307
- title = {{ANTILLES : An Open French Linguistically Enriched Part-of-Speech Corpus}},
308
- url = {https://github.com/qanastek/ANTILLES},
309
- version = {1.0.0},
310
- year = {2021}
311
- }
312
- ```
313
-
314
- UD_French-GSD corpora:
315
-
316
- ```latex
317
- @misc{
318
- universaldependencies,
319
- title={UniversalDependencies/UD_French-GSD},
320
- url={https://github.com/UniversalDependencies/UD_French-GSD}, journal={GitHub},
321
- author={UniversalDependencies}
322
- }
323
- ```
324
-
325
- {U}niversal {D}ependency Annotation for Multilingual Parsing:
326
-
327
- ```latex
328
- @inproceedings{mcdonald-etal-2013-universal,
329
- title = "{U}niversal {D}ependency Annotation for Multilingual Parsing",
330
- author = {McDonald, Ryan and
331
- Nivre, Joakim and
332
- Quirmbach-Brundage, Yvonne and
333
- Goldberg, Yoav and
334
- Das, Dipanjan and
335
- Ganchev, Kuzman and
336
- Hall, Keith and
337
- Petrov, Slav and
338
- Zhang, Hao and
339
- T{\"a}ckstr{\"o}m, Oscar and
340
- Bedini, Claudia and
341
- Bertomeu Castell{\'o}, N{\'u}ria and
342
- Lee, Jungmee},
343
- booktitle = "Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)",
344
- month = aug,
345
- year = "2013",
346
- address = "Sofia, Bulgaria",
347
- publisher = "Association for Computational Linguistics",
348
- url = "https://aclanthology.org/P13-2017",
349
- pages = "92--97",
350
- }
351
- ```
352
-
353
- LIA TAGG:
354
-
355
- ```latex
356
- @techreport{LIA_TAGG,
357
- author = {Frédéric Béchet},
358
- title = {LIA_TAGG: a statistical POS tagger + syntactic bracketer},
359
- institution = {Aix-Marseille University & CNRS},
360
- year = {2001}
361
- }
362
- ```
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ annotations_creators:
3
+ - machine-generated
4
+ - expert-generated
5
+ language_creators:
6
+ - found
7
+ languages:
8
+ - fr-FR
9
+ pretty_name: ANTILLES
10
+ size_categories:
11
+ - 100K<n<1M
12
+ source_datasets:
13
+ - original
14
+ task_categories:
15
+ - token-classification
16
+ task_ids:
17
+ - token-classification
18
+ ---
19
+
20
+ # ANTILLES : An Open French Linguistically Enriched Part-of-Speech Corpus
21
+
22
+ ## Table of Contents
23
+ - [Dataset Card for [Needs More Information]](#dataset-card-for-needs-more-information)
24
+ - [Table of Contents](#table-of-contents)
25
+ - [Dataset Description](#dataset-description)
26
+ - [Dataset Summary](#dataset-summary)
27
+ - [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
28
+ - [Languages](#languages)
29
+ - [Dataset Structure](#dataset-structure)
30
+ - [Data Instances](#data-instances)
31
+ - [sent_id = fr-ud-dev_00005](#sent_id--fr-ud-dev_00005)
32
+ - [text = Travail de trés grande qualité exécuté par un imprimeur artisan passionné.](#text--travail-de-trs-grande-qualit-excut-par-un-imprimeur-artisan-passionn)
33
+ - [Data Fields](#data-fields)
34
+ - [Data Splits](#data-splits)
35
+ - [Dataset Creation](#dataset-creation)
36
+ - [Curation Rationale](#curation-rationale)
37
+ - [Source Data](#source-data)
38
+ - [Initial Data Collection and Normalization](#initial-data-collection-and-normalization)
39
+ - [Who are the source language producers?](#who-are-the-source-language-producers)
40
+ - [Annotations](#annotations)
41
+ - [Annotation process](#annotation-process)
42
+ - [Who are the annotators?](#who-are-the-annotators)
43
+ - [Personal and Sensitive Information](#personal-and-sensitive-information)
44
+ - [Considerations for Using the Data](#considerations-for-using-the-data)
45
+ - [Social Impact of Dataset](#social-impact-of-dataset)
46
+ - [Discussion of Biases](#discussion-of-biases)
47
+ - [Other Known Limitations](#other-known-limitations)
48
+ - [Additional Information](#additional-information)
49
+ - [Dataset Curators](#dataset-curators)
50
+ - [Licensing Information](#licensing-information)
51
+ - [Citation Information](#citation-information)
52
+
53
+ ## Dataset Description
54
+
55
+ - **Homepage:** https://qanastek.github.io/ANTILLES/
56
+ - **Repository:** https://github.com/qanastek/ANTILLES
57
+ - **Paper:** [Needs More Information]
58
+ - **Leaderboard:** [Needs More Information]
59
+ - **Point of Contact:** [email protected]
60
+
61
+ ### Dataset Summary
62
+
63
+ `ANTILLES` is a part-of-speech tagging corpora based on [UD_French-GSD](https://universaldependencies.org/treebanks/fr_gsd/index.html) which was originally created in 2015 and is based on the [universal dependency treebank v2.0](https://github.com/ryanmcd/uni-dep-tb).
64
+
65
+ Originally, the corpora consists of 400,399 words (16,341 sentences) and had 17 different classes. Now, after applying our tags augmentation script `transform.py`, we obtain 60 different classes which add semantic information such as: the gender, number, mood, person, tense or verb form given in the different CoNLL-U fields from the original corpora.
66
+
67
+ We based our tags on the level of details given by the [LIA_TAGG](http://pageperso.lif.univ-mrs.fr/frederic.bechet/download.html) statistical POS tagger written by [Frédéric Béchet](http://pageperso.lif.univ-mrs.fr/frederic.bechet/index-english.html) in 2001.
68
+
69
+ <a rel="license" href="http://creativecommons.org/licenses/by-sa/4.0/"><img alt="Creative Commons License" style="border-width:0" src="https://i.creativecommons.org/l/by-sa/4.0/88x31.png" /></a><br />This work is licensed under a <a rel="license" href="http://creativecommons.org/licenses/by-sa/4.0/">Creative Commons Attribution-ShareAlike 4.0 International License</a>.
70
+
71
+ ### Supported Tasks and Leaderboards
72
+
73
+ `part-of-speech-tagging`: The dataset can be used to train a model for part-of-speech-tagging. The performance is measured by how high its F1 score is. A Flair Sequence-To-Sequence model trained to tag tokens from Wikipedia passages achieves a F1 score (micro) of 0.952.
74
+
75
+ ### Languages
76
+
77
+ The text in the dataset is in French, as spoken by [Wikipedia](https://en.wikipedia.org/wiki/Main_Page) users. The associated [BCP-47](https://tools.ietf.org/html/bcp47) code is `fr`.
78
+
79
+ ## Load the dataset
80
+
81
+ ### HuggingFace
82
+
83
+ ```python
84
+ from datasets import load_dataset
85
+ dataset = load_dataset("qanastek/ANTILLES")
86
+ print(dataset)
87
+ ```
88
+
89
+ ### FlairNLP
90
+
91
+ ```python
92
+ from flair.datasets import UniversalDependenciesCorpus
93
+ corpus: Corpus = UniversalDependenciesCorpus(
94
+ data_folder='ANTILLES',
95
+ train_file="train.conllu",
96
+ test_file="test.conllu",
97
+ dev_file="dev.conllu"
98
+ )
99
+ ```
100
+
101
+ ## Load the model
102
+
103
+ ### Flair ([model](https://huggingface.co/qanastek/pos-french))
104
+
105
+ ```python
106
+ from flair.models import SequenceTagger
107
+ tagger = SequenceTagger.load("qanastek/pos-french")
108
+ ```
109
+
110
+ ## HuggingFace Spaces
111
+
112
+ <table style="width: fit-content;">
113
+ <thead>
114
+ <tr>
115
+ <td>
116
+ <a href="https://huggingface.co/spaces/qanastek/French-Part-Of-Speech-Tagging">
117
+ <img src="https://huggingface.co/datasets/qanastek/ANTILLES/raw/main/imgs/en.png" width="160">
118
+ </a>
119
+ </td>
120
+ <td>
121
+ <a href="https://huggingface.co/spaces/qanastek/Etiqueteur-Morphosyntaxique-Etendu">
122
+ <img src="https://huggingface.co/datasets/qanastek/ANTILLES/raw/main/imgs/fr.png" width="160">
123
+ </a>
124
+ </td>
125
+ </tr>
126
+ </thead>
127
+ </table>
128
+
129
+ ## Dataset Structure
130
+
131
+ ### Data Instances
132
+
133
+ ```plain
134
+ # sent_id = fr-ud-dev_00005
135
+ # text = Travail de trés grande qualité exécuté par un imprimeur artisan passionné.
136
+ 1 Travail travail NMS _ Gender=Masc|Number=Sing 0 root _ wordform=travail
137
+ 2 de de PREP _ _ 5 case _ _
138
+ 3 trés trés ADV _ _ 4 advmod _ _
139
+ 4 grande grand ADJFS _ Gender=Fem|Number=Sing 5 amod _ _
140
+ 5 qualité qualité NFS _ Gender=Fem|Number=Sing 1 nmod _ _
141
+ 6 exécuté exécuter VPPMS _ Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part 1 acl _ _
142
+ 7 par par PREP _ _ 9 case _ _
143
+ 8 un un DINTMS _ Definite=Ind|Gender=Masc|Number=Sing|PronType=Art 9 det _ _
144
+ 9 imprimeur imprimeur NMS _ Gender=Masc|Number=Sing 6 obl:agent _ _
145
+ 10 artisan artisan NMS _ Gender=Masc|Number=Sing 9 nmod _ _
146
+ 11 passionné passionné ADJMS _ Gender=Masc|Number=Sing 9 amod _ SpaceAfter=No
147
+ 12 . . YPFOR _ _ 1 punct _ _
148
+ ```
149
+
150
+ ### Data Fields
151
+
152
+ | Abbreviation | Description | Examples | # tokens |
153
+ |:--------:|:--------:|:--------:|:--------:|
154
+ | PREP | Preposition | de | 63 738 |
155
+ | AUX | Auxiliary Verb | est | 12 886 |
156
+ | ADV | Adverb | toujours | 14 969 |
157
+ | COSUB | Subordinating conjunction | que | 3 007 |
158
+ | COCO | Coordinating Conjunction | et | 10 102 |
159
+ | PART | Demonstrative particle | -t | 93 |
160
+ | PRON | Pronoun | qui ce quoi | 667 |
161
+ | PDEMMS | Singular Masculine Demonstrative Pronoun | ce | 1 950 |
162
+ | PDEMMP | Plurial Masculine Demonstrative Pronoun | ceux | 108 |
163
+ | PDEMFS | Singular Feminine Demonstrative Pronoun | cette | 1 004 |
164
+ | PDEMFP | Plurial Feminine Demonstrative Pronoun | celles | 53 |
165
+ | PINDMS | Singular Masculine Indefinite Pronoun | tout | 961 |
166
+ | PINDMP | Plurial Masculine Indefinite Pronoun | autres | 89 |
167
+ | PINDFS | Singular Feminine Indefinite Pronoun | chacune | 136 |
168
+ | PINDFP | Plurial Feminine Indefinite Pronoun | certaines | 31 |
169
+ | PROPN | Proper noun | houston | 22 135 |
170
+ | XFAMIL | Last name | levy | 6 449 |
171
+ | NUM | Numerical Adjectives | trentaine vingtaine | 67 |
172
+ | DINTMS | Masculine Numerical Adjectives | un | 4 254 |
173
+ | DINTFS | Feminine Numerical Adjectives | une | 3 543 |
174
+ | PPOBJMS | Singular Masculine Pronoun complements of objects | le lui | 1 425 |
175
+ | PPOBJMP | Plurial Masculine Pronoun complements of objects | eux y | 212 |
176
+ | PPOBJFS | Singular Feminine Pronoun complements of objects | moi la | 358 |
177
+ | PPOBJFP | Plurial Feminine Pronoun complements of objects | en y | 70 |
178
+ | PPER1S | Personal Pronoun First Person Singular | je | 571 |
179
+ | PPER2S | Personal Pronoun Second Person Singular | tu | 19 |
180
+ | PPER3MS | Personal Pronoun Third Person Masculine Singular | il | 3 938 |
181
+ | PPER3MP | Personal Pronoun Third Person Masculine Plurial | ils | 513 |
182
+ | PPER3FS | Personal Pronoun Third Person Feminine Singular | elle | 992 |
183
+ | PPER3FP | Personal Pronoun Third Person Feminine Plurial | elles | 121 |
184
+ | PREFS | Reflexive Pronouns First Person of Singular | me m' | 120 |
185
+ | PREF | Reflexive Pronouns Third Person of Singular | se s' | 2 337 |
186
+ | PREFP | Reflexive Pronouns First / Second Person of Plurial | nous vous | 686 |
187
+ | VERB | Verb | obtient | 21 131 |
188
+ | VPPMS | Singular Masculine Participle Past Verb | formulé | 6 275 |
189
+ | VPPMP | Plurial Masculine Participle Past Verb | classés | 1 352 |
190
+ | VPPFS | Singular Feminine Participle Past Verb | appelée | 2 434 |
191
+ | VPPFP | Plurial Feminine Participle Past Verb | sanctionnées | 813 |
192
+ | VPPRE | Present participle | étant | 2 |
193
+ | DET | Determinant | les l' | 25 206 |
194
+ | DETMS | Singular Masculine Determinant | les | 15 444 |
195
+ | DETFS | Singular Feminine Determinant | la | 10 978 |
196
+ | ADJ | Adjective | capable sérieux | 1 075 |
197
+ | ADJMS | Singular Masculine Adjective | grand important | 8 338 |
198
+ | ADJMP | Plurial Masculine Adjective | grands petits | 3 274 |
199
+ | ADJFS | Singular Feminine Adjective | franéaise petite | 8 004 |
200
+ | ADJFP | Plurial Feminine Adjective | légéres petites | 3 041 |
201
+ | NOUN | Noun | temps | 1 389 |
202
+ | NMS | Singular Masculine Noun | drapeau | 29 698 |
203
+ | NMP | Plurial Masculine Noun | journalistes | 10 882 |
204
+ | NFS | Singular Feminine Noun | téte | 25 414 |
205
+ | NFP | Plurial Feminine Noun | ondes | 7 448 |
206
+ | PREL | Relative Pronoun | qui dont | 2 976 |
207
+ | PRELMS | Singular Masculine Relative Pronoun | lequel | 94 |
208
+ | PRELMP | Plurial Masculine Relative Pronoun | lesquels | 29 |
209
+ | PRELFS | Singular Feminine Relative Pronoun | laquelle | 70 |
210
+ | PRELFP | Plurial Feminine Relative Pronoun | lesquelles | 25 |
211
+ | PINTFS | Singular Feminine Interrogative Pronoun | laquelle | 3 |
212
+ | INTJ | Interjection | merci bref | 75 |
213
+ | CHIF | Numbers | 1979 10 | 10 417 |
214
+ | SYM | Symbol | é % | 705 |
215
+ | YPFOR | Endpoint | . | 15 088 |
216
+ | PUNCT | Ponctuation | : , | 28 918 |
217
+ | MOTINC | Unknown words | Technology Lady | 2 022 |
218
+ | X | Typos & others | sfeir 3D statu | 175 |
219
+
220
+ ### Data Splits
221
+
222
+ | | Train | Dev | Test |
223
+ |:------------------:|:------:|:------:|:-----:|
224
+ | # Docs | 14 449 | 1 476 | 416 |
225
+ | Avg # Tokens / Doc | 24.54 | 24.19 | 24.08 |
226
+
227
+ ## Dataset Creation
228
+
229
+ ### Curation Rationale
230
+
231
+ [Needs More Information]
232
+
233
+ ### Source Data
234
+
235
+ #### Initial Data Collection and Normalization
236
+
237
+ [Needs More Information]
238
+
239
+ #### Who are the source language producers?
240
+
241
+ [Needs More Information]
242
+
243
+ ### Annotations
244
+
245
+ #### Annotation process
246
+
247
+ [Needs More Information]
248
+
249
+ #### Who are the annotators?
250
+
251
+ [Needs More Information]
252
+
253
+ ### Personal and Sensitive Information
254
+
255
+ The corpora is free of personal or sensitive information since it has been based on `Wikipedia` articles content.
256
+
257
+ ## Considerations for Using the Data
258
+
259
+ ### Social Impact of Dataset
260
+
261
+ [Needs More Information]
262
+
263
+ ### Discussion of Biases
264
+
265
+ The nature of the corpora introduce various biases such as the names of the streets which are temporaly based and can therefore introduce named entity like author or event names. For example, street names such as `Rue Victor-Hugo` or `Rue Pasteur` doesn't exist before the 20's century in France.
266
+
267
+ ### Other Known Limitations
268
+
269
+ [Needs More Information]
270
+
271
+ ## Additional Information
272
+
273
+ ### Dataset Curators
274
+
275
+ __ANTILLES__: Labrak Yanis, Dufour Richard
276
+
277
+ __UD_FRENCH-GSD__: de Marneffe Marie-Catherine, Guillaume Bruno, McDonald Ryan, Suhr Alane, Nivre Joakim, Grioni Matias, Dickerson Carly, Perrier Guy
278
+
279
+ __Universal Dependency__: Ryan McDonald, Joakim Nivre, Yvonne Quirmbach-Brundage, Yoav Goldberg, Dipanjan Das, Kuzman Ganchev, Keith Hall, Slav Petrov, Hao Zhang, Oscar Tackstrom, Claudia Bedini, Nuria Bertomeu Castello and Jungmee Lee
280
+
281
+ ### Licensing Information
282
+
283
+ ```plain
284
+ For the following languages
285
+
286
+ German, Spanish, French, Indonesian, Italian, Japanese, Korean and Brazilian
287
+ Portuguese
288
+
289
+ we will distinguish between two portions of the data.
290
+
291
+ 1. The underlying text for sentences that were annotated. This data Google
292
+ asserts no ownership over and no copyright over. Some or all of these
293
+ sentences may be copyrighted in some jurisdictions. Where copyrighted,
294
+ Google collected these sentences under exceptions to copyright or implied
295
+ license rights. GOOGLE MAKES THEM AVAILABLE TO YOU 'AS IS', WITHOUT ANY
296
+ WARRANTY OF ANY KIND, WHETHER EXPRESS OR IMPLIED.
297
+
298
+ 2. The annotations -- part-of-speech tags and dependency annotations. These are
299
+ made available under a CC BY-SA 4.0. GOOGLE MAKES
300
+ THEM AVAILABLE TO YOU 'AS IS', WITHOUT ANY WARRANTY OF ANY KIND, WHETHER
301
+ EXPRESS OR IMPLIED. See attached LICENSE file for the text of CC BY-NC-SA.
302
+
303
+ Portions of the German data were sampled from the CoNLL 2006 Tiger Treebank
304
+ data. Hans Uszkoreit graciously gave permission to use the underlying
305
+ sentences in this data as part of this release.
306
+
307
+ Any use of the data should reference the above plus:
308
+
309
+ Universal Dependency Annotation for Multilingual Parsing
310
+ Ryan McDonald, Joakim Nivre, Yvonne Quirmbach-Brundage, Yoav Goldberg,
311
+ Dipanjan Das, Kuzman Ganchev, Keith Hall, Slav Petrov, Hao Zhang,
312
+ Oscar Tackstrom, Claudia Bedini, Nuria Bertomeu Castello and Jungmee Lee
313
+ Proceedings of ACL 2013
314
+ ```
315
+
316
+ ### Citation Information
317
+
318
+ Please cite the following paper when using this model.
319
+
320
+ ANTILLES extended corpus:
321
+
322
+ ```latex
323
+ @software{Labrak_ANTILLES_An_2021,
324
+ author = {Labrak, Yanis and Dufour, Richard},
325
+ month = {11},
326
+ title = {{ANTILLES : An Open French Linguistically Enriched Part-of-Speech Corpus}},
327
+ url = {https://github.com/qanastek/ANTILLES},
328
+ version = {1.0.0},
329
+ year = {2021}
330
+ }
331
+ ```
332
+
333
+ UD_French-GSD corpora:
334
+
335
+ ```latex
336
+ @misc{
337
+ universaldependencies,
338
+ title={UniversalDependencies/UD_French-GSD},
339
+ url={https://github.com/UniversalDependencies/UD_French-GSD}, journal={GitHub},
340
+ author={UniversalDependencies}
341
+ }
342
+ ```
343
+
344
+ {U}niversal {D}ependency Annotation for Multilingual Parsing:
345
+
346
+ ```latex
347
+ @inproceedings{mcdonald-etal-2013-universal,
348
+ title = "{U}niversal {D}ependency Annotation for Multilingual Parsing",
349
+ author = {McDonald, Ryan and
350
+ Nivre, Joakim and
351
+ Quirmbach-Brundage, Yvonne and
352
+ Goldberg, Yoav and
353
+ Das, Dipanjan and
354
+ Ganchev, Kuzman and
355
+ Hall, Keith and
356
+ Petrov, Slav and
357
+ Zhang, Hao and
358
+ T{\"a}ckstr{\"o}m, Oscar and
359
+ Bedini, Claudia and
360
+ Bertomeu Castell{\'o}, N{\'u}ria and
361
+ Lee, Jungmee},
362
+ booktitle = "Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)",
363
+ month = aug,
364
+ year = "2013",
365
+ address = "Sofia, Bulgaria",
366
+ publisher = "Association for Computational Linguistics",
367
+ url = "https://aclanthology.org/P13-2017",
368
+ pages = "92--97",
369
+ }
370
+ ```
371
+
372
+ LIA TAGG:
373
+
374
+ ```latex
375
+ @techreport{LIA_TAGG,
376
+ author = {Frédéric Béchet},
377
+ title = {LIA_TAGG: a statistical POS tagger + syntactic bracketer},
378
+ institution = {Aix-Marseille University & CNRS},
379
+ year = {2001}
380
+ }
381
+ ```