system HF staff commited on
Commit
5d5d5f0
0 Parent(s):

Update files from the datasets library (from 1.0.0)

Browse files

Release notes: https://github.com/huggingface/datasets/releases/tag/1.0.0

.gitattributes ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bin.* filter=lfs diff=lfs merge=lfs -text
5
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.model filter=lfs diff=lfs merge=lfs -text
12
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
13
+ *.onnx filter=lfs diff=lfs merge=lfs -text
14
+ *.ot filter=lfs diff=lfs merge=lfs -text
15
+ *.parquet filter=lfs diff=lfs merge=lfs -text
16
+ *.pb filter=lfs diff=lfs merge=lfs -text
17
+ *.pt filter=lfs diff=lfs merge=lfs -text
18
+ *.pth filter=lfs diff=lfs merge=lfs -text
19
+ *.rar filter=lfs diff=lfs merge=lfs -text
20
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
21
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
22
+ *.tflite filter=lfs diff=lfs merge=lfs -text
23
+ *.tgz filter=lfs diff=lfs merge=lfs -text
24
+ *.xz filter=lfs diff=lfs merge=lfs -text
25
+ *.zip filter=lfs diff=lfs merge=lfs -text
26
+ *.zstandard filter=lfs diff=lfs merge=lfs -text
27
+ *tfevents* filter=lfs diff=lfs merge=lfs -text
dataset_infos.json ADDED
@@ -0,0 +1 @@
 
1
+ {"lid_spaeng": {"description": "LinCE is a centralized Linguistic Code-switching Evaluation benchmark \n(https://ritual.uh.edu/lince/) that contains data for training and evaluating\nNLP systems on code-switching tasks.\n", "citation": "\n@inproceedings{molina-etal-2016-overview,\n title = \"Overview for the Second Shared Task on Language Identification in Code-Switched Data\",\n author = \"Molina, Giovanni and\n AlGhamdi, Fahad and\n Ghoneim, Mahmoud and\n Hawwari, Abdelati and\n Rey-Villamizar, Nicolas and\n Diab, Mona and\n Solorio, Thamar\",\n booktitle = \"Proceedings of the Second Workshop on Computational Approaches to Code Switching\",\n month = nov,\n year = \"2016\",\n address = \"Austin, Texas\",\n publisher = \"Association for Computational Linguistics\",\n url = \"https://www.aclweb.org/anthology/W16-5805\",\n doi = \"10.18653/v1/W16-5805\",\n pages = \"40--49\",\n}\n\n@inproceedings{aguilar-etal-2020-lince,\n title = \"{L}in{CE}: A Centralized Benchmark for Linguistic Code-switching Evaluation\",\n author = \"Aguilar, Gustavo and\n Kar, Sudipta and\n Solorio, Thamar\",\n booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n month = may,\n year = \"2020\",\n address = \"Marseille, France\",\n publisher = \"European Language Resources Association\",\n url = \"https://www.aclweb.org/anthology/2020.lrec-1.223\",\n pages = \"1803--1813\",\n language = \"English\",\n ISBN = \"979-10-95546-34-4\",\n}\n\nNote that each LinCE dataset has its own citation. Please see the source to see\nthe correct citation for each contained dataset.", "homepage": "http://ritual.uh.edu/lince", "license": "", "features": {"idx": {"dtype": "int32", "id": null, "_type": "Value"}, "tokens": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "lid": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": {"features": null, "resources_checksums": {"train": {}, "validation": {}, "test": {}}}, "supervised_keys": null, "builder_name": "lince", "config_name": "lid_spaeng", "version": {"version_str": "1.0.0", "description": "The Linguistic Code-switching Evaluation (LinCE) benchmark", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 4745003, "num_examples": 21030, "dataset_name": "lince"}, "validation": {"name": "validation", "num_bytes": 739950, "num_examples": 3332, "dataset_name": "lince"}, "test": {"name": "test", "num_bytes": 1337727, "num_examples": 8289, "dataset_name": "lince"}}, "download_checksums": {"https://ritual.uh.edu/lince/libaccess/eyJ1c2VybmFtZSI6ICJodWdnaW5nZmFjZSBubHAiLCAidXNlcl9pZCI6IDExMSwgImVtYWlsIjogImR1bW15QGVtYWlsLmNvbSJ9/lid_spaeng.zip": {"num_bytes": 1188861, "checksum": "f85c4166f858bb28c6bdf0b5062f9fe4645f5d67d24861fb04258424cccc19e4"}}, "download_size": 1188861, "post_processing_size": 0, "dataset_size": 6822680, "size_in_bytes": 8011541}, "lid_hineng": {"description": "LinCE is a centralized Linguistic Code-switching Evaluation benchmark \n(https://ritual.uh.edu/lince/) that contains data for training and evaluating\nNLP systems on code-switching tasks.\n", "citation": "\n@inproceedings{mave-etal-2018-language,\n title = \"Language Identification and Analysis of Code-Switched Social Media Text\",\n author = \"Mave, Deepthi and\n Maharjan, Suraj and\n Solorio, Thamar\",\n booktitle = \"Proceedings of the Third Workshop on Computational Approaches to Linguistic Code-Switching\",\n month = jul,\n year = \"2018\",\n address = \"Melbourne, Australia\",\n publisher = \"Association for Computational Linguistics\",\n url = \"https://www.aclweb.org/anthology/W18-3206\",\n pages = \"51--61\"\n}\n\n@inproceedings{aguilar-etal-2020-lince,\n title = \"{L}in{CE}: A Centralized Benchmark for Linguistic Code-switching Evaluation\",\n author = \"Aguilar, Gustavo and\n Kar, Sudipta and\n Solorio, Thamar\",\n booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n month = may,\n year = \"2020\",\n address = \"Marseille, France\",\n publisher = \"European Language Resources Association\",\n url = \"https://www.aclweb.org/anthology/2020.lrec-1.223\",\n pages = \"1803--1813\",\n language = \"English\",\n ISBN = \"979-10-95546-34-4\",\n}\n\nNote that each LinCE dataset has its own citation. Please see the source to see\nthe correct citation for each contained dataset.", "homepage": "http://ritual.uh.edu/lince", "license": "", "features": {"idx": {"dtype": "int32", "id": null, "_type": "Value"}, "tokens": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "lid": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": {"features": null, "resources_checksums": {"train": {}, "validation": {}, "test": {}}}, "supervised_keys": null, "builder_name": "lince", "config_name": "lid_hineng", "version": {"version_str": "1.0.0", "description": "The Linguistic Code-switching Evaluation (LinCE) benchmark", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 1662284, "num_examples": 4823, "dataset_name": "lince"}, "validation": {"name": "validation", "num_bytes": 268930, "num_examples": 744, "dataset_name": "lince"}, "test": {"name": "test", "num_bytes": 456850, "num_examples": 1854, "dataset_name": "lince"}}, "download_checksums": {"https://ritual.uh.edu/lince/libaccess/eyJ1c2VybmFtZSI6ICJodWdnaW5nZmFjZSBubHAiLCAidXNlcl9pZCI6IDExMSwgImVtYWlsIjogImR1bW15QGVtYWlsLmNvbSJ9/lid_hineng.zip": {"num_bytes": 432854, "checksum": "b36a28c1b05f8575952f8a245edce20ab8e21543c6e11feb80220681a4c81068"}}, "download_size": 432854, "post_processing_size": 0, "dataset_size": 2388064, "size_in_bytes": 2820918}, "lid_msaea": {"description": "LinCE is a centralized Linguistic Code-switching Evaluation benchmark \n(https://ritual.uh.edu/lince/) that contains data for training and evaluating\nNLP systems on code-switching tasks.\n", "citation": "\n@inproceedings{molina-etal-2016-overview,\n title = \"Overview for the Second Shared Task on Language Identification in Code-Switched Data\",\n author = \"Molina, Giovanni and\n AlGhamdi, Fahad and\n Ghoneim, Mahmoud and\n Hawwari, Abdelati and\n Rey-Villamizar, Nicolas and\n Diab, Mona and\n Solorio, Thamar\",\n booktitle = \"Proceedings of the Second Workshop on Computational Approaches to Code Switching\",\n month = nov,\n year = \"2016\",\n address = \"Austin, Texas\",\n publisher = \"Association for Computational Linguistics\",\n url = \"https://www.aclweb.org/anthology/W16-5805\",\n doi = \"10.18653/v1/W16-5805\",\n pages = \"40--49\",\n}\n\n@inproceedings{aguilar-etal-2020-lince,\n title = \"{L}in{CE}: A Centralized Benchmark for Linguistic Code-switching Evaluation\",\n author = \"Aguilar, Gustavo and\n Kar, Sudipta and\n Solorio, Thamar\",\n booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n month = may,\n year = \"2020\",\n address = \"Marseille, France\",\n publisher = \"European Language Resources Association\",\n url = \"https://www.aclweb.org/anthology/2020.lrec-1.223\",\n pages = \"1803--1813\",\n language = \"English\",\n ISBN = \"979-10-95546-34-4\",\n}\n\nNote that each LinCE dataset has its own citation. Please see the source to see\nthe correct citation for each contained dataset.", "homepage": "http://ritual.uh.edu/lince", "license": "", "features": {"idx": {"dtype": "int32", "id": null, "_type": "Value"}, "tokens": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "lid": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": {"features": null, "resources_checksums": {"train": {}, "validation": {}, "test": {}}}, "supervised_keys": null, "builder_name": "lince", "config_name": "lid_msaea", "version": {"version_str": "1.0.0", "description": "The Linguistic Code-switching Evaluation (LinCE) benchmark", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 3804156, "num_examples": 8464, "dataset_name": "lince"}, "validation": {"name": "validation", "num_bytes": 490566, "num_examples": 1116, "dataset_name": "lince"}, "test": {"name": "test", "num_bytes": 590488, "num_examples": 1663, "dataset_name": "lince"}}, "download_checksums": {"https://ritual.uh.edu/lince/libaccess/eyJ1c2VybmFtZSI6ICJodWdnaW5nZmFjZSBubHAiLCAidXNlcl9pZCI6IDExMSwgImVtYWlsIjogImR1bW15QGVtYWlsLmNvbSJ9/lid_msaea.zip": {"num_bytes": 803806, "checksum": "d082217628e702c53a63a2b3a6a7b764f2f9f2ed79d3598a48e3d327081e2913"}}, "download_size": 803806, "post_processing_size": 0, "dataset_size": 4885210, "size_in_bytes": 5689016}, "lid_nepeng": {"description": "LinCE is a centralized Linguistic Code-switching Evaluation benchmark \n(https://ritual.uh.edu/lince/) that contains data for training and evaluating\nNLP systems on code-switching tasks.\n", "citation": "\n@inproceedings{solorio-etal-2014-overview,\n title = \"Overview for the First Shared Task on Language Identification in Code-Switched Data\",\n author = \"Solorio, Thamar and\n Blair, Elizabeth and\n Maharjan, Suraj and\n Bethard, Steven and\n Diab, Mona and\n Ghoneim, Mahmoud and\n Hawwari, Abdelati and\n AlGhamdi, Fahad and\n Hirschberg, Julia and\n Chang, Alison and\n Fung, Pascale\",\n booktitle = \"Proceedings of the First Workshop on Computational Approaches to Code Switching\",\n month = oct,\n year = \"2014\",\n address = \"Doha, Qatar\",\n publisher = \"Association for Computational Linguistics\",\n url = \"https://www.aclweb.org/anthology/W14-3907\",\n doi = \"10.3115/v1/W14-3907\",\n pages = \"62--72\",\n}\n\n@inproceedings{aguilar-etal-2020-lince,\n title = \"{L}in{CE}: A Centralized Benchmark for Linguistic Code-switching Evaluation\",\n author = \"Aguilar, Gustavo and\n Kar, Sudipta and\n Solorio, Thamar\",\n booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n month = may,\n year = \"2020\",\n address = \"Marseille, France\",\n publisher = \"European Language Resources Association\",\n url = \"https://www.aclweb.org/anthology/2020.lrec-1.223\",\n pages = \"1803--1813\",\n language = \"English\",\n ISBN = \"979-10-95546-34-4\",\n}\n\nNote that each LinCE dataset has its own citation. Please see the source to see\nthe correct citation for each contained dataset.", "homepage": "http://ritual.uh.edu/lince", "license": "", "features": {"idx": {"dtype": "int32", "id": null, "_type": "Value"}, "tokens": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "lid": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": {"features": null, "resources_checksums": {"train": {}, "validation": {}, "test": {}}}, "supervised_keys": null, "builder_name": "lince", "config_name": "lid_nepeng", "version": {"version_str": "1.0.0", "description": "The Linguistic Code-switching Evaluation (LinCE) benchmark", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 2239014, "num_examples": 8451, "dataset_name": "lince"}, "validation": {"name": "validation", "num_bytes": 351649, "num_examples": 1332, "dataset_name": "lince"}, "test": {"name": "test", "num_bytes": 620512, "num_examples": 3228, "dataset_name": "lince"}}, "download_checksums": {"https://ritual.uh.edu/lince/libaccess/eyJ1c2VybmFtZSI6ICJodWdnaW5nZmFjZSBubHAiLCAidXNlcl9pZCI6IDExMSwgImVtYWlsIjogImR1bW15QGVtYWlsLmNvbSJ9/lid_nepeng.zip": {"num_bytes": 545342, "checksum": "a936df9a47beced14703e2b56e37b66cd596b6525b8ab7768ed608fc5639ea2e"}}, "download_size": 545342, "post_processing_size": 0, "dataset_size": 3211175, "size_in_bytes": 3756517}, "pos_spaeng": {"description": "LinCE is a centralized Linguistic Code-switching Evaluation benchmark \n(https://ritual.uh.edu/lince/) that contains data for training and evaluating\nNLP systems on code-switching tasks.\n", "citation": "\n@inproceedings{alghamdi-etal-2016-part,\n title = \"Part of Speech Tagging for Code Switched Data\",\n author = \"AlGhamdi, Fahad and\n Molina, Giovanni and\n Diab, Mona and\n Solorio, Thamar and\n Hawwari, Abdelati and\n Soto, Victor and\n Hirschberg, Julia\",\n booktitle = \"Proceedings of the Second Workshop on Computational Approaches to Code Switching\",\n month = nov,\n year = \"2016\",\n address = \"Austin, Texas\",\n publisher = \"Association for Computational Linguistics\",\n url = \"https://www.aclweb.org/anthology/W16-5812\",\n doi = \"10.18653/v1/W16-5812\",\n pages = \"98--107\",\n}\n\n@inproceedings{aguilar-etal-2020-lince,\n title = \"{L}in{CE}: A Centralized Benchmark for Linguistic Code-switching Evaluation\",\n author = \"Aguilar, Gustavo and\n Kar, Sudipta and\n Solorio, Thamar\",\n booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n month = may,\n year = \"2020\",\n address = \"Marseille, France\",\n publisher = \"European Language Resources Association\",\n url = \"https://www.aclweb.org/anthology/2020.lrec-1.223\",\n pages = \"1803--1813\",\n language = \"English\",\n ISBN = \"979-10-95546-34-4\",\n}\n\nNote that each LinCE dataset has its own citation. Please see the source to see\nthe correct citation for each contained dataset.", "homepage": "http://ritual.uh.edu/lince", "license": "", "features": {"idx": {"dtype": "int32", "id": null, "_type": "Value"}, "tokens": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "lid": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "pos": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": {"features": null, "resources_checksums": {"train": {}, "validation": {}, "test": {}}}, "supervised_keys": null, "builder_name": "lince", "config_name": "pos_spaeng", "version": {"version_str": "1.0.0", "description": "The Linguistic Code-switching Evaluation (LinCE) benchmark", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 5467832, "num_examples": 27893, "dataset_name": "lince"}, "validation": {"name": "validation", "num_bytes": 840593, "num_examples": 4298, "dataset_name": "lince"}, "test": {"name": "test", "num_bytes": 1758626, "num_examples": 10720, "dataset_name": "lince"}}, "download_checksums": {"https://ritual.uh.edu/lince/libaccess/eyJ1c2VybmFtZSI6ICJodWdnaW5nZmFjZSBubHAiLCAidXNlcl9pZCI6IDExMSwgImVtYWlsIjogImR1bW15QGVtYWlsLmNvbSJ9/pos_spaeng.zip": {"num_bytes": 819657, "checksum": "a829f883c47eb2a5f84936ed66912255b850b56b21011e2f8cb069b55f2f3995"}}, "download_size": 819657, "post_processing_size": 0, "dataset_size": 8067051, "size_in_bytes": 8886708}, "pos_hineng": {"description": "LinCE is a centralized Linguistic Code-switching Evaluation benchmark \n(https://ritual.uh.edu/lince/) that contains data for training and evaluating\nNLP systems on code-switching tasks.\n", "citation": "\n@inproceedings{singh-etal-2018-twitter,\n title = \"A Twitter Corpus for {H}indi-{E}nglish Code Mixed {POS} Tagging\",\n author = \"Singh, Kushagra and\n Sen, Indira and\n Kumaraguru, Ponnurangam\",\n booktitle = \"Proceedings of the Sixth International Workshop on Natural Language Processing for Social Media\",\n month = jul,\n year = \"2018\",\n address = \"Melbourne, Australia\",\n publisher = \"Association for Computational Linguistics\",\n url = \"https://www.aclweb.org/anthology/W18-3503\",\n doi = \"10.18653/v1/W18-3503\",\n pages = \"12--17\"\n}\n\n@inproceedings{aguilar-etal-2020-lince,\n title = \"{L}in{CE}: A Centralized Benchmark for Linguistic Code-switching Evaluation\",\n author = \"Aguilar, Gustavo and\n Kar, Sudipta and\n Solorio, Thamar\",\n booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n month = may,\n year = \"2020\",\n address = \"Marseille, France\",\n publisher = \"European Language Resources Association\",\n url = \"https://www.aclweb.org/anthology/2020.lrec-1.223\",\n pages = \"1803--1813\",\n language = \"English\",\n ISBN = \"979-10-95546-34-4\",\n}\n\nNote that each LinCE dataset has its own citation. Please see the source to see\nthe correct citation for each contained dataset.", "homepage": "http://ritual.uh.edu/lince", "license": "", "features": {"idx": {"dtype": "int32", "id": null, "_type": "Value"}, "tokens": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "lid": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "pos": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": {"features": null, "resources_checksums": {"train": {}, "validation": {}, "test": {}}}, "supervised_keys": null, "builder_name": "lince", "config_name": "pos_hineng", "version": {"version_str": "1.0.0", "description": "The Linguistic Code-switching Evaluation (LinCE) benchmark", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 537541, "num_examples": 1030, "dataset_name": "lince"}, "validation": {"name": "validation", "num_bytes": 80886, "num_examples": 160, "dataset_name": "lince"}, "test": {"name": "test", "num_bytes": 131192, "num_examples": 299, "dataset_name": "lince"}}, "download_checksums": {"https://ritual.uh.edu/lince/libaccess/eyJ1c2VybmFtZSI6ICJodWdnaW5nZmFjZSBubHAiLCAidXNlcl9pZCI6IDExMSwgImVtYWlsIjogImR1bW15QGVtYWlsLmNvbSJ9/pos_hineng.zip": {"num_bytes": 113872, "checksum": "ad8067145f19e4b350969fa3ee422d8f86a393170d6f448bdad2cd157075da53"}}, "download_size": 113872, "post_processing_size": 0, "dataset_size": 749619, "size_in_bytes": 863491}, "ner_spaeng": {"description": "LinCE is a centralized Linguistic Code-switching Evaluation benchmark \n(https://ritual.uh.edu/lince/) that contains data for training and evaluating\nNLP systems on code-switching tasks.\n", "citation": "\n@inproceedings{aguilar-etal-2018-named,\n title = {{Named Entity Recognition on Code-Switched Data: Overview of the CALCS 2018 Shared Task}},\n author = \"Aguilar, Gustavo and\n AlGhamdi, Fahad and\n Soto, Victor and\n Diab, Mona and\n Hirschberg, Julia and\n Solorio, Thamar\",\n booktitle = {{Proceedings of the Third Workshop on Computational Approaches to Linguistic Code-Switching}},\n month = jul,\n year = \"2018\",\n address = \"Melbourne, Australia\",\n publisher = \"Association for Computational Linguistics\",\n url = \"https://www.aclweb.org/anthology/W18-3219\",\n pages = \"138--147\"\n} \n\n@inproceedings{aguilar-etal-2020-lince,\n title = \"{L}in{CE}: A Centralized Benchmark for Linguistic Code-switching Evaluation\",\n author = \"Aguilar, Gustavo and\n Kar, Sudipta and\n Solorio, Thamar\",\n booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n month = may,\n year = \"2020\",\n address = \"Marseille, France\",\n publisher = \"European Language Resources Association\",\n url = \"https://www.aclweb.org/anthology/2020.lrec-1.223\",\n pages = \"1803--1813\",\n language = \"English\",\n ISBN = \"979-10-95546-34-4\",\n}\n\nNote that each LinCE dataset has its own citation. Please see the source to see\nthe correct citation for each contained dataset.", "homepage": "http://ritual.uh.edu/lince", "license": "", "features": {"idx": {"dtype": "int32", "id": null, "_type": "Value"}, "tokens": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "lid": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "ner": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": {"features": null, "resources_checksums": {"test": {}, "train": {}, "validation": {}}}, "supervised_keys": null, "builder_name": "lince", "config_name": "ner_spaeng", "version": {"version_str": "1.0.0", "description": "The Linguistic Code-switching Evaluation (LinCE) benchmark", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 9836312, "num_examples": 33611, "dataset_name": "lince"}, "validation": {"name": "validation", "num_bytes": 2980990, "num_examples": 10085, "dataset_name": "lince"}, "test": {"name": "test", "num_bytes": 6530956, "num_examples": 23527, "dataset_name": "lince"}}, "download_checksums": {"https://ritual.uh.edu/lince/libaccess/eyJ1c2VybmFtZSI6ICJodWdnaW5nZmFjZSBubHAiLCAidXNlcl9pZCI6IDExMSwgImVtYWlsIjogImR1bW15QGVtYWlsLmNvbSJ9/ner_spaeng.zip": {"num_bytes": 3075520, "checksum": "9ad57558575e156efc12cacf72f46f58ff4f1651808f55afe279495317fca461"}}, "download_size": 3075520, "post_processing_size": 0, "dataset_size": 19348258, "size_in_bytes": 22423778}, "ner_msaea": {"description": "LinCE is a centralized Linguistic Code-switching Evaluation benchmark \n(https://ritual.uh.edu/lince/) that contains data for training and evaluating\nNLP systems on code-switching tasks.\n", "citation": "\n@inproceedings{aguilar-etal-2018-named,\n title = {{Named Entity Recognition on Code-Switched Data: Overview of the CALCS 2018 Shared Task}},\n author = \"Aguilar, Gustavo and\n AlGhamdi, Fahad and\n Soto, Victor and\n Diab, Mona and\n Hirschberg, Julia and\n Solorio, Thamar\",\n booktitle = {{Proceedings of the Third Workshop on Computational Approaches to Linguistic Code-Switching}},\n month = jul,\n year = \"2018\",\n address = \"Melbourne, Australia\",\n publisher = \"Association for Computational Linguistics\",\n url = \"https://www.aclweb.org/anthology/W18-3219\",\n pages = \"138--147\"\n} \n\n@inproceedings{aguilar-etal-2020-lince,\n title = \"{L}in{CE}: A Centralized Benchmark for Linguistic Code-switching Evaluation\",\n author = \"Aguilar, Gustavo and\n Kar, Sudipta and\n Solorio, Thamar\",\n booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n month = may,\n year = \"2020\",\n address = \"Marseille, France\",\n publisher = \"European Language Resources Association\",\n url = \"https://www.aclweb.org/anthology/2020.lrec-1.223\",\n pages = \"1803--1813\",\n language = \"English\",\n ISBN = \"979-10-95546-34-4\",\n}\n\nNote that each LinCE dataset has its own citation. Please see the source to see\nthe correct citation for each contained dataset.", "homepage": "http://ritual.uh.edu/lince", "license": "", "features": {"idx": {"dtype": "int32", "id": null, "_type": "Value"}, "tokens": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "ner": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": {"features": null, "resources_checksums": {"train": {}, "validation": {}, "test": {}}}, "supervised_keys": null, "builder_name": "lince", "config_name": "ner_msaea", "version": {"version_str": "1.0.0", "description": "The Linguistic Code-switching Evaluation (LinCE) benchmark", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 3887684, "num_examples": 10103, "dataset_name": "lince"}, "validation": {"name": "validation", "num_bytes": 431414, "num_examples": 1122, "dataset_name": "lince"}, "test": {"name": "test", "num_bytes": 367310, "num_examples": 1110, "dataset_name": "lince"}}, "download_checksums": {"https://ritual.uh.edu/lince/libaccess/eyJ1c2VybmFtZSI6ICJodWdnaW5nZmFjZSBubHAiLCAidXNlcl9pZCI6IDExMSwgImVtYWlsIjogImR1bW15QGVtYWlsLmNvbSJ9/ner_msaea.zip": {"num_bytes": 938671, "checksum": "77d13c0a5e1b2923ec7fdd2941fda0d08b5fc690756feff12c411c9e73514b69"}}, "download_size": 938671, "post_processing_size": 0, "dataset_size": 4686408, "size_in_bytes": 5625079}, "ner_hineng": {"description": "LinCE is a centralized Linguistic Code-switching Evaluation benchmark \n(https://ritual.uh.edu/lince/) that contains data for training and evaluating\nNLP systems on code-switching tasks.\n", "citation": "\n@inproceedings{singh-etal-2018-language,\n title = \"Language Identification and Named Entity Recognition in {H}inglish Code Mixed Tweets\",\n author = \"Singh, Kushagra and\n Sen, Indira and\n Kumaraguru, Ponnurangam\",\n booktitle = \"Proceedings of {ACL} 2018, Student Research Workshop\",\n month = jul,\n year = \"2018\",\n address = \"Melbourne, Australia\",\n publisher = \"Association for Computational Linguistics\",\n url = \"https://www.aclweb.org/anthology/P18-3008\",\n doi = \"10.18653/v1/P18-3008\",\n pages = \"52--58\",\n}\n\n@inproceedings{aguilar-etal-2020-lince,\n title = \"{L}in{CE}: A Centralized Benchmark for Linguistic Code-switching Evaluation\",\n author = \"Aguilar, Gustavo and\n Kar, Sudipta and\n Solorio, Thamar\",\n booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n month = may,\n year = \"2020\",\n address = \"Marseille, France\",\n publisher = \"European Language Resources Association\",\n url = \"https://www.aclweb.org/anthology/2020.lrec-1.223\",\n pages = \"1803--1813\",\n language = \"English\",\n ISBN = \"979-10-95546-34-4\",\n}\n\nNote that each LinCE dataset has its own citation. Please see the source to see\nthe correct citation for each contained dataset.", "homepage": "http://ritual.uh.edu/lince", "license": "", "features": {"idx": {"dtype": "int32", "id": null, "_type": "Value"}, "tokens": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "lid": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "ner": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": {"features": null, "resources_checksums": {"test": {}, "train": {}, "validation": {}}}, "supervised_keys": null, "builder_name": "lince", "config_name": "ner_hineng", "version": {"version_str": "1.0.0", "description": "The Linguistic Code-switching Evaluation (LinCE) benchmark", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 474639, "num_examples": 1243, "dataset_name": "lince"}, "validation": {"name": "validation", "num_bytes": 121403, "num_examples": 314, "dataset_name": "lince"}, "test": {"name": "test", "num_bytes": 185220, "num_examples": 522, "dataset_name": "lince"}}, "download_checksums": {"https://ritual.uh.edu/lince/libaccess/eyJ1c2VybmFtZSI6ICJodWdnaW5nZmFjZSBubHAiLCAidXNlcl9pZCI6IDExMSwgImVtYWlsIjogImR1bW15QGVtYWlsLmNvbSJ9/ner_hineng.zip": {"num_bytes": 141285, "checksum": "a4abf0672adfbd3c50c9670ecb16df4950c5078473afd8af4b4c15954731772f"}}, "download_size": 141285, "post_processing_size": 0, "dataset_size": 781262, "size_in_bytes": 922547}, "sa_spaeng": {"description": "LinCE is a centralized Linguistic Code-switching Evaluation benchmark \n(https://ritual.uh.edu/lince/) that contains data for training and evaluating\nNLP systems on code-switching tasks.\n", "citation": "\n@inproceedings{patwa2020sentimix,\n title={SemEval-2020 Task 9: Overview of Sentiment Analysis of Code-Mixed Tweets},\n author=\"Patwa, Parth and\n Aguilar, Gustavo and\n Kar, Sudipta and\n Pandey, Suraj and\n PYKL, Srinivas and\n Garrette, Dan and\n Gamb{\"a}ck, Bj{\"o}rn and\n Chakraborty, Tanmoy and\n Solorio, Thamar and \n Das, Amitava\",\n booktitle = \"Proceedings of the 14th International Workshop on Semantic Evaluation ({S}em{E}val-2020)\",\n year = 2020,\n month = sep,\n address = \"Barcelona, Spain\",\n publisher = \"Association for Computational Linguistics\"\n}\n\n@inproceedings{aguilar-etal-2020-lince,\n title = \"{L}in{CE}: A Centralized Benchmark for Linguistic Code-switching Evaluation\",\n author = \"Aguilar, Gustavo and\n Kar, Sudipta and\n Solorio, Thamar\",\n booktitle = \"Proceedings of The 12th Language Resources and Evaluation Conference\",\n month = may,\n year = \"2020\",\n address = \"Marseille, France\",\n publisher = \"European Language Resources Association\",\n url = \"https://www.aclweb.org/anthology/2020.lrec-1.223\",\n pages = \"1803--1813\",\n language = \"English\",\n ISBN = \"979-10-95546-34-4\",\n}\n\nNote that each LinCE dataset has its own citation. Please see the source to see\nthe correct citation for each contained dataset.", "homepage": "http://ritual.uh.edu/lince", "license": "", "features": {"idx": {"dtype": "int32", "id": null, "_type": "Value"}, "tokens": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "lid": {"feature": {"dtype": "string", "id": null, "_type": "Value"}, "length": -1, "id": null, "_type": "Sequence"}, "sa": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": {"features": null, "resources_checksums": {"train": {}, "validation": {}, "test": {}}}, "supervised_keys": null, "builder_name": "lince", "config_name": "sa_spaeng", "version": {"version_str": "1.0.0", "description": "The Linguistic Code-switching Evaluation (LinCE) benchmark", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 3587783, "num_examples": 12194, "dataset_name": "lince"}, "validation": {"name": "validation", "num_bytes": 546692, "num_examples": 1859, "dataset_name": "lince"}, "test": {"name": "test", "num_bytes": 1349407, "num_examples": 4736, "dataset_name": "lince"}}, "download_checksums": {"https://ritual.uh.edu/lince/libaccess/eyJ1c2VybmFtZSI6ICJodWdnaW5nZmFjZSBubHAiLCAidXNlcl9pZCI6IDExMSwgImVtYWlsIjogImR1bW15QGVtYWlsLmNvbSJ9/sa_spaeng.zip": {"num_bytes": 1031412, "checksum": "e184ec996551e1bebe872910234685119750004da2787e2da132a0580ab74246"}}, "download_size": 1031412, "post_processing_size": 0, "dataset_size": 5483882, "size_in_bytes": 6515294}}
dummy/lid_hineng/1.0.0/dummy_data.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b48543b39a585b4af2e71ffe15897e978a2a9643bf39f0b3ad529b6b726223e9
3
+ size 1746
dummy/lid_msaea/1.0.0/dummy_data.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12750b265630fe89600de0cad8dccfb314ef4e93cb72f5994186b5f376314d44
3
+ size 1948
dummy/lid_nepeng/1.0.0/dummy_data.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4fa68a59d77a2755b993684a5d242666318fe539284e47e726511bdbe92afc93
3
+ size 1865
dummy/lid_spaeng/1.0.0/dummy_data.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7de58038e16a1a37e6f3a1edf86d0823f6c6e3e27055d28eb312af41e03b4fd5
3
+ size 1719
dummy/ner_hineng/1.0.0/dummy_data.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b670ddd070e165b6df6c0ef9ccc3817ee58c573b42e96a54edd97c9a4fdce90
3
+ size 1915
dummy/ner_msaea/1.0.0/dummy_data.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e99392b4723f84b8a9f5f0c24b60d1b28cdf7bc6dba04220d6d99b156e2932f
3
+ size 2126
dummy/ner_spaeng/1.0.0/dummy_data.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc76fd7ec200a0b20ab6e8f8aa97d692eee769885bda380eeb8caf9049f477b7
3
+ size 1974
dummy/pos_hineng/1.0.0/dummy_data.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:254d54932c548855ec7e867b2502fe487df822cdfeefc6cef654a438c096215d
3
+ size 2101
dummy/pos_spaeng/1.0.0/dummy_data.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8dc4e27e3d64e2968c6491ade18f9d27fac977aae39d0eb6a3a7bf1848c2d32
3
+ size 1735
dummy/sa_spaeng/1.0.0/dummy_data.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b5c474922a887ff340692874998835dff590bb8740bb0c11042d44d8c8ed5aea
3
+ size 1889
lince.py ADDED
@@ -0,0 +1,548 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """TODO(lince): Add a description here."""
2
+
3
+ from __future__ import absolute_import, division, print_function
4
+
5
+ import csv
6
+ import os
7
+ import re
8
+ import textwrap
9
+ from itertools import groupby
10
+
11
+ import six
12
+
13
+ import datasets
14
+
15
+
16
+ _CITATION = """\
17
+ @inproceedings{aguilar-etal-2020-lince,
18
+ title = "{L}in{CE}: A Centralized Benchmark for Linguistic Code-switching Evaluation",
19
+ author = "Aguilar, Gustavo and
20
+ Kar, Sudipta and
21
+ Solorio, Thamar",
22
+ booktitle = "Proceedings of The 12th Language Resources and Evaluation Conference",
23
+ month = may,
24
+ year = "2020",
25
+ address = "Marseille, France",
26
+ publisher = "European Language Resources Association",
27
+ url = "https://www.aclweb.org/anthology/2020.lrec-1.223",
28
+ pages = "1803--1813",
29
+ language = "English",
30
+ ISBN = "979-10-95546-34-4",
31
+ }
32
+
33
+ Note that each LinCE dataset has its own citation. Please see the source to see
34
+ the correct citation for each contained dataset."""
35
+
36
+ _DESCRIPTION = """\
37
+ LinCE is a centralized Linguistic Code-switching Evaluation benchmark
38
+ (https://ritual.uh.edu/lince/) that contains data for training and evaluating
39
+ NLP systems on code-switching tasks.
40
+ """
41
+
42
+ _LINCE_URL = "https://ritual.uh.edu/lince/libaccess/eyJ1c2VybmFtZSI6ICJodWdnaW5nZmFjZSBubHAiLCAidXNlcl9pZCI6IDExMSwgImVtYWlsIjogImR1bW15QGVtYWlsLmNvbSJ9"
43
+
44
+ _DATASET_CITATIONS = {
45
+ "lid_spaeng": textwrap.dedent(
46
+ """
47
+ @inproceedings{molina-etal-2016-overview,
48
+ title = "Overview for the Second Shared Task on Language Identification in Code-Switched Data",
49
+ author = "Molina, Giovanni and
50
+ AlGhamdi, Fahad and
51
+ Ghoneim, Mahmoud and
52
+ Hawwari, Abdelati and
53
+ Rey-Villamizar, Nicolas and
54
+ Diab, Mona and
55
+ Solorio, Thamar",
56
+ booktitle = "Proceedings of the Second Workshop on Computational Approaches to Code Switching",
57
+ month = nov,
58
+ year = "2016",
59
+ address = "Austin, Texas",
60
+ publisher = "Association for Computational Linguistics",
61
+ url = "https://www.aclweb.org/anthology/W16-5805",
62
+ doi = "10.18653/v1/W16-5805",
63
+ pages = "40--49",
64
+ }
65
+ """
66
+ ),
67
+ "lid_hineng": textwrap.dedent(
68
+ """
69
+ @inproceedings{mave-etal-2018-language,
70
+ title = "Language Identification and Analysis of Code-Switched Social Media Text",
71
+ author = "Mave, Deepthi and
72
+ Maharjan, Suraj and
73
+ Solorio, Thamar",
74
+ booktitle = "Proceedings of the Third Workshop on Computational Approaches to Linguistic Code-Switching",
75
+ month = jul,
76
+ year = "2018",
77
+ address = "Melbourne, Australia",
78
+ publisher = "Association for Computational Linguistics",
79
+ url = "https://www.aclweb.org/anthology/W18-3206",
80
+ pages = "51--61"
81
+ }
82
+ """
83
+ ),
84
+ "lid_msaea": textwrap.dedent(
85
+ """
86
+ @inproceedings{molina-etal-2016-overview,
87
+ title = "Overview for the Second Shared Task on Language Identification in Code-Switched Data",
88
+ author = "Molina, Giovanni and
89
+ AlGhamdi, Fahad and
90
+ Ghoneim, Mahmoud and
91
+ Hawwari, Abdelati and
92
+ Rey-Villamizar, Nicolas and
93
+ Diab, Mona and
94
+ Solorio, Thamar",
95
+ booktitle = "Proceedings of the Second Workshop on Computational Approaches to Code Switching",
96
+ month = nov,
97
+ year = "2016",
98
+ address = "Austin, Texas",
99
+ publisher = "Association for Computational Linguistics",
100
+ url = "https://www.aclweb.org/anthology/W16-5805",
101
+ doi = "10.18653/v1/W16-5805",
102
+ pages = "40--49",
103
+ }
104
+ """
105
+ ),
106
+ "lid_nepeng": textwrap.dedent(
107
+ """
108
+ @inproceedings{solorio-etal-2014-overview,
109
+ title = "Overview for the First Shared Task on Language Identification in Code-Switched Data",
110
+ author = "Solorio, Thamar and
111
+ Blair, Elizabeth and
112
+ Maharjan, Suraj and
113
+ Bethard, Steven and
114
+ Diab, Mona and
115
+ Ghoneim, Mahmoud and
116
+ Hawwari, Abdelati and
117
+ AlGhamdi, Fahad and
118
+ Hirschberg, Julia and
119
+ Chang, Alison and
120
+ Fung, Pascale",
121
+ booktitle = "Proceedings of the First Workshop on Computational Approaches to Code Switching",
122
+ month = oct,
123
+ year = "2014",
124
+ address = "Doha, Qatar",
125
+ publisher = "Association for Computational Linguistics",
126
+ url = "https://www.aclweb.org/anthology/W14-3907",
127
+ doi = "10.3115/v1/W14-3907",
128
+ pages = "62--72",
129
+ }
130
+ """
131
+ ),
132
+ "pos_spaeng": textwrap.dedent(
133
+ """
134
+ @inproceedings{alghamdi-etal-2016-part,
135
+ title = "Part of Speech Tagging for Code Switched Data",
136
+ author = "AlGhamdi, Fahad and
137
+ Molina, Giovanni and
138
+ Diab, Mona and
139
+ Solorio, Thamar and
140
+ Hawwari, Abdelati and
141
+ Soto, Victor and
142
+ Hirschberg, Julia",
143
+ booktitle = "Proceedings of the Second Workshop on Computational Approaches to Code Switching",
144
+ month = nov,
145
+ year = "2016",
146
+ address = "Austin, Texas",
147
+ publisher = "Association for Computational Linguistics",
148
+ url = "https://www.aclweb.org/anthology/W16-5812",
149
+ doi = "10.18653/v1/W16-5812",
150
+ pages = "98--107",
151
+ }
152
+ """
153
+ ),
154
+ "pos_hineng": textwrap.dedent(
155
+ """
156
+ @inproceedings{singh-etal-2018-twitter,
157
+ title = "A Twitter Corpus for {H}indi-{E}nglish Code Mixed {POS} Tagging",
158
+ author = "Singh, Kushagra and
159
+ Sen, Indira and
160
+ Kumaraguru, Ponnurangam",
161
+ booktitle = "Proceedings of the Sixth International Workshop on Natural Language Processing for Social Media",
162
+ month = jul,
163
+ year = "2018",
164
+ address = "Melbourne, Australia",
165
+ publisher = "Association for Computational Linguistics",
166
+ url = "https://www.aclweb.org/anthology/W18-3503",
167
+ doi = "10.18653/v1/W18-3503",
168
+ pages = "12--17"
169
+ }
170
+ """
171
+ ),
172
+ "ner_spaeng": textwrap.dedent(
173
+ """
174
+ @inproceedings{aguilar-etal-2018-named,
175
+ title = {{Named Entity Recognition on Code-Switched Data: Overview of the CALCS 2018 Shared Task}},
176
+ author = "Aguilar, Gustavo and
177
+ AlGhamdi, Fahad and
178
+ Soto, Victor and
179
+ Diab, Mona and
180
+ Hirschberg, Julia and
181
+ Solorio, Thamar",
182
+ booktitle = {{Proceedings of the Third Workshop on Computational Approaches to Linguistic Code-Switching}},
183
+ month = jul,
184
+ year = "2018",
185
+ address = "Melbourne, Australia",
186
+ publisher = "Association for Computational Linguistics",
187
+ url = "https://www.aclweb.org/anthology/W18-3219",
188
+ pages = "138--147"
189
+ }
190
+ """
191
+ ),
192
+ "ner_msaea": textwrap.dedent(
193
+ """
194
+ @inproceedings{aguilar-etal-2018-named,
195
+ title = {{Named Entity Recognition on Code-Switched Data: Overview of the CALCS 2018 Shared Task}},
196
+ author = "Aguilar, Gustavo and
197
+ AlGhamdi, Fahad and
198
+ Soto, Victor and
199
+ Diab, Mona and
200
+ Hirschberg, Julia and
201
+ Solorio, Thamar",
202
+ booktitle = {{Proceedings of the Third Workshop on Computational Approaches to Linguistic Code-Switching}},
203
+ month = jul,
204
+ year = "2018",
205
+ address = "Melbourne, Australia",
206
+ publisher = "Association for Computational Linguistics",
207
+ url = "https://www.aclweb.org/anthology/W18-3219",
208
+ pages = "138--147"
209
+ }
210
+ """
211
+ ),
212
+ "ner_hineng": textwrap.dedent(
213
+ """
214
+ @inproceedings{singh-etal-2018-language,
215
+ title = "Language Identification and Named Entity Recognition in {H}inglish Code Mixed Tweets",
216
+ author = "Singh, Kushagra and
217
+ Sen, Indira and
218
+ Kumaraguru, Ponnurangam",
219
+ booktitle = "Proceedings of {ACL} 2018, Student Research Workshop",
220
+ month = jul,
221
+ year = "2018",
222
+ address = "Melbourne, Australia",
223
+ publisher = "Association for Computational Linguistics",
224
+ url = "https://www.aclweb.org/anthology/P18-3008",
225
+ doi = "10.18653/v1/P18-3008",
226
+ pages = "52--58",
227
+ }
228
+ """
229
+ ),
230
+ "sa_spaeng": textwrap.dedent(
231
+ """
232
+ @inproceedings{patwa2020sentimix,
233
+ title={SemEval-2020 Task 9: Overview of Sentiment Analysis of Code-Mixed Tweets},
234
+ author="Patwa, Parth and
235
+ Aguilar, Gustavo and
236
+ Kar, Sudipta and
237
+ Pandey, Suraj and
238
+ PYKL, Srinivas and
239
+ Garrette, Dan and
240
+ Gamb{\"a}ck, Bj{\"o}rn and
241
+ Chakraborty, Tanmoy and
242
+ Solorio, Thamar and
243
+ Das, Amitava",
244
+ booktitle = "Proceedings of the 14th International Workshop on Semantic Evaluation ({S}em{E}val-2020)",
245
+ year = 2020,
246
+ month = sep,
247
+ address = "Barcelona, Spain",
248
+ publisher = "Association for Computational Linguistics"
249
+ }
250
+ """
251
+ ),
252
+ }
253
+
254
+
255
+ class LinceConfig(datasets.BuilderConfig):
256
+ """BuilderConfig for LinCE"""
257
+
258
+ def __init__(self, colnames, classes, label_column, **kwargs):
259
+ super(LinceConfig, self).__init__(
260
+ version=datasets.Version(
261
+ "1.0.0", description="The Linguistic Code-switching Evaluation (LinCE) benchmark"
262
+ ),
263
+ **kwargs,
264
+ )
265
+ self.colnames = colnames
266
+ self.classes = classes
267
+ self.label_column = label_column
268
+
269
+
270
+ class Lince(datasets.GeneratorBasedBuilder):
271
+ """TODO(lince): Short description of the LinCE dataset."""
272
+
273
+ BUILDER_CONFIG_CLASS = LinceConfig
274
+ BUILDER_CONFIGS = [
275
+ # ==========================================================================================
276
+ # Language Identification (LID) datasets
277
+ LinceConfig(
278
+ name="lid_spaeng",
279
+ data_dir="lid_spaeng",
280
+ colnames={"tokens": 0, "lid": 1},
281
+ classes={"lid": ["lang1", "lang2", "ne", "fw", "ambiguous", "mixed", "other", "unk"]},
282
+ label_column="lid",
283
+ description="Spanish-English language identification dataset (Latin script)",
284
+ ),
285
+ LinceConfig(
286
+ name="lid_hineng",
287
+ data_dir="lid_hineng",
288
+ colnames={"tokens": 0, "lid": 1},
289
+ classes={"lid": ["lang1", "lang2", "ne", "fw", "ambiguous", "mixed", "other", "unk"]},
290
+ label_column="lid",
291
+ description="Hindi-English language identification dataset (Latin script)",
292
+ ),
293
+ LinceConfig(
294
+ name="lid_msaea",
295
+ data_dir="lid_msaea",
296
+ colnames={"tokens": 0, "lid": 1},
297
+ classes={
298
+ "lid": ["ambiguous", "lang1", "lang2", "mixed", "ne", "other"],
299
+ },
300
+ label_column="lid",
301
+ description="Modern Standard Arabic-Egyptian Arabic language identification dataset (Persian script)",
302
+ ),
303
+ LinceConfig(
304
+ name="lid_nepeng",
305
+ data_dir="lid_nepeng",
306
+ colnames={"tokens": 0, "lid": 1},
307
+ classes={
308
+ "lid": ["ambiguous", "lang1", "lang2", "mixed", "ne", "other"],
309
+ },
310
+ label_column="lid",
311
+ description="Nepali-English language identification dataset (Latin script)",
312
+ ),
313
+ # ==========================================================================================
314
+ # Part-of-Speech (POS) Tagging datasets
315
+ LinceConfig(
316
+ name="pos_spaeng",
317
+ data_dir="pos_spaeng",
318
+ colnames={"tokens": 0, "lid": 1, "pos": 2},
319
+ classes={
320
+ "lid": ["UNK", "eng", "eng&spa", "spa"],
321
+ "pos": [
322
+ "ADJ",
323
+ "ADP",
324
+ "ADV",
325
+ "AUX",
326
+ "CONJ",
327
+ "DET",
328
+ "INTJ",
329
+ "NOUN",
330
+ "NUM",
331
+ "PART",
332
+ "PRON",
333
+ "PROPN",
334
+ "PUNCT",
335
+ "SCONJ",
336
+ "UNK",
337
+ "VERB",
338
+ "X",
339
+ ],
340
+ },
341
+ label_column="pos",
342
+ description="Spanish-English part-of-Speech tagging dataset (Latin script)",
343
+ ),
344
+ LinceConfig(
345
+ name="pos_hineng",
346
+ data_dir="pos_hineng",
347
+ colnames={"tokens": 0, "lid": 1, "pos": 2},
348
+ classes={
349
+ "lid": ["en", "hi", "rest"],
350
+ "pos": [
351
+ "ADJ",
352
+ "ADP",
353
+ "ADV",
354
+ "CONJ",
355
+ "DET",
356
+ "NOUN",
357
+ "NUM",
358
+ "PART",
359
+ "PART_NEG",
360
+ "PRON",
361
+ "PRON_WH",
362
+ "PROPN",
363
+ "VERB",
364
+ "X",
365
+ ],
366
+ },
367
+ label_column="pos",
368
+ description="Hindi-English part-of-Speech tagging dataset (Latin script)",
369
+ ),
370
+ # ==========================================================================================
371
+ # Named Entity Recongition (NER) datasets
372
+ LinceConfig(
373
+ name="ner_spaeng",
374
+ data_dir="ner_spaeng",
375
+ colnames={"tokens": 0, "lid": 1, "ner": 2},
376
+ classes={
377
+ "lid": ["lang1", "lang2", "ne", "fw", "ambiguous", "mixed", "other", "unk"],
378
+ "ner": [
379
+ "O",
380
+ "B-PER",
381
+ "I-PER",
382
+ "B-LOC",
383
+ "I-LOC",
384
+ "B-ORG",
385
+ "I-ORG",
386
+ "B-PROD",
387
+ "I-PROD",
388
+ "B-EVENT",
389
+ "I-EVENT",
390
+ "B-GROUP",
391
+ "I-GROUP",
392
+ "B-TITLE",
393
+ "I-TITLE",
394
+ "B-TIME",
395
+ "I-TIME",
396
+ "B-OTHER",
397
+ "I-OTHER",
398
+ ],
399
+ },
400
+ label_column="ner",
401
+ description="Spanish-English named entity recognition dataset (Latin script)",
402
+ ),
403
+ LinceConfig(
404
+ name="ner_msaea",
405
+ data_dir="ner_msaea",
406
+ colnames={"tokens": 0, "ner": 1},
407
+ classes={
408
+ "ner": [
409
+ "O",
410
+ "B-PER",
411
+ "I-PER",
412
+ "B-LOC",
413
+ "I-LOC",
414
+ "B-ORG",
415
+ "I-ORG",
416
+ "B-PROD",
417
+ "I-PROD",
418
+ "B-EVENT",
419
+ "I-EVENT",
420
+ "B-GROUP",
421
+ "I-GROUP",
422
+ "B-TITLE",
423
+ "I-TITLE",
424
+ "B-TIME",
425
+ "I-TIME",
426
+ "B-OTHER",
427
+ "I-OTHER",
428
+ ],
429
+ },
430
+ label_column="ner",
431
+ description="Modern Standard Arabic-Egyptian Arabic named entity recognition dataset (Persian script)",
432
+ ),
433
+ LinceConfig(
434
+ name="ner_hineng",
435
+ data_dir="ner_hineng",
436
+ colnames={"tokens": 0, "lid": 1, "ner": 2},
437
+ classes={
438
+ "lid": ["en", "hi", "rest"],
439
+ "ner": ["O", "B-PERSON", "I-PERSON", "B-ORGANISATION", "I-ORGANISATION", "B-PLACE", "I-PLACE"],
440
+ },
441
+ label_column="ner",
442
+ description="Hindi-English named entity recognition dataset (Latin script)",
443
+ ),
444
+ # ==========================================================================================
445
+ # Sentiment Analysis (SA) datasets
446
+ LinceConfig(
447
+ name="sa_spaeng",
448
+ data_dir="sa_spaeng",
449
+ colnames={"tokens": 0, "lid": 1, "sa": 2},
450
+ classes={
451
+ "lid": ["lang1", "lang2", "ne", "fw", "ambiguous", "mixed", "other", "unk"],
452
+ "sa": ["positive", "neutral", "negative"],
453
+ },
454
+ label_column="sa",
455
+ description="Spanish-English sentiment analysis dataset (Latin script)",
456
+ ),
457
+ ]
458
+
459
+ def _info(self):
460
+ features = {"idx": datasets.Value("int32"), "tokens": datasets.Sequence(datasets.Value("string"))}
461
+
462
+ if self.config.name != "ner_msaea":
463
+ features["lid"] = datasets.Sequence(
464
+ datasets.Value("string")
465
+ ) # excluding 'ner_msaea', all datasets have 'lid'
466
+
467
+ if self.config.name.startswith("pos_"):
468
+ features["pos"] = datasets.Sequence(datasets.Value("string"))
469
+
470
+ elif self.config.name.startswith("ner_"):
471
+ features["ner"] = datasets.Sequence(datasets.Value("string"))
472
+
473
+ elif self.config.name.startswith("sa_"):
474
+ features["sa"] = datasets.Value("string")
475
+
476
+ return datasets.DatasetInfo(
477
+ description=_DESCRIPTION,
478
+ features=datasets.Features(features),
479
+ supervised_keys=None,
480
+ homepage="http://ritual.uh.edu/lince",
481
+ citation=_DATASET_CITATIONS.get(self.config.name, "") + "\n" + _CITATION,
482
+ )
483
+
484
+ def _split_generators(self, dl_manager):
485
+ """Returns SplitGenerators."""
486
+ lince_dir = dl_manager.download_and_extract(f"{_LINCE_URL}/{self.config.name}.zip")
487
+ data_dir = os.path.join(lince_dir, self.config.data_dir)
488
+ return [
489
+ datasets.SplitGenerator(
490
+ name=datasets.Split.TRAIN,
491
+ gen_kwargs={
492
+ "filepath": os.path.join(data_dir, "train.conll"),
493
+ "colnames": self.config.colnames,
494
+ },
495
+ ),
496
+ datasets.SplitGenerator(
497
+ name=datasets.Split.VALIDATION,
498
+ gen_kwargs={
499
+ "filepath": os.path.join(data_dir, "dev.conll"),
500
+ "colnames": self.config.colnames,
501
+ },
502
+ ),
503
+ datasets.SplitGenerator(
504
+ name=datasets.Split.TEST,
505
+ gen_kwargs={
506
+ "filepath": os.path.join(data_dir, "test.conll"),
507
+ "colnames": {
508
+ key: self.config.colnames[key]
509
+ for key in six.iterkeys(self.config.colnames)
510
+ if key != self.config.label_column
511
+ },
512
+ },
513
+ ),
514
+ ]
515
+
516
+ def _generate_examples(self, filepath, colnames, delimiter="\t", metaregex=r"^# sent_enum = [0-9]+$"):
517
+ def is_empty_line(token_pack):
518
+ return all(field.strip() == "" for field in token_pack)
519
+
520
+ index = 0
521
+ for is_empty, pack in groupby(
522
+ csv.reader(open(filepath), delimiter=delimiter, quoting=csv.QUOTE_NONE), is_empty_line
523
+ ):
524
+ if is_empty is False:
525
+ # packed sentence -> [['tok_1', 'lid_1', 'ner_1'], ..., ['tok_n', 'lid_n', 'ner_n']]
526
+ pack = list(pack)
527
+ meta = []
528
+
529
+ if re.match(metaregex, pack[0][0]): # for sentence-level annotations (or meta)
530
+ meta = pack[0][1:] # keep the sentence-level labels
531
+ pack = pack[1:] # ignore meta fields to unzip
532
+
533
+ unpacked = list(zip(*pack))
534
+ if meta:
535
+ unpacked = unpacked + meta
536
+
537
+ row = {feature: unpacked[colnames[feature]] for feature in colnames}
538
+ row["idx"] = index
539
+ index += 1
540
+
541
+ # dummy labels for the test set
542
+ if self.config.label_column not in row:
543
+ if self.config.label_column == "sa":
544
+ row[self.config.label_column] = ""
545
+ else:
546
+ row[self.config.label_column] = [""] * len(row["tokens"])
547
+
548
+ yield index, row