Update files from the datasets library (from 1.0.0)

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

.gitattributes

@@ -0,0 +1,27 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bin.* filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zstandard filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

dataset_infos.json

@@ -0,0 +1 @@
+{"iwslt2017-en-it": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["en", "it"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-en-it", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 46648117, "num_examples": 231619, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 305254, "num_examples": 1566, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 200031, "num_examples": 929, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnmted//texts/DeEnItNlRo/DeEnItNlRo/DeEnItNlRo-DeEnItNlRo.tgz": {"num_bytes": 329331279, "checksum": "cebb3839b7580f8e4eecf66d75424caa29b9410a9a491a732b2d343626ee0243"}}, "download_size": 329331279, "dataset_size": 47153402, "size_in_bytes": 376484681}, "iwslt2017-en-nl": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["en", "nl"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-en-nl", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 42844125, "num_examples": 237240, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 311654, "num_examples": 1777, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 197822, "num_examples": 1003, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnmted//texts/DeEnItNlRo/DeEnItNlRo/DeEnItNlRo-DeEnItNlRo.tgz": {"num_bytes": 329331279, "checksum": "cebb3839b7580f8e4eecf66d75424caa29b9410a9a491a732b2d343626ee0243"}}, "download_size": 329331279, "dataset_size": 43353601, "size_in_bytes": 372684880}, "iwslt2017-en-ro": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["en", "ro"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-en-ro", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 44130134, "num_examples": 220538, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 316798, "num_examples": 1678, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 205036, "num_examples": 914, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnmted//texts/DeEnItNlRo/DeEnItNlRo/DeEnItNlRo-DeEnItNlRo.tgz": {"num_bytes": 329331279, "checksum": "cebb3839b7580f8e4eecf66d75424caa29b9410a9a491a732b2d343626ee0243"}}, "download_size": 329331279, "dataset_size": 44651968, "size_in_bytes": 373983247}, "iwslt2017-it-en": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["it", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-it-en", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 46648117, "num_examples": 231619, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 305254, "num_examples": 1566, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 200031, "num_examples": 929, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnmted//texts/DeEnItNlRo/DeEnItNlRo/DeEnItNlRo-DeEnItNlRo.tgz": {"num_bytes": 329331279, "checksum": "cebb3839b7580f8e4eecf66d75424caa29b9410a9a491a732b2d343626ee0243"}}, "download_size": 329331279, "dataset_size": 47153402, "size_in_bytes": 376484681}, "iwslt2017-it-nl": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["it", "nl"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-it-nl", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 43033360, "num_examples": 233415, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 309733, "num_examples": 1669, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 197782, "num_examples": 1001, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnmted//texts/DeEnItNlRo/DeEnItNlRo/DeEnItNlRo-DeEnItNlRo.tgz": {"num_bytes": 329331279, "checksum": "cebb3839b7580f8e4eecf66d75424caa29b9410a9a491a732b2d343626ee0243"}}, "download_size": 329331279, "dataset_size": 43540875, "size_in_bytes": 372872154}, "iwslt2017-it-ro": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["it", "ro"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-it-ro", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 44485345, "num_examples": 217551, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 314982, "num_examples": 1643, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 204997, "num_examples": 914, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnmted//texts/DeEnItNlRo/DeEnItNlRo/DeEnItNlRo-DeEnItNlRo.tgz": {"num_bytes": 329331279, "checksum": "cebb3839b7580f8e4eecf66d75424caa29b9410a9a491a732b2d343626ee0243"}}, "download_size": 329331279, "dataset_size": 45005324, "size_in_bytes": 374336603}, "iwslt2017-nl-en": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["nl", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-nl-en", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 42844125, "num_examples": 237240, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 311654, "num_examples": 1777, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 197822, "num_examples": 1003, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnmted//texts/DeEnItNlRo/DeEnItNlRo/DeEnItNlRo-DeEnItNlRo.tgz": {"num_bytes": 329331279, "checksum": "cebb3839b7580f8e4eecf66d75424caa29b9410a9a491a732b2d343626ee0243"}}, "download_size": 329331279, "dataset_size": 43353601, "size_in_bytes": 372684880}, "iwslt2017-nl-it": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["nl", "it"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-nl-it", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 43033360, "num_examples": 233415, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 309733, "num_examples": 1669, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 197782, "num_examples": 1001, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnmted//texts/DeEnItNlRo/DeEnItNlRo/DeEnItNlRo-DeEnItNlRo.tgz": {"num_bytes": 329331279, "checksum": "cebb3839b7580f8e4eecf66d75424caa29b9410a9a491a732b2d343626ee0243"}}, "download_size": 329331279, "dataset_size": 43540875, "size_in_bytes": 372872154}, "iwslt2017-nl-ro": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["nl", "ro"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-nl-ro", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 41338906, "num_examples": 206920, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 320960, "num_examples": 1680, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 202388, "num_examples": 913, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnmted//texts/DeEnItNlRo/DeEnItNlRo/DeEnItNlRo-DeEnItNlRo.tgz": {"num_bytes": 329331279, "checksum": "cebb3839b7580f8e4eecf66d75424caa29b9410a9a491a732b2d343626ee0243"}}, "download_size": 329331279, "dataset_size": 41862254, "size_in_bytes": 371193533}, "iwslt2017-ro-en": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["ro", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-ro-en", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 44130134, "num_examples": 220538, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 316798, "num_examples": 1678, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 205036, "num_examples": 914, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnmted//texts/DeEnItNlRo/DeEnItNlRo/DeEnItNlRo-DeEnItNlRo.tgz": {"num_bytes": 329331279, "checksum": "cebb3839b7580f8e4eecf66d75424caa29b9410a9a491a732b2d343626ee0243"}}, "download_size": 329331279, "dataset_size": 44651968, "size_in_bytes": 373983247}, "iwslt2017-ro-it": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["ro", "it"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-ro-it", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 44485345, "num_examples": 217551, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 314982, "num_examples": 1643, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 204997, "num_examples": 914, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnmted//texts/DeEnItNlRo/DeEnItNlRo/DeEnItNlRo-DeEnItNlRo.tgz": {"num_bytes": 329331279, "checksum": "cebb3839b7580f8e4eecf66d75424caa29b9410a9a491a732b2d343626ee0243"}}, "download_size": 329331279, "dataset_size": 45005324, "size_in_bytes": 374336603}, "iwslt2017-ro-nl": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["ro", "nl"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-ro-nl", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 41338906, "num_examples": 206920, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 320960, "num_examples": 1680, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 202388, "num_examples": 913, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnmted//texts/DeEnItNlRo/DeEnItNlRo/DeEnItNlRo-DeEnItNlRo.tgz": {"num_bytes": 329331279, "checksum": "cebb3839b7580f8e4eecf66d75424caa29b9410a9a491a732b2d343626ee0243"}}, "download_size": 329331279, "dataset_size": 41862254, "size_in_bytes": 371193533}, "iwslt2017-ar-en": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["ar", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-ar-en", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 56481251, "num_examples": 231713, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 2014304, "num_examples": 8583, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 241214, "num_examples": 888, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnted/texts/ar/en/ar-en.tgz": {"num_bytes": 27747532, "checksum": "f1b77caa115cf5cbd00e519b425e33615080c8735fd9dd59a0c811bacd10ab70"}}, "download_size": 27747532, "dataset_size": 58736769, "size_in_bytes": 86484301}, "iwslt2017-de-en": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["de", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-de-en", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 42608548, "num_examples": 206112, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 1608482, "num_examples": 8079, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 210983, "num_examples": 888, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnted/texts/de/en/de-en.tgz": {"num_bytes": 16751373, "checksum": "ebcd53c53c8df742437e3a0fb964772e78764c3f6d4e38df0b6a2f71abc3a8b4"}}, "download_size": 16751373, "dataset_size": 44428013, "size_in_bytes": 61179386}, "iwslt2017-en-ar": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["en", "ar"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-en-ar", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 56481251, "num_examples": 231713, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 2014304, "num_examples": 8583, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 241214, "num_examples": 888, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnted/texts/en/ar/en-ar.tgz": {"num_bytes": 29332212, "checksum": "7c9446ebf519fb9ca85d7fb7a6681918772eafc6a33bfdf9852f87442aabd3f7"}}, "download_size": 29332212, "dataset_size": 58736769, "size_in_bytes": 88068981}, "iwslt2017-en-de": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["en", "de"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-en-de", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 42608548, "num_examples": 206112, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 1608482, "num_examples": 8079, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 210983, "num_examples": 888, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnted/texts/en/de/en-de.tgz": {"num_bytes": 16751364, "checksum": "623caec69988b224dbe741f16ad90411b93430117a8021c4bb2bcab7ad1eb21e"}}, "download_size": 16751364, "dataset_size": 44428013, "size_in_bytes": 61179377}, "iwslt2017-en-fr": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["en", "fr"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-en-fr", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 49273478, "num_examples": 232825, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 1767473, "num_examples": 8597, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 207587, "num_examples": 890, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnted/texts/en/fr/en-fr.tgz": {"num_bytes": 27691674, "checksum": "1f5bd8cd242de15fb58a04af265adb8e604a7e1b805a42cb0de7dfd15447c5c6"}}, "download_size": 27691674, "dataset_size": 51248538, "size_in_bytes": 78940212}, "iwslt2017-en-ja": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["en", "ja"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-en-ja", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 48205171, "num_examples": 223108, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 1809015, "num_examples": 8469, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 208132, "num_examples": 871, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnted/texts/en/ja/en-ja.tgz": {"num_bytes": 26983540, "checksum": "27322c1170d057a92f27edd475cfd2c46f38bd1acdd52bbcffbe725e78b3d472"}}, "download_size": 26983540, "dataset_size": 50222318, "size_in_bytes": 77205858}, "iwslt2017-en-ko": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["en", "ko"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-en-ko", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 51678235, "num_examples": 230240, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 1869801, "num_examples": 8514, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 219303, "num_examples": 879, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnted/texts/en/ko/en-ko.tgz": {"num_bytes": 19363699, "checksum": "0ac6caa1fbfbb0c79f1e9ef2f8e5493b83003d6ad85ef504a5584e3b8b006f00"}}, "download_size": 19363699, "dataset_size": 53767339, "size_in_bytes": 73131038}, "iwslt2017-en-zh": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["en", "zh"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-en-zh", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 44271196, "num_examples": 231266, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 1605535, "num_examples": 8549, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 202545, "num_examples": 879, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnted/texts/en/zh/en-zh.tgz": {"num_bytes": 27596371, "checksum": "d178f2ede07d7d600e244649e2092fb5fcbdda9729a3bb016502512028e2615b"}}, "download_size": 27596371, "dataset_size": 46079276, "size_in_bytes": 73675647}, "iwslt2017-fr-en": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["fr", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-fr-en", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 49273478, "num_examples": 232825, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 1767473, "num_examples": 8597, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 207587, "num_examples": 890, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnted/texts/fr/en/fr-en.tgz": {"num_bytes": 26872507, "checksum": "a40ca3b9a0064bd1fd582df4ced260dd3e7ff9f6f7bf863f139f74b7a5cabd78"}}, "download_size": 26872507, "dataset_size": 51248538, "size_in_bytes": 78121045}, "iwslt2017-ja-en": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["ja", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-ja-en", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 48205171, "num_examples": 223108, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 1809015, "num_examples": 8469, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 208132, "num_examples": 871, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnted/texts/ja/en/ja-en.tgz": {"num_bytes": 26190272, "checksum": "19b394f0256e7398bc69fb27c0fde15cd2a6c372a910ad99707ae32d54371640"}}, "download_size": 26190272, "dataset_size": 50222318, "size_in_bytes": 76412590}, "iwslt2017-ko-en": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["ko", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-ko-en", "version": {"version_str": "0.0.0", "description": null, "datasets_version_to_prepare": null, "major": 0, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 51678235, "num_examples": 230240, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 1869801, "num_examples": 8514, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 219303, "num_examples": 879, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnted/texts/ko/en/ko-en.tgz": {"num_bytes": 19363701, "checksum": "c0cba8c055a9bd55902aff5d61783ddad281df86f2ebb673d97b126b734a0987"}}, "download_size": 19363701, "dataset_size": 53767339, "size_in_bytes": 73131040}, "iwslt2017-zh-en": {"description": "The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:\n\n  German, English, Italian, Dutch, Romanian.\n\nFor each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.\n", "citation": "@inproceedings{cettoloEtAl:EAMT2012,\nAddress = {Trento, Italy},\nAuthor = {Mauro Cettolo and Christian Girardi and Marcello Federico},\nBooktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},\nDate = {28-30},\nMonth = {May},\nPages = {261--268},\nTitle = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},\nYear = {2012}}\n", "homepage": "https://sites.google.com/site/iwsltevaluation2017/TED-tasks", "license": "", "features": {"translation": {"languages": ["zh", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": null, "builder_name": "iwsl_t217", "config_name": "iwslt2017-zh-en", "version": "0.0.0", "splits": {"train": {"name": "train", "num_bytes": 44271196, "num_examples": 231266, "dataset_name": "iwsl_t217"}, "test": {"name": "test", "num_bytes": 1605535, "num_examples": 8549, "dataset_name": "iwsl_t217"}, "validation": {"name": "validation", "num_bytes": 202545, "num_examples": 879, "dataset_name": "iwsl_t217"}}, "download_checksums": {"https://wit3.fbk.eu/archive/2017-01-trnted/texts/zh/en/zh-en.tgz": {"num_bytes": 26848340, "checksum": "8cecd4e8e196e5cd1e0af48b30e26c36acd33002d1e0f6e448570f68da290258"}}, "download_size": 26848340, "dataset_size": 46079276, "size_in_bytes": 72927616}}

dummy/iwslt2017-ar-en/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d72e7d76f4fcc1f7c65bff24dc0c601786239d40ea7a9ff9e223284bef441b91
+size 28959

dummy/iwslt2017-de-en/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6c937c25e1c1070f69aa92a768afc3b9925cd61ae95c64d9cb7412c0f1edf90
+size 26978

dummy/iwslt2017-en-ar/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a1a9cdba76ff0b24aa728d024a7fa6786144744773552a91afcdfb9f3a3498d
+size 28966

dummy/iwslt2017-en-de/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6803837cedf33387ab5dce24342f12c2c94ae44ea464010011721af74ecbe6a0
+size 26985

dummy/iwslt2017-en-fr/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e89da95a97f1d75be5d16ee5ef248f0c7796076455636252d2257038c0b48e84
+size 27005

dummy/iwslt2017-en-it/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09955c331c981409b0c297218c3c3fd99089441b7a9a1915cefce364d6d63b72
+size 9840

dummy/iwslt2017-en-ja/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ea1a70d4b24f366b5ab3e8c43f45e6ee56331e2bd06f25b6b18844816e14fac
+size 28841

dummy/iwslt2017-en-ko/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8a616ca969db95f8a7fc9abdda412d31852d141a970f58337de61a2c9feaf04
+size 27504

dummy/iwslt2017-en-nl/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89402df8d1d689cd96103f401220a0216cb7c248ee6e6a2d49bd0702d8b7cf98
+size 10254

dummy/iwslt2017-en-ro/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73ac95b8ca81c2b7411a564eb9663b41ecd669f058cd752bd297709f4a21691c
+size 10554

dummy/iwslt2017-en-zh/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8168cd4a0354a547377318f55308fe1760b04271f806bc31e0e7d31d444bdb06
+size 25303

dummy/iwslt2017-fr-en/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17a332aa25db413ada3b023be29274afcfa9933c9d39fdaf9be52775e8912bb6
+size 26998

dummy/iwslt2017-it-en/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3515f9f7731adfc434898b066b88e37d759bc5183d6c0ff00542616a183f124c
+size 9840

dummy/iwslt2017-it-nl/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4bf2341e18f6c01bc86c9c81dae6fb7b4ff569075926a6c9db849dc22ccd828
+size 10878

dummy/iwslt2017-it-ro/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7d625509ea821bf1064aba584217f82c678ca9a99618be5d452445f5e6d6c8cf
+size 11162

dummy/iwslt2017-ja-en/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86fe3c437924c9dadf344ae7a5128f4b9d13f14df838efd16261835c40f6b4de
+size 28848

dummy/iwslt2017-ko-en/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51875ac35365b50a42403ce96c89ae49c9f9f38c0e57d0f8b5ca42203c983a7b
+size 27497

dummy/iwslt2017-nl-en/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91f2f280ea24171448b7eead64e6e17df0d9b3eb321c3071f190363d3836d68c
+size 10250

dummy/iwslt2017-nl-it/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae95c6ff98712312e208508ee3e7b1ff8a67795d0cfb023a6bef0084e59429aa
+size 10874

dummy/iwslt2017-nl-ro/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56f3a1aa7f7dcf24e5525bdfb15642ca5a8c8c985b14430a8b2f869eb752e4bf
+size 10922

dummy/iwslt2017-ro-en/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1adee956b58dbb3963688ce74a153596f66f3364302c25dd5279c8b2024d34a1
+size 10556

dummy/iwslt2017-ro-it/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6932f9e3a4b7d671f1a04a3d3da9637ddb00840c4c5cadcbfbbf5f2cd6b11b18
+size 11164

dummy/iwslt2017-ro-nl/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a33a2eaa429bf7878c90bce3b5857c1d0dd7bd71481312a551f1e115d6bbb980
+size 10928

dummy/iwslt2017-zh-en/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:738c269897145efef7686164c74d870748e8619af1224f3ff45478efdfcf7626
+size 25303

iwslt2017.py

@@ -0,0 +1,214 @@
+# coding=utf-8
+# Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""IWSLT 2017 dataset """
+
+from __future__ import absolute_import, division, print_function
+
+import os
+
+import datasets
+
+
+_CITATION = """\
+@inproceedings{cettoloEtAl:EAMT2012,
+Address = {Trento, Italy},
+Author = {Mauro Cettolo and Christian Girardi and Marcello Federico},
+Booktitle = {Proceedings of the 16$^{th}$ Conference of the European Association for Machine Translation (EAMT)},
+Date = {28-30},
+Month = {May},
+Pages = {261--268},
+Title = {WIT$^3$: Web Inventory of Transcribed and Translated Talks},
+Year = {2012}}
+"""
+
+_DESCRIPTION = """\
+The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task. The languages involved are five:
+
+  German, English, Italian, Dutch, Romanian.
+
+For each language pair, training and development sets are available through the entry of the table below: by clicking, an archive will be downloaded which contains the sets and a README file. Numbers in the table refer to millions of units (untokenized words) of the target side of all parallel training sets.
+"""
+
+MULTI_URL = "https://wit3.fbk.eu/archive/2017-01-trnmted//texts/DeEnItNlRo/DeEnItNlRo/DeEnItNlRo-DeEnItNlRo.tgz"
+
+
+class IWSLT2017Config(datasets.BuilderConfig):
+    """ BuilderConfig for NewDataset"""
+
+    def __init__(self, pair, is_multilingual, **kwargs):
+        """
+
+        Args:
+            pair: the language pair to consider
+            is_multilingual: Is this pair in the multilingual dataset (download source is different)
+            **kwargs: keyword arguments forwarded to super.
+        """
+        self.pair = pair
+        self.is_multilingual = is_multilingual
+        super().__init__(**kwargs)
+
+
+# XXX: Artificially removed DE from here, as it also exists within bilingual data
+MULTI_LANGUAGES = ["en", "it", "nl", "ro"]
+BI_LANGUAGES = ["ar", "de", "en", "fr", "ja", "ko", "zh"]
+MULTI_PAIRS = [f"{source}-{target}" for source in MULTI_LANGUAGES for target in MULTI_LANGUAGES if source != target]
+BI_PAIRS = [
+    f"{source}-{target}"
+    for source in BI_LANGUAGES
+    for target in BI_LANGUAGES
+    if source != target and (source == "en" or target == "en")
+]
+
+PAIRS = MULTI_PAIRS + BI_PAIRS
+
+
+class IWSLT217(datasets.GeneratorBasedBuilder):
+    """The IWSLT 2017 Evaluation Campaign includes a multilingual TED Talks MT task."""
+
+    VERSION = datasets.Version("1.0.0")
+
+    # This is an example of a dataset with multiple configurations.
+    # If you don't want/need to define several sub-sets in your dataset,
+    # just remove the BUILDER_CONFIG_CLASS and the BUILDER_CONFIGS attributes.
+    BUILDER_CONFIG_CLASS = IWSLT2017Config
+    BUILDER_CONFIGS = [
+        IWSLT2017Config(
+            name="iwslt2017-" + pair,
+            description="A small dataset",
+            version=datasets.Version("1.0.0"),
+            pair=pair,
+            is_multilingual=pair in MULTI_PAIRS,
+        )
+        for pair in PAIRS
+    ]
+
+    def _info(self):
+        return datasets.DatasetInfo(
+            # This is the description that will appear on the datasets page.
+            description=_DESCRIPTION,
+            # datasets.features.FeatureConnectors
+            features=datasets.Features(
+                {"translation": datasets.features.Translation(languages=self.config.pair.split("-"))}
+            ),
+            # If there's a common (input, target) tuple from the features,
+            # specify them here. They'll be used if as_supervised=True in
+            # builder.as_dataset.
+            supervised_keys=None,
+            # Homepage of the dataset for documentation
+            homepage="https://sites.google.com/site/iwsltevaluation2017/TED-tasks",
+            citation=_CITATION,
+        )
+
+    def _split_generators(self, dl_manager):
+        """Returns SplitGenerators."""
+        source, target = self.config.pair.split("-")
+        if self.config.is_multilingual:
+            dl_dir = dl_manager.download_and_extract(MULTI_URL)
+            data_dir = os.path.join(dl_dir, "DeEnItNlRo-DeEnItNlRo")
+            years = [2010]
+        else:
+            bi_url = f"https://wit3.fbk.eu/archive/2017-01-trnted/texts/{source}/{target}/{source}-{target}.tgz"
+            dl_dir = dl_manager.download_and_extract(bi_url)
+            data_dir = os.path.join(dl_dir, f"{source}-{target}")
+            years = [2010, 2011, 2012, 2013, 2014, 2015]
+        return [
+            datasets.SplitGenerator(
+                name=datasets.Split.TRAIN,
+                # These kwargs will be passed to _generate_examples
+                gen_kwargs={
+                    "source_files": [
+                        os.path.join(
+                            data_dir,
+                            "train.tags.{}.{}".format(self.config.pair, source),
+                        )
+                    ],
+                    "target_files": [
+                        os.path.join(
+                            data_dir,
+                            "train.tags.{}.{}".format(self.config.pair, target),
+                        )
+                    ],
+                    "split": "train",
+                },
+            ),
+            datasets.SplitGenerator(
+                name=datasets.Split.TEST,
+                # These kwargs will be passed to _generate_examples
+                gen_kwargs={
+                    "source_files": [
+                        os.path.join(
+                            data_dir,
+                            "IWSLT17.TED.tst{}.{}.{}.xml".format(year, self.config.pair, source),
+                        )
+                        for year in years
+                    ],
+                    "target_files": [
+                        os.path.join(
+                            data_dir,
+                            "IWSLT17.TED.tst{}.{}.{}.xml".format(year, self.config.pair, target),
+                        )
+                        for year in years
+                    ],
+                    "split": "test",
+                },
+            ),
+            datasets.SplitGenerator(
+                name=datasets.Split.VALIDATION,
+                # These kwargs will be passed to _generate_examples
+                gen_kwargs={
+                    "source_files": [
+                        os.path.join(
+                            data_dir,
+                            "IWSLT17.TED.dev2010.{}.{}.xml".format(self.config.pair, source),
+                        )
+                    ],
+                    "target_files": [
+                        os.path.join(
+                            data_dir,
+                            "IWSLT17.TED.dev2010.{}.{}.xml".format(self.config.pair, target),
+                        )
+                    ],
+                    "split": "dev",
+                },
+            ),
+        ]
+
+    def _generate_examples(self, source_files, target_files, split):
+        """ Yields examples. """
+        id_ = 0
+        source, target = self.config.pair.split("-")
+        for source_file, target_file in zip(source_files, target_files):
+            with open(source_file, "r", encoding="utf-8") as sf:
+                with open(target_file, "r", encoding="utf-8") as tf:
+                    for source_row, target_row in zip(sf, tf):
+                        source_row = source_row.strip()
+                        target_row = target_row.strip()
+
+                        if source_row.startswith("<"):
+                            if source_row.startswith("<seg"):
+                                # Remove <seg id="1">.....</seg>
+                                # Very simple code instead of regex or xml parsing
+                                part1 = source_row.split(">")[1]
+                                source_row = part1.split("<")[0]
+                                part1 = target_row.split(">")[1]
+                                target_row = part1.split("<")[0]
+
+                                source_row = source_row.strip()
+                                target_row = target_row.strip()
+                            else:
+                                continue
+
+                        yield id_, {"translation": {source: source_row, target: target_row}}
+                        id_ += 1