Datasets:
ted_hrlr

- {"az_to_en": {"description": "Data sets derived from TED talk transcripts for comparing similar language pairs\nwhere one is high resource and the other is low resource.\n", "citation": "@inproceedings{Ye2018WordEmbeddings,\n author = {Ye, Qi and Devendra, Sachan and Matthieu, Felix and Sarguna, Padmanabhan and Graham, Neubig},\n title = {When and Why are pre-trained word embeddings useful for Neural Machine Translation},\n booktitle = {HLT-NAACL},\n year = {2018},\n }\n", "homepage": "https://github.com/neulab/word-embeddings-for-nmt", "license": "", "features": {"translation": {"languages": ["az", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": {"input": "az", "output": "en"}, "builder_name": "ted_hrlr", "config_name": "az_to_en", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 186540, "num_examples": 904, "dataset_name": "ted_hrlr"}, "train": {"name": "train", "num_bytes": 1226853, "num_examples": 5947, "dataset_name": "ted_hrlr"}, "validation": {"name": "validation", "num_bytes": 122709, "num_examples": 672, "dataset_name": "ted_hrlr"}}, "download_checksums": {"http://www.phontron.com/data/qi18naacl-dataset.tar.gz": {"num_bytes": 131005909, "checksum": "216a86c3df4d4f522856fe9b920ff5be6b394d769cc88974ae8f9f5546953bbc"}}, "download_size": 131005909, "dataset_size": 1536102, "size_in_bytes": 132542011}, "aztr_to_en": {"description": "Data sets derived from TED talk transcripts for comparing similar language pairs\nwhere one is high resource and the other is low resource.\n", "citation": "@inproceedings{Ye2018WordEmbeddings,\n author = {Ye, Qi and Devendra, Sachan and Matthieu, Felix and Sarguna, Padmanabhan and Graham, Neubig},\n title = {When and Why are pre-trained word embeddings useful for Neural Machine Translation},\n booktitle = {HLT-NAACL},\n year = {2018},\n }\n", "homepage": "https://github.com/neulab/word-embeddings-for-nmt", "license": "", "features": {"translation": {"languages": ["az_tr", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": {"input": "az_tr", "output": "en"}, "builder_name": "ted_hrlr", "config_name": "aztr_to_en", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 186540, "num_examples": 904, "dataset_name": "ted_hrlr"}, "train": {"name": "train", "num_bytes": 39834469, "num_examples": 188397, "dataset_name": "ted_hrlr"}, "validation": {"name": "validation", "num_bytes": 122709, "num_examples": 672, "dataset_name": "ted_hrlr"}}, "download_checksums": {"http://www.phontron.com/data/qi18naacl-dataset.tar.gz": {"num_bytes": 131005909, "checksum": "216a86c3df4d4f522856fe9b920ff5be6b394d769cc88974ae8f9f5546953bbc"}}, "download_size": 131005909, "dataset_size": 40143718, "size_in_bytes": 171149627}, "be_to_en": {"description": "Data sets derived from TED talk transcripts for comparing similar language pairs\nwhere one is high resource and the other is low resource.\n", "citation": "@inproceedings{Ye2018WordEmbeddings,\n author = {Ye, Qi and Devendra, Sachan and Matthieu, Felix and Sarguna, Padmanabhan and Graham, Neubig},\n title = {When and Why are pre-trained word embeddings useful for Neural Machine Translation},\n booktitle = {HLT-NAACL},\n year = {2018},\n }\n", "homepage": "https://github.com/neulab/word-embeddings-for-nmt", "license": "", "features": {"translation": {"languages": ["be", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": {"input": "be", "output": "en"}, "builder_name": "ted_hrlr", "config_name": "be_to_en", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 186606, "num_examples": 665, "dataset_name": "ted_hrlr"}, "train": {"name": "train", "num_bytes": 1176899, "num_examples": 4510, "dataset_name": "ted_hrlr"}, "validation": {"name": "validation", "num_bytes": 59328, "num_examples": 249, "dataset_name": "ted_hrlr"}}, "download_checksums": {"http://www.phontron.com/data/qi18naacl-dataset.tar.gz": {"num_bytes": 131005909, "checksum": "216a86c3df4d4f522856fe9b920ff5be6b394d769cc88974ae8f9f5546953bbc"}}, "download_size": 131005909, "dataset_size": 1422833, "size_in_bytes": 132428742}, "beru_to_en": {"description": "Data sets derived from TED talk transcripts for comparing similar language pairs\nwhere one is high resource and the other is low resource.\n", "citation": "@inproceedings{Ye2018WordEmbeddings,\n author = {Ye, Qi and Devendra, Sachan and Matthieu, Felix and Sarguna, Padmanabhan and Graham, Neubig},\n title = {When and Why are pre-trained word embeddings useful for Neural Machine Translation},\n booktitle = {HLT-NAACL},\n year = {2018},\n }\n", "homepage": "https://github.com/neulab/word-embeddings-for-nmt", "license": "", "features": {"translation": {"languages": ["be_ru", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": {"input": "be_ru", "output": "en"}, "builder_name": "ted_hrlr", "config_name": "beru_to_en", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 186606, "num_examples": 665, "dataset_name": "ted_hrlr"}, "train": {"name": "train", "num_bytes": 59953616, "num_examples": 212615, "dataset_name": "ted_hrlr"}, "validation": {"name": "validation", "num_bytes": 59328, "num_examples": 249, "dataset_name": "ted_hrlr"}}, "download_checksums": {"http://www.phontron.com/data/qi18naacl-dataset.tar.gz": {"num_bytes": 131005909, "checksum": "216a86c3df4d4f522856fe9b920ff5be6b394d769cc88974ae8f9f5546953bbc"}}, "download_size": 131005909, "dataset_size": 60199550, "size_in_bytes": 191205459}, "es_to_pt": {"description": "Data sets derived from TED talk transcripts for comparing similar language pairs\nwhere one is high resource and the other is low resource.\n", "citation": "@inproceedings{Ye2018WordEmbeddings,\n author = {Ye, Qi and Devendra, Sachan and Matthieu, Felix and Sarguna, Padmanabhan and Graham, Neubig},\n title = {When and Why are pre-trained word embeddings useful for Neural Machine Translation},\n booktitle = {HLT-NAACL},\n year = {2018},\n }\n", "homepage": "https://github.com/neulab/word-embeddings-for-nmt", "license": "", "features": {"translation": {"languages": ["es", "pt"], "id": null, "_type": "Translation"}}, "supervised_keys": {"input": "es", "output": "pt"}, "builder_name": "ted_hrlr", "config_name": "es_to_pt", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 343640, "num_examples": 1764, "dataset_name": "ted_hrlr"}, "train": {"name": "train", "num_bytes": 8611393, "num_examples": 44939, "dataset_name": "ted_hrlr"}, "validation": {"name": "validation", "num_bytes": 181535, "num_examples": 1017, "dataset_name": "ted_hrlr"}}, "download_checksums": {"http://www.phontron.com/data/qi18naacl-dataset.tar.gz": {"num_bytes": 131005909, "checksum": "216a86c3df4d4f522856fe9b920ff5be6b394d769cc88974ae8f9f5546953bbc"}}, "download_size": 131005909, "dataset_size": 9136568, "size_in_bytes": 140142477}, "fr_to_pt": {"description": "Data sets derived from TED talk transcripts for comparing similar language pairs\nwhere one is high resource and the other is low resource.\n", "citation": "@inproceedings{Ye2018WordEmbeddings,\n author = {Ye, Qi and Devendra, Sachan and Matthieu, Felix and Sarguna, Padmanabhan and Graham, Neubig},\n title = {When and Why are pre-trained word embeddings useful for Neural Machine Translation},\n booktitle = {HLT-NAACL},\n year = {2018},\n }\n", "homepage": "https://github.com/neulab/word-embeddings-for-nmt", "license": "", "features": {"translation": {"languages": ["fr", "pt"], "id": null, "_type": "Translation"}}, "supervised_keys": {"input": "fr", "output": "pt"}, "builder_name": "ted_hrlr", "config_name": "fr_to_pt", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 311650, "num_examples": 1495, "dataset_name": "ted_hrlr"}, "train": {"name": "train", "num_bytes": 8755387, "num_examples": 43874, "dataset_name": "ted_hrlr"}, "validation": {"name": "validation", "num_bytes": 212317, "num_examples": 1132, "dataset_name": "ted_hrlr"}}, "download_checksums": {"http://www.phontron.com/data/qi18naacl-dataset.tar.gz": {"num_bytes": 131005909, "checksum": "216a86c3df4d4f522856fe9b920ff5be6b394d769cc88974ae8f9f5546953bbc"}}, "download_size": 131005909, "dataset_size": 9279354, "size_in_bytes": 140285263}, "gl_to_en": {"description": "Data sets derived from TED talk transcripts for comparing similar language pairs\nwhere one is high resource and the other is low resource.\n", "citation": "@inproceedings{Ye2018WordEmbeddings,\n author = {Ye, Qi and Devendra, Sachan and Matthieu, Felix and Sarguna, Padmanabhan and Graham, Neubig},\n title = {When and Why are pre-trained word embeddings useful for Neural Machine Translation},\n booktitle = {HLT-NAACL},\n year = {2018},\n }\n", "homepage": "https://github.com/neulab/word-embeddings-for-nmt", "license": "", "features": {"translation": {"languages": ["gl", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": {"input": "gl", "output": "en"}, "builder_name": "ted_hrlr", "config_name": "gl_to_en", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 193213, "num_examples": 1008, "dataset_name": "ted_hrlr"}, "train": {"name": "train", "num_bytes": 1961363, "num_examples": 10018, "dataset_name": "ted_hrlr"}, "validation": {"name": "validation", "num_bytes": 137929, "num_examples": 683, "dataset_name": "ted_hrlr"}}, "download_checksums": {"http://www.phontron.com/data/qi18naacl-dataset.tar.gz": {"num_bytes": 131005909, "checksum": "216a86c3df4d4f522856fe9b920ff5be6b394d769cc88974ae8f9f5546953bbc"}}, "download_size": 131005909, "dataset_size": 2292505, "size_in_bytes": 133298414}, "glpt_to_en": {"description": "Data sets derived from TED talk transcripts for comparing similar language pairs\nwhere one is high resource and the other is low resource.\n", "citation": "@inproceedings{Ye2018WordEmbeddings,\n author = {Ye, Qi and Devendra, Sachan and Matthieu, Felix and Sarguna, Padmanabhan and Graham, Neubig},\n title = {When and Why are pre-trained word embeddings useful for Neural Machine Translation},\n booktitle = {HLT-NAACL},\n year = {2018},\n }\n", "homepage": "https://github.com/neulab/word-embeddings-for-nmt", "license": "", "features": {"translation": {"languages": ["gl_pt", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": {"input": "gl_pt", "output": "en"}, "builder_name": "ted_hrlr", "config_name": "glpt_to_en", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 193213, "num_examples": 1008, "dataset_name": "ted_hrlr"}, "train": {"name": "train", "num_bytes": 11734254, "num_examples": 61803, "dataset_name": "ted_hrlr"}, "validation": {"name": "validation", "num_bytes": 137929, "num_examples": 683, "dataset_name": "ted_hrlr"}}, "download_checksums": {"http://www.phontron.com/data/qi18naacl-dataset.tar.gz": {"num_bytes": 131005909, "checksum": "216a86c3df4d4f522856fe9b920ff5be6b394d769cc88974ae8f9f5546953bbc"}}, "download_size": 131005909, "dataset_size": 12065396, "size_in_bytes": 143071305}, "he_to_pt": {"description": "Data sets derived from TED talk transcripts for comparing similar language pairs\nwhere one is high resource and the other is low resource.\n", "citation": "@inproceedings{Ye2018WordEmbeddings,\n author = {Ye, Qi and Devendra, Sachan and Matthieu, Felix and Sarguna, Padmanabhan and Graham, Neubig},\n title = {When and Why are pre-trained word embeddings useful for Neural Machine Translation},\n booktitle = {HLT-NAACL},\n year = {2018},\n }\n", "homepage": "https://github.com/neulab/word-embeddings-for-nmt", "license": "", "features": {"translation": {"languages": ["he", "pt"], "id": null, "_type": "Translation"}}, "supervised_keys": {"input": "he", "output": "pt"}, "builder_name": "ted_hrlr", "config_name": "he_to_pt", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 361378, "num_examples": 1624, "dataset_name": "ted_hrlr"}, "train": {"name": "train", "num_bytes": 10627615, "num_examples": 48512, "dataset_name": "ted_hrlr"}, "validation": {"name": "validation", "num_bytes": 230725, "num_examples": 1146, "dataset_name": "ted_hrlr"}}, "download_checksums": {"http://www.phontron.com/data/qi18naacl-dataset.tar.gz": {"num_bytes": 131005909, "checksum": "216a86c3df4d4f522856fe9b920ff5be6b394d769cc88974ae8f9f5546953bbc"}}, "download_size": 131005909, "dataset_size": 11219718, "size_in_bytes": 142225627}, "it_to_pt": {"description": "Data sets derived from TED talk transcripts for comparing similar language pairs\nwhere one is high resource and the other is low resource.\n", "citation": "@inproceedings{Ye2018WordEmbeddings,\n author = {Ye, Qi and Devendra, Sachan and Matthieu, Felix and Sarguna, Padmanabhan and Graham, Neubig},\n title = {When and Why are pre-trained word embeddings useful for Neural Machine Translation},\n booktitle = {HLT-NAACL},\n year = {2018},\n }\n", "homepage": "https://github.com/neulab/word-embeddings-for-nmt", "license": "", "features": {"translation": {"languages": ["it", "pt"], "id": null, "_type": "Translation"}}, "supervised_keys": {"input": "it", "output": "pt"}, "builder_name": "ted_hrlr", "config_name": "it_to_pt", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 324726, "num_examples": 1670, "dataset_name": "ted_hrlr"}, "train": {"name": "train", "num_bytes": 8905825, "num_examples": 46260, "dataset_name": "ted_hrlr"}, "validation": {"name": "validation", "num_bytes": 210375, "num_examples": 1163, "dataset_name": "ted_hrlr"}}, "download_checksums": {"http://www.phontron.com/data/qi18naacl-dataset.tar.gz": {"num_bytes": 131005909, "checksum": "216a86c3df4d4f522856fe9b920ff5be6b394d769cc88974ae8f9f5546953bbc"}}, "download_size": 131005909, "dataset_size": 9440926, "size_in_bytes": 140446835}, "pt_to_en": {"description": "Data sets derived from TED talk transcripts for comparing similar language pairs\nwhere one is high resource and the other is low resource.\n", "citation": "@inproceedings{Ye2018WordEmbeddings,\n author = {Ye, Qi and Devendra, Sachan and Matthieu, Felix and Sarguna, Padmanabhan and Graham, Neubig},\n title = {When and Why are pre-trained word embeddings useful for Neural Machine Translation},\n booktitle = {HLT-NAACL},\n year = {2018},\n }\n", "homepage": "https://github.com/neulab/word-embeddings-for-nmt", "license": "", "features": {"translation": {"languages": ["pt", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": {"input": "pt", "output": "en"}, "builder_name": "ted_hrlr", "config_name": "pt_to_en", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 347803, "num_examples": 1804, "dataset_name": "ted_hrlr"}, "train": {"name": "train", "num_bytes": 9772911, "num_examples": 51786, "dataset_name": "ted_hrlr"}, "validation": {"name": "validation", "num_bytes": 207960, "num_examples": 1194, "dataset_name": "ted_hrlr"}}, "download_checksums": {"http://www.phontron.com/data/qi18naacl-dataset.tar.gz": {"num_bytes": 131005909, "checksum": "216a86c3df4d4f522856fe9b920ff5be6b394d769cc88974ae8f9f5546953bbc"}}, "download_size": 131005909, "dataset_size": 10328674, "size_in_bytes": 141334583}, "ru_to_en": {"description": "Data sets derived from TED talk transcripts for comparing similar language pairs\nwhere one is high resource and the other is low resource.\n", "citation": "@inproceedings{Ye2018WordEmbeddings,\n author = {Ye, Qi and Devendra, Sachan and Matthieu, Felix and Sarguna, Padmanabhan and Graham, Neubig},\n title = {When and Why are pre-trained word embeddings useful for Neural Machine Translation},\n booktitle = {HLT-NAACL},\n year = {2018},\n }\n", "homepage": "https://github.com/neulab/word-embeddings-for-nmt", "license": "", "features": {"translation": {"languages": ["ru", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": {"input": "ru", "output": "en"}, "builder_name": "ted_hrlr", "config_name": "ru_to_en", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 1459576, "num_examples": 5477, "dataset_name": "ted_hrlr"}, "train": {"name": "train", "num_bytes": 58778442, "num_examples": 208107, "dataset_name": "ted_hrlr"}, "validation": {"name": "validation", "num_bytes": 1318357, "num_examples": 4806, "dataset_name": "ted_hrlr"}}, "download_checksums": {"http://www.phontron.com/data/qi18naacl-dataset.tar.gz": {"num_bytes": 131005909, "checksum": "216a86c3df4d4f522856fe9b920ff5be6b394d769cc88974ae8f9f5546953bbc"}}, "download_size": 131005909, "dataset_size": 61556375, "size_in_bytes": 192562284}, "ru_to_pt": {"description": "Data sets derived from TED talk transcripts for comparing similar language pairs\nwhere one is high resource and the other is low resource.\n", "citation": "@inproceedings{Ye2018WordEmbeddings,\n author = {Ye, Qi and Devendra, Sachan and Matthieu, Felix and Sarguna, Padmanabhan and Graham, Neubig},\n title = {When and Why are pre-trained word embeddings useful for Neural Machine Translation},\n booktitle = {HLT-NAACL},\n year = {2018},\n }\n", "homepage": "https://github.com/neulab/word-embeddings-for-nmt", "license": "", "features": {"translation": {"languages": ["ru", "pt"], "id": null, "_type": "Translation"}}, "supervised_keys": {"input": "ru", "output": "pt"}, "builder_name": "ted_hrlr", "config_name": "ru_to_pt", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 409062, "num_examples": 1589, "dataset_name": "ted_hrlr"}, "train": {"name": "train", "num_bytes": 11882860, "num_examples": 47279, "dataset_name": "ted_hrlr"}, "validation": {"name": "validation", "num_bytes": 276866, "num_examples": 1185, "dataset_name": "ted_hrlr"}}, "download_checksums": {"http://www.phontron.com/data/qi18naacl-dataset.tar.gz": {"num_bytes": 131005909, "checksum": "216a86c3df4d4f522856fe9b920ff5be6b394d769cc88974ae8f9f5546953bbc"}}, "download_size": 131005909, "dataset_size": 12568788, "size_in_bytes": 143574697}, "tr_to_en": {"description": "Data sets derived from TED talk transcripts for comparing similar language pairs\nwhere one is high resource and the other is low resource.\n", "citation": "@inproceedings{Ye2018WordEmbeddings,\n author = {Ye, Qi and Devendra, Sachan and Matthieu, Felix and Sarguna, Padmanabhan and Graham, Neubig},\n title = {When and Why are pre-trained word embeddings useful for Neural Machine Translation},\n booktitle = {HLT-NAACL},\n year = {2018},\n }\n", "homepage": "https://github.com/neulab/word-embeddings-for-nmt", "license": "", "features": {"translation": {"languages": ["tr", "en"], "id": null, "_type": "Translation"}}, "supervised_keys": {"input": "tr", "output": "en"}, "builder_name": "ted_hrlr", "config_name": "tr_to_en", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 1026406, "num_examples": 5030, "dataset_name": "ted_hrlr"}, "train": {"name": "train", "num_bytes": 38607636, "num_examples": 182451, "dataset_name": "ted_hrlr"}, "validation": {"name": "validation", "num_bytes": 832358, "num_examples": 4046, "dataset_name": "ted_hrlr"}}, "download_checksums": {"http://www.phontron.com/data/qi18naacl-dataset.tar.gz": {"num_bytes": 131005909, "checksum": "216a86c3df4d4f522856fe9b920ff5be6b394d769cc88974ae8f9f5546953bbc"}}, "download_size": 131005909, "dataset_size": 40466400, "size_in_bytes": 171472309}}