Convert dataset to Parquet (#4)

- Convert dataset to Parquet (8539110e6f1adefe5f3325a47bdbb1d2d5b5a485)
- Add discofuse-wikipedia data files (f4a7713cd46f6331636d0f817a448e6e1ac2c487)
- Delete loading script (22d781962d6a36c07f4e4f04a5e1c4733d8b8873)
- Delete legacy dataset_infos.json (f2319c388343f2a1d75aa75313ae5493464a4334)

README.md

@@ -1,15 +1,14 @@
 ---
 annotations_creators:
 - machine-generated
-language:
-- en
 language_creators:
 - found
+language:
+- en
 license:
 - cc-by-sa-3.0
 multilinguality:
 - monolingual
-pretty_name: DiscoFuse
 size_categories:
 - 10M<n<100M
 source_datasets:
@@ -18,6 +17,7 @@ task_categories:
 - text2text-generation
 task_ids: []
 paperswithcode_id: discofuse
+pretty_name: DiscoFuse
 tags:
 - sentence-fusion
 dataset_info:
@@ -41,16 +41,16 @@ dataset_info:
     dtype: string
   splits:
   - name: train
-    num_bytes: 14736279993
+    num_bytes: 14736176073
     num_examples: 43291020
   - name: test
-    num_bytes: 151656323
+    num_bytes: 151655243
     num_examples: 445521
   - name: validation
-    num_bytes: 150207737
+    num_bytes: 150206657
     num_examples: 440902
-  download_size: 4326637746
-  dataset_size: 15038144053
+  download_size: 9422142544
+  dataset_size: 15038037973
 - config_name: discofuse-wikipedia
   features:
   - name: connective_string
@@ -71,16 +71,33 @@ dataset_info:
     dtype: string
   splits:
   - name: train
-    num_bytes: 6377924196
+    num_bytes: 6377885028
     num_examples: 16310585
   - name: test
-    num_bytes: 64008158
+    num_bytes: 64007750
     num_examples: 163657
   - name: validation
-    num_bytes: 65682035
+    num_bytes: 65681627
     num_examples: 168081
-  download_size: 1717422334
-  dataset_size: 6507614389
+  download_size: 3929336540
+  dataset_size: 6507574405
+configs:
+- config_name: discofuse-sport
+  data_files:
+  - split: train
+    path: discofuse-sport/train-*
+  - split: test
+    path: discofuse-sport/test-*
+  - split: validation
+    path: discofuse-sport/validation-*
+- config_name: discofuse-wikipedia
+  data_files:
+  - split: train
+    path: discofuse-wikipedia/train-*
+  - split: test
+    path: discofuse-wikipedia/test-*
+  - split: validation
+    path: discofuse-wikipedia/validation-*
 ---
 
 # Dataset Card for "discofuse"

dataset_infos.json

@@ -1 +0,0 @@
-{"discofuse-sport": {"description": " DISCOFUSE is a large scale dataset for discourse-based sentence fusion.\n", "citation": "@InProceedings{GevaEtAl2019,\n  title = {DiscoFuse: A Large-Scale Dataset for Discourse-Based Sentence Fusion},\n  author = {Geva, Mor and Malmi, Eric and Szpektor, Idan and Berant, Jonathan},\n  booktitle = {Proceedings of the 2019 Annual Conference of the North American Chapter of the Association for Computational Linguistics},\n  note = {arXiv preprint arXiv:1902.10526},\n  year = {2019}\n}\n\n", "homepage": "https://github.com/google-research-datasets/discofuse", "license": "", "features": {"connective_string": {"dtype": "string", "id": null, "_type": "Value"}, "discourse_type": {"dtype": "string", "id": null, "_type": "Value"}, "coherent_second_sentence": {"dtype": "string", "id": null, "_type": "Value"}, "has_coref_type_pronoun": {"dtype": "float32", "id": null, "_type": "Value"}, "incoherent_first_sentence": {"dtype": "string", "id": null, "_type": "Value"}, "incoherent_second_sentence": {"dtype": "string", "id": null, "_type": "Value"}, "has_coref_type_nominal": {"dtype": "float32", "id": null, "_type": "Value"}, "coherent_first_sentence": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "discofuse", "config_name": "discofuse-sport", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 14736279993, "num_examples": 43291020, "dataset_name": "discofuse"}, "test": {"name": "test", "num_bytes": 151656323, "num_examples": 445521, "dataset_name": "discofuse"}, "validation": {"name": "validation", "num_bytes": 150207737, "num_examples": 440902, "dataset_name": "discofuse"}}, "download_checksums": {"https://storage.googleapis.com/gresearch/discofuse/discofuse_v1_sports.zip": {"num_bytes": 4326637746, "checksum": "a390083c7923e11efeeea04a9a79074149e5ef9be614466f50aec28f1a5eec41"}}, "download_size": 4326637746, "post_processing_size": null, "dataset_size": 15038144053, "size_in_bytes": 19364781799}, "discofuse-wikipedia": {"description": " DISCOFUSE is a large scale dataset for discourse-based sentence fusion.\n", "citation": "@InProceedings{GevaEtAl2019,\n  title = {DiscoFuse: A Large-Scale Dataset for Discourse-Based Sentence Fusion},\n  author = {Geva, Mor and Malmi, Eric and Szpektor, Idan and Berant, Jonathan},\n  booktitle = {Proceedings of the 2019 Annual Conference of the North American Chapter of the Association for Computational Linguistics},\n  note = {arXiv preprint arXiv:1902.10526},\n  year = {2019}\n}\n\n", "homepage": "https://github.com/google-research-datasets/discofuse", "license": "", "features": {"connective_string": {"dtype": "string", "id": null, "_type": "Value"}, "discourse_type": {"dtype": "string", "id": null, "_type": "Value"}, "coherent_second_sentence": {"dtype": "string", "id": null, "_type": "Value"}, "has_coref_type_pronoun": {"dtype": "float32", "id": null, "_type": "Value"}, "incoherent_first_sentence": {"dtype": "string", "id": null, "_type": "Value"}, "incoherent_second_sentence": {"dtype": "string", "id": null, "_type": "Value"}, "has_coref_type_nominal": {"dtype": "float32", "id": null, "_type": "Value"}, "coherent_first_sentence": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "discofuse", "config_name": "discofuse-wikipedia", "version": {"version_str": "1.0.0", "description": "", "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 6377924196, "num_examples": 16310585, "dataset_name": "discofuse"}, "test": {"name": "test", "num_bytes": 64008158, "num_examples": 163657, "dataset_name": "discofuse"}, "validation": {"name": "validation", "num_bytes": 65682035, "num_examples": 168081, "dataset_name": "discofuse"}}, "download_checksums": {"https://storage.googleapis.com/gresearch/discofuse/discofuse_v1_wikipedia.zip": {"num_bytes": 1717422334, "checksum": "e8a5ec52cdd9820ce9b410b47c3a57a49a300470c976202cd7caab613658ebfe"}}, "download_size": 1717422334, "post_processing_size": null, "dataset_size": 6507614389, "size_in_bytes": 8225036723}}

discofuse-sport/test-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9183aace3727027a6d75eefc16a0704e91e2619c7c6d60d0b9266d1a793682c8
+size 94994467

discofuse-sport/train-00000-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a3d5f441e1e9440a600b01ac9fed6feda0ccaeb34f246c1e530eca5ee714427
+size 308126690

discofuse-sport/train-00001-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:832f1d483369a59516fbf97476bad6bfc63c7f063a6052236025b103a2e4e003
+size 305566158

discofuse-sport/train-00002-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f8aea2580843459fdcf70887b3cd35e32eaef98998a7dcbd4b664aaeb63202a
+size 307084313

discofuse-sport/train-00003-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d1c04bba7f326d4aca45a617166efc5857c5c26fafd14f06ee6b831c561e4ca
+size 308411269

discofuse-sport/train-00004-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e57430725c783061534001a1053bde61a30c3d9a75ca516fbeb7c56f3ff1acbe
+size 308394219

discofuse-sport/train-00005-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8116e965a5d4e61251a93089246eaad07844e8d4584278135236dd5514bba6e
+size 307987875

discofuse-sport/train-00006-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74707a3c885f618f1f4e7ba2b6b2bef599a6dd2cf6cd99fdd0b19d27a3bfb95a
+size 307391357

discofuse-sport/train-00007-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2adeb83ddff793dea972cc8c151c8ac1cee328fd9a95da57fab6856d49b7fd9
+size 308073498

discofuse-sport/train-00008-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5cfdfeb3c16c5682b60d672f901b05c87ea37f7c96b9c7015f332564a76e5ad1
+size 306444351

discofuse-sport/train-00009-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b6949ea9968bb292fb7a17d763c5f937816def0ca9f18d9db1342d2a3962ae7
+size 307844633

discofuse-sport/train-00010-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9bab2c216f601574da432e86be46ee54a6f2d0cce10c5dfa28c8f3d7c8cdab6
+size 307660360

discofuse-sport/train-00011-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d62754e6d50c6dbde00b0e874de3546c27c94b115ab0e49abc63506795ce158
+size 308278823

discofuse-sport/train-00012-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:726a39396b069281847c234b7271bc96ca7481942547f1cbc4b6c198d6451414
+size 308139806

discofuse-sport/train-00013-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f82faad524098b7a9ed94cc259a0caa5be5f00868597e481a74f8b12b28f5afb
+size 308261508

discofuse-sport/train-00014-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd44e9b92af8c408c10c030c35db0cbf7da00b3a1bb15935b2652cc3b715cb34
+size 308180069

discofuse-sport/train-00015-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:792aabdd3730c4e715ba017188105b6f8697e65c33ffb68349b0a10ff90c5284
+size 308447385

discofuse-sport/train-00016-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0fcdbd380e210e88d156071d65da4ac1d3c33ea0d25124f70915d9525c1ac43e
+size 307069435

discofuse-sport/train-00017-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3434df004d78eff99187032253287cb924d6b224b08021ac9beaa6960f06cefd
+size 306007238

discofuse-sport/train-00018-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dacc26d267b5dc1a178f9d83c27dcc72fc5f552b6a2dd821b8d9a04dc21c2d06
+size 308332171

discofuse-sport/train-00019-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9f694d0a92998a933b5bdf10dade891d2be026a416996a75fd123b367e384c2
+size 307019513

discofuse-sport/train-00020-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8cdfdabae8e3d33cd23bae749deb8b7a561cb65ff6bb2bad5648a20432647f34
+size 305922155

discofuse-sport/train-00021-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f5bff73bc9a14122d20f888454636ab73f4c904e6175e1d6f6dbab74e4782b9
+size 308234126

discofuse-sport/train-00022-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9afccf8fdbd9ccdb597df65ef2a2aaa3850de254a3f2794a4f5c1d318d1644dc
+size 307969671

discofuse-sport/train-00023-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9dd21930dfbcab6283bfca2ebd34edc106a5173484b25e6488ef51557cb63807
+size 308966119

discofuse-sport/train-00024-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab8ce0ae3b3893060a21ccf6cc4693ceac478d6227c6a613695ce09a9e939167
+size 308541880

discofuse-sport/train-00025-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82c84b9b9bd6bcdcd551b3943e9e59ed584a28d7681d436cc17f58354d347711
+size 308077997

discofuse-sport/train-00026-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c4a1df4097b8633f3b9ef267ab032757aa3e689194f362a45c7e4c1324c557b
+size 307562119

discofuse-sport/train-00027-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd759434a653d98043a1418ea4a81446f79611245ed63b3b79f5c262ce37f9d3
+size 307174197

discofuse-sport/train-00028-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09f812e111fa4ea491bd0ed93539d35ef88601daa653dec3108cf68a188f8588
+size 309604147

discofuse-sport/train-00029-of-00030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82f4c5a98311f03652293d5abaa72f19dad0891c2986e46503c4c20ce6f3cef5
+size 308296941

discofuse-sport/validation-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9892b7e1f60bb45e84633a9c123204a778b86a9698a2349561654fc7cd6321b6
+size 94078054

discofuse-wikipedia/test-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11be05ef93d439377e80d7045ea7e49e948c9b5db6e573018d7d694d81bba77f
+size 38687474

discofuse-wikipedia/train-00000-of-00013.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b30021c5c812644ed42c21dd5d972510cbd1a2f2d5d30c96deb00e5f38b100fe
+size 294960063

discofuse-wikipedia/train-00001-of-00013.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fe8882c85203c079411692407ae42c54ede9e7ff3f6d51c3f66441c2decfc47
+size 296884678

discofuse-wikipedia/train-00002-of-00013.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a222cea4f9b13e34a508d83da4a64cfe1b8c080b5f23b1627701fd86e0b1e12
+size 295549696

discofuse-wikipedia/train-00003-of-00013.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efc2501b52c9915acf8c646f1bf21b205647f66adeb464eae0855f574638da3f
+size 296199956

discofuse-wikipedia/train-00004-of-00013.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01bb21e87f6b1011984398261450874d0f013c83fa065c0666bf74c5b2bce966
+size 296117061

discofuse-wikipedia/train-00005-of-00013.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb84f132785f2ca2163a60d1322f47a1947a0f696bdadfece22cf46b1cb3670e
+size 296435505

discofuse-wikipedia/train-00006-of-00013.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d52c46977eefd77c9643b1a1d4eabd091f1f56b54a6489ae4491e0604ac9a6b
+size 296311233

discofuse-wikipedia/train-00007-of-00013.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0878b60331a698c4941053e5b3fc67762a417372b0c31e0993144526be0d7405
+size 296702371

discofuse-wikipedia/train-00008-of-00013.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:007ec0ee669a9ca7f00ebfcdf5ea0d8701e4ffed853992de8b78278892aaddde
+size 296235227

discofuse-wikipedia/train-00009-of-00013.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f6d932ca26b06fcbb8c95a1e60c6de5530772f88f50cee9e918926e4177793d
+size 296701954

discofuse-wikipedia/train-00010-of-00013.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fab8bf951b7170599d0a7de5f617e46e67542a2b35e0258b63f452ca774923c1
+size 296434475

discofuse-wikipedia/train-00011-of-00013.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea7cfa0f61672b6d7818a1110058984a041ad11dd31ca9894d16253f7947eb82
+size 296323688

discofuse-wikipedia/train-00012-of-00013.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98175d6c67d0afb7a993337c2dbb122f3a3ca2e21a89b1d32b64c35b82b900fa
+size 296146432

discofuse-wikipedia/validation-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07a2b6756c20327b202e58bb1a7975bfd4be4907d86f7b3c066f03b2bdca728a
+size 39646727

discofuse.py

@@ -1,194 +0,0 @@
-"""TODO(discofuse): Add a description here."""
-
-
-import csv
-import os
-
-import datasets
-
-
-_URL_ = "https://storage.googleapis.com/gresearch/discofuse/"
-_CITATION = """\
-@InProceedings{GevaEtAl2019,
-  title = {DiscoFuse: A Large-Scale Dataset for Discourse-Based Sentence Fusion},
-  author = {Geva, Mor and Malmi, Eric and Szpektor, Idan and Berant, Jonathan},
-  booktitle = {Proceedings of the 2019 Annual Conference of the North American Chapter of the Association for Computational Linguistics},
-  note = {arXiv preprint arXiv:1902.10526},
-  year = {2019}
-}
-
-"""
-
-# TODO(discofuse):
-_DESCRIPTION = """\
- DISCOFUSE is a large scale dataset for discourse-based sentence fusion.
-"""
-
-
-class DiscofuseConfig(datasets.BuilderConfig):
-
-    """BuilderConfig for Discofuse"""
-
-    def __init__(self, data_url, balanced=False, **kwargs):
-        """
-
-        Args:
-            balanced: to specify if we want to load the balanced file or the full file
-            **kwargs: keyword arguments forwarded to super.
-        """
-        super(DiscofuseConfig, self).__init__(version=datasets.Version("1.0.0", ""), **kwargs)
-        self.balanced = balanced
-        self.data_url = data_url
-
-
-class Discofuse(datasets.GeneratorBasedBuilder):
-    """TODO(discofuse): Short description of my dataset."""
-
-    # TODO(discofuse): Set up version.
-    VERSION = datasets.Version("1.0.0")
-    BUILDER_CONFIGS = [
-        DiscofuseConfig(
-            name="discofuse-sport", description="sentence fusion", data_url=_URL_ + "discofuse_v1_sports.zip"
-        ),
-        DiscofuseConfig(
-            name="discofuse-wikipedia", description="sentence fusion", data_url=_URL_ + "discofuse_v1_wikipedia.zip"
-        ),
-    ]
-
-    def _info(self):
-        # TODO(discofuse): Specifies the datasets.DatasetInfo object
-        return datasets.DatasetInfo(
-            # This is the description that will appear on the datasets page.
-            description=_DESCRIPTION,
-            # datasets.features.FeatureConnectors
-            features=datasets.Features(
-                {
-                    "connective_string": datasets.Value("string"),
-                    "discourse_type": datasets.Value("string"),
-                    "coherent_second_sentence": datasets.Value("string"),
-                    "has_coref_type_pronoun": datasets.Value("float32"),
-                    "incoherent_first_sentence": datasets.Value("string"),
-                    "incoherent_second_sentence": datasets.Value("string"),
-                    "has_coref_type_nominal": datasets.Value("float32"),
-                    "coherent_first_sentence": datasets.Value("string"),
-                    # These are the features of your dataset like images, labels ...
-                }
-            ),
-            # 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://github.com/google-research-datasets/discofuse",
-            citation=_CITATION,
-        )
-
-    def _split_generators(self, dl_manager):
-        """Returns SplitGenerators."""
-        # TODO(discofuse): Downloads the data and defines the splits
-        # dl_manager is a datasets.download.DownloadManager that can be used to
-        # download and extract URLs
-        if self.config.name == "discofuse-sport":
-            dl_dir = dl_manager.download_and_extract(self.config.data_url)
-            data_dir = os.path.join(dl_dir, "discofuse_v1/sports")
-            if self.config.balanced:
-                return [
-                    datasets.SplitGenerator(
-                        name=datasets.Split.TRAIN,
-                        # These kwargs will be passed to _generate_examples
-                        gen_kwargs={"filepath": os.path.join(data_dir, "train_balanced.tsv")},
-                    ),
-                    datasets.SplitGenerator(
-                        name=datasets.Split.TEST,
-                        # These kwargs will be passed to _generate_examples
-                        gen_kwargs={"filepath": os.path.join(data_dir, "test_balanced.tsv")},
-                    ),
-                    datasets.SplitGenerator(
-                        name=datasets.Split.VALIDATION,
-                        # These kwargs will be passed to _generate_examples
-                        gen_kwargs={"filepath": os.path.join(data_dir, "dev_balanced.tsv")},
-                    ),
-                ]
-            else:
-                return [
-                    datasets.SplitGenerator(
-                        name=datasets.Split.TRAIN,
-                        # These kwargs will be passed to _generate_examples
-                        gen_kwargs={"filepath": os.path.join(data_dir, "train.tsv")},
-                    ),
-                    datasets.SplitGenerator(
-                        name=datasets.Split.TEST,
-                        # These kwargs will be passed to _generate_examples
-                        gen_kwargs={"filepath": os.path.join(data_dir, "test.tsv")},
-                    ),
-                    datasets.SplitGenerator(
-                        name=datasets.Split.VALIDATION,
-                        # These kwargs will be passed to _generate_examples
-                        gen_kwargs={"filepath": os.path.join(data_dir, "dev.tsv")},
-                    ),
-                ]
-        else:
-            if self.config.name == "discofuse-wikipedia":
-                dl_dir = dl_manager.download_and_extract(self.config.data_url)
-                data_dir = os.path.join(dl_dir, "discofuse_v1/wikipedia")
-                if self.config.balanced:
-                    return [
-                        datasets.SplitGenerator(
-                            name=datasets.Split.TRAIN,
-                            # These kwargs will be passed to _generate_examples
-                            gen_kwargs={"filepath": os.path.join(data_dir, "train_balanced.tsv")},
-                        ),
-                        datasets.SplitGenerator(
-                            name=datasets.Split.TEST,
-                            # These kwargs will be passed to _generate_examples
-                            gen_kwargs={"filepath": os.path.join(data_dir, "test_balanced.tsv")},
-                        ),
-                        datasets.SplitGenerator(
-                            name=datasets.Split.VALIDATION,
-                            # These kwargs will be passed to _generate_examples
-                            gen_kwargs={"filepath": os.path.join(data_dir, "dev_balanced.tsv")},
-                        ),
-                    ]
-                else:
-                    return [
-                        datasets.SplitGenerator(
-                            name=datasets.Split.TRAIN,
-                            # These kwargs will be passed to _generate_examples
-                            gen_kwargs={"filepath": os.path.join(data_dir, "train.tsv")},
-                        ),
-                        datasets.SplitGenerator(
-                            name=datasets.Split.TEST,
-                            # These kwargs will be passed to _generate_examples
-                            gen_kwargs={"filepath": os.path.join(data_dir, "test.tsv")},
-                        ),
-                        datasets.SplitGenerator(
-                            name=datasets.Split.VALIDATION,
-                            # These kwargs will be passed to _generate_examples
-                            gen_kwargs={"filepath": os.path.join(data_dir, "dev.tsv")},
-                        ),
-                    ]
-
-    def _generate_examples(self, filepath):
-        """Yields examples."""
-        # TODO(discofuse): Yields (key, example) tuples from the dataset
-        with open(filepath, encoding="utf-8") as f:
-            data = csv.DictReader(f, delimiter="\t")
-            for id_, row in enumerate(data):
-                co_first_sent = row["coherent_first_sentence"]
-                co_second_sent = row["coherent_second_sentence"]
-                connect_str = row["connective_string"]
-                discourse_type = row["discourse_type"]
-                has_coref_pronoun = row["has_coref_type_pronoun"]
-                has_coref_nominal = row["has_coref_type_nominal"]
-                inco_first_sent = row["incoherent_first_sentence"]
-                inco_second_sent = row["incoherent_second_sentence"]
-                yield id_, {
-                    "connective_string": connect_str,
-                    "discourse_type": discourse_type,
-                    "coherent_second_sentence": co_second_sent,
-                    "has_coref_type_pronoun": has_coref_pronoun,
-                    "incoherent_first_sentence": inco_first_sent,
-                    "incoherent_second_sentence": inco_second_sent,
-                    "has_coref_type_nominal": has_coref_nominal,
-                    "coherent_first_sentence": co_first_sent,
-                }