Add loading script and dataset infos

dataset_infos.json

@@ -0,0 +1 @@
+{"de": {"description": "This is the MLSUM subset of the GEM benchmark. MLSUM is the first large-scale MultiLingual SUMmarization dataset.\nObtained from online newspapers, it contains 1.5M+ article/summary pairs in five different languages -- namely, French, German, Spanish, Russian, Turkish.\nTogether with English newspapers from the popular CNN/Daily mail dataset, the collected data form a large scale multilingual dataset which can enable new research directions for the text summarization community.\nWe report cross-lingual comparative analyses based on state-of-the-art systems.\nThese highlight existing biases which motivate the use of a multi-lingual dataset.\n", "citation": "@article{scialom2020mlsum,\n  title={MLSUM: The Multilingual Summarization Corpus},\n  author={Scialom, Thomas and Dray, Paul-Alexis and Lamprier, Sylvain and Piwowarski, Benjamin and Staiano, Jacopo},\n  journal={arXiv preprint arXiv:2004.14900},\n  year={2020}\n}\n", "homepage": "", "license": "", "features": {"gem_id": {"dtype": "string", "id": null, "_type": "Value"}, "gem_parent_id": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}, "topic": {"dtype": "string", "id": null, "_type": "Value"}, "url": {"dtype": "string", "id": null, "_type": "Value"}, "title": {"dtype": "string", "id": null, "_type": "Value"}, "date": {"dtype": "string", "id": null, "_type": "Value"}, "target": {"dtype": "string", "id": null, "_type": "Value"}, "references": [{"dtype": "string", "id": null, "_type": "Value"}]}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "mlsum", "config_name": "de", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 855411361, "num_examples": 220748, "dataset_name": "mlsum"}, "validation": {"name": "validation", "num_bytes": 49576087, "num_examples": 11392, "dataset_name": "mlsum"}, "test": {"name": "test", "num_bytes": 49018014, "num_examples": 10695, "dataset_name": "mlsum"}, "challenge_train_sample": {"name": "challenge_train_sample", "num_bytes": 1891220, "num_examples": 500, "dataset_name": "mlsum"}, "challenge_validation_sample": {"name": "challenge_validation_sample", "num_bytes": 2199723, "num_examples": 500, "dataset_name": "mlsum"}, "challenge_test_covid": {"name": "challenge_test_covid", "num_bytes": 19710589, "num_examples": 5058, "dataset_name": "mlsum"}}, "download_checksums": {"https://gitlab.lip6.fr/scialom/mlsum_data/-/raw/master/MLSUM/de_train.zip": {"num_bytes": 311059697, "checksum": "88e788437bae48af6b3d18a554af4b2794cc6143a137df3f56daa91a37e3ea7e"}, "https://gitlab.lip6.fr/scialom/mlsum_data/-/raw/master/MLSUM/de_val.zip": {"num_bytes": 17771216, "checksum": "732620c32e1d3f393ee3193f57f1217d8549499eb4906e144252aaab39aa910b"}, "https://gitlab.lip6.fr/scialom/mlsum_data/-/raw/master/MLSUM/de_test.zip": {"num_bytes": 17741147, "checksum": "447e3b1839ab94d5700cc2aedc0b52521404865b2589656acc90a654ed0de4ff"}, "https://storage.googleapis.com/huggingface-nlp/datasets/gem/gem_mlsum_bad_ids_fixed.json": {"num_bytes": 784429, "checksum": "7d1b5c340329da32b3a6c1b880e9d72b5193eb0782bc03261e4eaee08c3d5b64"}, "https://storage.googleapis.com/huggingface-nlp/datasets/gem/gem_challenge_sets/mlsum_de.zip": {"num_bytes": 15427039, "checksum": "7cc6751d7a76e833e0db27ce0b06a50be4df43dfd5d5284dd11888439a126310"}}, "download_size": 362783528, "post_processing_size": null, "dataset_size": 977806994, "size_in_bytes": 1340590522}, "es": {"description": "This is the MLSUM subset of the GEM benchmark. MLSUM is the first large-scale MultiLingual SUMmarization dataset.\nObtained from online newspapers, it contains 1.5M+ article/summary pairs in five different languages -- namely, French, German, Spanish, Russian, Turkish.\nTogether with English newspapers from the popular CNN/Daily mail dataset, the collected data form a large scale multilingual dataset which can enable new research directions for the text summarization community.\nWe report cross-lingual comparative analyses based on state-of-the-art systems.\nThese highlight existing biases which motivate the use of a multi-lingual dataset.\n", "citation": "@article{scialom2020mlsum,\n  title={MLSUM: The Multilingual Summarization Corpus},\n  author={Scialom, Thomas and Dray, Paul-Alexis and Lamprier, Sylvain and Piwowarski, Benjamin and Staiano, Jacopo},\n  journal={arXiv preprint arXiv:2004.14900},\n  year={2020}\n}\n", "homepage": "", "license": "", "features": {"gem_id": {"dtype": "string", "id": null, "_type": "Value"}, "gem_parent_id": {"dtype": "string", "id": null, "_type": "Value"}, "text": {"dtype": "string", "id": null, "_type": "Value"}, "topic": {"dtype": "string", "id": null, "_type": "Value"}, "url": {"dtype": "string", "id": null, "_type": "Value"}, "title": {"dtype": "string", "id": null, "_type": "Value"}, "date": {"dtype": "string", "id": null, "_type": "Value"}, "target": {"dtype": "string", "id": null, "_type": "Value"}, "references": [{"dtype": "string", "id": null, "_type": "Value"}]}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "mlsum", "config_name": "es", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 1208122300, "num_examples": 259888, "dataset_name": "mlsum"}, "validation": {"name": "validation", "num_bytes": 51491999, "num_examples": 9977, "dataset_name": "mlsum"}, "test": {"name": "test", "num_bytes": 71957172, "num_examples": 13366, "dataset_name": "mlsum"}, "challenge_train_sample": {"name": "challenge_train_sample", "num_bytes": 2363443, "num_examples": 500, "dataset_name": "mlsum"}, "challenge_validation_sample": {"name": "challenge_validation_sample", "num_bytes": 2655596, "num_examples": 500, "dataset_name": "mlsum"}, "challenge_test_covid": {"name": "challenge_test_covid", "num_bytes": 13553368, "num_examples": 1938, "dataset_name": "mlsum"}}, "download_checksums": {"https://gitlab.lip6.fr/scialom/mlsum_data/-/raw/master/MLSUM/es_train.zip": {"num_bytes": 466443036, "checksum": "a01f4b4b873aa6cdeae15952a22ede2146734d0b60e7297470a35956507c863a"}, "https://gitlab.lip6.fr/scialom/mlsum_data/-/raw/master/MLSUM/es_val.zip": {"num_bytes": 19483214, "checksum": "e38fce9950008ec4b48963692891c4c94d51a1e307286fb596e093aeb1230c92"}, "https://gitlab.lip6.fr/scialom/mlsum_data/-/raw/master/MLSUM/es_test.zip": {"num_bytes": 27386169, "checksum": "177cfcf358bc4aa9bce2753b8e9de4f6eb41d2c30b1a99ef29d64e70537a1c0d"}, "https://storage.googleapis.com/huggingface-nlp/datasets/gem/gem_mlsum_bad_ids_fixed.json": {"num_bytes": 784429, "checksum": "7d1b5c340329da32b3a6c1b880e9d72b5193eb0782bc03261e4eaee08c3d5b64"}, "https://storage.googleapis.com/huggingface-nlp/datasets/gem/gem_challenge_sets/mlsum_es.zip": {"num_bytes": 11524578, "checksum": "a254972fef695970aea0370be64fed6aec8c8b760f238fabd0e8f363bf8274cd"}}, "download_size": 525621426, "post_processing_size": null, "dataset_size": 1350143878, "size_in_bytes": 1875765304}}

mlsum.py

@@ -0,0 +1,159 @@
+import json
+import os
+
+import datasets
+
+_CITATION = """\
+@article{scialom2020mlsum,
+  title={MLSUM: The Multilingual Summarization Corpus},
+  author={Scialom, Thomas and Dray, Paul-Alexis and Lamprier, Sylvain and Piwowarski, Benjamin and Staiano, Jacopo},
+  journal={arXiv preprint arXiv:2004.14900},
+  year={2020}
+}
+"""
+
+_DESCRIPTION = """\
+This is the MLSUM subset of the GEM benchmark. MLSUM is the first large-scale MultiLingual SUMmarization dataset.
+Obtained from online newspapers, it contains 1.5M+ article/summary pairs in five different languages -- namely, French, German, Spanish, Russian, Turkish.
+Together with English newspapers from the popular CNN/Daily mail dataset, the collected data form a large scale multilingual dataset which can enable new research directions for the text summarization community.
+We report cross-lingual comparative analyses based on state-of-the-art systems.
+These highlight existing biases which motivate the use of a multi-lingual dataset.
+"""
+_URL = "https://gitlab.lip6.fr/scialom/mlsum_data/-/raw/master/MLSUM/"
+_LANG = ["de", "es"]
+_URLs = {
+    "de": {
+        "train": "https://gitlab.lip6.fr/scialom/mlsum_data/-/raw/master/MLSUM/de_train.zip",
+        "validation": "https://gitlab.lip6.fr/scialom/mlsum_data/-/raw/master/MLSUM/de_val.zip",
+        "test": "https://gitlab.lip6.fr/scialom/mlsum_data/-/raw/master/MLSUM/de_test.zip",
+        "bad_ids": "https://storage.googleapis.com/huggingface-nlp/datasets/gem/gem_mlsum_bad_ids_fixed.json",
+        "challenge_set": "https://storage.googleapis.com/huggingface-nlp/datasets/gem/gem_challenge_sets/mlsum_de.zip",
+    },
+    "es": {
+        "train": "https://gitlab.lip6.fr/scialom/mlsum_data/-/raw/master/MLSUM/es_train.zip",
+        "validation": "https://gitlab.lip6.fr/scialom/mlsum_data/-/raw/master/MLSUM/es_val.zip",
+        "test": "https://gitlab.lip6.fr/scialom/mlsum_data/-/raw/master/MLSUM/es_test.zip",
+        "bad_ids": "https://storage.googleapis.com/huggingface-nlp/datasets/gem/gem_mlsum_bad_ids_fixed.json",
+        "challenge_set": "https://storage.googleapis.com/huggingface-nlp/datasets/gem/gem_challenge_sets/mlsum_es.zip",
+    },
+}
+
+
+class Mlsum(datasets.GeneratorBasedBuilder):
+
+    BUILDER_CONFIGS = [
+        datasets.BuilderConfig(
+            name=lang,
+            version=datasets.Version("1.0.0"),
+            description="",
+        )
+        for lang in _LANG
+    ]
+
+    def _info(self):
+        return datasets.DatasetInfo(
+            description=_DESCRIPTION,
+            features=datasets.Features(
+                {
+                    "gem_id": datasets.Value("string"),
+                    "gem_parent_id": datasets.Value("string"),
+                    "text": datasets.Value("string"),
+                    "topic": datasets.Value("string"),
+                    "url": datasets.Value("string"),
+                    "title": datasets.Value("string"),
+                    "date": datasets.Value("string"),
+                    "target": datasets.Value("string"),
+                    "references": [datasets.Value("string")],
+                }
+            ),
+            supervised_keys=None,
+            homepage="",
+            citation=_CITATION,
+        )
+
+    def _split_generators(self, dl_manager):
+        """Returns SplitGenerators."""
+        dl_dir = dl_manager.download_and_extract(_URLs[self.config.name])
+        lang = str(self.config.name)
+        challenge_sets = [
+            ("challenge_train_sample", f"train_mlsum_{lang}_RandomSample500.json"),
+            ("challenge_validation_sample", f"validation_mlsum_{lang}_RandomSample500.json"),
+            ("challenge_test_covid", f"{lang}_test_covid19_cleaned.jsonl"),
+        ]
+        return [
+            datasets.SplitGenerator(
+                name=datasets.Split.TRAIN,
+                gen_kwargs={
+                    "filepath": os.path.join(dl_dir["train"], lang + "_train.jsonl"),
+                    "split": "train",
+                    "lang": lang,
+                    "filepaths": dl_dir["bad_ids"],
+                },
+            ),
+            datasets.SplitGenerator(
+                name=datasets.Split.VALIDATION,
+                gen_kwargs={
+                    "filepath": os.path.join(dl_dir["validation"], lang + "_val.jsonl"),
+                    "split": "validation",
+                    "lang": lang,
+                    "filepaths": dl_dir["bad_ids"],
+                },
+            ),
+            datasets.SplitGenerator(
+                name=datasets.Split.TEST,
+                gen_kwargs={
+                    "filepath": os.path.join(dl_dir["test"], lang + "_test.jsonl"),
+                    "split": "test",
+                    "lang": lang,
+                    "filepaths": dl_dir["bad_ids"],
+                },
+            ),
+        ] + [
+            datasets.SplitGenerator(
+                name=challenge_split,
+                gen_kwargs={
+                    "filepath": os.path.join(dl_dir["challenge_set"], f"mlsum_{self.config.name}", filename),
+                    "split": challenge_split,
+                },
+            )
+            for challenge_split, filename in challenge_sets
+        ]
+
+    def _generate_examples(self, filepath, split, filepaths=None, lang=None):
+        """Yields examples."""
+        if split in ["train", "validation", "test", "challenge_test_covid"]:
+            if split == "challenge_test_covid":
+                bad_ids = {}
+            else:
+                bad_ids_dct = json.load(open(filepaths, encoding="utf-8"))
+                bad_ids = dict((bad_url, True) for _, bad_url in bad_ids_dct[f"{lang}-{split}"])
+            with open(filepath, encoding="utf-8") as f:
+                id_ = -1
+                for line in f:
+                    data = json.loads(line)
+                    if data["url"] in bad_ids:
+                        continue
+                    else:
+                        id_ += 1
+                        yield id_, {
+                            "gem_id": f"{self.config.name}-{split}-{id_}",
+                            "gem_parent_id": f"{self.config.name}-{split}-{id_}",
+                            "text": data["text"],
+                            "target": data["summary"],
+                            "references": [] if split == "train" else [data["summary"]],
+                            "topic": data["topic"],
+                            "url": data["url"],
+                            "title": data["title"],
+                            "date": data["date"],
+                        }
+        else:
+            exples = json.load(open(filepath, encoding="utf-8"))
+            if isinstance(exples, dict):
+                assert len(exples) == 1, "multiple entries found"
+                exples = list(exples.values())[0]
+            for id_, exple in enumerate(exples):
+                if len(exple) == 0:
+                    continue
+                exple["gem_parent_id"] = exple["gem_id"]
+                exple["gem_id"] = f"{self.config.name}-{split}-{id_}"
+                yield id_, exple