Update files from the datasets library (from 1.3.0)

Release notes: https://github.com/huggingface/datasets/releases/tag/1.3.0

.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

README.md

@@ -0,0 +1,181 @@
+---
+annotations_creators:
+- no-annotation
+language_creators:
+- found
+languages:
+- id
+licenses:
+- unknown
+multilinguality:
+- monolingual
+size_categories:
+- 100K<n<1M
+source_datasets:
+- original
+task_categories:
+- conditional-text-generation
+task_ids:
+- summarization
+---
+
+# Dataset Card for Large-scale Indonesian Summarization
+
+## Table of Contents
+
+- [Dataset Description](#dataset-description)
+  - [Dataset Summary](#dataset-summary)
+  - [Supported Tasks](#supported-tasks-and-leaderboards)
+  - [Languages](#languages)
+- [Dataset Structure](#dataset-structure)
+  - [Data Instances](#data-instances)
+  - [Data Fields](#data-instances)
+  - [Data Splits](#data-instances)
+- [Dataset Creation](#dataset-creation)
+  - [Curation Rationale](#curation-rationale)
+  - [Source Data](#source-data)
+  - [Annotations](#annotations)
+  - [Personal and Sensitive Information](#personal-and-sensitive-information)
+- [Considerations for Using the Data](#considerations-for-using-the-data)
+  - [Social Impact of Dataset](#social-impact-of-dataset)
+  - [Discussion of Biases](#discussion-of-biases)
+  - [Other Known Limitations](#other-known-limitations)
+- [Additional Information](#additional-information)
+  - [Dataset Curators](#dataset-curators)
+  - [Licensing Information](#licensing-information)
+  - [Citation Information](#citation-information)
+  - [Contributions](#contributions)
+
+## Dataset Description
+
+- **Homepage:** [IndoLEM (Indonesian Language Evaluation Montage)](https://indolem.github.io/)
+- **Repository:** [Liputan6: Summarization Corpus for Indonesian](https://github.com/fajri91/sum_liputan6/)
+- **Paper:** https://arxiv.org/abs/2011.00679
+- **Leaderboard:**
+- **Point of Contact:** [Fajri Koto](mailto:[email protected]),
+[Jey Han Lau](mailto:[email protected]), [Timothy Baldwin](mailto:[email protected]), 
+
+### Dataset Summary
+
+In this paper, we introduce a large-scale Indonesian summarization dataset. We harvest articles from this http URL,
+an online news portal, and obtain 215,827 document-summary pairs. We leverage pre-trained language models to develop
+benchmark extractive and abstractive summarization methods over the dataset with multilingual and monolingual
+BERT-based models. We include a thorough error analysis by examining machine-generated summaries that have
+low ROUGE scores, and expose both issues with ROUGE it-self, as well as with extractive and abstractive
+summarization models.
+
+The dataset has two variants: "canonical" and "xtreme". The "xtreme" variant discards development and test 
+document–summary pairs where the summary has fewer than 90% novel 4-grams (the training data remains the same 
+as the canonical variant).
+
+You need to manually request the liputan6 dataset using the form in https://github.com/fajri91/sum_liputan6/
+and uncompress it. The liputan6 dataset can then be loaded using the following command 
+`datasets.load_dataset("id_liputan6", 'canonical', data_dir="<path/to/uncompressed_folder>")` or
+`datasets.load_dataset("id_liputan6", 'xtreme', data_dir="<path/to/uncompressed_folder>")`.
+### Supported Tasks and Leaderboards
+
+[More Information Needed]
+
+### Languages
+Indonesian
+
+## Dataset Structure
+```
+{
+  'id': 'string',
+  'url': 'string',
+  'clean_article': 'string',
+  'clean_article': 'string',
+  'extractive_summary': 'string'
+}
+```
+### Data Instances
+
+An example of the dataset:
+```
+{
+  'clean_article': 'Liputan6.com, Ambon: Partai Bulan Bintang wilayah Maluku bertekad membantu pemerintah menyelesaikan konflik di provinsi tersebut. Syaratnya, penanganan penyelesaian konflik Maluku harus dimulai dari awal kerusuhan, yakni 19 Januari 1999. Demikian hasil Musyawarah Wilayah I PBB Maluku yang dimulai Sabtu pekan silam dan berakhir Senin (31/12) di Ambon. Menurut seorang fungsionaris PBB Ridwan Hasan, persoalan di Maluku bisa selesai asalkan pemerintah dan aparat keamanan serius menangani setiap persoalan di Maluku secara komprehensif dan bijaksana. Itulah sebabnya, PBB wilayah Maluku akan menjadikan penyelesaian konflik sebagai agenda utama partai. PBB Maluku juga akan mendukung penegakan hukum secara terpadu dan tanpa pandang bulu. Siapa saja yang melanggar hukum harus ditindak. Ridwan berharap, Ketua PBB Maluku yang baru, Ali Fauzi, dapat menindak lanjuti agenda politik partai yang telah diamanatkan dan mau mendukung penegakan hukum di Maluku. (ULF/Sahlan Heluth).',
+  'clean_summary': 'Konflik Ambon telah berlangsung selama tiga tahun. Partai Bulan Bintang wilayah Maluku siap membantu pemerintah menyelesaikan kasus di provinsi tersebut.',
+  'extractive_summary': 'Liputan6.com, Ambon: Partai Bulan Bintang wilayah Maluku bertekad membantu pemerintah menyelesaikan konflik di provinsi tersebut. Siapa saja yang melanggar hukum harus ditindak.',
+  'id': '26408',
+  'url': 'https://www.liputan6.com/news/read/26408/pbb-siap-membantu-penyelesaian-konflik-ambon'
+}
+
+```
+
+### Data Fields
+- `id`: id of the sample
+- `url`: the url to the original article
+- `clean_article`: the original article
+- `clean_article`: the abstractive summarization
+- `extractive_summary`: the extractive summarization
+
+### Data Splits
+
+The dataset is splitted in to train, validation and test sets.
+
+## Dataset Creation
+
+### Curation Rationale
+
+[More Information Needed]
+
+### Source Data
+
+#### Initial Data Collection and Normalization
+
+[More Information Needed]
+
+#### Who are the source language producers?
+
+[More Information Needed]
+
+### Annotations
+
+#### Annotation process
+
+[More Information Needed]
+
+#### Who are the annotators?
+[More Information Needed]
+
+### Personal and Sensitive Information
+
+[More Information Needed]
+
+## Considerations for Using the Data
+
+### Social Impact of Dataset
+
+[More Information Needed]
+
+### Discussion of Biases
+
+[More Information Needed]
+
+### Other Known Limitations
+
+[More Information Needed]
+
+## Additional Information
+
+### Dataset Curators
+
+[More Information Needed]
+
+### Licensing Information
+
+[More Information Needed]
+
+### Citation Information
+```
+@inproceedings{Koto2020Liputan6AL,
+  title={Liputan6: A Large-scale Indonesian Dataset for Text Summarization},
+  author={Fajri Koto and Jey Han Lau and Timothy Baldwin},
+  booktitle={AACL/IJCNLP},
+  year={2020}
+}
+```
+### Contributions
+
+Thanks to [@cahya-wirawan](https://github.com/cahya-wirawan) for adding this dataset.

dataset_infos.json

@@ -0,0 +1 @@
+{"canonical": {"description": "In this paper, we introduce a large-scale Indonesian summarization dataset. We harvest articles from this http URL,\nan online news portal, and obtain 215,827 document-summary pairs. We leverage pre-trained language models to develop\nbenchmark extractive and abstractive summarization methods over the dataset with multilingual and monolingual\nBERT-based models. We include a thorough error analysis by examining machine-generated summaries that have\nlow ROUGE scores, and expose both issues with ROUGE it-self, as well as with extractive and abstractive\nsummarization models.\n", "citation": "@inproceedings{id_liputan6,\n  author    = {Fajri Koto, Jey Han Lau, Timothy Baldwin},\n  title     = {Liputan6: A Large-scale Indonesian Dataset for Text Summarization},\n  year      = {2020},\n  url       = {https://arxiv.org/abs/2011.00679},\n}\n", "homepage": "https://arxiv.org/abs/2011.00679", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "url": {"dtype": "string", "id": null, "_type": "Value"}, "clean_article": {"dtype": "string", "id": null, "_type": "Value"}, "clean_summary": {"dtype": "string", "id": null, "_type": "Value"}, "extractive_summary": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "id_liputan6", "config_name": "canonical", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"validation": {"name": "validation", "num_bytes": 20944658, "num_examples": 10972, "dataset_name": "id_liputan6"}, "test": {"name": "test", "num_bytes": 20526768, "num_examples": 10972, "dataset_name": "id_liputan6"}, "train": {"name": "train", "num_bytes": 382245586, "num_examples": 193883, "dataset_name": "id_liputan6"}}, "download_checksums": {}, "download_size": 0, "post_processing_size": null, "dataset_size": 423717012, "size_in_bytes": 423717012}, "xtreme": {"description": "In this paper, we introduce a large-scale Indonesian summarization dataset. We harvest articles from this http URL,\nan online news portal, and obtain 215,827 document-summary pairs. We leverage pre-trained language models to develop\nbenchmark extractive and abstractive summarization methods over the dataset with multilingual and monolingual\nBERT-based models. We include a thorough error analysis by examining machine-generated summaries that have\nlow ROUGE scores, and expose both issues with ROUGE it-self, as well as with extractive and abstractive\nsummarization models.\n", "citation": "@inproceedings{id_liputan6,\n  author    = {Fajri Koto, Jey Han Lau, Timothy Baldwin},\n  title     = {Liputan6: A Large-scale Indonesian Dataset for Text Summarization},\n  year      = {2020},\n  url       = {https://arxiv.org/abs/2011.00679},\n}\n", "homepage": "https://arxiv.org/abs/2011.00679", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "url": {"dtype": "string", "id": null, "_type": "Value"}, "clean_article": {"dtype": "string", "id": null, "_type": "Value"}, "clean_summary": {"dtype": "string", "id": null, "_type": "Value"}, "extractive_summary": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "id_liputan6", "config_name": "xtreme", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"validation": {"name": "validation", "num_bytes": 9652946, "num_examples": 4948, "dataset_name": "id_liputan6"}, "test": {"name": "test", "num_bytes": 7574550, "num_examples": 3862, "dataset_name": "id_liputan6"}}, "download_checksums": {}, "download_size": 0, "post_processing_size": null, "dataset_size": 17227496, "size_in_bytes": 17227496}}

dummy/canonical/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f631dc2bdb5daf6217b37c8b1d1e12c6653f88bd78e7c946b0e7aff90a6482c8
+size 14764

dummy/xtreme/1.0.0/dummy_data.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f631dc2bdb5daf6217b37c8b1d1e12c6653f88bd78e7c946b0e7aff90a6482c8
+size 14764

id_liputan6.py

@@ -0,0 +1,175 @@
+# 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.
+"""Large-scale Indonesian Summarization Dataset"""
+
+from __future__ import absolute_import, division, print_function
+
+import glob
+import json
+import logging
+import os
+import re
+from pathlib import Path
+
+import datasets
+
+
+_CITATION = """\
+@inproceedings{id_liputan6,
+  author    = {Fajri Koto, Jey Han Lau, Timothy Baldwin},
+  title     = {Liputan6: A Large-scale Indonesian Dataset for Text Summarization},
+  year      = {2020},
+  url       = {https://arxiv.org/abs/2011.00679},
+}
+"""
+
+_DESCRIPTION = """\
+In this paper, we introduce a large-scale Indonesian summarization dataset. We harvest articles from this http URL,
+an online news portal, and obtain 215,827 document-summary pairs. We leverage pre-trained language models to develop
+benchmark extractive and abstractive summarization methods over the dataset with multilingual and monolingual
+BERT-based models. We include a thorough error analysis by examining machine-generated summaries that have
+low ROUGE scores, and expose both issues with ROUGE it-self, as well as with extractive and abstractive
+summarization models.
+"""
+
+_HOMEPAGE = "https://arxiv.org/abs/2011.00679"
+
+_LICENSE = ""
+
+
+class IdLiputan6Config(datasets.BuilderConfig):
+    """BuilderConfig for IdLiputan6"""
+
+    def __init__(self, **kwargs):
+        """BuilderConfig for IdLiputan6.
+        Args:
+          **kwargs: keyword arguments forwarded to super.
+        """
+        super(IdLiputan6Config, self).__init__(**kwargs)
+
+
+class IdLiputan6(datasets.GeneratorBasedBuilder):
+    VERSION = datasets.Version("1.0.0")
+
+    BUILDER_CONFIGS = [
+        IdLiputan6Config(
+            name="canonical",
+            version=VERSION,
+            description="Canonical Liputan6 dataset",
+        ),
+        IdLiputan6Config(
+            name="xtreme",
+            version=VERSION,
+            description="Xtreme Liputan6 dataset",
+        ),
+    ]
+
+    @property
+    def manual_download_instructions(self):
+        return """\
+            You need to manually request the liputan6 dataset using the form in https://github.com/fajri91/sum_liputan6/
+            and uncompress it.  The liputan6 dataset can then be loaded using the following command
+            `datasets.load_dataset("id_liputan6", 'canonical', data_dir="<path/to/uncompressed_folder>")` or
+            `datasets.load_dataset("id_liputan6", 'xtreme', data_dir="<path/to/uncompressed_folder>")`.
+            """
+
+    def _info(self):
+        features = datasets.Features(
+            {
+                "id": datasets.Value("string"),
+                "url": datasets.Value("string"),
+                "clean_article": datasets.Value("string"),
+                "clean_summary": datasets.Value("string"),
+                "extractive_summary": datasets.Value("string"),
+            }
+        )
+        return datasets.DatasetInfo(
+            description=_DESCRIPTION,
+            features=features,
+            supervised_keys=None,
+            homepage=_HOMEPAGE,
+            license=_LICENSE,
+            citation=_CITATION,
+        )
+
+    def _split_generators(self, dl_manager):
+        data_dir = os.path.abspath(os.path.expanduser(dl_manager.manual_dir))
+        if not os.path.exists(data_dir):
+            raise FileNotFoundError(
+                "{} does not exist. Make sure you insert a manual dir via `datasets.load_dataset('id_liputan6', "
+                "'canonical', data_dir=...)`. Manual download instructions:\n{}".format(
+                    data_dir, self.manual_download_instructions
+                )
+            )
+        split_generators = [
+            datasets.SplitGenerator(
+                name=datasets.Split.VALIDATION,
+                gen_kwargs={
+                    "article_dir": os.path.join(data_dir, "{}/dev".format(self.config.name)),
+                    "split": "dev",
+                },
+            ),
+            datasets.SplitGenerator(
+                name=datasets.Split.TEST,
+                gen_kwargs={
+                    "article_dir": os.path.join(data_dir, "{}/test".format(self.config.name)),
+                    "split": "test",
+                },
+            ),
+        ]
+        if self.config.name == "canonical":
+            split_generators.append(
+                datasets.SplitGenerator(
+                    name=datasets.Split.TRAIN,
+                    gen_kwargs={
+                        "article_dir": os.path.join(data_dir, "{}/train".format(self.config.name)),
+                        "split": "train",
+                    },
+                )
+            )
+        return split_generators
+
+    def _generate_examples(self, article_dir, split):
+        detokenizers = [
+            [re.compile(r"([Ll])iputan6 . com "), r"\1iputan6.com"],
+            [re.compile(r" ([.,:])"), r"\1"],
+            [re.compile(r"\( ([^)]+) \)"), r"(\1)"],
+            [re.compile(r"\" ([^\"]+) \""), r'"\1"'],
+            [re.compile(r"\[ ([^]]+) ]"), r"[\1]"],
+        ]
+        logging.info("⏳ Generating %s examples from = %s", split, article_dir)
+        guid = 0
+        for path in sorted(
+            glob.glob(os.path.join(article_dir, "**/*.json"), recursive=True), key=lambda p: int(Path(p).stem)
+        ):
+            with open(path, encoding="utf-8") as f:
+                data = json.load(f)
+                clean_article = " ".join([" ".join(i) for i in data["clean_article"]])
+                for d in detokenizers:
+                    clean_article = d[0].sub(d[1], clean_article)
+                clean_summary = " ".join([" ".join(i) for i in data["clean_summary"]])
+                for d in detokenizers:
+                    clean_summary = d[0].sub(d[1], clean_summary)
+                extractive_summary = " ".join([" ".join(data["clean_article"][i]) for i in data["extractive_summary"]])
+                for d in detokenizers:
+                    extractive_summary = d[0].sub(d[1], extractive_summary)
+                yield guid, {
+                    "id": str(data["id"]),
+                    "url": data["url"],
+                    "clean_article": clean_article,
+                    "clean_summary": clean_summary,
+                    "extractive_summary": extractive_summary,
+                }
+            guid += 1