Duplicate from deepmind/pg19

Co-authored-by: Parquet-converter (BOT) <parquet-converter@users.noreply.huggingface.co>

.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 +1,211 @@
+---
+annotations_creators:
+- expert-generated
+language_creators:
+- expert-generated
+language:
+- en
+license:
+- apache-2.0
+multilinguality:
+- monolingual
+size_categories:
+- 10K<n<100K
+source_datasets:
+- original
+task_categories:
+- text-generation
+task_ids:
+- language-modeling
+paperswithcode_id: pg-19
+pretty_name: PG-19
+dataset_info:
+  features:
+  - name: short_book_title
+    dtype: string
+  - name: publication_date
+    dtype: int32
+  - name: url
+    dtype: string
+  - name: text
+    dtype: string
+  splits:
+  - name: train
+    num_bytes: 11453688452
+    num_examples: 28602
+  - name: validation
+    num_bytes: 17402295
+    num_examples: 50
+  - name: test
+    num_bytes: 40482852
+    num_examples: 100
+  download_size: 11740397875
+  dataset_size: 11511573599
+---
+
+# Dataset Card for "pg19"
+
+## Table of Contents
+- [Dataset Description](#dataset-description)
+  - [Dataset Summary](#dataset-summary)
+  - [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
+  - [Languages](#languages)
+- [Dataset Structure](#dataset-structure)
+  - [Data Instances](#data-instances)
+  - [Data Fields](#data-fields)
+  - [Data Splits](#data-splits)
+- [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:** [https://github.com/deepmind/pg19](https://github.com/deepmind/pg19)
+- **Repository:** [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
+- **Paper:** [Compressive Transformers for Long-Range Sequence Modelling](https://arxiv.org/abs/1911.05507)
+- **Point of Contact:** [More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
+- **Size of downloaded dataset files:** 11.74 GB
+- **Size of the generated dataset:** 11.51 GB
+- **Total amount of disk used:** 23.25 GB
+
+### Dataset Summary
+
+This repository contains the PG-19 language modeling benchmark.
+It includes a set of books extracted from the Project Gutenberg books library, that were published before 1919.
+It also contains metadata of book titles and publication dates.
+
+PG-19 is over double the size of the Billion Word benchmark and contains documents that are 20X longer, on average, than the WikiText long-range language modelling benchmark.
+Books are partitioned into a train, validation, and test set. Book metadata is stored in metadata.csv which contains (book_id, short_book_title, publication_date).
+
+Unlike prior benchmarks, we do not constrain the vocabulary size --- i.e. mapping rare words to an UNK token --- but instead release the data as an open-vocabulary benchmark. The only processing of the text that has been applied is the removal of boilerplate license text, and the mapping of offensive discriminatory words as specified by Ofcom to placeholder tokens. Users are free to model the data at the character-level, subword-level, or via any mechanism that can model an arbitrary string of text.
+To compare models we propose to continue measuring the word-level perplexity, by calculating the total likelihood of the dataset (via any chosen subword vocabulary or character-based scheme) divided by the number of tokens --- specified below in the dataset statistics table.
+One could use this dataset for benchmarking long-range language models, or use it to pre-train for other natural language processing tasks which require long-range reasoning, such as LAMBADA or NarrativeQA. We would not recommend using this dataset to train a general-purpose language model, e.g. for applications to a production-system dialogue agent, due to the dated linguistic style of old texts and the inherent biases present in historical writing.
+
+### Supported Tasks and Leaderboards
+
+[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
+
+### Languages
+
+[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
+
+## Dataset Structure
+
+### Data Instances
+
+#### default
+
+- **Size of downloaded dataset files:** 11.74 GB
+- **Size of the generated dataset:** 11.51 GB
+- **Total amount of disk used:** 23.25 GB
+
+An example of 'train' looks as follows.
+```
+This example was too long and was cropped:
+
+{
+    "publication_date": 1907,
+    "short_book_title": "La Fiammetta by Giovanni Boccaccio",
+    "text": "\"\\n\\n\\n\\nProduced by Ted Garvin, Dave Morgan and PG Distributed Proofreaders\\n\\n\\n\\n\\nLA FIAMMETTA\\n\\nBY\\n\\nGIOVANNI BOCCACCIO\\n...",
+    "url": "http://www.gutenberg.org/ebooks/10006"
+}
+```
+
+### Data Fields
+
+The data fields are the same among all splits.
+
+#### default
+- `short_book_title`: a `string` feature.
+- `publication_date`: a `int32` feature.
+- `url`: a `string` feature.
+- `text`: a `string` feature.
+
+### Data Splits
+
+| name  |train|validation|test|
+|-------|----:|---------:|---:|
+|default|28602|        50| 100|
+
+## Dataset Creation
+
+### Curation Rationale
+
+[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
+
+### Source Data
+
+#### Initial Data Collection and Normalization
+
+[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
+
+#### Who are the source language producers?
+
+[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
+
+### Annotations
+
+#### Annotation process
+
+[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
+
+#### Who are the annotators?
+
+[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
+
+### Personal and Sensitive Information
+
+[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
+
+## Considerations for Using the Data
+
+### Social Impact of Dataset
+
+[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
+
+### Discussion of Biases
+
+[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
+
+### Other Known Limitations
+
+[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
+
+## Additional Information
+
+### Dataset Curators
+
+[More Information Needed](https://github.com/huggingface/datasets/blob/master/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
+
+### Licensing Information
+
+The dataset is licensed under [Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0.html).
+
+### Citation Information
+
+```
+@article{raecompressive2019,
+  author = {Rae, Jack W and Potapenko, Anna and Jayakumar, Siddhant M and
+            Hillier, Chloe and Lillicrap, Timothy P},
+  title = {Compressive Transformers for Long-Range Sequence Modelling},
+  journal = {arXiv preprint},
+  url = {https://arxiv.org/abs/1911.05507},
+  year = {2019},
+}
+```
+
+
+### Contributions
+
+Thanks to [@thomwolf](https://github.com/thomwolf), [@lewtun](https://github.com/lewtun), [@lucidrains](https://github.com/lucidrains), [@lhoestq](https://github.com/lhoestq) for adding this dataset.

data/test_files.txt

@@ -0,0 +1,100 @@
+test/10146.txt
+test/10321.txt
+test/10356.txt
+test/10762.txt
+test/12204.txt
+test/15562.txt
+test/22424.txt
+test/24553.txt
+test/2544.txt
+test/25646.txt
+test/25773.txt
+test/25830.txt
+test/26183.txt
+test/26239.txt
+test/26493.txt
+test/26618.txt
+test/27454.txt
+test/28444.txt
+test/28988.txt
+test/29594.txt
+test/29973.txt
+test/30312.txt
+test/30752.txt
+test/30754.txt
+test/30909.txt
+test/30981.txt
+test/31065.txt
+test/3129.txt
+test/31974.txt
+test/3247.txt
+test/32761.txt
+test/3340.txt
+test/33426.txt
+test/33756.txt
+test/34467.txt
+test/35205.txt
+test/35246.txt
+test/3608.txt
+test/36256.txt
+test/37006.txt
+test/37328.txt
+test/37403.txt
+test/37443.txt
+test/3754.txt
+test/37702.txt
+test/38106.txt
+test/3890.txt
+test/38929.txt
+test/38955.txt
+test/4047.txt
+test/40579.txt
+test/40700.txt
+test/4128.txt
+test/41603.txt
+test/41607.txt
+test/42081.txt
+test/42655.txt
+test/43536.txt
+test/43845.txt
+test/44099.txt
+test/44557.txt
+test/45313.txt
+test/45881.txt
+test/45888.txt
+test/46915.txt
+test/47068.txt
+test/47558.txt
+test/47581.txt
+test/47676.txt
+test/48693.txt
+test/49078.txt
+test/49529.txt
+test/49596.txt
+test/50287.txt
+test/51410.txt
+test/53345.txt
+test/5396.txt
+test/54537.txt
+test/54624.txt
+test/55339.txt
+test/55871.txt
+test/56410.txt
+test/5734.txt
+test/5770.txt
+test/57791.txt
+test/58473.txt
+test/58553.txt
+test/58598.txt
+test/5956.txt
+test/5962.txt
+test/6412.txt
+test/6941.txt
+test/7412.txt
+test/7987.txt
+test/8197.txt
+test/8559.txt
+test/860.txt
+test/8788.txt
+test/9315.txt
+test/9931.txt

data/validation_files.txt

@@ -0,0 +1,50 @@
+validation/1022.txt
+validation/11155.txt
+validation/13089.txt
+validation/16959.txt
+validation/1925.txt
+validation/2383.txt
+validation/23956.txt
+validation/24360.txt
+validation/25066.txt
+validation/27688.txt
+validation/28213.txt
+validation/28776.txt
+validation/29981.txt
+validation/32629.txt
+validation/34016.txt
+validation/34056.txt
+validation/34100.txt
+validation/356.txt
+validation/35816.txt
+validation/36402.txt
+validation/37833.txt
+validation/38214.txt
+validation/38403.txt
+validation/4024.txt
+validation/41074.txt
+validation/42067.txt
+validation/42142.txt
+validation/42306.txt
+validation/43423.txt
+validation/44896.txt
+validation/44912.txt
+validation/4533.txt
+validation/48089.txt
+validation/48461.txt
+validation/48677.txt
+validation/49091.txt
+validation/50355.txt
+validation/51859.txt
+validation/5195.txt
+validation/5321.txt
+validation/53682.txt
+validation/54098.txt
+validation/555.txt
+validation/55658.txt
+validation/56719.txt
+validation/57843.txt
+validation/58093.txt
+validation/6404.txt
+validation/7510.txt
+validation/8545.txt

pg19.py

@@ -0,0 +1,146 @@
+"""PG-19 language modeling benchmark - a set of books extracted from the Project Gutenberg books library"""
+
+
+import csv
+import os
+
+import datasets
+
+
+# TODO(pg19): BibTeX citation
+_CITATION = """\
+@article{raecompressive2019,
+  author = {Rae, Jack W and Potapenko, Anna and Jayakumar, Siddhant M and
+            Hillier, Chloe and Lillicrap, Timothy P},
+  title = {Compressive Transformers for Long-Range Sequence Modelling},
+  journal = {arXiv preprint},
+  url = {https://arxiv.org/abs/1911.05507},
+  year = {2019},
+}
+
+"""
+
+# TODO(pg19):
+_DESCRIPTION = """\
+This repository contains the PG-19 language modeling benchmark.
+It includes a set of books extracted from the Project Gutenberg books library, that were published before 1919.
+It also contains metadata of book titles and publication dates.
+
+PG-19 is over double the size of the Billion Word benchmark and contains documents that are 20X longer, on average, than the WikiText long-range language modelling benchmark.
+Books are partitioned into a train, validation, and test set. Book metadata is stored in metadata.csv which contains (book_id, short_book_title, publication_date).
+
+Unlike prior benchmarks, we do not constrain the vocabulary size --- i.e. mapping rare words to an UNK token --- but instead release the data as an open-vocabulary benchmark. The only processing of the text that has been applied is the removal of boilerplate license text, and the mapping of offensive discriminatory words as specified by Ofcom to placeholder tokens. Users are free to model the data at the character-level, subword-level, or via any mechanism that can model an arbitrary string of text.
+To compare models we propose to continue measuring the word-level perplexity, by calculating the total likelihood of the dataset (via any chosen subword vocabulary or character-based scheme) divided by the number of tokens --- specified below in the dataset statistics table.
+One could use this dataset for benchmarking long-range language models, or use it to pre-train for other natural language processing tasks which require long-range reasoning, such as LAMBADA or NarrativeQA. We would not recommend using this dataset to train a general-purpose language model, e.g. for applications to a production-system dialogue agent, due to the dated linguistic style of old texts and the inherent biases present in historical writing.
+"""
+
+
+_SPLIT_FILES_PATH = "data/{split}_files.txt"
+_ASSET_ROOT_URL = "https://storage.googleapis.com/deepmind-gutenberg/"
+_METADATA_URL = _ASSET_ROOT_URL + "metadata.csv"
+
+
+def flat_map(fn, arr):
+    return [el for sub_arr in map(fn, arr) for el in sub_arr]
+
+
+class Pg19(datasets.GeneratorBasedBuilder):
+    """PG-19 dataset - books as plain text extracted from the Project Gutenberg library"""
+
+    # TODO(pg19): Set up version.
+    VERSION = datasets.Version("0.1.0")
+
+    def _info(self):
+        # TODO(pg19): 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(
+                {
+                    "short_book_title": datasets.Value("string"),
+                    "publication_date": datasets.Value("int32"),
+                    "url": datasets.Value("string"),
+                    "text": 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/deepmind/pg19",
+            citation=_CITATION,
+        )
+
+    def _split_generators(self, dl_manager):
+        """Returns SplitGenerators."""
+        splits = ["train", "validation", "test"]
+        files = dl_manager.download({split: _SPLIT_FILES_PATH.format(split=split) for split in splits})
+
+        for split, names_file in list(files.items()):
+            with open(names_file, encoding="utf-8") as f:
+                split_files = f.read().splitlines()
+            split_files = sorted(split_files)
+            split_files = {
+                os.path.splitext(os.path.basename(file))[0]: _ASSET_ROOT_URL + file
+                for file in split_files
+            }
+            files[split] = split_files
+
+        metadata = dl_manager.download(_METADATA_URL)
+        downloaded_files = dl_manager.download(files)
+        return [
+            datasets.SplitGenerator(
+                name=datasets.Split.TRAIN,
+                gen_kwargs={
+                    "ids": list(downloaded_files["train"]),
+                    "metadata": metadata,
+                    "files": downloaded_files["train"],
+                },
+            ),
+            datasets.SplitGenerator(
+                name=datasets.Split.VALIDATION,
+                gen_kwargs={
+                    "ids": list(downloaded_files["validation"]),
+                    "metadata": metadata,
+                    "files": downloaded_files["validation"],
+                },
+            ),
+            datasets.SplitGenerator(
+                name=datasets.Split.TEST,
+                gen_kwargs={
+                    "ids": list(downloaded_files["test"]),
+                    "metadata": metadata,
+                    "files": downloaded_files["test"],
+                },
+            ),
+        ]
+
+    def _generate_examples(self, ids, metadata, files):
+        """Yields examples."""
+        # TODO(pg19): Yields (key, example) tuples from the dataset
+
+        with open(metadata, encoding="utf-8") as f:
+            reader = csv.DictReader(f, fieldnames=["_id", "short_book_title", "publication_date", "url"])
+            id2metadata = {row["_id"]: row for row in reader}
+
+        for _id in ids:
+            data = id2metadata[_id]
+            file = files[_id]
+
+            with open(file, encoding="utf-8") as f:
+                text = f.read()
+
+            _id = data["_id"]
+            short_book_title = data["short_book_title"]
+            publication_date = int(data["publication_date"])
+            url = data["url"]
+
+            yield _id, {
+                "short_book_title": short_book_title,
+                "publication_date": publication_date,
+                "url": url,
+                "text": text,
+            }