M4LE.py

# 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.
import os

import datasets
import json


_DESCRIPTION = (
    "M4LE is a systematic and comprehensive long-context benchmark. It aims to"
    " evaluate LM performances in five long-context understanding abilities,"
    " across multiple domains, languages and task types."
)
_HOMEPAGE = "https://github.com/KwanWaiChung/M4LE"
_LICENSE = """MIT License
Copyright (c) 2023 Wai-Chung Kwan

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE."""
URL = r"https://huggingface.co/datasets/wckwan/M4LE/resolve/main/data.zip"


tasks = [
    "arxiv",
    "bigpatent_global_cls",
    "bigpatent_global_sum",
    "booksum",
    "c3",
    "cepsum",
    "clts+",
    "cnewsum",
    "cnnnews",
    "drcd_explicit-single_ret",
    "drcd_semantic-single_qa",
    "duorc",
    "dureader",
    "hotpotqa",
    "lcsts",
    "marc",
    "mnds-news_explicit-multiple_cls+ret",
    "mnds-news_explicit-single_cls+ret",
    "mnds-news_semantic-multiple_cls+cnt",
    "ncls",
    "news-commentary-en2zh",
    "news-commentary-zh2en",
    "news2016",
    "newsqa",
    "nq-open",
    "online-shopping",
    "open-subtitles-en2zh",
    "open-subtitles-zh2en",
    "pubmed",
    "tedtalks-en2zh",
    "tedtalks-zh2en",
    "thucnews_explicit-multiple_cls+ret",
    "thucnews_explicit-single_cls+ret",
    "thucnews_semantic-multiple_cls+cnt",
    "triviaqa",
    "wiki2019zh",
    "wikihow",
    "wikitext-103",
    "wow",
]


class M4LEConfig(datasets.BuilderConfig):
    def __init__(self, **kwargs):
        super().__init__(version=datasets.Version("1.0.0"), **kwargs)


class LongBench(datasets.GeneratorBasedBuilder):
    BUILDER_CONFIGS = [
        M4LEConfig(
            name=task,
        )
        for task in tasks
    ]

    def _info(self):
        features = datasets.Features(
            {
                "instruction": datasets.Value("string"),
                "input": datasets.Value("string"),
                "answers": [datasets.Value("string")],
                "input_length": datasets.Value("int32"),
                "total_length": datasets.Value("int32"),
                "length_bucket": datasets.Value("int32"),
            }
        )
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=features,
            homepage=_HOMEPAGE,
            license=_LICENSE,
        )

    def _split_generators(self, dl_manager):
        data_dir = dl_manager.download_and_extract(URL)
        task_name = self.config.name
        return [
            datasets.SplitGenerator(
                name=datasets.Split.TEST,
                gen_kwargs={
                    "filepath": os.path.join(
                        data_dir, "data", f"{task_name}.jsonl"
                    ),
                },
            )
        ]

    def _generate_examples(self, filepath):
        with open(filepath, encoding="utf-8") as f:
            for idx, line in enumerate(f):
                key = f"{self.config.name}-{idx}"
                item = json.loads(line)
                yield key, {
                    "instruction": item["instruction"],
                    "input": item["input"],
                    "answers": item["answers"],
                    "input_length": item["input_length"],
                    "total_length": item["total_length"],
                    "length_bucket": item["length_bucket"],
                }