phrase_retrieval.py

# coding=utf-8
# Copyright 2020 The TensorFlow Datasets Authors and the HuggingFace Datasets Authors.
#
# 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.

# Lint as: python3 
"""PiC: A Phrase-in-Context Dataset for Phrase Understanding and Semantic Search."""


import json
import os.path

import datasets


logger = datasets.logging.get_logger(__name__)


_CITATION = """\
@article{pham2022PiC,
  title={PiC: A Phrase-in-Context Dataset for Phrase Understanding and Semantic Search},
  author={Pham, Thang M and Yoon, Seunghyun and Bui, Trung and Nguyen, Anh},
  journal={arXiv preprint arXiv:2207.09068},
  year={2022}
}
"""

_DESCRIPTION = """\
Phrase in Context is a curated benchmark for phrase understanding and semantic search, consisting of three tasks of increasing difficulty: Phrase Similarity (PS), Phrase Retrieval (PR) and Phrase Sense Disambiguation (PSD). The datasets are annotated by 13 linguistic experts on Upwork and verified by two groups: ~1000 AMT crowdworkers and another set of 5 linguistic experts. PiC benchmark is distributed under CC-BY-NC 4.0.
"""

_HOMEPAGE = "https://phrase-in-context.github.io/"

_LICENSE = "CC-BY-NC-4.0"

_URL = "https://auburn.edu/~tmp0038/PiC/"
_SPLITS = {
    "train": "train-v1.0.json",
    "dev": "dev-v1.0.json",
    "test": "test-v1.0.json",
}

_PR_PASS = "PR-pass"
_PR_PAGE = "PR-page"


class PRConfig(datasets.BuilderConfig):
    """BuilderConfig for Phrase Retrieval in PiC."""

    def __init__(self, **kwargs):
        """BuilderConfig for Phrase Retrieval in PiC.
        Args:
          **kwargs: keyword arguments forwarded to super.
        """
        super(PRConfig, self).__init__(**kwargs)


class PhraseRetrieval(datasets.GeneratorBasedBuilder):
    """Phrase Retrieval in PiC dataset. Version 1.0."""

    BUILDER_CONFIGS = [
        PRConfig(
            name=_PR_PASS,
            version=datasets.Version("1.0.3"),
            description="The PiC Dataset for Phrase Retrieval at short passage level (~11 sentences)"
        ),
        PRConfig(
            name=_PR_PAGE,
            version=datasets.Version("1.0.3"),
            description="The PiC Dataset for Phrase Retrieval at Wiki page level"
        ),
    ]

    def _info(self):
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features(
                {
                    "id": datasets.Value("string"),
                    "title": datasets.Value("string"),
                    "context": datasets.Value("string"),
                    "query": datasets.Value("string"),
                    "answers": datasets.Sequence(
                        {
                            "text": datasets.Value("string"),
                            "answer_start": datasets.Value("int32"),
                        }
                    )
                }
            ),
            # No default supervised_keys (as we have to pass both question and context as input).
            supervised_keys=None,
            homepage=_HOMEPAGE,
            license=_LICENSE,
            citation=_CITATION,
        )

    def _split_generators(self, dl_manager):
        urls_to_download = {
            "train": os.path.join(_URL, self.config.name, _SPLITS["train"]),
            "dev": os.path.join(_URL, self.config.name, _SPLITS["dev"]),
            "test": os.path.join(_URL, self.config.name, _SPLITS["test"])
        }
        downloaded_files = dl_manager.download_and_extract(urls_to_download)

        return [
            datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"]}),
            datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={"filepath": downloaded_files["dev"]}),
            datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"filepath": downloaded_files["test"]}),
        ]

    def _generate_examples(self, filepath):
        """This function returns the examples in the raw (text) form."""
        logger.info("generating examples from = %s", filepath)
        key = 0
        with open(filepath, encoding="utf-8") as f:
            pic_pr = json.load(f)
            for example in pic_pr["data"]:
                title = example.get("title", "")

                answer_starts = [answer["answer_start"] for answer in example["answers"]]
                answers = [answer["text"] for answer in example["answers"]]

                # Features currently used are "context", "question", and "answers".
                # Others are extracted here for the ease of future expansions.
                yield key, {
                    "title": title,
                    "context": example["context"],
                    "query": example["question"],
                    "id": example["id"],
                    "answers": {
                        "answer_start": answer_starts,
                        "text": answers,
                    }
                }
                key += 1