Datasets:
Tasks:
Token Classification
Sub-tasks:
parsing
Languages:
English
Multilinguality:
monolingual
Size Categories:
unknown
Language Creators:
found
Annotations Creators:
expert-generated
Source Datasets:
original
File size: 3,493 Bytes
5aaf8e4 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 |
import glob
import json
import os
from dataclasses import dataclass
import datasets
from datasets import BuilderConfig, SplitGenerator
_CITATION = """\
@article{yang2018scidtb,
title={Scidtb: Discourse dependency treebank for scientific abstracts},
author={Yang, An and Li, Sujian},
journal={arXiv preprint arXiv:1806.03653},
year={2018}
}
"""
_DESCRIPTION = """Annotation corpus for discourse relations benefits NLP tasks such as machine translation and question
answering. SciDTB is a domain-specific discourse treebank annotated on scientific articles.
Different from widely-used RST-DT and PDTB, SciDTB uses dependency trees to represent discourse structure, which is
flexible and simplified to some extent but do not sacrifice structural integrity. We discuss the labeling framework,
annotation workflow and some statistics about SciDTB. Furthermore, our treebank is made as a benchmark for evaluating
discourse dependency parsers, on which we provide several baselines as fundamental work."""
_URL = "https://codeload.github.com/PKU-TANGENT/SciDTB/zip/refs/heads/master"
_HOMEPAGE = "https://github.com/PKU-TANGENT/SciDTB"
@dataclass
class SciDTBConfig(BuilderConfig):
"""BuilderConfig for SciDTB."""
def __init__(self, subdirectory_mapping, encoding, **kwargs):
super(SciDTBConfig, self).__init__(**kwargs)
self.subdirectory_mapping = subdirectory_mapping
self.encoding = encoding
class SciDTBDataset(datasets.GeneratorBasedBuilder):
"""Scientific Discourse Treebank Dataset"""
BUILDER_CONFIGS = [
SciDTBConfig(
name="SciDTB",
version=datasets.Version("1.0.0", ""),
description=_DESCRIPTION,
subdirectory_mapping={
"train": "SciDTB-master/dataset/train",
"dev": "SciDTB-master/dataset/dev/gold",
"test": "SciDTB-master/dataset/test/gold",
},
encoding="utf-8-sig",
),
]
def _info(self):
features = datasets.Features(
{
"root": datasets.Sequence(
{
"id": datasets.Value("int32"),
"parent": datasets.Value("int32"),
"text": datasets.Value("string"),
"relation": datasets.Value("string"),
}
),
"file_name": datasets.Value("string"),
}
)
return datasets.DatasetInfo(
description=_DESCRIPTION,
features=features,
homepage=_HOMEPAGE,
citation=_CITATION,
)
def _split_generators(self, dl_manager):
data_dir = dl_manager.download_and_extract(_URL)
return [
SplitGenerator(
name=split,
# These kwargs will be passed to _generate_examples
gen_kwargs={
"dir_path": os.path.join(data_dir, subdir),
},
)
for split, subdir in self.config.subdirectory_mapping.items()
]
def _generate_examples(self, dir_path):
_files = glob.glob(f"{dir_path}/*.dep")
for file_path in _files:
with open(file_path, mode="r", encoding=self.config.encoding) as f:
annotations = json.load(f)
annotations["file_name"] = os.path.basename(file_path)
yield annotations["file_name"], annotations
|