Datasets:
Tasks:
Question Answering
Modalities:
Text
Sub-tasks:
open-domain-qa
Languages:
English
Size:
100K - 1M
ArXiv:
License:
File size: 4,183 Bytes
08cf776 afaad9d 08cf776 7ff48d8 08cf776 7ff48d8 08cf776 afaad9d 08cf776 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 |
"""KQA Pro: A large-scale, diverse, challenging dataset of complex question answering over knowledge base."""
import json
import os
import datasets
logger = datasets.logging.get_logger(__name__)
_CITATION = """\
@inproceedings{KQAPro,
title={{KQA P}ro: A Large Diagnostic Dataset for Complex Question Answering over Knowledge Base},
author={Cao, Shulin and Shi, Jiaxin and Pan, Liangming and Nie, Lunyiu and Xiang, Yutong and Hou, Lei and Li, Juanzi and He, Bin and Zhang, Hanwang},
booktitle={ACL'22},
year={2022}
}
"""
_DESCRIPTION = """\
A large-scale, diverse, challenging dataset of complex question answering over knowledge base.
"""
_URL = "https://thukeg.gitee.io/kqa-pro/"
_DOWNLOAD_URL = "https://cloud.tsinghua.edu.cn/f/df54ff66d1dc4ca7823e/?dl=1"
_URLS = {
"train": "train.json",
"val": "val.json",
"test": "test.json"
}
_TRAIN_CONFIG_NAME = "train_val"
_TEST_CONFIG_NAME = "test"
class KQAProConfig(datasets.BuilderConfig):
"""BuilderConfig for KQA Pro."""
def __init__(self, **kwargs):
"""BuilderConfig for KQA Pro.
Args:
**kwargs: keyword arguments forwarded to super.
"""
super(KQAProConfig, self).__init__(**kwargs)
class KQAPro(datasets.GeneratorBasedBuilder):
"""KQAPro: A large scale knowledge-based question answering dataset."""
BUILDER_CONFIGS = [
KQAProConfig(
name=_TRAIN_CONFIG_NAME,
description="KQA Pro"
),
KQAProConfig(
name=_TEST_CONFIG_NAME,
description="KQA Pro"
),
]
def _info(self):
if self.config.name == _TEST_CONFIG_NAME:
return datasets.DatasetInfo(
description=_DESCRIPTION,
features=datasets.Features(
{
"question": datasets.Value("string"),
"choices": datasets.features.Sequence(datasets.Value("string")),
}
),
supervised_keys=None,
homepage=_URL,
citation=_CITATION,
)
return datasets.DatasetInfo(
description=_DESCRIPTION,
features=datasets.Features(
{
"question": datasets.Value("string"),
"sparql": datasets.Value("string"),
"program": datasets.features.Sequence(
{
"function": datasets.Value("string"),
"dependencies": datasets.features.Sequence(datasets.Value("int32")),
"inputs": datasets.features.Sequence(datasets.Value("string"))
}
),
"choices": datasets.features.Sequence(datasets.Value("string")),
"answer": datasets.Value("string")
}
),
# No default supervised_keys (as we have to pass both question
# and context as input).
supervised_keys=None,
homepage=_URL,
citation=_CITATION,
)
def _split_generators(self, dl_manager):
downloaded_files = dl_manager.download_and_extract(_URLS)
if self.config.name == _TEST_CONFIG_NAME:
return [
datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={
"filepath": downloaded_files["test"]})
]
return [
datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={
"filepath": downloaded_files["train"]}),
datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={
"filepath": downloaded_files["val"]})
]
def _generate_examples(self, filepath):
"""This function returns the examples in the raw (text) form."""
logger.info("generating examples from = %s", filepath)
with open(filepath, encoding="utf-8") as f:
kqa = json.load(f)
for idx, sample in enumerate(kqa):
yield idx, sample
|