File size: 2,988 Bytes
b33835e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
44c7a01
 
b418ffd
2df2d46
44c7a01
b33835e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2df2d46
b33835e
 
2df2d46
b33835e
 
2df2d46
 
 
b33835e
2df2d46
b33835e
fcd37c6
 
 
803f291
 
fcd37c6
 
b418ffd
803f291
fcd37c6
 
 
 
 
b418ffd
fcd37c6
 
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

import json
import pandas as pd
import datasets

logger = datasets.logging.get_logger(__name__)

_DESCRIPTION = """\
Klue Machine Reading Comprehension Data
"""

_URL = "https://huggingface.co/datasets/LeverageX/klue-mrc/resolve/main/"
_URLS = {
    "train_data": _URL + "klue-mrc-v1.1_train.json",
    "validation_data": _URL + "klue-mrc-v1.1_dev.json",
}

class KoreanNewspaper(datasets.GeneratorBasedBuilder):

    BUILDER_CONFIGS = [
        datasets.BuilderConfig(
            name="KLUE Machine Reading Comprehension",
            version=datasets.Version("1.0.0", ""),
            description="For LeverageX Project",
        ),
    ]

    def _info(self):
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features(
                {
                    "context": datasets.Value("string"),
                    "question": datasets.Value("string"),
                    "answers":dict,
                    "guid":datasets.Value("string"),

                }
            ),
            # No default supervised_keys (as we have to pass both question
            # and context as input).
            supervised_keys=None,
            homepage="https://klue-benchmark.com/tasks/70/overview/description",
        )

    def _split_generators(self, dl_manager):
        downloaded_files = dl_manager.download_and_extract(_URLS)
        return [
            datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train_data"]}),
            datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={"filepath": downloaded_files["validation_data"]}),
        ]

    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 :
            data = json.load(f)

        data = data['data']

        for info in data :
            title = info['title']  
            news_category = info['news_category']
            source = info['source']

            paragraphs = info['paragraphs']

            if len(paragraphs) == 0 :
                continue

            context = paragraphs[0]['context']
            qas = paragraphs[0]['qas']

            for q in qas :
                question = q['question']

                answer_key = 'answers' if len(q['answers']) > 0 else 'plausible_answers'            
                answer = q[answer_key][0]
                answer_text = answer['text']
                answer_start = answer['answer_start']
                answer_data = {'answer_start' : [answer_start], 'text': [answer_text]}
                guid = q['guid']

                yield key, {
                    "guid" : guid,
                    "context" : context,
                    "question" : question,
                    "answers" : answer_data
                }
                key += 1