Delete _wikitabletext.py
Browse files- _wikitabletext.py +0 -85
_wikitabletext.py
DELETED
@@ -1,85 +0,0 @@
|
|
1 |
-
#!/usr/bin/env python3
|
2 |
-
|
3 |
-
"""
|
4 |
-
The script used to load the dataset from the original source.
|
5 |
-
"""
|
6 |
-
|
7 |
-
|
8 |
-
import os
|
9 |
-
from collections import defaultdict
|
10 |
-
|
11 |
-
import json
|
12 |
-
import datasets
|
13 |
-
|
14 |
-
_CITATION = """\
|
15 |
-
@inproceedings{bao2018table,
|
16 |
-
title={Table-to-Text: Describing Table Region with Natural Language},
|
17 |
-
author={Junwei Bao and Duyu Tang and Nan Duan and Zhao Yan and Yuanhua Lv and Ming Zhou and Tiejun Zhao},
|
18 |
-
booktitle={AAAI},
|
19 |
-
url={https://www.aaai.org/ocs/index.php/AAAI/AAAI18/paper/download/16138/16782},
|
20 |
-
year={2018}
|
21 |
-
}
|
22 |
-
"""
|
23 |
-
|
24 |
-
_DESCRIPTION = """\
|
25 |
-
WikiTableText contains 5,000 tables from Wikipedia, each of which has at least 3 rows and 2 columns.
|
26 |
-
For each table, three rows are selected resulting in 15,000 rows that are further used for manual annotation."""
|
27 |
-
|
28 |
-
_URL = "https://github.com/msra-nlc/Table2Text"
|
29 |
-
_LICENSE = "CC BY 4.0"
|
30 |
-
|
31 |
-
class WikiTableText(datasets.GeneratorBasedBuilder):
|
32 |
-
VERSION = datasets.Version("1.0.0")
|
33 |
-
|
34 |
-
def _info(self):
|
35 |
-
return datasets.DatasetInfo(
|
36 |
-
description=_DESCRIPTION,
|
37 |
-
features=datasets.Features({
|
38 |
-
"headers": datasets.Value("string"),
|
39 |
-
"content": datasets.Value("string"),
|
40 |
-
"row_number": datasets.Value("string"),
|
41 |
-
"reference": datasets.Value("string"),
|
42 |
-
}),
|
43 |
-
supervised_keys=None,
|
44 |
-
homepage=_URL,
|
45 |
-
citation=_CITATION,
|
46 |
-
license=_LICENSE,
|
47 |
-
)
|
48 |
-
|
49 |
-
def _split_generators(self, dl_manager):
|
50 |
-
"""Returns SplitGenerators."""
|
51 |
-
return [
|
52 |
-
datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"split" : "train"}),
|
53 |
-
datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={"split" : "dev"}),
|
54 |
-
datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"split" : "test"}),
|
55 |
-
]
|
56 |
-
|
57 |
-
def _normalize(self, lst):
|
58 |
-
lst = lst.split("_||_")
|
59 |
-
lst = [x.replace("_$$_", " ") for x in lst]
|
60 |
-
lst = [x.replace("_", "").strip() for x in lst]
|
61 |
-
|
62 |
-
return lst
|
63 |
-
|
64 |
-
def _generate_examples(self, split):
|
65 |
-
"""Yields examples."""
|
66 |
-
id_ = 0
|
67 |
-
|
68 |
-
with open(f"MSRA_NLC.Table2Text.{split}") as f:
|
69 |
-
for line in f.readlines():
|
70 |
-
items = line.split("\t")
|
71 |
-
e = {
|
72 |
-
"row_number" : items[0],
|
73 |
-
"headers" : self._normalize(items[1]),
|
74 |
-
"content" : self._normalize(items[2]),
|
75 |
-
"reference" : self._normalize(items[3])[0],
|
76 |
-
}
|
77 |
-
|
78 |
-
id_ += 1
|
79 |
-
yield id_, e
|
80 |
-
|
81 |
-
|
82 |
-
|
83 |
-
if __name__ == '__main__':
|
84 |
-
dataset = datasets.load_dataset(__file__)
|
85 |
-
dataset.push_to_hub("kasnerz/wikitabletext")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|