Delete _cacapo.py
Browse files- _cacapo.py +0 -132
_cacapo.py
DELETED
@@ -1,132 +0,0 @@
|
|
1 |
-
#!/usr/bin/env python3
|
2 |
-
|
3 |
-
"""
|
4 |
-
The script used to load the dataset from the original source.
|
5 |
-
"""
|
6 |
-
|
7 |
-
import os
|
8 |
-
import xml.etree.cElementTree as ET
|
9 |
-
from collections import defaultdict
|
10 |
-
from glob import glob
|
11 |
-
from os.path import join as pjoin
|
12 |
-
from pathlib import Path
|
13 |
-
|
14 |
-
import datasets
|
15 |
-
|
16 |
-
_CITATION = """\
|
17 |
-
@inproceedings{van2020cacapo,
|
18 |
-
title={The CACAPO dataset: A multilingual, multi-domain dataset for neural pipeline and end-to-end data-to-text generation},
|
19 |
-
author={van der Lee, Chris and Emmery, Chris and Wubben, Sander and Krahmer, Emiel},
|
20 |
-
booktitle={Proceedings of the 13th International Conference on Natural Language Generation},
|
21 |
-
pages={68--79},
|
22 |
-
year={2020}
|
23 |
-
}
|
24 |
-
"""
|
25 |
-
|
26 |
-
_DESCRIPTION = """\
|
27 |
-
CACAPO is a data-to-text dataset that contains sentences from news reports for the sports, weather, stock, and incidents domain in English and Dutch, aligned with relevant attribute-value paired data. This is the first data-to-text dataset based on "naturally occurring" human-written texts (i.e., texts that were not collected in a task-based setting), that covers various domains, as well as multiple languages. """
|
28 |
-
_URL = "https://github.com/TallChris91/CACAPO-Dataset"
|
29 |
-
_LICENSE = "CC BY 4.0"
|
30 |
-
|
31 |
-
def et_to_dict(tree):
|
32 |
-
dct = {tree.tag: {} if tree.attrib else None}
|
33 |
-
children = list(tree)
|
34 |
-
if children:
|
35 |
-
dd = defaultdict(list)
|
36 |
-
for dc in map(et_to_dict, children):
|
37 |
-
for k, v in dc.items():
|
38 |
-
dd[k].append(v)
|
39 |
-
dct = {tree.tag: dd}
|
40 |
-
if tree.attrib:
|
41 |
-
dct[tree.tag].update((k, v) for k, v in tree.attrib.items())
|
42 |
-
if tree.text:
|
43 |
-
text = tree.text.strip()
|
44 |
-
if children or tree.attrib:
|
45 |
-
if text:
|
46 |
-
dct[tree.tag]["text"] = text
|
47 |
-
else:
|
48 |
-
dct[tree.tag] = text
|
49 |
-
return dct
|
50 |
-
|
51 |
-
|
52 |
-
def parse_entry(entry):
|
53 |
-
res = {}
|
54 |
-
otriple_set_list = entry["originaltripleset"]
|
55 |
-
res["original_triple_sets"] = [{"otriple_set": otriple_set["otriple"]} for otriple_set in otriple_set_list]
|
56 |
-
mtriple_set_list = entry["modifiedtripleset"]
|
57 |
-
res["modified_triple_sets"] = [{"mtriple_set": mtriple_set["mtriple"]} for mtriple_set in mtriple_set_list]
|
58 |
-
res["category"] = entry["category"]
|
59 |
-
res["eid"] = entry["eid"]
|
60 |
-
res["size"] = int(entry["size"])
|
61 |
-
res["lex"] = {
|
62 |
-
"comment": [ex.get("comment", "") for ex in entry.get("lex", [])],
|
63 |
-
"lid": [ex.get("lid", "") for ex in entry.get("lex", [])],
|
64 |
-
"text": [ex.get("text", "") for ex in entry.get("lex", [])],
|
65 |
-
}
|
66 |
-
return res
|
67 |
-
|
68 |
-
|
69 |
-
def xml_file_to_examples(filename):
|
70 |
-
tree = ET.parse(filename).getroot()
|
71 |
-
|
72 |
-
examples = et_to_dict(tree)["benchmark"]["entries"][0]["entry"]
|
73 |
-
return [parse_entry(entry) for entry in examples]
|
74 |
-
|
75 |
-
|
76 |
-
class CACAPO(datasets.GeneratorBasedBuilder):
|
77 |
-
VERSION = datasets.Version("1.0.0")
|
78 |
-
|
79 |
-
def _info(self):
|
80 |
-
return datasets.DatasetInfo(
|
81 |
-
description=_DESCRIPTION,
|
82 |
-
features=datasets.Features({
|
83 |
-
"category": datasets.Value("string"),
|
84 |
-
"lang": datasets.Value("string"),
|
85 |
-
"size": datasets.Value("int32"),
|
86 |
-
"eid": datasets.Value("string"),
|
87 |
-
"original_triple_sets": datasets.Sequence(
|
88 |
-
{"otriple_set": datasets.Sequence(datasets.Value("string"))}
|
89 |
-
),
|
90 |
-
"modified_triple_sets": datasets.Sequence(
|
91 |
-
{"mtriple_set": datasets.Sequence(datasets.Value("string"))}
|
92 |
-
),
|
93 |
-
"lex": datasets.Sequence(
|
94 |
-
{
|
95 |
-
"comment": datasets.Value("string"),
|
96 |
-
"lid": datasets.Value("string"),
|
97 |
-
"text": datasets.Value("string"),
|
98 |
-
}
|
99 |
-
),
|
100 |
-
}),
|
101 |
-
supervised_keys=None,
|
102 |
-
homepage=_URL,
|
103 |
-
citation=_CITATION,
|
104 |
-
license=_LICENSE,
|
105 |
-
)
|
106 |
-
|
107 |
-
def _split_generators(self, dl_manager):
|
108 |
-
"""Returns SplitGenerators."""
|
109 |
-
return [
|
110 |
-
datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filedirs": ["Incidents", "Sports", "Stocks", "Weather"], "split" : "train"}),
|
111 |
-
datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={"filedirs": ["Incidents", "Sports", "Stocks", "Weather"], "split" : "dev"}),
|
112 |
-
datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"filedirs": ["Incidents", "Sports", "Stocks", "Weather"], "split" : "test"}),
|
113 |
-
]
|
114 |
-
|
115 |
-
def _generate_examples(self, filedirs, split):
|
116 |
-
"""Yields examples."""
|
117 |
-
id_ = 0
|
118 |
-
|
119 |
-
for lang in ["en", "nl"]:
|
120 |
-
for filedir in filedirs:
|
121 |
-
xml_file = os.path.join(lang, filedir, f"WebNLGFormat{split.title()}.xml")
|
122 |
-
|
123 |
-
for exple_dict in xml_file_to_examples(xml_file):
|
124 |
-
exple_dict["category"] = filedir
|
125 |
-
exple_dict["lang"] = lang
|
126 |
-
id_ += 1
|
127 |
-
yield id_, exple_dict
|
128 |
-
|
129 |
-
|
130 |
-
if __name__ == '__main__':
|
131 |
-
dataset = datasets.load_dataset(__file__)
|
132 |
-
dataset.push_to_hub("kasnerz/cacapo")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|